Computer Physics Communications

Simphony is an open-source software package designed for the topological analysis of lattice vibrations based on Wannier tight-binding models. Its primary function is to classify the topology of novel materials by computing bulk and slab phonon band structures, extracting phonon surface spectra, and providing analysis tools such as Wilson loop calculations and Weyl node detection. The workflow is analogous to that of established electronic topology codes like Wannier90 and WannierTools. It also incorporates long-range polar interactions during the wannierization process, making Simphony one of the first tools capable of diagnosing topology in polar insulators.

We present the Multiscale Universal Interface 2.0 library, the next major release of the lightweight, header-only C++ library for scalable multiphysics or multiscale code coupling. Designed to lower barriers to integration, this release introduces key enhancements in non-conformal data mapping, coupling algorithms, performance scalability, and language interoperability. A new Radial Basis Function spatial sampler enables conservative data transfer across non-conformal interfaces, while built-in modules support robust relaxation schemes for strongly coupled systems with dynamic interfaces. The addition of a dedicated linear algebra subsystem eliminates previous external dependencies and provides a foundation for advanced interpolation and solver acceleration. Accessibility is further expanded through redesigned wrappers for C, Fortran, and Python, and smart communication strategy improves High-Performance Computing scalability via refined communication strategies. Benchmarks on supercomputers demonstrate near-linear scalability under heavy communication loads. This release represents a significant step toward enabling flexible, high-performance multiphysics simulations across diverse platforms and disciplines.

Understanding the spatial character of electronic excitations, in particular the distinction between local and charge-transfer (CT) contributions, is essential in the analysis of excited-state phenomena. However, commonly used approaches often rely on method-specific representations or implicit partitioning schemes, limiting their reproducibility and transferability across electronic-structure frameworks. Domain Assignment and Interface Solution in pYthon (DAISpY, pronounced ”daisy”) presents a standalone and format-agnostic tool for domain-based charge-transfer (CT) analysis of excited states. The method builds on a general representation of excited states in terms of configuration interaction CI-like amplitudes and aggregates their contributions into domain → domain CT matrices. Orbital contributions are assigned to user-defined spatial domains, enabling a direct and intuitive mapping of electron-hole redistribution between molecular fragments. The formalism consistently treats singles, pairs, and doubles excitations while preserving normalization and avoiding double counting. DAISpY is designed as a modular and reproducible analysis framework, supporting multiple input routes, including electronic-structure checkpoint files and portable data representations. The implementation is independent of any specific quantum chemistry package, ensuring broad applicability.

In this paper we present Open Multi-Processing (OpenMP) Fortran 90/95 programs to solve the Gross-Pitaevskii equation for a rotating dipolar Bose-Einstein condensate (BEC) in two and three dimensions, which is a new version of our previous published programs for a dipolar Bose-Einstein condensate without rotation. After the recent experimental study of a rotating dipolar BEC [L. Klaus et al., Nature Phys. 18, 1453 (2022)], the present programs will be useful tools for related theoretical investigation. The algorithm used is the split-step semi-implicit Crank-Nicolson scheme for imaginary- and real-time propagation to obtain stationary states and BEC dynamics, respectively, as in the previous version [L. E. Young-S. et al., Comput. Phys. Commun. 286 (2023) 108669].

We present binningverdict, an open-source Python package for quantitatively selecting logarithmic or linear binning in small-angle neutron scattering (SANS) data reduction. The framework combines correlation-aware variance–bias optimization with Fisher-information analysis in a computationally efficient workflow operating directly on measured (Q, I, σ) profiles. Central to the method is a closed-form reconstruction-error ratio, 𝑅_{MSE}=𝐸*_{log}/𝐸*_𝐿, obtained by extending the Freedman–Diaconis mean-squared-error decomposition to Q-dependent bin widths. The resulting expression depends only on four elementary moments of the data and is entirely model-free. An optional Fisher-information module evaluates finite-N parameter-estimation precision and demonstrates asymptotic scheme-independence between logarithmic and linear discretizations. The package provides convergence diagnostics, batch-processing capability, and millisecond-scale execution using only NumPy. Benchmarks on canonical SANS models, together with applications to measured EQ-SANS datasets, demonstrate robust performance and establish the framework as a practical tool for automated SANS reduction and preprocessing workflows.

We introduce WimPyC, a Python code for the calculation of the capture rate of Weakly Interacting Massive Particles (WIMPs) by celestial bodies through nuclear scattering in the optically thin regime. WimPyC is an extension of the WimPyDD code, that calculates WIMP–nucleus scattering signals in direct detection (DD) experiments, and allows to combine DD and capture in celestial bodies in virtually any scenario within the framework of Galilean–invariant non–relativistic effective theory (NREFT), including inelastic scattering, an arbitrary WIMP spin and a generic WIMP velocity distribution in the Galactic halo. WimPyDD and WimPyC are suitable for both top–down approaches, where the interaction operators of a high–energy physics model are matched to those of the NREFT, and for bottom–up studies, where the Wilson coefficients of the NREFT are explored in a model–independent way and/or where the velocity distribution is written in terms of a superposition of streams taken as free parameters. As in the case of WimPyDD WimPyC exploits the factorization of the three main components that enter in the calculation of the capture rate: i) the Wilson coefficients that encode the dependence of the signals on the ultraviolet completion of the effective theory; ii) a response function that depends on the nuclear physics; iii) the halo function that depends on the WIMP velocity distribution. In WimPyC these three components are calculated and stored separately for later interpolation and combined together only as the last step of the signal evaluation procedure. This makes the phenomenological study of the capture rate with WimPyC transparent and improves computational speed.

PACme is a C++ computer code developed and optimized for the simulation, data analysis, and visualization of experimental (Time Differential) Perturbed Angular Correlations (PAC) observables resulting from both static and dynamic (time-dependent) magnetic and electric quadrupole hyperfine fields. The program is fully generalized to simultaneously handle fractions of probe atoms in different environments, each subject to static or multiple transient states. The code works for an arbitrary probe nuclear spin embedded in single-crystal or polycrystalline materials. A distinctive feature of PACme is the analytical treatment of static hyperfine field distributions, which are incorporated directly into the Hamiltonian formulation, even in the study of dynamic interactions. This approach is motivated by the fact that distributions of electric field gradients or magnetic fields around probe atoms-often caused by uncorrelated defects-are common and have a significant impact on the resulting PAC spectra. This article presents the implementation, structure, and usage of the PACme code as a general-purpose tool for the simulation and analysis of TDPAC spectra.

Pulgon-tools is an open-source software package providing building blocks for the analysis and modeling of quasi-one-dimensional (quasi-1D) periodic systems based on line-group theory. While mature libraries exist for space-group detection in three-dimensional crystals, an automated and structure-based identification of line groups has so far been lacking. We present software that integrates four complementary components within a consistent line-group framework: (i) structure generation, (ii) symmetry detection, (iii) irreducible representations (irreps) and character tables and (iv) harmonic interatomic force constants (IFCs) correction. This paper introduces the general code structure and several examples that illustrate some relevant applications of the program.

Evolutionary algorithms provide a powerful route to exploring the structural complexity of molecules and materials, enabling efficient exploration of the system's energy landscape to identify the minima related to stable and metastable configurations. Yet, existing frameworks often trade flexibility for chemical fidelity or scalability, leading to unphysical structures or overly restricted search spaces. The here presented framework {is designed to address several of these limitations} through a modular evolutionary architecture that preserves chemical consistency while enabling scalable exploration. combines an interchangeable evolutionary pipeline with chemically grounded molecular and crystalline encodings, {ensuring physically meaningful sampling across different atomistic problem classes}. A hierarchical Supercell Escalation (HiSE) protocol propagates low-energy motifs from minimal cells to larger supercells, improving sampling efficiency in extended systems. The execution model integrates synchronous parallelism for physical evaluations with an asynchronous multi-agent island strategy, while a hybrid SQL–HDF5 archive ensures efficient, reproducible, and fault-tolerant data management. Across benchmarks spanning peptide conformations, Lennard–Jones nanoclusters, binary-oxide convex-hull reconstruction, and CuO/Cu2O surface phase diagrams, {recovers the targeted low-energy basins and thermodynamic envelopes across the tested benchmarks}, resolves complex phase boundaries, and d{shows low orchestration overhead and efficient strong-scaling behavior for the tested benchmark configuration}. High-level YAML workflows enable autonomous discovery campaigns {with reduced manual intervention in the tested workflows, positioning as a flexible framework for scalable, chemically consistent atomistic exploration.}

GHWC is a GPU-accelerated gyrofluid code for simulating quasi-two-dimensional turbulence with consistent finite Larmor radius (FLR) effects in magnetized plasmas. The simulation setup allows for fundamental studies of FLR effects on isothermal resistive drift waves, turbulence, and zonal flows. It includes the standard Hasegawa-Wakatani model in the limit of cold ions. It is a GPU-accelerated update of GHW, rewritten in CUDA. This version has already been used in a recent publication on zonal-flow merging.