A theoretical study of the structure and stability of borohydride on 3d transition metals

Abstract

The adsorption of borohydride on 3d transition metals (Cr, Mn, Fe, Co, Ni and Cu) was studied using first principles calculations within spin-polarized density functional theory. Magnetic effect on the stability of borohydride is noted. Molecular adsorption is favorable on Co, Ni and Cu, which is characterized by the strong s–d_zz hybridization of the adsorbate-substrate states. Dissociated adsorption structure yielding one or two H adatom fragments on the surface is observed for Cr, Mn and Fe.

Introduction

Understanding the structure, energetics, and mechanism of adsorption of borohydride (BH_4ads) (“ads” means adsorbed on the surface) on different metals is an important step in the design and engineering on the atomic scale of surface catalysts for reactions involving borohydride. This is essential, for example, to guide the choice of anode catalyst for Direct Borohydride Fuel Cell (DBFC).

DBFC is an alkaline-based fuel cell which has a potential to generate high power densities competitive to Direct Methanol Fuel Cell (DMFC) for portable power applications [1]. Over an electrocatalyst selective to direct oxidation, each borohydride molecule is capable of producing eight electrons via the suggested over-all cell reaction:

$$B{H}_4^{-}+8O{H}^{-}\to B{O}_2^{-}+6H_{2}O+8e^{-}\text{.}$$

However, the efficiency and power density of DBFCs are limited in part by the lack of an effective anode electrocatalyst [2], [3], [4] and the competing non-selective hydrolysis reaction which leads to undesirable hydrogen gas evolution [5], [6], [7], [8].

Many experimental studies on the electrooxidation of borohydride were carried out since the initial introduction of an aqueous sodium borohydride solution as an anode fuel for alkaline fuel cell in the 1960s [9], [10], [11], [12]. Low coulombic efficiency was reported for Ni, Pd, and Pt [5], [7] due to the production of hydrogen gas. Au [13], [14] and Ag [14] anodes were reported to have high coulombic efficiency but the slow electrode kinetics requires high overpotentials to attain a practical power density. Other electrochemical studies involved pure and alloy catalysts using a variety of experimental techniques [15], [16], [17], [18], [19], [20], [21], [22], [23], [24]. The general picture that emerged from these studies is that the initial adsorption of borohydride anion (BH₄⁻) on catalysts, accompanied by a simultaneous transfer of electron generating BH_yads + (4 − y)H_ads (with y = 1,2,3 for dissociative and y = 4 for molecular adsorption), and followed by electrocatalytic reactions of BH_yads with H₂O, OH_ads, and OH⁻ can explain, in principle, the electrooxidation process [25].

Despite the large number of experimental studies, only a few theoretical studies were made on the adsorption of borohydride on different metals [25], [26], [27], [28], [29]. For the case of Pt(111), Pd(111) and Ir(111) surfaces, we previously found that borohydride dissociates generating BH_ads and 3H_ads fragments [25]. The dissociated geometries differ only in the most stable sites for H_ads (i.e., Pd: fcc hollow site, Ir: top site, Pt: next-neighboring top site). These are the same sites that were proposed as the preferred locations for H adatoms on Pd, Ir and Pt in the absence of borohydride species [30], [31]. Surface diffusion of H_ads and associative desorption yielding H₂ is very likely. This explains the experimental observation of high H₂ evolution on Pt [5], [17]. Thus, a possible way to avoid hydrogen evolution is to promote the molecular adsorption of borohydride on the metal surface. This molecularly adsorbed state was found for Au [25], [27], Os, Ag, Rh, and Ru [25] surfaces (no H₂O_ads coadsorption). Low adsorption energy for the case of Au causes the experimentally reported low surface coverage by adsorbed species. Thus, high overpotentials are required to achieve an appreciable rate of oxidation on Au [14]. A desired anode catalyst for direct oxidation, must therefore promote a strong molecular adsorption of borohydride to produce high surface coverage by reactive species [29] and avoid hydrogen evolution.

In this paper, we study the adsorption of borohydride on 3d transition metals (Cr, Mn, Fe, Co, Ni, and Cu) using first principles calculations within density functional theory (DFT). These metals are usually employed as alloying materials for various catalytic surfaces for fuel cell applications [32], [33], [34], [35], [36], [37], [38], [39], [40], [41]. An understanding on how borohydride interacts with these alloying components can aid in the design of metal alloys or overlayers that are often considered experimentally. We present the adsorption energy of borohydride on these metals and then show the effect of the magnetic properties of the metals on the stability of borohydride on the surface. Then, we describe electronically the possible structures of borohydride on the surface and show that unique structures on 3d transition metals, different from those on noble metals, can be achieved. To the best of our knowledge, the available literature still lacks a thorough fundamental understanding of the interaction of borohydride on the 3d metal surfaces considered in the present work.