electronics
Article
Simulation Study on Heat Generation Characteristics of
Lithium-Ion Battery Aging Process
Rui Huang 1,2,3 , Yidan Xu 4 , Qichao Wu 4 , Junxuan Chen 1,2 , Fenfang Chen 4 and Xiaoli Yu 3,4, *
Citation: Huang, R.; Xu, Y.; Wu, Q.;
Chen, J.; Chen, F.; Yu, X. Simulation
Study on Heat Generation
Characteristics of Lithium-Ion
Battery Aging Process. Electronics
2023, 12, 1444. https://doi.org/
10.3390/electronics12061444
Academic Editors: Yi Xie, Dan Dan
and Jiahao Liu
Received: 1 March 2023
Revised: 14 March 2023
Accepted: 15 March 2023
Published: 17 March 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Electronics 2023, 12, 1444. https://doi.org/10.3390/electronics12061444
1 Key Laboratory of Clean Energy and Carbon Neutrality of Zhejiang Province, Hangzhou 310063, China
2 Jiaxing Research Institute, Zhejiang University, Jiaxing 314031, China
3 Key Laboratory of Smart Thermal Management Science & Technology for Vehicles of Zhejiang Province,
Taizhou 317200, China
4 College of Energy Engineering, Zhejiang University, Hangzhou 310027, China
* Correspondence: yuxl@zju.edu.cn
Abstract: Lithium-ion battery heat generation characteristics during aging are crucial for the creation
of thermal management solutions. The heat generation characteristics of 21700 (NCA) cylindrical
lithium-ion batteries during aging were investigated using the mathematical model that was created in
this study to couple electrochemical mechanisms, heat transfer, and aging loss. These findings indicate
that, at the same operating current, the heat generation power of the cell increased significantly with
battery aging. This increase was primarily due to the energy loss caused by the growth of the solid–
electrolyte interface (SEI) and a reduction in the negative porosity and other physical characteristics
of the SEI, such as its ionic conductivity and molar volume, which also had an impact on the heat
generation power. By investigating the variations in battery heat generation in different aging modes,
the electrochemical mechanisms underlying the effects of aging on battery heat generation can be
comprehended in depth.
Keywords: lithium-ion battery; electrochemical–aging–thermal model; heat generation characteristics;
SEI; aging modes
1. Introduction
Due to the rising energy density of lithium-ion batteries, their performance, lifetime,
and thermal stability are substantially influenced by temperature, and thermal management
has become a pressing concern [1–4]. The temperature of a lithium-ion battery is dictated by
its internal heat generation power and external heat transfer power; therefore, knowing the
heat generation characteristics of lithium-ion batteries is essential for effective and precise
thermal management [5,6]. The internal heat generation characteristics of the cell can be
studied thoroughly and efficiently using a numerical model based on the cell’s mechanism
and a vast amount of experimental data.
According to published research, the heat generation characteristics of lithium-ion
batteries are complex and influenced by many factors [7,8]. Bernardi et al. [9] carried out an
analysis of the thermodynamic energy balance inside the battery and pointed out that the
heat generated during the use of the battery mainly comes from Joule heat, reaction heat,
mixing heat, and phase change heat. On this basis, Bahiraei et al. [10] and Chiew et al. [11]
coupled the battery heat generation model with a pseudo-two-dimensional electrochemical
model for thermal management system scheme design. Lithium-ion batteries were subject
to significant aging during operation, and the aging of the battery had an important
impact on the parameters that relate to the internal chemistry of the lithium-ion battery.
Yi et al. [12,13] developed a two-dimensional distribution model of potential, current
density, and heat generation for pouch cells to enable the computation of the battery’s
age-related heat generation. As a major side reaction leading to cell aging, the SEI growth
reaction has an important influence on the heat generation of the cell during aging by
https://www.mdpi.com/journal/electronics
Electronics 2023, 12, 1444 2 of 20

its products [14–16]. Darcovich et al. [17] combined the finite volume model and single
particle model to establish the two-dimensional model of a prismatic cell and coupled it
with a simplified SEI growth model to study changes in the heat of polarization, ohmic
heat, and electrochemical reaction heat of the cell during aging. Tang et al. [18] established
an electrochemical–thermal coupling model for pouch cells, considering the electrolyte
reduction decomposition reaction and the growth of the SEI and analyzed the changes
in reversible heat, polarization heat, ohmic heat, and the total heat generation during
the charging and discharging of batteries in different aging states; they found that the
heat generation power increased during the charging and discharging of aged batteries,
but that the total heat generation decreased. Researchers have undertaken experimental
investigations on variations in heat generation during battery aging, but none of them
has focused on the impacts of the battery aging process on heat generation, nor have they
studied the effects of aging on the characteristics of heat generation.
In this paper, an electrochemical–aging–thermal coupling model of a lithium-ion
battery was proposed. Model parameters that are sensitive to temperature change were
estimated by comparing modeling findings with experimental data. After detailed model
validation at varied operating temperatures and charge–discharge rates, the changes in the
heat-generating components over the aging process and the influences of SEI physical char-
acteristics on the heat-generation process were simulated and reviewed. The mechanism of
the effect of different aging processes on heat generation was investigated.

2. Experiments
A commercial 21700 NCA battery (Φ21 mm × 70 mm cylinder battery, capacity 4.9 Ah,
Si-C anode material, Li (Ni0.8 Co0.15 Al0.05 ) O2 cathode material) was experimentally char-
acterized. A NEWARE (CT4008-5V12A; Shenzhen, China) test system, 8-channel A-to-D
converter, and a computer data logger were used to monitor the charge–discharge current
and battery voltage. A real-time graph was created and stored on a computer. The battery
was placed in a temperature control box for operating at different temperatures.
The complete set of temperature measurement equipment consisted of a T-type ther-
mocouple, an NI 9214 (NI, Austin, TX, USA) thermocouple board, an NI cRIO 9037 (NI,
Austin, TX, USA) data acquisition controller, etc. The T-type thermocouple probe was
stuck on the central site of the batteries, and the batteries were continuously charged and
discharged with a certain constant current. Lastly, the surface temperature change during
the operation was monitored. During the full process, the battery surface temperature was
measured and recorded in situ by a multi-channel temperature logger.
Pulse tests at different ambient temperatures (23 ◦ C, 40 ◦ C) and different currents
(0.5 C, 1 C, 2 C) were performed to characterize the dynamic performance of the lithium-ion
battery. It should be noted that the temperature-dependent parameters, such as lithium-ion
diffusivity in solid and solution phases, and the reaction rate could be extracted from these
experimental data.

3. Model Development
3.1. Model Assumptions and Calculation Domain
This study developed an electrochemical–aging–thermal model for cylindrical 21700 NCA
lithium-ion batteries based on the pseudo-two-dimensional model of Doyle et al. [19–21]. The
principal model assumptions are displayed as follows:
(1) The intricate winding structure within the cell was disregarded and the cell was
treated as a uniform cylinder.
(2) Current collector effects on lithium-ion transport and heat transfer were disregarded.
(3) The change in the entropic coefficient throughout the battery aging process was not
taken into account.
(4) Adverse effects during the operation were disregarded.
Figure 1 shows a sketch of the winding structure of the cylindrical cell [22] and the
schematic calculation domain of the one-dimensional (1D) battery model. In Figure 1, LCu ,
 (4) Adverse effects during the operation were disregarded.
Figure 1 shows a sketch of the winding structure of the cylindrical cell [22] an
schematic calculation domain of the one-dimensional (1D) battery model. In Figure
Electronics 2023, 12, 1444 Ln, Lsep, Lp, and LAl denote the thickness of the negative collector (copper foil), ne
3 of 20

electrode, separator, positive electrode, and positive collector (aluminum foil), r

tively.
Ln , Lsep , Lp, and LAl denote the thickness of the negative collector (copper foil), negative
electrode, separator, positive electrode, and positive collector (aluminum foil), respectively.

1. Intrinsic cell structure [22] and model calculation domain.

Figure 1.Figure
Intrinsic cell structure [22] and model calculation domain.
3.2. Electrochemical Model
3.2. 3.2.1. Lithium-Ion
Electrochemical Concentration Distribution
Model
Assuming that the positive and negative active material particles are spherical and that
3.2.1. Lithium-Ion Concentration
the lithium ions Distribution
only have a concentration gradient in the radial direction of the spherical
particles, the
Assuming distribution
that of lithium
the positive andions in the solid
negative phasematerial
active radial direction according
particles aretospheric
Fick’s second law is given by the equation:
that the lithium ions only have a concentration  
gradient in the radial direction
spherical particles, the distribution of D
= lithium r2 ions in the solid phase radial (1) directi
∂c s s,i ∂ ∂c s
∂t r2 ∂r ∂r
cording to Fick’s second law is given by the equation:
where cs is the lithium-ion concentration in the solid phase, Ds,i is the solid phase diffusivity,
and r is the radius distance variable of∂the D s , i ∂  2 ∂c s 
c s particle.
= r 
The following equation describes ∂tthe concentration
r 2 ∂r  ∂change
r  in lithium ions in the
electrolyte (solution phase) due to the combined effects of diffusion and electromigration:
where cs is the lithium-ion concentration
∂ ∂
 in the
eff ∂ce
 solid phase,
(1 − t+ ) jtot Ds,i is the solid phase dif
ε e,i ce = De,i + (2)
ity, and r is the radius distance ∂t variable ∂x of ∂xthe particle. F
Thewhere
following
εe,i is theequation describes
solution phase volume the concentration
fraction, change
ce is the lithium-ion in lithium
concentration in ions
the in th
trolyte (solution phase)
solution phase, due
x is the to the
distance combined
variable througheffects of diffusion
a cell component, and
t+ is the electromigrati
transference
number of Li-ion species dissolved in a solution, jtot is the total current density, and F is
Faraday’s constant.
∂
ε effective
∂
c = diffusivity
+
 eff ∂ce  eff 1 − t jtot ( )
The relationship between the D D+e,i and the diffusivity De,i of the
solution phase is ∂t e,i e ∂x  e,i ∂x  F
eff βi
De,i = De,i ε e (3)
where εe,i is the solution phase volume fraction, ce is the lithium-ion concentration
solution3.2.2.
phase,Electric Potential Distribution
x is the distance variable through a cell component, t+ is the transf
Based on Ohm’s law and the distribution of potential in solid and solution phases
numberalong
of Li-ion species dissolved in a solution, jtot is the total current density, an
the thickness direction of the electrode, the equation of solid phase potential Φs
Faraday’s constant.
distribution can be derived.  
∂ eff ∂ eff
The relationship between the effective σi φs diffusivity
= jtot De,i and the diffusivity
(4) De,i
∂x ∂x
solution phase is eff
The relationship between the effective conductivity σi and the conductivity σi of the
solid phase is
eff
D eeff, i = D e , i ε eβ i
σi = σi ε s,i (5)
Electronics 2023, 12, 1444 4 of 20

The potential distribution equation for the solution phase is derived using the law of
conservation of charge and Ohm’s law:
   
∂ ∂ ∂ eff ∂
κ eff φe = − κD ln ce − jtot (6)
∂x ∂x ∂x ∂x

2RTκ eff
eff
1 − t+


κD = (7)
F
κ eff = κε e,ii
β
(8)

where Φe is the solution phase potential, κ eff is the effective ionic conductivity of the
eff is the effective ionic diffusion conductivity, R is the gas constant, and T is
electrolyte, κD
the cell temperature.

3.2.3. Electrode Reaction Kinetics

The Butler–Volmer kinetic equation can be used to describe the current density jint of
the lithium embedding and de-embedding reaction:

αp,int F
    
α F
jint = as,i i0,i exp ηact,int − exp − n,int ηact,int (9)
RT RT

jtot
ηact,int = φs − φe − R − Ueq,i (θi ) (10)
as,i film
α  α
surf p surf n
α

i0,i = Fk0,i ce p cmax
s,i − c s,i c s,i (11)

where as,i is the specific surface area, i0,i is the exchange current density, αp, and αn are
the transfer coefficients for the positive and negative current, η act,int is the activation
overpotential of the reaction, Rfilm is the film resistance on the surface of the active material
particles, Ueq,i is the open-circuit voltage, and k0,i is the reaction rate constant.
The positive and negative open-circuit voltages are related to their local charge states
θ i , defined as the ratio of the lithium-ion concentration on the surface of the active material
csurf
s,i to its maximum lithium ion concentration cs,i .
max

csurf
s,i
θi = (12)
cmax
s,i

3.3. Aging Model

3.3.1. Loss of Lithium-Ion Inventory (LLI)
During the operation, solvent molecules in the electrolyte, such as ethylene carbonate
(EC), conducted a redox reaction on the surface of the negative electrode’s active material
to generate an SEI. Based on previous studies [23–26], this study assumed that the result of
the SEI growth reaction in a battery was Li2 CO3 . In addition to the reaction kinetics, the
diffusion of solvent molecules within the existing SEI limited the current density of the SEI
growth reaction. Safari et al. [27] established a model for the SEI growth reaction, including
diffusion constraints based on the Tafer formulation of the aforementioned process.
 
s αn,SEI F
jSEI = − Fk0,SEI cEC exp − ηact,SEI (13)
RT

Rfilm jtot
ηact,SEI = φs − φe − Ueq,SEI − (14)
as,n
Electronics 2023, 12, 1444 5 of 20

The diffusion process of the electrolyte determines the link between the csEC and
EC concentration:
cs − c0EC j
− DEC EC = − SEI (15)
δfilm F
where csEC is the EC concentration diffused to the surface of the negative active particle, c0EC
is the EC concentration in the electrolyte, δfilm is the thickness of the film on the surface of
the active particle, and DEC is the diffusion coefficient of EC in the film.
Combining Equations (13) and (15) produces

c0EC F
jSEI = − δ (16)
film 1
DEC + 
αn,SEI F

k0,SEI exp − RT ηact,SEI

The total reaction current density on the surface of the active material particles
is equal to the sum of the current densities of the SEI growth reaction and the de-
embedded lithium reaction.
jtot = jint + jSEI (17)
According to the principle of mass conservation, the SEI’s growth rate is

∂cSEI j
= − SEI (18)
∂t 2F
The SEI film growth response causes the thickening of the active material’s surface
film. According to the conservation of mass, and assuming that the negative active material
particles are spherical and the surface film is uniformly covered, the rate of change in the
active material surface film thickness δfilm is as follows:

∂δfilm ∂c MSEI j M
= SEI · = − SEI SEI (19)
∂t ∂t as,n ρSEI 2Fas,n ρSEI

∂Rfilm 1 j M
=− · SEI SEI (20)
∂t κSEI 2Fas,n ρSEI
where MSEI is the molar mass and ρSEI is the density of the SEI.
SEI growth leads to a decrease in the porosity of the negative electrode, as described
by the following equation:
∂ε s,n ∂δ
= − as,n film (21)
∂t ∂t

3.3.2. Loss of Active Material (LAM)

By analyzing the differential voltage of the battery aging process, a high linearity
was found for the active material loss indicator GDVA_LAM and the cumulative battery
power throughput ψ after taking the logarithm, respectively. Therefore, after a linear fit,
it can be converted into a power function to represent the relationship between the active
material loss index and the cumulative battery power throughput, and the rate of change
in the solid phase volume fraction as the active material loss rate can be expressed using
the following equation:
∆ε s,i = µA · ψ B (22)
where ψ is the cumulative battery power throughput, and both A and B are constants.

3.4. Thermal Model

The reversible reaction heat, heat of polarization, ohmic heat, heat of SEI growth, and
contact resistance heat are among the heat-generating factors that were taken into account
 Electronics 2023, 12, 1444 6 of 20

by the model [28]. When the aforementioned five heating elements were combined, the
battery’s overall heat generation power was
. . . . . .
qtot ( x, t) = qrea + qact + qohm + qfilm + qc (23)

The cylindrical coordinate system can be employed in the single-cell model since the
object of this investigation is a cylinder. The internal heat transfer equation for the cell
can be obtained by simplifying it to a cylinder with a homogeneous interior material and
equally distributed heat production:
     
∂T 1 ∂ ∂T 1 ∂ ∂T ∂ ∂T .
ρbatt Cp,batt = λbatt,r r + 2 λbatt,ϕ + λbatt,z + qbatt (24)
∂t r ∂r ∂r r ∂ϕ ∂ϕ ∂z ∂z
R Lbatt .
. 0 qtot Ac dx
qbatt = (25)
πR2batt lbatt
where ρbatt is the density of the cell, Cp,batt is the heat capacity of the cell, λbatt,r , λbatt,ϕ ,
and λbatt,z are the radial, circumferential, and axial thermal conductivities of the cell,
.
respectively, qbatt is the average power of heat generation per unit volume of the cell, Lbatt
is the total electrode thickness (including the collector), Rbatt is the outer diameter of the
cell, lbatt is the height of the cell, and Ac is the area of the positive (both sides) that has an
opposing negative.
The power of the external surface’s heat transmission can be defined as follows when
the external medium is air, coolant, and other fluids, with the core and fluid convective
heat transfer:
∂T
−λbatt,r = hcov [ T ( Rbatt , t) − Tamb ] (26)
∂r r= Rbatt
where hcov is the heat transfer coefficient and Tamb is the ambient temperature.

3.5. Solution Method

12, x FOR PEER REVIEW The coupling relationships between the above electrochemical, aging,7and thermal
of 20
models are shown in Figure 2. The model calculations were performed using the finite
element commercial software COMSOL MULTIPHYSICS 6.0.

Figure 2. CouplingFigure 2. Coupling

relationship relationship
between between submodels.
submodels.

4. Model Parameterization and Validation

4.1. Parameterization
4.1.1. Basic Design Parameters
For simulation research, the physical parameters of the simulation object are crucial.
The parameters of the model were determined using experimental measurements, com-
Electronics 2023, 12, 1444 7 of 20

4. Model Parameterization and Validation

4.1. Parameterization
4.1.1. Basic Design Parameters
For simulation research, the physical parameters of the simulation object are crucial.
The parameters of the model were determined using experimental measurements, com-
putations, and estimates. The electrochemical–aging–thermal design parameters for the
4.9 Ah NCA battery are provided in Tables 1–3.

Table 1. Parameters used in the electrochemical model.

Parameters Negative Electrode Separator Positive Electrode

Li /m 5 × 10−6 - 5 × 10−6
σs,i /S·m−1 6.3 × 107 [6] - 3.8 × 107 [6]
Li /m 88 × 10−6 8 × 10−6 62 × 10−6
Ri /m 9 × 10−6 - 5 × 10−6
εs,i 0.71 - 0.703
εe,i 0.21 0.45 0.19
ce,0 /mol·m−3 1200 [29] 1200 [29] 1200 [29]
cmax −3
s,i /mol·m 34,507 [29] - 49,000 [29]
αa,int , αc,int 0.5 [6] - 0.5 [6]
β 1.5 1.5 1.7
Ds,i /m2 ·s−1 9 × 10−14 [30] - 1.5 × 10−14 [30]
σs,i /S·m−1 100 [29] - 3.8 [29]
i0,ref /A·m2 0.75 - 2
EaD,i /J·mol−1 48,000 - 22,000
Eak,i /J·mol−1 36,000 - 30,000
t+ 0.363 [6]
Tref/ K 298.15
Rfilm,0 /Ω·m2 0.001 [28]
Ac /m2 0.12
F/C.mol−1 96,487
R/J.(mol·K)−1 8.314

Table 2. Parameters used in the aging model.

Parameters Value
c0EC /mol·m−3 4500 [31]
Ueq, SEI /V 0.4 [28]
αc, SEI 0.5 [28]
Ea,Dsei/ J·mol−1 30,000
Ea,ksei/ J·mol−1 35,000
σfilm /S·m−1 5 × 10−6 [31]
MSEI /kg·mol−1 0.07388
ρSEI /kg·m−3 2110 [25]

Table 3. Parameters used in the thermal model.

Parameters Value
Cp,batt /J·kg−1 ·K−1 880
λbatt,r /W·m−1 ·K−1 0.7
λbatt,ϕ /W·m−1 ·K−1 10.5
λbatt,z /W·m−1 ·K−1 10.5
ρbatt /kg·m−3 2846
hcov /W·m−2 ·K−1 18
lbatt /m 0.0700
Rbatt /m 0.0105
 Electronics 2023, 12, 1444 8 of 20

4.1.2. Kinetic Parameters

In this paper, the electrochemical–aging–thermal model was utilized to investigate the
features of lithium-ion battery heat generation. As the diffusion coefficients of lithium in
the positive and negative active materials and the reaction rates of the positive and negative
electrodes were strongly impacted by temperature, Arrhenius’ formula was typically used
to characterize their relationship with temperature.
  
EaX,i 1 1
Xi ( T ) = Xiref exp − (27)
R Tref T

Using the relationship described in [32,33], the diffusion coefficient De and conductiv-
ity κ of lithium ions in the electrolyte could be calculated.

−1690 −0.563ce
   
De = 1470 · exp(0.00133ce ) · exp · exp · 10−10 m2 /s (28)
T T

κ = 10−4 ce (−10.5 + 0.074T − 6.96 × 10−5 T 2 + 6.68 × 10−4 ce

(29)
−1.78 × 10−5 ce T + 2.8 × 10−8 ce T 2 + 4.94 × 10−7 c2e − 8.86 × 10−10 c2e T )2

4.1.3. Thermodynamic Parameters

The potential difference of the electrode material is not only determined by the
amount of lithium incorporated in different cell materials but also by the temperature.
The model uses the following equation to calculate the open-circuit voltage of the
electrode at various temperatures.

ref
∂Ueq,i
Ueq,i = Ueq,i + ( T − Tref ) (30)
∂T
Electronics 2023, 12, x FOR PEER REVIEW 9 of 2
Figure 3 depicts the open-circuit voltages (OCV) and entropic coefficient of Si–C
negative and NCA positive at different states of charge (SOC).

4.5 0.5
4.0 0.4 NCA
3.5 SiC
0.3
dU/dT /mV·K−1

3.0 NCA
OCV /V

2.5 SiC 0.2

2.0 0.1
1.5
0.0
1.0
0.5 -0.1
0.0 -0.2
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
SOC SOC
(a) (b)
Figure3.3. Thermodynamic
Figure Thermodynamicparameters
parametersatatdifferent
differentSOC:
SOC:(a)
(a)open-circuit
open-circuitvoltages
voltagesand
and(b)
(b)the
the entropi
coefficient.
entropic coefficient.

4.1.4. Active Material Loss Model Calibration

4.1.4. Active Material Loss Model Calibration
According to Equation (22), and in conjunction with the experimental data, a func-
tional According to established
link could be Equation (22), and the
between in conjunction
loss index ofwith the and
positive experimental data, a func
negative active
materials and the accumulated battery power throughput under different cycle settings, as activ
tional link could be established between the loss index of positive and negative
materials and
demonstrated bythe accumulated
Equations battery
(31)−(34). power
The loss throughput
of positive underactive
and negative different cycle
materials at settings
as demonstrated by Equations (31)−(34). The loss of positive and negative active material
at 40 °C was less affected by the charging and discharging multiplier of the cycle; hence
the same relationship could be utilized for different multipliers.
 Δ ε sn, i _ 23 C 0.5 D 1 = 4.73 × 10 −5 ⋅ψ 0.79
 p (31
 Δ ε s , i _ 23 C 0.5 D 1 = 6.93 × 10 ⋅ψ
− 20 4.58

 Δ ε sn, i _ 23 C 1 D 1 = 8.15 × 10 −5 ⋅ψ 0.90

 Electronics 2023, 12, 1444 9 of 20

40 ◦ C was less affected by the charging and discharging multiplier of the cycle; hence, the
same relationship could be utilized for different multipliers.
(
∆εns,i_23C0.5D1 = 4.73 × 10−5 · ψ0.79
p (31)
∆ε s,i_23C0.5D1 = 6.93 × 10−20 · ψ4.58
(
∆εns,i_23C1D1 = 8.15 × 10−5 · ψ0.90
p (32)
∆ε s,i_23C1D1 = 4.60 × 10−16 · ψ3.77
(
∆εns,i_23C1D2 = 4.08 × 10−4 · ψ0.84
p (33)
∆ε s,i_23C1D2 = 8.62 × 10−22 · ψ5.37
(
∆εns,i_40 = 7.5 × 10−4 · ψ0.55
p (34)
∆ε s,i_40 = 4.26 × 10−8 · ψ1.68

4.2. Model Validation

To validate the accuracy of the electrochemical–aging–thermal coupling model pro-
duced previously, the calculated results of the model were compared to the experimen-
tal test results. Figure 4 compares the experimental and simulated voltage and surface
temperature rise curves of the battery during constant current discharge under various
operating conditions. Figure 5 compares the cell’s heat production power under adiabatic
conditions (initial temperature of 31 ◦ C) with a continuous current discharge at a 0.5 C
multiplier. The foregoing comparison with the experimental results demonstrates that the
tronics 2023, 12, x FOR PEER REVIEW electrochemical–thermal coupling model proposed in this study could accurately simulate 10 of
the electrochemical and heat production characteristics of the new cell under varying
operating conditions.

4.4 4.4 4.4

0.5C_M 0.5C 0.5C_M 0.5C 0.5C_M 0.5C
1C_M 1C 1C_M 1C 1C_M 1C
4.0 2C_M 2C 4.0 2C_M 2C 4.0 2C_M 2C

Discharge@0 ℃ Discharge@23 ℃ Discharge@40 ℃

3.6 3.6 3.6
U/V

U/V

U/V

3.2 3.2 3.2

2.8 2.8 2.8

2.4 2.4 2.4

0 2000 4000 6000 8000 0 2000 4000 6000 8000 0 2000 4000 6000 8000
t/s t/s t/s
(a)Voltage of discharge process at 0 °C (c)Voltage of discharge process at 23 °C (e)Voltage of discharge process at 40 °C
50 60 70
0.5C_M 0.5C 0.5C_M 0.5C 0.5C_M 0.5C
1C_M 1C 1C_M 1C 1C_M 1C
40 65
2C_M 2C 2C_M 2C 2C_M 2C
50
60
30 Discharge@0 ℃ Discharge@23 ℃ Discharge@40 ℃
T/℃

T/℃

T/℃

40 55
20
50
30
10 45

0 20 40
0 2000
4000 6000 8000 0 2000 4000 6000 8000 0 2000 4000 6000 8000
t/s t/s t/s
(b)Temperature rise of discharge process at 0 °C (d)Temperature rise of discharge process at 23 °C (f)Temperature rise of discharge process at 40 °C

FigureFigure 4. Validation
4. Validation of new
of new battery
battery discharge
discharge voltageand
voltage andsurface
surfacetemperature
temperature rises
risesunder
under differ
different
conditions. conditions.

0.7
0.5C Discharge@Adiabatic
0.6
0.5 Model
Experiment
0.4
q/W
 0 10 20 45
40
0 2000 4000 6000 8000 0 2000 4000 6000 8000 0 2000 4000 6000 800
0 t/s 20 t/s 40 t/s
0
(b)Temperature 2000 4000
rise of discharge6000
process8000 0 2000
at 0 °C (d)Temperature 4000 6000
rise of discharge8000
process at 023 °C 2000 4000
(f)Temperature 6000 8000
rise of discharge process at
t/s t/s t/s
(b)Temperature rise of discharge process at 0 °C (d)Temperature rise of discharge process at 23 °C (f)Temperature rise of discharge process at 40 °C

Electronics 2023, 12, 1444

Figure 4. Validation of new battery discharge voltage and surface temperature 10 of 20
rises
Figure 4. Validation
conditions. of new battery discharge voltage and surface temperature rises under differen
conditions.

0.7
0.7
0.5C
0.5C Discharge@Adiabatic
Discharge@Adiabatic
0.6
0.6
0.5 Model
Model
0.5 Experiment
Experiment
0.4
0.4
q/W
q/W
0.3
0.3
0.2
0.2
0.1
0.1
0.0
0 2000 4000 6000 8000
0.0 t/s
0 2000 4000 6000 8000
Figure 5. Validation of newt/sbattery discharge heat generation power under adiabatic conditions.
Figure 5. Validation of new battery discharge heat generation power under adiabatic conditions.
Figure 5. Validation
Figure of new
6 depicts the battery
calculated discharge
findings heat generation
of battery power
capacity decline under
under adiabati
various op-
Figure
erating 6 depictsand
conditions the calculated
the testingfindings
results.ofAt
battery capacity
the end of thedecline underthe
age, when various
battery was
operating
cycledFigureconditions
at 23 °C 6 depictsand
C1D1 and the
the testing results.
23 calculated
°C C1D2 working At the
findings endofof the age,
battery
conditions, when
therecapacitythe battery
decline
was a dramatic und
decline
was cycled at 23 ◦ C C1D1 and 23 ◦ C C1D2 working conditions, there was a dramatic
in capacity.
erating Lithium precipitation
conditions from the
and the testing negative
results. Atelectrode
the end and
ofanthe
increase in the rate
age, when th
decline in capacity. Lithium precipitation from the negative electrode and an increase
of electrode
cycled active material
atof23electrode
°C C1D1 loss
and could
23 °Clossboth
C1D2 be the real causes of the abrupt decline in ca
in the rate active material couldworking
both be theconditions,
real causes of there was a dra
the abrupt
pacity [31].
decline in capacity [31].
in capacity. Lithium precipitation from the negative electrode and an incre
of 5.0
electrode active material loss could
5.0 both be the real causes of the abrupt
C0.5D1 C0.5D1_M
pacity [31]. C1D1 C1D1_M
C1D2 C1D2_M
Cycle@40 ℃
4.5 4.5
Capacity/Ah Capacity/Ah

Capacity/Ah

5.0 5.0
C0.5D1 C0.5D1_M
C1D1 C1D1_M
4.0 C1D2 4.0
Cycle@23C1D2_M
℃ Cycle@40 ℃
4.5 C0.5D1
4.5 C0.5D1_M
Capacity/Ah

C1D1 C1D1_M
C1D2 C1D2_M
3.5 3.5
0 200 400 600 800 0 200 400 600 800
4.0 Cycle index 4.0 Cycle index
(a) Cycle@23 ℃
(b)
C0.5D1 C0.5D1_M
C1D1 C1D1_M
Figure6.6. Validation
Figure Validationofofbattery
batterycapacity
capacityunder
underdifferent
differentconditions
conditionsofofcycle
cycleaging
agingatat(a)
(a)2323◦ C
°C and (b
C1D2 C1D2_M
40 3.5
°C. ◦
and (b) 40 C. 3.5
0 200 400 600 800 0 200 400 600
Cycle index
Since the heat exchange Cycle index rise
boundary conditions are consistent, the temperature
during battery charging(a)and discharging has a clear relationship with the(b)
battery’s heat
generation power, and the change in temperature could also be utilized to indicate the
change in
Figure 6. heat generation
Validation power. The
of battery experimental
capacity underand computed
different results ofof
conditions thecycle
temper-
aging at (
ature
40 °C. rise on the battery’s surface after aging were compared to validate the accuracy
of the model developed in this chapter regarding the heat-generating characteristics of
the battery after aging. Figure 7 depicts the temperature rise curves of the aged battery
at different ambient temperatures during discharge under different cycling conditions.
By comparing Figure 7a,b, it was found that the aged battery was more influenced by
the ambient temperature at a low multiplier discharge, and the internal resistance of
the battery decreased under a high-temperature environment (40 ◦ C), resulting in lower
temperature rises on the surface of the battery. Meanwhile, the simulation results of the
discharge temperature rise of the aged battery at different temperatures were consistent
with the trend in the experimental data, indicating that the model could be used for
subsequent research.

Electronics 2023, 12, 1444
ambient temperature at a low multiplier discharge, and the internal resistance of the b
tery decreased under a high-temperature environment (40 °C), resulting in lower temp
ature rises on the surface of the battery. Meanwhile, the simulation results of the discha
temperature rise of the aged battery at different temperatures were consistent with
trend in the experimental data, indicating that the model could be used11for subsequ
of 20
research.

60 80
C0.5C1 C0.5C1_M C0.5D1 C0.5D1_M
C1D1 C1D1_M C1D1 C1D1_M
C1D2 C1D2_M C1D2 C1D2_M
50 70

Discharge@23 ℃ Discharge@40 ℃

T/℃

T/℃
40 60

30 50

20 40
0 700 1400 2100 2800 3500 0 700 1400 2100 2800 3500
t/s t/s
(a) (b)
Figure7.7.Validation
Figure Validation of surface
of surface temperature
temperature risecells
rise of aging of aging
duringcells during
discharge 23 ◦ C and (b)
at (a)discharge 40 ◦23
at (a) C. °C and
40 °C.
5. Results and Discussion
The performance
5. Results and lifetime of lithium-ion batteries are particularly sensitive to
and Discussion
temperature [34,35], and the majority of studies solely consider the accuracy of the model’s
The performance
computation andignoring
of capacity loss, lifetimetheofchanges
lithium-ion batteries are particularly
in the heat-generating sensitive
characteristics
temperature [34,35], and the majority of studies solely consider the
during battery aging. However, the different values of some key parameters are the cause accuracy of
model’s
for computation
simulation ofresults
calculation capacity
thatloss, ignoring
outline the changes
the changes in the
in the heat heat-generating
generation charac- char
teristics of the battery aging process that deviate significantly from the actual
teristics during battery aging. However, the different values of some key parameters situation;
therefore, it is necessary to use simulations to study the influence of several key factors on
the cause for simulation calculation results that outline the changes in the heat generat
the changes in heat generation in the battery aging process.
characteristics of the battery aging process that deviate significantly from the actual si
ation;
5.1. therefore,
Changes in Heat it is necessary
Generation Power to use simulations
of Different Componentstoduring
study thethe influence
Aging Process of several k
factors on 8the
Figure changes
depicts in heat
the heat generation
generation power in the battery
variation agingcomponents
in different process. with the
number of cycles for the discharging process when the battery is cycled under 23 12 ◦ C C1D1
EER REVIEW of 20
◦
5.1. 40
and Changes inoperating
C C1D1 Heat Generation Power
conditions. of be
It can Different Components
observed during ages,
that, as the battery the Aging Process
SEI heat
generation increases the most, followed by negative solution phase ohmic heat, negative
Figure 8 depicts the heat generation power variation in different components w
polarization heat, and positive polarization heat.
the number of cycles for the discharging process when the battery is cycled under 23
15 C1D1 and 40 °C C1D1 operating 15 conditions. It can be observed that, as the battery ag
SEI heat generationqc increases the most, followed by negative solution phase
qc ohmic he
qnact n
10 negative polarization heat, and positive polarization heat. qact
qe,n
ohm
10 qe,n
qavg/kW·m−3

qavg/kW·m−3

ohm
qe,ohm
p
qe,ohm
p

qpact qpact
5 5
qsep
ohm
qsep
ohm

qsohm qsohm
0 qfilm 0 qfilm
qrea
qrea
C1D1@23 ℃ C1D1@40 ℃
-5 -5
0 200 400 600 0 200 400 600 800 1000
Cycle index Cycle index

(a) (b)
Figure 8. Variation in heat generation
Figure 8. Variation power for different
in heat generation components
power for during
different components cyclic
during cyclicaging at(a)(a)
aging at 23 23
◦C
◦
and (b) 40 C.
°C and (b) 40 °C.
As illustrated in Figure 9, the explanation for the rise in heat power in the SEI film
As illustrated in Figure
is that the9,thickening
the explanation for the
of the SEI makes therise
pathin
ofheat power
lithium-ion in thefrom
diffusion SEIitfilm is
longer,
that i.e., the resistance of the SEI increases. The reason for the increase
the thickening of the SEI makes the path of lithium-ion diffusion from it longer, i.e., in the ohmic thermal

the resistance of the SEI increases. The reason for the increase in the ohmic thermal power
of the negative solution phase is that, on the one hand, the thickening of the SEI consumes
the electrolyte, and on the other, its generation occupies the solution phase space of the
 °C and (b) 40 °C.

As illustrated in Figure 9, the explanation for the rise in heat power in the SEI f
that the thickening of the SEI makes the path of lithium-ion diffusion from it longe
Electronics 2023, 12, 1444 12 of 20
the resistance of the SEI increases. The reason for the increase in the ohmic thermal p
of the negative solution phase is that, on the one hand, the thickening of the SEI cons
the electrolyte, and on the other, its generation occupies the solution phase space o
power of the negative solution phase is that, on the one hand, the thickening of the SEI
negative
consumes theelectrode, resulting
electrolyte, in other,
and on the a decrease in negative
its generation occupiesporosity [31].
the solution In addition
phase space to t
crease in SEI heat
of the negative generation
electrode, resultingand
in anegative
decrease insolution
negativephase ohmic
porosity heat
[31]. In power,
addition to the po
the increase inpolarization
irreversible SEI heat generation
heat atand
the negative solution
end of aging phase ohmic
increases heat which
slightly, power, theis related
positive irreversible
change polarization
in the window heat at the
of electrode end of aging increases
de-embedded lithiumslightly, whichof
as a result is related
lithium-ion in
to the change in the window of electrode de-embedded lithium as a result of lithium-ion
tory loss and active material loss.
inventory loss and active material loss.

8 0.22 12 0.22
Pn Pn 40 ℃_C1D1
23 ℃_C1D1
RSEI RSEI
6 9 0.21
0.21

Rfilm/ mΩ
Rfilm/ mΩ

εe,neg
εe,neg
4 6 0.20

0.20
2 3 0.19

0 0.19 0 0.18
0.0 0.4 0.8 1.2 1.6 0.0 0.7 1.4 2.1 2.8
t/ × 105 s t/ × 105 s
(a) (b)
Figure
Figure 9.9.Variation
Variation inresistance
in SEI SEI resistance andporosity
and negative negative porosity
during during
cyclic aging 23 ◦ C aging
at (a)cyclic 40 ◦(a)
and (b) at C. 23 °C a
40 °C.
5.2. Effect of SEI Physical Parameters on Heat Generation during Aging
The investigation of changes in the heat generation power of different components
5.2. Effect of SEI Physical Parameters on Heat Generation during Aging
during the aging process revealed that the heat generation of the SEI growth reaction was
The investigation
the primary cause for the riseof in
changes in the power
heat generation heat generation power
during the aging of different
process of the cell. compo
On the one hand, the ionic conductivity of the SEI directly determined
during the aging process revealed that the heat generation of the SEI growth SEI resistance, while reaction
on
Electronics 2023, 12, x FOR PEER REVIEWthe other hand, its molar volume determined the growth of its own volume, which was 13 of
the primary cause for the rise in heat generation power during the aging process o
caused by the consumption of a unit amount of substance by the lithium ions, and the
cell.
volumeOngrowth
the one hand,
of the the ionic
SEI resulted in conductivity
a decrease in theofelectrode’s
the SEI directly
porosity, determined
which impacted SEI resist
while on the
the negative other hand,
electrode’s kineticits molar volume
performance [23]. determined the growth of its own vo
5.2.1.
whichIon Conductivity
was caused by the consumption of a unit amount of substance by the lithium
5.2.1. Ion Conductivity
and the volume growth
Figure 10 depicts ofbattery
the the SEIcapacity
resultedreduction,
in a decrease
whichin the
waselectrode’s
calculatedporosity, w
for variou
Figure 10 depicts the battery capacity reduction, which was calculated for various SEI
impacted
SEI the negative
conductivity electrode’sItkinetic
values [26,36–41]. can be performance
observed that [23].
the rate of capacity decline low
conductivity values [26,36–41]. It can be observed that the rate of capacity decline lowered
ered as the ion conductivity of the
as the ion conductivity of the SEI rose. SEI rose.

5.0 5.0

4.8 4.8
23 ℃_C1D1 40 ℃_C1D1
4.6 4.6
Capacity/Ah

Capacity/Ah

4.4 4.4

4.2 1 × 10−6 2 × 10−6

4.2 1 × 10−6 2 × 10−6
4 × 10−6 6 × 10−6 4 × 10−6 6 × 10−6
4.0 8 × 10−6 1 × 10−5
4.0 8 × 10−6 1 × 10−5
1 × 10−4 Unit: S/m 1 × 10−4 Unit: S/m
3.8 3.8
0 100 200 300 400 500 0 200 400 600 800
Cycle index Cycle index
(a) (b)
Figure 10. Simulation
Figure 10. Simulationresults
results
ofof capacity
capacity decline
decline when
when using
using different
different SEI conductivities
SEI ionic ionic conductivities
at at (
23 23 ◦and
(a) °C (b)(b)
C and ◦ C.
4040°C.

Figure 11 shows the SEI resistance and average negative porosity of aged cells fo
various SEI conductivity values. The results in the figure were derived from the discharg
process’ latest cycle. As the lithium-ion conductivity of SEI diminished, the resistance i
the aged SEI membrane increased, particularly when it decreased from 1 × 10−5 S/m to 1
 Figure 10. Simulation results of capacity decline when using different SEI io
23 °C and (b) 40 °C.

Electronics 2023, 12, 1444 Figure 11 shows the SEI resistance and average negative 13 of 20 poro

various SEI conductivity values. The results in the figure were deriv
process’ latest cycle. As the lithium-ion conductivity of SEI diminis
Figure 11 shows the SEI resistance and average negative porosity of aged cells for
the aged
various SEI membrane
SEI conductivity increased,
values. The particularly
results in the figure when
were derived from theit discharge
decreased fr
process’
−6 S/m, and the SEI resistance tended to accelerate growth.inIn add
latest cycle. As the lithium-ion conductivity of SEI diminished, the resistance
10
the aged SEI membrane increased, particularly when it decreased from 1 × 10−5 S/m to
conductivity
1 × 10−6 S/m, andof thethe SEI wastended
SEI resistance reduced withgrowth.
to accelerate age, negative porosity
In addition, when the dec
ionic conductivity of the SEI was reduced with age, negative porosity decreased.

60 0.21
Rfilm_23 ℃ Rfilm_40 ℃
50 εe,neg_23 ℃ εe,neg_40 ℃

40 0.20
Rfilm/ mΩ

εe,neg
30

20 0.19

10

0 0.18
0 2 4 6 8 10
Conductivity/ × 10 S·m−1
−5

Figure 11. Simulation results of the SEI resistance and negative porosity of the cell after aging when
Figure 11. Simulation results of the SEI resistance and negative porosity of t
using different SEI ionic conductivities.
using different SEI ionic conductivities.
As demonstrated in Figure 12, the substantial increase in SEI resistance led to a drastic
increase in SEI heat generation power following cell aging. Additionally, when the ionic
Electronics 2023, 12, x FOR PEER REVIEW As demonstrated
conductivity of the SEI membranein Figure
increased 12,
from 1 ×the
10−5substantial
S/m to 1 × 10−4increase in SEI14reo
S/m, there was
no substantial
tic increase increase
in SEIin theheat
SEI resistance and SEI heat
generation generation
power at the conclusion
following cell ofaging.
the Ad
cycle, showing that 1 × 10−5 S/m was already a higher value for the SEI ionic conductivity.
ionic conductivity of the SEI membrane increased from 1 × 10−5 S/m
9
was80no substantial increase in the SEI resistance and SEI
23 ℃_C1D1 heat genera
23 ℃_C1D1
40 ℃_C1D1 40 ℃_C1D1
of the
60 cycle, showing that 1 × 10 8S/m was already a higher value
−5
−3
qfilm /kW·m−3

ductivity.
ohm /kW·m

40 7
qe,n

20 6

0 5
0 2 4 6 8 10 0 2 4 6 8 10
Conductivity/ × 10−5 S·m−1 Conductivity/ × 10−5 S·m−1
(a) (b)
Figure 12.Simulation
Figure 12. Simulation results
results of heat
of the the heat generation
generation rate the
rate during during the discharge
discharge process
process of the of the cell
cell after
aging whenusing
aging when using different
different SEI SEI
ionicionic conductivities.
conductivities. (a) SEIheat
(a) SEI growth growth heat
and (b) and (b)
negative negative electro
electrolyte
ohmic heat.
ohmic heat.

Figure 13 demonstrates that as the ionic conductivity of the SEI decreases, the rate of
Figure 13 demonstrates that as the ionic conductivity of the SEI decreases, the rat
increase in the heat generation power of the SEI during aging accelerates, resulting in a
increase in the
large increase heat
in the generation
maximum power
surface of the rise
temperature SEI during
duringtheaging accelerates, resulting
discharge.
large increase in the maximum surface temperature rise during the discharge.

24
23 ℃_C1D1
22 40 ℃_C1D1
20
18
ΔT/℃

16
 ohmic heat.

Figure 13 demonstrates that as the ionic conductivity of the SEI decreas

Electronics 2023, 12, 1444
increase in the heat generation power of the SEI during aging accelerates,
14 of 20
large increase in the maximum surface temperature rise during the dischar

24
23 ℃_C1D1
22 40 ℃_C1D1
20
18
ΔT/℃
16
14
12
10
0 2 4 6 8 10
Conductivity/ × 10−5 S·m−1
Simulation
Figure 13.13.
Figure resultsresults
Simulation of maximum
of thesurface
SEI temperature
resistanceriseandduring the discharging
negative porosityprocess
of the cell af
of the cell after aging when using different SEI ionic conductivities.
using different SEI ionic conductivities.
The preceding calculation results highlight the important value of SEI ion conduc-
tivity when conducting the
The preceding coupled electrochemical–thermal–aging
calculation modeling of
results highlight the−important Li-ionof SEI i
value
6 − 5
batteries. Additionally, when the conductivity was between 1 × 10 to 1 × 10 S/m, the
ity when conducting
susceptibility theand
to SEI resistance coupled
cell heat electrochemical–thermal–aging
production increased. modeling
teries. Additionally, when the conductivity was between 1 × 10−6 to 1 × 10−5
5.2.2. Molar Volume
ceptibility to SEI resistance and cell heat production increased.
Figure 14 depicts the battery capacity reduction calculated for different SEI molar
volume values [25,28,36,37,39–41]. When the SEI molar volume increased, so did the rate
5.2.2.
at whichMolar Volume
the cell’s capacity declined. Owing to the rise in the SEI molar volume, the SEI
coating on the surface of the negatively active particles became thicker. The diffusion of
Electronics 2023, 12, x FOR PEER REVIEW Figure 14 depicts the battery capacity reduction calculated for differ
solvent molecules in the electrolyte became challenging in the SEI, resulting in a decrease 15 of
volume
in the pacevalues
at which[25,28,36,37,39–41]. When
the SEI experienced a growth the SEI
reaction; molar
hence, volume
the rate increased, s
of cell capacity
at which
decline was the cell’s
lowered [42].capacity declined. Owing to the rise in the SEI molar vo
coating
5.0 on the surface of the negatively
5.0 active particles became thicker. Th
solvent molecules in the electrolyte became challenging in the SEI, resulting
40 ℃_C1D1
in 4.5
the pace at which the23SEI 4.5 a growth reaction; hence, the rate o
experienced
℃_C1D1
Capacity/Ah

Capacity/Ah

decline was lowered [42].

4.0

4.0 1 × 10−6 1 × 10−5 1 × 10−6 1 × 10−5

2 × 10−5 4 × 10−5 3.5 2 × 10 −5
4 × 10−5
6 × 10−5 8 × 10−5 6 × 10−5 8 × 10−5
1 × 10−4 Unit: m3/mol 1 × 10−4 Unit: m3/mol
3.5 3.0
0 100 200 300 400 500 0 200 400 600 800
Cycle index Cycle index
(a) (b)
Figure 14.Simulation
Figure 14. Simulation results
results of capacity
of capacity loss when
loss when using different
using different SEI
SEI molar molar at
volumes (a) 23 ◦ Cat (a) 23
volumes
and 40◦ °C.
(b)40
and (b) C.

As illustrated in Figure 15, the rise in the molar volume of the SEI led to an increase
As illustrated in Figure 15, the rise in the molar volume of the SEI led to an increa
in the SEI resistance and a decrease in the cell’s negative porosity as a result of aging. An
in the SEI
increase resistance
in the and aenhanced
SEI resistance decreasethein SEI
theheat
cell’s negativepower
generation porosity assame
at the a result of aging. A
current.
increase
Figure 16 in the SEI resistance
demonstrates enhanced
that as the the SEI decreased,
negative porosity heat generation power
the ionic at the same
conductivity and curre
Figure 16 demonstrates
lithium-ion that as
diffusion coefficient the negative
decreased while theporosity decreased,
resistance and ohmic the ionic
heat conductivity
of the nega- a
tive electrolyte increased. As depicted in Figure 17, large increases in the heat generation
lithium-ion diffusion coefficient decreased while the resistance and ohmic heat of the ne of
the SEI
ative and negative
electrolyte electrolyteAs
increased. ohmic heat led
depicted in to an increase
Figure in the
17, large temperature
increases riseheat
in the of thegenerati
battery during operation after age. In contrast to the SEI’s ionic conductivity, the SEI molar
of the SEI and negative electrolyte ohmic heat led to an increase in the temperature rise
the battery during operation after age. In contrast to the SEI’s ionic conductivity, the S
molar volume values in the range of 1 × 10−6 to 1 × 10−4 m3/mol had a substantial effect
the cell capacity’s decline and heat generation after aging.
 Figure 16 demonstrates that as the negative porosity decreased, the ionic c
As illustrated in Figure 15, the rise in the molar volume of the SEI led to an increase
lithium-ion diffusion
in the SEI resistance and acoefficient
decrease in the decreased while
cell’s negative the resistance
porosity as a result ofand ohmi
aging. An
ative electrolyte
increase increased.
in the SEI resistance As depicted
enhanced the SEI heatin Figure
generation17,power
largeatincreases in the
the same current.
Figure
of the 16
SEIdemonstrates
and negative that as the negativeohmic
electrolyte porosity decreased,
heat led tothe
anionic conductivity
increase in the and
tem
Electronics 2023, 12, 1444 15 of 20
lithium-ion diffusion coefficient decreased while the resistance and ohmic heat of the neg-
the battery during operation after age. In contrast to the SEI’s ionic cond
ative electrolyte increased. As depicted in Figure 17, large increases in the heat generation
molar volume
of the SEI values
and negative in the range
electrolyte ohmic heatof 1led × 10
to anto
−6 1 × 10in−4the
increase m3temperature
/mol had arise sub
of
volume values in the range of 1 × 10 − 6 to 1 × 10 − 4 m 3 /mol had a substantial effect on the
the cell capacity’s decline and heat generation after aging.
the battery during operation after age. In contrast to the SEI’s ionic conductivity, the SEI
cell capacity’s decline and heat generation after aging.
molar volume values in the range of 1 × 10−6 to 1 × 10−4 m3/mol had a substantial effect on
the cell
20 capacity’s decline and heat generation
0.28 after aging.
Rfilm_23 ℃ Rfilm_40 ℃

20 εe,neg_23 ℃ εe,neg_23 ℃ 0.28

15 Rfilm_23 ℃ Rfilm_40 ℃ 0.24
/ mΩ εe,neg_23 ℃ εe,neg_23 ℃
15 0.24

εe,neg
10 0.20
R/ mΩfilm

εe,neg
10 0.20
Rfilm

5 0.16
5 0.16

00 0.12 0.12
00 2 2 4 46 68 108 10
Molar
Molarvolume/ × 10−5×m10
volume/
3
−5 −1 3
·mol m ·mol−1
Figure 15. Simulation results of SEI film resistance and anode porosity of cells after aging when
Figure 15. Simulation results of SEI film resistance and anode porosity of cells after aging when using
Figure
using 15. Simulation
different results of SEI film resistance and anode porosity of cells
different SEI filmSEI film
molar molar
volumes. volumes.
using different SEI film molar volumes.
25 10
23 ℃_C1D1 23 ℃_C1D1
25
20
40 ℃_C1D1
9
4010
℃_C1D1
23 ℃_C1D1 23 ℃_C1D1
−3
/kW·m−3

40 ℃_C1D1 40 ℃_C1D1
ohm /kW·m

20
15 8 9
−3
qfilm −3

10 7 ohm /kW·m
qfilm /kW·m

15 8
qe,n

5 6
10 7
qe,n

0 5
0 2 4 6 8 10 0 2 4 6 8 10
5 Molar volume/ × 10−5 m3·mol−1 Molar6volume/ × 10−5 m3·mol−1
(a) (b)
lectronics 2023, 12, x FOR PEER REVIEW 0 5
Figure 0 Simulation
16.
Figure 16. 2 results
Simulation 4 of SEI
results 6 and
of SEI
and 8 electrolyte
negative
negative 10 ohmicohmic
electrolyte 0 generation
heat 2 generation
heat 4 during
rates rates6the
during 8the
−5 3 −1 −5 3
discharge process of volume/
cells after× 10
aging m ·mol
when using different SEI filmMolar
molar volume/
volumes. × 10
(a) SEI ·mol−
discharge process of cells after aging when using different SEI film molar volumes. (a) SEI growthmgrowth
Molar
heat and(b)
heat and (b)negative
negative electrolyte
electrolyte ohmic
ohmic heat.heat.
(a) (b)
16 16. Simulation results of SEI and negative electrolyte ohmic heat generatio
Figure 23 ℃_C1D1
discharge process of cells after aging when using different SEI film molar volum
40 ℃_C1D1
heat and (b) negative electrolyte ohmic heat.
14
ΔT/℃

12

10
0 2 4 6 8 10
Molar volume/ × 10 m ·mol−1
−5 3

Figure
Figure The The
17. 17. maximum surface temperature
maximum rises during the discharging
surface temperature rises duringprocess
theofdischarging
the cell after proc
aging when using different SEI film molar volumes.
aging when using different SEI film molar volumes.

5.3. Effect of Aging Mode on the Change in Heat Generation during Battery Ag
The SEI heat generation, negative electrolyte ohmic heat, negative p
and positive polarization heat are the primary heat-generating factors tha
 Molar volume/ × 10 m ·mol
Figure 17. The maximum surface temperature rises during the discharging process of the cell af
Figure 17. The maximum surface temperature rises during the discharging process of the cell after
aging when using different SEI film molar volumes.
aging when using different SEI film molar volumes.
5.3. Effect of Aging Mode on the Change in Heat Generation during Battery Aging
Electronics
5.3.2023, 12, 1444
Effect of Aging Mode on the Change in Heat Generation during Battery Aging 16 of 20
The SEI heat generation, negative electrolyte ohmic heat, negative polarization he
The SEI heat generation,
and positive negative electrolyte
polarization heat areohmic heat, negative
the primary polarization
heat-generating factorsheat,
that are altered w
and positive polarization
cell heat
aging are
[9]; the primary
thus, the heat-generating
discussion focuses factors
mostly that
on are
these altered
5.3. Effect of Aging Mode on the Change in Heat Generation during Battery Agingfour with
heat-generating term
cell aging [9]; thus, Figure
the The
discussion
18
SEIcompares focuses mostly
the average
heat generation, on
negativeheat these four
ohmicheat-generating
generation
electrolyte power terms.heat
of the polarization
heat, negative primary heat,generati
Figure 18 comparesterms the average
and positive heat
at thepolarization generation
beginning-of-life power
(BOL)
heat are the of the primary
and heat-generating
primary end-of-life (EOL) heat generation
throughout
factors the discharge
that are altered with p
cell aging
terms at the beginning-of-life
cess. [9];
The aging thus,
(BOL) the discussion
and
mode end-of-life
of an aging focuses
(EOL)
batterymostly on these
throughout
consists fourthe
the
of only heat-generating
discharge
LLI andpro- terms. LLI a
concurrent
cess. The aging mode Figure
of an18aging
LAM. compares the average
battery consists heat
of generation
only the LLIpower
andof concurrent
the primary heat
LLI generation
and
terms at the beginning-of-life (BOL) and end-of-life (EOL) throughout the discharge process.
LAM. The aging mode of an aging battery consists of only the LLI and concurrent LLI and LAM.
25 25
BOL BOL
23 ℃_C1D1 40 ℃_C1D1
EOL_LLI&LAM 25 EOL_LLI&LAM
25 20
BOL 20 23 EOL_LLI
℃_C1D1 BOL EOL_LLI
EOL_LLI&LAM EOL_LLI&LAM 40 ℃_C1D1
20 20

q/kW·m−3
EOL_LLI
q/kW·m−3

EOL_LLI 15 15

q/kW·m−3
q/kW·m−3

15 10 15 10

10 5 10 5

5 0 5 0
act_n ohm_e,n act_p film act_n ohm_e,n act_p film

0 (a) 0 (b)
act_n ohm_e,n act_p film act_n ohm_e,n act_p film
Figure 18. Comparison of the main heat generation items during the discharging process when
(a) (b)
cell ages to EOL under different aging modes at (a) 23 °C and (b) 40 °C.
Figure 18. ComparisonFigure
of the18.main heat generation
Comparison of the main items during the
heat generation items discharging process when
during the discharging processthe
when the
cell ages to EOL undercell Figure
different
ages 19under
aging
to EOL demonstrates
modes at (a)
different that
23
aging °C when (a)just
andat(b)
modes 40 the
23 ◦°C. LLI
C and (b) occurs
40 C. in a battery, the negative el
◦

trolyte ohmic heat and SEI heat are greater after aging because all the lost lithium ions a
Figure 19 demonstrates that when just the LLI occurs in a battery, the negative electrolyte
involved in
Figure 19 demonstrates thewhen
that SEI growth
just thereaction, the SEI
LLI occurs inisa thicker,
battery,and thethe porosity
negative declines more af
elec-
ohmic heat and SEI heat are greater after aging because all the lost lithium ions are involved
trolyte ohmic heat andaging.
SEI
in the heat
SEI arereaction,
growth greaterthe after
SEI aging because
is thicker, and the all the lost
porosity lithium
declines more ions are
after aging.
involved in the SEI growth reaction, the SEI is thicker, and the porosity declines more after
25 0.24 25 0.24
aging. R _LLI&LAM R _LLI
23 ℃ 40 ℃
R _LLI&LAM
film R _LLI film film film
εe,neg_LLI&LAM εe,neg_LLI εe,neg_LLI&LAM εe,neg_LLI
20 0.22 20 0.22
25 Rfilm_LLI&LAM Rfilm_LLI
0.24 25 R _LLI&LAM Rfilm_LLI
0.24
15 40 ℃ 0.20
film 15 0.20
Rfilm/ mΩ

23 ℃
Rfilm/ mΩ
εe,neg

εe,neg
εe,neg_LLI&LAM εe,neg_LLI ε _LLI&LAM e,neg εe,neg_LLI
20 0.22 20 0.22
10 0.18 10 0.18
15 0.20 15 0.20
Rfilm/ mΩ

Rfilm/ mΩ

5 0.16 5 0.16
εe,neg

εe,neg

10 0 0.18 10 0.14 0 0.18 0.14

0.0 0.4 0.8 1.2 1.6 0.0 0.5 1.0 1.5 2.0 2.5
5 0.16t/ × 105 s 5 0.16
t/ × 105 s

0 0.14 0
(a) 0.14
(b)
0.0 0.4 0.8 1.2 1.6 0.0 0.5 1.0 1.5 2.0 2.5
Figure 19. Variations in SEI film resistance and anode porosity under different aging modes at
t/ × 105 s ◦ t/ × 105 s
(a) 23 C and (b) 40 ◦ C.
(a) (b)
In addition, only in the case of LLI was there a significant decrease in the heat of
the polarization of positive and negative electrodes after aging in the battery. As shown
in Figure 20, this is because the space available for lithium embedding in positive and
negative electrodes remains unchanged, but the amount of lithium ions migrating by
diffusion between positive and negative electrodes decreases, causing the maximum degree
of lithium embedding in positive and negative electrodes to decrease.
 In addition, only in the case of LLI was there a significant decrease in the heat of t
polarization of positive and negative electrodes after aging in the battery. As shown
Figure 20, this is because the space available for lithium embedding in positive and neg
tive electrodes remains unchanged, but the amount of lithium ions migrating by diffusi
Electronics 2023, 12, 1444 between positive and negative electrodes decreases, causing the maximum 17 degree
of 20 of li
ium embedding in positive and negative electrodes to decrease.

1.0 1.0
23 ℃_C1D1 40 ℃_C1D1
0.8 0.8

0.6 0.6

SOC

SOC
SOCn SOCn
0.4 SOCp 0.4 SOCp

0.2 0.2

0.0 0.0
0.0 0.4 0.8 1.2 1.6 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8
t/ × 105 s t/ × 105 s
(a) (b)
Figure
Figure 20.20.Variations
Variations in electrode
in electrode SOC during
SOC during theprocess
the aging agingwhen
process
onlywhen only LLI
LLI occurred occurred
in the cell at in the c
◦ ◦
at (a)
(a) 23 23 °C and
C and (b) 40(b)C.40 °C.

In conclusion, if the real aging pattern of the battery contains LAM but is not ac-
In conclusion, if the real aging pattern of the battery contains LAM but is not
counted for in the modeling, it results in a big computation of SEI heat generation, negative
counted
electrolytefor in the
ohmic modeling,
heat, and a minor it results in a of
calculation bigpolarization
computation heatof SEIthe
after heat generation,
battery has neg
tive
aged.electrolyte
This is due ohmic heat, and
to the greater a minor
sensitivity calculation
of SEI of polarization
heat generation and negative heat after the batte
electrolyte
has
ohmicaged.
heatThis
to theisaging
due to the greater
pattern, ignoringsensitivity
LAM results of SEI
in aheat generation
considerable and negative
increase in heat elect
lyte ohmicpower
generation heat and
to the aging
surface pattern, ignoring
temperature LAM
after the cell results
has aged, in a considerable
as depicted in Figure 21.increase
It cangeneration
heat also be seen power
from Figure 21 that the
and surface contour of the
temperature temperature,
after the cell hasrepresented
aged, as by the
depicted in F
increasing curve, during the discharge of the battery in different aging modes
ure 21. It can also be seen from Figure 21 that the contour of the temperature, represent was distinct.
When LLI and LAM occurred, the temperature rise rate at the conclusion of discharge
by the increasing curve, during the discharge of the battery in different aging modes w
was larger than when only LLI occurred. When the LAM is not taken into account, the
distinct.
maximumWhen amount LLI and LAM
of lithium occurred,
embedded in thethe temperature
positive rise electrodes
and negative rate at theduring
conclusion
the of d
charge was larger than when only LLI
battery cycle drops, as does the polarization heat. occurred. When the LAM is not taken into accou
the maximum amount of lithium embedded in the positive and negative electrodes duri
the battery cycle drops, as does the polarization heat.

40 56
D1@23 ℃_C1D1 D1@40 ℃_C1D1
35 52
T/℃

T/℃

30 48

25 BOL 44 BOL
EOL_LLI&LAM EOL_LLI&LAM
EOL_LLI EOL_LLI
20 40
0 1000 2000 3000 4000 0 1000 2000 3000 4000
t/s t/s
Figure 21. The comparison(a) (b) when the
of the surface temperature rises during the discharging process
cell ages to EOL under different aging modes at (a) 23 ◦ C and (b) 40 ◦ C.
Figure 21. The comparison of the surface temperature rises during the discharging process wh
the cell ages to EOL under different aging modes at (a) 23 °C and (b) 40 °C.
6. Conclusions
In this study, based on the aging mechanism of LLI and LAM, a mathematical model of
6.
theConclusions
electrochemical–aging–thermal interaction was constructed. The heat generation power
and surface temperature rise of the battery were determined by comparing the simulation
findings with the experimental data of combined working conditions with varying current
charging and discharging rates (0.5 C, 1 C, and 2 C) and temperatures (23 ◦ C and 40 ◦ C).
In terms of electrochemical characteristics, thermal properties, and aging properties, the
simulation findings were validated and shown to be in good agreement with the actual
Electronics 2023, 12, 1444 18 of 20

data. Based on this model, the heat generation characteristics of NCA batteries during
aging were examined and the results revealed that:
(1) As the battery ages, its overall heat generation power at the same operating current
grows dramatically. This is mostly due to the heat generation power increase in the
SEI and the ohmic heat of the negative electrolyte.
(2) The SEI ionic conductivity falls below 1 × 10−5 S/m, the SEI resistance and heat
generation power grow exponentially, and the negative electrolyte ohmic heat di-
minishes quickly. The increase in the SEI molar volume causes both the SEI heat
generation power and the negative electrolyte ohmic heat to rise. Therefore, in the
battery manufacturing process, reasonable formation technology should be adopted
to improve the ion conductivity of the SEI and accelerate the lithium-ion mass transfer
on the electrode surface and, at the same time, improve the compactness of the SEI to
inhibit the decrease in electrode porosity caused by volume growth.
(3) Compared with the occurrence of both LLI and LAM, when only LLI occurred in
the cell, the SEI film was thicker, and the negative porosity lower when the same
capacity declined, so the SEI heat generation and the negative electrolyte ohmic heat
could be greater. Simultaneously, the maximum amount of lithium embedded in the
positive and negative electrodes continued to diminish throughout the course of the
cycle, as did the polarized thermal power at the conclusion of battery charging and
discharging. Consequently, when just LLI occurred, the maximum temperature rise
during the operation at the same capacity decline was greater than when both LLI
and LAM occurred, and the temperature rise rate at the end of discharge was lower
due to the narrower window of positive and negative lithium de-embedding.

Author Contributions: Conceptualization, R.H., Y.X. and F.C.; methodology, R.H., Y.X. and Q.W.; val-
idation, Y.X., Q.W. and F.C.; formal analysis, J.C.; investigation, R.H., J.C. and X.Y.; writing—original
draft preparation, R.H. and Y.X.; writing—review and editing, R.H., Y.X. and J.C.; visualization, Y.X.
and Q.W.; supervision, R.H. and X.Y. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by the State Key Laboratory of Clean Energy Use Open Fund,
grant number ZJUCEU2022016.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Camargos, P.H.; dos Santos, P.H.; dos Santos, I.R.; Ribeiro, G.S.; Caetano, R.E. Perspectives on LI-ION battery categories for
electric vehicle applications: A review of state of the art. Int. J. Energy Res. 2022, 46, 19258–19268. [CrossRef]
2. Das, U.K.; Shrivastava, P.; Tey, K.S.; Idris, M.Y.I.B.; Mekhilef, S.; Jamei, E.; Seyedmahmoudian, M.; Stojcevski, A. Advancement of
lithium-ion battery cells voltage equalization techniques: A review. Renew. Sustain. Energy Rev. 2020, 134, 110227. [CrossRef]
3. Abdelkareem, M.A.; Maghrabie, H.M.; Abo-Khalil, A.G.; Adhari, O.H.K.; Sayed, E.T.; Radwan, A.; Elsaid, K.; Wilberforce, T.; Olabi,
A.G. Battery thermal management systems based on nanofluids for electric vehicles. J. Energy Storage 2022, 50, 104385. [CrossRef]
4. Jiang, Z.Y.; Li, H.B.; Qu, Z.G.; Zhang, J.F. Recent progress in lithium-ion battery thermal management for a wide range of
temperature and abuse conditions. Int. J. Hydrogen Energy 2022, 47, 9428–9459. [CrossRef]
5. Li, Y.; Zhou, Z.; Wu, W.-T. Three-dimensional thermal modeling of Li-ion battery cell and 50 V Li-ion battery pack cooled by
mini-channel cold plate. Appl. Therm. Eng. 2018, 147, 829–840. [CrossRef]
6. Bahiraei, F.; Ghalkhani, M.; Fartaj, A.; Nazri, G.-A. A pseudo 3D electrochemical-thermal modeling and analysis of a lithium-ion
battery for electric vehicle thermal management applications. Appl. Therm. Eng. 2017, 125, 904–918. [CrossRef]
7. Huang, R.; Li, Z.; Hong, W.; Wu, Q.; Yu, X. Experimental and numerical study of PCM thermophysical parameters on lithium-ion
battery thermal management. Energy Rep. 2020, 6, 8–19. [CrossRef]
8. Xie, Y.; He, X.-J.; Hu, X.-S.; Li, W.; Zhang, Y.-J.; Liu, B.; Sun, Y.-T. An improved resistance-based thermal model for a pouch
lithium-ion battery considering heat generation of posts. Appl. Therm. Eng. 2019, 164, 114455. [CrossRef]
9. Bernardi, D.M.; Pawlikowski, E.M.; Newman, J. A General Energy Balance for Battery Systems. J. Electrochem. Soc. 1985,
132, 5–12. [CrossRef]
10. Bahiraei, F.; Fartaj, A.; Nazri, G.-A. Electrochemical-thermal Modeling to Evaluate Active Thermal Management of a Lithium-ion
Battery Module. Electrochim. Acta 2017, 254, 59–71. [CrossRef]
Electronics 2023, 12, 1444 19 of 20

11. Chiew, J.; Chin, C.S.; Toh, W.D.; Gao, Z.; Jia, J.; Zhang, C.Z. A pseudo-three-dimensional electrochemical-thermal model of a
cylindrical LiFePO4/graphite battery. Appl. Therm. Eng. 2019, 147, 450–463. [CrossRef]
12. Yi, J.; Koo, B.; Shin, C.B.; Han, T.; Park, S. Modeling the effect of aging on the electrical and thermal behaviors of a lithium-ion
battery during constant current charge and discharge cycling. Comput. Chem. Eng. 2017, 99, 31–39. [CrossRef]
13. Yi, J.; Lee, J.; Shin, C.B.; Han, T.; Park, S. Modeling of the transient behaviors of a lithium-ion battery during dynamic cycling.
J. Power Sources 2015, 277, 379–386. [CrossRef]
14. Liu, J.; Duan, Q.; Ma, M.; Zhao, C.; Sun, J.; Wang, Q. Aging mechanisms and thermal stability of aged commercial 18650 lithium
ion battery induced by slight overcharging cycling. J. Power Sources 2019, 445, 227263. [CrossRef]
15. Zhao, X.; Yin, Y.; Hu, Y.; Choe, S.-Y. Electrochemical-thermal modeling of lithium plating/stripping of Li(Ni0.6Mn0.2Co0.2)O2/Carbon
lithium-ion batteries at subzero ambient temperatures. J. Power Sources 2019, 418, 61–73. [CrossRef]
16. Yang, S.; Hua, Y.; Qiao, D.; Lian, Y.; Pan, Y.; He, Y. A coupled electrochemical-thermal-mechanical degradation modelling
ap-proach for lifetime assessment of lithium-ion batteries. Electrochim. Acta 2019, 326, 134928. [CrossRef]
17. Darcovich, K.; MacNeil, D.D.; Recoskie, S.; Cadic, Q.; Ilinca, F.; Kenney, B. Coupled Numerical Approach for Automotive Battery
Pack Lifetime Estimates with Thermal Management. J. Electrochem. Energy Convers. Storage 2018, 15, 021004. [CrossRef]
18. Tang, Y.; Wu, L.; Wei, W.; Wen, D.; Guo, Q.; Liang, W.; Xiao, L. Study of the thermal properties during the cyclic process of lithium
ion power batteries using the electrochemical-thermal coupling model. Appl. Therm. Eng. 2018, 137, 11–22. [CrossRef]
19. Fuller, T.F.; Doyle, M.; Newman, J. Simulation and Optimization of the Dual Lithium Ion Insertion Cell. J. Electrochem. Soc. 1994,
141, 1–10. [CrossRef]
20. Doyle, M.; Fuller, T.F.; Newman, J. Modeling of Galvanostatic Charge and Discharge of the Lithium/Polymer/Insertion Cell.
J. Electrochem. Soc. 1993, 140, 1526–1533. [CrossRef]
21. Doyle, M.; Newman, J.; Gozdz, A.S.; Schmutz, C.N.; Tarascon, J.-M. Comparison of Modeling Predictions with Experimental Data
from Plastic Lithium Ion Cells. J. Electrochem. Soc. 1996, 143, 1890. [CrossRef]
22. Voelcker, J. Lithium Batteries Take to the Road. IEEE Spectr. 2007, 44, 26–31. [CrossRef]
23. Christensen, J.; Newman, J. A Mathematical Model for the Lithium-Ion Negative Electrode Solid Electrolyte Interphase.
J. Electrochem. Soc. 2004, 151, A1977–A1988. [CrossRef]
24. Li, J.; Landers, R.; Park, J. A comprehensive single-particle-degradation model for battery state-of-health prediction.
J. Power Sources 2020, 456, 227950. [CrossRef]
25. Arora, P.; Doyle, M.; White, R.E. Mathematical modeling of the lithium deposition overcharge reaction in lithium-ion batteries
using carbon-based negative electrodes. J. Electrochem. Soc. 1999, 146, 3543. [CrossRef]
26. Deshpande, R.D.; Verbrugge, M.W.; Cheng, Y.-T.; Wang, J.; Liu, P. Battery Cycle Life Prediction with Coupled Chemical
Degradation and Fatigue Mechanics. J. Electrochem. Soc. 2012, 159, A1730–A1738. [CrossRef]
27. Safari, M.; Morcrette, M.; Teyssot, A.; Delacourt, C. Multimodal Physics-Based Aging Model for Life Prediction of Li-Ion Batteries.
J. Electrochem. Soc. 2009, 156, A145–A153. [CrossRef]
28. Rao, L.; Newman, J. Heat-Generation Rate and General Energy Balance for Insertion Battery Systems. J. Electrochem. Soc. 1997,
144, 2697–2704. [CrossRef]
29. Lee, K.J.; Smith, K.; Pesaran, A.; Kim, G.H. Three dimensional thermal-, electrical-, and electrochemical-coupled model for
cy-lindrical wound large format lithium-ion batteries. J. Power Sources 2013, 241, 20–32. [CrossRef]
30. Dees, D.W.; Kawauchi, S.; Abraham, D.P.; Prakash, J. Analysis of the Galvanostatic Intermittent Titration Technique (GITT) as
applied to a lithium-ion porous electrode. J. Power Sources 2009, 189, 263–268. [CrossRef]
31. Yang, X.-G.; Leng, Y.; Zhang, G.; Ge, S.; Wang, C.-Y. Modeling of lithium plating induced aging of lithium-ion batteries: Transition
from linear to nonlinear aging. J. Power Sources 2017, 360, 28–40. [CrossRef]
32. Landesfeind, J.; Gasteiger, H.A. Temperature and Concentration Dependence of the Ionic Transport Properties of Lithium-Ion
Battery Electrolytes. J. Electrochem. Soc. 2019, 166, A3079–A3097. [CrossRef]
33. Valøen, L.O.; Reimers, J.N. Transport Properties of LiPF6-Based Li-Ion Battery Electrolytes. J. Electrochem. Soc. 2005,
152, A882. [CrossRef]
34. Lu, Z.; Yu, X.L.; Wei, L.C.; Cao, F.L.; Zhang, Y.; Meng, X.Z.; Jin, L.W. A comprehensive experimental study on tempera-ture-
dependent performance of lithium-ion battery. Appl. Therm. Eng. 2019, 158, 113800. [CrossRef]
35. Diao, W.; Saxena, S.; Pecht, M. Accelerated cycle life testing and capacity degradation modeling of LiCoO2-graphite cells.
J. Power Sources 2019, 435, 226830. [CrossRef]
36. Ramadass, P.; Haran, B.; Gomadam, P.M.; White, R.; Popov, B.N. Development of First Principles Capacity Fade Model for Li-Ion
Cells. J. Electrochem. Soc. 2004, 151, A196–A203. [CrossRef]
37. Kindermann, F.M.; Keil, J.; Frank, A.; Jossen, A. A SEI Modeling Approach Distinguishing between Capacity and Power Fade.
J. Electrochem. Soc. 2017, 164, E287–E294. [CrossRef]
38. Ning, G.; Popov, B.N. Cycle Life Modeling of Lithium-Ion Batteries. J. Electrochem. Soc. 2004, 151, A1584–A1591. [CrossRef]
39. Atalay, S.; Sheikh, M.; Mariani, A.; Merla, Y.; Bower, E.; Widanage, W.D. Theory of battery aging in a lithium-ion battery: Capacity
fade, nonlinear aging and lifetime prediction. J. Power Sources 2020, 478, 229026. [CrossRef]
40. Tahmasbi, A.A.; Kadyk, T.; Eikerling, M.H. Statistical Physics-Based Model of Solid Electrolyte Interphase Growth in Lithium Ion
Batteries. J. Electrochem. Soc. 2017, 164, A1307–A1313. [CrossRef]
Electronics 2023, 12, 1444 20 of 20

41. Liang, J.; Gan, Y.; Yao, M.; Li, Y. Numerical analysis of capacity fading for a LiFePO4 battery under different current rates and
ambient temperatures. Int. J. Heat Mass Transf. 2020, 165, 120615. [CrossRef]
42. Ekström, H.; Lindbergh, G. A Model for Predicting Capacity Fade due to SEI Formation in a Commercial Graphite/LiFePO4 Cell.
J. Electrochem. Soc. 2015, 162, A1003–A1007. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.