In October 2022, the UK XFEL project entered a new phase to explore how best to deliver the advanced XFEL capabilities identified in the project's Science Case. This phase includes developing a conceptual design for a unique new machine to fulfil the required capabilities and more. It also examines the possibility of investment opportunities at existing XFELs to deliver the same aims, and a comparison of the various options will be made. The desired next-generation capabilities include transform limited operation across the entire X-ray range with pulse durations ranging from 100 as to 100 fs; evenly spaced high repetition rate pulses for enhanced data acquisition rates; optimised multi-colour FEL pulse delivery and a full array of synchronised sources (XUV-THz sources, electron beams and high power/high energy lasers). The project also incorporates sustainability as a key criteria. This contribution gives an overview of progress to date and future plans.

Plasma based accelerators have achieved beams with multi-GeV energy, percent-level energy spread, micron emittance and stability over a full day however it remains a challenge to generate beams with all these properties simultaneously. External injection of a beam from a RF linac into a plasma-based accelerator holds the prospect of improving the beams from plasma accelerators by combining their high gradient with the high quality of RF accelerators. If the beam is matched to the plasma, then the initial beam emittance and energy spread can be preserved. This technique can also be used to investigate the staging of multiple plasma accelerator stages in a controlled manner by providing a stable beam to the plasma target being tested. We present results of an experiment performed at the CLARA accelerator in the UK investigating the external injection of the 35 MeV, 20 pC electron beam containing from the linac into a laser driven plasma wave with accelerating gradient ~100 MV/m. The beam length was larger than the plasma wavelength resulting in electrons experiencing both positive and negative accelerating fields across several plasma buckets which broadens the energy spectrum rather than a pure energy gain. This proof-of-principle experiment is part of preparatory work aiming towards acceleration of electron beams with near perfect beam quality preservation. Simulations are also presented for beam parameters after a scheduled upgrade to CLARRA which inform future experiments.

Electro-optic diagnostics are able to non-destructively resolve the longitudinal charge profile of highly relativistic bunches without complicated calibrations and ambiguous phase recovery techniques. The most implemented technique is EO spectral decoding as it is simple and reliable, and has an easy to interpret output. However, its resolution is limited to the geometric mean of the transform limited and stretched probe laser durations. Until very recently, efforts to improve on this have resulted in designs that lose the attractive properties of spectral decoding. On the CLARA accelerator at Daresbury Laboratory we have demonstrated a new EO system that exploits common-path spectral interferometry, 'EOSI', which removes the geometric mean limitation. The system was used to measure 35 MeV/c bunches live at 10 Hz, ranging from 150 pC down to 2 pC, and at a range of compressions from several ps down to ~300 fs rms. We explain the technique, describe the measurements, and outline issues and improvements. The technique differs from a spectral decoding system by only a single optical element, potentially allowing current EO systems to be upgraded.