Bootstrap embedding: An internally consistent fragment-based method.

Area of Science:

• Computational chemistry
• Electronic structure theory
• Quantum mechanics

Background:

• Strong correlation in large molecular systems presents a significant challenge for electronic structure theory due to rapidly increasing computational costs.
• Fragment embedding methods offer a promising approach by dividing systems into smaller, manageable fragments.
• Existing fragment embedding methods often suffer from slow convergence with fragment size, attributed to boundary errors between fragments and the surrounding environment (bath).

Purpose of the Study:

• To introduce and evaluate a novel electronic embedding method, Bootstrap Embedding, designed to overcome the limitations of existing fragment embedding techniques.
• To improve the description of fragment edges in embedding theories through the use of overlapping fragments.
• To demonstrate the method's efficacy in accurately calculating energies and populations for complex systems.

Main Methods:

• Development of Bootstrap Embedding, a self-consistent wavefunction-in-wavefunction embedding theory.
• Utilizing overlapping fragments to enhance the description of fragment boundaries.
• Application of the method to the one-dimensional Hubbard model and a translationally asymmetric variant.

Main Results:

• Bootstrap Embedding demonstrates rapid convergence with respect to the embedded fragment size.
• The method effectively overcomes the surface-area-to-volume-ratio error common in many embedding approaches.
• Accurate calculations of energies and populations were achieved for the tested models.

Conclusions:

• Bootstrap Embedding presents a significant advancement in electronic embedding methods.
• The rapid convergence and accuracy suggest its potential for treating electron correlation in large molecular systems with high fidelity.
• This method offers a pathway towards low-scaling, high-accuracy computational treatments in quantum chemistry.