Mitigating Heterogeneities in Lithium-Ion Battery Modules Under Fast Charging

Home > Resources > Mitigating Heterogeneities in Lithium-Ion Battery Modules Under Fast Charging

Mitigating Heterogeneities in Lithium-Ion Battery Modules Under Fast Charging

Mitigating Heterogeneities in Lithium-Ion Battery Modules Under Fast Charging Feb. 05, 2024

Mitigating Heterogeneities in Lithium-Ion Battery Modules Under Fast Charging

 

Temperature control is essential for fast-charging performance of Li-ion battery. Cold temperature leads to sluggish ion transportation of electrolyte, brittle polymeric cell components, changes in solid electrolyte interface (SEI) properties and associated resistance build-up, and Li plating and dendrite growth [1]. High temperature boosts fast charging but can also accelerate battery degradation and increase the rise of battery thermal failure. This is caused by the increased rate of side-reactions, resulting in loss of cyclable lithium and higher rate of attrition of active materials at higher temperatures.

Temperature and its gradient in a battery unit strongly affect performance and life of battery units [2]. It is recommended that pack temperature uniformity of a li-ion battery pack in electric vehicles shall be less than 3 ºC [3]. Battery thermal management systems (BTMS) are important for controlling battery pack temperature and minimizing temperature gradients to prevent thermal-related issues in Battery Energy Storage Systems (BESS). This thermal management goal is more critical for fast charging of battery modules made of large format, high-energy-density cells. Current BTMS in battery electric vehicles (BEVs) are inadequate in limiting the maximum temperature rise of the battery during extreme fast charge (i.e., 6C charge). To achieve fast-charge, the size of the battery thermal management system needs to increase from today's BEV average size of 1–5 kW to around 15–25 kW [4].

A combined experimental and modelling approach is employed to access thermal and electrochemical heterogeneities of a battery module under extreme fast charge conditions and develop corresponding mitigation approaches. The electrochemical-thermal model was built based on electrical characterization of 32 mAh pouch cells, including constant-current 1C to 9C rates of charging at varying temperatures. Predictive performance of the model in heat generation was validated by comparing results against measurements conducted using a microcalorimeter. Thereafter, the validated model is used to predict performance of a battery module consisting of six large format pouch cells. The large pouch cell has a capacity of 25 Ah and the identical electrode design of the 32 mAh cells. 3D simulation results suggest significant temperature and charge differences can be produced. The heterogeneous behavior was enlarged along charging. As shown in Figure 1, it was found that electrodes close to the tabs were preferentially charged.

Cell electrochemical heterogeneity can be reduced by reducing cell temperature difference. Two potential solutions are investigated using the developed 3D model, including the enhancement of heat transfer within cells, such as increasing cell thermal conductivities with thicker current collectors, and the optimal design of thermal management systems. The feasibility of state-of-the-art thermal management strategies for fast charging is evaluated, including liquid cooling using cold plate devices and direct liquid cooling.

References

[1] Pesaran, S. Santhanagopalan, G.H. Kim, Addressing the Impact of Temperature Extremes on Large Format Li-Ion Batteries for Vehicle Applications, 30th International Battery Seminar, Ft. Lauderdale, Florida, 2013.

[2] Garimella et. Al., A Critical Review of Thermal Issues in Lithium-Ion Batteries, Journal of The Electrochemical Society, ISSN 1945-7111, 158(3), R1-R25(2011)

[3] USABC, Li-Ion Battery Thermal Management System Requirements, 2018

[4] Keyser et. Al., Enabling Fast Charging – Battery Thermal Considerations, Journal of Power Sources, ISSN 0378-7753, 367(2017) 228-236

 

Figure 1

 

Homogenized modeling methodology for 18650 lithium-ion battery module under large deformation

Abstract

Effective lithium-ion battery module modeling has become a bottleneck for full-size electric vehicle crash safety numerical simulation. Modeling every single cell in detail would be costly. However, computational accuracy could be lost if the module is modeled by using a simple bulk material or rigid body. To solve this critical engineering problem, a general method to establish a computational homogenized model for the cylindrical battery module is proposed. A single battery cell model is developed and validated through radial compression and bending experiments. To analyze the homogenized mechanical properties of the module, a representative unit cell (RUC) is extracted with the periodic boundary condition applied on it. An elastic–plastic constitutive model is established to describe the computational homogenized model for the module. Two typical packing modes, i.e., cubic dense packing and hexagonal packing for the homogenized equivalent battery module (EBM) model, are targeted for validation compression tests, as well as the models with detailed single cell description. Further, the homogenized EBM model is confirmed to agree reasonably well with the detailed battery module (DBM) model for different packing modes with a length scale of up to 15 × 15 cells and 12% deformation where the short circuit takes place. The suggested homogenized model for battery module makes way for battery module and pack safety evaluation for full-size electric vehicle crashworthiness analysis.

Citation: Tang L, Zhang J, Cheng P (2017) Homogenized modeling methodology for 18650 lithium-ion battery module under large deformation. PLoS ONE 12(7): e0181882. https://doi.org/10.1371/journal.pone.0181882

Editor: Xiaosong Hu, Chongqing University, CHINA

Received: May 8, 2017; Accepted: July 8, 2017; Published: July 26, 2017

Copyright: © 2017 Tang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper.

Funding: This work is financially supported by the National Science Foundation of China under Grants No. 51605032, the National Science Foundation of Beijing under Grant No. 3174052 and the Fundamental Research Funds for the Central Universities No. 2017ZY31. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

With the strong support from the government [1] and major technology breakthrough for lithium-ion batteries (LIBs) [2, 3], electric vehicles (EVs) have been witnessed to boom over the past recent years [4–6]. The major reason for LIBs to become the primary choice for EVs is due to the combination advantage of high energy/power density, lightweight, and safety [6, 7]. Many research works were conducted on the optimization approach for electrified vehicles to increase cost competitiveness and reduce carbon emissions [6, 8–10]. Additionally, the battery management system plays an important role in maintaining battery lifetime without unduly sacrificing its performance. Some key technologies of battery management system that monitor the unmeasurable internal states of the battery have been extensively studied [11–14]. Battery models such as electrochemical [13], thermal [15], and high-order physics-based model [16] are also used for evaluation of existing charging strategies, estimator and controller development, simulation, and optimization. However, none of these works have considered the mechanical performance of the battery. Because of the inevitable crash or impact for vehicles during traffic accidents, there is a high possibility of internal short-circuit [15], thermal runaway [17], and even explosion [18, 19] for LIBs subjected to external mechanical loading. Thus, the crashworthiness issue of EVs by consideration of LIB safety remains a paramount concern in electric vehicle safety.

In the past years, many pioneering efforts were made to elucidate the mechanical safety behavior of LIBs on multiple length scales, ranging from component material scale to battery pack scale. On the component level, the mechanical properties of the case [20], separator [21], anode, and cathode foils [22] of the cell have been investigated. These results will help to understand the failure mechanism of internal short circuit of lithium-ion battery, but can not characterize the global mechanical behavior of the battery. Many experiments were conducted for the investigation of mechanical behaviors of the battery cells under various loading conditions [20, 23, 24]. The finite element (FE) models were used to understand the mechanical properties and predict extreme cases. Since coating of active material coating and separator are highly porous and soaked in the electrolyte, the detailed modeling including each component and interaction among them is very complicated. Therefore, it is reasonable and acceptable to take jellyroll as a homogeneous material [23, 25, 26]. Furthermore, the dynamic behavior [27] and SOC effect [28] of the battery have been studied and the results suggested that higher SOC leads to higher structure stiffness. In this paper, a computational model of a single cylindrical battery is established and validated based on homogeneous modeling technique.

For battery module and pack, the mechanical safety performance is closely related to sizes and packing modes of the module and modeling every single cell in detail would be costly. Therefore, equivalent homogenized model for the module is fully necessary for many applications such as vehicle level crashworthiness analyses and optimum design. Sahraei et al. [29] used a homogenized crushable foam core to simulate the interior containing battery cells of an idealized battery pack to model the drop test. The battery pack was taken as a linear elastic material in [30] for crash analysis of a conceptual electric vehicle. However, modeling the module by using a simple bulk material would result in sacrificing computational accuracy. Additionally, Lai et al. [31] adopted macro homogenized material models calibrated by the test data to simulate the punch test of a small-scale module specimen. Nevertheless, it is very difficult to directly measure the integrated mechanical properties of the battery module.

This paper responds to the challenge by extracting a representative unit cell (RUC) with the periodic boundary condition applied on it to analyze the homogenized mechanical properties of the module. An elastic–plastic constitutive model is established to describe the equivalent battery module (EBM) model. Further, a small-scale battery module is tested to compare the mechanical behavior with those obtained from the EBM model and detailed battery module (DBM) model. Upon the satisfactory comparison results, EBM model is further generalized for larger battery modules and different packing modes and the feasibility of the established model is discussed.

Results

Experiments, homogenized EBM model and corresponding DBM model were conducted on the case for the cubic dense packing mode, i.e. θ = π. The settings of boundary conditions for both computational models were exactly the same as in the experiment.

For the EBM computational model, four-node bilinear quadrilateral plane strain, which reduced the integration of 2D solid elements, was used; this was also verified via convergence study. Moreover, the proposed homogenized mechanical properties in Section 2.4 were adopted. For the DBM model, nine single battery cell models were included that developed in Section 2.1. The supporting platform and the indenter were set as discrete rigid. The indenter was designated a certain displacement with quasi-static compressive loading. A penalty-based contact was set for the contacting parts by reasonably assuming a friction coefficient of 0.1.

The stress evolution of the module during deformation is shown in Fig 7. The stress distribution of EBM model is mean, and the corners near the load are under high stress state. For the DBM model, the stress wave spreads from contact area along the load direction to the rest part of the cell, and the stress values of contacting areas of the neighboring cells are higher. In addition, the maximum stress value of the EBM model is smaller than that of DBM model under the same deformation because of larger contact area. It should be noted that, as the deformation increases, the stress growth of the EBM model slows down. As presented in the subsequent section, in the overall mechanical response, the deviation of the load between the EBM model and the DBM model will increase dramatically when the deformation exceeds a certain range. Therefore, the homogenized EBM model can only precisely predict the mechanical behavior of a battery module within a certain deformation range.

 

PPT

PowerPoint slide

PNG

larger image

TIFF

original image

Download:

 

Fig 7.

The stress evolution of the module during deformation.

 

 

Progression of the buildup under constrained compression condition for EBM model (top) and DBM model (bottom).

https://doi.org/10.1371/journal.pone.0181882.g007

Comparison of the mechanical responses of the battery module for cubic dense packing is shown in Fig 8A. The load-displacement curves are close to one another (where r2 = 0.92–0.96), thereby indicating that the EBM model can well predict the integral mechanical behavior of the battery module under constrained compression loading conditions on a small size. It is noted that the EBM model could not distinctively simulate the local mechanical behavior such as contacts between neighboring cells.

 

PPT

PowerPoint slide

PNG

larger image

TIFF

original image

Download:

 

Fig 8.

Mechanical responses of the battery module in small size.

 

 

Comparison of results of the module under constrained compression condition for (a) cubic dense packing and (b) hexagonal dense packing. The plots of deformed modules in the experiment and simulations are shown in the subplot of the figure.

https://doi.org/10.1371/journal.pone.0181882.g008

Additionally, as illustrated in Fig 8, the DBM model can well reflect the mechanical behavior and deformation of the module for cubic dense packing (where r2 = 0.96) and hexagonal dense packing (where r2 = 0.94), so it could be used as a method to validate the EBM model if it is difficult to conduct an experiment.

Concluding remarks

Correctly predicting the mechanical behavior of the module subjected to mechanical loading is critical to understanding the mechanical integrity of the battery pack and ensuring full vehicle crash safety. In this paper, a general method to establish the computational homogenized EBM model is proposed. For the convenience of study, we uniformly defined the packing size and mode of the module. Moreover, the RUC was extracted from the module and the periodic boundary condition was applied on it to acquire the homogenized mechanical properties. The experiments and corresponding DBM model were performed to validate the proposed homogenized EBM model under constrained compression loading conditions. For the cubic dense packing mode, the EBM model can well predict the mechanical behavior with length scale up to 15 × 15 and 15% deformation. For the hexagonal dense packing mode, the revised EBM model based on the packing density can represent well the mechanical properties of the module in the range of 12% deformation under different sizes. For asymmetrical packing modes where the structure of the module is unstable, after removing the load plateau stage that was caused by the relative motion of individual cells, the mechanical behavior of the module can be predicted well by the EBM model for hexagonal dense packing with the same size. Therefore, the homogenized modeling method is widely applicable for different packing modes.

Acknowledgments

This work is financially supported by the National Science Foundation of China under Grants No. 51605032, the National Science Foundation of Beijing under Grant No. 3174052 and the Fundamental Research Funds for the Central Universities No. 2017ZY31.

References

1.

Wang D, Zamel N, Jiao K, Zhou Y, Yu S, Du Q, et al. Life cycle analysis of internal combustion engine, electric and fuel cell vehicles for China. Energy. 2013;59:402–12.

2.

Chan CK, Peng H, Liu G, McIlwrath K, Zhang XF, Huggins RA, et al. High-performance lithium battery anodes using silicon nanowires. Nat Nano. 2008;3(1):31–5.

3.

Kang B, Ceder G. Battery materials for ultrafast charging and discharging. Nature. 2009;458(7235):190–3. pmid:19279634

4.

Vikström H, Davidsson S, Höök M. Lithium availability and future production outlooks. Applied Energy. 2013;110:252–66.

5.

Zhou Y, Wang M, Hao H. Plug-in electric vehicle market penetration and incentives: a global review. Mitigation and Adaptation Strategies for Global Change. 2015;20(5):777–95.

6.

Hu X, Zou Y, Yang Y. Greener plug-in hybrid electric vehicles incorporating renewable energy and rapid system optimization. Energy. 2016;111:971–80.

7.

Song MK, Zhang Y, Cairns EJ. A long-life, high-rate lithium/sulfur cell: a multifaceted approach to enhancing cell performance. Nano Letters. 2013;13(12):5891–9. pmid:24219588

8.

Hu X, Jiang J, Bo E, Cao D. Advanced Power-Source Integration in Hybrid Electric Vehicles: Multicriteria Optimization Approach. IEEE Transactions on Industrial Electronics. 2015;62(12):7847–58.

9.

Hu X, Moura SJ, Murgovski N, Bo E, Cao D. Integrated Optimization of Battery Sizing, Charging, and Power Management in Plug-In Hybrid Electric Vehicles. IEEE Transactions on Control Systems Technology. 2016;24(3):1036–43.

10.

Hu X, Martinez CM, Yang Y. Charging, power management, and battery degradation mitigation in plug-in hybrid electric vehicles: A unified cost-optimal approach. Mechanical Systems & Signal Processing. 2017;87:4–16.

11.

Sun F, Hu X, Yuan Z, Li S. Adaptive unscented Kalman filtering for state of charge estimation of a lithium-ion battery for electric vehicles. Energy. 2011;36(5):3531–40.

12.

Hu X, Jiang J, Cao D, Bo E. Battery Health Prognosis for Electric Vehicles Using Sample Entropy and Sparse Bayesian Predictive Modeling. IEEE Transactions on Industrial Electronics. 2016;63(4):2645–56.

13.

Zheng L, Zhang L, Zhu J, Wang G, Jiang J. Co-estimation of state-of-charge, capacity and resistance for lithium-ion batteries based on a high-fidelity electrochemical model. Applied Energy. 2016;180:424–34.

14.

Zou C, Manzie C, Nesic D, Kallapur AG. Multi-time-scale observer design for state-of-charge and state-of-health of a lithium-ion battery. Journal of Power Sources. 2016;335:121–30.

15.

Xu J, Wu Y, Yin S. Investigation of effects of design parameters on the internal short-circuit in cylindrical lithium-ion batteries. RSC Adv. 2017;7(24):14360–71.

16.

Zou C, Manzie C, Nešić D. A Framework for Simplification of PDE-Based Lithium-Ion Battery Models. IEEE Transactions on Control Systems Technology. 2016;24(5):1594–609.

17.

Lars H, Markus H. Lithium Ion Batteries for Hybrid and Electric Vehicles–Risks, Requirements and Solutions Out of the Crash Safety Point of View. 2011:11–0269.

18.

Dai H, Wei X, Sun Z, Wang D. A novel dual-inductor based charge equalizer for traction battery cells of electric vehicles. International Journal of Electrical Power & Energy Systems. 2015;67:627–38.

19.

Wang Z, Liu P, Cui J, Xi Y, Zhang L, Wang Z, et al. Research on Quantitative Models of Electric Vehicle Charging Stations Based on Principle of Energy Equivalence. Mathematical Problems in Engineering. 2013;69(3):831–42.

20.

Sahraei E, Campbell J, Wierzbicki T. Modeling and short circuit detection of 18650 Li-ion cells under mechanical abuse conditions. Journal of Power Sources. 2012;220:360–72.

21.

Xu J, Wang L, Guan J, Yin S. Coupled effect of strain rate and solvent on dynamic mechanical behaviors of separators in lithium ion batteries. Materials & Design. 2016;95:319–28.

22.

Lai W-J, Ali MY, Pan J. Mechanical behavior of representative volume elements of lithium-ion battery modules under various loading conditions. Journal of Power Sources. 2014;248:789–808.

23.

Greve L, Fehrenbach C. Mechanical testing and macro-mechanical finite element simulation of the deformation, fracture, and short circuit initiation of cylindrical Lithium ion battery cells. Journal of Power Sources. 2012;214:377–85.

24.

Sahraei E, Hill R, Wierzbicki T. Calibration and finite element simulation of pouch lithium-ion batteries for mechanical integrity. Journal of Power Sources. 2012;201:307–21.

25.

Wierzbicki T, Sahraei E. Homogenized mechanical properties for the jellyroll of cylindrical Lithium-ion cells. Journal of Power Sources. 2013;241:467–76.

26.

Xu J, Liu B, Wang X, Hu D. Computational model of 18650 lithium-ion battery with coupled strain rate and SOC dependencies. Applied Energy. 2016;172:180–9.

27.

Xu J, Liu B, Wang L, Shang S. Dynamic mechanical integrity of cylindrical lithium-ion battery cell upon crushing. Engineering Failure Analysis. 2015;53:97–110.

28.

Xu J, Liu B, Hu D. State of Charge Dependent Mechanical Integrity Behavior of 18650 Lithium-ion Batteries. Sci Rep. 2016;6:21829. pmid:26911922; PubMed Central PMCID: PMC4766446.

29.

Sahraei E, Wierzbicki T, Hill R, Luo M. Crash Safety of Lithium-Ion Batteries Towards Development of a Computational Model. Sae Technical Papers. 2010.

30.

Kukreja J, Nguyen T, Siegmund T, Chen W, Tsutsui W, Balakrishnan K, et al. Crash analysis of a conceptual electric vehicle with a damage tolerant battery pack. Extreme Mechanics Letters. 2016;9:371–8.

31.

Ali MY, Lai W-J, Pan J. Computational models for simulation of a lithium-ion battery module specimen under punch indentation. Journal of Power Sources. 2015;273:448–59.

32.

Li S. On the unit cell for micromechanical analysis of fibre-reinforced composites. Proceedings of the Royal Society A. 1999;455(1983):815–38.

33.

Li S. General unit cells for micromechanical analyses of unidirectional composites. Composites Part A Applied Science & Manufacturing. 2001;32(6):815–26.

34.

Li S. Boundary conditions for unit cells from periodic microstructures and their implications. Composites Science and Technology. 2008;68(9):1962–74.

35.

Xia Z, Zhang Y, Ellyin F. A unified periodical boundary conditions for representative volume elements of composites and applications. International Journal of Solids and Structures. 2003;40(8):1907–21.

 

The importance of design in lithium ion battery recycling – a critical review


Open Access Article
This Open Access Article is licensed under a

Open Access ArticleThis Open Access Article is licensed under a Creative Commons Attribution-Non Commercial 3.0 Unported Licence

 

The importance of design in lithium ion battery recycling – a critical review †

Dana L. Thompson ab, Jennifer M. Hartley ab, Simon M. Lambert bc, Muez Shiref bc, Gavin D. J. Harper db, Emma Kendrick db, Paul Anderson be, Karl S. Ryder ab, Linda Gaines f and Andrew P. Abbott *ab
aSchool of Chemistry, University of Leicester, Leicester, LE1 7RH, UK. E-mail: apa1@le.ac.uk
bThe Faraday Institution, Quad One, Harwell Science and Innovation Campus, Didcot, UK
cSchool of Engineering, Newcastle University, Merz Court, Newcastle upon Tyne, NE1 7RU, UK
dSchool of Metallurgy and Materials, University of Birmingham, Birmingham, B15 2TT, UK
eSchool of Chemistry, University of Birmingham, Birmingham, B15 2TT, UK
fReCell Center, Argonne National Laboratory, Lemont, IL, USA

 

Received 10th August 2020

, Accepted 20th October 2020

 

First published on 20th October 2020

Abstract

Recycling is always seen as an end-of-pipe process returning as much material as possible into a circular economy. There is a growing school of thought that suggests product design should be an important step in the recycling process. While this review is aimed specifically at one technological product, it contains facets that are applicable to the recycling of any complex product. Decarbonisation of energy production necessitates a proliferation of efficient electrical storage and a significant proportion of this, particularly in automotive propulsion, will use lithium ion batteries. The scale of the projected electric vehicle market means that a circular economy model needs to be established while the scale of end-of-life product is still manageable to prevent a build-up of hazardous waste. This critical review investigates the issues of lithium ion battery recycling and discusses the aspects of pack, module and cell design that can simplify battery dismantling and recycling. It highlights not only Green aspects of elemental recovery, but also technoeconomic features which may govern the appropriate direction for recycling. It also shows that as cell design changes, the approach to recycling can become more efficient.

 

Introduction

To create circular economy for a material it is important to have few components, a lower cost for the secondary process than the primary process, a simple purification flowsheet, valuable components and a collection and segregation mechanism. It also helps when the material has a significant environmental impact if not recycled as this tends to mandate its recycling.

These criteria are met by a variety of materials including glass, paper/card, steel, aluminium, plastic bottles, car catalysts and lead acid batteries. These are all well established and mature markets which grow at a manageable rate. The Green Chemistry credentials of these processes for decreasing waste in the environment must not be overlooked; recycling aluminium and steel saves 90% and 60% of the energy of the primary processes, respectively. Substitution and re-use are also a central part of the circular economy hierarchy. However, when new disruptive products and innovations come to the market with large growth potential, a product can rapidly become an environmental issue if a circular economy has not been designed with the product. This was the case in the 90s/00s with the growth in PET bottles, particularly for water,1 and has also been the case with the rapid growth in waste electrical and electronic equipment (WEEE) or the rapid demise of cathode ray tubes.

Disruptive growth in technologies can occur when a new product becomes easy to mass produce, bringing it within the reach of a larger potential market e.g. the internal combustion engine, or when a step change in the performance trajectory of a disruptive innovation reaches a critical inflection point, leading to an incumbent technology being superceded.2 Additionally, environmental regulation can be a driver for requiring technological change in some markets, e.g. the removal of the internal combustion engine, in areas with clean-air mandates.3 The decarbonisation of power production and transport has clear environmental benefits; however, a holistic approach needs to be developed for the materials which enable renewable energy to be harvested, e.g. with photovoltaic devices, electric motors, generators and storage batteries. Step changes in technology cause issues for supply and demand in critical elements, so the change from internal combustion engine to electric vehicles will cause an increase in the demand for metals such as Co, Ni, Li and Nd, and decrease the need for Pt and Pd. Such a step change in technology may enable a circular economy if there is a shortage in the raw material and the economics and logistics enable recycling.4

This critical review highlights just one of these technologies, lithium ion batteries, and tracks the issues that need to be addressed, as well as the lessons that can be learnt from some of the successful recycling industries. It highlights the importance of product design in circular economy and the aspects that can be included to simplify separation.

To evaluate the future of lithium ion battery recycling it is helpful to compare it with the successful lead acid battery market. This ubiquitous product is in most forms of automotive transport as a starter device providing the initiation of combustion. The early history of automotive propulsion was dominated by electric vehicles powered mostly using lead acid accumulators which were invented in 1859. The issues were their poor power density (250 W kg−1) and energy density (40 W h kg−1),5,6 resulting in short ranges. The advent of the internal combustion engine overcame these difficulties and consigned lead acid cells to niche, slow-moving, short-range vehicles. The proliferation of the internal combustion engine is the major contributor to greenhouse gas emissions (estimated at 18%).7 More recent developments of battery technology include the lithium ion battery, which outweighs the lead acid battery in both power density (800 W kg−1) and energy density (180 W h kg−1), making it the cell of choice for modern technology such as electric vehicles (Fig. 1).6,8


Fig. 1

Basic schematic of a lead acid and lithium ion battery. 6

While it has a relatively poor energy density, the lead acid cell has, through numerous iterations, become standardised and is designed for recycling. Modern lead acid batteries are able to reuse >98% by mass of the material.6 This is due in part to the simplicity of their design, where the anode and cathode are Pb and PbO2, respectively. The lead acid battery is self-contained in one unit, not assembled into modules and packs, so it needs no disassembly prior to recycling. Each automotive battery weights 12–21 kg of which lead is more than half the weight which make it economically worthwhile to recycle. The recycling process of a lead acid battery is a simple one: the case is crushed, allowing the sulphuric acid electrolyte to escape, and the lead electrodes are separated from the polypropylene casing and separator by density. The lead is smelted and the polypropylene can be reused in new casings.6

Both Gaines and May et al. state that recycling rates have reached almost 100% in the USA, Japan and most of Europe. This success is because of the incentive to recycle lead acid batteries – it is economical to do so due to the relatively high cost of lead, and the process is an efficient one due to the uniformity of the materials used and battery design.6,9 The materials recycled from the lead acid batteries are then used to manufacture new batteries, thus allowing for closed loop recycling. The success of the lead acid battery circular economy can be easily judged. In the UK in 2018, lead acid batteries accounted for 55% of all batteries collected despite being only 4% by mass of batteries produced. Data on the remaining 96% of batteries are hotly contested. There are a range of different recycling rates cited in the literature,10 however, the recycle rate is significantly lower than that of lead acid batteries due to the complexity of different designs, the low value of materials such as zinc and manganese oxide and the lack of legislation controlling their disposal. Research by Circular Energy Solutions10 critiques much of the existing literature and makes a case that recycling rates are not as pessimistic as the literature would indicate, however there remain challenges. It is important to draw a distinction between lithium ion batteries used in consumer electronics, which are small and easily misplaced, and vehicle batteries, which are large and enter dedicated waste facilities. Recycling processes for lithium ion batteries exist, but the problem lies with their collection rate. Whilst there is a market for viable lithium ion batteries that can be used in second-life applications, it is harder to incentivise the collection of lithium ion batteries for recycling with little economic value, or potentially an associated gate fee. Contrasted with lead acid batteries, this shows the success of a product which has a simple design, a relatively high cost, a low-cost recycling process, a structured collection program and a significant environmental impact if not recycled.

Achieving the same for lithium ion batteries is difficult, due to the more complex cell design and cell chemistry. The lack of any standardisation of cells and the predominance of cells from small portable devices means that initial recycling approaches will be more similar to solid municipal waste, producing streams of lower purity. Homogenisation of cell design and chemistry and the larger fraction of similar automotive cells will enable easier recycling with streams of higher purity and higher value. From a Green Chemistry perspective, it is the scale of the market growth that necessitates the manufacturing and recycling process to be as efficient as possible. As of 2017, there were 3 million electric vehicles in the global stock, which is expected to grow to 125 million by 2030, and 530 million by 2040.11,12 The task facing recycling of lithium ion batteries can be easily understood by comparing it to current recycling markets. Global markets are complex to analyse, so the UK is used as an example, assuming that scrapping rates remain roughly constant at ca. 1 million p.a. reaching end of life. Assuming that by 2040 electric vehicles achieve a market penetration of 50%, approximately 200 kt of lithium ion batteries will reach end of life from an automotive perspective. This may not, however, mean that this is the amount of material that will need recycling as there is significant potential for many of these batteries to reach second life for energy storage, particularly for renewable energy such as wind and solar.13 To put this into perspective, 200 kt is approximately 20 times the size of the current lead acid battery market in the UK. It is therefore clear that new infrastructure will need to be developed to cope with this volume of material and standardise its transport, handling and processing. Depending on process economics, this may then require legislation defining extended producer responsibility for batteries.14 While life cycle analysis has been carried out for the production of lithium ion batteries15 comparatively little analysis of recycling costs and throughputs has been performed. One of the aims of this critical review is to show how product design is an important factor that is often overlooked in Green Chemistry. The recycling metrics can be significantly altered by considering disassembly during the design process. The disassembly of lithium ion battery modules, albeit manually at present, has been shown to produce a high yield (ca. 80%) of total mass recovered in a purer state that was possible using shredded material. The active materials could be short-loop recycled into new electrodes with only minimal performance loss for the anode.16

The growth in lithium ion battery recycling can be judged from Fig. 2 which shows the number of articles and patents on the topic in the past few years. It also shows the actual and projected growth in automotive electric vehicles. This shows clearly how concerted international attention is focussed on the creation of a circular economy for battery materials.


Fig. 2

 

(a) Number of articles and patents on the topic of lithium ion battery recycling (b) Recent and projected growth in automotive lithium ion battery markets.

 

 

Lithium ion batteries

Unlike the lead acid battery, the structure of lithium ion batteries is much more complex, with a series of small cells being collected together to make a module and a number of modules are assembled to make the overall battery pack. An automotive battery pack is composed of hundreds or thousands of cells, which not only have to be individually opened but also disassembled from the ensemble. The complex structure and risks associated with electric shock and potential fires make safe dismantling slow and labour intensive. For this reason, many current approaches start with comminution (crushing) in the same approach to lead acid batteries, but this is poor from a Green metric perspective as it requires more steps, more energy and more ancillary processing chemicals.

The components of a lithium ion battery pack are shown in Table 1. Each component in a lithium ion battery may consist of different chemistries to those made by, for example, another company within the battery industry. There is a great variation in lithium ion battery cathode chemistries (such as variations of NCA, NMC, LMO, LCO and LFP). In current use, graphite dominates the supply of lithium ion battery anode materials, with low levels of SiOx and silicon being introduced in the high energy cells;17 however, lithium titanate (Li4Ti5O12)18 and more recently TiNb2O7 can be used in lower energy density but high power cells.19,20 One issue for recycling companies is that due to the rapid advances in technology, two versions of the same car model may even have different battery chemistries. Perhaps a desire for quick treatment and simple recycling, given present modest volumes, has led to processing techniques that are chemistry agnostic, resulting in lower purity products. While the hazard labelling of lithium ion batteries is strongly regulated21 the lack of compositional labelling for easy identification, coupled with the simplicity of shredding, is a barrier to more nuanced recycling schemes. Most battery packs contain no information about the chemistry of the anode, cathode or electrolyte, meaning that cells from the different packs need to be dealt with by the same process; this is why pyrometallurgy and comminution are the only acceptable methods of recycling at present. Improved battery labelling would enable different battery chemistries to be separated before processing and would prevent contamination between e.g. NMC and LFP chemistries. The issues of labelling are beginning to be addressed, with the Society of Automobile Engineers (SAE International)22 recently recommending a labelling scheme, and the Chinese Government is also considering mandating labelling of lithium ion batteries.

Table 1

Cell materials in a typical lithium ion battery 9,24–30

Cell component Materials Composition/wt% Cost/% Cathode active material Layered structures,

e.g.

LiCoO2 (LCO)/Li(Ni

x

Mn

y

Co1−

x

y

)O2 (NMC)/Li(Ni1−

x

y

Co

x

Al

y

)O2 (NCA) 22–25 65–70 Spinel structures,

e.g.

LiMn2O4 (LMO) Olivine structures,

e.g.

LiFePO4 (LFP) Cathode foil Al 4–5 1 Anode active material Carbonaceous materials (graphite, hard carbon), lithium titanate, or silicon-based materials 24–26 8–9 Anode foil Cu 3 2 Binder Polyvinylidine fluoride (PVDF)/polytetrafluoroethylene (PTFE)/polyvinyl alcohol (PVA)/carboxymethyl cellulose (CMC)/styrene butadiene rubber (SBR) 2–3 8–9 Electrolyte Mixtures of ethylene carbonate (EC)/propylene carbonate (PC)/dimethyl carbonate (DMC)/ethyl methyl carbonate (EMC)/diethyl carbonate (DEC) + additives

e.g.

fluoroethylene carbonate (FEC)/vinylene carbonate (VC) 10–12 1 Conductive additive Acetylene black (AB) 1 0.1 Conductive salt LiPF6 1.5–2 8 Separator Polyethylene (PE)/polypropylene (PP) 4–5 4 Cell case Varies (metal or laminate) 4–6 4

Table 1 shows a number of different chemistries for each component, including a rough idea of the cost of each component. This is clearly only a guide, as cost depends on the chemistry of the cathode material, the type of binder, the solvent and additives. Crucially, the main drivers for cost are scale and purity, which can only be roughly estimated here, but other similar studies have similar breakdowns.23

The following equations denote the half redox charge and discharge reactions in terms of Li ion movement in the anode and cathode, respectively:

 

 

Li

x

C

n

(s) ⇌

x

Li+(soln) +

x

e− +

n

C(s)

 

(1) 

 

Li1−

x

MO2(s) +

x

Li+(soln) +

x

e− ⇌ LiMO2(s)

 

(2)where

 

The importance of product design in recycling efficiency

This is an often-overlooked concept in circular economy discussions. Recycling is seen as an end of pipe activity, processing a device which has been developed for optimal performance often without thought of dismantling or recycling protocols. Lead acid cells represent a technology which is easy to open (mechanical crushing), rapidly liberating a significant proportion (

ca.

60% by mass) of metal per unit with a suitable value which can be separated from the other components due to significant density differences. By contrast, alkaline zinc–carbon batteries account for about 80% of cells produced worldwide (over 10 billion units p.a.) but each contains 25% steel casing, and only 15% of Zn, and 17% of MnO2. The similar densities, 7.14 and 5.03 g cm−3, coupled with the low cost of virgin material, makes recycling less economically viable than lead acid batteries.

 

 

Macroscopic design issues

For automotive applications, battery packs need to be both power and energy dense, which can only be achieved by aggregating cells into modules, and modules into packs. 31 Increasing the number of cells in a module decreases the ratio of active material: cell case and complicates the issue of opening cells. A Tesla S85 Mk 1 battery pack, for example, contains 7104 cylindrical cells, whereas a Nissan Leaf Mk 1 22 kW h battery pack is made up of 192 pouch cells, and a BMW i3 Mark 1 22 kW h battery pack contains 96 prismatic cells. This is shown pictorially in Fig. 3 . The active energy storage mass of the pack can be calculated as 64%, 60% and 82% of the total pack mass for the Tesla, Nissan and BMW respectively.


Fig. 3

Different types of battery cells and how they are organised to form modules and packs. 32

where eqn (1) describes the anodic half reaction in a typical lithium ion battery, and eqn (2) shows the cathodic half redox reaction (where M is the transition metal(s) of choice). Discharging takes place from left to right, whilst charging takes place from right to left.

One of the main issues is the way in which the cells, modules and packs are assembled. The cells themselves are hermetically sealed and the modules and packs are often glued together with adhesives.32 This provides rigidity, but means that they can often only be dissolved in molecular organic solvents. This precludes disassembly as a viable recycling method due to the time and solvent requirements. The assembly of packs and modules is probably the largest barrier to disassembly and hence efficient cell dismantling and recycling. Marshall et al. describe the complex steps required to separate cells.16 While the disassembly approach is more successful at recovering more material, and in a purer state, this is naturally at the sacrifice of the speed at which material can be processed – which is limited by the pack, module and cell opening. The structures are clearly established for safety and potentially cell longevity, but at the expense of recycling efficiency. When dismantling is slow and costly, the only method of recycling becomes pyrometallurgy, which is expensive and inefficient. Recycling is therefore in a “Catch 22” situation, where cell and pack design controls recycling strategy. The lack of binding legislation or policy does not enforce improvements in recycling efficiency and this in turn does not influence cell design.

 

The ideal recycling process

K

i

, of a specific phase in a separation step could be expressed mathematically as: 

 

(3)where

q

i

is the selectivity coefficient for a specific separation property of each component and

m

i

is the mass fraction of that component. To achieve a high degree of separation, the selectivity coefficient needs to be high for the selected component compared to the other components, and the amount of material needs to be as high as possible. The larger the number of components, the less likely it is that

q

i

(for the component of interest) will be larger than the other components. Comminution increases the number of components, decreases the mass fraction of the target element and complicates separation. For a lithium ion battery, there are the 10 components listed in −3, separation is simple. Carrying out the same approach with lithium ion batteries is not possible due to the similarities in density between the cathode materials and current collectors. Accordingly, separation is based on a variety of steps including redox, solubility, electrostatic and magnetic properties where the differences in selectivity coefficients is smaller.

Theoretically, the aim of a recycling process is to divide the constituents of a device into chemically pure, distinct phases. Separation of components in a mixture depends upon a difference in the properties of the components which can be utilised to bring about a separation. These differences include size, density, wettability, magnetism, redox behaviour, surface charge, solubility, appearance, phase change, adsorption, and combustion. The ability to bring about a separation depends upon the relative affinity of the components to the property being distinguished. The separation factor,, of a specific phase in a separation step could be expressed mathematically as:whereis the selectivity coefficient for a specific separation property of each component andis the mass fraction of that component. To achieve a high degree of separation, the selectivity coefficient needs to be high for the selected component compared to the other components, and the amount of material needs to be as high as possible. The larger the number of components, the less likely it is that(for the component of interest) will be larger than the other components. Comminution increases the number of components, decreases the mass fraction of the target element and complicates separation. For a lithium ion battery, there are the 10 components listed in Table 2 , whereas in a lead acid battery there are only 4. For the latter, the separation is based on density and, given that lead and polypropylene have values of 11.3 and 0.9 g cm, separation is simple. Carrying out the same approach with lithium ion batteries is not possible due to the similarities in density between the cathode materials and current collectors. Accordingly, separation is based on a variety of steps including redox, solubility, electrostatic and magnetic properties where the differences in selectivity coefficients is smaller.

Table 2

Types of electrode lixiviants and issues associated with their use

 

Component Solvent Issues Advantages Binder Organic solvent Higher cost, flammability Targets binder and does not etch metal Collectors Oxidising agents Aq. acids – poor LCA Lower cost Aq. base – low cost Metal oxide Acidic solutions Reactivation of cathode material Lower cost Carbon None Insoluble Easily reactivated

A more ideal scenario would be to dismantle the pack and remove the liquid phase (2 components), followed by splitting the solids into individual bins of electrodes, packaging and separators as shown schematically in Fig. 4. Each electrode bin would only contain the current collector, active material and a binder. Use of binder which was soluble in water or a Green solvent such as alcohol would enable simple separation of the current collector from the active material without significant use of ancillary chemicals.33 Separating the powdered active material from the current collector sheet can be done on size. This ensures the minimum number of components in each bin and the largest possible selectivity coefficient. Li et al. used two water miscible binders; a carbon black (CB)/carboxymethyl cellulose (CMC)/styrene butadiene rubber for the anode and a CB/CMC/PVDF binder for the cathode. The normal PVDF binder would be cast using N-methyl-2-pyrrolidone which is toxic and expensive. Water with binders which can be dispersed in aqueous solutions were used to produce a cell which showed similar charge–discharge properties. The water-soluble component of the binder (CMC) enabled the cell to be recycled using water. The recovered NCM523 cathode material was re-lithiated, and shown to have a performance similar to the unused material.34


Fig. 4

 

Schematic diagram of an idealised

vs.

a real battery recycling process.

 

 

 

Macroscopic recycling issues

The first design issue that needs to be addressed is how to open the pack, module and cell easily. Clearly the outer pack design needs to be as robust as possible, so it does not fail in service, but this does not preclude mechanisms which are easier to open. Metallic tools need to be avoided to decrease the possibility of shorting the cell and igniting the contents. 35 This could be a particular issue for prismatic and cylindrical cells, which are already used in primary battery products.

The importance of a simple disassembly mechanism has been highlighted by several authors and some attempts have been made to automate the opening of pouch cells.36–38 Many groups agree that module and pack disassembly tasks should be carried out using smart robots.39 While this can be achieved at pack level, it is complicated by the myriad pack designs and fixings and glues that are used to assemble packs.40 While pack shapes and configurations will clearly change with each manufacturer, standardisation in fixing type would simplify disassembly as it would only require one tool. It should be noted that for robotic disassembly, flexible cables present a challenge, and a move to solid busbars that are fixed in a predictable position (vs. a flexible cable) may simplify disassembly.

Pack and module designs vary significantly, even within a manufacturer's own fleets. It is common, regardless of the form factor of the cell, to assemble groups of cells into modules but the variety of the number of cells and the series/parallel configurations is broad (as indicated in Fig. 4). Typically, regardless of the arrangement, cells in a module are permanently affixed to one another and are not intended to be broken down as part of a servicing activity. A total of nine joining technologies are identified,41 all of which are, in the sense of serviceability or disassembly, permanent processes, and attempts to disassemble these joints are likely to be a destructive process. An example of pouch cell tabs welded to copper busbars in a 2S2P module for a first-generation Nissan Leaf module is shown in Fig. 5(a).


Fig. 5

 

(a) Pouch cell tabs ultrasonically welded to copper busbar, (b) module interconnections utilising threaded bolts through solid busbars (c) dismantling of pack sub-assemblies using standard shop tooling – all Nissan Leaf 1st generation.

 

Connections between modules are typically more serviceable and may employ technologies such as threaded or torqued connections or bespoke, mechanically constrained push-fit connections. The repeatable functionality of these connections makes the modules both more easily replaceable in a service situation and simpler to disassemble at the end of life. Fig. 5(b) shows solid busbar interconnections between modules of a first-generation Nissan Leaf held in place by threaded bolts. The mechanical fixings between modules are often equally serviceable, Fig. 5(c) shows manual disassembly of a sub-pack architecture which would more easily lend itself to automation.

At cell level, standardisation is more difficult to achieve since cylindrical, pouch and prismatic cells are commonly used. Recent studies have shown that the polymer separator between the electrodes can be used as a method to separate the active components of the cell. An apparatus built by Li et al. used a vacuum conveyor equipped with pinch grips and a series of skimmers to separate the anode from the cathode and the separator.42 A Z-fold uses a single sheet of separator wound alternately between the anode and cathode in a cell. As the Z-fold is flattened out the anode and cathode will automatically be partitioned onto opposite sides of the separator. This is shown schematically in Fig. 6. It could work equally well with prismatic, cylindrical or pouch cells, but automated methods of cell opening need to be built into the design. It should, however, be noted that these separators can be very fragile, particularly at end-of-life (EoL), and often rip during disassembly. Separators are also becoming more complex; early versions were homopolymers of polypropylene, but later ones are composites or mixed polymer (PE/PP) and some even contain ceramic layers, complicating the recycling process.


Fig. 6

Schematic diagram of a prototype device for separating anodes, cathodes and separators from a lithium ion battery. 42

An alternative approach to dismantling could also be achieved by simply changing the geometry of the electrode connector tabs which connect the electrodes of the same polarity together. This is beginning to be done by some manufacturers. The incorporation of a failure mode, probably cut as notches post EoL, as shown in Fig. 7(a) (dashed line) could enable simple separation of anode and cathode stacks and enable the polymer components to be easily segregated.


Fig. 7

 

(a) Schematic diagram showing how opposite tab alignment could simplify cell disassembly (dashed lines show potential cut points), (b) photograph of a BYD pack where lithium ion battery cells (blades) form a structural element of the vehicle and offer easier disassembly.

 

A recent development by Chinese battery manufacturer BYD has developed long, thin cells called blades, which fit into a grid and can impart structural strength to the battery pack. Although details are currently unpublished, this could remove the need for separate modules and glues, enabling simple disassembly and exchange for cells when faults develop. The manufacturers claim that the optimised pack structure enables an increase of 50% in space utilisation and leads to safer operation of cells.43 This shows that battery development is a fast-moving subject.

An important aspect of pack design is the standardisation of fixings. Each manufacturer places pack fixings in different locations which need different tools to open them. Standardisation of pack opening is fundamental to cell dismantling.40

 

Comminution

vs.

dismantling

 

Currently, the over-riding factor with lithium ion battery use and recycling is safety. The potential consequences of the battery chemistry coming into contact with moisture necessitates hermetic sealing of the cell, which naturally complicates end of life processing. To date, the majority of processes have either shredded the cell under an inert atmosphere or used high temperature pyrometallurgy.

Shredding the cells effectively dilutes all the constituents, which then need to be physically separated by froth floatation, electrostatic, magnetic or density separation techniques in combination with wet chemical processes. All of these have inherently low selectivity coefficients leading to products with lower purity, lower value and necessitate extensive purification to re-enter the manufacturing circle.44

Pyrometallurgical processing is currently seen as the pragmatic approach to recycling as it already functions for a range of disparate materials; however, it is only economically viable through implementation of significant gate fees to process the material. It is fast and generally safe but will lose the volatile elements and have a higher energy and ancillary input. Nonetheless it will lead to a higher purity of the highest value metals. From an LCA perspective, opening and separating the cell into the electrodes, separators and electrolyte and delaminating the separated electrodes will require fewer steps and potentially fewer ancillary chemicals than shredding. This naturally has the barriers that the cells are not easily opened and there are safety issues with the current cell designs.

From a pragmatic perspective, the future of lithium ion battery processing will probably require a mixture of processing, including pyro- and hydro-metallurgical methods blended with mechanical separation, as the design enables it. This will lead to different grades of

Delete
  • wechat

    +86 18686976230: +86 18686976230