Skip to main content
QUICK REVIEW

[Paper Review] GW quasiparticle calculations with spin-orbit coupling for the light actinides

Towfiq Ahmed, R. C. Albers|arXiv (Cornell University)|Sep 16, 2013
Physics of Superconductivity and Magnetism4 citations
TL;DR

This study presents one-shot GW quasiparticle calculations with spin-orbit coupling for light actinides (Np, U, Pu) using a full-potential LAPW method, demonstrating significant band renormalization and enhanced electronic correlation in the intermediate coupling regime. The inclusion of spin-orbit coupling improves agreement with experimental photoemission spectra, particularly near the Fermi level, and reveals increasing screened Coulomb interactions with lattice expansion.

ABSTRACT

We report on the importance of GW self-energy corrections for the electronic structure of light actinides in the weak-to-intermediate coupling regime. Our study is based on calculations of the band structure and total density of states of Np, U, and Pu using a one-shot GW approximation that includes spin-orbit coupling within a full potential LAPW framework. We also present RPA screened effective Coulomb interactions for the f-electron orbitals for different lattice constants, and show that there is an increased contribution from electron-electron correlation in these systems for expanded lattices. We find a significant amount of electronic correlation in these highly localized electronic systems.

Motivation & Objective

  • To investigate the role of GW self-energy corrections in the electronic structure of light actinides with strong spin-orbit coupling.
  • To assess the impact of electron-electron correlation in the weak-to-intermediate coupling regime for 5f electrons in Np, U, and Pu.
  • To evaluate the RPA-screened Coulomb interaction W(ω=0) for f-orbitals as a function of lattice constant to quantify correlation strength.
  • To benchmark GW+SO calculations against experimental photoemission spectroscopy and LDA results, emphasizing accuracy in spectral weight and valence electron count.
  • To provide a first-principles foundation for future GW+DMFT studies by establishing a reliable starting point with spin-orbit coupling.

Proposed method

  • Employed a one-shot GW approximation within a full-potential linearized augmented plane wave (LAPW) framework to compute quasiparticle band structures.
  • Included spin-orbit coupling via a relativistic single-particle Hamiltonian, treating the SO interaction as a perturbation added to the Kohn-Sham potential.
  • Used the random phase approximation (RPA) to compute the screened Coulomb interaction W(ω) and projected it onto f-orbitals at ω=0 to obtain effective U-like interactions.
  • Calculated photoemission spectra using the real-frequency method for the self-energy, avoiding analytical continuation issues common in other GW implementations.
  • Performed systematic calculations across Np, U, Pu, and an extended Pu system with varying lattice constants to probe the itinerant-to-localized crossover.
  • Ensured conservation of spectral weight and occupied valence electron count by using a real-frequency approach, improving reliability over methods requiring analytical continuation.

Experimental results

Research questions

  • RQ1How do GW self-energy corrections with spin-orbit coupling affect the band structure and density of states in light actinides?
  • RQ2To what extent does increasing lattice spacing enhance electronic correlation in 5f systems, as measured by the screened Coulomb interaction?
  • RQ3How does the inclusion of spin-orbit coupling in GW calculations improve agreement with experimental photoemission spectra compared to LDA or non-relativistic GW?
  • RQ4What is the role of the RPA-screened Coulomb interaction W(ω=0) in quantifying static correlation effects in partially filled f-electron systems?
  • RQ5Can one-shot GW with spin-orbit coupling serve as a reliable starting point for future GW+DMFT studies of strongly correlated actinides?

Key findings

  • The one-shot GW+SO calculation significantly renormalizes the band structure, particularly broadening the spectral peak near the Fermi level, which improves agreement with experimental photoemission data.
  • For δ-Pu with expanded lattice spacing, the SO-coupled peak at 1.2 eV shifts 1.5 eV to higher binding energy, indicating strong correlation effects in the localized regime.
  • The RPA-screened Coulomb interaction W(ω=0) for f-orbitals increases with lattice expansion, showing a clear trend of enhanced electronic correlation in more dilute systems.
  • The effective Coulomb interaction W(ω=0) scales with the bare Coulomb interaction V_ff, confirming that correlation strength grows as f-orbitals become more localized.
  • The real-frequency GW method preserves spectral weight and electron count, unlike analytical continuation-based approaches, ensuring better consistency with LDA results.
  • GW+SO calculations yield better agreement with experimental PES than LDA or non-relativistic GW, especially for the three main peaks near the Fermi edge, validating the importance of including both correlation and spin-orbit effects.

Better researchstarts right now

From reading papers to final review, dramatically reduce your research time.

No credit card · Free plan available

This review was created by AI and reviewed by human editors.