21–25 Sept 2026
Paul Scherrer Institut
Europe/Zurich timezone
Due to shipping delays, we will not be able to hand out conference giveaways on Monday. We hope they will arrive later this week. Otherwise, we will send them to you. Sorry for that.

Train-Resolved Statistical Recovery of Weak SAXS Signals in Liquids at the European XFEL

Not scheduled
25m
Auditorium (WHGA/001) (Paul Scherrer Institut)

Auditorium (WHGA/001)

Paul Scherrer Institut

Invited Chemistry Chemistry

Speaker

Carles Serrat (Polytechnic University of Catalonia)

Description

Small-angle X-ray scattering (SAXS) measurements of aqueous L-cysteine solutions were performed at the SPB/SFX instrument of the European XFEL using train-resolved acquisition with the AGIPD detector. Detecting weak sample-dependent signals under XFEL conditions is challenging because the measured contrast is often dominated by normalization uncertainties, beam fluctuations, detector effects, and liquid-jet variability. To address this problem, we developed a train-resolved analysis based on independent scale-plus-offset fitting of cysteine and water profiles, combined with transmission-matched water--water controls. This approach isolates residual scattering contributions that cannot be explained by global intensity variations and exceed the residual structure observed in matched water controls.

Top panel: Control-corrected residual SAXS profiles after train-resolved scale-plus-offset correction and subtraction of transmission-matched water controls.

Bottom panel: Corresponding statistical significance (ΔR/σΔR), demonstrating the statistical robustness of the recovered signal.

For 0.5 M aqueous L-cysteine, we observe a reproducible control-corrected residual SAXS signal whose amplitude increases strongly with XFEL transmission. The residual exhibits a characteristic sign-changing dependence on momentum transfer, with a positive contribution at low q and a negative contribution at higher q. Closely similar residual shapes are observed independently at the two highest transmission settings, consistent with a common fluence-dependent contribution. Bootstrap and convergence analyses show that the signal emerges progressively from the accumulation of many train pairs and exhibits uncertainty scaling close to the expected 1/√N behaviour.
Beyond the observation of a weak fluence-dependent residual SAXS signal, the work establishes a methodological result of broader relevance for XFEL solution scattering: high repetition rate alone is insufficient to achieve maximum sensitivity. Train-resolved analysis together with closely matched water controls proved critical for recovering statistically robust weak signals and distinguishing them from slow experimental drifts and common-mode fluctuations, a challenge that also arises in nonlinear X-ray measurements.

Author

Carles Serrat (Polytechnic University of Catalonia)

Co-authors

Presentation materials

There are no materials yet.