Experimental adaptive optimization with genetic algorithms: Solving problems in mass spectrometry, optics, photoelectron spectroscopy, and quantum control

Experimental searches for global optima in multidimensional spaces are common in many calls of science. In our experiments we use a genetic algorithm for the optimization of laser pulses for coherent control, ion optic voltages for mass spectrometry, and a phase modulator for focusing laser light. The thesis also includes an example of a system that reacts to shaped laser pulses and a new design for a reflectron electron gun for femtosecond pulses.; Laser pulse shaping combined with a global optimization using experimental feedback has been proposed recently as a solution to molecular quantum control problems. Theory has shown that it is possible to control the outcome of chemical reactions by shaping the amplitude and phase of the spectral components of laser pulses. We show molecular control using shaped laser pulses found with experimental feedback and a genetic algorithm. The algorithm optimizes the fluorescence efficiency and effectiveness of a large laser dye molecule by finding the optimal laser pulse chirp, bandwidth and wavelength.; Photoelectron spectroscopy with pulse duration and bandwidth modulation is used to study the ultrashort lived S₂ state of phenol. We find that the S₂ state relaxes on a femtosecond time scale to the S₁ state by internal conversion. Molecular orbital calculations explain the predominance of the X state over the A state in the spectra. Spectra taken at several wavelengths show changes in the vibrational overlap between the intermediate and excited states. Because of the pulse dependent properties of the spectra this system presents itself as a possible system for laser pulse shaping and control based on feedback similar to the mass spectra.; We show that the genetic search technique is useful in other areas, including adaptive optics and mass spectrometry. In the adaptive optics experiment we use the genetic algorithm to find the optimal settings of a phase modulator to focus a laser pulse on a moving target. The motion of the target simulates experimental noise. For the optimization of the mass spectrometer the genetic algorithm uses measured mass spectra to find up to 8 ion optic voltages, optimizing the mass resolution and ion count rate.; These examples demonstrate the general applicability of the technique of experimental adaptive optimization to many laboratory experiments in several different areas of interest.