Un microscopio computazionale per lo studio di materiali innovativi per il fotovoltaico del futuro

The importance of using photovoltaics (PVs) to obtain clean and reliable energy hasbeen widely recognized. The scientific research on new PV materials is crucial to identifynew approaches to the concept of solar energy conversion and to increase the photonversionefficiency. Among the different approaches for PV, the quantum dot (QD) solarcell is one of the most promising. A single layer QD consists of a distribution ofnanocrystals embedded into an amorphous dielectric matrix. It is in principle possibleto affect the optoelectronic properties of QD layers by controlling the morphology ofthe nanograins. The complex atomic scale structure of such materials makes it quitedifficult to predict theoretically their optoelectronic properties. The theoretical challengeis two-fold; on the one hand, it is necessary to generate large scale atomistic models;on the other hand, the corresponding electronic properties must be computed over alarge number of atoms. While the first issue can be coped with large-scale moleculardynamics (MD) simulations, the second issue is a major computational bottleneck andit requires a class of advanced algorithms for the electronic structure calculations inwhich the computational workload scales linearly with the number of atoms. Within acollaboration among CASPUR and the SLACS (Sardinian Laboratory for Computationalmaterials science) we have implemented a new numerical method to investigate theoptoelectronic properties of silicon based QD layers. By using such a theoretical frameworkit is possible to assist the design of new efficient materials for PV