Barry on Batteries: Digitalization Tackles Battery Recycling's Operational Challenges

As you may know, I started writing this column in 2022. Since then, the lithium-ion and electrification supply chain has grown worldwide. Over these four years, we have experienced rapid changes in the market, including battery demand and supply chain shifts, technological advancement with the scaling and implementation of AI and the global government commitments to investment.  

In fact, I was speaking at the International Battery Seminar and Conference, and as I was walking around the exhibit hall — much to my surprise — a vendor called out: “Barry on Batteries, I read your column all of the time.” Wow! This got me thinking about where the state of the battery market was back then and where we are now. Let me recap and provide updates about the new best practices for battery recycling as well as forecast what I see on the horizon.  

I’ve covered the complete supply chain, including lithium production from geothermal brines, battery materials production and unit operations of slurry mixing and reactions, solid-liquid separation and process dryers, electrolytes and their effectiveness, innovations in battery types, chemistry and designs and lithium-ion battery recycling. 

This now brings us to the operational and technical challenges these process silos present when designing a battery recycling facility. Digitalization can help operators overcome some of these barriers. Each silo falls into three main categories: Collection, transport and storage; diagnosis, discharge and disassembly; and material recovery. 

Battery Collection, Transport and Storage

The supply chain begins with the feed material and collection, transport and storage. Different battery types present various challenges to ensure maximum recovery. Cloud-based advanced modeling and simulations can help achieve batch-to-batch integrity. For the battery materials to be produced and reused, the required quality is as strict as pharmaceutical because any feed material variability can impact how a cell works in an EV. Modeling and simulations are used to optimize the collection by types, such that the feed material to each process is reproducible and the quality output can be controlled.  

All the necessary data and documentation need to be recorded including integration of monitoring sensors for critical parameters of temperature, pressure and off-gases in real-time. This addresses the reproducibility of the process to ensure that each process step is meeting the quality targets.  Finally, strategically placed cameras can detect early signs of fire hazards, gas leaks or battery structural damage. These cameras are based upon charge-coupled device technology, which uses specialized sensors to capture light and convert it into high-fidelity digital images. 

The cameras can look at the actual battery and compare it to the original battery documentation to ensure the structural quality. We now have the feed material stored in a safe location and the next group of steps is diagnosis, discharge and disassembly. 

Diagnosis, Discharge, Disassembly

To begin, you must examine the battery to determine if it’s suitable for second use or recycling. The diagnostic process involves assessing key parameters, such as remaining capacity, internal resistance, voltage stability and physical integrity. The process also involves diagnosing the physical structural state as well as advanced cell testing. You can perform this analysis using vision cameras and machine learning to determine the extent of any battery damage.  

Prior to recycling, the battery needs to be safely discharged using brine soaking or electrical discharge back to the grid or energy storage systems. Electrical discharge approaches are more energy efficient and environmentally sound.  

Machine-learning (ML) algorithms and process simulation can significantly reduce the time and complexity of the selected process and can predict the optimal discharge profile including the temperature rise to prevent thermal runaway and possible battery rebound charge or over-discharge danger. Machine learning also can develop standardize discharge protocols based on the battery’s age, chemistry, usage history, residual charge and conditions. A discharge curve is reproducible, similar to the process drying curve, such that from the same initial parameters the process will be the same. The recycler can input the battery parameters, and machine learning will produce the discharge curve to ensure the correct cycle time.  

Currently, manual operations for disassembly involve over 50 individual steps depending upon the type of battery and how the battery cell, module or pack is designed. This disassembly continues to be a barrier for scalability. Robots can perform the initial work to increase efficiencies. For the more detailed disassembly, battery processors can incorporate vision systems, sensors and AI. These technologies provide real-time insights for identifying variations or damages in the batteries, as well as analyzing the information for adjusting the disassembly process.  

Digital Twins for Material Recovery and Commissioning

The following pretreatment and the hydrometallurgical steps can benefit from the use of digital twins for process control and derisking the scale-up. All these steps have interdependent physical and chemical parameters that influence each other, such as equipment designs, pH, temperature, reagent flow rates and metal concentrations. 

The traditional design of experiments approach to process optimization requires a great deal of time and effort, while digital twins enable rapid hypothesis interactional testing. Digital twins also incorporate computational fluid dynamics (CFD) models for physical behavior analysis as well as machine learning. The combination of digital twins and CFD models with system integration testing provide valuable improvements in safety and reliability.

Lastly, once the plant is mechanically complete, you can use digital twins to conduct virtual commissioning to minimize delays before the actual operations begin.  

As the demand and pressures for efficiency increase on battery recycling it’s essential that recycling plants are designed like chemical plants. Chemical engineers and recyclers must create something that is safe, efficient, scalable and sustainable to meet the demand of industry across the globe. The chemical engineering techniques used for specialty chemicals, pharmaceutical and similar plants for process design, optimization, safety and reliability are transferrable to the recycling industry.