Featured Publications & Research Highlights using the UC San Diego NE-MRC

We are hoping to grow this page into a periodically rotating showcase of the research and researchers from various different fields that have used the UC San Diego NanoEngineering Materials Research Center (NE-MRC). If you have used the NE-MRC for your research and would like to have your research featured here (and on the respective instrument page) please contact us!

If you have used the NE-MRC for your research please send us an email with the reference for the publication so we can update our publication list.

NanoEngineering / Chemical Engineering / Materials Engineering

Illustration of the inner workings of a convolutional neural network. Each convolutional layer extracts features from its preceding layer, using filters learned from training the model, to form feature maps. These feature maps are then down-sampled by a pooling layer to exploit data locality. A traditional dense neural network computes the probability that the input diffraction pattern belongs to a given Bravais lattice is computed.
Visualizing the features utilized for classification of a diffraction pattern to face-centered cubic (FCC). Each heatmap displays the importance of information in the image for correctly classifying it as face centered cubic.

Research Highlights

Identifying molecular structure is a crucial step in the analysis of proteins, pharmaceuticals, geology, and materials science because crystal structure often plays an important role in the properties exhibited. Electron diffraction is one of the most common methods of structure analysis in each of these fields. One of the major challenges within the world of electron diffraction is analysis of the collected diffraction patterns.

Kevin Kaufmann et al. have developed a computer algorithm that independently analyzes diffraction patterns utilizing the same type of deep learning (a subset of artificial intelligence and machine learning) algorithm as is applied in facial recognition and self-driving cars. In this work, they demonstrate the method using a scanning electron microscope (SEM) to collect electron backscatter diffraction (EBSD) patterns.

A modern EBSD system enables determination of fine-scale grain structures, crystal orientations, relative residual stress/strain, and other information. The drawback of commercial EBSD systems is the software’s inability to determine the atomic structure of the crystalline lattices present within the material being analyzed. Using the newly designed method, the deep neural network independently analyzes each diffraction pattern to determine the lattice, out of all possible lattice structure types, with a high degree of accuracy.

Link to article: https://science.sciencemag.org/content/367/6477/564.full

Full citation: Kevin Kaufmann, Chaoyi Zhu, Alexander S. Rosengarten, Daniel Maryanovsky, Tyler J. Harrington, Eduardo Marin, and Kenneth S. Vecchio, “Crystal symmetry determination in electron diffraction using machine learning” Science, 2020, 367, 564-568.

Researcher Profile

Kevin Kaufmann is a fifth year PhD Candidate in NanoEngineering under the guidance of Dr. Kenneth Vecchio at UC San Diego, where he also completed his Bachelors and Masters degrees. His doctoral studies are focused on the application of data science, machine learning, and artificial intelligence to solve materials science problems. He is currently involved in the development of a suite of tools for accelerating each step in the material development process from design to analysis. Kevin’s interest in these topics developed from an interest in data-driven materials science and the opportunity to streamline the material discovery loop.

Photo credit: Liezel Labios

SIO Oceanography / Geosciences / Marine Sciences

Research Highlights

Researcher profile

Other Disciplines

Research Highlights

Researcher profile