Rapid in-situ calibration of computational micro-spectrometer with few-shot meta-learning.

Computational micro-spectrometers comprised of detector arrays and encoding structure arrays, such as on-chip Fabry-Perot (FP) cavity filters, have great potential in many in-situ applications owing to their compact size and snapshot imaging ability. Given manufacturing deviation and environmental influence are inevitable, easy and effective calibration for spectrometer is necessary, especially for in-situ applications. Currently calibration strategies based on iterative algorithms or neural networks require accurate measurements of pixel-level (spectral) encoding functions through monochromator or large amounts of standard samples. These procedures are time-consuming and expensive, thereby impeding in-situ applications. Meta-learning algorithms with few-shot learning ability can address this challenge by incorporating the prior knowledge in the simulated dataset. In this work, we propose a meta-learning algorithm free of measuring encoding function or large amounts of standard samples to calibrate a micro-spectrometer with manufacturing deviation effectively. Our micro-spectrometer comprises 16 types of FP filters covering a wavelength range of 550-720 nm. The center wavelength of each filter type deviates from the design up to 6 nm. After calibration with 15 different color data, the average reconstruction error on the test dataset decreased from 7.2 × 10-3 to 1.2 × 10-3, and further decreased to 9.4 × 10-4 when the calibration data increased to 24. The performance is comparable to algorithms trained with measured encoding function both in reconstruction error and generalization ability. We estimated that the cost of in-situ calibration through reflectance measurements of color chart decreased to one percent of the cost through monochromator measurements. By exploiting prior deviation information in simulation data with meta-learning, the efficiency and cost of calibration are significantly improved, thereby facilitating the large-scale production and in-situ application of micro-spectrometers.

Electron injection and defect passivation for high-efficiency mesoporous perovskite solar cells.

Printable mesoscopic perovskite solar cells (p-MPSCs) do not require the added hole-transport layer needed in traditional p-n junctions but have also exhibited lower power conversion efficiencies of about 19%. We performed device simulation and carrier dynamics analysis to design a p-MPSC with mesoporous layers of semiconducting titanium dioxide, insulating zirconium dioxide, and conducting carbon infiltrated with perovskite that enabled three-dimensional injection of photoexcited electrons into titanium dioxide for collection at a transparent conductor layer. Holes underwent long-distance diffusion toward the carbon back electrode, and this carrier separation reduced recombination at the back contact. Nonradiative recombination at the bulk titanium dioxide/perovskite interface was reduced by ammonium phosphate modification. The resulting p-MPSCs achieved a power conversion efficiency of 22.2% and maintained 97% of their initial efficiency after 750 hours of maximum power point tracking at 55 ± 5°C.

Highly oriented MAPbI3 crystals for efficient hole-conductor-free printable mesoscopic perovskite solar cells.

Highly crystalline perovskite films with large grains and few grain boundaries are conducive for efficient and stable perovskite solar cells. Current methods for preparing perovskite films are mostly based on a fast crystallization process, with rapid nucleation and insufficient growth. In this study, MAPbI3 perovskite with inhibited nucleation and promoted growth in the TiO2/ZrO2/carbon triple mesoscopic scaffold was crystallized by modulating the precursor and the crystallization process. N-methylformamide showed high solubility for both methylammonium iodide and PbI2 and hampered the formation of large colloids in the MAPbI3 precursor solution. Furthermore, methylammonium chloride was added to reduce large colloids, which are a possible source of nucleation sites. During the crystallization of MAPbI3, the solvent was removed at a slow controlled speed, to avoid rapid nucleation and provide sufficient time for crystal growth. As a result, highly oriented MAPbI3 crystals with suppressed non-radiative recombination and promoted charge transport were obtained in the triple mesoscopic layer with disordered pores. The corresponding hole-conductor-free, printable mesoscopic perovskite solar cells exhibited a highest power conversion efficiency of 18.82%. The device also exhibited promising long-term operational stability of 1000 h under continuous illumination at maximum power point at 55 ± 5 °C and damp-heat stability of 1340 h aging at 85 °C as well as 85% relative humidity.