Quantum control and detection: Theory to reality, gas phase to condensed phase.

摘要

Quantum control of molecular dynamics, defined as the use of tailored laser pulses to optimally drive a quantum system to a desired final outcome, has now been realized both theoretically and experimentally for several types of chemically interesting systems. Our goal is to find the light field which optimally drives a sample to a designated target. We formulate our control theory in terms of Liouville space density matrix language. The density matrix formalism has great advantages for dealing with mixed state systems, and allows a smooth transition from exact quantum to semiclassical and classical mechanics. The latter aspect is extremely important when we go beyond simple molecular systems. In the weak response regime, we study one-photon optimal pump and two-photon optimal pump-dump control. In both cases, the optimal control can be cast as an eigenvalue problem. The theory shows that for a simple system the dynamics can be efficiently controlled by optimal fields at low temperature. Since most of chemistry occurs in condensed phases, we also explore the controllability of these larger systems. Because the large dimensionality precludes exact quantum calculations, we employ nearly classical and semiclassical Gaussian wave packet dynamics. In condensed phase systems, many phenomena such as caging, solvent induced dephasing, energy relaxation and nonadiabatic curve crossing play important roles in chemical reactions. To better understand the system dynamics in the condensed phase, in particular the vibrational dynamics, we develop a phase space Redfield theory to study the relaxation and dephasing processes and compare it with the classical generalized Langevin equation (GLE) and the quantum Master equation. To prove that the target state has been realized, we employ pump-probe spectroscopy to monitor the effect of the optimal pump pulse on the subsequent molecular dynamics. For a thermally populated experimental sample, the final signal has to be properly averaged according to its thermal distribution. We develop an efficient way of calculating the total laser induced fluorescence (LIF) signal of a thermal sample quantum mechanically, and also implement it with classical molecular dynamics. This allows us to simulate the experimental observable under realistic conditions.