Dissertation / PhD Thesis

PUBDB-2023-06144

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Imaging via photon-photon correlation of X-ray fluorescence

Trost, F. (Corresponding author)CFEL* ; Chapman, H. N. (Thesis advisor)CFEL*XFEL.EU*DESY* ; Roehlsberger, R. (Thesis advisor)XFEL.EU*DESY*

2023

This record in other databases:

Please use a persistent id in citations: urn:nbn:de:gbv:18-ediss-112508  doi:10.3204/PUBDB-2023-06144

Abstract: This work discusses and demonstrates the imaging method ‘photon-photon correlation of X-ray fluorescence’, also called ‘Incoherent Diffraction Imaging (IDI)’. This method corresponds to the ‘intensity interferometry’, known from astronomy. However, this method requires measurement or exposure times that are on the order of the coherence time of the measured radiation; for Kα fluorescence of light transition metals, this is on the order of a femtosecond – a temporal length that can be achieved by modern X-ray free electron lasers (XFELs).In addition to the exposure time requirement, this work discusses other influencing factors that can cause a reduction in the useful correlation signal. These are the temporal shape of the excitation pulse, the sample size, the (non) polarization of the detected photons, and others. These factors, collectively called ‘visibility factor’, also manifest as ‘speckle contrast’ and can be measured without the need to perform intensity correlation. A weighting method is presented to determine the speckle contrast from a data set consisting of images with very low photon counts that fluctuate significantly from image to image. This method is applied to compare the speckle contrast of iron Kα fluorescence excited by 3 fs and 15 fs XFEL pulses. An increase in speckle contrast was found for the short pulses compared to the longer ones – a fundamental requirement for the IDI method.Furthermore, inherent noise sources affecting the IDI are discussed. A model is derived to estimate the dependence of the signal-to-noise ratio (SNR) on the photon number per pixel, temporal coherence (or visibility factor), and the shape of the imaged object. In addition, simulations in two and three dimensions were performed to validate the model’s predictions. Unlike coherent imaging methods, more detected photons do not always result in higher SNR. Moreover, larger and more complex objects generally yield poorer SNR, even when the number of measured photons is proportional to the object size or complexity.Finally, an experiment that uses the photon-photon correlation of X-ray fluorescence photons for the first time to reconstruct a nontrivial (noncontinuous) fluorescence emitter distribution is presented. In the course of this experiment, the application of IDI to determine XFEL beam parameters such as focus and temporal pulse length is demonstrated.

Note: Dissertation, University of Hamburg, 2023

---

Contributing Institute(s):

1. CFEL-Coherent X-Ray Imaging (FS-CFEL-1)

Research Program(s):

1. 633 - Life Sciences – Building Blocks of Life: Structure and Function (POF4-633) (POF4-633)
2. DFG project 390715994 - EXC 2056: CUI: Advanced Imaging of Matter (390715994) (390715994)

Experiment(s):

1. Experiments at CFEL
2. MID: Materials Imaging and Dynamics (SASE2)
3. Measurement at external facility

Appears in the scientific report 2023

---

Database coverage:
;

---