Skip to content

Advances in Quantum Defect Embedding Theory

MetadataDetails
Publication Date2025-08-13
JournalJournal of Chemical Theory and Computation
AuthorsSiyuan Chen, Victor Yu, Yu Jin, Marco Govoni, Giulia Galli
InstitutionsUniversity of Chicago, University of Modena and Reggio Emilia
Citations1
AnalysisFull AI Review Included

This research details significant methodological advances in Quantum Defect Embedding Theory (QDET) aimed at improving the accuracy and reliability of simulating correlated electronic states in solid-state quantum defects.

  • Refined Double-Counting (DC2025): A new double-counting correction was derived that consistently incorporates the frequency dependence of the screened Coulomb interaction, leading to VEE shifts up to 0.25 eV and improved agreement with experimental data for neutral Group IV vacancies.
  • Unoccupied Orbitals Impact: Including unoccupied Kohn-Sham (KS) orbitals in the active space was found to be non-negligible for the Nitrogen-Vacancy (NV-) center, improving its Vertical Excitation Energy (VEE) by approximately 0.1 eV, while having a negligible effect (< 0.01 eV) on neutral Group IV vacancies.
  • Hybridization Effects: An approximate method to include hybridization between the active space and the environment (using bath orbitals) was introduced. The resulting VEE shifts were small, ranging from 12 meV (NV-) to 0.07 eV (PbV0).
  • Functional Dependence: QDET results showed moderate dependence on the initial Density Functional Theory (DFT) functional. The dielectric dependent hybrid (DDH) functional yielded VEEs up to 0.2 eV higher than the PBE functional, resulting in better overall agreement with experimental NV- VEEs.
  • Impurity Solver Validation: Consistency checks across four different impurity solvers (FCI, Selected CI, MR-CISD, AFQMC) confirmed that the results are within chemical accuracy (0.04 eV).
  • Molecular Qubit Application: The refined QDET method was successfully applied to the Cr(o-tolyl)4 molecular qubit, achieving excellent agreement with the experimental VEE (1.225 eV computed vs. 1.241 eV experimental).
ParameterValueUnitContext
Supercell Size (Max)1,727atomsDemonstrated scalability of QDET
Plane Wave Cutoff60RyDFT calculations (Quantum ESPRESSO code)
PDEP Convergence2 to 3 times NeN/ANumber of projective dielectric eigenpotentials (Ne = total electrons)
VEE Convergence Threshold10meVRequired PDEP convergence accuracy
DDH Mixing Parameter (α)0.18N/AInverse of macroscopic dielectric constant of bulk diamond
NV- VEE (DC2025, DDH, Unocc.)2.235eVFinal computed VEE (compared to Exp. 2.18 eV)
Cr(o-tolyl)4 VEE (QDET)1.225eVFirst singlet excited state VEE (compared to Exp. 1.241 eV)
Hybridization Effect (Group IV Max)< 0.07eVMaximum VEE shift due to approximate hybridization (PbV0)
Hybridization Effect (NV-)12meVVEE shift due to ensemble-averaged hybridization
AFQMC Precision Limit~0.01eVEstimated precision error of Auxiliary Field Quantum Monte Carlo solver

The computational workflow integrates DFT, many-body perturbation theory (G0W0), and high-level quantum chemistry impurity solvers.

  1. DFT Starting Point:
    • Restricted DFT calculations performed using the Quantum ESPRESSO (QE) code with PBE or dielectric dependent hybrid (DDH) functionals.
    • SG15 PBE norm-conserving pseudopotentials and a 60 Ry kinetic energy cutoff were used.
  2. Active Space Selection:
    • Orbitals partitioned using the Minimum Model plus KS Energy (MM+KSE) approach.
    • The active space includes minimum model orbitals plus occupied KS orbitals close to the Valence Band Maximum (VBM), and optionally, unoccupied orbitals close to the Conduction Band Minimum (CBM).
  3. Low-Level Environment (G0W0):
    • G0W0 calculations performed using the WEST code to obtain quasiparticle energies, screening (W0), and polarizabilities (P0).
    • Convergence ensured by setting the number of Projective Dielectric Eigenpotentials (PDEPs) to three times the number of electrons (Ne).
  4. Effective Hamiltonian Construction (Heff):
    • The effective one-body term (teff) and two-body term (veff) were constructed.
    • The novel “refined double-counting” scheme (DC2025) was applied to ensure consistent treatment of frequency dependence in the effective Hamiltonian.
  5. Hybridization Modeling (Haux):
    • An auxiliary Hamiltonian (Haux) was constructed to account for hybridization effects (Δ(ω)).
    • Bath orbitals were selected based on their contribution (Sb) to the hybridization term, using a threshold of T = 2/3 (66.67%).
  6. Impurity Solving:
    • The effective Hamiltonian was diagonalized using various high-level solvers: Full Configuration Interaction (FCI), Selected Configuration Interaction (Selected CI), Multi-reference Configuration Interaction Single and Doubles (MR-CISD/CIS(D)), and Auxiliary Field Quantum Monte Carlo (AFQMC).
    • For states split by hybridization, an “ensemble-averaged” VEE was computed to mimic the expected excited state in the full system.

This advanced QDET methodology is critical for the accurate theoretical characterization and engineering of quantum materials, particularly those used in solid-state quantum technology.

  • Quantum Computing:
    • Solid-State Qubits: Precise prediction of electronic and optical properties (VEEs) for diamond defects (NV-, SiV0) which serve as leading candidates for spin qubits.
    • Molecular Qubits: Validation and design of novel molecular systems, such as Cr(o-tolyl)4, for use in scalable quantum information processing architectures.
  • Quantum Sensing and Metrology:
    • Optimization of defect properties (e.g., energy levels, spin states) in wide-bandgap semiconductors (like diamond) to enhance their performance as nanoscale sensors for magnetic fields and temperature.
  • Materials Discovery and Engineering:
    • Enables ab initio simulation of strongly correlated states in large, realistic supercells, accelerating the discovery and integration of new point defects in semiconductors and insulators for various technological applications.
  • High-Performance Simulation Software:
    • The methodology relies on and drives improvements in large-scale computational codes (e.g., Quantum ESPRESSO, WEST, PySCF), benefiting the broader computational materials science community.
View Original Abstract

Quantum defect embedding theory (QDET) is a many-body embedding method designed to describe condensed systems with correlated electrons localized within a given region of space, for example spin defects in semiconductors and insulators. Although the QDET approach has been successful in predicting the electronic properties of several point defects, several limitations of the method remain. In this work, we propose multiple advances to the QDET formalism. We derive a double-counting correction that consistently treats the frequency dependence of the screened Coulomb interaction, and we illustrate the effect of including unoccupied orbitals in the active space. In addition, we propose a method to describe hybridization effects between the active space and the environment, and we compare the results of several impurity solvers, providing further insights into improving the reliability and applicability of the method. We present results for defects in diamond and for molecular qubits, including a detailed comparison with experiments.