Scanning Transmission Ion Microscopy and Diffraction Imaging

Abstract

In this paper we investigate the possibility of applying the "ORION" (ZeissSMT : Peabody, MA) Helium ion scanning microscope (HIM) to imaging in the scanning transmission mode of operation. In particular because the interaction of He+ ions with solids differs in many ways from that for electrons it is necessary to determine the changes in operating conditions, and in image interpretation, that may be required.An important first question is what constitutes a 'thin' sample for a beam of 30-50keV He+ ions? Monte Carlo simulations of He+ ion transport through thin foils show that for a given energy the signal falling on to an axial bright field (BF) detector remains close to 100% transmission for a range of thickness but then rapidly falls as plural scattering begins to dominate. Choosing that critical thickness as an estimate of the maximum usable range for transmission imaging, then (figure 1) for 40keV operation useful foils would be 250nm thick for carbon, 150nm for silicon, 50nm for silver and 25nm for gold. At 100keV these value are about 2 to 3 times higher. These calculations do not take into account the enhanced transmission (channeling) of ions in crystals and so are likely an underestimate of the performance. Monte Carlo simulations to determine the broadening of the ion beam at the exit surface of the specimen were also made (figure 2). For silicon at 40keV the Full Width at Half Maximum height (FWHM) of the beam, assuming an incident probe of zero diameter, is less than 3nm for all thicknesses up to 150nm, although a skirt of scattered ions that gradually increase in both intensity and width is visible but remains relatively insignificant. Similar results are predicted for other elements confirming that, within the constraints imposed on the usable sample thickness, scanning transmission ion microscopy will be capable of offering high resolution imaging. In a STEM system detectors are typically provided for on axis Bright Field (BF) and the off axis High Angle Annular Dark Field (HAADF) observations. As predicted by Cowley's principle of reciprocity the acceptance angle of the BF detector has a significant effect on the appearance of the image. If the BF detector acceptance is confined to a small fraction of a Bragg angle then the BF STEM image will appear similar to a conventional TEM bright field image of the same material, but at the expense of poor signal to noise ratio. A larger BF detector will collect more signal, but the form of the image will increasingly deviate from the conventional TEM version. The annular HAADF detector has an inner radius chosen to be large enough (50 millirads or more) to exclude any Bragg scattered electrons and the collected signal is then solely a function of the specimen thickness and atomic number. The same considerations apply to the detectors for the He+ STIM, but practical considerations intervene. For a 40keV He+ beam a typical Bragg 604