Two-dimensional Ruddlesden−Popper perovskites (2D RPPs) have attracted tremendous research interest because of their excellent photoelectric performance and environmental stability. However, a fundamental understanding of the intrinsic fluorescence characteristics is still limited, and the origin of dual-emission peaks in 2D RPPs is under intense debate. In this work, high-quality 2D (BA) 2 PbBr 4 single crystals that were synthesized using a novel one-step cooling crystallization method exhibited obvious dual blue emission peaks at 412 and 432 nm under wide-field excitation. Space-and time-resolved photoluminescence (PL) spectra of mechanically exfoliated flakes under one-and two-photon excitation reveal that a single high-energy exciton emission was observed from the excitation spot, which underwent a remarkable redshift after propagation. The results can be well fit by the photon recycling model. Our results confirmed that the PL peak at 412 nm originates from the exciton emission at the excitation spot, while the PL peak at 432 nm was induced by photon recycling and long-range waveguide assisted by total internal reflection. Our findings emphasize the importance of photon recycling in the optical spectrum or the exciton kinetics in 2D layered perovskites. Moreover, the 2D perovskite with inner-plane directional photon recycling and a waveguide effect exhibits great potential in constructing highperformance optoelectronic devices.

The low-energy layer edge states (LESs) from quasi 2D hybrid perovskite single crystals have shown great potential because of their nontrivial photoelectrical properties. However, the underlying formation mechanism of the LESs still remains controversial. Also, the presence or creation of the LESs is of high randomness due to the lack of proper techniques to manually generate these LESs. Herein, using a single crystals platform of quasi-2D (BA) 2 (MA) n−1 Pb n I 3n+1 (n > 1) perovskites, the femtosecond laser ablation approach to design and write the LESs with a high spatial resolution is reported. Fundamentally, these LESs are of smaller bandgap 3D MAPbI 3 nanocrystals which are formed by the laser-induced BA escaping from the lattice and thus the lattice shrinkage from quasi-2D to 3D structures. Furthermore, by covering the crystal with tape, an additional high-energy emission state corresponding to the reformation of (BA) 2 PbI 4 (n = 1) within the irradiation region is generated. This work presents a simple and efficient protocol to manually write LESs on single crystals and thus lays the foundation for utilizing these LESs to further enhance the performance of future photoelectronic devices.

Manipulating the propagation of photons is meaningful for optical data recognition but extremely challenging. Recently, the metal halide perovskite semiconductors present great potential in photonics and optoelectronics applications, such as waveguide, lasing, photoluminescence, electroluminescence, photovoltaic, and photodetection. However, so far there are only few research studies that have focused their attention on the management of photons applying this type of emerging and promising semiconductors. Herein, we propose a conceptually new perovskite microconfiguration which could delay the photon propagation and guide the photon flow direction with at least 30 times multiplication of communication bandwidth. The defect localization effect on photo-excited carriers slows down the photon transmission on the microsecond scale. Furthermore, we propose a controllable photon communication speed ascribing to a continuous wave laser irradiation-tunable recombination rate. The present results indicate the great potential of inorganic perovskite in all-optical information processing.

The carrier diffusion length (LD) is of great relevance to the quantum efficiencies of organic-inorganic perovskites. However, so far there is no direct and noncontact measurement of the carrier diffusion length in this emerging material. Herein, we directly visualize the carrier diffusion length of organic-inorganic perovskites via spatial mapping of photoluminescence. Our results reveal that the carrier diffusion length (LD) is 6.25 μm for three-dimensional FAPbBr3 single crystals. Further, we discover that nonlinear and nonradiative effects can be neglectable during the diffusion process of photogenerated carriers in FAPbBr3 single crystals. In contrast, two-dimensional BA2PbI4 single crystals display shorter LD (2.40 μm) due to the transport barrier from their insulating organic layer. Our work simplifies the complicated operation for carrier diffusion length measurements, which is fundamentally important for technical development and scientific research on organic-inorganic perovskites.

DEtection TRansformer (DETR) started a trend that uses a group of learnable queries for unified visual perception. This work begins by applying this appealing paradigm to LiDAR-based point cloud segmentation and obtains a simple yet effective baseline. Although the naive adaptation obtains fair results, the instance segmentation performance is noticeably inferior to previous works. By diving into the details, we observe that instances in the sparse point clouds are relatively small to the whole scene and often have similar geometry but lack distinctive appearance for segmentation, which are rare in the image domain. Considering instances in 3D are more featured by their positional information, we emphasize their roles during the modeling and design a robust Mixed-parameterized Positional Embedding (MPE) to guide the segmentation process. It is embedded into backbone features and later guides the mask prediction and query update processes iteratively, leading to Position-Aware Segmentation (PA-Seg) and Masked Focal Attention (MFA). All these designs impel the queries to attend to specific regions and identify various instances. The method, named Position-guided Point cloud Panoptic segmentation transFormer (P3Former), outperforms previous state-ofthe-art methods by 3.4% and 1.2% PQ on SemanticKITTI and nuScenes benchmark, respectively. The source code and models are available at https://github.com/SmartBot-PJLab/P3Former.