Journal of Energy Storage 55 (2022) 105734

Contents lists available at ScienceDirect

Journal of Energy Storage

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

Research papers

Electrochemical discharge of Li-ion batteries - A methodology to evaluate

the potential of discharge electrolytes without corrosion
Hassan Rouhi a, Rodrigo Serna-Guerrero b, Annukka Santasalo-Aarnio a, *
a
Research group of Energy Conversion, Department of Mechanical Engineering, School of Engineering, Aalto University, PO Box 14400, FI-00076, AALTO, Finland
b
Research group of Mineral Processing and Recycling, Department of Chemical Engineering and Metallurgy, School of Chemical Engineering, Aalto University, PO Box
16200, FI-00076, AALTO, Finland

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

Keywords: The demand for Lithium-ion batteries (LIB) is expected to increase exponentially due to the electrification of
Li-ion battery society. Thus, recycling LIBs will be essential to support this activity and ensure the availability of the limited
Electrochemical discharge raw material. End-of-life batteries entering the recycling process may still contain energy, representing a safety
Voltage rebound
risk for processing. The energy can be drained in an electrochemical discharge before mechanical crushing, and it
Recycling
has been proposed that an aqueous NaCl solution would be suitable media. The field has lacked an accurate tool
to evaluate the suitability of an electrolyte for LIB discharge. In this work, we intoduce two methods: external
and internal electrochemical discharge. We also validate the methodology selection with ammonia-based elec­
trolytes and provide a corrosion-free evaluation of the role of NaCl as an electrochemical discharge medium. The
new methodology results confirmed that the batteries could be discharged to low voltage levels in NaCl elec­
trolytes. Still, this electrolyte is highly corrosive, preventing the recovery of active materials in high purity. Thus,
with the new methodology, finding alternative salt solutions that provide an efficient and non-corrosive
discharge environment will be fast.

1. Introduction
To decarbonize our energy systems and limit the global temperature
rise, some authors estimate that net-zero greenhouse gas emissions must
be reached within the next 15 years [1]. Different energy storage sys­
tems play a vital role in balancing the production and consumption of
energy. Specifically, lithium-ion batteries (LIBs) have become a critical
part of stationary energy storage systems and the electrification of
transportation with the rise of electric vehicles (EVs) [2]. As a result of
global awareness, the number of EVs is expected to increase from 75,000
to 24.4 million between 2016 and 2030 globally. Therefore, >1 million
batteries are expected to reach their end-of-life by 2030 [3]. They will
need to be recycled due to the limited sources of valuable natural ma­
terials. Improperly disposing of batteries threatens human health due to
harmful components like organic electrolytes and Co-based powders
[4]. Additionally, recycling some materials like Cu, Al, and Co is
economically feasible, while other materials like Li, Ni, and Mn are
currently lost. On the other hand, recycling LIB materials is vital to avoid
a shortage of critical raw materials such as lithium [5,6].
When batteries enter the recycling facilities, they can still have
energy that causes fire hazards during transport and storage. During the
crushing stage, there is a risk of explosion due to the possibility of a
short-circuit between the cathode and the anode, releasing an enormous
amount of energy in a brief time [7,8]. The risk of explosion is not
limited to batteries in use but is also present for piles of stored batteries
in recycling plants, warehouses, and transportation [9,10]. As a result,
to ensure safe and efficient recycling, the excess energy needs to be
removed from the battery before dismantling. Different techniques have
been proposed to reduce the risk of explosion during the crushing of
LIBs, such as freezing the batteries with liquid nitrogen before crushing
[7] or recycling them in an inert atmosphere [11,12]. Nevertheless,
these methods are expensive and involve further use of materials [10].
Electrochemical discharge of batteries is a method introduced almost
15 years ago [13], and since then, various authors have claimed it to be
efficient [10,14–17]. In electrochemical discharge, the batteries are
typically submerged into an aqueous salt solution that acts as a primitive
resistor or controlled short-circuit to discharge the batteries. When pure
water is used, the water-splitting half-reactions could be responsible for
discharging the batteries [18,19]. Although the required voltage for the
water-splitting reaction is 1.23 V in theory, in practice, values of up to

* Corresponding author.
E-mail address: annukka.santasalo@aalto.fi (A. Santasalo-Aarnio).

https://doi.org/10.1016/j.est.2022.105734
Received 9 May 2022; Received in revised form 1 August 2022; Accepted 20 September 2022
Available online 28 September 2022
2352-152X/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
H. Rouhi et al.
1.68 V have been reported [20]. Therefore, we would need a reaction
with a lower theoretical voltage than water splitting to achieve deep
discharge of LIBs. One option could be an ammonia-based solution, as
splitting the ammonia molecule has a lower onset potential than the
electrolysis of water [18]. Additionally, the LIB casing corrosion in
ammonium carbonate and ammonium hydroxide solutions was not re­
ported to be severe, at least within 24 h [19].
Furthermore, a phenomenon called “voltage rebound” or “voltage
relaxation” could impact the safety of the battery recycling process.
Voltage rebound refers to a process where the battery voltage increases
toward its steady-state value when there is no current pulse [21]. During
the discharge of a LIB, the internal state of the battery is non-linear with
heterogeneities in the concentration of the Li-ions in both electrodes and
the electrolyte. When battery discharge is terminated, the current in the
circuit is switched off, and the Li-ions move from an area of higher
concentration to a lower concentration area. The underlining phenom­
ena behind the rebound effect are not fully understood, but previous
studies reported factors like temperature, state-of-charge (SOC),
discharge speed, and battery age [22–24]. The voltage rebound phe­
nomenon has been reported to last up to 24 h [25]. As we showed in our
previous work [26], when the battery is electrochemically discharged in
aqueous carbonate electrolytes (K2CO3 and Na2CO3) and removed from
the solution, it will rebound to close to 2.5 V. Therefore; it is of interest
to study the effect of this phenomenon with other aqueous electrolytes
and additionally to investigate what is the impact of the rate of discharge
on the rebound voltage. Another critical concept in battery recycling is
the state of charge (SOC) which estimates how much energy is available
in the battery. Technically, the SOC is the ratio of the remaining capacity
to the battery's maximum capacity [27,28]. The voltage rebound could
significantly impact the SOC estimation, and ignoring it leads to an error
in the SOC estimation circuit. As a result, it can cause harm and addi­
tional safety risks during recycling.
The previous efforts from our group presented the first experimental
evidence that the batteries could efficiently be discharged in salt elec­
trolytes, although strong corrosion occurs when NaCl is used [29]. After
that, there was a systematic study with 26 inorganic salts by Shaw-
Steward et al. [30], where the batteries were submerged in various
aqueous salt solutions for 24 h. They also reported the minimum voltage
obtained after 24 h of discharge in each electrolyte, but they did not
monitor the voltage changes after removal from the electrolyte. On the
other hand, Nembhard continued monitoring the voltage of the batteries
after removal from the electrolyte and found an increase in the battery
voltage value up to 2.5 V corroborating the aforementioned rebound
effect in LIBs [31].
Recently, two groups explored different aqueous solutions as dis­
charging media [32,33]. They investigated NaCl, FeSO4, MnSO4,
NaNO3, KCl, and MgSO4 as potential discharge media. The results
confirmed that chlorine-based salts (both NaCl and KCl) cause severe
corrosion of the battery casing, leading to electrolyte leakage into the
discharge solution and potentially causing harm to the valuable active
materials [32]. Regardless of these severe shortcomings, Yao et al.
proposed this solution to be used in recycling processes and claim that
discharge in sulfate solutions is more environmentally friendly due to
lower toxic gas emissions, even though severe corrosion was also
observed [30,32].
Yao et al. and Xiao et al. [32,33] also state that the voltage clearly
drops below 0.5 V residual voltage in the aqueous salt solutions. Yao
et al. also measured the solution's voltage with a BST-3008-5V50mA
battery testing system [32]. However, they do not describe how the
voltage was measured through poles that have experienced the forma­
tion of metal oxide layers and passivation.
Due to the inconsistencies of discharging methods encountered in the
current literature, this work presents a systematic analysis of different
voltage measurement methods and proposes an alternative methodol­
ogy to obtain accurate and reliable results. We provide two experimental
methods: an internal method where the batteries are submerged in the
2
Journal of Energy Storage 55 (2022) 105734
discharged electrolyte and properly describe how their voltage can be
monitored accurately after removal from the solution. Secondly, we
demonstrate an external method where the potential corrosiveness of
discharge electrolytes can be ruled out to properly evaluate the
discharge capabilities of the studied electrolyte. Additionally, we
experimentally prove that LIBs cannot be fully discharged to 0 V in NaCl
electrolytes, even though this has been claimed in previous reports
[29,33]. Nevertheless, NaCl electrolytes can discharge LIBs under the 2
V voltage level but with the severe corrosion of the casing materials.
Thus, a less corrosive discharge medium must be found.
2. Materials and methods
2.1. Equipment
The LIBs used for this work were Panasonic Cameron Sino CS-
NCR18650B (China, capacity 3250 mAh). The cells were cylindrical
with LiNiCoAlO2 as the positive electrode material. To ensure the reli­
ability of the results, a new battery was used for each experiment.
Table 1 summarises the properties of the battery.
To measure the voltage of the battery during electrochemical
discharge of the batteries, two different devices were used: a digital
voltmeter (Fluke 87 V TRMS Industrial Multimeter) and an IviumStat
potentiostat (Teamator, Sweden).
For the discharge experiments, aqueous salt solutions at various
concentrations were used to study the effect of the electrolyte conduc­
tivity on the reaction rate with the aim of obtaining the lowest possible
discharge voltage. The salts used were NaCl (VWR International, quality
min. 98 %) and (NH4)2CO3 (Sigma-Aldrich). For NaCl, three different
concentrations, i.e., 5, 10, and 20 wt%, and the solutions were prepared
with de-ionized water (<0.5 μS− 2). For (NH4)2CO3, 0.5 M (corre­
sponding 5 wt% utilized by [30] for accurate comparison) and most
concentrated saturated solutions were prepared to study the effect of
concentration. In the experiment, ammonia hydroxide (NH4OH) was
ready-made 5 M solutions (Fluka).
2.2. Methodology to measure the battery voltage during electrochemical
discharge
There are several possibilities to measure the voltage of the battery
during the battery discharge in the electrolyte. As that occurs in the
aqueous phase, the direct connection of the measurement device to the
battery poles is not possible. In Fig. 1, we present a few options: most of
the experiments published by other authors [30,32,33] were conducted
with the internal electrochemical discharge method – whereby the batteries
are directly submerged into the electrolyte (Fig. 1A) – following the
expected industrial practice. We also used this method in our previous
study [26] to explore the discharge effect of Na2CO3 and K2CO3 elec­
trolytes. In this procedure, the batteries were placed into an electrolyte
to perform the electrochemical discharge. For measuring the voltage of
the batteries during internal electrochemical discharge, the batteries
were lifted from the electrolyte, the poles were rinsed with de-ionized
water, dried, and measured with a digital voltmeter at specific time
intervals.
However, the corrosion of the poles might interfere with this
method. Therefore, to evaluate the fundamental discharge behavior
without the influence of corrosion, we also applied the external
Table 1
Properties of the Panasonic battery used for this study.
Nominal voltage 3.7 V
Max voltage 4.2 V
Initial voltage measured 3.7 V
Capacity 3250 mAh
Cathode material LiNiCoAlO2
H. Rouhi et al.
Fig. 1. (A) Internal electrochemical discharge of the LIB (B) External electrochemical discharge of the LIB in an aqueous electrolyte.
electrochemical discharge (Fig. 1B) method. In this method, the battery is
placed above the electrolyte, and Pt wires with a diameter of 0.5 mm,
and a length of 50 cm are connected to the poles, and the opposite ends
are submerged into the electrolyte. Pt wires were selected as they do not
corrode in aqueous salt electrolytes. Finally, the poles of the batteries
were connected to the measurement device (digital multimeter or
potentiostat), which enabled us to monitor the voltage of the batteries
continuously.
As there were recent publications that claimed LIB discharge to reach
low voltage levels using NaCl electrolyte with the internal method [32],
despite having a significant corrosion effect on the battery poles, such
experiment was replicated with a slightly modified internal electro­
chemical discharge method presented in Fig. 2 to follow the set-up of the
original publication as closely as possible. To measure the voltage of the
LIBs during the experiment, the battery poles were connected to the Pt-
wires, and the opposite end of the Pt-wires connected to the digital
voltmeter using crocodile-type clips (Fig. 2), and the voltage between
the poles of the batteries could be read easily.
All the measurements conducted in this work were completed in a
fume hood due to the potential evolution of gases like H2, O2, NH3, or
Cl2. Due to the length of some experiments, water evaporation happened
occasionally. As a result, the volume of the electrolyte solution was
constantly monitored, and additional water was added as needed.
3. Results and discussion
To increase the material recovery from LIBs, they should be
Fig. 2. Modified internal method to evaluate battery voltage during the elec­
trochemical discharge.
3
Journal of Energy Storage 55 (2022) 105734
discharged prior to mechanical recycling. One of the most proposed
methods is to utilize a conductive liquid medium to discharge batteries
of different sizes and shapes efficiently. To better compare the potential
of different electrolytes to discharge the batteries, it would be essential
to unify the methodology and ascertain that the voltage reading repre­
sents the actual energy remaining in the battery to ensure the safety of
the processes. In chapter 3.1, we present an accurate and valid meth­
odology and validate that with ammonia-based electrolytes. In chapter
3.2, we focus the study on the most often proposed electrolyte, NaCl, and
show the effect of electrolyte corrosion on the results obtained.
3.1. Methodology validation to accurately observe battery voltage in the
electrochemical discharge process
In the previous years, several research groups have submerged the
batteries into inorganic aqueous salt solutions to discharge the batteries
[10,11,22,27,28], a procedure referred to in this work as “internal
electrochemical discharge”. While this would be the method used in
actual recycling processes, the main problem is the likely corrosion of
the battery casing, which might release toxic electrolytes and damage
valuable metal parts in the batteries. Additionally, the battery poles can
corrode, and this has an impact on the volatage reading accuracy. In
order to quantify the discharge efficiency of a particular electrolyte, we
presented a new external electrochemical discharge method in our
previous work [29], wherein the battery is placed outside the electrolyte
and only connected to the electrolyte with Pt-wires to exclude the in­
fluence of corrosion. In previous studies, a variety of different in­
struments have been utilized to measure the voltage. Still, it is crucial to
identify the advantages and disadvantages of different equipment,
which is particularly important in the case of external discharge. In this
work, we compare two devices for the external voltage measurement: a
potentiostat that we used in one of our previous papers [29] and a digital
voltmeter.
To show the impact of the different voltage measurement techniques,
ammonia-based aqueous salt solutions were selected. These species have
not been studied broadly in the literature, but as they are alkaline, they
might exhibit a comparatively less corrosive nature. We tested two
different ammonia-based discharge media that have shown some of the
lowest corrosion rates on LIB casings [30]: 5 M ammonium hydroxide
(NH4OH) and ammonium carbonate ((NH4)2CO3). For the latter, two
concentrations (0.5 M and saturated solutions) were used to observe the
effect of the ionic strength of the solution. The results from the tradi­
tional internal electrochemical discharge experiments in these electro­
lytes are presented in Fig. 3.
H. Rouhi et al.
Fig. 3. Internal electrochemical discharge of LIBs with ammonia-based aqueous electrolytes.
As can be seen in Fig. 3, during the first 8 h of the measurement, the
battery discharged slowly in the lower concentration electrolytes, and
only a tiny decrease in voltage was observed from the starting voltage of
3.74 V. The discharge rate of the battery increased after around 30 h
when more bubbles from the electrodes were observed. The overall
discharge time was about 20–60 h, depending on the electrolyte type.
However, we continued the experiments for 100 h to reach a steady-
state value of around 1.7–2.0 V in the different electrolytes. While dis­
charging the battery with a more concentrated ammonium carbonate
solution led to a comparatively faster discharging rate in the beginning,
the final voltage value was almost the same in all cases.
2 V limit was consistently reached after roughly 60 h with all the salts
studied. Shaw-Steward et al. discharged batteries for 24 h in 5 wt%
(NH4)2CO3 (~0.5 M) solution [30] and reported a voltage of 2.28 V, a
value almost 0.5 V lower compared to the result in Fig. 3. The main
likely reason is that they used a different battery type that can have
different pole surface areas. Overall, neither of the two studies reported
voltage values below 2 V.
To ensure that corrosion of the battery poles did not affect the
reading of the battery voltage, the voltage behavior during external
discharge also was tested. The 5 wt% (NH4)2CO3 electrolyte was
selected as the medium for technology validation. The voltage mea­
surements with three different set-ups are presented in Fig. 4.
As seen in Fig. 4, the results vary significantly depending on the
different experimental methods; therefore, there might be different
Fig. 4. A comparison of internal and external discharge of LIBs with 0.5 M (NH4)2CO3 discharge electrolytes with two separate voltage reading devices.
4
Journal of Energy Storage 55 (2022) 105734
physical phenomena involved. As we know, the internal electrochemical
discharge method is the closest to the expected industrial discharge
practice, and the results indicate that the voltage does not drop below
1.7 V with an of the ammonia-based electrolytes (Fig. 3). For this reason,
it was unexpected that external electrochemical discharge with the
potentiostat (Fig. 4 - green line) indicated that the voltage decreases
down to near 0 V in around 22 h. On the other hand, external discharge
with a voltameter provided an end voltage of 2.1 V (Fig. 4 – black line),
closer to the internal discharge results. Thus, the discharge rate during
external discharge with the voltameter was clearly slower than the rate
observed during internal discharge. This can also be related to the sur­
face area of the Pt-wires.
It was further unclear why such a large discrepancy in behavior was
observed between the two external experimental set-ups, as no evidence
of corrosion was seen during the internal electrochemical discharge
experiments. Therefore, we conducted an experiment where the battery
was only connected to the potentiostat and the wires were left in the air
(not in connection with the electrolyte), and the voltage of the battery
was monitored simultaneously with the potentiostat and the digital
voltmeter. The battery discharged immediately after connecting the
potentiostat to 0 V after around 22 h. Based on these results, it was
evident that the potentiostat was not monitoring pure electrochemical
discharge but a combination of electrochemical discharge and electrical
discharge with the measurement set-up.
When the battery is discharged, a small (but not negligible) current
H. Rouhi et al.
flows into the circuit. Since there is only one closed path in the set-up,
the current leaks into the measurement device (potentiostat) due to its
lower input impedance than a digital voltmeter (typically in the range of
10 M Ω). Based on Ohm's law, a small current value will flow into the
measurement device when the input impedance value is not large
enough. The value of the current flowing into the measurement device
depends on the value of the voltage of the LIB and the input impedance
of the measurement device. Because of this flowing current, a voltage
drop will occur inside the potentiostat and across the voltage probes.
Consequently, in the external electrochemical discharge with a poten­
tiostat, the battery will work as a power supply until it discharges to
almost 0 V over time, and the electrochemical reaction in the electrolyte
does not entirely consume the power, but a combination of electrical and
electrochemical phenomena is responsible for the full discharge of the
batteries. In a real end-of-life battery discharge process, connecting
wires to individual batteries would not be technically feasible, but in the
case of large battery packs (such as the ones used in EVs), this is most
likely the best strategy for efficient discharge.
To avoid interference from the measurement equipment, we propose
using a digital voltmeter to measure the LIB voltage during external
discharge due to its almost infinite input impedance. This analysis
confirmed that the reason for the faster and deeper discharge of the
battery in the external discharge with a potentiostat (Fig. 4 – green line)
was that the discharge in this case is a combination of both electrical and
electrochemical discharge. Therefore, this data cannot provide a proper
view of the electrolyte discharge efficiency. When comparing the in­
ternal and external electrochemical discharge with the voltameter in
Fig. 4, we can see that the final voltages are close to each other, but the
internal discharge is more rapid, which is most likely due to the lower
contact surface between the battery and discharge solution during
external electrochemical discharge. Consequently, the external electro­
chemical discharge method can provide an idea of the final voltage level
of the LIB after the electrochemical discharge but might not provide
useful information in the context of industrial electrochemical discharge
about the discharge rate and can therefore not be used solely as an
experimental tool. Thus, this experimental procedure possesses a
significantly lower risk and is therefore recommended to rule out elec­
trolytes that do not provide the required voltage level for discharge.
In our first research paper on LIB discharge, we presented data in
NaCl electrolytes [29], but at that time, we were not aware of the effect
of the measurement device on batteries' discharge. In that work, we first
tested the internal discharge, but as the corrosion rate was very signif­
icant, we also tried external electrochemical discharge for LIBs in 5 wt%,
10 wt%, and 20 wt% of NaCl electrolytes. The results seemingly showed
that the batteries would be discharged to 0 V, but in the light of the
methodology information, we now know that the batteries discharged in
a mixed electrical and electrochemical discharge process. These findings
highlight the importance of an accurate battery voltage measurement
method during electrochemical discharge and prove that some mea­
surement methods do not only provide information about pure elec­
trochemical discharge. With the new experimental methodology, we
will present NaCl electrolyte's pure electrochemical discharge effect
compared to a combined method of electrical and electrochemical
discharge presented in our previous publication [29]. For the other
experiment in that previous work, the voltage drop during the experi­
ments was not that significant, as a solid product was formed in the
electrolytes that contained metal particles. Thus, it is also advised to
keep in mind that these experiments reflect the join electrical and
electrochemical discharge results.
3.2. Electrochemical discharge of LIBs in NaCl electrolyte
Aqueous NaCl electrolyte is one of the most common media proposed
for the electrochemical discharge of LIBs before recycling. Our previous
work [29] also stated that this could be an efficient medium to discharge
the batteries fully. Since we have shown above that this external
5
Journal of Energy Storage 55 (2022) 105734
measurement method with a potentiostat does not reveal the pure
electrochemical discharge capacity of the electrolyte (Fig. 4), we will
reproduce these experiments using a method providing information on
the pure electrochemical discharge behavior.
As it was previously reported that NaCl electrolytes would cause
severe corrosion to battery casing and pole materials [32,33], we carried
out external electrochemical discharge experiments to study the actual
behavior of these batteries without a corrosive environment. Further­
more, the battery's voltage was measured directly from the poles using a
digital multimeter (Fig. 1B). To better compare the previous results
[29,32], we used the same concentrations (5 wt%, 10 wt%, and 20 wt%)
of NaCl electrolytes to observe if electrolyte conductivity will have a role
in the discharge process. The results for external discharges are shown in
Fig. 5.
During the experiments, bubbles were generated on both electrodes,
and deposition of solids was not detected. As the onset voltage of H2 and
Cl2 evolution is quite similar, neither of these reactions can be ruled out.
As seen in Fig. 5, the discharge rate is higher at the early stage of the
process, but the discharge rate decelerates, eventually reaching a con­
stant voltage. As expected, at a higher concentration of the NaCl elec­
trolyte (10 wt% and 20 wt%), the electrolyte conductivity increases, and
the discharge rate of the battery is more efficient compared to the lower
concentration of NaCl (5 wt%). Overall, when the concentration of a
solution increases, the ion conductivity of that solution will also in­
crease. However, when the concentration of NaCl reaches a certain level,
the interaction between negative and positive ions will increase. As a
result, the conductivity does not increase with concentration anymore
[34]. That is why there is no significant difference between the steady-
state voltage of the batteries during the discharge with 10 wt% and 20
wt% of NaCl electrolyte. Therefore, increasing the concentration will
only result in a higher cost of chemicals without a significant advantage
on performance.
Moreover, the steady-state voltage of the battery in 5 wt% of NaCl
solution is around 1.7 V, and for the higher concentrations, the steady-
state voltage value is around 1 V (Fig. 5). Thus, all of them are below 1.7
V, and because the water-splitting reaction has a high overpotential and
most likely cannot discharge batteries below 1.7 V, some additional
reaction is probably taking place. The voltage readings of the 10 and 20
wt% were 0.978 V and 1.084 V, respectively. Considering the species
involved, it is likely that Cl2 gas formation occurs at the positive pole of
the battery, which enables the voltage to reach lower values than with
many other reported electrolytes [30]. This makes the NaCl electrolyte
an interesting option for electrochemical discharge. As it would be
important to find a discharge medium to withdraw the energy from the
battery, we also continued monitoring the batteries' voltage after the
discharge process. Fig. 6 presents the rebound voltage after removal
from each discharge media.
As seen in Fig. 6, the voltage of the batteries increases rapidly in the
first hours after removal from the electrolyte, but it diffuses to its steady-
state value and reaches its final value after one week. The same phe­
nomenon occurred in our previous studies on electrochemical discharge
in carbonate electrolytes [26]. In the 5 wt% solutions, the voltage of the
battery rose around 0.45 V, reaching 1.88 V. In contrast, in the 20 wt%
NaCl electrolytes, the rebound voltage increased by >0.65 V. However,
the final voltage is only 1.63 V. Thus, the experiments prove the hy­
pothesis that faster discharge leads to a higher rebound voltage.
Overall, with higher NaCl concentrations, it was possible to keep the
final steady-state voltage below 2 V, indicating that it would be safe to
shred the batteries in a mechanical recycling process. However, in a
NaCl medium, corrosion is still a significant challenge that affects the
recovery of valuable metals from batteries and causes emissions from
the recycling process by releasing toxins.
Other studies on discharging the batteries with a NaCl electrolyte
claimed that the LIBs could be discharged to voltages lower than 1 V
[32,33]. However, they also reported severe metal shell corrosion and
electrolyte leakage [30]. Therefore, questions remain on the
H. Rouhi et al. Journal of Energy Storage 55 (2022) 105734

Fig. 5. External electrochemical discharge of LIBs with a voltameter as a voltage measurement device in three different NaCl aqueous electrolytes concentrations.

Fig. 6. Voltage rebound effect after discharging the battery with three different concentrations of NaCl electrolyte.

measurement protocol used in those studies and what is the reliability of
such measurements. To further clarify this, we decided to repeat the
internal electrochemical discharge with NaCl solution to understand the
voltage reading procedure better. For an accurate comparison, we
selected the same 7 wt% concentration of NaCl electrolyte used in ref.
[32]. To read the voltage of the battery continuously while it discharges
internally, the poles of the battery were connected to the Pt-wires (as
shown in Fig. 2). Then, the battery was submerged into the solution, and
the other side of the Pt-wires was connected to a multimeter with
crocodile clips. Fig. 7 displays the battery's voltage during the modified

Fig. 7. Modified internal electrochemical discharge of LIB with 7 wt% NaCl aqueous electrolyte.

6
H. Rouhi et al.
internal discharge with a 7 wt% NaCl aqueous electrolyte. In the figure,
the blue line is the voltage of the battery measured by the multimeter
when the battery was submerged into the electrolyte, the red line is the
voltage of the battery immediately after removing it from the solution,
and finally, the red dashed line is presented as a guide to observing what
most likely has been the accurate voltage of the battery.
Fig. 7 shows that the voltage of the battery started to decrease from
its initial value (3.7 V) and reached 1 V after around 80 min. Yao et al.
also reported the discharge voltage slightly above 1 V for the 1.2 M (7 wt
%) NaCl electrolyte in the same time interval [32]. As shown in Fig. 7,
after around 20 min, the battery discharged abruptly down to 1.86 V
from 3.64 V. Yao et al. also reported the quick voltage discharge of the
battery at the initial phase of the experiment [32]. Thenceforward, the
voltage of the battery fluctuated until the end of the experiments
because of the severe corrosion on the battery poles, leading to loss of
the connection between the poles and the multimeter. At the last minute
of the internal electrochemical discharge experiment, the multimeter
showed 0.61 V as the final voltage of the battery, which is similar to the
previous study on the internal discharge in a NaCl solution [32].
Fig. 8 displays the corrosion observed around 3 h into the experiment
(Fig. 8A) and when the battery was lifted from the electrolyte (Fig. 8B).
These Figures confirm corrosion on the battery connector poles and
battery electrolyte leakage into the aqueous solution. Shaw-Steward
et al. reported the same phenomenon after around 5 h internal
discharge with a NaCl electrolyte when they used cylindrical LIB cells
[30]. The high speed of bubbles forming at the battery's negative elec­
trode shows that some gas evolution reaction was ongoing (even though
these many bubbles are not normally observed below the potential of
2.5 V). Fig. 8B shows the positive electrode of the battery that was
completely damaged after the experiment, showing the NaCl electro­
lyte's strong corrosivity. The battery's voltage was measured immedi­
ately after lifting the battery with the multimeter and finding a proper
metallic contact, and the reading was 3.5 V (Fig. 7). That is a very similar
value to the one that was measured at 20 min before the significant drop
in the voltage (Fig. 7). Therefore, it is most likely that after around 20
min, the battery casing was compromised. Most likely, some of the
electrolytes leaked into the discharge medium, and the discharge pro­
cess almost stopped. Also, due to the completely damaged positive pole,
the connection of the measurement circuit was most likely compromised
after 20 min, and no conclusions can be drawn about the battery voltage
after that time.
Shaw-Steward et al. discharged LIB in 5 wt% of NaCl electrolyte
[30], and they also reported that it was impossible to determine the
battery's final voltage due to strong corrosion at the positive terminal of
Fig. 8. LIB discharge in NaCl electrolyte (A) measurement and corrosion after 3 h of experiment (B) Corrosion at the positive electrode of the battery at the end of
the experiment.
7
Journal of Energy Storage 55 (2022) 105734
the battery, which is in line with the result of this study. Lu et al. [19]
claimed that the battery discharged dramatically in 10 wt% of NaCl
solution after only 7 min. They also claimed that the solution penetrated
the casing, making batteries discharge faster. Although the method of
voltage measurement in such work was not mentioned, based on the
observations of the present study, it is reasonable to associate it with
severe corrosion of the battery casing. Our finding supports the con­
clusions of Shaw-Steward et al. [30] ‘s conclusion that reading the
battery's voltage in the NaCl electrolyte should be performed very
carefully as the value might not represent the actual voltage.
In summary, while the external electrochemical discharge of the
batteries with 10 wt% NaCl solution seems a reasonable option for
battery discharging, internal discharge can lead to a wrong interpreta­
tion of the voltage reading and potentially compromise the valuable
metallic, active materials in the cells. The accurate method presented in
this work will enable the exploration of different electrolytes for effi­
cient electrochemical discharge: it is recommended to use external
discharge and only internal discharge for the most interesting candi­
dates to reveal the rate of discharge and the effect of corrosion on the
discharge process.
4. Conclusion
Due to the limited resources of the main battery materials, end-of-life
batteries must be recycled to recover their raw materials. Out of the
different LIB discharge methods, electrochemical discharge is widely
accepted among scientists as a robust method capable of the large-scale
discharge of small batteries. Accuracy of the voltage reading is critical,
as it can affect the safety of the crushing process. In this work, we proved
that the equipment used for voltage measurement could affect the
discharge process of the battery: it has been shown that using the
potentiostat could lead to a combination of the electrical and electro­
chemical discharge in the batteries. We also proposed a measurement
that solely describes the electrochemical discharge properties of the
electrolyte.
To circumvent the effect of corrosion, we propose a more accurate
external electrochemical discharge method alongside the internal elec­
trochemical discharge method. The external electrochemical discharge
method allows us to study the discharge potential of an electrolyte even
if the medium is corrosive, which allows us to gather an initial under­
standing of the final voltage the battery can reach in the given discharge
medium. Internal discharge experiments should also be conducted for
the most interesting candidates to obtain further information about the
discharge time that is difficult to estimate via the external discharge
H. Rouhi et al.
method. Additionally, it is vital to use a multimeter as an instrument for
voltage readings due to its high inner impedance.
Finally, the study investigated the possibility of using ammonia and
NaCl electrolytes as LIB discharge media. The advantage of the
ammonia-based solutions is their low corrosivity, but we found out that
these solutions cannot discharge batteries low enough for recycling
purposes after the voltage rebound is considered. The discharge in NaCl
was efficient, and the voltage remained below 2 V even after voltage
rebound, but NaCl electrolyte is very corrosive to the LIB casing mate­
rial, and it can cause severe damage to the recoverable metals. Also, due
to the strong corrosivity of NaCl, the measurements during the batteries'
internal discharge are unreliable. However, this study demonstrates that
it is possible to discover aqueous salt electrolytes to discharge batteries
below 2 V even after the voltage rebound. This would be of high
importance to ensure sustainable and efficient recycling processes.
CRediT authorship contribution statement
Hassan Rouhi: Conceptualization, Data curation, Formal analysis,
Visualization, Writing – original draft. Rodrigo Serna-Guerrero:
Conceptualization, Investigation, Funding acquisition, Writing – review
& editing. Annukka Santasalo-Aarnio: Conceptualization, Investiga­
tion, Funding acquisition, Visualization, Writing – original draft.
Declaration of competing interest
The authors declare the following financial interests/personal re­
lationships which may be considered as potential competing interests:
Annukka Santasalo-Aarnio reports financial support was provided by
Academy of Finland.
Data availability
Data will be made available on request.
Acknowledgments
The author would like to acknowledge the work of Aurelia Aminoff
for conducting ammonia-based experiments with a potentiostat and
Arpad Toldy for language editing. The work was funded by the Academy
of Finland with two separated project: Hyper-sphere (341629) and
Profiling funding 5 for Energy Storage (326346).
References
[1] S. Solomon, Intergovernmental panel on climate change, and intergovernmental
panel on climate change, in: Climate Change 2007: The Physical Science Basis:
Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University Press,
Cambridge; New York, 2007.
[2] B. Scrosati, J. Garche, Lithium batteries: status, prospects and future, J. Power
Sources 195 (9) (2010) 2419–2430, https://doi.org/10.1016/j.
jpowsour.2009.11.048.
[3] E. Drabik, V. Rizos, Prospects for electric vehicle batteries in a circular economy,
2018. CEPS Research Report No 2018/05, July.
[4] X. Zheng, Z. Zhu, X. Lin, Y. Zhang, Y. He, H. Cao, Z. Sun, A mini-review on metal
recycling from spent lithium-ion batteries, Engineering 4 (2018) 361–370, https://
doi.org/10.1016/j.eng.2018.05.018.
[5] A. Boyden V.K. Soo M. Doolan 'The environmental impacts of recycling portable
lithium-ion batteries'. Procedia CIRP. VOL. 48.pp. 188-193, 016.
[6] J. Heelan, E. Gratz, Z. Zheng, Q. Wang, M. Chen, D. Apelian, Y. Wang, Current and
prospective Li-ion battery recycling and recovery processes, JOM 68 (2016)
2632–2638, https://doi.org/10.1007/s11837-016-1994-y.
[7] L. Li, J. Ge, F. Wu, R. Chen, S. Chen, B. Wu, Recovery of cobalt and lithium from
spent lithium-ion batteries using organic citric acid as leachant, J. Hazard. Mater.
176 (2010) 288–293, https://doi.org/10.1016/j.jhazmat.2009.11.026.
[8] R. Clancy, Tesla shares fall after car fire and rating downgrade. https://www.te
legraph.co.uk/finance/newsbysector/industry/10352809/Tesla-shares-fall-after
-car-fire-and-rating-downgrade.html, 2018. (Accessed 3 February 2018).
[9] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun, C. Chen, Thermal runaway caused fire
and explosion of lithium-ion battery, J. Power Sources 208 (2012) 210–224,
https://doi.org/10.1016/j.jpowsour.2012.02.038.
Journal of Energy Storage 55 (2022) 105734
[10] A. Sonoc, J. Jeswiet, V.K. Soo, Opportunities to improve recycling of automotive
lithium-ion batteries, Procedia CIRP 29 (2015) 752–757, https://doi.org/10.1016/
j.procir.2015.02.039.
[11] M.J. Lain, Recycling of lithium-ion cells and batteries, J. Power Sources 97–98
(2001) 736–738, https://doi.org/10.1016/S0378-7753(01)00600-0.
[12] US 20070196725 A1, in: F. Tedjar, J.-C. Foudraz (Eds.), Method for the Mixed
Recycling of Lithium-based Anode Batteries and Cells. Recupyl. France, 2007,
p. 13. US 10/593,332, 4.4.2005. 23.8.2007.
[13] D.C.R. Espinosa, A.M. Bernardes, J.A.S. Tenório, An overview on the current
processes for the recycling of batteries, J. Power Sources 135 (2004) 311–319,
https://doi.org/10.1016/j.jpowsour.2004.03.083.
[14] B. Scorsati, J. Garche, Lithium batteries: status, prospects and future, J. Power
Sources 195 (2010) 2419–2430, https://doi.org/10.1016/j.jpowsour.2009.11.048.
[15] S. Krüger, C. Hanisch, A. Kwade, M. Winter, S. Nowak, Effect of impurities caused
by a recycling process on the electrochemical performance of Li
[Ni0.33Co0.33Mn0.33]O2, J. Electroanal. Chem. 726 (2014) 91–96, https://doi.
org/10.1016/j.jelechem.2014.05.017.
[16] X. Zhang, Y. Xie, X. Lin, H. Li, H. Cao, An overview on the processes and
technologies for recycling cathodic active materials from spent lithium-ion
batteries, J.Mater.Cycles Waste Manag. 15 (2013) 420–430, https://doi.org/
10.1007/s10163-013-0140-y.
[17] T. Zang, Y. He, L. Ge, R. Fu, X. Zhang, Y. Huang, Characteristics of wet and dry
crushing methods in the recycling process of spent lithium-ion batteries, J. Power
Sources 240 (2013) 766–771, https://doi.org/10.1016/j.jpowsour.2013.05.009.
[18] B.K. Boggs, G.G. Botte, On-board hydrogen storage and production: an application
of ammonia electrolysis, J. Power Sources 192 (2009) 573–581, https://doi.org/
10.1016/j.jpowsour.2009.03.018.
[19] M. Lu, H. Zhang, B. Wang, X. Zheng, C. Dai, The re-synthesis of LiCoO2from spent
lithium-ion batteries separated by vacuum-assisted heat-treating method, Int. J.
Electrochem. Sci. 8 (2013) 8201–8209.
[20] P. Millet, R. Ngameni, S.A. Grigoriev, N. Mbemba, F. Brisset, A. Ranjbari,
C. Etiévant, PEM water electrolysers: from electrocatalysis to stack development,
Int. J. Hydrog. Energy 35 (2010) 5043–5052, https://doi.org/10.1016/j.
ijhydene.2009.09.015.
[21] A. Li, S. Pelissier, P. Venet, P. Gyan, Fast characterization method for modeling
battery relaxation voltage, Batteries 2 (2) (2016), https://doi.org/10.3390/
batteries2020007.
[22] K. Qian, B. Huang, A. Ran, Y. He, B. Li, F. Kang, State-of-health (SOH) evaluation
on lithium-ion battery by simulating the voltage relaxation curves, Electrochim.
Acta 303 (2019) 183–191. https://doi-org.libproxy.aalto.fi/10.1016/j.electacta.20
19.02.055.
[23] D. Hauck, M. Kurrat, Overdischarging lithium-ion batteries, in: A. Kwade,
J. Diekmann (Eds.), Recycling of Lithium-Ion Batteries the LithoRec Way 2018,
Springer, Cham, Switzerland, 2018, ISBN 978-3-319-70572- 9, pp. 53–81, https://
doi.org/10.1007/978-3-319-70572-9 (electronic).
[24] L. Canals Casals A.M. Schiffer Gonzalez B. Garcia J. Llorca PHEV battery aging
study using voltage recovery and internal resistance from onboard data, IEEE
Trans.Veh.Technol. Vol. 65, No. 6, pp. 4209-4216. ISSN 0018-9545. DOI: 10.1109/
TVT.2015.2459760.
[25] M.A. Roscher, D.U. Sauer, Dynamic electric behavior and open circuit voltage
modelling of lifepo4-based lithium-ion secondary batteries, J. Power Sources 196
(2011) 331–336.
[26] H. Rouhi, E. Karola, A. Santasalo-Aarnio, R. Serna-Guerrero, Voltage behaviour in
lithium-ion batteries after electrochemical discharge and its implications on the
safety recycling process, J. Energy Storage 35 (2021), 102323, https://doi.org/
10.1016/j.est.2021.102323.
[27] K. Ogura, M.L. Kolhe, 4 - battery technologies for electric vehicles, in:
Tariq Muneer, Mohan Lal Kolhe, Aisling Doyle (Eds.), Electric Vehicles: Prospects
and Challenges, Elsevier, 2017, ISBN 9780128030219, pp. 139–167, https://doi.
org/10.1016/B978-0-12-803021-9.00004-5.
[28] H. Mu, R. Xiong, Chapter 1 - modelling, evaluation, and state estimation for
batteries, in: Hui Zhang, Dongpu Cao, Haiping Du (Eds.), Modelling, Dynamics and
Control of Electrified Vehicles 2018, Woodhead Publishing, 2018, ISBN
9780128127865, pp. 1–38, https://doi.org/10.1016/B978-0-12-812786-5.00001-
X.
[29] S. Ojanen, M. Lundström, A. Santasalo-Aarnio, R. Serna-Guerrero, Challenging the
concept of electrochemical discharge using salt solution for lithium-ion batteries
recycling, J.Waste Manag. 76 (2018) 242–249, https://doi.org/10.1016/j.
wasman.2018.03.045.
[30] J. Shaw-Stewart, A. Alvarez-Reguera, A. Greszta, M. Masood, R. Sommerville,
E. Kendrick, Aqueous solution discharge of cylindrical lithium-ion cells, J.Sustain.
Mater.Technol. 22 (2019), e00110, https://doi.org/10.1016/j.susmat.2019.
e00110.
[31] N. Nembhard, Safe, Sustainable Discharge of Electric Vehicle Batteries as a Pre-
treatment Step to Crushing in the Recycling Process, Master's thesis, in: KTH
Industrial Engineering and Management, 2019. Stockholm: 66 p.
[32] L.P. Yao, Q. Zeng, T. Qi, J. Li, An environmentally friendly discharge technology to
pretreat spent lithium-ion batteries, J. Clean. Prod. 245 (2020), 118820, https://
doi.org/10.1016/j.jclepro.2019.118820.
[33] J. Xiao, J. Guo, L. Zhan, Z. Xu, A cleaner approach to the discharge process of spent
lithium-ion batteries in different solutions, J. Clean. Prod. 255 (2020), 120064,
https://doi.org/10.1016/j.jclepro.2020.120064.
[34] C. Liu, L. Xu, X. Xian, Study on the relationship between concentration of salt
solution and its conductivity, Environ. Monit. China 15 (1999) 21–24.