After scattering from the surface region of a sample, ions in a horizontal scattering plane are collected into a toroidal electrostatic analyzer (TEA), shown above. Opposite polarity voltages applied to the deflecting plates steer ions (within energy and angular windows) up to the position sensitive detector (PSD). The angular range collected from the scattering plane is 22°. The energy range is given as 0.01949×E_pass. E_pass is the center energy passing through the TEA, given in keV by V/0.06, where ±V are the voltages (in kV) applied to the plates of the TEA. For a 100 keV beam, the energy window is then about 2 keV, and the applied voltages to the TEA are ±6 kV. The distance between the sample and entrance slit, the distance between the exit slit and PSD, and the radius of curvature and distance to center axis of the toroidal plates are optimized to provide two key focussing characteristics of the TEA [1]. Ions leaving the surface at the same point and same energy, but at slightly different angles in the vertical direction will be focussed to the same point on the PSD. The height of the entrance slit controls the accepted angular range, and the height of the beam spot is made small so to give good energy resolution. Ions leaving the surface with the same energy and parallel directions, but at different points in the horizontal direction, will be focussed to the same point on the PSD. This allows the beam spot to be widened, in turn giving a good beam current to the target without a corresponding loss of angular resolution.

To analyze the energy and angle of the ions passing through the TEA, the PSD utilizes a pair of micro-channel plates in a chevron mounting, followed by a multi-anode charge-dividing collector. The channel plates operate in the pulse-saturated mode, thus enabling single event counting. Particles impinging on the first micro-channel plate create secondary electrons. The number of these electrons is amplified to approximately 10⁶ per incident ion by the two micro-channel plates, giving a charge cloud at the exit side of about 1 pC. This charge cloud impinges on the multi-anode collector shown schematically below.

Two linear arrays of interweaving triangular electrodes, each set coupled with resistors and capacitors, lead to four electrodes at the corners. The angle and energy of incident ions are computed from the relative amounts of charge collected at each corner by charge sensitive amplifiers. Since the charge cloud overlaps several triangles, the angular resolution is actually much better than that given by the width of one triangle. The total system angular resolution is given as 0.1° for ~100 keV protons.

The total system instrumental energy resolution for ~100 keV protons is given as about 110 eV. DE/E = 0.0011 is essentially a constant function of E. This resolution can be found by examining the signal from an element with submonolayer coverage on a very flat surface. For example, Au deposition onto the clean (7´7)-Si(111) surface, followed by a short anneal to 700 °C [2,3], provides a very flat, single atomic layer of Au. An example Au surface peak signal is shown below.

The total system energy resolution is found from the full width at half maximum (FWHM) of this signal, ~190 eV. This width is given by a combination of instrumental effects, and intrinsic effects of the proton-Au interaction. The low energy tail of this distribution is not due to subsurface Au, as can be proven by examining the angular yields. It results from two independent effects. The energy loss from scattering in a single layer is small, so the statistics of the energy losses give the asymmetric Vavilov distribution. A more significant effect in Au, however, is that several large (~100 eV) energy losses are possible due to excitation of inner-shell electrons, leading to the tail at lower energy. This intrinsic ion-atom interaction effect has been included in the measurement of the total system energy resolution. Other intrinsic effects include low energy electron losses of protons travelling in and out of the Au layer, and Doppler (vibrational) broadening. Factors influencing the instrumental energy resolution are fluctuations in the beam energy, beam spot and entrance slit heights, and energy spreading due to the PSD. Removing the intrinsic energy broadening effects of the proton-Au interaction gives the system instrumental energy resolution of ~110 eV. This was determined from simulations of ion trajectories in the TEA (given factors such as beam and entrance slit heights) and measured quantities such as the accelerator energy stability. A discussion of the relative sizes of the various contributions to the total system energy resolution is given in appendix E of Metal and Alloy Surface Structure Studies Using MEIS.