Visualizing the MIT in V2O3
Despite 60 years of research, the metal-to-insulator transition in complex oxide materials still holds some mysteries. In a study published in Nature Physics, we show that the electronic transition in V2O3 evolves in real space by forming intertwined `mazes' of insulating and metallic patches.
The first order metal-to-insulator in has been widely studied as a typical example of a phase transition driven by strong correlation effects. In this transition the resistivity changes by several decades turning the material from bad metal at high temperature into a so-called Mott insulator at low temperatures. During a period as visiting scholar in the group of Dimitri Basov at the University of California: San Diego I was involved in the first low temperature measurements made with an upcoming technique in materials science: scattering-type scanning near-field optical spectroscopy. Near field optical spectroscopy makes use of a standard atomic force microscope, coupled to a michelson-morley type interferometer, to probe materials on length scales much smaller than the diffraction limit. In our experiment we used infrared radiation with a wavelength of 11 μm and used the near-field interaction between a metallic AFM needle and a material to probe an area of 25 nm radius. Together with PhD student Alex S. McLeod, we visualize the spatial dependence of the electronic response in thin films of V2O3 as the material is slowly cycled through the transition. Unexpectedly, the electronic response shows stark metallic and insulating patterns. By analyzing these patterns we find that the macroscopic response of the system results from a percolative type of transition: as temperature increases metallic patches start to grow until they form a connected network of metallic stripes (the percolation threshold). At this temperature the resistivity shows the strongest decrease and starts to behave metallic. Finally, we show that the structural transition, which was thought to occur concomitantly, is decoupled from the electronic transition. By careful thermometry between different experiments we show that structural transition temperature is higher than the electronic transition temperature.
Read more on complex oxides.
Switching speed limit in magnetite.
Magnetite is a model system for understanding correlated oxides. Nevertheless, the exact mechanism of the insulator–metal, or Verwey, transition has long remained inaccessible. In our Nature Materials paper published in the summer of 2013 we show that we can switch this material from insulating to metallic on unprecedented short time scales.
The Verwey transition was investigated with pump–probe X-ray diffraction and optical reflectivity techniques. The low-T, insulating phase of magnetite, named after the UvA professor and Philips NatLab director E.J.W. Verwey - is due to charges freezing into a pattern also involving ordering of the Fe3d orbitals. Recent X-ray diffraction work showed showed that the Verwey phase possesses 'trimerons', trapping the otherwise mobile charges at three-site, distorted centres in the crystal. Out pump-probe work shows how trimerons become mobile across the insulator–metal transition. We find this to be a two-step process. After an initial 300 fs destruction of individual trimerons, phase separation occurs on a 1.5 +/- 0.2 ps timescale to yield residual insulating and metallic regions. This work establishes the speed limit for switching in future oxide electronics. The research was conducted by a consortium of groups led by Hermann Dürr's group at SIMES in SLAC. Hermann is adjunct professor attached to the QMat group and Mark got involved in this work while on sabbatical at SIMES in the summer of 2012.
Read more on the Nature website or in the pdf.