Unlocking hidden information in sparse small-angle neutron scattering measurements.

Area of Science:

• Soft matter physics
• Materials science
• Neutron scattering techniques

Background:

• Small-Angle Neutron Scattering (SANS) provides nanoscale structural insights into soft matter.
• Low neutron flux limits SANS effectiveness, causing long acquisition times and noisy data.

Purpose of the Study:

• To develop a Bayesian statistical inference method using Gaussian Process Regression (GPR) for reconstructing high-fidelity SANS data.
• To enhance SANS data quality and reduce measurement times.

Main Methods:

• Benchmarking computationally and validating through SANS experiments on diverse soft matter systems.
• Applying GPR-based inference to both experimental and synthetic SANS data.
• Leveraging intensity smoothness and continuity for data reconstruction.

Main Results:

• GPR significantly enhances SANS data quality, reducing measurement times by up to two orders of magnitude.
• The method effectively reduces noise and reconstructs scattering intensity.
• Demonstrated applicability to wormlike micelles, colloidal suspensions, polymers, and lyotropic phases.

Conclusions:

• GPR-based Bayesian inference offers a cost-effective approach to maximize experimental efficiency in SANS.
• This framework is beneficial for weakly scattering, time-sensitive studies, and applicable to other low-SNR techniques.
• Potential for transformative impact on compact neutron sources and structural analysis in resource-limited settings.