As researchers continue to push the boundaries of battery design, seeking to pack more power and energy into a given space or weight, one of the more promising technologies being investigated is the lithium-ion battery, which uses a solid electrolyte material between two electrodes instead of the typical liquid.
However, this type of battery has been plagued by a tendency for dendritic metal protrusions to form on one of the electrodes, eventually connecting to the electrolyte and causing a short circuit. Now, researchers at MIT and elsewhere have found a way to prevent this dendrite formation, which could unlock the potential of this new high-energy battery.
The findings of this study, published in the journal Nature Energy, were completed by MIT graduate students Richard Parker, Professor Huiming Jiang, and Professor Craig Carter, along with seven other researchers from MIT, Texas A&M University, Brown University, and Carnegie Mellon University.
Jiang explained that solid-state batteries are a long-sought-after technology for two reasons: safety and energy density. However, he said, "The only way to achieve this interesting energy density is to use metal electrodes." He added that while it's possible to combine metal electrodes with liquid electrolytes and still achieve good energy density, this doesn't offer the same safety advantages as solid electrolytes.
He said that solid-state batteries only make sense on metal electrodes, but attempts to develop such batteries have been hampered by dendrite growth, which eventually connects the gap between the two electrode plates, causing short circuits, weakening or deactivating the cells in the battery.
We already know that dendrites form faster at higher currents, which is typically necessary for fast charging. So far, the current density achieved in experimental solid-state batteries is far lower than that required for practical commercially available rechargeable batteries. But Jiang says this prospect is worth pursuing because experimental versions of this battery can already store twice the energy of conventional lithium-ion batteries.
The team solved the dendrite problem by making a compromise between solid and liquid states. They fabricated a semi-solid electrode that contacts a solid electrolyte material. The semi-solid electrode provides a self-healing surface at the interface, rather than the brittle surface of a solid, which can lead to tiny cracks, providing the initial seeds for dendrite formation.
This idea was inspired by experimental high-temperature batteries, where one or both electrodes are composed of molten metal. According to Park, the paper's first author, molten metal batteries at hundreds of degrees Celsius are impractical for portable devices, but this work does demonstrate that high current densities can be achieved at liquid interfaces without dendrite formation. "Our motivation was to develop electrodes based on carefully selected alloys to introduce a liquid phase that can serve as a self-healing element for the metal electrodes," Park said.
He explained that the material is more solid than liquid, but similar to the amalgam dentists use to fill cavities, it can still flow and take shape. At the battery's normal operating temperature, "it's in a state where both solid and liquid phases exist simultaneously," in which case the solid phase consists of a mixture of sodium and potassium. Jiang said the research team demonstrated that it's possible to operate the system at 20 times the current used with solid lithium without forming any dendrites. The next step is to replicate this performance with an actual lithium-containing electrode.
In the second version of the solid-state battery, the team introduced a very thin layer of liquid sodium-potassium alloy between the solid-state lithium electrode and the solid-state electrolyte. They showed that this method can also overcome the dendrite problem, providing another approach for further research.
Jiang said this new method can be easily applied to many different versions of solid-state lithium batteries, which are currently being studied by researchers around the world. He said the team's next step is to demonstrate the system's applicability in various battery architectures. Co-author Viswanathan, a professor of mechanical engineering at Carnegie Mellon University, said, "We believe we can translate this method into any solid-state lithium-ion battery. We believe it can be immediately applied to battery development, with broad applications in handheld devices, electric vehicles, and electric vehicles."
