You are on page 1of 10

Resources, Conservation & Recycling 133 (2018) 76–85

Contents lists available at ScienceDirect

Resources, Conservation & Recycling


journal homepage: www.elsevier.com/locate/resconrec

Full length article

Modeling the potential impact of lithium recycling from EV batteries on T


lithium demand: A dynamic MFA approach

Saskia Ziemanna, , Daniel B. Müllerb, Liselotte Schebekc, Marcel Weila,d
a
Karlsruhe Institute of Technology (KIT), P.O. Box 3640, 76021 Karlsruhe, Germany
b
Industrial Ecology Programme (IndEcol), Department of Hydraulic and Environmental Engineering, Norwegian University of Science and Technology (NTNU), S.P.
Andersens veg 5, NO-7491 Trondheim, Norway
c
Technical University Darmstadt, Faculty of Civil and Environmental Engineering, Institute IWAR, Material Flow Management and Resource Economy, Franziska-Braun-
Straße 7, 64287 Darmstadt, Germany
d
Helmholtz Institute Ulm for Electrochemical Energy Storage, (HIU) Albert-Einstein-Allee 11, 89081 Ulm, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Electric mobility is a key element in the transition to a more sustainable transport system. Already today Li-ion
Electric mobility batteries (LIB) are used in many stationary and mobile applications. If electric vehicles (EV) based on LIB reach a
Lithium-ion batteries strong market penetration the recycling of lithium from growing numbers of automotive batteries will be de-
Automotive battery recycling cisive for future availability of this resource. However, the amount that EV battery recycling can contribute to
Lithium recovery
lithium availability is unclear since the ability of secondary lithium to be reused in batteries is uncertain. Here
Scrap surplus
Material flow analysis
we assess the trend in demand for lithium and the possible effects of automotive battery recycling by applying
dynamic material flow analysis to the global lithium cycle. We found that lithium recovery from EV battery
recycling could result in a significant oversupply of secondary material if its quality is not high enough to allow
for reprocessing in battery production. In this case the application of secondary lithium would be limited to other
use sectors and thus also restrict the possible savings of this resource. If secondary lithium only substitutes a
relatively small quantity of virgin material whereas total lithium demand keeps rising, the production of primary
raw materials will have to increase strongly.
The findings of our work reveal the challenge that might result from a quality problem of recycled battery
materials and thereby emphasize the importance of developing cost-effective recycling technologies with effi-
cient material recovery processes for traction batteries as well as the timely development of a functioning re-
cycling infrastructure.

1. Introduction Angerer et al., 2009; Roskill, 2013; Weil et al., 2009; Tahil, 2008). The
majority of these studies have determined there is sufficient lithium to
Electric vehicles (EV) can be an alternative to cars using fossil fuel meet the possible future demand for EV in the coming decades. How-
and can play an important role in a more sustainable development of ever, some of these analyses also detected a potential premature de-
the transport sector. In scenarios featuring increasing electric mobility, pletion of current lithium reserves (Pehlken et al., 2017; Weil and
the use of traction batteries such as lithium-ion batteries (LIB) is ex- Ziemann, 2014, Gruber et al., 2011; IEA, 2009; Tahil, 2008). Specifi-
pected to grow significantly, raising the demand for the component cally, there is a considerable risk of a shortage due to the concentration
materials (Simon et al., 2015; Weil and Ziemann, 2014; Navigant, 2013; of lithium deposits in certain regions and countries characterized by
Buchert et al., 2011). political instability such as Bolivia and China (DERA, 2014; Worldbank,
As a contribution to the ongoing discussion of future lithium 2011). Due to the risk of supply chain disruptions, the status of a sig-
availability, several studies have analyzed the impact of EV penetration nificant part of the lithium reserves can be classified as critical (cf.
on lithium demand and compared it with data on lithium reserves and Fig. 1, see also Oliveira et al., 2015; Weil and Ziemann, 2014; Grosjean
resources (Pehlken et al., 2017; Weil and Ziemann, 2014; Vikström et al., 2012).
et al., 2013; Mohr et al., 2012; Kushnir and Sandén, 2012; Gruber et al., Recycling is an important measure to mitigate potential supply risks
2011; Gaines and Nelson, 2010; IEA, 2009; Yaksic and Tilton, 2009; and also to reduce the demand for primary raw materials (Hagelüken,


Corresponding author.
E-mail addresses: saskia.ziemann@kit.edu (S. Ziemann), Daniel.mueller@ntnu.no (D.B. Müller), l.schebek@iwar.tu-darmstadt.de (L. Schebek), marcel.weil@kit.edu (M. Weil).

https://doi.org/10.1016/j.resconrec.2018.01.031
Received 14 July 2017; Received in revised form 28 January 2018; Accepted 29 January 2018
Available online 22 February 2018
0921-3449/ © 2018 Elsevier B.V. All rights reserved.
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

Fig. 1. Dynamic model with lithium flows.

2014; UNEP, 2013; Buchert et al., 2011). Recycling is usually con- sources. The cost of recovering lithium (in the form of Li2CO3) from the
sidered to be beneficial for the economy and the environment because it slag using the currently available recycling processes is estimated to be
reduces primary production, energy consumption, and the associated in the same range as that of producing battery-grade lithium carbonate
environmental impact (Zeng et al., 2014; Hagelüken, 2014; UNEP, (Li2CO3) from pegmatite minerals such as spodumene (Tytgat, 2013
2013; Li et al., 2013). In addition, domestic recycling can reduce the pers. comm.). Specifically, the variable1 cost of extracting Li2CO3 from
quantity of imported materials (Gaines, 2014; Richa et al., 2014) and spodumene can be estimated at 3.1 US$ per kg Li2CO3 (Hykawy and
thus the vulnerability of supply chains. Therefore, the majority of the Thomas, 2009) and are thus actually higher than the cost of the current
above-mentioned studies emphasize the relevance of battery recycling procedure for extracting Li2CO3 from brine deposits corresponding to
for future lithium supply security and include recycled material as an 1.2 US$/kg (Muhl, 2014; Yaksic and Tilton, 2009; Hykawy and
important additional future source of lithium that could significantly Thomas, 2009). To achieve lower recycling costs, the whole recycling
decrease the need for primary raw materials (Idjis et al., 2013; Mohr chain must become more efficient, and more cost-efficient recycling
et al., 2012; Kushnir and Sandén, 2012; Gruber et al., 2011; Gaines and technologies must be developed. Furthermore, subsidies could also be
Nelson, 2010; IEA, 2009; Yaksic and Tilton, 2009; Angerer et al., 2009). an important factor for increasing recycling whose potential role will be
However, the recycling of batteries does not directly imply the re- investigated further.
cycling of lithium from spent batteries (Gardener, 2017; McCormick, An additional challenge is posed by the quality of Li2CO3 recovered
2016; Kushnir, 2015). The European Union already defined legal re- from EV batteries: the secondary Li2CO3 needs to be of high enough
quirements for recycling of batteries, where collection rates of at least quality to find a market for its original purpose, or it has to find an
45% should be reached by 26 September 2016 (Richtlinie, 2013/56/ alternative market (Gaines, 2014). Even if there is still relatively little
EU). Since the number of end-of-life (EOL) batteries from EVs is ex- information about the application of recovered Li2CO3 in battery
pected to rise significantly over the coming decades (Gaines, 2014; manufacture some research results showed that performance and dur-
Richa et al., 2014; Georgi-Maschler et al., 2012), companies such as ability of batteries produced from recycled active materials did not
Toxco (Thompson, 2011), Umicore (Tytgat et al., 2008), and Accurec reach the same quality level as if they were produced from virgin ma-
(Weyhe, 2010) have developed recycling processes for batteries, in- terials (Kwade et al., 2013). In this case, the recirculation of secondary
cluding LIB. These technologies are focused, however, on the recovery Li2CO3 may be limited to use in other sectors such as in ceramics, glass,
of more valuable materials such as cobalt and nickel (Heelan et al., and alloys (not for rechargeable batteries). But here a problem could be
2016; Kushnir, 2015; Sonoc et al., 2015). Even though these processes arising, because the demand for lithium in these sectors is very likely to
can be applied to batteries from consumer electronics, recycling tech- grow at a slower rate than that for EV batteries. It has to be stated, that
nology for lithium-based traction batteries is still the subject of ongoing the stationary applications of LIB are not considered in this work, but
research (e.g., Zeng et al., 2014; Li et al., 2013; Georgi-Maschler et al., would further deepen the mentioned problem.
2012; Kwade et al., 2013; Buchert et al., 2011). This research has led to The effects of the quantities and qualities of Li2CO3 recovered from
the development of recycling processes, including lithium recovery, on EV batteries on the overall recycling system have not been investigated
a laboratory scale, but the reuse of recycled lithium in the manufacture so far.
of new EV batteries has only been partly successful (Kwade et al., Our study analyses the impact of a rising demand for lithium-based
2013). traction batteries in electric mobility on the potential for lithium re-
Today, almost no lithium is recovered because it is not economical, covery from EOL-EV batteries and the options for as well as the lim-
and this resource remains instead in the slag that is used in the con- itations of using secondary lithium for EV batteries and other lithium
struction sector as a cement additive (McCormick, 2016; Kushnir, 2015; products. The following questions are addressed. (1) What impact does
Sonoc et al., 2015; Tytgat, 2013 pers. comm.; Dewulf et al., 2010). This
means, based on our current knowledge, that the economics of lithium
1
recovery is dependent on the costs for producing lithium from primary Variable costs do not include the amortization of capital cost of the mine and plant
construction that would additionally amount to $ 3.333 per ton of Li2CO3.

77
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

strong growth in demand for EV batteries have on the recycling of (cf. Fig. 1). Stock development and vehicle lifetimes determine the
traction batteries, especially the potential recovery of secondary li- input of new vehicles and output of retired ones.
thium? (2) To what extent can recovered lithium from traction battery
recycling contribute to a reduction in the demand for primary raw Vehicle stock forms the basis for deriving the stock of batteries in
materials as a way to save lithium resources? (3) What are the options use, which incorporates four vehicle types equipped with traction
and limitations of lithium recycling from a systems perspective? batteries: FCV, HEV, PHEV, and FEV. The calculated inflow of new EVs
into vehicle stock is the primary driver of battery stock (cf. Fig. 1). The
2. Methodology input of new batteries and output of EOL batteries is computed by stock
development and battery lifetime. The useful lifetime of batteries is
2.1. System definition generally expressed as the loss of a battery's ability to provide a specific
amount of its original nominal capacity, usually 80%. Based on current
The system taken into consideration in our study includes all the technology of LIB, battery lifetime in EV is shorter than the average
relevant processes of the global lithium cycle: the processes of lithium vehicle lifetime (Keil et al., 2015), which means that the model im-
production including primary production from brines and minerals as plicitly assumes that EV batteries are replaced after reaching their EOL
well as secondary production from recycling, manufacture and use, and by new batteries for each of the different types of EVs. Such exchange is
end of life (EOL) (cf. Fig. 1). already carried out in the EVs currently available on the market
The use process is divided into two subprocesses: (Voelcker, 2014; Ingram, 2012). Thus, the inflow of lithium for EV
batteries is determined by the inflow of new EVs that need to be
• EV batteries: subdivided further into different drive technologies equipped with traction batteries and also by the new batteries required
(type-component extension) as replacements for EOL batteries in vehicles that are still running. In
• Other use sectors: nonautomotive batteries, glass and ceramics, lu- this regard the model does not assume a reuse of functioning batteries
bricants, air conditioning, aluminum production, pharmaceuticals, from EOL vehicles, but always generates the need for new batteries in
alloys, and other uses. new vehicles and for new batteries in used cars. This assumption can be
justified by the fact that such a further utilization of used batteries
A potentially promising application of LIB in home power back-up would require perfect markets for the used but not obsolete batteries as
systems for the storage of electricity produced from, for example, solar well as appropriate replacement opportunities, which is not the case.
cells is not included in the modeling, just as the larger application in
stationary storage systems is not. (2) The lithium stocks for other uses – these are nonautomotive bat-
The EOL-process includes two subprocesses: battery recycling and teries (primary and rechargeable), glass, ceramics, glass-ceramics,
waste management of other lithium products. Two types of battery lubricants, air conditioning, aluminum production, chemicals,
recycling are taken into consideration, i.e., recycling with lithium re- pharmaceuticals, alloys, and other uses (later referred to as basic
covery and recycling without lithium recovery, in which case lithium demand) – is modeled using a defined and constant annual growth
accumulates in the slag used in the construction sector. Most lithium rate of lithium consumption in all these use sectors.
products other than EV batteries are treated in waste management
processes without lithium recovery if at all (Lu et al., 2016; USGS, 2.2.2. The production process
2012). Lithium flows from waste management either end up in the The lithium inflow to the use process3 (cf. Fig. 1) is met by primary
construction sector or in the environment (e.g. landfills). Lithium losses and secondary lithium production. Lithium can be extracted as a pri-
occur predominantly within the use phase through dissipative appli- mary raw material from brines and mineral deposits and can then be
cations like pharmaceuticals or lubricants (cf. Fig. 1). However, the used to manufacture lithium-based traction batteries for EVs as well as
total losses can be regarded as very small in comparison to the antici- to fabricate other lithium products. Secondary lithium is assumed to
pated demand for lithium for EV batteries and are therefore neglected originate exclusively from the recycling of EV batteries leaving out
here (Ziemann et al., 2012b). other use sectors, where mainly product recycling without lithium re-
The model is projected for a time frame of 40 years (2010–2050). covery takes place (cf. Ziemann et al., 2012a).

2.2. Model description


2.2.3. The end-of-life process
Based on the present recycling situation of LIB (cf. Section 1), it is
2.2.1. The use process
possible to recover recyclable grade lithium from obsolete traction
A dynamic model of the global lithium cycle was developed based
batteries provided that suitable processes are used. In this case the
on the stock dynamics concept developed by Muöller (2006), applying
actual recovery rates determine the amount of secondary lithium that
the in-use stock as the primary driver for material flows.
reaches the production process (cf. Fig. 1). If lithium is not recovered, it
The global lithium in-use stock consists of two parts, which are
is used in the construction sector (as slag) or landfills instead.
modeled in different ways: the lithium stock in future electric vehicles
(1) and the lithium stock of other lithium applications (2).
2.3. Recycling scenarios
(1) A stock-driven material flow analysis (MFA) model for vehicles
built up by population (pop) and car per capita (car/cap) forms the The lithium outflow from the EV battery stock provides the basis for
basis of the development of the lithium stock in traction batteries as a quantitative estimate of the amount of secondary lithium. In the
well as of the respective inflows and outflows (cf. Fig. 1). This MFA- modeling we assumed different recovery rates for secondary lithium of
model is extended by a type component approach to integrate dif- either 40% or 80% determined by the efficiency of the whole battery
ferent drive technologies: fuel cell vehicles (FCV), natural gas ve- recycling chain (cf. Weil and Ziemann 2014, see also Fig. A.2 and ex-
hicles (NGV),2 diesel, gasoline, hybrid electric vehicles (HEV), plug- planation in SI).
in hybrid electric vehicles (PHEV), and fully electric vehicles (FEV) A recovery rate of 80% can be seen as the optimistic case where

3
For reasons of simplification, manufacturing processes using various lithium com-
2
Natural gas vehicles use compressed natural gas (CNG) or liquefied petroleum gas pounds for fabricating a wide variety of lithium products are not regarded separately
(LPG) as fuel. since these are not the focus of the model.

78
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

every single process step (collection, dismantling, preprocessing, and production and on the usage of primary resources.
recovery) needs very high values regarding their efficiencies. Such an
optimistic rate (or even higher) for lithium recycling from EV batteries (A) The first scenario assumes open-loop recycling where the recovered
is assumed by other studies about future lithium availability as well Li2CO3 does not reach battery grade quality, but the recycled ma-
(e.g. Kushnir and Sandén, 2012; Gruber et al., 2011; Gaines and Nelson, terial can be used instead of primary Li2CO3 in basic demand sec-
2010; Yaksic and Tilton, 2009; Angerer et al., 2009). Since presently tors such as glass, glass-ceramics, ceramics, alloys, aluminum pro-
feasible lithium recycling processes do not imply a high recovery effi- duction, and some other minor applications.
ciency of lithium from LIB, but rather a moderate of approx. 50–60% (B) In the second scenario, closed-loop recycling is assumed, meaning
(cf. Kushnir, 2015) a lower recovery rate of 40% might be achievable in that secondary Li2CO3 reaches a quality that allows its reuse in the
medium term (see also Fig. A.2 and explanation in SI). Through ap- fabrication of new Li-ion batteries for EV as well as non-automotive
plying different recovery rates of 40 and 80% in comparison to the applications. Recycled material can thus substitute primary Li2CO3
current RR of lithium from traction batteries being still 0% the potential in these products.
for EV battery recycling contributing to future resource availability can
be assessed within a certain range. Within the sector of Basic Demand internal recycling of LIB in
The current status of battery recycling technologies comprises two electronics was not considered since at the moment only a small frac-
different approaches in essence. tion of these batteries are collected for recycling and because pyrolysis
as the solely available recycling technology on an industrial scale in
1. “Recovery of metals” is mainly characterized by pyrometallurgical Europe does currently not recover secondary lithium.
separation of metals and incineration of organics. Resulting metal
alloys can be used as such or further refined to the level that they 2.4. Quantitative estimate of parameters
can be transformed into new battery materials. Such processes allow
for a large scale metal recovery with a high energy efficiency and Detailed documentation of the quantitative estimates of the para-
high process robustness for different battery types. However lithium meters is available in the Supporting information (Appendix). An
is separated from the recovered metal alloys and remains in the overview of the input parameters and their variation is given in Table 1.
resulting slag which needs further treatment to extract lithium The parameters to build up the demand driven MFA model of the
(Kushnir, 2015; Tytgat, 2009). global lithium stock in EV (Fig. 1) are as follows: population and vehicle
2. “Recovery of compounds” consists primarily of physical separation ownership as car per capita, vehicle lifetime, market penetration of
of fractions (plastics, steel, foils, electrolyte) followed by a propriety vehicle types as projection, battery lifetime, and lithium content in
technology to extract compounds from fractions. Recycling pro- different vehicle types.
cesses aim at extensive recovery of valuable compounds such as The specific lithium content of each EV depends on the battery size
electrolyte, separator, and active cathode material. But these tech- and the proportion of lithium in the battery materials (cf. Table 1).
nologies usually are more energy intensive, require several process Different types of LIBs are currently used in EVs, and further develop-
chemicals, and might be less robust since they are dedicated for ment of Li-based traction batteries is anticipated. Thus, three battery
well-defined battery types (Kushnir, 2015; Georgi-Maschler et al., types with differing lithium content were examined as alternative de-
2012; Kwade et al., 2013; Tytgat, 2009) velopments: Lithium-Nickel-Mangan-Cobalt-Oxide/Graphite (NMC-G)
as an example of a state-of-the-art battery, Lithium-Nickel-Cobalt-Alu-
In our study we focus on the first approach since it is already suc- minium-Oxide/Graphite (NCA-G) which is increasingly being preferred
cessfully implemented in recycling companies (Kushnir, 2015; Gaines, by many car manufacturers, and Lithium-Sulfur (Li-S) as a promising
2014; Tytgat, 2009). The slag resulting from this process could theo- type of battery for the future (cf. Table 1, more information in Ap-
retically contain up to 1% lithium by mass − a concentration roughly pedix).
equivalent to marginal spodumene mines for lithium and thus it may be
possible to recover it. Based on this we assume that the effort to recover 3. Results
secondary Li2CO3 from slags is similar to extracting primary Li2CO3
from spodumene, but the quality of recovered lithium remains unclear 3.1. Lithium consumption in electric vehicles
(Tytgat, 2013 pers. comm.). Due to different battery materials and
chemicals contained the slag is a fairly complex material for which The annual lithium inflow into the use sector of EV batteries in the
processing is difficult and recovering lithium from it only a theoretical base case is predicted to show a continuous and strong increase until
possibility until now (Kushnir, 2015). Therefore we cannot expect that 2042, and then growth of lithium demand is predicted to level out. This
the quality of potentially recovered Li2CO3 from EV battery recycling development is caused by the number of HEV starting to decline after
will reach the same level as primary produced Li2CO3. reaching its highest value in 2030 and additionally by a slowed down
To assess the implications of a growing quantity of secondary increase for PHEV from 2040 on (cf. Fig. 2). Lithium inflow of 0.85 Mt
Li2CO3 from EOL-EV batteries against this background, we investigated in 2050 corresponds to approximately 27 times the current lithium
two distinct recycling options: closed-loop recycling and open-loop production of 0.032 Mt in 2012. The annual material inflow into EV
recycling. In closed-loop recycling, the intrinsic properties of materials batteries is expected to develop in the same way regardless of whether
do not undergo any changes. This implies that the inherent character- low (NMC batteries) or high (Li-S batteries) lithium content is em-
istics of the recycled materials are not considerably different from those ployed, reaching either 0.37 Mt or 1.43 Mt in the year 2050, respec-
of the virgin material. The recycled material or product can still be used tively. This will be equivalent to 12 or 45 times the current level of
in the same product system and thus replace primary materials production (cf. Fig. 2). In addition, the lithium demand for other (non-
(Huysman et al., 2015; Williams et al., 2015). In open-loop recycling, in traction battery related) use sectors (i.e., basic demand) evolves at a
contrast, the intrinsic properties and the inherent characteristics of the constant annual growth rate and will amount to approximately 0.2 Mt
material are altered in such a way (i.e., often a degradation in quality) in 2050. LIB in consumer electronics will continue to have the largest
that reuse in the same product is not possible. Instead, the recycled share in this and account for lithium demand of almost 0.05 Mt in 2050
material can only be used for other product applications, which is also and thus roughly 5 times that of 2012.
termed downcycling (Williams et al., 2015). Looking more closely at this development, it can be seen that the
Using this distinction, we defined two main scenarios to evaluate lithium required for EV batteries might outpace lithium consumption
the potential effects of lithium recycling from EV batteries on lithium from basic demand as soon as 2016 in the base case for NCA batteries

79
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

Table 1
Overview of input data and references for modeling lithium demand in electric vehicles.

Input parameters Data Reference(s)

2010 2050

Population Medium case 6.5 bn 8.6 bn UN (2011)


Car ownership (car/1000 capita) Medium case 144 320 Modaresi and Müller (2012)
Type share of EV in light duty vehiclesa [%] FCV 0 18.6 IEA (2009): Blue Map scenario
HEV 1.2 7.6
PHEV 0 33.3
FEV 0 25.9
Vehicle lifetime 15 years Ball et al. (2011), Mueller et al. (2007)
Battery lifetime 10 years Keil et al. (2015), Kushnir and Sandén (2012)
Lithium content [kg] Battery HEVb/FCV PHEV FEV Weil et al. (2013)
Battery size composition 1.5 kWh 12 kWh 25 kWh
Low: NMC-G 0.13 kg/kWh 0.2 kg 1.6 kg 3.3 kg Majeau-Bettez et al. (2011), own calculations
Medium: NCA-G 0.25 kg/ 0.4 kg 3 kg 6.3 kg Gaines and Nelson (2009), own calculations
kWh
High: Li-S 0.41 kg/kWh 0.6 kg 5 kg 10.5 kg Zhang et al. (2010), own calculation
Recovery rate (recovery of Li2CO3) Low 40% Assumption (based on Weil and Ziemann. 2014)
High 80%
Basic demand (other use sectors, incl. LIB in electronics) 5% annual growth rate 28 202 Roskill (2013), Yaksic and Tilton (2009), own
[Mt Li] calculation

a
Type share also includes diesel and gasoline vehicles as well as natural gas vehicles whose market penetration will decrease significantly in the BlueMap projection of IEA (more
information is provided in the SI).
b
For reasons of simplification it is assumed in the modeling that every new produced HEV will be equipped with a LIB. In fact, there are still car manufacturers selling their hybrids
with the more expensive NiMH-batteries, but LIB are set to displace NiMH in the near future due to the advantages they provide as can already be observed for market leader Toyota or
also Nissan.

(2015 for Li-S and 2021 for NMC). Between 2011 and 2014 there could batteries it would be higher at 0.41 and 0.81 Mt, respectively. These
be seen an average increase of 27% per year in global lithium-ion volumes of secondary lithium can in principle serve as a substitute for a
battery consumption reaching an estimated 55 gigawatthours (GWh) in considerable part of the primary raw materials required for the lithium
2014 (USGS, 2014). inflow into the in-use stock.
The outflow of lithium from EOL-EV batteries occurs with a time lag
of the assumed battery lifetime of 10 years. The annual material out- 3.2. Lithium recycling and reuse
flow from the EV battery stock is continuously growing and will reach
approximately 0.6 Mt (NCA) in 2050. The development for the other Within the context of the different recycling scenarios, the extent to
battery types should be similar, but result in different outflows of either which secondary Li2CO3 from EV battery recycling can meet the
0.26 Mt (NMC) or 1.01 Mt (Li-S). The outflow is lower than the inflow growing demand for lithium was evaluated. Also investigated was how
as long as the stock of EV batteries is growing. In 2028, however, the this might change with a permanent increase in the number of EOL
annual lithium outflow from EV batteries of NCA type will amount to automotive batteries becoming available for recycling as a function of
0.08 Mt and exceed the inflow into the sectors of basic demand. The the quality and the reuse options for recycled lithium.
timing of this switch for the alternative battery types can be as late as The results of the different recycling scenarios are shown in Fig. 3. If
2038 (NMC) or as early as 2025 (Li-S). the Li2CO3 recovered from obsolete traction batteries does not reach
The amount of secondary lithium that could be gained from auto- battery-grade quality and can thus only be used for specific use sectors
motive battery recycling is usually determined from the lithium outflow that do directly require primary Li2CO3 (Scenario A), the supply of
of EV batteries from the use phase by using an applied recovery rate of secondary Li2CO3 from EV battery recycling will soon exceed the de-
either 40% or 80%. This would result in as much as 0.24 or 0.48 Mt for mand for Li2CO3 in the use sectors of glass, glass-ceramics, ceramics,
NCA batteries in 2050. For NMC batteries, the amount would usually be alloys, aluminum production, and some other minor applications. The
smaller, approximating 0.1 or 0.21 Mt, respectively, and for Li-S first year of Li2CO3 oversupply would be 2021 for LiS batteries and a

Fig. 2. Annual Li inflow into and Li outflow from use in the base case: basic demand and lithium demand for EV batteries by type according to the BLUE Map scenario (BM).

80
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

Fig. 3. Simulated future demand of Li and the relative share covered by primary and secondary Li production assuming different recovery rates and possibilities for reusing recycled Li.

recovery rate of 80%; whereas the oversupply would be delayed until comparable effects: lithium demand increases much faster if high li-
2030 in the case of NMC batteries and a recovery rate of 40%. If a lower thium content and low battery lifetime are assumed (in the same
growth rate of 2.5% for Basic Demand is applied the oversupply of amount), but it is only slightly lower if low lithium content and longer
Li2CO3 occurs already approximately one year earlier, but will be ad- battery lifetime are applied (detailed information can be found in the
ditionally delayed when the growth rate for Basic Appendix).
Demand amounts to 7.5% (see also Appendix for detailed results). The demand for lithium from other use sectors (c.f. Fig. 2 Basic
The gap between demand and supply can grow to considerably different Demand) accounts for a much smaller fraction of total lithium demand
levels, ranging from 0.16 to 0.77 Mt per year in 2050, depending on than that for EV batteries despite the annual growth rate of 5%. This
battery type and recovery rate. The higher this difference between demand increase is also primarily driven by lithium consumption for
demand and supply of secondary Li2CO3 the more virgin material is LIB, in this case for consumer electronics for which the market ex-
needed tapping lithium sources that require more energy for extracting panded rapidly within the last years reaching high growth rates. In fact,
lithium than from brines at the moment. further demand increase in electronic batteries is expected on a lower
In contrast, if the recovered Li2CO3 from EV battery recycling scale since saturation effects can already be observed in industrial
reaches battery-grade quality, thereby enabling its reuse in the manu- countries in Europe and North America for e.g. cellular phones or
facture of new rechargeable batteries (Scenario B), the supply of sec- portable PCs reducing market forecasts to approximately 6% growth
ondary Li2CO3 could meet a certain part of the required inflow in rate for the next years (GFK 2016, Pillot 2015). However, even if the
battery applications. Besides covering the lithium inflow into non- increase of LIB demand for portable electronic devices might be higher
automotive it could also sustain a limited number of new Li-based the potential development of Basic Demand also depends from other
batteries for EVs irrespective of which battery type is favored. At a use sectors such as glass, ceramics, greases or aluminum production,
recovery rate of 40%, the amount of secondary Li2CO3 could cover some of which are predicted to have smaller growth rates or even de-
approximately 25% of the required annual lithium inflow for EV bat- cline over the coming decades e.g. greases and aluminum production
teries in 2050. The amount could even reach 53% in 2050 if the re- (c.f. Roskill, 2013; Yaksic and Tilton, 2009).
covery rate is 80%. The lithium content in EV batteries is determined by the composi-
tion of the individual battery type and the battery capacity. The di-
rection of battery research suggests a trend to the use of more lithium-
4. Discussion intense active materials such as NCA, LTO, and Li-S (cf. Simon et al.,
2014). In addition, developments in the automotive industry are con-
4.1. Development of demand for lithium centrated on increasing battery capacity, which tends to result in a
higher lithium content of traction batteries. So far FEVs are produced
The future demand for lithium and the availability of scrap are with a battery capacity of approximately 20–25 kWh (Honda, 2014;
highly uncertain, depending mainly on the degree of EV penetration BMW, 2013; Nissan, 2013) (cf. Table 1). Still, there are trends toward
and the lithium content in batteries. The influence of changes in po- the use of larger batteries (60–85 kWh) within the automotive industry
pulation development and vehicle ownership on lithium demand is less – e.g. Tesla Roadster already has a Li-ion battery with a capacity of 56
intense than the effects of low or high lithium content and short or long kWh that enables a driving range of 320 km (Tesla, 2013). This
battery lifetime respectively. Lithium content and battery lifetime have

81
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

intended development is triggered by some important advantages of recovered from the slag of currently applied pyrometallurgical re-
larger batteries including longer driving range, which increase the at- cycling processes in a high enough quality since these slags also contain
tractiveness of electric mobility, or improved battery endurance pos- other materials and certain impurities that may interfere with a po-
sibly resulting in a longer battery lifetime. Increased battery capacity tential extraction process. This could result in higher requirements for
may even bring about better possibilities for EV to be integrated in energy and extraction chemicals reducing the efficiency and thus
surrounding systems such as vehicle-to-grid and smart cities. Moreover, profitability of the process (Sonoc et al., 2015). It is, therefore, con-
the material composition is decisive since even a change in lithium- ceivable that the recovery of Li2CO3 from slags might be executed in a
based traction batteries from the NCA type to the Li-S type would in- way where the focus is on economics instead of quality issues. Such
crease the lithium content substantially, namely from 6.25 to 10.5 kg downgrading of secondary lithium constrains the contribution of EV
per FEV (cf. Table 1). If this trend continues, further increases in the battery recycling to future lithium availability, because it only sub-
lithium content per vehicle are to be expected during the next 20 years. stitutes limited amounts of virgin material needed for some lithium
Another factor influencing the outcome is the lifetime of EV bat- applications that can use lower qualities of recovered Li2CO3. However,
teries which is assumed to be shorter (around 10 years) than that of in this case an oversupply of secondary Li2CO3 from open-loop re-
vehicles (approximately 15 years) based on the warranties most car cycling will occur. The timing of such oversupply depends on the re-
manufacturers give for their LIB at the moment of approximately 8–10 covery rate, the lithium content in batteries, and the possibilities for
years (cf. Morris, 2017; Shahan, 2016). However, the battery lifetime is reusing recovered Li2CO3 (cf. Fig. 3). For investigating to what extent
dependent on many “parameters” like e.g. working temperature, Depth these factors influence the availability of secondary lithium to suffice
of Discharge (DoD), the quantity of charge cycles and use frequencies, future raw material demand recovery rates of 40 and 80% as well as
but also calendar degradation and environmental conditions making it three battery technologies with different lithium contents were applied.
difficult to give general values for the useful lifespan of LIB in EV Surplus of recovered material could occur as early as 2021 and
(Smith 2017). In addition to that there exist only little practical ex- could add up to a cumulative amount of leftover secondary lithium in
perience with aging EV batteries until now, but different statements can the range of 7.6–9 Mt by 2050, which corresponds to 39%–47% of the
be found in current reports about how long the battery pack in an currently known lithium reserves or even outperforms the sum of cur-
electric car is to be used ranging from 5 to 20 years (Morris, 2017; rent reserves in countries exposing high risks for supply disruption such
Smith, 2017; Shahan, 2016). Nevertheless battery lifespan in bigger as China and Bolivia (cf. Weil and Ziemann, 2014). Likewise, the con-
cities such as Indianapolis was found to be slightly lower than 10 years siderable amount of recovered but not recycled lithium would be as
for the majority of cars according to a recent study (TRB, 2016). This much as 11%–22% (at a moderate recovery rate of 40%) or 30%–47%
mismatch of vehicle lifetime and battery lifetime can have different (at 80%) of the required primary lithium inflow in 2050. Although the
implications. potential oversupply of secondary lithium could result in an accumu-
If the battery lifetime is assumed to be 10 years while the vehicle lation of surplus material that might be reprocessed in the future, op-
lifetime is taken to be 15 years two batteries will be required for each portunities and regulations for deposition (e.g., of recovered Li2CO3 or
EV: one battery with a 10 year lifetime and one battery with a lifetime lithium-containing slags from battery recycling) would be required in
of 5 years. For future lithium demand this would imply the shorter the the meantime.
battery lifetime the more batteries have to be replaced during a vehi- In contrast a low recovery rate could delay and reduce a potential
cle's lifetime and more material is thus required to fabricate EV bat- oversupply of recovered lithium. The same applies for a higher growth
teries. rate of 7.5% in the relevant use sectors of BD (see SI for details).
Furthermore there also is the possibility that people will not replace Connected to this approach is the prospect of a decrease in the amount
the traction battery of their EV after it reached the end of its useful of lithium used in EV batteries (see also Zeng and Li, 2013).
service life since the battery can still have at least 80% of its original Altogether the results of the different scenarios (cf. Fig. 3) show that
capacity by then. It is conceivable that people might instead buy a new open-loop recycling could only make a limited contribution to the
car containing a new battery with 100% capacity or they adapt to the substitution of virgin materials as well as to saving primary resources.
situation of a probably decreasing driving range for the remaining ve- This would lead to a faster increase in lithium mining to satisfy growing
hicle lifetime. Another option would be the reuse of functioning bat- lithium demand for all applications. Such a development results in the
teries (aged approximately 5 years) from retired vehicles in new EV, but necessity to exploit lower quality deposits with presumably higher ex-
this has not been implemented in the model. Currently such reuse is traction costs. Higher costs for primary lithium production are caused
unrealistic mainly due to the fact that battery replacement would re- by e.g. lower lithium concentrations in brines or minerals and un-
quire identical vehicle construction and functioning markets for used favorable magnesium-to-lithium ratios in brines. The exploitation of
batteries. There is constant development in vehicle construction (model lower quality deposits is usually accompanied by higher energy re-
cycles in the automotive industry are approximately four years) and quirements and emissions resulting from e.g. lower evaporation rates of
further battery research. In contrast, it is implicitly assumed that bat- brines or difficult accessibility of deposits and also the environmental
teries from retired vehicles that are still functioning are put to some impacts of resource extraction are higher as well as possible supply risks
kind of second use. One promising possibility is the reuse of such (Muhl, 2014; Wanger, 2011). Since a considerable part of current li-
second-life batteries for stationary applications, such as are already thium reserves is situated in high risk countries (e.g. Bolivia, China) fast
implemented in the USA for the individual cells of EV batteries that are increase of lithium demand without efficient lithium recycling could
still active (Gaines, 2014). Since the obsolete batteries removed from result in a potential dependence on critical reserves with a higher po-
cars are still capable of significant energy storage, repurposing LIB into tential of supply disruption (cf. Weil and Ziemann, 2014).
stationary applications (e.g., residential home power backup and large Following these findings, the focus for developing efficient lithium
industrial electrical load shifting) provides a valuable extension of life recycling technology should be placed on closed-loop recycling because
providing both environmental and economic benefits (Ahmadi et al., the market potential of battery grade lithium is larger in the longer
2014; Heymans et al., 2014). term. If EV penetration grows strongly and increasing numbers of EOL
batteries are recycled, the market for LIB traction batteries alone will be
4.2. Impact of battery recycling large enough to absorb the amount of secondary lithium that is gen-
erated. This implies the development of suitable structures to be able to
The quality of the recovered lithium from battery recycling pro- process recovered Li2CO3 to a quality level that allows for recirculation
cesses is decisive for its reutilization as secondary raw material in the in the manufacture of new rechargeable batteries. In fact, the only re-
fabrication of lithium products. But it is still unclear if lithium can be cycling technology currently available for LIB on an industrial scale in

82
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

Europe is pyrolysis, where lithium remains in the slag and recovering it delay lithium material recycling and thereby restrict the availability of
is only a theoretical possibility that would be expensive and thus in- secondary lithium (see also Ziemann et al., 2016).
efficient at the moment (Gardener, 2017; McCormick, 2016; Kushnir, Furthermore, the contribution of recycling in a growing economy is
2015; Tytgat, 2013). However, pyrolysis is highly effective at re- limited (Graedel, 2015). This is also the case regarding the potential
covering Nickel, Cobalt, and Copper in a concentrated and relatively lithium outflow from battery recycling covering only a limited amount
clean alloy with high efficiency (see Kushnir, 2015; Tytgat, 2013). of annual lithium demand as long as the lithium stock is growing fast,
Despite that the revenue from recycled material for the recycling which is to be expected until saturation of the EV market (see also
companies is still low at the moment making legislation the main factor Pehlken et al., 2017). The effect that the share of secondary Li pro-
for the decision to recycle and thus giving recyclers the possibility to duction remains low as long as total raw material demand keeps rising
charge a premium gate fee for used batteries (Kushnir, 2015). If already can be observed for many metals and results mainly from the residence
LIB recycling as executed now is not economic lithium recovery from time of products in the use phase delaying the outflow to recycling
currently available recycling processes does not seem likely to be cost (UNEP, 2013).
effective in the short term. Instead there is the need for developing The model’s limitations refer to the resolution of the system defi-
alternative recycling processes to get to closed loop recycling, espe- nition and parameter assumptions. The MFA-model we used employs a
cially regarding circular flows of lithium (Gardener, 2017; McCormick, global system definition to calculate lithium flows on a global level. The
2016; Kushnir, 2015). In this regard other recycling strategies more stock of vehicles worldwide from which the lithium stock in future
dedicated to recover valuable compounds such as electrolyte, separator, electric vehicles is derived was built up using global averages of po-
and active cathode material (e.g. hydrothermal or direct physical re- pulation growth and car ownership development. Such a stock-driven
cycling) could offer a promising though more complicating approach approach is frequently used in MFA-Studies to determine the long-term
(cf. chapter 2.3, see also Kushnir, 2015; Georgi-Maschler et al., 2012, development of material stocks and flows in the global system of ve-
Gaines, 2011; Tytgat, 2009). The hydrometallurgical recycling process hicle use (cf. Løvik et al., 2014; Modaresi and Müller, 2012; Kushnir
seems to be more beneficial for Li recovery, but the efforts are relatively and Sandén, 2012; Carles, 2010). Since the data applied was originally
high and not seldom toxic substances are used (Rodriguez Garcia and built up as aggregates from a more regional scale it takes also account
Weil, 2015). of variances in future development of car ownership in different parts of
Furthermore, in order to facilitate the closed-loop recycling of EV the world (see detailed explanation of parameter estimation in the SI).
batteries, a functioning infrastructure (e.g., collection, transport) must Likewise the IEA projection of the global future EV market is based on
be established (Zeng et al., 2014; Wang et al., 2014; Zeng and Li, 2013), regional sales forecasts (cf. IEA, 2009). Although it predicts a very
which requires effective cooperation between battery researchers, bat- strong increase of EV market shares this expansion is in line with EV
tery manufacturers, and recycling companies. Such collaboration sales targets announced by governments all over the world. The IEA
should allow the participants to design efficient recycling processes projection thus constitutes a level of market penetration that govern-
adapted to LIB, in contrast to the rather robust recycling technologies ments are willing to stimulate or might be obligated to for reaching
employed at the moment to include a broad variation in batteries. ambitious climate targets. It probably makes this forecast an established
There are several strategies to expand EV battery recycling: a standard scenario for several investigations of future impacts from EV develop-
configuration of batteries to develop appropriate recycling equipment; ment on material demand and recycling (cf. Vikström et al., 2013;
further chemistry standardization to reduce the necessity for sorting; Modaresi and Muöller, 2012). Other studies applied optimistic sce-
labeling of cells to allow for better sorting; and advanced battery design narios for future EV market penetration, too (e.g. Mohr et al., 2012;
for disassembly makes separation of contained materials possible (cf. Kushnir and Sandén, 2012; Gruber et al., 2011; Carles, 2010; Yaksic and
Sonoc et al., 2015; Zeng et al., 2015; Kushnir, 2015; Gaines, 2014; Zeng Tilton, 2009), thus reaching comparatively high values of future li-
and Li, 2013; Gaines, 2011). With such improvements recycling of EV thium demand for EV-batteries, but their implications on future lithium
batteries could yield valuable, high-purity materials, also in the case of availability depend on the individual scope of the respective study.
secondary Li2CO3. Beyond this, a credible and practical framework is Furthermore parameters such as car ownership or sales forecasts are
required to facilitate the development of an economic recycling system likely to be influenced at a much smaller scale and might result in re-
with efficient material recovery since there are still uncertainties with gional differences of EV numbers. However, such differences and re-
respect to, for example, the safe and economic transport of spent bat- gional flows of lithium for EV batteries that may arise are not con-
teries. In this context, there is an additional need for reliable framework sidered in the modeling.
conditions for the processing of spent batteries so that investment in
efficient recycling processes can pay off within a limited period of time 5. Conclusions
(Zeng et al., 2014; Wang et al., 2014). Against this background gov-
ernment intervention in the form of subsidies might be needed since EV Through our analysis it could be shown that lithium recycling is
recycling technologies with lithium recovery are still the object of re- important for future availability of this resource, especially in the light
search and have not been proven to be economic yet (IM, 2016; of growing demand by electric mobility and potentially high risks for
Kushnir, 2015). It is likewise difficult to anticipate a high efficiency in supply disruptions on the production side in politically instable coun-
EV battery recycling and lithium recovery because of the ongoing re- tries. However the contribution of lithium recycling to save primary
search in battery technology, specifically by further improvements to resources is dependent on the process efficiency as well as the quality of
LIB and changes in battery chemistry within the next twenty years the recovered materials, which is decisive for potential recirculation
(Pehlken et al., 2017; Hanisch et al., 2015; Ziemann et al., 2012b). into new products. The quality of secondary lithium decides whether an
The realization of closed-loop recycling of lithium-based batteries is open-loop or closed-loop recycling for specific applications can be
even more relevant if stationary applications for LIB such as home carried out. If the quality of recovered lithium from EV batteries is not
batteries for the storage of energy from solar cells (e.g., Tesla Power high enough for reutilization in manufacture of new batteries open-loop
Wall (Tesla, 2016)) gain a considerable share of the market and put recycling will be applied. In this case secondary lithium could only be
additional pressure on lithium resources. However this was not taken reused in other use sectors, but such reuse is limited by the develop-
into account in the analysis at this stage, because the market integration ment of basic demand. If these use sectors cannot take up the amounts
of such storage devices is currently very complex and diverse and dif- of recovered lithium from EV battery recycling a surplus of secondary
ficult to quantify. The same applies for the reuse of batteries in second lithium could easily occur and thus only a limited part of potential
life applications which could also provide for stationary electricity future lithium demand would be met. Conversely if secondary lithium
storage. But further extension of battery lifetime beyond 10 years could reaches battery grade quality closed-loop recycling of lithium-based EV

83
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

batteries can be realized. Such closed-loop recycling would be able to References


provide larger amounts of secondary lithium as substitute for virgin
materials in manufacture of new batteries. This is necessary to sustain a Ahmadi, L., Yip, A., Fowler, M., Young, S.B., Fraser, R.A., 2014. Environmental feasibility
significant part of growing lithium demand once electric vehicles with of re-use of electric vehicle batteries. Sustain. Energy Technol. Assess. 6, 64–74.
Angerer, G., Marscheider-Weidemann, F., Wendl, M., Wietschel, M., 2009. Lithium für
LIB reach high market shares. Thus, closed loop-recycling cannot only Zukunftstechnologien. Nachfrage und Angebot unter besonderer Berücksichtigung
prevent a potential oversupply of secondary lithium, but also save a der Elektromobilität. ISI–Fraunhofer-Institut für System-und Innovationsforschung,
larger amount of primary resources compared to the open-loop re- Karlsruhe.
BMW, 2013. BMW i3 Technische Daten. http://www.bmw.com/com/de/newvehicles/i/
cycling. Since closed-loop recycling of EV batteries is crucial for a sig- i3/2013/showroom/technical_data.html. (Accessed 24 January 2014).
nificant reduction of demand for primary raw materials different op- Ball, R., Keers, N., Marcus, A., Bower, E., 2011. Mobile energy resources in grids of
tions to facilitate this type of recycling can be defined: electricity. Deliverable D2.1, 2.1, 2011.
Buchert, M., Jenseit, W., Dittrich, S., Hacker, F., Schüler-Hainsch, E., Ruhland, K., Knöfel,

• Structures and framework conditions to produce battery grade


S., Goldmann, D., Rasenack, K., Treffer, F., 2011. Ressourceneffizienz und re-
ssourcenpolitische Aspekte des Systems Elektromobilität: Arbeitspaket 7 des
Li CO for reutilization in fabrication of new batteries;
2 3 Forschungsvorhabens OPTUM: Optimierung der Umweltentlastungspotenziale von

• Standardization of batteries to develop adapted processes for re- Elektrofahrzeugen. Anhang zum Schlussbericht im Rahmen der Förderung von
Forschung und Entwicklung im Bereich der Elektromobilität des Bundesministeriums
cycling of EV batteries concentrating on the valuable compounds; für Umwelt. Naturschutz und Reaktorsicherheit (BMU), Darmstadt, pp. 2011.
• Subsidies for retrofitting recycling processes or to cover investment Carles, L., 2010. Modelling Long-Term Global Lithium Stocks and Flows. In: Master
Thesis, ´Ecole Polytechnique F´ed´erale de Lausanne. Lausanne, Switzerland.
costs of recycling companies so that lithium can e.g. be recovered
DERA-Rohstoffliste, 2014. Angebotskonzentration Bei Metallen Und Industriemineralen
from the slags of pyrometallurgical recycling processes despite high –Potenzielle Preis- Und Lieferrisiken. Deutsche Rohstoffagentur (DERA) in der
costs; Bundesanstalt für Geowissenschaften und Rohstoffe (BGR), Berlin. http://www.bgr.

• Regulations for the processing of EOL-EV batteries to go beyond bund.de/DE/Gemeinsames/Produkte/Downloads/DERA_Rohstoffinformationen/


rohstoffinformationen-10.pdf?__blob=publicationFile&v=6. (Accessed 21 May
overall recycling rates for EOL-vehicles or parts of such vehicles to 2016).
be more material specific. Dewulf, J., van der Vorst, G., Denturck, K., van Langenhove, H., Ghyoot, W., Tytgat, J.,
et al., 2010. Recycling rechargeable lithium ion batteries: critical analysis of natural
resource savings. Resour. Conserv. Recycl. 54, 229–234.
However the efficiency of EV battery recycling − though still an GFK, 2016. Global Smartphone Sales Growth Driven by China. Press release. http://www.
object of research − appears as a complex interaction of the quality of gfk.com/insights/press-release/global-smartphone-sales-growth-driven-by-china/.
the recovered lithium and of recycling costs. Usually the costs for re- (Accessed 22 September 2017).
Gaines, L., Nelson, P., 2009. Lithium Ion Batteries: Possible Materials Issues. Argonne
covery of secondary materials compete with production costs for pri- National Laboratory. https://www.anl.gov/energy-systems/publication/lithium-ion-
mary materials. Since the recycling costs are merely resulting from the batteries-examining-material-demand-and-recycling-issues. (Accessed 25 July 2015).
effort to recover materials during recycling processes they represent Gaines, L., Nelson, P., 2010. Lithium-ion Batteries: Examining Material Demand and
Recycling Issues. Minerals, Metals and Materials Society/AIME.
expenditures in case of i.a. energy and process requirements as well as
Gaines, L., 2011. Recycling of Li-Ion Batteries. Illinois Sustainable Technology Center,
emissions. In order to achieve high quality recycling products the re- University of Illinois. http://www.istc.illinois.edu/about/SeminarPresentations/
quired energy might be higher and so would be the costs for recovery. 20111115.pdf. (Accessed 20 January 2017).
A closer look at pyrometallurgical recycling technologies being Gaines, L., 2014. The future of automotive lithium-ion battery recycling: charting a
sustainable course. Sustain. Mater. Technol. 1–2, 2–7.
more frequently used than others reveals that currently lithium re- Gardener, J., 2017. The Rise of Electric Cars Could Leave Us with a Big Battery Waste
covery from the slags of such battery recycling processes is not carried Problem. The Guardian. https://www.theguardian.com/sustainable-business/2017/
out because it lacks profitability for the recycling companies. This is aug/10. (Assessed 20 September 2017).
Georgi-Maschler, T., Friedrich, B., Weyhe, R., Heegn, H., Rutz, M., 2012. Development of
due to the efforts to recover lithium from these slags being approxi- a recycling process for Li-ion batteries. J. Power Sources 207, 173–182.
mately at the same level as extracting primary Li2CO3 from minerals, Graedel, T., 2015. Defining and determining criticality. In: ESM Workshop and Criticality
but the major part of primary lithium production still comes from Debate at World Resources Forum Davos. October. pp. 13.
Grosjean, C., Mirandaa, P., Perrina, M., Poggi, P., 2012. Assessment of world lithium
brines exposing lower extraction costs at present resulting in a com- resources and consequences of their geographic distribution on the expected devel-
paratively low raw material price. opment of the electric vehicle industry. Renew. Sustain. Energy Rev. 16, 1735–1744.
It can be concluded that lithium recycling depends strongly on Gruber, P.W., Medina, P.A., Keoleian, G.A., Kesler, S.E., Everson, M.P., Wallington, T.J.,
2011. Global lithium availability. J. Ind. Ecol. 15 (5), 760–775.
economic issues, meaning that the profitability of lithium recovery from
Hagelüken, C., 2014. Recycling of (critical) metals. In: Gunn, G. (Ed.), Critical Metals
EV battery recycling is determined by the raw material price for pri- Handbook. John Wiley & Sons, Oxford, pp. 41.
mary lithium and its potential development. Based on the expected Hanisch, C., Diekmann, J., Stieger, A., Haselrieder, W., Kwade, A., 2015. Recycling of
lithium-Ion batteries. In: In: Yan, J. (Ed.), Handbook of Clean Energy Systems, vol.
demand increase and possible threats by tight supply situations a rising
2015. Wiley, pp. 1–24.
price can be assumed even if its development in the short to medium Heelan, J., Gratz, E., Zheng, Z., Wang, Q., Chen, M., Apelian, D., Wang, Y., 2016. Current
term is hardly predictable. Access to materials may be an important and prospective Li-ion battery recycling and recovery processes. J. Miner. Met. Mater.
policy driver therefore making the development of cost-efficient re- Soc. (TMS) 68 (10), 2632–2638.
Heymans, C., Walker, S.B., Young, S.B., Fowler, M., 2014. Economic analysis of second
cycling processes for EV batteries considering also alternative recycling use electric vehicle batteries for residential energy storage and load-levelling. Energy
strategies and technological approaches vital for future lithium re- Policy 71, 22–30.
source availability in industrial countries. Honda, 2014. Fit EV Specifications. http://automobiles.honda.com/fit-ev/specifications.
aspx. (Accessed 24 January 2014).
Huysman, S., Debaveye, S., Schaubroeck, T., De Meester, S., Ardente, F., Mathieux, F.,
Dewulf, J., 2015. The recyclability benefit rate of closed-loop and open-loop systems:
Acknowledgements a case study on plastic recycling in Flanders. Resour. Conserv. Recycl. 101, 53–60.
Hykawy, J., Thomas, A., 2009. Lithium: The Next Strategic Material. Industry Report,
Byron Capital Markets.
The authors want to thank Roja Modaresi, Stephan Pauliuk and International Energy Agency (IEA), 2009. Energy Technology Transitions for Industry –
Gang Liu for their great support and excellent discussions regarding the Strategies for the Next Industrial Revolution. (Paris).
Idjis, H., Attias, D., Ghaffari, A., Richet, S., Bocquet, J.C., 2013. Recycling lithium bat-
modeling during the research stay at NTNU Trondheim and Marco
teries: a key issue for electro-mobility. In: 1 st Armand Peugeot Chair Conference
Muhl whose thesis contributed to the modeling. Electromobility: Challenging Issues. Paris.
Ingram, A., 2012. Toyota Hybrid Battery Replacement Cost Guide. Green Cart Reports.
www.greencarreports.com/news/1078138_toyota-hybrid-battery-replacement-cost-
guide. (Accessed 18 April 2014).
Appendix A. Supplementary data Keil, P., Schuster, S., v. Lüders, C., Hesse, H., Arunachala, A., Jossen, A., 2015. In:
Lifetime Analyses of Lithium-Ion EV Batteries, 3rd Electromobility Challenging Issues
Supplementary data associated with this article can be found, in the Conference (ECI). Singapore, 1st–4th December.
Kushnir, D., Sandén, B.A., 2012. The time dimension and lithium resource constraints for
online version, at https://doi.org/10.1016/j.resconrec.2018.01.031.

84
S. Ziemann et al. Resources, Conservation & Recycling 133 (2018) 76–85

electric vehicles. Resour. Policy 37, 93–103. Sonoc, A., Jeswiet, J., Soo, V.K., 2015. Opportunities to improve recycling of automotive
Kushnir, D., 2015. Lithium Ion Battery Recycling Technology 2015: Current State and lithium ion batteries. Procedia CIRP 29, 752–757.
Future Prospects. Environmental Systems Analysis. Chalmers University, Göteborg, TRB, 2016. Quantifying the Impacts of Electric Vehicle Travel Patterns on Battery Life
Sweden, pp. 18 ESA REPORT # 2015. Span. Transportation Research Board. https://trid.trb.org/view.aspx?id=1437290.
Kwade, A., Bärwaldt, G., Meinfelder, T., Steinbild, M., Kramer, K., Barenschee, B., et al., (Accessed 24 September 2017).
2013. Recycling von Lithium-ionen-batterien. http://www.erneuerbar-mobil.de/ Tahil, W., 2008. The Trouble with Lithium 2: Under the Microscope. http://www.
projekte/foerderprojekte-aus-dem-konjunkturpaket-ii-2009-2011/batterierecycling/ meridian-int-res.com/Projects/Lithium_Microscope.pdf. (Accessed 12 March 2012).
abschlussberichte-recycling/abschlussbericht-lithorec.pdf. (Accessed 18 April 2014). Tesla Motors, 2013. Roadster Technology – Battery. http://www.teslamotors.com/
Løvik, A.N., Modaresi, R., Müller, D.B., 2014. Long-term strategies for increased recycling roadster/technology/battery. (Accessed 24 January 2014).
of automotive aluminum and its alloying elements. Environ. Sci. Technol. 48 (8), Tesla Motors, 2016. Powerwall – Energiespeicher für einen nachhaltigen Haushalt.
4257–4265. https://www.teslamotors.com/de_DE/powerwall. (Accessed 20 December 2016).
Li, L., Dunn, J.B., Zhang, X.X., Gaines, L., Jie Chen, R., Wu, F., Amine, K., 2013. Recovery Thompson, S., 2011. Kinskursky/toxco overview. In: US China Electric Vehicle and
of metals from spent lithium-ion batteries with organic acids as leaching reagents and Battery Technology Workshop. Lemont.
environmental assessment. J. Power Sources 233, 180–189. Tytgat, J., Lopez, I., van Damme, G., 2008. End of Life Management and Recycling of
Lu, B., Liu, J., Yang, J., 2016. Substance flow analysis of lithium for sustainable man- Rechargeable Lithium Ion, Lithium-polymer and Nickel Metal Hydride Batteries. An
agement in mainland China: 2007–2014. Resour. Conserv. Recycl. 80, 58–63. Industrial Award-winning Comprehensive Solution. Umicore. http://www.
Majeau-Bettez, G., Hawkins, T.R., Strømman, A.H., 2011. Life cycle environmental as- batteryrecycling.umicore.com/download/. (Accessed 14 May 2015).
sessment of lithium-ion and nickel metal hydride batteries for plug-in hybrid and Tytgat, J., 2009. In: Li-ion and NiMH Battery Recycling at Umicore, Joint European
battery electric vehicles. Environ. Sci. Technol. 45, 4548–4554. Commission/EPoSS/ERTRAC Expert Workshop: Batteries and Storage Systems for the
McCormick, M., 2016. Lithium Recycling Still ‘Too Expensive’. Industrial Minerals. Fully Electric Vehicle. Brussels, 19 June.
http://www.indmin.com/Article/3648970/Lithium-recycling-still-too-expensive/. Tytgat, J., 2013. Personal Communication with Marcel Weil, Stuttgart.
(Assessed 20 September 2017). World Population in 2300, Department of Economic and Social Affairs, United Nations
Modaresi, R., Müller, D.B., 2012. The role of automobiles for the future of aluminum (UN), New York, 2011.
recycling. Environ. Sci. Technol. 46, 8587–8594. UNEP, 2013. Environmental Risks and Challenges of Anthropogenic Metals Flows and
Mohr, S.H., Gavin, M., Giurco, D., 2012. Lithium resources and production: critical as- Cycles, A Report of the Working Group on the Global Metal Flows to the International
sessment and global projections. Minerals 2, 65–84. Resource Panel. In: van der Voet, E., Salminen, R., Eckelman, M., Mudd, G., Norgate,
Morris, C., 2017. How Long Will A Tesla, LEAF, Or Other EV Battery Last? Inside EVs. T., Hischier, R. (Eds.).
https://insideevs.com/tesla-leaf-ev-battery-degradation/. (Assessed 22 September USGS, 2012. Lithium Mineral Commodity Summaries 2012. US Department of the
2017). Interior/U.S. Geological Survey, Reston.
Müller, D.B., 2006. Stock dynamics for forecasting material flows-case study for housing USGS, 2014. Lithium: 2014 Minerals Yearbook. US Department of the Interior/U.S.
in the Netherland. Ecol. Econ. 59, 152–156. Geological Survey, Reston.
Mueller, D.B., Cao, J., Kongar, E., Altonji, M., Weiner, P.-H., Graedel, T.E., 2007. Service Vikström, H., Davidsson, S., Höök, M., 2013. Lithium availability and future production
Lifetimes of Mineral End Uses. U.S. Geological Survey (USGS). outlooks. Appl. Energy 110, 252–266.
Muhl, M., 2014. Modellierung der zukünftigen Verfügbarkeit von Lithium für neue Voelcker, J., 2014. Nissan Leaf New Battery Cost: $5, 500 For Replacement With Heat-
Technologien unter Berücksichtigung geographischer und geopolitischer Aspekte. Resistant Chemistry. http://www.greencarreports.com/news/. (Accessed 12 March
Technische Universität Berlin, Berlin (Diplomarbeit). 2015).
Navigant Research, 2013. Lithium Ion Batteries for Electric Vehicles Will Reach $22 Wang, X., Gaustad, G., Babbitt, C.W., Richa, K., 2014. Economies of scale for future li-
Billion in Annual Market Value by 2020. Press release 22.01.2013. http://www. thium-ion battery recycling infrastructure. Resour. Conserv. Recycl. 83, 53–62.
navigantresearch.com/newsroom/lithium-ion-batteries-for-electric-vehicles-will- Wanger, T.C., 2011. The lithium future—resources, recycling, and the environment.
reach-22-billion-in-annual-market-value-by-2020. (Accessed 04 March 2015). Conserv. Lett. 4 (3), 202–206.
Nissan, 2013. The Nissan Leaf. http://www.nissanusa.com/electric-cars/leaf/charging- Weil, M., Ziemann, S., 2014. Recycling of traction batteries as a challenge or chance for
range/battery/. (Accessed 24 January 2014). future lithium availability. In: Pistoia, G. (Ed.), Lithium-Ion Batteries: Advances and
Oliveira, L., Messagie, M., Rangaraju, S., Sanfelix, J., Rivas, M.H., Van Mierlo, J., 2015. Applications. Elsevier, Amsterdam, pp. 509–528.
Key issues of lithium-ion batteries –from resource depletion to environmental per- Weil, M., Ziemann, S., Schebek Lithium, L., 2009. a strategic metal for emerging tech-
formance indicators. J. Clean. Prod. 108, 354–362. nologies – scarce of abundant? In: R'09 Twin World Congress Resource Management
Pehlken, A., Albach, S., Vogt, T., 2017. Is there a resource constraint related to lithium ion and Technology for Material and Energy Efficiency. Davos, 15/16.09.
batteries in cars? Int. J. Life Cycle Assess. 22, 40–53. Weil, M., Ziemann, S., Zimmermann, B., 2013. New materials and components for future
Pillot, C., 2015. Future trends in the rechargeable battery market. In: 20th International batteries – A resource perspective. In: Vortrag Auf Der Fachmesse World of Energy
Congress for Battery Recycling (ICBR). Montreux, Switzerland, 23.–25. September Solutions Battery & Storage – Materials and Components. Stuttgart, 30.09.−02.10.
2015. Weyhe, R., 2010. Stoffliche Verwertung moderner Batteriesysteme. In: Teipel, U. (Ed.),
Richa, K., Babbitt, C.W., Gaustad, G., Wang, X., 2014. A future perspective on lithium-ion Rohstoffeffizienz und Rohstoffinnovationen Bd.1. Fraunhofer Verlag, Stuttgart
battery waste flows from electric vehicles. Resour. Conserv. Recycl. 83, 63–76. (pp XX).
Richtlinie 2013/56/EU des Europäischen Parlaments und des Rates vom 20. November Williams, T.G.J.L., Heidrich, O., Sallis, P.J., 2015. A case study of the open-loop recycling
2013 über Batterien und Akkumulatoren sowie Altbatterien und Altakkumulatoren of mixed plastic waste for use in a sports-field drainage system. Resour. Conserv.
hinsichtlich des Inverkehrbringens von Cadmium enthaltenden Gerätebatterien und Recycl. 55, 118–128.
−akkumulatoren, die zur Verwendung in schnurlosen Elektrowerkzeugen bestimmt Worldbank, 2011. The Worldwide Governance Indicators (WGI) Project. Washington, DC,
sind, und von Knopfzellen mit geringem Quecksilbergehalt sowie zur Aufhebung der http://info.worldbank.org/governance/wgi/index.aspx#home. (Accessed 02
Entscheidung 2009/603/EG der Kommission, L 329/5 vom 10.12.2013. September 2015).
Rodriguez Garcia, G., Weil, M., 2015. Toxicity assessment of extractants used in LiB re- Yaksic, A., Tilton, E., 2009. Using the cumulative availability curve to assess the threat of
cycling. In: Presentation Symposium Chair Mines Urbaines: Challenges and mineral depletion: the case of lithium. Resour. Policy 34 (4), 185–194.
Innovation in Recycling Processes. Paris, France, 02.–03. June 2015. Zeng, X., Li, J., 2013. Implications for the carrying capacity of lithium reserve in China.
Roskill, 2013. Lithium—Market Outlook to 2017. Roskill Information Services Ltd., Resour. Conserv. Recycl. 80, 58–63.
London. http://www.roskill.com/reports/minor-and-light-metals/lithium/leaflet. Zeng, X., Li, J., Singh, N., 2014. Recycling of spent lithium-ion battery: a critical review.
(Accessed 22 February 14). Environ. Sci. Technol. 44 (10), 1129–1165.
Shahan, Z., 2016. Tesla CTO Says EV Batteries Should Last 10–15 Years. Reneweconomy. Zeng, X., Li, J., Liu, L., 2015. Solving spent lithium-ion battery problems in China: op-
http://reneweconomy.com.au/tesla-cto-says-ev-batteries-should-last-10-15-years- portunities and challenges. Renew. Sustain. Energy Rev. 52, 1759–1767.
minimum-78060/. (Assessed 25 September 2017). Zhang, J.-G., Wang, D., Xu, W., Xiao, J., Williford, R.E., 2010. Ambient operation of Li/
Simon, B., Ziemann, S., Weil, M., 2014. Criticality of metals for electrochemical energy Air batteries. J. Power Sources 195 (13), 4332–4337.
storage systems − Towards a technology specific indicator. Metall. Res. Technol. Ziemann, S., Weil, M., Schebek, L., 2012a. Dissipative applications of lithium – lost for the
111, 191–200. future? Revue de Metallurgie 109, 341–347.
Simon, B., Ziemann, S., Weil, M., 2015. Potential metal requirement of active materials in Ziemann, S., Weil, M., Schebek, L., 2012b. Tracing the fate of lithium – the development
lithium-Ion battery cells of electric vehicles and its impact on reserves: focus on of a material flow model. Resour. Conserv. Recycl. 63, 26–34.
europe. Resour. Conserv. Recycl. 104, 300–310. Ziemann, S., Ketzer, D., Young, S.B., Weil, M., Poganietz, W.R., 2016. Increased lifetime
Smith, K., 2017. Battery Lifespan. National Renewable Energy Laboratory, Transportation and resource efficiency in electric mobility – linking material flow analysis with
Research. https://www.nrel.gov/transportation/battery-lifespan.html. (Accessed 29 system dynamics. In: Teipel, U., Reller, A. (Eds.), Symposium Rohstoffeffizienz und
September 2017). Rohstoffinnovationen 4. Fraunhofer Verlag, Stuttgart, pp. 153–168.

85