Two-dimensional (2D) materials are attractive as potential separation membranes due to their highly controllable structure and single atom thickness. The careful creation of nanopores through controlled perforation can result in highly selective transport properties. Additionally, the selectivity and permeation flux benefit from the extreme thinness of the membrane [1]. Here, we present an investigation of controllably synthesized nanometer-scale pores using aberration-corrected STEM imaging. Monolayer materials such as hexagonal boron nitride (hBN) and graphene can be difficult to image with all but the most specialized of microscopes [2]. Due to their small interatomic spacings (1-2Å), low-Z composition, and single atom thickness, imaging must be performed at the highest resolutions, adequately low accelerating voltages, and high electron doses. While lowering acceleration voltage helps mitigate kinetic damage, it also generally decreases ultimate resolution. This relationship has cemented 60-80 keV accelerating voltages as the standard for low-Z 2D materials. Even at these lower voltages, monolayers can damage under the beam during image acquisition. Imaging nanopores in these monolayer materials presents an additional challenge. Small, atomic scale nanopores must be quickly identified and studied before the electron beam causes damage to the as-synthesized structure. For this reason, it is critical to record images using as low a dose as necessary. All imaging was performed using a JEOL NEOARM transmission electron microscope equipped with a CEOS ASCOR probe corrector. All imaging was performed at 80 keV to minimize damage to the monolayers while retaining atomic resolution. Images were formed on a JEOL annular detector with camera lengths from 8-20 cm. The nanoporous monolayers were stored in an inert environment to minimize potential growth of the nanopores. Prior to imaging, monolayers were heated on a benchtop hotplate to gently clean the sample. Pixel dwell times varied from 32-128 μs to maximize the signal coming from low-Z elements. STEM images were acquired at 512 x 512 and 1024 x 1024 pixel resolution. Hundreds of STEM images were acquired over the course of the study to obtain statistically significant data on nanopore size and distribution. The size of individual nanopores was calculated using the average of their triangular edge lengths. Synthesized nanopores occupied a range of sizes, from 0.5-5.0 nm. As the nanopores are beam sensitive, some amount of damage during imaging is expected. For this reason, the range of nanopore sizes should be taken as an upper limit. Pores identified near tears or edges of the monolayer were not included in the statistical analysis as they were not reflective of the nanopores identified in the bulk of the monolayer. In the presented materials we will compare and contrast similar results from nanopores in monolayer hBN and graphene. The inherent strengths of 2D materials come from their high degree of crystalline perfection. Nanopores in 2D materials necessarily involve local rearrangements and dangling bonds which make those areas inherently more beam sensitive. While researching and imaging these structures is of the utmost importance to these nanopore membranes, it represents a challenging frontier for the study of structure property relationships [3]. High resolution STEM (left) image of nanometer scale nanopore in hBN. Lower magnification STEM image (right) of the same nanopore.
No takes yet. Share an insight, caveat, or question.
Hall et al. (2026) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: