Simulation study of a new InGaN p-layer free Schottky based solar cell

Abstract

On the road towards next generation high efficiency solar cells, the ternary Indium Gallium Nitride (InGaN) alloy is a good passenger since it allows to cover the whole solar spectrum through the change in its Indium composition. The choice of the main structure of the InGaN solar cell is however crucial. Obtaining a high efficiency requires to improve the light absorption and the photogenerated carriers collection that depend on the layers parameters, including the Indium composition, p-and n-doping, device geometry. . . Unfortunately, one of the main drawbacks of InGaN is linked to its p-type doping, which is very difficult to realize since it involves complex technological processes that are difficult to master and that highly impact the layer quality.In this paper, the InGaN p-n junction (PN) and p-i-n junction (PIN) based solar cells are numerically studied using the most realistic models, and optimized through mathematically rigorous multivariate optimization approaches. This analysis evidences optimal efficiencies of 17.8% and 19.0% for the PN and PIN structures. It also leads to propose, analyze and optimize p-layer free InGaN Schottky-Based Solar Cells (SBSC): the Schottky structure and a new MIN structure for which the optimal efficiencies are shown to be a little higher than for the conventional structures: respectively 18.2% and 19.8%.The tolerance that is allowed on each parameter for each of the proposed cells has been studied. The new MIN structure is shown to exhibit the widest tolerances on the layers thicknesses and dopings. In addition to its being p-layer free, this is another advantage of the MIN structure since it implies its better reliability. Therefore, these new InGaN SBSC are shown to be alternatives to the conventional structures that allow removing the p-type doping of InGaN while giving photovoltaic (PV) performances at least comparable to the standard multilayers PN or PIN structures.

SLALOM: Open-source, portable, and easy-to-use solar cell optimizer. Application to the design of InGaN solar cells

“…This mathematically secured procedure is largely used in physical engineering areas such as mechanical, electrical, or civil engineering [7][8][9]. We have used this method for the design and the optimization of solar cells [10][11][12] and propose with this paper, to fully detail and completely release it and the software behind it for the free use of all, under the MIT License.…”

“…In this context, we used SLALOM to precisely seek the optimal efficiency that can be achieved by InGaN PN junction solar cells, using as-realistic-as-possible physical models with parameters extracted from experimental data, with the aim to simultaneously optimize the cell 5 following parameters : the Indium composition alongside the P and N layers dopings and thicknesses. These parameters and the used models have been detailed previously [10,11]. Table 1 and equation 2give the values that were taken for the bandgap, the affinity, the density of states, and the absorption.…”

“…Solving this problem requires optimizing the solar cell while taking into account the limits of material technology, mainly related to InGaN elaboration and to solar cell fabrication. This can be done in SLALOM by setting the optimization parameters range to technology limits in slalomDevice.py: the corresponding optimizations and results can be found in previously published literature [11,12].…”

Effects of structural defects and polarization charges in InGaN-based double-junction solar cell

“…For relatively high-Indiumcontent structures, Doolittle et al [5] theoretically showed that a p-i-n structure having an Indium composition of 60% exhibits * Corresponding author Email addresses: abdoulwahab.adaine@univ-lorraine.fr (Abdoulwahab Adaine), sidi.hamady@univ-lorraine.fr (Sidi Ould Saad Hamady), nicolas@fressengeas.net (Nicolas Fressengeas) a maximum efficiency of about 19%. Our previous studies [6,7] have shown that single junction structures have conversion efficiencies not exceeding 20%. The p-n structure presents an optimal efficiency of 17.8% while that of a p-i-n structure is 19.0%.…”

“…When the direct experimental data are not available, the InGaN parameters values are determined using Vegard's law linear interpolation between InN and GaN binaries values, except for the bandgap (E g ) and the electronic affinity (χ), where we used a quadratic modified Vegard's law with a bowing parameter of b = 1.43 eV for the bandgap and b = 0.8 eV for the electronic affinity [18]. All these physical models and dependent material parameters are detailed in our previous work [6,7] and summarized in table 1.…”