An all-inorganic, stable, and highly active tetraruthenium homogeneous catalyst for water oxidation

Article abstract of DOI:10.1002/anie.200705652

(Chemical Equation Presented) Oxidation without organics: A tetraruthenium polyoxometalate (see picture; Ru blue, O red, Si yellow, W black) catalyzes the rapid oxidation of H₂O to O₂ in water at ambient temperature, and shows considerable stability under turnover conditions. The complex was characterized by several methods, including X-ray crystallography and cyclic voltammetry.

Full text of DOI:10.1002/anie.200705652

Communications
DOI: 10.1002/anie.200705652
Oxygen Generation
An All-Inorganic, Stable, and Highly Active
Tetraruthenium Homogeneous Catalyst for Water
Oxidation**
Yurii V. Geletii, Bogdan Botar,* Paul Kögerler, Daniel A. Hillesheim,
Djamaladdin G. Musaev, and Craig L. Hill*
Angewandte
Chemie
3896
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3896 –3899

Angewandte
Chemie
The design of viable and well-defined molecular catalysts for
water oxidation in conjunction with the conversion of solar
into chemical energy is both intellectually and practically
important.[1–19] Increasingly, high-resolution crystal structures
and physicochemical investigations of the Mn4Ca-centered
water oxidation–oxygen evolving center (OEC) in photo-
system II, and studies of model compounds, are providing
more insight into the properties of this multimetal biocata-
lyst.[11,14,16–18,20–25] In early work, Shafirovich and co-workers
documented that several nonmolecular materials (metal
oxides/hydroxides) catalyzed H2O oxidation to O2 over a
wide pH range.[1,2] Classic work by Meyer and co-workers on
the soluble and structurally defined functional H2O oxidation
catalyst, [(bpy)2(H2O)RuORu(H2O)(bpy)2]4+,[3] and subse-
quent work by several groups[15–17,19] has clarified some
mechanistic aspects of water oxidation and defined key
challenges. Despite all these achievements and ongoing
research, however, stable and rapid molecular and homoge-
neous catalysts for H2O oxidation and O2 generation have yet
to be achieved. Some organic ligands are adequately stable in
photovoltaic devices (for example, [Ru(bpy)3]3+/2+ units in
Grätzel cells last 10000 + h),[10] but they are not stable in
water-splitting devices. Extensive homogenous catalytic oxi-
dation studies suggest that likely intermediates in H2O
oxidation would degrade all organic ligands, a point consistent
with the findings involving molecular H2O oxidation catalysts
reported to date.[3,7,8,14–17,19] Thus, the need to develop highly
active and stable H2O oxidation catalysts remains of consid-
erable importance. Based on the reported {Ru2} catalysts for
H2O oxidation, documented polyoxometalate (POM) com-
plexes with multinuclear d-electron-containing centers capa-
ble of accepting several electrons needed for H2O oxida-
tion,[26–31] and the report by Shannon and co-workers of
electrocatalytic O2 evolution by the Neumann–Khenkin
to as yet unknown products,[34] but these findings led us to
further, related studies. Herein we report the synthesis and
characterization of Rb8K2[{Ru4O4(OH)2(H2O)4}(g-
SiW10O36)2]·25H2O (1), an oxidatively and hydrolytically
stable complex that addresses some of the core challenges:
it catalyzes the rapid oxidation of H2O to O2, does so in
aqueous solution at pH 7, and is quite stable under turnover
conditions.
Reaction of [g-SiW10O36]8ꢀ with two equivalents of RuIII in
acidic aqueous solutions (pH 1.6) at ambient temperature,
followed by addition of RbCl, affords crystals of polyanion
salt 1 in ca. 40% yield. The X-ray crystal structure of 1 reveals
the same “out-of-pocket” d-metal coordination polyhedra
observed in water-soluble g-diiron derivatives;[35,36] namely,
the ruthenium centers are corner-sharing and not ligated to
the central SiO4 unit (Figure 1). The two “out-of-pocket” g-
{SiW10Ru2} monomeric units are rotated by 908 around the
Figure 1. Structure of the polyanion in 1, highlighting the central
{Ru4(m-O)4(m-OH)2(H2O)4}6+ core (ball-and-stick representation,
Ru blue, m-O red, O(H2) orange; hydrogen atoms omitted for clarity)
and the slightly distorted {Ru4} tetrahedron (transparent blue). The
polytungstate fragments are shown as gray polyhedra, and Si as yellow
spheres.
complex, [WZnRu2(OH)(H2O)(ZnW9O34)2]11ꢀ [32]
we initially
,
prepared [{RuIII2(OH)2(H2O)2}(g-SiW10O36)]4ꢀ [33] and demon-
strated that it did catalyze H2O oxidation.[34] Unfortunately,
this complex is unstable in aqueous solution, and transforms
vertical C2 axis relative to one another, defining overall D2d
symmetry for the polyanion. The staggered structure facili-
tates incorporation of a {Ru4(m-O)4(m-OH)2(H2O)4}6+ core, in
which the four ruthenium centers span a slightly distorted
tetrahedron with Ru···Ru distances of 3.47–3.66 Š. The
adjacent ruthenium centers within each g-{SiW10Ru2} unit
are bridged by hydroxo ligands (bond valence sum 1.29), and
oxo ligands bridge the ruthenium centers of different mono-
meric units.
[*] Dr. B. Botar, Dr. P. Kögerler
Institut für Festkörperforschung
Forschungszentrum Jülich GmbH
52425 Jülich (Germany)
Fax: (+49)2461-612-620
E-mail: b.botar@fz-juelich.de
Dr. Y. V. Geletii, D. A. Hillesheim, Prof. Dr. C. L. Hill
Department of Chemistry
Emory University
ꢀ ꢀ
1515 Dickey Drive, Atlanta, GA 30322 (USA)
Fax: (+1)404-727-6076
E-mail: chill@emory.edu
The presence of m-oxo Ru O Ru bridges (indicated by a
band at 487 cmꢀ1 in the Raman spectrum) is consistent with
other structural reports on dimeric ruthenium-containing
POMs.[31,37–39] Five lines of evidence indicate that, during the
synthesis of 1, RuIII (RuCl3·H2O reactant) is oxidized by O2 to
give a {RuIV4} complex: 1) magnetic properties are consistent
with diamagnetic d4 RuIV and not paramagnetic d5 RuIII
Dir. Dr. D. G. Musaev
Cherry L. Emerson Center for Scientific Computation
Emory University
1515 Dickey Drive, Atlanta, GA 30322 (USA)
[**] This work was supported by the following grants to C.L.H. &
D.G.M.: U.S. Department of Energy (DE-FG02-03ER15461 and DE-
FG02-06ER06-15). We thank Ken Hardcastle and Rui Cao for X-ray
crystallographic efforts.
ꢀ
centers; 2) bond valence sums on the Ru O bond lengths
yield oxidation states of 4.04 and 4.17 for the two crystallo-
graphically independent ruthenium atoms; 3) elemental anal-
ysis (number of countercations) and charge balance consid-
erations are in accord with a {RuIV4} core; 4) 1 is EPR silent;
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 3896 –3899
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3897

Communications
and 5) the rest potentials of the cyclic voltammograms are
RuV/IV couples in 1. At potentials lower than the rest
potential, three reduction peaks at 530, 370, and ꢀ170 mV
and the corresponding re-oxidations at 640, 480, and 290 mV,
respectively, are observed, which is consistent with three RuIV/
III reductions in 1. A peak for a fourth RuIV/III reduction at ca.
ꢀ350 mVoverlaps with an intense peak for WVI/V reduction at
ca. ꢀ460 mV (not shown in Figure 2a). Thus, under acidic
conditions, two RuIV centers are electrochemically oxidized to
RuV, while all four RuIV centers can be reduced to RuIII. At
higher pH, the rest potentials are lower, but the CV traces are
less informative because the ruthenium complexes exhibit
complex acid–base equilibria under these conditions (Fig-
ure 2a). Significantly, at pH 7.0 there is a very sharp increase
in current at E > 900 mV; a very weak peak at circa 930 mV is
hardly seen. The currents at 950–1050 mV are several-fold
higher in the presence of 0.6 mm 1; this behavior is consistent
with electrocatalytic H2O oxidation at these unusually low
potentials (Figure 2a). This result led us to evaluate 1 as a
catalyst for homogeneous H2O oxidation in aqueous solution.
The test reaction for H2O oxidation was the well-studied
process in Equation (1).[2,11,14] A CV of [Ru(bpy)3]2+ at pH 7.0
consistent with a {RuIV4} core. In addition, the oxidation of
RuIII in aqueous solution to give a RuIV-containing POM is
known.[31,38]
In preparation for catalytic studies, several techniques
were used to further characterize oxidation states and
potentials of the ruthenium centers and the protonation
states of the {Ru4(m-O)4(m-OH)2(H2O)4}6+ core. Repeated
acid–base titrations in both directions and monitored both by
pH values and the UV/Vis spectra indicate that 1 has two pKa
values in the pH range 3.5–4.5, and these titrations are
reversible from pH 2.5–7.5. Thermogravimetric analysis and
differential thermal calorimetry indicate the anion in 1 is
stable to 4508C.
Cyclic voltammograms (CVs) of aqueous solutions of 1
are pH-dependent (Figure 2a). At pH 1.0, two oxidation
peaks, at ca. 940 and at 1050 mV are observed in a scan from
the rest potential (800 mV) to positive potentials and
corresponding reduction peaks at ca. 750 and 965 mV are
observed on the reverse scan. These are assigned to two
4 ½RuðbpyÞ3Š3þ þ 2 H2O ! 4 ½RuðbpyÞ3Š2þ þ O2 þ 4 Hþ
ð1Þ
shows reversible behavior (Ea = 1100, Ec = 940 mV, and Ia/Ic
ꢁ 1, Figure 2b), and this potential is higher than both the
most-positive peak observed for 1 (at pH 1.0) and the
standard potential for the four-electron oxidation of H2O to
O2 (Eo = 0.82 V vs. NHE at pH 7).
At a very low concentration of 1, 0.006–0.029 mm,
catalytic currents are observed at potentials corresponding
to the oxidation of [Ru(bpy)3]2+ to [Ru(bpy)3]3+ [see Equa-
tion (1) and Figure 2b]. The peak current increases almost
linearly with [1]. An increase in the anodic peak is accom-
panied by a complete disappearance of a cathodic peak. The
contribution of 1 to the catalytic current in the absence of
[Ru(bpy)3]3+ is negligible (Figure 2b).
These promising electrochemical findings led us to inves-
tigate catalysis of Equation (1) by 1, monitored by both
accumulated [Ru(bpy)3]2+ spectrophotometrically and O2
chromatographically (GC with TC detector). In the absence
of 1 the typical reaction time, t1/2, for Equation (1) is
> 30 min. Addition of very small amounts of 1 (ca. 0.5–
1.5 mm) considerably shortens the reaction time. The reaction
does not obey a simple exponential law (the apparent rate
constant decreases with time) and is complete in 30–40 s after
addition of 1.5 mm 1. In a typical (unoptimized) reaction,
1.2 mm (12 mmol) [Ru(bpy)3]3+ in the presence of 10 mm
(0.1 mmol) of 1 gives 1.07 mm (10.7 mmol) [Ru(bpy)3]2+ (ca.
90% based on initial [Ru(bpy)3]3+) and 1.78 mmol of O2 (ca.
66% based on initial [Ru(bpy)3]3+). To rule out catalysis by
RuO2, a well-known catalyst for H2O oxidation[5] and a
conceivable decomposition product of 1, we examined the
activity of RuCl3 under otherwise identical conditions. The
lifetime of [Ru(bpy)3]3+ was only half as long in the presence
of 10 mm RuCl3 than with no catalyst. At the equivalent
ruthenium concentration of 1 (2.5 mm), the rate of Equa-
tion (1) was ca. 102 times higher. The data are consistent with
Figure 2. a) CVs of 1 mm 1 in 0.1m HCl (pH 1.0, red curve), in 0.4m
sodium acetate buffer (pH 4.7, blue curve), and of 0.6 mm 1 (pH 7.0,
black curve). Scans start at the rest potentials (800, 600, and 400 mV,
respectively), and potentials are relative to a Ag/AgCl (3m NaCl)
reference electrode. b) CVs of 1 mm [Ru(bpy)3]2+ at different concen-
trations of 1 (0, 0.006, 0.012, and 0.029 mm: black, red, green, and
blue curves, respectively) and of 0.029 mm 1 in the absence of
[Ru(bpy)3]2+ (pink curve), pH 7.0. Conditions: scan rate 25 mVsꢀ1, in
0.025m sodium phosphate buffer and 0.15m NaCl for pH 7.0.
3898
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3896 –3899

Angewandte
Chemie
[3] S. W. Gersten, G. J. Samuels, T. J. Meyer, J. Am. Chem. Soc.
catalysis by 1 and not RuO2. Addition of 10 mm K8[g-
SiW10O36] to a 10 mm solution of RuCl3 (precursors of 1) has
no effect on the reaction rate, because 1 forms more slowly at
this high buffered pH. Although the turnovers (TON) and
stoichiometries (ratios of oxygen atoms present in the system)
make it clear that the oxygen atoms in O2 must derive from
H2O and not another sources, we nonetheless, conducted a
reaction according to Equation 1 using isotopically labeled
water (18O, 5 and 10%). A proportionally enriched O2 was
confirmed by gas chromatography using EI-mass spectromet-
ric detection (see the Supporting Information, Figure S9 for
details). In addition, the ratio of 16O18O:18O18O (m/z ratio
34:36) also confirmed that the oxygen atoms in O2 were from
H2O.
Thus, the present study clearly documents the formation
of O2 from H2O, confirmation of which has been a major issue
for most of the reported water oxidations by soluble
complexes.[8,11,14,15] This work demonstrates that H2O is not
only the solvent but also the source of oxygen atoms in O2.
Indeed, the number of oxygen atoms in the O2 produced far
exceeds the number of oxygen atoms in 1.
1982, 104, 4029.
[4] F. P. Rotzinger, S. Munavalli, P. Comte, J. K. Hurst, M. Grätzel,
F.-J. Pern, A. J. Frank, J. Am. Chem. Soc. 1987, 109, 6619.
[5] A. Harriman, M.-C. Richoux, P. A. Christensen, S. Mosseri, P.
Neta, J. Chem. Soc. Faraday Trans. 1 1987, 83, 3001.
[6] T. Meyer, Acc. Chem. Res. 1989, 22, 163.
[7] W. Ruettinger, G. C. Dismukes, Chem. Rev. 1997, 97, 1.
[8] J. Limburg, J. S. Vrettos, L. M. Liable-Sands, A. L. Rheingold,
R. H. Crabtree, G. W. Brudvig, Science 1999, 283, 1524.
[9] M. Hara, C. C. Waraksa, J. T. Lean, B. A. Lewis, T. E. Mallouk, J.
Phys. Chem. A 2000, 104, 5275.
[10] M. Grätzel, Nature 2001, 414, 338.
[11] M. Yagi, M. Kaneko, Chem. Rev. 2001, 101, 21.
[12] J. Messinger, Phys. Chem. 2004, 6, 4764.
[13] A. S. Polo, M. K. Itokazu, N. Y. M. Iha, Coord. Chem. Rev. 2004,
248, 1343.
[14] S. Mukhopadhyay, S. K. Mandal, S. Bhaduri, W. H. Armstrong,
Chem. Rev. 2004, 104, 3981.
[15] J. K. Hurst, Coord. Chem. Rev. 2005, 249, 313.
[16] R. Zong, R. Thummel, J. Am. Chem. Soc. 2005, 127, 12802.
[17] X. Yang, M.-H. Baik, J. Am. Chem. Soc. 2006, 128, 7476.
[18] J. Yano, J. Kern, K. Sauer, M. J. Latimer, Y. Pushkar, J.
Biesiadka, B. Loll, W. Saenger, J. Messinger, A. Zouni, V. K.
Yachandra, Science 2006, 314, 821.
Although the kinetics of H2O oxidation catalyzed by 1 are
complicated, which might indicate involvement of multiple
intermediates, four lines of evidence indicate that 1 stays
intact in many oxidation states in neutral aqueous media:
1) the acid–base titrations are fully reversible up to pH 7.5;
2) the cyclic voltammograms are reproducible and show
reversible oxidation and reduction of multiple ruthenium
centers; 3) the RuCl3 control experiment shows H2O oxida-
tion activity two orders of magnitude lower than 1; and 4) the
turnover number for 1 in the catalytic H2O oxidation experi-
ments is ca. 18 (moles O2/mol 1) or 120 (moles [Ru(bpy)3]3+/
mol 1). In summary, we document a rapid and stable catalyst
for H2O oxidation to O2 that is operational in H2O under
ambient conditions.
[19] F. Liu, T. Cardolaccia, B. J. Hornstein, J. R. Schoonover, T. J.
Meyer, J. Am. Chem. Soc. 2007, 129, 2446.
[20] T. G. Carrell, S. Cohen, G. C. Dismukes, J. Mol. Catal. A 2002,
187, 3.
[21] H. Chen, J. W. Faller, R. H. Crabtree, G. W. Brudvig, J. Am.
Chem. Soc. 2004, 126, 7345.
[22] K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber, S. Iwata,
Science 2004, 303, 1831.
[23] A. M. Tyryshkin, R. K. Watt, S. V. Baranov, J. Dasgupta, M. P.
Hendrich, G. C. Dismukes, Biochemistry 2006, 45, 12876.
[24] R. Tagore, R. H. Crabtree, G. W. Brudvig, Inorg. Chem. 2007, 46,
2193.
[25] W. Hillier, T. Wydrzynski, Coord. Chem. Rev. 2008, 252, 306.
[26] POM reviews are given in the Supporting Information.
[27] B. Keita, I. M. Mbomekalle, Y. W. Lu, L. Nadjo, P. Berthet, T. M.
Anderson, C. L. Hill, Eur. J. Inorg. Chem. 2004, 3462.
[28] B. Keita, I. M. Mbomekalle, L. Nadjo, T. M. Anderson, C. L.
Hill, Inorg. Chem. 2004, 43, 3257.
[29] T. M. Anderson, R. Cao, W. A. Neiwert, K. I. Hardcastle, C. L.
Hill, M. Ammam, B. Keita, L. Nadjo, Eur. J. Inorg. Chem. 2005,
1770.
[30] B. Keita, P. Mialane, F. SØcheresse, P. d. Oliveira, L. Nadjo,
Electrochem. Commun. 2007, 9, 164.
Experimental Section
Rb8K2[{Ru4O4(OH)2(H2O)4}(g-SiW10O36)2]·25H2O (1): The synthesis
of 1 and its characterization by single crystal X-ray diffraction and
eight other techniques are described in detail in the Supporting
Information.
Catalytic water oxidation and O2 generation: The kinetics of
Equation (1) monitored by production of both [Ru(bpy)3]2+ at 454 nm
and O2 by GC are described in detail in the Supporting Information.
[31] M. Sadakane, D. Tsukuma, M. H. Dickman, B. S. Bassil, U.
Kortz, M. Capron, W. Ueda, Dalton Trans. 2007, 2833.
[32] A. R. Howells, A. Sankarraj, C. Shannon, J. Am. Chem. Soc.
2004, 126, 12258.
[33] D. Quiꢀonero, Y. Wang, K. Morokuma, L. A. Khavrutskii, B.
Botar, Y. V. Geletii, C. L. Hill, D. G. Musaev, J. Phys. Chem. B
2006, 110, 170.
Received: December 10, 2007
Revised: February 14, 2008
Published online: March 19, 2008
[34] C. L. Hill et al., unpublished results.
[35] B. Botar, Y. V. Geletii, P. Kꢁgerler, D. G. Musaev, K. Morokuma,
I. A. Weinstock, C. L. Hill, J. Am. Chem. Soc. 2006, 128, 11268.
[36] B. Botar, P. Kꢁgerler, C. L. Hill, Inorg. Chem. 2007, 46, 5398.
[37] C. Rong, M. T. Pope, J. Am. Chem. Soc. 1992, 114, 2932.
[38] W. J. Randall, T. J. R. Weakley, R. G. Finke, Inorg. Chem. 1993,
32, 1068.
Keywords: dioxygen generation · electrochemistry ·
homogeneous catalysis · oxidation · polyoxometalates
.
[1] V. Y. Shafirovich, A. E. Shilov, Kinet. Katal. 1979, 20, 1156.
[2] V. Y. Shafirovich, N. K. Khannanov, V. V. Strelets, Nouv. J.
Chim. 1980, 4, 81.
[39] M. Sadakane, M. Higashijima, Dalton Trans. 2003, 659.
Angew. Chem. Int. Ed. 2008, 47, 3896 –3899
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3899