Optimal Dynamic Discrimination of Similar Molecules through Quantum Learning Control

Abstract

A paradigm for discriminating similar quantum systems in the laboratory is presented based on optimal control principles with the aid of closed loop learning algorithms. The optimal dynamic discrimination (ODD) process is simulated for a noninteracting mixture of up to three similar finite-dimensional quantum systems. The optimal control field giving rise to species discrimination, that considers the presence of field and observation noise, is deduced with a genetic algorithm (GA). The similar quantum systems yield distinct dynamics and detection signals, although influenced by the same control laser pulse. The ODD process is shown to operate by drawing on constructive and destructive interference effects to simultaneously maximize or minimize the signals from each of the species in the mixture. The ODD technique may have applications to the analysis and separation of possibly even complex chemical species.