Elsevier

Catalysis Today

Volume 67, Issue 4, 15 June 2001, Pages 341-355
Catalysis Today

Combinatorial discovery of bifunctional oxygen reduction — water oxidation electrocatalysts for regenerative fuel cells

https://doi.org/10.1016/S0920-5861(01)00327-3Get rights and content

Abstract

Electrode arrays containing 715 unique combinations of five elements (Pt, Ru, Os, Ir, and Rh) were prepared by borohydride reaction of aqueous metal salts, and were screened for activity as oxygen reduction and water oxidation catalysts. Using a consensus map, catalysts that showed high activity for both reactions and good resistance to anodic corrosion were identified in the Pt–Ru rich region of the Pt–Ru–Ir ternary. The ternary catalyst Pt4.5Ru4Ir0.5 (subscripts indicate atomic ratios) is significantly more active than the previously described Pt1Ir1 bifunctional catalyst for both reactions. While the best ternary catalyst is close to Pt1Ru1 in composition, the latter is unstable with respect to anodic corrosion. A detailed kinetic comparison of anodically stable catalysts Pt4.5Ru4Ir0.5 and Pt1Ir1 showed that the addition of the oxophilic element Ru increases the reaction rate by stabilizing S–O bonds (Ssurfaceatom) and accelerating the oxidative deprotonation of S–OH groups.

Introduction

Combinatorial chemistry, while most popularly used in the discovery of biochemicals and pharmaceuticals, has also for many years been used to identify and optimize inorganic materials of complex compositions [1], [2]. The combinatorial method entails the synthesis of large libraries of compounds, in which composition or processing conditions are systematically varied, followed by screening for a particular property of interest. Recently, this approach has been used with some success in the discovery of new catalysts [3], [4], [5], [6], and sophisticated methods for rapid screening of catalyst libraries have now been developed [7], [8], [9], [10].

The combinatorial approach differs from other systematic experimental methods [11], [12], [13], [14], [15] in one particularly important respect: combinatorial analysis relies on parallel or rapid serial testing of many catalyst samples, in such a way that the figure of merit (reaction rate, selectivity, stability, etc.) is determined without bias for all members of the library. More efficient optimization methods do the testing in a serial or serial–parallel manner and use the information from earlier experiments to determine where to look next. For example, in simplex optimization of catalysts, the next composition to be formulated and tested is determined by the relative merit of the data point obtained in the previous cycle. While systematic methods are particularly efficient for evaluating catalyst compositions — particularly when testing is the most time-consuming step in the process — they are put at a serious disadvantage when two or more variables (such as activity and selectivity) must be optimized simultaneously. This is because these variables will rarely optimize to the same composition. However, the combinatorial method, which provides data on all members of a large catalyst library, can be used to find consensus regions of parameter space that reasonably optimize all the variables of interest. In this paper, we describe the combinatorial discovery of bifunctional electrocatalysts. These catalysts are simultaneously optimized, using consensus maps, for activity in the reduction of oxygen to water and for the reverse reaction, the oxidation of water to oxygen. In addition, we restrict the search to catalysts that are stable in the latter anodic process in acidic electrolytes.

Bifunctional electrocatalysts are of particular interest for use in regenerative fuel cells. The possible applications of these fuel cells include unmanned platforms in low-earth orbit (LEO) and in geosynchronous-earth orbit (GEO) [16]. The permanently manned LEO space station and orbital transfer vehicles between LEO and GEO require them to be self-sufficient. They should have the versatility to convert solar energy during the sunlit portion of the orbit into fuel and to generate power during the dark portion by means of the same systems. Similar considerations apply to fuel cells for other long-term space applications, such those intended for use on planetary missions. There are also terrestrial applications for regenerative fuel cells that include back-up power, electrical load-leveling, stand-alone photovoltaic power systems [17], [18], [19], [20], and remote or portable power plants [21], [22]. In all these applications, the combined requirements for economy of space, weight, materials, and design simplicity lead to devices in which an electrolyzer and a fuel cell work in tandem as an electrical energy storage and generation system.

A regenerative fuel cell is a battery-like hydrogen/oxygen system, which offers the possibility of splitting water by electrolysis, storing the hydrogen gas (and in some applications the oxygen gas as well) and then generating electricity by the fuel cell process. It has the distinct advantage over state-of-the-art battery systems in that power and energy are separated, since the energy is directly related to the fuel storage, while the rated power depends on the electrode area. This means that simply increasing reactant storage, without changing the reactor stack(s) can increase the stored energy. The result is that the mass advantage of typical advanced batteries loses out to that of regenerative fuel cells when the discharge time is increased beyond a few tens of minutes.

Despite their apparent advantages, regenerative fuel cells are still in their early stages of development because of two limiting factors: system efficiency and cost. An important step towards the reduction of cost is the concept of using only one converter for the electrolyzer and the fuel cell. In these systems, one electrode is used solely for the oxygen reactions (oxygen evolution in the electrolysis mode, oxygen reduction in the fuel cell mode), whereas the other operates as the corresponding hydrogen electrode (hydrogen evolution in the electrolysis mode, hydrogen oxidation in the fuel cell mode). With favorable kinetics of hydrogen electrocatalysis on platinum, improvement of the kinetics of oxygen electrocatalysis for both reactions — oxygen reduction and water oxidation — is crucial. Previous studies have found that at the oxygen electrode in regenerative fuel cells, the preferred oxygen reduction catalysts have poor oxygen evolution performance, and the preferred oxygen evolution catalysts have poor fuel cell performance. Moreover, catalysts at the oxygen electrodes tend to corrode at the very positive potentials used in the electrolysis mode. This effect decreases the performance of the catalyst quickly, and so the lifetime of the catalyst is not adequate for most applications. Thus, there is a need to discover new electrocatalytic materials that can resist corrosion and also catalyze both the oxygen reduction and oxygen evolution reactions efficiently.

The first attempt to develop oxygen electrocatalysts was made using nickel and nickel oxide catalysts in the form of ‘valve-electrodes’ [23]. Since then, many studies pertaining to oxygen electrocatalysis in regenerative fuel cells have been published. The catalytic materials used are in general metal alloys and oxides. A corrosion study of several mixed oxides in alkaline electrolyte solutions has been identified sodium–platinum oxide and lead–iridium oxide compositions as promising catalysts [24]. More recent studies have shown that Pt50/Ir50 (numbers in subscripts indicate atomic ratios) or 50% Pt/50% IrO2 (wt.%) can be used in regenerative fuel cells. Reasonable efficiencies and lifetimes were achieved [25], [26], [27]. Other studies have shown that Rh/Ru (1:1)-oxide and Ir/Rh (1:2)-oxide are also promising oxygen electrode catalysts in regenerative fuel cells.

Section snippets

Synthesis of combinatorial arrays

Electrode arrays were prepared by dispersing aqueous solutions of five metal salts (RhCl3, H2PtCl6, RuCl3, OsCl3 and IrBr3) onto a Teflon-coated Toray carbon sheet, using a robotic plotter (Cartesian Technologies, PixSys 3200). The completed array contained the same total number of moles of metal at each spot. A 40-fold molar excess of 5% aqueous sodium borohydride was added to each spot, and the reduced array was thoroughly washed with nanopure water (18.3 MΩ/cm).

Screening of catalysts for anode (water oxidation) and cathode (oxygen reduction) reactions

The screening process was

Mapping and fabrication of electrode arrays

Combinations of several different elements can be tested systematically as catalysts by mapping the composition space in the manner of a phase diagram. We prepared combinations of five different elements (Pt, Ru, Rh, Ir, and Os) by mapping the four-dimensional composition space into a planar array of discrete spots. At the resolution of 10 spots along each binary edge, this full pentanary map contains 715 unique combinations. The mapping and fabrication strategies have been described in detail

Conclusions

We have shown that combinatorial discovery can be used to identify better dual use catalysts, by superimposing activity maps to create a consensus map. Because the combinatorial method provides activity data for the entire composition space, it is more straightforwardly adapted to this task other than systematic optimization methods. While in this example the two figures of merit were catalytic activity for different reactions, in principle the same method could be used to optimize the activity

Acknowledgements

We thank the Army Research Office for support of this work under grant DAAH04-94-G-0055, subcontract SA151-298 from Illinois Institute of Technology. DAD thanks the NSF Solid State Chemistry Program for support of his summer research at Penn State. We thank Prof. Eugene Smotkin for helpful discussions about kinetic isotope effects, and for providing a copy of Ref. [42] prior to publication. We also thank Dr. Renxuan Liu for designing the gas diffusion cell used to test bulk catalysts, and Mr.

References (43)

  • C.L. Hill et al.

    J. Mol. Catal. A

    (1996)
  • E.A. Dawson et al.

    Appl. Catal.

    (1992)
  • J.E. Baker et al.

    Appl. Catal.

    (1989)
  • A.A.G. Lima et al.

    Appl. Catal. A

    (1998)
  • P.A. Lehman et al.

    Int. J. Hydrogen Energy

    (1997)
  • L.A. Torres et al.

    Int. J. Hydrogen Energy

    (1998)
  • O. Ulleberg et al.

    Solar Energy

    (1997)
  • L.L. Swette et al.

    J. Power Sources

    (1988)
  • L.L. Swette et al.

    J. Power Sources

    (1994)
  • G. TamizhMani et al.

    J. Electroanal. Chem.

    (1998)
  • C. Iwakura et al.

    Electrochim. Acta

    (1977)
  • A. Damjanovic et al.

    Electrochim. Acta

    (1966)
  • J.J. Hanak

    J. Mater. Res.

    (1970)
  • B. Jandeleit et al.

    Angew. Chem. Int. Ed. Engl.

    (1999)
  • M.T. Reetz et al.

    Angew. Chem. Int. Ed. Engl.

    (1998)
  • A.C. Cooper et al.

    J. Am. Chem. Soc.

    (1998)
  • C. Hinderling et al.

    Angew. Chem.

    (1999)
  • F.C. Moates et al.

    Ind. Eng. Chem. Res.

    (1996)
  • S.M. Senkan

    Nature

    (1998)
  • S.J. Taylor et al.

    Science

    (1998)
  • P. Cong et al.

    Angew. Chem. Int. Ed. Engl.

    (1999)
  • Cited by (203)

    • Characterization and preparation of Pt-Ru-Ir alloys produced by thermal pyrolysis for hydrogen production

      2022, Chemical Physics Impact
      Citation Excerpt :

      Another suggested strategy is the formation of platinum alloys, allowing maintaining the high catalytic ability of platinum to carry out the HER, but with a smaller amount of platinum. Research by Chen et al. [17] has suggested using alloys with noble metals such as Pt, Ru, Ir, Os, and Rh, to carry out both HER and OER. According to some studies, and based on the combinatorial libraries’ technique, Pt, Ir, and Ru could be used to build electrodes for HER.

    • Influence of the Pt<inf>x</inf>RuIr<inf>y</inf> electrocatalysts composition toward oxygen evolution and reduction reactions for unitized regenerative fuel cell

      2021, International Journal of Hydrogen Energy
      Citation Excerpt :

      Different studies have been focusing on the use of Pt, Ru, Ir metals (or its oxides) and their combinations, Pt–Ru–Ir, Pt–RuO2–IrO2 [13,30], Pt/IrO2 [42], Ir or IrO2 supported by Pt [11,23], and Ir@Pt, i.e., Ir is the core-shell of Pt electrocatalysts [43]. Chen et al. [44] used the combinatorial chemistry with Pt, Ru, Os, Ir, and Rh. He encountered that Pt4·5Ru4·0Ir0.5 is the best combination as a bifunctional electrocatalyst.

    View all citing articles on Scopus
    View full text