PSI Bericht Nr. 17-07 December 2017

Particle Accelerator Activities at the Paul Scherrer Institut

Terence GARVEY

Paul Scherrer Institut, Villigen, Switzerland 5232

Division of Large Research Infrastructures

Invited paper presented at

Les Journées Accélérateurs de la Société Française de Physique,

Roscoff, 3rd – 6th October, 2017.

Resumé

Les Activités Accélérateur à l’Institut Paul Scherrer.

L'Institut Paul Scherrer exploite deux complexes d'accélérateurs en tant que ‘centre- serveurs’ pour une grande communauté de chercheurs. Il s'agit de l'Accélérateur de Protons à Haute Intensité (HIPA) et de la Source de Lumière Suisse (SLS). HIPA est un cyclotron à protons de 590 MeV. Il sert à produire des neutrons par spallation pour la recherche en physique de la matière condensée et à produire des muons et d'autres particules secondaires pour la recherche en magnétisme et en physique des particules. Le SLS est un anneau de stockage d’électrons de 2,4 GeV utilisé comme source de rayonnement synchrotron de 3ème génération fournissant des photons pour une large gamme de disciplines scientifiques. En plus de ces deux installations, l'Institut met progressivement en service un laser à électrons libres en rayons X (SwissFEL) qui fournira aux chercheurs des impulsions femto-seconde intenses à partir d’une ligne de rayons X ‘dur’ (ARAMIS) et de rayons X ‘mou’ (ATHOS). L'Institut exploite également un cyclotron supraconductrice à 250 MeV (COMET) aux fins de la thérapie par proton. Le centre de thérapie a récemment été équipé d'un troisième « gantry » rotatif qui est en cours de mise en service. Un « upgrade » du système de radiofréquence de HIPA et des projets pour "SLS-2" seront présentés. L'accélérateur SwissFEL et son état de mise en service seront également décrits.

Particle Accelerator Activities at the Paul Scherrer Institut

Terence GARVEY,

Paul Scherrer Institut, Villigen, Switzerland 5232, Division of Large Research Infrastructures Abstract The Paul Scherrer Institute operates two accelerator complexes as user facilities for a large community of researchers. These are the High Intensity Proton Accelerator (HIPA) and the Swiss Light Source (SLS). HIPA is a 590 MeV proton cyclotron. It is used to produce neutrons by spallation for research in condensed matter physics and to produce muons and other secondary particles for research in magnetism and particle physics. The Swiss Light Source is a 2.4 GeV electron storage ring used as a 3rd generation synchrotron radiation source providing photons for a broad range of scientific disciplines. In addition to these two facilities the Institute is currently progressively bringing into operation an X-ray Free Electron Laser, SwissFEL, which will provide researchers with intense femtosecond pulses from hard (ARAMIS) and soft (ATHOS) X-ray beam lines. The Institute also operates a 250 MeV superconducting cyclotron (COMET) for the purposes of proton therapy. The therapy center has recently been equipped with a third rotating gantry which is currently being commissioned. An up-grade of the HIPA RF system and plans for “SLS-2” will be presented. The SwissFEL accelerator and its commissioning status will also be described.

Introduction

We will present an overview of the different activities in progress at the Paul Scherer Institut (PSI) in the field of particle accelerators. The treatment is not intended to be ‘complete’ or highly detailed but aims rather to give the reader a flavour of the accelerators currently operating as user facilities, as well as those activities aimed at their improvement, or ‘up-grade’.

There are currently two operating user facilities (i) the High Intensity Proton Accelerator (HIPA) and (ii) the Swiss Light Source (SLS). In addition to these two research facilities, PSI operates a superconducting cyclotron (COMET) for proton therapy.

There is also a new accelerator currently under construction, the Swiss Free Electron Laser (SwissFEL) linac. As will be seen below, this facility is, in practice, divided between two distinct research sectors; the first, ARAMIS, for research with hard x-rays; and the second, ATHOS, for research with soft x-rays. As the heart of the SwissFEL facility is a radio-frequency (RF) linac, we shall restrict our description of the new machine essentially to the RF linac.

We will complete the overview by discussing some of the accelerator research activities which have taken place in parallel with accelerator operation and construction. All that will be described here has been reported previously and an extensive reference list is given for the interested reader.

The Ring Cyclotron

The oldest accelerator on the PSI site is the Ring Cyclotron of the HIPA complex. This is a 590 MeV, sector- type, proton cyclotron initially built in 1974 as a meson factory for studies in high-energy physics. The cyclotron has an injection chain consisting of an 870 keV Cockroft-Walton accelerator and a 72 MeV injector cyclotron, Injector-II (so named as it replaced an earlier commercial injector cyclotron of weaker intensity). Since 1996, the Ring Cyclotron provides beam to a tungsten target (SIN-Q) producing neutrons by spallation for research in neutron scattering. However, the machine continues to provide beams of secondary particles for precision studies in particle physics at low energies. In addition, muons produced from the decay of these secondary particles are used as probes for the study of the magnetic properties of materials within the Laboratory for Muon Spin Spectroscopy. The nominal current provided by the ring cyclotron is 2.2 mA corresponding to a beam power of ~ 1.3 MW. Approximately two hundred turns are required by the beam to reach full energy. The relative loss of beam power, essentially at the extraction elements is ~ 10-4. These losses result in activation of machine components rendering maintenance difficult. Currently a shut-down of approximately four months each year is required to maintain the machine while limiting radiation exposure to service personnel below regulatory limits. A view of the Ring Cyclotron is shown in Fig.1. The machine consists of 8 main magnets with four 50 MHz accelerating resonators and one 150 MHz “flat-top” resonator interspersed between them.

Figure 1: View of the Ring Cyclotron.

An up-grade of HIPA has been taking part in stages in order to increase the beam current to 3 mA (1.8 MW beam power). The major part of the up-grade is concerned with a replacement of much of the RF power equipment. The beam-current in cyclotrons is known to scale with the inverse cube of the number of turns required to reach extraction energy [1]. The fewer the number of turns the greater is the distance between proton bunches on adjacent turns and thus the beam can be extracted more cleanly without particle loss. Fewer turns implies the need for increased energy gain per turn. To this end four new copper resonators were installed on the ring cyclotron to increase the energy per turn. The last of these new resonators was installed in 2008. A photo of one such resonator, prior to installation in the ring, is shown in Fig.2a. New resonators for Injector II have also been built, Fig.2b, and installation of these will take place in the near future. A detailed description of the RF up-grade plans for the HIPA complex can be found in [2].

Figure 2a: One of the 50 MHz resonator of the ring cyclotron. Figure 2b: New resonator for the Injector II cyclotron

The Swiss Light Source

The Swiss Light Source is a 2.4 GeV, 400 mA electron storage ring for the production of synchrotron radiation. Synchrotron radiation sources have emerged as one of the most important tools for the study of the structure of materials at the atomic, molecular and mesosopic scales today [3]. The SLS serves a broad research community in the fields of condensed matter physics, material sciences, biology and chemistry. The storage ring has an injection complex consisting of a 100 MeV S-band linac and a full energy booster synchrotron. The machine has been in operation since 2001 [4]. The full energy booster permits “top-up” operation of the storage ring, whereby current losses due to finite beam life-time can be replenished by injecting additional current at given intervals of time. In practice the injection of ~ 1 mA every ~ 100 seconds maintains the current effectively constant, Fig. 3. The constant beam current, and, consequently, constant heat-load on optical elements of the synchrotron radiation beam lines is an important element in the positional stability of the photon beams received by users. In addition to top-up operation, a fast-orbit feedback system taking information from high precision beam position monitors ensures that the stability of both position (sub-micron) and intensity of the SLS beam satisfies the needs of its user community. The SLS demonstrates an excellent reliability. The availability of the machine is typically 99% of its projected 5’000 hours per year user time with mean times between failures of ~ 135 hours (2015 figures). A schematic of the SLS storage rings and its associated beam lines is shown in Fig. 4.

Figure 3. Variation of beam current and vertical beam size in the SLS during user operation.

Currently, many operating light sources are planning to up-grade to higher brightness machines through the installation of new lattices employing multi-bend achromats. These new diffraction limited light sources (DLSR) produce electron beams with two orders of magnitude reduction in emittance and with a corresponding increase in photon beam brightness. In order to remain competitive with other sources it is imperative that the SLS should up-grade to a DLSR. A novel lattice layout has been elaborated which allows a factor of 40 emittance reduction on the SLS despite its relatively small circumference (288 m) in comparison with other 3rd generation light sources [5]. A Conceptual Design (CDR) report for SLS-2 has recently been published [6]. The CDR has been enthusiastically received by an international committee of experts who recommend that we should now proceed to a full Technical Design Report. A comparison of the SLS and SLS-2 beam parameters is given in Table 1 [7].

Figure 4: Schematic of the SLS layout and beamlines (image provided by A. Streun).

Table 1. Comparison of present SLS beam parameters with SLS-2 design values at 2.4 GeV and 400 mA. The SLS-2 emittance includes the contribution of intra-beam scattering. SLS SLS-2 Emittance (pm) 5069 126 Lattice type TBA 7BA Circumference (m) 288 290.4

Tunes Qx/y 20.43/8.22 39.2 / 15.3 Natural chromaticities x/y -67/-19.8 -95 / -35.2 Horizontal damping partition 1.00 1.71 Momentum compaction factor (10-4) 6.56 - 1.33 Radiated power (kW) 208 222 RMS energy spread (10-3) 0.86 1.07 Damping times x/y/E (ms) 8.9/8.9/4.4 4.9/8.4/6.5

The COMET cyclotron

In addition to accelerators for basic research, PSI operates an accelerator dedicated to proton therapy [8]. The advantage of protons, with respect to X-rays, for the irradiation of tumors is well documented [9]. The existence of the “Bragg peak” whereby protons deposit the majority of their energy close to the end of their range in matter allows one to deliver the dose precisely at the location of the tumour while reducing the dose to the healthy cells which surround it, Fig. 5.

Figure 5: Comparison of the dose distribution in matter between proton and X-ray beams.

The accelerator used for therapy is a 250 MeV superconducting cyclotron, COMET. The machine was manufactured by the company ACCEL Instruments GmbH in collaboration with PSI. Although the accelerator was purchased commercially, the beamlines which allow protons to be delivered to patients were built by PSI. A “degrader” is used to allow the proton energy to be adjusted down to 70 MeV, in order to provide a ‘spread-out’ Bragg peak adapted to match the spatial extent of a tumour. Beamline components which allow transport of the beam to the patient are mounted on gantries which can rotate around the patient thus allowing multiple treatment ‘fields’ to be used. Until recently the accelerator provided beams to two gantries and to a dedicated beamline, OPTIS, for the treatment of ocular tumours. A schematic layout of the currently operating proton therapy facility is shown in Fig 6.

Figure 6: Schematic layout of the proton therapy center at PSI.

Recently, a third gantry, appropriately named Gantry-3, financed by the Canton of Zürich, has been installed which will deliver beam to its first patient in spring of 2018 [10]. For illustration, a photo of Gantry-2 is shown in Fig 7. It is worth noting that these gantries are physically very large which makes their production relatively expensive, a point to which we shall return later.

Figure 7: Gantry-2

The SwissFEL Linac

The construction of the Swiss Free Electron Laser at PSI was motivated by a very strong scientific case prepared by potential users who wanted access to Ångstrom scale-length wavelengths coupled with femtosecond resolution timing and extremely high brightness [11]. The project was to be built as a national, rather than international, facility and thus within strict financial constraints. As the linac is a cost driver, for which cost is roughly proportional to the linac energy, the energy was chosen to be as small as possible while being compatible with operation at 1 Å. In the case of SwissFEL the chosen energy was 5.8 GeV. Additional constraints were the length of the facility (< 900 m) and the average power consumption (< 5 MW).

The basic building blocks of a SASE FEL are; (i) a high brightness electron source, (ii) an RF linac to accelerate the electron beam to the required energy, (iii) a series of undulators in which to generate the X-rays, (iv) high quality X-ray transport and focusing system, and (v) experimental end stations to exploit the X-ray beams. High brightness electron sources and short period undulators were identified as areas requiring R&D whereas it was initially felt that the linac could be built with radio-frequency structures commercially available at S-band frequency. Novel short period undulators (λu = 15 mm, K = 1.2) where developed, allowing hard X-ray wavelengths to be generated at comparatively modest linac energy in accordance with the FEL resonance equation;

λ  K 2  λ = u  +  2 1  2γ  2 

A low emittance RF photo-injector gun was designed (0.2 mm-mrad normalized has been measured in tests) allowing high coherence at wavelengths down to 1 Å (ε ~ λ/4π).

Before proceeding to build SwissFEL, an Injector Test Facility (SITF) was designed and built to develop some of the technologies (low emittance source, diagnostics, RF phase and amplitude control..) needed for the SASE laser and to demonstrate some of the beam properties (slice emittance, energy spread, bunch length compression) [12,13]. The Injector was operated from 2010 to 2014, after which it was closed to free up resources for the main project. A view looking downstream of the Test Facility is shown in Fig.8. A detailed summary of the beam studies carried out, and the results obtained, on the SITF can be found in [14].

Figure 8: Downstream view of the SwissFEL Injector Test Facility

The radio-frequency linac

A schematic of SwissFEL is shown in Fig. 9 with the main nominal beam parameters in table 2. The lower (purple) beam line in the figure, ARAMIS, is built for hard X-ray experiments and is currently being commissioned. The upper (red) beam line, ATHOS, is for soft X-ray experiments and will be discussed below. The RF systems of the linac consist of the 2.5 cell RF photo-injector gun (S-band) followed by the injector, comprising six S-band RF structures. The main linac is divided into three parts (Linacs, 1, 2 and 3) containing, respectively, thirty-six, sixteen and fifty-two C-band accelerating structures (see below). Magnetic bunch compression is applied after the injector and after Linac 1 in order to reach the high peak current required for SASE operation. Two X-band structures are used at the end of the injector to linearize the longitudinal phase space before bunch compression. Wavelength changes on ARAMIS are made by adjusting the electron beam energy gain in linacs 2 and 3.

Figure 9: Schematic of the SwissFEL linac and the hard (ARAMIS) and soft (ATHOS) X-ray beamlines.

Table 2. Beam parameters for SwissFEL linac and wavelength ranges for ARAMIS and ATHOS beam-lines

Early developments of an electron source for SwissFEL centered on a laser assisted field emission source [15]. However, after LCLS had demonstrated that 1 Å lasing could be achieved with an RF gun, studies began on an RF gun for SwissFEL [16]. The gun developed, incorporated ideas from both the LCLS design and the PHIN gun developed at LAL-Orsay for CTF, such as elliptical irises to minimize surface fields, racetrack iris shape to reduce quadrupole field components and symmetric feeding of RF power to the gun from left and right [17]. A large iris thickness ensures 15 MHz separation between the zero and π modes of the cavity. A schematic of the gun is shown in Fig. 10.

Figure 10: 3-D CAD drawing of the SwissFEL RF gun.

The six S-band (3 GHz) travelling-wave structures, providing an injector energy of 350 MeV, were purchased from Research Instruments following a PSI design [18]. Although it was initially felt that similar structures could be used for the main linac, a comparison of construction cost and ten-year operational cost showed a clear economic advantage in favour of C-band (5.7 GHz) structures. Consequently, it was decided to build the main linac from C-band structures. However, C-band structures are not so readily available commercially. Therefore, the decision was taken to design and develop a new two-meter long structure in-house [19]. The structure cells were precision machined in industry. After reception at PSI the cells were robotically stacked vertically and then vacuum brazed in a dedicated oven procured for this purpose. The cell fabrication was performed with sufficient precision (sub-micron) such as to avoid the need for post-braze “dimple” tuning cell-by-cell. This was an audacious choice and is a notable success for the project. The complete linac requires 102 two-meter long structures. The production of over 100 structures from more than 10’000 cells was a major enterprise for the laboratory. The last structure was produced in August 2016 and the last girder installed in the machine tunnel the following month. A view of the structures in the SwissFEL tunnel can be seen in Fig. 11. The basic SwissFEL RF modules consists of four structures powered by a single 50 MW / 3µs pulse Toshiba klystron, driven by a solid-state modulator (370 kV, 344 A) with 20 ppm voltage stability at 100 Hz repetition rate. The klystron output fills the RF structures via an RF pulse compression structure (of the Barrel Open Cavity type), designed and built at PSI [20]. Thus the C-band linac consists of 26 such modules. The nominal structure operating gradient for SwissFEL is 28 MV/m. However, they have been comfortably conditioned as high as 52 MV/m with a breakdown rate of ~2×10-6 per pulse [21]. At the nominal gradient the breakdown rate is well below the specified 1×10-8 per pulse.

The higher gradients of C-band w.r.t. S-band also allow some reduction in facility length. However, this is not a strong motivation for the choice of C-band. The active acceleration length of the SwissFEL C-band structures is 208 meters. However, with 60 m of undulators, 50 m needed for experimental areas, 100 m of photon transport (distance required to allow the photon beam to diverge enough not to damage the first mirror) and ~ 300 m of other electron beam line elements, including the injector, the C-band structures amount to only ~ 25% of the length of the building! This is an important point for advocates of high gradient schemes such as plasma accelerators for ‘compact’ FEL’s. A reduction of the linac length, even to zero, would permit only 25% reduction in building length.

Figure 11: C-band structures installed in the SwissFEL tunnel. An RF pulse compression structure is visible to the upper left of the picture.

The undulator

Although we have been concentrating on RF systems it would be remiss not to mention the other element of the machine developed for SwissFEL. The ARAMIS undulator, U15, is an in-vacuum undulator producing linearly polarized radiation [22]. Although wavelength changes are mainly foreseen with changing electron beam energy the ARAMIS undulator has sufficient gap variation to allow ~ 5% wavelength change. The 4 m long undulator consists of 1060 permanent magnet blocks. The nominal value, of the undulator parameter, K = 1.2, is obtained for an undulator gap of 4.2 mm. Thirteen undulators are now installed in the SwissFEL tunnel, Fig 12.

Figure 12: U15 undulators installed in the SwissFEL tunnel (upstream view).

SwissFEL commissioning

At the time of writing SwissFEL has not yet reached its full energy. Although all elements of the accelerator and the ARAMIS beam line are in place, not all of the RF power is installed due to delays in the delivery of the modulators. The linac is being commissioned progressively however at ever increasing energies. A first lasing was obtained on the 2nd December 2016. The beam energy was 345 MeV (half the injector plus one C-band module). Even so a first gain curve could be measured using a silicon-diode detector, Fig. 13(l). The wavelength was too long to be measured on any available spectrometer but from the beam energy and the undulator parameter it could be calculated to be 24 nm. This modest initial success was obtained in time to inaugurate the facility on the morning of the 5th December 2016 in the presence of the President of the Swiss Confederation. In May of this year a second gain curve was measured within the wavelength range of the ARAMIS beam line, 4.1 nm using an electron beam of 910 MeV, Fig 12(r). At the time of writing (October 2016) SwissFEL has lased down to 1.3 nanometres.

Fig 13: Gain curves obtained during SwissFEL commissioning: (l) at 345 MeV beam energy, (r) at 910 MeV energy.

We have not discussed many interesting developments that have been made in advanced diagnostics, low level RF phase and amplitude control, femtosecond timing and synchronisation, among others, which have been developed for SwisssFEL. For further details of these and the SwissFEL facility the reader is referred to reference [23] and to references therein.

ATHOS

As illustrated schematically in Fig. 9, the SwissFEL linac will provide beam to a second undulator array for the production of soft X-rays, in the wavelength range 0.65 nm to 5.2 nm. A pulsed kicker system is used to deflect the beam coming from linac 2 to the second (ATHOS) beamline, parallel with ARAMIS but laterally displaced by 3.75 m. The beam line contains one single C-band module which is used as an energy “vernier” to vary the electron beam energy by ± 250 MeV through appropriate phasing of the RF structures. By adjusting the beam energy at the switch-yard from 2.9 GeV to 3.15 GeV a total energy excursion from 2.65 GeV to 3.4 GeV is available on the ATHOS beamline. The ATHOS undulators can be adjusted in K vale from 1.0 to 3.6 and the combination of this K range with the electron beam energy range allows one to cover the required soft X-ray wavelength range. The ATHOS undulators have a periodicity of 38 mm. They are of the APPLE X type and allow complete control of the polarisation of the X-ray beam [24]. The arrangement of the undulators is designed to allow several novel operating modes which are discussed in [24] and in references therein. Downstream of the undulators an X-band (~ 12 GHz) deflecting structure will be used to measure the temporal structure of the beam emerging from the undulator. ATHOS will provide its first beam to users in 2020.

PSI and the European X-FEL

The Accelerator Division of PSI is also very much involved with the European collaboration building the European X-FEL in Germany. Indeed the Swiss “in-kind”contribution to this ambitious project has been delivered by the Diagnostics section of PSI [25, 26]. Approximately 460 beam position monitor electronics, for both cavity BPMs and button-type BPMs, were delivered to the project. In addition, for the transverse intra- bunch feedback (IBFB) system the pulsed kickers were designed, tested and delivered. The RF pulse amplifiers for the IBFB system were also specified, procured, tested and delivered to E-XFEL. In contrast to SwissFEL, E- XFEL employs superconducting RF cavities installed in cryogenic modules. A view through the E-XFEL tunnel is shown in Fig. 14.

Figure 14: View through the European XFEL tunnel.

Accelerator R&D activities at PSI

The need to operate and maintain the SLS, HIPA and COMET while simultaneously constructing SwissFEL largely exhausts the resources of PSI’s accelerator division in terms of manpower and budget. Therefore R&D activities have been, by necessity, limited during the SwissFEL construction period. Nevertheless, through the provisions of grants from Swiss National Funding (SNF), the Swiss Commission for Technology and Innovation (CTI) and through European Union supported programs (such as TIARA and EuCARD) a respectable level of R&D continues. Recent grant awards ACHIP and CHART have also made funds available for R&D work. Some examples of recent R&D activities are given in this section.

CHART - Swiss (CH) Accelerator Research and Technology

The SBFI (Swiss Ministry for Education, Research and Innovation) has made funding available for accelerator research under the program CHART – Swiss (CH) Accelerator Research and Technology. This program will support R&D for the Future Circular Collider project including beam dynamics studies and the development of superconducting magnet technology. This last topic has synergies with PSI projects as superconducting magnets are strongly considered for the SLS-2 project discussed above. At least some SLS-2 users would be interested in having the hardest X-ray wavelengths possible and, for this, a design of a 6 T dipole super-bend has been elaborated, Fig 15 [27]. Superconducting magnets are also under consideration for future proton therapy gantry designs as they have the promise of making possible lighter and cheaper gantries than that shown in Fig 7 [28]. A second topic covered by CHART is the development of high field, laser driven, acceleration using THz radiation. This again has strong synergies with another R&D collaboration, ACHIP (see below), in which PSI is a participant. Several institutes including CERN, EPFL, PSI, the ETH-Zurich and the University of Geneva collaborate on CHART activities.

Figure 15. (Left) Schematic of a 6T superconducting longitudinal gradient bend dipole; (Right) Magnetic field profile along the beam axis (image provided by C. Calzolaio)

ACHIP

The ACHIP (Accelerator on a CHIP) collaboration has obtained a 13.5 M$ grant from the Moore Foundation to carry out research on acceleration using minute dielectric structures powered by THz radiation [29]. This collaboration of US and European Universities and Institutes includes the EPFL and PSI. Such structures are of interest to PSI not only for acceleration but also for transverse “streaking” of short electron bunches to determine their temporal profile. Presently electron bunch lengths on linac driven FELs are measured using transverse deflecting cavities at RF frequencies. However, when attosecond pulses become available, these structures will no longer have the required resolution and THz structures could be an interesting alternative. The PSI contribution to ACHIP includes experiments carried out on the SwissFEL linac. First experiments on the injector have already taken place and future tests are foreseen on the ATHOS beamline for 2019 [30]. The COSAMI study

There is a general consensus within the semiconductor community that Extreme Ultra-Violet Lithography (EUVL) will be the next generation high volume manufacturing technique for producing smaller and faster integrated circuits. Advances in multi-layer silicon-carbide mirrors with high reflectivity (~70%) and large (2%) bandwidth make 13.5 nm the wavelength of choice for this technology. The development of metrology methods at EUV wavelengths for mask inspection will be indispensable for the success of EUVL. A mask inspection tool is currently being developed on an SLS beamline with the support of private industry. However, the development of such a tool only makes sense if a source of EUV radiation, having the required properties, can be built and operated in an industrial environment at an acceptable cost. Following an initial study by Wrulich et. al. [31], we have carried out a study (supported by a CTI grant) of a compact (5x12 m2) synchrotron radiation source that could potentially meet the needs of industry. The source is made compact by having the booster synchrotron and the storage ring mounted on a common support, at different heights, in a quasi-concentric fashion. The injector linac and transfer line are built within the perimeter of the rings in order to minimize the source “foot-print”. To provide a small emittance (~ 10 nm-rad) the technique currently being applied to DLSR designs, i.e. the use of a multi-bend achromat is employed. The optical functions of the storage ring are shown in Fig.16 [32]. The storage ring energy is 430 MeV and the required brightness is obtained with a beam of 200 mA with an undulator 3.2 m in length and with a periodicity of 16 mm. The undulator design is based on experience and technology of the ARAMIS undulator. A three-dimensional CAD image of the COSAMI complex is shown in Fig. 17 illustrating the storage ring superimposed above the booster and with the injector linac and transfer line nestled inside the rings. Further details of the COSAMI project can be found in references [33, 34].

0.3 14 0.2

12 0.1

10 0.0

-0.1 8 -0.2 6

-0.3 Dispersion [m] Betafunctions [m] 4 -0.4

2 -0.5

-0.6 0 0 5 10 15 20 25

Figure 16: Storage ring optical functions (β , β , D , D ) x y x y

Figure 17: Three-dimensional CAD image of COSAMI

X-band structure developments

By virtue of FORCE/FLARE grants from the SNF we have been able to participate in the development of X- band RF structures for the CLIC linear collider project led by CERN. As part of this work we have financed the construction of a passive structure, CLASSE (CLIC Accelerating Structure Setup), used for experiments at FACET (SLAC) to map out the structure wakefields. The structure in question contained silicon damping loads integrated into the structure cells to dampen the excitation of higher order modes excited by the beam and reduce the wakefield amplitude to an acceptable value for the bunch train employed in the CLIC design [35]. Details of the structure design and the simulations can be found in [36]. The experiments at FACET yielded results well in agreement with simulations [37]. The CLASSE structure consisted of a string of six 26-cell sections one of which is shown in Fig. 18.

Figure 18: One section of the CLASSE structure used for wakefield measurements at FACET.

Following the successful production of C-band structures for SwissFEL, with their relatively fast conditioning and low breakdown rate we have been producing X-band structures as part of the PSI-CERN collaboration on CLIC R&D. The approach is to build an X-band structure according to the CERN design but to fabricate it, using as much as is possible, the production protocol of the C-band structures. Two structures have been built supported by an SNF-FLARE grant. A schematic of the brazed structure awaiting a low power bead-pull test is shown in Fig. 19. Both structures are currently under high power test. Preliminary results from the first were reported in [38].

Figure 19: Prototype X-band structure ready for high power test.

Solid-state amplifier developments

As is the case of most synchrotron radiation sources, the booster synchrotron and the SLS storage ring uses klystron amplifiers to power the RF cavities. The SOLEIL-synchrotron has pioneered the use of solid state amplifiers for their light source. In these devices a large number of transistor modules, delivering a few hundred watts of power, are combined to provide several tens of kilowatts. The advantages of solid-state amplifiers are numerous; no need for a focusing solenoid as in the case of the klystron; no need for the extensive lead shielding for the spent beam of a klystron; no need for a vacuum system in which to propagate the klystron beam; no need for heated filaments for the klystron gun; no need for the high voltage power supply of the klystron gun; and, perhaps most important of all, the solid-state amplifier can continue to run even in the event of a failure of one transistor module. This last advantage is in contrast to the klystron amplifier where a failure of the solenoid, filament, vacuum or high voltage supply means that the klystron is down. The eventual use of solid-state amplifiers for SLS has been under discussion for some time. However stronger priorities meant that developments were limited to producing a transistor providing 350 W at 500 MHz and to manufacturing co- axial combining elements. With the support of a CTI grant it was possible to develop a full scale amplifier suitable for application to the SLS. Detailed information on the PSI amplifier can be found in [39, 40]. The project was carried out in collaboration with a Swiss industry partner. The amplifier has been fully tested on the SLS booster and will eventually be employed there routinely, making the klystron available for powering a 500 MHz cavity test bench. The industry partner has since obtained an order for the delivery of solid-state amplifiers to a major European light source. The amplifier is shown in Fig 20.

Figure 20: Solid-state amplifier developed at PSI

Closing remarks

With the Swiss Light Source and the Ring Cyclotron, PSI provides users with beams of photons, neutrons and muons for fundamental research. Timely up-grades to the SLS and HIPA will ensure that PSI users will continue to have access to first class research infrastructures. The Swiss Free Electron Laser will add femtosecond X-ray beams to the available probes for the study of matter. The superconducting cyclotron COMET provides proton beams for patient treatment. In addition, the laboratory pursues a healthy and vigorous program of accelerator R&D.

References

1. W. Joho, “High Intensity Problems in Cyclotrons”, Proceedings of the 9th International Conference on Cyclotrons and their Applications (Caen), pp337 – 347, 1981. http://accelconf.web.cern.ch/AccelConf/c81/papers/ei-03.pdf 2. L. Stingelin, M. Bopp, M. Broennimann, J. Cherix, H. Fitze, M. Schneider and W. Tron, “High Power RF Systems and Resonators for Sector Cyclotrons”, Proceedings of the 19th International Conference on Cyclotrons and their Applications (Lanzou), pp332 – 337, 2010. http://accelconf.web.cern.ch/AccelConf/Cyclotrons2010/papers/wem2cio01.pdf 3. G. Garcia, T. Garvey and L. Rivkin, Accelerators and Photon Sources in “Applications of Particle Accelerators in Europe”, EuCARD-2 monograph, pp 88 -101, 2017. http://eucard2.web.cern.ch/ 4. A. Luedeke, A. Andersson, M. Boege, B. Kalantari, B. Keil, M. Pedrozzi, T. Schilcher, V. Schlott and A. Streun, “Status of the Swiss Light Source”, Proceedings of the Tenth European Particle Accelerator Conference (Edinburgh), pp3424 – 3426, 2006. http://accelconf.web.cern.ch/AccelConf/e06/PAPERS/THPLS061.PDF 5. A. Streun in “Future Research Infra-structures: Challenges and Opportunities”, Proceedings of the International School of Physics “Enrico Fermi”, S. Bertolucci and L. Palumbo (Eds.), 2017. ISBN 978-1-61499-731-3. https://www.iospress.nl/book/future-research-infrastructures-challenges-and-opportunities/ 6. A. Streun (Ed.), Masamitsu Aiba, Alexander Anghel, Uwe Barth, Michael Böge, Ciro Calzolaio, Marco Calvi, Alessandro Citterio, Micha Dehler, Karsten Dreyer, Terence Garvey, Christopher Gough, Michael Hahn, Dominique Hauenstein, Juri Honegger, Boris Keil, Philippe Lerch, Stephan Maag, Fabio Marcellini, Marco Negrazus, Beat Ronner, Stephane Sanfilippo, Corina Sattler, Volker Schlott, Thomas Schmidt, Lothar Schulz, Lukas Stingelin, Vjeran Vranković, Johan Wickström, Albin Wrulich, Elmar Zehnder, Elke Zimoch “The SLS-2 Conceptual Design Report”, PSI report 17-03. http://www.lib4ri.ch/institutional-bibliography/psi/psi-berichte.html 7. A. Streun, T. Garvey, L. Rivkin, V. Schlott, T. Schmidt, P. Willmott and A. Wrulich, “SLS-2 The Up- grade of the Swiss Light Source” paper submitted to Journal of Synchrotron Radiation, 2017. 8. J.M. Schippers, J. Duppich, G. Goitein, M. Jermann, A. Lomax, E. Pedroni, H. Reist, B. Timmermann and J. Verweij, “The use of protons in cancer therapy at PSI and related instrumentation”, Journal of Physics: Conference Series, 41, pp61 -71, 2006. http://iopscience.iop.org/article/10.1088/1742-6596/41/1/005/pdf 9. R.R. Wilson, “Radiological Use of Fast Protons”, Radiology, Vol. 47, pp 487 – 491, 1946. https://doi.org/10.1148/47.5.487 10. A. Koschik, C. Baumgarten, C. Bula, J. Duppich, A. Gerbershagen, M. Grossmann, V. Rizzoglio, J. Welte, “PSI Gantry 3:Integration of a New gantry into an Existing Proton Therapy Facility”, Proceedings of the 7th International Particle Accelerator Conference (Busan, Korea) pp 1927 – 1929 (2016). http://accelconf.web.cern.ch/AccelConf/ipac2016/html/author.htm 11. B.D. Patterson (Ed.), “Ultrafast Phenomena at the Nanoscale: Science opportunities at the SwissFEL X-ray Laser”, PSI Report 09-10, 2009. https://www.psi.ch/swissfel/CurrentSwissFELPublicationsEN/SwissFEL_Science_Case_small.pdf 12. Y. Kim, A. Adelmann, R. J. Bakker, M. Dehler, R. Ganter, T. Garvey, A. Oppelt, M. Pedrozzi, J.-Y. Raguin, L. Rivkin, V. Schlott, A. Streun, F. Stulle and A. F. Wrulich, “Start-to-end Simulations of the PSI 250 MeV Injector Test Facility”, Proceedings of the 11th European Particle Accelerator Conference (Genoa), pp 100-102, 2008. http://accelconf.web.cern.ch/AccelConf/e08/papers/mopc015.pdf 13. M. Pedrozzi (Ed.), “The SwissFEL Injector Conceptual Design Report – Accelerator Test facility for SwissFEL”, PSI Report Nr. 10-05, July 2010. https://www.psi.ch/swissfel/HomeEN/SwissFEL_Injector_CDR_310810.pdf 14. T. Schietinger, M. Pedrozzi, M. Aiba, V. Arsov, S. Bettoni, B. Beutner, M. Calvi, P. Craievich, M. Dehler, F. Frei, R. Ganter, C. P. Hauri, R. Ischebeck, Y. Ivanisenko, M. Janousch, M. Kaiser,B. Keil, F. Löhl, G. L. Orlandi, C. Ozkan Loch, P. Peier, E. Prat, J.-Y. Raguin, S. Reiche, T. Schilcher, P. Wiegand, E. Zimoch, D. Anicic, D. Armstrong, M. Baldinger, R. Baldinger, A. Bertrand, K. Bitterli, M. Bopp, H. Brands, H. H. Braun, M. Brönnimann, I. Brunnenkant, P. Chevtsov, J. Chrin, A. Citterio, M. Csatari Divall, M. Dach, A. Dax, R. Ditter, E. Divall, A. Falone, H. Fitze, C. Geiselhart, M.W. Guetg, F. Hämmerli, A. Hauff, M. Heiniger, C. Higgs, W. Hugentobler, S. Hunziker, G. Janser, B. Kalantari, R. Kalt, Y. Kim, W. Koprek, T. Korhonen, R. Krempaska, M. Laznovsky, S. Lehner, F. Le Pimpec, T. Lippuner, H. Lutz, S. Mair, F. Marcellini, G. Marinkovic, R. Menzel, N. Milas, T. Pal, P. Pollet, W. Portmann, A. Rezaeizadeh, S. Ritt, M. Rohrer, M. Schär, L. Schebacher, St. Scherrer, V. Schlott, T. Schmidt, L. Schulz, B. Smit, M. Stadler, B. Steffen, L. Stingelin, W. Sturzenegger, D.M. Treyer, A. Trisorio, W. Tron, C. Vicario, R. Zennaro, and D. Zimoch, “Commissioning experience and beam physics measurements at the SwissFEL Injector test Facility”, Phys. Rev. Accel. Beams 19 100702 (2016). https://journals.aps.org/prab/pdf/10.1103/PhysRevAccelBeams.19.100702 15. R. Ganter, B. Beutner, S. Binder, H.H. Braun, T. Garvey, C. Gough, C. Hauri, R. Ischebeck, S. Ivkovic, F. Le Pimpec, K. Li, M. L. Paraliev, M. Pedrozzi, T. Schietinger, B. Steffen, A. Trisorio, and A. Wrulich, “Electron Beam Characterization of a Combined Diode RF Electron Gun”, PRST-AB 13, 093502 (2010). https://journals.aps.org/prab/pdf/10.1103/PhysRevSTAB.13.093502 16. J.Y. Raguin, M. Bopp, A. Citterio, A. Scherrer, “The SwissFEL RF Gun: RF Design and Thermal Analysis”, Proceedings of the 26th International Linear Accelerator Conference (Tel-Aviv) pp 442-444, 2012. http://accelconf.web.cern.ch/AccelConf/LINAC2012/papers/tuplb01.pdf 17. R. Roux, “Conception of Photoinjectors for the CTF3 experiment”, International Journal of Modern Physics A, Vol. 22, No. 22, pp 3925 – 3941(2007). https://doi.org/10.1142/S0217751X07037524 18. J.Y. Raguin, “The SwissFEL S-Band Accelerating Structure: RF Design”, Proceedings of the 26th International Linear Accelerator Conference (Tel-Aviv) pp 498-500, 2012. http://accelconf.web.cern.ch/AccelConf/LINAC2012/papers/tupb011.pdf 19. J.Y. Raguin and M. Bopp, “The SwissFEL C-Band Accelerating Structure: RF Design and Thermal Analysis”, Proceedings of the 26th International Linear Accelerator Conference (Tel-Aviv) pp 501-503, 2012. http://accelconf.web.cern.ch/AccelConf/LINAC2012/papers/tupb012.pdf 20. R. Zennaro, M. Bopp, A. Citterio, R. Reiser, T. Stapf,”C-Band RF Pulse Compressor for SwissFEL”, Proceedings of the 4th International Particle Accelerator Conference (Shanghai), pp 2827-2829, (2013). http://accelconf.web.cern.ch/AccelConf/IPAC2013/papers/wepfi059.pdf 21. F. Loehl, J. Alex, H. Blumer, M. Bopp, H. Braun, A. Citterio, U. Ellenberger, H. Fitze, H. Joehri, T. Kleeb, L. Paly, J. Raguin, L. Schulz, R. Zennaro, “Status of the SwissFEL C-Band Linear Accelerator”, Proceedings of the 35th International Free Electron laser Conference ( New York), pp 317- 321, 2013. http://accelconf.web.cern.ch/AccelConf/FEL2013/papers/tupso43.pdf 22. M.Calvi, M.Brügger, S.Danner, A.Imhof, H. Jöhri, T.Schmidt and C.Scoular, “SwissFEL U15 Magnet Assembly: First Experimental Results”, Proceedings of the 34th International Free Electron laser Conference (Nara, Japan), pp 662 – 665, 2012. http://accelconf.web.cern.ch/AccelConf/FEL2012/papers/thpd63.pdf 23. Christopher J Milne, Thomas Schietinger, Masamitsu Aiba, Arturo Alarcon, Jürgen Alex, Alexander Anghel, Vladimir Arsov, Carl Beard, Paul Beaud, Simona Bettoni, Markus Bopp, Helge Brands, Manuel Brönnimann, Ingo Brunnenkant, Marco Calvi, Alessandro Citterio, Paolo Craievich, Marta Csatari Divall, Mark Dällenbach, Michael D’Amico, Andreas Dax, Yunpei Deng, Alexander Dietrich, Roberto Dinapoli, Edwin Divall, Sladana Dordevic, Simon Ebner, Christian Erny, Hansrudolf Fitze, Uwe Flechsig, Rolf Follath, Franziska Frei, Florian Gärtner, Romain Ganter, Terence Garvey, Zheqiao Geng, Ishkhan Gorgisyan, Christopher Gough, Andreas Hauff, Christoph P Hauri, Nicole Hiller, Tadej Humar, Stephan Hunziker, Gerhard Ingold, Rasmus Ischebeck, Markus Janousch, Pavle Juranić, Mario Jurcevic, Maik Kaiser, Babak Kalantari, Roger Kalt, Boris Keil, Christoph Kittel, Gregor Knopp, Waldemar Koprek, Henrik T Lemke, Thomas Lippuner, Daniel Llorente Sancho, Florian Löhl, Carlos Lopez-Cuenca, Fabian Märki, Fabio Marcellini, Goran Marinkovic, Isabelle Martiel, Ralf Menzel, Aldo Mozzanica, Karol Nass, Gian Luca Orlandi, Cigdem Ozkan Loch, Ezequiel Panepucci, Martin Paraliev, Bruce Patterson, Bill Pedrini, Marco Pedrozzi, Patrick Pollet, Claude Pradervand, Eduard Prat, Peter Radi, Jean-Yves Raguin, Sophie Redford, Jens Rehanek, Julien Réhault, Sven Reiche, Matthias Ringele, Jochen Rittmann, Leonid Rivkin, Albert Romann, Marie Ruat, Christian Ruder, Leonardo Sala, Lionel Schebacher, Thomas Schilcher, Volker Schlott, Thomas Schmidt, Bernd Schmitt, Xintian Shi, Markus Stadler, Lukas Stingelin, Werner Sturzenegger, Jakub Szlachetko, Dhanya Thattil, Daniel M Treyer, Alexandre Trisorio, Wolfgang Tron, Seraphin Vetter, Carlo Vicario, Didier Voulot, Meitian Wang, Thierry Zamofing, Christof Zellweger, Riccardo Zennaro, Elke Zimoch, Rafael Abela, Luc Patthey, Hans-Heinrich Braun, “ SwissFEL: The Swiss X-Ray Free Electron Laser”, Appli. Sci. 2017 7(7). http://www.mdpi.com/2076-3417/7/7/720 24. R. Ganter (Ed.), H. H. Braun, T. Garvey, F. Löhl, B. Patterson, M. Pedrozzi, L. Patthey, S. Reiche (review board), “Athos Conceptual Design Report”, PSI Report Nr. 17-02, September 2017, ISSN 1019-0643. https://www.psi.ch/swissfel/HomeEN/Athos_CDR_2017_V2.pdf 25. Boris Keil, Raphael Baldinger, Robin Ditter, Waldemar Koprek, Reinhold Kramert, Goran Marinkovic, Markus Roggli, Markus Stadler, Daniel Marco Treyer, “A Generic BPM Electronics Platform For European XFEL, SwissFEL and SLS”, Proceedings of the 1st International Beam Instrumentation Conference (Tsukuba) pp 11 – 15, 2012. http://accelconf.web.cern.ch/AccelConf/IBIC2012/papers/mocb02.pdf 26. Boris Keil, Raphael Baldinger, Robin Ditter, Mathias Gloor, Waldemar Koprek, Fabio Marcellini,Goran Marinkovic, Markus Roggli, Martin Rohrer, Markus Stadler, Daniel Marco Treyer, “Status of the European X-FEL Transverse Intra Bunch Train Feedback System”, First International Beam Instrumentation Conference (Melbourne) pp 492 – 496, 2015. http://accelconf.web.cern.ch/AccelConf/IBIC2015/papers/tupb064.pdf 27. Ciro Calzolaio, Stephane Sanfilippo, Serguei Sidorov, Alexander Anghel, and Andreas Streun, “Design of a Superconducting Longitudinal Gradient Bend Magnet for the SLS Upgrade” I.E..E.E Trans. On Applied Superconductivity, 27 No. 4, 4000305, June 2017. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7755773 28. Alexander Gerbershagen, David Meer, Jacobu s Maarten Schippers, Mike Seidel, “A novel beam optics concept in a particle therapy gantry utilizing the advantages of superconducting magnets”, Zeitschrift für Medizinische Physik, 26, Issue 3, pp 224 – 237, September 2016. https://www.sciencedirect.com/science/article/pii/S0939388916300010 29. K. P. Wootton, R. J. England, S. Tantawi, R.W. Asmann, I. Hartl, F. X. Kartner, W. Kuropka, F. Mayet, A. Ruehl, L. Rivkin, P. Hommelhoff, J. Illmer, A. Li, J. McNeur, A. Mittelbach, N. Schonenberger, A. Tafel, R. Ischebeck, Y.-J. Lee, M. Qi, Purdue, D. S. Black, R. L. Byer, H. Deng, S. Fan, J. S. Harris, T. Hughes, N. Sapra, O. Solgaard, J. Vučković, B. M. Cowan, O. Boine-Frankenheim, T. Egenolf, U. Niedermayer, P. Musumeci, “TOWARDS A FULLY INTEGRATED ACCELERATOR ON A CHIP: DIELECTRIC LASER ACCELERATION (DLA) FROM THE SOURCE TO RELATIVISTIC ELECTRONS”. The 8th International Particle Accelerator Conference (Copenhagen), pp 2520 – 2525 (2017), http://accelconf.web.cern.ch/AccelConf/ipac2017/papers/weyb1.pdf 30. Eugenio Ferrari, Rasmus Ischebeck, Nicole Hiller, Simona Borrelli, Franziska Frei, Cigdem Ozkan- Loch, Micha Dehler, Eduard Prat, Simona Bettoni, Sven Reiche, Albert Romann, Jörg Raabe, Blagoj Sarafinov, Saha Susmita, Marco Calvi, Terence Garvey, Volker Schlott, Vitaliy A. Guzenko, Christian David, Martin Bednarzik, Hans-Heinrich Braun, Leonid Rivkin, “The ACHIP Experimental Chambers at PSI”, Proceedings of the 38th International Free Electron Laser Conference (Santa Fe), 2017. 31. A. Wrulich, Y. Ekinci, S. Lee and L. Rivkin, “A Feasibility Study for COSAMI – A Compact EUV Source for Actinic Mask Inspection”, PSI internal report (unpublished). 32. A. Streun, “COSAMI lattices: ring, booster and transfer line”, PSI internal note, 28th June 2016 (unpublished). 33. R.M. Bergmann, T. Bieri, P. Craievich, Y. Ekinci, T. Garvey, C. Gough, M. Negrazus, L. Rivkin, C. Rosenberg, L. Schulz, T. Schmidt, L. Stingelin, A. Streun. V. Vrankovic, A. Wrulich, A. Zandanello Callegher and R. Zennaro, “A Compact Storage Ring for the Production of EUV Radiation”, Proceedings of the 13th International Topical Meeting on Nuclear Applications of Particle Accelerators (Quebec) 2017. http://accapp17.org/wp-content/uploads/2017/09/AppAcc-paper.pdf 34. M. Aiba, R.M. Bergmann, T. Bieri, P. Craievich, M. Ehrlichman, Y. Ekinci, T. Garvey, C. Gough, P. Lerch, A. Müller, M. Negrazus, L. Rivkin, C. Rosenberg, L. Schulz, T. Schmidt, L. Stingelin, A. Streun. V. Vrankovic, A. Wrulich, A. Zandanello Callegher and R. Zennaro, “COSAMI Feasibility Study” PSI internal report (unpublished), January 2017. 35. M. Aichler, P. Burrows, M. Draper, T. Garvey, P. Lebrun, K. Peach, N. Phinney, H. Schmickler, D. Schulte, N. Toge (Eds.), “A Multi-TeV Linear Collider based on CLIC Technology”, CERN-2012-007. http://project-clic-cdr.web.cern.ch/project-clic-cdr/CDR_Volume1.pdf 36. G. De Michele, “Wakefield Simulations and Measurements for the CLIC RF Accelerating Structure”, Ecole Polytechnique Federale de Lausanne Doctoral Thesis No. 6185, 2014. https://infoscience.epfl.ch/record/198491 37. Hao Zha, Andrea Latina, Alexej Grudiev, Giovanni De Michele, Anastasiya Solodko, Walter Wuensch, Daniel Schulte, Erik Adli, Nate Lipkowitz, and Gerald S. Yocky, “Beam based measurements of long range transverse wakefields in the Compact Linear Collider main-linac accelerating structure”, Phys. Rev. Accel. Beams, 19 011001, (2016). https://journals.aps.org/prab/pdf/10.1103/PhysRevAccelBeams.19.011001 38. R. Zennaro, H. Blumer, M. Bopp, T. Garvey, L. Rivkin, N. Catalan-Lasheras, A.Grudiev, G. McMonagle, A. Solodko, I. Syratchev, R. Wagner, B. Wooley, W. Wuensch, T.G. Lucas, M. Volpi, T. Argyropoulos and D. Esperante-Pereira, “High Power test of a Prototype X-Band Accelerating Structure for CLIC”, Proceedings of the 8th International Particle Accelerator Conference (Copenhagen), pp 4318-4320, 2017. http://accelconf.web.cern.ch/AccelConf/ipac2017/papers/thpik097.pdf 39. M. Gaspar, “Study and Development of Solid State High Power RF Sources for Synchrotron Light Sources”, Ph.D. Thesis of the University of Coimbra, September 2014. 40. M. Gaspar and T. Garvey, “A Compact 500 MHz 65 kW Solid-State Power Amplifier for Accelerator Amplifications”, I.E.E.E. Transactions on Nuclear Science, 63, No. 2, pp 699 – 706, April 2016. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7321071

Acknowledgements

I should like to thank the Bureau of the Accelerator Division of the Societé Française de Physique for their kind invitation to make this presentation and for their hospitality during the Roscoff workshop. It should be clear that the work described here represents the efforts of a large number of individuals from the Division of Large Research Infrastructures, the SwissFEL Project group and other PSI Divisions. My thanks go to my PSI colleagues who provided “slides” for the oral presentation. All photos are credited to PSI except those shown in figure 14 (D. Nölle, DESY) and figures 18 and 19 (CERN).

We acknowledge the following financial support for the R&D activities described above: the solid-state amplifier (CTI grant no. 13192.1 PFFLM-IW), the COSAMI design study (CTI grant No. 19193.1 PFNM-NM), the X-band structure development (SNF grants 200021-126838, 206620_135012 and 20FL20_147463), the ACHIP studies (Moore Foundation Grant).