Skip to main content
Ir a la página de inicio de la Comisión Europea (se abrirá en una nueva ventana)
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS
Contenido archivado el 2024-06-18

Optical properties of hybrid organic/inorganic nano-particles for photovoltaic applications: toward a predictive computational approach

Final Report Summary - HY-SUNLIGHT (Optical properties of hybrid organic / inorganic nanoparticles for photovoltaic applications: toward a predictive computational approach)

Photovoltaic (PV) sets itself as a key technology to fight climate change and enhance supply of electricity, particularly where grid access is not permitted/too costly. Spreading of present PV based on semiconductor solar cells, however, is hindered by their high cost and high weight. Hopes for a new PV generation are currently based on organic materials for their light weight and broad absorption spectrum, with a potential for disruptive pricing and new products, such as building integrated PVs.

In this respect, fascinating perspectives are provided by exploiting hybrid organic / inorganic nanoparticles as the light-absorbing active layer. Typical hybrid PV (HPVs) are composite films of semiconductor nanoparticles (also called quantum dots) coupled to organic cromophores or polymers active in the visible range. Simulating the optical properties of these systems by present computational tools, however, is an extremely challenging task. On the one hand, this is due to the need to treat on a pure quantum level the very large molecular systems involved. On the other hand, the very different nature of light-matter interactions in the different segments of the HPV material, makes the optical response of HPVs not just the sum the constituents' responses, and much more difficult to describe theoretically.

This project moved decisive steps toward the development of the theoretical background and implementation of computational methodologies for the quantitative description of the electronic and optical properties of hybrid systems with PV applications. We focused our attention on ab-initio modelling, i.e. atom-by-atom fully quantum description of dye-sensitised solar cells (DSSCs), which are being intensively investigated due to their competitive efficiency transforming solar energy into electricity. These are multi-scale systems with three components: an organic or inorganic dye-molecule that defines the optically active medium, whose optical absorption should overlap as much as possible to the solar radiation spectrum, a semiconductor nanoparticle, often made from titanium dioxide, where photo-excited electrons should rapidly be injected and transported to the anode, and the electrolyte solution containing a redox couple to reduce efficiently the oxidised chromophore.

In a step-by-step strategy toward this goal, we have first developed the computational tools to investigate the optical properties of isolated dye molecules by accurate ab initio methods. Then, we have set the theoretical / computational background to model the hybrid system, that is, the dye molecule close to the nanoparticle and embedded in a solvent (electrolyte) solution, and demonstrated the new scheme with a prototype system.

Of the many possible molecular compounds for PVs, the best performance are typically shown by ruthenium complexes. We successfully completed the first ever ab initio calculation of the optical absorption of a ruthenium dyes (a molecule known as black dye) in gas phase, employing a method known as configuration interaction, which allows to take into account accurately the instantaneous interactions between the many electrons of the molecule. Previous studies were limited to so-called mean-field methods, where electrons are treated as an average electrostatic field, an approximation which may typically lead to strong overestimate, e.g. of the optical gap, the threshold frequency for light absorption. For this goal to be achieved, it has been necessary to first compare several ab initio quantum-chemistry software, and assess their performance in terms of memory requirements and scalability on massive parallel architectures. Final calculations have been performed exploiting the High Performance Computing facility at CINECA (see http://www.cineca.it(se abrirá en una nueva ventana) online for further details), an HPC-Europe Tier-0 node, and required several tens of thousands of CPU hours to be completed. Our results allowed not only to demonstrate that the optical absorption of ruthenium complexes can be predicted on a fully quantum basis, but also to establish which of the many available approximate (and therefore computationally far less intensive) methods could be used, and with which confidence. This study will therefore constitute a benchmarking calculation for future works in this field.

The next step toward modelling optical properties of DSSCs has been to include the effect of a dielectric nanoparticle and of the solvent. This requires to take into account that electrons in the molecular dye interact differently from the well-known Coulomb law if they are immersed in a polarisable medium and / or are in the presence of a dielectric nanoparticle. This can be described by a so-called Green's function appropriate to the problem at hand, which has been implemented within the integral equation formalism method of the 'polarisable continuum model'. Here, the molecular system is embedded in an imaginary surface with an appropriate surface charge. The necessary numerical algorithms have been coded in a set of highly optimised Fortran 90 modules and implemented into GAMESS, one of the most popular and efficient software for quantum chemistry calculations.

We have applied this new methodology to study the optical and redox properties of the L0 organic sensitiser close to a titanium dioxide nanoparticle in a acetonitrile solution, the typical solvent used in DSSC. Our method allowed to calculate by ab initio methods the so-called oxidation potentials, the driving forces to separate the photo-generated charges and for dye regeneration, the key parameters for solar cells efficiency, as a function the distance of the molecular system to the surface of the nanoparticle and for different relative orientations of the dye and the nanoparticle.

Predicting these types of dependencies with quantitative accuracy are clearly crucial to design new HPVs, avoiding a costly trial-and-error process. The implemented model is general and can be applied to organic or inorganic dye-molecules, to simulate the spectral and electrochemical properties of such systems attached or closed to continuum mediums. This may clearly have a large impact in driving the choice for new molecular systems for HPV applications, but it may also find important use in nanomedicine and biology, where the optical properties of hybrid organic / inorganic nanoparticles may find ground-breaking applications.
Mi folleto 0 0