LAUSR

DOI: 10.1002/aenm.201802790 their optical and electrical properties. A common optical-based approach to estimate bandgaps of direct-gap semiconductors is to employ absorption spectroscopy. In this approach, one captures absorption spectra of materials and obtains absorption thresholds from Tauc plots.[3] The Tauc analysis provides a consistent way to compare among materials. However, as mentioned by Green et al.,[4] these thresholds represent optical bandgaps, not electronic bandgaps, as the Tauc analysis underestimates the presence of excitonic transitions and broadening factors such as temperatures, defect densities, and material disorders.[4,5] Alternatively, one can capture the so-called band-to-band luminescence spectra from materials, which are results of radiative recombination between free electrons and holes in the conduction and valence bands, respectively. The peak wavelengths (or energies) of the luminescence spectra have been reported to be close to the absorption thresholds at room temperature by numerous authors[6–10] (also see Figure S1, Supporting Information). Thus, the two optical methods can be used to estimate the material optical bandgap although they often underestimate the true electronic bandgap. However, both of these approaches measure the spatially average optical bandgap over a certain area. Therefore, capturing spatially resolved optical bandgaps of the entire sample area requires point-by-point mapping in both X and Y directions, and thus it is time consuming and impractical for large-area device imaging or high-throughput production-line environments. In PV research and manufacturing, luminescence imaging, including both electroluminescence (EL) and photoluminescence (PL), is a mainstream characterization tool for crystalline silicon solar cells.[11–17] This camera-based method can provide fast diagnostics of PV materials and devices with micrometerscale spatial resolution. Recently, luminescence imaging has increased in significance for perovskite device and material characterization. Many works have extracted various device parameters of PSCs such as correlations between luminescence intensities and open-circuit voltages,[18–20] spatial distributions of nonradiative recombination centers,[21,22] series resistances from EL images,[23] and lateral inhomogeneities on each subcell in monolithic perovskite-silicon tandem structures.[24] It is also a powerful tool for monitoring long-term performance and A fast, nondestructive, camera-based method to capture optical bandgap images of perovskite solar cells (PSCs) with micrometer-scale spatial resolution is developed. This imaging technique utilizes well-defined and relatively symmetrical band-to-band luminescence spectra emitted from perovskite materials, whose spectral peak locations coincide with absorption thresholds and thus represent their optical bandgaps. The technique is employed to capture relative variations in optical bandgaps across various PSCs, and to resolve optical bandgap inhomogeneity within the same device due to material degradation and impurities. Degradation and impurities are found to both cause optical bandgap shifts inside the materials. The results are confirmed with micro-photoluminescence spectroscopy scans. The excellent agreement between the two techniques opens opportunities for this imaging concept to become a quantified, high spatial resolution, large-area characterization tool of PSCs. This development continues to strengthen the high value of luminescence imaging for the research and development of this photovoltaic technology.