In a great example of a low-cost research solution that could deliver big results, University of Michigan scientists have created a window for lithium-based batteries in order to film them as they charge and discharge.
The future of lithium-ion batteries is limited, says University of Michigan researcher Neil Dasgupta, because the chemistry cannot be pushed much further than it already has. Next-generation lithium cells will likely use lithium air and lithium sulfur chemistries. One of the big hurdles to be overcome in making these batteries practical is dendrites - tiny branch-like structures of lithium that form on the electrodes.
Dendrites can pierce cell walls or limit the lithium's potential. Neither is good for battery efficiency and can even lead to fires and explosions like those that have plagued the Samsung Galaxy Note 7 – though the jury is still out on exactly what caused the Note 7's well documented woes.
Battery cells are normally tested through cycles of charge and discharge, testing the capacity and flow potential of the cells before being dissected. Dasgupta and his team took a different approach. They added a window to a lithium cell so that they could film the dendrites forming and deforming during charge and discharge cycles. This has helped the team gain new insight into how dendrites form in real-time and how they can be mitigated.
They found that as the battery cycled, dendrites would grow during lithium accumulation on the electrode and then shrink on the reverse cycle. As the lithium pulled away from the electrode's surface, small pits in the metal were left behind. These pits became nucleation sites for dendrites during the next cycle. With each cycle, the dendrites grew larger. The team describes the dendrites growing as "organic, like plants growing and withering over the course of a battery cycle."
Dendrites can also be beneficial to the battery's process, the team found. If the dendrites grow in an even carpet over the battery's interior, remaining small, they can keep more lithium in play during the process, keeping battery performance stable. The U of M team believes that learning to control dendrites in an optimal way should be the goal of battery design going forward, because eliminating them altogether is not very likely.
The team says that the window solution is usable by researchers globally and costs about US$100 to implement. The findings have been published in ACS Central Science.
Dasgupta discusses the research in the following video.
Source: University of Michigan
