First-Principles Assessment of the Reactions of Boric Acid on NiO(001) and ZrO2(1̅11) Surfaces

The present study investigates the adsorption and dissociation reaction pathways of boric acid, B(OH)3, and the reaction kinetic descriptors on NiO(001) and ZrO2(1 ̅11) surfaces. Density functional theory is employed for ground-state calculations, while the nudged elastic band method is used for obtaining reaction barriers. Strong electron correlations in the case of NiO are included using the DFT + U approach. Adsorption of boric acid on clean ZrO2(1 ̅11) is found to be more favorable compared with that on NiO(001), in agreement with prior experiments. Dissociative adsorption is observed to dominate over molecular adsorption in the case of ZrO2(1̅11), whereas NiO(001) favors molecular adsorption. The most stable configuration for B(OH)3 on NiO(001) is a hydrogen-bonded molecular structure, Nis-(OH)B(OH)(OH)···Os (s = surface atom), with an adsorption energy of −0.74 eV. On ZrO2(1 ̅11), a single O−H dissociated structure, Zrs-(O)B(OH)(HO)-Zrs + Os-H, with an adsorption energy of −1.61 eV, is the most stable. Our results reveal lower activation barriers for B(OH)3 dissociation on NiO(001) than on ZrO2(1̅11). We demonstrate the importance of both the surface transition-metal atom and oxygen states and discuss bonding mechanisms leading to different adsorption configurations on such metal oxides. The analysis of surface reactivity presented here is useful in designing metal oxides for catalytic applications and is of significant importance in fuel materials durability in nuclear energy systems. ■ INTRODUCTION Transition-metal oxide (TMO) surfaces have been of significant importance for a wide range of applications, including catalysis, thin-film coatings, fuel cells, and gas sensors. In such applications, a critical understanding of the surface reactivity and adsorption/dissociation reactions in various environments becomes necessary. In addition, understanding and control of surface reaction kinetics is of significant relevance in the field of corrosion and chemical sensors. Adsorption of boric acid on TMOs has been of interest for a broad range of reasons in the past. For example, boric acid adsorption on TiO2 has been studied in the context of dyesensitized solar cells, to convert sunlight into electricity. Boric acid incorporation in soils has been investigated as boron is an important micronutrient for plants. In the present work, our focus is on understanding the process of boron poisoning in nuclear reactors that leads to the safety issue of unequal axial power distribution along the fuel rods. The accumulation and incorporation of boron inside the corrosion deposits (namely, the CRUD) on the nuclear fuel cladding is believed to play a major role in neutron flux depression. The source of boron is the boric acid that is added to control the neutron activity in nuclear reactors. Previous work reveals that boron incorporation and deposition into the corrosion deposits could occur by adsorption and solid-state reactions or precipitation from water. In particular, experiments on CRUD oxides report the following trend for the adsorption strength of boron among the substrates: Fe3O4 > NiFe2O4 > ZrO2 > NiO. It is suggested that the adsorbing boron species is likely to be the neutral boric acid molecule and that the formation of surface complexes on CRUD oxides could lead to precipitation of boron containing compounds, such as the bonaccordite (Ni2FeBO5). Although these experiments report the collective adsorption behavior on CRUD oxides, an atomic scale mechanistic description of boric acid adsorption and dissociation is still lacking. Rate-theory modeling at the continuum level has helped to explain certain aspects of heat, momentum, and mass transfer with respect to CRUD deposition in general. However, in order to predict the kinetics of such mechanisms, the present study aims to describe the surface reaction mechanisms of boric acid on relevant oxides at the atomic level using first-principles quantum mechanical calculations. We choose NiO and ZrO2 substrates as the model systems in this work. Bulk ZrO2 exhibits several polymorphs for different ranges of temperature and pressure. However, studies have shown that ZrO2 deposits found in CRUD have a monoclinic structure. 18 Hence, we study adsorption characteristics on the monoclinic ZrO2 phase. On the other hand, we choose to study NiO in its usual rocksalt structure and antiferromagnetic state. Reactor environments are complex in terms of the chemistry of solid materials as well as of water. The oxide surfaces are Received: February 17, 2012 Revised: April 16, 2012 Published: May 1, 2012 Article