Real-Space-BSE

**Real-Space-BSE** is a specialized Fortran 2003 library implementing the **Bethe-Salpeter Equation (BSE)** in a **real-space** formalism. Unlike many plane-wave or k-space codes, this tool is specifically designed for large **molecular…

2. TDDFT & EXCITED-STATE 2.4 BSE Methods VERIFIED 1 paper
Back to Mind Map Official Website

Overview

**Real-Space-BSE** is a specialized Fortran 2003 library implementing the **Bethe-Salpeter Equation (BSE)** in a **real-space** formalism. Unlike many plane-wave or k-space codes, this tool is specifically designed for large **molecular systems**, utilizing a localized basis set to efficiently compute optical absorption spectra and optical band gaps for systems with thousands of atoms.

Reference Papers (1)

Full Documentation

Official Resources

  • Source Code: https://github.com/AlexBuccheri/Bethe-Salpeter
  • Developer: Alex Buccheri
  • License: See Repository

Overview

Real-Space-BSE is a specialized Fortran 2003 library implementing the Bethe-Salpeter Equation (BSE) in a real-space formalism. Unlike many plane-wave or k-space codes, this tool is specifically designed for large molecular systems, utilizing a localized basis set to efficiently compute optical absorption spectra and optical band gaps for systems with thousands of atoms.

Theoretical Methods

  • Real-Space Formulation: Solves the BSE directly in real-space coordinates.
  • Minimal Tight-Binding Basis: Uses a simplified basis for efficiency.
  • Two-Center Approximation: Approximates integrals to speed up computation.
  • Full BSE: Solves the full equation, not restricted to the Tamm-Dancoff Approximation (TDA).
  • Monopole Approximation: Used for Coulomb terms beyond nearest neighbors.

Capabilities

  • Optical Spectra: Calculation of optical absorption cross-sections.
  • Optical Band Gaps: Direct determination of optical gaps including excitonic binding.
  • Large Systems: Tested on molecular systems up to 6000 atoms (as of 2017).
  • Molecule-Only: Specifically for finite molecular systems, not periodic solids.

Implementation & Tech Stack

  • Language: Fortran 2003 (F2003 library).
  • Status: Research code / Library.
  • Parallelization: Real-space methods are naturally parallelizable, though specific MPI/OpenMP details depend on the driver.

Application Areas

  • Large Molecules: Organic semiconductors, polymers, large biomolecules.
  • Nanostructures: Finite quantum dots or clusters where periodic boundaries are artificial.
  • Excitonics: Study of spatial exciton extent in real space.

Related Tools in 2.4 BSE Methods