Accelerating the ceramic synthesis of a classic superconductor

The Sun group is leading an international team in developing a new computational model to design ceramic reactions, which can be broadly applied to accelerate the manufacturing and development of functional materials.
Accelerating the ceramic synthesis of a classic superconductor

Assistant Professor Wenhao Sun

Ceramics are found in more than just pottery—they also comprise the active materials in Li-ion batteries, superconductors, solid-oxide fuel cells and more. These advanced ceramics have complicated crystal structures and chemical compositions, and are often difficult to synthesize in the laboratory.

Published this week in Advanced Materials, an international team of scientists at the University of Michigan, Hokkaido University and UC Berkeley revealed new insights into how ceramic reactions proceed. These fundamental insights will guide chemists in better preparing functional materials, which is crucial for accelerating materials manufacturing and deployment.   

Using the high-temperature superconductor YBa2Cu3O6+x (YBCO) as a model system, the team used advanced electron microscopy and X-ray diffraction techniques, along with state-of-the-art computational modeling, to examine how different precursors affect the formation of this classic ceramic material.

“Typically, YBCO is synthesized from BaCO3, Y2O3 and CuO, which can take upwards of 12 hours at 940°C,” says Assistant Professor Wenhao Sun from the University of Michigan. “However, there is an interesting report that replacing BaCO3 with BaO2 reduces the reaction to four hours, and results in a synthesis product without impurities.”

By watching this reaction in a high-temperature transmission electron microscope, the team observed a surprising liquid melt that remarkably led to YBCO formation in only 30 minutes. This observation inspired a new computational model to design ceramic reactions, which can be broadly applied to accelerate the manufacturing and development of functional materials. 

Breaking down ceramic reactions

To understand why ceramic synthesis typically takes so long, the team first thought about how these reactions might look microscopically. “If we think of ceramic precursors like grains of sand, it becomes clear that three different precursors touch only at a single point,” says Sun. "This means that reactions wouldn’t happen between all 3 precursors at once. Instead, reactions should occur much more readily at the interfaces between two precursors at a time.”

With three precursors, a reaction cannot be analyzed as A + B + C → ABC. There should be a sequence of two reactions; with the first reaction being A + B → AB, and then AB + C → ABC. The team then built a thermodynamic model to predict which pair of precursors react first.

When starting with the traditional BaCO3 precursor, the Y2O3 + CuO reaction was expected to occur first. Using powerful X-ray beams at the SPring-8 synchrotron in Japan, the experimental team confirmed this prediction, and found the Y2O3 + CuO reaction to be extremely slow.

When starting with the alternative BaO2 precursor, BaO2 + CuO was calculated to become the preferable first reaction. This prediction was confirmed by the X-ray characterization, but the successful synthesis of YBCO in only 30 minutes shocked the team. By watching the reaction in the electron microscope, the team saw the surprising formation of a liquid melt, which resolved the mystery of the fast reaction.

The unexpected role of the liquid

The conventional intuition in ceramic synthesis is that atomic diffusion is slow because the precursors are all solids. To accelerate diffusion, synthesis usually occurs at very high temperatures (greater than 700°C), and even then may take half a day or more to complete.

However, when starting with the BaO2 precursor, the reaction proceeded through a number of liquid phase intermediates. These liquid phases gave way to fast atomic diffusion, which provided the fast reaction kinetics for YBCO formation. "By looking at the thermodynamic phase diagram, we can anticipate which precursors can provide these low-temperature liquid melts," says Sun. "With our new theoretical model, we can design better precursors to drive reactions through reaction pathways with favorable thermodynamics and kinetics for time- and energy-efficient reactions."

Akira Miura, the lead experimentalist on the work, commented on the importance of the combined experimental and computational approach: “Without a model, our observations would not be generalizable to other systems," he said. "Without the observation, it would not be possible to validate the computational model. Only together could we build new fundamental insights on the synthesis of this classic ceramic system.”

Ultimately, the team aims to continue working together to understand how materials form, and how to control chemical reactions to produce desired materials. “Computational materials predictions routinely identify new materials for safer batteries, better solar cells, stronger steels, etc. Now the crucial question is—how do we make them? Our research aims to resolve this critical synthesis bottleneck.”

Article: “Observing and modeling the sequential pairwise reactions that drive solid-state ceramic synthesis,” Advanced Materials (2021), DOI: 10.1002/adma.202100312

Funding for this work came from the Department of Energy Office of Science through an Early Career Award for Sun, and GENESIS (A Next GENeration SyntheESIS Center), a Department of Energy Frontier Research Center. The experimental work was partially supported by KAKENHI Grant Nos. JP16K21724, JP19H04682, and JP20KK0124.