Hybrid Gauge Approach for Accurate Real-Time TDDFT Simulations with Numerical Atomic Orbitals

H. W. Zhao, Lixin HeJournal of Chemical Theory and Computation 2025 · DOI

TL;DR

This paper introduces a "hybrid gauge" approach for simulating ultrafast electron dynamics in materials using real-time time-dependent density functional theory (rt-TDDFT) with numerical atomic orbitals (NAOs). This new method overcomes significant errors in existing velocity gauge approaches, leading to more accurate and reliable simulations of light-matter interactions.

What Problem Does This Paper Address?

Simulating the behavior of electrons in materials when exposed to extremely short and intense laser pulses (ultrafast dynamics) is crucial for understanding phenomena like chemical reactions, material responses to light, and energy transfer. Real-time time-dependent density functional theory (rt-TDDFT) is a powerful computational tool for this. However, when dealing with periodic systems (like crystals), the standard "velocity gauge" approach, which is necessary to maintain the system's periodicity, introduces errors when using numerical atomic orbitals (NAOs) as the basis set. This is because the velocity gauge formulation neglects how the vector potential of the electromagnetic field affects the phase of atomic orbitals, leading to inaccuracies, especially when calculating electron currents.

Methods

The authors propose a "hybrid gauge" that combines the strengths of both the "length gauge" (which uses the electric field) and the "velocity gauge" (which uses the vector potential). This hybrid approach explicitly includes both the electric field and the vector potential in the Hamiltonian, ensuring that the crucial position-dependent phase information within the atomic orbitals is preserved. This is achieved by modifying the basis functions to incorporate the time-dependent vector potential. The derivation involves careful mathematical formulation to ensure the correct inclusion of these phase factors. They then implement this hybrid gauge within the ABACUS code, which uses NAO basis sets.

Datasets

Not discussed in the paper. The paper focuses on theoretical development and computational simulations using established physical systems (H2CO3 molecule and bulk silicon) rather than training on specific datasets.

Key Insights and Ideas

  • The "phase problem" in velocity gauge rt-TDDFT with NAOs is real and significant: Neglecting the position-dependent phase variations induced by the vector potential leads to substantial errors, particularly in current calculations.
  • A hybrid gauge can elegantly solve this problem: By incorporating both the electric field and the vector potential, the hybrid gauge preserves essential phase information, leading to accurate results.
  • Hybrid gauge is computationally efficient: Unlike the velocity gauge, it avoids computationally expensive grid-point integrations for nonlocal potentials, relying instead on two-center integrals that can be pre-calculated for fixed ions.
  • The hybrid gauge is easily implementable: It requires only minimal modifications to existing length-gauge frameworks.
  • Periodic systems benefit greatly: While the length gauge is unsuitable for periodic systems, the hybrid gauge offers a viable and accurate alternative to the velocity gauge.

Key Results

  • Accurate time-dependent currents and optical spectra: For both molecular (H2CO3) and periodic (bulk silicon) systems, the hybrid gauge yields results that are in excellent agreement with the length gauge (for molecules) and with established literature, outperforming the uncorrected velocity gauge.
  • Elimination of unphysical low-frequency divergences: The velocity gauge often produces unphysical features in absorption spectra, which are absent in the hybrid gauge results.
  • Improved accuracy with larger basis sets: While increasing the basis set size reduces errors in the velocity gauge, the hybrid gauge provides accurate results even with smaller basis sets.
  • Reliable nonlinear response simulations: The hybrid gauge accurately captures nonlinear optical phenomena in bulk silicon under strong laser fields, including harmonic generation, matching experimental trends and literature values.
  • Numerical correction for velocity gauge helps but doesn't fully fix: A proposed numerical correction for the velocity gauge improves its accuracy but cannot entirely eliminate errors, especially at strong field intensities.

Limitations and Future Work

  • Numerical correction limitations: While the numerical correction for the velocity gauge improves results, it doesn't fully eliminate errors, particularly at strong field intensities and with smaller basis sets.
  • Future work: The authors suggest that this robust and efficient hybrid gauge framework paves the way for studying more complex ultrafast phenomena, such as strong-field excitations, high-harmonic generation, and light-induced phase transitions, in a wider range of systems. They also mention the potential for studying systems with moving ions, which would require further development.

Summary updated

Abstract

Ultrafast real-time dynamics are critical for understanding a broad range of physical processes. Real-time time-dependent density functional theory (rt-TDDFT) has emerged as a powerful computational tool for simulating these dynamics, offering insight into ultrafast processes and light-matter interactions. In periodic systems, the velocity gauge is essential because it preserves the system's periodicity under an external electric field. Numerical atomic orbitals (NAOs) are widely employed in rt-TDDFT codes due to their efficiency and localized nature. However, directly applying the velocity gauge within the NAO basis set neglects the position-dependent phase variations within atomic orbitals induced by the vector potential, leading to significant computational errors - particularly in current calculations. To resolve this issue, we develop a hybrid gauge that incorporates both the electric field and the vector potential, preserving the essential phase information in atomic orbitals and thereby eliminating these errors. Our benchmark results demonstrate that the hybrid gauge fully resolves the issues encountered with the velocity gauge in NAO-based calculations, providing accurate and reliable results. This algorithm offers a robust framework for future studies on ultrafast dynamics in periodic systems using NAO bases.
BibTeX
@article{zhao2025hybrid,
  author = {H. W. Zhao and Lixin He},
  title = {{Hybrid Gauge Approach for Accurate Real-Time TDDFT Simulations with Numerical Atomic Orbitals}},
  journal = {Journal of Chemical Theory and Computation},
  year = {2025},
  doi = {10.1021/acs.jctc.5c00111},
}

Loading related content...

Figure