Prof David McKenzie - University of Sydney
Participating Sydney University Node Researchers:
Dr Nigel Marks
Mr Hugh Wilson (PhD student)
A/Prof Michelle Simmons - University of New South Wales
Dr Neil Curson - University of New South Wales
Prof Steven Prawer - University of Melbourne
Prof Phil Smith - University of Newcastle
Dr Marian Radny - University of Newcastle
The molecular modelling program, at the Nanoscale Simulation Laboratory at the University of Sydney, provides theoretical support to the experimental programs of the Centre, by performing atomistic simulations of fabrication processes. The aims of these simulations are twofold: to provide a sound theoretical basis for predictions concerning new or existing fabrication processes, and to aid in the interpretation of experimental data, where the underlying mechanisms giving rise to the observed results are unclear.
The Molecular Modelling program has ongoing collaborations with the Atomic Scale Fabrication and the Single Ion Implantation programs. Current projects include:
Determination of Si Surface Structures:
Using state-of-the-art ab initio density functional theory codes we determine the stable structures and likely dynamic paths for defects on the silicon surface. We focus on structures likely to be produced during the bottom-up fabrication process, in particular those involved in the adsorption of phosphine. The simulations give a realistic idea of how the fabrication process can be expected to proceed on the atomic level.
STM Image Simulation:
Scanning tunnelling microscope (STM) images of atomic-scale surface features combine both geometric and electronic features of the surface into a single quantity and so interpretation is problematic. For this reason, we have been working on determining predicted STM images of the defects we expect to see on the surface, in order to aid in their identification in experiment. By combining our electronic structure calculations from Density Functional Theory with the Tersoff-Hamann approximation, we can construct simulated images of the surface. Some discrepancies still remain between the predicted and observed images of certain structures, and we are working to resolve these by developing methods more sophisticated than Tersoff-Hamann.
Surface Defect Dynamics:
In addition to our static calculations of defect energetics, we are also interested in dynamical behaviour at the surface. Using explicit molecular dynamics methods we can explore picosecond-scale phenomena such as adsorption events. For longer timescale phenomena (nanosecond and above) we use nudged-elastic-band-type methods to determine activation barriers and transition rates. We have also begun work on the hyperdynamics method, one of a class of 'rare-event algorithms', which can speed up the simulation of dynamics by orders of magnitude by reducing the amount of time spent simulating uninteresting thermal vibrations.
Phosphorus Implantation Simulation:
In our collaboration with the Single Ion Implantation group, we predict the trajectories of single ions implanted at low (few keV) energies. Using the state of the art empirical potentials, we can simulate implantations into lattices containing tens of thousands of atoms. Recent work has focussed on dual-ion implantation in which two phosphorus atoms are implanted together as a P-P dimer.