Artemis offers the following end stations for experiments in condensed matter:
• Time-resolved angle-resolved photoemission spectroscopy (TR-ARPES)
• Time-resolved spin-resolved time-of-flight (TR-spin-ToF)
The analysis chamber—which can dock to either the 1-kHz or the 100-kHz beamline—is built around a SPECS Phoibos-100 hemispherical analyzer and a FeSuMa analyzer by Fermiologics. The chamber offers the capabilities for pump-probe ARPES, static ARPES with a He lamp, static core-level spectroscopy with non-monochromatized lab source, and low-energy electron diffraction (LEED). The manipulator can be cooled or heated (see specifications below)..
There is also a preparation chamber, which is connected to the analysis chamber by a gate valve. Evaporation is possible in the preparation chamber (DN40 and DN63 flanges are available for mounting evaporators), as is sample exposure to non-toxic gases via a leak valve and sample cleaving. The sample garage can hold up to five samples at a time, and the load lock can hold one sample at a time.
A UHV “suitcase" is available, if you wish to prepare samples in your home lab and ship them in UHV to your beamtime.
The system uses flag-style sample holders. If you would like, we can send you sample plates in advance of your beamtime for mounting samples.
The pressure in the preparation chamber is in the low-10-10 mbar range. The pressure in the analysis chamber is in the low-10-10 - high 10-11 mbar range depending on the manipulator temperature.
Please discuss your experimental requirements with the beamline scientists.
Please note that the end station described here will be soon be upgraded as part of the “HiLUX” upgrade project. When the new end station is available for user access, this section will be updated accordingly. You can read more about the upgrade project at
https://www.clf.stfc.ac.uk/Pages/HiLUX.aspx.
Table 1: ARPES end-station parameters
Parameters
| ARPES end-station
|
Angle of incidence of beam on sample at normal emission
| 45 deg
|
Hemispherical analyzer slit direction
| Horizontal—radial from Γ
|
Sample cooling
| ~20K by open-cycle liquid-He cooling
|
Sample heating
| Analysis chamber: radiative heating; or, up to 1270 K by electron bombardment of the back of the sample. Preparation chamber: radiative heating; or, up to 1270 K by electron bombardment of the back of the sample.
|
Manipulator degrees of freedom
| 4 motorized (x, y, z, polar angle); 1 manual (azimuthal)
|
The goal of this development project is to offer pump-probe measurements of k vs E dependency with spin resolution, for the study of material systems with spin-polarized band structures. The system consists of a ToF analyzer (25-cm-long drift tube) with a mini-Mott polarimeter for spin-resolved measurements. A microchannel plate can be inserted in front of the polarimeter, by means of a linear translator, when spin-integrated measurements are needed. The sample manipulator allows measurement temperatures down to 20 K, and up to approximately 1000 K. There is a preparation chamber separated by a gate valve from the analysis chamber, for simple sample preparation such as sputtering and annealing.
The system has so far been used in successful experiments with a 250-kHz-repetition-rate laser source [1,2], and we are now looking to commission the system on the new 100-kHz beamline. If you have are interested in taking part in this effort as a commissioner, please contact
yu.zhang@stfc.ac.uk.
Photon parameters for the XUV beamlines The parameters listed below are intended to give a general idea of the capabilities of the XUV beamlines, when used for photoemission with condensed matter samples. There is some flexibility in these, but some are coupled and cannot be independently controlled. (For example, time resolution sets limits on energy resolution, and vice versa.) Therefore, please take the numbers below as a starting point for discussion with the beamline scientists, who will try to help you find the best arrangements for your particular experiment.
Table 2: Photon parameters for the 1 kHz and new 100 kHz beamlines
Photon parameters
|
1 kHz beamline
|
100 kHz beamline
|
Probe energy range
| 15 ~ 45 eV
| ~17 - 41 eV
Higher photon energies will be available in future calls.
|
Pump energy range
| λ = 235 nm – 16 μm, tunable
| The signal beam and its harmonics: λ = 1750 nm, 875 nm, or 438 nm (i.e., 0.71, 1.42 and 2.84 eV).
The idler beam and its harmonics: λ = 2500 nm, 1250 nm, or 625 nm (i.e., 0.50, 0.99 and 1.98 eV).
Additional options will be available in future calls.
|
Energy resolution
(limited by short-pulse characteristics and by space charge)
| 250 ~ 400 meV
(Photon-energy and pulse-length dependent; discuss with beamline staff what resolution is available for your experimental parameters.)
| ~ 100 - 150 meV (Photon-energy and pulse-length dependent; discuss with beamline staff what resolution is available for your experimental parameters.)
|
| Time resolution
(limited by pulse lengths of pump and probe)
| ~ Few tens of fs
| ~ 50 fs
|
| Beam spotsize
at normal emission
| ~ Few hundred μm diameter
| Typically ~ 70 μm ; ~ 20 μm available with lower flux
|
Probe flux
| ~107 photons/pulse (~1010 photons/second) at about 25 eV
| ~105 photons/pulse (~1010 photons/second) at about 32 eV
|
Pump fluence
| Adjustable. Max fluence typically several mJ/cm2/pulse (depending on pump wavelength).
| Adjustable. Max fluence typically several mJ/cm2/pulse (depending on pump wavelength).
|
Probe polarization
| Linear polarization,
s or
p
| Linear polarization,
s or
p
|
Further technical notes for preparing a user proposal
Requirements for samples (crystallinity, domain size, etc.)
Sample orientation, crystallinity, and domain size
If
k-vs.-E dispersions are to be measured, the sample needs to be single-crystalline within the region of analysis (i.e., within the size of the beamspot on the sample). If more than one crystalline domain is present inside the analysis region, the measured dispersion will be the sum of the dispersions of all the domains within the region.
Cleanliness
Within the photon energy range available at Artemis, the universal mean free path is very short [3]—meaning that measurements are extremely surface-sensitive. It is therefore important that the sample surface be extremely clean. Typically, samples are prepared in vacuum by cleaving, annealing, decapping, or sputtering/annealing.
Sample quality
Photoemission linewidth becomes broad when defect density becomes high. Good sample quality is key to acquiring data of sufficient quality for successful analysis.
Sample electronic properties
Insulating samples are, typically, inappropriate for photoemission spectroscopy, since the photoemission process then leads to sample charging. Some exceptions exist—e.g., atomically thin insulating layers on conducting substrates are sometimes feasible. E-mail the beamline scientists to discuss this issue.
Probe photon energy
The photoemitting (probe) laser beam is generated by high-harmonic generation in a gas jet. You must specify your required probe photon energy in your proposal (see the beamline capabilities above) so that the probe beam can be set up and optimised before you arrive for your beamtime. Please communicate with the beamline staff before you submit your proposal, to confirm that the probe photon energy and flux you need is available and that your requested energy tolerance is feasible. The beam photon energy can be scanned during your beamtime (in steps of 2ω).
Pump-probe issues
Pump photon energy and polarization
Available photon energies are shown above. Any polarization is possible, and polarization-dependent measurements are also possible. Both the pump photon energy and the polarization need to be specified in the beamtime application, so that the beamline can be set up in advance of the experiment. For polarization-dependent measurements, a motorized polarizer will be inserted into the beamline and incorporated into the experimental acquisition and control software.
Pump photon energy cannot be changed on-the-fly. If you will require more than one pump photon energy for your experiment, contact beamline staff before submitting your proposal, so we can discuss possible options.
Acquisition time
Pump-probe techniques are intrinsically time consuming, since each delay time will constitute one full spectral acquisition—potentially on the time scale of up to a few hours, depending on laser repetition rate, on fluence limitations (especially to prevent space charge [4, 5]), and on statistics. It is typically feasible to probe, e.g., 10 - 20 delay points selected to span the expected time scales of phenomena of interest. If polarization dependence is to be studied, then the number of measurement points will be multiplied accordingly. In building a feasible proposal, you should take acquisition time into account.
References
[1] Cacho, et al., “Absolute spin calibration of an electron spin polarimeter by spin-resolved photoemission from the Au(111) surface states,” Rev. Sci. Instr. 80 (2009) 043904.
https://doi.org/10.1063/1.3115213[3] Seah and Dench, “Quantitative Electron Spectroscopy of Surfaces: A Standard Data Base for Electron Inelastic Mean Free Paths in Solids,” Surf. Interface Analysis 1 (1979) 2.
https://doi.org/10.1002/sia.740010103
