What if a scanning electron microscope could reveal the same dislocation lines as transmission electron microscopy, while examining steel under conditions that do not require the same thin-sample approach? This paper tests that idea by precisely steering the electron beam.

Electron channeling in scanning electron microscopy is directly related to crystallographic orientation relative to the incident electron beam. Electron channeling SEM images can provide information on the crystallographic properties of a sample.

If electron channeling can observe local crystal orientation changes, changes in backscattered electron intensity should be possible due to lattice defects such as grain boundaries and dislocations in a well-polished bulk sample. Several examples have already been reported for observing lattice defects in bulk samples using electron channeling contrast imaging.

The study controls the incident electron beam direction with high precision using a piezoelectric element as a driving unit. It investigates dislocation contrast in steel materials using SEM and ECCI, and experimentally measures the depth information of that dislocation contrast.

SEM and ECCI were performed from a Kikuchi map by tilting the sample in a scanning electron microscope with a special stage containing three piezoelectric axes, precisely controlling the incident beam direction to the crystal plane. ECC images used a bottom-column backscattered electron detector, a four-millimetre working distance, a thirty-micrometre aperture, and accelerating voltages from fifteen to thirty kilovolts.

A focused-ion-beam thin-film sample was also observed in a transmission electron microscope at three hundred kilovolts, and SEM and ECCI were then applied to the same observed area. Figure one shows electron-channeling contrast images as the sample is tilted relative to the ferrite grain’s (200) crystal plane, alongside the corresponding backscattered-electron intensity profiles at fifteen and thirty kilovolts.

The marked Bragg positions indicate orientations where channeling changes the measured signal, while the image series makes the evolving contrast visible directly. This matters because it demonstrates how beam energy and precise crystal alignment shape the signal used to interpret dislocations in SEM-based electron-channeling contrast imaging.

As the incident primary electron beam direction angle varied concerning the (200) crystal plane, the backscattered-electron intensity changed significantly near the Bragg conditions. At thirty kilovolts, the tilt-angle width between the Bragg positions of (200) and (200) was approximately two point eight degrees, while the width between the channeling positions was approximately three point six degrees.

The channeling condition was the incident condition where the backscattered-electron intensity reached a local minimum, and it was satisfied on the higher-angle side rather than at the Bragg condition. Few studies have examined how accelerating voltage affects dislocation contrast in SEM and ECCI, and whether high accelerating voltages help or hurt remains under discussion.

Although SEM and ECCI observations have used accelerating voltages from ten to thirty kilovolts, the optimal accelerating-voltage conditions remain unclear. Figure five shows the same field of view in electron channeling contrast images acquired at fifteen, twenty, twenty-five, and thirty kilovolts, with channeling conditions maintained for the g equals zero one one excitation.

The dislocation network is visible across the images, and the opened arrows in the thirty-kilovolt panel identify ten dislocations used to evaluate the width, W d c. This comparison matters because changing accelerating voltage also changes the channeling position, so directional control is required when comparing the same area.

Figure six plots dislocation contrast width, W d c, against accelerating voltage from fifteen to thirty kilovolts. The plotted points show measurements at fifteen, twenty, twenty-five, and thirty kilovolts, with the width changing across these settings. This matters because the authors emphasize that channeling position varies with accelerating voltage, so directional control must be repeated when comparing the same field of view; the measured width also relates to how SEM/ECCI depth information is interpreted for dislocation-density quantification.

Bright-field TEM observations were performed on the same ferrite-grain thin film to validate SEM and ECCI as a method for observing dislocation contrast. The bright-field TEM image used g equals one one zero excitation from the front side, while ECC images used the same excitation from the front and back sides of the thin film.

The dislocation contrast between bright-field TEM and SEM and ECCI was obtained at the same diffraction vector in the same field of view, with good correspondence between them. The experiment demonstrated that SEM and ECCI, with the channeling condition strictly satisfied by controlling the incident beam direction, provides contrast due to a single dislocation.

Figure seven directly compares dislocation contrast in bright-field TEM and ECC imaging of the same ferrite thin film, using g equal to one-one-zero. Panel (a) shows the BF-TEM view from the front at three hundred kilovolts, while panels (b) and (c) show ECC views from the front and back at thirty kilovolts; labeled dislocations, including those with endpoints, can be traced across the images.

Panel (d) superimposes the two ECC views, supporting the authors’ validation of SEM/ECCI for observing dislocation contrast. Understanding the depth of the image is necessary to quantify dislocation density by SEM and ECCI. The paper places this depth question in relation to a calculation by Wilkinson and colleagues, who considered the depth at which more than ninety percent of the backscattered-electron intensity was removed.

Accordingly, SEM and ECCI with the channeling condition strictly satisfied by controlling the incident electron beam direction can suggestively obtain dislocation contrast deeper than five extinction distances. Figure eleven shows a stereo pair of ECC images acquired at thirty kilovolts, tilted twelve degrees, with the diffraction axis g equal to zero one one.

The labeled dislocation features appear in both views, while panel c represents their geometry as a screw dislocation line between the sample surface and the depth direction. This arrangement matters because comparing the measured two-dimensional length with the depth provides the basis for interpreting dislocation contrast in three dimensions.

The paper measures dislocation density with the grid-intercept method, using the numbers of intersections, the total lengths of vertical and horizontal grid lines, and the visibility depth of the observation field. The measured and calculated depth values used for the visibility depth were one hundred seventy and eighty nanometres.

The grid consisted of ten vertical and ten horizontal lines, each one micrometre long. Because dislocations with b equals a over two times plus or minus [111] disappeared in the g equals one zero one image, the g equals zero one one image was used to measure dislocation density.

Table two reports how many times dislocations intersect the horizontal and vertical grid lines in Figure twelve. For h one through h ten, the counts are seven, five, four, five, seven, four, five, four, four, and three; the corresponding primed h lines have counts of one, zero, zero, zero, zero, one, one, one, two, and two.

The v and primed v lines are listed separately, providing the intersection counts needed for the authors’ grid-intercept measurement of dislocation density. The study shows that carefully controlled SEM electron channeling contrast can reproduce dislocation positions seen by bright-field TEM, distinguish dislocation characteristics, and provide depth information useful for estimating dislocation density in steel.