Quantum transport and far from equilibrium response in nano-junctions

Aim and Plan
This project is about an entirely new low-voltage approach whose electron trans- port characteristics will be investigated theoretically. The experimental partners and collaborators at IBM are building the first nanoscale realization of the PET which makes the theoretical and the experimental progress somewhat hand in hand. The objective is the study of thin film heterostructure at nanometre scales for switching operations including the many-body effects and evaluation of scalability of such technology to low dimensions.

The ultimate aim is the development of a state-of-the-art theoretical frame- work for modelling non-equilibrium quantum transport in nanojunctions.

Methods and Objectives
This theoretical project includes the development and application of atomic scale electronic structure and quantum transport computational techniques to model the conductance of piezoresistive materials in contact with metal electrodes in a device setup. The used methods are based on the density functional theory (DFT), combined with the non-equilibrium Greens functions (NEGF) technique for quantum transport, as implemented in the Smeagol code.
Although the electronic structure calculations, in general, are well under- stood and controlled especially using the Density Functional Theory(DFT), the proper description of quantum transport requires more advanced mathematical descriptions e.g., non-equilibrium Keyldysh Green's Functions. In the litera- ture usually Non-Equilibrium Green's Function methods are employed in the framework of DFT to describe transport. It predicts correctly the qualitative nature of reduction in bandgap of these materials with pressure. But the DOS, bandstructure and f-electron conductnces do not agree with experiments. This is because, strong electron-electron correlations are not captured correctly for the highly localised Sm f-shells.
Piezoresistive materials such as SmS or SmSe typically include strongly cor- related f-electrons, so that their transport properties cannot be captured by state of the art DFT methods. To be able to describe such systems new algo- rithms are therefore required, which will be based on the dynamical mean field theory (DMFT) embedded in the NEGF framework. DMFT is supposed to re- duce the over-delocalistion of f-electrons and may thus, reduce the conductance overestimation.
To date such methods are only in their infancy, and an important aspect of the project will be the development of many-body DMFT+NEGF compu- tational algorithms into a mature software, which will also be distributed for use to the community. Part of the calculations will also be performed with the CASTEP software, and a project to implement DMFT in CASTEP is under way in collaboration with King's College London. The atomic scale con- ductive properties of the piezoresist/metal nanodevices will then be passed to our collaborators and embedded in a multiscale approach encompassing finite elements simulations and analytical models, which will allow to develop new device designs for the experimental partners at IBM.
This project is a collaboration between the National Physical Laboratory and Royal Holloway, University of London, as centres for the development of electronic structure methods, where Ivan Rungger and Keith Refson are based
as core developers of the Smeagol and CASTEP software packages, respectively. Collaborators in this project include: IBM TJ Watson (USA), IBM Ruesch- likon (Switzerland), CASTEP developers group (UK), Nano-bio spectroscopy group in the University of the Basque Country (Spain), University of Augsburg (Germany), Kings College London (UK).