Electrochemical oxidation of methanol using alcohol-soluble Ru/Pt and Ru/Pd catalysts

Article abstract of DOI:10.1016/j.ica.2007.10.046

The heterobimetallic Ru/Pt and Ru/Pd complexes [η⁵-C₅H₄CH₂CH₂N(CH₃)₂ · HI]Ru(PPh₃)(μ-I)(μ-dppm)PtCl₂ (7), [η⁵-C₅H₄CH_2<

Full text of DOI:10.1016/j.ica.2007.10.046

Available online at www.sciencedirect.com
Inorganica Chimica Acta 361 (2008) 3237–3246
www.elsevier.com/locate/ica
Electrochemical oxidation of methanol using alcohol-soluble
Ru/Pt and Ru/Pd catalysts
*
Daniel Serra, Lisa McElwee-White
Department of Chemistry and Center for Catalysis, University of Florida, Gainesville, FL 32611-7200, USA
Received 26 September 2007; accepted 28 October 2007
Available online 6 November 2007
Dedicated to Robert J. Angelici
Abstract
The heterobimetallic Ru/Pt and Ru/Pd complexes [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-dppm)PtCl2 (7), [g5-C5H4CH2CH2-
N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-dppm)PtI2 (8), [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-dppm)PdCl2 (9), and [g5-C5H4CH2CH2-
N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-dppm)PdI2 (10) were prepared by the reaction of [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)I(j1-dppm) (6)
with Pt(COD)Cl2, Pt(COD)I2, and Pd(COD)Cl2, respectively. Electronic interaction between the two metals is significant for the
iodide-bridged compounds 7–10, as evidenced by the shifts of their redox potentials in comparison to the mononuclear complexes.
The electrochemical oxidation of methanol was carried out with heterobimetallic complexes 7–10 and leads to the formation of dim-
ethoxymethane (DMM) and methyl formate (MF) as the major oxidation products. The chloride complexes 7 and 9 are the most active
catalysts, as evidenced by their TON and current efficiencies. Addition of water at the beginning of the electrolysis results in increased
formation of the more oxidized product MF along with higher current efficiencies and TON.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Heterobimetallic complexes; Methanol oxidation; Electrocatalysis
1. Introduction
the heterobimetallic complexes Cp(PPh3)Ru(l-Cl)(l-
dppm)PtCl2 (1) [20] Cp(PPh3)Ru(l-Cl)(l-dppm)PdCl2 (2)
[21] and Cp(PPh3)RuCl(l-dppm)AuCl (3) [21], followed
by investigation of their electrochemical properties. Elec-
trooxidation of methanol with complexes 1–3 provided evi-
dence that interaction of the two metal centers was
beneficial as these catalysts showed significant increases
in the current efficiencies compared to those of the mono-
nuclear model compounds CpRu(j2-dppm)Cl [22] or
CpRu(PPh3)2Cl [23].
Previous studies on electrooxidation of methanol by 1–3
and related compounds [22] were carried out in non-polar
aprotic media (1,2-dichloroethane) due to the insolubility
of the complexes in polar solvents. In order to facilitate
the electrochemistry by moving to protic media, we have
now prepared derivatives of 1–3 in which the cyclopenta-
dienyl ligand is functionalized with a quaternized amino
group to improve solubility in polar solvents. This
Interest in potential catalytic applications has led to
extensive research on heterobimetallic complexes, due to
the possibility of exploiting the different reactivity of the
two metals in chemical transformations [1–7]. The close
proximity of two adjacent metal centers offers the possibil-
ity of cooperative reactivity and/or different mechanistic
roles of the metal centers [2,8–14]. It has been shown that
the cooperative effect in heterobimetallic complexes can
enhance catalytic activities or can result in unique proper-
ties not observed in monometallic models [12,15–19].
The interest in the possibility of such a cooperative effect
between two different metal centers in homogeneous elec-
trochemical oxidation of alcohols led to preparation of
*
Corresponding author. Tel.: +1 352 392 8768; fax: +1 352 846 0296.
E-mail address: lmwhite@chem.ufl.edu (L. McElwee-White).
0020-1693/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2007.10.046

3238
D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
approach has been used in the past to adapt organometallic
catalysts for use in water [24–28]. Although the compounds
presented in this work are only minimally soluble in water,
they are soluble in methanol, allowing the electrooxidation
of methanol to be carried out in the absence of additional
solvent. We now report the synthesis of new heterobimetal-
lic Ru/Pt and Ru/Pd complexes and their catalytic behav-
ior in the electrochemical oxidation of neat methanol.
trode. The Eo values for the ferrocenium/ferrocene couple
in DCE/0.1 M TBAT and MeOH/0.1 M TBABF4 were
+0.67 and +0.61 V.
Electrolysis products were analyzed by gas chromatog-
raphy on a Shimadzu GC-17A chromatograph using a
15 m · 0.32 mm column of ATꢁ-WAX (Alltech, 0.5 lm
film) on fused silica. The column was attached to the injec-
tion port with a neutral 5 m · 0.32 mm ATꢁ-WAX deacti-
vated guard column. The products produced during the
electrolysis of methanol were quantitatively determined
with the use of a known amount of n-heptane as an internal
standard. Product identification was confirmed by compar-
ison of the retention times of the oxidation product with
authentic samples.
2. Experimental
2.1. General considerations
All reactions and manipulations were performed under
an argon atmosphere using standard Schlenk techniques.
Pentane and ethyl ether were dried by distillation from
Na/Ph2CO. Acetonitrile and 1,2-dichloroethane were dried
by distillation from CaH2. Methanol was dried by distilla-
tion from magnesium. Benzene and dichloromethane were
dried using an MBraun solvent purification system. All
solvents were saturated with argon prior to use. All deuter-
ated solvents (Cambridge Isotope Laboratories) for NMR
2.2. Synthesis of the complexes
2.2.1. [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)2I (5)
A solution of 4 (1.52 g, 1.91 mmol) in methylene chlo-
ride (100 mL) was washed with 150 mL solution of
degassed 1 M HCl. The organic layer was dried under
MgSO4 and the solvent was evaporated. The resulting
orange solid, after drying, was reacted with sodium iodide
(2.85 g, 19.1 mmol) in degassed dichloromethane at room
temperature. After 48 h, the resulting mixture was washed
with H2O (3 · 50 mL). After drying the organic layer,
evaporation of the solvent and recrystallization from
CH2Cl2/hexanes, compound 5 was obtained as a red brown
solid (1.68 g, 85.5% yield). IR (neat film): m(N–H)
measurements were degassed via freeze–pump–thaw cycles
1
and stored over molecular sieves (4 A). H and 31P{1H}
˚
NMR spectra were recorded at room temperature on a
Varian Mercury 300 spectrometer operating at 300 and
121 MHz, respectively, with chemical shifts (d, ppm)
reported relative to residual solvent peaks (1H NMR) or
85% H3PO4 (31P NMR). IR spectra were obtained as neat
films on NaCl using a Varian 2000 FT-IR spectrophotom-
eter. [(g5-C5H4CH2CH2N(CH3)2 Æ HCl]Ru(PPh3)2Cl (4)
was prepared according to published procedures [29]. All
other starting materials were purchased in reagent grade
purity and used without further purification.
Electrochemical experiments were performed at ambient
temperature in a glove box using an EG&G PAR model
263A potentiostat/galvanostat. Cyclic voltammograms
(CV) were recorded in 3.5 mL of DCE/0.1 M tetrabutylam-
monium trifluoromethanesulfonate (TBAT) or MeOH/
0.1 M tetrabutylammonium tetrafluoroborate (TBABF4)
at ambient temperature under nitrogen. All potentials are
reported versus NHE and referenced to Ag/Ag+. The refer-
ence electrode consisted of a silver wire immersed in an ace-
tonitrile solution containing freshly prepared 0.01 M
AgNO3 and 0.1 M TBAT or TBABF4. The Ag+ solution
and silver wire were contained in a 75 mm glass tube fitted
at the bottom with a Vycor tip. Cyclic voltammetry was
performed in a three-compartment H-cell separated by a
medium-porosity sintered glass frit in 2.5–3.5 mL of
DCE/0.1 M TBAT or MeOH/0.1 M TBABF4 at room
temperature under nitrogen. A highly polished glassy car-
bon electrode (diameter 3 mm) was the working electrode
and a platinum flag was used as the counter electrode.
All electrochemical measurements were performed inside
the glove box. The constant potential electrolysis was per-
formed in similar equipment except that the glassy carbon
working electrode was replaced by a vitreous carbon elec-
2669 cmꢀ1
.
1H NMR (CDCl3): d 2.77–2.90 (m, 2H,
CH2CH2N(CH3)2 Æ H+), 2.85 (s, 6H, N(CH3)2 Æ H+), 3.20–
3.30 (m, 2H, CH2CH2N(CH3)2 Æ H+), 3.53 (br s, 2H,
C5H4), 3.82 (br s, 2H, C5H4), 7.10–7.40 (m, 30H,
P(C6H5)3), 12.38 (br s 1H, N–H). 31P{1H} NMR (CDCl3):
d 37.85. HRMS (ESI): calcd for C45H45INP2Ru m/z
890.1115 (MꢀI)+, found 890.1125.
2.2.2. [g5 ꢀ C5H4CH2CH2N(CH3)2 Æ HI]
Ru(PPh3) (j1 ꢀ dppm)I (6)
A 500 mL Schlenk flask was charged with 5 (1.68 g,
1.65 mmol), bis(diphenylphosphino)methane (dppm) (1.25 g,
3.25 mmol) and 250 mL of THF. The orange mixture was
stirred for 7 days at ambient temperature and the solvent
was evaporated on a vacuum line. The resulting crude
mixture contained j1-dppm compound 6, excess dppm
and PPh3. After evaporation of the solvent on a rotary
evaporator, purification was achieved by recrystallization
from dichloromethane/hexane to afford 1.71 g of a red-
orange solid (92% yield). IR (neat film): m(N–H)
1
2666 cmꢀ1. H NMR (CDCl3): d 2.79 (s, 6H, N(CH3)2 Æ
H+), 2.87–3.30 (m, 4H, CH2CH2N(CH3)2 Æ H+), 3.34 (br
s, 1H, Ph2PCHHPPh2), 3.47 (br s, 1H, Ph2PCHHPPh2),
4.22 (br s, 2H, C5H4), 4.52 (br s, 2H, C5H4), 6.89–7.79
(m, 35H, (C6H5)2PCH2P(C6H5)2 + P(C6H5)3), 10.67 (br s,
1H, N–H). 31P{1H} NMR (CDCl3): d 43.85 (dd, Ru–
PPh2CH2PPh2, JPP = 3, 41 Hz), 33.89 (dd, Ru–PPh3,
JPP = 21, 41 Hz), ꢀ26.64 (dd, Ru–PPh2CH2PPh2, JPP
=

D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
3239
3, 21 Hz). HRMS (ESI): calcd for C52H52INP3Ru m/z
1012.1401 (MꢀI)+, found 1012.1414.
2.2.3. [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-
dppm)PtCl2 (7)
4.40 (br s, 2H, C5H4), 4.50 (br s, 1H, Ph2PCHHPPh2),
5.26 (br s, 2H, C5H4), 6.19 (br s, 1H, N–H), 6.06–8.07
(m, 35H, (C6H5)2PCH2P(C6H5)2 + P(C6H5)3). 31P{1H}
NMR (CDCl3): d 51.22 (dd, Ru–PPh2CH2PPh2, JPP =
23, 35 Hz), 41.26 (dd, Ru–PPh3, JPP = 7, 35 Hz), 11.73
(dd, Ru–PPh2CH2PPh2–Pd, JPP = 7, 23 Hz). HRMS
(ESI): calcd for C52H52Cl2INP3PdRu m/z 1187.9813
(MꢀI)+, found 1187.9836.
2.2.6. [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)
(l-dppm)Pdl2 (10)
In
a 50 mL Schlenk flask, complex 6 (0.300 g,
0.263 mmol) and Pt(COD)Cl2 (98.5 mg, 0.263 mmol) were
dissolved in 30 mL of benzene. The yellow solution was
stirred at ambient temperature for 24 h then the solution
was dried under vacuum and recrystallized from CH2Cl2/
pentane to yield 0.151 g (41%) of yellow solid. IR (neat
film): m(N–H) 2672 cmꢀ1
.
1H NMR (CDCl3): d 2.73
In a 50 mL Schlenk flask, complex 6 (0.254 g,
(m, 1H, Ph2PCHHPPh2), 3.03 (s, 6H, N(CH3)2 Æ H+),
3.13–3.81 (m, 4H, CH2CH2N(CH3)2 Æ H+ + 1H, Ph2PCH-
HPPh2), 4.44 (br s, 2H, C5H4), 5.19 (br s, 2H, C5H4),
5.65–8.13 (m, 35H, (C6H5)2PCH2P(C6H5)2 + P(C6H5)3),
7.93 (br s, 1H, N–H). 31P{1H} NMR (CDCl3): d 47.37
(dd, Ru–PPh2CH2PPh2, JPP = 13, 35 Hz), 40.98 (dd, Ru–
PPh3, JPP = 3, 35 Hz), ꢀ7.21 (dd, Ru–PPh2CH2PPh2–Pd,
JPP = 3, 13 Hz, JPtP = 1872 Hz). HRMS (FAB): calcd for
C52H52Cl2INP3PtRu m/z 1277.0425 (MꢀI)+, found
1277.0426.
0.223 mmol) and Pd(COD)Cl2 (63.8 mg, 0.223 mmol) were
dissolved in 30 mL of dichloromethane. The yellow solu-
tion was stirred at ambient temperature for 2 h, then
10 equiv. of NaI were added to the solution and stirred
for 24 h at room temperature. The solution was filtered
under N2 and the filtrate was dried under vacuum and
recrystallized from CH2Cl2/pentane to yield 0.195 g
(66.4%) of a dark purple solid. IR (neat film): m(N–H)
2668 cmꢀ1
. d 2.62 (br s, 1H,
1H NMR (CDCl3):
Ph2PCHHPPh2), 3.08 (s, 6H, N(CH3)2 Æ H+), 3.47–3.78
(m, 4H, CH2CH2N(CH3)2 Æ H+), 4.41 (br s, 2H, C5H4),
4.50 (br s, 1H, Ph2PCHHPPh2), 5.27 (br s, 2H, C5H4),
6.11 (br s,1H, N–H), 5.71–8.12 (m, 35H, (C6H5)2P-
CH2P(C6H5)2 + P(C6H5)3). 31P{1H} NMR (CDCl3): d
49.45 (dd, Ru–PPh2CH2PPh2, JPP = 22, 35 Hz), 39.38
(dd, Ru–PPh3, JPP = 6, 35 Hz), 13.83 (dd, Ru–PPh2CH2-
PPh2–Pd, JPP = 6, 22 Hz). HRMS (ESI): calcd for
C52H52I3NP3PdRu m/z 1371.8525 (MꢀI)+, found
1371.8558.
2.2.4. [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-
dppm)PtI2 (8)
In
a 50 mL Schlenk flask, complex 6 (0.200 g,
0.175 mmol) and Pt(COD)I2 (97.8 mg, 0.175 mmol) were
dissolved in 30 mL of dichloromethane. The yellow-orange
solution was stirred at ambient temperature for 2 h then
the solution was dried under vacuum and recrystallized
from CH2Cl2/pentane to yield 0.185 g (64.5%) of an orange
solid. IR (neat film): m(N–H) 2669 cmꢀ1
.
1H NMR
(CDCl3): d 2.58 (m, 1H, Ph2PCHHPPh2), 3.04 (s, 6H,
N(CH3)2 Æ H+), 3.49–3.86 (m, 4H, CH2CH2N(CH3)2 Æ H+),
4.40 (br s, 2H, C5H4), 4.60 (m, 1H, Ph2PCHHPPh2), 5.24
(br s, 2H, C5H4), 6.81–8.12 (m, 35H, (C6H5)2PCH2-
P(C6H5)2 + P(C6H5)3), 10.96 (br s, 1H, N–H). 31P{1H}
3. Results and discussion
3.1. Synthesis of complexes 5 and 6
The preparation and spectroscopic data for [g5-
C5H4CH2CH2N(CH3)2 Æ HCl]Ru(PPh3)2Cl (4) have been
previously reported [29]. Treatment of 4 with sodium
iodide in dichloromethane at room temperature for two
days affords [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)2I
(5) as a red brown solid in greater than 85% yield (Scheme
1). Preparation of the j1-dppm complex 6 is a direct reac-
tion of the iodide compound 5 with bis(diphenylphos-
phino)methane (dppm) in tetrahydrofuran (Scheme 1).
This step required several days of reaction at room temper-
ature. Increasing the temperature does speed up disappear-
ance of the starting material but also decreases the yield of
6 due to the formation of the j2-dppm side product [g5-
C5H4CH2CH2N(CH3)2 Æ HI](PPh3)Ru(j2-dppm)I. Crude
product from room temperature reaction can be recrystal-
lized from dichloromethane/hexane to obtain pure 6 (92%
yield), although several recrystallizations may be required.
Impure material can also be purified by column chroma-
tography. The red-orange solid complex 6 is air stable in
the solid state but a solution of the complex slowly
degrades after several hours of exposure to air.
NMR (CDCl3): d 46.65 (dd, Ru–PPh2CH2PPh2, JPP
=
12, 35 Hz), 39.36 (dd, Ru–PPh3, JPP = 3, 34 Hz), ꢀ1.84
(dd, Ru–PPh2CH2PPh2–Pt, JPP = 3, 12 Hz, JPPt = 1872
Hz). HRMS (FAB): calcd for C52H52I3NP3PtRu m/z
1460.9138 (MꢀI)+, found 1460.9116.
2.2.5. [g5-C5H4CH2CH2N(CH3)2 Æ HI]Ru(PPh3)(l-I)(l-
dppm)PdCl2 (9)
In
a 50 mL Schlenk flask, complex 6 (0.300 g,
0.263 mmol) and Pd(COD)Cl2 (75.2 mg, 0.26 mmol) were
dissolved in 30 mL of dry methanol. The dark red solution
was stirred at ambient temperature for 1 h until a
red-orange precipitate formed. The solution was concen-
trated under vacuum to a small volume (4–5 mL), filtered
through a medium swivel frit, washed with hexanes and
dried under vacuum. The product was recrystallized from
CH2Cl2/pentane to yield 0.243 g (65.9%) of a red-orange
solid. IR (neat film): m(N–H) 2672 cmꢀ1
.
1H NMR
(CDCl3): d 2.69 (br s, 1H, Ph2PCHHPPh2), 3.07 (s, 6H,
N(CH3)2 Æ H+), 3.34–3.88 (m, 4H, CH2CH2N(CH3)2 Æ H+),

3240
D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
Scheme 1.
3.2. Synthesis of the Ru/Pt complexes 7 and 8
3.4. Cyclic voltammetry of complexes 5–10
The reaction of the Ru complex 6 with Pt(COD)Cl2 or
Pt(COD)I2 at room temperature afforded the heterobime-
tallic complexes 7 and 8 as a yellow and a red-orange
powder, respectively (Scheme 2). The formation of complex
7 is usually accompanied by a small amount of Pt(j2-
dppm)Cl2 formed by dppm transfer from [g5-C5H4CH2-
CH2N(CH3)2 Æ HI](PPh3)Ru(j1-dppm)I (6) to the Pt center.
Complexes 7 and 8 are air stable in the solid state but show
signs of decomposition when exposed to air for several days.
The CV data for complexes 5–10 are listed in Table 1.
The cyclic voltammograms of the monometallic ruthenium
complexes generally exhibit two waves. The first one is
assigned to the Ru(II/III) redox couple while the second
one is assigned to the Ru(III/IV) couple. In the case of
the amino-substituted compounds a third wave is also
observed due to the oxidation of the amino moiety. This
electrochemical process has been previously described for
aliphatic amines [30,31]. Protonation of the amine moiety
should give electrochemically inactive ammonium ions that
exhibit the regular waves observed for the mono and het-
erobimetallic complexes [21,32]. However, it appears that
the cyclic voltammograms of all the ammonium complexes
performed in 1,2-dichloroethane (DCE) always exhibit a
small wave due to the oxidation of the amine moiety simi-
lar to the oxidation wave of N,N-dimethylphenethylamine
(Table 1, Figs. 1 and 2). This phenomenon is possibly
due to the presence of small amounts of non-protonated
amine in the samples. When the cyclic voltammetry is per-
formed in methanol, this amine oxidation wave is not
observed.
3.3. Synthesis of the Ru/Pd complexes 9 and 10
The Ru/Pd complex 9 was prepared in a similar manner
to complexes
7
and
8
by reaction of [g5-
C5H4CH2CH2N(CH3)2 Æ HI](PPh3)Ru(j1-dppm)I (6) with
Pd(COD)Cl2 in dichloromethane at room temperature
(Scheme 3). The I-bridged complex can be obtained as a
yellow powder after several successive recrystallizations
from dichloromethane/pentane (66% yield). The synthesis
of complex 10 requires an additional step since Pd(COD)I2
is not commercially available. Instead, 10 was synthesized
by generating complex 9 in situ in dichloromethane, fol-
lowed by 48 h reaction at room temperature with five equiv
of NaI (Scheme 3). The resulting compound can be purified
by filtration and successive recrystallizations (CH2Cl2/pen-
tane) yielding a dark purple solid in 65% yield. Complexes
9 and 10 show very similar stabilities as compared to their
Ru/Pt analogues 7 and 8.
Electronic interactions between the metals in the I-
bridged heterobimetallic complexes 7–10 are evidenced by
significant redox potential shifts compared to those of the
monometallic models compounds 5 or 6. The cyclic vol-
tammograms of the Ru/Pt complexes 7 and 8 performed
in DCE exhibit four redox waves each. The second and
fourth waves are due to the Ru(II/III) and Ru(III/IV) cou-
Scheme 2.

D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
3241
Scheme 3.
Table 1
Formal potentials for complexes 5–10
b
b
Complexes
Solvent Couple Epa (V) Couple
Epa (V) E1/2 (V) Couple
Epa (V) E1/2 (V) Couple
Epa (V)
N,N-Dimethylphenethylamine DCE
amine
1.35
Ru(II/III)
Ru(III/IV)
5
DCE
MeOH amine
aminea 0.60
Ru(II/III) 0.83
Ru(II/III) 0.82
0.76
0.76
Ru(III/IV) 1.64
Ru(III/IV)
c
6
DCE
MeOH amine
aminea 0.78
Ru(II/III) 1.05
Ru(II/III) 1.00
0.96
0.94
Ru(III/IV) 1.95
Ru(III/IV)
c
7
DCE
MeOH amine
aminea 1.26
Ru(II/III) 1.36
Ru(II/III) 1.32
Pt(II/IV) 1.72
Pt(II/IV)
Ru(III/IV) 1.98
Ru(III/IV)
c
c
8
DCE
MeOH amine
aminea 0.65
Ru(II/III) 0.93
Ru(II/III) 0.83
0.82
Pt(II/IV) 1.56
Pt(II/IV) 1.55
1.45
1.52
1.47
Ru(III/IV) 2.01
Ru(III/IV)
c
9
DCE
MeOH amine
aminea 1.12
Ru(II/III) 1.35
Ru(II/III) 1.33
Pd(II/IV) 1.67
Ru(III/IV) 2.04
c
c
Pd(II/IV)
Ru(III/IV)
10
DCE
MeOH amine
aminea 0.64
Ru(II/III) 1.00
Ru(II/III) 0.93
Pd(II/IV) 1.57
Pd(II/IV) 1.46
Ru(III/IV) 2.09
c
Ru(III/IV)
All potentials obtained in 0.1 M DCE/TBAT (tetrabutylammonium triflate) or in 0.1 M MeOH/TBABF4 (tetrabutylammonium tetrafluoroborate),
reported vs. NHE and referenced to Ag/Ag+.
a
Due to the presence of non protonated amine in samples.
E1/2 reported for reversible waves.
Not observed.
b
c
ples, while the third wave is assigned to the Pt(II/IV) redox
couple. A small first wave is also observed and has been
assigned to the oxidation of the residual non-protonated
amines. As expected, electronic interactions between the
metal centers are observed for those complexes due to the
presence of the iodide bridge. This behavior has already
been described for the analogous complexes Cp(L)Ru
(l-X)(l-dppm)MX2 (L = PPh3, X = Cl, I, M = Pt, Pd
[20,21,32]; L = CO, X = I, M = Pd, Pt [33]). The potentials
for the Ru(III/IV) redox couple are not affected by the
changes of the ligands around the Pt center for 7 and 8.
In contrast, comparison of the Pt substituted chloride 7
with the iodide complex analogue 8 reveals a significant
430 mV negative shift of the Ru(II/III) redox potential as
a result of an increase of the electron density at the Pt
center. A similar effect is also observed at the Pt center,
with the Pt(II/IV) redox couple easier to oxidize by
160 mV upon substitution of chloride ligands by iodides.

3242
D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
complexes show the effect of increasing the electronic den-
sity at the Pd center since the amine oxidation is shifted
negative by 480 mV when chlorides are substituted by
iodides at the Pd center.
Comparisons of the redox potentials of heterobimetallic
complexes 7–10 with the mononuclear complexes 5 indicate
that the binuclear complexes exhibit shifts that are consis-
tent with electron donation through the iodide bridges. Sig-
nificant electron donation from the Ru center to the more
electron poor Pt or Pd is evident in chloride complexes 7
and 9 as evidenced by the higher Ru(II/III) redox potential
when compared to the monomeric complex 5 (1.36 V,
1.35 V and 1.05 V for 7, 9 and 5, respectively). In contrast,
the presence of the iodide ligands at the Pt or Pd center of 8
and 10 disfavors electron donation from the Ru through
the iodide bridge. As a result the Ru(II/III) oxidation
waves are shifted to lower potential due to a more elec-
tron-rich ruthenium center for complexes 8 and 10.
Fig. 1. Cyclic voltammograms of complex 9 in DCE; glassy carbon
working electrode; Ag/Ag+ reference electrode.
3.5. Electrochemical oxidation of methanol with
heterobimetallic complexes 7–10
Cyclic voltammetry in the presence of methanol is a con-
venient way to screen for catalytic activity for methanol
electrooxidation. The cyclic voltammograms of the Ru/Pt
chloride complex 7, and the Ru/Pd complexes 9 and 10
show current increases starting at the oxidation wave of
the second metal center (Pt(II/IV) and Pd(II/IV) couples,
respectively) indicating the onset of catalytic methanol oxi-
dation (Figs. 1 and 2). On the other hand, the cyclic vol-
tammogram of the iodide heterobimetallic Ru/Pt complex
8 shows only a small increase of the current at the Pt(II/
IV) wave.
Previous bulk electrolysis experiments with heterobime-
tallic complexes 1 [20], 2 [21] and 3 [21] and related
complexes [22] were performed in the non-polar solvents
1,2-dichloroethane or dichloromethane due to poor solu-
bility of complexes 1–3 in polar media. The use of a chlo-
rinated solvent for electrochemical oxidation is
problematic since a high concentration of the supporting
electrolyte (0.7 M TBAT) is necessary to reach an accept-
able conductivity and electron transfer rate during the elec-
trolysis processes. Complexes 7–10 were prepared with the
expectation that their increased solubility in alcoholic or
aqueous solution would improve the electron transfer
properties during the electrolysis.
Although the amino-substituted complexes 7–10 were
found to have poor solubility in water, these compounds
are soluble in methanol. This allows the use of methanol
as both reactant and solvent for the electrolysis experi-
ments. For complexes 7–10, using the high potential of
1.7 V versus NHE for bulk electrolyses was necessary in
DCE but unnecessary in methanol. The lower fixed poten-
tial of 1.5 V versus NHE and lower electrolyte concentra-
tion of 0.1 M TBABF4 afforded good results in methanol.
Control experiments without catalysts did not show evi-
dence of methanol oxidation products at that potential.
Fig. 2. Cyclic voltammograms of complex 10 in DCE; glassy carbon
working electrode; Ag/Ag+ reference electrode.
It is interesting that changing the terminal halides on plat-
inum affects not only the Ru(II/III) and Pt(II/IV) redox
waves but also affects the redox potentials of the amine oxi-
dation waves of the cyclopentadienyl ring. A considerable
610 mV negative shift is observed for the oxidation of the
amine when the electron density is increased at the Pt
center.
It is not surprising to observe the same behavior for the
analogous palladium complexes 9 and 10 (Figs. 1 and 2).
The cyclic voltammograms for the Ru/Pd compounds 9
and 10 exhibit irreversible oxidations at 1.35 V and
1.00 V versus NHE, respectively, assigned to their Ru(II/
III) couples. As also seen for 7 and 8, a 350 mV negative
shift is observed for diiodide complex 10 when compared
to dichloride 9. Complexes 9 and 10 exhibit also quasi-
reversible Pd(II/IV) redox waves at 1.67 V and 1.57 V ver-
sus NHE, respectively. The Pd(II/IV) couple for the iodide
compound 10 is found to be easier to oxidize by 100 mV
when compared to the Pd(II/IV) couple for the chloride
compound 9. Again the residual non-protonated amine

D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
3243
Table 2
Bulk electrolysis data for the oxidation of methanol by complexes 7–10 and 5
Charge (C)
Product ratios (DMM/MF)a,b
Complex: Ru/Pt (7)
Ru/Pt (8)
Ru/Pd (9)
Ru/Pd (10)
Ru (5)
25
50
75
100
150
200
1
0.25
0.48
0.69
0.94
0.41
1.18
1.89
4.39
1.66
1.76
2.34
3.17
3.82
0.13
1.59
2.09
2.44
3.80
4.50
0.45
0.64
0.86
1.01
1.08
6.27
(At charge X)
1.16 (300 C)
12.99 (300 C)
1.02 (250 C)
Current efficiencies (%) after 200 C
TON after 200 C
51
11
16
4
44
9
23
4
10
2
Current efficiencies (%) (at charge X)
TON (at charge X)
57 (300 C)
19 (300 C)
23 (300 C)
9 (300 C)
27 (250 C)
6 (250 C)
a
Electrolyses were performed at 1.5 V vs. NHE in pure methanol. A 10 mM catalyst concentration was used for each experiment.
Determined by GC with respect to n-heptane as an internal standard. Each ratio is reported as an average of 2–3 experiments.
b
As previously described [20–23,32], the homogeneous
electrochemical oxidation of methanol results in formation
of formaldehyde, formic acid, dimethoxymethane (DMM),
and methyl formate (MF) as oxidation products. Formal-
dehyde and formic acid are the initial two-electron and
four-electron oxidation products from the oxidation of
methanol; however in the presence of excess methanol they
undergo fast condensation leading to the formation of
DMM and MF, respectively. The product ratios, current
efficiencies and turnover numbers (TON) for the electrooxi-
dation of methanol by catalysts 7–10 are summarized in
Table 2. For comparison purposes, results for the mono-
meric model complex 5 are also presented. The current effi-
ciencies are the ratio of the charge necessary to produce the
observed yields of DMM and MF to the total charge
passed during the bulk electrolysis.
Fig. 3. Product evolution for the electrolysis of Ru/Pt complex 7 in 0.1 M
MeOH/TBABF4.
The chloride Ru/Pt complex 7 and Ru/Pd complex 9
gave moderately higher current efficiencies, (51% and
44%, respectively) when compared to compounds 1, 2
and 3. Experiments in pure methanol have demonstrated
that this solvent is more favorable for the catalysis since
the electron transfer kinetics are improved in more polar
solvent and the lower potential of 1.5 V can be used. The
product ratios and product evolutions (Table 2, Figs. 3
and 4) from complexes 7 and 9 show also that the partition-
ing of oxidation products for those two complexes is differ-
ent. In the early stage of the catalysis, complex 7 forms the
four-electron oxidation product MF in higher concentra-
tion. However at a late stage of the process it appears that
the production of the two-electron oxidation product
DMM becomes more favored. The overall process shows
formation of the two oxidation products in similar concen-
trations. Complex 10 shows similar behavior compared to
7 but with a lower activity (23% CE and 4 TON for com-
plex 10 versus 51% CE and 11 TON for 7) (Fig. 5). For
Ru/Pd complex 9, the behavior is different as observed in
Fig. 4. From the early stage of the catalysis until the end,
the two-electron oxidation product DMM is favored as
Fig. 4. Product evolution for the electrolysis of Ru/Pd complex 9 in 0.1 M
MeOH/TBABF4.

3244
D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
described by the increase of the product ratio from 1.66 to
3.82. This behavior seems to be similar to the monomeric
model complex 5 but with better current efficiency for
Ru/Pd complex 9. Complex 8 was found to be the least
active catalyst with only 16% current efficiency. Close
observation of the data shows fast increases of the product
ratios from 0.41 to 12.99 until 300 C of charge are passed.
In fact this complex selectively favors the formation of the
two-electron oxidation product DMM while the concentra-
tion of MF stagnates during the process (Fig. 6). The
results observed for those four compounds show that the
chloride-substituted Pt or Pd complexes 7 and 9 exhibit
better electrocatalytic results as observed by the higher cur-
rent efficiencies and TON. In contrast the iodide complexes
8 and 10 prove to be less active.
The electrochemical oxidation of methanol by com-
plexes 7–10 was also performed under ‘‘wet conditions’’
by adding 10 lL of water into the system before electrolysis
(Table 3, Figs. 7–10). Previous results showed that the pres-
ence of water at an earlier stage of the electrolysis (before it
is generated by condensation) usually favors the formation
of the more oxidized product MF [22]. For the Ru/Pt com-
plexes 7 and 8, the presence of water (Table 3) decreases the
DMM/MF product ratios when compared to those from
the ‘‘dry conditions’’ (Table 2). Obviously for those two
compounds, water favors the formation of MF in higher
concentration (Figs. 7 and 8). Increases of the current effi-
ciencies and TON are also observed for complexes 7 and 8
when the electrolysis is performed under ‘‘wet conditions’’.
Again comparison of complex 7 with compound 8 shows
that chloride substitution at the Pt center significantly
enhances the activity for electrocatalytic oxidation of meth-
anol (65% CE and 13 TON for 7 versus 21% CE and 4
TON for 8).
Fig. 5. Product evolution for the electrolysis of Ru/Pd complex 10 in
0.1 M MeOH/TBABF4.
In the case of the Ru/Pd complexes 9 and 10, similar
behavior is observed (Figs. 9 and 10). About twice the cur-
rent efficiency and twice the TON are obtained with the
chloride substituted complex (64% CE and 13 TON for 9
Fig. 6. Product evolution for the electrolysis of Ru/Pt complex 8 in 0.1 M
MeOH/TBABF4.
Table 3
Bulk electrolysis data for the oxidation of methanol under wet conditionsc by complexes 7–10 and 5
Charge (C)
Product ratios (DMM/MF)a,b,c
Complex: Ru/Pt (7)
Ru/Pt (8)
Ru/Pd (9)
Ru/Pd (10)
Ru (5)
25
50
75
1
1
1
1
0.16
1
1.53
1
0.28
0.45
2.53
0.27
100
0.40
0.71
3.50
0.44
150
0.63
1.18
4.07
0.86
200
0.78
1.73
4.24
1.26
(At charge X)
0.92 (300 C)
3.37 (800 C)
4.11 (555 C)
1.79 (250 C)
Current efficiencies (%) after 200 C
TON after 200 C
65
13
21
4
63
13
30
6
Current efficiencies (%) (at charge X)
TON (at charge X)
76 (300 C)
24 (300 C)
43 (800 C)
41 (800 C)
98 (555 C)
58 (555 C)
34 (250 C)
9 (250 C)
a
Electrolyses were performed at 1.5 V vs. NHE in pure methanol. Ten mM catalyst concentration was used for each experiment.
Determined by GC with respect to n-heptane as an internal standard. Each ratio is reported as an average of 2–3 experiments.
Ten lL of water was added into the system before electrolysis.
b
c

D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
3245
Fig. 7. Product evolution for the electrolysis of Ru/Pt complex 7 in 0.1 M
MeOH/TBABF4 in wet methanol.
Fig. 10. Product evolution for the electrolysis of Ru/Pd complex 10 in 0.1
M MeOH/TBABF4 in wet methanol.
and 30% CE and 6 TON for 10). Among the catalysts,
complex 7 is again more active followed closely by the anal-
ogous Ru/Pd chloride complex 9.
Experiments were carried out to determine catalyst life-
time and how much product could be formed before the
catalyst decomposed. Compounds 8 and 9 were used for
the experiments and deactivation of the catalysts was deter-
mined by the loss of color in the reaction side of the cell
(Table 3). For complex 8, 800 C of charge were passed in
about 26 h. In this case 41 turnovers were obtained, form-
ing about 100 lL of DMM (1.127 mmol) and 21 lL of MF
(0.3 mmol) with an efficiency of 43% (GC determination)
(Fig. 8). For complex 9, only 555 C of charge could be
passed into the system in about the same time but with
superior results. About 98% current efficiency and 58
TON were reached with complex 9 (24 h), forming about
167 lL of DMM (1.897 mmol) and 28 lL of MF
(0.46 mmol) (Fig. 9).
Fig. 8. Product evolution for the electrolysis of Ru/Pt complex 8 in 0.1 M
MeOH/TBABF4 in wet methanol.
4. Conclusion
The catalytic activities for the electrochemical oxidation
of methanol were investigated for a series of heterobimetal-
lic Ru/Pt and Ru/Pd complexes bearing amino-substituted
Cp rings. Although these compounds were initially synthe-
sized for application in aqueous media, they were found to
have poor solubility in water. However, in contrast to the
original catalysts 1–3, compounds 5–10 are soluble in
methanol.
Addition of methanol during cyclic voltammetry exper-
iments with 7–10 showed an enhancement of the current
typical of catalytic activity for the electrooxidation of
methanol. As previously described [21,23], DMM and
MF are the oxidation products detected during bulk elec-
trolyses in pure methanol. The differences among the com-
plexes can be observed in the product ratios, current
efficiencies and TON. The chloride-substituted Pt or Pt
complexes 7 and 9 are the most active catalysts, having
Fig. 9. Product evolution for the electrolysis of Ru/Pt complex 9 in 0.1 M
MeOH/TBABF4 in wet methanol.

3246
D. Serra, L. McElwee-White / Inorganica Chimica Acta 361 (2008) 3237–3246
[11] R.D. Adams, T.S. Barnard, Organometallics 17 (1998) 2885.
[12] L. Quebatte, R. Scopelliti, K. Severin, Eur. J. Inorg. Chem. (2006)
231.
[13] C. Cornelissen, G. Erker, G. Kehr, R. Fro¨hlich, Organometallics 24
(2005) 214.
[14] J.A. McCleverty, M.D. Ward, Acc. Chem. Res. 31 (1998) 842.
[15] M. Shibasaki, Y. Yamamoto (Eds.), Multimetallic Catalysts in
Organic Synthesis, Wiley-VCH Weinheim, 2004.
[16] N.L. James, J.J. Dennett, G. Nilsson, A. Rosborough, M.J. Fergu-
son, M. Wang, R. McDonald, J. Takats, Organometallics 23 (2004)
4478.
[17] J.H. Sinfelt, in: Bimetallic Catalysts: Discovery, Concepts, and
Applications, Wiley, New York, 1983.
[18] J. Halpern, Inorg. Chim. Acta 62 (1982) 31.
[19] S.Y. Ng, L.Y. Goh, L.L. Koh, W.K. Leong, G.K. Tan, S. Ye, Y. Zhu,
Eur. J. Inorg. Chem. (2006) 663.
the highest TON and current efficiencies (51% CE and 11
TON for 7 and 44% CE and 9 TON for 9). The iodide-
substituted complexes 8 and 10 are less active but still more
active than the mononuclear model complex [g5-C5H4-
CH2CH2NMe2 Æ HI]Ru(PPh3)2I (5). Increasing the elec-
tronic density at the non-Ru metal appears to deactivate
the catalyst as demonstrated by the poorer results from
complexes 8 and 10 (16% CE and 4 TON for 8 and 23%
CE and 4 TON for 10). Experiments in which water is
added at the beginning of the catalysis show that the for-
mation of the more oxidized product MF is favored in
the presence of water along with higher current efficiencies
and TON (up to 98% CE and 58 TON with catalyst 9).
[20] M.E. Tess, P.L. Hill, K.E. Torraca, M.E. Kerr, K.A. Abboud, L.
McElwee-White, Inorg. Chem. 39 (2000) 3942.
[21] G. Matare, M.E. Tess, K.A. Abboud, Y. Yang, L. McElwee-White,
Organometallics 21 (2002) 711.
[22] Y. Yang, L. McElwee-White, Dalton Trans. (2004) 2352.
[23] C.R. Anthony, L. McElwee-White, J. Mol. Catal. A-Chem. 227
(2005) 113.
Acknowledgement
This work was supported by NASA Glenn Research
Center under Grant No. NAG 3-2930.
References
[24] A. Avey, D.R. Tyler, Organometallics 11 (1992) 3856.
[25] A.I. Philippopoulos, R. Bau, R. Poilblanc, N. Hadjiliadis, Inorg.
Chem. 37 (1998) 4822.
[26] P. Jutzi, U. Siemeling, J. Organomet. Chem. 500 (1995) 175.
[27] O.R. Allen, L. Croll, A.L. Gott, R.J. Knox, P.C. McGowan,
Organometallics 23 (2004) 288.
[28] P.W. Causey, M.C. Baird, Organometallics 23 (2004) 4486.
[29] H.-S. Chu, C.-P. Lau, K.-Y. Wong, W.-T. Wong, Organometallics 17
(1998) 2768.
[30] L.C. Portis, J.T. Klug, C.K. Mann, J. Org. Chem. 39 (1974)
3488.
[1] P. Braunstein, J. Rose, in: E.W. Abel, F.G.A. Stone, G. Wilkinson
(Eds.), Comprehensive Organometallic Chemistry II, vol. 10, Perg-
amon, Oxford, 1995, p. 351.
[2] R.D. Adams, F.A. Cotton (Eds.), Catalysis by Di- and Polynuclear
Metal Cluster Complexes, Wiley-VCH, New York, 1998.
[3] G. Suss-Fink, G. Meister, in: F.G.A. Stone, R. West (Eds.), Advances
in Organometallic Chemistry, vol. 35, Academic Press, 1993, p. 41.
[4] E.K. Van Den Beuken, B.L. Feringa, Tetrahedron 54 (1998) 12985.
[5] N. Wheatley, P. Kalck, Chem. Rev. 99 (1999) 3379.
[31] A. Adenier, M.M. Chehimi, I. Gallardo, J. Pinson, N. Vila, Langmuir
20 (2004) 8243.
[6] B.H. Xia, H.X. Zhang, C.M. Che, K.H. Leung, D.L. Phillips, N.Y.
Zhu, Z.Y. Zhou, J. Am. Chem. Soc. 125 (2003) 10362.
[32] Y. Yang, K.A. Abboud, L. McElwee-White, Dalton Trans. (2003)
4288.
[33] D. Serra, K.A. Abboud, C.R. Hilliard, L. McElwee-White, Organo-
metallics 26 (2007) 3085.
[7] N.D. Jones, B.R. James, Adv. Synth. Catal. 344 (2002) 1126.
[8] D.G. McCollum, B. Bosnich, Inorg. Chim. Acta 270 (1998) 13.
[9] B. Bosnich, Inorg. Chem. 38 (1999) 2554.
[10] D.W. Stephan, Coord. Chem. Rev. 95 (1989) 41.