Healing of dendrites in secondary metal batteries

Abstract

Over the past few decades, lithium ion batteries (LIBs) have dominated the rechargeable battery landscape. LIBs involve Li+ ions shuttling between electrodes on charging-discharging. Traditionally, graphite is used as the anode, serves as a cheap and reliable electrode but is limited by its capacity of 372 mAh g-1. Lithium metal, with a theoretical capacity of 3780 mAh g-1, is the ideal choice for the anode in a lithium battery. However, utilization of Li metal as the electrode has been known to be plagued by the formation of dendrites on electrochemical cycling. These dendrites are associated with a number of problems such as drying of the electrolyte, reduction in columbic efficiency, but most dangerously, the ability of these dendrites to short the battery, leading to thermal runaway and eventual fires. In our work, we show that these dendrites can be healed in-situ in a Li-metal battery by means of using a lithium iron phosphate (LFP) cathode and a Li metal anode. The healing is achieved by extensive surface diffusion of Li-atoms from the dendritic surface caused by the rise in local temperatures due to joule-heating at high current densities. This phenomenon has been corroborated by computational thermal modelling to predict rise in temperatures and density functional theory (DFT) to understand the diffusion characteristics of Li atoms on dendrite surfaces.

Potassium (K) batteries offer a sustainable and low cost alternative to Li-ion technology. The utilization of K metal anodes allows for the development of high power potassium batteries. However, like Li, K metal anodes are observed to develop dendrites on electrochemical cycling that pose serious safety concerns. In this work, we show that potassium dendrites can be healed in a potassium metal battery by means of joule-heat assisted self-diffusion. It was observed that the potassium dendrites could be healed at a current density an order of magnitude lower than that required to heal lithium dendrites. We show that the reason for this is the far greater mobility of surface atoms of K metal compared to that of Li metal. Such in-situ healing of dendrites could eliminate the risk of short circuiting and enable safe deployment of metal batteries for the next generation of high performance energy storage devices.