Block2: A comprehensive open source framework to develop and apply state-of-the-art DMRG algorithms in electronic...

Area of Science:

• Computational Physics
• Quantum Chemistry
• Materials Science

Background:

• Density Matrix Renormalization Group (DMRG) and Matrix Product States (MPS) are powerful numerical methods for strongly correlated quantum systems.
• Implementing these algorithms efficiently, especially for ab initio electronic structure Hamiltonians, presents significant computational challenges.

Purpose of the Study:

• To introduce block2, an open-source framework designed for implementing and performing DMRG and MPS algorithms.
• To provide a flexible, extensible, and efficient computational tool for researchers in quantum many-body physics and quantum chemistry.
• To showcase the framework's capabilities through numerical examples across various applications.

Main Methods:

• The block2 framework is built upon core DMRG and MPS algorithms.
• It incorporates specialized optimizations for ab initio electronic structure Hamiltonians.
• The framework supports various algorithmic extensions, including dynamical correlation theories and quantum chemistry methods.

Main Results:

• block2 provides out-of-the-box support for key algorithms: eigenstate, time-dependent, response, and finite-temperature calculations.
• The framework demonstrates high efficiency and flexibility, enabling integration with external numerical packages.
• Numerical examples illustrate successful applications in diverse scientific domains.

Conclusions:

• block2 offers a robust and versatile open-source platform for advanced quantum many-body simulations.
• Its design facilitates research in strongly correlated systems, quantum chemistry, and materials science.
• The framework empowers researchers with efficient tools for tackling complex quantum problems.