Guangcai Chang

X-ray phase contrast imaging based on propagation is a crucial technique for achieving non-destructive detection at the micro- and nano-scales. However, the recovery of phase information from intensity measurements presents a typical ill-posed inverse problem. Traditional iterative algorithms often necessitate multi-distance measurements, which increases both the complexity and time cost of experiments. Although deep learning has introduced a novel paradigm for phase retrieval, supervised methods depend on paired data that are challenging to obtain. Furthermore, existing approaches frequently ignore wavefront distortion of the illumination probe, leading to significant background artifacts in reconstruction results. To address these challenges, we propose a probe-aware self-supervised neural network (PASNet), designed to jointly estimate the complex object and probe wavefields from single-distance holograms for phase reconstruction. PASNet employs a compact U-Net architecture that takes as joint input holograms representing “object + probe” and “probe only.” A physically-driven loss function is utilized to integrate the Fresnel diffraction into the network optimization process, enabling self-supervised learning without requiring labeled data. Evaluation on simulated and experimental datasets demonstrates that PASNet effectively decouples object and probe wavefronts while suppressing background artifacts. With only single-distance double-exposure, our proposed method achieves superior phase structural fidelity and background stability compared with traditional iterative algorithms and other neural networks. Additionally, robustness analysis indicates that PASNet maintains strong performance across varying propagation distances and noise levels.

A systematic opto-mechanical design approach is developed for a long Nano-AKB mirror system aimed at stable sub-50 nm hard X-ray nanofocusing. The approach integrates deformation–stiffness-balanced mirror clamping, hierarchical precision alignment, and thermomechanical stability management and has been applied to nanofocusing instruments at HEPS. During beamline commissioning at HEPS ID19, the system achieved a focal spot of 34.6 ± 1.31 nm × 41.6 ± 0.67 nm (H × V) at 20 keV while maintaining Z-displacement stability below 5 nm RMS. The results meet the design expectations, strongly attesting to the efficacy and practical viability of the proposed approach.

Overglaze colour decoration represents a seminal innovation in ancient Chinese ceramic technology, yet the technical and artistic mechanisms underlying its earliest forms remain poorly understood. Honglvcai wares from the Jin dynasty (1115–1234 CE) are a primary example of this development. However, the absence of detailed surface characterisation has hindered insights into painting techniques. Here we combine elemental mapping, phase characterisation, and quantitative surface topography to reconstruct the canonical brush-handling methods employed by artisans at the Dangyangyu kiln. The red pigment is confirmed as hematite, dispersed in a lead-based flux that diffuses into the underlying white glaze, ensuring adhesion. The Danxian Pingtu (‘single-outline and flat wash’) and Mogu (‘boneless’) painting styles were recognised. Critically, roughness gradients across lines act as physical records of deliberate pressure modulation for pigment deposition, indicating Zhongfeng (centred-tip) and Cefeng (side-brush) techniques and revealing exceptional motor control. These findings shed light on how brush dynamics, material behaviours, and firing processes converged for achieving exquisite Honglvcai overglaze decoration. This work establishes surface metrology as a vital tool for craftsmanship reconstruction in ancient Chinese ceramics.

Corrigendum to the article by Hu et al. [(2026). J. Synchrotron Rad. 33 , 454–466].

A comprehensive understanding of working and degradation mechanisms of energy materials is essential for enhancing their performance and tackling contemporary energy and environmental challenges, particularly in the transition from fossil fuels to renewables. Emerging nanoscale X‐ray characterization techniques based on synchrotron radiation provide robust, multi‐scale, and multi‐modal analytical capabilities, making them a suite of powerful tools for revealing operational and failure mechanisms, especially of bulk materials with features ranging from nano to micrometers under realistic operating conditions. This Perspective initially highlights the advantages of X‐ray methods based on synchrotron light sources and then summarizes X‐ray‐matter interactions and corresponding characterization methods. Three pivotal approaches that facilitate nanoscale X‐ray imaging are subsequently summarized. Representative examples illustrate the versatility and power of these prominent techniques, including X‐ray imaging, scattering/diffraction, and spectroscopy, in probing morphologies, compositions, atomic structures, and electronic properties. Integration of multi‐scale and multi‐modal X‐ray techniques is underscored to achieve synergistic enhancements in analytical depth. It concludes with a discussion of current challenges, emerging opportunities, and future directions, providing insights into potential advancements in this field.

X-ray ptychography is a coherent diffraction imaging technique that allows for the quantitative retrieval of both the amplitude and phase information of a sample in diffraction-limited resolution. However, traditional reconstruction algorithms require a large number of iterations to obtain phase and amplitude images exactly, and the expensive computation precludes real-time imaging. To solve the inverse problem of ptychography data, PtychoNN uses deep convolutional neural networks for real-time imaging. However, its model is relatively simple, and its accuracy is limited by the size of the training dataset, resulting in lower robustness. To address this problem, a series of W-Net neural network models have been proposed which can robustly reconstruct the object phase information from the raw data. Numerical experiments demonstrate that our neural network exhibits better robustness, superior reconstruction capabilities and shorter training time with high-precision ptychography imaging.

The Laue-type multilayer monochromator (LMM) is a promising optical element with a small size and high efficiency in a synchrotron radiation facility. By the dynamical diffraction theory, using DC magnetron sputtering technology, an LMM with a total thickness of 47 µm and a periodic thickness of 4.7 nm WSi₂/Si multilayer at 26 keV is designed and fabricated. During the preparation, the total number of layers is up to 20000, and every 300th layer of Si is replaced by WSi2 as the marker, so the multilayer is divided into 67 areas. The cross section of the multilayer is measured by a scanning electron microscope (SEM), and the marker region thickness error is 0.28% (RMS). The diffraction test experiment of the LMM is carried out at the Shanghai synchrotron radiation facility (SSRF). The 1st-order peak angle is 5.05 mrad, and the efficiency is 75.0%, which is close to the theoretical calculation result of 5.1 mrad and 79.1%. The Darwin width of the LMM is 0.17 mrad which is equal to the theoretical calculation. Based on the Bragg’s diffraction equation, the energy resolution (ΔE/E) is 3.3%.

Lateral graded multilayer can realize reflection, collimation and focusing of hard X-ray, and are currently the research frontier and hotspot of synchrotron radiation and high-performance X-ray sources. To reduce the d-spacing error of graded multilayers, a root mean square error optimization method based on double genetic algorithm (DGA-RMSE) is proposed. The theoretical d-spacing distribution is obtained by optical design, and the range is 1.9 ∼ 3.1 nm. The optimized d-spacing distribution is obtained by convolution of particle beam function and continuous monotonic rate distribution line (RDL) which is constructed in the form of a polynomial. The GA is applied to optimize variables from the polynomial twice, and the RMSE of thickness error is optimized and converged to 0.0065 nm. The final thickness error which is measured by the grazing incidence X-ray reflectivity (GIXRR) is consistent with the theoretical calculation. The results show that DGA-RMSE can precisely select polynomial function of RDL, reducing the error in high-precision magnetron sputtering and mask technology.

In synchrotron facilities, optics with multilayer coatings are used for beam monochromatization, focusing, and collimation. These coatings might be damaged by high heat load, poor film adhesion, high internal stress, or poor vacuum. Optical substrates always need high quality, which is expensive and has a long processing cycle. Therefore, it is desired to make the substrate reusable and the refurbished coating as good as a brand-new one. In this study, a W/B₄C multilayer coating with a 2 nm Cr buffer layer was prepared on a Si substrate. The coating was successfully stripped from the Si substrate by dissolving the Cr buffer layer using an etchant. The roughness and morphology after the different etching times were investigated by measuring the GIXRR and 3D surface profiler. It is shown that the time required to etch the W/B₄C multilayer coating with a Cr buffer layer, is quite different compared with etching a single Cr film. A layer of silicon dioxide was introduced during the fitting. After the new etching process, the roughness of the sample is as good as the one on a brand-new substrate. The W/B₄C multilayer coatings with a Cr buffer layer were recoated on the etched samples, and the interface roughness was not damaged by the etching process.

A new, to the best of our knowledge, figuring method for a multilayer Laue lens (MLL) in the hard X-ray region is proposed in this paper. Theoretical simulation at 20 keV shows that the figuring method can compensate for the structure error. The phase errors of the first-order diffracted wave decrease from 0.85π to 0.26π after figuring. The spatial resolution changes from 45 nm to 26 nm after figuring, which is almost the same as that of the ideal MLL with a spatial resolution of 24 nm. The figured MLL can achieve 36% of the ideal MLL's first-order diffraction efficiency. Such method may reduce the requirements for the fabrication of the MLLs, and may make it possible to manufacture the larger numerical aperture MLL with the longer working distance in the future.