Novel CuSi2P3-Based Semi-Solid Anolyte for Redox Flow Batteries

Abstract

Introduction Redox flow batteries (RFBs), as a kind of low-cost, high-efficiency energy storage system, have received increasing attention in recent years. However, the energy density of RFBs is much lower than that of most commercialized lithium-ion batteries, which not only makes RFBs inconvenient in stationary energy storage but also greatly limits their application in many fieldsIn this work, a low-potential CuSi2P3-based composite is synthesized through a simple high-energy mechanical ball milling and impregnation method; then, it is used to prepare a semisolid anolyte, which is able to achieve a high volumetric capacity (320- 400 Ah L-1). The effect of adding binder to a composite is also discussed for the first time. The synthesis method of adding a binder to the composite provides a direction for optimizing the suspension for other active materials to be applied to semi-solid flow battery.Results and DiscussionFigure 1 compares the 1st and 10th galvanostatic discharge-charge cycling tests of two semisolid anolytes, 8 vol% CSP@C-10 vol% Super P (8CSP@C-10SP) and 8 vol% CSP@C-LiPAA-10 vol% Super P (8CSP@C-LiPAA-10SP). The specific capacity of 8CSP@CLiPAA-10SP anolyte at the first cycle can reach up to >2000 mAh g-1, which approaches the theoretical capacity (2079 mAh g-1) based on the corresponding binary Li-alloy products (Li3.75Si and Li3P), with the capacity retention greater than 98% from the 2nd to the 10th cycle. However, the specific capacity of 8CSP@C-10SP anolyte without LiPAA at the first cycle is only about 1000 mAh g-1, and severe capacity fading is encountered. Only a specific capacity of 420 mAh g-1 is left at the 10th cycle, leading to capacity retention of only 40%. The presence of LiPAA makes a more stable CSP composite, which can obtain a better cycling performance. We concluded that the cycling stability of LiPAA binder-based CSP composite is much better for RFBs compared with the one without binder.Figure 1b-e reveals the comparison of SEM images between 8CSP@C-10SP and 8CSP@C-LiPAA-10SP at pristine stage and after cycles respectively. It can be seen from the images that these two semisolid anolytes present very porous structure at the pristine stage (Figure 1b, d). However, after cycles, the surface of 8CSP@C-10SP produces very large film-like precipitates (Figure 1c), which greatly hinders the reaction of active materials in the subsequent cycles and further results in serious capacity fading. On the other hand, the porous structure of the 8CSP@C-LiPAA-10SP after cycles enables sound performance of the anolyte during the continuous cycles. The addition of LiPAA in the CSP composite can improve the contact between CSP and carbon, which confirms the electrical conductivity uniformity of CSP-based suspension.The addition of LiPAA binder to the composite enables the CSP semisolid anolytes to maintain better uniformity and rheological performance. Figure 2 shows the viscosity test results of electrolyte (1 M LiPF6 EC/DEC) and various CSP-based semisolid anolytes. It is indicated in the results that the non-Newtonian fluid semisolid anolytes present greater viscosity than the Newtonian fluid electrolytes, which results in larger flow resistance and more pump loss; therefore, viscosity is an important issue to be considered in the application of SSFBs. We find the viscosity of CSP@C-LiPAA anolyte significantly smaller than that of CSP@C anolyte at the same concentration, which is favorable to reduce the flow loss of the anolyte. The addition of LiPAA binder improves the rheological performance of the semisolid anolyte in an effective way and enormously enhances its cycling stability. Further investigations on the electrochemical reversibility, energy efficiency, cycle life, full cell design, and flow cell performance will be discussed. Acknowledgments The work described was substantially supported by a grant from the the Natural Science Foundation of Guangdong Province (2023A1515010954) and the Shenzhen Science and Technology Fund (JCYJ20220531101801002). Figure 1