The UK XFEL (United Kingdom X-ray Free Electron Laser) is a proposed accelerator facility that will use predominantly superconducting RF structures to accelerate and manipulate the transiting electron bunches. Deviations from design longitudinal length of these Niobium RF cavities, or cells within, caused by errors in manufacturing, fabrication tolerances, assembly and installation can lead to certain modes becoming frequency degenerate. This gives rise to mode mixing, whereby the component modes combine to form additive and subtractive field distributions. Here we look at the mixing of transverse magnetic (TM) modes and locate problematic combinations where field strength enhancements can lead to increased on-axis loss factors and dipole kick factors, both of which can have adverse effects on the beam emittance and energy.

The Ghost Collider is a proposal for a 550 GeV center-of-mass (275 GeV per beam) linear collider with four interaction regions, each with the design luminosity. The primary innovation is the use of “ghost bunches” containing equal numbers of electrons and positrons, therefore being electrically neutral. In the linacs, energy is transferred between electrons and positrons in the same bunch, decelerating one type of particle and using the energy to accelerate the other; a new class of Energy Recovery Linac. At the interaction points (IPs), collisions between two neutral ghost bunches occur. Historically this approach has been referred to as "charge compensation of beam-beam". To avoid instabilities, round beams with small disruption parameter are arranged at the IPs, ensuring particles and their energy can be recycled with minimal loss. Four “serial IPs” are incorporated, where chromatic errors produced in one IP are canceled in the following IP. All interaction points have the nominal luminosity per IP of $2.8 \times 10^{34}$ cm$^{-2}$ s$^{-1}$ for a facility luminosity of $11 \times 10^{34}$ cm$^{-2}$ s$^{-1}$ @ 100 MW total electrical power for SR replacement, linac RF, cryogenic and damping ring systems. The result is a totally original concept for an electron-positron collider.

A 550 GeV centre-of-mass Higgs factory is presented, the Linear Ghost Collider (LGC). Acceleration and deceleration are performed within SRF linacs where the bunches trans- ported are net neutral, comprising equal charges of electrons and positrons, termed ghost bunches. Within these, one charge partner accelerates, and the other decelerates. En- ergy is recovered after a collision, and all particles recycled. An accompanying paper - Ghost Collider (GC) - introduces this concept. LGC comprises an alternative configuration to GC that eliminates turn-around arcs. This enables a large re- duction in energy lost to synchrotron radiation, and in bunch degradation, in comparison to GC. Two variants of LGC are presented: a pulsed version realisable with proven SRF technology with instantaneous luminosity $35 \times 10^{34}$ cm$^{−2}$ s$^{−1}$ @ 100 MW electrical power; and a continuous-wave (CW) version based on expected parameters for thin-film Nb$_3$Sn-on-copper SRF technology, capable of $348 × 10^{34}$ cm$^{−2}$ s$^{−1}$ @ 160 MW electrical power.

The GHOST collider final focus system is a pure quadrupole-drift beamline targeting $\beta^* = 2$ mm at four serial interaction points, with peak $\beta$-functions reaching ${\sim}65$ km in the final triplet. Beam tracking simulations reveal that a nominal bunch develops a pronounced C-shape in longitudinal phase space at IP1, where $\sigma_z$ grows from $0.15$ mm to ${\sim}2.3$ mm—a factor of ${\sim}15\times$ increase that directly reduces luminosity. This mechanism is identified as chromatically amplified betatron path length, where off-momentum particles acquire enlarged betatron amplitudes in the high-$\beta$ final triplet, generating excess path length via the geometric $\Delta z = - \frac{1}{2} \int (x'^2 + y'^2) \, ds$ integral. Within the monoenergetic Balandin framework, $\epsilon^2\xi_2$ with $\xi = - \frac{1}{2} \int \gamma \, ds$ dominates $\epsilon^2W^2$ by over four million times; the full beam $\sigma_z$ is a further factor of ${\sim}27$ larger, driven by $\sigma_\delta$-induced chromatic amplitude growth. Phase-advance scans confirm that $\sigma_z$ is insensitive to the apochromatic ($W \approx 0$) condition, and when combined with the geometric hourglass effect, this distortion poses a significant challenge to maximizing luminosity within the current lattice design.

The UK XFEL (United Kingdom X-ray Free Electron Laser) is a proposed accelerator facility that will use pre- dominantly superconducting RF structures to accelerate and manipulate the transiting electron bunches. Deviations from design longitudinal length of these Niobium RF cavities, or cells within, caused by errors in manufacturing, fabrica- tion tolerances, assembly and installation can lead to modes becoming frequency degenerate. This gives rise to mode mixing, whereby the component modes combine to form ad- ditive and subtractive field distributions. Here we look at the mixing of transverse magnetic (TM) modes and locate prob- lematic combinations where field strength enhancements can lead to increased on-axis loss factors and dipole kick factors, both of which can have adverse effects on the beam emittance and energy.

The spectral brightness of a free-electron laser (FELs) is strongly dependent on the properties of the electron beam used to generate the radiation. Bunches used to drive FELs must have high currents, which are produced through longitudinal bunch compression. However, collective effects (such as coherent synchrotron radiation) degrade the emittance and energy spread during the bunch compression process, which in turn reduces the spectral brightness. A compression scheme for an FEL must be designed to reduce the impact of collective effects on the electron distribution in order to generate high spectral brightness FEL pulses. In this paper we compare commonly used four-dipole chicanes to two alternative bunch compressor configurations that are designed to mitigate coherent synchrotron radiation. We show that compression can be achieved with much reduced emittance degradation compared to 4-dipole chicanes in two cases: arc compression of positively chirped bunches, and 5-dipole compression of negatively chirped bunches. Each is suited to different FEL schemes.

The UK-XFEL spreader will distribute 1 MHz bunches between ten undulator beamlines. Each of the different paths through the spreader must transport an electron bunch to the undulator with minimal increase in the emittance and energy spread, whilst also conforming to various geometry constraints. To minimise the energy spread/emittance growth, the design was optimised to mitigate the effect of coherent synchrotron radiation (CSR). This was done by control of the phase advance between repeated sections, and by cancelling CSR kicks using a single kick CSR model. Due to the number of constraints that needed to be adhered to, an optimisation tool was developed to aid in the design of the optics. This paper will give an overview of the optimisation tool, show the design of different sections of the spreader, and then show the results of beam tracking.

The baseline design for the proposed UK XFEL (United Kingdom X-ray Free Electron Laser) facility includes main linear accelerating linacs, which are comprised of 600 L-band 9-cell TESLA style superconducting (SC) RF cavities, which will accelerate a 1 MHz repetition-rate irregularly spaced composite electron beam comprised of varying bunch charges up to an energies of 8 GeV. Here the TESLA cavities are simulated and the dipole modes excited by the beam are characterised in order to calculate the long-range transverse wakefields (LRTWs). We track the particles through 1 km under the influence of these LRTWs using the PLACET code and evaluate the transverse emittance dilution of the electron beam.

The beam arrival time jitter is an important parameter for many advanced electron accelerators, including next-generation XFEL drivers. Accurate jitter modelling is particularly important during the design of a new facility, where it is often used to determine the tolerances of the RF sources and photoinjector laser. Conventionally, jitter is modeled using computationally intensive start-to-end simulations; in this approach, the entire accelerator is simulated many times, while the parameter associated with each jitter source is varied within its expected tolerance. This approach is time-consuming and scales poorly with the size and complexity of the accelerator. Semi-analytic jitter models therefore have significant advantages, owing to their speed and greater insight into the underlying physics. In this contribution, we propose a simple matrix-based model for the arrival time jitter in a linear accelerator with an arbitrary layout. We validate the model against simulations of CLARA at Daresbury Laboratory, and use it to explore the jitter tolerances of different operating modes.

The UK XFEL conceptual design and options analysis (CDOA) process has recently been completed. The UK XFEL design uses a superconducting linac operating with up to a 1 MHz repetition rate electron beam and demands beyond state of the art electron beam quality. This leads to strict requirements on the injector. The requirement for MHz repetition rates restricts the electron source technology to a number of operations: DC guns, VHF guns, L-band SRF guns, VHF SRF guns and DC-SRF guns. Two of these injector technologies were investigated in detail: a VHF gun and a high field L-band RF gun. In this proceeding the simulated beam dynamics performance and the technical readiness of the options will be compared and a number of conclusions will be drawn for the UK XFEL project.

UK scientists are amongst the leading users of pioneering X-ray free-electron laser (XFEL) facilities, which enable revolutionary atomic-level imaging and ultrafast time resolution far beyond other methods. Looking to the future for the UK community, the UK XFEL project has developed a science and technology case and a conceptual design for a next-generation XFEL that can utilise maturing technologies to deliver vast increases in research productivity through enhanced data rates, throughput, and x-ray quality. To achieve this, the design couples extremely high repetition rate operation (100 kHz – 1 MHz) developments of a range of accelerator, laser and end station technologies with the emergence of artificial intelligence and significantly increased multiplexing capabilities. Such a facility would drive deep and wide-ranging impact across all areas of science and technology and major sectors of the economy. The project has identified and analysed options to realise these next-generation capabilities through either hosting a new facility or co-developing upgrades at existing international projects. This contribution provides an overview of the project and its next steps.