Oxidation-Budgeted Design of Lead-Free RbGeI3 Solar Cells: Back-Surface-Field Enhancement, Realistic Loss Budgeting and SHAP Sensitivity From SCAPS-1D Optimization to Manufacturing Tolerances, Benchmarked Against CsGeI3

The rapid rise of lead-halide perovskite solar cells has been tempered by lead toxicity, motivating a search for lead-free absorbers with comparable optoelec- tronics. Rubidium germanium iodide (RbGeI3) offers a direct, tunable bandgap, strong absorption, and improved environmental compatibility. Here we report a SCAPS-1D optimization of an RbGeI3 cell organized around its central liability, Ge2+ oxidation. After screening eight ETL–HTL combinations, the FTO/ TiO2/RbGeI3/CuI/Au architecture was carried forward on stability and cost grounds (a D2/NiO cross-check through the same pipeline reaches 28.16%, so this stack is not the efficiency optimum). Eleven parameters were varied se- quentially on one frozen deck; the consolidated device reaches 27.06% PCE. A Cu2O-like back-surface field raises the ideal point to 28.01% and recovers about one point on the identical full-realism stack (20.78% → 21.76%). Mapping VGe vacancy thermodynamics onto the defect ladder converts air exposure into a fab spec (Ef ≳ 0.46 eV holds the headline at 300 K equilibrium, 0.57–0.65 eV for anneal freeze-in; ambient falls below 20.76%), with a 15.53% current-state rung marking what first fabrication must beat. The solver pipeline is calibration- checked against experimental β-CsPbI3 [55], and SHAP analysis recovers bulk defects, bandgap and electron mobility as the dominant levers.