Advanced thermoelectrics at nanoscale: from materials to devices
CECAM-FR-IDF, FIAP Jean Monnet, Paris, France
The key objective of this interdisciplinary workshop is to identify challenges in modeling and simulation towards a fundamental understanding of charge and heat transport phenomena in advanced thermoelectric materials, devices, and circuits. Even though the great relevance of this research field, both for basic research and technological applications, an international workshop on these topics is still lacking. By bringing the most recent viewpoints together from different research communities (physicists, chemists, and engineers) the role of multiscale modeling across different scales (from atoms to circuits) in uncovering complex thermoelectric phenomena will be directly addressed and exploited.
Leading scientists in this field, both theoreticians and experimentalists, from a wide range of countries (as many as possible), will identify and cover the achieved milestones and outline research efforts in determining the essential science of transport mechanisms. Specific topics and issues that will be carefully considered are:
i) How much physics do we lose within the Single Particle Scattering Approach?
Many theoretical calculations of electronic and thermal transport as well as thermo-power phenomenon in nanoscale materials and devices are based on the approach developed by Landauer often coupled with ground state first-principles methods like Density Functional Theory. Though computationally and conceptually simple, the validity of this approach under particular experimental conditions and for nanoscale junctions is still a hot topic in debate. “Closed system”, “ideal leads” and “ground state” assumptions seems not enough to get deep insight into out of equilibrium and non-linear processes like thermoelectric energy conversion in nanostructures.
ii) How can we include all phonon and electron scattering mechanisms?
Even if most of the studies are based on mean field approximations, in order to get the physics of thermo-power phenomena in nanostructures and devices, all the scattering mechanisms between electrons and phonons should be considered. Advanced quantum mechanical approaches, like NEGF, Time Dependent DFT or open quantum systems theories, though computationally demanding, can take into account all these many body effects. However, due to the large number of scattering events per unit of volume and unit of time, it seems that the inclusion of all the relevant particle interactions in thermoelectric transport phenomena is still a hard task both from the theoretical and programming point of view.
iii) Which is the best way to couple first principles calculations to larger scale simulations?
Multiscale modeling approaches based on hierarchies of overlapping scales offer the possibility to calculate macroscopic properties from an accurate quantum mechanical perspective. In particular the advent of efficient and accurate quantum mechanical methods (i.e. DFT), the development of new empirical and semi-empirical potentials and the enormous growth of computing power make this possibility concrete and realistic. However, some limitations represent a general barrier to the progress in this field such as i) the current theoretical huge gap between atomistic methods, empirical descriptions and macroscopic models, ii) the absence of an integration of libraries and codes which could span all the scales, from the quantum mechanical to the continuous one, and which include visualization and analysis tools, iii) the lack of a forum of discussion about computational methods, techniques, as well as for novel philosophies of scale coupling and their implementation.
Michele Amato (Université Paris Sud) - Organiser
Philippe Dollfus ( Institut dElectronique Fondamentale - CNRS) - Organiser
Stefano Ossicini (University of Modena and Reggio Emilia, Modena) - Organiser
Riccardo Rurali (ICMAB-CSIC) - Organiser
Gyeong S Hwang (Department of Chemical Engineering, University of Texas, Austin) - Organiser