We report all-inorganic solar cells based on solution-processed BiI3. Two-electron donor solvents such as tetrahydrofuran and dimethylformamide were found to form adducts with BiI3, which make them highly soluble in these solvents. BiI3 thin films were deposited by spin-coating. Solvent annealing BiI3 thin films at relatively low temperatures (≤100 °C) resulted in increased grain size and crystallographic reorientation of grains within the films. The BiI3 films were stable against oxidation for several months and could withstand several hours of annealing in air at temperatures below 150 °C without degradation. Surface oxidation was found to improve photovoltaic device performance due to the formation of a BiOI layer at the BiI3 surface which facilitated hole extraction. Nonoptimized BiI3 solar cells achieved the highest power conversion efficiencies of 1.0%, demonstrating the potential of BiI3 as a nontoxic, air-stable metal-halide absorber material for photovoltaic applications.

The remarkable performance of Pb halide perovskites in optoelectronic devices is complicated by concerns over their toxicity, which has motivated a search for Pb-free alternatives that have similar performance. Bi halides and halide perovskites have been predicted to be among the most promising Pb-free alternatives; however, their performance in devices has fallen short of expectations. One of the major challenges in fabricating efficient devices based on these Bi-based alternatives has been poor control over the morphology of thin films. Using BiI 3 as a model system, we demonstrate that the film morphology and surface coverage are strongly dependent on the Lewis basicity of solvents that are used during deposition. We demonstrate that coordinating BiI 3 with strong Lewis bases in tetrahydrofuran results in conformal films that have been difficult to achieve using conventional deposition techniques.

Bismuth-based halide perovskites have been proposed as a potential nontoxic alternative to lead halide perovskites; however, they have not realized suitable performance. Their poor performance has been attributed to substandard film morphologies and too wide of a band gap for many applications. Herein we used a two-step deposition procedure to convert BiI3 thin films into A3Bi2I9 (A = FA+, MA+, Cs+, or Rb+), which resulted in a substantial improvement in film morphology, a larger band gap, and greater compositional tunability compared toresults when using aconventional single-step deposition technique. Additionally, we attempted to reduce the undesirably wide band gap in Rb3Bi2I9 thin films by inducing chemical pressures through cation-size mismatch, with an underlying hypothesis that cation-size mismatch could induce compressive strain within the 2D Rb3Bi2I9 lattice. However, we found that all A x Rb3–x Bi2I9 compositions with x > 0 adopted the 0D structure, and no changes to the band gap were observed with alloy. These results imply that the band gap of A x Rb3–x Bi2I9 is insensitive to A-site alloying.

The ability to engineer the surface chemistry of complex ternary nanocrystals is critical to their successful application in photovoltaic, thermoelectric, and other energy conversion devices. For many years, several studies have shed light into the surface chemistry of unary and binary semiconductor nanocrystals, as well as their surface modification with monodentate and multidentate ligands in a variety of applications. In contrast, our understanding of the surface chemistry and ligand modification of ternary and other complex multinary nanocrystals remains relatively limited. Recently, our group reported the synthesis of colloidal NaBiS2 semiconductor nanocrystals with sizes tunable between 2-60 nm, and a light absorption edge of ca. 1.4 eV. Here, we use a combination of infrared and nuclear magnetic resonance spectroscopies to show that the as-made NaBiS2 nanocrystals are capped by oleylamine and neodecanoate ligands. We investigate biphasic liquid-liquid exchange as a means to replace these native ligands with either carboxylate-terminated lipoic acid or with small iodide ligands, leading in both cases to solubility in polar solvents-such as methanol, water, and dimethylformamide. We also investigate a layerby-layer, biphasic solid-liquid exchange approach to prepare films of NaBiS2 nanocrystals capped with halide ligands-iodide, bromide, chloride. Upon exchange and removal of the native ligands, we show that the resistance of NaBiS2 nanocrystal films greatly decreases, with their measured conductivity being comparable to that of films made of isostructural PbS nanocrystals, which have been used in solar cells. Lastly, we report the first solar cell device made of NaBiS2 nanocrystal films with a limited power conversion efficiency (PCE) of 0.07. Further nanostructuring and ligand optimization may enable the preparation of much more efficient energy conversion devices based on NaBiS2 as well as other non-toxic and Earth-abundant, biocompatible multinary semiconductors.

Germanium nanocrystals (Ge NCs) have potential to be used in several optoelectronic applications such as photodetectors and light-emitting diodes.