The general layout of FERMI is reported below. There are two 1 m-separated undulators (FEL1 and FEL2) delivering the radiation into the photon transport system. The radiation is collected by PADReS (Photon Analysis Delivery and REduction System), and then directed to the selected endstation in the experimental hall. So, any experiment (except MagneDYN) can have access to the full energy spectrum of the two FELs.; MagneDYN, n the contrary, can access only FEL2. It is then necessary to characterize the photon radiation emitted by both the undulator chains. For this reason two identical systems are installed. They consist of a shutter, a beam defining aperture, a beam position monitor, an intensity monitor, a differential pumping system, and a gas adsorption cell. After the gas adsorption cell the system is symmetrically repeated, therefore the light meets a second differential pumping system, a second intensity monitor, a second beam position monitor and, after that, the first mirror. Inside the safety hutch a system of 4 mirrors, two for FEL 1 and two for FEL 2, delivers the radiation to the online photon energy spectrometers that have to analyze, shot by shot, the energy spectrum of the emitted radiation. After that, for the main part of PADReS, an insertable switching mirror gives the possibility of directing the photon beam to the EIS-TIMER endstation. On the central path, on the other way, a beam splitting and delaying section is inserted before the 3-ways switching chamber. Starting from this point 3 branchlines, leading to three endstations (EIS-TIMEX, DiProI and LDM), start and host the refocusing sections. MagneDYN, on the other side, receives the radiation after its energy spectrometer and has only one possible path to the endstation.

Main targets

The main target of PADReS is to analyze the photon beam in terms of absolute intesity, position and spectral distribution of each pulse, working online shot-to-shot. Besides this, there is also the possibility to measure invasively the transversal coherence (on the first part of the EIS-TIMER branchline). Another important section of PADReS is devoted to split the photon beam and to introduce a controllable delay between the two generated semi-beams. Each endstation, at the end, has a dedicated refocusing section made by grazing incidence x-ray mirrors.

PADReS Layout

The PADReS includes the entire photon beam transport and diagnostics system at FERMI, going from the front-end shutters to the valves before the endstations. Part of it is located in the undulator hall (left of the main wall - see figure), part in the so-called safety hutch (between the two walls - see figure), and part in the main experimental hall. The five light-ports provide the photon beam to the respective experiments, and each of them benefit from the analysis and the manipulation performed on the beam itself.


This element defines the acceptance angle of the incoming photon beam. Due to the very high peak power (or fluence) characterizing the FEL photon beam impinging on this element (about 40 mJ/cm2 for FEL1 at 20 nm, and about 400 mJ/cm2 for FEL2 at 5 nm), particular care has been used in designing it. The radiation incidence angle on the BDA is 7.6º. This value is the minimum possible according to mechanical constrains, and it increases the illuminated area by a factor of ~8, decreasing the fluence of the same amount. Mechanically, the BDA is formed by two trunks of pyramid as shown in  the figure below . The central aperture of each trunk is 20x20 mm2, and they are moved one respect to the other in order to select the effective aperture. A complete closure of the total aperture is possible, and the central aperture itself can be moved ±12.5 mm around the ideal axis (equivalent to about ±0.6 mrad).


The BPMs are based on the interception of the tails of the photon beam transversal intensity distribution by four metallic blades collecting a drain current. The distribution has been calculated and simulated by the FERMI@Elettra Machine Physics Group, and it results Gaussian in both transverse directions. Consequently, it is possible to calculate precisely the centroid of the horizontal-vertical transverse intensity distributions. Each blade can travel 30 mm transversally, and a complete closure in both directions is possible. All the four blades are electrically insulated, and made by Copper. By reading simultaneously the four currents it is possible to determine, pulse-by-pulse, the relative displacement of each single pulse with respect to the others, and the initial nominal position. The spatial resolution is expected to be better than 2 µm rms. Moreover, by the concurrent readings of the second BPM positioned about 9 meters after, it is possible to determine the angular movement of the photon beam shot-by-shot with sub-µrad precision.


The working principle of the intensity monitors is the atomic photo-ionization of a rare gas at low particle density in the range of 1011 cm-3 (p ≈ 10-5 mbar). The photon beam traveling through a rare gas-filled chamber generates ions and electrons, which are extracted and collected separately. From the currents generated this way it is possible to derive the absolute number of photons per pulse, shot by shot. The I0Ms have the following characteristics:
  • almost completely transparent
  • wide dynamical range 
  • no saturation effects 
  • independent from the beam position fluctuations
  • usable on the whole wavelength of FERMI
  • usable continuously for on-line shot-to-shot measurements
  • provide the absolute number of photons per single pulse

Gas Absorber

The pulses delivered by the FELs have high peak powers, and some experiments need a way to reduce the FEL intensity over many orders of magnitude (example: for sample alignment); consequently, a gas absorbing (GA) section is installed between the two BPM-I0M sections. This GA represents a reliable way to reduce the intensity without changing the photon beam characteristics. It consists of a windowless gas-filled cell differentially pumped with respect to the other sections. Different gases can be used in order to cover the whole spectral range emitted by the FELs, and different pressures have to be used to guarantee the attenuation factor requested by the users. In particular, by using a maximum gas pressure of about 0.1 mbar on a 6 meters-long cell, it is possible to reduce the overall intensity by 4 orders of magnitude. The calculated transmission for Nitrogen, Xenon, and Krypton is reported in figure (Transmission data calculated using CXRO online toolset).
It is clear that Nitrogen is enough to cover almost the entire range, excluding the low wavelength part between 5 and 20 nm that can be covered by Xenon and Krypton. The GA length (6 m) is 2 times the safety wall thickness. In this way we can insert the gas directly in the center of the pipe, and symmetrically pump it at the two edges, simplifying considerably the mechanics of the systems. Recently, the GA has been endowed by Neon and Helium, besides Nitrogen.

Energy Spectrometers

A single energy spectrometer (PRESTO) is designed to acquire the FEL spectrum for both FEL1 and FEL2 in the wavelength range 100-3 nm. The optical part is made by three identical plane Silicon substrates. All of them have the central part ruled in order to realize a Variable Line Spacing (VLS) diffraction grating. Each grating is designed to deliver and focus a very small part of the incoming radiation (about 1%) onto a YAG crystal, imaged by a CCD detector, while most of the incoming photons (>97%) are reflected to the following beamlines. The grazing angle of incidence is fixed to 2.5°, while the distance from the source slightly depends on the selected wavelength, and is about 45 m. The focus position changes as a function of the photon energy for both angle and distance. The minimum and maximum collectable diffraction angles are limited by the mechanical system, and are 9° and 19°, respectively, while the focal distance ranges from 2500 to 3100 mm.
A similar spectrometer (TARDI) serves only MagneDYN, which receives light only from FEL2. There are two gratings covering the 60-2.1 nm range, working at a grazing incidence angle of 2°.
The gratings parameters are reported in the tables below (D0, D1, D2nad D3 are the groove density variation parameters).



Grating parameters


Parameter G1 G2 G3
Wavelength range m = 1 (nm) 100 - 24.8 27.6 - 6.7 12.7 - 3.1
Energy resolution (meV) 0.2 -2.9 0.3 - 9.5 0.4 - 8.1
D0 (l/mm) 500 1800 3750
D1 (l/mm2) 0.35 1.26 2.68
D2 (l/mm3) 1.75 10-4 6.28 10-4 1.4 10-3
Groove profile Laminar Laminar Laminar
Groove heigth (nm) 12 4 8
Groove ration (w/d) 0.60 0.65 0.65
Coating / Thickness (nm) Graphite / 50 Gold / 50 Nickel / 50

Expected resolutions


Parameter LE HE
Wavelength range m = 1 (nm) 60.5 - 13.5 9.6 - 2.1
D0 (l/mm) 600 3750
D1 (l/mm2) 0.42 2.6
D2 (l/mm3) 2.10 10-4 1.30 10-3
D3 (l/mm4) 9.50 10-11 5.90 10-10
Groove profile Laminar Laminar
Groove heigth (nm) 10 4
Groove ration (w/d) 0.70 0.70
Coating / Thickness (nm) a-C / 30 Gold / 30




In order to clean the photon beam emitted by the source from unwanted radiation such as seed laser, higher harmonics or first stage (for FEL2), a system of filtering stations is installed along PADReS. In particular, each FEL has a dedicated filtering station in the undulator hall, with two discs hosting up to 11 filters each. Moreover, another 2-disc filtering station is installed in the experimental hall serving the following endstations: EIS-TIMEX, DiProI and LDM.
The filters can be mounted and replaced before each experimental run, and several materials and filter-thicknesses are available for users.
For each filtering station it is possible to select up two different filters, one on the "IN" disc and the other on the "OUT" disc.

In the following figure the location of the three filtering stations is reported.

The mechanical layout of a filtering station is reported in the following left figure.
The right figure reports typical examples of FEL radiation filtering for FEL1 and FEL2. In oboth cases only the fundamental FEL radiation is selected, while all the rest (including seed laser, higher harmonics, different stage photons) is cut.
NOTE: for each experiment a proper selection of filter can/should be done. Depending on the wavelength, it could be not possible to find a single filter that cut everything except the wavelength of interest. In that case, multiple filters can be use (affecting also the final intensity).
For some cases also the Gas Absorber, normally used just to reduce the FEL intensity, could have a filtering effect depending on the used gas.


Filters FEL1

In the following the currently-installed filters (divided between IN and OUT discs) and their transmissions are reported for the FEL1 filtering station in Undulator Hall.




Filters FEL2

In the following the currently-installed filters (divided between IN and OUT discs) and their transmissions are reported for the FEL1 filtering station in Undulator Hall.




Filters Experimental Hall

In the following the currently-installed filters (divided between IN and OUT discs) and their transmissions are reported for the FEL1 filtering station in Undulator Hall.




Last Updated on Friday, 15 November 2019 12:07