Machine learning reveals hidden components of X-ray pulses

Machine learning reveals hidden components of X-ray pulses

An X-ray pulse (white line) is made up of “real” and “imaginary” components (red and blue dashes), which define quantum effects. A neural network analyzes low-resolution measurements (black shadow) to reveal the high-resolution pulse and its components. Credit: SLAC National Accelerator Laboratory

Ultrafast pulses from X-ray lasers reveal how atoms move on femtosecond time scales. That’s a quadrillionth of a second. However, measuring the properties of the pulses themselves is challenging. Although determining the maximum pulse strength or “amplitude” is easy, the time at which the pulse reaches its maximum or “phase” is often hidden. A new study trains neural networks to analyze the pulse to reveal these hidden subcomponents. Physicists also call these subcomponents “real” and “imaginary”. Starting from low-resolution measurements, neural networks reveal finer details with each pulse and can analyze pulses millions of times faster than previous methods.

The new analysis method is up to three times more accurate and millions of times faster than existing methods. Knowing the components of each radiograph pulse results in better, clearer data. This will expand the science possible using ultrafast X-ray lasers, including fundamental research in chemistry, physics and material science and applications in areas such as quantum computing. For example, the additional pulse information could enable simpler and higher-resolution time-resolved experiments, reveal new areas of physics, and open the door to new studies of quantum mechanics. The neural network approach used here could also have wide applications in X-ray and accelerator science, including studying the shape of proteins or the properties of an electron beam.

Characterization of system dynamics is an important application for X-ray free-electron lasers (XFELs), but measuring the time-domain properties of the X-ray pulses used in these experiments has been a long-standing challenge. Diagnosing the properties of each individual XFEL pulse may enable a new class of simpler and potentially higher resolution dynamical experiments. This research by scientists at SLAC National Accelerator Laboratory and Deutsches Elektronen-Synchrotron is a step toward that goal. The new approach is training Neural Networks, a form of machine learning, to combine low-resolution measurements in both the time and frequency domains and recover the properties of X-ray pulses at high resolution. The physics-based neural network architecture can be trained directly on unlabeled experimental data and is fast enough for real-time analysis of the new generation of megahertz XFELs. Critically, the method also recovers the phase, opening the door to XFEL coherent control experiments, modeling the complex motion of electrons in molecules and condensed matter systems.

The study was published in Optics Express.


Machine learning is paving the way for smarter particle accelerators


More info:
D. Ratner et al, Phase and amplitude recovery of X-ray FEL pulses using neural networks and differentiable models, Optics Express (2021). DOI: 10.1364/OE.432488

Quote: Machine learning reveals hidden components of X-ray pulses (2022, August 5), retrieved August 6, 2022, from https://phys.org/news/2022-08-machine-reveals-hidden-components-x- ray.html

This document is subject to copyright. Except for any fair dealing for the purposes of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.

Related Posts

Leave a Reply

Your email address will not be published.