Scientists Uncover “Mouse Bite” Defects Inside Computer Chips

A stunning new imaging breakthrough lets scientists see — and fix — the atomic flaws hiding inside tomorrow’s computer chips.

Researchers at Cornell University have achieved something chipmakers have long wanted. Using advanced high-resolution 3D imaging, they have directly observed atomic-scale defects inside computer chips for the first time. These tiny flaws can interfere with performance and reliability in modern electronics.

The new imaging approach was developed in collaboration with Taiwan Semiconductor Manufacturing Company (TSMC) and Advanced Semiconductor Materials (ASM). Because computer chips power everything from smartphones and cars to AI data centers and quantum computers, the impact of this work could extend across nearly all modern technology.

The findings were published on February 23 in Nature Communications. Doctoral student Shake Karapetyan is the lead author of the study.

“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, the Samuel B. Eckert Professor of Engineering in the Cornell Duffield College of Engineering, who led the project.

Why Atomic Scale Defects Matter

For decades, microscopic defects have challenged the semiconductor industry. As chips have become more complex and components have shrunk to the scale of individual atoms, even the smallest structural irregularity can have measurable consequences.

At the center of every computer chip is the transistor, a tiny switch that controls the flow of electricity. Each transistor contains a channel that opens and closes to allow electrons to move through it.

“The transistor is like a little pipe for electrons instead of water,” Muller said. “You can imagine, if the walls of the pipe are very rough, it’s going to slow things down. And so measuring how rough the walls are and which walls are good and which walls are bad is now even more important.”

Modern high-performance chips can contain billions of these transistors. Today, a transistor channel may be only about 15 to 18 atoms wide, making precise measurement extremely difficult.

“These days, a transistor channel can be only about 15 to 18 atoms wide, which is super, super tiny, and they’re extremely intricate,” Karapetyan said. “At this point, it matters where every atom is, and it’s really hard to characterize.”

From Early Transistors to 3D Chip Structures

Muller’s experience in semiconductor research stretches back to his time at Bell Labs from 1997 to 2003, where transistors were originally invented. There, he studied the physical limits that determine how small a transistor can be.

In the early days, transistors were built in flat layouts that spread outward across the chip surface. As manufacturers ran out of space, engineers began stacking them vertically, creating complex three-dimensional structures.

“The problem is these 3D structures are smaller than the size of a virus. And these days, it’s a lot smaller. It’s more like a molecule-in-the-cell kind of scale,” Muller said.

As devices have become more compact, diagnosing performance problems has grown increasingly challenging.

The Rise of Advanced Electron Microscopy

While at Bell Labs, Muller and colleague Glen Wilk ’90, now vice president of technology at ASM, worked on replacing silicon dioxide, which leaked too much current at small scales, with hafnium oxide as a gate material. Their research helped pave the way for hafnium oxide to become the industry standard in computers and cell phones in the mid-aughts.

“The papers we published on how to use electron microscopes to characterize these materials, I can tell you, a lot of the semiconductor folks had read those very, very carefully,” said Muller, who co-directs the Kavli Institute at Cornell for Nanoscale Science and the Cornell Center for Materials Research (CCMR). “When we got back into this project, that was very clear. And the microscopy has gone a very long way. Back then, it was like flying biplanes. And now you’ve got jets.”

The “jet” he refers to is electron ptychography, a powerful computational imaging technique. It relies on an electron microscope pixel array detector (EMPAD), a technology co-developed by Muller’s group, to capture detailed electron scattering patterns after electrons pass through a transistor.

By analyzing how those patterns shift from one scan point to another, scientists can reconstruct images with exceptional clarity. The detector is so precise that it has produced the highest resolution images ever recorded, revealing atoms in unprecedented detail, a feat recognized by Guinness World Records.

Detecting “Mouse Bite” Defects

More than 25 years after their previous collaboration, Muller and Wilk teamed up again, this time with support from TSMC and its Corporate Analytical Laboratories group. Their goal was to apply EMPAD technology to modern semiconductor devices.

“You can think of this imaging technique like solving a massive puzzle, both in terms of taking the experimental data and doing the computational reconstruction,” Karapetyan said.

After gathering and processing the data, the team tracked the positions of individual atoms and identified subtle roughness along the transistor channels. Karapetyan referred to these tiny irregularities as “mouse bites.” The defects formed during the optimized growth process used to manufacture the structures. Test samples were produced at the nanoelectronics research center Imec.

“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,” Karapetyan said. “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.”

Implications for AI and Quantum Computing

The ability to directly visualize atomic-scale defects could influence nearly every technology that depends on advanced chips, including smartphones, laptops, and data centers. It may be especially valuable for next-generation systems such as quantum computers, which demand extraordinary precision in material structure.

“I think there’s a lot more science we can do now, and a lot more engineering control, having this tool,” Karapetyan said.

Reference: “3D atomic-scale metrology of strain relaxation and roughness in Gate-All-Around transistors via electron ptychography” by Shake Karapetyan, Steven E. Zeltmann, Glen Wilk, Ta-Kun Chen, Vincent D.-H. Hou and David A. Muller, 23 February 2026, Nature Communications.
DOI: 10.1038/s41467-026-69733-1

Co-authors of the study include Steven Zeltmann of the Platform for the Accelerated Realization, Analysis and Discovery of Interface Materials (PARADIM), along with Ta-Kun Chen and Vincent Hou of TSMC.

The research was funded by TSMC. Microscopy facilities were supported by CCMR and PARADIM, both funded by the National Science Foundation.

Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.