**1. Introduction**

Electrification of the transport sector is mandatory to achieve the Paris climate targets, as it currently accounts for around 24% of all global CO2 emissions [1]. However, battery electric vehicles (BEVs) currently cause almost twice the greenhouse gas emissions in the manufacturing phase compared to equivalent combustion vehicles, mainly due to the resource-intensive production of the battery [2]. Nevertheless, electric vehicle batteries (EVBs) show a significantly better environmental performance if, first, renewable energy is used to charge the battery during the use phase and, second, if electrification occurs within a circular economy. In this context, the battery plays the most important role, as it is the most expensive component in BEVs and contains valuable materials and components suitable for reuse A battery pack generally consists of several modules made up of battery cells. Currently, Li-ion cells are the most common. They can be found in three shapes (cylindrical, pouch, and prismatic) with different cell chemistries (NMC, LCO, LMO, LFP, and NCA). However, the general structure of a Li-ion battery cell is independent of the cell

**Citation:** Baazouzi, S.; Rist, F.P.; Weeber, M.; Birke, K.P. Optimization of Disassembly Strategies for Electric Vehicle Batteries. *Batteries* **2021**, *7*, 74. https://doi.org/10.3390/ batteries7040074

Academic Editor: Catia Arbizzani

Received: 21 July 2021 Accepted: 2 November 2021 Published: 7 November 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

format and the used chemistries. Its main components are anodes, cathodes, a separator, an electrolyte, and housing. In batteries for the automotive sector, all cell formats and a wide range of chemistries are used. Thereby, materials' purchase costs are the primary cost driver.

The circular use of components and materials offers big economic opportunities and has great potential to secure the supply of strategic raw materials for cell manufacturers [3]. The work of Sato and Nakata [4] showed that by 2035, high quantities of critical materials for the production of new Li-ion batteries in Japan will be obtained from the recycling of batteries at the end-of-life (EoL) stage (34% of lithium (Li), 50% of cobalt (Co), 28% of nickel (Ni), and 52% of manganese (M)). However, according to Kotak et al. [5], alternative circular economy strategies such as reuse and remanufacturing would extend the use phase of batteries and thus avoid the resource-intensive production of new batteries. Furthermore, they allow postponing recycling, which will result in improving the recycling efficiency due to the fact that the recycling processes are continuously being developed. In this context, disassembly plays a key role in the implementation of all alternative circular economy strategies at the EoL phase [6]. In addition, by using advanced disassembly technologies and strategies, material recyclers can significantly reduce the mix of materials to be handled in resource-intensive downstream material recycling processes.

Nevertheless, today, disassembly represents a bottleneck process that has to be performed faster [7]. Currently, EVB disassembly is done manually [8], which leads to high costs and poses safety risks to human workers. For these reasons, industrial and highly automated disassembly is mandatory in the future [6]. Automated disassembly is required to handle future quantities of returning battery systems in an economically viable and secure manner. Based on the review of several literature sources, Tan et al. [9] divided the battery disassembly process at the module-level into four steps. It starts with removing the battery casing, followed by the extraction of the battery management system (BMS), power electronics, and the thermal management system. After that, wires, cables, and connectors are removed. Finally, the modules are obtained after disassembling the securing holders. The modules can be further disassembled to obtain the battery cells. Thereby, the five main components have to be removed from the modules. These are cell contacting, cell fixation, housing, thermal management, and the BMS [10]. Gerlitz et al. [10] classified the challenges for automated disassembly at the module level into product-related and process-related challenges. Thereby, the main challenge posed by the product is the design variety. The main process-related challenges are the non-detachable joints and the hazards related to Li-ion batteries.

Figure 1 shows the different players in the life cycle of EVBs and their role in implementing a circular economy. Thereby, disassemblers specify the material flow at the EoL phase. Remanufacturers are very important for implementing high-value circularity solutions at the different system levels. Recyclers are obligatory to close the loop. Here, it is worth mentioning that there are different opinions in the ongoing research projects on who will carry out disassembly, remanufacturing, and recycling. It is assumed that these operations will either be performed by the same stakeholder, such as recyclers, or that new actors will be established due to the expected enormous return volumes and the diversified skills necessary to establish economic and high-quality circularity of EVBs. In this work, we adopt the second opinion. Disassemblers play a decisive role, whatever the recovery option at the EoL phase is. However, EVB disassemblers have to deal with several challenges in the future, such as:


**Figure 1.** Participants along the life cycle of battery systems and their role in establishing a circular economy.

At the EoL phase, remanufacturers and recyclers are also crucial to extend the life of battery components or to recycle the battery parts if recycling is the only recovery option due to advanced aging or if recycling is the most suitable recovery option. The big challenge here is to find out the optimal route for an EVB at the EoL phase, as there are multiple alternative circular economy strategies and diverse recycling paths. Furthermore, this decision depends on the market demand. Therefore, a multicriteria decision platform is extremely essential [1].

Once the optimal EoL strategy has been determined, how should the battery be disassembled to implement the selected route economically? This question leads to an optimization problem that must be solved for individual batteries to significantly increase the economic efficiency of disassembly as the most expensive processing step in the current state [11]. Ke et al. [12] performed disassembly tests on the same battery type with the same skilled workers. They observed that the workers could disassemble the battery at least 11.5% faster when they had an optimized disassembly sequence.

Disassembly cannot be seen as the reverse of assembly because, first, disassembly is subject to many uncertainties and, second, there are different ways to perform disassembly. Here, different disassembly modes can be distinguished using several criteria. such as the disassembly depth (Complete/incomplete), the disassembly techniques (Destructive/nondestructive), the number of used manipulators (Sequential/parallel), and the automation level (Manual/automated). Thereby, the disassembly of complex products cannot be performed when it is only experience-based. Disassembly planning solutions that are adaptive and use optimization algorithms are necessary to determine optimal disassembly strategies.

This paper aims to contribute to designing adaptive disassembly planners for battery systems by combining the autonomous disassembly planner presented by Choux et al. [13] with a disassembly strategy optimizer, which will be implemented and tested using an Audi A3 Sportback e-tron hybrid battery pack. The battery, instructions about its disassembly, and several essential data for the disassembly planning, such as the disassembly times and revenues at component level after applying a specific circular economy strategy, have been described in detail in [14]. In this paper, the optimal disassembly strategy maximizes the optimal economic profit. It consists of the following decisions: (1) the optimal disassembly sequence, (2) the optimal disassembly depth, and (3) the optimal circular economy strategy for each component (reuse, remanufacturing, repurposing, and recycling). The proposed disassembly planner can significantly contribute to implementing high-value circularity levels at the EoL phase of EVBs in automated disassembly solutions in the future.

The following sections are organized as follows: Section 2 describes the main components of an automated disassembly solution. Thereby, the disassembly planner with an integrated disassembly strategy optimizer represents a core building block. Section 3 describes our methodology by presenting our design for an adaptive disassembly planner and a disassembly strategy optimizer. Finally, we present and discuss our use case results in Section 4.
