Monte Carlo Simulation of Nanostructures: Semiconductor Devices to Ion Channels

This dissertation discusses the transport Monte Carlo model and its application for simulation of semiconductor devices and biological ion channels. A quantum corrected 3-D Monte Carlo simulator has been developed to investigate the various physical phenomena underlying nanoscale semiconductor devices. The results for two-dimensional capacitors obtained from MC simulations have been compared with the results obtained from the Schrodinger-Poisson solver to test the validity of the model. Device simulations of finFET devices show that quantum effects do not have significant effect on device currents. Furthermore, the presence of long fin extension regions can significantly downgrade the performance of finFET devices. The effects of dielectric boundary force, caused by image charges, have also been investigated for nanoscale devices. A new simulator based on the Boltzmann transport Monte Carlo model has been developed to study the ion channel problem. This simulator uses a coarse grained approach to simulate ion channels, thereby allowing for long simulation times necessary for calculating channel currents. Focus grid methodology has been applied for ion channel simulations to further speed up the simulations. The simulator was validated by comparing results of bulk simulations for model electrolyte solutions with those obtained from equilibrium Monte Carlo method. Full-scale channel simulations were performed for the gramicidin and OmpF porin channels. The large size of chloride ions and electrostatic factors were found to be the main reasons for the absolute selectivity shown toward sodium ions by gramicidin channel. The gramicidin channel currents correlated very well with the potential energy profiles for sodium ions. Porin simulations under various bias and bath concentration conditions were performed. We were able to reproduce many of the properties exhibited by OmpF porins. We found that the large negative charge on the porin molecule was mainly responsible for the highly selective nature of porin channels. The channel currents obtained from our simulations matched experimental data for a range of bias conditions and bath concentrations. Use of focus grid scheme resulted in a speedup of 200% and 150% for gramicidin and porin simulations, respectively.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Kathawala, Gulzar Ahmed

Contributors dc:contributor

• Ravaioli, Umberto

• Engineering, Electronics and Electrical

Language dc:language

eng

Identifier

(MiAaPQ)AAI3199042

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/80927