Simulations of Quantum Dot Growth on Semiconductor Surfaces: Morphological Design of Sensor Concepts

摘要

Due to quantum confinement, superlattice detectors may conceivably be tuned to detect electromagnetic signatures difficult to detect with technology available today. The challenge in creating new superlattice detectors is in the growth process where the confluence of chemistry, physics, and mechanics make control of parameters nontrivial. Furthermore, the time scales involved are often prohibitive for deterministic modeling approaches. We have developed and validated a new modeling method for probabilistic modeling of superlattice growth that spans the appropriate time scales relevant for experiments. The predictions of the models have been tested successfully against numerous independent experiments. Through a fundamentally new Green's function formulation for strain interactions among quantum dots, we present a new kinetic Monte Carlo methodology that can successfully predict pattern formation on surfaces while requiring only experimentally-viable parameters as input. The new method also can be used for engineering design of dot correlations in-plane and in-bulk to predict dot alignment in general superlattices.