(Funded by the U.S. Department of Energy)
Scientists at the U.S. Department of Energy’s (DOE) Argonne National Laboratory have unveiled a novel approach to understanding stochasticity in tiny magnetic structures. Their work explores the intricate decision-making processes of nanomagnetic Galton boards, a modern take on a classical concept in statistics and computing. Their insights have the potential to transform computing architectures, leading to more sophisticated neural networks and enhancing encryption technologies to secure data against cyber threats. A Galton board uses a triangular array of pegs. As balls fall through the grid, they randomly bounce left or right, eventually landing somewhere along the bottom. In a nanomagnetic version of the Galton board instead of pegs, the boards use tiny magnetic structures made from a nickel-iron alloy. Instead of balls, they use domain walls, which are boundaries that separate regions with different magnetic orientations within a material. Nanostructures in this work were fabricated at the Center for Nanoscale Materials, a DOE Office of Science user facility at Argonne.

(Funded by the U.S. Department of Energy)
Researchers at the U.S. Department of Energy’s Lawrence Livermore National Laboratory have developed a new method to deposit quantum-dot films on corrugated surfaces. The researchers used electrophoretic deposition, which drives the quantum dots through a solution with an electric field toward an electrode with the opposite charge. When they reach that electrode, the quantum dots assemble into a film. Traditionally, quantum dots are made with long organic ligands – molecules that bind to the dots and stabilize them in solution. But after the quantum dots are deposited as a film, those long ligands act as insulators and limit device performance, so they are removed with post-processing. In this study, the researchers made quantum dot films using short ligands, which are more conductive and negate the need for post-processing.

(Funded by the U.S. Department of Energy)
Researchers from the Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) and the University of California, Los Angeles, have created a heat pump that consists of stacked layers of electrocaloric materials, which temporarily change temperature in response to an electric field. Six polymer film discs, each about an inch in diameter and coated with carbon nanotubes, serve as a heat pump, moving warmth from the layer closest to the heat source away to the outermost layer. The nanotubes function as conductors for the electric field that stimulates the material. A proof of concept lowered ambient temperatures by 16 degrees Fahrenheit within 30 seconds, and readings at the edge of the device dipped as low as 25 degrees Fahrenheit.

(Funded by the U.S. Department of Energy)
Scientists at Lawrence Berkeley National Laboratory (Berkeley Lab) and SLAC National Accelerator Laboratory have revealed the fundamental mechanisms that limit the performance of copper nanocatalysts – critical components in chemical reactions that transform carbon dioxide and water into valuable fuels and chemicals. Copper’s catalytic properties quickly degrade during these reactions, diminishing its performance over time. The researchers identified and observed two competing mechanisms that drive the copper nanoparticles that make up the nanocatalysts to the brink of degradation: nanoparticle migration and coalescence, in which smaller particles combine into larger ones, and Ostwald ripening, where larger particles grow at the expense of smaller particles. These findings suggest mitigation strategies to protect the copper nanocatalysts by limiting either mechanism. Part of the research was conducted at the Molecular Foundry, a DOE Office of Science national user facility at Berkeley Lab.

(Funded by the U.S. Department of Energy and the U.S. National Science Foundation)
For more than 100 years, scientists have used a method called crystallography to determine the atomic structure of materials, but this technique only works well when researchers have large, pure crystals. For a powder of nanocrystals, the method only hints at the unseen structure. Now, scientists at Columbia Engineering have created a machine learning algorithm that can observe the pattern produced by a powder of nanocrystals to infer their atomic structures. The scientists began with a dataset of 40,000 crystal structures and jumbled their atomic positions until they were indistinguishable from random placement. Then, they trained a deep neural network to connect these almost randomly placed nanocrystals with their associated X-ray diffraction patterns. Lastly, the algorithm was able to determine the atomic structure from nanocrystals of various shapes in the powder.