High-resolution single-photon imaging with physics-informed deep learning.

High-resolution single-photon imaging remains a big challenge due to the complex hardware manufacturing craft and noise disturbances. Here, we introduce deep learning into SPAD, enabling super-resolution single-photon imaging with enhancement of bit depth and imaging quality. We first studied the complex photon flow model of SPAD electronics to accurately characterize multiple physical noise sources, and collected a real SPAD image dataset (64 × 32 pixels, 90 scenes, 10 different bit depths, 3 different illumination flux, 2790 images in total) to calibrate noise model parameters. With this physical noise model, we synthesized a large-scale realistic single-photon image dataset (image pairs of 5 different resolutions with maximum megapixels, 17250 scenes, 10 different bit depths, 3 different illumination flux, 2.6 million images in total) for subsequent network training. To tackle the severe super-resolution challenge of SPAD inputs with low bit depth, low resolution, and heavy noise, we further built a deep transformer network with a content-adaptive self-attention mechanism and gated fusion modules, which can dig global contextual features to remove multi-source noise and extract full-frequency details. We applied the technique in a series of experiments including microfluidic inspection, Fourier ptychography, and high-speed imaging. The experiments validate the technique's state-of-the-art super-resolution SPAD imaging performance.

U-Shape Transformer for Underwater Image Enhancement.

The light absorption and scattering of underwater impurities lead to poor underwater imaging quality. The existing data-driven based underwater image enhancement (UIE) techniques suffer from the lack of a large-scale dataset containing various underwater scenes and high-fidelity reference images. Besides, the inconsistent attenuation in different color channels and space areas is not fully considered for boosted enhancement. In this work, we built a large scale underwater image (LSUI) dataset, which covers more abundant underwater scenes and better visual quality reference images than existing underwater datasets. The dataset contains 4279 real-world underwater image groups, in which each raw image's clear reference images, semantic segmentation map and medium transmission map are paired correspondingly. We also reported an U-shape Transformer network where the transformer model is for the first time introduced to the UIE task. The U-shape Transformer is integrated with a channel-wise multi-scale feature fusion transformer (CMSFFT) module and a spatial-wise global feature modeling transformer (SGFMT) module specially designed for UIE task, which reinforce the network's attention to the color channels and space areas with more serious attenuation. Meanwhile, in order to further improve the contrast and saturation, a novel loss function combining RGB, LAB and LCH color spaces is designed following the human vision principle. The extensive experiments on available datasets validate the state-of-the-art performance of the reported technique with more than 2dB superiority. The dataset and demo code are available at https://bianlab.github.io/.

Image-free single-pixel object detection.

Recently developed image-free sensing techniques have achieved remarkable performance in various vision tasks. However, existing image-free methods still cannot simultaneously obtain the category, location, and size information of all objects. In this Letter, we report a novel image-free single-pixel object detection (SPOD) technique. SPOD enables efficient and robust multi-object detection directly from a small number of measurements, eliminating the requirement for complicated image reconstruction. Different from the conventional full-size pattern sampling method, the reported small-size optimized pattern sampling method achieves higher image-free sensing accuracy with fewer pattern parameters (∼1 order of magnitude). Moreover, instead of simply stacking CNN layers, we design the SPOD network based on the transformer architecture. It can better model global features and reinforce the network's attention to the targets in the scene, thus improving the object detection performance. We demonstrate the effectiveness of SPOD on the Voc dataset, which achieves a detection accuracy of 82.41% mAP at a sampling rate of 5% with a refresh rate of 63 f.p.s.

Photoluminescence Induced by Substitutional Nitrogen in Single-Layer Tungsten Disulfide.

The electronic and optical properties of two-dimensional materials can be strongly influenced by defects, some of which can find significant implementations, such as controllable doping, prolonged valley lifetime, and single-photon emissions. In this work, we demonstrate that defects created by remote N2 plasma exposure in single-layer WS2 can induce a distinct low-energy photoluminescence (PL) peak at 1.59 eV, which is in sharp contrast to that caused by remote Ar plasma. This PL peak has a critical requirement on the N2 plasma exposure dose, which is strongest for WS2 with about 2.0% sulfur deficiencies (including substitutions and vacancies) and vanishes at 5.6% or higher sulfur deficiencies. Both experiments and first-principles calculations suggest that this 1.59 eV PL peak is caused by defects related to the sulfur substitutions by nitrogen, even though low-temperature PL measurements also reveal that not all the sulfur vacancies are remedied by the substitutional nitrogen. The distinct low-energy PL peak suggests that the substitutional nitrogen defect in single-layer WS2 can potentially serve as an isolated artificial atom for creating single-photon emitters, and its intensity can also be used to monitor the doping concentrations of substitutional nitrogen.

Agile wide-field imaging with selective high resolution.

Wide-field and high-resolution (HR) imaging are essential for various applications such as aviation reconnaissance, topographic mapping, and safety monitoring. The existing techniques require a large-scale detector array to capture HR images of the whole field, resulting in high complexity and heavy cost. In this work, we report an agile wide-field imaging framework with selective high resolution that requires only two detectors. It builds on the statistical sparsity prior of natural scenes that the important targets locate only at small regions of interest (ROI), instead of the whole field. Under this assumption, we use a short-focal camera to image a wide field with a certain low resolution and use a long-focal camera to acquire the HR images of ROI. To automatically locate ROI in the wide field in real time, we propose an efficient deep-learning-based multiscale registration method that is robust and blind to the large setting differences (focal, white balance, etc) between the two cameras. Using the registered location, the long-focal camera mounted on a gimbal enables real-time tracking of the ROI for continuous HR imaging. We demonstrated the novel imaging framework by building a proof-of-concept setup with only 1181 gram weight, and assembled it on an unmanned aerial vehicle for air-to-ground monitoring. Experiments show that the setup maintains 120° wide field of view (FOV) with selective 0.45mrad instantaneous FOV.

Amorphous to Crystal Phase Change Memory Effect with Two-Fold Bandgap Difference in Semiconducting K2Bi8Se13.

Chalcogenide-based phase change memory (PCM) is a key enabling technology for optical data storage and electrical nonvolatile memory. Here, we report a new phase change chalcogenide consisting of a 3D network of ionic (K···Se) and covalent bonds (Bi-Se), K2Bi8Se13 (KBS). Thin films of amorphous KBS deposited by DC sputtering are structurally and chemically homogeneous and exhibit a surface roughness of 5 nm. The KBS film crystallizes upon heating at ∼483 K. The optical bandgap of the amorphous film is about 1.25 eV, while its crystalline phase has a bandgap of ∼0.65 eV shows 2-fold difference between the two states. The bulk electrical conductivity of the amorphous and crystalline film is ∼7.5 × 10-4 and ∼2.7 × 10-2 S/cm, respectively. We have demonstrated a phase change memory effect in KBS by Joule heating in a technologically relevant vertical memory cell architecture. Upon Joule heating, the vertical device undergoes switching from its amorphous to crystalline state of KBS at 1-1.5 V (∼50 kV/cm), increasing conductivity by a factor of ∼40. Besides the large electrical and optical contrast in the crystalline and amorphous KBS, its elemental cost-effectiveness, stoichiometry, fast crystallization kinetics, as determined by the ratio of the glass transition and melting temperature, Tg/Tm ∼ 0.5, as well as the scalable synthesis of the thin film determine that KBS is a promising PC material for next general phase change memory.

Creation of Single-Photon Emitters in WSe2 Monolayers Using Nanometer-Sized Gold Tips.

Due to their tunable bandgaps and strong spin-valley locking, transition metal dichalcogenides constitute a unique platform for hosting single-photon emitters. Here, we present a versatile approach for creating bright single-photon emitters in WSe2 monolayers by the deposition of gold nanostars. Our molecular dynamics simulations reveal that the formation of the quantum emitters is likely caused by the highly localized strain fields created by the sharp tips of the gold nanostars. The surface plasmon modes supported by the gold nanostars can change the local electromagnetic fields in the vicinity of the quantum emitters, leading to their enhanced emission intensities. Moreover, by correlating the emission energies and intensities of the quantum emitters, we are able to associate them with two types of strain fields and derive the existence of a low-lying dark state in their electronic structures. Our findings are highly relevant for the development and understanding of single-photon emitters in transition metal dichalcogenide materials.

Defect creation in WSe2 with a microsecond photoluminescence lifetime by focused ion beam irradiation.

Defect engineering is important for tailoring the electronic and optical properties of two-dimensional materials, and the capability of generating defects of certain types at specific locations is meaningful for potential applications such as optoelectronics and quantum photonics. In this work, atomic defects are created in single-layer WSe2 using focused ion beam (FIB) irradiation, with defect densities spanning many orders of magnitude. The influences of defects are systematically characterized. Raman spectroscopy can only discern defects in WSe2 for a FIB dose higher than 1 × 1013 cm-2, which causes blue shifts of both A'1 and E' modes. Photoluminescence (PL) of WSe2 is more sensitive to defects. At cryogenic temperature, the low-energy PL induced by defects can be revealed, which shows redshifts and broadenings with increased FIB doses. Similar Raman shifts and PL spectrum changes are observed for the WSe2 film grown by chemical vapor deposition (CVD). A four microsecond-long lifetime is observed in the PL dynamics and is three orders of magnitude longer than the often observed delocalized exciton lifetime and becomes more dominant for WSe2 with increasing FIB doses. The ultra-long lifetime of PL in single-layer WSe2 is consistent with first-principles calculation results considering the creation of both chalcogen and metal vacancies by FIB, and can be valuable for photo-catalytic reactions, valleytronics and quantum light emissions owing to the longer carrier separation/manipulation time.

Ultrafast Exciton Trapping at sp3 Quantum Defects in Carbon Nanotubes.

Semiconducting single-walled carbon nanotubes (SWCNTs) constitute an ideal platform for developing near-infrared biosensors, single photon sources, and nanolasers due to their distinct optical and electrical properties. Covalent doping of SWCNTs has recently been discovered as an efficient approach in enhancing their emission intensities. We perform pump-probe studies of SWCNTs that are covalently doped with sp3 quantum defects and reveal strikingly different exciton formation dynamics and decay mechanisms in the presence of the defect sites. We show that, in highly doped SWCNTs, ultrafast trapping of excitons at the defect sites can outpace other photodynamic processes and lead to ground-state photobleaching of the quantum defects. Our fitting of the transient data with a kinetic model also reveals an upper limit in the quantum defect density for obtaining highly luminescent SWCNTs without causing irreversible damage. These findings not only deepen our understanding of the photodynamics in covalently doped SWCNTs but also reveal critical information for the design of bright near-infrared emitters that can be utilized in biological, quantum information, and nanophotonic applications.

Multistates and Polyamorphism in Phase-Change K2Sb8Se13.

The phase-change (PC) materials in the majority of optical data storage media in use today exhibit a fast, reversible crystal → amorphous phase transition that allows them to be switched between on (1) and off (0) binary states. Solid-state inorganic materials with this property are relatively common, but those exhibiting an amorphous → amorphous transition called polyamorphism are exceptionally rare. K2Sb8Se13 (KSS) reported here is the first example of a material that has both amorphous → amorphous polyamorphic transition and amorphous → crystal transition at easily accessible temperatures (227 and 263 °C, respectively). The transitions are associated with the atomic coordinative preferences of the atoms, and all three states of K2Sb8Se13 are stable in air at 25 °C and 1 atm. All three states of K2Sb8Se13 exhibit distinct optical bandgaps, Eg = 1.25, 1.0, and 0.74 eV, for the amorphous-II, amorphous-I, and crystalline versions, respectively. The room-temperature electrical conductivity increases by more than 2 orders of magnitude from amorphous-I to -II and by another 2 orders of magnitude from amorphous-II to the crystalline state. This extraordinary behavior suggests that a new class of materials exist which could provide multistate level systems to enable higher-order computing logic circuits, reconfigurable logic devices, and optical switches.

All-Electrical Determination of Crystal Orientation in Anisotropic Two-Dimensional Materials.

The crystal orientation of an exfoliated black phosphorous flake is determined by purely electrical means. A sequence of three resistance measurements on an arbitrarily shaped flake with five contacts determines the three independent components of the anisotropic in-plane resistivity tensor, thereby revealing the crystal axes. The resistivity anisotropy ratio decreases linearly with increasing temperature T and carrier density reaching a maximum ratio of 3.0 at low temperatures and densities, while mobility indicates impurity scattering at low T and acoustic phonon scattering at high T.

The hypoxia-mimetic agent CoCl2 induces chemotherapy resistance in LOVO colorectal cancer cells.

Hypoxia, which is an important factor that mediates tumor progression and poor treatment response, is particularly associated with tumor chemoresistance. However, the molecular mechanisms underlying hypoxia-induced colorectal cancer chemoresistance remain unclear. The present study aimed to explore the mechanism underlying hypoxia­induced chemotherapy resistance in LOVO colorectal cancer cells. LOVO cells were cultured in a hypoxic environment simulated by cobalt chloride (CoCl2), which is a chemical inducer of hypoxia­inducible factor­1α (HIF­1α). HIF­1α is a transcription factor that has an important role in tumor cell adaptation to hypoxia, and controls the expression of several genes. Various CoCl2 concentrations are often used to simulate degrees of hypoxia. In the present study, following treatment with CoCl2, an MTT assay was conducted to determine the growth and drug sensitivity of LOVO cells. Reverse transcription­polymerase chain reaction and western blotting were used to detect the mRNA and protein expression levels of HIF­1α and factors associated with chemotherapy resistance, including multidrug resistance protein (MRP) and multidrug resistant 1 (MDR1), which encodes the major transmembrane efflux transporter P­glycoprotein (P­gp). In addition, the expression levels of apoptosis­related proteins, including B­cell lymphoma 2 (Bcl­2), Bcl­2­associated X protein (Bax) and Bcl­2­associated agonist of cell death (Bad) were detected by western blotting. Flow cytometry (FCM) was used to visually observe Adriamycin (ADR) accumulation and retention, thus analyzing intracellular drug transportation in cells under hypoxic and normoxic conditions. CoCl2­simulated hypoxia was able to inhibit tumor cell proliferation, and upregulate the expression levels of HIF­1α, MDR1/P­gp and MRP. In addition, proapoptotic members of the Bcl­2 protein family, Bax and Bad, were downregulated. The anti­apoptotic member Bcl­2 exhibited no significant change in expression, whereas the ratio of Bcl­2/Bax was increased. Results of FCM demonstrated that the intracellular retention of ADR was significantly decreased in the hypoxia group cells. In conclusion, the present study revealed that a CoCl2-simulated hypoxic microenvironment was able to effectively induce chemoresistance and reduce apoptosis in LOVO cells.

Evaporative thinning: a facile synthesis method for high quality ultrathin layers of 2D crystals.

The palette of two-dimensional materials has expanded beyond graphene in recent years to include the chalcogenides among other systems. However, there is a considerable paucity of methods for controlled synthesis of mono- and/or few-layer two-dimensional materials with desirable quality, reproducibility, and generality. Here we show a facile top-down synthesis approach for ultrathin layers of 2D materials down to monolayer. Our method is based on controlled evaporative thinning of initially large sheets, as deposited by vapor mass-transport. Rather than optimizing conditions for monolayer deposition, our approach makes use of selective evaporation of thick sheets to control the eventual thickness, down to a monolayer, a process which appears to be self-stopping. As a result, 2D sheets with high yield, high reproducibility, and excellent quality can be generated with large (>10 µm) and thin (∼ 1-2 nm) dimensions. Evaporative thinning promises to greatly reduce the difficulty involved in isolating large, mono- and few-layers of 2D materials for subsequent studies.