Mark Baertschy

Gaining mechanistic insight from closed loop learning control: The importance of basis in searching the phase space

This paper discusses different routes to gaining insight from closed loop learning control experiments. We focus on the role of the basis in which pulse shapes are encoded and the algorithmic search is performed. We demonstrate that a physically motivated, nonlinear basis change can reduce the dimensionality of the phase space to one or two degrees of freedom. The dependence of the control goal on the most important degrees of freedom can then be mapped out in detail, leading toward a better understanding of the control mechanism. We discuss simulations and experiments in selective molecular fragmentation using shaped ultrafast laser pulses.

Interpreting closed-loop learning control of molecular fragmentation in terms of wave-packet dynamics and enhanced molecular ionization

We interpret a molecular fragmentation experiment using shaped, ultrafast laser pulses in terms of enhanced molecular ionization during dissociation. A closed-loop learning control experiment was performed to maximize the CF3+CH3+ production ratio in the dissociative ionization of CH3COCF3. Using ab inito molecular structure calculations and quasistatic molecular ionization calculations along with data from pump-probe experiments, we identify the primary control mechanism which is quite general and should be applicable to a broad class of molecules.

Phase-matching conditions for nonlinear frequency conversion by use of aligned molecular gases.

Transient birefringence can be induced in a gas of anisotropic molecules by an intense polarized laser pulse. We propose to use this birefringence to phase match nonlinear optical frequency-conversion processes. The conditions for anisotropic phase matching are derived, and experimental conditions required for phase-matched third-harmonic generation in a gas-filled hollow-core fiber are presented. We show that these conditions are experimentally feasible over a significant parameter range, making possible a new type of nonlinear optics.

Solution of a Three-Body Problem in Quantum Mechanics Using Sparse Linear Algebra on Parallel Computers

A complete description of two outgoing electrons following an ionizing collision between a single electron and an atom or molecule has long stood as one of the unsolved fundamental problems in quantum collision theory. In this paper we describe our use of distributed memory parallel computers to calculate a fully converged wave function describing the electron-impact ionization of hydrogen. Our approach hinges on a transformation of the Schrödinger equation that simplifies the boundary conditions but requires solving very ill-conditioned systems of a few million complex, sparse linear equations. We developed a two-level iterative algorithm that requires repeated solution of sets of a few hundred thousand linear equations. These are solved directly by LU-factorization using a specially tuned, distributed memory parallel version of the sparse LU-factorization library Super-LU. In smaller cases, where direct solution is technically possible, our iterative algorithm still gives significant savings in time and memory despite lower megaflop rates.

Optimal single-pulse excitation of rotational impulsive molecular phase modulation

The full transient macroscopic linear optical susceptibility tensor induced in a transiently aligned molecular gas by a single, linearly polarized intense alignment pulse is studied. We determine the optimal properties of the pulse that forms the rotational wave packet. Significantly, we demonstrate that the optimal pulse for phase modulation differs from the optimal alignment pulse. Finally, we show that the limited information about rotational wave packets obtained by measuring the linear optical susceptibility can be augmented by also measuring the time-varying nonlinear optical susceptibilities.

Transient Optical Susceptibility Induced by Nonperturbative Rotational Wave Packets

Rotational wave packets formed in a gas of linear molecules can produce a time-dependent index of refraction that can be used as a phase modulator for an ultrafast pulse [1]. A collection of anisotropic molecules can be made to align along the polarization direction of an intense, linearly polarized laser pulse [1]–[4]. Although the molecules will quickly go out of alignment, the periodic rephasing of the wave packet causes the molecules to come back into alignment at regular intervals. During these so-called rotational revivals, the wave packet evolves through states where the molecules are both aligned and antialigned (i.e., perpendicular to) the polarization of the laser pulse.

Controlling molecular fragmentation: Adiabatic rapid passage as a mechanism for charge transfer

Control over CF₃ ⁺/CHBr₂ ⁺ in laser driven fragmentation of CHBr₂COCF₃, is observed. Pump-probe spectroscopy reveals a charge transfer mechanism. This mechanism may allow for a measurement of the wave function for a dissociating polyatomic molecule.

Understanding Learning Control in a Molecular Family

Experiments demonstrate coherent control over a wide variety of atomic and molecular systems using shaped ultrafast laser pulses. In many cases, where the system Hamiltonian is not known well enough to predict optimal control fields a priori, learning algorithms are used. Although learning algorithms are successful at discovering optimal laser pulses for control, it is generally difficult to understand the physical mechanisms underlying control. Although a few pioneering experiments have been able to uncover the control mechanism exploited by an optimal pulse, the lack of systematic techniques to gain insights into control mechanisms and to develop predictive models for control remains a significant barrier to achieving the ultimate goal of laser selective chemistry. This chapter uses a combination of learning algorithm modifications, pump-probe spectroscopy, ab initio molecular structure calculations, and quasi-static molecular ionization calculations to decipher the mechanism underlying control in l, l, l-trifluoroacetone. This predicts control in similar molecules such as tri-deuterated acetone and 1,1,1-trichloroacetone.

Interpreting learning control of molecular fragmentation

This presentation will focus on progress toward uncovering the physical mechanisms underlying control in a series of closed loop molecular fragmentation experiments using shaped ultrafast laser pulses.

Systematic Behavior in Chemical Reactions Driven by Shaped Ultrafast Laser Pulses

We investigate learning control over molecular fragmentation in a series of similar molecules. We interpret the physical mechanism underlying control in terms of an intuitive picture that is based upon ab initio molecular structure calculations.