Exploring Human Tooth Enamel with Electron Counting 4D STEM

The K3^® IS camera offers simultaneous low-dose imaging via real-time electron counting, alongside a large field of view (FOV) and a rapid rate of continuous data capture.

The STEMx™ system precisely synchronizes the speed of the scanning probe when used in a 4D STEM experiment, matching this to the camera frame rate to eliminate the potential for data loss and ensure high-speed data acquisition.

Context

As the component making up the outer layer of human teeth, dental enamel comprises a sophistically, hierarchically-structured bio-composite material.

A person’s quality of life will be substantially impaired if their dental enamel is damaged in any way. Its structural details are essential to a number of human health outcomes.

Enamel crystallites contain tiny amounts of the mineral apatite, and this can be investigated using (scanning) transmission electron microscopy ((S)TEM). This material only exhibits minimal sensitivity to the electron beam. However, performing such studies is historically challenging.

Atomic resolution STEM imaging has recently demonstrated its capacity to provide detailed insight into the coherent atomic structure of crystallites (Figure 1a). It was noted that to be successful, the following was essential:

1. Cooling the sample to the temperature of liquid nitrogen
2. The use of significant electron doses of ~6 x 10³ e^-/Ų to attain high signal-to-noise (SNR) images while ensuring there was no sample decay
3. A comparatively small FOV of the specimen meant that only a limited number of crystallites could be captured with each image

Materials and Methods

A combination of the K3 IS camera, and a STEMx system was utilized to capture 4D STEM diffraction datasets at 300 frames per second. This was done in electron counting mode on a JEOL JEM-ARM 300F (S)TEM.

Having been prepared using a focused ion beam, a sample of outer tooth enamel was imaged at room temperature rather than the more common approach of performing STEM imaging in cryogenic conditions.

It took around 240 seconds to acquire the complete dataset. It was 512 x 512 x 262 x 266 pixels (132.5 x 132.5 nm²) with a total dose of only ~200 e^-/Ų (Figure 1b and Figure 1c).

Python scripting and the DigitalMicrograph^® software were employed on an offline computer to process the binned-by-two data, allowing the creation of segmented virtual detectors (Figure 1c).

Qualitative analysis was undertaken to evaluate the difference in intensity between two opposing segmented detectors. This revealed that single crystallites exhibited a tilt within the apatite lattice (Figure 1e and Figure 1f).

 Figure 1. a) Cryo-STEM image displaying part of an enamel crystallite. Image adapted with permission from the below reference. b) Average diffraction pattern of the 4D STEM data set. Inset: Single diffraction pattern of the same dataset. c) Average diffraction pattern with 12 segmented virtual detectors (indicated by white lines). d) Virtual image created from the average diffraction pattern, combining intensities from all segmented virtual detectors. e,f) Maps displaying the differences in intensity between two selected opposite partial virtual detectors (indicated in inset). The black arrows indicate places within single grains that show variations in intensity between the opposed virtual detectors, which signify a tilt in the crystallographic directions of the apatite lattice with respect to one other. Image Credit: Gatan Inc.

The presence of this tilt indicates that enamel crystallites are unlikely to be as coherent as the cryo-STEM images suggest. This could impact the crystallites’ mechanical properties, which in turn could offer improved insight into crystallite growth during enamel formation.

Summary

The K3 IS with STEMx is able to confidently facilitate hardware-synchronized 4D STEM diffraction experiments at high speeds (>3500 fps at 256 x 256-pixel resolution) as well as maintaining the highest possible SNR (via electron counting).

These capabilities allow samples to be studied at lower dose conditions, helping to limit the occurrence of beam damage artifacts.

Unlike HRSTEM, using the K3 IS for 4D STEM allows detailed studies of larger FOV (~12x in this case) while providing access to a wide range of angles in the reciprocal space.

This additional information can be employed in the generation of orientation maps, enabling improved characterization and understanding of these types of materials.

Reference

1. DeRocher, K.A., Smeets, P.J.M., Goodge, B.H. et al. Chemical gradients in human enamel crystallites. Nature 583, 66–71 (2020). https://doi.org/10.1038/s41586-020-2433-3

This information has been sourced, reviewed and adapted from materials provided by Gatan Inc.

For more information on this source, please visit Gatan Inc.