My Ph.D. was based on the use of molecular dynamics techniques to study the atomic-level processes that occur at interfaces between silicon-based devices and their external environment. The adhesive properties of such devices are determined by the ultrathin native oxide layer that spontaneously forms on the silicon surface under normal conditions and so precise characterisation of its structure is important for a range of technological applications. I have used Density Functional Theory to monitor the structure, charge distribution and stress development during oxidation of the silicon surface and investigated the behaviour of phosphorous and boron dopants in the oxide layer through quantum mechanical static and dynamic approaches. These techniques are limited by the small system size and the short simulation time addressable, but combined they provide us with detailed information on the thermodynamically stable dopant positions and reveal reaction mechanisms that would be difficult to access experimentally.
The industrial preparation of silicon-on-insulator devices takes advantage of the strong adhesion between hydrophilic silicon surfaces to bond together crystalline wafers at room temperature. I have used information from first principles simulations of the natively oxidised silicon surface to develop a new classical potential for oxidised silicon systems. By parameterising the interactions of the surface with individual water molecules, I am able to study the atomic-level processes that determine the strength of adhesion between silicon wafers in a wet environment.
Meanwhile, the implantation of silicon-based medical devices into tissues or into the bloodstream results in immediate protein and cell adsorption onto the material surface. The key to successful device implantation is the ability of the surface to control protein adsorption and, hence, guide cell assembly and promote compatibility with the surrounding tissue. By further extending the classical potential to include interactions between biomolecules and the natively oxidised silicon surface, I have performed large-scale computer simulations of the adsorption of proteins, such as collagen (left) and human serum albumin (right), onto realistic models of device surfaces in the presence of water.
The high carrier mobilities and 2D nature of graphene make it a promising candidate for applications in carbon-based nanoelectronics. In the solution-gate field effect transistor, modulation of the channel conductance is achieved by applying a gate potential from a reference electrode across an electrolyte, which acts as the dielectric. We have used classical molecular dynamics simulations, employing polarisable force fields, to rationalise the response of such devices to changes in solution pH at varying gate voltages.