Renewable and Sustainable Energy Reviews 192 (2024) 114191

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Review article

Challenges and opportunities for second-life batteries: Key technologies and

economy
Xubo Gu a , Hanyu Bai a , Xiaofan Cui b , Juner Zhu c , Weichao Zhuang d , Zhaojian Li e ,
Xiaosong Hu f , Ziyou Song a ,∗
a
Department of Mechanical Engineering, National University of Singapore, Singapore, 117575, Singapore
b Department of Electrical and Computer Engineering, University of California, Los Angeles, Los Angeles, CA, 90095, USA
c Department of Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA
d
School of Mechanical Engineering, Southeast University, Nanjing, 211189, China
e
Department of Mechanical Engineering, Michigan State University, East Lansing, MI, 48824, USA
f
Department of Mechanical and Vehicle Engineering, Chongqing University, Chongqing, 400044, China

ARTICLE INFO ABSTRACT

Keywords: Due to the increasing volume of electric vehicles in automotive markets and the limited lifetime of onboard
Second-life battery lithium-ion batteries, the large-scale retirement of batteries is imminent. The battery packs retired from electric
Degradation model vehicles still own 70%–80% of the initial capacity, thus having the potential to be utilized in scenarios
Battery regrouping
with lower energy and power requirements to maximize the value of batteries. However, spent batteries
Economy analysis
are commonly less reliable than fresh batteries due to their degraded performance, thereby necessitating a
Battery management system
comprehensive assessment from safety and economic perspectives before further utilization. To this end, this
paper reviews the key technological and economic aspects of second-life batteries (SLBs). Firstly, we introduce
various degradation models for first-life batteries and identify an opportunity to combine physics-based theories
with data-driven methods to establish explainable models with physical laws that can be generalized. However,
degradation models specifically tailored to SLBs are currently absent. Therefore, we analyze the applicability
of existing battery degradation models developed for first-life batteries in SLB applications. Secondly, we
investigate a representative process of dealing with retired batteries and discuss the regrouping standards for
cell-to-cell variation for the first time to guide the classification procedure and enhance the performance and
safety of SLBs. Thirdly, we scrutinize the economic analysis of SLBs and summarize the potentially profitable
applications. Finally, we comprehensively examine and compare power electronics technologies that can
substantially improve the performance of SLBs, including high-efficiency energy transformation technologies,
active equalization technologies, and technologies to improve reliability and safety.

1. Introduction
Electric vehicles (EVs) have become increasingly popular in recent
years. In 2022, global sales of EVs reached 10.5 million units, account-
ing for 13% of total light vehicle sales. This represents a 55% increase
from 2021, indicating a strong growth trend [1]. Most commercial EVs
adopt lithium-ion batteries (LIBs) because of their excellent properties,
such as high energy density and high power density. Typically, the
lifespan of the LIB pack in an EV is around 8–10 years, after which
the battery is retired when its remaining capacity decreases to 70%–
80% of its initial value. Due to the high volume of EVs being in
service and the limited lifespan of LIBs, a significant volume of retired
batteries is expected in the near future. To reduce the cost of EVs
and mitigate their environmental impacts, the retired LIBs should be
reused and ultimately recycled. These retired batteries can still retain
70%–80% of their original capacity and can be utilized in scenarios
with lower energy and power requirements, such as energy storage
stations or communication base stations [2]. In this way, the value of
LIBs can be maximized in their second-life applications. Meanwhile,
the upfront cost of EVs will be reduced since car owners can recover
some of their value from selling their retired batteries. The metals and
electrolytes in LIBs would cause heavy metal and chemical pollution
to the environment if they were directly discarded [3]. Hence, it is
important to recycle LIBs after their echelon utilization to extract
valuable materials for further battery production. The development
of an effective echelon utilization and recycling system is crucial to

∗ Corresponding author.
E-mail address: ziyou@nus.edu.sg (Z. Song).

https://doi.org/10.1016/j.rser.2023.114191
Received 19 March 2023; Received in revised form 14 October 2023; Accepted 9 December 2023
Available online 26 December 2023
1364-0321/© 2023 Elsevier Ltd. All rights reserved.
X. Gu et al. Renewable and Sustainable Energy Reviews 192 (2024) 114191

(FLBs) has been decreasing in recent years [5,6] and the refurbish-
Nomenclature ment cost of SLBs is still high [7,8]. The economic viability of SLBs
depends on various factors, including technical performance (i.e., bat-
List of Abbreviations tery degradation), which directly influences the service life of SLBs,
AC Alternating Current refurbishment costs depending on the size of SLB packs, and potential
revenues for all stakeholders (e.g., SLB sellers and buyers), which can
BESS Battery Energy Storage System
vary between countries due to different policies. Finally, SLBs have
COE Cost of Energy
lower power efficiency, heterogeneous electrical characteristics, and
DC Direct Current lower reliability compared to new batteries. Therefore, it is essential to
DDM Data-driven Model design appropriate power electronics for SLB applications. Power elec-
DSPV Distributed Solar Photovoltaic tronics are the key energy conversion interfaces that can improve the
EOL End-of-life performance of SLBs. However, designing power electronics for SLBs
ESS Energy Storage System can be challenging, particularly for large SLB packs, due to significant
EV electric vehicle cell-to-cell variation.
FLB First-life Battery In a comprehensive review by Li et al. [9], the SLB market was ex-
FPP Full Power Processing tensively evaluated from technical, economic, and policy perspectives.
The authors discussed technical topics such as degradation models, the
LAMNE Loss of Anode Active Material
aging knee, and regrouping strategies, and summarized the economic
LAMPE Loss of Cathode Active Material
performance of SLBs in practical applications. However, this review did
LCOE Levelized Cost of Electricity
not cover several key issues, including the specific application of hybrid
LCOS Levelized Cost of Storage physics-based and data-driven approaches for SLBs to characterize
LIB Lithium-ion Battery their degradation, the formulation of a grouping standard to balance
LLI Loss of Lithium Inventory refurbishment costs and SLB performance, an updated economic assess-
MMC Multilevel Modular Converters ment of SLBs that highlights more recent results, and a comprehensive
NE Negative Electrode introduction to the advanced power electronics techniques that have
NPV Net Present Value been developed for SLBs. To this end, we aim to address these issues in
OEM original equipment manufacturer this review.
The novelty and contributions of this review are as follows:
PBM Physics-based Model
(1) Key technological and economic aspects of SLBs are comprehen-
PDE Partial Differential Equation
sively reviewed.
PE Positive Electrode
(2) The applicability of existing battery degradation models devel-
PINN Physics-informed Neural Network oped for FLBs in the context of SLB applications is analyzed.
PPP Partial Power Processing (3) Comprehensive processes, state-of-the-art, challenges along with
PV Photovoltaic potential solutions, and regulations and standards revolving around
SEI Solid Electrolyte Interphase the whole assessment procedure of retired batteries are summarized.
SLB Second-life Battery For the first time, the regrouping standard of cell-to-cell variation for
SOH State of Health SLBs is discussed to guide the classification procedure and enhance
performance and safety.
(4) The economic feasibility of SLBs is discussed, where potentially
profitable applications are summarized, detailed modeling approaches
support the sustainable growth of the EV industry and has broad are scrutinized, and uncertainties and underlying assumptions are ana-
societal significance worldwide. lyzed. (5) Power electronics technologies that can substantially improve
However, the effective utilization of second-life batteries (SLBs) is a the performance of SLBs are comprehensively examined and compared.
multifaceted problem. Firstly, the determination of SLB’s internal status The remaining sections of this work are organized as follows. We
is complicated. The status of SLBs consists of the internal state (e.g., SEI first introduce various degradation mechanisms and models in Sec-
layer and lithium plating), the external characteristics (e.g., capacity tion 2. In Section 3, the comprehensive process of dealing with re-
and power capability), and the remaining life expectancy. Note that tired batteries is presented, and the regrouping standard for cell-to-cell
measuring and characterizing the internal changes of SLBs is gen- variation is first discussed. In Section 4, the economic benefits of
SLBs in different applications are reviewed and classified from the
erally not straightforward, as different SLBs from various EVs have
aspects of understanding the potential benefits of SLBs, economic as-
experienced different operational conditions, resulting in varying in-
sessment strategies with critical assumptions, integration of battery
ternal statuses in anodes, electrolytes, cathodes, and current collectors,
degradation modeling, representative economic indicators, and the
thereby making it difficult to establish common testing protocols to
importance of market and government incentives. In Section 5, we
assess the status of SLBs. For example, the used battery cells with
discuss advanced power electronics technologies that enable the battery
the same capacity may have significantly different remaining life due
management system for SLBs. In Section 6, the conclusions of this work
to various dominant degradation mechanisms inside. Nondestructive
are summarized.
detection methods, such as in-situ neutron diffraction, have been used
in academia [4]. However, their widespread application in large-scale 2. Degradation model
retired LIBs can be costly. Secondly, reusing a large number of re-
tired batteries will lead to enormous demand for the disassembly and Main degradation mechanisms and models for FLBs have been
sorting of heterogeneous battery packs/modules/cells, which can be extensively studied. Fig. 1 shows three types of degradation models:
labor-intensive and expensive. Moreover, sorting batteries becomes physics-based models (PBMs), data-driven models (DDMs), and hybrid
challenging due to the loss of historical data, which necessitates exten- models. Specifically, PBMs are explainable but may not be applicable
sive new testing, increasing the cost of SLBs. Thirdly, from an economic when degradation pathways are poorly understood. In contrast, DDMs
point of view, the profitability of SLBs in different applications still rely on high-quality data and bypass mechanisms but are not inter-
needs to be determined, provided that the price of first-life batteries pretable. The hybrid model leverages the advantages of both PBMs and

2
X. Gu et al.
Fig. 1. Schematic illustration of Section 2 on battery degradation models.
DDMs and has the potential to establish an explainable model with
physical laws that can be generalized. The missing historical data over
the first-life operation and the unknown knee point have posed notice-
able challenges to accurately characterizing the degradation behaviors
of SLBs. Currently, the degradation models particularly focusing on
SLBs are still absent due to their unknown internal health status, which
indicates the interplay of various aging mechanisms (e.g., SEI growth
and lithium plating) and cannot be easily characterized using individual
parameters (e.g., resistance or capacity). This section introduces the
existing battery degradation models for FLBs and investigates their
potential applicability in SLBs after appropriate modifications. With
sufficient experimental data, hybrid models offer decent opportuni-
ties to reveal hidden mechanisms in SLBs while maintaining good
interpretability.
2.1. Introduction of degradation models
The battery can be regarded as a dynamic electrochemical sys-
tem including intricate internal reactions. The system will gradually
deteriorate as the cycle number increases, caused by various mecha-
nisms, such as physical stresses and chemical side reactions. Common
battery degradation mechanisms can be classified into three main
modes [10,11]: (1) Loss of lithium inventory (LLI). Lithium ions in
the battery are consumed by surface film formation (SEI) growth,
lithium plating, decomposition reactions, etc. These consumed lithium-
ions cannot transport between the positive electrode (PE) and the
negative electrode (NE) anymore, leading to a reduction in capacity.
Another manner in which lithium-ion loss occurs is through trapping
in isolated active material particles; (2) Loss of anode active material
(LAMNE ) and Loss of cathode active material (LAMPE ). Particle crack-
ing and loss of electric contact can lead to LAMNE , which increases
the difficulty of lithium insertion into the active materials of NE.
Particle cracking, structural disordering, and loss of electric contact
can lead to LAMPE , making it difficult for lithium to enter the active
materials of PE; (3) Loss of electrolyte. Loss of electrolyte results
from electrolyte consumption due to side reactions like SEI formation
and thickening, as well as electrolyte decomposition caused by high
potential or high temperature. The loss of electrolytes will lead to an
increase in resistance and affect the capacity and power capability of
3
Renewable and Sustainable Energy Reviews 192 (2024) 114191
the battery. Significant electrolyte loss may even result in a considerate
capacity drop [11]. In addition to these intricate internal changes inside
the battery, the operation conditions (e.g., temperatures and current
amplitudes) also strongly affect the degradation rate of the battery
and may induce different aging mechanisms. These factors make it a
challenging problem to quantify battery degradation under different
working conditions.
Given the challenge of tracing degradation, its modeling has been
an active research area, with significant studies available in the litera-
ture [12]. Various kinds of degradation models can be summarized into
three main categories:
• PBMs. PBMs explore the fundamental processes in the battery to
trace the degradation pathway. For SEI formation, efforts can be
seen in the literature to describe its creation from macroscopic
levels [13] and molecular levels, such as kinetic Monte Carlo
simulation [14], molecular dynamics [15], and density functional
theory [16]. Considering both SEI growth and lithium plating,
Yang et al. [17] built an electrochemical–thermal model and
explained that SEI growth should be responsible for the linear
degradation and lithium plating causes the nonlinear aging trends
in the second stage. Particle fracture is also a typical physical
process inside batteries, which exposes an extra surface area
for further SEI growth. O’Kane et al. [18] established a PBM
that incorporates these three mechanisms and showed that LLI
significantly increases when particle cracking is considered.
PBMs offer several natural advantages. Firstly, degradation mod-
els provide valuable insights into battery aging owing to the
well-understood internal processes. For instance, Yang et al. [17]
demonstrated that an internal mechanism such as SEI growth or
lithium plating could dominate each degradation phase. Secondly,
the knowledge gained from PBMs can be leveraged for future en-
gineering design. For example, Han et al. [11] demonstrated that
the internal side reactions were directly influenced by the battery
design. By integrating physical models into the design optimiza-
tion, it is possible to mitigate internal side reactions and enhance
the battery lifespan. Thirdly, this kind of method can well extrap-
olate to other cases if the dominant aging mechanism remains the
same. However, degradation characterization for SLBs can still be
X. Gu et al.
challenging as the electrochemical states/parameters are difficult
to measure and estimate [19], especially when the historical data
is missing, leaving the degradation pathway over the first-life
application largely unknown.
• DDMs. DDMs take advantage of historical data to train a model to
predict degradation trends. Machine learning frameworks involv-
ing models like Naive Bayes [20], Support Vector Regression [21],
Gaussian Process Regression [22,23], and Neural Networks [24,
25] were well practiced. Severson et al. [26] collected a com-
prehensive dataset of 124 cells and built a feature-based linear
model, which can predict the remaining life using the first 100
cycles with an error of 9.1%. Using the same data, Hsu et al. [27]
developed a deep neural network that can predict the remaining
life by the first cycle in a lower test error of 6.46%, indicating the
promising performance of the extracted features.
DDMs have shown great potential to predict battery degradation
based on available high-quality data. Such datasets, however, are
challenging to acquire in the academic community because of
limited resources. Besides, DDMs require parameter calibration
or training for new batteries, and the extracted features can be
challenging to interpret, even with high prediction accuracy.
In addition, the empirical model is also established based on
data along with engineering experience. This kind of model pro-
duces a regression model that represents the degradation trend
considering several important factors, such as time, tempera-
ture, C-rate [28,29], internal resistance [30], battery discharge
curve [31], etc. Empirical models are relatively simple and are
often adopted in the literature to evaluate the remaining value
for cells [32], as will be discussed in Section 4.
• Hybrid models. Hybrid approaches aim to combine the strengths
of both PBMs and DDMs to better predict battery lifetime. The
reviews in [33,34] discussed various integration structures for
predicting battery lifetime, which can be broadly divided into
serial integration structures and compact integration structures,
as shown in Fig. 2. For the serial integration structures, one
approach is residual learning, where DDM can learn the residuals
between the PBM outputs and the experimental outputs, as shown
in structure A1 of Fig. 2. Park et al. [35] integrated a recurrent
neural network with the single particle model to recover model-
measurement mismatch and improve the prediction accuracy of
the output voltage. Another approach is transfer learning, where
the DDM learns from the data generated by a PBM and a few
experimental data, as shown in structure A2 of Fig. 2. In this
structure, the PBM performs as the data generator to reduce the
dependency of a DDM on experimental data. Li et al. [36] used an
experimentally verified pseudo 2D (P2D) model to generate much
data to train a deep neural network for estimating battery states.
Parameter identification is also a typical hybrid structure, where
the DDM estimates the best parameters for a PBM to explain the
experimental data, as shown in structure A3. Reports of such an
approach can be widely seen in the literature. The work by Li
et al. [37] applied the cuckoo search algorithm to identify 26
parameters for the P2D model through a two-step identification
process and a multi-objective fitness function.
Besides the three serial integration structures, PBMs and DDMs
can be integrated into compact structures, as shown in structures
B1 and B2. Specifically, structure B1 is typically accomplished
either by a physics-guided loss function and/or model architec-
ture on the basis of the DDM, while in structure B2 part of the
physical model (usually costly or inaccurate) is replaced with a
DDM. The PINN is an example of structure B2, the essence of
which is to solve partial differential equations (PDEs) using deep
learning trained by automatic differentiation [38].
By incorporating information on physical laws, boundary con-
ditions, and initial conditions, PINN can effectively solve PDEs,
and it offers several advantages over traditional methods, such as
4
Renewable and Sustainable Energy Reviews 192 (2024) 114191
being meshless and able to solve PDEs for all possible parameters,
fast inference, and the ability to solve inverse problems. Zubov
et al. [39] conducted an initial investigation of the use of PINN
to solve the single particle model and the reduced-order P2D
model. They obtained satisfactory results for most variables but
experienced large errors for certain state variables, such as elec-
trolyte potential. However, in light of the rapid advancements in
this field and the emergence of more effective training methods,
it is promising to see the PINN solution for a full P2D model
across a broad range of parameters. As the battery degrades, the
parameters of the governing equations will change as well, which
can make the degradation model more complex. Since no related
research has been found yet to exploit the adaptive PINN as degra-
dation models, this can be a future research direction in terms of
organically leveraging both explainable degradation mechanisms
and learning capabilities for high-dimensional correlations with
limited degradation data.
2.2. Degradation models for SLBs
While substantial degradation models have been provided in the lit-
erature, there is still a dearth of studies focused on degradation models
specifically designed for SLBs. This is a more challenging task for two
reasons: firstly, the lack of historical data for first-life applications, and
secondly, the uncertainty surrounding the ‘‘knee point’’. This section
provides a comprehensive overview of these issues.
2.2.1. First-life operation data
The historical operation data of SLBs over their first-life applications
significantly affect how the battery ages in second-life applications.
For example, the fast charging in a cold environment will cause sig-
nificant lithium plating, which leads to capacity loss and even takes
the risk of thermal runaway caused by lithium dendrites penetrating
the separator [41]. The research in [42] explored the effects of work-
ing conditions on battery degradation throughout its entire lifetime.
Through experimentation on multiple 18650-type cells at two different
temperatures (0 ◦ C and 45 ◦ C), it was observed that lithium plating
occurred at 0 ◦ C/0.5 C-rate, while SEI growth and LAMNE occurred at
45 ◦ C/0.5 C-rate. These findings suggest that the internal degradation
mechanism of the battery was significantly impacted by the operating
condition.
Although historical data is critical for understanding the degrada-
tion pathway of SLBs, it is often unavailable after they are disassembled
from EV battery packs. Note that the simple parameters (e.g., remaining
capacity and internal resistance) are not enough to well describe how
the battery has aged. For example, two battery cells, both with 80%
remaining capacities, may have totally different remaining lifetimes
in their second-life applications due to different internal statuses after
the first life. A potential research direction to address this issue is
to develop simple testing protocols, which can be used to provide
informative aging information on SLBs by estimating and assessing the
key parameters related to various aging mechanisms.
2.2.2. Knee point
The degradation curve of a cell can be drawn as the change in
capacity over cycle number, as depicted in Fig. 3. During the early
stages of operation, the curve will exhibit a linear trend. However,
beyond a certain point, the performance of the cell will experience a
sharp decline, and the transition point from a mostly linear region to
a highly non-linear region is commonly referred to as the knee point.
After the knee point, the battery capacity will rapidly decrease, and
severe safety issues may arise. The internal mechanisms responsible for
the knee are convoluted and not yet fully understood.
X. Gu et al.
Fig. 2. Battery hybrid degradation model. Serial structures: A1 - residual learning, A2 - transfer learning, A3 - parameter identification; compact structures: B1 - physics-guided
model, B2 - data-driven assisted model.
Source: Recreated from [40].
Fig. 3. The knee point during battery degradation.
The review by Attia et al. [43] covered 303 cells from 17 datasets
and calculated the relationship between the knee point and the EOL
point (80% of nominal capacity), which can be represented by Eq. (1).
𝑦 = 0.984𝑥 + 93.918 (𝑅2 = 0.874)
where 𝑥 represents the cycle number to the knee point and 𝑦 is the cycle
number to EOL. The equation has shown a clear linear relationship and
demonstrated that the occurrence of the knee is always accompanied
by the EOL point, i.e., the retired batteries take higher risks of encoun-
tering the knee point. For safe SLB usage, more light should be shed on
accurate knee point prediction. However, a comprehensive framework
with accurate models for predicting knees of battery cells with various
chemistries is still missing. In [43], the failure modes leading to knees
were categorized into six pathways: lithium plating, electrode sat-
uration, percolation-limited connectivity, resistance growth, additive
depletion, and mechanical deformation. Meanwhile, the knees can
be classified into one of three ‘‘internal state trajectories’’: snowball,
hidden, and threshold. Because each mechanism is represented by
a different pair of pathway-trajectory, the problem of modeling and
predicting knees is extremely challenging.
2.2.3. Applicability of the existing degradation models for SLBs
As mentioned above, it is still unclear whether the degradation
models mentioned above for FLBs are still applicable to SLBs. The DDM,
Renewable and Sustainable Energy Reviews 192 (2024) 114191
due to its nature of fitting an approximate function and bypassing
mechanisms, could still perform well if specific lab test datasets are
established for SLBs. Some research like [44] can be conducted for
SLBs to build degradation models by data-driven approaches. Besides,
such methods will greatly benefit from open-source datasets. However,
very limited data about SLBs can be available online, which hampers
the development of effective DDMs for SLBs. It is recommended that
academia and industry make valuable datasets open-source to provide
abundant data for degradation modeling in SLBs (similar to current
famous FLB datasets, including but not limited to NASA datasets [45,
46], CALCE datasets [47], Stanford/Toyota datasets [44,48], Oxford
battery degradation datasets [49], Panasonic 18650PF Li-ion battery
datasets [50], and NCM/NCA datasets [51]).
For PBMs, the prerequisite is understanding the dominant mech-
anism responsible for the degradation pathway over the first life.
Therefore, a simple testing protocol, which can effectively distinguish
the degradation mechanisms for SLBs, would be useful. Some modifi-
cations are required to enable the implementation of PBMs in SLBs.
As for hybrid models, it remains unclear whether structures A1–A3
can accurately capture the SLB degradation trend, as there is a lack
of research in this area, while the use of new SLB experimental data
could potentially mitigate prediction errors caused by unknown inter-
nal mechanisms. To reveal unknown mechanisms in SLBs, a hybrid
(1)
model with the B2 structure (e.g., PINN) can be applied to solve the
inverse problem by leveraging observation data. Finegan et al. out-
lined PINN approaches to extracting physical insights from cell failure
events, which can improve the safety of batteries [52]. Thus, hybrid
models show promise as a solution for better understanding the internal
mechanisms of SLB degradation.
3. Technical assessment procedures for retired batteries
The potential availability of second-life batteries is significant. Ac-
cording to the joint report by McKinsey and the Global Battery Alliance,
the projections estimate the global supply of second-life batteries will
reach 15 GWh by 2025 and further increase to 112–227 GWh by
2030 [53]. Besides, McKinsey also reported that the global demand for
Li-ion batteries is expected to skyrocket in the next decade [54]. The
demand is projected to rise from about 700 GWh in 2022 to around 4.7
TWh by 2030. The majority of this demand in 2030, approximately 4.3
TWh, will be for mobility applications like EVs. This indicates a greater
potential supply of second-life batteries in the next decade (2030 -). The
enormity of these figures underscores the urgency in devising strategies
5
X. Gu et al.
Fig. 4. Schematic illustration of Section 3 on the technical assessment procedures for retired batteries.
for the cost-effective reutilization of these batteries. Thus, a technical
assessment procedure for retired batteries is imperative. Given that
most of the retired Li-ion battery packs will come from EVs, this section
will mainly discuss the technical procedure in the context of retired EV
batteries.
The unified assessment procedure of retired lithium-ion batteries
can be difficult, considering that the form factor, design, and chemistry
of the existing modules and cell degradation vary greatly from one
another. Zhu et al. [55] summarized the common features of technical
procedure of dealing with various battery chemistries, and proposed
a representative five-step assessment procedure. Fig. 4 shows the five
pivotal stages for assessing retired batteries: (1) initial assessment:
grounded on the battery’s historical data; (2) disassembly: retired bat-
tery packs or modules are taken apart; (3) screening: the batteries
undergo rigorous testing, spanning mechanical, electrochemical, and
safety measures; (4) regrouping: following testing, batteries are sorted
and regrouped based on their performance metrics; (5) management
strategy: tailored control and management strategies are designed for
the batteries’ second-life applications. This section will cover the topics
from (1) to (4). After regrouping, specific management strategies are
necessary to deal with the low energy and power capabilities, large
inconsistencies, and potential safety concerns when integrating retired
batteries from different EVs into second-life applications. Section 5
provides a comprehensive introduction to advanced power electronics
to address these issues.
3.1. Initial assessment
Initial evaluation of retired EV batteries is crucial for their potential
repurposing. Upon arrival at the repurposing facility, comprehensive
data on these batteries, encompassing manufacturer details, model
specifications, production batches and dates, battery types (including
nominal capacities, voltages, and chemical compositions), operational
history, and recycled content should be meticulously cataloged [55].
This information can be easier to access in the near future through
the battery passport initiative: the European Commission has initiated
the battery passport starting in 2026 to improve sustainable battery
materials sourcing practices and enable a circular economy. All bat-
teries having a capacity of greater than 2 kWh will be covered under
this legislation [56,57]. However, while existing battery passport data
requirements may help maintain minimum performance guarantees for
EVs within their first 8 years of use, they may not enable accurate
prediction of remaining useful life beyond 8 years. Without accurate
remaining useful life predictions, battery passports can only give lim-
ited guidance concerning resale (refurbishment may be required) or
6
Renewable and Sustainable Energy Reviews 192 (2024) 114191
repurposing of the EV packs. Three key points should be considered to
ensure battery passports can inform the decisions of resale or repurpos-
ing [58]: (1) at least three state of health (SOH) measurements should
be made over different time points to establish an aging trajectory, and
each SOH measurement should be sufficiently accurate; (2) a 5% SOH
inaccuracy cannot provide confident remaining useful life predictions;
(3) the battery management system must enable the accurate prediction
of remaining useful life under changing use cases, which may require
the adoption of physics-based state estimation algorithms in the battery
management system.
Attention is warranted during the transportation of retired batteries
to reuse entities. The transportation regulations for LIBs in SAE J2950-
202006 should be obeyed [59]. Given the diminished reliability of
SLBs, more severe regulation should be formulated, the Society of
Automotive Engineers has been developing the standards for battery
secondary use, i.e., SAE J2997 [60]. The primary objective of this
endeavor is to formulate standards for the evaluation and testing of
batteries, encompassing a myriad of secure reuse options. This initia-
tive seeks to incorporate existing or forthcoming standards including
transportation and labeling, and augment these foundational standards
with pertinent information to ensure safe and dependable utilization.
3.2. Disassembly
After the initial assessment, those less reliable packs will be disman-
tled. Each disassembly step is related to one of the four categories [61]:
(1) chassis and battery safety components; (2) electronic and electric
system components; (3) dismantling of other battery parts and (4)
separation of modules. Given the heterogeneity in pack design from
different manufacturers, the disassembly process, time, and cost vary.
Table 1 lists the disassembly analysis for 5 top-selling commercial EVs
with various pack designs, including the disassembly time and cost
under manual, semi-automated, and fully automated cases. Table 1
indicates that: (1) the disassembly time varies with the battery pack
due to different designs. Specifically, the Nissan Leaf is the most time-
consuming, and the manual disassembly time from pack to cell can
reach as high as 468.1 min. This can be explained by its excessive use
of screws (1810 in total), mostly due to the complex bus bar design.
This indicates the potential to improve the dissemble efficiency by
an optimized design [61]; (2) the disassembly cost of pack-to-module
is significantly larger than the module-to-cell. For example, this ratio
reaches around 9 times in the case of Peugeot 208; this demonstrates
the retired battery system is preferably used in the pack/module levels
for cost considerations (at a safety premise); (3) the introduction of
fully automated disassembly can significantly reduce the cost by around
X. Gu et al. Renewable and Sustainable Energy Reviews 192 (2024) 114191

Table 1
Disassembly analysis for 5 commercial EVs. Data extracted and summarized from [61].
Vehicle Disassembly time per pack (to cell level) Manual disassembly Disassembly cost (to cell level) per kWha
cost per kWha
Man- Semi- Time Fully Time Pack to Module Manual Semi- Cost Fully Cost
ual automated saving automated saving module to cell ($/kWh) automated saving automated saving
(min) (min) (%) (min) (%) ($/kWh) ($/kWh) ($/kWh) (%) ($/kWh) (%)
Renault ZOE 323.8 81.5 74.8 47.3 85.4 0.39 2.19 2.58 0.45 82.6 0.076 97.1
Nissan Leaf 468.1 107.5 77 68.4 85.4 0.78 4.07 4.85 0.79 83.7 0.14 97.1
Peugeot 208 449.4 132.6 70.5 65.7 85.4 0.39 3.34 3.73 0.87 76.7 0.11 97.1
BAIC BJEV 198.5 43.6 78 29 85.4 0.62 0.99 1.61 0.21 87 0.05 96.9
BYD Han 173.6 44.6 74.3 25.4 85.4 0.32 0.62 0.94 0.17 81.9 0.027 97.1

a
The calculated cost is based on the UK’s labor cost ($24.88/h); given the significant cost decrease from automation, the conclusions hold for other main EV markets, including China and the
US (the hourly manufacturer labor cost in 2022 are $6.78/h and $25.43/h, respectively [62]).

97% and the time by 85% in all cases. This shows the potential of
developing automatic disassembly to replace humans.
Besides the time and cost, safety risks during disassembly will be
another notable challenge. Retired EV-LIBs often contain hazardous
materials, including heavy metals, toxic chemicals, and volatile organ-
ics, posing risks like electric shocks, short-circuits, thermal runaways,
fires, and toxic gas emissions during the disassembly process [63].
Given these safety factors and the cost decrease shown in Table 1,
opportunities exist to design fully automated disassembly machines to
deal with the dismantling of retired battery systems. The difficulty may
come from the various pack structures, but AI could benefit the whole
disassembly process, particularly in addressing the design uncertainty
and safety issues [64].
3.3. Fast screening
Screening, sorting and regrouping are the subsequent stages after
disassembly. The overall purpose of these steps is to screen out the
cells that cannot meet the requirements of second-life applications and
regroup the batteries with a close level of degradation and similar
electrochemical performances. Screening involves assessing mechanical
integrity, evaluating electrochemical performance, and assessing safety.
In mechanical integrity evaluation, the shape, size, and deformation
of battery cells will be initially evaluated in the visual inspection, and
those cells having unqualified appearances (e.g., leakage or damage)
will be disposed of or recycled. Electrochemical performance eval-
uation includes capacity, voltage, power capability, impedance, and
temperature. These can be synergized to assess battery conditions.
Table 2 lists the measurement methods for these parameters. Assessing
the internal status of SLBs is also crucial. Lai et al. [65] proposed a
heating impedance method to indicate the lithium plating severity. The
third step is the safety assessment. All battery cells operate within a
predefined range of conditions to ensure safety, encapsulated in the
‘‘safety envelope’’ concept [66]. However, compared to fresh lithium-
ion batteries, retired batteries potentially pose higher safety threats due
to prolonged use and internal anomalies like gas generation and lithium
plating. Challenges arise when assessing the safety performance of
retired batteries since they have typically undergone complex degrada-
tion processes. While conventional safety tests (thermal, electrical, and
mechanical) are still relevant, there is a pressing need for specialized
tests or algorithms to identify minor defects within these batteries. It
should be noted that during tests, LIBs may emit harmful gases, and
appropriate safety equipment must be deployed to ensure personnel
safety. Safety standards like the CFR 1910 and e-CFR 1926 by the
Occupational Safety and Health Administration in the US should be
followed when evaluating the SLBs [67].
To efficiently deal with a large number of heterogeneous battery
cells, a fast screening procedure is urgently required for SLBs. For
example, internal resistance can be quickly measured, and those bat-
teries with normal resistance can go to the next procedure, while the
abnormal batteries will be directly recycled [68]. An efficient way to
predict SOH is also helpful. Takahashi et al. [69] estimated the capacity
using a small window of current and voltage charging data by the
bagging of Gaussian Process Regression models. The SOH prediction
efficiency was improved by the short period of data. However, these
approaches only adopt one indicator and would not give a compre-
hensive evaluation of the cells. In some other advancements, statistical
methodologies are being employed to analyze extensive datasets related
to battery performance assessment. Lai et al. [70] devised a neural
network to scrutinize 5,000 cells, focusing on voltage and capacity
profiles. Their findings indicated an efficiency enhancement, surpassing
traditional methods reliant on full charge–discharge by over fivefold.
Braco et al. [71] introduced a two-phase method for estimating capacity
and internal resistance of SLBs, empirically validated on 506 cells, 203
modules, and 3 battery packs from Nissan Leaf EVs, achieving optimal
accuracy with error under 2% for capacity estimations and reduced
testing times from over a day to 2 min per cell.
At the end of the screening, the ratio of cells with qualified per-
formance for a second-life application can be determined. By counting
those cells passing the screening, together with those directly echeloned
used packs, an approximate ratio of suitable SLBs can be calculated.
This ratio offers valuable insights into the available quantity of SLBs,
and can be adopted to assess their potential to meet future capacity
demands for various second-life applications.
3.4. Sorting and regrouping
3.4.1. Methods
Using the assessment data, batteries can be sorted and categorized
based on their performance. The primary objective is minimizing vari-
ations between individual cells, as these discrepancies can adversely
impact battery longevity. The UL Solutions formulated the sorting
standard ANSI/CAN/UL 1974, which encompasses the requirements for
sorting and grading batteries intended for repurposing [72]. This can
be a good reference for industrial practices. Batteries can be potentially
classified by their parameters and categorized into different groups by
human-made rules. For example, according to the remaining voltage
values and fast/slow charge/discharge performance, the study in [73]
classified NiMH and Li-ion batteries based on the remaining voltage,
and retired batteries were then repurposed for use in a portable power
supply and illuminator. Given that the performance of retired batteries
is highly related to the working conditions in second-life applications,
the effective capacity, i.e., the capacity of batteries at a specific dis-
charge rate and temperature, was proposed in [74] to tackle this
issue. This allows retired batteries to be effectively regrouped to meet
application requirements. These works achieve classification by fol-
lowing specific artificial rules requiring expert experience and manual
classification. To automatically regroup heterogeneous batteries, some
authors proposed using machine learning approaches, including but not
limited to the K-means algorithm [75,76], self-organized network [68],
and support vector machine regression analysis [77]. Machine learning
approaches can synergize the measured parameters comprehensively
and achieve an automatic classification. However, the interpretability
of these methods needs to be improved to account for the intertwined
degradation mechanisms in SLBs. The work presented in [65] devel-
oped a more comprehensive regrouping method for SLBs. This method

7
X. Gu et al.
Table 2
Key parameters to evaluate cells and test methods.
Parameter Test method
Capacity CC-CV test [83]
Voltage quasi-OCV vs SOC test [83]
DC resistance Internal resistance tester [68], Hybrid pulse power
characterization [83]
Power capability Hybrid pulse power characterization [83]
Impedance Electrochemical impedance spectroscopy [84]
Temperature Surface sensors [68]
utilized a one-dimensional classification algorithm to evaluate side
reactions (e.g., lithium plating) of batteries. The batteries with the same
side reaction characteristics are further classified according to different
applications for SLBs, including energy scenarios, power scenarios, and
energy–power scenarios. In an energy scenario, priority was given to
capacity and remaining life, whereas in a power scenario, the main
concerns were internal resistance and remaining life. For a power–
energy scenario, capacity, internal resistance, and remaining life were
all considered.
It is essential to accelerate the classification efficiency for reduced
refurbishment cost, which directly determines the economic perfor-
mance of SLBs [78]. One class of acceleration approach is to apply
simple test protocols and avoid the cycle test for SLBs. For instance,
the research in [79] classified large amounts of retired batteries ef-
fectively by combining a fast pulse test with an improved bisecting
K-means algorithm. The classification efficiency was improved without
impairing performance by using pulse test curves as features. This
method reduced the time cost from 5 h to 2 min. Some work searches
for acceleration by improvement in current measurement methods.
For example, the work by [80] proposed a method that uses offline
hybrid pulse power characterization micro-test data to benchmark cells’
response over an SOC range. The recovering time after the pulse tests
reflects the strength of the battery, with stronger batteries recovering
more quickly than weaker ones, indicating higher power capability
and capacity. The proposed technique efficiently distinguishes different
batteries in 80 s.
3.4.2. Regrouping standards for cell-to-cell variation
Although repurposing batteries has been investigated in the ex-
isting work, the tolerance of cell-to-cell variation within a battery
pack remains unclear. It is essential to establish a clear and reason-
able standard that can serve as a guideline for repurposing batter-
ies to enhance the performance of SLBs, as it is still unclear how
cell-to-cell variation impacts pack-level performance (e.g., efficiency
and safety). An accurate quantitative standard of regrouping batteries
should be established by incorporating the self-balancing mechanism
among parallel-connected battery cells, meaning that the progression
of cell-to-cell variation regarding capacities will stabilize under the
convex or linear degradation curves [81]. Therefore, the cell-to-cell
variation is tolerable to some extent when regrouping batteries. This
will benefit battery classification because cells with mild variation can
be reasonably clustered into the same group. However, the allowable
variation should be further quantified for different battery designs
and operation scenarios. In [82], cell-to-cell variation within parallel-
connected cells under different cooling structures was analyzed, and it
found that the round cooling structure with relatively uniform cooling
conditions can significantly reduce the cell-to-cell variation compared
with the sequential structure with a non-uniform cooling structure.
However, there is a trade-off between system cost and suppressing cell-
to-cell variation, as uniform cooling structures are generally complex
and costly. Given the above analysis, the stabilized cell-to-cell variation
can be adopted as the upper limit of parameter variation for SLBs.
8
Renewable and Sustainable Energy Reviews 192 (2024) 114191
4. Economic assessment of second life batteries
4.1. Trade-off between recycling and reuse
Recycle or reuse? Which option maximizes residual values of SLBs
with minimal environmental pressure has become a global issue. But
actually, these two options are not contradictory to each other, as
the batteries can always be recycled after they get retired from their
second-life applications. The traditional method of burying waste bat-
teries in the soil can cause serious pollution to soil and water bodies,
and under the double constraints of resources and the environment,
the technical route of ‘‘production-use-disposal’’ is not sustainable. In
recent years, the ‘‘produce-use-recycle-reproduce’’ technology route,
also known as the ‘‘cradle-to-cradle’’ method, has received more and
more attention due to its resource and environmental advantages [85].
It can reduce the exploitation of local resources and has good economic
and ecological values. Currently, there are two mainstream technolog-
ical routes for handling decommissioned lithium-ion batteries: direct
recycling and repurposing. In terms of recycling, due to the complex
structure and number of materials in LIBs, they must first be classified,
disassembled, and separated, after which they can be subjected to direct
recycling, pyrometallurgy, hydrometallurgy, biometallic recovery, or a
combination of these methods [86–88]. While the technical readiness
of most of the methods is relatively high, except for the biometallic
recovery one, its main drawback is that it is generally unfriendly to the
environment and has a higher cost. For example, while the hydrometal-
lurgy method offers high material recovery efficiency with the desired
purity, it is a complicated process that leads to high costs, great pollu-
tion, and also makes it difficult to separate Co and Ni [89,90]. While
the biometallic method is the greenest solution for battery recycling, it
has really low recovery efficiencies and the lowest technical readiness.
In terms of reusing batteries, LIBs retired from electric vehicles still
retain 70%–80% of their initial capacity [91,92]. Although they may
not be able to meet the requirements of electric vehicles, they can
be reused in other less demanding applications and enter the route of
secondary utilization. After secondary utilization, material recycling is
performed when the capacity of retired LIBs decays to about 30%–40%
of the initial capacity [93]. Such a technological route of second-life
utilization is in line with the concept of the 4Rs of environmental
protection, and this technological route has considerable economic and
environmental value and can maximize the life cycle of LIBs [90,94,95].
Repurposing retired LIBs holds significant economic, environmen-
tal, and social significance, as highlighted by studies [95–98]. Firstly,
reusing LIBs aids in curbing CO2 emissions. Ahmadi et al. [99] demon-
strated a 56% reduction in CO2 emissions compared to natural gas
usage by constructing a grid energy storage system employing retired
LIBs. Secondly, excessive lithium extraction depletes water sources
and can cause ground subsidence. Utilizing SLBs reduces direct dis-
posal, lessening soil and water pollution. Third, incorporating retired
LIBs into energy storage enhances efficiency and facilitates greater
utilization of renewable energy, fostering clean energy and low-carbon
societies [100]. Fourth, it extends LIBs’ lifecycle, reduces costs, and
promotes sustainable electric vehicle development. Additionally, repur-
posing mitigates energy security risks and societal concerns. The global
demand for electric vehicles intensifies the importance of cobalt and
lithium, yet these resources are unevenly distributed, creating geopo-
litical tensions [76,88,100]. Certain countries, like the Democratic
Republic of the Congo, dominate reserves, raising resource security
issues and ethical concerns related to resource extraction. SLBs address
these challenges by meeting energy storage needs without straining
resources or exploiting labor [101,102].
Although SLBs have the above-mentioned advantages, concerns
such as the additional costs and environmental impact induced by the
transportation and refurbishment processes of SLBs and safety risks
should be carefully evaluated to prove their’ reused values. Whether
the industry is willing to use SLBs highly depends on their values, and
X. Gu et al.
the key players in the decision of recycling or reusing are the costs of
using SLBs, the benefits gained by the SLBs, policies, and incentives. To
this end, as shown in Fig. 5, this section will first discuss the service
life of the SLB, which is the basis of all second-life battery value-based
analysis. Then, the economic feasibility of the SLB will be reviewed,
which covers both the costs and benefits gained by SLBs using different
applications. Assessment techniques and the current state-of-the-art are
summarized with details regarding the modeling methods, economic
indicators, and critical assumptions underlying the assessments. Last
but not least, different SLB business models will be introduced, as well
as the legal framework associated with the coordination of different
stakeholders.
4.2. Service-life assessment
The costs and benefits associated with SLBs are inextricably linked
to their anticipated service life. Second-life batteries, while providing
a valuable opportunity to extend the life of lithium-ion cells beyond
their initial application, demand meticulous assessment. Before using
retired batteries in the energy storage system (ESS), the remaining ca-
pacities of batteries need to be examined or estimated to initiate a safe
and economical operation in second-life applications. As mentioned
in Section 3, batteries with different SOH levels would be available
for second-life applications. Typically, SLBs with a higher remaining
capacity yield more revenue, but they may come at a higher cost. To
make effective use of SLBs, the cost of maintaining and refurbishing
these batteries must be outweighed by their benefits. The primary
hurdle in this domain is accurately predicting the remaining lifespan
of SLBs. This estimation directly influences their utility and safety
for prolonged use. To overcome this challenge, advanced modeling
methods capturing both linear and non-linear degradation patterns,
substantial data for model calibration, and a clear operational boundary
for safe second-life applications must be developed.
4.2.1. Aging-conscious assessment
Understanding battery degradation processes is crucial for opti-
mizing usage and prolonging battery lifespan, ultimately maximizing
benefits. Investors encounter a dilemma: choosing between larger,
costlier SLBs with extended lifespans or smaller, cheaper batteries with
limited service life while meeting the same storage demand. Finding
the optimal SLB size is essential for maximizing benefits. However,
accurately estimating SLBs’ remaining useful life relies on specific op-
erating profiles and initial health status. Previous studies calculated the
cost of degradation for both FLBs and SLBs by adding the replacement
cost in economic analyses, assuming a fixed battery lifetime (e.g., five
years), while ignoring the operating cost caused by battery degra-
dation. Considering the profitability of SLBs, accounting for battery
degradation is essential as it directly impacts the long-term benefits
of SLBs, which are linked to energy throughput. Operating profiles
of stationary applications exhibit dynamic behavior, characterized by
fluctuations in current over time. Battery degradation patterns vary
based on different applications and control strategies. Kim et al. [103]
have studied these variations, primarily focusing on FLBs rather than
SLBs. Therefore, it is crucial to analyze SLB degradation under various
conditions to accurately estimate their contribution over their service
life. This analysis ensures precise benefit estimation and guarantees that
the revenue from the SLB project will sufficiently cover the capital cost.
Recently, more and more studies have considered the dynamic
degradation process when evaluating the economic performance of
SLBs and recognized that battery aging has highly non-linear char-
acteristics, with rapid capacity loss and resistance rise [104,105]. To
incorporate the dynamic degradation process in the dispatching cal-
culations or optimization process, researchers have employed simpli-
fication, linearization, and iteration technologies to fit existing cell-
level battery degradation models. For example, previous studies have
developed simplified semi-empirical linear degradation models using
9
Renewable and Sustainable Energy Reviews 192 (2024) 114191
health factors [106], constant degradation rates [107], and lifetime
throughput [106,108,109] to quantify battery degradation. While these
simplified models are easy to implement and perform well under spe-
cific cycling conditions, they are neither accurate enough for economic
analysis nor able to reflect actual battery degradation mechanisms
under dynamic operating conditions. Some studies linearized semi-
empirical degradation models using techniques such as the rainflow
counting algorithm to approximate the cycling aging behavior of SLBs
and incorporated them into their economic analysis [110–112]. Only
a few studies have directly applied non-linear degradation models
in their analysis as an offline degradation estimation method to get
accurate estimations of capacity losses under specific mission pro-
files [104,113,114]. Physical models offer detailed insights into the
electrochemical reactions in batteries [115], enabling precise behav-
ior modeling. When combined with aging models, they can simulate
lifetime degradation. However, fine-tuning these models is challeng-
ing as one set of parameters applies to only one cell, leading to
unreliable results across different cells. Parameter variation among
cells necessitates multiple sets, making single-parameter modeling less
dependable. Furthermore, individually parameterizing and modeling
each cell is computationally inefficient and often impossible due to
the lack of historical cell data. A major hurdle seems to be accessing
battery historical data from automotive manufacturers, since it is often
protected by automakers as trade secrets. This data would be useful
for comparing computer aging simulation results with battery module
specifications certified by cell makers. Moreover, physical models, con-
structed through partial differential equations, pose challenges when
integrated into model-based optimization problems as they introduce
computationally intractable nonlinear parts. Despite advancements in
various models, accurately predicting battery degradation in dynamic
stationary applications remains elusive, leaving the service life of SLBs
uncertain. Studies claiming SLBs’ profitability might have relied on
overly optimistic hypotheses or assumptions about SLBs’ battery degra-
dation behaviors. Currently, there exists a substantial gap in precisely
modeling SLBs’ degradation behavior and integrating it into the SLBs
assessment model.
4.2.2. Warranty and insurance
In the realm of battery technology, the warranty is intricately
tied to both the economic viability and the durability of batteries
during their secondary usage. The warranty terms are meticulously
formulated to ensure that, within a specific timeframe, a predetermined
percentage of the product’s performance remains consistently optimal.
This determination is not haphazard; rather, it is a result of the man-
ufacturer’s meticulous calculations aimed at estimating the expenses
associated with repairing and rectifying any defective components.
This expense, borne by the manufacturer, is invariably embedded in
the product’s overall price. In the case of SLBs, establishing appropri-
ate warranty terms necessitates a rigorous, evidence-based approach.
This commences with a comprehensive scientific assessment of the
remaining operational life of these SLBs. Subsequently, an assessment
of the failure rate for these batteries is conducted, enabling us to
pinpoint the costs that will be incurred to rectify these failures. Further-
more, it is worth noting that beyond the product warranty, additional
safety measures are often considered, and insurance companies may
step in to provide coverage for severe second-life battery incidents or
safety-related concerns. This multi-faceted approach underscores the
complexity and precision required in the battery industry to warrant
a product’s performance and longevity.
4.3. Cost assessment
SLB price setting determines the selling price of the SLBs, and it
is considered the monetary cost of using SLBs in economic assess-
ments. The biggest advantage of decommissioned LIBs over new LIBs
is the procurement price. Currently, the price of a decommissioned
X. Gu et al.
LFP battery is about $0.5/kWh, and the price of a decommissioned
NCM battery is about $0.45/kWh; by comparison, the price of a new
LFP battery is about $1.3/kWh, and the price of a new NCM battery
is about $0.9/kWh, and the price of a lead–acid battery is about
$0.3/kWh [91,92,98,116]. The price of a retired lithium-ion battery
is estimated to be only half the price of a new battery and close to
the price of a lead–acid battery, which is widely used for all stationary
energy applications where there is a huge market demand that makes
the economic value of second-life batteries very obvious.
4.3.1. Price-setting
Determining SLB pricing for large-scale energy storage involves
considering FLB prices, refurbishment costs, expenses from auxiliary
devices like power electronics, and operational costs. From the perspec-
tive of price makers, as FLB prices decrease, SLB profitability becomes
highly sensitive to refurbishment and auxiliary device costs. Refur-
bishment expenses include dismantling, sorting, and refurbishment
costs, influenced by testing equipment and labor costs. The cost per
kilowatt-hour storage capacity of SLBs decreases with scale, utilizing
standardized processing equipment adaptable to various EV battery
modules. While sorting is costly due to detailed battery investigations,
current expenses mainly stem from testing and research for classifi-
cation methodologies. Sorting costs vary based on the sorting level,
with cell-level sorting being more expensive but vital for optimal
performance. Identifying the most cost-effective sorting level is crucial,
emphasizing advanced testing methods as well as the importance of
historical data. While precise classification increases refurbishment
costs, the extended lifespan enhances their value, enabling them to
be sold at higher prices. Striking a balance between maximizing SLB
longevity and managing reasonable refurbishment expenses remains a
fundamental challenge in sustainable energy storage.
4.3.2. Cost analysis
The pricing of reused batteries is intricate [117,118], emphasizing
the need for economic analysis beyond comparing new and retired
battery costs [119]. To address price uncertainties in SLBs, methods
like market surveys, sensitivity analysis, and evaluating SLB benefits
are used. Market surveys calculate SLB costs by subtracting refur-
bishment expenses from the original price [106,120,121]. According
to Li et al. [74], the price of SLBs for energy storage is $72/kWh,
and the selling price of FLBs is $232/kWh. Some statistics in China
show that SLBs are originally sold at about $23-31/kWh, and resell
for about $62-70/kWh after testing, screening, and reconfiguration. It
is also predicted that by 2025, the price of SLBs will be 30%–70% of
the price of new batteries [53]. Some studies conducted a sensitivity
analysis on the cost of SLBs and found the profitable purchase cost of
SLBs [8,122–124]. Mathews et al. [113] assessed the profitability of
a PV-SLBESS standalone system under different operating conditions
and suggested that retailers sell SLBs for less than 60% of the FLBs’
price. Moreover, studies also estimated the benefits of FLBs during
their SOH drops from 100% to 80% to determine the profit margin
and appropriate cost of SLBs [104]. Considering the residual value of
batteries in retired EVs, Neubauer et al. [125] constructed a framework
to estimate the selling price of retired EV batteries. They estimated
the selling price of a battery health model to be $44-180/kWh for a
SLB and $150-250/kWh for a FLB. However, the cost considerations in
existing studies are not precise, as the costs related to transportation,
storage, and replacement are either ignored or under assumptions that
contain huge uncertainties, resulting in a lack of reliability in current
techno-economic evaluations [107,126]. Taking transportation costs as
an example, SLBs are not well-defined in most countries’ transportation
regulations, leading to complications and high transportation costs due
to safety concerns. It is estimated that the inbound logistic costs of
the battery are approximately 1.3 pounds per kWh [98]. All existing
regulations are established for FLBs and security is a major issue in the
transportation of SLBs [127]. With regard to air transport, the United
10
Renewable and Sustainable Energy Reviews 192 (2024) 114191
Nations International Civil Aviation Organization has implemented a
ban on the transport of individual LIBs (UN3480) and individual LIBs
(UN3090) on passenger aircraft. However, packaged LIBs (UN3091
and UN3481) stored in equipment may be transported on commercial
aircraft under the regulations. In addition, US railroads are reluctant to
accept failed LIBs. This restriction is due to the spontaneous combustion
nature of the organic liquid electrolyte in LIBs. However, bulk trans-
portation prevents fires in this area. Strict regulation of transportation
increases transportation costs. In addition, if the batteries are stored
in containers or immersed in water, the chance of ignition may be
reduced. Transportation rules in Europe are more stringent than in the
US, and batteries must follow UN manual tests and standards. In short,
the regulation of the transport of LIBs varies from country to country,
and there is a pressing desire to formalize standards for the transport
of LIBs. The uncertainties in repurposing costs, unreliable SLB market
alignment, and lack of financial justification are the main barriers
that impact decisions on deploying SLBs. Tighter cooperation between
the industry and the research parties is the key to closing current
information gaps, and the transparency of costs related to SLBs plays a
significant role in improving the accuracy of economic assessments.
4.4. Benefits assessment
Before integrating SLBs into secondary applications, it is vital to
conduct economic studies and promote profitable uses. By emphasizing
second-life benefits, they also indirectly enhance the utilization of
FLBs, contributing to the circular economy of LIBs. It has been found
that repurposing EV batteries in stationary applications can lower the
upfront cost of EVs and the cost of FLBs [106]. Debnath et al. [128]
evaluated the effect of reusing batteries in the smart grid on the price
of FLBs and the benefits of EV owners. The results indicated that
the revenue generated by SLBs compensates for 19.56% of the FLB’s
purchase cost. A similar conclusion was obtained in [8] that the profit
of repurposing SLBs for stationary energy storage applications would
reduce the EV’s upfront cost by 14.3% to 36.3%.
Similar to other stationary energy storage devices, using SLBs helps
improve power system reliability and the quality of electricity supply.
Overall, SLB applications can be categorized into three types: on-grid,
off-grid, and mobile applications. The first techno-economic analysis
for SLBs was presented by the US Advanced Battery Consortium [129].
Later, Cready et al. [130] estimated the investment costs of various
SLB applications, considering the costs associated with acquisition,
transportation, battery testing, and refurbishment. Transmission sup-
port, light commercial load following, residential load following, and
distributed node telecommunications backup power were considered
note-worthy and might be profitable for reusing batteries at that time.
4.4.1. Application schemes
On-grid applications SLBs play a crucial role in advancing renew-
able energy technologies by addressing intermittency issues [131,132].
They increase end-user participation in electricity supply and ancillary
services [133,134]. In regions like California and New York, incentive
programs encourage consumers to enhance energy self-generation. For
instance, in New York, the storage incentive program supports systems
of 50 kW or more, reducing summer peak demand significantly. Tradi-
tional power plants struggle with rapid load changes due to high inertia
and base load requirements. SLBs provide an excellent solution, absorb-
ing excess energy during periods of decreased demand and supplying
energy promptly during periods of high demand. Demonstrated by the
3 MW/2.8 MWh energy storage system at the Johan Cruyff Arena in
Amsterdam, SLBs enable effective energy storage and grid services. This
system, comprising 590 battery packs, demonstrates SLBs’ capacity to
meet varying demands, showcasing their adaptability and efficiency in
real-time applications.
Off-grid applications Using SLBs in micro-grid and smart-grid
applications has been proven to be feasible in work [108]. SLB systems
X. Gu et al.
can supply power to independently operated electrical systems without
the need to connect to the main electrical system, e.g., to supply
power to remote areas [135]. Another promising market is energy-
demanding telecommunication towers and remote mining, which are
usually powered by diesel generators operating 24 h a day, 7 days a
week, and which can be powered by SLBs together with solar energy.
Fresh food distribution centers with many refrigerators and freezers
have high energy needs and are also potential customers for building
energy storage systems that can be powered by used batteries [136].
Mobile applications While few studies have delved into SLBs for
mobile applications, promising opportunities have emerged in various
areas [67,137]. One notable challenge in electric vehicle deployment
is infrastructure, particularly the need to enhance the electrical system
and construct charging stations [138,139]. SLBs offer a solution. Re-
search has explored using SLBs in off-grid photovoltaic vehicle charging
systems, successfully integrating them into structures with solar cells,
inverters, and power managers. Companies like FreeWire are repur-
posing Nissan batteries for mobile vehicle charging, demonstrating
SLBs’ practicality. SLBs also support vehicle-to-grid services, enabling
parked electric vehicles to supply power to the grid when idle, reduc-
ing dependency and enhancing grid reliability. This approach signif-
icantly reduces upfront battery costs for EV consumers and logistics
companies. Despite the potential, challenges exist, such as battery
degradation, energy loss, the need for robust communication protocols,
and additional infrastructure requirements. Successful vehicle-to-grid
implementation requires bi-directional chargers and communication
standards, involving substantial investment. Additionally, SLBs find
applications in short-range vehicles like forklifts, low-speed vehicles in
shopping centers, apartments, boats, and electric scooters, highlighting
their versatility in off-grid scenarios [67,140].
4.4.2. Economic modeling
To determine the actual profitability of SLBs in different application
schemes, economic models are used. Various economic models can be
found in existing studies to assess the economic benefits of SLBs, such
as the net present value (NPV) model, which estimates the net present
worth of the SLB project, including time-varying costs and revenue; the
levelized cost of electricity (LCOE) model, which stands for the average
cost of electricity considering time-varying investment costs and energy
production; the levelized cost of storage (LCOS) model, which is similar
to the LCOE model but only includes energy storage-related costs in
calculations; and the cost–benefit model, which represents the ratio
between the investment cost and the economic benefit obtained from
SLBs. In the aforementioned economic models, the capital recovery
factor, calculated by the discount rate and the estimated lifetime of the
project, is used to convert future cash flows into present values.
NPV The NPV model considers the present value of cash inflows and
outflows throughout the project and is widely used in capital budgeting
and investment planning [141]. In the case of SLB applications, a posi-
tive NPV indicates profitability. For example, Bai et al. [124] evaluated
the NPV of distributed solar photovoltaic systems in China and found
that reusing batteries is profitable nationally for both the industrial and
commercial sectors. However, Gur et al. [142] reached the opposite
conclusion when analyzing the European electricity market.
LCOE The LCOE model can provide an explicit comparison between
the economic performance of SLBs and FLBs, with the flexibility to
be reformulated to incorporate other forms of expenditure [143]. For
example, in fast charging applications, the LCOE could be reduced
by 12%–14% using SLBs, although this result is dependent on several
factors, including the cost and lifetime of the SLBs, their efficiency, and
the discount rate [144]. Steckel et al. [145] proposed LCOS and found
that the LCOS of SLB-ESS (i.e., $234-278/MWh) is higher than that of
FLB-ESS (i.e., $211/MWh), but the upfront costs for SLB-ESS are only
64.3%–78.9% of those of FLB-ESS.
Cost–benefit The cost–benefit model is useful for analyzing the ex-
penditures and gains of using SLBs in ESS. Studies by Fallah et al. [123]
11
Renewable and Sustainable Energy Reviews 192 (2024) 114191
and Rallo et al. [109] used this module to calculate return-on-
investment in energy arbitrage and frequency regulation services. The
results showed that a higher number of arbitrage trades might lead to
a lower cycle life for SLBs but generate a higher cash inflow and return
on investment. The cost–benefit ratio generally decreases as the price
of SLB drops.
4.4.3. State-of-the-art
Grid-connected applications are considered more profitable than
off-grid applications [126,133,146,147] when used to renewable firm-
ing, area and frequency regulation, and peak shaving, as summarized in
Table 3. The study by [148] has demonstrated that using SLBs for load
leveling reduces the electricity bill by up to 39.7% because integrating
BESS with renewable generation facilities results in a higher self-
consumption rate [144]. However, the efficacy of COE reduction from
SLB integration depends on the scale of the renewable power plant.
SLBs are more suited for large-to-medium-scale applications in China
in terms of promoting a self-consumption rate compared to residential
applications [124,149]. Similar conclusions are found in [142], which
suggested that SLBs would be crucial for utility-scale BESS, also known
as front-of-the-meter battery storage [150], due to more intense energy
demand [151]. Using SLBs in a BESS can also reduce the cost of energy
(COE), as proved in the studies considering local tariffs in different
countries, such as Portugal [122], Germany [7], and China [124].
Overall, these studies highlight that SLBs would be an alternative to
FLBs in various applications, especially supply-side management, as
they help lower both the capital investment of ESS and COE. Compared
with a fresh battery ESS, the investment cost can be reduced by up to
60% when SLBs are integrated with photovoltaic (PV) panels [113],
70% when SLBs are used as energy backup sources [108], and 73.62%
when SLBs are in a grid-connected renewable energy system [152].
Collectively, these studies outline the critical role of economic
models, and the model should be selected based on the purpose of the
study. However, it is challenging to compare the economic performance
of various SLB applications across studies due to differences in the fi-
nancial indicators and economic model selections. Most existing studies
compare the economic performance of ESS designs with or without
SLBs in a single application [122,142,144]. Due to inconsistencies in
the methodology used for system modeling and economic assessment
across various studies, a comparison of the economic performances of
SLBs in different applications is still missing. Therefore, standardization
of economic evaluation methodologies for SLB-ESS assessments is es-
sential to showcase the advantages of using SLBs in specific application
schemes in order to promote their adoption in the market, which is
still absent in current studies [107]. Besides, despite numerous studies
on the benefits gained by using SLBs in different applications, they
are all performed under a specific scheme with underlying assump-
tions, which means that these assessment results may lack applicability
in reality where the actual profitability of SLB applications remains
unclear. Thus, it is crucial to identify the root of uncertainties and
the deterministic factors involved in the SLB assessment, as well as
understand the advantages and disadvantages of different assessment
techniques, as shown in Fig. 5.
Assumptions made during the modeling process and parameter
values would have a significant impact on the evaluation results. The
profitability of SLBs is intricately linked to several factors, including
varying electricity prices across countries, the discount rate at the time,
and the predicted service life of the SLB when using the assigned
applications.
Electricity prices China’s industrial sector experiences higher elec-
tricity prices than residential areas, whereas the opposite is true in the
US, where residential electricity costs are higher. This price discrepancy
significantly impacts SLB profitability, particularly in applications like
energy arbitrage. Gur et al. [142] noted that the profitability of PV-
SLBESS varied across European countries and might differ in other
regions.
X. Gu et al.
Fig. 5. SLBs economic assessment strategies and the current state-of-the-art.
Discount rate In order to convert future cash flow into present
value, a discount rate is applied, and its value mainly depends on the
project risk, interest rate, and duration of the project. The value of the
discount rate is mainly determined based on past studies, varying from
1% to 15%, as shown in Table 3. The discount rate has a significant
impact on the evaluation results [107,144], as a higher discount rate
reduces the future cash inflow [142].
Service-life Another factor that plays a major role in assessing
the benefit of SLBs is the EOL projection/assumption, i.e., the entire
lifespan of batteries [55]. Current studies assume a static SOH at EOL of
around 50%–60% [104,144], but this may not be accurate. It was found
that the lower the SOH at EOL, the higher the benefit–cost ratio, since
greater contributions from batteries accumulate over a longer period
of time. So it is crucial to investigate the estimation of EOL points for
SLBs in various second-life applications.
Incentives and policies Additionally, SLB profitability is heavily
influenced by government incentives and policies [55]. Mirzaei Om-
rani and colleagues [148] underscored the effect of SLB costs and
subsidies on applications like load leveling in residential, industrial,
and photovoltaic power plant sectors. Their analysis demonstrated
that a 20% reduction in SLB costs could increase the internal rate
of return by 164%. Government support, especially in the residential
sector, is crucial for promoting SLB usage. Repurposing batteries in
households can generate significant net profits and reduce expenses
related to constructing on-peak thermal power plants. This approach
12
Renewable and Sustainable Energy Reviews 192 (2024) 114191
offers substantial savings for governments. Additionally, continuous
government subsidies, enhancements in the electricity price market and
adjustment mechanisms, and more precise battery safety management
further increases the likelihood of economic benefits from widespread
application of large-scale SLB.
4.5. Industry outlook
The lithium-ion battery industry operates within an intricate chain
involving manufacturers, electric vehicle producers, consumers, refur-
bishment firms, and recycling companies. Refurbishment and recycling
companies, specializing in retired batteries, seek profits from low-
priced units but struggle with lowering refurbishment costs. Consumers
holding property rights demand high recycling prices. Battery manufac-
turers, in determining manufacturing processes, aim for low-cost raw
materials from recycling firms [153]. Electric vehicle manufacturers
seek cost-effective, durable batteries for warranties. Decisions on SLB
responsibility raise questions: should specialized firms, car manufactur-
ers, or battery producers handle it? Safety concerns add complexity,
demanding a balance between accountability and profitability. The
success of second-life batteries hinges on creating a profitable industry
chain through strategic reuse efforts, where an effective business plan
for echelon use is crucial [154,155].
X. Gu et al. Renewable and Sustainable Energy Reviews 192 (2024) 114191

Table 3
A summary of the economic performances and assessment methodologies of SLBs in various applications. Green: Profitable, white: might
be profitable, pink: not profitable.

Application Economic Degradation model SLBs cost considera- Discount

model tion [$/kWh] rate [%]
On-grid applications
[126] Peak-shaving NPV, cost– Constant decline rate 15%–40% of the FLB 8
benefit cost
model
[122] PV-BESS NPV, payback Semi-empirical model [114] Nissan Leaf: 1496 3, 5, 7
period $/cell, Citroen Co:
2175 $/cell
[120] Uninterrupted power Cost–benefit Constant throughput degradation 38-132 10
supply model model
[149] Demand responds COE DOD-based aging model, rain-flow – –
cycle counting algorithm
[113] PV-BESS Cost–benefit Semi-empirical data-based model 104.5 7
model [114]
[151] Peak shaving NPC,LCOE Lifetime throughput model [108] 65[130] 6.9
[131] Grid-support NPV Weighted average Amp-throughput 46 4.5
model
[144] Fast charging energy LCOE – 95 6.7
storage
[142] PV-BESS NPV – 126-473 1–10
[106] Utility-scale NPV Constant throughput degradation 30% of the FLB cost 10
applications model
[132] Load-leveling NPV – 55–132 1
[8] Energy arbitrage Cost–benefit Gaussian process regression model 112–128 2–8
model [105]
[146] Regulation market NPV DOD-based model, rain-flow cycle 126 5
counting algorithm
[133] Energy, non-energy Cost–benefit DOD-based model 73 7
service model
[109] Energy arbitrage, Annual saving, Equivalent circuit model 52.6 –
peak shaving return-on-
investment
[107] Utility-scale LCOE, LCOS Degradation rate at 1%–3% per year 64.3%–78.9% of the –
application FLB cost
[104] Wind farm Equivalent Semi-empirical model [112] 100 –
price model
Off-grid applications
[108] Smart grid Capital invest- – 100 5–10
generation side asset ment, COE
management
[134] Hybrid wind farm Revenue, annu- Dynamic degradation model 90 8.42
alized cost
[124] PV-BESS NPC,LCOE – 48 –
[147] Smart-grid Cost, revenue Equivalent circuit model 40–147 –
[152] Microgrid NPV,COE Modified kinetic battery model – –
Mobile applications
[138] Fast charging station Total cost of – 100 5
ownership,
NPV
[139] EV charging station COE DOD-based linear model – 4

4.5.1. Business models
The market for SLBs mirrors the used automotive parts market,
involving intricate relationships among various entities. There are three
distinct business models [156]: closed market, intermediary-based mar-
ket, and open market. In these markets, stakeholders engage and collab-
orate differently. After refurbishment, original equipment manufactur-
ers (OEMs) can either sell them or repurpose them for less demanding
applications, using them in final products.
Closed market A closed market model is mostly due to data
confidentiality concerns among battery and automobile manufactur-
ers. Here, relationships revolve around customer/company and com-
pany/customer interactions. While blockchain technology enhances
trust and data security and opens doors to innovative business models,
further research into blockchain is demanded to resolve supply chain
conflicts, boost customer confidence in used products, and enhance
supply chain transparency through battery traceability [157]. In a
closed-market scenario, original equipment manufacturers can lease
batteries for EV use. Once these batteries no longer meet vehicle
requirements, the original equipment manufacturers reclaim, test, sort,
and repurpose them for a second application. In this model, the re-
sponsibility for battery recycling also falls on the battery or vehicle
manufacturer [154]. Companies focusing on conservation technologies
will embrace this approach, benefiting from the refurbishing and re-
cycling markets. They have to either develop their own refurbishment
processes or collaborate with third-party firms for these services.
Intermediary-based market In the middle market model, original
equipment manufacturers could form partnerships with other compa-
nies [154]. Partnerships between different stakeholders in the battery
industry, such as automakers, battery manufacturers, logistic compa-
nies, recycling companies, and refurbishment firms, are formed to
establish circular channels for the smooth and effective use of retired
batteries. These collaborations help create comprehensive and efficient
systems for SLB repurposing [98]. In this setup, the intermediary acts
between the OEM and the customer. The OEM may choose to share
their knowledge, data, and uncertainties about the SLB business with

13
X. Gu et al.
the intermediary, or they might simply sell the SLB outright. Trans-
ferring ownership of used batteries to an intermediary presents risks
for both parties: the OEM risks losing control of the used EV bat-
tery market, data breaches, technical confidentiality issues, and brand
damage due to accidents or low performance [154]. Simultaneously,
the intermediary may buy batteries of lower quality than expected,
depending on the OEM, who might be uncertain about the available
quantity, quality, or performance of the batteries. Ensuring well-drafted
contracts on secure platforms becomes essential to safeguarding the
rights of both parties, where possible.
Open market In the open marketplace model, a marketplace op-
erator facilitates an online platform connecting customers and sellers
and manages inventory based on supply and demand for batteries used
in second-life applications [158]. This operator acts as an intermedi-
ary for battery transactions. However, only 21.95% of manufacturers
analyzed in a study provide platforms enabling customers to specify
energy storage systems before purchase [154]. Without blockchain
technology safeguarding against opportunistic behavior, the open mar-
ket poses challenges. Understanding each battery’s history becomes
crucial, complicating individual pricing due to varying aging patterns
under similar conditions. This increases uncertainty about battery per-
formance, longevity, reliability, and safety in secondary applications.
Hence, it is challenging for EV owners to market these batteries to one
another.
Different regions adopt different methods based on their industrial
setups and regulations. For instance, Japan primarily relies on the
closed market, where manufacturers handle the end-of-life manage-
ment process directly. In contrast, the European Union and the US often
utilize the intermediary-based market, where industry alliances coor-
dinate recycling efforts. Meanwhile, the rise of third-party recycling
companies is becoming prominent as the demand for battery recycling
increases with the growing number of decommissioned lithium-ion
batteries.
4.5.2. Legal framework
In the ever-evolving landscape of battery technology, the legalis-
tic framework for SLBs is intricately woven into existing regulations
crafted for FLBs. The existing legal infrastructure forms the scaffold
upon which innovative measures for SLBs are constructed, covering
the entire spectrum from manufacturing intricacies to meticulous end-
of-life management protocols. The pivotal challenge of ensuring envi-
ronmentally responsible disposal looms large [95]. This concern has
spurred a global response, prompting nations across the world to
devote attention and resources to drafting comprehensive regulations
for recycling and systematic utilization of retired LIBs [34,154].
Current standards and regulations mainly focus on battery recycling
processes. However, considering the above-mentioned business model,
the responsibility of battery reusing may be in the same place as the
recycling processes. Therefore, the legalistic framework of SLBs may
contain similarities. In Denmark, a stringent policy mandates all battery
retailers to collect and recycle discarded batteries [159]. To finance
the costs incurred during the collection, transportation, and processing
stages, a tax ranging from 6% to 8% is levied on the sale of new batter-
ies. On a larger scale, the EU introduced comprehensive policies for the
handling and recycling of waste batteries in 2006 (2006/66/EC) [160].
A trailblazing approach was adopted back in 2008, when stringent
mandates were imposed on battery manufacturers. These mandates
required the establishment of recycling systems specifically tailored for
used automotive batteries. Clear obligations were outlined for manu-
facturers, importers, sellers, and consumers within the battery industry
chain. Simultaneously, a deposit system was introduced, ingeniously
nudging consumers toward proactive involvement in recycling endeav-
ors. Building on this foundation, the European Parliament’s Committee
on Environment, Public Health, and Food Safety ratified a groundbreak-
ing battery regulation in 2022. The European Union has initiated the
‘‘BATTERY 2030+’’ program, a visionary effort aimed at addressing
14
Renewable and Sustainable Energy Reviews 192 (2024) 114191
challenges in future battery research and development [161]. This
program advocates for a holistic approach to battery design, focusing
on the entire lifecycle of batteries, including second-life management.
The core principle revolves around creating a closed-loop, synergistic
system from raw material extraction to recycling [94,156]. Meanwhile,
the US implemented a multi-tiered regulatory approach at the fed-
eral, state, and local levels. Regulations mandated battery retailers to
proactively recycle used batteries and introduced a deposit system to
incentivize consumers to turn in their used battery products. Japan has
enacted the ‘‘Law for Promotion of Effective Utilization of Resources’’
to implement the 3R plan, which stands for recycling, reusing, and
reducing [162]. These laws specifically mandate the establishment of
a system for recycling batteries. In the context of China, a nascent
yet robust dismantling, refurbishment, and reuse industry for batteries
is emerging. The Chinese government has mandated automakers to
establish intricate recycling channels for retired batteries, employing
strategies like buyback, trade-in, and subsidies [163]. Moreover, col-
laborative efforts among automakers, battery manufacturers, recycling
and dismantling entities, refurbishment firms, and battery recycling
companies are actively encouraged, forming circular channels for the
efficient and effective utilization of batteries [156].
Amidst this global tapestry of legal frameworks, testing standards
have become a linchpin. In China, for instance, the management of
decommissioned LIBs operates at three levels: production, use-phase,
and recycling. To enhance efficiency and sustainability, the government
mandates automakers establish reuse and recycling channels through
buybacks, trade-ins, and subsidies [164]. Collaborations between au-
tomakers, battery manufacturers, and recycling entities are encouraged
to ensure circular and effective battery utilization. Specific regulations,
like the ‘‘Measures for the Administration of the Cascade Utilization
of Traction Batteries for New Energy Vehicles’’, enforce collaboration
between SLB enterprises, vehicle manufacturers, and battery manufac-
turers [164]. Enterprises must adhere to national standards, focusing on
reusing waste battery packs and modules. Strict testing standards eval-
uate residual capacity, enhancing product performance and reliability.
Owners of cascade products must responsibly transfer end-of-life items
to designated recycling outlets. Unauthorized actions are prohibited,
holding both enterprises and owners accountable. This comprehen-
sive framework ensures a streamlined, responsible, and sustainable
approach to LIB lifetime management in China [165].
In short, the legalistic framework surrounding SLBs is not a static
entity but a vibrant, responsive organism, adapting and evolving as
technologies and challenges shift. It represents a harmonious synergy
between stringent regulations, innovative incentives, and collaborative
efforts, all aimed at realizing a sustainable, circular economy for bat-
teries. As we delve deeper into the 21st century, this dynamic legal
framework will undoubtedly continue to evolve, steering the course
toward a greener, more sustainable energy future.
5. Power electronics technologies
In a BESS, power electronics play the key energy conversion inter-
faces among the batteries, other distributed energy resources, loads,
and power distribution networks. The performance of BESS with SLBs
(SL-BESS) is inadequate because the SLBs are low in power efficiency,
heterogeneous in health conditions, and insufficient in reliability. Well-
designed power converters can substantially amend this performance
inadequacy of SL-BESS: (1) SLBs are heterogeneous in power capabil-
ities, energy capacities, and health conditions. Although the grouping
can lower the heterogeneity, the costs of transportation and inventory
in preparation for regrouping are significant. Power converters are
a beneficial alternative to the grouping because the converters can
compensate for the discrepancy among SLBs instantaneously [166]; (2)
SLBs are more unreliable and unsafe than new batteries. The power-
electronics-based intelligent devices can swiftly protect the SL-BESS
from detrimental damages such as catching fire and short circuits; (3)
X. Gu et al.
Fig. 6. Core power electronic technologies that enable more efficient, reliable, and
safer SL-BESS.
SLBs and power converters are cascaded in an SL-BESS. Given the lossy
SLBs, more efficient power conversion circuits and architectures can
enhance the overall power efficiency of SL-BESS. Additionally, when
considering the substantially lower price of SL batteries compared to
fresh batteries and more use of higher-rated power converters, the
cost associated with power electronics can become a more significant
portion of the overall SL-BESS cost. However, to the best of the authors’
knowledge, there have been no review papers that extensively sum-
marize power electronics technologies for SLBs. While [67] reviewed
power converter topologies for SL-BESS, it did not specifically address
power processing architectures. Therefore, we present a comprehensive
review of four power-electronics technologies that offer advantages
to the SL-BESS: (1) efficient energy transformation, (2) active equal-
ization, (3) network reconfiguration, and (4) system integration, as
summarized in Fig. 6.
5.1. Efficient energy transfer
Different power electronics circuits are needed to transform the DC
energy of batteries into DC, Pulse, and alternating current (AC) forms in
the DC microgrid, pulse-charging of batteries, and AC grid, respectively.
In all transformations, energy efficiency is a key performance indica-
tor for battery storage systems. SLBs are lossier due to the increase
in resistance. Moreover, SLBs are more complicated in the output
impedance due to the SEI layer growth [167]. This impedance signifi-
cantly varies during the cycling process. By deploying more efficient
power converters [168], the thermal management cost of the entire
SL-BESS can be decreased, the energy efficiency improves, and battery
life can be considerably extended. Soft-switching techniques [169],
variable operation modes [170], and design optimization [171] have
been utilized to improve the efficiency of power converters as well as
the entire SL-BESS.
5.2. Active equalization
Equalization can improve the power and energy utilization of bat-
tery systems. The energy system with SLBs, which are heterogeneous
in health conditions, can especially benefit from equalization. Passive
equalization dissipates the excess energy as heat to homogenize the bat-
tery units. The speed and efficiency of equalization can be significantly
15
Renewable and Sustainable Energy Reviews 192 (2024) 114191
improved by the emerging active equalization technologies. In active
equalization, the power flows among batteries are actively regulated by
power converters. Active equalization can uniform the energy, power,
and SOH. In active energy equalization, power electronics networks,
together with state estimation and adaptive algorithms, can equally
deplete the energy of all heterogeneous units at the same time [166,
172]. The ratings of the power converters can be customized so that the
power capability of each battery unit is utilized and the power output of
BESS can be maximized [173,174]. The active SOH equalization lever-
ages the information on the battery degradation dynamics. With the
information on the battery degradation dynamics, the power converters
can drive the unhealthy battery unit less, drive the healthy unit more,
and allow the degradation conditions of all units to converge [175].
In comparison to the existing literature that focuses mostly on cell-
level equalization, this paper reviews the equalization architectures in
a way that is agnostic to the battery cells, modules, and packs. State-
of-the-art equalization circuit architectures can be categorized into full
power processing (FPP) and partial power processing (PPP).
5.2.1. Full power processing
In FPP, each battery unit delivers power to the load through a
power converter. The power that permeates the SLBs needs to flow
through the power converters. The power converter is designed to
compensate for the heterogeneous electrical characteristics of SLBs. The
circuit architectures of FPP will be reviewed in both DC–DC and DC–AC
conversions as shown in Figs. 7 and 8, respectively. In both conversions,
the SLB-converter units can be connected in series or parallel [176] to
the output.
In the DC–DC conversion, each SLB unit is connected to the DC bus
through a DC–DC converter, as shown in Fig. 7(a). The topology of DC–
DC converters varies with the performance specifications [176,177].
Mukherjee et al. [177] used the boost topology because it can step up
the low battery voltage, decrease the number of series batteries, and
carry a small ripple current to batteries. The work by [176] discussed an
isolated 𝐿𝐶-resonant topology for wider output step-up range, higher
power efficiency, and galvanic isolation. Two architectures are typically
used in DC–AC conversion. One architecture is a two-stage that includes
a DC–DC stage and a DC–AC stage, as shown in Fig. 8(b). The first
DC–DC stage is the same as the DC–DC conversion and the second DC–
AC stage usually contains a bulk inverter [177]. The other architecture
is the series or parallel combinations of microinverters [178,179], as
shown in Fig. 8(a). A DC–DC converter can be cascaded in the front of
each modular inverter so that the power flow of each battery can be
individually regulated [178–180], as shown in Fig. 8(c).
5.2.2. Partial power processing
In PPP, the major power flow is directly delivered from batteries
to the load. The power converters only process the power mismatch
among batteries, therefore the energy utilization and efficiency of
the system are increased. Most PPP reported in the literature is for
series-output, DC–DC conversion architecture [181–183]. According
to the layers of power conversion networks, PPP can be classified as
flat PPP and hierarchical PPP. Flat PPP architecture contains a single
layer of power conversion network, as shown in Fig. 7(b). There are
a variety of implementations of the flat PPP architecture for DC–
DC conversion [178,179,181,181,182,182,183,183–185]. The research
in [181–183] connected a power converter to two adjacent batteries to
exchange the mismatched energy. In [186–189], the authors connected
a converter between each battery and the output bus. Compared to the
FPP, this converter only processes the power mismatch between the
battery and the output bus. [184] proposed a virtual bus architecture
that substitutes this output bus with an extra DC bus that is formed by
a capacitor. In [185], batteries exchange energy through an AC bus. A
few papers have investigated the PPP solutions for parallel SLB units.
For example, La et al. [190] used the dynamical resistor to control the
X. Gu et al. Renewable and Sustainable Energy Reviews 192 (2024) 114191

Fig. 7. Different types of DC–DC power processing architectures: (a) partial power processing; (b) full power processing.

Fig. 8. Different types of DC–AC power processing architectures: (a) one-stage full power processing; (b) two-stage full power processing with a bulk inverter; (c) two-stage full
power processing with microinverters; (d) one-stage partial power processing.

16
X. Gu et al.
current of each SLB unit. The papers that demonstrated PPP for DC–
AC conversion majorly utilized Multilevel Modular Converters (MMC)
architectures, as shown in Fig. 8(d), and adaptive battery balancing
algorithms [178,179].
The performance of flat PPP degrades as the heterogeneity of SLBs
grows. This motivates the hierarchical PPP architectures, where multi-
ple layers of power converters are utilized to equalize the heterogeneity
among not only neighbor batteries but also far-end batteries [173,174,
191,192]. Zhang et al. [191] demonstrated that a two-layer series-
output hierarchical PPP can reduce energy loss, lower the current
rating, and decrease the cost and volume of the system. The design
of hierarchical PPP is complicated. The work in [173,174] proposed
a systematic data-driven design methodology that is based on the
statistical distribution of SLB supply. The power converter cost and
system power efficiency are significantly decreased compared to the
flat PPP. Cui et al. [192] derived a parallel-output hierarchical PPP for
managing the SLB packs as EV charging buffering.
5.3. Network reconfiguration
SLBs are more dispersed in their capacity and capability. The life
of the whole SL-BESS should not be affected by the earlier termination
of a unit. Converters should be designed so that the weakest reliability
link is not on power converters. In the existing literature, redundancy,
dynamic reconfigurability, and hot-swapping technologies have been
used to enhance the reliability of SL-BESS. Mukherjee et al. [178] cre-
ated redundancy by using a ‘‘k-out-of-n’’ system, where the SL-BESS can
operate as long as k battery-power converter units out of n modules are
working properly. The work by [193] proposed an adaptive reconfig-
urable multi-cell structure to improve reliability. Maughan et al. [176]
proposed a hot-swapping approach that allows the whole system to
operate seamlessly and without any interruption when battery packs
need to be disconnected in faults.
SLBs are highly degraded and more vulnerable to thermal runaway,
short circuits, and overcharge. The power-electronics-based intelligent
devices play major roles in fault protection, diagnostics, and prog-
nostics. From the fault protection perspective, as exhibited in [193],
the solid-state breakers can break the short-circuit faults much faster
than the mechanical contractors. From the prognostics perspective, as
demonstrated in [194,195], the power conversion network can be dual-
used to perform the online Electrochemical Impedance Spectroscopy
and online SOH so that the degradation condition is monitored.
5.4. System integration
The major application of SL-BESS is in stationary energy storage
systems [196]. According to the mission profiles that the SL-BESS
performs in the system, SL-BESS can be classified into four categories:
energy-intensive DC–DC conversion, energy-intensive DC–AC conver-
sion, power-intensive DC–DC conversion, and power-intensive DC–AC
conversion.
In energy-intensive DC–DC conversion, SL-BESS colocated with the
solar can support DC microgrids [197–199]. The energy-intensive DC–
AC conversion case, for example, includes the SLBs directly integrated
into the AC power grid or AC microgrids for the services of peak shav-
ing, spinning reserve, or renewable firming [67,200,201]. In both cases,
the system-level control question is whether SLBs with different energy
levels can be proportionally utilized and depleted simultaneously.
In power-intensive DC–DC conversion, SLBs act as a buffer for
the EV’s fast charging [197,198,202]. The power-intensive DC–AC
conversion case, for example, includes the SLBs integrated with the
grid-forming inverters for transmission stabilization [67,200] as well as
frequency regulation [203] and uninterrupted power supply for public
building backups [181]. In both cases, the system-level control question
is whether BESS can push high enough power support in a short period
of time.
17
Renewable and Sustainable Energy Reviews 192 (2024) 114191
6. Conclusions
Due to the high volume of EVs being in service and the limited
lifetime of the onboarding LIBs, the large-scale retirement of LIBs will
come in the near future. Since SLBs retired from EVs still own 70%–
80% of the initial capacity, they have the potential to be utilized in
scenarios with lower energy and power requirements. In this way, the
value of LIBs can be maximized in their second-life applications. This
paper reviews critical technologies and the economics of SLBs. The
main conclusions are as follows:
(1) The degradation models for FLBs mainly include PMBs, DDMs,
and hybrid models. While physics-based models are explainable, they
cannot be applied to situations where the degradation modes are not
well understood. In contrast, DDMs circumvent this issue by leveraging
high-quality data, but their interpretability is limited. Hence, there is
an opportunity to apply hybrid models to establish explainable models
with physical laws that can be generalized. The missing historical data
over the first-life operation and poorly understood knee points have
posed two challenges to accurately modeling the degradation behaviors
of SLBs. At present, there is a dearth of degradation models specifically
designed for SLBs. Therefore, we investigate the feasibility of utilizing
battery degradation models developed for FLBs in the context of SLBs.
(2) The representative procedure for assessing retired batteries
encompasses five pivotal stages: initial assessment, disassembly, fast
screening, regrouping, and management strategy. This work summa-
rizes comprehensive processes, state-of-the-art, challenges and poten-
tial solutions, along with regulations and standards revolving around
the whole assessment procedure of retired batteries.
(3) For economic considerations, only those profitable SLB applica-
tions are worth further promotion. Therefore, it is crucial to conduct
a thorough economic analysis of SLBs prior to their implementation
in second-life applications. This paper provides a summary of the
profitable applications of SLBs and those that hold significant potential.
Specifically, SLBs have proven to be beneficial in a variety of applica-
tions, including renewable firming, area and frequency regulation, and
peak shaving. Additionally, the integration of SLBs in micro-grid and
smart-grid applications has also shown feasibility.
(4) Finally, power electronics serve as the critical energy conver-
sion interfaces among batteries in SLB applications, and appropriately
designed power converters have the potential to significantly enhance
the performance of SLB applications in terms of power, electrical char-
acteristics, and reliability. This paper provides a comprehensive review
of advanced power electronics technologies, including high-efficiency
energy transformation, active equalization, network reconfiguration,
and system integration.
Declaration of competing interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Data availability
No data was used for the research described in the article.
Acknowledgments
This work is supported by the National University of Singapore
Start-Up Grant (A-0009527-01-00).
X. Gu et al.
References
[1] Roland I. EV-Volumes - The electric vehicle world sales database. 2023, URL
https://www.ev-volumes.com/country/total-world-plug-in-vehicle-volumes/.
[2] Lai X, Huang Y, Gu H, Deng C, Han X, Feng X, et al. Turning waste into
wealth: A systematic review on echelon utilization and material recycling of
retired lithium-ion batteries. Energy Storage Mater 2021;40(March):96–123.
http://dx.doi.org/10.1016/j.ensm.2021.05.010.
[3] Harper G, Sommerville R, Kendrick E, Driscoll L, Slater P, Stolkin R, et al. Recy-
cling lithium-ion batteries from electric vehicles. Nature 2019;575(7781):75–86.
http://dx.doi.org/10.1038/s41586-019-1682-5.
[4] Wandt J, Jakes P, Granwehr J, Eichel R-A, Gasteiger HA. Quantitative and
time-resolved detection of lithium plating on graphite anodes in lithium ion
batteries. Mater Today 2018;21(3):231–40. http://dx.doi.org/10.1016/j.mattod.
2017.11.001.
[5] Vykhodtsev AV, Jang D, Wang Q, Rosehart W, Zareipour H. A review of
modelling approaches to characterize lithium-ion battery energy storage systems
in techno-economic analyses of power systems. In: Renewable and sustainable
energy reviews, vol. 166, Elsevier Ltd; 2022, http://dx.doi.org/10.1016/j.rser.
2022.112584.
[6] Gruber K, Gauster T, Laaha G, Regner P, Schmidt J. Profitability and investment
risk of Texan power system winterization. Nat Energy 2022;7(5):409–16. http:
//dx.doi.org/10.1038/s41560-022-00994-y.
[7] Madlener R, Kirmas A. Economic viability of second use electric vehicle
batteries for energy storage in residential applications. In: Energy Procedia,
vol. 105, Elsevier Ltd; 2017, p. 3806–15. http://dx.doi.org/10.1016/j.egypro.
2017.03.890.
[8] Wu W, Lin B, Xie C, Elliott RJ, Radcliffe J. Does energy storage provide
a profitable second life for electric vehicle batteries? Energy Econ 2020;92.
http://dx.doi.org/10.1016/j.eneco.2020.105010.
[9] Li J, He S, Yang Q, Wei Z, Li Y, He H. A comprehensive review of second
life batteries towards sustainable mechanisms: Potential, challenges, and future
prospects. IEEE Trans Transp Electrif 2022. http://dx.doi.org/10.1109/TTE.
2022.3220411.
[10] Birkl CR, Roberts MR, McTurk E, Bruce PG, Howey DA. Degradation diagnostics
for lithium ion cells. J Power Sources 2017;341:373–86. http://dx.doi.org/10.
1016/J.JPOWSOUR.2016.12.011.
[11] Han X, Lu L, Zheng Y, Feng X, Li Z, Li J, et al. A review on the key issues of
the lithium ion battery degradation among the whole life cycle. eTransportation
2019;1. http://dx.doi.org/10.1016/j.etran.2019.100005.
[12] Hu X, Xu L, Lin X, Pecht M. Battery lifetime prognostics. Joule
2020;4(2):310–46. http://dx.doi.org/10.1016/j.joule.2019.11.018.
[13] Röder F, Braatz RD, Krewer U. Direct coupling of continuum and kinetic Monte
Carlo models for multiscale simulation of electrochemical systems. Comput
Chem Eng 2019;121:722–35. http://dx.doi.org/10.1016/J.COMPCHEMENG.
2018.12.016.
[14] Methekar RN, Northrop PWC, Chen K, Braatz RD, Subramanian VR. Kinetic
Monte Carlo simulation of surface heterogeneity in graphite anodes for lithium-
ion batteries: Passive layer formation. J Electrochem Soc 2011;158(4):A363.
http://dx.doi.org/10.1149/1.3548526, URL https://iopscience.iop.org/article/
10.1149/1.3548526/meta.
[15] Leung K, Budzien JL. Ab initio molecular dynamics simulations of the initial
stages of solid–electrolyte interphase formation on lithium ion battery graphitic
anodes. Phys Chem Chem Phys 2010;12(25):6583–6. http://dx.doi.org/
10.1039/B925853A, URL https://pubs.rsc.org/en/content/articlehtml/2010/cp/
b925853a. https://pubs.rsc.org/en/content/articlelanding/2010/cp/b925853a.
[16] Leung K. Electronic structure modeling of electrochemical reactions at
electrode/electrolyte interfaces in lithium ion batteries. J Phys Chem C
2013;117(4):1539–47. http://dx.doi.org/10.1021/jp308929a, URL https://pubs.
acs.org/sharingguidelines.
[17] Yang XG, Leng Y, Zhang G, Ge S, Wang CY. Modeling of lithium plating induced
aging of lithium-ion batteries: Transition from linear to nonlinear aging. J
Power Sources 2017;360:28–40. http://dx.doi.org/10.1016/j.jpowsour.2017.05.
110.
[18] O’Kane SEJ, Ai W, Madabattula G, Alonso-Alvarez D, Timms R, Sulzer V, et
al. Lithium-ion battery degradation: How to model it. Phys Chem Chem Phys
2022;24(13):7909–22. http://dx.doi.org/10.1039/D2CP00417H.
[19] Ulrike Krewer. Review-dynamic models of Li-Ion batteries for diagnosis and op-
eration: A review and perspective. J Electrochem Soc 2018;165(16):A3656–73.
http://dx.doi.org/10.1149/2.1061814jes.
[20] Ng SS, Xing Y, Tsui KL. A naive Bayes model for robust remaining useful
life prediction of lithium-ion battery. Appl Energy 2014;118:114–23. http:
//dx.doi.org/10.1016/J.APENERGY.2013.12.020.
[21] Wang S, Zhao L, Su X, Ma P. Prognostics of lithium-ion batteries based
on battery performance analysis and flexible support vector regression. Ener-
gies 2014;7(10):6492–508. http://dx.doi.org/10.3390/EN7106492, URL https:
//www.mdpi.com/1996-1073/7/10/6492/htm.
[22] Richardson RR, Osborne MA, Howey DA. Gaussian process regression for
forecasting battery state of health. J Power Sources 2017;357:209–19. http:
//dx.doi.org/10.1016/j.jpowsour.2017.05.004.
18
Renewable and Sustainable Energy Reviews 192 (2024) 114191
[23] Li L, Wang P, Chao KH, Zhou Y, Xie Y. Remaining useful life pre-
diction for lithium-ion batteries based on Gaussian processes mixture.
PLoS One 2016;11(9):e0163004. http://dx.doi.org/10.1371/JOURNAL.PONE.
0163004, URL https://journals.plos.org/plosone/article?id=10.1371/journal.
pone.0163004.
[24] Zhang Y, Xiong R, He H, Pecht MG. Long short-term memory recurrent neural
network for remaining useful life prediction of lithium-ion batteries. IEEE
Trans Veh Technol 2018;67(7):5695–705. http://dx.doi.org/10.1109/TVT.2018.
2805189.
[25] Liu D, Xie W, Liao H, Peng Y. An integrated probabilistic approach to
lithium-ion battery remaining useful life estimation. IEEE Trans Instrum Meas
2015;64(3):660–70. http://dx.doi.org/10.1109/TIM.2014.2348613.
[26] Severson KA, Attia PM, Jin N, Perkins N, Jiang B, Yang Z, et al. Data-
driven prediction of battery cycle life before capacity degradation. Nat Energy
2019;4(5):383–91. http://dx.doi.org/10.1038/s41560-019-0356-8.
[27] Hsu C-w, Xiong R, Chen N-y, Li J, Tsou N-t. Deep neural network battery
life and voltage prediction by using data of one cycle only. Appl Energy
2022;306(July 2021).
[28] Wright RB, Motloch CG, Belt JR, Christophersen JP, Ho CD, Richardson RA, et
al. Calendar- and cycle-life studies of advanced technology development pro-
gram generation 1 lithium-ion batteries. J Power Sources 2002;110(2):445–70.
http://dx.doi.org/10.1016/S0378-7753(02)00210-0.
[29] Bloom I, Cole BW, Sohn JJ, Jones SA, Polzin EG, Battaglia VS, et al. An
accelerated calendar and cycle life study of Li-ion cells. J Power Sources
2001;101(2):238–47. http://dx.doi.org/10.1016/S0378-7753(01)00783-2.
[30] Tseng KH, Liang JW, Chang W, Huang SC. Regression models using fully
discharged voltage and internal resistance for state of health estimation of
lithium-ion batteries. Energies 2015;8(4):2889–907. http://dx.doi.org/10.3390/
EN8042889, URL https://www.mdpi.com/1996-1073/8/4/2889.
[31] Lu J, Wei L, Malek Pour M, Mekonnen Y, Sarwat AI. Modeling discharge
characteristics for predicting battery remaining life. In: IEEE transportation
electrification conference and expo. Chicago; 2017, p. 468–73.
[32] Seger PV, Thivel P-X, Riu D. A second life Li-ion battery ageing model with
uncertainties: From cell to pack analysis. J Power Sources 2022;541:231663.
http://dx.doi.org/10.1016/J.JPOWSOUR.2022.231663, URL https://linkinghub.
elsevier.com/retrieve/pii/S0378775322006619.
[33] Aykol M, Gopal CB, Anapolsky A, Herring PK, van Vlijmen B, Berliner MD, et
al. Perspective—Combining physics and machine learning to predict battery
lifetime. J Electrochem Soc 2021;168(3):030525. http://dx.doi.org/10.1149/
1945-7111/abec55.
[34] Guo W, Sun Z, Vilsen SrB, Meng J, Stroe DI. Review of ‘‘grey box’’ lifetime
modeling for lithium-ion battery: Combining physics and data-driven methods. J
Energy Storage 2022;56:105992. http://dx.doi.org/10.1016/j.est.2022.105992,
URL https://linkinghub.elsevier.com/retrieve/pii/S2352152X22019806.
[35] Saehong Park, Dong Zhang, Moura S. Hybrid electrochemical modeling with
recurrent neural networks for li-ion batteries. In: 2017 American control con-
ference. Seattle, WA, USA: IEEE; 2017, p. 3777–82. http://dx.doi.org/10.23919/
ACC.2017.7963533, URL https://ieeexplore.ieee.org/document/7963533/.
[36] Li W, Zhang J, Ringbeck F, Jöst D, Zhang L, Wei Z, et al. Physics-
informed neural networks for electrode-level state estimation in lithium-
ion batteries. J Power Sources 2021;506:230034. http://dx.doi.org/10.
1016/j.jpowsour.2021.230034, URL https://linkinghub.elsevier.com/retrieve/
pii/S0378775321005619.
[37] Li W, Demir I, Cao D, Jöst D, Ringbeck F, Junker M, et al. Data-driven system-
atic parameter identification of an electrochemical model for lithium-ion bat-
teries with artificial intelligence. Energy Storage Mater 2022;44:557–70. http://
dx.doi.org/10.1016/j.ensm.2021.10.023, URL https://linkinghub.elsevier.com/
retrieve/pii/S2405829721004864.
[38] Karniadakis GE, Kevrekidis IG, Lu L, Wang S, Yang L. Physics-informed machine
learning ability. Nat Rev Phys 2021;3. ISBN: 0123456789.
[39] Zubov K, McCarthy Z, Ma Y, Calisto F, Pagliarino V, Azeglio S, et al.
NeuralPDE: Automating physics-informed neural networks (PINNs) with error
approximations. 2021, p. 77, arXiv.
[40] Aykol M, Gopal CB, Anapolsky A, Herring PK, van Vlijmen B, Berliner MD, et
al. Perspective—Combining physics and machine learning to predict battery
lifetime. J Electrochem Soc 2021;168(3):030525. http://dx.doi.org/10.1149/
1945-7111/abec55, Publisher: IOP Publishing.
[41] Hu X, Deng X, Wang F, Deng Z, Lin X, Teodorescu R, et al. A review
of second-life lithium-ion batteries for stationary energy storage applica-
tions. Proc IEEE 2022;110(6):735–53. http://dx.doi.org/10.1109/JPROC.2022.
3175614, URL https://ieeexplore.ieee.org/document/9788509/.
[42] Waldmann T, Quinn JB, Richter K, Kasper M, Tost A, Klein A, et al. Electro-
chemical, post-mortem, and ARC analysis of Li-Ion cell safety in second-life
applications. J Electrochem Soc 2017;164(13):A3154–62. http://dx.doi.org/10.
1149/2.0961713jes.
[43] Attia PM, Bills A, Brosa Planella F, Dechent P, dos Reis G, Dubarry M, et al.
Review—‘‘Knees’’ in lithium-ion battery aging trajectories. J Electrochem Soc
2022;169(6):060517. http://dx.doi.org/10.1149/1945-7111/ac6d13.
[44] Severson KA, Attia PM, Jin N, Perkins N, Jiang B, Yang Z, et al. Data-
driven prediction of battery cycle life before capacity degradation. Nat. Energy
2019;4(5):383–91. http://dx.doi.org/10.1038/s41560-019-0356-8.
X. Gu et al.
[45] Saha B, Goebel K. Battery data set. NASA Prognostics Data Repository, NASA
Ames Research Center, Moffett Field, CA; 2007.
[46] Bole B, Kulkarni CS, Daigle M. Adaptation of an electrochemistry-based Li-Ion
battery model to account for deterioration observed under randomized use. In:
Annual Conference of the PHM Society, vol. 6, (no. 1). 2014, http://dx.doi.
org/10.36001/PHMCONF.2014.V6I1.2490, URL https://papers.phmsociety.org/
index.php/phmconf/article/view/2490.
[47] University of Maryland. CALCE Battery Datasets. 2014, URL https://web.calce.
umd.edu/batteries/data.
[48] Attia PM, Grover A, Jin N, Severson KA, Markov TM, Liao YH, et al. Closed-
loop optimization of fast-charging protocols for batteries with machine learning.
Nature 2020;578(7795):397–402. http://dx.doi.org/10.1038/s41586-020-1994-
5.
[49] Birkl C. Oxford battery degradation dataset 1. University of Oxford; 2017.
[50] Kollmeyer P. Panasonic 18650PF Li-ion battery data. Mendeley Data 2018;1.
http://dx.doi.org/10.17632/WYKHT8Y7TG.1.
[51] Zhu J, Wang Y, Huang Y, Bhushan Gopaluni R, Cao Y, Heere M,
et al. Data-driven capacity estimation of commercial lithium-ion batteries
from voltage relaxation. Nature Commun 2022;13(1):2261. http://dx.doi.org/
10.1038/s41467-022-29837-w, URL https://www.nature.com/articles/s41467-
022-29837-w. Number: 1 Publisher: Nature Publishing Group.
[52] Finegan DP, Zhu J, Feng X, Keyser M, Ulmefors M, Li W, et al. Perspective the
application of data-driven methods and physics-based learning for improving
battery safety. Joule 2021;5(2):316–29. http://dx.doi.org/10.1016/j.joule.2020.
11.018.
[53] Engel H, Hertzke P, Siccardo G. Second-life EV batteries: The newest value
pool in energy storage. 2019, URL https://www.mckinsey.com/industries/
automotive-and-assembly/our-insights/second-life-ev-batteries-the-newest-
value-pool-in-energy-storage#/.
[54] McKinsey & Company. Battery 2030: Resilient, Sustainable, and Circular.
Tech. Rep., 2023, URL https://www.mckinsey.com/industries/automotive-and-
assembly/our-insights/battery-2030-resilient-sustainable-and-circular#/.
[55] Zhu J, Mathews I, Ren D, Li W, Cogswell D, Xing B, et al. End-of-life or second-
life options for retired electric vehicle batteries. Cell Rep Phys Sci 2021;2(8).
http://dx.doi.org/10.1016/j.xcrp.2021.100537.
[56] European Commission. Green Deal: Sustainable batteries for a circular and
climate neutral economy. 2020, URL https://single-market-economy.ec.
europa.eu/news/green-deal-sustainable-batteries-circular-and-climate-neutral-
economy-2020-12-10_en.
[57] Global implications of the EU battery regulation 373.
[58] Weng A, Dufek E, Stefanopoulou A. Battery passports for promoting electric
vehicle resale and repurposing. Joule 2023;7(5):837–42. http://dx.doi.org/
10.1016/j.joule.2023.04.002, URL https://linkinghub.elsevier.com/retrieve/pii/
S2542435123001368.
[59] SAE. Recommended practices for shipping transport and handling of
automotive-Type battery system - lithium ion J2950_202006. 2020, URL https:
//www.sae.org/standards/content/j2950_202006.
[60] SAE. Standards for Battery secondary use J2997. URL https://www.sae.org/
standards/content/j2997/.
[61] Lander L, Tagnon C, Nguyen-Tien V, Kendrick E, Elliott RJ, Abbott AP, et
al. Breaking it down: A techno-economic assessment of the impact of battery
pack design on disassembly costs. Appl Energy 2023;331:120437. http://
dx.doi.org/10.1016/j.apenergy.2022.120437, URL https://linkinghub.elsevier.
com/retrieve/pii/S0306261922016944.
[62] Trading Economics. Wages in manufacturing. URL https://tradingeconomics.
com/country-list/wages-in-manufacturing.
[63] Rallo H, Benveniste G, Gestoso I, Amante B. Economic analysis of the disassem-
bling activities to the reuse of electric vehicles Li-ion batteries. Resour Conserv
Recy 2020;159:104785. http://dx.doi.org/10.1016/j.resconrec.2020.104785,
URL https://linkinghub.elsevier.com/retrieve/pii/S0921344920301063.
[64] Meng K, Xu G, Peng X, Youcef-Toumi K, Li J. Intelligent disassembly of
electric-vehicle batteries: A forward-looking overview. Resour Conserv Recy
2022;182:106207. http://dx.doi.org/10.1016/J.RESCONREC.2022.106207.
[65] Lai X, Huang Y, Deng C, Gu H, Han X, Zheng Y, et al. Sorting, regrouping,
and echelon utilization of the large-scale retired lithium batteries: A critical
review. Renew Sustain Energy Rev 2021;146(March):111162. http://dx.doi.org/
10.1016/j.rser.2021.111162.
[66] Li W, Zhu J, Xia Y, Li W, Zhu J, Xia Y, Gorji MB, et al. Data-driven safety
envelope of lithium-ion batteries for electric vehicles data-driven safety enve-
lope of lithium-ion batteries for electric vehicles. Joule 2019;3(11):2703–15.
http://dx.doi.org/10.1016/j.joule.2019.07.026.
[67] Hossain E, Murtaugh D, Mody J, Faruque HMR, Sunny MSH, Mohammad N.
A comprehensive review on second-life batteries: Current state, manufactur-
ing considerations, applications, impacts, barriers potential solutions, business
strategies, and policies. IEEE Access 2019;7:73215–52. http://dx.doi.org/10.
1109/ACCESS.2019.2917859.
[68] Garg A, Yun L, Gao L, Putungan DB. Development of recycling strategy for
large stacked systems: Experimental and machine learning approach to form
reuse battery packs for secondary applications. J Clean Prod 2020;275:124152.
19
Renewable and Sustainable Energy Reviews 192 (2024) 114191
[69] Takahashi A, Allam A, Onori S. Evaluating the feasibility of batteries for
second-life applications using machine learning. iScience 2023;26(4):106547.
http://dx.doi.org/10.1016/j.isci.2023.106547, URL https://linkinghub.elsevier.
com/retrieve/pii/S2589004223006247.
[70] Lai X, Qiao D, Zheng Y, Ouyang M, Han X, Zhou L. A rapid screening
and regrouping approach based on neural networks for large-scale retired
lithium-ion cells in second-use applications. J Clean Prod 2019;213:776–91.
http://dx.doi.org/10.1016/j.jclepro.2018.12.210.
[71] Braco E, Martín IS, Sanchis P, Ursúa A. Fast capacity and internal resistance
estimation method for second-life batteries from electric vehicles. Appl Energy
2023;329:120235. http://dx.doi.org/10.1016/j.apenergy.2022.120235.
[72] UL. ANSI/CAN/UL standard for evaluation for repurposing batteries. 2018,
URL https://www.shopulstandards.com/ProductDetail.aspx?productId=UL1974_
1_S_20181025&ShowFreeviewModal=1.
[73] Schneider EL, Oliveira CT, Brito RM, Malfatti CF. Classification of discarded
NiMH and Li-Ion batteries and reuse of the cells still in operational conditions
in prototypes. J Power Sources 2014;262:1–9. http://dx.doi.org/10.1016/J.
JPOWSOUR.2014.03.095.
[74] Li J, Wang Y, Tan X. Research on the classification method for the secondary
uses of retired lithium-ion traction batteries. Energy Procedia 2017;105:2843–9.
http://dx.doi.org/10.1016/j.egypro.2017.03.625.
[75] Jiang Y, Wang Y, Zhang C, Su G, Liu J. Research on group methods of
second-use Li-ion batteries based on k-means clustering model. In: IEEE
transportation electrification conference and expo, ITEC Asia-Pacific 2014 -
Conference proceedings. Institute of Electrical and Electronics Engineers Inc.;
2014, http://dx.doi.org/10.1109/ITEC-AP.2014.6941098.
[76] Celebi ME, Kingravi HA, Vela PA. A comparative study of efficient ini-
tialization methods for the k-means clustering algorithm. Expert Syst Appl
2013;40(1):200–10. http://dx.doi.org/10.1016/J.ESWA.2012.07.021.
[77] Choubin B, Moradi E, Golshan M, Adamowski J, Sajedi-Hosseini F, Mosavi A.
An ensemble prediction of flood susceptibility using multivariate discriminant
analysis, classification and regression trees, and support vector machines.
Sci Total Environ 2019;651:2087–96. http://dx.doi.org/10.1016/J.SCITOTENV.
2018.10.064.
[78] Song Z, Feng S, Zhang L, Hu Z, Hu X, Yao R. Economy analysis of second-
life battery in wind power systems considering battery degradation in dynamic
processes: Real case scenarios. Appl Energy 2019;251. http://dx.doi.org/10.
1016/J.APENERGY.2019.113411.
[79] Zhou Z, Ran A, Chen S, Zhang X, Wei G, Li B, et al. A fast screening
framework for second-life batteries based on an improved bisecting K-means
algorithm combined with fast pulse test. J Energy Storage 2020;31(101739).
http://dx.doi.org/10.1016/j.est.2020.101739.
[80] Muhammad M, Attidekou PS, Ahmeid M, Milojevic Z, Lambert S. Sorting
of spent electric vehicle batteries for second life application. In: Proceedings
of 2019 the 7th international conference on smart energy grid engineering.
Ontario, Canada: IEEE; 2019, p. 325–9. http://dx.doi.org/10.1109/SEGE.2019.
8859921.
[81] Song Z, Yang XG, Yang N, Delgado FP, Hofmann H, Sun J. A study
of cell-to-cell variation of capacity in parallel-connected lithium-ion battery
cells. eTransportation 2021;7:100091. http://dx.doi.org/10.1016/j.etran.2020.
100091.
[82] Song Z, Yang N, Lin X, Pinto Delgado F, Hofmann H, Sun J. Progression of
cell-to-cell variation within battery modules under different cooling structures.
Appl Energy 2022;312(November 2021):118836. http://dx.doi.org/10.1016/j.
apenergy.2022.118836.
[83] Martinez-Laserna E, Sarasketa-Zabala E, Villarreal Sarria I, Stroe DI, Swier-
czynski M, Warnecke A, et al. Technical viability of battery second life: A
study from the ageing perspective. IEEE Trans Ind Appl 2018;54(3):2703–13.
http://dx.doi.org/10.1109/TIA.2018.2801262.
[84] Li X, Yuan C, Li X, Wang Z. State of health estimation for Li-ion battery
using incremental capacity analysis and Gaussian process regression. Energy
2020;190. http://dx.doi.org/10.1016/J.ENERGY.2019.116467.
[85] Rossi F, Tosti L, Basosi R, Cusenza MA, Parisi ML, Sinicropi A. Environmental
optimization model for the European batteries industry based on prospective
life cycle assessment and material flow analysis. Renew Sustain Energy Rev
2023;183:113485. http://dx.doi.org/10.1016/j.rser.2023.113485, URL https://
linkinghub.elsevier.com/retrieve/pii/S1364032123003428.
[86] Arat HT. Recycling and reusing batteries: A significant way for effective
sustainability of FCEVs and EVs. Int J Hydrogen Energy 2023;48(60):23272–
8. http://dx.doi.org/10.1016/j.ijhydene.2023.01.189, URL https://linkinghub.
elsevier.com/retrieve/pii/S0360319923003828.
[87] Tian G, Yuan G, Aleksandrov A, Zhang T, Li Z, Fathollahi-Fard AM, et
al. Recycling of spent Lithium-ion Batteries: A comprehensive review for
identification of main challenges and future research trends. Sustain Energy
Technol Assess 2022;53. http://dx.doi.org/10.1016/j.seta.2022.102447.
[88] Chen M, Ma X, Chen B, Arsenault R, Karlson P, Simon N, et al. Recycling end-
of-life electric vehicle lithium-ion batteries. Joule 2019;3(11):2622–46. http://
dx.doi.org/10.1016/j.joule.2019.09.014, URL https://linkinghub.elsevier.com/
retrieve/pii/S254243511930474X.
X. Gu et al.
[89] Zhao Y, Pohl O, Bhatt AI, Collis GE, Mahon PJ, Rüther T, et al. A review
on battery market trends, second-life reuse, and recycling. Sustain Chem
2021;2(1):167–205. http://dx.doi.org/10.3390/suschem2010011.
[90] Modak NM, Sinha S, Ghosh DK. A review on remanufacturing, reuse, and
recycling in supply chain—Exploring the evolution of information technology
over two decades. Int J Inf Manag Data Insights 2023;3(1):100160. http://
dx.doi.org/10.1016/j.jjimei.2023.100160, URL https://linkinghub.elsevier.com/
retrieve/pii/S2667096823000071.
[91] Seger PV, Thivel P-X, Riu D. A second life Li-ion battery ageing model with
uncertainties: From cell to pack analysis. J Power Sources 2022;541:231663.
http://dx.doi.org/10.1016/j.jpowsour.2022.231663, URL https://linkinghub.
elsevier.com/retrieve/pii/S0378775322006619.
[92] Casals LC, Amante García B, Canal C. Second life batteries lifespan: Rest
of useful life and environmental analysis. J Environ Manag 2019;232:354–
63. http://dx.doi.org/10.1016/j.jenvman.2018.11.046, URL https://linkinghub.
elsevier.com/retrieve/pii/S0301479718313124.
[93] Sathre R, Scown CD, Kavvada O, Hendrickson TP. Energy and climate effects
of second-life use of electric vehicle batteries in California through 2050. J
Power Sources 2015;288:82–91. http://dx.doi.org/10.1016/j.jpowsour.2015.04.
097, URL https://linkinghub.elsevier.com/retrieve/pii/S0378775315007545.
[94] Hossain E, Murtaugh D, Mody J, Faruque HMR, Sunny MSH, Mohammad N.
A comprehensive review on second-life batteries: Current state, manufactur-
ing considerations, applications, impacts, barriers potential solutions, business
strategies, and policies. IEEE Access 2019;7:73215–52. http://dx.doi.org/10.
1109/ACCESS.2019.2917859, Publisher: Institute of Electrical and Electronics
Engineers Inc..
[95] Lai X, Chen J, Chen Q, Han X, Lu L, Dai H, et al. Comprehensive
assessment of carbon emissions and environmental impacts of sodium-ion
batteries and lithium-ion batteries at the manufacturing stage. J Clean Prod
2023;423:138674. http://dx.doi.org/10.1016/j.jclepro.2023.138674, URL https:
//linkinghub.elsevier.com/retrieve/pii/S0959652623028329.
[96] Zhao E, Walker PD, Surawski NC, Bennett NS. Assessing the life cycle cumu-
lative energy demand and greenhouse gas emissions of lithium-ion batteries. J
Energy Storage 2021;43:103193. http://dx.doi.org/10.1016/j.est.2021.103193,
URL https://linkinghub.elsevier.com/retrieve/pii/S2352152X21008926.
[97] Lai X, Huang Y, Gu H, Deng C, Han X, Feng X, et al. Turning waste into wealth:
A systematic review on echelon utilization and material recycling of retired
lithium-ion batteries. Energy Storage Mater 2021;40:96–123. http://dx.doi.org/
10.1016/j.ensm.2021.05.010, URL https://linkinghub.elsevier.com/retrieve/pii/
S2405829721002142.
[98] Colarullo L, Thakur J. Second-life EV batteries for stationary storage applica-
tions in local energy communities. Renew Sustain Energy Rev 2022;169:112913.
http://dx.doi.org/10.1016/j.rser.2022.112913, URL https://linkinghub.elsevier.
com/retrieve/pii/S1364032122007948.
[99] Ahmadi L, Yip A, Fowler M, Young SB, Fraser RA. Environmental feasibility of
re-use of electric vehicle batteries. Sustain Energy Technol Assess 2014;6:64–74.
http://dx.doi.org/10.1016/j.seta.2014.01.006.
[100] Sultana I, Chen Y, Huang S, Rahman MM. Recycled value-added circular energy
materials for new battery application: Recycling strategies, challenges, and
sustainability-a comprehensive review. J Environ Chem Eng 2022;10(6):108728.
http://dx.doi.org/10.1016/j.jece.2022.108728, URL https://linkinghub.elsevier.
com/retrieve/pii/S2213343722016013.
[101] Sharmili N, Nagi R, Wang P. A review of research in the Li-ion battery
production and reverse supply chains. J Energy Storage 2023;68:107622. http:
//dx.doi.org/10.1016/j.est.2023.107622, URL https://linkinghub.elsevier.com/
retrieve/pii/S2352152X23010198.
[102] Börner MF, Frieges MH, Späth B, Spütz K, Heimes HH, Sauer DU, et al.
Challenges of second-life concepts for retired electric vehicle batteries. Cell Rep
Phys Sci 2022;3(10):101095. http://dx.doi.org/10.1016/j.xcrp.2022.101095,
URL https://linkinghub.elsevier.com/retrieve/pii/S2666386422003976.
[103] Kim S, Yi Z, Kunz R, Dufek EJ, Tanim R, Chen B-r, et al. Article accelerated
battery life predictions through synergistic combination of physics-based models
and machine learning through synergistic combination. Cell Rep Phys Sci
2022;101023. http://dx.doi.org/10.1016/j.xcrp.2022.101023.
[104] Song Z, Feng S, Zhang L, Hu Z, Hu X, Yao R. Economy analysis of second-
life battery in wind power systems considering battery degradation in dynamic
processes: Real case scenarios. Appl Energy 2019;251. http://dx.doi.org/10.
1016/j.apenergy.2019.113411.
[105] Yang D, Zhang X, Pan R, Wang Y, Chen Z. A novel Gaussian process regression
model for state-of-health estimation of lithium-ion battery using charging
curve. J Power Sources 2018;384(September 2017):387–95. http://dx.doi.org/
10.1016/j.jpowsour.2018.03.015.
[106] Neubauer J, Pesaran A. The ability of battery second use strategies to impact
plug-in electric vehicle prices and serve utility energy storage applications.
J Power Sources 2011;196(23):10351–8. http://dx.doi.org/10.1016/j.jpowsour.
2011.06.053.
[107] Steckel T, Kendall A, Ambrose H. Applying levelized cost of storage methodol-
ogy to utility-scale second-life lithium-ion battery energy storage systems. Appl
Energy 2021;300. http://dx.doi.org/10.1016/j.apenergy.2021.117309.
20
Renewable and Sustainable Energy Reviews 192 (2024) 114191
[108] Debnath UK, Ahmad I, Habibi D. Gridable vehicles and second life batteries for
generation side asset management in the smart grid. Int J Electr Power Energy
Syst 2016;82:114–23. http://dx.doi.org/10.1016/j.ijepes.2016.03.006.
[109] Rallo H, Canals Casals L, De La Torre D, Reinhardt R, Marchante C, Amante B.
Lithium-ion battery 2nd life used as a stationary energy storage system: Ageing
and economic analysis in two real cases. J Clean Prod 2020;272. http://dx.doi.
org/10.1016/j.jclepro.2020.122584.
[110] Xu B, Zhao J, Zheng T, Litvinov E, Kirschen DS. Factoring the cycle aging
cost of batteries participating in electricity markets. IEEE Trans Power Syst
2018;33(2):2248–59. http://dx.doi.org/10.1109/TPWRS.2017.2733339.
[111] Marques P, Garcia R, Kulay L, Freire F. Comparative life cycle assessment of
lithium-ion batteries for electric vehicles addressing capacity fade. J Clean Prod
2019;229:787–94. http://dx.doi.org/10.1016/j.jclepro.2019.05.026.
[112] Song Z, Li J, Han X, Xu L, Lu L, Ouyang M, et al. Multi-objective optimization of
a semi-active battery/supercapacitor energy storage system for electric vehicles.
Appl Energy 2014;135:212–24. http://dx.doi.org/10.1016/J.APENERGY.2014.
06.087.
[113] Mathews I, Xu B, He W, Barreto V, Buonassisi T, Peters IM. Technoeconomic
model of second-life batteries for utility-scale solar considering calendar and
cycle aging. Appl Energy 2020;269. http://dx.doi.org/10.1016/j.apenergy.2020.
115127.
[114] Xu B, Oudalov A, Ulbig A, Andersson G, Kirschen DS. Modeling of lithium-
ion battery degradation for cell life assessment. IEEE Trans Smart Grid
2018;9(2):1131–40. http://dx.doi.org/10.1109/TSG.2016.2578950.
[115] Pozzato G, Lee SB, Onori S. Modeling degradation of Lithium-ion batteries
for second-life applications: Preliminary results. In: 2021 IEEE conference
on control technology and applications. San Diego, CA, USA: IEEE; 2021,
p. 826–31. http://dx.doi.org/10.1109/CCTA48906.2021.9659267, URL https:
//ieeexplore.ieee.org/document/9659267/.
[116] Casals LC, Amante García B, Canal C. Second life batteries lifespan: Rest of
useful life and environmental analysis. J Environ Manag 2019;232:354–63.
http://dx.doi.org/10.1016/j.jenvman.2018.11.046.
[117] Song J, Chen X. Eco-efficiency of grain production in China based on water
footprints: A stochastic frontier approach. J Clean Prod 2019;236:117685. http:
//dx.doi.org/10.1016/j.jclepro.2019.117685, URL https://linkinghub.elsevier.
com/retrieve/pii/S0959652619325351.
[118] Neubauer J, Pesaran A. The ability of battery second use strategies to impact
plug-in electric vehicle prices and serve utility energy storage applications.
J Power Sources 2011;196(23):10351–8. http://dx.doi.org/10.1016/j.jpowsour.
2011.06.053.
[119] Lai X, Huang Y, Deng C, Gu H, Han X, Zheng Y, et al. Sorting, regrouping,
and echelon utilization of the large-scale retired lithium batteries: A critical
review. In: Renewable and sustainable energy reviews, vol. 146, Elsevier Ltd;
2021, http://dx.doi.org/10.1016/j.rser.2021.111162.
[120] Neubauer J, Pesaran A, Williams B, Ferry M, Eyer J. Techno-economic analysis
of PEV battery second use: Repurposed-battery selling price and commercial and
industrial end-user value. 1, 2012, http://dx.doi.org/10.4271/2012-01-0349,
URL https://www.osti.gov/biblio/1044470.
[121] Arshad F, Lin J, Manurkar N, Fan E, Ahmad A, Tariq MuN, et al. Life
cycle assessment of lithium-ion batteries: A critical review. In: Resources,
conservation and recycling, vol. 180, Elsevier B.V.; 2022, http://dx.doi.org/
10.1016/j.resconrec.2022.106164.
[122] Assunção A, Moura PS, de Almeida AT. Technical and economic assessment
of the secondary use of repurposed electric vehicle batteries in the residential
sector to support solar energy. Appl Energy 2016;181:120–31. http://dx.doi.
org/10.1016/j.apenergy.2016.08.056.
[123] Fallah N, Fitzpatrick C, Killian S, Johnson M. End-of-life electric vehicle
battery stock estimation in Ireland through integrated energy and circular
economy modelling. Resour Conserv Recy 2021;174. http://dx.doi.org/10.
1016/j.resconrec.2021.105753.
[124] Bai B, Xiong S, Song B, Xiaoming M. Economic analysis of distributed solar
photovoltaics with reused electric vehicle batteries as energy storage systems
in China. Renew Sustain Energy Rev 2019;109:213–29. http://dx.doi.org/10.
1016/j.rser.2019.03.048.
[125] Neubauer J, Smith K, Wood E, Pesaran A. Identifying and overcoming critical
barriers to widespread second use of PEV batteries. Tech. rep., 2015, URL
www.nrel.gov/publications.
[126] Sun B, Su X, Wang D, Zhang L, Liu Y, Yang Y, et al. Economic analysis of
lithium-ion batteries recycled from electric vehicles for secondary use in power
load peak shaving in China. J Clean Prod 2020;276. http://dx.doi.org/10.1016/
j.jclepro.2020.123327.
[127] Pistoia G, Liaw B, editors. Behaviour of lithium-ion batteries in electric vehicles.
In: Green energy and technology, Cham: Springer International Publishing;
2018, http://dx.doi.org/10.1007/978-3-319-69950-9, URL http://link.springer.
com/10.1007/978-3-319-69950-9.
[128] Debnath UK, Ahmad I, Habibi D. Quantifying economic benefits of second life
batteries of gridable vehicles in the smart grid. Int J Electr Power Energy Syst
2014;63:577–87. http://dx.doi.org/10.1016/j.ijepes.2014.05.077.
[129] Naum Pinsky. Electric vehicle battery 2nd use study. Tech. rep., Argonne:
Argonne National Laboratory, Electrochemical Technology Program; 1998.
X. Gu et al.
[130] Cready E, Lippert J, Pihl. Technical and economic feasibility of applying used
EV batteries in stationary applications. ’’ 2003. http://dx.doi.org/10.2172/
809607, URL https://www.osti.gov/biblio/809607.
[131] Wolfs P. An economic assessment of ‘‘second use’’ lithium-ion batteries for grid
support. In: 2010 20th Australasian universities power engineering conference.
2010, p. 1–6.
[132] Heymans C, Walker SB, Young SB, Fowler M. Economic analysis of second
use electric vehicle batteries for residential energy storage and load-levelling.
Energy Policy 2014;71:22–30. http://dx.doi.org/10.1016/j.enpol.2014.04.016.
[133] Fallah N, Fitzpatrick C. How will retired electric vehicle batteries perform in
grid-based second-life applications? A comparative techno-economic evaluation
of used batteries in different scenarios. J Clean Prod 2022;361:132281. http:
//dx.doi.org/10.1016/j.jclepro.2022.132281.
[134] Alimisis V, Hatziargyriou ND. Evaluation of a hybrid power plant comprising
used EV-batteries to complement wind power. IEEE Trans Sustain Energy
2013;4(2):286–93. http://dx.doi.org/10.1109/TSTE.2012.2220160.
[135] Veilleux G, Potisat T, Pezim D, Ribback C, Ling J, Krysztofiński A, et al.
Techno-economic analysis of microgrid projects for rural electrification: A
systematic approach to the redesign of Koh Jik off-grid case study. Energy
Sustain Dev 2020;54:1–13. http://dx.doi.org/10.1016/j.esd.2019.09.007, URL
https://linkinghub.elsevier.com/retrieve/pii/S0973082619309160.
[136] Heymans C, Walker SB, Young SB, Fowler M. Economic analysis of second
use electric vehicle batteries for residential energy storage and load-levelling.
Energy Policy 2014;71:22–30. http://dx.doi.org/10.1016/j.enpol.2014.04.016,
Publisher: Elsevier Ltd.
[137] Martinez-Laserna E, Gandiaga I, Sarasketa-Zabala E, Badeda J, Stroe DI,
Swierczynski M, et al. Battery second life: Hype, hope or reality? A critical
review of the state of the art. Renew Sustain Energy Rev 2018;93:701–18.
http://dx.doi.org/10.1016/J.RSER.2018.04.035.
[138] Funke S, Jochem P, Ried S, Gnann T. Fast charging stations with stationary
batteries: A techno-economic comparison of fast charging along highways and
in cities. In: Transportation research procedia, vol. 48, Elsevier B.V.; 2020, p.
3832–49. http://dx.doi.org/10.1016/j.trpro.2020.08.036.
[139] Graber G, Calderaro V, Galdi V, Piccolo A. Battery second-life for dedicated
and shared energy storage systems supporting EV charging stations. Electronics
(Switzerland) 2020;9(6). http://dx.doi.org/10.3390/electronics9060939.
[140] Neto PBL, Saavedra OR, De Souza Ribeiro LA. A dual-battery storage
bank configuration for isolated microgrids based on renewable sources.
IEEE Trans Sustain Energy 2018;9(4):1618–26. http://dx.doi.org/10.1109/
TSTE.2018.2800689.
[141] Net present value (NPV): What it means and steps to calculate It. 2022, URL
https://www.investopedia.com/terms/n/npv.asp.
[142] Gur K, Chatzikyriakou D, Baschet C, Salomon M. The reuse of electrified
vehicle batteries as a means of integrating renewable energy into the European
electricity grid: A policy and market analysis. Energy Policy 2018;113:535–45.
http://dx.doi.org/10.1016/j.enpol.2017.11.002.
[143] Bai H, Song Z. Lithium-ion battery, sodium-ion battery, or redox-flow bat-
tery: A comprehensive comparison in renewable energy systems. J Power
Sources 2023;580:233426. http://dx.doi.org/10.1016/j.jpowsour.2023.233426,
URL https://linkinghub.elsevier.com/retrieve/pii/S0378775323008029.
[144] Kamath D, Arsenault R, Kim HC, Anctil A. Economic and environmental
feasibility of second-life lithium-ion batteries as fast-charging energy storage.
Environ Sci Technol 2020;54(11):6878–87. http://dx.doi.org/10.1021/acs.est.
9b05883.
[145] Steckel T, Kendall A, Ambrose H. Applying levelized cost of storage methodol-
ogy to utility-scale second-life lithium-ion battery energy storage systems. Appl
Energy 2021;300. http://dx.doi.org/10.1016/J.APENERGY.2021.117309.
[146] Xu X, Hu W, Liu W, Wang D, Huang Q, Chen Z. Study on the economic benefits
of retired electric vehicle batteries participating in the electricity markets. J
Clean Prod 2021;286. http://dx.doi.org/10.1016/j.jclepro.2020.125414.
[147] Canals Casals L, Barbero M, Corchero C. Reused second life batteries for
aggregated demand response services. J Clean Prod 2019;212:99–108. http:
//dx.doi.org/10.1016/j.jclepro.2018.12.005.
[148] Mirzaei Omrani M, Jannesari H. Economic and environmental assessment of
reusing electric vehicle lithium-ion batteries for load leveling in the residential,
industrial and photovoltaic power plants sectors. Renew Sustain Energy Rev
2019;116. http://dx.doi.org/10.1016/j.rser.2019.109413.
[149] Saez-De-Ibarra A, Martinez-Laserna E, Koch-Ciobotaru C, Rodriguez P, Stroe DI,
Swierczynski M. Second life battery energy storage system for residential
demand response service. In: Proceedings of the IEEE international conference
on industrial technology, vol. 2015-June, Institute of Electrical and Electronics
Engineers Inc.; 2015, p. 2941–8. http://dx.doi.org/10.1109/ICIT.2015.7125532.
[150] Utility-scale batteries innovation landscape brief. International Renewable
Energy Agency; 2019, URL www.irena.org.
[151] Kamath D, Shukla S, Arsenault R, Kim HC, Anctil A. Evaluating the cost
and carbon footprint of second-life electric vehicle batteries in residential and
utility-level applications. Waste Manag 2020;113:497–507. http://dx.doi.org/
10.1016/j.wasman.2020.05.034.
21
Renewable and Sustainable Energy Reviews 192 (2024) 114191
[152] Bhatt A, Ongsakul W, Madhu M. N. Optimal techno-economic feasibility
study of net-zero carbon emission microgrid integrating second-life battery
energy storage system. Energy Convers Manage 2022;266:115825. http://dx.
doi.org/10.1016/J.ENCONMAN.2022.115825, URL https://linkinghub.elsevier.
com/retrieve/pii/S0196890422006215.
[153] Malhotra A, Battke B, Beuse M, Stephan A, Schmidt T. Use cases for stationary
battery technologies: A review of the literature and existing projects. Renew
Sustain Energy Rev 2016;56:705–21. http://dx.doi.org/10.1016/j.rser.2015.11.
085, URL https://linkinghub.elsevier.com/retrieve/pii/S1364032115013520.
[154] Brauer S, Monhof M, Klor B, Plenter F, Beverungen D, Siemen C. Residential
energy storage from repurposed electric vehicle batteries: Market overview
and development of a service-centered business model. In: 2016 IEEE 18th
conference on business informatics. Paris, France: IEEE; 2016, p. 143–52. http:
//dx.doi.org/10.1109/CBI.2016.24, URL http://ieeexplore.ieee.org/document/
7780309/.
[155] Chen H, Zhang T, Gao Q, Han Z, Jin Y, Li L, et al. Assessment and
management of health status in full life cycle of echelon utilization for retired
power lithium batteries. J Clean Prod 2022;379:134583. http://dx.doi.org/10.
1016/j.jclepro.2022.134583, URL https://linkinghub.elsevier.com/retrieve/pii/
S0959652622041555.
[156] Antônio Rufino Júnior C, Sanseverino ER, Gallo P, Koch D, Schweiger H-
G, Zanin H. Blockchain review for battery supply chain monitoring and
battery trading. Renew Sustain Energy Rev 2022;157:112078. http://dx.doi.org/
10.1016/j.rser.2022.112078, URL https://linkinghub.elsevier.com/retrieve/pii/
S1364032122000089.
[157] Rufino Júnior CA, Riva Sanseverino E, Gallo P, Koch D, Kotak Y, Schweiger H-
G, et al. Towards a business model for second-life batteries: Barriers,
opportunities, uncertainties, and technologies. J Energy Chem 2023;78:507–
25. http://dx.doi.org/10.1016/j.jechem.2022.12.019, URL https://linkinghub.
elsevier.com/retrieve/pii/S2095495622006842.
[158] Peck ME, Wagman D. Energy trading for fun and profit buy your neighbor’s
rooftop solar power or sell your own-it’ll all be on a blockchain. IEEE
Spectr 2017;54(10):56–61. http://dx.doi.org/10.1109/MSPEC.2017.8048842,
URL https://ieeexplore.ieee.org/document/8048842. Conference Name: IEEE
Spectrum.
[159] Stolz B, Held M, Georges G, Boulouchos K. Techno-economic analysis
of renewable fuels for ships carrying bulk cargo in Europe. Nat Energy
2022;7(2):203–12. http://dx.doi.org/10.1038/s41560-021-00957-9.
[160] IEA G-SS. Grid-scale storage – analysis. IEA; 2022, URL https://www.iea.org/
reports/grid-scale-storage.
[161] Europe invests in research on the batteries of the future | shaping Europe’s
digital future. 2020, URL https://digital-strategy.ec.europa.eu/en/news/europe-
invests-research-batteries-future.
[162] Law for promotion of effective utilization of resources. URL https://lpdd.org/
resources/law-for-promotion-of-effective-utilization-of-resources/.
[163] ESG S. China’s NEV battery recycling: Two main methods and current prob-
lems. 2021, URL https://senecaesg.com/insights/chinas-nev-battery-recycling-
two-main-methods-and-current-problems/. Section: 2021 02 04.
[164] State Council Gazette Issue No.33 Serial No.1752 (Nov 30, 2021). URL
https://english.www.gov.cn/archive/statecouncilgazette/202111/30/content_
WS61a5749fc6d0df57f98e5bfe.html.
[165] Lai X, Chen Q, Tang X, Zhou Y, Gao F, Guo Y, et al. Critical review
of life cycle assessment of lithium-ion batteries for electric vehicles: A
lifespan perspective. eTransportation 2022;12:100169. http://dx.doi.org/10.
1016/j.etran.2022.100169, URL https://linkinghub.elsevier.com/retrieve/pii/
S2590116822000157.
[166] Abdel-Monem M, Hegazy O, Omar N, Trad K, Bossche PVD, Mierlo JV. Lithium-
ion batteries: Comprehensive technical analysis of second-life batteries for smart
grid applications. In: 2017 19th European conference on power electronics
and applications. Institute of Electrical and Electronics Engineers Inc.; 2017,
http://dx.doi.org/10.23919/EPE17ECCEEUROPE.2017.8099385.
[167] Roder P, Stiaszny B, Ziegler JC, Baba N, Lagaly P, Wiemhofer HD.
The impact of calendar aging on the thermal stability of a LiMn2O4-
Li(Ni1/3Mn1/3Co1/3)O2/graphite lithium-ion cell. J Power Sources
2014;268:315–25. http://dx.doi.org/10.1016/j.jpowsour.2014.06.040.
[168] Schimpe M, Piesch C, Hesse HC, Pab J, Ritter S, Jossen A. Power flow
distribution strategy for improved power electronics energy efficiency in battery
storage systems: Development and implementation in a utility-scale system.
Energies 2018;11(3):533. http://dx.doi.org/10.3390/en11030533.
[169] Qian H, Zhang J, Lai JS, Yu W. A high-efficiency grid-tie battery energy storage
system. IEEE Trans Power Electron 2011;26(3):886–96. http://dx.doi.org/10.
1109/TPEL.2010.2096562.
[170] Mukherjee N, Strickland D. Analysis and comparative study of different con-
verter modes in modular second-life hybrid battery energy storage systems.
IEEE J Emerg Sel Top Power Electron 2016;4(2):547–63. http://dx.doi.org/10.
1109/JESTPE.2015.2460334.
[171] Fang Z, Cai T, Duan S, Chen C. Optimal design methodology for llc resonant
converter in battery charging applications based on time-weighted average
efficiency. IEEE Trans Power Electron 2015;30(10):5469–83. http://dx.doi.org/
10.1109/TPEL.2014.2379278.
X. Gu et al.
[172] Liu W, Geng G, Jiang Q, Fan H, Yu J. Model-free fast frequency control support
with energy storage system. IEEE Trans Power Syst 2020;35(4):3078–86. http:
//dx.doi.org/10.1109/TPWRS.2019.2961955.
[173] Cui X, Ramyar A, Siegel J, Peyman M, Stefanopoulou A, Avestruz A-T, et al.
Lite-sparse hierarchical partial power processing for second-use battery energy
storage systems. IEEE Access 2021;1–7, arXiv:2106.11749.
[174] Cui X, Ramyar A, Siegel JB, Mohtat P, Stefanopoulou AG, Avestruz AT.
Comparing power processing system approaches in second-use battery energy
buffering for electric vehicle charging. J Energy Storage 2022;49:104017. http:
//dx.doi.org/10.1016/J.EST.2022.104017.
[175] Azimi V, Allam A, Joe WT, Choi Y, Onori S. Fast charging-minimum degrada-
tion optimal control of series-connected battery modules with DC/DC bypass
converters. In: Proceedings of the American control conference, vol. 2021-
May, Institute of Electrical and Electronics Engineers Inc.; 2021, p. 231–6.
http://dx.doi.org/10.23919/ACC50511.2021.9482982.
[176] Maughan B, Mustafa A, Zane R, Wang H. Hot swapping analysis and sys-
tem operation of series connected converters in active battery reconditioning
systems for second life battery applications. In: 2022 IEEE 23rd workshop
on control and modeling for power electronics. IEEE; 2022, p. 1–6. http://
dx.doi.org/10.1109/COMPEL53829.2022.9830014, URL https://ieeexplore.ieee.
org/document/9830014/.
[177] Mukherjee N, Strickland D. Control of second-life hybrid battery energy storage
system based on modular boost-multilevel buck converter. IEEE Trans Ind
Electron 2015;62(2):1034–46. http://dx.doi.org/10.1109/TIE.2014.2341598.
[178] Mukherjee N, Strickland D. Second life battery energy storage systems: Con-
verter topology and redundancy selection. In: 7th IET international conference
on power electronics, machines and drives. IET; 2014, p. 1–6.
[179] Ma Z, Gao F, Gu X, Li N, Wu Q, Wang X, et al. Multilayer SOH equalization
scheme for MMC Battery energy storage system. IEEE Trans Power Electron
2020;35(12):13514–27. http://dx.doi.org/10.1109/TPEL.2020.2991879.
[180] Mukherjee N, Strickland D. Modular ESS with second life batteries operating
in grid independent mode. In: Proceedings - 2012 3rd IEEE International
symposium on power electronics for distributed generation systems. 2012, p.
653–60. http://dx.doi.org/10.1109/PEDG.2012.6254072.
[181] Milas NT, Tatakis EC. Fast battery cell voltage equaliser based on the bi-
directional flyback converter. IEEE Trans Transp Electrif 2022;1. http://dx.doi.
org/10.1109/TTE.2022.3186520, URL https://ieeexplore.ieee.org/document/
9807338/.
[182] Wang X, Cheng KWE, Fong YC. Series-parallel switched-capacitor balancing
circuit for hybrid source package. IEEE Access 2018;6:34254–61. http://dx.doi.
org/10.1109/ACCESS.2018.2849864.
[183] Mubenga NS, Salami B, Stuart T. Bilevel vs. Passive equalizers for sec-
ond Life EV batteries. Electricity 2021;2(1):63–76. http://dx.doi.org/10.3390/
ELECTRICITY2010004, URL https://www.mdpi.com/2673-4826/2/1/4.
[184] Ziegler A, Oeser D, Hein T, Ackva A. Development and application of an
active balancing system for lithium-ion cells. In: 2019 IEEE vehicle power and
propulsion conference. IEEE; 2019, p. 1–6.
[185] Hua Y, Zhou S, Cui H, Liu X, Zhang C, Xu X, et al. A comprehensive review
on inconsistency and equalization technology of lithium-ion battery for electric
vehicles. Int J Energy Res 2020;44(14):11059–87. http://dx.doi.org/10.1002/
ER.5683, URL https://onlinelibrary-wiley-com.proxy.lib.umich.edu/doi/full/10.
1002/er.5683.
[186] Einhorn M, Permann R, Kral C, Conte FV, Guertlschmid W, Blochberger T, et
al. Current equalization of serially connected battery cells for a possible second
life application. In: 2011 IEEE vehicle power and propulsion conference. IEEE;
2011, p. 1–5.
[187] Imtiaz AM, Khan FH. ‘‘Time shared flyback converter’’ based regenerative cell
balancing technique for series connected Li-ion battery strings. IEEE Trans
Power Electron 2013;28(12):5960–75. http://dx.doi.org/10.1109/TPEL.2013.
2257861.
22
Renewable and Sustainable Energy Reviews 192 (2024) 114191
[188] Einhorn M, Guertlschmid W, Blochberger T, Kumpusch R, Permann R, Conte FV,
et al. A current equalization method for serially connected battery cells
using a single power converter for each cell. IEEE Trans Veh Technol
2011;60(9):4227–37. http://dx.doi.org/10.1109/TVT.2011.2168988.
[189] Nguyen N-A, La P-H, Choi S-J. High-speed target SOC alignment algorithm
for second-life battery pack maintenance. In: 2022 IEEE transportation elec-
trification conference and expo. IEEE; 2022, p. 131–6. http://dx.doi.org/
10.1109/ITEC53557.2022.9813947, URL https://ieeexplore.ieee.org/document/
9813947/.
[190] La PH, Choi SJ. Novel dynamic resistance equalizer for parallel-connected
battery configurations. Energies 2020;13(13):3315. http://dx.doi.org/10.3390/
EN13133315, URL https://www.mdpi.com/1996-1073/13/13/3315/htm.
[191] Zhang Z, Gui H, Gu DJ, Yang Y, Ren X. A hierarchical active balancing architec-
ture for lithium-ion batteries. IEEE Trans Power Electron 2017;32(4):2757–68.
http://dx.doi.org/10.1109/TPEL.2016.2575844.
[192] Cui X, Ramyar A, Contreras V, Judge G, Siegel J, Stefanopoulou A, et
al. Lite-sparse hierarchical partial power processing for parallel batteries in
heterogeneous energy storage systems. In: 2021 IEEE 22nd workshop on control
and modelling of power electronics. Institute of Electrical and Electronics
Engineers Inc.; 2021, http://dx.doi.org/10.1109/COMPEL52922.2021.9645943.
[193] Ci S, Lin N, Wu D. Reconfigurable battery techniques and systems: A sur-
vey. IEEE Access 2016;4:1175–89. http://dx.doi.org/10.1109/ACCESS.2016.
2545338.
[194] Liu M, Chen Y, Elasser Y, Chen M. Dual frequency hierarchical mod-
ular multilayer battery balancer architecture. IEEE Trans Power Electron
2021;36(3):3099–110. http://dx.doi.org/10.1109/TPEL.2020.3015768.
[195] Oji T, Zhou Y, Ci S, Kang F, Chen X, Liu X. Data-driven methods for battery
SOH estimation: Survey and a critical analysis. IEEE Access 2021;9:126903–16.
http://dx.doi.org/10.1109/ACCESS.2021.3111927.
[196] Song Z, Nazir MS, Cui X, Hiskens IA, Hofmann H. Benefit assessment
of second-life electric vehicle lithium-ion batteries in distributed power
grid applications. J Energy Storage 2022;56:105939. http://dx.doi.org/10.
1016/j.est.2022.105939, URL https://www.sciencedirect.com/science/article/
pii/S2352152X22019272.
[197] Hamidi A, Weber L, Nasiri A. EV charging station integrating renewable energy
and second-life battery. In: Proceedings of 2013 international conference on
renewable energy research and applications. IEEE Computer Society; 2013, p.
1217–21. http://dx.doi.org/10.1109/ICRERA.2013.6749937.
[198] Deng Y, Zhang Y, Luo F, Mu Y. Operational planning of centralized charging
stations utilizing second-life battery energy storage systems. IEEE Trans Sustain
Energy 2021;12(1):387–99. http://dx.doi.org/10.1109/TSTE.2020.3001015.
[199] Fares AM, Klumpner C, Sumner M. A novel modular multiport converter
for enhancing the performance of photovoltaic-battery based power systems.
Appl Sci 2019;9(19):3948. http://dx.doi.org/10.3390/APP9193948, URL https:
//www.mdpi.com/2076-3417/9/19/3948/htm.
[200] Silvestri L, De Santis M, Bella G. Techno-economic evaluation of a second-
life battery energy storage system enabling peak shaving and PV integration
in a ceramic manufacturing plant. In: 2021 IEEE international conference
on industrial engineering and engineering management. Institute of Electrical
and Electronics Engineers Inc.; 2021, p. 1566–70. http://dx.doi.org/10.1109/
IEEM50564.2021.9672803.
[201] Heymans C, Walker SB, Young SB, Fowler M. Economic analysis of second
use electric vehicle batteries for residential energy storage and load-levelling.
Energy Policy 2014;71:22–30. http://dx.doi.org/10.1016/J.ENPOL.2014.04.
016.
[202] D’Arpino M, Cancian M. Design of a grid-friendly DC fast charge station with
second life batteries. Tech. rep., SAE International, Ohio State University; 2019,
http://dx.doi.org/10.4271/2019-01-0867.
[203] Janota L, Králík T, Knápek J. Second life batteries used in energy storage
for frequency containment reserve service. Energies 2020;13(23):6396. http:
//dx.doi.org/10.3390/EN13236396, URL https://www.mdpi.com/1996-1073/
13/23/6396.