A Big Data Computational Framework for Simulation and Reconstruction of Nanocomposite Hybrid Photovoltaics

Addressing Big Data and Energy issues through research on Hybrid Photovoltaics

Our research aims at extending the X-ray scattering theory to the more recent hybrid organic / quantum dot devices through the development of a flexible Big Data computational framework capable of simulating, reconstructing and visualizing the nanostructures within fully functional hybrid photovoltaic devices.

In collaboration with scientists at the Lawrence Berkeley National Laboratory (ALS) and the Swiss National Supercomputing Centre (CSCS) we do develop a variety of Big Data model systems aimed at studying the morphology of hybrid organic / quantum dot photovoltaic systems. We apply some of the most advanced Big Data software algorithms (HipGISAXS)  to get a deeper understanding of the structure / function relationship within hybrid photovoltaic model systems and devices.

Synchrotron scattering studies have proven to be a successful approach to understand and thus control the structural properties in both organic and hybrid nanocomposite blends, which is crucial for the optimization of photovoltaic device performances. The goal is to understand the interplay of the structure formation mechanisms within the nanocomposite materials forming the active layer of the photovoltaic devices, thus in order to methodically study their structural properties we employing both Grazing Incidence Small and Wide-angle X-ray Scattering (GISAXS and GIWAXS). Analyzing GISAXS and GIWAXS has traditionally been restricted to shapes with analytic solutions and limited distributions of shapes. We aims at developing a flexible Big Data computational framework for GISAXS / GIWAXS simulations, capable of simulating a wide variety of analytical shapes, providing accurate models aimed at expanding the fundamental theoretical background in both photovoltaic and energy fields. Standard computational frameworks can only account for extracting a limited amount of information from X-ray scattering data. Our joint research project with Dr. Alexander Hexemer’s group at ALS and CSCS aims at overcoming this limitation through the development of a sufficiently flexible computational framework capable of interpret and model the structural arrangements within the hybrid photovoltaics.

Synchrotron data are conventionally analyzed within the framework of the distorted-wave Born approximation (DWBA) theory. Due to the large amount of samples and correspondingly large data volumes, the development of automatized data modeling capable of autonomously extracting the majority of information from the X-ray scattering data will be crucial. Further development of the HipGISAXS Big Data architecture will address this problem by implementing artificial intelligence routines. HipGISAXS is a simulation software for modeling GISAXS and GIWAXS data, it was developed from the ground up with the intention and focus of very high performance and flexibility. Currently HipGISAXS is supporting multi CPU and multi GPU clusters and it has been designed to accommodate the most diverse model systems ranging from simple analytical shapes in a variety of arrangements and orientations to the development of completely free forms described by a triangular surface. Furthermore the HipGISAXS Big Data framework is part of a wider effort to develop high performance mathematics and algorithms to address the data deluge currently experienced by synchrotron users. This Big Data framework is part of the Center for Applied Mathematics for energy Research Applications located at Lawrence Berkeley National Laboratory. The Center for Applied Mathematics for Energy Research Applications (CAMERA) is an integrated, cross disciplinary center aimed at inventing, developing, and delivering the fundamental new mathematics required to capitalize on experimental investigations at DOE User Facilities. Our collaboration with ALS and CAMERA sets thus the strategy for the development of new mathematics and algorithms and their use across different platforms.

The purpose of this project is to lay the scientific groundwork for a range of hybrid organic based devices.  Despite the maturity of research of polymer based photovoltaics, the structure-function relationship in these devices is not well known. We thus aim at further extending the fundamental understanding on the structural properties and thermodynamics of the hybrid photovoltaics, which are indispensable to further develop hybrid solar cells into a new concept with low fundamental losses. Organic and related hybrid materials progressed rapidly from exotic niche materials to vital players in the arena of future photovoltaic materials. However, the interplay of the respective components in the photoactive layer is still not understood. Further understanding the interplay between structural and electronic properties will open new ways towards self-assembled nano-morphologies in solar cells.

Altogether, by gaining a fundamental understanding of how the structure formation influences performance in hybrid organic / quantum dot photovoltaics, our research aims at contributing towards scientific knowledge and the rational design of devices, which will pave the way to performance enhancement and improve their life-times.

Adolphe Merkle Institute - Chemin des Verdiers 4 - CH-1700 Fribourg - Phone +41 26 300 9254