Cavity-based X-ray free-electron lasers present a promising path toward fully coherent, high-brightness X-ray sources with enhanced stability and spectral purity. By using Bragg-reflecting crystal cavities to recirculate and amplify an X-ray seed pulse over multiple passes, CBXFELs offer the potential for orders-of-magnitude improvements in coherence and brightness compared to single-pass FELs. This talk will present an overview of the CBXFEL concept and the proof-of-principle experiment currently under development at SLAC. Recent progress will be presented, along with ongoing efforts in beam–X-ray overlap diagnostics and cavity alignment. The talk will also address the key technical challenges ahead for CBXFELs and briefly explore alternative cavity-based XFEL designs as promising paths forward.

Formation of current spike in electron bunch has direct implication for attosecond pulse generation in XFEL. In this paper, we present start-to-end simulation for tunable, short current spike generation in the LCLS copper linac using photocathode laser shaping. Our approach uses two stacked laser pulses—a long and a short pulse—to imprint a small modulation in the electron bunch as it is created in the injector. This initial modulation is then amplified as the bunch travels through the downstream bunch compressors, ultimately forming a sharp current spike. We also discuss how different shapes of the initial laser pulses influence the final current profile and the efficiency of spike generation.

Shaping ultraviolet (UV) laser beams is critical for optimizing photoinjector performance for applications in free-electron lasers (FELs). It has been shown that a 50% truncated Gaussian beam can achieve the lowest emittance via space charge compensation at LCLS-I. However, conventional shaping techniques to prepare this beam are limited by significant power losses or are not adapted for UV light. Here we report a high-precision transverse-shaping technique based on custom fused-silica phase plates with >99 % transmission at 253 nm. This approach enables spatial beam profile tailoring and significantly enhances beam stability at the photocathode. Using IMPACT-T simulations, we predict a 33% (from 0.67um to 0.45um) reduction in normalized emittance for a 250 pC bunch at LCLS-I. Experimental implementation at FACET-II demonstrated a 37% emittance reduction (from 5.4um to 3.4um) at 1.6 nC. These results establish phase-plate beam shaping as a high-fidelity, low-loss approach for high-brightness photoinjectors. Implementation at LCLS-II which will enable stable operation at megahertz repetition rates is underway.