Electrochemical Performance for Ordered Intermetallics

                
                        Figure 1: Schematic of direct formic acid PEMFCs                                    Figure 2: Crystal structures         

Polymer electrolyte membrane fuel cells (PEMFCs) that can run directly on liquid fuels of small organic molecules such as methanol, ethanol, formic acid (FA) and others, featuring higher volametric and gravimetric energy densities and better energy efficiency have recently attracted considerable interest as wide variety of power applications because of the simple handling of liquid fuels in storage and transfer (Fig. 1). In order to develop commercially available PEMFCs using liquid fuels which satisfy demands from both economical and environmental points of view, we must overcome difficulties in the development of electrocatalysts in the fuel cells. Serious losses of energy due to the slow kinetics in the oxidation of SOMs and gradual drops of catalytic activity by carbon monoxide (CO) poisoning occur in the anode process. The improvement of electrocatalytic activity and tolerance to CO in the electro-oxidation of SOMs with Pt-based bimetallic and trimetallic electrodes, and Pt-group metals has been attempted. For example, the Pt-Ru alloys are among some of the most efficient catalysts for oxidation of methanol and have shown relative tolerance to CO poisoning, increased current densities, and a decreased overpotential for fuel oxidation when compared to pure Pt. However, Pt-Ru alloy has inherently poorly defined surface structures, for example, with its surface sites occupied by Pt or Ru in a random fashion (Fig. 2-(a) and (c)).  In addition, the use of Ru-based alloys in place of pure Pt introduces a stability problem: during extended periods of operation, particularly under non-optimal usage (especially at high temperatures and current densities), the alloy surface becomes depleted of Ru. We have considered that these properties of ordered intermetallic compounds demand necessities for improving the performance of anode materials with bimetallic alloys. In the surface of PtPb and PtBi ordered intermetallic phases whose employments have been proposed by us for first time as anode electrocatalyst materials, atomically ordering of two element atoms (Fig. 2-(b) and (d)) for showing higher activity are meted.

                 

      Figure 3: Cyclic Voltammograms of Pt, PtBi, PtPb                          Figure 4: Cyclic Voltammograms for CO Oxidation Studied

PtBi ordered intermetallic bulk phases have been shown to exhibit remarkable electrocatalytic activity for the oxidation of FA in terms of onset potential and current density (Fig. 3 (b)).  The PtBi phases showed an onset of oxidation potential (-0.12 V) of approximately 300 mV less positive than pure Pt (Fig. 3 (a)).  The steady-state current density of the PtBi phases is 40 times larger than that of pure Pt in the FA oxidation.  Moreover, oxidation curve observed at PtPb ordered intermetallic phases (Fig. 3 (c)) also showed negative potential shift and increased steady-state current density, compared to PtBi ordered intermetallic phases.  The onset potential and the oxidation current of the PtPb phases are -0.2 V and 3.6 mAcm^-2, respectively. CO is known to be among most problematic poison in PEM direct fuel cells with SOMs.  The CO intermediate species strongly adsorb onto the Pt atoms and seriously degrade the oxidation of FA and methanol.  Eliminating the CO intermediate and inhibiting the adsorption of the CO intermediate are the major factors in high anode performance.  As shown in Fig. 4 (B), the Pt surface show no current up to about +0.65 V and a sharp peak about +0.65 V ascribed to the oxidation of adsorbed CO to CO₂ after exposing CO to the surfaces, compared to (A) before adsorption of CO on Pt. On the other hand, the effect of CO exposure to PtBi ordered intermetallic phases is different to compared pure Pt surfaces (C).  The voltammograms remained essentially unchanged between before and after CO exposure, indicating that CO molecules do not irreversibly adsorb on the surface of PtBi ordered intermetallic phases.  The dramatic drop in the affinity of CO for PtBi ordered imtermetallic surface is a direct consequence of its structure, specially, the difference in the Pt-Pt distance between Pt and PtBi. The expansion of the Pt-Pt distance in PtBi might make it very difficult for CO to bind in bridge site or three-fold hollow site configurations.

Ordered intermetallic phases need to be prepared in nanoparticle form, so as to have increased surface area, which, in turn, should lead to enhanced electrochemical activity per platinum atom for incorporating ordered intermetallic phases into actual PEMFCs. We have prepared nanoparticles of ordered intermetallic phases by reducing metal precursors salts with reducing agents at room-temperature. For example, we prepared PtPb ordered intermetallic nanoparticles with H₂PtCl₆∙6H₂O, Pb(C₂H₃O₂)₂∙3H₂O, and sodium borohydride (NaBH₄) whose BET surface areas were 15 m²g^-1. In addition, the pXRD pattern (Fig. 5 (a)) of the PtPb nanoparticles showed the formation of single phase of ordered intermetallic PtPb. STEM image of synthesized PtPb nanoparticles (Fig. 5 (b)) showed the individual particles about 15-30 nm in size, and these aggregate to form networks.  Lattice fringes are clearly evident (Fig. 5(c)). When the activity of the PtPb nanoparticles was compared to that of commercially available Pt and Pt-Ru alloy nanoparticles for the oxidation of FA, the PtPb nanoparticles showed remarkable superiority of electrochemical catalytic activity in terms of oxidation onset potential (-0.18 V vs. Ag/AgCl) and current density (3 mAcm^-2).