SLAC’s LCLS-II is pioneering high-repetition-rate attosecond X-ray science, enabling new opportunities to optimize X-ray generation by controlling the electron beam at its source—the photoinjector. LCLS-II employs a 20 ps Gaussian UV laser pulse to drive the photocathode, with an added narrow modulation to induce microbunching for extended modes.* Recent advances in laser pulse shaping and frequency upconversion now allow for more sophisticated tailoring of the electron beam at the injector. We present a novel approach using spectral amplitude and phase shaping of the IR laser, followed by dispersion-controlled nonlinear synthesis—relying on phase-modulated noncollinear sum-frequency generation—for UV upconversion.** This enables diverse UV temporal profiles, including flattop and double/triple spikes, offering new degrees of freedom for shaping. Preliminary results from LCLS-II beam time show these modulations produce effective downstream perturbations to the electron bunch at the undulators, demonstrating feasibility for programmable bunch formation. We are integrating this shaping into a start-to-end simulation framework,*** enabling digital twin modeling of the XFEL chain—from photoinjector laser to X-ray output—laying the groundwork for fully tunable, end-to-end optimized, application-specific X-ray pulses.

SLAC’s LCLS-II is pioneering high-repetition-rate attosecond X-ray science, enabling new opportunities to optimize X-ray generation by controlling the electron beam at its source—the photoinjector. LCLS-II employs a 20 ps Gaussian UV laser pulse to drive the photocathode, with an added narrow modulation to induce microbunching for extended modes.* Recent advances in laser pulse shaping and frequency upconversion now allow for more sophisticated tailoring of the electron beam at the injector. We present a novel approach using spectral amplitude and phase shaping of the IR laser, followed by dispersion-controlled nonlinear synthesis—relying on phase-modulated noncollinear sum-frequency generation—for UV upconversion.** This enables diverse UV temporal profiles, including flattop and double/triple spikes, offering new degrees of freedom for shaping. Preliminary results from LCLS-II beam time show these modulations produce effective downstream perturbations to the electron bunch at the undulators, demonstrating feasibility for programmable bunch formation. We are integrating this shaping into a start-to-end simulation framework,*** enabling digital twin modeling of the XFEL chain—from photoinjector laser to X-ray output—laying the groundwork for fully tunable, end-to-end optimized, application-specific X-ray pulses.