|
Description
The analyser has a 5-element entrance lens and is mounted either on a single-axis or on a two axis goniometer
as a bolt-on instrument with mounting flange NW 200 CF (10'' OD). On request the single axis goniometer can be mounted on a NW 150 CF (8'' OD) flange. The control units Analyser Control and Lens Control cover the kinetic energy range from 0 to 1680 eV. The SHA 50 is a
simulated hemispherical analyser
with Jost-type fringing-field correction (K. Jost: J. Phys. E Vol.12 (1979) , p. 1001- 1011). An electrostatic entrance lens (5 elements) with three selectable apertures for the angular acceptance (±1°, ±3°, or ±5°) is adapted to the SHA. Although its mean-path radius is 50mm, the SHA 50 needs a vacuum chamber of 300 mm I.D. only and has a width of only 56mm. The entrance lens is designed to match a sample to lens distance of 25±1 mm.
The SHA 50 is designed for highest transmission at selectable resolution. Easily exchangable entrance apertures
allow to define the accepted angular range of ±1°, ±3°, or ±5°. For improved signal-to-noise ratio the SHA 50 can be upgraded to a double-pass analyser by fitting a subsequent 90°-deflector. The rotational axis of the SHA is accessible through a coaxial NW 100 CF (6'' OD) port offering a wide range of unique options. A sample manipulator can be built into the port. This compact arrangement facilitates a
precise ex-situ alignment of sample and analyser axis. As a special option (or as a retrofit) the SHA 50 can be equipped with the electron spin-polarization detector Focus SPLEED
for spin-resolved UPS or ESCA, e.g. photo- or Auger-emission from ferromagnetic systems. For this option the 90°-deflector is required.
The control units Analyser Control and Lens Control cover the energy range from 0 to 1680eV. Analyser Control supplies all
voltages for the analyser and the optional 90° deflector; Lens Control supplies the voltages for the three variable lens elements and an optional sample bias. High stability voltage supplies
, in combination with a mu-metal shielded analyser chamber (optional a mu-metal adaptation of the analyser base flange is available) facilitates
high resolution applications below 10 meV FWHM. The Analyser Control Unit can be operated either under front-panel control or under full computer control. The Lens Control Unit is operated either
manually with three reference voltages of ±5V or under full computer control to employ user selectable nonlinear lens curves of e.g. constant magnification or maximum intensity. Dedicated interface
boards and software for AT-compatible are available.
The figures represent drawings of the SHA 50 mounted on a double-axis goniometer on flange NW 200 CF (8'') at a working distance
of 305 mm (12''). The single axis goniometer can optionally mounted on a NW 150 CF flange. Other working distances (e.g. 8'' or 10'') can be delivered on request.
 
Analyser
The SHA50 is an electrostatic Simulated Hemispherical sector Analyser with 50 mm mean path radius.
This type of energy analyser for charged particles has first been described by Jost, who also performed a
detailed optimization of the fringing-field correction elements. It is evident that the size of that novel analyser
is substantially reduced. Since the electrode shape and the correction electrodes exactly produce a spherical
field in the region of the beam trajectories, the size reduction causes no loss in performance.
The compact design of the SHA 50 allows for small angles between the exciting beam and the observation
direction of down to ±15° in the plane of the main rotation or down to 15° in the perpendiculare plane. As
an option, an aperture stack for 0° excitation (i.e. the exciting beam passes from the back of the analyser coaxially through the entrance lens) can be mounted.
The SHA features an optional 90° deflector, which improves the signal-to-noise ratio beyond the usual
level and is of special interest when employing an electron spin analyser, such as the Focus SPLEED
detector. The 90°-geometry allows easy access to the spin analyser and, more important, the 90° deflection
converts a longitudinal spinpolarisation component into a transversal one which can be detected by the spin
analyser. This facilitates an analysis of all three electron spin polarisation components by a simple sample
rotation. Hence this configuration allows, e.g. the simultaneous measurement of both the in-plane and the out-of-plane (perpendicular) components of the surface magnetisation in a magnetic material.
Entrance Lens
The SHA 50 is equipped with an electrostatic 5 element entrance lens. It adapts the kinetic energy of the
analysed electrons to the pass energy of the analyser. The first lens element is at ground potential and the
fifth lens element is kept at entrance slit potential. The second to fourth lens elements (called L1,L2 and L3) are excited by voltages generated by the power supply Lens Control.
Owing to its zoom capability the lens can be operated at constant magnification in a wide range of
energies. Typical magnifications from the sample to the entrance aperture are 0.6 or 1. The lens voltages
follow general nonlinear characteristics which can be software selected and altered by the user.
An energy scan over the range from Ekin = 0 to 160 eV at a pass energy of the analyser of e.g. 4 eV works
as follows: Between 4eV and 160eV the electrons are decelerated to 4eV via the appropriate action of the
entrance lens. The final retardation factor is R = 40 and stays in the allowed range of most lens curves.
Independent of this retardation, the lens always produces an image of the source spot on the sample at the
entrance aperture of the analyser with the choosen magnification. The entrance aperture w1 of the analyser thus defines the size of the accepted source region on the sample.
In XPS with standard laboratory sources (e.g. twin anode X-Ray tubes) the excitation area on the sample is
rather large. Trajectory calculations have shown that in this case two points are important:
- The high retard ratios Ekin / Epass for XPS lines at high energies principally reduce the intensities in
XPS (and AES) spectroscopy (reduction of the effective angular acceptance). For XPS the pass energy should therefore be as large as possible, taking into account the needed minimum energy
resolution.
- The angular deviations caused by the finite source size will generally reduce the energy resolution of the analyser.
In the case of low lying lines (shallow core levels) it is recommended to work in the 0 - 163 eV range.
Electron Detection
For both electron and ion detection the SHA employs a Channeltron .
Energetic Considerations
The sample is normally at (or near) earth potential. The kinetic energy of Auger- or
photoelectrons is thus measured with respect to ground. Electrons leaving the sample with a certain kinetic energy Ekin are accelerated or decelerated by a retarding voltage Uret to the
selected pass energy Epass. Then according to the energy diagram:
where W is a constant arising from the work function of the analyser materials and from a possible bias on the sample.
The SHA acts as a narrow pass filter letting through only electrons within a small energy interval of width D E around Epass (see sect. 3.2).
For Auger lines Ekin is independent of the energy of the exciting photons or electrons. The Auger process
gains its energy through an electronic transition of an outer-shell electron to an inner-shell vacancy. This process does not depend on how the inner vacancy has been produced.
In contrast a direct phototransition in XPS or UPS is governed by the energy relation:
where hn is the photon energy, EB is the electron's binding energy referenced to the Fermi level, and f is the
work function of the sample (i.e. the minimum energy required to bring an electron from the Fermi level of the sample to the detector placed at a macroscopic distance away from the sample).
Ultraviolet Photoelectron Spectroscopy (UPS)
For UPS typical photon energies are hn =21.22 eV (for He I), hn = 40.8 eV (for the He II
resonance wavelength) or synchrotron radiation in the vacuum ultraviolet (VUV) region. The SHA 50 is operated in the low-energy mode:
Ekin = 0 to 163.8 eV (step resolution 2.5 meV)
(or 0 to 65.5 eV with step resolution of 1 meV) Epass = 0.5 eV, 1 eV, 2 eV or 4 eV
In order to directly observe work-function changes, e.g. due to absorption or contamination processes, it is desirable to scan the spectrum down to the low-energy cut off Ekin = 0. To obtain this cut-off correctly, it is
advantageous to apply a small negativ bias voltage (typically a few V) to the sample which facilitates a better
handling of the low-energy electrons. Note, however, that thia bias voltage shifts the kinetic energy scale
accordingly and a ripple on the bias voltage (as well as a ripple on the ground potential) fully shows up in an increased peak width, i.e. a reduced energy resolution.
Note that a bias voltage reduces the angular resolution. The electrons are accelerated in the (normally field
free) region between the sample and the lens front end. This distorts the initial starting angle of the electrons.
X-Ray Photoelectron Spectroscopy (XPS)
For XPS typical photon energies are hn =1486.6 eV (for Al Ka ) or hn =1253.6 eV (for Mg Ka ) or synchrotron radiation with energies beyond the UPS range.
The SHA 50 is operated in the high-energy mode:
Ekin = 0 to 1638 eV (step resolution 25 meV)
(or 0 to 655 eV with step resolution of 10 meV,
selectable by jumper setting) Epass = 5 eV, 10 eV, 20 eV or 50 eV
For XPS a sample bias is usually not required.
To take spectra in the XPS mode, proceed as described in sec. 4.2, steps 1. to 9.. Note that the standard Al/Mg Ka source should be as close to the sample as possible.
In XPS spectra the kinetic energy Ekin scale is conventionally often replaced by the binding energy EB
referred to the Fermi energy. The advantage of referring energies to EF is that this is a fixed energy level, whereas the vacuum level differs from EF by the workfunction, which is influenced, e.g., by surface
contaminations and residual gas adsorptions on the sample surface.
The latter effect shows up in photoemmission spectra:
The Fermi edge stays fixed on the energy scale (constant sample bias provided), whereas the position of the Ekin = 0 low-energy cut-off of the spectra depends on the condition of the sample surface. Since a shift of
this cut-off directly reflects a corresponding work-function change, its measurement provides a powerful method to measure work function changes Df , e.g. induced by adsorption or epitaxy. This is mostly done
in the UPS mode.
Auger Electron Spectroscopy (AES)
The energy of an Auger line is independent of the initial excitation process (photon- or
electron-induced) and can be found in Auger-tables. For photon-induced AES the operation of the SHA is identical to XPS, see preceding section. In case a tuneable photon source (e.g.
Synchrotron radiation) or a twin-anode (Al/Mg Ka ) X-ray source is used, an easy distinction
between Auger and XPS lines in an electron spectrum is possible by changing the excitation energy (e.g. switching from Al to Mg). All XPS photoemission lines will shift in Ekin accordingly
by the photon-energy difference, whereas the Auger line positions will remain fixed.
In electron-induced AES an electron beam of typically 3 to 10 keV energy is used to create
the primary core hole. The deexcitation via Auger-electron emission then happens as in the first case. An essential difference is, however, that here the Auger lines are riding on a
substantially higher background of secondary electrons, released by the primary beam. The sensitivity of Auger-line detection can be strongly improved by a differentiating techique. The
optimisation of the position and size of the primary beam focus on the sample usually requires a few cycles of steps 6 to 9 thereby gradually improving the settings of the x/y-deflection of the
electron beam and the focus of the electron gun together with optimisation of the sample position and orientation. After accumulation of the spectrum, the differentiation is done
numerically. Alternatively, the Auger lines can be displayed after background subtraction without differentiation. This way is often preferred for a quantitative analysis of Auger line intensities.
|
