High-throughput thickness gradient screening reveals thickness and light-intensity dependent efficiency in indoor...

Indoor organic photovoltaics (OPVs) are promising power sources for Internet-of-Things devices, but optimizing performance under diverse indoor lighting is challenging because the optimal thickness of the active layer depends on the competition between charge transport and recombination, as well as on the incident light spectrum in a complex manner. Here, we use a high-throughput customized blade-coating system to generate continuous active-layer thickness gradients (50-450 nm) for five binary blends comprising three wide-bandgap donors (PTQ10, PM6, D18) and three non-fullerene acceptors (o-IDFBR, eh-IDTBR, FCC-Cl). Devices were characterized under four LED spectra (2700 K, 5200 K, 6500 K, B4) using a spectrum-on-demand source. Across 600 devices, intermediate thicknesses maximize shunt resistance (R _P) and fill factor, whereas thin and thick layers suffer from leakage and recombination. PTQ10:FCC-Cl shows broad thickness tolerance (≈230-410 nm) and moderate spectral stability, with PCE varying by only ≈2.6% across indoor spectra and reaching a maximum of ≈21.7% under 2700 K. Conversely, PM6:FCC-Cl attains higher PCE (≈26.4%) but is strongly thickness-sensitive, with peak efficiency realized within a narrow range (∼305 nm) and varying by ≈15% across all four spectra. Analysis of intensity- and thickness-dependent charge transport indicates that performance is governed by photon absorption, spectral overlap and insufficient R _P. This work demonstrates a rapid screening method and highlights the importance of thickness- and spectrum-optimized active layers for efficient indoor OPVs.