[Paper Review] Effects of electronic correlation on the high harmonic generation in helium: a time-dependent configuration interaction singles vs time-dependent full configuration interaction study

Aleksander P. Woźniak, Michał Przybytek+2 more|arXiv (Cornell University)|Jan 23, 2022

Laser-Matter Interactions and Applications143 references15 citations

TL;DR

This study compares time-dependent configuration interaction with single excitations (TD-CIS) and full configuration interaction (TD-FCI) to assess the impact of electronic correlation on high harmonic generation (HHG) in helium. Using real-time wavefunction propagation with Gaussian basis sets augmented for continuum states, it demonstrates that including dynamical electron correlation significantly improves the plateau intensity stability in HHG spectra, but only when the basis set adequately describes the uncorrelated cutoff position.

ABSTRACT

In this paper, we investigate the effects of full electronic correlation on the high harmonic generation in the helium atom subjected to laser pulses of extremely high intensity. To do this, we perform real-time propagations of the helium atom wavefunction using quantum chemistry methods coupled to Gaussian basis sets. The calculations are done within the real-time time-dependent configuration interaction framework, at two levels of theory: time-dependent configuration interation with single excitations (TD-CIS, uncorrelated method) and time-dependent full configuration interaction (TD-FCI, fully correlated method), and analyse obtained HHG spectra. The electronic wavefunction is expanded in Dunning basis sets supplemented with functions adapted to describing highly excited continuum states. We also compare the TD-CI results with grid-based propagations of the helium atom within the single-active-electron approximation. Our results show when including the dynamical electron correlation, a noticeable improvement to the description of HHG can be achieved, in terms of e.g. a more constant intensity in the lower energy part of the harmonic plateau. However, such effects can be captured only if the basis set used suffices to reproduce the most basic features, such as the HHG cutoff position, at the uncorrelated level of theory.

Motivation & Objective

• To investigate the role of electronic correlation in high harmonic generation (HHG) in helium under intense laser fields.
• To compare the accuracy of uncorrelated TD-CIS versus fully correlated TD-FCI methods in describing HHG spectra.
• To assess whether the computational cost of full correlation is justified by improved physical description.
• To determine the minimal basis set requirements for meaningful HHG results at the correlated level.

Proposed method

• Performs real-time time-dependent configuration interaction (TDDCI) wavefunction propagation using Gaussian basis sets.
• Employs two levels of theory: TD-CIS (uncorrelated, single excitations only) and TD-FCI (fully correlated, including all excitations).
• Uses Dunning-style basis sets augmented with functions tailored for continuum states and Rydberg excitations.
• Compares results with grid-based single-active-electron (SAE) calculations to validate the correlated approach.
• Applies complex absorbing potentials (CAPs) to simulate ionization and prevent wavefunction reflection.
• Computes HHG spectra from the time-dependent dipole acceleration.

Experimental results

Research questions

• RQ1 Does including full electronic correlation in TD-FCI improve the description of HHG in helium compared to TD-CIS?
• RQ2 How does the inclusion of double excitations affect the harmonic plateau intensity and cutoff position?
• RQ3 Can the improved physical description from correlation be achieved only with sufficiently large basis sets?
• RQ4 To what extent do basis set limitations at the uncorrelated level limit the benefits of correlation at the correlated level?
• RQ5 How do the results compare to single-active-electron models in terms of spectral shape and cutoff?

Key findings

• Including dynamical electron correlation via TD-FCI leads to a more stable and constant intensity in the lower-energy part of the HHG plateau compared to TD-CIS.
• The improvement in plateau structure is only observable when the basis set at the TD-CIS level already correctly reproduces the HHG cutoff position.
• TD-FCI results show better agreement with grid-based SAE calculations than TD-CIS, especially in the plateau region.
• The inclusion of double excitations in the CI expansion significantly enhances the description of electron correlation effects in strong-field ionization and recombination.
• The basis set must be sufficiently flexible at the uncorrelated level to allow meaningful gains from correlation at the correlated level.
• The study confirms that basis set completeness is a prerequisite for capturing correlation effects in HHG, even when using highly correlated methods.