Diamond Annual Review 2023/24

58 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 3 / 2 4 a wet laboratory and the other a dry one. They house two glove boxes, a high temperature furnace, an anaerobic Coy chamber, microscopes, centrifuges and other standard lab equipment. Furthermore, there is a counting room with a Gamma spectrometer and liquid scintillation counter. The Laboratory building also houses a storage room for storing active materials in a safe and secure manner. In 2022 the lab welcomed its first users and since January 2023 has been accessed by 10 user groups which typically have five to six members each. Usage so far has predominantly been for the preparation of samples to take to the synchrotron beamlines and much of this inside a glovebox. Examples of glovebox use include the design and testing of a custom-made flange and transport system for use on B18 and preparing samples to study rare actinide- actinide bonding of trivalent thorium. Another noteworthy study focussed on the corrosion behaviour of uranium metal as it would be found in legacy UK storage ponds. Ada Lovelace Centre The Ada Lovelace Centre is a centre of expertise in research software engineering and data management. It was setup to maximise the science impact of the large scale STFC national facilities, Diamond Light Source, the Central Laser Facility, and the ISIS Neutron and Muon Source, by transforming and optimising the interpretation and utilisation of the data from the facilities. At Diamond, the Scientific Software, Controls and Computation (SSCC) group works on new software interfaces to control the experiments, capture and analyse data resulting from the experimental work at the facility. The Ada Lovelace Centre is complimentary to these activities, providing expertise in Materials Modelling, Computational Maths, AI, Computational Engineering and Computational Biology as well as extensive expertise in research software engineering and platforms and services for large scale data management and computing. The Ada Lovelace Centre previously funded responsive projects to address facilities needs but is now moving towards a programme structure to address longer term needs and strategic priorities and is working closely with Diamond to shape this new programme of work. The ALC programme will allow for the development of integrated, cross- disciplinaryactivities,pullingexpertisefromseveralareastotacklecritical science oroperationalproblems togenerateimpact.Asanexample,ALCareworkingwith the Imaginggroupalongwithmembersof theALCAI,mathsand imaginggroups to develop more robust solutions for ptychography experiments at Diamond. Many of the challenges experienced across ISIS, CLF and Diamond, such as providing tools and platforms for imaging, scattering or modelling, are similar and would benefit for greater collaboration The ALC is also seeking to increase coordination of activities across the site to prevent duplication, ensure wider knowledge of existing solutions and more effective use of resources. To drive research and development in scientific computing the ALC supports several PhD Studentships at Diamond. Studentships funded to date cover areas such as the machine optimisation, theoretical simulations related to materials, and new computational imaging methods. National Synchrotron Light Source-II collaboration on Bluesky Bluesky is a library for experiment control and collection of scientific data and metadata. It emphasises the live streaming of data, reusable experiment control procedures, experiment suspend/pause/rewind, adaptive experiment control, pluggable I/O, customisability and integration with scientific Python. The Bluesky library is supported by an international light source community Data acquisition for Diamond-II will be managed by Diamond’s newAthena platform, a service-based experiment orchestration system built on top of NSLS-II’s Python-based Bluesky/Ophyd data collection framework which can easily be customised to the needs of individual beamlines yet also offer reuse of software components for common beamline capabilities. In April 2024, Diamond hosted the Bluesky Hackathon event aimed at developers of the framework to come together and work on tasks and issues with the software alongside the NSLS-II. Hackathons (also known as codingmarathons or code sprints) are popular in many open source projects as away of making progress on certain development efforts. At the Bluesky Hackathon, attendees worked on a large number of open issues in the Bluesky software collaboration, which gave Diamond engineers the opportunity to become more familiar with Bluesky and foster international collaboration with NSLS-II. LEAPS Diamond is part of the League of European Accelerator-based Photon Sources (LEAPS), a strategic consortium initiated by the directors of the synchrotron radiation and free electron laser (FEL) user facilities in Europe. Its primary aim is to ensure and promote the quality and impact of fundamental, applied and industrial research carried out at each facility to the greater benefit of European science and society. As part of its remit, the collaboration aims to strengthen interactions with industry, to exploit more fully the potential of synchrotron and FEL facilities for industrial research and to develop enabling technology. This in turn will have economic impacts while solving pressing problems facing humanity. In November 2023, Diamond collaborated with STFC and LEAPS to deliver a supplier information day. The event gave delegates the occasion to explore and discuss emerging opportunities at facilities across Europe. AI Collaborations In the past year, the diagnostics group have taken tentative steps towards controlling the stability of the synchrotron electron beam using lessons learnt and code shared with Advanced Light Source at the US Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab). Both synchrotrons deliver different types of light for measurements and experimentation. Tweaking of the electron beam position can produce X-ray beam fluctuations on the beamlines, which can create challenges in certain experiments, especially during events such as top-up. The project between Berkeley Lab and UC Berkeley, started in 2019, sought to use Machine Learning (ML) algorithms to improve the stability of the beam’s size and orbit to largely cancel out electron beam instabilities. Applying Machine Learning in this form made the overall system a form of Artificial Intelligence (AI), where computer systems analysed a large set of data to build algorithms that solve complex problems. The Machine Learning algorithm used at the ALS was a neural network, such algorithms are designed to recognise patterns in data and‘learn’from this in order to address given tasks. Their project has had a lasting impact on performance enhancements of the facility. The electron beam data, including the positions of the storage ring magnets, their steering parameters and information on insertion devices and bending magnets, is fed into the neural network which recognises patterns in the data and identifies how different device parameters affect the width of the electron beam. The algorithm then recommends adjustments to optimise beam parameters. Diamond has also recently welcomed visitors from the US Department of Energy (DoE) facilities to conduct an experiment using shared tomographic segmentation algorithms using a shared platform, MLExchange, on the DIAD beamline. One element of this endeavour is to grow our AI and ML links, Integrated Facilities and Collaborations