The CRYSTAL code, 1976-2020 and beyond, a long story

Area of Science:

• Computational Physics
• Quantum Chemistry
• Materials Science

Background:

• Periodic ab initio calculations are crucial for understanding crystalline materials.
• Accurate electronic structure methods are needed for predicting material properties.
• Efficient and scalable computational codes are essential for modern research.

Purpose of the Study:

• To introduce CRYSTAL, a periodic ab initio code for electronic structure calculations.
• To highlight its capabilities in handling diverse systems and properties.
• To emphasize its parallel implementations for high-performance computing.

Main Methods:

• Utilizes Gaussian-type basis sets for crystalline orbitals (Bloch functions).
• Supports all-electron and pseudopotential calculations with various density functionals, including Hartree-Fock.
• Fully exploits symmetry and offers tools for structural modification and property evaluation.

Main Results:

• CRYSTAL treats 3D, 2D, 1D, and 0D systems uniformly.
• Enables calculation of numerous tensorial properties and vibrational spectra (IR, Raman) with analytical intensities.
• Provides serial, parallel, and massive-parallel versions for efficient execution on HPC clusters.

Conclusions:

• CRYSTAL is a powerful, flexible, and scalable code for electronic structure studies.
• Its comprehensive features facilitate the investigation of complex materials and properties.
• Efficient parallelization ensures its applicability on large-scale computing architectures.