Oxide Electronics beyond Moore
The rapidly approaching “end” of Moore’s law is full of opportunities for the incorporation of new technologies which have been in the background so far. The primary mover behind this paradigm shift has been the march towards 3D circuits where wafers of circuits are bonded with intermediate thru wafer vias to enhance the circuit density without compromising speed. One of the interesting fall offs of this approach is the fact that all the wafers do not have to go through the same device process and further the wafers need no longer all be silicon! This provides a new opportunity for hetero-integration of different material systems and for novel device technologies based on 2D materials and oxides and heterostructures that have manifested new/ novel physical phenomena, of relevance to the information processing/ communication/ storage industry.[more]
While oxides in their bulk form have shown a tremendous spectrum of physical characteristics, oxide heterostructures provide a versatile pathway to combine these various physical phenomena and create novel electronic phases and eventually devices. We are now at the stage where we can produce customized oxide heterostructures with atomic precision. Well-controlled thin films, heterostructures, interfaces and quantum wells based on oxide materials have been fabricated and are being explored for a possible new generation of oxide electronic devices. A highly sophisticated epitaxial deposition technique, namely PLD-MBE is available in our lab, will be a key enabler for the preparation of our atomically precise controlled systems.
In many ionic materials, including the oxides, surfaces created along specific directions become electrically charged. Equivalently, such electronic charging, or ‘polarization’, can also occur at the interface connecting two materials. Theoretically, in certain systems this could lead to the build up of an ever increasing voltage in the materials, a situation known as a ‘polarity catastrophe’. Of course, this cannot occur for energetic reasons in practical systems, and nature deals with this situation by reconstructing the electronic configuration of the interface. This can occur by the shifting of charge across the interface or by structural reconstructions, meaning the displacement of atoms. For the oxide materials especially, a tantalizing consequence of these reconstructions is that it provides a means to create novel electronic phases, stabilized by the interface and which cannot exist in the bulk. Studies have already shown 2-dimensional conducting planes to emerge between otherwise insulating materials. Using the state of the art thin-film deposition techniques, we will study interfaces involving various other interesting systems, including other titanates, as well as cuprates, manganates and vanadates. In the cuprate materials, the multivalency and charge transfer properties of the Cu-O system are at the heart of high–Tc superconductivity. The vanadates are of interest for their wide range of possible valence states and the possibility to also realize p-type doping, and the manganates for their magnetic and Colossal Magneto-Resistance properties.
Quantum Phase Transition
High-T c superconductivity in cuprates results from doping charge carriers into Mott insulators. The magnetic and superconducting properties of Mott insulators doped by holes (p-type carriers) are different from those doped by electrons (n-type carriers), showing asymmetry in the phase diagrams. Moreover, their transport properties depend on the type of carriers, for example, in-plane normal-state resistivity exhibits quadratic temperature (T ) dependence for n-type cuprates while linear T dependence is seen for p-type cuprates with optimal doping. Such comparisons of electron- and holedoping asymmetry (symmetry) in cuprates should help to further our understanding of the cuprate superconductors. The typical electron-hole asymmetry (symmetry) investigation is usually based on cuprates with different crystallographic structures such as NCCO (T-structure) and La2−xSrxCuO4 (T-structure). These materials have different parent Mott insulators and thus exhibit different properties even without doping. Therefore, it is desirable to dope electrons and holes into a Mott insulator without changing the crystallographic structure and address the physics in both sides of such “ambipolar” cuprate.
Doping charge carriers will causes the change of cuprates from antiferromagnetic Mott insulators to high-Tc superconductors, and then normal metal. Among these transitions, superconductor-insulator transition (SIT) in the two-dimensional limit is of particular interest. Continuous changing of carrier density is necessary to understand the nature of such phase transition, and thus, further our understanding of cuprate superconductors. Electric field-effect doping is a potential avenue for this investigation and it is reversible and exempt from chemical disorders, enabling the study of intrinsic properties of cuprates.
One of the great unresolved issues in resistive switching is its driving mechanism. A deep understanding of this mechanism is required in order to optimize Resistive Random Access Memory (ReRAM) device characteristics and develop guidelines for scaling, reliability, and reproducibility. Based on our experiments, we have proposed a new model based on the so-called “quasi conduction band” which can give a more stable characteristic, thereby enabling device scaling with enhanced reproducibility.