We developed a fast and scalable method to improve 3D current collectors for negative electrodes in lithium-ion batteries. When covered with a silicon-based coating, these electrodes showed excellent performance, with only a small capacity loss of 0.2% per cycle—much better than standard electrodes. However, the cost to implement this technology in production is beyond the project's budget.
We also created polycarbonate formulations for gel electrolytes, which were applied to the battery's anode and cathode layers. Unfortunately, the lab-scale cells were not stable enough, so further improvements are necessary.
On the positive side, we successfully tested a more eco-friendly manufacturing process for the cathode, which uses water instead of the harmful NMP solvent. This green process worked just as well as the traditional method, but the final drying step needs special attention because the water-based binders tend to absorb moisture. However, these binders also perform better at high temperatures. We achieved the same moisture level in the final cathode by adding a small amount of acid, which didn’t affect the thickness of the electrode. While production on a pilot line has been confirmed, larger-scale manufacturing will require longer processing times.
On the anode side, we developed electrodes with 70% silicon content, using next-generation materials on a mass-production scale. These anodes retained 70% of their capacity after 1,000 charge cycles in full cells. The manufacturing process was scaled up for larger battery formats, like the 21700 type, achieving an energy density of 800 Wh/L. By the end of the project, we had produced 261 of these large cells to build a home energy storage module.
We thoroughly tested the new electrodes, electrolytes, and battery cells, both when they were new and after use, using the multi-technique ECO2LIB approach. We found that the high silicon content in the anodes led to the formation of a protective layer (SEI) on the surface, which changed over time and affected performance. Electrolytes without fluorine provided a more stable SEI, which continued to grow over time. The drop in capacity was mainly caused by the formation of this SEI layer and uneven lithium distribution, which led to defects and damage at the cell level.
We imaged the degradation heterogeneities at the cell level and obtained detailed 3D maps of trapped lithium and inactivated regions after long-term ageing. We found specific deformations and defects caused by silicon agglomeration during the slurry preparation and electrode manufacturing. We also discovered correlations between reaction kinetics, particle motions/volume changes, and fluid dynamics inside industry-grade cylindrical cells, providing key insights into complex chemomechanics during high silicon content battery charge and discharge.
We also developed multi-physics models to understand the effects of mixing silicon with graphite and the internal stresses in the battery. These models were tested with real data and simplified to predict how the battery's capacity would decrease over time. Additionally, we explored an eco-friendly process to recycle batteries, separating different materials. The process successfully separated fine and coarse fractions, but it didn't perform as well as expected due to a specific binder used in the tested cells. We recommend switching to an aqueous binder in future ECO2LIB cells. We also investigated ways to purify the electrolyte solution for reuse, but side reactions with water limited its use in battery production. Water is essential for safety, but the organic components from this process can be repurposed in other areas. In the end, we achieved a 76% battery recovery rate.
A Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) analysis compared ECO2LIB NMC622 lithium-ion cells to current NMC622 reference cells. Despite higher material costs, mainly due to the silicon anode and extra NMC622 oxide, ECO2LIB cells had comparable production costs per kWh due to a higher cell capacity and the employment of water as the cathode solvent, leading to the omission of cost-intensive NMP recovery. While the production of ECO2LIB cells has a higher environmental footprint, the improved energy density reduces the overall impact per kWh provided over the battery's lifetime. Compared to standard cells, ECO2LIB cells lower CO2 emissions by 10% and total costs by 3% over their lifetime. These calculations assume large-scale production and an equal number of cycles for both cell types, though long-term testing may be needed for confirmation. The same recycling rates were assumed for both the standard and ECO2LIB processes, which could be achieved if only aqueous binders are used in future batteries.
