Numerical simulation of high-efficiency silicon solar cells with different anti-reflective coatings

Crystalline silicon solar cells remain the most widely adopted photovoltaic technology; however, their performance is constrained by optical reflection losses and non-optimized device parameters such as layer thickness, doping concentration, and anti-reflective coatings. This study aims to numerically optimize the performance of high-efficiency crystalline silicon solar cells by investigating the effects of absorber and window layer thicknesses, doping carrier concentrations, and single-layer anti-reflective coating (ARC) materials. Numerical simulations were carried out using the Personal Computer One-Dimensional (PC1D) simulation tool, employing a p-type silicon absorber and an n-type silicon window layer, where the absorber thickness was varied from 60 μm to 240 μm, the window layer thickness from 0.5 μm to 5.0 μm, the absorber doping concentration from 1×1014 to 1×1019 cm-3, and the window doping concentration from 1×1016 to 1×1020 cm-3. In addition, single-layer ARC materials including silicon nitride (Si3N4), titanium dioxide (TiO2), zinc oxide (ZnO), silicon dioxide (SiO2), and magnesium fluoride (MgF2) were analysed with thicknesses ranging from 50 nm to 140 nm. The results show that under ideal anti-reflective conditions (zero reflectance), a maximum power conversion efficiency (PCE) of 25.35% was achieved, while a baseline efficiency of 16.85% was obtained without any ARC. Among the ARC materials studied, ZnO with an optimized thickness of 80 nm produced the highest efficiency of 23.41% at an absorber thickness of 240 μm and a window layer thickness of 0.5 μm, yielding a maximum open-circuit voltage (Voc) of 0.7457 V and a short-circuit current (Isc) of 4.054 A. These findings highlight that systematic optimization of structural parameters and ARC selection, particularly the use of ZnO, plays a critical role in minimizing optical losses and enhancing the overall efficiency of crystalline silicon solar cells.