Improved spatial resolution microanalysis in a scanning transmission electron microscope

Abstract

The attachment of energy-dispersive analysers to scanning transmission electron microscopes makes possible accurate high spatial resolution microanalysis. This paper quantifies the X-ray spatial resolution obtainable for good sensitivity microanalysis as a function of microscope lens settings and primary electron beam accelerating voltage in a 200 kV scanning transmission electron microscope. It is shown that at 200 kV a resolution of approximately 500 A is possible with a 1500 A thick nickel-iron-chromium alloy specimen. At 100 kV the resolution drops to nearer 1000 A These figures are supported by Monte Carlo calculations of electron trajectories in thin films. Comparisons of experimentally measured and theoretically predicted X-ray intensities suggest a severe loss of signal intensity within the microscope; this loss becomes greater at higher accelerating voltages. The influence on microanalysis of angle of specimen inclination to beam and to the detector is considered and precautions necessary to limit beam drift during 1000 s analyses are discussed. The application of high sensitivity, high resolution microanalysis methods to a grain boundary segregation effect in an austenitic creep resistant steel is described. It is shown that titanium segregates in a zone 1000-1 500 di thick on either side of the grain boundary to a level of approximately three times the matrix concentration.When considering high energy electron beams (i.e. 100

The measurement and calculation of the X‐ray spatial resolution obtained in the analytical electron microscope

“…There have been many definitions and theoretical calculations of b. Some definitions of b include a diameter which contains only 75-800;, of the emerging electrons (Faulkner et al, 1977;Hutchings et al, 1979;Brown, 1981), while other definitions include 90% (Newbury & Myklebust, 1980;Cliff & Lorimer, 1981;Reed, 1982). However, the most widely used delinition of 6, and the one used here, is that derived from the single-scattering model developed by Reed (1982) from the original paper by Goldstein et al (1977).…”

A consistent definition of probe size and spatial resolution in the analytical electron microscope

“…Many other definitions of b have been proposed using the diameters which contain 75-80% of the emerging electrons (e.g. Faulkner et al, 1977;Hutchings et al, 1979;Brown, 1981). Most of these approaches ignored the role of the electron probe diameter and came out 0 1987 The Royal Microscopical Society with expressions for b alone.…”

The role of specimen and instrumental parameters in STEM‐EDS X‐ray microanalysis of thin foils

“…More recently resolution has been related to the diameter of a cylindrical volume, parallel to the axis of the incident electron probe, within which an arbitrarily fixed proportion, usually 90",, of X-ray generation events occur ( Fig. 1) (Kyser, 1979;Doig et al, 1980Doig et al, , 1981bFaulkner et al, 1977;Hall et al, 1981;Hutchings et al, 1979;Stephenson et a]., 1981).…”

“…Perhaps the most precise descriptions of electron scattering and therefore spatial resolution of microanalysis in thin foils have been achieved using Monte Carlo methods of calculation (Green, 1963;Bishop, 1965;Kyser, 1979;Hutchings et al, 1979;Heinrich et al, 1976). However, there remain some discrepancies in the numerical value obtained between various workers (Faulkner et al, 1977;Stephenson et al, 1981) that originate from differences in the input parameters and calculation method. Such procedures, however, are relatively time consuming in computation and, as such, are not readily available on-line to most operators undertaking microanalysis on a given feature within a foil sample.…”