Navigation

Recent Publications

More...

Home

Parameterisation of density functional tight binding theory for transport modelling in nanoscale devices: including the repulsive potential

figaro - Posted on 07 April 2016

Project Description: 

The project aims to advance the state of the art in atomic level modelling of electron and phonon transport in solid state nano-devices. The goal is to deliver a novel comprehensive tool for semi-automatic parameterisation of density-functional tight binding theory (DFTB). This is necessary to accurately and efficiently model atomic structure, electronic and dielectric properties, carrier transport and interaction with light in a single, self-consistent formalism, in systems that are inaccessible to ab initio or semi-empirical methods. For example, accurate DFTB parameters are prerequisite for a quantitative enquiry are three new, diverse, and very relevant contemporary problems: 1) impact of dopant-segregation on current-voltage characteristics of ultra-scaled Si nanowire MOSFETs; 2) fundamental processes of carrier separation in novel photovoltaic materials based on metal-halide perovskite; 3) mechanism of photo-catalytic water-splitting at the interface between different phases of crystalline gallium oxide.

Researcher name: 
Stanislav Markov
Researcher position: 
Research Assistant Professor
Researcher email: 
figaro@hku.hk
Research Project Details
Project Duration: 
05/2016 to 09/2017
Project Significance: 
The primary long-term impact is in the delivery of enabling technology for research in quantum transport by means of DFTB. Although DFTB is not an ab initio theory, its relevance is growing as it can intrinsically capture the structural, electronic and dielectric properties of nano-devices whose characteristics are dominated by surfaces, interfaces and defects. Such devices are very challenging for ETB, which is typically parametrised for bulk-like materials. DFT on the other hand is prohibitively expensive for larger structures, and suffers the well-known band-gap deficiency inherent to simpler density functionals. The capabilities of DFTB however pivot on accurate parameterisation. Therefore, developing SKOPT to automate the parameter optimisation opens a cost efficient way (2-3 orders faster), or sometimes the only way, for quantitative investigation of certain problems.