Chirped Pulse Control of Raman Coherence in Atoms and Molecules

TL;DR

This paper introduces a robust chirped pulse control scheme based on C-CARS for maximizing vibrational coherence in atoms and molecules, with applications in remote sensing and imaging, demonstrated through numerical simulations and deep learning analysis.

Contribution

It presents a novel chirped pulse control method using C-CARS that enhances vibrational coherence and selectivity, incorporating theoretical derivation, numerical simulations, and deep learning techniques.

Findings

Effective control of vibrational coherence demonstrated in simulations

Spectral selectivity achieved through chirp rate optimization

Deep learning used to analyze control pulse characteristics

Abstract

A novel chirped pulse control scheme is presented based on Coherent Anti-Stokes Raman Spectroscopy (C-CARS) aiming at maximizing the vibrational coherence in atoms and molecules. The scheme utilizes chirping of the three incoming pulses, the pump, the Stokes and the probe, in the four-wave mixing process of C-CARS to fulfill the adiabatic passage conditions. The derivation of the scheme is based on simplifying the four-level system into a 'super-effective' two level system via rotating wave approximation and adiabatic elimination of the excited state manifold. The robustness, spectral selectivity and adiabatic nature of C-CARS method may prove useful for sensing, imaging, and detection. It is demonstrated that the selectivity in excitation of vibrational degrees of freedom can be controlled by carefully choosing the spectral chirp rate of the pulses. The C-CARS control scheme is applied to a surrogate methanol molecule to generate an optimal anti-Stokes signal backscattered from a cloud of molecules a kilometer away. The theory is based on the solution of the coupled Maxwell-Liouville von Neumann equations and focuses on the quantum effects induced in the target molecules by the control pulse trains. The propagation effects of pulses through the medium are evaluated and the buildup of the molecular-specific anti-Stokes signal is demonstrated numerically. A deep learning technique, using Convolutional Neural Networks (CNN), is implemented to characterize the control pulses and evaluate time-dependent phase characteristics from them. The effects of decoherence induced by spontaneous decay and collisional dephasing are also examined. Additionally, we present the technique of Fractional Stimulated Raman Adiabatic Passage (F-STIRAP) and demonstrate that it can be utilized for remote detection in a multi-level system by creation of a maximally coherent superposition state.