Computational electron microscopy; measuring multiple properties at once; hexagonal boron nitride defects.
Researchers from Cornell University, TSMC, and ASM used electron ptychography for atomic-scale defect inspection of transistors.
The computational imaging method uses an extremely precise electron microscope pixel array detector (EMPAD) to collect detailed scattering patterns of electrons after they pass through transistors and compare how the patterns change from one scan position to another.
“Since there’s really no other way you can see the atomic structure of these defects, this is going to be a really important characterization tool for debugging and fault-finding in computer chips, especially at the development stage,” said David Muller, professor of engineering in the Cornell Duffield College of Engineering, in a press release.
The approach enabled the researchers to detect interface roughness in the channel structures arising from defects that formed during the optimized growth process. The researchers propose using the method to check the fabrication process after each step.
“Fabrication of modern devices takes hundreds, if not thousands, of steps of chemical etching and deposition and heating, and then every single step does something to your structure,” said Shake Karapetyan, a doctoral student at Cornell, in a press release. “Before you used to look at projective images to try to figure out what was really going on. Now you have a direct probe to actually see after every single step and have a better grasp of, oh, I put the temperature this high, and then this is what it looks like.” [1]
Researchers from Leiden University and QuantaMap built a superconducting quantum interference device (SQUID) microscope that can measure the temperature, magnetism, structure, and electrical properties of a material in a single scan with nanoscale precision.
“In a real quantum device, all physical properties are closely intertwined. If you only study one aspect at a time, you never get ahead. With our microscope, we reveal the relationships between non-equilibrium properties such as current and dissipation, and how those relate to chip structure,” said Matthijs Rog, a PhD student at Leiden, in a press release. “Most existing microscopes only work well with very flat samples. That’s limiting, because many of the most interesting effects occur at the edges of materials or at the boundary between two different quantum materials. Our microscope has no trouble with that at all – it can examine a bumpy chip just as easily as a flat crystal.”
The ‘Tapping Mode SQUID-on-Tip’ (TM-SOT) microscope enables multimodal imaging to be performed extremely close to the sample surface using tapping mode feedback. This allows for stability during extended measurement campaigns and imaging of realistic, nanostructured devices, including operational quantum chips and exotic quantum materials.
“Current chip testing relies mainly on electrical characterization of fully finished chips inside quantum computers – a process that takes weeks per chip and if the performance of some qubits is reduced, it cannot reveal the underlying reason,” said Johannes Jobst, founder and CEO of QuantaMap, in a press release. “This new imaging platform addresses that gap. The ability to perform root‑cause analysis at the nanoscale makes it possible to identify reasons for failure in quantum chips at any fabrication stage, correlating device performance with local material behavior, and faster design–fabrication–test feedback loops.” [2]
Researchers from Rice University combined electron microscopy, cathodoluminescence mapping, and force-based measurements to detect defects in hexagonal boron nitride (hBN) films.
Long, narrow misalignments known as stacking faults can occur when flakes of hBN are peeled from a bulk crystal and transferred to silicon and silicon dioxide wafers. When imaged using regular optical or atomic force microscopes, the flakes looked smooth. However, cathodoluminescence spectroscopy revealed the stacking defects.
“hBN emits deep ultraviolet light that many labs cannot easily excite. This emission map revealed bright, narrow stacking faults that other methods miss ⎯ one reason they have been overlooked,” said Hae Yeon Lee, an assistant professor of materials science and nanoengineering at Rice, in a statement. “These hidden defects act like tiny charge pockets and weaken insulation: The same hBN can start leaking electricity at much lower voltage along the defects than nearby areas.”
The researchers say the approach can also be applied to other layered materials. [3]
[1] S. Karapetyan, S.E. Zeltmann, G. Wilk, et al. 3D atomic-scale metrology of strain relaxation and roughness in Gate-All-Around transistors via electron ptychography. Nat Commun (2026). https://doi.org/10.1038/s41467-026-69733-1
[2] M. Rog, T. J. Blom, D. B. Boltje, et al. Tapping-Mode SQUID-on-Tip Microscopy with Proximity Josephson Junctions. Nano Letters 2026 26 (5), 1608-1615 https://dx.doi.org/10.1021/acs.nanolett.5c04571
[3] T. Lang, Y. Liu, A. Chugh, et al. Hidden Stacking Fault Charge Traps in Hexagonal Boron Nitride and Their Impact on Dielectric Breakdown. Nano Letters 2026. https://doi.org/10.1021/acs.nanolett.5c06347
 |
 |
 |
 |
 |  |
 |  |
 |
Leave a Reply