AI improves electron microscopy
Their approach, say the researchers, enables scientists to get even more detailed information about materials and the microscope itself, which can further expand its uses.
“Our method,” says Argonne assistant scientist and lead author of a paper on the research Tao Zhou, “is helping improve the resolution of existing instruments so people don’t need to upgrade to new expensive hardware so often.”
With resolution 1,000 times greater than a light microscope, electron microscopes are exceptionally good at imaging materials and detailing their properties. Electrons act like waves when they travel, and electron microscopes exploit this knowledge to create images.
Images are formed when a material is exposed to a beam of electron waves. Passing through, these waves interact with the material, and this interaction is captured by a detector and measured. These measurements are used to construct a magnified image.
Along with creating magnified images, electron microscopes also capture information about material properties, such as magnetization and electrostatic potential, which is the energy needed to move a charge against an electric field. This information, say the researchers, is stored in a property of the electron wave known as “phase,” which describes the location or timing of a point within a wave cycle, such as the point where a wave reaches its peak.
When measurements are taken, information about the phase is seemingly lost. As a result, scientists cannot access information about magnetization or electrostatic potential from the images they acquire.
“Knowing these characteristics is critical to controlling and engineering desired properties in materials for batteries, electronics and other devices,” says Argonne material scientist and group leader Charudatta Phatak, a co-author of the paper. “That’s why retrieving phase information is important.”
Retrieving such phase information is a decades-old problem, say the researchers. It originated in X-ray imaging and is now shared by other fields, including electron microscopy. To resolve this problem, the researchers propose leveraging tools built to train deep neural networks, a form of AI.
Neural networks are essentially a series of algorithms designed to mimic the human brain and nervous system. When given a series of inputs and output, these algorithms seek to map out the relationship between the two. But to do this accurately, neural networks have to be trained. That’s where training algorithms come into play.
“Tech companies like Google and Facebook have developed packages of software that are designed to train neural networks,” says said Cherukara. “What we’ve essentially done is taken those and applied them to the scientific challenge of phase retrieval.”
Using these training algorithms, the researchers demonstrated a way to recover phase information. But what makes their approach unique, they say, is that it also enables scientists to retrieve essential information about their electron microscope.
“Normally when you’re trying to retrieve the phase, you presume you know your microscope parameters perfectly,” says Zhou. “However, that knowledge might not be accurate. With our method, you don’t have to rely on this assumption. Instead, you actually get the conditions of your microscope – that’s something other phase retrieval methods can’t do.”
Their method also improves the resolution and sensitivity of existing equipment. This means that researchers will be able to recover tiny shifts in phase, and in turn, get information about small changes in magnetization and electrostatic potential, all without requiring costly hardware upgrades.
“Just doing a software upgrade we were able to improve the spatial resolution, accuracy and sensitivity of our microscopy,” says Zhou. “The fact that we didn’t need to add any new equipment to leverage these benefits is a huge advantage from an experimentalist’s point of view.”