batteries

Review
Methodological Approaches to End-Of-Life
Modelling in Life Cycle Assessments of
Lithium-Ion Batteries
Anders Nordelöf 1, * , Sofia Poulikidou 2 , Mudit Chordia 1 , Felipe Bitencourt de Oliveira 1 ,
Johan Tivander 1 and Rickard Arvidsson 1
1 Environmental Systems Analysis, Chalmers University of Technology, SE 412 96 Gothenburg, Sweden
2 Mechanics and Maritime Sciences, Chalmers University of Technology, SE 412 96 Gothenburg, Sweden
* Correspondence: anders.nordelof@chalmers.se; Tel.: +46-317728611




Received: 12 April 2019; Accepted: 24 June 2019; Published: 2 July 2019


Abstract: This study presents a review of how the end-of-life (EOL) stage is modelled in life cycle
assessment (LCA) studies of lithium-ion batteries (LIBs). Twenty-five peer-reviewed journal and
conference papers that consider the whole LIB life cycle and describe their EOL modelling approach
sufficiently were analyzed. The studies were categorized based on two archetypal EOL modelling
approaches in LCA: The cutoff (no material recovery, possibly secondary material input) and EOL
recycling (material recovery, only primary material input) approaches. It was found that 19 of the
studies followed the EOL recycling approach and 6 the cutoff approach. In addition, almost a third of
the studies deviated from the expected setup of the two methods by including both material recovery
and secondary material input. Such hybrid approaches may lead to double counting of recycling
benefits by both including secondary input (as in the cutoff approach) and substituting primary
materials (as in the EOL recycling approach). If the archetypal EOL modelling approaches are not
followed, it is imperative that the modelling choices are well-documented and motivated to avoid
double counting that leads to over- or underestimations of the environmental impacts of LIBs. Also,
21 studies model hydrometallurgical treatment, and 17 completely omit waste collection.

Keywords: life cycle assessment; lithium-ion; battery; end-of-life; recycling; cutoff

1. Introduction
Lithium-ion batteries (LIBs) have been widely used as power and energy storage components in
consumer electronics and portable devices, such as mobile phones and laptops, since the 1990s [1].
Compared to other commercially available battery types, LIBs offer, for example, high energy density,
long calendar and cycle lives, as well as high energy efficiency [2]. Rapid adoption of LIBs in a range
of applications within the energy and transport sectors has placed them at the core of several emerging
and growing technologies addressing energy security, climate change, and fossil fuel dependency.
For the development of electromobility, and road-bound electric vehicles in particular, LIBs have
grown to become the dominant battery technology [2–4]. In the wake of the LIB technology success,
there are environmental and resource challenges. With global demand almost doubling every five
years, concerns about material availability have been raised, especially for lithium and cobalt [2,4–8].
In addition, earlier research points towards high energy intensity in battery manufacturing, albeit with
significant variability between studies, which offsets some of the environmental benefits associated
with LIB utilization [9–11]. For this reason, it has been suggested that the use of recycled battery
materials can reduce energy use and emissions associated with LIB production [12,13], and that battery
recycling is a good way to ensure future availability of battery raw materials [2,5,14,15].

Batteries 2019, 5, 51; doi:10.3390/batteries5030051 www.mdpi.com/journal/batteries

Batteries 2019, 5, 51 2 of 15

As a response, a number of life cycle assessment (LCA) studies have been carried out with the
aim to uncover the environmental impacts of LIBs [9–11,16,17]. LCA is as a well-established tool used
for analyzing the environmental impacts of products, services, and technologies. In LCA, a life-cycle
perspective is employed, and different stages can be identified: Raw material acquisition, production,
use and end-of-life (EOL). In the final stage, batteries are scrapped, and go to waste treatment and
possible recovery of materials. However, different case study results vary notably, and earlier literature
reviews of the research field, covering traction batteries in general or LIBs in particular, relate those
variations to methodological choices, such as the modelling approaches, system boundaries, and data
sources, including inconsistencies [9–11,16]. Among these reviews, only Ellingsen et al. [9] discuss
the EOL stage, albeit briefly. The discussion includes summarizing the technical scope of the EOL
stage for each study and mentioning the main modelling approaches [9], but no in-depth analysis.
However, lack of methodological awareness and rigor in EOL modelling may constitute a risk for
under- or overestimating the benefits of battery recycling. There are also clear examples of differing
results for the EOL of LIBs. For example, Li et al. [18] report that the EOL stage stands for 3% of the
overall LIB contribution to climate change and over 28% in the case of ozone depletion impact. In
contrast, Sanéflix et al. [19] report that recycling of LIBs can reduce the life-cycle climate change impact
by approximately one quarter.
Considering the expectation that recycling will reduce both production emissions and the need for
primary raw materials, together with significant and notably different life-cycle impacts reported for
recycling of LIBs, it can be concluded that little attention has been given to EOL modelling approaches
applied in LCAs of LIBs. To address this gap, the purpose of this paper is to review the modelling
approaches used for the EOL stage in peer-reviewed LCA studies of LIBs. The focus is to investigate
methodological consistency (and inconsistency) and to discuss qualitatively how EOL modelling
influence results, by giving examples. Additional aims are to complement previous review work in
the research field, assist interpretation of existing studies and support the EOL modelling in future
LCA studies of LIBs. Before presenting the review procedure (Section 4) and the findings (Section 5),
we provide a brief description of relevant treatment and material recovery processes for spent LIBs
(Section 2), as well as a description of standard approaches for modelling the EOL stage according
to generic LCA methodology (Section 3). At the end, results are summarized into main findings
(Section 6).

2. Technical Background
LIBs consist of battery cells containing an electrolyte, a separator, a cathode, and an anode [2,20].
Various types of LIBs are available, denoted after the active cathode material, including lithium
iron phosphate (LFP), lithium cobalt oxide (LCO), lithium manganese oxide (LMO), nickel cobalt
aluminum oxide (NCA), and lithium nickel manganese cobalt oxide (NMC) [2]. Each of these active
materials results in different battery properties and characteristics, making them suitable for different
applications. The cathode also includes an aluminum current collector adjoining the active material.
For the anode, the dominant design is graphite attached to a copper current collector, and for electrolyte
materials, a lithium salt is mixed with organic solvents [2,3]. The separator, which serve as an electrical
isolation between the two electrodes to prevent short circuits, is commonly made of polyethylene
or polypropylene [2]. For larger applications, cells are often combined into packs, which provides
structural elements, cooling equipment, internal conductors and wiring, external connectors, sensors,
control electronics, and a protective enclosure [17,21,22]. Typically, the materials dominating in these
other pack components are aluminum, steel, or copper [17,21,22]. Although the technologies and
networks required for effective EOL handling of LIBs are still evolving [13,14], several schemes have
been developed for recovering various materials [23].
Batteries 2019, 5, 51 3 of 15

2.1. Waste Treatment of LIBs

The first step in the EOL of batteries, as for any product, is waste collection. A high collection
rate is a key to effective recycling, but its realization can be a challenge. Annual national statistical
waste collection data is usually based on what is placed on the market compared to what reaches the
waste handling facilities the same year [24]. Such data for LIBs in the EU shows that the total battery
waste collection rates currently are low (0–25% in different countries). However, for products with
growing production volumes, such as LIBs, the average product’s life must pass before the waste
collection volumes are able to catch up, meaning that current data is likely not representative for
the future EOL collection rate. Another explanation for low collection rates is that smaller LIBs are
often embedded in products such as laptops, where they end up in unreported channels for trade and
recycling, e.g., in China and South Korea [24,25]. In addition, for other small LIBs used in portable
electronic devices, mobile phones and laptops, there is also a significant hibernating stock in society [24].
Still, battery waste collection constitutes a challenge both in the real world as well as in terms of LCA
data availability. In contrast, for well-established product systems, such as steel and aluminum metal,
representative average collection data is available.
Nevertheless, for collected batteries, a number of pretreatment steps typically follows, with the
aim to separate waste streams that later may take different routes in the EOL processing [14,26,27].
Pretreatment often includes dismantling, sorting and shredding [23,26]. The latter means that a
large mill crushes the component into several material fractions, which are sorted automatically [28].
LIB packs and cells are thus disaggregated into different residues: electronics, plastics, containers,
current collectors and conductors, anode materials, and cathode materials. Some of these material
streams are subject to further waste treatment (e.g., incineration or landfilling), whereas others undergo
material recovery.

2.2. Material Recovery from LIBs

For material recovery from LIBs, chemical processes and “direct recycling” procedures are
available [23,26,27,29]. The chemical metal extraction methods can be divided into two main types:
Pyrometallurgical and hydrometallurgical treatment. Pyrometallurgical schemes, employing pyrolysis,
were initially the most common for industrial recycling of spent LIBs [23]. Dismantled modules or
shredded batteries are smelted in a furnace along with limestone as a slag-forming agent. Plastics
and solvents burn to provide energy for a reaction where valuable metals such as nickel, cobalt, and
copper form a relatively pure alloy [23,29]. However, other metals, including lithium, aluminum,
and manganese end up in the slag, making them much more challenging and costlier to recover.
Hydrometallurgical schemes often involve acid leaching of metals from battery residues, which are
then separated and recovered as pure metals or metal salts by solvent extraction, precipitation, or
electroextraction [27,30]. A benefit of hydrometallurgy is that almost any material can be recovered,
but drawbacks include needs for large quantities of hot water, acids, and solvents [23].
Direct recycling is a material recovery process where the cell constituents are detached without
altering their chemical form, allowing direct reuse in battery production, or requiring only minor
refurbishment [29,31]. Following disassembly and breaching, cells are treated in chamber with fluid
carbon dioxide, after which the electrolyte is extracted for reuse and the remaining materials are
pulverized and separated. This method enables most materials of the cell to be recycled, although
minor upgrading can be required [23,29,31].

3. EOL Modelling Approaches in LCA

Several approaches have been suggested for the modelling of EOL in LCA [32,33]. However,
in practice, there are currently two main methods being used: The cutoff approach (also called
the “recycled content approach”) and the EOL recycling approach (also called the “avoided burden
Batteries 2019, 5, 51 4 of 15

Batteries 2019, 4, x FOR PEER REVIEW 4 of 15

approach”) [34,35]. Note that waste treatment, at least in the form of collection, is a step in the EOL
modelling thatEOL
3.1. Cutoff and is generally included in both approaches.
Recycling
In theand
3.1. Cutoff cutoff
EOLapproach,
Recycling the fates of recyclable materials are not modelled all the way to where
they recirculate into new production. Instead, commonly only collection and the steps referred to as
In the cutoff approach, the fates of recyclable materials are not modelled all the way to where
pretreatment in the battery recycling literature (see Section 2) are included. In the typical setup, this
they recirculate into new production. Instead, commonly only collection and the steps referred to as
means that the recovery and upgrading of recyclable materials are “cut off” from the product system,
pretreatment in the battery recycling literature (see Section 2) are included. In the typical setup, this
and that no credits are awarded for providing waste streams containing materials that are recycled
means that the recovery and upgrading of recyclable materials are “cut off” from the product system,
back to the same or other product systems, i.e., as secondary raw materials. Instead, benefits are given
and that no credits are awarded for providing waste streams containing materials that are recycled
in the upstream modelling of input raw materials, when a certain share (x) of input for a specific
back to the same or other product systems, i.e., as secondary raw materials. Instead, benefits are given
material is assumed to consist of secondary input, while the remainder (1 − x) comes from primary
in the upstream modelling of input raw materials, when a certain share (x) of input for a specific
material extraction. The input of secondary raw materials is charged only with the burdens caused
material is assumed to consist of secondary input, while the remainder (1 − x) comes from primary
by the recovery and upgrading processes. In total, this implies an overall reduced production impact
material extraction. The input of secondary raw materials is charged only with the burdens caused by
for products with recycled content, and the assumed share x1, x2, etc. of different secondary materials
the recovery and upgrading processes. In total, this implies an overall reduced production impact for
are key parameters in this approach. Note also that the recycled content does not need to match the
products with recycled content, and the assumed share x1, x2, etc. of different secondary materials
recyclable output from the product system, since it could just as well originate from material recovery
are key parameters in this approach. Note also that the recycled content does not need to match the
of some other product. The cutoff approach is illustrated graphically in alternative (a) of Figure 1.
recyclable output from the product system, since it could just as well originate from material recovery
of some other product. The cutoff approach is illustrated graphically in alternative (a) of Figure 1.

Secondary raw
materials used as input

Recycling and
upgrading No credit - cutoff of
secondary raw materials
(varying point after use)
Recycled content
(a.)
Raw material
Waste collection
extraction and Production Use Credits are given for the
and treatment
processing displacement of primary
materials and
corresponding raw
material extraction
End-of-life modelling

Raw material
(b.) Waste collection Recycling and Production of
extraction and Production Use
and treatment upgrading new product
processing

End-of-life modelling

Figure 1. Illustration of (a) the cutoff approach, and (b) the end-of-life recycling approach. Boxes show
processes, full arrowsofshow
Figure 1. Illustration (a) thematerial flows, broken
cutoff approach, arrows
and (b) show scrap
the end-of-life material
recycling flows, and
approach. crossed
Boxes show
arrows show
processes, fullsubstituted material
arrows show flows.
material In alternative
flows, (a), the
broken arrows position
show scrapofmaterial
the cutoff point
flows, may
and vary,
crossed
and as a result, the scope of upstream “recycling and upgrading” varies correspondingly. Although
arrows show substituted material flows. In alternative (a), the position of the cutoff point may vary, not
shown, upgrading generally requires some blending with primary materials.
and as a result, the scope of upstream “recycling and upgrading” varies correspondingly. Although
not shown, upgrading generally requires some blending with primary materials.
There are different opinions in the LCA research field on where to set the cutoff point. Suggestions
include directly
There are when the use
different phase in
opinions hasthe
ceased,
LCAorresearch
after collection
field on and transportation
where to set thetocutoff
a scrapyard,
point.
supposedly when the value of the scrap is the lowest [35,36]. However, many
Suggestions include directly when the use phase has ceased, or after collection and transportation to used products contain
both recyclable
a scrapyard, and non-recyclable
supposedly when thematerials,
value of the where theislatter
scrap proceed[35,36].
the lowest to further treatment
However, manysuch as
used
incineration or disposal
products contain in a landfill.
both recyclable andIf such waste treatment
non-recyclable is required,
materials, where the it islatter
always attributed
proceed to the
to further
product under study [33], and for this reason it is common to model the cutoff point
treatment such as incineration or disposal in a landfill. If such waste treatment is required, it is always after sufficient
separation and
attributed to sorting
the product to make
underastudy
clear [33],
distinction
and forbetween the flows
this reason recirculating
it is common intothe
to model a new product
cutoff point
life cycles, and remaining waste flows. In agreement with this setup, waste
after sufficient separation and sorting to make a clear distinction between the flows recirculating intotreatment processes
included in cutoff
a new product life system model
cycles, and of the Ecoinvent
remaining waste flows. database place the
In agreement with cutoff point after
this setup, wasteseparation
treatment
and sorting, for example after vehicle shredding [37]. For LIBs,
processes included in cutoff system model of the Ecoinvent database place the cutoff a similar scheme would include
point after
pretreatment, and then make the cutoff before materials are restored by subsequent
separation and sorting, for example after vehicle shredding [37]. For LIBs, a similar scheme would recovery processes.
include pretreatment, and then make the cutoff before materials are restored by subsequent recovery
processes. Importantly, to be consistent, the EOL cutoff should be made at a point which harmonizes
with the input data used for the secondary raw materials in the production stage.
Batteries 2019, 5, 51 5 of 15

Importantly, to be consistent, the EOL cutoff should be made at a point which harmonizes with the
input data used for the secondary raw materials in the production stage.
In the EOL recycling approach, shown as alternative (b) in Figure 1, the material recovery is
explicitly modelled instead of being cut off. In particular, the recovered quantity of a material is
typically assumed to directly replace an equivalent amount of primary material upstream in the
product system, or in a different product system. In essence, this recycling reduces the need for
primary material production and the environmental burdens linked to the use of a corresponding
amount of primary material can be avoided. Hence, these avoided impacts are credited to the product
studied in the EOL stage as negative impacts. A key feature of this approach is that, regardless of any
actual recycled content in products in the real world, if the EOL crediting is based on avoided primary
material production, then all of the material input upstream in the LCA model also should bear the
load of primary metal production [35]. Otherwise, benefits of recycling are accounted for twice in the
same life cycle.

3.2. Closed-Loop, Open-Loop and Closed-Loop Approximation

Another way to categorize EOL modelling is to specify whether the recycling occurs in a closed or
an open loop [38]. In closed-loop recycling, recovered material is recirculated back to the same product
system, i.e., the same production (or type of production) that made the product from which the material
was recovered [38]. An example could be lithium from a scrapped LIB, which is recycled into a LIB of
the same type. In alternative (a) of Figure 1, this would occur if the broken arrow representing scrap
going out of waste treatment is connected back upstream directly to the other broken arrow entering
the recycling and upgrading process. In such a case, only net raw materials need to be accounted
for. However, even though promising methods are emerging for such “functional recycling” back to
original use for many LIB battery materials, the current situation in many large-scale recycling systems
is very different. For example, in real EOL for vehicles, closed-loop recycling is non-existent for most
materials [28].
In open-loop recycling, the recycled material is instead used in a different product, an example
being the use of cobalt from recycled LIBs in samarium-cobalt magnets for sensors and electric motors.
However, since large scale waste treatment, e.g., shredding, often involves some degree of material
intermixing and dispersion of metals present in small quantities into recovered bulk metals, such as
steel and aluminum, open-loop recycling is typically not functional [28]. Instead, materials have lower
quality and are down-cycled into products with lower material quality requirements. To compensate
the lower quality, secondary and primary materials are commonly mixed in varying shares to match
material requirements for different applications [39]. In LCA modelling, open-loop recycling has for
long been regarded as a challenge, since the environmental burden of the initial material production,
the waste treatment and the material recovery can be allocated in many different ways between two
or more cascaded product systems [40]. A number of allocation procedures have been suggested to
solve this issue for open-loop recycling, with the cutoff approach described above being the simplest to
apply and traditionally the most common [32,33,40]. Whereas the cutoff approach always pairs with
open-loop recycling, since the recycled material feeds into the larger market for recycled materials
rather than going directly back to the production, the EOL recycling approach can be modelled both as
the closed- or open-loop type depending on type of LCA and the argument for displacement.
There is sometimes a distinction made between attributional, accounting type of LCA and
consequential, change-oriented type of LCA. In brief, attributional LCA quantifies life cycle impacts
attributable to a product in an established status-quo system, whereas a consequential LCA quantifies
changes in life cycle impacts due to a certain decision [41]. The cutoff approach, which is based on
fixed average market data for recycled contents, may thus be particularly suitable for attributional
LCA. In contrast, the EOL recycling method builds on avoiding burdens by substituting (i.e., changing
the demand for) primary materials, and it is the preferred way of accounting for byproducts in
consequential LCA [33,42]. The use of substitution in attributional LCA is disputed in the LCA research
Batteries 2019, 5, 51 6 of 15

field because it alters the otherwise status-quo system assumed in attributional LCA [42–44]. One
reason is that the dynamics of substitution depends on the market situation for the specific material [44].
However, there are arguments for using the EOL recycling approach also in attributional LCA to
model open-loop situations, especially if the recovered material has not been subject to any changes of
inherent properties and quality, i.e., the recycling is argued to create functionally equivalent materials,
because then it can be approximated as occurring in a closed loop [38,40]. Another typical assumption
when using such a closed-loop approximation for EOL modelling is that the recovered materials in the
studied product system do not displace recycled materials from other products [40] (i.e., not affecting
the secondary material market), since the point of the approximation is to circumvent the open-loop
allocation problem [38].

4. Method
The collection of appropriate articles for the analysis started with a literature search using the
Scopus [45] and Web of Science [46] databases, specified for the stated time frame and based on the
following search string: (“life cycle assessment” OR “life cycle analysis”) AND (“lithium ion battery”
OR “lithium ion batteries”) AND (“recycling” OR “end of life”). First, LCA case studies of LIBs were
identified and selected. Other work was excluded, such as material or substance flow analyses of LIBs,
case studies of other battery types, as well as review articles and other overviewing LCA literature.
Publications not clearly stated to have been peer-reviewed were also excluded. Next, studies not
including the full life cycle from cradle-to-grave, i.e., assessments of LIB production or the EOL stage
only, were removed from the collection. This omission was done because the raw material input can
otherwise not be put into context with waste treatment practices and material recovery procedures,
which is important since batteries can both contain recycled materials and be a source of recovered
materials. At this stage, 29 articles remained, and they were all read with special attention to the
description of the scope and modelling approach in the method description, the reported inventory
data use and impact assessment results for the EOL stage. To complete the review, it was necessary to
find at least some information about the EOL modelling. Three studies [47–49], although covering the
full life cycle of the battery, do not provide this, and one study [50] directs the reader to another article
without any further description. It was therefore decided to exclude these four studies from the review.
As a result, the review includes 25 peer-reviewed scientific journal and conference articles which
assess the cradle-to-grave environmental impacts of LIBs using LCA, published from January 2010
until February 2019. All studies include at least one LIB among the technology options, on cell or
pack level, or put the LIB into focus as a part of a larger technical system (typically an electrical
vehicle product system). The studies were sorted into two main groups representing the two main
EOL modelling approaches outlined in Section 3: Those employing the EOL recycling approach and
those employing the cutoff approach. Unless explicitly reported, this was done by going through
the processes modelled in the EOL stage, classifying those that only included pretreatment processes
or less as “cutoff”, and those that also included material recovery in the form of pyrometallurgy,
hydrometallurgy, direct recycling or other technical methods as “EOL recycling”. The inventory data
description was checked for recycled content in the material inputs to production, including references
to the GREET model [51] and the Ecoinvent database [37]. In addition, the method section and impact
assessment results were scanned for reporting of avoided burden and crediting as negative impacts,
typical for the EOL recycling.

5. Results and Discussion

5.1. Cutoff, EOL Recycling and Hybrid Approaches

Table 1 presents a summary of the review results. It shows that 19 of the 25 studies use the EOL
recycling approach and 6 the cutoff approach. Among the studies, 15 were marked for closed-loop
approximation, including both Sanfélix et al. [19] and Sanfélix et al. [52], who argue for substitution in
Batteries 2019, 5, 51 7 of 15

their attributional study setup. In contrast, only one consequential LCA study by Vandepaer et al. [53]
was found, where the avoided burden is looked upon as substitution in an open loop scheme.

Table 1. Summary of the peer-reviewed articles included in this review, sorted into “EOL recycling” or
“cutoff” for the EOL modelling approach, “primary only” or “recycled content” for the production
of raw materials, and “avoided burden” or “no crediting” for the EOL crediting. Materials having a
recycled content are stated in parenthesis when reported in the study. The final two rows show (in bold)
the expected study setups in order to be consistent with the two main EOL modelling approach as
described in Section 3.

EOL Modelling
Study Production of Raw Materials EOL Crediting
Approach
Ahmadi et al. [54] EOL recycling Recycled content (Al) No crediting
Ciez and Whitacre [55] 1,2 EOL recycling Recycled content (Fe, Ni) Avoided burden
Cusenza et al. [56] EOL recycling Primary only Avoided burden
de Souza et al. [57] EOL recycling Primary only Avoided burden
Dunn et al. [21] 1 EOL recycling Recycled content (Al) Avoided burden
Dunn et al. [58] 1 EOL recycling Not reported Avoided burden
Faria et al. [59] EOL recycling Not reported Not reported
Gaines et al. [12] Cutoff Recycled content No crediting
Hendrickson et al. [60] 1 EOL recycling Not reported Avoided burden
Li et al. [18] 2 EOL recycling Recycled content (Al) No crediting
Notter et al. [61] 2 Cutoff Recycled content (Al) No crediting
Oliveira et al. [62] Cutoff Recycled content No crediting
Oliveira et al. [4] EOL recycling Not reported Avoided burden
Raugei and Winfield [63] 2 EOL recycling Recycled content (Al, Cu) Avoided burden
Richa et al. [64] EOL recycling Recycled content (Al, Cu, Fe) Avoided burden
Ryan et al. [65] 1,3 Cutoff Recycled content (Al, Fe, Li) No crediting
Sanfélix et al. [19] EOL recycling Not reported Avoided burden
Sanfélix et al. [52] EOL recycling Not reported Avoided burden
Swart et al. [66] 4 EOL recycling Primary only Avoided burden
Tagliaferri et al. [67] 2 EOL recycling Not reported Avoided burden
Unterreiner et al. [68] 5 EOL recycling Primary only Avoided burden
Vandepaer et al. [69] Cutoff Not reported No crediting
Vandepaer et al. [53] 2,6 EOL recycling Not reported Avoided burden
Zackrisson et al. [36] 2 Cutoff Recycled content No crediting
Zhao and You [70] EOL recycling Recycled content No crediting
EOL recycling approach EOL recycling Primary only Avoided burden
Cutoff approach Cutoff Recycled content No crediting
1 2 3
GREET model [51] stated as a data source; Ecoinvent database [37] stated as a data source; Modelled with a
closed loop for the material recovery of lithium salt, open loop for other materials; 4 Modelled with a closed loop for
all material recovery, and only net primary material input for production; 5 Reports credits in the production stage
instead of the EOL stage; 6 Consequential LCA study.

Two studies, Ryan et al. [65] and Swart et al. [66], were found to model closed loops in the
foreground system, making their studies challenging to classify as the cutoff and EOL recycling
approaches then coincide as one. Especially, Swart et al. [66] is an excellent example of a closed-loop
model where only net primary raw materials are considered in terms of inputs to the product system, as
described in Section 3. Still, it was sorted among the EOL recycling studies since it includes modelling
of the material recovery processes, in line with the principle for sorting the studies as reported in
Batteries 2019, 4, x FOR PEER REVIEW 8 of 15
Batteries 2019, 5, 51 8 of 15

reported in Section 4. Nonetheless, also when taking these two studies into account [65,66], this
review 4.
Section confirmed
Nonetheless,the general
also when dominance of the
taking these twotwo main
studies EOL
into modelling
account [65,66],approaches
this review(cutoff
confirmed and
EOL recycling) also in peer-reviewed LCAs of batteries, specifically. In
the general dominance of the two main EOL modelling approaches (cutoff and EOL recycling) also in fact, no instances of other
approaches were
peer-reviewed LCAsfound in the literature
of batteries, collection.
specifically. In fact, no instances of other approaches were found in
In addition to the
the literature collection. studies reviewed, Table 1 also includes two rows at the bottom indicating the
expected typical to
In addition setup
the of the cutoff
studies and EOL
reviewed, recycling
Table approaches,
1 also includes whichatare
two rows theinbottom
line with the general
indicating the
arguments for using either method as presented in Section 3. It was found
expected typical setup of the cutoff and EOL recycling approaches, which are in line with the general that 7 of the reviewed
studies are not
arguments consistent
for using eitherwith these two
method conventional
as presented setups.3.Consequently,
in Section It was found there that 7isofa risk of under-
the reviewed
or overestimation
studies of the environmental
are not consistent with these twobenefits of recovering
conventional battery materials
setups. Consequently, there inisthese
a riskstudies,
of under- as
illustrated in Figure 2. For the EOL recycling approach, Richa et al. [64]
or overestimation of the environmental benefits of recovering battery materials in these studies, as constitutes an example of
potential double counting (see example 5 in Figure 2), where the upstream
illustrated in Figure 2. For the EOL recycling approach, Richa et al. [64] constitutes an example of raw material input
includes recycled
potential content (see
double counting for steel
example(37%),
5 inaluminum (32%) and
Figure 2), where copper (15%).
the upstream At the same
raw material inputtime,
includesthe
recovered materials include aluminum and copper, which are credited
recycled content for steel (37%), aluminum (32%) and copper (15%). At the same time, the recovered for avoiding burden
corresponding
materials include only to primary
aluminum and material production
copper, which [64,71],
are credited forwithout
avoidingclearly
burdenexplaining
corresponding the logic.
only
Similarly, Ciez and Whitacre [55] model avoided energy use and coupled
to primary material production [64,71], without clearly explaining the logic. Similarly, Ciez and reductions of carbon
dioxide emissions
Whitacre [55] model brought
avoided about
energyby material recoveryreductions
use and coupled of copper,of iron, nickel,
carbon cobalt,
dioxide and cement
emissions slag
brought
from pyrometallurgical
about by material recovery treatment;
of copper,lithium, nickel,
iron, nickel, cobalt,
cobalt, andmanganese,
cement slag and from iron from hydro-
pyrometallurgical
treatment; lithium, nickel, cobalt, manganese, and iron from hydro-metallurgical treatment; orprecursor
metallurgical treatment; or complete cell canisters, current collectors, and cathode complete
materials
cell fromcurrent
canisters, direct collectors,
recycling. Even so, theyprecursor
and cathode also report the use from
materials of secondary steel andEven
direct recycling. nickel so,inthey
the
original
also material
report the useproduction.
of secondary It is difficult
steel to judge
and nickel to original
in the what extent this implies
material production. doubleIt iscounting
difficult toof
benefits from recycling in the same product life cycle, and the degree that environmental
judge to what extent this implies double counting of benefits from recycling in the same product life and resource
impacts
cycle, ofthe
and thedegree
studied thatLIBs are underestimated,
environmental and resourcebut impacts
it createsofuncertainty
the studied about
LIBs are theunderestimated,
validity of the
results.
but it creates uncertainty about the validity of the results.

Figure 2. Schematic illustration of how environmental impacts vary depending on the selection of EOL
Figure 2. Schematic illustration of how environmental impacts vary depending on the selection of
modelling approach, if the study uses the expected setup for cutoff (1) or EOL recycling (4), or apply a
EOL modelling approach, if the study uses the expected setup for cutoff (1) or EOL recycling (4), or
hybrid version (2, 3, or 5). Using the EOL approach without giving any credits for recycling (2 or 3) leads
apply a hybrid version (2, 3, or 5). Using the EOL approach without giving any credits for recycling
to a conservative estimation of the environmental and resource impacts of LIBs, whereas modelling
(2 or 3) leads to a conservative estimation of the environmental and resource impacts of LIBs, whereas
both recycled content and avoided burden (5) implies double counting which may underestimate the
modelling both recycled content and avoided burden (5) implies double counting which may
impact. Bar sizes are indicative only and do not reflect actual results.
underestimate the impact. Bar sizes are indicative only and do not reflect actual results.
In contrast, Ahmadi et al. [54], Li et al. [18], and Zhao and You [70] make conservative estimations
for a In contrast,
number Ahmadi et al.
of environmental [54], categories
impact Li et al. [18], and recycling
by taking Zhao andprocedures
You [70] into
make conservative
account in both
estimations for a number of environmental impact categories by taking recycling procedures into
ends of the battery life cycle, i.e., including recycled content as well as modelling the material recovery
account in both ends of the battery life cycle, i.e., including recycled content as well as modelling the
Batteries 2019, 5, 51 9 of 15

processes in the EOL stage, but without giving any credits (see example 2 in Figure 2). At large, this is
in line with the cutoff approach, but it violates the convention of setting the cutoff point before or at
the place where recyclables and non-recyclables are separated, by also including material recovery
and upgrading processes in the studied system. The main argument against this violation is that the
subsequent product system, which is the driver for the material recovery and upgrading, should carry
the environmental impacts of the recovery and upgrading processes, especially if the same principle
governs the upstream cutoff point preceding the studied product system. On the other hand, since there
are regulations making battery recycling compulsory, for example the EU Battery Directive enforcing
at least 50% recycling by mass for LIBs [72], it can be argued that the life cycle of the product under
study should include material recovery even for the cutoff approach. But such detailed discussions
about system boundaries for the EOL stage are rarely encountered in the reviewed LCAs of batteries.
Thus, it is reasonable to claim that this conservative approach rather overestimates the environmental
and resource impacts of LIBs.
Returning to Ryan et al. [65], it can also be observed that hybrid versions of the cutoff and EOL
recycling approaches are sometimes purposefully used. In this study, they report having applied the
cutoff approach, including waste handling for the EOL stage and recycled content included for the
input of aluminum and steel, but also for a lithium salt (Li2 CO3 ). However, since their main data
source, the GREET model, does not contain lithium salt as an option for recycled content, they explicitly
point out that they modelled a closed loop for the lithium salt and that the burden of those material
recovery processes is ascribed to the production stage. Hence, the study was classified as using the
cutoff approach and closed-loop recycling at the same time, even though the cutoff generally implies
open-loop recycling.
This observation also brings attention to the fact that the default settings of the GREET model
include recycled content [21,55], i.e., making it suitable for use in combination with the cutoff
approach. Similarly, the often-used Ecoinvent database contains recycled content in average market
and production mix datasets for the attributional systems models. In order to use these data sources in
combination with the standard setup of the EOL recycling approach, primary raw materials should be
selected deliberately. However, such databases are complex and may include aggregated datasets with
recycled fractions for multiple materials, e.g., in electronic units in Ecoinvent [37]. This makes them
difficult to combine with the EOL recycling approach and its requirement of only primary raw material
input. However, the problem of ‘hidden’ recycled content in market datasets becomes less problematic
to use for the EOL recycling approach if the recycled content is minor (e.g., <10%), whereby the dataset
still can be seen as a decent proxy for primary material production. Any such approximation should
preferably be stated clearly to enable users of the study’s results to interpret the model. Carefully
applied hybrid approaches, where specific materials are modelled with different approaches, is another
possible solution, but it also requires the study to be reported with detailed information about the
modelling in order for the reader to interpret results. Ryan et al. [65] do this in a commendable way.
Another option is argued for by Raugei and Winfield [63], namely to calculate the credits using
production mix data containing both primary and secondary materials. By doing so, they claim to be
consistent even though they combine recycled content with crediting for avoided burdens. However,
this implies a break with a common assumption for the closed-loop approximation, which in turn
is a key justification for the use of avoided burdens in attributional LCA: The assumption that no
secondary materials from other products are displaced. Even though there are supporters for this way
of crediting [33], an important objection is that it becomes conceptually difficult to understand the
modelling in the open loop context: the studied product system alters the secondary material markets
underlying the average production mixes used for the crediting, and thus makes the quantified credit
invalid, in a sort of circular reference. In addition, on a more fundamental level, a key purpose of
recycling is to decrease the use of natural resources and environmental burden of primary material
extraction. This also makes it questionable to replace secondary material with other secondary material.
Moreover, in the real world, it is likely that recycling is economically viable mainly when there is a
Batteries 2019, 5, 51 10 of 15

demand for recycled materials, and not when different recycled materials are competing [42]. Replacing
secondary material with secondary material thus seems unlikely from that perspective too. It thus
remains vital to clearly motivate specific modelling choices when disembarking the more established
versions of the EOL modelling approaches.

5.2. Recovery Procedures, Recycled Cell Materials and Waste Collection

For the technical scope of the cell material recovery procedures, inclusion of pyro- and
hydrometallurgical treatment is most common, which is the case in Cusenza et al. [56], where
aluminum, cobalt, copper, nickel, manganese, and steel are recovered, and awarded with primary
material credits and coupled avoided burden for a broad set of environmental impact categories.
Table 2 shows the share of the studies that apply the EOL recycling approach and include the recovery
procedures described in Section 2, as well as to what extent they model recovery of the following four
elements: cobalt, lithium, manganese and nickel. These metals are typical constituents of cell materials
recovered in current real-world industrial practices in China and South Korea [25]. Expectedly, most
studies focus on the recovery of cobalt and lithium, which have received attention regarding their
future resource availability. In addition, Table 2 displays that schemes based on hydrometallurgy are
commonly modelled, and this too is in agreement with ongoing activities in East Asia [25]. Aluminum,
copper, and steel belong to the metals which can be recovered from battery cells, but even more so from
the casings, conductors and electronics of the entire battery pack, which is separated mechanically
from the cells during pretreatment.

Table 2. Summarized results for the number of studies accounting for waste collection of LIBs (relevant
for all 25 studies) and modelling specific procedures and materials in the recovery of LIB cells (relevant
for the 19 studies modelling the EOL recycling approach).

No. of Studies Accounting for Waste Collection

Include or discuss collection rate 2
Model the transportation required for collection 6
Report no aspects of waste collection 17
No. of Studies Modelling Specific Cell Material Recovery Procedures in the EOL Recycling Approach
Direct recycling 4
Hydrometallurgy 16
Pyrometallurgy 12
Other procedures 1 2
No. of Studies Modelling Recovery of Specific Cell Materials in the EOL Recycling Approach
Cobalt 11
Lithium 12
Manganese 6
Nickel 7
1 This includes procedures not described in Section 2, for example “intermediate” recycling where lithium salt is
recovered from a liquid stream which is diverted during the post-shredder sorting process.

Table 2 also presents the share of all reviewed studies taking waste collection into account, by
including losses in the collection step and the transports required. Notably, very few studies include or
even discuss the collection rate, i.e., do not acknowledge that a significant number of batteries may
never reach dedicated battery EOL handling. A key explanation is that there is a knowledge gap
around the collection of LIBs, not only in LCA research, but in general [25], as indicated in Section 2.
Nevertheless, given the importance of the collection step for the overall impact of recycling, this is
clearly a weakness in current EOL modelling.
Batteries 2019, 5, 51 11 of 15

6. Conclusions and Implications for Stakeholders

This review has shown that the majority of the selected LCA case studies of LIBs apply the EOL
recycling approach, while the remaining share apply the cutoff approach. The results of this study thus
confirm the dominance of these two methods also in battery LCA studies. In addition, some hybrid
approaches exist. For key stakeholders of LIB technology who rely on information from LCA studies
for strategic decision support, such as industrial managers and policy makers, our main message is
that these two approaches address the EOL stage in fundamentally different ways, which influences
both quantitative results and what information the study provides.
LCA practioners focusing on LIBs are recommended to use the EOL recycling approach for
evaluating emerging and developing recycling processes for cells, since it allows for a detailed study of
these processes themselves. This may explain its dominance among the reviewed studies. The modelled
material recovery procedures generally correspond well to current evolving industry practices, except
for the collection step where there are gaps in terms of both inclusion in the study scope as well as in
data availability. However, it is important to be aware that the use of a closed-loop approximation,
employed in some of the attributional LCA studies reviewed, entails little consideration of potential
material down-cycling, or that there is no guarantee that a 1:1 replacement of primary material occurs
in a market-mediated open-loop recycling procedure [44]. However, down-cycling can be accounted
for by including additional upgrading procedures within the EOL system boundaries and lower
displacement rates by crediting for avoided burden using a realistic correction factor [43].
In contrast, both LCA practioners and users of LCA study results should be aware that the cutoff
approach is typically appropriate to use in well-established waste treatment and material recovery
settings, where the material waste streams supply a noticeable share of the recycled content for specific
material markets, typical examples being the common battery pack materials aluminum and steel.
Corresponding average market data with appropriate recycled contents is often available in LCA
databases, while they usually do not include all new and unexplored material recovery processes. In
the conducted review, the cutoff approach was found to be more common among earlier studies, where
less was known about potential future recycling processes, or where the EOL stage was in less focus.
As a general observation, inconsistencies compared to the archetypal cutoff and EOL recycling
approaches occur when two methods are blended in hybrid versions. Crediting for avoided burden by
replacing primary materials in combination with recycled content brings the risk of under-estimating
the environmental burden of LIBs. Contrary, giving no credits and at the same time include all
material recovery and upgrading instead brings the risk of overestimating burden. Consequently, LCA
practioners are recommended to be cautious with such blending. Still, depending on the data available
for modelling the production stages, a hybrid approach can sometimes be motivated. But in such
cases, when the expected setups of the two archetypal approaches are not followed, it is vital that
the modelling choices are well-documented and clearly motivated in order for the target audience
of the study to be able to interpret the results. As receivers of LCA results, industrial managers
and policy makers are recommended to carefully scrutinize methodological explanations, since they
may otherwise receive incorrect decision support regarding the environmental impacts and resource
requirements for LIBs.

Author Contributions: Conceptualization, A.N., R.A., S.P. and M.C.; methodology, A.N., R.A. and J.T.; formal
analysis, A.N., R.A., M.C. and F.B.d.O.; Writing—original draft preparation, A.N., R.A. and S.P.; Writing—review
and editing, A.N., R.A., M.C., F.B.d.O. and J.T.
Funding: This review was carried out as a joint effort involving participants funded in several different projects
concerning batteries and LCA. The authors would like to express their gratitude for funding to the Swedish
Electromobility Centre, the Swedish Energy Agency, the Swedish Research Council Formas and the Area of
Advance Energy at Chalmers University of Technology.
Conflicts of Interest: The authors declare no conflict of interest.
Batteries 2019, 5, 51 12 of 15

References
1. Whittingham, M.S. History, Evolution, and Future Status of Energy Storage. Proc. IEEE 2012, 100, 1518–1534.
[CrossRef]
2. Zubi, G.; Dufo-López, R.; Carvalho, M.; Pasaoglu, G. The lithium-ion battery: State of the art and future
perspectives. Renew. Sustain. Energy Rev. 2018, 89, 292–308. [CrossRef]
3. Pehlken, A.; Albach, S.; Vogt, T. Is there a resource constraint related to lithium ion batteries in cars? Int. J.
Life Cycle Assess. 2017, 22, 40–53. [CrossRef]
4. Oliveira, L.; Messagie, M.; Rangaraju, S.; Sanfelix, J.; Hernandez Rivas, M.; Van Mierlo, J. Key issues of
lithium-ion batteries – from resource depletion to environmental performance indicators. J. Clean. Prod. 2015,
108, 354–362. [CrossRef]
5. Miedema, J.H.; Moll, H.C. Lithium availability in the EU27 for battery-driven vehicles: The impact of
recycling and substitution on the confrontation between supply and demand until 2050. Resour. Policy 2013,
38, 204–211. [CrossRef]
6. Speirs, J.; Contestabile, M.; Houari, Y.; Gross, R. The future of lithium availability for electric vehicle batteries.
Renew. Sustain. Energy Rev. 2014, 35, 183–193. [CrossRef]
7. Simon, B.; Ziemann, S.; Weil, M. Potential metal requirement of active materials in lithium-ion battery cells
of electric vehicles and its impact on reserves: Focus on Europe. Resour. Conserv. Recycl. 2015, 104, 300–310.
[CrossRef]
8. Kushnir, D.; Sandén, B.A. The time dimension and lithium resource constraints for electric vehicles.
Resour. Policy 2012, 37, 93–103. [CrossRef]
9. Ellingsen, L.A.-W.; Hung, C.R.; Strømman, A.H. Identifying key assumptions and differences in life cycle
assessment studies of lithium-ion traction batteries with focus on greenhouse gas emissions. Transp. Res. D
Transp. Environ. 2017, 55, 82–90. [CrossRef]
10. Peters, J.F.; Weil, M. Providing a common base for life cycle assessments of Li-Ion batteries. J. Clean. Prod.
2018, 171, 704–713. [CrossRef]
11. Peters, J.F.; Baumann, M.; Zimmermann, B.; Braun, J.; Weil, M. The environmental impact of Li-Ion batteries
and the role of key parameters—A review. Renew. Sustain. Energy Rev. 2017, 67, 491–506. [CrossRef]
12. Gaines, L.; Sullivan, J.; Burnham, A.; Belharouak, I. Life-Cycle Analysis of Production and Recycling of
Lithium Ion Batteries. Transp. Res. Rec.: J. Transp. Res. Board 2011, 2252, 57–65. [CrossRef]
13. Gaines, L. To recycle, or not to recycle, that is the question: Insights from life-cycle analysis. MRS Bull. 2012,
37, 333–338. [CrossRef]
14. Zeng, X.; Li, J.; Singh, N. Recycling of Spent Lithium-Ion Battery: A Critical Review. Crit. Rev. Environ. Sci.
Technol. 2014, 44, 1129–1165. [CrossRef]
15. Dewulf, J.; Van der Vorst, G.; Denturck, K.; Van Langenhove, H.; Ghyoot, W.; Tytgat, J.; Vandeputte, K.
Recycling rechargeable lithium ion batteries: Critical analysis of natural resource savings. Resour. Conserv.
Recycl. 2010, 54, 229–234. [CrossRef]
16. Nordelöf, A.; Messagie, M.; Tillman, A.-M.; Ljunggren Söderman, M.; Van Mierlo, J. Environmental impacts
of hybrid, plug-in hybrid, and battery electric vehicles—what can we learn from life cycle assessment? Int. J.
Life Cycle Assess. 2014, 19, 1866–1890. [CrossRef]
17. Kim, H.C.; Wallington, T.J.; Arsenault, R.; Bae, C.; Ahn, S.; Lee, J. Cradle-to-Gate Emissions from a Commercial
Electric Vehicle Li-Ion Battery: A Comparative Analysis. Environ. Sci. Technol. 2016, 50, 7715–7722. [CrossRef]
18. Li, B.; Gao, X.; Li, J.; Yuan, C. Life Cycle Environmental Impact of High-Capacity Lithium Ion Battery with
Silicon Nanowires Anode for Electric Vehicles. Environ. Sci. Technol. 2014, 48, 3047–3055. [CrossRef]
19. Sanfélix, J.; Messagie, M.; Omar, N.; Van Mierlo, J.; Hennige, V. Environmental performance of advanced
hybrid energy storage systems for electric vehicle applications. Appl. Energy 2015, 137, 925–930. [CrossRef]
20. Ellingsen, L.A.-W.; Majeau-Bettez, G.; Singh, B.; Srivastava, A.K.; Valøen, L.O.; Strømman, A.H. Life Cycle
Assessment of a Lithium-Ion Battery Vehicle Pack. J. Ind. Ecol. 2014, 18, 113–124. [CrossRef]
21. Dunn, J.B.; Gaines, L.; Sullivan, J.; Wang, M.Q. Impact of Recycling on Cradle-to-Gate Energy Consumption
and Greenhouse Gas Emissions of Automotive Lithium-Ion Batteries. Environ. Sci. Technol. 2012, 46,
12704–12710. [CrossRef] [PubMed]
22. Saw, L.H.; Ye, Y.; Tay, A.A.O. Integration issues of lithium-ion battery into electric vehicles battery pack.
J. Clean. Prod. 2016, 113, 1032–1045. [CrossRef]
Batteries 2019, 5, 51 13 of 15

23. Kushnir, D. Lithium Ion Battery Recycling Technology 2015. In Current State and Future Prospects; ESAReport
# 2015:18; Chalmers University of Tehcnology: Gothenburg, Sweden, 2015.
24. Huisman, J.; Leroy, P.; Tertre, F.; Söderman, M.L.; Chancerel, P.; Cassard, D.; Løvik, A.N.; Wäge, P.; Kushnir, D.;
Rotter, V.S. Prospecting Secondary Raw Materials in the Urban Mine and Mining Wastes (ProSUM)—Final
Report. 21 December 2017: Brussels, Belgium. ISBN: 978-92-808-9061-7 (electronic). Available online:
http://www.prosumproject.eu/sites/default/files/DIGITAL_Final_Report.pdf. (accessed on 28 March 2019).
25. Melin, H.E. Forskningsöversikt om återvinning och återbruk av litiumjonbatterier; Circular Energy Storage,
Swedish Energy Agency: Eskilstuna, Sweden. Available online: http://www.energimyndigheten.se/
globalassets/forskning-innovation/overgripande/forskningsoversikt-om-atervinning-och-aterbruk-av-
litiumjonbatterier-2019.pdf. (accessed on 29 March 2019).
26. Zheng, X.; Zhu, Z.; Lin, X.; Zhang, Y.; Hec, Y.; Cao, H.; Sun, Z. A Mini-Review on Metal Recycling from Spent
Lithium Ion Batteries. Engineering 2018, 4, 361–370. [CrossRef]
27. Lv, W.; Wang, Z.; Cao, H.; Sun, Y.; Zhang, Y.; Sun, Z. A Critical Review and Analysis on the Recycling of
Spent Lithium-Ion Batteries. ACS Sustain. Chem. Eng. 2018, 6, 1504–1521. [CrossRef]
28. Andersson, M.; Ljunggren Söderman, M.; Sandén, B.A. Are scarce metals in cars functionally recycled?
Waste Manag. 2017, 60, 407–416. [CrossRef] [PubMed]
29. Gaines, L. The future of automotive lithium-ion battery recycling: Charting a sustainable course. Sustain.
Mater. Technol. 2014, 1–2, 2–7. [CrossRef]
30. Ekberg, C.; Petranikova, M. Chapter 7—Lithium Batteries Recycling. In Lithium Process Chemistry; Chagnes, A.,
Światowska, J., Eds.; Elsevier: Amsterdam, The Netherlands, 2015; pp. 233–267.
31. Dunn, J.B.; Gaines, L.; Barnes, M.; Sullivan, J.; Wang, M. Material and Energy Flows in the Materials
Production, Assembly, and End-of-Life Stages of the Automotive Lithium-Ion Battery Life Cycle. In Energy
Systems Division, Argonne National Laboratory; ANL/ESD/12-3 Rev; U.S. Department of Energy: Argonne, IL,
USA, 2014.
32. Ekvall, T.; Tillman, A.-M. Open-loop recycling: Criteria for allocation procedures. Int. J. Life Cycle Assess.
1997, 2, 155. [CrossRef]
33. Guérin, B.L.E. Modelling of Recycling in Life Cycle Assessment. Master’s Thesis, Department of
Environmental Engineering, Technical University of Denmark, Copenhagen, Denmark, 2017.
34. Atherton, J. Declaration by the Metals Industry on Recycling Principles. Int. J. Life Cycle Assess. 2007, 12,
59–60. [CrossRef]
35. Frischknecht, R. LCI modelling approaches applied on recycling of materials in view of environmental
sustainability, risk perception and eco-efficiency. Int. J. Life Cycle Assess. 2010, 15, 666–671. [CrossRef]
36. Zackrisson, M.; Avellán, L.; Orlenius, J. Life cycle assessment of lithium-ion batteries for plug-in hybrid
electric vehicles – Critical issues. J. Clean. Prod. 2010, 18, 1519–1529. [CrossRef]
37. Wernet, G.; Bauer, C.; Steubing, B.; Reinhard, J.; Moreno-Ruiz, E.; Weidema, B. The ecoinvent database
version 3 (part I): overview and methodology. Int. J. Life Cycle Assess. 2016, 21, 1218–1230. [CrossRef]
38. ISO. Environmental Management—Life Cycle Assessment—Requirements and Guidelines; ISO 14044:2006;
International Organization for Standardization: Geneva, Switzerland, 2016; p. 46.
39. Classen, M.; Althaus, H.-J.; Blaser, S.; Scharnhorst, W.; Tuchschmid, M.; Jungbluth, N.; Emmenegger, M.F.
Life Cycle Inventories of Metals; Ecoinvent report No. 10. (Data v2.1); Swiss Centre for Life Cycle Inventories:
Dübendorf, Switzerland, 2009.
40. Baumann, H.; Tillman, A.-M. The Hitch Hiker’s Guide to LCA—An Orientation in Life Cycle Assessment
Methodology and Application; Studentlitteratur: Lund, Sweden, 2004.
41. Finnveden, G.; Hauschild, M.Z.; Ekvall, T.; Guinée, J.; Heijungs, R.; Hellweg, S.; Koehler, A.; Pennington, D.;
Suh, S. Recent developments in Life Cycle Assessment. J. Environ. Manag. 2009, 91, 1–21. [CrossRef]
[PubMed]
42. Schrijvers, D.L.; Loubet, P.; Sonnemann, G. Critical review of guidelines against a systematic framework
with regard to consistency on allocation procedures for recycling in LCA. Int. J. Life Cycle Assess. 2016, 21,
994–1008. [CrossRef]
43. Vadenbo, C.; Hellweg, S.; Astrup, T.F. Let’s Be Clear(er) about Substitution: A Reporting Framework to
Account for Product Displacement in Life Cycle Assessment. J. Ind. Ecol. 2017, 21, 1078–1089. [CrossRef]
44. Geyer, R.; Kuczenski, B.; Zink, T.; Henderson, A. Common Misconceptions about Recycling. J. Ind. Ecol.
2016, 20, 1010–1017. [CrossRef]
Batteries 2019, 5, 51 14 of 15

45. Scopus. Online Subscription-Based Scientific Multi-Disciplinary Database 2019. Available online: https:
//www.scopus.com (accessed on 27 February 2019).
46. Web of Science. Online Subscription-Based Aggregated Scientific Multi-Disciplinary Database. 2019.
Available online: http://webofknowledge.com/WOS (accessed on 27 February 2019).
47. Wang, Y.; Yu, Y.; Huang, K.; Chen, B.; Deng, W.; Yao, Y. Quantifying the environmental impact of a Li-rich
high-capacity cathode material in electric vehicles via life cycle assessment. Environ. Sci. Pollut. Res. 2017,
24, 1251–1260. [CrossRef] [PubMed]
48. Yu, Y.; Chen, B.; Huang, K.; Wang, X.; Wang, D. Environmental Impact Assessment and End-of-Life Treatment
Policy Analysis for Li-Ion Batteries and Ni-MH Batteries. Int. J. Environ. Res. Public Health 2014, 11, 3185–3198.
[CrossRef] [PubMed]
49. Casals, L.C.; García, B.A.; Aguesse, F.; Iturrondobeitia, A. Second life of electric vehicle batteries: relation
between materials degradation and environmental impact. Int. J. Life Cycle Assess. 2017, 22, 82–93. [CrossRef]
50. Bobba, S.; Mathieux, F.; Ardente, F.; Blengini, G.A.; Cusenza, M.A.; Podias, A.; Pfrang, A. Life Cycle
Assessment of repurposed electric vehicle batteries: An adapted method based on modelling energy flows.
J. Energy Storage 2018, 19, 213–225. [CrossRef]
51. Argonne National Laboratory. GREET® Model—The Greenhouse Gases, Regulated Emissions, and Energy
Use in Transportation Model. 2019. Available online: https://greet.es.anl.gov (accessed on 12 March 2019).
52. Sanfélix, J.; de la Rúa, C.; Schmidt, J.H.; Messagie, M.; Van Mierlo, J. Environmental and economic performance
of an li-ion battery pack: A multiregional input-output approach. Energies 2016, 9, 584. [CrossRef]
53. Vandepaer, L.; Cloutier, J.; Bauer, C.; Amor, B. Integrating Batteries in the Future Swiss Electricity Supply
System: A Consequential Environmental Assessment. J. Ind. Ecol. 2018, 23, 709–725. [CrossRef]
54. Ahmadi, L.; Young, S.B.; Fowler, M.; Fraser, R.A.; Achachlouei, M.A. A cascaded life cycle: reuse of electric
vehicle lithium-ion battery packs in energy storage systems. Int. J. Life Cycle Assess. 2017, 22, 111–124.
[CrossRef]
55. Ciez, R.E.; Whitacre, J.F. Examining different recycling processes for lithium-ion batteries. Nat. Sustain. 2019,
2, 148–156. [CrossRef]
56. Cusenza, M.A.; Bobba, S.; Ardente, F.; Cellura, M.; Di Persio, F. Energy and environmental assessment of
a traction lithium-ion battery pack for plug-in hybrid electric vehicles. J. Clean. Prod. 2019, 215, 634–649.
[CrossRef] [PubMed]
57. de Souza, L.L.P.; Lora, E.E.S.; Palacio, J.C.E.; Rocha, M.H.; Renó, M.L.G.; Venturini, O.J. Comparative
environmental life cycle assessment of conventional vehicles with different fuel options, plug-in hybrid
and electric vehicles for a sustainable transportation system in Brazil. J. Clean. Prod. 2018, 203, 444–468.
[CrossRef]
58. Dunn, J.B.; Gaines, L.; Kelly, J.C.; James, C.; Gallagher, K.G. The significance of Li-ion batteries in electric
vehicle life-cycle energy and emissions and recycling’s role in its reduction. Energy Environ. Sci. 2015, 8,
158–168. [CrossRef]
59. Faria, R.; Marques, P.; Garcia, R.; Moura, P.; Freire, F.; Delgado, J.; de Almeida, A.T. Primary and secondary
use of electric mobility batteries from a life cycle perspective. J. Power Sources 2014, 262, 169–177. [CrossRef]
60. Hendrickson, T.P.; Kavvada, O.; Shah, N.; Sathre, R.; Scown, C.D. Life-cycle implications and supply chain
logistics of electric vehicle battery recycling in California. Environ. Res. Lett. 2015, 10, 014011. [CrossRef]
61. Notter, D.A.; Gauch, M.; Widmer, R.; Wäger, P.; Stamp, A.; Zah, R.; Althaus, H.-J. Contribution of Li-Ion
Batteries to the Environmental Impact of Electric Vehicles. Environ. Sci. Technol. 2010, 44, 6550–6556.
[CrossRef]
62. Oliveira, L.; Messagie, M.; Mertens, J.; Laget, H.; Coosemans, T.; Van Mierlo, J. Environmental performance
of electricity storage systems for grid applications, a life cycle approach. Energy Convers. Manag. 2015, 101,
326–335. [CrossRef]
63. Raugei, M.; Winfield, P. Prospective LCA of the production and EoL recycling of a novel type of Li-ion
battery for electric vehicles. J. Clean. Prod. 2019, 213, 926–932. [CrossRef]
64. Richa, K.; Babbitt, C.W.; Nenadic, N.G.; Gaustad, G. Environmental trade-offs across cascading lithium-ion
battery life cycles. Int. J. Life Cycle Assess. 2017, 22, 66–81. [CrossRef]
65. Ryan, N.A.; Lin, Y.; Mitchell-Ward, N.; Mathieu, J.L.; Johnson, J.X. Use-Phase Drives Lithium-Ion Battery
Life Cycle Environmental Impacts When Used for Frequency Regulation. Environ. Sci. Technol. 2018, 52,
10163–10174. [CrossRef] [PubMed]
Batteries 2019, 5, 51 15 of 15

66. Swart, P.; Dewulf, J.; Biernaux, A. Resource demand for the production of different cathode materials for
lithium ion batteries. J. Clean. Prod. 2014, 84, 391–399. [CrossRef]
67. Tagliaferri, C.; Evangelisti, S.; Acconcia, F.; Domenech, T.; Ekins, P.; Barletta, D.; Lettieri, P. Life cycle assessment
of future electric and hybrid vehicles: A cradle-to-grave systems engineering approach. Chem. Eng. Res. Des.
2016, 112, 298–309. [CrossRef]
68. Unterreiner, L.; Jülch, V.; Reith, S. Recycling of Battery Technologies – Ecological Impact Analysis Using Life
Cycle Assessment (LCA). Energy Procedia 2016, 99, 229–234. [CrossRef]
69. Vandepaer, L.; Cloutier, J.; Amor, B. Environmental impacts of Lithium Metal Polymer and Lithium-ion
stationary batteries. Renew. Sustain. Energy Rev. 2017, 78, 46–60. [CrossRef]
70. Zhao, S.; You, F. Comparative Life-Cycle Assessment of Li-Ion Batteries through Process-Based and Integrated
Hybrid Approaches. ACS Sustain. Chem. Eng. 2019, 7, 5082–5094. [CrossRef]
71. Richa, K.; Babbitt, C.W.; Nenadic, N.G.; Gaustad, G. Supplementary Material for Environmental Trade-Offs Across
Cascading Lithium-Ion Battery Life Cycles; Golisano Institute for Sustainability, Rochester Institute of Technology:
Rochester, NY, USA, 1 January 2017; p. 35. Available online: https://doi.org/10.1007/s11367-015-0942-3
(accessed on 12 March 2019).
72. European Union. Directive 2006/66/EC of the European Parliament and of the Council; Official Journal of the
European Union: Luxembourg, 2006.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).