Environmental impacts, pollution sources and pathways of spent lithium-ion batteries

There is a growing demand for lithium-ion batteries (LIBs) for electric transportation and to support the application of renewable energies by auxiliary energy storage systems. This surge in demand requires a concomitant increase in production and, down the line, leads to large numbers of spent LIBs. The ever-increasing battery waste needs to be managed accordingly. Currently, there are no universal or unified standards for waste disposal of LIBs around the globe. Each country uses one or a combination of practices such as landfilling, incineration and full or partial recycling depending on the number of batteries leaving the market, current legislation and infrastructures. Informal disposal or reprocessing is not a rare activity. This review records, identifies and categorises the environmental impacts, sources and pollution pathways of spent LIBs. The drawbacks of the disposal practices are highlighted and the threats associated with them are discussed. The evidence presented here is taken from real-life incidents and it shows that improper or careless processing and disposal of spent batteries leads to contamination of the soil, water and air. The toxicity of the battery material is a direct threat to organisms on various trophic levels as well as direct threats to human health. Identified pollution pathways are via leaching, disintegration and degradation of the batteries, however violent incidents such as fires and explosions are also significant. Finally, the paper discusses some of the main knowledge gaps for future assessments. The current study oﬀers a comprehensive overview of the threats and hazards that need to be managed in order to ensure the design and implementation of safe disposal and processing options for spent LIBs.


Introduction
Lithium-ion batteries (LIBs) are found in all aspects of our livesfrom small portable electronic devices through electric vehicles (EVs) to battery energy storage systems (BESS).LIBs are perceived as crucial to support the wide adoption of renewable energy sources as these do require BESS to manage the intermittency in their power supply for a reliable operation of the electricity grid.The application of LIBs in electric traction has initiated a revolution in the automotive industry that is motivated to decarbonise the transport sector and reduce local air pollution.In 2009, there were nearly 25.6 GW h (about 134 000 tons) of LIB energy storage capacity placed on the global market, in 2019 it was about 218 GW h (over 1. 2

million tons), and it has been
This journal is © The Royal Society of Chemistry 2021 estimated that in 2030 more than 2500 GW h (over 12.7 million tons) will be placed on the market, which subsequently results in a vast amount of waste from the spent LIBs. 1 Various international and domestic initiatives like the European Green Deal in the EU, 2 the UK Government pledge to ban new ICE cars by 2030 (hybrids by 2035) 3 or the need to establish zero-emission mobility as an urgent priority, 4 require a very large increase in the production of new LIBs.Moreover, transitioning to a low carbon economy 5 requires building new, more powerful BESS all around the world.These installations will be used as storage of intermittent solar, wind, or wave electricity, grid support, frequency stabilisation and behindthe-meter-storage, replacing conventional technologies.On one hand, the increased number of LIBs has created concerns over the availability of some critical metals, and their wild shortterm price volatility which could affect the sustainable supply of such energy storage devices. 6There are also some ethical and environmental concerns with the extraction of some metals for LIBs. 7On the other hand, this surge in demand for batteries will require a concomitant increase in production and, down the line, large numbers of LIBs reaching end-of-life (EoL).Hence, that will cause an ever-increasing battery waste that needs to be managed accordingly. 8However, many types of batteries currently do end up in landfills or are incinerated, primarily due to the lack of adequate standards; enforcement of regulatory controls, inefficient, or absence of, national battery

Wojciech Mrozik
Wojciech Mrozik is a Faraday Institution Research Fellow at Newcastle University, UK.He is part of the ''SafeBatt -Science of Battery Safety'' and previously ''Reuse and Recycling of lithiumion Batteries'' projects funded by Faraday Institution.He is an expert in environmental and analytical chemistry with a particular interest in environmental fate assessment and removal of anthropogenic pollutants.Last few years Wojciech's research focused on the safe application, operation and utilization of lithium-ion batteries.That concerns understanding of battery thermal runaway, fire prevention and mitigation.

Mohammad Ali Rajaeifar
Mohammad Ali Rajaeifar is currently a research associate at Newcastle University.He studies technologies and management strategies and provides sustainability analysis of different production systems.He started his MSc in the area of sustainable agriculture and biofuel production in 2013.Then he extended his research to sustainability assessment of biorefineries, low carbon transport systems and waste management systems during his PhD and post-doctoral studies.More recently, he has developed his research area by focusing on the sustainability of EVs and lithium-ion batteries.Mohammad is also an expert in data collection and analysis, compiling data inventories, and assessing the environmental impacts of complicated industrial systems.

Oliver Heidrich
Since 1992 Oliver has worked with national and international companies, local authorities and Universities in waste, recycling, environmental, life cycle assessment, stakeholder and corporate management.He was a director of Econoplas, a plastic manufacturing business and SEQM a business consultancy business.He provides detailed understanding of the inter-relationships between resource use and their spatial distribution within and outside cities, in relation to climate change, the built environment, and associated resource requirements.Besides his many expert reports and research papers, Oliver is concerned to see the results of his research disseminated through the wider press and digital media.collection and recycling schemes. 9As a result, human health and environmental quality could be placed at risk as a wide range of pollutants could be released like heavy metals or hydrofluoric acid (HF) when batteries are disposed of inappropriately.
The main aim of this study is to provide an up-to-date overview of the environmental impact and hazards of spent batteries.This paper reports and discusses the fate, disposal routes and potential pollution sources and pathways from spent LIBs.Despite the clear importance of this area, the data on the environmental impact of EoL LIBs is extremely limited, thus, this study identifies the possible environmental impacts associated with battery EoL.There is an urgent need from manufacturers, waste handling companies, recycling firms and public stakeholders to understand this matter as wrong disposal practices, lack of data or improper identification of hazards will drastically increase the number of incidents, fires and potentially fatal accidents.The possible emission routes and pollution pathways e.g.air, water and land, are also evaluated through the course of this study.We have also recognised possible hazards to human health and ecosystem quality due to different battery treatment and disposal practices.Finally, the knowledge gaps are identified that need to be covered if the right disposal practices and management systems are to be implemented.

Disposal of spent batteries
The management of spent LIBs incorporates reuse, recycling, landfilling, processing or (unfortunately) illegal disposal (Fig. 1).The local disposal routes depend on national legislation and regulations, recycling capacities, collection systems, consumer behaviours and the battery retail markets. 10Clearly, different approaches are required for small batteries compared to large ones.
Small LIBs are returned by the customer either through battery drop-off points (LIBs that can be easily detached from the device) or given to waste electrical and electronic equipment sites (WEEEs) if the user cannot remove them from the electrical device.From these collection points, they are transported to sorting facilities and later to be recycled or landfilled.Through ignorance of the collection system or idleness, small spent LIBs may be put into general waste or into recycle bins by customers, which led to damaging fires during transportation or in material recovery facilities (MRFs).Veolia, one of the global waste handling companies, has noted a 38% increase of incidents since 2017, 11 due to the presence of the LIBs in the waste stream.At MRFs, LIBs will be crushed or penetrated during the standard processing, which can also lead to fires or even explosions.The Environmental Services Association estimates that 12 there were ca.250 fires in the UK waste treatment facilities caused by small LIBs between April 2019 and March 2020, compared to ca. 130 in the year to March 2020, a rise of 25% of all fires in these facilities. 13,14The German steel recyclers confederation (BDSV) has claimed that 90% of fires at their associated sites in the last years were caused by LIBs. 15ome specific examples of such fires are the incident at the Great Blakenham scrapyard, UK (15th Sep 2017) where 100 tonnes of metal and a crane were caught up in fire; 16 multiple incidents at Tokyo's Musashino Clean Center garbage disposal facility, Japan, in 2017 and 2018; 17 and some other regular incidents in the UK, USA, Scotland, and Germany. 18arge LIBs are typically transported by professional services once removed from EVs, 19 however, their health status is usually unknown.1][22][23][24] These events may take place during transportation, storage, recycling or landfilling.

Recycling
2.1.1Low recovery rates.Material recovery from the EoL batteries is the preferred option over the other disposal routes 10,25,26 since LIBs are now considered a strategically valuable waste stream containing relatively valuable metals such as cobalt, nickel or lithium. 27,28][31][32] Material recovery is also the way to be less dependent on raw mining elsewhere or to protect the country's own small reserves.For instance, the lithium demand for LIBs produced in China by 2050 could meet up 60% by recycling. 33Currently, China is the largest consumer and producer of LIBs and recycling of spent LIBs has only started recently. 34Although some 14 pieces of legislation try to manage the emission pathways of all types of batteries waste, effective regulation of controlling spent LIB is still in its infancy. 34In the USA the management and regulation of LIBs is also fragmented, due to its Federal States, with some federal policies guiding battery disposal, specifically the 1996 Mercury-Containing and Rechargeable Battery Management Act to manage certifications and standards of batteries. 35However, overall, it is patchwork of state by state regulations with different levels of maturity and control. 36In the EU, new legislation is coming into force in 2022 to manage and control the raw material sourcing, disposal and recycling of LIBs. 1 Clearly, material recovery should avert the environmental burden from ore mining and beneficiation of metals used in LIBs by replacing the recycled ones in the production chain, 37 and could also avoid the environmental burdens from the other harmful disposal routes.In 2018, it is estimated that 97 000 tonnes of LIBs were recycled globally, 19 including LIBs from portable electronics as well as LIBs from EVs and BESS applications.Small LIBs currently dominate recycling (i.e. more than 80% of the recycled batteries) as the number of batteries reaching actual EoL from EVs is relatively low. 38This is due to the relatively nascent application of LIBs in EVs, and the fact that when EV LIBs have reached EoL they still have potential use in other applications, such as supporting the Grid or renewable energy generation, as they typically retain up to 80% of their original capacity. 39Thus, LIBs from EVs may be finally discarded up to 10 years after their first life. 40 closer look at the recovery of LIBs reveals that compared to the amount of LIBs placed on the global market, the actual recovery rates are quite low.For example, the average collection rate in the EU, one of the well-regulated markets, is nearly 40% while only 12% belongs to LIBs. 41The average recovery rates are also lower for some other less-regulated markets such as Australia and the US. 29,42However, there are still few markets that perform better than the EU.More specifically, South Korea and China together showed better recovery rates.Actually, these countries are the destination of most of the exported batteries (either in a separate battery form or as part of electronic devices) from many parts of the world especially Europe and North America.This is also reflected in the total quantity of LIBs recycled: from 97 000 tonnes treated globally in 2018, 67 000 were processed in China and 18 000 in South Korea. 19owever, it is estimated that less than 40% of the total battery materials actually can be recycled under the current materials flow scheme. 43For instance, in 2016, only around 30% of nickel, 33% of cobalt, 23% of lithium and 5% of graphite were recovered in China. 44Such poor management resulted in economic loss due to the free discharge of valuable materials, as well as damage to the environment and human health. 45,46arious national and international initiatives have been implemented to tackle this low recovery problem such as the 2006 EU battery directive 47 and the proposed new EU regulation which concerns batteries and waste batteries. 48The key challenge, however, is the lack of recycling infrastructure around the world, there are only a few full-scale recycling facilities (Table 1). 49,50Illustration of this problem is the UK situation.Prior to exiting the EU on the 31st December 2020, the regular practice was to ship batteries to mainland Europe for recycling, e.g. to Umicore in Belgium.However, with the restricted access to the EU market and strict regulations treating large LIBs as hazardous materials, the UK, without proper recycling infrastructure, may either have to landfill its batteries 51 or temporarily store them in dedicated facilities until suitable facilities are built or until an appropriate trade deal with the EU is agreed.
Both landfilling and stockpiling will have a potentially negative effect on the surrounding environment as they can release pollutants, contaminants and cause accidental fires, which are discussed below.
2.1.2Recycling methods.Material recovery from the EoL LIBs is currently carried out by one of three major processes: pyrometallurgy, hydrometallurgy, and direct recycling. 25,40yrometallurgy employs smelting in a high-temperature process, which usually involves burning and subsequent separation to produce a mixed metal alloy of Co, Cu, Fe and Ni. 52That is an established technology commonly applied to small and large LIBs, especially to cobalt-rich batteries.Usually this method does not require pre-sorting of battery types, however, the resulting alloy requires further processing.Hydrometallurgy recovers the desired metals from cathode material via leaching in an acidic or basic aqueous solution.Additional steps such as subsequent concentration and purification are later required.This process allows the recovery of almost all LIBs components with high purity. 53However, the separation of the metals from each other requires additional purification steps: in addition, unlike pyrometallurgy, sorting is required.Finally, direct recycling involves the direct re-use of the cathode and/or anode material from the electrodes of spent LIBs after reconditioning.Although promising in principle, as it allows the simple recovery and re-use (without further processing) of most of the battery constituents, 25,54,55 it is still not fully matured as a technology.There are also alternative technologies under investigation such as plasma smelting technology, 37 bioleaching, 56 redox targeting-based material recycling, 57 etc.
Each of the listed technologies has its own limitations in recovering LIB's materials.Therefore, the final process should involve a combination of various techniques with a well-defined material flow chart to assure the highest efficiency of recycling. 58The enormous research and development effort should tackle this issue, however, the gradual but constant change in battery chemistry will require the facilities to shift their operational mode in just a few years.Is the process safe to operate with new chemistry?Are there any hazardous by-and end-products that need to be treated accordingly?These and other questions must be answered to ensure the safety of the workers and mitigate the environmental impact of the process.
2.1.3Pollution specific to recycling and the potential environmental impacts.Materials recovery is not a pollutionfree activity.For example, pyrometallurgy is a highly energydemanding process, resulting in GHG emissions and the generation of toxic gases or hazardous slag that may need to be landfilled. 31Moreover, the intermittent product, so-called ''black mass'' (a sludgy mixture of lithium, manganese and cobalt), may also contain other hazardous substances such as alkylfluorophosphates that are a serious health concern. 59yrometallurgical recycling process could impose environmental risks in global warming, photochemical ozone creation, carcinogenic and non-carcinogenic effect, ozone layer depletion and eutrophication impact categories. 18A recent study on plasma smelting recycling showed that the employed energy source (for the smelter), recycling efficiency, the quality of the recycled materials, aluminium recovery and adopting a battery waste improvement process could significantly affect the overall environmental impacts of pyrometallurgy. 37n contrast, hydrometallurgy produces far less GHG emissions but requires supplementary wastewater treatment to ensure that receiving waters do not suffer additional pollution, e.g. from acids.Hydrometallurgical recycling processes were reported to impose environmental risks on freshwater and terrestrial acidification. 18ohr et al. (2020) compared the environmental impacts of recycling of different battery chemistries (i.e.NMC, NCA, LFP and solid-state) using a conventional pyrometallurgical, a conventional hydrometallurgical process and an advanced hydrometallurgical process (that recovers graphite and electrolyte). 60The results indicated that the advanced hydrometallurgical processes could perform the best (in terms of global warming impact) due to additional recovery of graphite and electrolyte, while the pyrometallurgical process performs the worst due to high energy consumption and lack of Li recovery.The authors also suggested that a cell chemistry specific approach for recycling should be taken into consideration since some recycling processes could lead to unfavourable impact on the environment when treating some specific chemistries, for example, hydrometallurgical recycling of LFP and sodium-ion batteries (SIB) could add more burden in abiotic resource depletion potential; or pyrometallurgical recycling of LFP cells could increase the global warming impact.
As direct recycling is still in its very early stages of development, there are no real-world data on potential pollution.According to Dunn et al. (2015), using a direct recycling process for NMC, LCO, LFP and LMO batteries could significantly reduce the emissions of GHGs and SO x (from the production of batteries) and thus mitigate global warming and acidification potential of the whole battery life cycle. 61It should be noted that when talking about the environmental impacts of a specific recycling process, the net impact on the environment is usually considered which is calculated by subtracting the released emissions from the avoided emissions (recycling benefits).Thus, a recycling process itself may have a negative impact on the environment in a specific environmental impact category while the credits from the recycled materials (environmental impacts avoided due to returning the recovered material to the market) might compensate those negative impacts and make the net environmental impact favourable.Overall, it should be noted that the environmental impacts of different recycling processes could vary based on the processed cell chemistry, the employed emission reduction technology, recycling efficiency and the quality of the materials recovered.
2.1.4Comparison to non-lithium ion battery recycling methods.Finally, there is a lesson to be learnt from the already established recycling processes for other battery types.For example, the processing of lead-acid batteries is of major concern especially in developing, countries 62 as, in such locations, recycling is often performed without controlling lead emissions. 63In addition, there is also often a lack of appropriate regulation and the consequent existence of informal and polluting recycling industries.Thus, in the absence of adequate standards and/or lack of enforcement of regulatory controls, the lead recycled from batteries results in significant environmental contamination and human exposure even in established, industrial-scale recycling facilities. 64It has been estimated that over 495 000 deaths and more than 9 million disability-adjusted life years (DALYs) occur as a result of the long-term impact of lead exposure on health.Reports also show that low-and middleincome countries have the highest death rates associated with lead exposure. 65Comparative analysis reveals that current environmental and social influence such as consumption of raw material/metal, energy demand, scrap emissions, recycling or disposal cost; of the lead-acid battery system is greater than that of the LIB system. 66However, it should be highlighted that such studies are performed on retrospective data (i.e. for China in 2016) and do not fully reflect the future with much higher numbers of production and recycling/disposal of LIBs.Therefore, it is essential to assess in detail all the gaps in knowledge and technology, as well as socio-economic issues, and apply these to the future LIB recycling industry.This is important not only for developed countries but even more for emerging economies that usually have lax or weak environmental legislation. 67

Landfill
Landfilling is the main method of disposing of solid waste 68 with the rates of deposition of municipal waste ranging from 53% in the USA, 69 79% in China 70 to 94% in Malaysia. 71f these, around 4% includes electronic wastes (e-waste), often containing batteries. 724][75][76] This is the case not only for many emerging economies but also in developed ones due to (as stated above) lax or weak regulations as well as the lack of recycling infrastructure.Therefore, in the short-term at least, it is certain that LIBs (especially those from small portable devices) will be buried.Currently, due to the relatively small number of EVs, large LIBs are usually recovered. 44owever, with the rapid growth in volume and still inadequate infrastructure, even high-powered batteries may be either landfilled or temporarily stored.Yet, in the long-term, it is more likely they will follow the recovery route rather than landfilling, as it would be a waste of a significant resource of valuable materials. 67he landfill itself can be both a sink or source of pollution that can have negative impacts on surrounding communities. 68he main sources of emissions from landfills are surface dust, landfill gases (e.g.biogas), leachate and fires. 77,780][81] However, in the future it's possible that landfills may be a place of valuable resources for urban mining -the sink of the rare and needed metals that can be extracted without actual mining.
2.2.1 Landfill fires.3][84][85][86] There are two types of fire, surface and subsurface (cavity): and in most cases, they are due to the spontaneous auto-ignition of methane. 87Surface fire spreads outside the landfill area but is relatively easy to extinguish.Cavity fires are a form of combustion (pyrolysis) where the thermal reaction takes place under anoxic conditions deep below the landfill surface. 77,88These are difficult to detect and may create large voids in the landfill, which can cause the landfill surface to cave in.Buried LIBs, which may be holding a charge and contain flammable electrolytes, may trigger, amplify and prolong landfill fires.
Landfill fires caused by (usually small) LIBs are a major emerging problem.The Environmental Services Association (ESA), the trade body representing the UK's resource and waste management industry, reported that 25% of total landfill fires (4500) in the UK in the period 2017-18, were attributed to LIBs: a significant 20% increase in comparison to the previous year. 89Examples of such fires include that at the Dunbar landfill site on 22nd January 2019, where around 300 tonnes of waste burst into flames in waste sorting building. 90Fires in pre-landfilling facilities may be illustrated by Shoreway MRF, USA.At least 25 fires at this facility were caused by LIBs between April 2013 and September 2017.The most severe fire, on the 9th September 2017, required over 100 firefighters to extinguish it, and the fire shut the facility for 3 months and cost over $8.5M in restoration. 91ccurate prediction of the ignition potential of LIBs is challenging, as spent batteries will have various compositions, States of Health (SoH) and charge (SoC); besides, it is likely that cells, modules and packs will be dumped and landfilled.In landfills, batteries may be damaged during compacting and hence ignite either immediately or after burial: in the latter case, serious fires could result due to surrounding flammable materials and existing methane in the landfill environment.Even if damaged batteries do not ignite, the pyrolysis that accompanies thermal runaway will generate significant heat and toxic and potentially explosive gases. 92It should be noted that fires caused by batteries may occur months or even years after they were buried.
The hazards of damaged LIBs are in addition to conventional landfill fires, which produce a mixture of toxic gases and smoke 77,93 consisting of polycyclic aromatic hydrocarbons (PAHs), [94][95][96] dioxins/furans, [97][98][99][100][101] volatile organic compounds (VOCs), 102,103 heavy metals 104,105 polychlorinated biphenyls or organochlorine pesticides 104 and finally, particulate matter (PM) with an aerodynamic diameter smaller than 2.5 mm. 106,1071][22][23][24] Landfill fires do not only affect the air around the site as the smoke can carry particulate matter and chemicals to further distances: for example, it has been shown that there may be a short term rise in the concentrations of heavy metals and Chemical Oxygen Demand (COD) or PAHs 86 in water bodies next to burning sites.Therefore, avoiding battery landfilling or at least neutralization/immobilization of hazardous content must be seriously considered.That could be for instance (i) complete discharging-no excess of the energy, immobilization of copper on aluminium foil; (ii) removal of the flammable electrolyte from the battery -reduced fire risk, formation of hazardous gasses and vapour cloud explosion; (iii) using additional liners (bentonite clay etc.) capable of binding heavy metals -no transport through landfill layers.

Illegal disposal and processing
Wherever there is a potential of making profit there are attempts to bypass official routes of making business.As the recycling of LIBs will be profitable at least to some degree, there is a big chance that some illegal processing will occur, as it happened for waste electronic equipment.Such activity will result in pollution surrounding the processing site, poor working conditions of workers and thus worsening their health and quality of life.Illegal disposal will occur if the recycling or official landfilling will be to ''too expensive'' for some ''business''.The dumpsites will be located in very random places, completely not fit for purpose and that will result in severe pollution, unexpected fires and huge additional costs to local residents/ councils -utilization and remediation of the area.Unexpected incidents (e.g.fires) may even occur months after the disposal making the process of finding responsible/guilty person very tricky.
It is certain that LIBs will be illegally recycled and disposed of. 67,108Moreover, the burden of illegal processing will be mostly put on emerging economies due to the high costs of labour, lack of recycling facilities and strict environmental laws in developed countries. 6A good example of this exploitation is the export of e-waste facilitated by rich and developed nations to poor and developing ones. 109,110This is also reflected in the statistics: for instance, two-thirds of global e-waste collected in 2014 were exported, but more importantly, half of it was through illegal routes. 111This resulted in over 3 million tonnes of e-waste exported outside of regulated schemes.Data for 2019 show that 82.6% of global e-waste flows were neither formally collected, properly managed nor documented. 72The EU generates ca.44.3 Mt of e-waste annually, however only 0.6 Mt is known to reach landfills whilst the rest is dumped, traded, informally exported or recycled in a non-environmentally appropriate way. 112Hence there is no guarantee that exported EoL LIBs will be recycled or processed in regulated, safe and environmentally friendly ways. 44.3.1 E-Waste problem.Interestingly, even now with a relatively low number of EoL EVs, it is possible to buy used LIB cells, modules on the popular, online, auction portals.This means that they are sold outside official collection/recycling schemes, even though they have reached their first EoL.With the growing demand from individuals, e.g. for behind-themeter energy storage, some of the batteries may be illegally sold, especially in countries with weak or no regulations, and the obvious question arises: what happens to such self-made systems when they reach real EoL?Will they be passed to accredited recycling centres or dumped?In addition, what would be the impact of the illegal processing of LIBs on human health and the natural environment?][115][116][117][118][119][120][121] The findings include the fact that heavy metals (i.e.lead or cadmium) and polycyclic aromatic hydrocarbons in the soils or watercourses surrounding such centres are significantly above national safety limits.Cobalt, copper and nickel are also heavy metals included in LIBs that might cause hazards if LIBs are inappropriately treated.Lead-acid battery informal processing can highlight potential issues for LIBs in the future.For instance, estimates have shown that there could be between 10 000 and 30 000 informal lead-acid battery recyclers in 90 low-and middleincome countries that pose a major threat to up to 16 million people. 122Such sites elevate levels of lead in soils and plants 123 and have caused higher concentrations of the metal in children's blood in Serbia, 124 Australia, 125 China 126 and India. 127,128Informal sites rarely control their harmful emissions, are subject to weak or no regulation, and are often located within residential areas. 124,129,130

Incidents involving LIBs
Incidents involving LIBs are mostly focused on fires and the release of toxic gases.In addition to the risk of these hazards to first responders and spectators at, for example, a road traffic accident, there is also the broader spectrum of the environmental impact of such fires.Fig. 2 131,132 shows typical emission pathways from a burning LIB to the surroundings.In addition to heat, fire or a toxic and potentially explosive vapour cloud there is also a possibility of air transportation of substances and with time, their deposition in other, distant places.The close surroundings are also affected by the fire debris, transportation of pollutants by fire extinguishing agents and release of remaining contamination from burnt carcasses -depending on the type and place of the disposal.
The sometimes explosive failure of LIBs is always due to some form of abuse which can be from heat, penetration, blunt trauma or overcharging, and examples of the results of these can be found in the literature citing examples on the land, sea and in the air. 133,134Spontaneous ignition of LIBs has also been reported, particularly concerning electric vehicles; 135 the events leading to this remain essentially unknown, but contamination Energy & Environmental Science Review during production or poor design is often quoted.An increasing and slightly worrying phenomenon is the incidence of fires and explosions in large LIB energy storage systems (LIBESS).The first such incident occurred in Arizona in 2012 in a new 1.5 MW LIBESS owned by Arizona Public Services. 136Several design flaws were identified following the incident including a lack of proper ventilation and inadequate monitoring all of which were supposed to have been addressed 137 when a second LIBESS owned by APS exploded in April 2019 injuring four first responders, two of whom badly.The incident is now believed to have been a vapour cloud explosion. 138,139We listed some other examples of LIBESS incidents in Table 2 to illustrate the rising problem.
In essence, LIBs in thermal runaway produce gases that are a mixture including hydrogen, carbon monoxide, carbon dioxide, HF, short-chain alkanes and alkenes, HCN, NO x and droplets of solvent, 21,23,138,[140][141][142] the last giving the mixture the appearance of a white cloud. 143In broad terms, the composition of the white vapour varies little as a function of cell chemistry, form factor or manufacturer. 138The gas mixture will be vented either when safety valves/blast caps activate or when a pouch cell bursts: at high SOC and in the presence of sufficient oxygen, the white vapour ignites to produce flare-like flames.At low SOC or in the absence of sufficient air (e.g. the air is displaced by the vapour or by fire suppressant) the white vapour does not ignite, 144 in which case, in a confined space, there could be a possibility of a flash fire, fireballs developing, or in extreme cases, even a vapour cloud explosion.
LIBs are also more and more used in marine transportation.They can work as a propulsion system, powering ferries or even submarines, 145 energy recapture systems or hybrid systems.The advantage is usually lower weight and better efficiency in comparison to i.e. lead-acid systems. 146As in the case of the EVs and BESS, there is a possibility of faults, incidents, fires and thus pollution of water ecosystems.An example of such an incident is a fire and subsequent gas explosion on board a diesel-electric ferry in Norway on the 10th of October 2019 147 -the fire was spotted during the crossing, but the ferry returned to the harbour under its power where everyone was safely evacuated.Overnight, however, a serious gas explosion rocked the battery room causing significant damage.It was reported that twelve firefighters were hospitalized due to exposure to hazardous gases associated with the burning batteries.Table 2 also shows some other known up-to-date incidents.
Fire extinguishing agents (water, foam etc.) used to mitigate the incidents will be a carrier to all pollutants released from LIBs.That can broader the area of the impact.For instance, HF released during a fire may form a solution with the fire extinguishing agent and thus, at concentrations as low as 0.01%, could cause injury to the more sensitive areas of human skin following exposures as short as 5 min. 148If we add that between 20 and 200 mg of HF could be released per W h of an EV battery pack, 22 such solution or even higher concentrations may be of no surprise.This reaction should be also taken into account when designing and using any extinguishing systems in places dealing with larger amounts of LIBs.The first responders must also be aware of the additional hazards, thus proper personal protective equipment (PPE) and operating procedures should be in place.
When the fire is gone there are left carcasses of burnt batteries.Except probably some steel parts (pack cover, module case etc.) they would not represent any value for recycling thus probably discarded/landfilled.These carcasses may be treated as fire debris according to Fig. 2. Currently, there are no data directly revealing what kind of pollutants may be released from such waste.Given the typical post-fire analysis 132,149 and the composition of LIBs, we can suspect these may be remains of nanooxides, metals, PAHs and VOCs.

Sources and pollution pathways
Spent LIBs are considered hazardous wastes (especially those from EVs) due to the potential environmental and human  Testing a new 1.5 MW BESS linked to a solar energy system when it was destroyed by fire.Subsequent investigation concluded that lack of ventilation and inadequate monitoring were the major causes of the incident and recommended improved ventilation, 24/7 monitoring and the ability to send remote alarms.a Wisconsin, USA August 2016 The damage is estimated up to $4M.The fire started in the DC power and control system rather than the lithium ion batteries.b Drogenbos, Belgium November 2017 Fire badly damaged the container of a 1 MW BESS after the fire detection and control system failed to control the blaze.c Various places across South Korea 2018-2019 Up to 30 fires involving LG Chem lithium ion BESS.The causes of these fires have not been made public as yet, but defective lithium ion cells were not believed to be a major contributing factor.The concatenation of multiple factors have been postulated.d Surprise Arizona April 2019 Vapour cloud explosion.Four first responders were injured, 2 very badly.9 first responders contaminated with hydrogen cyanide.e Nathan campus of Griffiths University, Brisbane, Australia March 2020 LFP BESS exploded causing major damage to buildings.The cause of the fire and explosion has been stated as having been an internal short circuit.f Carnegie Road, Merseyside, UK September 2020 One of three containers exploded in the early hours of 15 September, followed by a fire.The ventilation units were blown off the roof of the container by the blast, with debris (including one of the doors) scattered across a wide area.The cause of the explosion remains under investigation.g Perles-et-Castelet, Arie `ge France December 2020 A BESS containing 60 ''batteries'' ignited.The container was situated on top of a second container and presented some difficulties to the fire service, which had to pull the burning container down.The cause of the fire is under investigation.The explosion occurred on board a pleasure craft whilst moored in the harbour.A fireball was observed above the boat by local residents.p Russian nuclear submarine Losharik, Barents Sea July 2019 An electrical short circuit occurred while the nuclear submarine was docking with the Podmoskovye, its mothership.This caused lithium ion cells to go into thermal runaway resulting in explosions and fire, killing 14 high-ranking officers on board.q Hybrid electric ferry Ytterøyningen, Norway October 2019 Coolant leaked from the 1989 kW h BESS on the vessel causing a small fire.This was extinguished by the seawater sprinkler system.However, the seawater then caused short circuits in the battery which exploded on the following day.12 firefighters were hospitalized due to exposure to hazardous gases.Firefighter' knocked on his back' in fire blast at Griffith University.Brisbane, Australia: Brisbane Times; 2020.g https://www.energy-storage.news/news/fire-at-20mw-uk-battery-storage-plant-in-liverpool.h https://www.ouest-france.fr/region-occitanie/ariege/ariege-un-incendie-dans-un-container-debatteries-au-lithium-une-quarantaine-de-pompiers-mobilises-7070012.i https://www.datacenterdynamics.com/en/news/ovh-fire-update-four-halls-sbg1-destroyed-well-all-sbg2/.j https://www.datacenterknowledge.com/industry-perspectives/give-ovh-break-and-use-data-center-fire-teachable-moment.k Fire at Solar Plant in Hongseong Involves EV Batteries from LG Energy Solution -Businesskorea.l https://www.pv-magazine.com/2021/04/21/two-firefighters-killed-and-one-missing-after-beijing-battery-blaze/.m https://www.shawlocal.com/mywebtimes/news/local/2021/07/19/no-evacuationsfor-battery-fire-at-energy-storage-facility-east-of-grand-ridge/.n https://www.theage.com.au/national/victoria/blaze-at-tesla-big-battery-extinguished-after-three-daybattle-for-control-20210802-p58f6x.html.o https://www.professionalmariner.com/battery-related-fire-damages-famed-hybrid-tug-puts-it-out-of-service/.p https://112-water.nl/inzetten-verslagen/201-brand-aan-boord-bij-hybride-boot.q https://www.wsws.org/en/articles/2019/07/12/russ-j12.html.r https:// gcaptain.com/fire-and-gas-explosion-in-battery-room-of-norwegian-ferry-prompts-lithium-ion-power-warning/.s https://maritime-executive.com/article/newly-built-hybrid-passenger-vessel-suffers-battery-fire-under-way.
1][152][153][154] The disposal and processing of LIBs, as well as their properties (e.g.chemistries), will have a significant impact on various environmental compartments (Fig. 3).LIBs contain a variety of chemicals including reactive salts, volatile organic electrolytes and additives: the latter are often commercial secrets and hence their toxicity and combustion products are largely unknown. 155,156oreover, battery fires, in combination with biogas from landfills, may release toxins into the air or leach the harmful contents into the soil, groundwater and surface water. 118,157,158][161][162][163][164][165] Fig. 3 shows likely pollution routes (originated from EoL LIBs disposal/processing) to the environment and possible cross-contamination within different compartments.Pollutants once released from LIBs may enter one compartment first and later be transported to another one.For instance, heavy metals may be collected by dumpsite leachate and transported into the soil or groundwater outside the initial site.On the other hand, if a battery leaches onto open ground, heavy metals may run-off with the rain to a nearby river or lake.However, due to knowledge and data gaps, not all routes could be fully reviewed in this work.

Emissions to air
3.1.1Dust.Fine particles may be released from LIBs to the air during disassembly and recycling processes; from a landfill or dumpsite by degradation or fire/explosion; and are considered as a part of the total dust emissions. 166Generally, dust is a mixture of solid particles and liquid droplets of a wide range of sizes (nanometres to few micrometres) and chemical composition, all together suspended in the atmosphere. 167,168][177][178][179][180][181][182][183][184][185][186][187] During disassembly and material recovery of LIBs, shredding is the main mechanical processing option 25 that could generate dust emissions.These emissions could potentially cause a respiratory hazard to workers in recycling centres as well as those living or working in the vicinity.Moreover, particles and chemicals (e.g.PAHs) released from batteries may aggregate together in the atmosphere, be transported on larger distances and settled down causing for example soil pollution.As LIBs particles may be of different materials and sizes, it is mandatory to assess and investigate their possible toxicity and respiratory hazard.
The potential negative effect of three battery materials: lithium iron phosphate (LFP), lithium titanium oxide (LTO) and lithium cobalt oxide (LCO) was studied utilizing mouse bioassays. 188The mixed metal oxides present in the cathodes of LIBs could release particles small enough to penetrate the lungs and induce inflammation.The extent of the impact varies depending on the chemistry of the LIB.From the materials tested by Sironval et al. (2018), LTO was the least harmful, whereas LCO induced the strongest inflammation.As a result, the fibrotic responses were observed in cells with higher inflammation rates. 188nother generator of dust and particles are explosions and fires.During thermal runaway the particles are first ejected from the safety vent together with other gases, resulting in heavy smoke with black colour. 189,190After cooling down, they settled as surface dust.The composition of these powders from lithium nickel cobalt manganese (NMC) cells was mainly C, O, Al, Mn, and F. This result suggests that it is mostly the negative carbon particles of the battery, as well as the oxidized positive substances of the electrode, that are ejected during gas venting. 190Another study 191 found that thermally abused prismatic automotive NMC cells released PM that contained heavy metals.The authors found that nickel and copper were detected in all collected fractions (1.45 to 500 mm) whereas zinc and chromium settled with bigger fractions (50 to 500 mm).Ni had the largest mass percentage followed by Cu, Zn and Cr.In terms of the total emitted quantity, PM accounted for 1.7% of the cell mass.Once settled, such particulates may interact with surrounding waters and soil and ultimately be of threat to crops.
3.1.2Hydrofluoric acid.Out of various hazardous gases released from LIBs to air, HF is of the highest concern and must be treated with high caution.It can enter the human body via the skin or respiratory systems and causes severe corrosive effects and systemic toxicity. 192Inhalation of just a few ppm of HF can result in serious toxic effects.HF readily penetrates the skin and moves quickly to the deeper tissue layers where it releases the freely dissociable fluoride ion.This ion is extremely toxic, due to its strong reactivity. 193Moreover, HF gas is hygroscopic and readily soluble in water. 158,194,195Concentrated solutions of HF are highly corrosive to the skin and underlying tissues and accidental dermal exposure has been reported to cause death in humans. 148,196][199][200] The acid can also be produced due to accidental spillage and hence the exposure of the LiPF 6 in the electrolyte to humid air 158 according to: although this is disputed. 201s well as the direct reaction of LiPF 6 with water, at ambient temperature in the organic carbonate solvents employed, LiPF 6 exists in equilibrium with its decomposition products according to: PF 5 is prone to degradation upon exposure to traces of water, moisture or alcohols 202 to form HF which is also highly toxic: 158 To mitigate the effect of reaction (3), commercial LIB electrolytes often contain additives, up to 5% by mass or volume, some of which suppress PF 5 formation.Others are added to protect against overcharge, provide fire-retarding protection or improve Solid Electrolyte Interphase (SEI) formation or quality. 203It is not clear what effect these may have on reaction (2) or other decomposition reactions, or their toxicity or the identity of the products from their combustion.
The second likely source of HF from LIBs is a thermal runaway and fire.The amount of HF produced during an incident depends on the size of the cell and its chemistry 21,23,142,[204][205][206] and may incorporate endothermal decomposition of the binderpolyvinylidene fluoride (PVDF). 159turk et al. (2015) detected significant quantities of HF from burning NMC and LFP type pouch cells. 141Surprisingly, LFP cells produced the highest levels of hydrofluoric acid fumes.So far this type of chemistry was regarded as one of the safest as having the highest cathode decomposition temperature; 135 however, the cells employed in the LIBESS that exploded at Griffiths University in March 2020 were LFP cells. 207arsson et al. (2017) have estimated that between 20 and 200 mg of HF could be released per W h of an EV battery pack. 22hus, the amount of HF could be greater than 80-800 times the US National Institute for Occupational Safety and Health Immediate Danger to Life or Health (IDLH) level. 208.1.3Vapours and gases.It is now apparent that thermal runaway generates a white vapour during venting the cells, essentially independent of chemistry, form factor and manufacturer.139 This white vapour contains H 2 , SO 2 , NO 2 , HF, HCl, CO, CO 2 , droplets of organic solvent (giving the white colour) and a large range of small chain alkanes and alkenes.7,[20][21][22][23][24] The exact composition of the white vapour appears to depend on the State of Charge (SoC) and battery chemistry.20,142,209,210 The white vapour is produced when cell is venting as a result of various chemical processes, triggered at successively higher temperatures.135 There is insufficient oxygen produced when the cathode structure collapses to sustain fire, but when the white vapour vents from the cells it may or may not ignite: at SoC 4 50% and if sufficient oxygen is present, the vapour can ignite to produce jet-like flames.144 However, if there is insufficient oxygen, e.g.inside a LIBESS with air displaced by the vapour and/or a suppressant, 138 the vapour represents a wholly novel hazard concerning LIBs; if the white vapour contacts an ignition source, there could be a possibility of a flash fire, fireballs developing, or in extreme cases even a vapour cloud explosion.211

Emissions to soil and water
Once physically disposed of, LIBs will leach their contents into the surrounding soils, infiltrate to deeper layers, pollute groundwater and can run-off into surface waters.As water is very often a carrier of pollutants in the soil, it is challenging to ascertain the individual impact of LIBs on these two spheres alone, and hence we discuss the impact on them jointly.
3.2.1 Landfill/dump leachate.One of the most important hazards associated with landfilling and illegal processing is the generation of leachate: it is formed as a result of various biological and chemical degradation processes, and rain percolating through waste.][214] Leachate could impose a serious environmental impact as it can act as a transportation medium for hazardous species.It may pollute soils or groundwater and transport contaminates over considerable distances, often many kilometres from the site [215][216][217][218][219] potentially contaminating groundwater. 212IB leachates will potentially carry various pollutants such as heavy metals, additives, electrolyte degradation products but also dissolved gases.The latter, such as HF, HCl or SO 2 (in form of sulphuric acid) not only have a straightforward toxic effect but also alter the properties of receiving waters and solutions in soil.The acidification caused by these corrosive species may have the same effect as ''acid rain'', lowering the pH and thus strongly impacting plants and animals. 220They would also enhance the release of soil-bound heavy metals.Under more acidic conditions the actual toxicity of heavy metals is higher than it would be predicted from the metal concentration alone. 221.2.2Heavy metals.The content of metals in batteries depends upon their design and size (i.e.cylindrical, prismatic or pouch cells), as well as their chemistry (i.e.NMC, LCO etc.).Typically they are in the range of 0.05-0.37kg Co per kW h, 0.25-0.86kg Ni per kW h and 0.46-0.9][224][225] These figures translate to tens of kilograms of such materials in EV packs that potentially may be released to the environment if improperly disposed of, representing a series of threats to human health and the environment.
7][228] Heavy metals polluting crops, fruits or dust are known to induce genetic damage in children 229,230 and cause lower academic performance. 231

Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2021 Nickel released into the environment binds strongly to small solid particles, especially containing iron or manganese.Ni has a low tendency to accumulate in fish and small animals.With respect to human health, nickel is the most common allergic metal.It may also cause respiratory disorders and even cancer at higher concentrations. 232Lithium has relatively low toxicity and small dosages are beneficial for various trophic levels of organisms. 233Nevertheless, it is quite an active element and, in higher concentrations, may cause digestive and neurologic systems disorders. 234,235Small dosages of cobalt are beneficial to humans (synthesis of vitamin B12) but larger dosages cause respiratory problems and skin rashes. 236In general, adverse reactions to Co exposure depend primarily on its chemical form. 237he leachate from LIBs penetrating through landfill or illegal dumpsites will collect and carry various metals, including lithium, cobalt, nickel, manganese, iron, chromium and copper. 1559][240][241] Metals found in leachates (both real and artificial) that exceeded toxicity limits were lead, mercury, cobalt, copper, nickel, chromium and thallium -depending on the type of test used.One of the studies 242 analysed the leachate from a landfill containing NMC batteries: the authors found that less than 4% of the total cobalt, nickel, aluminium, copper, and iron from the battery were in solution, whilst 11.45% of the manganese and 42.50% of the lithium were present.
It is unlikely that metals will be released only in metallic or ionic form from buried LIBs: they will likely be mobilized as nanoparticles (NPs) as well.Such nanomaterials (mostly of cathode origin) and their ecotoxicological potential towards various organisms were assessed by several studies.Thus Hang et al. (2017) studied the impact of NMC on the soil bacteria Shewanella oneidensis MR-1 and found that exposure to 5 mg L À1 NMC significantly impaired bacterial population growth and respiration. 243The authors showed that these effects were due to the release of metal ions from the NMC, with the largest effects arising from Ni 2+ and Co 2+ species.Gunsolus et al. (2017) also investigated the toxicity of NMC towards the same bacteria, varying the relative composition of the metals in the mixed metal oxide. 244The authors found that lowering the Ni and Co content and increasing the Mn content, lowered the biological impact significantly.In fact, reducing the cobalt content and increasing Ni has been a key aspect of NMC battery development over the last few years, driven by safety considerations 245,246 and global supply chain issues. 247,248Another study assessed the biological impact of NMC on a Gram-positive model bacterium, Bacillus subtilis. 249rowth and cellular respiration of the bacteria were inhibited significantly by free Ni 2+ and Co 2+ ions released from the incongruent dissolution of the NMC.DNA damage tests at the single-cell level confirmed that the toxicity caused by the redesigned NMC (enriched in Mn) was lower in comparison to solutions containing either free Ni and Co or original NMC material.A recent paper 250 reported a study on the influence of multiphase lithiated cobalt phosphate (mLCP) nanomaterial on Shewanella oneidensis MR-1 and Bacillus subtilis.The authors found that mLCP was toxic to both species due to the release of cobalt ions.
The toxicity of NMC and LCO materials towards eukaryotic organisms was studied using Daphnia magna. 251The work showed that there is negligible acute toxicity to daphnid at the highest concentrations of 25 mg L À1 , whereas there is a significant impact on daphnid reproduction and survival during chronic exposure (21 days) at concentrations of 0.25 mg L À1 for LCO and 1.0 mg L À1 for NMC.They confirmed that toxicity depends on chemical composition -the replacement of Co by Ni and Mn resulted in better daphnid survival, reproduction, and body size.Finally, the authors highlighted there was a strong nanomaterial-specific impact -probably due to adhesion. 252lthough, the effect could not be fully confirmed by experiments with just free Li, Ni, Mn, and Co ions dissolved in suspension metal.
LCO and NMC have been shown to have adverse effects on the benthic invertebrate Chironomus riparius 253 by inhibiting growth and development.The increased toxicity of these materials was associated with quick sedimentation of nanoparticles and thus faster interaction with benthic organisms.As in previous studies, a reduction in the biological impact of NMC versus LCO was observed.
On the higher trophic level, the fish class, there is an evident impact of LCO on rainbow trout gill epithelial cells. 254It was found that exposure to Li + and Co 2+ alone, did not reduce cell viability, however strong toxic effects were observed for LCO nanoparticles and these were found to increase the concentration of reactive oxygen species (ROS).Overproduction of ROS may cause biochemical damage, overstress to cells and ultimately, their damage. 255These negative effects were measured at sub-toxic concentrations of LCO which are close to predicted environmental levels.Finally, it was found that the nanomaterials were internalized in cellular organelles, which was not the case for bacteria. 243A subsequent paper 256 reported studies on the response of the same type of cells to LCO-induced oxidative stress.The authors postulated a ''two-hit'' model for LCO toxicity.Initially, the intact LCO material induces high levels of ROS leading to gradual engagement of stress response genes (''first hit'').In the next step, the gradual release of metal ions from NPs starts to suppress the expression of these genes, disrupting the cell response to elevated ROS levels (''second hit'').This mode of action ultimately confuses the cell response to stress and may lead to severe damage.
The take-home message from all ecotoxicity studies is that the replacement of toxic metals (Co and Ni) with less toxic ones (Mn) may increase the environmental compatibility and sustainability of the next generation of LIB materials.
Heavy metals from LIBs may also enter the natural environment due to materials recovery processes, as well as legal and illegal disposal routes.Due to the lack of valid data concerning the recycling of LIBs, for analytical purposes, we can consider other well-established processing of automotive batteries.Critical analysis of the lead-acid batteries recovery reveals that large quantities of potentially toxic slag have to be dealt with.In most countries, such wastes are often disposed of in waste dumps that may lack properly designed engineered landfills, especially in emerging economies.Analysis of slag revealed that apart from high pH (12.22) it contained also high concentrations (g kg À1 ) of Pb (4101), Cu (42.5),Cr (41.2),Zn (0.5), Cd (40.5) and Ni (0.1), indicating a poor metal recovery in the recycling processes. 257The authors suggested that there was a greater risk of release of these heavy metals over the long-term following disposal into the environment.Fujimori et al. (2016) found elevated levels of Pb in some China soils located next to recycling plants. 258In the case of LIB recycling, we also have to deal with the slag resulting from the pyrometallurgical route, mainly containing Mn, Li, and Al. 259.2.3Battery degradation products.The study undertaken by Gru ¨tzke et al. (2015) may give indications of the fate of processed battery material when kept in storage (i.e.prior to recycling) or landfilled.59 The authors shredded NMC LIBs from EVs and enclosed them in sealed tinplate and plastic cans.Their main goal was to assess which compounds are released from spent, processed material, and if there is any further degradation of original compounds, and the possible emergence of hazardous transformation products.The so-called ''black mass'' was monitored for over 20 months with regular analyses of gases, liquids and remaining salts.Their results showed that the tinplate cans underwent fast corrosion -first spots were visible after a few days, and after five months some of the cans were already perforated by rusting.Plastic cans did not show any visual signs of degradation.In addition to HF, the authors found traces of phosphoric acids, LiPF 6 , other additives (i.e.cyclohexylbenzene) and solvents -DMC, EC, EMC and their degradation products.More importantly, they identified several alkylfluorophosphates, including dimethyl fluorophosphate (DMFP) and diethylfluorophosphate (DEFP). Alylfluorophosphates are of concern as they have similar structures to the chemical warfare agents such as sarin.260 DMFP and DEFP were also detected by Kraft et al. (2015 and 2016) who investigated the thermal degradation of LIB electrolytes.160,261 Although present in manageable low concentrations, they still may have toxic potential both to humans and the environment.
These studies highlight the important issue of storing spent and processed battery materials.If real-life scenarios are envisaged, such as temporary storage in the canister or landfilling, certainly, the corrosion of containers or battery metal case (module or pack) will be speeded up by external environmental factors such as leachate or humid air.It also indicates that spent degraded battery already contains harmful contentthese products may enter the air or be transported with the solution to other compartments.
3.2.4Additives.Finally, LIBs may also leach the remains of electrolyte and its additives. 262,263The most common additives are polymers, Lewis acids, sulphur-containing and phosphorous containing additives, polyfluoroalkyl substituted ethylene carbonates or ionic liquids (ILs).The latter, ILs, are particularly interesting as they can play many roles in the battery system: as a pure electrolyte; in mixture with conventional solvents; as a gel polymer electrolyte or as a binder. 264,265ILs are non-volatile and a nonflammable class of organic salts consisting of an organic cation and an organic or inorganic anion and there is a broad spectrum of cations (i.e.imidazolium, pyridinium, choline etc.) and anions (organic and inorganic). 266Therefore, there is a very broad spectrum of the environmental impact of these compounds depending on their structure.However, due to their properties, the routes of exposure and potential hazards are limited to interactions with solutions.The distribution of ionic liquids in the environment is governed by their structure e.g.ILs with longer alkyl chains absorb more easily on the soil surface.Therefore, some ionic liquids may be quite mobile in water or soil columns whereas others may bind strongly to various environmental compartments and may become persistent pollutants.8][269][270][271][272][273][274][275] Although their use is usually restricted to small quantities, ILs could still have considerable negative effects on wildlife and human health.For instance, most of the ionic liquids in battery systems are considered toxic and poorly biodegradable compounds. 276,277s in the case of the mobility and transport in environmental media, the toxicity of ILs depends on the lipophilicity and alkyl chain length, the direct interaction with the biological system (membrane, cell or organism) and trophic levels (bacteria, plants, invertebrates or human cells).[280][281][282][283][284][285][286][287]

Recommendations for future studies
Table 3 illustrates the possible routes of emission and pollutants released from LIBs.Although partially supported by the existing literature (see Sections 2 and 3), this is more of an indicative list.There are lots of unknowns, incomplete data, not yet researched specific topics or even contradictory results that need to be clarified to mitigate any negative impact of spent batteries.
Therefore, we identified some of the main knowledge gaps and probe the following questions on the environmental impacts of spent LIBs that might help to manage these better in the future: (1) What are the current and prospective volumes of spent LIBs?
(2) How much spent batteries reach the relevant disposal stream?
(3) Where spent batteries will be processed/disposed of/abandoned?(4) In what form: as a whole, partitioned, shredded etc. LIBs will be processed/disposed of/abandoned?
(5) What is the impact of changing chemistries on the waste streams in terms of needed technology but also in terms of the life span?Although, some of these questions could be partially answered already based on approximations i.e. prospective volumes are estimated on sales number; but there is an urgent need for reliable and verified data.Obtaining reliable and accurate data will have serious implications for the safety of the recycling and disposal facilities and helps to shape, implement and regulate appropriate practices and legislations.The real data should be supported by various Life Cycle Assessments methodology, cost-benefit studies and the use of more modelling in the prediction of i.e. life cycle of the specific battery in a given application.
As the growing demand for new LIBs will result in increased numbers of spent batteries, we recommend that this quantity must be appropriately managed and controlled across the various disposal routes.The best-case scenario assumes high collection rates of spent batteries with an almost full recovery of materials that can be reused (boosting the circular economy).It is evident that domestic i.e. decentralised recycling will be more profitable than centralised recycling, however, it will take time for the industry to scale up collection, processes and infrastructures. 6It is also argued that latecomers will be threatened by the existing industry that may benefit more quickly from the larger economies of scale.Decentralised systems do require a substantial amount of spent LIBs (especially large EV) to reach economy of scale as it was estimated that recycling costs will decrease if between 1000 and 15 000 tonnes per year are recycled at one plant. 6Collection of spent and ''second-life'' large batteries, might be linked with some business model e.g.sale of the car but the lease of battery -then the responsibility is on dealer/seller or can be governed by local councils with i.e. specialized collection centre -responsibility on the user.To improve the collection and later decision on the re-use, recycling or landfilling knowing the usage history of ''first life'' of EV/BESS batteries would be of great use -the battery's condition can be determined much faster and more economically.That will speed up the streamlining of LIBs to an accurate route of processing.However, we must bear in mind that modules in the pack can age differently 288 -so possible that the selection will be on the modules not pack level.Nevertheless, improving the reuse of the materials could reduce the environmental, economic and social burden of the existing battery supply chain including disposal practices is inevitable.However, the challenges in the collection, dismantling and recycling of LIBs must be overcome.It is of crucial importance to develop innovative technologies to lower the cost of material recovery and reduce the environmental impacts of this industry. 6Furthermore, illegal disposal and informal processing that leads to serious pollution must be prevented.This could be achieved through better collection schemes, expansions, and improvements in the current recycling infrastructure and posing legal obstacles for exporting second-hand EVs or LIBs.The last anybody wants is that LIBs to become, for instance, a source of soil pollution by heavy metals.Moreover, there should be a constant drive to produce LiBs containing less hazardous and more ''green'' materials 289 that not only improve the capacity and efficiency of the battery but also prevent destructive behaviour and have minimal negative impact on the environment if improperly disposed of. 290New LIBs should be designed for recycling meaning the materials can be easy separate from each other and able to be re-used in new batteries (direct recycling).That links to new technologies of material recovery that are less energy intensive, cheaper and produce less or no secondary pollution.The improvements must be guided by appropriate legislation, regulations and management structures.These must encourage customers and manufacturers to properly handle spent batteries, highlighting the benefits of recycling (in light of the obligations for economic operators as regards the sourcing of raw materials) and promote the circular economy model and finally, must also clearly ensure where each liability lays. 48inally, we must address the problem of penetration of large ''2nd life'' LIBs into society.There is a general perception, particularly in Europe, that the re-use (using an EV battery without change in an EV), remanufacture (using an EV battery after replacing defective modules in an EV) and repurposing (using modules from an EV at end-of-life to assemble a battery for a purpose other than traction, e.g.stationary storage) of LIBs can make a positive contribution to the decarbonization of the planet. 48The safety of new cells, modules and battery packs can be assessed using the ''type tests'' that form the basis of all codes, standards and regulations governing, for example, domestic and industrial battery energy storage systems.Type testing relies on testing a sample of cells from a batch, usually to destruction, and accepting the result as representative of the batch.The use of type testing on new cells is valid due to the extremely tight quality control of their manufacture, however, such type testing is now accepted as inappropriate to assess the safety of second life LIBs due to the wide variation in SoH across an EV pack during its life and possible exposure to extremes of temperature, overcharging and/or charging at high currents, [291][292][293] all of which can increase the potential for thermal runaway. 294,295The draft standards IEC63330 296 (Requirements for reuse of secondary batteries) and IEC63338 297 (General guidance for reuse of secondary cells and batteries) fully acknowledge the fact that type tests cannot be employed to assess the safety of second life LiBs.The key challenge is determining the SoH, or more correctly, the State-of-Safety of second life LIBs. 298part of the latter must be full knowledge of the first life of the LIB in the EV, including full details of any abuse (e.g.overcharging) maximum charge and discharge currents & operational temperatures.The draft standards IEC63330 and IEC63338 were intended to address this problem and hence facilitate the safe application of second life LiBs, but have failed to do so, as they do not require any testing and rely completely on the original manufacturers of the EVs passing on full BMS data (and hence potentially valuable intellectual property) to remanufacturers & repurposers, which is perceived as highly unlikely.The draft EU Batteries Regulation also sidesteps the safety issue of second-life batteries by requiring only that BMS data is made fully available.
It is generally accepted that the re-use, remanufacturing and repurposing of complete packs where the original EV manufacturer is retained in the supply chain is of low risk. 299owever, second life cells, modules and even full battery packs are freely available from online retailers, such that inexperienced organisations and even the general public can repurpose LIBs.Thus, a major concern is the use of these suppliers by homeowners to build do-it-yourself domestic BESS: domestic battery energy storage systems are currently completely unregulated in the UK with respect to the application of lithium-ion batteries.Concerns over the hazards posed by such systems in the home environment have been addressed in other countries: thus the US NFPA 855 standard 300 and the draft DR2 AS/NZS 5139 : 2019 Australian and New Zealand standard 301 do not permit domestic lithium-ion BESS inside the home, and NFPA 855 does not permit do-it-yourself lithium-ion BESS at all.
The online trade-in 2nd life LIBs is completely unregulated, and delivery of these devices is often offered by means illegal under UN 38.3 302 as they are classified as dangerous goods and hence require transportation in accordance with the European agreement ''Accord europe ´en relatif au transport international des marchandises dangereuses par route'' (ADR) requirements. 303

Conclusion
There is no doubt that the urgent need to decarbonise transportation puts LIBs at the forefront of the action.However, the growing stream of spent LIBs would impose an enormous threat to the natural environment and human health, as batteries contain hazardous materials.In this review, the current, possible and likely waste management practices of LIBs were identified -from collection and recycling to landfilling, through the EoL incidents up to illegal disposal.Currently, landfilling is the most common practice but there is a growing share of recycling.The current review also assessed the most likely hazards and incidents during each EoL practice.The fire and explosion incidents are currently the most common events that have been evidenced by real-life incidents.Leaching is another pollution pathway that will co-dominate in the future.Identified hazards released from LIBs contain vapours and gases (i.e.HF, CO or HCN), metal nanooxides (i.e.LMO, NMC), degradation products of the electrolyte This is © The Royal Society of Chemistry 2021 (alkylfluorophosphates) and possible traces of additives.These pollutants may be released to soil, water (groundwater) and air, depending on recycling, disposal method or abuse incident.Released pollutants may pose a serious threat to wildlife and humans with often immediate effects like in the case of contact with HF during EV fire.Degradation of the battery content (especially electrolyte) in some cases may lead to the emergence of chemicals structurally similar to chemical warfare agents.The initial studies on the (eco)toxicity of the cathode nanomaterials showed that LIBs may pose a threat to living organisms and human health.But the replacement of toxic metals (Co and Ni) with less toxic ones (Mn) may increase the environmental compatibility and sustainability of the next generation of lithium battery materials.Finally, we identified several urgent knowledge gaps that need to be covered to mitigate the negative impact on the environment of LIBs, and recommended some actions to tackle these issues.
(EP/S003053/1) as part of its ''Recycling of Li-Ion Batteries (ReLIB)'' project (FIRG005) ''SafeBatt -Science of Battery Safety'' (FIRG028).The funders had no input or role to play in the conceptualization, design, literature collection, analysis, decision to publish or preparation of the study of this manuscript.
h SBG-2 OVHcloud data centre in Strasbourg, France March 2021 One data centre was completely destroyed and 4 of the 12 rooms of a second centre were also destroyed.Unconfirmed reports that fire started in BESS.ij Hongseong, South Chungcheong Province, China April 2021 Fire in the BESS supporting a solar energy facility, supplied by LG Energy Solution (LGES).The fire destroyed the facility causing ca.440 M Won damage k .Shopping mall in Beijing, China April 2021 Explosion and fire in a 25 MW h LFP BESS situated on the roof of a shopping mall.The BESS supported a solar array. 2 firefighters were killed, 1 injured.The BESS consisted of two sections, North and South: whilst dealing with a fire in the Southern section and explosion occurred in the Northern section, killing the firefighters.l Invenergy storage facility, LaSalle County, USA July 2021 Fire in one of the 18 containers at a LFP BESS near the La Salle Generation Station nuclear power plant.m ''Victorian Big Battery'', Moorabool, Geelong, Australia July/August 2021 The Moorabool installation is the largest LiBESS in Australia.The fire started in one container and spread to a second, paired container which was adjacent to the first.The firefighters adopted a defensive strategy due to the challenges of the installation, and allowed the containers to burn out in a controlled manner over 3 days.n Maritime incidents Place Date Comments Hybrid tug Campbell Foss, USA November 2012 Fire in BESS of the vessel after it had finished assisting a second vessel in the port of Long Beach.The fire started in the battery compartment and spread to the engine room via a ventilation duct.The vessel's engineer was hospitalized.The incident has been attributed to long term overcharging of the battery.o Boat Carila Sis, Sneek harbour, Netherlands May 2019

( 6 )
What, how, where and in which volumes are hazards released from the spent batteries?(7)What is the prevalence and distribution of battery pollutants once being released into the natural environment?Energy & Environmental Science Review (8) What is the circulation and interactions of battery pollutants among the land-water-air-emission pathways and the wildlife nexus?(9)What is the (eco)toxicity and (bio)accumulation of LIBs materials to various organisms and humans?(10) How should or would environmental studies support the design and disposal of spent LIBs?

Table 1
Current recycling facilities around the world adopted from ref.49 and 50 a P -pyrometallurgy, H -hydrometallurgy, M -physical separation.

Table 2
List of some major involving LIBESS and Maritime incidents

Table 3
Possible pollutants and routes of their emission released from LiBs e. C 2 H 4 ; CH 3 COCHO etc.) -Ionic liquids -Unknown additives/ degradation products