X-ray ptychography has the advantage of nanoscale resolution and extended field-of-view (FOV), which has led to its widespread application in many cutting-edge research fields. Currently, the technique faces the following challenges: greatly increased time cost for large FOV and increased reconstruction difficulty due to large data volume. In this study, we present a synergistic method to improve the efficiency of large FOV ptychography by integrating a large-probe large-step scanning strategy with an iterative partition reconstruction (IPR) algorithm. To increase the probe size beyond the detector Fourier window, we develop an upsampling-multimode fusion algorithm that enables ptychography scan with a step size over 5 µm using only stepper motors, thus improving the acquisition efficiency by more than 12 times compared to conventional ptychography and STXM. Meanwhile, we propose an IPR method with an adaptive sample splicing mechanism for raster scan or partition scan, which effectively suppresses the artifacts due to probe variations, significantly reduces the hardware requirements for reconstruction while improving reconstruction efficiency, and eliminates 10% overlapping redundancy in conventional tile scan, further improving efficiency. Verification experiments using probes of 20 and 25 µm demonstrated the performance of the developed strategy, which provides a new solution for high-resolution, large FOV ptychography.
Li et al. (Thu,) studied this question.