Recycling of lithium-ion batteries: a novel method to separate coating and foil of electrodes

Highlights

• A new process to recycle lithium-ion batteries is proposed.

• The binder is thermally decomposed to weaken the adhesion of electrode coatings.

• An air-jet-separator detaches coating powder from current collector foils.

• Binder decomposition and impacts produce primary particles instead of agglomerates.

• 97% w/w of the coating can be regained with aluminum impurities of only 0.1% w/w.

Abstract

Lithium-ion batteries will play a crucial role in the development of mobile consumer devices, stationary energy storage systems, and electric mobility. The growth in these fields will bring about a surge in the lithium-ion battery market. This leads experts to agree that more effective recycling processes are needed in conjunction with the recycling of lithium. This calls for an entirely revolutionary recycling process which we here have attempted to develop.

Our approach uses thermal decomposition of the polyvinylidene fluoride binder to lessen the cohesion of coated active material particles and weaken the adhesion between coating and foil. Then, an air-jet-separator is able to detach the coating powder from the current collector foils while stressing remaining particulate agglomerates. This separation process named ANVIIL (Adhesion Neutralization via Incineration and Impact Liberation) was tested on a laboratory scale with electrode rejects. We compared this to the widely used mechanical recycling process that utilizes a cutting mill to separate the current collector and coating. Intermediates and products were characterized using thermogravimetric analysis, tape adhesion tests, atomic absorption spectroscopy, particle size analysis, and gravimetric sieve analysis. We found that 97.1% w/w of the electrode coating can be regained with aluminum impurities of only 0.1% w/w, 30 times purer than the comparative process. This demonstrates a more effective recycling process than is currently available that also enables the recapture of lithium from the electrode coating.

Introduction

The transition from combustion engines to electric vehicles is one highly pursued path in the effort to reduce oil consumption. Electric mobility can also reduce overall energy consumption if it is applied as part of a bigger transition to a more sustainable economy (Bauer et al., 2015, van Vliet et al., 2011). Battery-powered support in hybrid vehicles will enable the use of high-efficiency technology that combustion engines cannot adopt on their own such as regenerative braking (Lv et al., 2015) and electrical driving in slow city or traffic jam situations (Lee et al., 2013). Furthermore, electric vehicles (EV) can contribute to the overall reduction of air pollution in mega-cities (Bauer et al., 2015, Baumann et al., 2012). The demand for lithium ion batteries (LIBs) is expected to rise even further in the field of portable electronics as well as stationary energy storage as LIBs will play a substantial role in enabling the efficient use of renewable energy sources (Hammond and Hazeldine, 2015, Scrosati and Garche, 2010). In all fields, electrochemical storage technology – predominantly lithium ion batteries – determine the lifetime of some products or are the component which has to be replaced first. These factors will lead to a growing demand in spent battery collection systems (Sun et al., 2015) and recycling technologies for electrochemical storage devices (Hanisch et al., 2015a, Hoyer et al., 2014, Richa et al., 2014, Zeng and Li, 2014, Ziemann et al., 2013).

Recent research states that the demand of cobalt for the production of 20 million EV batteries per year – with averaged recent material compositions – would be about equal to current world mine production of cobalt and would deplete current cobalt reserves in less than 60 years. The nickel needed for the production of 20 million EV batteries per year would be larger by 170 fold than today's existing production capacity (Delucchi et al., 2014). These figures alone signify a need for an alternative method of resource acquisition. The recycling of products containing Ni and Co has already shown benefits in terms of scarce and expensive material conservation as well as in the reduction of material production emissions like SOx (Sullivan and Gaines, 2012). Energy consumption during lithium-ion battery production can be reduced due to the use of recycled materials (Simon and Weil, 2013, Sullivan and Gaines, 2012). Additionally, in-production recycling of active materials can result in substantial economic and ecological savings (Hanisch et al., 2015b).

However nowadays, recycling processes focus on the recycling of cobalt and nickel, not lithium. This is related to the higher raw material costs of these transition metals compared with the costs of the less expensive lithium (Dunn et al., 2012, Ziemann et al., 2012). No real lithium scarcity is foreseen until 2050 when easily extractable reserves in stable countries could decrease significantly (Weil and Ziemann, 2014). Recent debates often conclude that there is sufficient lithium available in terms of resources, but the potential increase in the production rate has yet to be properly regarded. Lithium resources, excluding the lithium in the oceans, are calculated at 30 Mt (Kushnir and Sandén, 2012) or 38.68 Mt (Gruber et al., 2011). Thus, Lithium will only be in critical supply within this century if EV batteries become used on large scale “or if batteries are not recycled” (Kushnir and Sandén, 2012) with recycling rates of at least 90% (Gruber et al., 2011). A substantial potential for lithium recycling is seen in Europe (Miedema and Moll, 2013), in China (Zeng and Li, 2013) and in the United States (Wang et al., 2014). All aforementioned experts agree on the necessity of lithium recycling for long-term sustainability.

Long and diverse process chains have to be applied to recycle lithium-ion batteries efficiently especially in the case of those from electric vehicles (Hanisch et al., 2015a):

First, a deactivation step can lower the dangers resulting from the stored energy and chemical reaction potential of charged lithium-ion batteries. The batteries should be discharged to minimize the stored energy. Thermal pretreatment steps then can be used to volatize the electrolyte or even to decompose all organic compounds of the battery cells (Georgi-Maschler, 2009, Vezzini, 2014a). Alternatively, lithium-ion batteries could be frozen thus inactivating the galvanic elements. The temperature of all flammable components is reduced far below their flashing points so that an ignition is prevented.

Bigger battery systems (e.g. from electric vehicles) can be disassembled to decrease the size and remaining electrochemically stored energy. At present, the variety of battery system designs is wide due to a low degree of standardization. Consequently, manual disassembly steps are usually employed (Herrmann et al., 2014).

In most cases, the next step of the recycling process is crushing. The battery cells are opened and valuable components of the cells are released. Dry and wet crushing operations can be applied (Zhang et al., 2013a).

The resulting fragments of electrodes, separator, case, and a powder fraction consisting of coating agglomerates have to be separated for further recycling. This can be accomplished with a combination of sorting and sieving steps such as magnetic separation and air separation methods like cross-flow classification or zigzag sifting (Hanisch et al., 2015a).

A more intensive, second milling step then can be applied to remove the remaining particulate coating from the current collector foil. This increases the yield of coating fragments and foil with fewer impurities in each of the fractions (Hanisch et al., 2011a). Due to the fact that the mechanical stressing not only loosens the coating from the foil but also results in some smaller foil fragments, these fragments have to be screened.

Alternatively, other mechanisms such as dissolving the binder in a solvent (Contestabile et al., 2001, Hanisch et al., 2011b, Hanisch et al., 2015b, Li et al., 2009c, Li et al., 2013) or decomposing the binder under high temperatures (Hanisch et al., 2013, Hanisch et al., 2014) work to separate the current collector and the particulate coating.

In this work, a combination of thermal and mechanical processes is used to separate current collector foil and coating. Depending on the binder chemistry the electrode compound is heated up to 400–800 °C and the binder (mostly polyvinylidene fluoride) decomposes releasing hydrogen fluoride and other gaseous components. A vacuum pyrolysis at 600 °C with a pressure below 1 kPa for 30 min can be used to evaporate organic components (Sun and Qiu, 2011). Other processes using thermal decomposition of the binder components are discussed in literature (Gu and Nie, 2011, Lee and Rhee, 2002, Li et al., 2009c, Li et al., 2013, Lu et al., 2013, Song et al., 2013, Sun and Qiu, 2011, Sun and Qiu, 2012).

The regained powder fraction still contains all of the components of the original coating: the desired lithium-rich battery active material, graphite, conductive carbon, binder, and even some components from the electrolyte – like organic residues such as ethylene carbonate, dimethyl carbonate, and conductivity salt. In the event of a previous thermal decomposition, the organic fractions and conductivity salt would have already reacted.

Metallurgical processes have to be applied in order to regain the valuable metals cobalt, nickel and lithium from the former-coating powder. In pyro-metallurgical processes several components of battery cells or the battery cells themselves are melted. The transition metals nickel, cobalt, and copper can be recycled from the cast while lithium and aluminum remain in the slag. Copper, cobalt, and nickel can be regained after further processing of smelter output via leaching and solvent extraction. 98% w/w of cobalt and 93% w/w nickel can be recovered pyro-metallurgically (Swart et al., 2014). Due to the fact that pyro-metallurgical recovery of lithium and aluminum is currently not possible from pyro-metallurgical processes, further treatment is necessary to recover lithium (Elwert et al., 2012, Georgi-Maschler et al., 2012, Li et al., 2013).

Alternatively or additionally to pyro-metallurgy hydro-metallurgical processes can be applied. These processes like leaching, extraction, crystallization and precipitation are needed to recover pure metals, e.g. lithium, from separated coating materials or from slag of pyro-metallurgical processes.

Furthermore, products of the hydrometallurgical processes can be used to re-synthesize battery active materials in battery grade. More comprehensive reviews on hydro-metallurgical processes in the field of recycling of lithium-ion batteries are reviewed in literature (Chagnes and Pospiech, 2013, Hanisch et al., 2015a, Joulié et al., 2014) and included in general reviews about lithium-ion battery recycling (Xu et al., 2008, Zeng et al., 2014, Zhang et al., 2013b). The first step of a hydro-metallurgical treatment is the leaching of the regained powder fractions including the active material fraction. In this process, the lithium-transition metal compounds are leached in inorganic or organic acids in order to prepare them for the later separation steps. The reduction of impurities and organic residues as well as the separation of the different product metals as purely as possible are the goals of the leaching process step. Nitric acid (HNO₃) can be used with the addition of hydrogen peroxide (H₂O₂) to leach LiCoO₂ (Lee and Rhee, 2002) as well as sulfuric acid (H₂SO₄), hydrochloric acid (HCl), citric acid (C₆H₈O₇*H₂O) (Li et al., 2014), malic acid (C₄H₆O₅), and aspartic acid (C₄H₇NO₄) (Li et al., 2013). Further leaching processes are discussed in detail in scientific publications (Kim et al., 2012, Li et al., 2009a, Li et al., 2009b, Weng et al., 2013).

After the leaching step, the metals can be precipitated selectively from the resulting leaching solutions of hydro-chloric, sulfuric, and nitric acid (Joulié et al., 2014). Furthermore, electrochemical processes can be used to regain LiCoO₂ (Ra and Han, 2006) and to separate mixtures of cobalt and lithium ions (Iizuka et al., 2013). Cobalt can as well be regained by solvent extraction (Granata et al., 2012a, Jha et al., 2013). Further research about hydro-metallurgical processes and processes to obtain high quality raw materials for battery active material synthesis from metal solutions can be found in literature (Gao and Li, 2012, Intaranont et al., 2014, Li et al., 2008, Lupi et al., 2005, Ma et al., 2013, Nan et al., 2006, Sun and Qiu, 2012, Swain et al., 2007, Zhu et al., 2012, Zou et al., 2013).

More information about the state of the art of recycling of lithium-ion batteries is summarized in review publications (Bernardes et al., 2004, Espinosa et al., 2004, Georgi-Maschler et al., 2012, Granata et al., 2012b, Hanisch et al., 2015a, Nan et al., 2006, Swain et al., 2007, Vezzini, 2014b, Wen et al., 2012, Xu et al., 2008, Zeng et al., 2014, Zhang et al., 2013b). An exemplary process chain to recycle lithium-ion battery cells is illustrated in Fig. 1.

The knowledge of the above mentioned processes (Section 1.2) along with the following facts led the authors to suggest a new process to separate current collector foils and coating.

The binder polyvinylidene fluoride (PVDF) is the component to facilitate adhesion between coating and current collector foil as well as cohesion between the single electrode coating particles (Chen et al., 2013, Liu et al., 2012). PVDF is broadly used in lithium-ion electrodes (Li et al., 2011) and has a lower decomposition temperature than graphite, carbon black, and the aluminum and copper foils (see Section 3.1). Therefore, this prompted us to incorporate a heating step in our process.

Regaining pure intermediate materials is necessary for the re-synthesizing of new active materials (Krüger et al., 2014, Weng et al., 2013). Impurities stemming from the mixture of materials and various electrochemical reactions during the battery's life as well as impurities from the recycling process itself have to be reduced to a minimum to avoid a downcycling of these high-tech materials. Impurities could be organic components like organic carbonates of the electrolyte, their decomposition products, graphite, and carbon black. Impurities could also be inorganics like copper and aluminum from current collector foils, iron, fluorine compounds, and other additives from the electrolyte such as magnesium.

In experiments to re-synthesize LiNi_0.33Mn_0.33Co_0.33O₂, particularly high aluminum concentrations in leaching solutions disturbed material conditioning and therefore influenced the secondary particle shape in an undesirable way. The BET surface was “more than twice” of the sample with lower aluminum contents. SEM images showed smaller non-spherical secondary particles. The electrochemical cycling performance of recycled material from spent lithium-ion batteries at 1C was slightly decreased due to lower electrode kinetics (Krüger et al., 2014). This motivates to find a process to regain active materials purely.

Mechanical stressing, e.g. applying a cutting mill, results in broad particle size distributions. In the case of the separation of electrode compounds this causes small 50–200 μm fragments of aluminum and copper foils as shown in Section 3.3 on the one hand.

On the other hand, simple mechanical separation of lithium-ion electrodes reduces the coating to agglomerates with diameters up to 250 μm. These agglomerates could have a larger diameter than the smallest copper and aluminum shreds.

However, impact stresses can be applied to transform kinetic energy into breakage energy of agglomerates (Mishra and Thornton, 2001, Schilde et al., 2011). This effect can be used to separate two or more fractions, and especially to loosen coatings from their substrate and break agglomerates. When the agglomerates are broken, the resulting primary particles can pass fine mesh sizes and a fine sieving process works to reduce impurities as shown below. The finest mesh sizes with a dry powder are obtained using air jet sieve technology (Hanisch et al., 2011a, Schmidt, 1980). As a result less mechanical stressing, would result in less smaller foil fragments, while a deagglomaration, using impacts stressing, reduces the possible sieving mesh size. Fine sieving can further separate particles by micrometers for highly accurate classification. Alltogether this will lead to a purer powder fraction.

Our method, the adhesion neutralization via incineration and impact liberation (ANVIIL) approach, combines all the aforementioned (see Section 1.3) facts:

A thermal decomposition of PVDF results in lower adhesion between coating and foil as well as lower cohesion between the coated active material particles. A new designed air-jet-separator separates coating powder from current collector foils using mesh sizes down to 50 μm to guarantee high powder purities preventing the foil fractions >50 μm from passing through the sieve. This mechanism is based on air jet sieve technology normally used for sieve analysis. It is equipped with a higher air speed and a lid to create impact stress on foils and agglomerates as illustrated in Fig. 2.

It can be embedded in a recycling process for lithium ion batteries as shown in Fig. 1. At first, batteries are crushed under inert atmosphere (or at least under conditions with reduced oxygen and water content) in order to prevent the ignition of flammable components in the presence of activation energy from the electric charge of the battery.

After an optional extraction of the conductivity salt LiPF₆, the electrode fragments are dried in a dryer under inert drying gas from which the organic solvents can be condensed and regained afterward. Following magnetic separation, air classifiers separate heavy case materials, electrode compound pieces and the light separator. So electrode pieces are fed to a gas-tight rotary furnace where the binder is thermally decomposed. Afterward, the heated electrode pieces are stressed and sieved in an air impact separator resulting in a clean fraction of the former current collector foils and a gas stream loaded with active material and graphite particles. These desired transition metal oxide and graphite particles are separated from the air stream using a cyclone or filter mechanisms.

In the following, the ANVIIL process is described in detail. Intermediates and products were characterized using thermogravimetric analysis, tape adhesion tests, atomic absorption spectroscopy, particle size analysis via laser diffraction, and gravimetric sieve analysis. The results were compared to a mechanical process and were found to give an improved recycling yield as well as a significantly higher purity.