Skip to Main Content U.S. Department of Energy
Physical Sciences

Physics and Chemistry of Epitaxial Oxides

Researchers studying the physics and chemistry of epitaxial oxides
Researchers studying the physics and chemistry of epitaxial oxides.

The Physical Sciences Division has signature capabilities in epitaxial film growth and properties of oxides. Of particular note is the ability to relate electronic, magnetic, optical, and chemical properties to film structure and composition. Our interests focus most generally on properties modification by selective doping and interface formation. Our specific interests include magnetic-cation-doped transition-metal oxides for semiconductor spintronics, electronic and magnetic properties of oxide interfaces, and anion doped oxides for fundamental studies of visible light photocatalysis.

We use oxygen plasma-assisted molecular beam epitaxy and pulsed laser deposition to grow a wide variety of crystalline oxides. Our use of these growth methods results in crystalline oxides of the highest structural quality. In addition to in situ reflection high-energy electron diffraction and high-energy-resolution x-ray photoelectron spectroscopy, we use ex situ lab-based methods such as high-resolution x-ray diffraction and transmission electron microscopy, as well as synchrotron-based techniques such as x-ray absorption near-edge spectroscopy, extended x-ray absorption fine structure, and x-ray linear and magnetic circular dichroism to fully characterize these films. Additionally, we work closely with solid-state theorists at Pacific Northwest National Laboratory to explore these materials and their properties using first-principles electronic structure calculations.

Using these capabilities, we have explored structure-property relationships for materials such as Co- and Cr-doped TiO2 anatase, Co- and Mn-doped ZnO, Ti-doped Fe2O3, H-doped ZnO, and N-doped TiO2 anatase and rutile. From these studies, we have learned

Electron Energy
Left—trimethyl acetate bound to the surface of N-doped TiO2 anatase (001). Right—visible light of energy less than the bandgap of bulk anatase is absorbed at a N site, resulting in electron-hole pair creation. Contrary to expectation, the holes are not trapped at isolated N sites, but can diffuse to the surface via hopping conductivity and photo-decompose the trimethyl acetate.
  • Room-temperature ferromagnetism in Co- and Cr-doped TiO2 anatase is mediated either by defects and/or spinoidal decomposition
  • Room-temperature ferromagnetism in Co-doped ZnO is caused by defects and/or the reduction of substitutional Co(II) to make either Co metal nanoparticles or CoZn intermetallic
  • Dopant substitution in Mn-doped ZnO is not random, but rather is partially correlated, resulting in locally higher dopant concentrations than expected based on the incident atom fluxes
  • Dopant distributions in nanoscale materials deviate from the expectations of bulk statistical predictions, and are well described by Monte Carlo methods we have developed
  • Visible-light photochemistry occurs in N-doped TiO2 anatase (001), but not N-doped rutile (110).

The last point is illustrated in the drawing. Here we show in schematic fashion what happens when electron-hole pairs are generated using a photon energy that is less than the bandgap of pure anatase, but more than the optical absorption threshold at a N site in N-doped anatase. N substitution for O in the lattice at the ~1 at. % level results in localized N 2p states at the top of the valence band. Holes associated with electron-hole pair creation at a N site are not expected to be mobile because the localized N 2p states should act as hole traps. And, this is precisely what happens in N-doped rutile (110); the hole-driven decomposition of small carboxylic acids does not occur because the holes cannot get to the rutile surface. In contrast, visible-light generated holes in N-doped anatase (001) are quite effective at photo-decomposing carboxylic acids. The difference may be due to different extents of N - N overlap in the two polymorphs; density functional calculations suggest that hole hopping to the anatase (001) surface is more facile than that to the rutile (110) surface due to differences in crystal symmetry and surface orientation and their effect of N - N spin density overlap. Thus, there may be orientations of rutile that show higher hole mobility than certain orientations of anatase. In any event, this work shows for the first time that visible light photocatalytic activity in N-TiO2 can be traced to specific orientations of specific polymorphs, and results in a plausible mechanism for the process. This work is of interest for harvesting visible light to carry out organic waste destruction and photochemically splitting water to generate H2 as an energy source.

Contact: Scott Chambers

Physical Sciences

Seminar Series

Science at PNNL

PSD Research

Research Highlights

View All PSD Highlights