Femtochemistry and Laser Control of Photochemical Reactions

Since the advent of the laser in the 1960’s, the possibility of using coherent light sources to control molecules and molecular reactions has been a hot topic in photochemistry. The ability to generate laser pulses on ever shorter time scales, down to femto- and even atto-second durations, as well as the development of optical techniques for actively controlling the time-dependent features of these pulses, has driven the field forward into new lines of theoretical and experimental inquiry.

At higher field strengths, polarization forces, i.e., the distortion of electronic states leading to induced dipole moments related to the polarizability, can play an important role. The application and control of these polarization forces (via ultrashort optimized laser pulses) has previously been exploited experimentally in order to control, e.g., the rotational, vibrational, and dissociation dynamics of molecules through the so-called dynamic Stark effect.

This doctoral thesis contains a series of theoretical works where the concept of exploiting the dynamic Stark effect to control quantum molecular wave packets through dynamic Stark control (DSC) is introduced and explored through a variety of different applications. In the second chapter, a clarifying interpretation is presented of how the dynamic Stark effect operates by shifting the energy levels of the potential surfaces present in a molecule using a simplified 1D harmonic model.

Using this model, general properties and relationships are derived that paint an intuitive picture of this type of second order excitation process from a temporal, as well as a spectral, perspective. In the third chapter, the concept of DSC is extended to include diatomic molecular rotations as well as vibrations, and it is shown that the spectral interpretation of the second order excitation process associated with DSC can be exploited to exert state selective control over ro-vibrational transitions.

Inspired and verified by impulsive alignment experiments performed on gas phase I2 molecules, the aforementioned rotational model is then further extended to include the coupling between the quadrupole moments of the atomic nuclei and the molecular rotational states. It is demonstrated that this so-called quadrupole coupling will significantly perturb the time dependent alignment traces in molecules with large quadrupole coupling constants when they are prepared in superpositions of rotational states by ultrashort laser pulses.

In the fourth chapter, the concept of vibrational DSC is revisited and extended, and it is demonstrated that this approach can be used to control the delicate process of laser induced enantiomeric conversion (deracemization) in a racemic mixture of biaryl molecules. This is achieved using a simulated closed loop experimental setup to optimize the phases of a pair of nonresonant, linearly polarized Gaussian laser pulses.

By carefully modelling the optimization process to accurately reflect current experimental techniques and capabilities, this work represents an advance towards a first-ever experimental implementation of laserinduced enantiomeric conversion. This result is then extended by outlining a simplified approach to the deracemization task that does not require appreciable optimization of the laser pulse shape.

While less effective than the full closed-loop approach, this method has the advantage of being significantly less challenging to implement experimentally. In the fifth chapter, a different approach to experimental DSC is outlined. This method is based on the application of a trained neural network to generate a control field based on dynamic experimental feedback from the molecular system.