Ni diffusion in ceria lattice: A combined experimental and theoretical study

Abstract

Cation diffusion in ceria is of significant importance in many fundamental processes, but the mechanism by which 3d-transition metals migrate in ceria has not yet been elucidated. Here, we select Ni as a representative transition metal and combine experimental and theoretical analyses to determine where Ni is located in ceria and how it migrates. Bulk Ce_0.995Sm_0.005O_1.9975-δ pellets are prepared with a sputtered NiO source on top, and the diffusion coefficient of the Ni impurity is measured at temperatures from 1250 to 1350 °C by analyzing the depth profiles of Ni via time-of-flight secondary ion mass spectrometry. In addition, corresponding activation enthalpy values for Ni diffusion when Ni is present in substitutional and interstitial sites are calculated by density functional theory and compared with the measured values. We reveal that the diffusion of Ni impurity mainly occurs via the interstitial mechanism in ceria and propose an atomic-scale interpretation of Ni migration.

Introduction

Ceria has a fluorite structure with space group $Fm\overline{3}m$ over the entire temperature range, from room temperature to its melting point [1]. In this material, oxygen ions have a more open sublattice than cation ions, readily forming oxygen defects (typically vacancies) [2,3]. Thus, the migration of oxygen ions in ceria is relatively easy, and, therefore, ceria has been intensively studied as an oxygen-ion conducting solid electrolyte [1,[4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]–17]. On the other hand, cation transport in ceria is quite sluggish; it is challenge to reliably analyze and interpret the diffusion of cations, and there have only been a few reports about this phenomenon [15,16]. Moreover, an atomic-scale understanding of cation diffusion in ceria has rarely been achieved, despite the fact that cation migration has a significant impact on many fundamental processes, such as sintering [18], creep [19], and internal friction [20]. In particular, ceria is the most widely used material as an active support for heterogeneous catalysts and as an electrolyte and electrode for ceramic electrochemical devices. In many devices, ceria is in contact with 3d-transition metal components such Ni, Fe, and Co at high temperatures. Therefore, the migration of the transition metal into the ceria is technically crucial, as it can greatly affect the performance and durability of these devices.

To elucidate the cation diffusion mechanism, Beschnitt et al. using both classical and quantum-mechanical simulation techniques, recently conducted a theoretical investigation of the self-diffusion of Ce cations in a ceria super cell [21]. It was concluded that Ce ions must migrate via a vacancy mechanism. Moreover, following that study, the impurity diffusion of Hf and Zr ions into Gd-doped (0.5 at%) ceria was examined by means of time-of-flight secondary ion mass spectrometry (ToF-SIMS). It was found that both Hf and Zr cations diffuse predominantly by the vacancy mechanism, as Hf and Zr are isovalent with Ce and their Shannon radii are comparable with the radius of Ce [15].

On the other hand, thus far there have been only a few reports on the diffusion of 3d-transition metals in ceria [16]. Diffusion of relatively small atoms that normally occupy interstitial sites in solvent crystals generally occurs by the interstitial mechanism. However, although 3d-transition metal ions are smaller in size (0.55–0.77 Å) than the Ce ion (0.97 Å) [22], they are not small enough ideally to fill interstitial sites in a cation-sublattice. Accordingly, transition metals may occupy both substitutional and interstitial sites, and contradictory experimental results on their positions have been reported [[23], [24]–25]. Therefore, it is as yet unclear how 3d-transition metals are transported in ceria. Moreover, the diffusion mechanism in the oxide depends not only on the size of diffusing ion but also on many internal factors, such as the oxidation number and coordination chemistry of the ion and the crystal structure of the host oxide, as well as external factors such as the impurity content, temperature, and gas atmosphere [18].

In this work, therefore, to elucidate the diffusion mechanism of transition metal ions in ceria, we combine diffusion experiments and theoretical calculations. Ni is selected as a representative transition metal impurity because it is widely used with ceria in various high-temperature applications, such as catalytic converters [26], hydrocarbon reformers [27], and solid oxide fuel cells [14]. First, we measure the depth profiles of Ni impurities diffused into bulk ceria pellets by means of ToF-SIMS, which determines the magnitude and the activation enthalpy of the Ni impurity diffusion coefficient at temperatures from 1250 to 1350 °C under air. Then, using density functional theory (DFT) calculations with a 2 × 2 × 2 super cell, the defect formation energy and migration energy values are calculated when Ni is located at substitutional and interstitial sites, respectively. By comparing the activation energy values for diffusion obtained experimentally and theoretically, we are able to state that Ni cations in lightly doped ceria predominantly diffuse via the interstitial mechanism.