Periodic Density Matrix Embedding for CO Adsorption on the MgO(001) Surface

Area of Science:

• Computational Chemistry
• Materials Science
• Surface Science

Background:

• Heterogeneous catalysis relies on gas molecule adsorption to metal oxide surfaces.
• Accurate quantum chemical modeling is crucial but computationally expensive.
• Quantum-embedding methods offer a balance of accuracy and efficiency.

Purpose of the Study:

• To apply periodic density matrix embedding theory (DMET) for calculating CO adsorption energy on MgO(001).
• To assess the accuracy of DMET compared to traditional quantum chemical methods.
• To develop a memory-efficient procedure for periodic DMET calculations.

Main Methods:

• Periodic density matrix embedding theory (DMET).
• Coupled-cluster theory with single and double excitations (CCSD).
• Second-order Møller-Plesset perturbation theory (MP2).
• Embedding cluster sizes up to 266 electrons and 306 orbitals.

Main Results:

• DMET calculations achieved agreement within 1.2 kcal/mol of non-embedding references for CO adsorption energy.
• Demonstrated the efficacy of DMET for localized chemical phenomena on surfaces.
• Introduced a memory-efficient method for handling electron repulsion integrals in periodic DMET.

Conclusions:

• Periodic DMET is a viable and accurate method for modeling gas adsorption on metal oxide surfaces.
• The developed memory-efficient procedure enhances the practicality of DMET for large systems.
• This work advances computational approaches for heterogeneous catalysis research.