UCLA

Coherent diffractive imaging (CDI) is an imaging technique which uses reconstructionalgorithms in place of lenses to avoid aberration, which becomes vital in the x-ray regime. Over the last two decades, CDI has served as an invaluable tool to reveal the structure and chemical composition of a wide range of material and biological systems, providing insight across many disciplines. As part of the modern scientific frontier, CDI continues to evolve as an integral tool for the investigation of increasingly complex systems, while pushing the spatiotemporal limit. This work contains three studies which demonstrate extended capabilities of CDI through the utilization of fundamental light-matter interactions in tandem with advanced algorithms. First, the development of a novel broadband reconstruction algorithm called “Spectrum Probe and Image Reconstruction” (SPIRE) is presented in the context of attosecond imaging. An understanding of incoherent scattering permits the development of SPIRE which is shown to outperform other broadband reconstruction algorithms through the use of novel constraints. SPIRE is tested with a set of numerical simulations and a tabletop experiment with a white LED source. The results demonstrate the viability of CDI with broadband illumination and on ultrafast time scales. Second, x-ray magnetic circular dichroism (XMCD) is used to probe the magnetization of a geometrically confined ferromagnet on the nanoscale. A novel vector tomography algorithm titled “Vector REal Space Iterative Reconstruction Engine (Vector RESIRE)" is used to directly reconstruct the 3D magnetization vector field in addition to the sample structure, without any prior assumptions. Topological analysis of the reconstruction result reveals a network of non-trivial magnetic point defects known as hedgehogs. The 3D spatial distribution of hedgehogs serve to validate the current theory of hedgehog confinement and potential of metalattices in spintronics applications. In the appendix, polarization-dependent imaging contrast (PIC) is used to map the orientation of the crystal c-axis of calcium carbonate in a coral skeleton. Hierarchical clustering is applied to 4D scanning transmission electron microscope (STEM) data from the same sample to gain additional information about the crystal grain structure. The PIC and 4D STEM results are correlated to elucidate growth and nucleation conditions for coral.