A novel free software allows improved calculation of transport properties in materials with potential for green technologies

by Virginia Greco

Elphbolt is the first software for simulating charge and thermal transport in solid materials that allows integrating the reciprocal effect of electrons and phonons into the calculations. This new free tool will enable detailed studies of transport phenomena in solids, with probable impact on the search for thermopower materials for green energy applications.

Achieving a deep understanding of the transport properties of solid materials is crucial, not only in itself for comprehending the underlying physics, but also for designing and forging materials with specific properties for different applications. Advanced computer simulation programmes are extremely valuable tools, which allow researchers to perform complex computations of real systems and their evolution over time starting from theory –i.e. equations.

Charge and heat transport in materials are typically determined by the interactions of electrons and phonons (pseudo-particles associated with the lattice vibrations) among themselves and each other. The dynamics of these phenomena can be calculated –using the Boltzamann transport equations. The standard approach solves the equation for the phononic and the electronic systems separately, assuming that the other remains in equilibrium. However, in 1930 Rudolf Peierls conceived the idea that the dynamics of charge (electrons) and vibrations (phonons) in materials are actually interconnected and affect their transport properties, such as thermopower, thermal conductivity, and charge conductivity. Even though many theoretical calculations have been done to explain this mutual effect and experiments have demonstrated it, no free code has been available so far that could capture it in a fully self-consistent manner.

Researchers from the ICN2 Theory and Simulation Group, led by Prof. Pablo Ordejón, and from the Department of Physics of the Boston College (US) developed a software, called elphbolt (from electron-phonon Boltzmann transport), that enables incorporating in simulations the reciprocal effect of electron and phonon interactions on the transport properties of materials. The theory on which elphbolt is based and its implementation are thoroughly explained in a paper published in Computational Materials (a journal of the Nature Partner Journals series), signed by Dr Nakib Protik as first author and corresponding author, together with Prof. Ordejón. Dr Miguel Pruneda, CSIC scientist in the same group, and Dr Chunhua Li and Prof. David Broido, from Boston College, also participated in this study.

By introducing the effect of non-equilibrium electrons on phonons (which is called electron drag) and vice versa (phonon drag), the elphbolt software finally allows achieving a fully self-consistent solution to this long-standing and challenging transport problem. In addition, it uses an ab initio –i.e., from first principles— computational method, which means that the only input into a calculation are physical constants, no tunable parameters are required. The code is provided for free (under a GNU General Public License), is compact, well-documented and extensible.

“This work opens a new avenue for the detailed investigation of the rich, underlying transport physics in solid materials, and promises to be useful in the search for materials for green technologies,” explains Dr Nakib Protik. Indeed, this software will facilitate the study of heat transport in solids and, thus, the search for high thermopower materials, among others. These materials are particularly interesting because –applying conveniently the thermoelectric effect, by which a temperature difference creates an electric potential and the other way around– they can be used to convert waste heat to useful electricity.

Reference article:

Nakib H. Protik, Chunhua Li, Miguel Pruneda, David Broido & Pablo Ordejón, The elphbolt ab initio solver for the coupled electron-phonon Boltzmann transport equations. npj Comput Mater 8, 28 (2022). DOI: 10.1038/s41524-022-00710-0