Competition between clustering and dispersion of cobalt atoms on perovskite surfaces: SrTiO3(001) and KTaO3(001)

Problem Statement
Perovskite oxides (ABO₃) are promising materials for catalysis, photocatalysis, and electrocatalysis due to their tunable chemical properties and unique physical characteristics. However, their practical application is hindered by inherent surface instability, which often leads to the deactivation of catalytic properties and material degradation. While extrinsic doping is a common strategy to counteract this, an atomic-scale understanding of how dopants interact with, incorporate into, or cluster on perovskite surfaces is currently missing. Specifically, there is a need to understand how uncompensated surface polarity can be managed by extrinsic doping with ions of suitable valency to stabilize the surface and disperse co-catalysts at the atomic scale.

Methodology
The study investigates the interaction of cobalt (Co) atoms with two prototypical perovskite surfaces: the weakly polar SrTiO₃(001) and the polar KTaO₃(001). The experimental approach combines microscopic and spectroscopic techniques:

• Sample Preparation: SrTiO₃(001) was prepared via wet chemical etching followed by UHV annealing to create a mixture of surface reconstructions. KTaO₃(001) was prepared by in-situ cleaving and subsequent hydroxylation to form a (2×1) periodic termination.
• Deposition and Annealing: Cobalt was deposited via physical vapor deposition (PVD) at sub-monolayer coverages (0.05–0.43 ML) at room temperature. Samples were subjected to stepwise annealing cycles up to 700°C.
• Characterization:
  • Microscopy: Non-contact atomic force microscopy (ncAFM) and scanning tunneling microscopy (STM) were used to resolve atomic-scale adsorption, dispersion, clustering, and surface reconstruction. Kelvin Probe Force Microscopy (KPFM) measured local work function variations. Low-energy electron diffraction (LEED) monitored long-range periodicity.
  • Spectroscopy: X-ray photoelectron spectroscopy (XPS) and synchrotron radiation-excited photoelectron spectroscopy (SRPES) were employed to determine the chemical state (oxidation states) and depth distribution of cobalt species.

Key Results

1. Chemical State: Upon deposition at room temperature, cobalt exists predominantly in ionic states (+2 and +3), with a minor metallic component. Annealing induces a competition between the formation of metallic clusters and the retention of ionic species.
2. SrTiO₃(001) Behavior:
   • Clustering vs. Incorporation: Annealing promotes the coalescence of cobalt into clusters. However, unlike simple clustering, a significant portion of cobalt incorporates into the surface and subsurface regions.
   • New Reconstruction: At high temperatures (700°C), a new cobalt-stabilized surface reconstruction forms, covering approximately 20% of the surface. This phase exhibits a lower work function (+0.23 V shift relative to the clean surface) and a quasi-periodic structure distinct from the original surface phases.
   • Work Function: KPFM reveals two distinct surface phases: the original reconstruction and the new cobalt-stabilized phase, both showing higher local potentials than the clean surface.
3. KTaO₃(001) Behavior:
   • Surface Retention: The polar KTaO₃(001) surface retains its bulk-terminated (1×1) periodicity after annealing. Atomic-resolution microscopy does not show direct evidence of dispersed cobalt atoms on the topmost layer at high temperatures.
   • Subsurface Incorporation: Despite the lack of surface features, statistical analysis of cluster sizes and depth-sensitive XPS data indicate that cobalt incorporates into the near-surface region (several layers deep). This suggests cobalt participates in the polarity compensation mechanism, likely via cation exchange or accommodation in subsurface layers, similar to clean annealed KTaO₃.
4. Mechanistic Differences: The divergent behaviors are attributed to the reducibility of the B-site cation. Ti⁴⁺ in SrTiO₃ can reduce to Ti³⁺ and change coordination (octahedral to tetrahedral), facilitating surface reconstruction and cobalt incorporation. In contrast, Ta⁵⁺ in KTaO₃ strongly prefers the octahedral configuration and is less reducible, favoring polarity compensation through subsurface rearrangement rather than surface reconstruction.

Significance and Claims
The paper claims to provide a missing atomic-scale view of how extrinsic dopants impact perovskite surface structure and stability. The key contributions are:

• Identification of Competing Mechanisms: The study demonstrates that upon annealing, cobalt on perovskites undergoes a competition between coalescing into metallic/ionic clusters and incorporating into the lattice.
• Surface Stabilization: It identifies a specific cobalt-stabilized surface reconstruction on SrTiO₃, linking the presence of cobalt to a new, stable surface phase with distinct electronic properties (work function).
• Polarity Compensation: The work suggests that for highly polar surfaces like KTaO₃, extrinsic doping may stabilize the surface not by forming surface clusters, but by incorporating into the near-surface region to compensate polarity, a mechanism consistent with findings on clean KTaO₃.
• Catalytic Relevance: The findings shed light on atomic-scale processes relevant to oxidative coupling and photocatalysis. The ability to disperse co-catalysts (like cobalt) and stabilize specific surface phases is presented as a strategy to reduce dependence on scarce, high-cost materials in single-atom catalytic systems. The authors note that record photocatalytic efficiencies in doped perovskites are often achieved after high-temperature annealing, a condition under which these specific surface phenomena were observed.