Contrast-variation resonant soft X-ray scattering for partial scattering functions of multicomponent soft matter systems

Resonant soft X-ray scattering (RSoXS) is a powerful synchrotron-based tool for characterization due to its intrinsic element and bond sensitivities, large accessible size scale, and unique bond orientation sensitivity. In small angle scattering, any pattern from more than two isotropic materials can be split into separate contributions from each material that vary with the contrasts of those materials. In small angle neutron scattering (SANS), contrast variation is achieved by synthetically replacing hydrogens with deuterium in a monotonic radiolabeled series and applying a singular value decomposition to obtain partial scattering functions. We propose an analogous quantitative decomposition approach to RSoXS data analysis to isolate and measure the distribution of individual materials within a system by exploiting the energy-dependent scattering length densities (complex indices of refraction) of each component across the soft X-ray energy range. This "stainless," chemistry-specific contrast is created simply by changing the incident X-ray energy, thus eliminating the need for radiolabeling and ensuring a consistent structure factor throughout the process. The decomposed structure functions of a triblock copolymer system are evaluated using GPU-accelerated forward simulations of RSoXS patterns based on real-space models, allowing for the characterization of nontrivial and nonequilibrium structures. This quantitative analysis using RSoXS flips the paradigm from using spectroscopy as a fingerprint to understand variations in a "stack" of RSoXS patterns, to instead "baking in" the energy dependence at an early stage of analysis to produce energy-agnostic partial scattering functions of individual materials and greatly accelerate the extraction of meaningful material structure information in multicomponent soft materials.