Li-ion-conducting solid electrolytes can simultaneously overcome two grand challenges for Li-ion batteries: the severe safety concerns that limit the large-scale application and the poor electrolyte stability that forbids the use of high-voltage cathodes. Nevertheless, the ionic conductivity of solid electrolytes is typically low, compromising the battery performances.

Precisely determining the ionic transport mechanism(s) is a prerequisite for the rational design of highly conductive solid electrolytes. For decades, the research on this subject has primarily focused on the atomic and microscopic scales, where the main features of interest are unit cells and microstructures, respectively.

Here, it is shown that the largely overlooked mesoscopic scale lying between these extremes could be the key to fast ionic conduction. In a prototype system, (Li_0.33La_0.56)TiO₃, a mesoscopic framework is revealed for the first time by state-of-the-art scanning transmission electron microscopy. Corroborated by theoretical calculations and impedance measurements, it is demonstrated that such a unique configuration maximizes the number of percolation directions and thus most effectively improves the ionic conductivity. This discovery reconciles the long-standing structure–property inconsistency in (Li_0.33La_0.56)TiO₃ and also identifies mesoscopic ordering as a promising general strategy for optimizing Li⁺ conduction.

—Ma et al.

To directly observe the atomic arrangement in the solid electrolyte, the researchers used aberration-corrected scanning transmission electron microscopy to send electrons through a sample. To observe an extremely small feature in a three-dimensional (3D) material with a method that essentially provides a two-dimensional (2D) projection, they needed a sample of extraordinary thinness. To prepare one, they relied on comprehensive materials processing and characterization capabilities of the Center for Nanophase Materials Sciences, a DOE Office of Science User Facility at ORNL.

Usually the transmission electron microscopy specimen is 20 nanometers thick, but Ma developed a method to make the specimen ultra-thin (approximately 5 nanometers). That was the key because such a thickness is comparable to the size of the hidden feature we finally resolved.

—Miaofang Chi, senior author

The researchers examined a prototype system called LLTO, shorthand for its lithium, lanthanum, titanium and oxygen building blocks. LLTO possesses the highest bulk conductivity among oxide systems.

In this material, lithium ions move fastest in the planar 2D pathways resulting from alternating stacks of atomic layers rich in either lanthanum or lithium. The ORNL-led team was the first to see that, without hurting this superior 2D transport, tiny domains, or fine features approximately 5 to 10 nanometers wide, throughout the 3D material provided more directions in which lithium ions could move. The domains looked like sets of shelves stacked at right angles to others. The smaller the shelves, the easier it was for ions to flow in the direction of an applied current.

ORNL’s Yongqiang Cheng and Bobby Sumpter performed molecular dynamics simulations that corroborated the experimental findings.

Previously, scientists looked at the atomic structure of the simplest repeating unit of a crystal—called a unit cell—and rearranged its atoms or introduced different elements to see how they could facilitate ion transport. Unit cells are typically less than 1 nanometer wide. In the material that the ORNL scientists studied for this paper, the unit cell is nearly half a nanometer. The team’s unexpected finding—that fine features, of only a few nanometers and traversing a few unit cells, can maximize the number of ionic transport pathways—provides new perspective.

The finding adds a new criterion. This largely overlooked length scale could be the key to fast ionic conduction.

—Miaofang Chi

Researchers will need to consider phenomena on the order of several nanometers when designing materials for fast ion conduction.

The prototype material has high ionic conductivity because not only does it maintain unit-cell structure, but also it adds this fine feature, which underpins 3D pathways. We’re not saying that we shouldn’t be looking at the unit-cell scale. We’re saying that in addition to the unit cell scale, we should also be looking at the scale of several unit cells. Sometimes that outweighs the importance of one unit cell.

—Cheng Ma

For several decades, when researchers had no explanation for certain material behaviors, they speculated phenomena transcending one unit cell could be at play. But they never saw evidence. This is the first time we proved it experimentally, Ma said. “This is a direct observation, so it is the most solid evidence.”

The DOE Office of Science supported electron microscopy, theory calculations and electrochemical analysis. Work was performed at the Center for Nanophase Materials Sciences, a DOE Office of Science User Facility at ORNL. Researchers at Tsinghua University in China participated in materials synthesis and electrochemical analysis.

Resources

• Ma, C., Cheng, Y., Chen, K., Li, J., Sumpter, B. G., Nan, C.-W., More, K. L., Dudney, N. J. and Chi, M. (2016), “Mesoscopic Framework Enables Facile Ionic Transport in Solid Electrolytes for Li Batteries.” Adv. Energy Mater., 1600053 doi: 10.1002/aenm.201600053