Cornell Tool Maps Single Atoms in Chips to Find Flaws

The technique, developed in a collaboration between Cornell, Taiwan Semiconductor Manufacturing Company (TSMC), and Advanced Semiconductor Materials (ASM), is a form of computational imaging called electron ptychography. It utilizes an advanced electron microscope pixel array detector (EMPAD) to analyze the scattering patterns of electrons as they pass through the chip's transistor structures. This method is so precise it has been recognized by Guinness World Records for producing the highest-resolution images ever captured. The research was led by David Muller, a professor of engineering at Cornell who previously worked at Bell Labs, where the transistor was invented, and doctoral student Shake Karapetyan, the lead author of the paper published in *Nature Communications*. Their work allows for the 3D imaging of individual atoms, revealing the atomic structure of defects that have long been a challenge for the semiconductor industry as components shrink to the atomic scale. This high-resolution imaging has identified specific imperfections nicknamed "mouse bites" at the interfaces within transistor channels. These rough patches, which occur during the manufacturing process, disrupt the flow of electrons through channels that are only about 15 to 18 atoms wide, ultimately slowing down the chip's performance. By pinpointing these previously invisible atomic-scale flaws, chip designers and fabrication engineers can get direct feedback to refine the manufacturing process. This capability is crucial for debugging and fault-finding during the development of next-generation semiconductors, especially for complex 3D architectures. For an electrical engineering student, this represents a significant leap in metrology for failure analysis. The ability to visualize individual impurity atoms and other defects directly impacts the yield and performance of chips designed for AI and quantum computing, where structural precision at the atomic level is paramount.