Inverse Compton scattering (ICS) at shallow angles can be used to tune the spectrum and increase the brightness of the emitted radiation. Here, we investigate the geometric scaling laws governing this interaction and experimentally validate them by demonstrating shallow-angle ICS of a 780 nm laser crossing a 4.7 MeV electron beam at 5.8°. Angular and spectral measurements of the emitted radiation show excellent agreement with analytical and numerical predictions. To further optimize this interaction, we consider the use of velocity-matched pulse-front tilt to decouple the total yield from the interaction bandwidth. The resulting framework is used to evaluate the performance scaling of three distinct driving accelerator architectures: a radio-frequency linac, a laser wakefield accelerator, and a storage ring synchrotron. These results establish a scalable, practical pathway for deploying next-generation, high-brightness compact light sources across diverse accelerator platforms.

A novel concept is introduced to reduce the longitudinal emittance in a low momentum compaction factor (low-alpha) storage ring, enabling the stable storage of electron bunches shorter than 100 nm. This design strategy can be applied to any quasi-isochronous storage ring to achieve very high radiation power through the longitudinal coherence of the emitted radiation. This paper addresses the nonlinear dynamics and aperture optimization for this type of storage ring. An optimal design example based on our analysis is presented. Single-particle tracking results demonstrate that an electron beam with an equilibrium root-mean-square bunch length of approximately 80 nm can be maintained in this ring, featuring a dynamic aperture larger than 1 mm. Additionally, preliminary considerations of intrabeam scattering effects in this ring are discussed.

Permanent magnet electron optics offer many advantages over electromagnets and are being increasingly used in high-energy (GeV) electron accelerator designs. Here, we identify the advantages of permanent magnet electron optics for low-energy (keV) electron accelerators. We explore the applications of a class of designs based on axially magnetized permanent magnets, which offer a variety of advantages such as short focal lengths (few mm), while also preventing apparent emittance growth resulting from starting particles in a magnetic field. The proposed design philosophy is applied to an accelerator based on the ultracold electron source. The design is shown to be “emittance preserving” even for very short focal lengths (approximately 5 mm) at an emittance level better than 1 nm rad, while the short beamline ($\approx 12\mathrm{cm}$) limits space-charge effects. Two remedies for the mitigation of typical manufacturing and alignment challenges are considered. The performance of the design (related to parasitic aberrations) is enhanced by the proposed techniques. Applications of this design philosophy can improve the performance of ultrafast electron diffraction setups with minimal manufacturing effort.

We demonstrate that graph neural network embeddings of injector configurations provide a practical operational reference space for the Continuous Electron Beam Accelerator Facility injector at Jefferson Lab. Using 137,389 snapshots spanning January 2022 through March 2023, we show that injector operation occupies a small number of persistent, well-separated neighborhoods in a 16-dimensional learned state space rather than a featureless continuum. Density-based clustering identifies ten persistent operating regimes with strong operational run alignment, and regime persistence statistics confirm that these regimes are stable over timescales of hours to weeks. Large relocations between neighborhoods are rare and episodic; 99.6% of 1-h operating windows fall within an empirically derived jitter baseline. Geometric outlier screening narrows a year-long dataset to a small set of intervals warranting operational review, and nearest-neighbor retrieval enables case-based reasoning over the historical archive. A controlled beam study validates that deliberate injector reconfiguration traces coherent, interpretable trajectories in embedding space. Together these capabilities demonstrate that machine-state embeddings support holistic operational monitoring in ways that single-channel inspection cannot.

We demonstrate that multipoint Bayesian algorithm execution (multipointBAX) can overcome fundamental computational challenges in storage ring design optimization. Dynamic aperture and momentum aperture optimization is a multipoint, multiobjective design task for storage rings, ultimately informing the flux of x-ray sources and luminosity of colliders. Current production black-box optimization methods require extensive particle-tracking simulations for each trial configuration; the high computational cost restricts the extent of the search to $\sim {10}^3$ configurations and therefore limits the quality of the final design. We remove this bottleneck using multipointBAX, which selects, simulates, and models each trial configuration at the single-particle level. We demonstrate our approach on a design for a fourth-generation light source, with neural-network-powered multipointBAX achieving equivalent Pareto front results using more than 2 orders of magnitude fewer tracking computations compared to genetic algorithms. The significant reduction in cost positions multipointBAX as a promising alternative to black-box optimization, and we anticipate multipointBAX will be instrumental in the design of future light sources, colliders, and large-scale scientific facilities.

The first experimental observation of a crabbing orbit induced by beam-beam interaction in the presence of a finite crossing angle in a high-energy proton beam is presented. This measurement required both the design of a dedicated experiment in the large hadron collider (LHC) and careful calibration and optimization of the beam instrumentation to produce and detect such a subtle effect. To also measure the dependence of this effect on the crossing angle, this parameter is varied from positive to negative values, further showing the reproducibility of the effect. Tracking simulations were performed to corroborate the experimental results, showing good agreement with the measured crabbing amplitudes. This experiment highlights the potential of the existing wideband beam position monitors to diagnose crabbing effects, which will be crucial in the High-Luminosity LHC (HL-LHC) upgrade.

We show that plasma-accelerated electron beams enable single-cycle control of undulator radiation, including control of its polarization and time-frequency structure as described by the Wigner distribution. Using the ultrashort, high-current electron bunch produced by a plasma-wakefield accelerator reported in Emma et al. Terawatt attosecond x-ray source driven by a plasma accelerator, APL Photonics 6, 076107 (2021), we consider the regime in which the resonant radiation wavelength exceeds the effective bunch length, so that emission is coherently generated within each undulator period and follows the undulator magnetic waveform. This allows subcycle control of the electric field through undulator design, effectively turning the undulator into a tunable source of waveform-shaped radiation. Using numerical simulations, we model two compact undulator configurations: a linearly tapered undulator that produces a chirped optical waveform, and an undulator that generates radiation with controlled, cycle-to-cycle polarization variation. Together, these results establish a route to subcycle waveform and polarization control of ultrashort radiation.

Fourth-generation storage-ring light sources have achieved transverse emittances approaching the diffraction limit at x-ray wavelengths, while their longitudinal coherence remains limited. Existing laser-modulation schemes can induce strong microbunching, but most storage-ring implementations produce only one useful coherent output from a given longitudinal region during each revolution, thereby underutilizing the intrinsic multiuser capability of storage rings. We propose a multiple-echo-enabled harmonic generation (multi-EEHG) scheme that applies successive excitation-echo cycles to the same longitudinal slice of a stored bunch within one revolution, enabling coherent-radiation delivery to multiple beamlines at different wavelengths. A general formulation of the n-stage EEHG bunching factor and a corresponding optimization procedure are derived. As an example, a triple-EEHG configuration is designed for the southern advanced photon source storage ring. Simulations demonstrate coherent radiation at multiple wavelengths with single-pulse photon numbers up to ${10}^9$, corresponding to an enhancement of approximately three orders of magnitude over synchrotron radiation for the same spectral bandwidth, while achieving few-meV bandwidth without a monochromator. The proposed scheme offers a scalable approach for multibeamline coherent operation in next-generation storage-ring light sources.

The use of an intense ultrashort laser pulse to induce electron polarization has been proposed previously in the literature. We used Python to reproduce the local constant crossed-field approximation to calculate the transverse polarization obtained from repeated laser interactions, starting from nonzero initial polarization, as well as the associated energy kicks. We also studied how the collision-geometry affects both the magnitude of ${\chi }_e$ and the effective field asymmetry experienced by the electrons, identifying several possible ways to generate useful asymmetry even when the laser wavelength does not admit pulse lengths that directly reach the plane-wave optimum. We show that, over multiple laser shots, lower values of the quantum efficiency parameter ${\chi }_e$ are associated with higher transverse polarization output but require a greater number of shots to reach saturation. We also show that the mean energy loss scales approximately linearly with ${\chi }_e$. Based on this framework, we developed a preliminary Better MAD (methodical accelerator design) (BMAD) laser module to simulate the full 6D dynamics in a proposed Future Circular Collider Z pole lattice and to identify the threshold ${\chi }_e$ value above which particle losses begin to occur.

The ion composition of dual-species beams in synchrotrons is typically inferred from invasive measurements performed after beam extraction. This paper introduces a complementary noninvasive method to determine the ion composition of such beams directly inside the synchrotron. The approach is applicable to low- and medium-energy ion synchrotrons and to small relative mass-to-charge ratio offsets, typically at the ${10}^{-4}$ level. The method exploits dispersive orbit offsets between the two species and corresponding frequency corrections applied by the synchrotron rf radial regulation loop. This capability is of particular interest for ongoing research on online monitoring in carbon ion beam therapy using mixed ${{}^4\text{He}}^{2+}$ and ${{}^{12}\mathrm{C}}^{6+}$ beams, which feature a relative mass-to-charge ratio offset of 0.065%. The proposed method is analytically derived and tested with particle tracking simulations using xsuite. Its applicability under realistic experimental conditions is demonstrated at the MedAustron facility using mixed ${{}^4\text{He}}^{2+}$ and ${{}^{12}\mathrm{C}}^{6+}$ beams. The results show good agreement with independent postextraction measurements.

Electron coolers with high-density electron bunches obtained via rf acceleration provide a feasible path for extending electron cooling to high energies. In such coolers, the dispersive redistribution of cooling rates between the longitudinal and transverse directions is crucial and must be rigorously calculated. This paper presents exact formulas for cooling redistribution in high-energy electron coolers. The most general formulas are derived in terms of Carlson elliptic integrals, which are further reduced to elementary functions for a key special case. These formulas enable an efficient and accurate calculation of cooling rates and were instrumental in designing the electron coolers for the Electron-Ion Collider (EIC) by enabling rapid parameter identification and optimization. Consequently, these results offer a valuable tool for the design and development of future high-energy electron coolers.

We demonstrate a new 11.424 GHz SLED-type rf pulse compressor for powering high-gradient x-band photoinjectors with pulse lengths around 20 ns. rf pulse compression provides a practical path to higher peak power at the cost of pulse length for various applications such as rf deflectors for electron beam diagnostics on free-electron lasers. Our new compact pulse compressor uses spherical cavities supporting axially symmetric TE modes which have minimal electric fields on the cavity surfaces, intended to improve high-power robustness as compared to existing compact spherical SLEDs which use a TE dipole mode. We present the design of this pulse compressor composed of two spherical cavities and a waveguide hybrid. The two cavities and hybrid have ${\mathrm{TE}}_{01}$ circular waveguide ports. This pulse compressor was built and high power tested at SLAC. These tests demonstrated a peak power of 317 MW by compressing 52 MW, $1\mathrm{\mu }\mathrm{s}$ pulses generated by a SLAC XL-4 klystron representing, to our knowledge, the highest peak power achieved from a single-stage compact SLED rf pulse compressor at x-band. The full width at half maximum of this compressed pulse was $(27.1\pm 1.1)\mathrm{ns}$. We conjecture that this development demonstrates a viable route to reaching the high-gradient, short pulse regime for accelerating structures and rf photoinjectors.

Beam breakup (BBU) instability limits the performance of energy recovery linacs (ERLs). This study presents a new approach to estimating the BBU threshold current using a physics-inspired neural network. The application of neural networks to the highly nonlinear beam dynamics of ERLs opens a new paradigm for the development of efficient future particle accelerators. It is found that figure-of-merit-based data can predict the threshold current with higher accuracy than directly feeding the proposed network with raw parameter-based data. This study investigates how embedding such a physically meaningful feature influences predictive performance and generalization. In this work, the effect of the variation in higher-order modes (HOM) parameters due to the geometrical tolerances in the rf cavity fabrication on the BBU threshold current is analyzed with the proposed multilayered bidirectional long short-term memory (BiLSTM) neural network.

A compact, in-vacuum switchable beam dump is proposed as a first-of-a-kind engineering solution to improve operational flexibility in high-energy accelerator facilities. The design integrates beam-pass and beam-dump modes into a single wall-embedded unit, eliminating the need for external shielding and movable components outside the vacuum system. The concept is specifically developed for the superconducting fragment separator (Super-FRS) at Facility for Antiproton and Ion Research, making use of an existing opening in a modular radiation-shielding wall between experimental areas. It employs synchronous rotation to switch between an unobstructed vacuum beamline and a fully shielded beam-dump configuration. This architecture enables personnel access to downstream experimental areas while beam delivery continues in the upstream section. Two absorber configurations were investigated: an all-iron design and a mixed configuration incorporating the high-density tungsten alloy Densimet. Iron provides cost-effective bulk shielding and strong attenuation of secondary gamma radiation, whereas Densimet enables compact energy deposition with superior stopping power. The required thickness of additional downstream boron carbide shielding to adequately suppress slow neutrons differs between the two options. FLUKA Monte Carlo simulations confirm that both configurations satisfy the stringent personnel safety limit of $<3\mathrm{\mu }\mathrm{Sv}/\mathrm{h}$ in all relevant scenarios. Complementary vacuum simulations demonstrate that the beamline pressure remains below $5\times {10}^{-6}\mathrm{mbar}$ during operation. The results indicate that the proposed switchable beam dump offers a viable path to enhance facility availability and throughput at Super-FRS. The compact, modular concept is scalable and may be adapted to other high-energy accelerator facilities facing similar space and access constraints.

Coherent terahertz (THz) sources based on storage rings offer a compelling pathway to high-average-power radiation, leveraging the high revolution frequency of the electron beam. However, current achievable repetition rates are severely constrained by the radiation damping required to dissipate phase-space perturbations introduced during generation. To address this challenge, we propose a novel scheme incorporating an active demodulation mechanism. This method utilizes a phase-reversed laser interaction to actively compensate for the laser-induced energy modulation immediately following the radiation emission. By rapidly restoring the beam quality rather than waiting for natural radiation damping, the recovery time is significantly shortened. Simulations based on HLS-II, combined with an analytical damping estimate, suggest the feasibility of coherent 10 THz generation at repetition rates on the order of 100 kHz, corresponding to an average power of 6.2 mW.

In SuperKEKB, mitigation of sudden beam losses (SBLs) is essential to improve the luminosity. SBL refers to the rapid loss of a substantial portion of the stored beam current within a few revolutions. During SuperKEKB operation from 2019 to 2024, over 200 SBL events were observed, in which a beam abort with 5%–40% beam loss occurred within 2–3 turns, making high beam current operation difficult. The frequency of SBLs was very high, especially in the low-energy ring (LER). The occurrence of SBLs was highly concentrated during beam-current upgrade phases, presenting a severe operational bottleneck. While SBLs were rare during stable operations at lower currents, the frequency escalated to more than five events per day when operating at higher beam currents. These frequent events triggered intense beam losses, resulting in 11 quenches of the final-focus superconducting quadrupole magnets (QCS) within a period of less than three weeks in December 2024. Furthermore, SBL events could damage the collimator heads and the Belle II detector, causing devastating impacts on operation. SBLs occurred under both noncollisional and collisional conditions, but only under circumstances involving a small beam size at the interaction point and a high beam current. Most SBL events were accompanied by vacuum pressure bursts, primarily in the wiggler section of the LER, and an increase in beam size was also observed during these events. To investigate this phenomenon, SBL was artificially reproduced during beam operation and subsequently discharging the clearing electrodes. Inspecting the wiggler beam pipe, we found black stains, and removal of these stains led to a significant reduction in the frequency of SBLs. The black stains we discovered were identified as degraded vacuum sealant through observations using synchrotron radiation. This paper reports the results of an artificial reproduction test of the SBL phenomenon (knocking on the beam pipe, discharge inside the beam pipe), and an investigation of these black stains, including their composition and formation mechanism.

A 300–600 MeV high-intensity scaling fixed-field alternating gradient (FFA) facility is proposed as a multipurpose proton driver for the China Spallation Neutron Source (CSNS) Phase-II upgrade. To identify a viable lattice configuration under strict engineering constraints, a statistical analysis methodology based on large-sample ray-tracing simulations comprising over 360,000 parameter configurations is applied. A two-stage scanning pipeline evaluates the topological limits of both FDF triplet and FD doublet structures under realistic multisource error budgets. Factoring in phase-space geometry, error tolerance, and facility integration, the findings indicate that a spiral FD doublet lattice with a super-periodicity of $N=16$ is preferred as the structural baseline; it avoids parameter bottlenecks and yields a large, resonance-sparse stable region. To characterize the coupled high-dimensional dynamics quantitatively, a machine learning feature attribution framework is applied. The framework identifies the spiral edge angle and the full-ring vertical tune as the primary drivers of beam survivability, and maps the underlying resonance stopbands from tracking data alone. Based on these results, a baseline configuration is established with a wide dynamic tune allowance that maintains long-term beam survivability within the physical and engineering boundaries of the facility.

The design and optimization of a broadband electron gun with a semi-immersed cathode for an Electron Beam Ion Source (EBIS) operating with a 5 T solenoidal magnetic field is presented. The work contributes to the development of high-performance, tunable EBIS for next-generation all-linac medical accelerators for cancer therapy, as well as other applications in accelerator physics. The system is primarily designed to meet the requirements for the production of fully stripped carbon ions. An acceptance criterion for the injection of the electron beam into a full magnetic field is derived and used to validate the design. The gun position is optimized for the transmission of an electron beam with a current of 2.0 A and a current density of $1000\mathrm{A}/{\mathrm{cm}}^2$ inside the solenoidal field. The operating range of the electron beam can be broadened from 0.2 to 3 A by using a nonadiabatic magnetic field on the beam path close to the gun, which transfers the transverse component of the electron energy into the longitudinal direction and therefore reduces the amplitude of the beam scalloping. The current density can be adjusted by the use of another coil surrounding the cathode.

The physics and ring design of the storage ring proton electric dipole moment (sr-pEDM) experiment, including the symmetries used to mitigate first-order and second-order systematic effects, are described in [Z. Omarov et al. Comprehensive symmetric-hybrid ring design for a proton edm experiment at below ${10}^{-29}\mathrm{e}·\mathrm{cm}$, Phys. Rev. D 105, 032001 (2022)]. The sr-pEDM experiment leverages a novel symmetric-hybrid, frozen-spin storage ring design using radial electric bending and alternate magnetic focusing to minimize systematic errors, enabling precise measurement of the vertical rotation of the polarization of polarized stored proton beams. In this work, we present a more systematic and detailed analysis of second-order effects, focusing in particular on the interplay between stray radial magnetic fields and electric quadrupole and sextupole fields. We demonstrate how the symmetries of the symmetric-hybrid ring design, including beam reversal and magnetic quadrupole polarity switching, can be used to suppress these effects to below the targeted sensitivity of the sr-pEDM experiment.

The x-ray free-electron laser (XFEL) facility SACLA uses a thermionic pulsed electron gun and a beam chopper to produce a low-emittance, short-bunch beam. Recently, a new type of fast pulse generator, which utilized wideband rf amplifiers and a new chopper chamber with high-impedance strip-line electrodes, was developed to reliably and accurately cut out a short bunch from the long-pulsed beam emitted from a single-crystal ${\text{CeB}}_6$ hot cathode. Using these apparatuses, we succeeded in forming nanosecond multibunch structure in the thermionic electron beam and generating a double-bunch XFEL pulse with a total pulse energy of 1 mJ at 10 keV. The results will provide new possibilities for XFEL beams: higher intensity, higher coherence, and higher pulse repetition rate.