Sung Yun Lee

Machine learning is attracting surging interest across nearly all scientific areas by enabling the analysis of large datasets and the extraction of scientific information from incomplete data. Data-driven science is rapidly growing, especially in X-ray methodologies, where advanced light sources and detection technologies produce vast amounts of data that exceed meticulous human inspection capabilities. Despite the increasing demands, the full application of machine learning has been hindered by the need for data-specific optimizations. In this study, we introduce a new deep-learning-based phase retrieval method for imperfect diffraction data. This method provides robust phase retrieval for simulated data and performs well on partially damaged and noisy single-pulse diffraction data from X-ray free-electron lasers. Moreover, the method significantly reduces data processing time, facilitating real-time image reconstructions that are crucial for high-repetition-rate data acquisition. This approach offers a reliable solution to the phase problem to be widely adopted across various research areas confronting the inverse problem.

Structured light has sparked interest by introducing a parameter to adjust light–matter interactions in photonics applications. While phase structures applied to their wavefronts enable functional imaging probes, their application in X-ray microscopy has been limited due to their complex configuration. Here, we present X-ray vortex beam ptychography using an off-axis spiral zone plate, showcasing quantitative nanoscale imaging with straightforward manipulation of the phase structure. The simple experimental setup enhances adaptability to most light sources. This coherent X-ray microscopy, capable of resolving nontrivial topological states with additional phase structures of light, is poised to advance photonics research by leveraging topological features.

Mesoporous nanoparticles provide rich platforms to devise functional materials by customizing the three-dimensional (3D) structures of nanopores. With the pore network as a key tuning parameter, the noninvasive and quantitative characterization of these 3D structures is crucial for the rational design of functional materials. This has prompted researchers to develop versatile nanoprobes with a high penetration power to inspect various specimens sized a few micrometers at nanoscale 3D resolutions. Here, with adaptive phase retrievals on independent data sets with different sampling frequencies, we introduce multidistance coherent X-ray tomography as a noninvasive and quantitative nanoprobe to realize high-resolution 3D imaging of micrometer-sized specimens. The 3D density distribution of an entire mesoporous silica nanoparticle was obtained at 13 nm 3D resolution for quantitative physical and morphological analyses of its 3D pore structure. The morphological features of the whole 3D pore network and pore connectivity were examined to gain insight into the potential functions of the particles. The proposed multidistance tomographic imaging scheme with quantitative structural analyses is expected to advance studies of functional materials by facilitating their structure-based rational design.

Phase recovery is a well-known inverse problem prevalent across science disciplines and attracts active research interests to develop a number of theoretical and experimental methods. Recent developments in artificial intelligence have further prompted research activities in processing the experimentally collected imperfect data, but applications have been limited to slow-varying data such as real images. Experimental noise present in largely fluctuating diffraction data, in particular, adds practical challenges to hamper consistent phase recovery. Here, we introduce a convolutional neural-network assisted k-space denoising method that can directly manage noisy diffraction signals. It showed superior performance on denoising the diffraction data, which promote improved phase recovery from noise-buried single-pulse diffraction signals obtained by the x-ray free-electron laser. Adapting our method to general diffraction data can expand boundaries of interpretable data and enhance observability of faint objects with weak signals.