In this paper, a cathode laser pulse shaper at 515 nm is presented that will be used for emittance optimizations. In case alkali antimonide photocathodes are used, the shaped green pulses can be applied directly for photoemission while Cs₂Te photocathodes requires second harmonic generation to provide UV laser pulses. Recent tests of CsK₂Sb photocathodes in the high gradient RF gun at PITZ are first steps for the future usage of green laser pulses, which would simplify the requirements for the photocathode laser system, especially for CW operation cases envisioned in future. As long the alkali antimonide photocathodes are not in regular use yet, the laser pulses need to be converted into the UV. The green pulse shaper still simplifies the laser system since two conversion stages from IR to green to UV were needed in the past, which dilutes the quality of the shaped laser pulses. In this paper, a pulse shaper for direct generation of 515nm 3D ellipsoidal pulses is presented that is expected to further improve the beam emittance generated by ellipsoidal laser shaping.

The slice energy spread of the electron beam is one of the key parameters for high performance of linac-driven free electron lasers (FELs). The simulated uncorrelated energy spread in modern XFEL photoinjectors with beam energies of many tens of MeV is on the order of a few keV or even less. Thus, accurate measurement of the slice energy spread is not trivial. Two recent studies on high energy (>100 MeV) photoinjectors at SwissFEL and European XFEL have reported much higher slice energy spread than expected at their XFEL working points (200 – 250 pC). A new method for measuring slice energy spread at a lower beam energy (∼20 MeV) is proposed and demonstrated at the Photo Injector Test facility at DESY Zeuthen (PITZ). The contribution will summarize previous results obtained on high energy injectors and then review the details of the technique used at PITZ as well as the experimental results for 250 pC, which are considerably lower than the results measured at high energy injectors.

A new generation of high-gradient normal conducting 1.3 GHz RF gun with 1% duty factor was developed to provide a high-quality electron source for superconducting linac driven free-electron lasers like FLASH and European XFEL. Compared to the Gun4 series, Gun5 aims for a ~50% longer RF pulse length (RF pulse duration of up to 1 ms at 10 Hz repetition rate) combined with high gradients (up to ~60 MV/m at the cathode). In addition to the improved cell geometry and cooling concept, the new cavity is equipped with an RF probe to measure and control the amplitude and phase of the RF field inside the gun. The first characterization of Gun5.1 included measurements of RF amplitude and phase stability (pulse-to-pulse and along 1 ms RF pulse). The dark current was measured at various peak power levels. The results of this characterization will be reported.

Multi-alkali antimonide photocathodes can have high quantum efficiency similar as UV sensitive (Cs2Te) photocathodes, but with the advantages of photoemission sensitivity in the visible region of the light spectrum and a significant reduction in the mean transverse energy of photoelectrons. A batch of three KCs2Sb photocathodes was grown on molybdenum substrates via a sequential deposition method in a new preparation system at INFN LASA. Afterwards, the cathodes were successfully tested in the high gradient RF gun at PITZ. This contribution summarizes the experimental results obtained in both the preparation chamber and the RF gun. Based on those findings, we are now optimizing the recipe of KCs2Sb and NaKSb(Cs) photocathodes for lower field emission and longer lifetime, and the measurements for the latest photocathodes with the improved recipe are also presented.

A numerical study is carried out on the quality of the electron bunch train produced from a photoinjector based on a frequency-detuning dependent gun coupler kick. The impact of the kick on the emittance of the bunch train is modelled via three-dimensional electromagnetic field maps calculated at detuned frequencies of the gun cavity within long radio-frequency pulses. Beam dynamics simulations are performed in the so-called frequency-detuning regime. Preliminary results are presented and discussed.

The photoinjectors of FLASH at DESY (Hamburg, Germany) and the European XFEL are operated by laser driven RF-guns. In both facilities cesium telluride photocathodes are successfully used since several years. We present recent data on the lifetime, quantum efficiency (QE), and dark current of the photocathodes currently in operation. In addition we present recent design changes in the photocathode transfer systems in order to further improve the cathode handling.

ERMI is the seeded Free Electron Laser (FEL) user facility at Elettra laboratory in Trieste, operating in the VUV to soft X-rays spectral range. In order to extend the FEL spectral range to shorter wavelengths, an upgrade plan for increasing the Linac energy from 1.5 GeV to 2.0 GeV is actually going on. After successful testing of the short prototype of new high gradient S-band accelerating structure up to an accelerating gradient of 40 MV/m, a full length 3.0 m HG structure has been built in collaboration with Paul Scherrer Institute (PSI). In the first step, two such new structures would be installed in place of S0a and one deflector at K15 increasing the beam energy to 1.7 GeV. In the next phase 14 new HG structures would replace the present Backward Travelling Wave sections reaching to the final goal of 2.0 GeV. Currently first 3.0 m HG structure is under conditioning and high power testing at Cavity Test Facility of Elettra. In this paper we report the low power measurement results as well as conditioning results of 3.0 m HG structure.

The biggest benefit of DC photoelectron-gun driven by the sub-picosecond laser is that such type of guns can be operated with the current density much higher than the Child’s low limitation. We demonstrated 0.3 nC bunch generation by irradiating a 100 fs Ti:sapphire laser focused to 0.1 square-cm area onto a tangsten photocathode installed in a diode type 40 kV DC gun. The drawback is the strong Coulomb repulsive force by which electrons may suffer the emittance degradation in the vicinity of the cathode. To reduce the repulsive force at the cathode surface, we are trying to generate a “sheet-like” photoelectron bunch. In our experiments, electron bunches are generated by irradiating the laser pulse shaped in an ellipse onto the photocathode. The ellipticity is set in the range of 0.03-0.05 while the most of sheet-beam experiments were conducted with the ellipticity about 0.1. The smaller the ellipcity, the longer the circumference; this may reduce the radial electric field on the electron bunch side-wall. Moreover the electron bunch shape is rather a “line” than a “sheet” due to the short duration of the drive laser pulse. We conducted a preliminaly experiment and observed that the elliptical photo-electron bunch had much larger divergence angle in the minor axis direction. In the presentation, experimental results, the numerical simulation on the particle motion and the design of the sheet-photo-electron DC gun will be discussed.

Next-generation electron photoinjector accelerators, such as the LCLS-II photoinjector, have increasingly tight requirements on the excitation lasers, often calling for tens of picosecond, temporally flat-top, ultraviolet (UV) pulse trains to be delivered at up to 1 MHz*. We present an experimental demonstration of temporal pulse shaping for the LCLS-II photoinjector laser resulting in temporally flat-top pulses with 24 ps durations. Our technique is a non-colinear sum frequency generation scheme wherein two identical infrared optical pulses are imparted with equal and opposite amounts of spectral dispersion. The mixing of these dispersed pulses within a thick nonlinear crystal generates a second harmonic optical pulse that is spectrally narrowband with a designed temporal profile**. In experiment we achieve upwards of 40% conversion efficiency with this process allowing this to be used for high average and peak power applications. These narrowband pulses can then be directly upconverted to the UV towards use in driving free electron laser photocathodes. Additionally, we present a theoretical framework for adapting this method to shape optical pulses driving other photoinjector based applications.

The next generation of augmented brightness XFELs, such as LCLS-II, promises to address current challenges associated with systems with low X-ray cross-sections. Typical photoinjector lasers produce coherent ultraviolet (UV) pulses via nonlinear conversion of an infrared (IR) laser. Fast and active beam manipulation is required to capitalize on this new generation of XFELs, and controlling the phase space of the electron beam is achieved by shaping the UV source. However current techniques for such shaping in the UV rely on stacking pulses in time, which leads to unavoidable intensity modulations and hence space-charge driven microbunching instabilities [1]. Traditional methods for upconversion do not preserve phase shape and thus require more complicated means of arriving at the desired pulse shapes after nonlinear upconversion [2]. Upconversion through four-waving mixing (FWM) allows direct phase transfer, convenient wavelength tunability by easily changeable phase matching parameters, and also has the added advantage of greater average power handling than traditional χ(2) nonlinear processes [3, 4,]. Therefore, we examine a possible solution for e-beam shaping using a machine learning (ML) implementation of real-time photoinjector laser manipulation which shapes the IR laser source and then uses FWM for the nonlinear upconversion and shaping simultaneously. Our presentation will focus on the software model of the photoinjector laser, the associated ML models, and the optical setup. We anticipate this approach to not only enable active experimental control of X-ray pulse characteristics but could also increase the operational capacity of future e-beam sources, accelerator facilities, and XFELs. References: [1] S. Bettoni, et al. “Impact of laser stacking and photocathode materials on microbunching instability in photoinjectors”, Phys. Rev. Accel. Beams 23, 024401 (2020) [2] Lemons, Randy, et al. “Dispersion-controlled Temporal Shaping of Picosecond Pulses via Non-colinear Sum Frequency Generation.” Phys. Rev. Accel. Beams 25, 013401 (2022) [3] P. Zuo, T. Fuji, and T. Suzuki, "Spectral phase transfer to ultrashort UV pulses through four-wave mixing," Opt. Express 18, 16183-16192 (2010) [4] John E. Beetar, M. Nrisimhamurty, Tran-Chau Truong, Yangyang Liu, and Michael Chini, "Thermal effects in molecular gas-filled hollow-core fibers," Opt. Lett. 46, 2437-2440 (2021)

We are developing an ultrafast and Ultracold Electron Source (UCES), based on near-threshold, two-step, femtosecond photoionization of laser-cooled rubidium gas in a grating Magneto Optical Trap (MOT). This source delivers stable ultrafast electron bunches with a unique combination of high bunch charge and low transverse emittance ~1.9 nm·rad, demonstrating the cold electron temperature ~25 K. Recent development focused on long term stabilizing the electron beam. By pulsing the high voltage accelerator potential, the effects of surface charge buildup in the accelerator structure are mitigated and secondary electron emission as a result of ion impacts on the cathode is prevented. This made a high resolution ponderomotive scattering measurement possible, in which a 1.1 mJ, 25 fs, 800 nm laser pulse is focused onto the electron bunch to a waist of 5.9 μm in vacuum. The ponderomotive force scatters the electrons which can be detected in the transverse profile. In this way the electron bunch length inside the self-compression point of the UCES has been measured to be 735±7 fs. Some wavelength dependent temporal structure originating from the ionization process could be observed.

At Eindhoven university an inverse Compton scattering (ICS) source is being built. The ICS source consists of a 100kV photo gun electron injector, X-band accelerator, and interaction laser. One of the first upgrades for this ICS source is operating in a so-called burst mode. In burst mode, the electron injector is replaced by the advanced continuous electron (ACE) injector and a Fabry-Perot cavity is added to the laser. Both systems work in a 100 nanosecond long burst. Significantly increasing the current x-ray yield and the brilliance of the ICS source. The ACE injector works by generating a continuous beam with a high current and low emittance through thermionic emission. The continuous electron beam is then chopped into a pulsed beam by a combination of a dual-mode elliptical RF cavity and a knife-edge. The dual-mode cavity uses both the fundamental mode (1.5 GHz) and its second harmonic (3.0 GHz) to increase the duty cycle of the chopping process to approximately 30% with a minimal loss of beam quality. Finally, a second dual-mode elliptical RF cavity compresses the pulse length of the bunches, preparing the beam for injection into an X-band linear accelerator.

A unique platform for a Tera Hertz Transmission Line design for a superradiant FEL is present. The smart line is controlled by Artificial Intelligence (AI) intended for a wide tunable broad-spectrum THz radiation propagation. The main goal is to transfer radiation in the most efficient way. A 3D analysis and diagnostic of radiation space-frequency tool was developed. The AI changes the functions of the mirrors in such a way that all the reflected rays will reach the target. The rays represent the electromagnetic field similar to a light field. The representation of the field in terms of rays was carried out using the Wigner Distribution Function. It allows describing the dynamics of field evolution in future propagation. This in turn helps with the initial design of the transmission line and facilitates the use of a Ray Tracing method for future processing. Thus, working in the linear and non-linear regimes. The Ray Tracing method and code is greatly enhanced using parallel processing with graphics cards.

Laser Wakefield Accelerators are now sufficiently mature to provide GeV scale/high-brightness electron beams capable of driving Free Electron Laser (FEL) sources. Here, we show start-to-end simulations carried out in the framework of the EuPRAXIA project of a Free Electron Laser driven by an LWFA accelerator in the Resonant Multi-Pulse Ionisation Injection (ReMPI) framework. Simulations with this model using a 1 PW Ti:Sa laser system and a 20 cm long capillary, show the injection and acceleration of an electron beam up to 4.5 GeV, with a slice energy spread and a normalized emittance below $4\times 10^{-4}$ and 80 $nm \times rad$, respectively. The transport of the beams from the capillary exit to the undulator is provided by a matched beam focusing with a marginal beam-quality degradation. Finally, 3D simulations of the FEL radiation generated inside an undulator show that $\approx 10^{10}$ photons with central wavelength of $0.15\, nm$ and peak power of $\simeq 0.3\, GW$ can be produced for each bunch. Our start-to-end simulations indicate that a single-stage ReMPI accelerator can drive a high-brightness electron beam having quality large enough to be efficiently transported to a FEL undulator, thus generating X-ray photons of brilliance exceeding $10^{25} ph/s/mm^2/0.1\%bw$

Attosecond pulse production is an important development focus for most major FEL facilities. Chirp/taper and eSASE schemes, both of which will shorten the pulses well below the femto-second level for both hard and soft x-rays, are proposed for implementation at EuXFEL. As a high repetition rate super conducting linac that feeds three 200m long undulator lines for parallel operation, EuXFEL presents distinct challenges but also unique opportunities for the proposed schemes.

We report on the operation of the DRACO Laser Driven electron source for stable multi-day operation for FEL applications. The nC-class accelerator delivers charge densities around 10 pC/MeV , <1 mrad rms divergence at energies up to 0.5 GeV and peak currents of over 10 kA [1]. Precise characterisation is paramount for controlled operation, including: spectrally resolved charge diagnostic, coherent optical transition radiation (TR) to resolve microbunch beam structures [2] and TR-based multioctave high-dynamic range spectrometry for sub-fs resolved characterisation of the 10 fs rms electron bunches [3]. Achieved stability allows for systematic exploration of demanding applications, resulting in the recent demonstration of the first LWFA based Beam-driven Plasma Wakefield Accelerator [4]. Fulfilling the high demands required for FEL operation, the COXINEL manipulation line developed at Synchotron SOLEIL has recently been installed at our facility. Based on successful beam transport of over 13000 shots within 9 experimental days during commissioning, we were able to demonstrate the very first operation of a seeded FEL driven by a laser plasma accelerator [5].

Frequency mixing was studied experimentally at SASE3, the soft X-ray undulator of the European XFEL. Two frequencies were generated in the first part of the undulator in alternating K configuration. The mixing process occurred in the second part with detuned undulator segments used to generate R56. Finally, the difference frequency was radiated and amplified in a third part of the SASE3 undulator. Experiments were performed at several electron energies (11.5 GeV, 14 GeV, and 16.5 GeV) with frequency mixing generation at photon energies between 500 eV and 1.1 keV. Pulse energies were on the mJ level, depending on the length of the radiator part. A practical application of frequency mixing at European XFEL is a possible extension of the operating range of the SASE3 undulator towards lower photon energies, by using a relatively short afterburner with longer period.

The generation of x-ray pulses carrying orbital angular momentum from an x-ray free-electron laser (FEL) has attracted considerable attention due to the ability to directly change atomic states and develop new material characterization techniques. In this contribution, we report a new method for generating intense x-ray vortices. The method is based on the widely used self-amplified spontaneous emission scheme and does not require additional helical undulators or external laser systems. It can therefore in principle be employed by all existing XFEL facilities with limited hardware additions.

High-brightness extreme ultraviolet (EUV) light source is strongly required for high-resolution photoelectron spectroscopy, imaging experiments, and EUV lithography. In this work, the self-modulation technique is introduced into seeded FELs, such as high-gain harmonic generation (HGHG), to significantly reduce the requirement of the seed laser power by enhancing coherent energy modulation. Numerical simulations demonstrated that the modified HGHG configuration with the self-modulation technique could generate high-repetition-rate, fully coherent, stable, and kilowatt-scale EUV pulses at a more compact linac-based light source.

Beam-driven plasma wakefield accelerators (PWFAs) offer a unique regime for the generation and acceleration of high-quality electron beams to multi-GeV energies. Here we present an innovative hybrid staging approach, deploying electron beams generated in a laser-driven wakefield accelerator (LWFA) as drivers for a PWFA, integrated in a particularly compact setup. This scenario exploits the capability of LWFAs to deliver shortest, high peak-current electron bunches [1] with the prospects for high-quality witness beam generation in PWFAs [2]. The feasibility of the concept is presented through exemplary particle-in-cell simulations, before describing experimental results from extensive campaigns performed at high-power laser facilities; ATLAS (LMU, Munich), SALLE-JAUNE (LOA, Paris) and DRACO (HZDR, Dresden). Using few-cycle optical probing we captured clear images of beam-driven plasma waves in a dedicated plasma stage, allowing us to identify a non-linear plasma-wave excitation regime. Trailing the plasma waves, the impact of ion motion to the transverse modulation of the plasma density was observed over many picoseconds [3]. Furthermore, we demonstrate for the first time the acceleration of distinct witness beams in such LWFA-driven PWFA (LPWFA) setup [4,5], showcasing an accelerating gradient on the order of 100 GV/m. These milestones pave the way towards compact sources of energetic ultra-high brightness electron beams as well as a miniature model for large scale PWFA facilities.

Attosecond laser pulses in the extreme ultraviolet/soft X-ray (XUV/SXR) spectral regions are presently available for attosecond pump-probe spectroscopy and extreme ultraviolet lithography for chip manufacturing, ultrafast atomic-scale microscopy, and nonlinear X-ray optics. There are two main approaches to produce attosecond light pulses: high-harmonic generation (HHG) in gas-phase or solid-state matter based on the three-step model, and X-ray free-electron lasers (XFELs) based on self-amplified spontaneous emission (SASE) and laser seeding processes of relativistic free electrons traveling through an undulator. Here, we propose a novel route of producing attosecond laser pulses, based on the generation of attosecond electron pulse trains in photon-induced near-field electron microscopy (PINEM), combined with the SASE principle for light amplification. Our scheme relies on high-density nanotip arrays emitting dense electron bunches that are subsequently modulated with a PINEM-type interaction, enabling high-gain for amplification of XUV/SXR high harmonic radiation. Our PINEM-HHG mechanism using attosecond electron pulses can serve as promising ultra-bright extreme ultraviolet/soft X-ray attosecond laser sources.

The possibility of using a plasma accelerated electron beam to generate Free Electron Laser (FEL) radiation has recently been proven. In the plasma acceleration process an intense broadband spectrum radiation in the X ray region, the betatron radiation, is produced by the electron beam passing through the ionized gas. In this paper it is proposed to use this radiation, suitably monochromatised, as a seed to stimulate the emission in the Free Electron Laser on the fundamental frequency and on the higher harmonics. This scheme could be adopted from all FEL injected by plasma accelerated electron beams via particle or laser wakefield acceleration.

S.Y. Teng(1,2), S.H. Chen(1), W.Y. Chiang(2), M.C. Chou(2), H.P. Hsueh(2), W.K. Lau(2), A.P. Lee(2), P.T. Lin(3) 1 Department of Physics, NCU, Taoyuan, Taiwan 2 NSRRC, Hsinchu, Taiwan 3 Department of Engineering and System Science, NTHU, Hsinchu, Taiwan. Intense coherent THz radiation has been generated from an 18-period, hybrid-type U100 planar undulator as it is driven by short relativistic electron pulses produced from the NSRRC photoinjector. However, it is observed that the number of output optical pulse cycles is much less than the number of undulator periods and therefore the radiation spectral bandwidth has been broadened. It is found that the dispersion of undulator with excessive energy spread is responsible for this undesired broadening of THz radiation spectrum. In this study, instead of using rectilinear rf bunch compression (i.e. velocity bunching) in photoinjector linac, we investigate the feasibility of using nonlinear magnetic bunch compression for spectral bandwidth control of coherent THz undulator radiation.

The LAPLACIAN (Laser Acceleration Platform as a Coordinated Innovation Anchor) experimental facility inside the MIRAI project framework is the Japanese answer to the global effort for the development of compact accelerators based on laser plasma acceleration (LPA) for its application to free electron laser (FEL). Situated in the SPRING-8 site, LAPLACIAN aims for the generation of X-ray FEL with relativistic electrons (GeV) from an LPA source in a beamline of under 10 m. Even after the recent demonstration of by the SIOM group, achieving proper electron beam parameters in a consistent manner and a reliable coupling with the undulator is non-trivial and still under research. In LAPLACIAN, multiple gas targets and LPA schemes are being studied, including a planned multiple plasma stages setup for GeV electron energies combined with magneto-optics for coupling. In this talk, an overview of the current facility status and some future plans will be given. In addition, we will report in some of the already achieved results and the new planned beamline.

Shaping techniques traditionally used to produce few femtosecond and even sub femtosecond soft X-ray FEL pulses at LCLS do not scale well to high repetition rates. Here we present the progress of the LCLS-II X-ray temporal shaping project which uses infrared and ultraviolet picosecond lasers to shape the electron beam of the LCLS-II superconducting linac. Quickly switching these shaping lasers on and off will enable multiplexing different beams to different beamlines.

Cavity-based X-ray Free Electron Lasers (FELs) such as the X-ray regenerative amplifier FEL (XRAFEL) [1] and the X-ray FEL oscillator [2] have drawn great interest as a means of producing high-brightness, fully coherent and stable hard x-ray pulses for high-repetition rate FELs [3]. However, high efficiency outcoupling of the stored cavity x-ray radiation remains challenging. Here we present a novel XRAFEL design to achieve efficient cavity outcoupling or Q-switching by introducing energy chirp in the electron beam while leaving the high-quality X-ray optics intact. During the FEL interaction, electron beam with an linear energy chirp can be slightly compressed or decompressed by the undulator, which leads to a gradual shift of radiation frequency outside the bandwidth of the Bragg crystal for efficient outcoupling. Our simulation results show that substantial power can be outcoupled from the X-ray cavity driven by chirped electron beams at 100 kHz repetition rate. We also discuss parameter tradeoff in such an XRAFEL scheme and a practical way to achieve the desired fast chirp control by a small, normal-conducting RF station in the LCLS-II [4]. [1] Z. Huang and R. D. Ruth. PRL96, 144801 (2006). [2] K.-J. Kim, Y. Shvyd'ko, S. Reiche, PRL100 244802 (2008). [3] G. Marcus, et al., PRL125, 254801 (2020). [4] M. Nasr, et al., in proceedings of IPAC'16 (Busan, Korea,2016).

In early 2019, the UK initiated a project to develop the science case for a UK XFEL, featuring a diverse team of UK scientists and international advisors. Accelerator scientists were engaged to highlight potential future accelerator developments and to develop concept outlines for a facility design meeting the requirements for world-leading capabilities. The UK XFEL Science Case, featuring the concept outlines, was published in late 2020. Subsequent exercises further demonstrated the support of the UK community and the project is anticipated to enter a more detailed design phase. The concept outlines are reviewed and potential next steps are outlined.

In this paper we'll describe the status of the FAST-GREENS experimental program where a 4 m-long strongly tapered helical undulator with a seeded prebuncher is used in the high gain TESSA regime to convert a significant fraction (up to 10 %) of energy from the 240 MeV electron beam from the FAST linac to coherent 515 nm radiation. We'll also discuss the longer term plans for the setup where by embedding the undulator in an optical cavity matched with the high repetition rate from the superconducting accelerator (3,9 MHz), a very high average power laser source can be obtained. Eventually, the laser pulses can be redirected onto the relativistic electrons to generate by inverse compton scattering a very high flux of circularly polarized gamma rays for polarized positron production.

Plasma photocathode injectors may enable electron beams with normalised emittance at the nm-rad level from a Plasma Wakefield Acceleration (PWFA) stage [1]. These electron beams typically have kA-level peak currents leading to ultrahigh 5D brightness beams with the potential to drive advanced light sources [1]. The feasibility of the plasma photocathode was demonstrated at FACET-I at SLAC [2]. Further experimental campaigns are gradually aiming toward ultrahigh 5D and 6D brightness beams at FACET-II [3]. However, a series of milestones must be reached before these beams can be utilised for XFELs. For example, electron beams accelerated in plasma-based accelerators inherently have a significant energy chirp due to the GV/m accelerating gradients involved. Since energy chirp and energy spread can be detrimental to the high-gain FEL interaction, advanced approaches have been developed for energy spread minimisation of the initially ultrahigh 5D brightness beams towards ultrahigh 6D brightness [4]. Here we show within the framework of the PWFA-FEL project that it may also be possible to produce ultrahigh 5D brightness beams with reduced energy spread using beam-loading. We present results aiming at a trade-off between reduced energy spread, increased peak current, and increased emittance and their application to a soft XFEL in the water window.

We present a theory for free electrons interaction with radiation in both classical and quantum regimes and delineate their transition, based on a model of quantum electron wavepacket (QEW). The theory has general validity for a wide range of free electron interaction and radiation sources, including Free Electron Lasers, Cerenkov radiation, and transition radiation. We exemplify our analysis with the schemes of Smith-Purcell radiation and dielectric laser acceleration (DLA). These interactions, which were studied in terms of point particle physics, have a quantum nature in a phenomenon known as “photon-induced near-field electron microscopy” (PINEM). Our QEW model identifies three universal distinct interaction regimes: (i) near-point-particle acceleration/deceleration DLA regime, (ii) PINEM regime of multiphoton induced electron energy sidebands, and (iii) anomalous PINEM regime (APINEM) of a newly reported periodic spectral bunching. See the three regimes in Fig.2. The formulation displays the transition of the FEL stimulated gain expression from the quantum to classical limit. Elsewhere we provided extension of the semiclassical model to quantum electrodynamics to include spontaneous emission and spontaneous superradiance by modulated QEW similar to the classical prebunched-beam superradiant FEL in the classical point-particle picture。

A tunable and multicolor light source with near Fourier-limited pulses, controlled delay, and fully coherent beam with precisely adjustable phase profiles enables state-of-the-art measurements and studies of femtosecond dynamic processes with high elemental sensitivity and contrast. The start-to-end simulations efforts aim to take advantage of the available global pool of software and past and present extensive efforts to provide realistic simulations, particularly for cases where precise and fine manipulation of the beam phase space is concerned. Since, for such cases, tracking of beams with billions of particles through magnetic structures and handover between multiple codes are required, extensive realistic studies for such cases are limited. Here we will describe a workflow that reduces the needed computational resources and share studies of the EEHG seed section for the FLASH2020+ [1] project.

A THz FEL is in preparation at PITZ as a proof-of-principle experiment for a high power and high repetition rate THz source and as an option for THz-driven experiments at the European XFEL. Some of these experiments require excellent coherence and CEP stable THz pulses. In SASE regime the coherent properties of the FEL radiation are limited. A seeding scheme can be used instead of SASE to improve the coherent properties and shot-to-shot stability. Several options for seeding are considered in simulation for the THz FEL at PITZ: external laser pulse, pre-bunched electron beam, energy modulated electron beam and additional short spike. The results of the simulations for each method of seeding are evaluated and compared. The improvements over SASE in energy, spectral and temporal stability of the THz pulse are presented.

The FLASH facility houses a superconducting linac powering two FEL beamlines with MHz repetition rate in 10 Hz bursts. Within the FLASH2020+ project, which is taking care of facility development, one major aspect is the transformation of one of the two FEL beam lines to deliver externally seeded fully coherent FEL pulses to photon user experiments. At the same time the second beam line will use the SASE principle to provide photon pulses of different properties to users. Since the electron beam phase space conducive for SASE or seeded operation is drastically different, here a proof-of-principle experiment using the existing experimental seeding hardware has been performed demonstrating the possibility of simultaneous operation. In this contribution we will describe the setup of the experiment and accelerator, and discuss the chances and limitations of the experimental seeding hardware. Finally, we will discuss the results and their implications also for the FLASH2020+ project.

In the framework of the FLASH2020+ project, the FLASH1 beamline will be upgraded to deliver seeded FEL pulses for users. This upgrade will be achieved by combining high gain harmonic generation and echo-enabled harmonic generation with a wide-range wavelength-tunable seed laser, to efficiently cover the 60-4 nm wavelength range. The undulator chain will also be refurbished entirely using new radiators based on the APPLE-III design, allowing for polarization control of the generated light beams. With the superconducting linac of FLASH delivering electron beams at MHz repetition rate in burst mode, laser systems are being developed to seed at full repetition rates. In the contribution, we will report about the progress of the project.

Within the FLASH2020+ project the FLASH VUV/XUV FEL facility at DESY (Hamburg, Germany) is currently undergoing a major upgrade to become the first high repetition rate, fully coherent FEL light source worldwide [1]. To reach this goal, one of the two in parallel operated FEL branches will be seeded at a fixed wavelength at 343 nm in a first step (SEED 1) and tunable between 297 nm to 317 nm in a second step (SEED 2) following the two-color Echo-Enhanced Harmonic Generation (EEHG) scheme [2]. The seed laser system is designed to deliver UV pulse energies > 50 µJ and > 100 µJ for SEED 1 and SEED 2, respectively, and with 6000 pulses in one second (1 MHz pulse trains in 600 µs - 10 Hz bursts). In combination with the EEHG seeding principle, this will allow for the generation of high harmonics corresponding to XUV FEL pulses with photon energies of more than 300 eV (down to 4 nm in wavelength). In order to exploit the full capabilities of the narrow-band fully coherent FEL pulses for 24/7 scientific user experiments, the seed laser needs to provide broadly tunable, high power UV laser pulses with pulse durations of 50 fs, excellent beam quality and exceptional high short and long-term stability in respect to the seeding wavelength (< 2e-4), pulse – pulse energy (< 2%) and pointing jitter (< 20 µrad). Altogether, the requirements on the laser system are beyond state-of-the-art. We will present the concept as well as the first experimental results of our novel high-power seed laser system based on a 5 kW Inno-Slab CPA pump laser system, optical parametric chirped pulse amplification (OPCPA) and a highly efficient UV conversion scheme. An extensive numerical study based on a 3+1 dimensional start-to-end simulation code (chi3D) allows for a precise predictions of system performance in terms of output power, tunability, beam quality and stability in respect to the measured input parameters and respective statistical and systematic fluctuations. The theoretical results are confirmed by first experimental studies being in excellent agreement in terms of UV conversion efficiency, beam quality and the predicated improvement of the pulse-to-pulse stability compared to the OPCPA stability. The insides of this study had major impact on the conceptual design of the laser system, especially the dispersion concept and the best implementation of user controls, such as power attenuation and fast wavelength control, etc. [1] M. Beye, ed., “FLASH2020+: Making FLASH brighter, faster and more flexible : Conceptual Design Report.” Deutsches Elektronen-Synchrotron, DESY, Hamburg, 2020. DOI: 10.3204/PUBDB-2020-00465 [2] L. Schaper, ed.al, “Flexible and Coherent Soft X-ray Pulses at High Repetition Rate: Current Research and Perspectivesal” Appl. Sci. 2021, 11, 9729. https://doi.org/10.3390/app11209729

Phase-locked pulses are important for coherent control experiments. Here we present theoretical analyses and start-to-end simulation results for the generation of phase-locked pulses using the Hard X-ray Self-Seeding (HXRSS) system at the European XFEL. As proposed in Ref. [1], the method is based on a combination of self-seeding and fresh-slice lasing techniques. However, at variance with Ref. [1], here we exploit different transverse centroid offsets along the electron beam. In this way we may first utilize part of the electron beam to produce SASE radiation, to be filtered as seed and then generate HXRSS pulses from other parts of the beam applying appropriate transverse kicks. The final result consists in coherent radiation pulses with fixed phase difference and tunable time delay within the bunch length. This scheme should be useful for applications such as coherent x-ray pump-probe experiments.

FLASH2020+ is an upgrade project for the FLASH facility at Hamburg. A main goal of the project is to generate fully coherent soft X-ray FEL radiation at a high repetition rate (MHz). The project will utilize two external laser seeding principles in order to produced Seeded FEL with tunable wavelength from 4-60 nm. In order to achieve this goal, both HGHG (High Gain Harmonic Generation) and EEHG (Echo-Enhanced Harmonic Generation) methods provide FEL emission at harmonics of a seed laser. For HGHG, a tunable UV laser system (297-317 nm) and for EEHG a combination of the tunable UV laser and fixed wavelength (343 nm) laser system would be used to cover the whole range of wavelengths between 4-60 nm. In this contribution, we will describe the requirements of the seed laser to initiate the seeding process and will explain the concept of seed laser beam transport and incoupling into the modulators for FEL radiation production. The first seed laser (Seed1) with fixed wavelength is transported about 28 meters from laser lab to the incoupling chicane. The second seed laser (Seed2) with a tunable UV wavelength is transported about 35 meters. Our concept uses a full relay imaging system and in vacuum components for the laser transport in addition to high repetition rate diagnostics to deliver, monitor and control the beam and pulse parameters at the interaction with electron beam. We investigate the technical and engineering limitations for the design and address those challenges to provide the demanding seed laser parameters for generating high repetition rate seeded FEL.

Over the last few years tremendous progress has been gained in the theoretical understanding and experimental demonstration of seeded FELs . The ultimate spectral limit of seeded FEL, however, remains unclear, because of the broadening and distortions induced in the output spectrum by residual broadband energy modulations in the electron beam. In this talk, we present the mathematical descriptions of the impact of broadband energy modulations on the EEHG, HGHG and self seeding bunching spectrums produced by the microbunching instability through both the accelerator and the FEL line. We will show the agreement of our models with the systematic experimental characterization seeded FEL spectrums in FERMI and Eu-XFEL. Using experimental data of EEHG FEL performance in FERMI in the photon energy range 130–210 eV, we demonstrate that amplification of electron beam energy distortions primarily in the EEHG dispersive sections explains an observed reduction of the FEL spectral brightness proportional to the EEHG harmonic number. Local maxima of the FEL spectral brightness and of the spectral stability are found for a suitable balance of the dispersive sections’ strength and the first seed laser pulse energy[1]. [1] Physical Review Accelerators and Beams 24, 8, 2021

Recently, Hard x-ray self-seeding (HXRSS) operations at the European X-ray free-electron laser (EuXFEL) opened a pathway towards the application of pulses with high spectral density (in terms of ph/eV per pulse) in the fields of applied physics, chemistry and biology, where the coherent radiation spectrum is essential. The spectrum of hard x-ray self seeding pulses is generally accompanied by a pedestal around the central seeded photon energy. The pedestal contains two separate components: normal self-amplified spontaneous (SASE) and sideband emissions that can be ascribed to long-wavelength modulations of the electron beam. The pedestal limits the spectral purity and can impact some user applications. In this report, we analyze the purity of HXRSS pulses in the presence of microbunching instability. We look at the spectral contents after and before saturation, and display the contribution of the pedestal in the HXRSS spectrum.

External seeding schemes allow the generation of stable and fully coherent free electron laser (FEL) radiation but can be limited in repetition rates in orders of tens of Hz. This limitation is mainly posed by limited average power of the seed lasers that are required to provide hundreds of MW peak power to modulate the electron bunches. An optical-klystron-based high gain harmonic generation (HGHG) scheme, which can be implemented in several existing and upcoming seeded FEL beamlines with minimal to no additional installations, overcomes this limitation by greatly reducing the required seed laser power. In this work, we carefully study the scheme with detailed simulations that include imperfections of electron beam properties such as a quadratic electron beam energy chirp that characterizes existing FEL facilities. We discuss the optimization steps that in these conditions ensure successful operation, opening the path towards exciting science at FELs with fully coherent and high repetition rate FEL radiation.

An X-ray free-electron laser oscillator (XFELO) is a next generation X-ray source promising radiation with full three-dimensional coherence, nearly constant pulse to pulse stability and more than an order of magnitude higher spectral flux compared to SASE FELs. In this contribution, the concept of an R&D project for installation of an XFELO demonstrator experiment at the European XFEL facility is conceptually presented. It is composed of an X-ray cavity design in backscattering geometry of 133 m round trip length with four undulator sections of 20 m total length producing the FEL radiation. It uses cryocooled diamond crystals and employs the concept of retroreflection to reduce the sensitivity to vibrations. Start to end simulations were carried out which account for realistic electron bunch distributions, inter RF-pulse bunch fluctuations, various possible errors of the X-ray optics as well as the impact of heat load on the diamond crystals. The estimated performance and stability derived from these simulations shall be reported and foreseen issues shall be discussed.

The FLASH2020+ project has started to transform the FLASH facility to broaden the facility profile and meet demands of future user experiments. In a nine-month lasting shutdown until August 2022 the linear accelerator of the FLASH facility has, among others, been upgraded with a laser heater, new bunch compressors and new modules. The latter results in an energy upgrade to 1.35 GeV allowing to reach sub 4 nm wavelength. In the following 14-month lasting shutdown starting mid 2024 the FLASH1 FEL beamline will be completely rebuild. The design is based on external seeding at MHz repetition rate in burst mode allowing for coherent tuneable FEL radiation in wavelength and polarization by installation new APPLE-III undulators. Post compression of the beam downstream of the radiators will allow for high quality THz generation and together with the new experimental endstations and pump probe lasers provide a unique portfolio for next generation user experiments.

The 2nd stage of the FLASH2020+ project at DESY will be an upgrade of the FLASH1 beamline to enable HGHG and EEHG seeding with two modulator-chicane stages, and a radiator section with 11 Apple-III undulators to enable FEL radiation with controllable polarization. A key feature of FLASH, namely the capability of providing several thousand FEL pulses in the extreme UV and soft X-ray must not be compromised. Downstream of the radiator the beamline houses longitudinal diagnostics, a double bend (quasi-) achromat to separate the electrons from the photons and divert the electron beamline from the photon diagnostics, a post-compressor, a THz-Undulator (requires an electron beam that is compressed more strongly than for seeding), and finally the dumpline, capable of safely aborting up to 100 kW electron beam power. This article describes the conceptional and some technical details of the beamline with emphasis on the upstream part (modulators and radiator) designed for seeding.

ABSTRACT: FERMI is implementing a development plan to keep the facility in a world-leading position on the base of the requests coming from the user community and the advises from the Scientific Advisory Council and the Machine Advisory Committee. The ultimate goal of this plan consists in doubling the maximum photon energy available and in reducing the pulse duration below the characteristic lifetime of the atomic core levels in the source spectral range. An upgrade of FERMI aimed at reaching the oxygen K-edge requires a profound modification of the FEL configurations and of the main components of the machine, including the linac and the undulator lines. One of the most promising approaches for this upgrade is to implement the echo-enabled harmonic generation (EEHG) scheme, relying on two external lasers to precisely control the spectrotemporal properties of the FEL pulse. The conversion to EEHG of the first stage of the double-stage harmonic cascade presently in use on FEL-2, would allow to reach harmonics as high as 120, enabling to generate coherent pulses down to 2 nm. The main aspects of the upgrade strategy will be discussed in this contribution.

In optical conventional lasers, chirped pulse amplification (CPA) has become an extremely powerful technique for the generation of ultrashort pulses in the infrared and visible spectral ranges. In this contribution we report the successful implementation of CPA in a seeded XUV FEL. A second experiment, using a two-stage harmonic generation scheme (FERMI FEL-2) has the objective to generate coherent and phase-tailored few-femtosecond FEL pulses, with gigawatt peak power in the sub-10 nm spectral range. This second experiment is still in progress. We will discuss the main scientific and technical bottlenecks and the implications.

The local laser synchronization is known to be of high importance for Free Electron facilities, affecting both machine performance and pump-probe FEL-external laser experiments. So, there has been a continuous effort to improve the timing jitter of all machine lasers. One of the main contributions to the overall timing jitter comes from the locking of the local laser oscillators to the reference signal of the facility. Here we describe the latest developments and progress in this direction related to the FERMI seed laser system. The first investigated aspects includes the characterization and optimization of the locking performance of the commercial Ti:Sapphire oscillators Vitara T and HP (Coherent). We present data on the performance of three different oscillators of this type, as well as on the effect of adding an additional cavity length control actuator. The second presented aspect is related to the plan to extend the optical synchronization layout: for some planned seed laser operation modes two Ti:Sapphire oscillator need to be synchronized simultaneously. For this purpose, studies of optimum schemes for locking the two oscillators are in progress, first results are presented.

The advent of free electron lasers (FELs) in the soft and hard X-ray spectral region has opened the possibility to probe electronic, magnetic and structural dynamics, in both diluted and condensed matter samples, with femtosecond time resolution. In particular, FELs strongly enhanced the capabilities of several analytical techniques, which took advantage of the high degree of transverse coherence provided. FELs based on the harmonic up-conversion of an external seed laser are characterised also by a high degree of longitudinal coherence, since electrons inherit the coherence properties of the seed. At the present state of the art, the shortest wavelength delivered to user experiments by an externally seeded FEL light source is about 4 nm. We show here that pulses with a high longitudinal degree of coherence (first and second order) covering the water window and with photon energy extending up to 790 eV can be generated by exploiting the so-called nonlinear harmonic regime, which allows generation of radiation at harmonics of the resonant FEL wavelength. Moreover, we report the results of two proof-of-principle experiments: one measuring the oxygen K-edge absorption in water ($\sim$ 530 eV), the other analysing the spin dynamics of Fe and Co through magnetic small angle x-ray scattering at their L-edges (707 eV and 780 eV)

Free-electron lasers producing ultrashort pulses with high peak power are a resource to extend ultrafast non-linear spectroscopic techniques into the extreme-ultraviolet–X-ray regime. A super radiant cascade was proposed as a method to shorten the pulse duration in seeded FEL. Pulses shorter than the typical duration supported by the FEL gain bandwidth of the FEL amplifier in the linear regime were measured at FERMI. In these conditions we also observed a strong frequency pulling phenomenon that that will be discussed in this contribution.

FEL basic theory indicates that the output wavelength of a seeded FEL operated in the HGHG configuration is determined by the wavelength of the seed laser and light is emitted when undulators are tuned to one of the harmonics of the seed laser. In a more realistic case, when taking into account the electron beam imperfections and the finite bandwidths of the seed and of the amplification process, the output wavelength is influenced by these factors and there is a small variation from this rule. In this work, we consider the effects of the dispersive section, the curvature of the electron beam longitudinal phase-space and the frequency pulling as major contributors. We show how these quantities influence the effective final FEL wavelength. Furthermore, we show how one can reconstruct the electron beam longitudinal phase-space from the analysis of the FEL wavelength sensitivity to the seed laser delay with respect to the beam arrival time.

In order to meet the user request of extending the FERMI FEL spectral range over the whole water window, we are developing an upgrade strategy that is based on the implementation of the Echo Enabled Harmonic Generation (EEHG) scheme. The FERMI upgrade strategy is structured as follow: during a first phase, the single cascade FEL-1 branch will be adapted to operate either in EEHG or in HGHG. This upgrade can be achieved with relatively low cost and impact on FERMI operations and will improve the spectral range, spectral quality and scheme flexibility of FEL-1. Furthermore, it will provide a versatile test bench opening the possibility to explore in details the EEHG scheme potentialities and address many of the possible issues related to the second and more critical phase of the upgrade project: the upgrade of FEL-2. These two phases will proceed in parallel to the LINAC upgrade to increase the nominal energy. Solutions aiming at a peak beam energy of 1.8 and 2.0 GeV are under study. In this contribution we will focus on the upgrade of the FEL-1 branch that has already started and is foreseen to provide light to users with the new configuration by spring 2023.

Improvement of the longitudinal coherence in the proposed Soft Xray FEL, the SXL, for the MAX IV Laboratory is an important design aspect to enhance the user case. One of the main considered methods is HBSASE. However the final compression in the MAX IV acceleratos is done at full energy, and thus leaving an energy chirp in the electron pulse. This chirp in longitudinal phase space has to be removed for an efficient implementation of HBSASE. In this paper we show in simulations how the phase space is improved by first overcompressing the pulse, and then correct it by a two-plate wake field de-chirper. The resulting pulse is then shown to have qualities such that, by HBSASE, a significant narrowing of the FEL bandwidth is achieved at 1 nm.

In order to improve the brightness and coherence of the soft x-ray FEL line of SwissFEL (Athos), components for an Echo Enabled Harmonic Generation (EEHG) scheme are currently in preparation. The first components have been installed to allow first ESASE operation test in Spring 2022. This first stage consists in a 10 mJ class seed laser, a U200 modulator with individual control of each half period and a four electromagnets dipole chicane (R56 < 800 um). The large magnetic chicane and the second modulator are still in preparation for an installation by end 2022. This paper will give a technical description of the different systems as well as preliminary results of the commissioning with beam.

In the scope of the HERO ERC project, we are implementing a laser-based seeding scheme at the SwissFEL soft X-ray Athos beamline to generate fully coherent X-ray FEL pulses. With this perspective, we designed and built a new laser facility. It consists of a terawatt-class, femtosecond laser system based on Titanium Sapphire technology with wavelength tuning capability, an optical transfer line as well as a launching optical setup and diagnostics to spatially and temporally overlap the laser and the electron bunch inside the modulator, where the seeding process occurs. We present an overview of the facility with details of the laser performance as well as first commissioning results with the electron beam.

Nearly fully coherent hard X-ray self-seeded (HXRSS) free-electron laser (FEL) pulses with an unprecedented peak-brightness and a narrow spectrum using the forward Bragg-diffraction (FBD) monochromator has been provided. We have achieved outstanding performance of HXRSS FEL over photon energy range covering from 3.5 keV to 14.6 keV at PAL-XFEL. Furthermore, an averaged energy of seed FEL of ~1mJ is obtained in the range from 5 keV to 10 keV. With these pulses single-shot coherent imaging (SSI) experiment and serial femtosecond crystallography (SFX) were performed. We developed x-ray energy scanning program with the help of double crystal monochromator (DCM), which results in improved spectral impurity and fully calibrated energy scale. With this energy scanning program, we have conducted test experiments such as resonant inelastic X-ray scattering (RIXS) and X-ray emission spectroscopy (XES), femtosecond time resolved X-ray absorption near edge structure (TR-XANES). In this presentation, we present recent experimental results by using the hard X-ray self-seeded FEL with energy scanning at PAL-XFEL.

The FAST-GREENS FEL experiment is aimed at demonstrating extraction efficiencies of greater than 10%. This is accomplished with a high-power seed laser and an aggressively tapered undulator to compensate for the energy loss in the electron beam. A proof of concept experiment will be conducted at the Fermilab Accelerator Science and Technology Facility (FAST) using an undulator specifically built for this purpose. To support this experiment, the LINAC requires a unique setup that optimizes the longitudinal current distribution while preserving emittance in the presence of CSR and space-charge effects. This paper summarizes the beam dynamics optimization performed in support of TESSA and provides the nominal working point for the FEL experiment.

External seeding techniques like high-gain harmonic generation (HGHG) and echo-enabled harmonic generation (EEHG) have been proven to be able to generate fully coherent radiation in the EUV and X-ray range. However, towards seeding at a high repetition rate, the repetition rate of current laser systems with sufficient power for seeding is limited to the kilohertz range. One attractive solution to this limitation is to reduce the required seed laser power. In this contribution, we will present a harmonic optical klystron scheme with high gain harmonic generation. With the harmonic optical klystron scheme as the seeding technique, the required seed laser power is decreased, and higher harmonics than in a standard single-stage HGHG can be achieved.

At DELTA, a 1.5-GeV electron storage ring operated by the TU Dortmund University, the seeding scheme CHG (coherent harmonic generation), the counterpart to HGHG (high-gain harmonic generation) without FEL gain, is used to provide ultrashort pulses in the femtosecond regime at harmonics of the seedlaser wavelength. To provide higher harmonics and thus shorter wavelengths, it is planned to upgrade the short-pulse facility to the EEHG (echo-enabled harmonic generation) scheme, which has yet not been implement at any storage ring. To install the needed three undulators and two chicanes, about a quarter of the storage ring needs to be modified. The paper presents the layout of the envisaged EEHG facility and the demo project SPEED (Short-Pulse Emission via Echo at DELTA) where all components are realized in a single undulator.

At the 1.5 GeV synchrotron light source DELTA operated by the TU Dortmund University, the short-pulse facility employs the seeding scheme coherent harmonic generation (CHG) to produce ultrashort pulses in the vacuum ultraviolet and terahertz regime. This is achieved via a laser-induced electron energy modulation and a subsequent microbunching in a dispersive section. The spectro-temporal properties of the CHG pulses as well as the coherently emitted terahertz radiation are influenced by the seed laser parameters and can be manipulated by varying the laser pulse shape and the strength of the dispersive section. CHG spectra for different parameter sets were recorded and compared with the results of numerical simulations to reconstruct the spectra. A convolutional neural network was employed to extract the spectral phase information of the seed laser from the recorded spectra. In addition, the shaping of the coherently emitted THz pulses by controlling the seed pulse spectral phase using a spatial light modulator was also demonstrated.

The external seeding scheme Echo-Enabled Harmonic Generation (EEHG) utilizes two modulators and two chicanes to manipulate the longitudinal phase space of an electron beam to achieve bunching at higher harmonics of the seed laser wavelength. Different combinations of energy modulation and longitudinal dispersion can result in the same amount of bunching at a certain harmonic. This study investigates the impact of the choice of the energy modulation amplitudes on the bunching properties and the spectro-temporal characteristics of the free-electron laser (FEL) radiation. Finally, a comparison between EEHG and the single modulator-chicane seeding scheme High-Gain Harmonic Generation (HGHG) at the 15th harmonic of the seed laser wavelength is presented. The corresponding numerical modelling and simulations are performed within the parameter range of the future upgrade of the FEL user facility FLASH at DESY.

The external seeding technique Echo-Enabled Harmonic Generation (EEHG) consists of two undulators which are used to imprint energy modulations to an electron bunch via interaction with a seed laser. Each of these so-called modulators is followed by a chicane introducing longitudinal dispersion. Proper adjustment of the amplitudes of the energy modulations and dispersive strengths allows to achieve bunching at high harmonics of the seed laser wavelength. In the near future, this seeding scheme will be utilized in one of the beamlines of the free-electron laser (FEL) user facility FLASH at DESY to provide stable seeded radiation down to the soft X-ray regime at high repetition rate. Dedicated numerical simulations are carried out within the foreseen parameter space to investigate how variations of the energy modulations due to power fluctuations of the two seed lasers affect the bunching properties and the stability of the generated FEL radiation.

Externally seeded FELs can produce fully coherent short-wavelength pulses with the advantage of higher shot-to-shot stability and spectral intensity than SASE radiation. For the FLASH2020+ project, the Echo-Enabled Harmonic Generation (EEHG) seeding technique achieves seeded FEL radiation in the XUV and soft X-ray range down to wavelengths of 4 nm. The implementation of the EEHG requires precise phase space manipulations in the seeding section of the beamline, which would make the performance of the EEHG sensitive to the collective effects, such as Coherent Synchrotron Radiation (CSR) in some working range. Therefore, it is essential to consider the CSR in EEHG simulations and to understand its impact on the electron beam properties. In this work, we compare different methods for calculating CSR and investigate the mechanism of its effect on the EEHG performance.

EuPRAXIA@SPARC_LAB is a new Free Electron Laser (FEL) facility that is currently under construction at the Laboratori Nazionali di Frascati of the INFN. The electron beam driving the FEL will be delivered by an X-band normal conducting LINAC followed by a plasma wakefield acceleration stage. It will be characterized by a small footprint and include two different plasma-driven photon beamlines. In addition to the soft-X-ray beamline, named AQUA and delivering ultra-bright photon pulses for experiments in the water window to the user community, a second beamline, named ARIA, has been recently proposed and included in the project. ARIA is a seeded FEL line in the High Gain Harmonic Generation configuration and generates coherent and tunable photon pulses in the range between 50 and 180 nm. Here we present the potentiality of the FEL radiation source in this low energy range, by illustrating both the layout of the FEL generation scheme and simulations of its performances.

The extension of four-wave mixing (FWM) technique to the extreme ultraviolet and soft X-ray ranges allows to monitor the dynamics of coherent excitations of matter, when realized with the exquisite coherent property of bright FEL pulses. We show for the first time a scheme to provide transversally separated pulses with parallel or crossed linear polarizations, realized at FERMI FEL. This configuration paves the way to explore additional features of pump&probe and FWM techniques, and, in particular, the possibility to excite a transient polarization grating on the sample. For this reason, such a technique is important the detection of circular dichroism and chiral properties of matter and the characterization of spin waves and magnons. By tailoring the electrons trajectory along the undulator line, we demonstrate the possibility of deliver balanced and stable couple of pulses with an horizontal separation of the order of millimeters at the experimental station.

A new concept of a high repetition rate VUV FEL is discussed. The FEL is envisioned to operate in the wavelength range from 50 to 250 nm with pulse energies of about 30 µJ throughout the wavelength range, and a pulse length of a few 100 fs. The SRF LINAC technology developed and used at the Helmholtz-Zentrum Dresden-Rossendorf for the Radiation Source ELBE is planned to be used for the driver-accelerator. This allows operating an electron beam with an average current of 1 mA on the order of magnitude, pulse repetition rate of up to 10 MHz, and the bunch charge of 100 pC, as used for the FEL design. We consider using the HGHG to allow the generation of fully coherent pulses. The high repetition rate electron beam makes it possible to construct an FEL oscillator that would be used as the high repetition rate seed of the HGHG amplifier. In the proposed scheme, the SRF LINAC provides beams for the seeding oscillator and the HGHG amplifier simultaneously. The described FEL would create new experimental regimes, not available at any other photon source. These could result in transformative changes in physical chemistry studies in the gas phase and at the interfaces, e.g., heterogeneous catalysis.