The new technique makes modeling molecules easier

c2H4 rotational barrier potential energy surfaces obtained from CASSCF(12,12)/ACSE, CCSD(T), PBE-RDMFT, SCAN-RDMFT, PBE-DFT, SCAN-DFT, and CASSCF(12,12)/tPBE calculations with in the cc -pVDZ basis set. Credit: Physical Review Letters (2023). DOI: 10.1103/PhysRevLett.131.243003

Like the humans who created them, computers struggle with physics, but quantum mechanics is even harder. But a new technique developed by three scientists at the University of Chicago allows computers to simulate some challenging quantum mechanical effects in complex electronic materials with little effort.

By making these simulations more accurate and efficient, scientists hope that the technique will help discover new molecules and materials, such as new types of solar cells or quantum computers.

“This development has great potential for advancing our understanding of molecular phenomena, with significant implications for chemistry, material science, and related fields,” said scientist Daniel Gibney, a University of Chicago Ph.D. chemistry student and first author of the paper, published December 14 at Physical Review Letters.

Electrons and energy

A leaf or a solar panel looks neat and simple from the outside, but zoom in on the molecular level and you see a more complex dance of electrons and molecules.

To make new advances in sustainability, manufacturing, agriculture, and other fields, scientists model the behavior of these chemical and molecular interactions. This has helped reveal new design possibilities for the future for everything from new ways to sequester carbon dioxide to new types of quantum bits.

Much progress has been made in the past decades, but one of the areas that remains stubbornly difficult to simulate is when molecules begin to exhibit complex quantum mechanical behaviors that scientists call lig -on correlation.

The problem is that once the electrons start showing their most quantum-mechanical effects like becoming “entangled” the calculations immediately require more computing power. Even supercomputers struggle to handle the implications.

One of the most commonly used calculations is called density functional theory. “This is basically the most ubiquitous technique for predicting electronic structure, but it’s an approximation where all the electrons are considered as a function of one electron,” explained David Mazziotti, professor of chemistry and senior author of the study.

The new technique makes modeling molecules easier

Relative energies of meta- and para-benzyne relative to ortho-benzyne from RDMFT and DFT with SCAN and PBE functionals, MC-PDFT using the tPBE functional, and CCSD(T). Credit: Physical Review Letters (2023). DOI: 10.1103/PhysRevLett.131.243003

For many calculations, an approximation will work. But this begins to break down as the behavior of electrons becomes more relevant, as happens when quantum mechanics comes into play. In quantum mechanics, these electrons can be in many places, or orbitals, simultaneously. It restricts not only the human brain, but also the theory of the density function.

“And this is an important problem, because many issues that we are dealing with in the 21st century such as new molecules and materials for renewable energy and sustainability require us to take advantage of the quantum nature of materials,” Mazziotti said.

Mazziotti, Gibney, and third author Jan-Niklas Boyn found that they could add a universal correction to density functional theory that allowed electrons to be involved in multiple orbitals at once.

“This allows the calculation of orbitals that are not only completely filled or completely empty, but anywhere in between,” Mazziotti said. “We arrived at a picture of an electron that is still able to capture the behavior arising from the correlated many-body effects of the electron.”

A ‘universal’ adaptation

As a bonus, the scientists said, the code can be added to existing algorithms without having to rewrite the code. “Basically, the correction kicks in when it’s needed, but doesn’t interfere with the rest of the code otherwise,” Gibney said.

It is also universal that it can be added to the code simulating many types of electronic behavior, it is photovoltaic solar panels or carbon sequestration or superconducting materials or even biology.

For example, Boyn explained, one application could be to understand the chemistry that occurs using enzymes that contain metal atoms, known as metalloenzymes.

“There are many metalloenzymes that are responsible for a lot of the chemistry in your cells, for example, but they are very difficult to describe with current models,” he said. “This theory may in the future allow us to deal with this chemistry in a way that is impossible now.”

More information:
Daniel Gibney et al, Universal Generalization of Density Functional Theory for Static Correlation, Physical Review Letters (2023). DOI: 10.1103/PhysRevLett.131.243003

Provided by the University of Chicago

Citation: New technique makes modeling molecules easier (2023, December 18) retrieved 24 December 2023 from https://phys.org/news/2023-12-technique-molecules-easier. html

This document is subject to copyright. Except for any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. Content is provided for informational purposes only.


#technique #modeling #molecules #easier
Image Source : phys.org

Leave a Comment