Andrei C. Florean

Control of retinal isomerization in bacteriorhodopsin in the high-intensity regime

A learning algorithm was used to manipulate optical pulse shapes and optimize retinal isomerization in bacteriorhodopsin, for excitation levels up to 1.8 × 1016 photons per square centimeter. Below 1/3 the maximum excitation level, the yield was not sensitive to pulse shape. Above this level the learning algorithm found that a Fourier-transform-limited (TL) pulse maximized the 13-cis population. For this optimal pulse the yield increases linearly with intensity well beyond the saturation of the first excited state. To understand these results we performed systematic searches varying the chirp and energy of the pump pulses while monitoring the isomerization yield. The results are interpreted including the influence of 1-photon and multiphoton transitions. The population dynamics in each intermediate conformation and the final branching ratio between the all-trans and 13-cis isomers are modified by changes in the pulse energy and duration.

Multiphoton control of the 1,3-cyclohexadiene ring-opening reaction in the presence of competing solvent reactions.

Although physical chemistry has often concentrated on the observation and understanding of chemical systems, the defining characteristic of chemistry remains the direction and control of chemical reactivity. Optical control of molecular dynamics, and thus of chemical reactivity provides a path to use photon energy as a smart reagent in a chemical system. In this paper, we discuss recent research in this field in the context of our studies of the multiphoton optical control of the photo-initiated ring-opening reaction of 1,3-cyclohexadiene (CHD) to form 1,3,5- cis-hexatriene (Z-HT). Closed-loop feedback and learning algorithms are able to identify pulses that increase the desired target state by as much as a factor of two. Mechanisms for control are discussed through the influence of the intensity dependence, the nonlinear power spectrum, and the projection of the pulses onto low orders of polynomial phase. Control measurements in neat solvents demonstrate that competing solvent fragmentation reactions must also be considered. In particular, multiphoton excitation of cyclohexane alone is capable of producing hexatriene. Statistical analyses of data sets obtained in learning algorithm searches in neat cyclohexane and for CHD in hexane and cyclohexane highlight the importance of linear and quadratic chirp, while demonstrating that the control features are not so easily defined. Higher order phase components are also important. On the basis of these results the involvement of low-frequency ground-state vibrational modes is proposed. When the population is transferred to the excited state, momentum along the torsional coordinate may keep the wave packet localized as it moves toward the conical intersections controlling the yield of Z-HT.

Spectral phase effects on nonlinear resonant photochemistry of 1,3-cyclohexadiene in solution

We have investigated the ring opening of 1,3-cyclohexadiene to form 1,3,5-cis-hexatriene (Z-HT) using optical pulse shaping to enhance multiphoton excitation. A closed-loop learning algorithm was used to search for an optimal spectral phase function, with the effectiveness or fitness of each optical pulse assessed using the UV absorption spectrum. The learning algorithm was able to identify pulses that increased the formation of Z-HT by as much as a factor of 2 and to identify pulse shapes that decreased solvent fragmentation while leaving the formation of Z-HT essentially unaffected. The highest yields of Z-HT did not occur for the highest peak intensity laser pulses. Rather, negative quadratic phase was identified as an important control parameter in the formation of Z-HT.

Control of 1,3-Cyclohexadiene Ring-Opening

Control over the cyclohexadiene ring-opening is achieved via multiphoton excitation. A learning algorithm searched for excitation pulses while pulse effectiveness was evaluated using transient absorption. Control parameters are identified through analysis of the search set.

Probing excited state dynamics of retinal isomerization in bacteriorhodopsin

We study the effects of excitation pulse intensity, chirp and bandwidth on the ultrafast dynamics of retinal isomerization in bacteriorhodopsin. Variation of these parameters alters the decay times of the I and J conformations.

Coherent control of excited state vibrational coherences in the dye LD690 in solution

We demonstrate control of excited state wave packets in LD690 by spectral and phase shaping of the pump pulses. Blue detuned, positively chirped pulses maximize the excited state coherences and slow down the vibrational dephasing.

On the Search for Quantum Control of Electronic Spin by Shaped Ultrafast Optical Pulses

We report on the search for shaped ultrafast pulses to enhance the rate of inter‐system crossing in metalloporphyrins for NMR polarization enhancement. We found that chirp is the dominant feature in pulses that maximize triplet state population and conclude that the pulses enhanced singlet absorption rather than inter‐system crossing.

Learning Feedback Control Analysis

This chapter reviews learning algorithms and discusses principal control analysis (PCA) and most correlated feature analysis (MCFA) in the context of two control experiments: stimulated Raman scattering in liquid methanol and the multiphoton photoisomerization of 1,3-cyclohexadiene in solution. Learning algorithms can be used to find optimal control fields and to investigate the dynamics of physical systems. Although the algorithms have proven effective at controlling a variety of experimental systems, the mechanism of control is often unclear. The optimal pulse shapes discovered by such learning algorithms are usually too complicated to be easily interpreted and often contain unnecessary features that further obscure their interpretation. An increasing emphasis on understanding the control mechanisms has led to recent improvements to the algorithms and how they are used. PCA and MCFA are statistical techniques that use information collected during a learning search to determine important features of the control pulses and search space. The techniques are general and should be applicable to any fitness-directed search.

Spectral phase and detuning effects in high-power chirped pulse excitation of a dye solution

In a generalized picture of high-power chirped-pulse interactions in solutions, excited-state population depends on detuning from the vertical transition of the absorber. The power spectrum of the ultrafast pulse determines which vibronic states are accessed on the ground and excited-state manifold, as well as the relative intensity of transitions across the spectrum of the dye. The optical response of the laser dye LD690 clearly depends on both the power spectrum and spectral phase of a high-power ultrafast pulse. In a vertical transition, chirp can control excited-state population and enhance vibrational coherence in at least the ground state. Chirped pulses detuned to the blue edge of the dyes spectrum are less effective at controlling fluorescence and ground state vibrational coherences, probably due to the role of excited-state absorption.