In the quest to improve on the lithium-ion batteries that power today's mobile devices and electric vehicles, lithium-metal batteries hold a lot of potential. Although they promise to hold much more energy per charge, they do have shortcomings in their current form, particularly when it comes to safety. Engineers at Northwestern University are now claiming to have overcome these failings by making use of crumpled balls of graphene that fall into line as a scaffold when the battery is charging.

"In current batteries, lithium is usually atomically distributed in another material such as graphite or silicon in the anode," explains Northwestern Engineering's Jiaxing Huang. "But using an additional material 'dilutes' the battery's performance. Lithium is already a metal, so why not use lithium by itself?"

The reason is dendrites, which are microscopic lithium fibers that accumulate on the surface of the anode as the battery charges. As they spread, the dendrites hinder the performance of the battery and eventually cause it to short-circuit or catch fire. We have seen researchers look to counter this problem by using different electrolytes, reinforcing batteries with Kevlar and creating forests of 3D nanotubes that stop dendrites from forming.

The team at Northwestern University's approach actually builds on a type of graphene it developed back in 2011. Literally taking inspiration from a trashcan of scrunched up paper, the engineers came up with a way of crumpling graphene sheets into balls by first atomizing them into tiny water droplets. When the water evaporates it then creates a capillary force that crumples the sheets into the rough ball shape.

At the time, the engineers noted that graphene in this form had good potential for energy storage because its retains the same electrical properties and surface area as graphene sheets, but larger amounts can be packed within a tight structure.

Now they are putting their crumpled graphene balls to work as a porous scaffold for an anode in a new type of lithium metal battery. Porous scaffolds are seen as a promising part of the mix for these types of experimental devices because as the battery is charging, lithium deposits preferentially onto their surface in a way that avoids dendrite growth.

But in solving one problem, they create another. As the lithium coats the scaffold surface and then dissolves as the battery cycles, it causes the structure to expand and contract like a wet and dry sponge, in turn creating stress that can break it apart all together. While creating stronger scaffolds might seem the obvious solution, Huang and his team have taken a different approach.

"Our strategy is a reverse thinking," Huang tells New Atlas. "The particles are not bonded together. When soaking up lithium, it is a continuous piece of conductive material (lithium and graphene). When lithium is stripped, the particles can reorganize or "snuggle" together to form a continuous and uniform layer of graphene particles."

With this new scaffold that can withstand the volume fluctuations of the battery cycles, the team put it to work in a full cell battery, where it acted as a uniform, continuous network for the lithium ions to flow across the surface. The researchers say the battery is much lighter than a typical lithium battery, and where traditionally they can encapsulate lithium that is tens of microns thick, their solution can hold lithium stacked 150 microns high. The engineers have filed a provisional patent for the technology, though there still remains plenty to learn about the intricacies of it.

"There are a number of very intriguing fundamental science questions based on the results," Huang tells us. "For example, I am especially interested in why lithium-metal seems to still grow smoothly, instead of forming dendrite, when they grow outside the layer of crumpled graphene particles. Your own discoveries and observations are often the best inspiration for your next work."

The research was published in the journal Joule.