GeauxCTQMC

Official Resources

• Homepage: https://github.com/GeauxCTQMC/GeauxCTQMC (or LA-SiGMA links)
• Source Repository: https://github.com/GeauxCTQMC/GeauxCTQMC
• License: Open Source (BSD 3-Clause likely, check header)

Overview

GeauxCTQMC (pronounced "Go-CTQMC") is a highly optimized Continuous-Time Quantum Monte Carlo (CT-QMC) code based on the Hybridization Expansion (CT-HYB) algorithm. Developed under the LA-SiGMA (Louisiana Alliance for Simulation-Guided Materials Applications) project, it is designed to solve single-impurity Anderson models (SIAM) effectively, serving as the computational engine for Dynamical Mean Field Theory (DMFT) studies of strongly correlated materials.

Scientific domain: Strongly Correlated Materials, DMFT, Quantum Impurity Solvers Target user community: DMFT practitioners, Materials Scientists (Lanthanides/Actinides)

Theoretical Methods

• CT-HYB: Continuous-Time Hybridization Expansion QMC.
• Segment/Matrix Updates: Techniques for handling density-density and general interactions.
• Retarded Interactions: Ability to treat dynamically screened interactions $U(\omega)$.
• Measurement: Accumulation of Green's functions in imaginary time $G(\tau)$ and Legendre coefficients.

Capabilities (CRITICAL)

• Impurity Solver: Core function is to solve the effective impurity problem within DMFT.
• General Interactions: Can handle full Coulomb interactions including spin-flip and pair-hopping terms (Matrix formalism).
• Performance: High-performance C++ implementation using MPI for massive parallelization over Monte Carlo walkers.
• Cluster DMFT: Support for multi-orbital and small cluster impurities.

Key Features

Performance:

• MPI Parallelism: Scales to thousands of cores.
• Optimized Moves: Efficient proposal and acceptance of updates (insert/remove segments).

Advanced Physics:

• Dynamic U: Handles frequency dependent interactions, important for screened Coulomb interactions in solids (cRPA+DMFT).
• Superconductivity: Can measure Nambu spinors for superconducting phases.

Inputs & Outputs

• Input formats:
  • Uses standard text/JSON or HDF5 inputs for hybridization function $\Delta(\tau)$ and interaction parameters.
• Output data types:
  • Green's function $G(\tau)$.
  • Self-energy (if post-processed).
  • Measurement statistics.

Interfaces & Ecosystem

• Upstream: Interfaces with DFT+DMFT codes (e.g., Haule's code, or generic DMFT loops).
• Downstream: MaxEnt codes for analytic continuation.

Workflow and Usage

Typically called as an external executable within a DMFT self-consistency loop. The loop generates $\Delta(\tau)$, GeauxCTQMC calculates $G(\tau)$, and the loop updates the Weiss field.

Performance Characteristics

• Speed: Optimized for modern HPC architectures.
• Accuracy: Exact within statistical error (no discretization error in time).

Comparison with Other Codes

Verification & Sources

Primary sources:

1. GitHub Repository: https://github.com/GeauxCTQMC/GeauxCTQMC
2. LA-SiGMA Project website.
3. Publications referencing the code (e.g., PRB papers from Louisiana State University group).

Verification status: ✅ VERIFIED

• Source code: OPEN
• Origin: Academic research code (LA-SiGMA).