SIESTA

SIESTA
Name	SIESTA
Developer	Universidad Autónoma de Madrid; collaborators include Instituto de Ciencia de Materiales de Madrid, CIC nanoGUNE, University of Cambridge, Oak Ridge National Laboratory
Released	1996
Latest release	4.1 (example)
Programming language	Fortran
Operating system	Linux, Unix, macOS
Genre	Ab initio electronic structure software
License	GPL-compatible; commercial options

SIESTA

SIESTA is an ab initio electronic structure code for materials modeling that emphasizes efficiency and scalability for large systems. It uses pseudopotentials and a linear combination of numerical atomic orbitals to compute properties of solids, molecules, surfaces, and nanostructures. The code has been developed and applied by researchers across institutions such as Universidad Autónoma de Madrid, University of Cambridge, Oak Ridge National Laboratory, Lawrence Berkeley National Laboratory, and Max Planck Society.

Overview

SIESTA implements density functional theory techniques to predict electronic, structural, and response properties of matter, integrating approaches from Perdew–Burke–Ernzerhof exchange–correlation functionals, norm-conserving pseudopotentials inspired by Troullier–Martins, and basis-set ideas related to LCAO methods. The software targets applications spanning condensed matter investigations relevant to CERN experiments, NASA materials research, and industrial projects at organizations like Siemens and IBM Research. Its design facilitates studies of defects, interfaces, and large supercells that are computationally demanding for plane-wave codes used at facilities such as Argonne National Laboratory and Oak Ridge National Laboratory.

History and Development

Development began in the mid-1990s at groups including Universidad Autónoma de Madrid and collaborators such as Instituto de Ciencia de Materiales de Madrid and researchers formerly associated with Cavendish Laboratory, influenced by earlier electronic-structure efforts at Bell Labs and algorithmic advances from Los Alamos National Laboratory. Key milestones include adoption of norm-conserving pseudopotentials similar to work by N. Troullier and J. L. Martins, incorporation of linear-scaling algorithms with conceptual roots traceable to methods used at IBM Research and algorithmic ideas appearing in literature from University of California, Berkeley. Over time, contributors from institutions such as University of Cambridge, Max Planck Institute for Solid State Research, and EPFL expanded capabilities for transport, molecular dynamics, and hybrid-functional treatments.

Methodology and Features

SIESTA uses self-consistent field implementations of density functional approximations like Perdew–Burke–Ernzerhof and local density approximations related to work by John P. Perdew and collaborators. The primary basis is a finite-support numerical atomic orbital set enabling sparse Hamiltonians and efficient real-space integrals, in the spirit of methodologies developed at Oak Ridge National Laboratory and Lawrence Livermore National Laboratory. Core features include norm-conserving pseudopotentials inspired by Troullier–Martins, real-space grid for charge density integration akin to approaches used at Argonne National Laboratory, support for Monkhorst–Pack k-point sampling related to schemes by H. J. Monkhorst and James D. Pack, linear-scaling (O(N)) options leveraging ideas from Martin Head-Gordon-type locality, and modules for non-equilibrium Green’s functions influenced by formalisms used at Stanford University and University of Twente. Additional capabilities parallel developments at Max Planck Society and ETH Zurich for phonons, Born–Oppenheimer molecular dynamics, and transition-state searches.

Applications and Use Cases

Researchers have used the code for investigations of two-dimensional materials similar to studies at MIT and Columbia University, organic semiconductors researched at University of Cambridge and ETH Zurich, catalytic surfaces examined by teams at Caltech and Argonne National Laboratory, and defect physics relevant to Sandia National Laboratories projects. SIESTA has been employed in simulations supporting experiments at facilities like European Synchrotron Radiation Facility, Diamond Light Source, and SLAC National Accelerator Laboratory, and in collaborations with industrial partners such as Intel and Samsung for device-related modeling. Case studies include electronic transport through nanowires analogous to work at IBM Research, adsorption on metal surfaces studied in groups at University of California, Los Angeles, and battery-material modeling in projects related to Lawrence Berkeley National Laboratory.

Performance and Validation

Benchmarking efforts compare SIESTA’s performance and accuracy with plane-wave codes used at Princeton University, University of Texas at Austin, and Paul Scherrer Institute, and with all-electron codes developed at University of Oxford and Rutgers University. Validation studies often reference testbeds and cross-code comparisons involving pseudopotential sets similar to those from Hamann, and exchange–correlation benchmarks associated with Perdew and Kohn–Sham formulations. Scalability assessments report strong parallel performance on clusters at Oak Ridge National Laboratory and supercomputers at National Energy Research Scientific Computing Center, with linear-scaling options demonstrating favorable behavior for large systems relevant to projects at Max Planck Institute for Coal Research and Brookhaven National Laboratory.

Licensing and Availability

SIESTA is distributed under licenses that allow academic use and community contributions, comparable to models adopted by projects at Free Software Foundation and collaborations such as Quantum ESPRESSO and GPAW. Binary packages and source code historically have been handled through institutional portals and developer-maintained repositories involving contributors from Universidad Autónoma de Madrid and partner institutes like CIC nanoGUNE and Instituto de Ciencia de Materiales de Madrid. Commercial support and licensed distributions have been available through third-party vendors similar to arrangements seen for codes associated with Synopsys and Accelrys.

Category:Electronic structure software