From Fundamental Physics to Multi-functional Electronic Devices: Santander and Colleagues Publish in Nature
Today's microelectronic components consist of layers of semiconductors on a silicon substrate. In order to sustain the pace of miniaturization and upgrade in the performance of microelectronic devices beyond 2020, alternative technological solutions are being investigated. Researchers are increasingly turning their attention to transition metal oxides, which offer promising physical properties such as superconductivity, magnetoresistance, thermoelectricity, multiferroicity and photocatalytic capacity.
Within this family of materials, strontium titanate (SrTiO3) has been the subject of extensive research. This insulating material becomes a good conductor when it is doped, for example by creating a few oxygen vacancies. The interfaces between SrTiO3 and other oxides (LaTiO3 or LaAlO3) are conductive, even though the two materials are insulators. Moreover, they display properties like superconductivity, magnetoresistance and room-temperature thermoelectricity. The problem, however, is that interfaces between oxides are very difficult to produce.
Now an unexpected discovery has burst through this technological barrier. An international team led by researchers at CNRS and Université Paris-Sud (France) has produced a two-dimensional metallic electron gas (2DEG) on the surface of SrTiO3. This conductive layer, approximately two nanometers thick, was obtained by vacuum-cleaving a piece of strontium titanate, a very simple and economical process. The constituent elements of SrTiO3 are natural resources available in large quantities, and the compound is non-toxic, unlike e.g. the materials most widely used today in thermoelectric applications for microelectronics (bismuth tellurides). In addition, 2DEGs could probably be created on the surface of other transition metal oxides using a similar technique.
The discovery of a conductive layer of this type (not requiring the addition of a layer of another material) is a significant step forward for oxide-based microelectronics. It could make it possible to combine the intrinsic multifunctional properties of transition metal oxides with those of the two-dimensional metal on their surface. Possible developments could include the coupling of a ferroelectric oxide with the electron gas on its surface to produce non-volatile memories, or the inclusion of transparent circuits on the surface of solar cells or touch screens.
The 2DEG on the surface of strontium titanate was discovered and studied in experiments using angle-resolved photoemission spectroscopy (ARPES) at the Synchrotron Radiation Center at the University of Wisconsin, USA, and the SOLEIL synchrotron in France.
The researchers stress that the working philosophy of SRC, allotting a large amount of beamtime (3 weeks/semester) per project during a time lapse of two years, was essential for this discovery. First, it allowed us to focus our attention in an accidental observation: when we detected the first spectra of a metallic state on insulating strontium titanate, it occurred on a sample that was erroneously labelled as another material we intended to study. Second, the large allotted beamtime allowed exploring in detail, over a large number of samples and by varying a large number of parameters, a subject that goes against the commonly accepted "dogmas", namely, that a large-gap insulating material cannot present a metallic surface and cannot be studied by photoemission.
Two-dimensional electron gas with universal subbands at the surface of SrTiO3. A. F. Santander-Syro, O. Copie, T. Kondo et al. Nature 469, 189-193 (January 2011).