The University of Southampton

Ultrafast Laser X-Ray

Working in collaboration with other departments within the university, the group are developing femtosecond laser-based soft X-ray sources. 

The X-ray source can be used for high-resolution imaging, with the aim of probing the shape of single proteins and other nanoscale objects, and also for ultrafast time-resolved measurements, as the pulses are less that 10fs long. The source is based on a high-power femtosecond laser, which produces ultrafast soft X-ray pulses via high harmonic generation.

Group webpage


X-ray Image reconstruction using coherent diffraction and machine learning

Supervisor:  Dr W.S. Brocklesby 
Co-Supervisor:  Professor J Frey

The technique of coherent imaging involves illuminating an object with coherent light, and recording its diffraction pattern. The reverse process, reconstructing the image from the diffraction pattern, is not immediately possible because the phase of the diffraction pattern is lost in recording on a camera. Over the last few years, mathematical techniques have been successful in using algorithms to reconstruct images, making imaging possible even in wavelength regimes where no good lenses exist, such as the X-ray region. In Southampton we have been successful in developing soft X-ray imaging using ultrafast laser-generated x-rays, and demonstrated the first imaging of real biological samples using this technique.

Recently, the idea of combining algorithmic reconstruction techniques with techniques based on machine learning has been proposed. Many of the problems of reconstruction are similar to those addressed by machine learning, particularly using convolutional neural networks. Significant research expertise in machine learning exists in Southampton, and this project will involve working with members of the EPSRC-funded AI for Scientific Discovery network.  The project aims to advance soft X-ray imaging by developing and applying new phase reconstruction techniques to ultrafast x-ray imaging based on applications of neural networks. 

Femtosecond coherent X-ray imaging of biological structures

Supervisor:  W.S. Brocklesby 
Co-Supervisor:  John Chad/Katherine Deinhart, IfLS

Developments in femtosecond laser technology have made the generation of ultrafast X-ray pulses using nonlinear optics a viable alternative to X-ray sources such as synchrotrons. A coherent lab-based X-ray source allows the development of new techniques such as coherent X-ray imaging, which combines the availability of new X-ray sources with new developments in computational algorithms to create images without the use of lenses. Our group has developed a femtosecond X-ray microscope based on radiation from high harmonic generation, and we have recently demonstrated its first ever application to imaging of biological structures. This project will look at the development of coherent X-ray microscopy of biological samples, such as biofilms or neurons, where the high resolution and new contrast mechanisms will be able to find new solutions to biological imaging problems. The project will combine high-power femtosecond lasers with cutting-edge computational algorithm design.


We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.