CASSIOPEE

Two undulators are necessary to cover the large photon energy range available at Cassiopée. They were designed and characterized by the SOLEIL Insertion device Group (Antoine Daël, Oleg Chubar, Fabrice Marteau, Chamseddine Benabderrahmane, Marie-Emmanuelle Couprie and Fabien Briquez).

For low photon energies, Cassiopée uses a HU256 electromagnetic undulator, with a 256 mm period, which was designed at SOLEIL and built at the Budker Institute of Nuclear Physics in Novosibirsk (Russia). This undulator provides linearly (vertical or horizontal) or circularly (left or right) polarized photon in the 8 eV-155 eV range (see Figure 1).

Figure 1: Photo of the low photon energy HU256 undulator.

For the high photon energies, Cassiopée uses a HU60 permanent magnet Apple II type undulator, with a 60 mm period, which was designed and built at SOLEIL. This undulator provides linearly (vertical or horizontal) or circularly (left or right) polarized photon in the 100 eV-1500 eV range (see Figure 2).

Figure 2: Photo of the high photon energy HU60 undulator.

The first optical elements play a key role in the beamline's stability and performance. Cassiopée entrance optics is made of three silicon mirrors (M1a is a plane mirror and M1b and M1c are two spherical mirrors) attached to the same rotating table whose rotation axis is vertical, in the plane of M1a surface. Ideally, the surfaces of the three mirrors should be parallel (see Figure 3).

These three mirrors (and especially M1a) have to absorb the power radiated by the ring and the insertion devices, without suffering any deformation. In the worst condition, this absorbed power can be as high as 250W. Therefore, these three mirrors are cooled down at a temperature of 100 K, where the silicon dilatation coefficient is zero, to avoid any thermal bump at the impact point of the photon beam. This is done by a 9 bars liquid nitrogen closed loop circulation.

The entrance optics has two working positions:

• A low energy position, in the 8 eV-155 eV photon energy range, when HU256 is used. In this case, the rotating table is set for an incident angle of 5.02° on M1a. The photon beam is then reflected towards M1b, whose radius is calculated to focus the HU256 source point onto the monochromator’s gratings.
• A high energy position, in the 100 eV-1500 eV photon energy range, when HU60 is used. The rotating table is then set for an incident angle of 2.44° on M1a. The photon beam is then reflected towards M1c, whose radius is calculated to focus the HU60 source point onto the monochromator’s gratings.

When going from one position to the other, an in-vacuum motor can also be used to rotate only M1a, in order to adjust the parallelism between M1a and the other concerned mirror. These two reflections shifted the useful VUV radiation axis from 62 mm in the horizontal plane. Hence, a tungsten rod can be placed along the undulators axis to block the gamma rays emitted by the ring (which go through the silicon mirror).

Figure 3: Principle of the Cassiopée entrance optics, with the low- and high-energy working positions. The entrance optics was built by the Jobin-Yvon company (Longjumeau, France).

Monochromator Geometry

The design of the CASSIOPEE plane grating monochromator was done by the SOLEIL Optics Group. The starting point is the well-known SX700 monochromator, but several modifications have been made.

Figure 1: Picture showing the inside of the CASSIOPEE monochromator, with the plane mirror cage above the four available grating cages. The monochromator was built by the BesTec company (Berlin, Germany)

The working principle of the monochromator is shown in Figure 2. One can choose the incident angle α of the white beam on the plane grating by rotating the grating around an horizontal axis. The β angle can be chosen by rotating the plane mirror, above the grating, around a horizontal axis located below the grating plane, to ensure a constant height of the outgoing beam. For a given couple (α, β), the selected wavelength λ is given by the gratings law:

cos α - cos β = N.k.λ

where k is the diffraction order (-1 for CASSIOPEE) and N is the number of lines/mm on the grating. The CASSIOPEE monochromator contains three different gratings with 400 l/mm, 800 l/mm and 1600 l/mm.

The CASSIOPEE monochromator is designed to work in the “modified Petersen mode”, which means that α and β are always chosen to fulfil the following equation:

sin β ⁄ sin α = constant ≈ 0.2

to ensure an optimal resolution on the whole photon energy range.

Figure 2: Working principle of the monochromator

Monochromator Gratings

The gratings of the CASSIOPEE beamline focus the photon beam by using varied line spacing (VLS). The groove depth also varies along the direction perpendicular to the beam. Hence, by choosing the position of the grating with respect to the beam, one can choose the groove depth seen by the beam to enhance the grating reflectivity at each photon energy. They were made by the Jobin-Yvon company (Lonjumeau, France).

Figure 3: Schematic view of the variable groove depth grating.

The three Cassiopée experiments are:

• Spin-resolved Photoemission experiment
• High Resolution Angle-resolved Photoemission experiment
• Molecular Beam Epitaxy chamber

These chambers are connected to a Crossway chamber, and samples can be transferred in UHV from one chamber to the other as shown on Figure 1.

Figure 1: General view of the Cassiopée experiments, with the Spin-Resolved Photoemission experiment on top, the MBE chamber (bottom right) and the High Resolution Angle-resolved Photoemission experiment (bottom left). These three chambers are connected to the Crossway chamber, in the middle.

The samples are mounted on Omicron-type plates. See this section Users information for practical details.

Detectors

The Spin-Resolved Photoemission experiment is equipped with a Scienta SES2002 electron analyzer, which has been modified by the Scienta company to incorporate a Mott detector for spin analysis (see Figure 2). Part of the electron beam is sent towards the usual Scienta 2D detector, while the rest goes towards the Mott detector through transfer optics. The analyzer is mounted with horizontal slits. Angular resolution is then achieved in the horizontal direction. To accept the Mott detector, the 2D detector has been reduced to 1”, thus reducing the angular acceptance of the analyzer to ±8°.

Figure 2: Schematic view of the Scienta SES 2002 and its transfer optics to the Mott detector.

Inside the Mott detector, the electron beam is scattered by a gold target, and the scattering asymmetry is measured parallel and perpendicular to the sample surface, thus measuring the magnetization components in these two directions. The Mott detector was built in the University of Edinburgh, by Prof. Murray Campbell (see Figure 3).

Figure 3: The Mott detector without one of the four channelplates.

Sample Environment

The Spin-Resolved Photoemission experiment is equipped with a 4-axis manipulator, with a motorized Z translation, manual X and Y translations, and manual θ rotation (see Figure 4). The sample can be cooled down to 40 K with a liquid He cryostat.

Figure 4: View of the Spin-Resolved endstation manipulator with the sample holder and the Scienta SES2002 electron analyzer.

A new in-situ magnetization system is underway.

Beam Size

The beam is sent to the Spin-Resolved Photoemission experiment by a toroidal mirror which also focuses the beam. The beam size is quite large, not to damage the sample during long acquisitions.

Figure 5: Theoretical beam sizes for two different photon energy.

Detectors

The Cassiopée High Resolution Angle Resolved Photoemission experiment is equipped with a Scienta R4000 electron analyzer, whose angular acceptance is ±15° (Scienta Wide Angle Lens). This analyzer can be mounted with its slits either vertical or horizontal.

Figure 6: View of the ARPES endstation manipulator with the sample holder and the Scienta R4000 electron analyzer.

Sample Environment

The High Resolution Angle Resolved Photoemission experiment is equipped with a fully-motorized 4 axis sample holder with X, Y and Z translation and a θ rotation. The sample can be cooled down to 5 K with liquid He cryostat. The Scienta analyzer is mounted with vertical slits. The 30° vertical angular acceptance and the θ rotation of the sample allow to scan a large part of the reciprocal space (see Figure 6).

Beam Size

The beam is focussed by a pseudo-Wolter optics located just before the experimental chamber. It consists in a spherical mirror followed by a toroidal mirror (see Figure 7).

Figure 7: Schematic view of the pseudo-Wolter, focussing the beam in the CASSIOPEE High Resolution Angle Resolved Photoemission experiment.

The size of the beam depends on the photon energy, but is always of tens of microns. Ray tracing calculations are show in Figure 8.

Figure 8: Theoretical beam sizes as a function of photon energy.

The Cassiopée MBE chamber is equipped for surface preparation and characterization, as well as for thin film deposition (see Figure 9):

• 3-keV ion gun (OCI IPS3-D)
• Cylindrical mirror analyzer for Auger spectroscopy (Staib ESA100)
• LEED (Omicron SpectaLEED)
• RHEED (SPECS RHD30)
• 10 crucibles 3kW e-gun evaporator (Thermionics)
• Transferable electron bombardment evaporator (Omicron EFM3)
• Quartz microbalances (Inficon IC/5 controller)

Sample Holder

The MBE sample holder has three positions for the sample (see Figure 10):

• Position 1: Electron bombardment heating (up to 1000°C) and azimuthal rotation for RHEED experiment.
• Position 2: the sample can be cooled down to 80K by a liquid nitrogen circulation.
• Position 3: Direct current heating (for semi-conductors).

Figure 10: Inside view of the MBE chamber showing the sample holder and a quartz microbalance on the right.