The Indium Gallium Nitride (InGaN) III-Nitride ternary alloy has the potentiality to allow achieving high efficiency solar cells through the tuning of its band gap by changing the Indium composition. It also counts among its advantages a relatively low effective mass, high carriers' mobility, a high absorption coefficient along with good radiation tolerance. However, the main drawback of InGaN is linked to its p-type doping, which is difficult to grow in good quality and on which ohmic contacts are difficult to realize. The Schottky solar cell is a good alternative to avoid the p-type doping of InGaN. In this report, a comprehensive numerical simulation, using mathematically rigorous optimization approach based on state-of-the-art optimization algorithms, is used to find the optimum geometrical and physical parameters that yield the best efficiency of a Schottky solar cell within the achievable device fabrication range. A 18.2% efficiency is predicted for this new InGaN solar cell design.

L'alliage de Nitrure de Gallium et d'Indium (InGaN) est un matériau semi-conducteur qui présente aujourd’hui un grand intérêt pour la réalisation de cellules solaires à très haut rendement grâce à sa large et modulable bande interdite en fonction de la composition d’Indium. Néanmoins, la croissance de couches épitaxiales d’InGaN, le dopage et le contact ohmique au niveau de la couche “P” reste un challenge. La cellule solaire Schottky est une bonne alternative pour éviter la couche “P”. Nous avons simulé et optimisé, avec des méthodes mathématiques rigoureuses, une cellule solaire Schottky afin de trouver les paramètres géométriques et physiques optimaux donnant ainsi le rendement optimal de la cellule solaire. Nous avons obtenu un rendement optimal de 18.2%. Ce dernier montre le mérite d’une telle structure qui pourrait être une alternative aux structure PN et PIN afin d’éviter les difficultés liées à la couche P de la cellule solaire.

The synthesis of new efficient low cost solar cells made with earth abundant materials requires on the one hand the development of such materials and, on the other hand, an adequate design of the solar cell that optimizes their use. This latter part is all the more challenging as the materials can be tuned in a variety of ways, adjusting for instance their doping, composition, bandgap and thickness. The high number of parameters that can be adjusted implies that finding their optimal combination is a task in itself, which is presently addressed by a parametic optimisation: optimising one parameter at a time with respect to the cell efficiency, the others being kept constant. Despite its apparent simplicity, this method is time consuming and is not proven mathematically to allow to find the optimal parameter set. Mathematically proven methods can now be easily used in the field of photovoltaics, which allow a simultaneous optimisation of an increased number of parameters, implying a susbtantial decrease in the computation time, along with an increase in the precision of the prediction. A review of present methods and how we have improved them for speed, correctness and precision will be provided.

The development of new thin films solar cells is highly accelerated by using rigorous optimization approach. A large set of parameters is involved in the solar cell operation: active layer thickness, composition, bandgap, doping, contacts, etc. The study of the solar cell performances with respect to these linked parameters is necessary to better understand the underlying physics and to optimize the final device. We propose in this poster presentation (*) a new open-source solar cell optimizer: SLALOM for SoLAr ceLl multivariate OptiMizer. SLALOM implements, for the first time for solar cells, a rigorous multivariate approach while the standard optimization work used to use the one-by-one parameter procedure. SLALOM is implemented to be easily extended to any simulator, the core code itself does not depend on any particular engine. It can be adapted for any solar cell structure, runs locally or remotely on a calculation server, using the SSH protocol and includes a graphical user interface to real-time monitor the optimization. Two study cases, InGaN and CZTS solar cells, are presented to show the SLALOM ability to become a useful tool for the development of novel solar cells.

S

The design and optimization of novel structures is an essential part of the next-generation solar cells development. Indeed, the technological steps involved in the development of high-performance solar cells involve a huge set of interdependent physical and geometrical parameters: layers thicknesses, dopings, compositions, and defect characteristics. In this work, we propose a new open-source and free solar cell optimizer: SLALOM À for SoLAr ceLl multivariate OptiMizer À that implements a rigorous multivariate approach, which improves from the one-parameter-at-a-time procedure that is traditionally used in the field to a state-of-the-art multivariate approach. Applied to indium gallium nitride (InGaN) solar cells, it shows its potential to become a useful tool for the development of novel solar cells. SLALOM is implemented to be extended to any semiconductor simulation engine. Several models for solar cells have been implemented in SLALOM, including, for instance, InGaN. One can adapt these models to any solar cell technology by changing the parameter set, the here proposed generic code structure remaining unchanged.