Back to search

FRINATEK-Fri mat.,naturv.,tek

Relativistic Multiresolution Chemistry: Heavy-Element Compounds at the Complete Basis-set Limit

Alternative title: Relativistisk multiresolusjon kjemi: forbindelser med tunge elementer til basis-set grense.

Awarded: NOK 12.0 mill.

Chemical compounds that make up everything we know are all build from the approximately hundred elements in the periodic table. It is fascinating to think that the variety of materials, objects, life forms, from the most minute crystals to the largest planets are all made of the same building blocks. At the heart quantum chemistry is the study of how such elements interact with each other to form compounds and to predict the compounds' behavior, aided by computer simulations. All atoms in the periodic table are constituted by electrons moving around a nucleus. As the nuclear charge becomes larger, the electrons around it need to move faster to escape the nuclear attraction. For the heaviest of elements, such speed becomes comparable to the speed of light and Einstein's theory of relativity must be taken into account. Heavy elements are extremely important: their properties, originating also from relativity, are steadily being exploited in technological applications: efficient solar panels, high capacity batteries, LED lights, solid state drives all benefit from such properties. ReMRChem (Relativistic MultiResolution Chemistry) will make use of Multiwavelets, a novel mathematical concept, to perform precise calculations of energy and properties for heavy-element compounds. Multiwavelets are special functions, which allow to keep the error of a calculation under rigorous control, but likewise provide a simple framework to translate the theory to a program running on a supercomputer. A simple development framework and rigorous error control are crucial features to model heavy elements including relativity: the theory is more complicated, and traditional methods show limitations in terms of precison which can be achieved. With ReMRChem, it will be possible to generate highly precise results for energy and properties for heavy-element compounds including relativistic effects. This will in turn impact development in material science, and the energy sector.

Heavy elements are essential: they catalyze many chemical processes, they enhance the properties of materials, they improve the yield of solar panels, the capacity of batteries, and the quality of Organic Light-Emitting Diodes. It is therefore important to understand the structure and properties of compounds containing heavy elements. Computer modeling of the electronic structure is able to provide insight that cannot be obtained from experiments alone. A cornerstone of such simulations is the representation of the functions describing the electrons: the so-called orbitals. The two mainstream approaches are Atomic Orbitals (AOs), which represent the structure of isolated atoms, and Plane Waves (PWs), which are ideal for perfect crystals. Heavy-element compounds are, however, very challenging, exposing the limitations and weaknesses of these two approaches. The ReMRChem project will transform the field of electronic structure calculations of heavy-element compounds by developing a novel approach based on Multiwavelets (MWs), instead of AOs or PWs. MWs are a kind of Wavelets: the functions for which Yves Meyer received the Abel Prize in 2017. Our ambitious goal will be achieved by implementing: (1) the full spectrum of relativistic Hamiltonians (1-, 2- and 4-components) for Density Functional Theory. (2) the open-ended response formalism for arbitrary-order properties using MWs. Combining these two developments together, within a MW framework, it will for the first time be possible to generate relativistic results with unprecedented precision at the Density Functional Theory level, for energies and properties of any order. ReMRChem will make our MW code for Quantum Chemistry a unique tool for the simulation of heavy-element compounds: a significant step forward in the state of the art.


FRINATEK-Fri mat.,naturv.,tek