Abstract: Chemically stable SrO 1+ x surface oxide with c(6×2) superstructure is formed, creating space charge layer and surface band bending effects on undoped SrTiO 3 (001), under elevated oxygen pressure and temperature conditions.
Introduction
Strontium titanate ($SrTiO_3$, or STO) is often referred to as the "silicon" of oxide electronics. It provides the essential substrate for groundbreaking phenomena, including two-dimensional electron gases (2DEGs) and high-temperature superconductivity. Traditionally, researchers have relied on Niobium (Nb) doping to make STO conductive enough for standard surface analysis. However, this doping can obscure the material's intrinsic behavior, particularly how the undoped surface responds to the high temperatures and oxygen pressures typical of thin-film growth. This study investigates the "defect chemistry" of undoped (001) STO surfaces to understand how oxygen and strontium vacancies rearrange the electronic structure.
Experimental Methodology
To observe the surface in real-time under growth-like conditions, the researchers utilized Ambient Pressure X-ray Photoelectron Spectroscopy (AP-XPS). Unlike conventional XPS, which requires ultra-high vacuum (UHV), AP-XPS allows for chemical shift measurements in oxygen environments (up to 0.1 mbar) at temperatures reaching 600°C.
The team also employed depth-resolved XPS by varying photon energies to distinguish surface atoms from the bulk, and Low Energy Electron Diffraction (LEED) to visualize surface superstructures. Samples were prepared using an acid-etchant-free method to ensure a pristine $TiO_2$-terminated starting surface.
Key Insights: The $SrO_{1+x}$ Passive Layer
The study revealed that under oxygen pressure at 600°C, strontium atoms migrate from the subsurface to the top, forming a stable $SrO_{1+x}$ surface oxide. This layer acts as a chemical barrier, inhibiting further oxygen diffusion into the bulk.
Intrinsic Band Bending and Space Charge
The formation of this surface oxide layer consumes electrons from the underlying lattice, creating a "space charge layer"—a region depleted of electrons. This results in an upward bending of the electronic bands by approximately 0.54 eV. This finding is significant because it demonstrates that significant band bending occurs intrinsically, without the need for intentional doping.
UHV vs. Oxygen Environments
While both UHV and oxygen-rich environments eventually produce a $c(6 \times 2)$ surface reconstruction, their underlying mechanisms differ. In UHV, the reconstruction is driven by oxygen vacancies ($V_O$), whereas in oxygen environments, it is driven by strontium vacancies ($V_{Sr}$). The UHV-grown surfaces also tend to be less ordered due to a higher concentration of surface defects.
Conclusion and Future Outlook
The identification of a pre-existing space charge layer on undoped STO substrates has profound implications for the growth of complex heterostructures. This intrinsic electronic state likely influences the initial stages of film deposition and may explain variations in interface conductivity, such as the 2DEG, across different growth conditions. Future work will focus on refining atomic-scale models of the $c(6 \times 2)$ reconstruction to better understand the interplay between vacancies and surface stability.
Summary updated