DMSE Tenure Talk

A Journey to the Atomic Scale: How Quantitative Electron Microscopy Enables New Physical Insights 

Electron microscopy is one of the most powerful techniques available to investigate materials at the micro, nano, and even picometer length scales. After nearly 100 years of innovation, the capabilities of the tool continue to advance rapidly. Within the past twenty years, for example, electron microscopy has been revolutionized by the advent of the aberration corrector, new detector technologies, and dramatic improvements to electron sources. At the same time, direct measurements from the images had remained largely qualitative or semi-quantitative. This was particularly true for scanning transmission electron microscopy (STEM) images, where accuracy and precision had been significantly hampered by sample drift and scan distortion which prevented the ability to fully characterize and quantify the atomic structure changes that can ultimately define material properties. Moreover, these instruments require months or even years of practice to drive efficiently and to acquire reproducibly atomic scale, quantitative information. As a result, the necessary experience (and often luck) prevents even experts from being able to capture more than a few dozen datasets in a session on the microscope, thereby limiting statistical significance. 

In this talk, I will highlight approaches that we developed to make electron microscopy absolutely quantitative and how we’ve applied those techniques to solve a variety of materials challenges in metals, semiconductors, and functional oxide systems. I will describe the various approaches and demonstrate that the techniques are capable of achieving sub-picometer accuracy and enable real-space crystallographic measurements in STEM.  Multiple case studies will be presented to demonstrate the power of these techniques to understand the physical behavior of materials. I will show, for example, how picometer precise measurements enable the capability to directly observe static atomic displacements within complex oxide solid solutions and to understand how short range order gives rise to the properties of relaxor ferroelectric materials. Looking to the future, I will also discuss our work on applying artificial intelligence to both determine, in realtime, sample thickness/tilt and to automate the scanning transmission electron microscopy workflows, reducing the barrier to reproducibly capture statistically significant data.

Reception to follow talk in 6-104, Chipman Room.