Energy Conversion and Management 150 (2017) 304–330

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Review

Thermal issues about Li-ion batteries and recent progress in battery thermal MARK
management systems: A review
⁎
Huaqiang Liua, Zhongbao Weib, Weidong Hec, Jiyun Zhaoa,
a
Department of Mechanical and Biomedical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong Special Administrative Region
b
Energy Research Institute, Nanyang Technological University, Singapore 637459, Singapore
c
School of Energy Science and Engineering, University of Electronic Science and Technology, Chengdu, Sichuan 611731, PR China

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

Keywords: Electrical vehicles have the capability to lessen the severe threats of energy crisis and environment pollution. The
Li-ion battery Lithium ion battery as a promising solution for the energy storage in vehicular applications is briefly introduced
Temperature effects in this paper. The adverse effects of improper temperature, including performance degradation, potential
Battery modeling thermal runaway, temperature non-uniformity and low temperature performance are described afterwards. The
Thermal management system
thermal model, electrochemical model, equivalent circuit model and electrochemical/electrical-thermal cou-
Phase change
pling methods are also elaborated for the accurate battery modeling. More importantly, this review detailedly
summarizes the progress on battery thermal management systems (BTMSs) including the air, liquid, boiling, heat
pipe and solid-liquid phase change based strategies during recent years. Influence factors and development focus
of different BTMSs are stated elaborately. Passive cooling systems utilizing the latent heat during the phase
change process are more attractive options compared to the conventional single phase forced air and liquid
cooling methods. However, there still exist some challenges to be addressed before commercializing. In addition,
different methods could be combined to meet the requirements of various applications. The improvement of
these existing BTMSs is supposed to be paid more attention to enhance Li-the performance and safety of Li-ion
batteries.

1. Introduction
Nowadays, people all over the world are being posed to the threat of
energy crisis and severe environment pollution. However, driven by the
growing improvements in economic levels, the global demands for
energy resources, especially for the fossil fuels (e.g. oil), are still rising
rapidly. According to the statistics from BP Statistical Review of World
Energy 2016 [1] and BP Energy Outlook 2016 [2], the global oil con-
sumption grew by 1.9 million barrels per day, nearly two-thirds of
which was attributed to the transportation sector. To reduce the
emissions of pollutants and greenhouse gases, increasing attentions
have been paid to the development of clean energy vehicles, including
pure electric vehicles (EV), hybrid electric vehicles (HEV) and plug-in
hybrid electric vehicles (PHEV), which are more energy-efficient and
environment friendly than the traditional internal combustion engine
systems [3]. Anderson et al. [4] indicated that the employment of EVs
could bring about a remarkable reduction of greenhouse gas emissions
by almost 40% if the electric power resources are renewable (e.g. solar,
wind, etc.).
The key task of developing clean energy vehicles is to find a suitable
energy storage system which can support high driving mileage and fast
acceleration. Various power batteries have been proposed, such as lead-
acid (Pb-acid), nickel based, sodium based and lithium-ion (Li-ion)
batteries [9]. Some key indicators for evaluating these battery systems
are listed in Table 1. Among all these electrochemical systems, Li-ion
batteries are the most promising choices for EVs and HEVs due to their
superior characteristics of high specific energy and power, long cycling
life, and low self-discharge rates, as indicated in Table 1. Therefore, Li-
ion batteries have been widely employed in our daily lives, including
wrist watches, smart phones, laptops and many other portable elec-
tronics [10].
However, in spite of its attractive prospect, Li-ion battery based EVs
have not been widely adopted by consumers. The major hindrances in
regard with the battery system include the high cost, limited calendar
life, safety concerns, and the temperature-caused degradation [11].
Currently, extensive efforts have been made to find better electrode
materials to enhance the specific energy, specific power and lifetime.
However, the research in the thermal issues of Li-ion batteries under
various conditions and the development of battery thermal manage-
ment systems (BTMSs) have not been adequately addressed although

⁎
Corresponding author.
E-mail address: jiyuzhao@cityu.edu.hk (J. Zhao).

http://dx.doi.org/10.1016/j.enconman.2017.08.016
Received 30 May 2017; Received in revised form 1 August 2017; Accepted 5 August 2017
0196-8904/ © 2017 Elsevier Ltd. All rights reserved.
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

Nomenclature velocity of solvent

T temperature, K
as specific interfacial area of solid phase, m2 m−3 UL load voltage, V
A surface area of active material, m2 UOC open-circuit potential, V
As cell surface area exposed to the convection coolant, m2 URC voltage across the RC network, V
Bi Biot number V cell operating voltage, V
ce concentration of lithium ions in electrolyte phase,
mol m−3 Greek symbols
cs concentration of lithium ions in solid phase, mol m−3
C0 battery capacity, A h αa transfer coefficient of anode
Cp battery specific heat capacity, J kg−1 K−1 αc transfer coefficient of cathode
Ds mass diffusion coefficient of lithium in solid phase, m2 s−1 σ Stefan-Boltzmann constant, 5.67 × 10−8 W m−2 K−4
Ea activation energy, J σ eff effective electronic conductivity of the solid phase, S m−1
f± molecular salt activity coefficient ρ battery density, kg m−3
F faraday constant, 96,485 C mol−1 λ thermal conductivity, W m−1 K−1
h heat transfer coefficient, W m−2 K−1 ε emissivity of the cell surface
hs cell heat transfer coefficient, W m−2 K−1 εe volume fraction/porosity of electrolyte
i current per unit volume A m−3 φ angle in cylindrical-coordinate system, rad
ie current density in electrolyte phase, A m−2 ∅s potential in the solid phase, V
in transfer current density, A m−2 ∅e potential in the electrolyte phase, V
io exchange current density, A m−2 ηs surface overpotential, V
is current density in solid phase, A m−2 Ψ transport and kinetic parameters
I applied current density, A
IL current in the load circuit, A Subscripts
jn pore-wall flux across interface, mol m−2 s−1
ka anodic reaction rate constant, m3 s−1 ref with respect to a reference state
kc cathodic reaction rate constant, m3 s−1 amb ambient environment or coolant
k eff effective ionic conductivity, S m−1 avg with respect to average temperature
L total thickness of the cell, m rad equivalent radiative coefficient
Ln thickness of negative electrode, m t equivalent coefficient which consists of the convection
Lsep thickness of separator, m and radiation
n number of electrons participating in the reaction σ with respect to temperature difference
Pfluid fluid pressure drop, Pa x x coordinate
q̇ rate of internal heat generation per unit volume W m−3 y y coordinate
Q̇ heat generation rate of the battery, W z z coordinate
r distance in radial direction, m r radial coordinate
R1 equivalent resistance in the load circuit, Ω
Rg universal gas constant, 8.3145 J mol−1 K−1 Superscripts
Ri internal equivalent resistance of unit volume, Ω m−2
Rs radius of solid active material particle, m max the maximum value in the particle of electrode
ΔS entropy change, J mol−1 K−1 surf the value on the surface of the particle of electrode
t time, s avg the average value in the particle of electrode
t+0 transference number of the lithium ions with respect to the

they have a large impact on the performance, lifespan and security of temperature range between 15 °C and 35 °C is recommended by Pesaran
battery. The acceptable operating temperature range for Li-ion batteries et al. [13]. Temperatures outside the desired range will result in sig-
is −20 to 60 °C [12] and to maintain its optimal performance, a narrow nificant capacity loss. Motloch et al. indicated that every degree’s

Table 1
Some properties of different electric vehicle batteries.
adapted from [5–8]

Type Specific energy (W h kg−1) Specific power (kW kg−1) Nominal cell voltage (V) Life cycles1 Charge time (h) Self-discharge rate2 (%)

Lead-acid 25–40 150–250 2 200–700 8 5

Nickel-cadmium 45–80 200 1.2 500–2000 1 20
Nickel-metal hydride 60–120 200 1.2 500–1000 1 30
Nickel-iron 50 100 1.2 2000 / 20–40
Nickel-zinc 70 150 1.7 300 / 20
Sodium-sulphur 100 150–200 2 ∼10003 8 Quite low
Sodium metal chloride 100 150 2 > 10004 8 /
Li-ion 110–180 300 3.6 > 1000 2–3 10
Li-polymer 100–130 300 3.6 300–500 2–4 10

1
Up to 80% capacity.
2
Loss per month.
3
∼1000 means no more than 1000.
4
> 1000 means no less than 1000.

305
H. Liu et al.
increment of working temperature in the temperature range of 30–40 °C
decreases the calendar life of the battery by almost two months [14]. It
is worth noting that when the cell temperature exceeds the threshold
thermal runaway will be triggered which may in turn lead to a terrible
catastrophe [15]. In addition, the maximum temperature difference
among cells and modules in the battery pack is desired to be less than
5 °C [16]. Safety is the top priority that should be ensured when de-
signing electric vehicles. Therefore, an effective BTMS is required for
the Li-ion based EVs, HEVs and PHEVs to rapidly dissipate the heat
generated inside the battery pack. Under the low temperature scenario,
preheating the battery is also needed to ensure a desired driving per-
formance.
2. Lithium ion battery
In this section, typical structures and operating mechanisms of Li-
ion batteries are briefly introduced. Then the battery thermal issues are
discussed to highlight the effects of working temperature on the battery
performance. What follows is an elaboration about different categories
of battery models including the thermal model, electrochemical model,
equivalent circuit model and multi-physics coupled model. Finally and
more importantly, the state-of-the-art developments about various
BTMSs for battery cooling and heating in recent years are summarized.
2.1. Battery framework
A lithium ion cell is typically comprised of positive electrode, ne-
gative electrode, electrolyte, separator, current collector and a case.
Electrodes are the key components inside Li-ion batteries which de-
termine the capacity and energy density of the battery. A variety of
positive electrode materials have been developed and these materials
can be generally divided into three types according to their structures:
the layered structure, the spinel structure and the olivine structure [17].
The first type materials refer to those lithium transition metal oxides
with layered structure, such as LiCoO2 (LCO) and LiNiO2 (LNO). The
term spinel normally refers to LiMn2O4 (LMO) and the third is the group
of transition metal phosphates containing lithium (e.g. LiFePO4 (LFP)
and LiMnPO4 (LMP)). The characteristics of different positive electrode
materials are summarized in Table 2. Among these materials, LiCoO2
has been the most common choice since the first Li-ion battery was
marketed [18]. Utilizing LiCoO2 as the positive electrode material has
the merits of high capacity and mature preparation process. However, it
has the limitations of toxicity and high material cost as well as poor
thermal stability [19]. For the reason of safety, LiFePO4 which is more
resistant to thermal abuse [20] is regarded as a promising substitute
although its effective capacity is barely satisfactory yet [21]. Unlike the
positive electrodes, the negative electrodes are usually made of carbo-
naceous compounds including coke, graphite and Mesocarbon Mi-
crobeads (MCMB), among which the most commercially popular one is
graphite based carbon [22]. Li4Ti5O12 (LTO) has also been put in
practice as the negative electrode material but it faces the problem of
potential side reaction with the electrolyte. Researchers are making
Table 2
Characteristics of different positive electrode materials.
adapted from [19,24,25]
Positive Electrode LiCoO2 LiNiO2 Li(Ni0.8Co0.15Al0.05)O2
Abbr. LCO LNO NCA
Cell voltage (V) 3.7–3.9 3.6 3.65
Specific energy III III IV
Power II II III
Safety I II II
Lifespan II II III
Cost I III II
Excellent: IV; Good: III; Moderate: II; Bad: I.
306
Energy Conversion and Management 150 (2017) 304–330
great efforts to develop better materials to improve the batteries’ per-
formance. For example, silicon is viewed as an attractive option to re-
place the graphite carbons [23].
Electrolytes of Li-ion batteries are non-aqueous solutions of lithium-
containing salts (e.g. LiPF6) dissolved in solvent of organic liquid
mixtures, such as ethylene carbonate (EC), propylene carbonate (PC),
diethyl carbonate (DEC) or ethyl methyl carbonate (EMC) [26]. The
separator is used to separate the positive and the negative electrodes in
case of internal short circuit [27]. Although the Li ions can pass through
it without any obstruction, the separator is electrical insulator which
means that the electrons are blocked and not allowed to cross over
under normal circumstances. Moreover, when the temperature inside
the cell is excessive the separator will shut the cell down, which serves
as a fuse to prevent the occurrence of thermal runaway [28]. The
function of current collector is to collect the currents generated by
batteries. They are generally made of cuprum in the negative side and
aluminum in the positive side, respectively. The cell case is a seal
container which props up the whole cell and the raw material is usually
steel or aluminum to achieve satisfactory mechanical and thermo-
physical properties.
Two normal configurations of Li-ion cells in electrical vehicles by
winding and stacking up the anode-separator-cathode sandwich layer
are: the cylindrical shape and the prismatic shape. Their typical struc-
tures are shown in Fig. 1. The prismatic design improves the space
utilization and increases the flexibility while the cylindrical cells are
more economical and easy to produce due to the manufacturing ma-
turity even though at the expense of low space efficiency [31].
Automotive manufacturers will choose different battery shapes ac-
cording to their specific demands. What’s more, in practical applica-
tions, a single cell is obviously not adequate to provide enough energy
and power for the vehicle usage. Therefore, numerous cells are con-
nected in series and/or in parallel to construct the battery modules and
then quantities of modules are linked with electric wires to constitute a
battery pack. For instance, Tesla model S employs 7104 cylindrical
18,650 (18 means that the diameter is 18 mm; 65 indicates that the
height is 65 mm; 0 is the code for cylindrical cell) NCA-based Li-ion
cells in 16 modules to form its battery pack with an energy capacity up
to 85 kW h [32]. Each module contains 6 groups of 74 cells wired in
parallel and the 6 groups are then wired in series within the module
[33]. Pictures of batteries inside Tesla model S at module and pack level
are shown in Fig. 1(c) and (d).
2.2. Working principles
The schematic representation of the mechanisms of a LiCoO2/gra-
phite Li-ion battery is shown in Fig. 2. As can be seen, the charge/
discharge process is characterized by the transport of lithium ions and
electrons. During charge, lithium ions are extracted from the positive
electrode particles and flow through the electrolyte and the separator to
the negative side. To keep the electric equilibrium, the same amount of
electrons are released simultaneously at the surface of positive elec-
trode particles. These electrons are then collected by the positive
Li(Ni1-x-yMnxCoy)O2 LiMn2O4 LiFePO4
NMC LMO LFP
3.8–4.0 4.0 3.3
IV II II
II III III
II III IV
II II III
II III III
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

Fig. 1. Schematics and pictures of different battery configurations: (a) cylindrical cell; (b) prismatic cell; (c) module level; (d) pack level [29,30].

3. Adverse effects of temperature

There are two main temperature concerns about Li-ion batteries: (1)
the operating temperature goes beyond the acceptable scope; (2) the
low temperature uniformity leads to a localized degradation which
shortens the battery duration life. To obtain the optimum performance
the operating temperature of Li-ion battery needs to be kept within a
narrow range (15–35 °C). Working in temperatures out of this scale, Li-
ion batteries’ performance, lifespan and safety will all be deteriorated.
However, dreadful ambient environments (i.e. elevated or extremely
cold temperatures) are unavoidable in practical applications. Under
these adverse conditions, the battery will see its performance worsen
significantly due to the improper operating or storage temperature.
Thermal runaway of the battery may even be triggered under some
extreme conditions. In this section, the adverse impacts of improper
temperatures on Li-ion batteries are outlined briefly to elucidate the
significance of the BTMSs. The unexpected thermal behaviors including
the high temperature-caused performance degradation, thermal run-
Fig. 2. Schematic illustration of lithium ion battery [34].
away, temperature maldistribution, and low temperature behavior. As
the mechanisms are beyond the scope of this review, readers may
current collector and travel via the external circuit to the negative consult relevant papers for more details [35–41].
electrode to form the charge current. The lithium ions react with the
electrons and are intercalated into the positive electrode materials ul-
timately. During discharge, the electrochemical process is inverse to 3.1. Performance degradation
what happens during charge. An equal amount of lithium ions and
electrons escape from the negative electrode at the same time and mi- The performance degradation of Li-ion batteries herein refers to the
grate back to the positive electrode through the internal and the ex- capacity fade and power loss under high temperatures. The effects of
ternal passage respectively. low temperature will be discussed as an individual part in Section 3.4.
By using the LiFePO4/graphite as the positive/negative electrode USABC (The United States Advanced Battery Consortium) suggests that
materials, the electrochemical reactions occurring inside the Li-ion a calendar life for at least 10 years is expected for EVs [42]. However,
battery during charge and discharge processes can be expressed as the battery performance degrades rapidly with the increase of tem-
follows. perature and thus few EVs can meet the expectation. Considering the
massive combinations of diverse electrode materials and electrolyte
charge composites, it’s impractical to cover the performance degradation me-
Positive electrode: LiFePO4 ⇄ x Li+ + x e− + Li (1 − x ) FePO4
discharge (1) chanisms of all these electrochemical batteries. Generally speaking,
however, the capacity fade is attributed to the lithium loss and active
charge
material reduction [38] inside the battery while the origin of power
Negative electrode: 6 C+ x Li+ + x e− ⇄ Lix C6 abatement is the increasing cell internal resistance [43] due to the
discharge (2) elevated temperature.
The battery performance degradation can also be classified into the
cycle performance loss and calendar performance loss based on whether

307
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

the battery is being operated [44]. Cycle performance loss occurs when
the battery is cycled (i.e. charged/discharged). Ramadass et al. [45]
experimentally tested the capacity drop of Sony 18650 cells with
LiCoO2 positive electrode and carbon negative electrode by using the
cell voltage range of 2.0–4.2 V. Experiments were conducted under
different temperatures, including the room temperature, 45 °C, 50 °C
and 55 °C. The cells cycled at room temperature and 45 °C showed a
capacity reduction of 22.5% and 26.46% respectively after 500 cycles,
while this value reached as high as 70.56% at 55 °C. In the study of
Shim et al. [35], the cycling performance of LiNi0.8Co0.15Al0.05O2/gra-
phite lithium-ion pouch cells at 100% DOD (depth of discharge) under
the temperature of 25 °C and 60 °C were experimentally compared.
Results also indicated that the capacity loss increased significantly due
to the elevated temperature (4% at 25 °C compared to 65% at 60 °C
after 140 cycles).
The capacity/power fade when the Li-ion batteries are stored
without use is called the calendar performance loss. Wu et al. [46]
reported that the capacity of a Li-ion battery, which was 650 mA h after
being soaked at room temperature, dropped to 20 mA h after 60 °C
storage for 60 days. Jiang and Zhang [47] listed the requirements for
the storage of lithium-ion batteries:
(a) The storing environment should be dry and clean with a good
ventilation and desired ambient temperature
(b) The direct exposure to sunlight should be avoided
(c) The distance from the battery to any heat source should be at least
2m
(d) Being placed upside sown or any mechanical stress is not allowed.
Although the coating positive electrode materials [48] and adopting
proper electrolyte composites [49] can improve the battery perfor-
mance under the elevated temperature, the adverse effects cannot be
totally eliminated. Therefore, a BTMS that can well control the battery
temperature is of great significance.
3.2. Thermal runaway
Thermal runaway is the an undesirable condition when the elevated
battery temperature induces a series of exothermic reactions which in
turn raises the temperature further until irreparable incidents happen
[50]. In Ref. [51], Feng et al. experimentally investigated the thermal
runaway behavior of a large format prismatic Li-ion battery. They
found that if the thermal runaway took place, the internal temperature
of the 25-A h battery would reach as high as 870 °C. As a consequence,
fire and explosion of the battery will be triggered with the generation of
large amount of heat and deleterious gas if no actions is taken. In order
to figure out the thermal stability and abuse tolerance of Li-ion cells, a
variety of abuse tests have been developed including the oven exposure
[52], short-circuit [53], overcharge [54], nail penetration [55] and
crush [56] tests.
The undergoing reactions inside the Li-ion battery (taking LiCoO2 as
positive electrode material, carbon as negative electrode material and
LiPF6/solvent composites as electrolyte) and the relevant possible
equations are summarized in Table 3. Note that these reactions don’t
take place in a given sequence. The occurrence of these reactions varies
with the battery design, the battery chemistry and the state of charge
(SOC) as well as the abuse condition.
A typical path of thermal runaway is given in Ref. [58]. The first
step is the decomposition of the SEI (solid electrolyte interface) layer.
The passivating SEI layer is a thin film covering on the negative elec-
trode which prevents the reaction between the lithiated carbon and the
organic solvent in electrolyte. It contains stable components (e.g.
Li2CO3) and metastable substances (e.g. (CH2OCO2Li)2) [59], the latter
of which exothermically decompose as illustrated by Eq. (3) in Table 3
when the battery temperature rises to about 90 °C. Without the SEI’s
protection, the negative electrode will react with the electrolyte at
100 °C. This is a highly exothermic reaction which generates a large
amount of heat to further raise the temperature. And then the polymer
separator melts down at the temperature around 130 °C which causes a
potential short-circuit between the electrodes. If the temperature grows
to more than 200 °C, a chain of reactions may occur. Firstly, the lithium
metal oxide electrode decomposes and gives off oxygen. Then the sol-
vent in the electrolyte may react with the oxygen (Eq. (6a–c)) or di-
rectly react with the positive electrode materials (Eq. (8a–b)). Due to
the presence of lithium salt (e.g. LiPF6), the electrolyte is likely to
disassemble into the low molecular weight compounds, such as LiF,
C2H4, C2H5F, making the reactions more complicated. All these afore-
mentioned processes generate gaseous products leads to the swelling-up
of batteries during the thermal runaway. These processes also increase
the internal pressure of the battery thus enlarges the possibility of ex-
plosion and catastrophe. Besides, the lithiated carbon is possible to
react with the fluorinated binder. However, this process seldom occurs
because the reaction between the negative electrode and electrolyte is
more prone to take place and runs out the available lithium.

Table 3
Reactions and relevant equations of Li-ion battery during thermal runaway [57].

Reactions Possible chemical equations Temperature range (°C)

SEI decomposition (CH2OCO2Li)2 → Li2CO3 + C2H4 + CO2 +

1
O2 90–120 (3)
2

Negative electrode/electrolyte 2Li + C3H4O3 (EC) → Li2CO3 + C2H4 > 100 (4a)
2Li + C4H6O3 (PC) → Li2CO3 + C3H6 (4b)
2Li + C3H6O3 (DMC) → Li2CO3 + C2H6 (4c)

Separator meltdown / Around 130

Positive electrode decomposition 1 1

LixCoO2 → xLiCoO2 + (1 − x)Co3O4 + (1 − x)O2 196–230 (5a)
3 3
Co3O4 → 3CoO + 0.5O2 CoO → Co + 0.5O2 (5b)

Solvent/O2 2.5O2 + C3H4O3 (EC) → 3CO2 + 2H2O (6a)

4O2 + C4H6O3 (PC) → 4CO2 + 3H2O (6b)
3O2 + C3H6O3 (DMC) → 3CO2 + 3H2O (6c)

Electrolyte decomposition LiPF6 → LiF+PF5 200–300 (7a)

C2H5OCOOC2H5 + PF5 → C2H5OCOOPF4 + HF + C2H4 (7b)
C2H4 + HF → C2H5F (7c)
C2H5OCOOPF4 → PF3O + CO2 + C2H4+HF (7d)

Positive electrode/electrolyte 2Li + 2EC → Li–O–(CH2)4–O–Li + 2CO2 200–240 (8a)

Li–O–(CH2)4–O–Li + PF5 → Li–O–(CH2)4–F + 2LiF + POF3 (8b)

Negative electrode/binder –CH2–CF2 + Li → LiF + –CH=CF– + 0.5H2 > 260 (9)

308
H. Liu et al.
To lessen the potential hazards of thermal runaway, a diversity of
safety mechanisms are utilized in Li-ion batteries such as safety vents,
thermal fuses, self-resetting devices, shutdown separators, chemical
shuttles, coatings and thermal retardant electrolytes and electrodes
[60]. The development of an effective BTMS is also a critical task that
should be focused on.
3.3. Temperature maldistribution
The expectations of adequate driving mileage and fast acceleration
of the battery-based vehicles requires the development of large-scale
battery packs. However, during the charge/discharge process, heat is
produced as a result of the electrochemical reactions and internal re-
sistances. The enlargement of battery systems inevitably brings about
much more heat generation inside the pack especially during the fast
charging/discharging scenario. The heat will accumulate inside the
pack if not dissipated effectively. Besides, the convection coefficient is
higher at the surfaces of the outer batteries thus provides better dis-
sipation conditions than inner cells in the pack. Therefore, there is a
great temperature discrepancy between the batteries at the center and
those near the edges. As we have talked about, the performance of an
individual cell is a strong function of its temperature which means that
there will be a huge capacity variation from cell to cell. Furthermore,
Li-ion cells show low tolerance to the overcharge. Even a slight over-
charge will cause capacity fade [61] and the lithium plating due to high
overcharge may give rise to the occurrence of thermal runaway [62].
Thus, the overall charge capacity of the battery pack is limited by the
cell with the poorest performance [63].
Apart from the uneven temperature distribution among cells, the
temperature imbalance within a cell is an important concern as well.
The heat generation rate is not the identical at different locations inside
a Li-ion cell as the majority of electrochemical reactions occur at the
electrodes. It is reported that a LiCoO2 positive electrode generates four
times as much heat as the overall whole battery [64]. Hence, the
temperature rises are not uniform inside the cell. This will cause the
local deterioration which has an adverse effect on the performance of
the whole cell and may result in the battery failure.
In summary, the temperature non-uniformity both from cell to cell
and within cells will pose negative effects on the overall performance of
the battery pack. Therefore, severe temperature maldistribution should
be avoided and the maximum temperature deviation inside the battery
pack is usually expected to be below 5 °C [65]. The electrode mod-
ification can improve the intra-cell temperature uniformity [66] while a
well-designed BTMS can alleviate the inter-cell temperature in-
homogeneity more effectively [67]. Mohammadian et al. [68] studied
the feasibility of using an internal cooling method which uses the
electrolyte as the coolant to cut down the temperature inhomogeneity
inside the Li-ion cell. Results showed that the cell temperature uni-
formity was improved significantly.
3.4. Low temperature performance
In Section 3.1, it has been indicated that the high temperature leads
to the performance degradation of Li-ion batteries. Many studies have
implied that the cell performance is also poor under a low temperature.
The goal set by FreedomCAR that at least 5 or 7 kW (depending on
mode) of battery power is available in HEV at −30 °C [69] is beyond
the capability of almost all the HEVs in the automotive market. When
the temperature is low, especially below −20 °C, both the energy and
power of the battery are markedly reduced [41,70]. For instance, Na-
gasubramanian [71] compared the characteristics of 18,650 Li-ion cells
at different temperatures from 25 °C to −40 °C. Results showed that the
power and energy densities of the Panasonic cells at −40 °C fell to
1.25% and 5% of those at 25 °C (10 W L−1 and 5 W h L−1 vs.
800 W L−1 and 100 W h L−1).
The exact mechanisms related to the poor performance under the
309
Energy Conversion and Management 150 (2017) 304–330
low temperature is still under debate and need to be studied thor-
oughly. The slower ionic conductivity of the electrolyte at low tem-
peratures was ever thought to be the main cause that worsens the
performance of Li-ion cells [72]. Hence, numerous studies were dedi-
cated to the development of electrolyte mixtures with low freezing
points and high ionic conductivities [73]. The electrolyte mixtures in-
deed improved the low temperature performance at some circum-
stances. However, Zhang et al. [74] found that the increased charge
transfer resistance between the graphite and electrolyte becomes the
dominant factor if the temperature decreases to below −10 °C. In ad-
dition, it is indicated that charging a fully discharged cell is much more
difficult than discharging a fully charged one [75]. It means that the
charge performance degrades more quickly than discharge performance
under cold conditions.
The thermal runaway due to successive exothermal reactions caused
by elevated temperature is not likely to occur at a low temperature.
However, the formation of the lithium plating may arise [76]. The li-
thium dendrites grow on the negative electrode surface and decrease
the charge capacity of the cell as some of the cyclable lithium inter-
calated in the graphite is lost. Furthermore, it may penetrate the se-
parator thus results in internal short-circuits. In Ref. [76], the authors
pointed out that the production of lithium plating can be explained by
the negative polarization of the electrode with respect to Li/Li+ in the
low temperature range.
It is obviously impractical to avoid the adoption of Li-ion batteries
under the lower temperature in light of the demands of consumers
living in the cold environments (e.g. high latitude regions). To alleviate
the impacts of low temperature, batteries could be preheated to achieve
an acceptable performance [77]. Nevertheless, it consumes the addi-
tional energy which declines the overall efficiency of the battery pack.
4. Battery modeling
Since the performance and safety issues of Li-ion batteries are
strongly linked to the temperature, it is of great significance to analyze
the battery thermal behavior (e.g. temperature profile) at cell and pack
level. Despite the existence of other choices, experiments and numerical
thermal simulations are the most prevalent methods. However, the
experimental approach has its own limitations. For instance, one ex-
periment is only for a specific condition making it time-consuming and
laborious to investigate the battery behaviors under different ambient
environments and/or discharge rates. Additionally, the surface tem-
perature cannot accurately indicate the maximum temperature of the
battery due to the difference between the surface and the core. As the
battery case is a sealed container, it is almost impossible to acquire the
temperature data inside the battery using existent thermal detectors
such as thermocouples. Comparatively, numerical simulations can be
easily fitted to different conditions by altering the model parameters.
Moreover, the thermal behavior of any portion in the battery model can
be displayed. Therefore, numerical method is widely adopted by re-
searchers regardless of the slight discrepancy with practical situation
sometimes. It is also frequently used for the cell configuration optimi-
zation and the thermal management system design.
The first step of numerical simulations is the establishment of bat-
tery thermal models which will be discussed below in the first subsec-
tion. This section also reviews the literature about the commonly uti-
lized electrochemical and equivalent circuit models for thermal
analysis. The electrochemical and equivalent circuit models are always
coupled with thermal models to gain a more accurate representation of
the heat generation inside the battery.
4.1. Thermal model
During the charge/discharge process, heat is generated within the
battery. Majority of the heat will be conducted from the internal portion
to the surface and then dissipated by ambient air or other coolant if an
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

effective thermal management system is attached. The remaining en- and irreversible heat according to the production origins [81,82]. The
ergy is stored inside the battery. Thus, the building of battery thermal reversible heat originates from the entropy change associated with
models is mainly about the selection of expressions including the en- electrochemical reactions. Thus, it is also known as reaction heat or
ergy balance equation, heat generation equation and boundary condi- entropic heat. The irreversible heat is comprised by active polarization
tion equation. heat and ohmic heat [83,84]. The polarization is the deviation between
the cell open-circuit potential and the operating potential. At the solid-
4.1.1. Energy balance electrolyte interface, there is a resistance to hinder the charge transfer
The energy balance equation is based on the energy conservation process. The energy that needed to overcome the barrier during the
law and can be expressed as Eq. (1) which gives the temperature dis- lithium intercalation and de-intercalation process is referred to as the
tribution within the battery [78]. active polarization heat [85]. The ohmic heat or joule heat is the energy
loss caused by the transport resistance in solid and electrolyte phases.
∂T ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞
ρCp = λx + λy + ⎜ λz ⎟+ Q̇ Various equations have been introduced so far because of the in-
∂t ∂x ⎝ ∂x ⎠ ∂y ⎝ ∂y ⎠ ∂z ⎝ ∂z ⎠ (1) tricacy of heat generation process inside batteries. Several most com-
The term on the left hand side represents the energy accumulated monly used expressions are reviewed here. A localized heat generation
inside the battery while the right side terms are the three dimensional equation is shown by Eq. (7) which is derived from electrochemical
heat conduction and heat generation rate expressions respectively. analysis. The first term in the right side refers to the polarization heat
For cylindrical batteries, the equation can be reformed in the cy- and the second is the reaction heat. The third term is the joule heat
lindrical coordinates as below: because of resistance to the electronic conduction in the solid phase
while the fourth and fifth terms are the ohmic losses due to the im-
∂T 1 ∂ ⎛ ∂T ⎞ 1 ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞ pedance to the ionic transport in the electrolyte phase.
ρCp = λr r + 2 ⎜λ φ ⎟ + λz + Q̇
∂t r ∂r ⎝ ∂r ⎠ r ∂φ ⎝ ∂φ ⎠ ∂z ⎝ ∂z ⎠ (2)
∂U ⎞
Q̇ = as in (∅s −∅e −UOC ) + as in ⎛T + σ eff (∇∅s )2 + κ eff (∇∅e )2
To note, the overall effective battery thermal capacity ρCp is cal- ⎝ ∂T ⎠
culated based on the volume of each part of the cell given by [79]: 2Rg Tκ eff 0 d lnf ± ⎞
+ (t+−1) ⎜⎛1 + ⎟ ∇lnce · ∇∅e
∑i ρi Cp,i Vi F ⎝ d lnce ⎠ (7)
ρCp =
∑i Vi (3) This equation gives a sophisticated and precise description of the
where Vi denotes the volume of each component. In addition, a battery heat generated inside the battery while its deficiency is that the con-
cell is coined or stacked by multiple components with different thermal taining excessive electrochemical parameters increase the computation
properties. Thus, the thermal conductivity of the battery is anisotropic. cost. Therefore, a simplified form [86] of the formula proposed by
In the in-plane (prismatic) or axial (cylindrical) direction, the thermal Bernardi et al. [87] from thermodynamic perspective is widely used in
conductivity can be calculated by the literature:
λ dUOC
∑i ti Q̇ = I (UOC−V )−I (T )
λ|| = dT (8)
t
∑i (4)
The energy loss due to the charge transfer over-potential at the
In the through-plane (prismatic) or radial (cylindrical) direction, the solid-electrolyte interface, ohmic heat and mass transfer limitations
thermal conductivity is given by constitute the first term on the right side. The latter term shows the
t reversible entropy heat. To note, this equation is only applicable in the
∑i
λ⊥ = ti
case of no phase change or mixing heat release term, no concentration
∑ λi (5) gradients in the electrolyte, no open-circuit potential variation and only
one reaction heat source being considered.
It is also worth emphasizing that the abovementioned energy bal- More recently, Karimi and Li [88] rewrote Eq. (8) into the following
ance equation is under some assumptions for the simplification [78]. format:
For example, the battery thermos-physical properties are regarded as
independent of the temperature which is rational if being operated ΔS
q ̇ = Ri i 2−iT
nF (9)
within a narrow temperature range. Besides, the convective heat
transfer inside the battery is omitted as the flow velocity of electrolytes Based on experimental data, the internal resistance Ri and entropy
is greatly confined. change ΔS of a cylindrical SONY-US18650G3 battery [89] can be es-
The detailed three dimensional simulation occupies massive com- timated using empirical correlations as below:
puting resources especially under the intensive computation conditions.
−6 −0.3952 T = 20 °C
To reduce the complexity and simulation time, the temperature gra- ⎧ 2.258 × 10 SOC
Ri = 1.857 × 10−6SOC −0.2787 T = 30 °C
dients in one or two directions may be ignored without compromise of ⎨ −6 −0.1692 T = 40 °C
simulation accuracy. Beyond that, lumped capacitance thermal models ⎩1.659 × 10 SOC (10)
with confinement of three dimensions are commonly employed as well.
By assuming a uniform temperature distribution in all directions, Eq. ⎧ 99.88SOC −76.67 0 ⩽ SOC ⩽ 0.77
ΔS = 30 0.77 < SOC ⩽ 0.87
(1) can be expressed in the form as follow: ⎨
⎩− 20 70.87 < SOC ⩽ 1 (11)
d (ρCp T )
= hs As (T −Tamb) + Q̇ The SOC is defined as
dt (6)
it
The precondition of adopting the lumped model is that the Biot SOC = 1−
C0 (12)
number is much lower than 0.1 (i.e. Bi = hL/ λ ≪ 0.1) [80]. That is to
say, the requirement that either λ is much higher than the heat transfer The equations reviewed above are the most commonly used ex-
coefficient h or the cell thickness is sufficiently small should be fulfilled. pressions to calculate heat generation rate. The selection of the ap-
propriate calculating methods depends on the practical situations. For
4.1.2. Heat generation instance, if the accurate heat distribution of the cell is of consideration
The heat generation of Li-ion batteries includes the reversible heat and an electrochemical model is linked to the thermal model, the local

310
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

heat generation equation is the optimal choice. The thermodynamic
equation may be used when knowing the operating over-potential and
entropic heat coefficients. In contrast, the empirical formulas are ap-
plicable under the condition when the internal resistance can be ob-
tained.
4.1.3. Boundary conditions
Heat can be transported by conduction, convection and radiation
[90] all of which are involved in the heat dissipation process within the
cell. Heat is firstly conducted through the cell to the outer boundary
surface. Then it is released to the ambient environment or the cooling
media normally by convection and radiation.
For thermal management systems relying on the direct contact be-
tween battery and the fluids coolant (i.e. air or liquid coolant), no
matter passive or active cooling, the heat transfer between the cell
boundary surface and the coolant is mainly by the convective and ra-
diative process as the conduction generally plays a poor role during
normal operations. The boundary condition can be expressed by the
following correlation:
∂T 4 csavg−cssurf =
−λ = hs (T −Tamb) + εσ (T 4−Tamb )
∂n
where the term on the left-hand side denotes the conductive heat from
the cell and n is the direction perpendicular to the boundary surface.
The right-hand side terms refer to the convective and radiative heat
transfer rates respectively.
To simplify the calculation, the radiation heat dissipation rate is
usually adapted into the form of Newton’s law of cooling which is si-
milar to the convective equation. The radiative heat transfer coefficient
can be defined as:
4
εσ (T 4−Tamb )
hrad = ≈ 4εσTamb when T is slightly higher than Tamb
T −Tamb
As a consequence, Eq. (13) can be rewritten into:
∂T
−λ = hs (T −Tamb) + hrad (T −Tamb) = ht (T −Tamb)
∂n
For cooling conditions in which batteries are attached to solid ob-
jects (e.g. heat sink or solid phase change materials), heat is transferred
by the conduction between the cell and the cooling agents. The corre-
sponding relation is given by:
∂T ∂T
−λ = −λ amb
∂x ∂x
positive electrode and carbon negative electrode afterwards. A good
agreement was observed which verifies the effectiveness of the P2D
model.
However, as can be observed from Table 4, there are quantities of
non-linear partial differential equations (PDEs) which makes the si-
mulation occupy too much computational resources. In order to cut
down the computational duration and the simulation complexity, var-
ious mathematical and physical simplification methods have been
proposed. One idea is to integrate a polynomial approximation into the
porous electrode model (PP model). In the PP model, the concentration
in the solid phase is approximated by a parabolic profile while the re-
maining expressions retain the same as the P2D model [94]. As a
consequence, the PDE represented by Eq. (17) is substituted by an or-
dinary differential equation (ODE) represented by Eq. (24) and an al-
gebraic equation represented by Eq. (25), the derivation process of
which is elaborated in Refs. [95,96].
∂csavg
= −3as jn
∂t (24)
as jn Rs
5Ds (25)
(13)
Another method is referred to the single particle (SP) model. This
model rejects the porous electrode theory and considers each electrode
as a single bulk particle whose surface area is equal to the active area of
the solid porous electrode [97]. Besides, the diffusion (i.e. the con-
centration deviation) of Li species in the electrolyte phase is neglected
[98]. As a result, Li+concentrations and the solid phase potential are
only functions of the time t and remain constant along the electrode
thickness direction. A schematic representation of the SP model is
shown in Fig. 3(b).
Santhanagopalan et al. [99] discussed the simulation accuracy and
(14)
(15)
(16)

4.2. Electrochemical model

The electrochemical model accurately predicts the behavior of Li-

ion batteries and is firstly presented by the pioneering work of
Newman’s group [91]. They developed the commonly-used electro-
chemistry based model, namely Pseudo Two-dimensional (P2D) model,
which employs the concentrated solution theory and the porous elec-
trode theory [91,92]. In the model, only one-dimensional transport
from the negative electrode through the separator to the positive
electrode is considered and it is assumed that the active solid phase
material consists of mono-sized spherical particles. A schematic of the
P2D model is displayed in Fig. 3(a). The governing equations of the P2D
model are listed in Table 4 including the mass conversation of lithium
species in solid/electrolyte phases, the charge conversation in solid/
electrolyte phases, the corresponding boundary conditions and the
electrochemical kinetic equations. Lithium conservation equation in
solid phase is solved in the particle r-dimension—the pseudo second
dimension, which is the reason why it is named as P2D model. Doyle
and Newman [93] compared the simulation results of the P2D model Fig. 3. Schematic representation of electrochemical models: (a) the P2D model; (b) the SP
model [101].
and experimental data for a Li-ion cell assembled by manganese oxide

311
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

Table 4 showed a higher accuracy than the SP model at the expense of higher
Governing equations of the P2D model [91,92]. complexity. In 2009, Cai and White [105] proposed a reduced-order
model based on the proper orthogonal decomposition method. The si-
Physical Expressions Eq.
mechanism mulation time was reduced and the results fit well with the P2D model.
The drawback is that the incorporation of empirical data or higher-
Mass conservation of Li+species order simulation results is needed. More recently, Luo et al. [106] ex-
Solid phase (17)
∂cs 1 ∂
− 2
∂t
∂c
r ∂r (
r 2Ds s = 0
∂r ) tended the SP model by considering the non-uniformity of potential/
Boundary ∂c ∂c (17a) concentration distribution in the electrolyte phase. The simulation error
Ds s = 0; −Ds s = jn
conditions ∂r r = 0 ∂r r = Rs was maintained below 0.55% at charge/discharge rates below 4 C.
Electrolyte ∂
εe ce−∇·(Deeff ∇ce )−
i e·∇t+
+ as jn (1−t+0) = 0 (18) In general, the P2D model is by far the most representative elec-
phase ∂t F
trochemical model. However, its complexity restricts its application in
Boundary ∂ce ∂ce (18a)
= =0 most cases. Due to the significant reduction of computing cost, the SP
conditions ∂x x = 0 ∂x x = L
model is the most prevalent simplification method yet among various
Charge conservation
Solid phase (19) simplifying models introduced in this section. The polynomial approx-
is = −σ eff ∇∅s
Boundary ∂∅ ∂ ∅s I (19a) imation technique is usually employed as well to further accelerate the
−σ eff s = −σ eff = ;
conditions ∂x x = 0 ∂x x = L A computation speed. It is expected that a precise and computationally
∂ ∅s ∂ ∅s
= =0 efficient electrochemical model could be developed.
∂x x = Ln ∂x x = Ln + Lsep

Electrolyte 2Rg Tκ eff ∂lnf ± (20)

i e + κ eff ∇∅e − ⎛1 + ⎞ (1−t+0) ∇lnce = 0
phase F ⎝ ∂lnce ⎠ 4.3. Equivalent circuit model
Boundary ∂ ∅e ∂ ∅e (20a)
= =0
conditions ∂x x = 0 ∂x x = L
Different from using the complex partial differential equations to
Butler-Volmer kinetics equation describe the electrochemical mechanisms inside the battery, equivalent
(21)
jn = io ⎡exp
⎣
a s
RT (
−exp −
α Fη
c s
RT
⎤
⎦ ) ( α Fη
) circuit models provide an insight to the cell’s behaviors by means of
α c αa max α α
io = Fka kc (cs −cs ) c ce a (22) equivalent circuits which are constructed with simple electrical com-
ηs = ∅s −∅e −UOC (23) ponents such as voltage sources, resistors, and capacitors [107]. On the
other hand, quantities of parameters and assumptions are required for
the electrochemical models while only a few parameters are needed for
the computation cost of these three models for cycling performance. It the equivalent circuit models [108]. Hence, equivalent circuit models
was shown that due to the employment of simple ODEs instead of PDEs, are also widely adopted.
the calculation speed of the SP model is thousands of times faster than Most of the commonly used equivalent circuit models can be cate-
that of the P2D model to solve the model equations for over 800 cycles. gorized into the Thévenin based and the impedance based [109]. Im-
However, significant discrepancies between the SP model and the P2D/ pedance models usually incorporate a constant phase element (nor-
PP model occurred when the charge/discharge rate was higher than mally Warburg element) whose value is characterized by
1 C, although the simulation results were matched quite well at lower electrochemical impedance spectroscopy (EIS) [110]. Thévenin models
charge/discharge rates. Therefore, the SP model is a better choice than are consisted of a voltage source, a series resistor and resistor-capacitor
the others concerning the calculation speed and accuracy if the charge/ (RC) pairs based on Thévenin’s theorem [111]. Andre et al. [112] tested
discharge rate is less than 1 C. And compared to the P2D model, the PP and compared the Thévenin model and the impedance model. Results
model saves large amount of time without compromising the accuracy. showed that both models gave precise prediction of the battery per-
In addition, the SP model can be further simplified if the solid phase formance while at least a doubled computation speed is achieved by the
diffusion is approximated by a parabolic polynomial which is similar to Thévenin model. Therefore, the Thévenin model is more appropriate
the PP model [100]. and popular in the application of battery performance prediction and
In light of the complexity of the P2D/PP model and the limitations battery management system design. This subsection focuses mainly on
of the SP model, efforts have been made to develop novel simplification the utilization of Thévenin models.
models to predict the battery performance. For instance, a simplified The internal resistance model shown in Fig. 4(a) is the simplest
approach of the P2D model is introduced by Di Domenico et al. [102] in equivalent circuit model. It is comprised by an ideal voltage source UOC
year 2008, which is called the average model (AM) or average electrode connected with a resistor R1 which represents the battery internal
(AE) model. In the AM, the solid concentration distribution and the equivalent resistance [113]. The battery terminal voltage can be ex-
current density inhomogeneity are ignored. Instead, the average values pressed by Eq. (24) where IL > 0 for discharging and IL < 0 for char-
of the concentration and the current density across the solid phase are ging. This model can be integrated into any BMS as its structure is
adopted [103]. The AM is similar to the SP model but the key difference simple. However, the simulation accuracy is usually not adequate.
is that the AM deals with PDEs rather than first order ODEs in the SP
model. The simulation results of the AM, SP model and P2D model were UL = UOC−IL·R1 (24)
compared by Lin and Tang [104]. They demonstrated that the AM One method that can increase the exactitude is integrating the RC

Fig. 4. Schematic of the internal resistance model and the first order
RC equivalent circuit model.

312
H. Liu et al.
branches into the model to simulate the transient battery responses
[109,114]. The number of the RC networks determines the fidelity of
the model. Typically, more RC networks lead to a higher simulation
accuracy while raises the model complexity simultaneously. Therefore,
a trade-off between the accuracy and the complexity should be man-
aged. The schematic of the first order RC equivalent circuit model is
shown in Fig. 4(b). According to Kirchoff’s voltage law, the model
equation can be expressed by Eq. (25).
UL = UOC−IL·R1−URC
Rahimian et al. [115] compared the electrochemical SP model with
the first order equivalent circuit model by using them to predict the cell
voltage. Data indicated that the SP model is superior to the first order
RC model in terms of accuracy while the simulation time is in the same
order. However, this is the case for first order model. Integrating more
RC branches can correspondingly improve the fidelity. Lam et al. [116]
proposed a second order equivalent circuit model which uses two RC
pairs to represent the activation and concentration polarization re-
spectively. No significant error between the simulation results and the
experimental data was observed. On the other hand, the accuracy can
be further improved by considering the dependence of UOC and R on the
SOC and temperature [117,118]. Apart from that, the effect of hyster-
esis is sometimes taken into account, especially for LiFeO4 based li-
thium ion batteries [111,119]. Hu et al. [119] demonstrated their
evaluation of twelve equivalent circuit models used for LiNMC and
LiFePO4 batteries. Their results showed that the first order RC model
achieves sufficient accuracy with relatively low complexity for LiNMC
batteries. However, the effect of hysteresis is recommended to be added
for LiFePO4 batteries.
In conclusion, the equivalent circuit model is commonly adopted for
the battery modeling for the simplicity and adequate accuracy. Various
models have been proposed and verified, among which the Thévenin
model elaborated above is more appropriate in the application of bat-
tery thermal management design.
Fig. 5. Schematics of coupled electrochemical-thermal and electro-chemical models framework in [120,125].
Energy Conversion and Management 150 (2017) 304–330
4.4. Electrochemical/electrical-thermal coupled model
Electrochemical models can be linked with thermal models by the
Arrhenius law (Eq. (26)) which gives the relation between transport
and kinetic parameters Ψ and temperature T . For example, the solid
phase diffusion coefficient Ds follows the correlation of Eq. (27). Song
and Evans [120] developed a model for the thermal management of
lithium polymer batteries by coupling a two-dimensional thermal
model with a one-dimensional electrochemical model. The calculation
(25) framework of the proposed electrochemical-thermal coupled model is
shown in Fig. 5(a). As can be seen from the diagram, the electro-
chemical and physical reactions determine the heat generation rate
term while the temperature derived from the thermal model in turn
influences the electrochemical parameters.
⎡ E (Ψ) ⎛ 1 1 ⎞ ⎤
Ψ = Ψref ·exp ⎢ a ⎜ − ⎟
R ⎝ Tref T ⎠ ⎥ (26)
⎣ ⎦
⎡ E (D ) 1 1 ⎞ ⎤
Ds = Ds,ref ·exp ⎢ a s ⎜⎛ − ⎟
R ⎝ Tref T ⎠ ⎥ (27)
⎣ ⎦
In fact, in spite of the aforementioned simplification methods, the
electrochemical model is still complex for numerical simulations.
Therefore, electrochemical models are always coupled with the simple
lumped thermal model. The P2D model was coupled with a lumped
thermal model by Smith and Wang [121] for the power and thermal
behavior analysis of Li-ion battery pack. Guo et al. [97] constructed a
model that combines the SP model and a lumped thermal model in
which the spatial temperature distribution is ignored. More recently,
Prada et al. [122] coupled the average model with a lumped energy
balance to predict the temperature response of LiFeO4/graphite bat-
teries under fast charging conditions. However, the lumped model ne-
glects the cell internal temperature gradient which may play an im-
portant role in the design of thermal management system. In recent
years, the higher dimensional thermal models becomes the mainstream
313
H. Liu et al.
trend due to the improvement in computation tools. Greco and Jiang
[123] coupled the AM and a two-dimensional thermal model while they
extended the AM by considering the solid and solution phase diffusion
using the nodal network method. Their results showed that the model
could predict the thermal behavior accurately. Basu et al. [124] pro-
posed a three-dimensional electrochemical-thermal model to validate
the effectiveness of their novel BTMS.
Lin et al. [125] built an electro-thermal model for the cylindrical
LiFePO4/graphite battery consisted of an RC-based equivalent circuit
sub-model and a thermal sub-model. The coupling approach is outlined
in Fig. 5(b). The parameters of the equivalent circuit are dependent on
the temperature obtained from the thermal model. At the same time,
the calculation of the heat generation rate is based on the electrical
parameters. The authors validated the model at wide-range SOC and
temperature and a high accuracy was achieved. In Ref. [126], a third
order RC equivalent circuit model is coupled with a one dimensional
thermal model to predict the Li-ion cell’s behavior under different
conditions. A favorable agreement between the modeling and experi-
mental results was achieved.
5. Battery thermal management system
As have been discussed in Section 3, improper temperature would
bring about the unexpected performance degradation and may even
lead to the thermal runaway of Li-ion batteries. Therefore, a BTMS is
required when the battery pack is assembled. The primary aim of a
BTMS is to maintain the batteries at the optimum operating tempera-
ture range with even temperature distribution inside the cell and the
pack. Apart from that, some factors have to be considered for practical
vehicle applications such as compactness, weight, cost, complexity and
reliability of the system [77].
5.1. Battery cooling
A BTMS for cooling could be either passive (only the ambient en-
vironment is used) or active (the cooling media is forced by built-in
devices). According to the heat removal position, BTMSs are categor-
ized into internal or external systems. Currently, most thermal man-
agement strategies are external which means that heat is taken away at
the battery surface. However, in almost all cases, the maximum tem-
perature locates at the internal portion of the cell because of the
thermal resistance between the battery core and the cooling medium.
Raising the convection velocity may further increase the temperature
difference within the cell [78]. The internal cooling technique which
directly removes heat from inside is a promising option to improve the
temperature uniformity [127]. Mohammadian et al. [68] embedded
microchannels into the electrode area of Li-ion cells. The liquid elec-
trolyte which flows in the channels is employed as the coolant to re-
move the heat generated inside the battery. As the heat generation rate
near the electrodes is higher than the other areas [128], it is proved that
the proposed internal cooling method observably promotes the tem-
perature uniformity compared to the external cooling (the standard
Fig. 6. Different designs of inlet and outlet ducts compared in [131].
314
Energy Conversion and Management 150 (2017) 304–330
deviation decreased for more than 5 times at pumping power of
0.024 W). However, the internal cooling is rarely studied and one of the
causes attributes to the closure of batteries making the method difficult
to implement.
In terms of cooling medium, thermal management methods can be
classified into the air cooling, liquid cooling, phase change (liquid to
gas) cooling, phase change (solid to liquid) cooling and the combination
of them. All these types of BTMSs are introduced elaborately in this
section. However, it is unrealistic to review all the relevant BTMS
techniques considering the numerous works in this field. As a result, the
following parts will focus on the progress in BTMSs during the past five
years.
5.1.1. Air cooling
As the most traditional approach, air cooling has been studied in-
tensively and is widely adopted in commercial applications. For ex-
ample, the Honda Insight and Toyota Prius HEVs directly blows the
conditioned cabin air into the battery pack to cool the batteries [129].
Both the natural convection (i.e. passive) and forced convection (i.e.
active) can be utilized as air cooling strategies. Generally, the con-
vective heat transfer coefficient of the natural air cooling is much lower
than that of the forced air cooling. Thus, the natural convection cooling
is only effective for low energy density batteries [130]. It is scarcely
studied now despite its distinct advantages of simplicity, lightweight
and low cost [6]. By comparison, due to the installation of fans/blowers
the heat transfer coefficient of forced air flow is greatly enhanced which
is essential for heat dissipation requirements. In the past few years, the
improvement of active air cooling is primarily focused on three aspects:
geometric structure, parametric optimization and thermal model de-
velopment.
Park [131] compared the effects of geometric configurations with
different types of forced air flow inlet and outlet ducts (shown in Fig. 6)
by simulations using a thermal resistance model. He found that the
maximum temperature was much lower by adopting a tapered manifold
(Fig. 6(e)) than the other air flow shapes. The introduction of a pressure
relief ventilation in the outlet air duct can further decrease the max-
imum temperature and improve the temperature uniformity. In addi-
tion, it can be useful when the battery gives out hazardous gases. The
conclusion is validated by Sun and Dixon’s work in Ref. [132]. In this
paper, the authors developed a transient three-dimensional thermal
model which was used to simulate the thermal performance of a “Z-
type” battery pack. Results showed that the tapered flow ducts with
orifices at the outlet ducts and corrugation between cooling plates
could significantly reduce the cell temperature and improve the uni-
formity. Mohammadian and Zhang [133] proposed a special design of
air-cooled thermal management strategy for prismatic Li-ion cells by
inserting pin fins into the air flow channels. A lower and more uniform
temperature field was achieved. Afterwards, they embedded an alu-
minum porous metal foam into the flow channels to further improve the
Li-ion batteries’ thermal performance [134,135]. The effects of different
parameters including different design cases, pin fin arrangements, dis-
charge rates, inlet air velocity and temperature, porous insert length as
H. Liu et al.
well as permeability and porosity of the metal foam are also in-
vestigated in their studies. Yu et al. [136] developed an original BTMS
which sucks the two directional air flow with independent air channels
and fans. In their design as shown in Fig. 7, the air flow from the bottom
air duct rises up along the height direction, leading to a more uniform
distribution of the cell temperature. The maximum temperature
(33.1 °C vs. 42.3 °C) and the temperature difference (5 °C vs. 10 °C)
were remarkably decreased compared to the regular design. Wang et al.
[137] studied the influences of different cell structures and fan loca-
tions on the battery module thermal performance by numerical simu-
lations. Results indicated that among the investigated structures the
5 × 5 cubic arrangement is the most desired one as a trade-off between
the thermal performance and cost while the 19 cells hexagonal array is
the optimal structure with respect to the space efficiency. Besides, they
found that the best cooling effect was achieved with the fan placed on
the top. Fathabadi [138] proposed a battery pack design by using the
distributed thin air ducts for cooling. An excellent thermal performance
of the battery pack was obtained.
Fan et al. [139] analyzed the impacts of gap spacing and air flow
rate on the cooling performance of a forced air cooling thermal man-
agement system. Different cooling strategies including the one-side vs.
two-side cooling, even vs. uneven gap spacing were also compared. It
was discovered that the two-side cooling with even gap spacing of 3
mm and air flux of 40.8 m3 h−1 was an excellent choice for cooling the
investigated battery module. Employing the uneven gap design im-
proved the temperature uniformity but had limited effects on the
315
Energy Conversion and Management 150 (2017) 304–330
maximum temperature. Yang et al. [140] compared the aligned and
staggered cell structures of cylindrical battery pack. The parametric
optimization of longitudinal and transverse intervals as well as the
cooling air inlet width was investigated. Tong et al. [141] developed a
thermo-electrochemical model to explore the thermal performance of
forced air cooling under different cooling conditions. Factors including
the air inlet velocity, cell arrangement, inter-cell spacing and reversal
frequency were taken into account. A parametric study of axial flow air-
cooled cylindrical battery pack was carried out by Yang et al. [142]. It
was found that raising the radial spacing increased the average tem-
perature but decreased the temperature difference and power con-
sumption at the expense of space utilization. Moreover, a higher air
flow rate led to lower maximum temperature and a better temperature
uniformity.
The model developed in Ref. [132] is composed of a three-dimen-
sional battery pack flow sub-model indicating the flow parameters in
the cooling channels, a one-dimensional battery pack network sub-
model demonstrating the electrical relations and a three-dimensional
battery cell/module thermal sub-model showing the temperature dis-
tribution inside the battery pack. Saw et al. [143] numerically studied
the thermal behavior of the air-cooled battery pack using the CFD
method and derived the exponential expression between Nu number
and Re number. The relation can be used to predict the cooling per-
formance of BTMS. The authors in Ref. [144] built a electrochemical-
thermal model for battery pack using the air cooling approach based on
the porous electrode theory and the concentrated solution theory. An
Fig. 7. Structure of two-directional air flow
channels in [136].
H. Liu et al.
optimum fan power to balance the cost and compactness was re-
commended. Wang et al. [145] introduced an empirical heat source
based three-dimensional model to investigate the thermal behaviors of
the 5 × 5 cubic cell module shown in Ref. [137] under different con-
ditions. The heat generation rate in the thermal model was measured
experimentally under the adiabatic condition. It is found that the forced
air cooling was not necessary if the discharge rate is lower than 3 C with
the ambient temperature no more than 20 °C. For the investigated cell
module, the upper limit of ambient temperature for forced air cooling
was 35 °C. If the ambient temperature was above 35 °C, it made no
sense to cool down the battery pack by air convection because the
power required to decrease the cell temperature to below 40 °C was too
high.
Although the air cooling method has advantages in many aspects
such as the cost, complexity, maintenance and weight, the poor thermal
conductivity of air makes it improper for applications in abuse condi-
tions (e.g. high ambient temperature and/or high discharge rate). In
this case, more effective cooling strategies such as the liquid cooling are
necessary to be implemented.
5.1.2. Liquid cooling
Compared to the natural or forced air cooling, the liquid coolant has
a higher thermal conductivity which make it a more attractive option as
the cooling medium. Liquid BTMSs can be classified into the direct
cooling and indirect cooling. The direct liquid cooling is to submerge
the battery module into the circulated dielectric coolant (e.g. oil) while
the indirect cooling is realized by the fluid flowing through discrete
tubing/jacket around the module or cold plate where the modules are
placed on [77]. Because the heat must pass through the surface of
tubes/plates before being taken away by the coolant, the thermal re-
sistance of indirect cooling is inevitably increased. In contrast, the heat
transfer process in direct liquid cooling is more efficient due to the
direct contact between coolant and batteries. However, the high visc-
osity of dielectric fluids such as mineral oil leads to heavy power con-
sumption when high forced flow rate is needed. Things are different
with the indirect cooling which can employ low-viscosity liquids like
water. Chen et al. [146] compared the thermal performance of four
cooling strategies, namely, forced air cooling, fin cooling, indirect li-
quid cooling and direct liquid cooling. It was reported that the air
cooling method required the highest parasitic power cost, more accu-
rately, 2–3 times higher than the others. Additionally, the indirect li-
quid cooling was found to be a better choice than the direct cooling in
vehicle applications. Therefore, the direct liquid cooling is seldom
studied since the indirect liquid cooling is more practical.
Lan et al. [147] introduced a novel design of BTMS with discrete
Fig. 8. Structure of the cold plate with segmented minichannels in Ref. [152].
316
Energy Conversion and Management 150 (2017) 304–330
aluminum flat tubes twining around the prismatic cells. Each tube
contained a number of minichannels for water flow to pass through so
that the heat transfer efficiency was improved. Results indicated that
both the maximum temperature and the temperature difference were
well controlled with only few parasitic power consumption. In addition
to employing minichannels, Zhang et al. [148] wrapped the aluminum
tube around the prismatic cells with staggered arrangement. More im-
portantly, they inserted a sheet of flexible graphite with high thermal
conductivity between the cell wall and the tubes. Adding the flexible
graphite makes the heat from the cell surface conducted away quickly
even at the locations with poor contact to the flat tubes. As a result, the
temperature uniformity is significantly improved (the maximum tem-
perature difference decreased from 7 °C to 2 °C by using the flexible
graphite).
Another frequently adopted idea is to sandwich the Li-ion cells by
cold plates in which the liquid coolant such as water is circulated inside
the containing minichannels. A water cooling based BTMS for the
prismatic bipolar Li-ion battery pack was investigated numerically by
Tong et al. [149]. Between two cold plates lies a battery module which
contains a series of stacks. By increasing the coolant flow rate or the
plate wall thickness, the average temperature and the temperature
uniformity are ameliorated at the expense of a higher pumping power.
Elevating the number of stacks inside the cold plates will result in worse
thermal performance. It is worth noting that the geometry of the
coolant flow path has a great impact on the cooling effects of liquid cold
plate BTMSs. Therefore, the optimization of the minichannel structure
has been focused on in recent years. Huo et al. [150] analyzed an
aluminum cold plate based BTMS with straight minichannels. Para-
metric analysis about the number of coolant channels, flow direction,
inlet liquid water velocity and ambient temperature at 5 C discharging
rate were carried out. The optimal flow direction for the given BTMS is
to flow from the electrode side to another side. Besides, the higher the
number of channels and the flow rate were, the lower the temperature
rise was. The inlet flow rate had a critical value (5 × 10−4 kg s−1 for
the studied system in the reference) that showed the best cooling effect.
After that, Qian et al. [151] improved the straight design by employing
only one liquid inlet port. Water flows into the entrance and is then
divided into several branches which enters into different minichannels.
It is obvious that the flow rate of the water streams in the central
channels is higher than that of the flows in the outer channels. It helps
to diminish the temperature divergence as a large amount of heat is
generated near the electrode area. The proposed system could control
the maximum temperature below 40 °C for most of the discharge time
even if only two minichannels were used. Jin et al. [152] presented a
novel U-shape cold plate design by adding oblique cuts across the
H. Liu et al.
regular straight channel walls to disturb the boundary layer thus to
promote the heat transfer coefficient. The structure schematic of the
cold plate with segmented minichannels is shown in Fig. 8. Experi-
mental results showed that the cooling performance of the proposed
segmented minichannel structure was notably enhanced compared to
the conventional straight minichannel cold plate. Jarrett and Kim [153]
discussed the geometry optimization of a serpentine-channel cold plate
design by adjusting the channel width and location using a mathema-
tical algorithm. The effects of different boundary conditions on the
optimization were analyzed afterwards [154]. The optimized config-
urations in terms of pressure drop, mean temperature and temperature
uniformity are shown in Fig. 9. As can be seen, the Pfluid and Tave op-
timized geometries are similar while the optimized structure based on
the temperature uniformity has a shape of confined inlet and wider
outlet. In addition, both the heat generation distribution and the fluid
flow rate have impacts on the Tσ optimization. The average temperature
is only sensitive to the variation of heat flux field and the investigated
operating conditions do not affect the structure for Pfluid optimization.
It is noteworthy that the liquid cooling BTMSs either through dis-
crete tubes or by minichannel cold plates are developed for prismatic
Li-ion cells. Indeed, the indirect liquid cooling is more suitable for
battery cells with the prismatic geometry due to the poor contact be-
tween the cylindrical surface and the flat tube/plate walls. Only a few
liquid based BTMSs designed for cylindrical cells are raised. For in-
stance, based on the idea of liquid cold plate, Zhao et al. [155] proposed
a liquid cold cylinder design. Similar to the cold plate, each cylinder
which embraces a battery cell contains a number of minichannels.
Results illustrated that the maximum temperature was maintained at no
more than 40 °C with 4 minichannels in each cold cylinder and with a
flow rate of 1 × 10−3 kg s−1. In Ref. [124] Basu et al. implemented
conduction elements made of the highly conductive metal to envelop
the cylindrical cells. The heat flux is conducted from the cell surface to
the conduction elements and then to the coolant flow channel which is
attached to the conduction elements. The novel design can effectively
control the temperature rise and temperature distribution. However,
the contact resistance still has a great impact on the thermal perfor-
mance of BTMSs.
Other than the advancement of coolant path geometry and battery
pack structure, the promotion of liquid medium thermal property has
also been investigated. Al2O3 nanoparticles were added into the water
to be utilized as the coolant by Huo and Zhao [156]. Simulation results
demonstrated that the nanofluid based BTMS with nanoparticle volume
fraction of 0.04 decreased the battery mean temperature by 7% com-
pared to the pure water based BTMS but the standard deviation may be
increased. Yang et al. [157] studied the feasibility of using the liquid
metal for battery thermal management. The coolant flow is driven by an
electromagnetic pump instead of the conventional mechanical pump. It
was found that the liquid metal coolant achieved a better cooling per-
formance than water with less parasitic power cost, making it a pro-
mising alternative for the battery thermal management under abuse
Fig. 9. Optimization design of flow channels inside the cold plate [153] (The channels are in black color).
317
Energy Conversion and Management 150 (2017) 304–330
conditions. Zhao and Zhang et al. [158,159] proposed a hydrogel based
BTMS for the Li-ion batteries. Both experimental and simulation results
manifested its effectiveness in dissipating the battery heat generation
under abnormal conditions.
The example of a liquid BTMS in market employed by Tesla is
shown in Fig. 10. The cooling tube made of aluminum coated with
dielectric material is bent into a predetermined specific shape to wrap
the 18650 form cylindrical cells within each sheet. Liquid coolant of
water-glycol mixtures flows into the four channels of the cooling tube.
The addition of glycol is to lower the freezing point of water in case of
coolant freezing under cold environment. It should be noted that a
counter flow design is utilized within the cooling tube to achieve a
uniform temperature distribution between batteries (Fig. 10(d)).
In conclusion, the liquid cooling is more effective for battery
thermal management although it may add more weight and increase
the structure complexity compared to the air cooling. Besides, the po-
tential problem of coolant leakage should be considered. Therefore, the
liquid cooling is promising in light of the increasing demands for fast
charging and the utilization of lithium ion batteries in abuse conditions.
5.1.3. Phase change (liquid to gas) cooling
Active air and liquid thermal management methods transfer the
generated heat in a convective manner. The circulation of coolants
requires the installation of ducts, valves, fans or pumps which causes
extra weight and space consumption. It also increases the parasitic
power costs so that the cooling efficiency is decreased. Therefore,
passive cooling methods utilizing latent heat during the phase change
process of substances are massively investigated during the past years.
It is well known that matters exist in three phases: solid, liquid and
gas. When a substance is transformed from solid state to liquid state or
from liquid state to gas state, a large amount of energy will be absorbed
and vice versa. This phenomenon can be used to enhance the heat
dissipation capacity to remove the heat inside the battery. This section
concentrates on the passive BTMSs using the latent heat of liquid-vapor
phase change process, including the boiling systems and heat pipes.
In fact, the boiling thermal management can be viewed as a type of
passive liquid cooling method. However, the phase change is involved
during the heat transfer process. Therefore, it is categorized into the
phase change (liquid to gas) cooling in this paper. In 2013, Bandhauer
and Garimella [127] creatively proposed an internal BTMS by using the
phase-change fluid in the microchannels. The generated heat is trans-
ported to the microchannel evaporators which are integrated inside the
current collectors to transform the working fluid R134a into vapor
phase. The vaporized fluid flows up to the condenser due to the energy
density difference and is then condensed there. Subsequently, van Gils
et al. [161] introduced their research on BTMS for Li-ion cells using
boiling process. The battery is immersed into the dielectric fluid,
Novec7000 whose dielectric property and cooling ability were experi-
mentally tested in their study. The authors also investigated the impact
of pressure regulation in the boiling container on the intensity of
H. Liu et al.
Fig. 10. Schematic of Liquid BTMS used in Tesla: (a) the cooling tube inter-engaged with a sheet of cells; (b) the cooling tube setup; (c) the cooling tube; (d) the side view and end view of
the inlet and outlet [160].
boiling process which in turn affects the performance of thermal
management. Hirano et al. [162] developed a direct boiling cooling
system for prismatic lithium ion batteries. In their experiments, Li-ion
cells are sandwiched by porous materials, micro fibers or plastics, for
hydrofluoroether liquid and vapor fluid to flow through. Results im-
plied that the maximum temperature can be maintained at approxi-
mately 35 °C which is within the desired operating range even with a
charged/discharged rate of 20 C. Instead of using porous materials, An
et al. [163] recently took advantage of the flow boiling process in
minichannels to cool a battery module. They also chose the hydro-
fluoroether whose boiling temperature is 34 °C as the phase change
fluid. The battery temperature was kept around 40 °C with temperature
difference below 4 °C due to the heat transport enhancement brought
by the fluid boiling process.
As can be seen in Fig. 11(a), a typical capillary heat pipe consists of
two components, i.e., working fluid and a container with wick structure
on the inner surface. The wick structures adopted in heat pipes could be
sintered metal powder, screen, grooved or mesh design [164]. A heat
pipe also can be divided into three regions: evaporator section, adia-
batic section and condenser section. The working principle is as follows.
The heat flux is firstly transported from the container into the eva-
porator to vaporize the working fluid into the vapor phase. Then the
vapor spontaneously travels to condenser where it is cooled and
318
Energy Conversion and Management 150 (2017) 304–330
releases the latent heat owing to the pressure difference in the pipe. The
vapor is then transformed into the liquid state again which moves back
to the evaporator section through the wick structure under the effect of
capillary force. The advantages of heat pipe cooling systems, including
the lightweight, low cost, no need for maintenance, flexibility and
especially high thermal conductivity, appeal to extensive application
interests for them. Heat pipes have been widely used in electronics
[165], spacecraft [166], solar energy collectors [167] and nuclear
power plants [168]. However, the heat pipe based thermal management
system has not been fully investigated. Rao et al. [169] experimentally
studied the feasibility of using heat pipe for the battery thermal man-
agement in electric vehicle. Four copper heat pipes are placed between
two battery-sized prismatic heaters and the condenser section is cooled
by a water bath. It is found that the battery temperature can be
maintained below 50 °C for the given system if the input thermal power
is no more than 50 W. To keep the temperature difference lower than
5 °C, the maximum heat generation rate is 30 W.
The oscillating (also named as pulsating) heat pipe (OHP or PHP),
appearing as a new type of heat pipe, also attracts many attentions on
its application in the thermal management. Different from conventional
heat pipes, OHP is easier to produce as the wick structure is not needed
but its working principle is more complex and it is sensitive to gravity
[170]. According to the layouts, OHPs are classified into three types,
H. Liu et al.
including looped, looped with check valve and unlooped, as shown in
Fig. 11(b) –(d). Burban et al. [171] investigated the performance of an
unlooped PHP for the hybrid vehicle application. More recently, Rao
et al. [172] proposed a looped PHP based BTMS for prismatic power
batteries. The influences of different operating conditions such as
working fluids and orientations on the thermal performance were tested
in these two papers.
As one of the key components, the option of working fluid has a
great impact on the cooling effect of heat pipes. Different working
fluids, i.e. distilled water, alcohol and acetone, were utilized and
compared experimentally by Putra et al. [173] for a BTMS using the flat
plate loop heat pipe. Results showed that employing the acetone as
working fluid achieved better thermal performance than the other
choices when the input heat load is 1.61 W cm−2. Burban et al. [171]
tested the performance of acetone, methanol, water and n-pentane
under different input heat powers (25–550 W). The acetone and n-
pentane are recommended for low heat generation rate and low am-
bient temperature while the water and methanol are better choices
under abuse conditions (in other words, high thermal power load and/
319
Energy Conversion and Management 150 (2017) 304–330
Fig. 11. Working structures of (a) capillary heat pipe
and different PHPs: (b) looped; (c) looped with check
valve; (d) unlooped [164,170].
or high ambient temperature). Besides, R134a is found to be improper
for the OHP operation.
The cooling method of the condenser end is another important
factor that affects the performance of heat pipe BTMSs. Zhao et al.
[174] introduced a wet cooling method for a flat heat pipe BTMS by
adding water spray at the condenser sections. The authors experimen-
tally compared the water spray method with the forced air and ther-
mostat cooling and results proved the superiority of the proposed wet
cooling idea over the other strategies. Ye et al. [175] developed a delay
quench method in which there exists a short delay between the moment
when the coolant quenches the battery with higher temperature and the
time when the battery starts charging. Besides, cooling fins were fitted
and soldered to the cooling end to further enhance the heat transfer
coefficient. The battery temperature was kept within the optimum
range of 25–40 °C. It is also found that water and methanol are ap-
propriate working fluids for heat pipe BTMSs while ethanol is not.
Apart from working fluids and cooling methods, the influence of the
heat pipe inclined positions has been taken into consideration. Tran
et al. [176] placed the flat capillary heat pipe into different orientations
H. Liu et al.
and found that the effect of inclination angle on the thermal perfor-
mance is negligible. Similar conclusion was drawn in their another
study [177]. However, Burban [171] discovered that a better perfor-
mance for OHP based BTMS was achieved at the situation where the
condenser was placed above the evaporator. The main reason may be
that the key acting force of wick-structured heat pipes is the capillary
force while OHPs rely on the gravity. In addition, heat pipes should stay
in good contact with the cell surface to remove the heat of battery cells
effectively. For tubular heat pipes, however, a large contact area with
the battery is difficult to be ensured. Therefore, evaporator sections of
tubular heat pipes are generally flattened and then soldered or attached
to a rectangular plate to enhance the contact [171,175,177,178]. The
thermal grease is always utilized to further reduce the contact re-
sistance.
As the mechanism of heat transfer and two phase flow process in
heat pipes are complicated and difficult to model, studies of heat pipe
BTMSs are primarily experiment based. An effective method to nu-
merically analyze BTMSs using the heat pipe is needed so that the
system design can be optimized more conveniently. Greco et al. [179]
introduced a thermal circuit model for the capillary heat pipe by re-
placing the heat pipe using an analytical thermal network with thermal
resistances but it lacked the validation from the experimental data. Ye
et al. [180] simplified the heat pipe structure by assuming that the wick
was saturated within the liquid and neglecting the boiling process so
that only the thermal conduction through the three sections was con-
sidered. A good agreement was achieved between the experimental
data and transient simulation results.
To summarize, the development of phase change (liquid to gas)
cooling based BTMS is worth investigating. Due to the effect of latent
heat, the liquid-vapor cooling can obtain higher heat dissipation coef-
ficients compared to the single phase liquid convection cooling. The
boiling heat transfer method is a promising alternative but its practi-
cality has to be further improved. As for heat pipe based BTMSs, the
development directions could be to further improve the heat transfer
efficiency (e.g. adding nanoparticles into the working fluid) and extend
its application into real battery packs. Furthermore, the development of
an effective and simple numerical model is of great importance as well.
5.1.4. Phase change (solid to liquid) cooling
Apart from the liquid-vapor cooling system, the phase change pro-
cess from solid state to liquid state has also been adopted as a thermal
management strategy. Since the boiling and heat pipe cooling are
classified into the passive liquid thermal management by many re-
searchers, we refer to solid-liquid BTMSs normally when we talk about
the phase change material (PCM) cooling method. For convenience,
PCM is used to represent the materials that are transformed into liquid
phase from solid phase in the following parts.
As a passive cooling method, PCM thermal management systems
possess many merits including simplicity, lightness, high efficiency and
no need for extra components such as blowers. Compared to liquid-
vapor systems, solid-liquid phase change systems are more compact and
Table 5
Characteristics of selected paraffin waxes with proper melting temperature.
adapted from [187]
Material Melting point (°C) Latent heat Thermal conductivity
(kJ kg−1) (W m−1 K−1)
Hexadecane 18 236 0.21
Heptadecane 22 213 0.21
Octadecane 28 244 0.21
Nonadecane 32 222 0.21
Eicosane 37 246 0.21
Docosane 44 249 0.21
Tricosane 48 232 0.21
Tetracosane 51 255 0.21
320
Energy Conversion and Management 150 (2017) 304–330
practical which make them more attractive in vehicular applications.
Therefore, the PCM cooling has received extensive attentions during the
past few years.
Obviously, the first step of building a PCM system is to select a
phase change substance. Rao and Wang listed six requirements for
choosing an appropriate PCM for the thermal management as follows
[181]:
(a) The melting point within the desirable temperature scope;
(b) High latent heat, specific heat capacity and thermal conductivity;
(c) Low volume dilatation after phase change process;
(d) Negligible sub-cooling effect when freezing;
(e) Properties including stable, nontoxic, nonflammable and non-
explosive;
(f) Commercially economical.
Before the phase change point, heat is transported by the solid PCM
in a conduction manner. Once the temperature reaches the transition
point, the PCM melts and absorbs the heat flux from the battery.
Therefore, the phase change temperature is one of the key factors. Yan
et al. [182] numerically investigated a composite board based BTMS
which is comprised of the PCM, a heat conducting plate and an in-
sulation panel. It is reported that the desired phase transition tem-
perature range for Li-ion battery thermal management is 30–50 °C. And
results also implied that the proposed BTMS had good thermal perfor-
mance and the ability to prevent thermal runaway propagation under
abuse conditions. Paraffin wax is the most widely used PCM for battery
thermal management due to its properties of no corrosion effect, low
cost and availability in large quantities. Besides, it has a broad melting
range with different length of alkane chains (shown in Table 5) making
it proper for various cooling conditions. It is reported that a broader
phase change temperature range and a higher latent heat of fusion can
be obtained by mixing different paraffin [183]. Thus, paraffin mixtures
have been widely used in BTMSs as well [184–186].
Despite the advantages mentioned above, one of the biggest chal-
lenges for PCMs is their poor thermal conductivities (normally less than
0.4 W m−1 K−1) which determine the thermal transport efficiency. It
limits the PCMs’ applicability in conditions where the fast response to
thermal surge is required. To address this problem, numerous methods
have been implemented to enhance the overall thermal conductivity in
recent years. These techniques can be divided into three aspects: (1)
suspending thermally conductive fillers, such as carbon fibers
[185,186], graphene [188], nanoparticles [184] and carbon nanotubes
(CNT) [189] into PCMs; (2) embedding PCMs into porous media, in-
cluding polyurethane foam [190], metal foams [191–195] and ex-
panded graphite (EG) matrix [123,196–199]; (3) attaching metal me-
shes [200,201], graphite sheet [202] or cooling fins [123,199].
The thermal conductivity enhancement brought by adding carbon
fibers into paraffin mixtures and its improvement on the thermal per-
formance over the air and pure PCM cooling were studied by Samimi
et al. [185]. Results indicated that the thermal conductivity was greatly
Density (solid) Density (liquid) Specific heat (kJ kg−1 K−1)
(kg m−3) (kg m−3)
770 2.2
778 2.2
814 774 1.90
782 1.91
778 1.92
791 1.93
764 1.93
796 1.93
H. Liu et al.
enhanced by carbon fibers (105% average enhancement). Babapoor
et al. [186] also investigated the thermal performance of PCM-carbon
fiber composited based BTMS and experimentally tested the effect of
mass fraction and length of the carbon fiber. It is found that the addi-
tion of 0.46 wt% carbon fibers with length of 2 mm achieved the best
thermal performance. The thermal conductivities of carbon materials
including graphene and CNT could reach as high as over 3000 W m−1
K−1 [203] which is ten thousands higher than that of paraffin waxes.
Taking graphene and CNTs as additives is expected to improve the
thermal conductivity. Goli et al. [188] suspended graphene fillers into
IGI-1260 paraffin wax by magnetic stirring and the highest thermal
conductivity of the PCM composite at 20 wt% loading was found to be
about 200 times the value of pure paraffin (45 vs. 0.25 W m−1 K−1).
Nanoparticle is another kind of commonly used additive to enhance the
thermal conductivity. Ag, Cu and Fe3O4 nanoparticles as well as the
metal matrix were dispersed into a PCM respectively by Karimi et al.
[184] to work as BTMSs for cylindrical batteries. SEM images of hybrid
PCM composites with the addition of different metal nanoparticles are
shown in Fig. 12. Experimental results demonstrated that all the in-
vestigated nanoparticles could improve the thermal conductivity com-
pared to the pure paraffin system while Ag nanoparticles realized the
lowest battery bulk temperature among the three used nanoscale ma-
terials.
In addition to adding fillers, saturating PCMs into porous media has
been paid more attention during the past few years. It is a more pro-
mising way as metal foams and EGs not only increase the thermal
conductivity but also enhance the mechanical and thermos-physical
properties by acting as skeletons. Qu’s group compared the cooling
effect of copper foam-paraffin composite BTMS with that of air cooling
and pure PCM systems [193]. They developed a two dimensional
transient model for the analysis of battery thermal management [194].
It was reported that an acceptable battery temperature with uniform
distribution was obtained by this foam-paraffin system. Besides, it was
found that the surface temperature grew if the porosity and/or pore
density were raised. Aluminum foams also can be added into PCM
Fig. 12. SEM images of (a) pure paraffin and PCM composites with different metal nanoparticles: (b) Fe3O4; (c) Ag; (d) Cu [184].
321
Energy Conversion and Management 150 (2017) 304–330
systems [191,192]. Results proved the enhancement of PCM on the heat
storage properties and thermal conductivity by adding aluminum
foams. The thermal conductivity of the PCM increased from
0.29 W m−1 K−1 to 46.12 W m−1 K−1 after using the aluminum foam-
paraffin composite. Hussain et al. [195] conducted an experimental
study to investigate the thermal performance of a BTMS by in-
corporating the PCM into nickel foam. Reductions of 31% and 24%
were observed in the battery surface temperature after employing the
nickel foam-paraffin composite compared to the natural air and pure
PCM cooling systems.
Apart from enhancing the thermal conductivity, another benefit of
utilizing EG is that the capillary forces inside can prevent the PCM
composites from leaking after the paraffin melts, which makes it by far
the most attractive alternative for BTMS application. Jiang et al. [196]
incorporated the RT44HC paraffin into the EG to enhance the thermal
management of Li-ion batteries and studied the influence of EG mass
fraction. An EG loading fraction of 30 wt% is found to raise the PCM
thermal conductivity to 13.85 W m−1 K−1 which is 69 times higher
than that of the pure PCM. At the same time, the leakage ratio was
significantly decreased to 0.38 wt%. Ling et al. [197] introduced a
BTMS based on the paraffin/EG for cylindrical batteries and in-
vestigated the effects of different thermos-properties on the thermal
performance. The composite with phase transition point between 40 °C
and 45 °C was revealed to show the best performance for the tested
system. Additionally, the growth of composite density led to a higher
thermal conductivity and latent heat simultaneously. Yan et al. [198]
tested the paraffin/EG composite BTMS in dynamic cycling conditions
and recommended 45 °C as the optimal melting point for battery system
usage. Lv et al. [199] integrated the low-density polyethylene into the
usual EG/PA composite PCM. The LDPE/EG/PA showed better me-
chanical characteristics in terms of bending strength, impact strength
and Shore hardness.
In addition to the aforementioned techniques, the metal mesh en-
hanced PCM composites have also been studied by some groups. The
copper mesh was sandwiched using two PA/EG plates by Wu et al.
H. Liu et al.
[200] to constitute the copper mesh-enhanced PCM composite (shown
in Fig. 13) for the prismatic battery thermal management. The novel
copper mesh enhanced system performed better in cooling effects than
the conventional paraffin/EG composites as the maximum cell tem-
perature was reduced by 4 °C. Azizi and Sadrameli [201] discussed the
utilization of aluminum mesh plates for the PCM based LiFePO4 battery
BTMS at high temperature levels (50–55 °C). It is indicated that six
hours’ desired working condition could be maintained by the as-de-
veloped composite BTMS. Graphite sheets with pretty high in-plane
thermal conductivities and the EG were employed by Lin et al. [202] to
decrease the cell temperature difference. A temperature variation below
5 °C was attained even at the discharge rate of 2 C. Greco and Jiang
[123] introduced an electrochemical coupled thermal model to char-
acterize the thermal performance of the novel PCM/EG composite
BTMS with twelve fins. The addition of fins remarkably lowered the
maximum battery temperature (67.55 °C vs. 75.57 °C at 8 C discharge
rate).
Another potential problem that should be taken into account is the
consequence after the PCMs completely melt. As mentioned above,
phase change cooling methods take advantage of the latent heat during
the phase change process to absorb heat. If the heat loads of Li-ion
batteries are too high for the PCM to dissipate the absorbed heat timely,
energy will accumulate in the PCMs. When all the solid PCMs transform
into the liquid state, a higher thermal resistance may occur between
cells and the ambient environment. In this case, the cooling perfor-
mance will become even worse than the direct air cooling. Therefore,
other cooling techniques such as the forced air cooling and liquid
cooling are always incorporated to the PCM systems to solidify PCMs
and recover its latent heat. Ling et al. [205] introduced the forced air
method to improve the cooling effect of a PCM thermal management
system for the Li-ion battery pack. It was reported that the natural air
convection was not sufficient to remove the heat accumulated in the
PCM while the forced air cooling strengthened the safety and reliability
of the PCM system. Besides, experimental results showed that the al-
teration of air velocity had negligible effect on the battery temperature
distribution which means that the function of the forced air flow is only
322
Energy Conversion and Management 150 (2017) 304–330
to restore the latent heat of PCM. A hybrid BTMS that integrates the
PCM system with a forced liquid cooling system was proposed by
Hémery et al. [206]. The Li-ion cells and PCM are placed into aluminum
cans which are in contact with the liquid cooling plates with circulated
coolants inside to solidify the PCM. Afterwards, Rao et al. [207] cou-
pled the PCM with minichannels for the thermal management of pris-
matic batteries and investigated the performance of the as-proposed
system numerically. A maximum temperature of 320.6 K was predicted
for the PCM/minichannel coupled BTMS, while the maximum tem-
perature was 335.4 K for the pure PCM system. The liquid mass flow
rate and channel numbers both had impacts on the cooling effects.
Additionally, the feasibility of coupling PCM systems with heat pipes
including heat pipe with fins [208] and oscillating heat pipes [209,210]
has been studied experimentally and numerically. In [208], the tem-
perature difference of PCM/heat pipe BTMS was 28.9% lower than that
of pure PCM system. The duration of keeping the system less than 50 °C
was longer due to the integration of heat pipe with circular fins. More
recently, Wu et al. [204] also analyzed the thermal performance of the
PCM when it was coupled with the heat pipe for the battery pack. Heat
pipes were flattened and sandwiched by PCM plates which were then
inserted between Li-ion cells as shown in Fig. 14. The cell temperature
could be maintained below 50 °C even at the 5 C discharge rate. With
regard to the OCP assisted PCM systems, the response time of the PCM
to heat flux was significantly declined [210]. It was also found that the
start-up point of OHP should be lower than the phase change tem-
perature of PCM to improve uniformity of the battery temperature
distribution [209].
To summarize, the advancement of PCM based BTMSs has been
brought into focus and extensive related studies have been conducted
during the past several years. The largest challenge is to enhance its
thermal conductivity while the problem of latent heat recovery should
also be considered. In addition, the integration of PCM system in-
evitably adds the space requirement, weight and cost. Recent years’
researches on the development of the PCM based BTMSs are summar-
ized in Table 6.
Fig. 13. Schematic of copper mesh-enhanced
PCM: (a) fabrication scheme; (b) real photo [200].
H. Liu et al.
Fig. 14. Schematic of the PCM-heat pipe combined thermal management system in [204].
5.2. Battery heating
Despite that battery cooling has attracted much attention during the
past years, seldom literature about battery heating methods can be
found. The reason may be that a catastrophic thermal runaway is
overwhelmingly probable to be caused by high temperature at extreme
conditions while sub-zero temperature can only lead to performance
degradation at most of the time. However, as indicated in Section 3.4,
the Li-ion battery performance is significantly reduced under sub-zero
temperature. With the popularization of battery based vehicles espe-
cially pure EVs, the capability of vehicles is more dependent on the
performance of power batteries. Thus, the problem of poor battery
performance in cold environment is required to be addressed. In other
words, the batteries must be heated in an efficient way before being
cycled. The main criteria for selection of appropriated heating system
are listed below:
(a) The heating time;
(b) The power consumption for heating;
(c) The overall cost including the cost for system, energy expense and
maintenance;
(d) The complexity addition, such as the extra devices, weight and
space needed due to the integration of heating system.
Vlahinos and Pesaran [211] compared the heating effect and the
input power of several battery heating methods including core heating,
electric heating at the module and cell level as well as air heating to
find the most energy efficient battery heating approach. The core
heating method is to directly warm up the battery core by electric
heating while air heating takes advantage of warm air flowing in the
gaps between adjacent battery cells to transfer heat energy. The electric
heating at the module and cell level is to place electric heaters around
the module or each cells respectively. It was found that core heating is
the most efficient technique.
Ji and Wang [212] distinguished different battery heating techni-
ques according to their energy origins as shown in Fig. 15. Battery
heating approaches using internal battery power include self-internal
heating, convective heating and mutual pulse heating. Methods taking
advantage of external power supply are external convective heating and
AC heating. As we have stated in Section 3.4, the battery internal re-
sistance is greatly increased at low temperatures. Therefore, the cor-
responding heat generation becomes higher which is the only heating
source of the self-internal heating. Convective heating by battery power
supplies the power from the battery itself to the electric heater and the
fan. Air is blew in by the fan and then heated by the electric heater to
323
Energy Conversion and Management 150 (2017) 304–330
warm the battery in convective way as Fig. 15(b). The mutual pulse
heating divided the battery pack into two groups with same capacity.
One group is discharged to provide power for charging the other group.
The charge-discharge cycles are conducted alternately in the two bat-
tery groups. In this case, shorter heating time and higher energy effi-
ciency than self-internal heating as well as more uniform temperature
and higher reliability than internal convective heating are achieved.
However, special control system and circuit are required which increase
the system complexity and cost.
If the heating power is available from external resources, the con-
sumption of the battery power can be avoided thus extending the
driving range. The external convective heating only substitute the en-
ergy origin into external supply compared to the convective heating
using battery power as represented in Fig. 15(d). Stuart and Hande
[213] pioneered the method to heat the batteries utilizing alternating
current (AC) shown in Fig. 15(e) and studied its feasibility for lead acid
and nickel metal hydride batteries in 2004. They found that the heating
speed increase with the growth of the amplitude of the currents. It was
reported by Jin and Wang [212] that raising the signal frequency can
also decline the heating time. Recently, Zhu et al. [214] found that
apart from the current amplitude and frequency, the waveform affects
the heating efficiency as well. For instance, the current with sinusoidal
waveform warms up the battery from -24 °C to 7.79 °C while it is heated
from -24 °C to 25.6 °C for the rectangular pulse waveform in the same
condition in their investigation.
Wang et al. [215,216] designed an ‘all climate Li-ion battery’
structure that can self-heat at cold environments which is demonstrated
in Fig. 16. A nickel foil with two tabs was inserted in the cell in addition
to conventional electrodes and electrolyte. One tab is electrically con-
nected to the anode terminal while the other one forms the activation
terminal. When the battery temperature is low, the switch between the
activation terminal and negative terminal is open. In that case, current
is produced in the metal foil which will generate heat inside the cell due
to ohmic loss to warm up internally. Once the heating process is
achieved or in normal conditions, the switch is kept closed.
In addition, phase change materials are promising to perform well
in the environment where the ambient temperature experiences a
sudden drop due to its benefits of heat storage ability [217]. The stored
energy in PCM can be transferred to batteries if the temperature is
below the PCM melting point. Huo and Rao [218] found that a lower
thermal conductivity, greater latent heat of PCMs and higher ambient
temperature contribute to keeping the battery temperature. However, it
was also found that increasing the latent heat of PCM reduces the
battery temperature uniformity, which decreases the cycle life of bat-
tery.
 H. Liu et al.

Table 6
Summarization of developments on PCM based BTMSs in recent years.

Authors Material Cooling system Mass ratio of fillers Thermal conductivity Latent heat Melting point (°C) Solid density Specific heat capacity Research method
(wt.%) (W m−1 K−1) (kJ kg−1) (kg m−3) (kJ kg−1 K−1)

Pure PCM Composite Pure PCM Composite

Yan [182] PCM/heat conducting 0.358 225 30 814 2.15 Simulation

shell/insulation panel
Karimi [184] Paraffin mixture PCM/nanoparticles 2 0.22 239 39–45 911 Experiment
Samimi [185] Paraffin mixture PCM/carbon fibers 32–69 0.21 0.39–0.54 242 42–49 912 Simulation
Babapoor Paraffin mixture PCM/carbon fibers 32–69 0.21 242 42–49 912 Experiment
[186]
Goli [187] Paraffin wax (IGI- PCM/graphene 0.5–20 0.25 14–45 2 2–5.5 Experiment/
1260) Simulation
Shirazi [188] Paraffin PCM/CNTs 1–5* 0.25 0.8–2.4 900 2.5 Simulation
Javani [189] n-Octadecane PCM/polyurethane foam 0.358 814 2.15 Simulation
Wang [190] Paraffin PCM/Al foam 0.29 46.12 46–52 Experiment
Alipanah Octadecane PCM/Al foam 0.358 6.56–25.15 770 1.91 1.69–1.76 Simulation
[191] (kJ m−3 K−1)
Li [192] Paraffin (RT44HC) PCM/copper foam 0.2 0.8–11.33 270.7 42.76–49.24 810 2.25 Experiment
Qu [193] Paraffin PCM/copper foam 0.3 866 1.77 Simulation
Hussain [194] Paraffin PCM/Nickle foam 0.2 11.6 38–41 880 2 Experiment
Jiang [195] Paraffin (RT44HC) PCM/EG 3–30 0.25 275/266.8–192.5^ 41–46 2.000 0.963–1.631 Experiment/
Simulation

324
^
Ling [196] RT36/RT44/RT52 PCM/EG 75,85 5.3–11.2 50.3–73.4 36–52 700–1040 2.5,2.9 Experiment/
Simulation
Yan [197] RT36/RT45/RT58 PCM/EG 0.16–0.37 21.4–163.3 36–58 Experiment
Greco [123] n-Heptadecane PCM/EG/fins 214 21.9 775 2226.25 Simulation
Lv [198] Paraffin PCM/EG/LDPE/fins 7(EG), 30(LDPE) 0.16 1.38 182/87.4^ 44.0–50.2 910/856^ 2.68 2.48 Experiment
Wu [199] Paraffin PCM/EG/copper mesh PA:EG = 4:1 0.26 7.65 186.4/141.6^ 42 Experiment
Azizi [200] PEG1000 PCM/aluminum mesh 0.23 35–40 1092.7 2.142 Experiment/
Simulation
^
Lin [201] Paraffin PCM/EG/graphite sheet 24 0.16 3.95 174.4/132.6 23.9–27.1/ 910–842 2.39(with melted Experiment/
21.6–25.5^ paraffin) Simulation
Ling [203] Paraffin (RT44HC) PCM/EG/active air 20 7.85 181 714^ 2.5 Simulation
cooling
Hémery [204] Paraffin (RT28HC) PCM/EG/active liquid 245 28 Experiment
cooling
Rao [205] PCM/liquid minichannel 0.2–1.2 190 35–50 910 2.1 Simulation
cooling
Zhao [206] Paraffin PCM/EG/heat pipe/ 16 0.283 3.41 170.59^ 40 Experiment
circular fins
Wang [207] Paraffin PCM/OHP 0.21 195 40–42 822 1.77 Experiment
Zhao [208] Paraffin PCM/EG/OHP 7 0.37 2.21 161.54/142.67^ 62–66 Experiment
Wu [209] Paraffin PCM/EG/HP/forced air 0.268 7.654 186.4/141.7^ 42.02 Experiment
cooling

* Volume fraction.
^
Value of composite.
Energy Conversion and Management 150 (2017) 304–330
H. Liu et al. Energy Conversion and Management 150 (2017) 304–330

Fig. 16. Schematic of the self-heating battery structure in [215].

Table 7
Comparisons of different battery cooling methods.

Advantages Disadvantages

Forced air • Low cost • Poor thermal

cooling • Simplicity of appliance performance

Fig. 15. Battery heating methods using battery power: (a) self-internal heating (b) con-
• Easy
media
accessibility for cooling • Low temperature
uniformity
vective heating (c) mutual pulse heating and external power: (d) external convective • Applicability for different • Large space need
heating and (e) AC heating [212]. cell types • Noise problem
• High lifespan • Low efficiency
6. Conclusions
• Can be used for battery
heating
• Highly commercialized
The Li-ion battery has been proved to be a promising alternative for Liquid cooling • Large temperature drop • Potential leakage
electric vehicles. Although extensive efforts have been made on the • High temperature • High cost
selections of electrodes and electrolyte materials as well as the se- uniformity • Extra weight
• High efficiency • Increased complexity
parator development to increase the battery capacity and specific
power, the advancement of BTMSs is overlooked in a manner. However, • Can be used for battery
heating
• Maintenance difficulty
the performance, lifespan and safety of Li-ion battery are strongly as- • Applicability
cell types
for different
sociated with their operating temperature. Improper temperature will
lead to the power/capacity degradation, shorten the cycle life and may • Highly commercialized
even cause the thermal runaway thus results in fatal disasters. Large Heat pipe • High compactness • Relatively high cost
cooling • High efficiency • Contact resistance
• Relatively • More
temperature imbalance among cells or within each cell may also cause
high thermal suitable for prismatic
the failure of battery. Therefore, it is suggested that Li-ion batteries performance cells
should be maintained in a temperature range of 20–60 °C with the
• High lifespan • Not commercialized
temperature difference no more than 5 °C. To this end, an effective
PCM cooling • Low cost • Poor thermal
thermal management system is critical to control the batteries within a
• High compactness conductivity
desired temperature range and address the corresponding problems. • Large temperature drop • Flowability
To have a better understanding of the battery thermal behaviors and • Relatively uniform
temperature distribution
• Potential liquid leakage
after melting
• High • Volume
give insights to the thermal effects of BTMSs, it is essential to develop
lifespan high efficiency expansion
an accurate thermal model for Li-ion batteries to conduct numerical
• Energy storage ability • Extra weight
simulations. The key task of establishing a battery thermal model is to
formulate the expressions of energy conservation, heat generation and
• Not commercialized
boundary conditions which have been reviewed in this paper. However,
a thermal model alone is normally not precise enough to predict the

325
H. Liu et al.
thermal behavior. Therefore, the thermal model is always coupled with
an electrochemical or equivalent circuit model. The most prevalent
electrochemical model is the P2D model and a variety of simplification
methods have been proposed to reduce the computation complexity
while the most popular equivalent circuit model is the Thévenin model
which utilizes RC networks.
BTMSs which could effectively control the battery temperature are
of great significances in ensuring the performance and safety of Li-ion
batteries. The cooling strategies are categorized into air, liquid, phase
change (liquid-vapor) and PCM based systems according to their
cooling media. Among these approaches, the air and liquid cooling are
conventional methods which have been widely adopted in commercial
electrical vehicles. The merits of active air cooling include the low cost,
simplicity and easy availability of coolant materials. Developments of
air cooling have been focused on the optimization of geometric layouts
and operating parameters in recent years. The introduction of forced
liquid systems greatly enhances the efficiency of thermal management
compared to the poor heat removal coefficient of air cooling. Similarly,
researches on the indirect liquid cooling were primarily about how to
improve the channel geometry, system structures and coolant heat
transfer coefficient. The mutual problem of the forced air and liquid
thermal management is that the addition of extra components such as
valves, blowers/pumps and cooling ducts/channels will raise the
weight and space utilization of BTMSs. To address this issue, passive
thermal management techniques that take advantages of latent heat
during the phase change process can be employed. Boiling cooling is
capable of removing a large amount of heat in a short time but its
feasibility remains to be validated. As far as heat pipes, they have been
extensively used in many fields such as electronics thermal manage-
ment but have not been put into practice in BTMSs. The selection of
proper working fluids, a good contact between heat pipe and batteries
and an effective cooling method are the preconditions of excellent
thermal performance of the heat pipe based BTMSs. Besides, a new type
of heat pipe, oscillating heat pipe, attracts the attentions of some re-
searchers. The phase change (solid-liquid) cooling, referred to as PCM
cooling, is a promising alternative for conventional thermal manage-
ment strategies that has an excellent performance of temperature re-
duction and uniformity although the application of them still faces
some challenges. A prominent problem is the poor thermal conductivity
of PCMs, the proposed solutions of which have been reviewed in detail
in this paper. Another problem is the potential failure caused by the
complete melting of phase change substances. To address this problem,
the forced air and liquid cooling as well as heat pipe techniques are
integrated into the PCM systems to recover the latent heat but at the
expense of compromising the simplicity and compactness. The increase
of weight and cost also restricts its commercialization yet. Comparisons
of different battery cooling methods including forced air cooling, liquid
cooling, heat pipe cooling and PCM cooling are listed in Table 7.
Compared to the great attention paid on battery cooling, battery
heating techniques are comparatively neglected. It limits the applica-
tion of Li-ion batteries in cold climates and high-altitude drones such as
Canada and Russia. Each of the existing battery heating methods has
their own advantages and deficiencies. Therefore, an effective system to
warm up the batteries quickly in an economical and simple way is still
needed to be developed.
All in all, an efficient battery model with a high accuracy and an
effective thermal management system are highly desired. The BTMSs
should be extended to the battery pack level with the dynamic cycling
which is more similar to practical applications so that the heat accu-
mulation during charge/discharge cycles is better considered.
References
[1] British Petroleum. BP statistical review of world. Energy 2016.
[2] British Petroleum. BP Energy Outlook 2016;53:2016. http://dx.doi.org/10.1017/
CBO9781107415324.004.
326
Energy Conversion and Management 150 (2017) 304–330
[3] Wada M. Research and development of electric vehicles for clean transportation. J
Environ Sci 2009;21:745–9. http://dx.doi.org/10.1016/S1001-0742(08)62335-9.
[4] Andersen PH, Mathews JA, Rask M. Integrating private transport into renewable
energy policy: the strategy of creating intelligent recharging grids for electric ve-
hicles. Energy Policy 2009;37:2481–6. http://dx.doi.org/10.1016/j.enpol.2009.
03.032.
[5] Larminie J, Lowry J. Front matter. Electr Veh Technol Explain 2012:i–.
[6] Westbrook MH, Engineers I of E, Engineers S of A. The electric car: development
and future of battery, hybrid and fuel-cell cars. vol. 38; 2001.
[7] Reddy TB. Lindens handbook of batteries. 4th ed. Bronxville (New York): McGraw-
Hill Education; 2010.
[8] Aifantis KE, Hackney SA, Kumar RV. High energy density lithium batteries. Wiley-
VCH Verlag GmbH & Co. KGaA; 2010. http://dx.doi.org/10.1002/
9783527630011.
[9] Budde-Meiwes H, Drillkens J, Lunz B, Muennix J, Rothgang S, Kowal J, et al. A
review of current automotive battery technology and future prospects. Proc Inst
Mech Eng Part D J Automob Eng 2013;227:761–76. http://dx.doi.org/10.1177/
0954407013485567.
[10] Warner J. Introduction. Handb lithium-ion batter pack des. Elsevier; 2015. p. 1–8.
http://dx.doi.org/10.1016/B978-0-12-801456-1.00001-4.
[11] Conte FV. Battery and battery management for hybrid electric vehicles: a review. E
I Elektrotechnik Und Informationstechnik 2006;123:424–31. http://dx.doi.org/10.
1007/s00502-006-0383-6.
[12] Väyrynen A, Salminen J. Lithium ion battery production. J Chem Thermodyn
2012;46:80–5. http://dx.doi.org/10.1016/j.jct.2011.09.005.
[13] Pesaran A, Santhanagopalan GK. Addressing the impact of temperature extremes
on large format Li-ion batteries for vehicle applications. In: 30th int batter semin;
2013.
[14] Motloch CG, Christophersen JP, Belt JR, Wright RB, Hunt GL, Sutula RA, et al.
High-power battery testing procedures and analytical methodologies for HEV’s.
Futur Car Congr 2002;2002:1–7. http://dx.doi.org/10.4271/2002-01-1950.
[15] Sun J, Li J, Zhou T, Yang K, Wei S, Tang N, et al. Toxicity, a serious concern of
thermal runaway from commercial Li-ion battery. Nano Energy 2016;27:313–9.
http://dx.doi.org/10.1016/j.nanoen.2016.06.031.
[16] Pesaran Aa. Battery thermal models for hybrid vehicle simulations. J Power
Sources 2002;110:377–82. http://dx.doi.org/10.1016/S0378-7753(02)00200-8.
[17] Dahlin GR, Strom KE. Lithium batteries: research, technology and applications;
2010.
[18] Amatucci GG. CoO2, The end member of the LixCoO2 solid solution. J Electrochem
Soc 1996;143:1114. http://dx.doi.org/10.1149/1.1836594.
[19] Nitta N, Wu F, Lee JT, Yushin G. Li-ion battery materials: present and future. Mater
Today 2015;18:252–64. http://dx.doi.org/10.1016/j.mattod.2014.10.040.
[20] Sivakumar M, Muruganantham R, Subadevi R. Synthesis of surface modified
LiFePO4 cathode material via polyol technique for high rate lithium secondary
battery. Appl Surf Sci 2015;337:234–40. http://dx.doi.org/10.1016/j.apsusc.
2015.02.100.
[21] Lee J, Kumar P, Moudgil BM, Singh RK. Electrochemical enhancement of LiFePO4
as a cathode material by incorporating Cu flakes for lithium ion rechargeable
battery. Solid State Ionics 2013;231:18–24. http://dx.doi.org/10.1016/j.ssi.2012.
10.015.
[22] Shen CW, Ko TH, Liu CH, Chiu KF, Leu HJ, Lin PS, et al. Low temperature process
modified MCMB for lithium ion batteries. Solid State Ionics 2014;268:282–7.
http://dx.doi.org/10.1016/j.ssi.2014.07.001.
[23] Zuo X, Zhu J, Müller-Buschbaum P, Cheng Y-J. Silicon based lithium-ion battery
anodes: a chronicle perspective review. Nano Energy 2017;31:113–43. http://dx.
doi.org/10.1016/j.nanoen.2016.11.013.
[24] Wang Q, Jiang B, Li B, Yan Y. A critical review of thermal management models and
solutions of lithium-ion batteries for the development of pure electric vehicles.
Renew Sustain Energy Rev 2016;64:106–28. http://dx.doi.org/10.1016/j.rser.
2016.05.033.
[25] Doughty D, Roth EP. A general discussion of Li ion battery safety. Electrochem Soc
Interface 2012;21:37–44.
[26] Younesi R, Veith GM, Johansson P, Edstrom K, Vegge T. Lithium salts for advanced
lithium batteries: Li-metal, Li-O2, and Li-S. Energy Environ Sci 2015;8:1905–22.
http://dx.doi.org/10.1039/C5EE01215E.
[27] Jiang F, Nie Y, Yin L, Feng Y, Yu Q, Zhong C. Core-shell-structured nanofibrous
membrane as advanced separator for lithium-ion batteries. J Membr Sci
2016;510:1–9. http://dx.doi.org/10.1016/j.memsci.2016.02.067.
[28] Liu Y, Ma H, Hsiao BS, Chu B, Tsou AH. Improvement of meltdown temperature of
lithium-ion battery separator using electrospun polyethersulfone membranes.
Polymer 2016;107:163–9. http://dx.doi.org/10.1016/j.polymer.2016.11.020.
[29] Panasonic Industrial devices. Lithium ion batteries; 2007. https://industrial.
panasonic.com/cdbs/www-data/pdf/ACA4000/ACA4000PE3.pdf.
[30] Wk057. Pics/Info: Inside the battery pack. Tesla Mot Club; 2014. https://
teslamotorsclub.com/tmc/threads/pics-info-inside-the-battery-pack.34934/.
[31] Messina C. Lithium-ion cylindrical cells vs. prismatic cells. Reli Batter; 2015.
http://www.relionbattery.com/blog/lithium-cells-should-i-go-cylindrical-or-
prismatic.
[32] Roper LD. Tesla Model S; 2016. http://www.roperld.com/science/TeslaModelS.
htm.
[33] Tesla Model S. Wikipedia.org 2017. https://en.wikipedia.org/wiki/Tesla_Model_
S#Battery.
[34] Goodenough JB, Park K-S. The Li-ion rechargeable battery: a perspective. J Am
Chem Soc 2013;135:1167–76. http://dx.doi.org/10.1021/ja3091438.
[35] Shim J, Kostecki R, Richardson T, Song X, Striebel KA. Electrochemical analysis for
cycle performance and capacity fading of a lithium-ion battery cycled at elevated
H. Liu et al.
temperature. J Power Sources 2002;112:222–30. http://dx.doi.org/10.1016/
S0378-7753(02)00363-4.
[36] Sato N. Thermal behavior analysis of lithium-ion batteries for electric and hybrid
vehicles. J Power Sources 2001;99:70–7. http://dx.doi.org/10.1016/S0378-
7753(01)00478-5.
[37] Broussely M, Biensan P, Bonhomme F, Blanchard P, Herreyre S, Nechev K, et al.
Main aging mechanisms in Li ion batteries. J Power Sources 2005;146:90–6.
http://dx.doi.org/10.1016/j.jpowsour.2005.03.172.
[38] Ning G, Haran B, Popov BN. Capacity fade study of lithium-ion batteries cycled at
high discharge rates. J Power Sources 2003;117:160–9. http://dx.doi.org/10.
1016/S0378-7753(03)00029-6.
[39] MacNeil DD, Lu Z, Chen Z, Dahn JR. A comparison of the electrode/electrolyte
reaction at elevated temperatures for various Li-ion battery cathodes. J Power
Sources 2002;108:8–14. http://dx.doi.org/10.1016/S0378-7753(01)01013-8.
[40] Huang C-K, Sakamoto JS, Wolfenstine J, Surampudi S. The limits of low-tem-
perature performance of li-ion cells. J Electrochem Soc 2000;147:2893–6. http://
dx.doi.org/10.1149/1.1393622.
[41] Lin H-P, Chua D, Salomon M, Shiao H-C, Hendrickson M, Plichta E, et al. Low-
temperature behavior of Li-ion cells. Electrochem Solid-State Lett 2001;4:A71.
http://dx.doi.org/10.1149/1.1368736.
[42] Nagpure SC, Dinwiddie R, Babu SS, Rizzoni G, Bhushan B, Frech T. Thermal dif-
fusivity study of aged Li-ion batteries using flash method. J Power Sources
2010;195:872–6. http://dx.doi.org/10.1016/j.jpowsour.2009.08.025.
[43] Belt JR, Ho CD, Motloch CG, Miller TJ, Duong TQ. A capacity and power fade
study of Li-ion cells during life cycle testing. J Power Sources 2003;123:241–6.
http://dx.doi.org/10.1016/S0378-7753(03)00537-8.
[44] Nunotani K, Yoshida F, Kamiya Y, Daisho Y, Abe K, Kono M, et al. Development
and performance evaluation of lithium iron phosphate battery with superior rapid
charging performance—second report: evaluation of battery capacity loss char-
acteristics. Veh power propuls conf (VPPC) IEEE; 2011. p. 1–4. http://dx.doi.org/
10.1109/VPPC.2011.6042998.
[45] Ramadass P, Haran B, White R, Popov BN. Capacity fade of Sony 18,650 cells
cycled at elevated temperatures: Part I. Cycling performance. J Power Sources
2002;112:606–13. http://dx.doi.org/10.1016/S0378-7753(02)00474-3.
[46] Wu M-S, Chiang P-CJ. High-rate capability of lithium-ion batteries after storing at
elevated temperature. Electrochim Acta 2007;52:3719–25. http://dx.doi.org/10.
1016/j.electacta.2006.10.045.
[47] Jiang J, Zhang C. Fundamentals and applications of lithium-ion batteries in elec-
tric drive vehicles. John Wiley & Sons Singapore Pte. Ltd; 2015. http://dx.doi.org/
10.1017/CBO9781107415324.004.
[48] Xiao X, Ahn D, Liu Z, Kim JH, Lu P. Atomic layer coating to mitigate capacity
fading associated with manganese dissolution in lithium ion batteries. Electrochem
Commun 2013;32:31–4. http://dx.doi.org/10.1016/j.elecom.2013.03.030.
[49] Zhou Z, Ma Y, Wang L, Zuo P, Cheng X, Du C, et al. Triphenyl phosphite as an
electrolyte additive to improve the cyclic stability of lithium-rich layered oxide
cathode for lithium-ion batteries. Electrochim Acta 2016;216:44–50. http://dx.
doi.org/10.1016/j.electacta.2016.09.008.
[50] Shah K, Chalise D, Jain A. Experimental and theoretical analysis of a method to
predict thermal runaway in Li-ion cells. J Power Sources 2016;330:167–74. http://
dx.doi.org/10.1016/j.jpowsour.2016.08.133.
[51] Feng X, Fang M, He X, Ouyang M, Lu L, Wang H, et al. Thermal runaway features
of large format prismatic lithium ion battery using extended volume accelerating
rate calorimetry. J Power Sources 2014;255:294–301. http://dx.doi.org/10.1016/
j.jpowsour.2014.01.005.
[52] Peng P, Jiang F. Thermal behavior analyses of stacked prismatic LiCoO2 lithium-
ion batteries during oven tests. Int J Heat Mass Transf 2015;88:411–23. http://dx.
doi.org/10.1016/j.ijheatmasstransfer.2015.04.101.
[53] Zhao R, Liu J, Gu J. Simulation and experimental study on lithium ion battery
short circuit. Appl Energy 2016;173:29–39. http://dx.doi.org/10.1016/j.
apenergy.2016.04.016.
[54] Ye J, Chen H, Wang Q, Huang P, Sun J, Lo S. Thermal behavior and failure me-
chanism of lithium ion cells during overcharge under adiabatic conditions. Appl
Energy 2016;182:464–74. http://dx.doi.org/10.1016/j.apenergy.2016.08.124.
[55] Hatchard TD, Trussler S, Dahn JR. Building a “smart nail” for penetration tests on
Li-ion cells. J Power Sources 2014;247:821–3. http://dx.doi.org/10.1016/j.
jpowsour.2013.09.022.
[56] Wang H, Kumar A, Simunovic S, Allu S, Kalnaus S, Turner JA, et al. Progressive
mechanical indentation of large-format Li-ion cells. J Power Sources
2017;341:156–64. http://dx.doi.org/10.1016/j.jpowsour.2016.11.094.
[57] Wang Q, Ping P, Zhao X, Chu G, Sun J, Chen C. Thermal runaway caused fire and
explosion of lithium ion battery. J Power Sources 2012;208:210–24. http://dx.doi.
org/10.1016/j.jpowsour.2012.02.038.
[58] Bandhauer TM, Garimella S, Fuller TF. A critical review of thermal issues in li-
thium-ion batteries. J Electrochem Soc 2011;158:R1. http://dx.doi.org/10.1149/
1.3515880.
[59] Spotnitz R, Franklin J. Abuse behavior of high-power, lithium-ion cells. J Power
Sources 2003;113:81–100.
[60] Balakrishnan PG, Ramesh R, Prem Kumar T. Safety mechanisms in lithium-ion
batteries. J Power Sources 2006;155:401–14. http://dx.doi.org/10.1016/j.
jpowsour.2005.12.002.
[61] Ouyang M, Ren D, Lu L, Li J, Feng X, Han X, et al. Overcharge-induced capacity
fading analysis for large format lithium-ion batteries with LiyNi1/3Co1/3Mn1/
3O2 + LiyMn2O4 composite cathode. J Power Sources 2015;279:626–35. http://
dx.doi.org/10.1016/j.jpowsour.2015.01.051.
[62] Yuan Q, Zhao F, Wang W, Zhao Y, Liang Z, Yan D. Overcharge failure investigation
of lithium-ion batteries. Electrochim Acta 2015;178:682–8. http://dx.doi.org/10.
327
Energy Conversion and Management 150 (2017) 304–330
1016/j.electacta.2015.07.147.
[63] Chiu K-C, Lin C-H, Yeh S-F, Lin Y-H, Huang C-S, Chen K-C. Cycle life analysis of
series connected lithium-ion batteries with temperature difference. J Power
Sources 2014;263:75–84. http://dx.doi.org/10.1016/j.jpowsour.2014.04.034.
[64] Huang Q, Yan M, Jiang Z. Thermal study on single electrodes in lithium-ion bat-
tery. J Power Sources 2006;156:541–6. http://dx.doi.org/10.1016/j.jpowsour.
2005.05.083.
[65] Park C, Jaura AK. Dynamic thermal model of Li-ion battery for predictive behavior
in hybrid and fuel cell vehicles. SAE Int 2003. http://dx.doi.org/10.4271/2003-
01-2286.
[66] Zhao R, Zhang S, Liu J, Gu J. A review of thermal performance improving methods
of lithium ion battery: electrode modification and thermal management system. J
Power Sources 2015;299:557–77. http://dx.doi.org/10.1016/j.jpowsour.2015.09.
001.
[67] Cosley MR, Garcia MP. Battery thermal management system. INTELEC 2004 26th
annu int telecommun energy conf 2004; 2004. p. 38–45. http://dx.doi.org/10.
1109/INTLEC.2004.1401442.
[68] Mohammadian SK, He Y-L, Zhang Y. Internal cooling of a lithium-ion battery using
electrolyte as coolant through microchannels embedded inside the electrodes. J
Power Sources 2015;293:458–66. http://dx.doi.org/10.1016/j.jpowsour.2015.05.
055.
[69] Environmental Idaho National Engineering and Laboratory. FreedomCAR battery
test manual for power-assist hybrid electric vehicles. Doe/Id-11069; 2003.
doi:11069.
[70] Fan J. On the discharge capability and its limiting factors of commercial 18,650 Li-
ion cell at low temperatures. J Power Sources 2003;117:170–8. http://dx.doi.org/
10.1016/S0378-7753(03)00354-9.
[71] Nagasubramanian G. Electrical characteristics of 18650 Li-ion cells at low tem-
peratures. J Appl Electrochem 2001;31:99–104. http://dx.doi.org/10.1023/
A:1004113825283.
[72] Jaguemont J, Boulon L, Dubé Y. A comprehensive review of lithium-ion batteries
used in hybrid and electric vehicles at cold temperatures. Appl Energy
2016;164:99–114. http://dx.doi.org/10.1016/j.apenergy.2015.11.034.
[73] Senyshyn A, Mühlbauer MJ, Dolotko O, Ehrenberg H. Low-temperature perfor-
mance of Li-ion batteries: the behavior of lithiated graphite. J Power Sources
2015;282:235–40. http://dx.doi.org/10.1016/j.jpowsour.2015.02.008.
[74] Zhang SS, Xu K, Jow TR. The low temperature performance of Li-ion batteries. J
Power Sources 2003;115:137–40. http://dx.doi.org/10.1016/S0378-7753(02)
00618-3.
[75] Zhang SS, Xu K, Jow TR. Electrochemical impedance study on the low temperature
of Li-ion batteries. Electrochim Acta 2004;49:1057–61. http://dx.doi.org/10.
1016/j.electacta.2003.10.016.
[76] Waldmann T, Wilka M, Kasper M, Fleischhammer M, Wohlfahrt-Mehrens M.
Temperature dependent ageing mechanisms in Lithium-ion batteries – a Post-
Mortem study. J Power Sources 2014;262:129–35. http://dx.doi.org/10.1016/j.
jpowsour.2014.03.112.
[77] Pesaran A, Vlahinos A, Stuart T. Cooling and preheating of batteries in hybrid
electric vehicles. 6th ASME-JSME therm eng jt conf. 2003; 2003. p. 1–7.
[78] Chen Y, Evans JW. Thermal analysis of lithium-ion batteries. J Electrochem Soc
1996;143:2708–12. http://dx.doi.org/10.1149/1.1837095.
[79] Chen SC, Wan CC, Wang YY. Thermal analysis of lithium-ion batteries. J Power
Sources 2005;140:111–24. http://dx.doi.org/10.1016/j.jpowsour.2004.05.064.
[80] Holman JP. Heat transfer. 10th ed. McGraw-Hill Higher Education; 2010.
[81] Xiao M, Choe S-Y. Theoretical and experimental analysis of heat generations of a
pouch type LiMn2O4/carbon high power Li-polymer battery. J Power Sources
2013;241:46–55. http://dx.doi.org/10.1016/j.jpowsour.2013.04.062.
[82] Ye Y, Shi Y, Cai N, Lee J, He X. Electro-thermal modeling and experimental vali-
dation for lithium ion battery. J Power Sources 2012;199:227–38. http://dx.doi.
org/10.1016/j.jpowsour.2011.10.027.
[83] Lai Y, Du S, Ai L, Ai L, Cheng Y, Tang Y, et al. Insight into heat generation of
lithium ion batteries based on the electrochemical-thermal model at high dis-
charge rates. Int J Hydrogen Energy 2015;40:13039–49. http://dx.doi.org/10.
1016/j.ijhydene.2015.07.079.
[84] Zhang X. Thermal analysis of a cylindrical lithium-ion battery. Electrochim Acta
2011;56:1246–55. http://dx.doi.org/10.1016/j.electacta.2010.10.054.
[85] Nyman A, Zavalis TG, Elger R, Behm M, Lindbergh G. Analysis of the polarization
in a Li-ion battery cell by numerical simulations. J Electrochem Soc
2010;157:A1236. http://dx.doi.org/10.1149/1.3486161.
[86] Gu WB, Wang CY. Thermal-electrochemical modeling of battery systems. J
Electrochem Soc 2000;147:2910. http://dx.doi.org/10.1149/1.1393625.
[87] Bernardi D. A general energy balance for battery systems. J Electrochem Soc
1985;132:5. http://dx.doi.org/10.1149/1.2113792.
[88] Karimi G, Li X. Thermal management of lithium-ion batteries for electric vehicles.
Int J Energy Res 2013;37:13–24. http://dx.doi.org/10.1002/er.1956.
[89] Inui Y, Kobayashi Y, Watanabe Y, Watase Y, Kitamura Y. Simulation of tempera-
ture distribution in cylindrical and prismatic lithium ion secondary batteries.
Energy Convers Manag 2007;48:2103–9. http://dx.doi.org/10.1016/j.enconman.
2006.12.012.
[90] Bergman TL, Lavine AS, Incropera FP, DeWitt DP. Introduction to heat transfer.
6th ed. John Wiley & Sons, Inc.; 2011.
[91] Fuller TF. Simulation and optimization of the dual lithium ion insertion cell. J
Electrochem Soc 1994;141:1. http://dx.doi.org/10.1149/1.2054684.
[92] Doyle M. Modeling of galvanostatic charge and discharge of the lithium/polymer/
insertion cell. J Electrochem Soc 1993;140:1526. http://dx.doi.org/10.1149/1.
2221597.
[93] Doyle M. Comparison of modeling predictions with experimental data from plastic
H. Liu et al.
lithium ion cells. J Electrochem Soc 1996;143:1890. http://dx.doi.org/10.1149/1.
1836921.
[94] Abada S, Marlair G, Lecocq A, Petit M, Sauvant-Moynot V, Huet F. Safety focused
modeling of lithium-ion batteries: a review. J Power Sources 2016;306:178–92.
http://dx.doi.org/10.1016/j.jpowsour.2015.11.100.
[95] Subramanian VR, Ritter JA, White RE. Approximate solutions for galvanostatic
discharge of spherical particles I. Constant diffusion coefficient. J Electrochem Soc
2001;148:E444–9. http://dx.doi.org/10.1149/1.1409397.
[96] Subramanian VR, Tapriyal D, White RE. A boundary condition for porous elec-
trodes. Electrochem Solid-State Lett 2004;7:A259. http://dx.doi.org/10.1149/1.
1773751.
[97] Guo M, Sikha G, White RE. Single-particle model for a lithium-ion cell: thermal
behavior. J Electrochem Soc 2011;158:A122. http://dx.doi.org/10.1149/1.
3521314.
[98] Dubey AM, Ram R, Yadav AK. Ion cell performance using single particle re-
presentation of battery electrode. J Electroanal Chem 2016;768:11–7. http://dx.
doi.org/10.1016/j.jelechem.2016.02.018.
[99] Santhanagopalan S, Guo Q, Ramadass P, White RE. Review of models for pre-
dicting the cycling performance of lithium ion batteries. J Power Sources
2006;156:620–8. http://dx.doi.org/10.1016/j.jpowsour.2005.05.070.
[100] Wang CY. Micro-macroscopic coupled modeling of batteries and fuel cells. J
Electrochem Soc 1998;145:3407. http://dx.doi.org/10.1149/1.1838820.
[101] Jokar A, Rajabloo B, Désilets M, Lacroix M. Review of simplified pseudo-two-di-
mensional models of lithium-ion batteries. J Power Sources 2016;327:44–55.
http://dx.doi.org/10.1016/j.jpowsour.2016.07.036.
[102] Di Domenico D, Fiengo G, Stefanopoulou A. Lithium-ion battery state of charge
estimation with a Kalman Filter based on a electrochemical model. 2008 IEEE int
conf control appl, vol. 32. IEEE; 2008. p. 702–7. http://dx.doi.org/10.1109/CCA.
2008.4629639.
[103] Di Domenico D, Stefanopoulou A, Fiengo G. Lithium-ion battery state of charge
and critical surface charge estimation using an electrochemical model-based ex-
tended Kalman filter. J Dyn Syst Meas Control 2010;132:61302. http://dx.doi.
org/10.1115/1.4002475.
[104] Lin C, Tang A. Simplification and efficient simulation of electrochemical model for
Li-ion battery in EVs. Energy Procedia 2016;104:68–73. http://dx.doi.org/10.
1016/j.egypro.2016.12.013.
[105] Cai L, White RE. Reduction of model order based on proper orthogonal decom-
position for lithium-ion battery simulations. J Electrochem Soc 2009;156:A154.
http://dx.doi.org/10.1149/1.3049347.
[106] Luo W, Lyu C, Wang L, Zhang L. A new extension of physics-based single particle
model for higher charge–discharge rates. J Power Sources 2013;241:295–310.
http://dx.doi.org/10.1016/j.jpowsour.2013.04.129.
[107] Chen X, Shen W, Cao Z, Kapoor A. Adaptive gain sliding mode observer for state of
charge estimation based on combined battery equivalent circuit model. Comput
Chem Eng 2014;64:114–23. http://dx.doi.org/10.1016/j.compchemeng.2014.02.
015.
[108] Liaw BY, Nagasubramanian G, Jungst RG, Doughty DH. Modeling of lithium ion
cells? A simple equivalent-circuit model approach. Solid State Ionics
2004;175:835–9. http://dx.doi.org/10.1016/j.ssi.2004.09.049.
[109] Seaman A, Dao TS, McPhee J. A survey of mathematics-based equivalent-circuit
and electrochemical battery models for hybrid and electric vehicle simulation. J
Power Sources 2014;256:410–23. http://dx.doi.org/10.1016/j.jpowsour.2014.01.
057 Review.
[110] Greenleaf M, Li H, Zheng JP. Application of physical electric circuit modeling to
characterize Li-ion battery electrochemical processes. J Power Sources
2014;270:113–20. http://dx.doi.org/10.1016/j.jpowsour.2014.07.083.
[111] Zhang H, Chow M-Y. Comprehensive dynamic battery modeling for PHEV appli-
cations. IEEE PES gen meet IEEE; 2010. p. 1–6. http://dx.doi.org/10.1109/PES.
2010.5590108.
[112] Andre D, Meiler M, Steiner K, Walz H, Soczka-Guth T, Sauer DU. Characterization
of high-power lithium-ion batteries by electrochemical impedance spectroscopy.
II: Modelling. J Power Sources 2011;196:5349–56. http://dx.doi.org/10.1016/j.
jpowsour.2010.07.071.
[113] Chan HL, Sutanto D. A new battery model for use with battery energy storage
systems and electric vehicles power systems. 2000 IEEE power eng soc winter
meet conf proc 2000; 2000. p. 470–5. http://dx.doi.org/10.1109/PESW.2000.
850009.
[114] Chen M, Rincon-Mora GA. Accurate electrical battery model capable of predicting
runtime and I-V performance. IEEE Trans Energy Convers 2006;21:504–11. http://
dx.doi.org/10.1109/TEC.2006.874229.
[115] Rahimian SK, Rayman S, White RE. Comparison of single particle and equivalent
circuit analog models for a lithium-ion cell. J Power Sources 2011;196:8450–62.
http://dx.doi.org/10.1016/j.jpowsour.2011.06.007.
[116] Lam Long, Bauer P, Kelder E. A practical circuit-based model for Li-ion battery
cells in electric vehicle applications. 2011 IEEE 33rd int telecommun energy conf
2011; 2011. p. 1–9. http://dx.doi.org/10.1109/INTLEC.2011.6099803.
[117] Liaw BY, Jungst RG, Nagasubramanian G, Case HL, Doughty DH. Modeling ca-
pacity fade in lithium-ion cells. J Power Sources 2005;140:157–61. http://dx.doi.
org/10.1016/j.jpowsour.2004.08.017.
[118] Szumanowski A, Yuhua C. Battery management system based on battery nonlinear
dynamics modeling. IEEE Trans Veh Technol 2008;57:1425–32. http://dx.doi.org/
10.1109/TVT.2007.912176.
[119] Hu X, Li S, Peng H. A comparative study of equivalent circuit models for Li-ion
batteries. J Power Sources 2012;198:359–67. http://dx.doi.org/10.1016/j.
jpowsour.2011.10.013.
[120] Song L, Evans JW. Electrochemical-thermal model of Lithium polymer batteries. J
328
Energy Conversion and Management 150 (2017) 304–330
Electrochem Soc 2000;147:2086. http://dx.doi.org/10.1149/1.1393490.
[121] Smith K, Wang C-Y. Power and thermal characterization of a lithium-ion battery
pack for hybrid-electric vehicles. J Power Sources 2006;160:662–73. http://dx.
doi.org/10.1016/j.jpowsour.2006.01.038.
[122] Prada E, Di Domenico D, Creff Y, Bernard J, Sauvant-Moynot V, Huet F. Simplified
electrochemical and thermal model of LiFePO4-graphite Li-ion batteries for fast
charge applications. J Electrochem Soc 2012;159:A1508–19. http://dx.doi.org/
10.1149/2.064209jes.
[123] Greco A, Jiang X. A coupled thermal and electrochemical study of lithium-ion
battery cooled by paraffin/porous-graphite-matrix composite. J Power Sources
2016;315:127–39. http://dx.doi.org/10.1016/j.jpowsour.2016.03.018.
[124] Basu S, Hariharan KS, Kolake SM, Song T, Sohn DK, Yeo T. Coupled electro-
chemical thermal modelling of a novel Li-ion battery pack thermal management
system. Appl Energy 2016;181:1–13. http://dx.doi.org/10.1016/j.apenergy.2016.
08.049.
[125] Lin X, Perez HE, Mohan S, Siegel JB, Stefanopoulou AG, Ding Y, et al. A lumped-
parameter electro-thermal model for cylindrical batteries. J Power Sources
2014;257:1–11. http://dx.doi.org/10.1016/j.jpowsour.2014.01.097.
[126] Saw LH, Ye Y, Tay AAO. Electro-thermal characterization of Lithium Iron
Phosphate cell with equivalent circuit modeling. Energy Convers Manag
2014;87:367–77. http://dx.doi.org/10.1016/j.enconman.2014.07.011.
[127] Bandhauer TM, Garimella S. Passive, internal thermal management system for
batteries using microscale liquid-vapor phase change. Appl Therm Eng
2013;61:756–69. http://dx.doi.org/10.1016/j.applthermaleng.2013.08.004.
[128] Giuliano MR, Advani SG, Prasad AK. Thermal analysis and management of li-
thium-titanate batteries. J Power Sources 2011;196:6517–24. http://dx.doi.org/
10.1016/j.jpowsour.2011.03.099.
[129] Kelly KJ, Mihalic M, Zolot M. Battery usage and thermal performance of the
Toyota Prius and Honda Insight during chassis dynamometer testing. 17th annu
batter conf appl adv proc conf (Cat. No. 02TH8576) IEEE; 2006. p. 247–52. http://
dx.doi.org/10.1109/BCAA.2002.986408.
[130] Kim G-H, Pesaran A. Battery thermal management system design modeling. In:
22nd int batter hybrid fuel cell electr veh conf exhib EVS22, vol. 1; 2006. p.
126–33.
[131] Park H. A design of air flow configuration for cooling lithium ion battery in hybrid
electric vehicles. J Power Sources 2013;239:30–6. http://dx.doi.org/10.1016/j.
jpowsour.2013.03.102.
[132] Sun H, Dixon R. Development of cooling strategy for an air cooled lithium-ion
battery pack. J Power Sources 2014;272:404–14. http://dx.doi.org/10.1016/j.
jpowsour.2014.08.107.
[133] Mohammadian SK, Zhang Y. Thermal management optimization of an air-cooled
Li-ion battery module using pin-fin heat sinks for hybrid electric vehicles. J Power
Sources 2015;273:431–9. http://dx.doi.org/10.1016/j.jpowsour.2014.09.110.
[134] Mohammadian SK, Rassoulinejad-Mousavi SM, Zhang Y. Thermal management
improvement of an air-cooled high-power lithium-ion battery by embedding metal
foam. J Power Sources 2015;296:305–13. http://dx.doi.org/10.1016/j.jpowsour.
2015.07.056.
[135] Mohammadian SK, Zhang Y. Cumulative effects of using pin fin heat sink and
porous metal foam on thermal management of lithium-ion batteries. Appl Therm
Eng 2017;118:375–84. http://dx.doi.org/10.1016/j.applthermaleng.2017.02.121.
[136] Yu K, Yang X, Cheng Y, Li C. Thermal analysis and two-directional air flow thermal
management for lithium-ion battery pack. J Power Sources 2014;270:193–200.
http://dx.doi.org/10.1016/j.jpowsour.2014.07.086.
[137] Wang T, Tseng KJ, Zhao J, Wei Z. Thermal investigation of lithium-ion battery
module with different cell arrangement structures and forced air-cooling strate-
gies. Appl Energy 2014;134:229–38. http://dx.doi.org/10.1016/j.apenergy.2014.
08.013.
[138] Fathabadi H. A novel design including cooling media for Lithium-ion batteries
pack used in hybrid and electric vehicles. J Power Sources 2014;245:495–500.
http://dx.doi.org/10.1016/j.jpowsour.2013.06.160.
[139] Fan L, Khodadadi JM, Pesaran AA. A parametric study on thermal management of
an air-cooled lithium-ion battery module for plug-in hybrid electric vehicles. J
Power Sources 2013;238:301–12. http://dx.doi.org/10.1016/j.jpowsour.2013.03.
050.
[140] Yang N, Zhang X, Li G, Hua D. Assessment of the forced air-cooling performance
for cylindrical lithium-ion battery packs: a comparative analysis between aligned
and staggered cell arrangements. Appl Therm Eng 2015;80:55–65. http://dx.doi.
org/10.1016/j.applthermaleng.2015.01.049.
[141] Tong W, Somasundaram K, Birgersson E, Mujumdar AS, Yap C. Thermo-electro-
chemical model for forced convection air cooling of a lithium-ion battery module.
Appl Therm Eng 2016;99:672–82. http://dx.doi.org/10.1016/j.applthermaleng.
2016.01.050.
[142] Yang T, Yang N, Zhang X, Li G. Investigation of the thermal performance of axial-
flow air cooling for the lithium-ion battery pack. Int J Therm Sci
2016;108:132–44. http://dx.doi.org/10.1016/j.ijthermalsci.2016.05.009.
[143] Saw LH, Ye Y, Tay AAO, Chong WT, Kuan SH, Yew MC. Computational fluid
dynamic and thermal analysis of lithium-ion battery pack with air cooling. Appl
Energy 2016;177:783–92. http://dx.doi.org/10.1016/j.apenergy.2016.05.122.
[144] Zhu C, Li X, Song L, Xiang L. Development of a theoretically based thermal model
for lithium ion battery pack. J Power Sources 2013;223:155–64. http://dx.doi.
org/10.1016/j.jpowsour.2012.09.035.
[145] Wang T, Tseng KJ, Zhao J. Development of efficient air-cooling strategies for li-
thium-ion battery module based on empirical heat source model. Appl Therm Eng
2015;90:521–9. http://dx.doi.org/10.1016/j.applthermaleng.2015.07.033.
[146] Chen D, Jiang J, Kim GH, Yang C, Pesaran A. Comparison of different cooling
methods for lithium ion battery cells. Appl Therm Eng 2016;94:846–54. http://dx.
H. Liu et al.
doi.org/10.1016/j.applthermaleng.2015.10.015.
[147] Lan C, Xu J, Qiao Y, Ma Y. Thermal management for high power lithium-ion
battery by minichannel aluminum tubes. Appl Therm Eng 2016;101:284–92.
http://dx.doi.org/10.1016/j.applthermaleng.2016.02.070.
[148] Zhang T, Gao Q, Wang G, Gu Y, Wang Y, Bao W, et al. Investigation on the pro-
motion of temperature uniformity for the designed battery pack with liquid flow in
cooling process. Appl Therm Eng 2017;116:655–62. http://dx.doi.org/10.1016/j.
applthermaleng.2017.01.069.
[149] Tong W, Somasundaram K, Birgersson E, Mujumdar AS, Yap C. Numerical in-
vestigation of water cooling for a lithium-ion bipolar battery pack. Int J Therm Sci
2015;94:259–69. http://dx.doi.org/10.1016/j.ijthermalsci.2015.03.005.
[150] Huo Y, Rao Z, Liu X, Zhao J. Investigation of power battery thermal management
by using mini-channel cold plate. Energy Convers Manag 2015;89:387–95. http://
dx.doi.org/10.1016/j.enconman.2014.10.015.
[151] Qian Z, Li Y, Rao Z. Thermal performance of lithium-ion battery thermal man-
agement system by using mini-channel cooling. Energy Convers Manag
2016;126:622–31. http://dx.doi.org/10.1016/j.enconman.2016.08.063.
[152] Jin LW, Lee PS, Kong XX, Fan Y, Chou SK. Ultra-thin minichannel LCP for EV
battery thermal management. Appl Energy 2014;113:1786–94. http://dx.doi.org/
10.1016/j.apenergy.2013.07.013.
[153] Jarrett A, Kim IY. Design optimization of electric vehicle battery cooling plates for
thermal performance. J Power Sources 2011;196:10359–68. http://dx.doi.org/10.
1016/j.jpowsour.2011.06.090.
[154] Jarrett A, Kim IY. Influence of operating conditions on the optimum design of
electric vehicle battery cooling plates. J Power Sources 2014;245:644–55. http://
dx.doi.org/10.1016/j.jpowsour.2013.06.114.
[155] Zhao J, Rao Z, Li Y. Thermal performance of mini-channel liquid cooled cylinder
based battery thermal management for cylindrical lithium-ion power battery.
Energy Convers Manag 2015;103:157–65. http://dx.doi.org/10.1016/j.
enconman.2015.06.056.
[156] Huo Y, Rao Z. The numerical investigation of nanofluid based cylinder battery
thermal management using lattice Boltzmann method. Int J Heat Mass Transf
2015;91:374–84. http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.07.128.
[157] Yang X-H, Tan S-C, Liu J. Thermal management of Li-ion battery with liquid metal.
Energy Convers Manag 2016;117:577–85. http://dx.doi.org/10.1016/j.
enconman.2016.03.054.
[158] Zhao R, Zhang S, Gu J, Liu J, Carkner S, Lanoue E. An experimental study of
lithium ion battery thermal management using flexible hydrogel films. J Power
Sources 2014;255:29–36. http://dx.doi.org/10.1016/j.jpowsour.2013.12.138.
[159] Zhang S, Zhao R, Liu J, Gu J. Investigation on a hydrogel based passive thermal
management system for lithium ion batteries. Energy 2014;68:854–61. http://dx.
doi.org/10.1016/j.energy.2014.03.012.
[160] Adams DT, Berdichevsky G, Colson TE, Hebert A, Kohn S, Lyons D, et al. Battery
pack thermal management system. US Patent 20090023056, vol. A1; 2009.
[161] Van Gils RW, Danilov D, Notten PHL, Speetjens MFM, Nijmeijer H. Battery thermal
management by boiling heat-transfer. Energy Convers Manag 2014;79:9–17.
http://dx.doi.org/10.1016/j.enconman.2013.12.006.
[162] Hirano H, Tajima T, Hasegawa T, Sekiguchi T, Uchino M. Boiling liquid battery
cooling for electric vehicle. IEEE conf expo transp electrif Asia-Pacific (ITEC Asia-
Pacific) IEEE; 2014. p. 1–4. http://dx.doi.org/10.1109/ITEC-AP.2014.6940931.
[163] An Z, Jia L, Li X, Ding Y. Experimental investigation on lithium-ion battery
thermal management based on flow boiling in mini-channel. Appl Therm Eng
2017;117:534–43. http://dx.doi.org/10.1016/j.applthermaleng.2017.02.053.
[164] Manimaran R, Palaniradja K, Alagumurthi N, Hussain J. Factors affecting the
thermal performance of heat pipe – a review. Therm Eng 2011;37A:4143–6.
[165] Krishna J, Kishore PS, Solomon AB. Heat pipe with nano enhanced-PCM for
electronic cooling application. Exp Therm Fluid Sci 2017;81:84–92. http://dx.doi.
org/10.1016/j.expthermflusci.2016.10.014.
[166] Shukla KN. Heat pipe for aerospace applications—an overview. J Electron Cool
Therm Control 2015;5:1–14. http://dx.doi.org/10.4236/jectc.2015.51001.
[167] Pise GA, Salve SS, Pise AT, Pise AA. Investigation of solar heat pipe collector using
nanofluid and surfactant. Energy Procedia 2016;90:481–91. http://dx.doi.org/10.
1016/j.egypro.2016.11.215.
[168] Wang C, Chen J, Qiu S, Tian W, Zhang D, Su GH. Performance analysis of heat pipe
radiator unit for space nuclear power reactor. Ann Nucl Energy 2017;103:74–84.
http://dx.doi.org/10.1016/j.anucene.2017.01.015.
[169] Rao Z, Wang S, Wu M, Lin Z, Li F. Experimental investigation on thermal man-
agement of electric vehicle battery with heat pipe. Energy Convers Manag
2013;65:92–7. http://dx.doi.org/10.1016/j.enconman.2012.08.014.
[170] Fasula C. Oscillating heat pipes (OHP). University of Rhode Island; 2009.
[171] Burban G, Ayel V, Alexandre A, Lagonotte P, Bertin Y, Romestant C. Experimental
investigation of a pulsating heat pipe for hybrid vehicle applications. Appl Therm
Eng 2013;50:94–103. http://dx.doi.org/10.1016/j.applthermaleng.2012.05.037.
[172] Rao Z, Huo Y, Liu X. Experimental study of an OHP-cooled thermal management
system for electric vehicle power battery. Exp Therm Fluid Sci 2014;57:20–6.
http://dx.doi.org/10.1016/j.expthermflusci.2014.03.017.
[173] Putra N, Ariantara B, Pamungkas RA. Experimental investigation on performance
of lithium-ion battery thermal management system using flat plate loop heat pipe
for electric vehicle application. Appl Therm Eng 2016;99:784–9. http://dx.doi.
org/10.1016/j.applthermaleng.2016.01.123.
[174] Zhao R, Gu J, Liu J. An experimental study of heat pipe thermal management
system with wet cooling method for lithium ion batteries. J Power Sources
2015;273:1089–97. http://dx.doi.org/10.1016/j.jpowsour.2014.10.007.
[175] Ye Y, Shi Y, Saw LH, Tay AAO. Performance assessment and optimization of a heat
pipe thermal management system for fast charging lithium ion battery packs. Int J
Heat Mass Transf 2016;92:893–903. http://dx.doi.org/10.1016/j.
329
Energy Conversion and Management 150 (2017) 304–330
ijheatmasstransfer.2015.09.052.
[176] Tran T-H, Harmand S, Sahut B. Experimental investigation on heat pipe cooling for
hybrid electric vehicle and electric vehicle lithium-ion battery. J Power Sources
2014;265:262–72. http://dx.doi.org/10.1016/j.jpowsour.2014.04.130.
[177] Tran TH, Harmand S, Desmet B, Filangi S. Experimental investigation on the
feasibility of heat pipe cooling for HEV/EV lithium-ion battery. Appl Therm Eng
2014;63:551–8. http://dx.doi.org/10.1016/j.applthermaleng.2013.11.048.
[178] Wang Q, Jiang B, Xue QF, Sun HL, Li B, Zou HM, et al. Experimental investigation
on EV battery cooling and heating by heat pipes. Appl Therm Eng 2015;88:54–60.
http://dx.doi.org/10.1016/j.applthermaleng.2014.09.083.
[179] Greco A, Cao D, Jiang X, Yang H. A theoretical and computational study of li-
thium-ion battery thermal management for electric vehicles using heat pipes. J
Power Sources 2014;257:344–55. http://dx.doi.org/10.1016/j.jpowsour.2014.02.
004.
[180] Ye Y, Saw LH, Shi Y, Tay AAO. Numerical analyses on optimizing a heat pipe
thermal management system for lithium-ion batteries during fast charging. Appl
Therm Eng 2015;86:281–91. http://dx.doi.org/10.1016/j.applthermaleng.2015.
04.066.
[181] Rao Z, Wang S. A review of power battery thermal energy management. Renew
Sustain Energy Rev 2011;15:4554–71. http://dx.doi.org/10.1016/j.rser.2011.07.
096.
[182] Yan J, Wang Q, Li K, Sun J. Numerical study on the thermal performance of a
composite board in battery thermal management system. Appl Therm Eng
2016;106:131–40. http://dx.doi.org/10.1016/j.applthermaleng.2016.05.187.
[183] Quanying Y, Chen L, Lin Z. Experimental study on the thermal storage perfor-
mance and preparation of paraffin mixtures used in the phase change wall. Sol
Energy Mater Sol Cells 2008;92:1526–32. http://dx.doi.org/10.1016/j.solmat.
2008.07.002.
[184] Karimi G, Azizi M, Babapoor A. Experimental study of a cylindrical lithium ion
battery thermal management using phase change material composites. J Energy
Storage 2016. http://dx.doi.org/10.1016/j.est.2016.08.005.
[185] Samimi F, Babapoor A, Azizi M, Karimi G. Thermal management analysis of a Li-
ion battery cell using phase change material loaded with carbon fibers. Energy
2016;96:355–71. http://dx.doi.org/10.1016/j.energy.2015.12.064.
[186] Babapoor A, Azizi M, Karimi G. Thermal management of a Li-ion battery using
carbon fiber-PCM composites. Appl Therm Eng 2015;82:281–90. http://dx.doi.
org/10.1016/j.applthermaleng.2015.02.068.
[187] Malik M, Dincer I, Rosen MA. Review on use of phase change materials in battery
thermal management for electric and hybrid electric vehicles. Int J Energy Res
2016;40:1011–31. http://dx.doi.org/10.1002/er.3496.
[188] Goli P, Legedza S, Dhar A, Salgado R, Renteria J, Balandin AA. Graphene-en-
hanced hybrid phase change materials for thermal management of Li-ion batteries.
J Power Sources 2014;248:37–43. http://dx.doi.org/10.1016/j.jpowsour.2013.08.
135.
[189] Shirazi AHN, Mohebbi F, Azadi Kakavand MR, He B, Rabczuk T. Paraffin nano-
composites for heat management of lithium-ion batteries: a computational in-
vestigation. J Nanomater 2016;2016:1–10. http://dx.doi.org/10.1155/2016/
2131946.
[190] Javani N, Dincer I, Naterer GF, Rohrauer GL. Modeling of passive thermal man-
agement for electric vehicle battery packs with PCM between cells. Appl Therm
Eng 2014;73:305–14. http://dx.doi.org/10.1016/j.applthermaleng.2014.07.037.
[191] Wang Z, Zhang Z, Jia L, Yang L. Paraffin and paraffin/aluminum foam composite
phase change material heat storage experimental study based on thermal man-
agement of Li-ion battery. Appl Therm Eng 2015;78:428–36. http://dx.doi.org/10.
1016/j.applthermaleng.2015.01.009.
[192] Alipanah M, Li X. Numerical studies of lithium-ion battery thermal management
systems using phase change materials and metal foams. Int J Heat Mass Transf
2016;102:1159–68. http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.07.010.
[193] Li WQ, Qu ZG, He YL, Tao YB. Experimental study of a passive thermal manage-
ment system for high-powered lithium ion batteries using porous metal foam sa-
turated with phase change materials. J Power Sources 2014;255:9–15. http://dx.
doi.org/10.1016/j.jpowsour.2014.01.006.
[194] Qu ZG, Li WQ, Tao WQ. Numerical model of the passive thermal management
system for high-power lithium ion battery by using porous metal foam saturated
with phase change material. Int J Hydrogen Energy 2014;39:3904–13. http://dx.
doi.org/10.1016/j.ijhydene.2013.12.136.
[195] Hussain A, Tso CY, Chao CYH. Experimental investigation of a passive thermal
management system for high-powered lithium ion batteries using nickel foam-
paraffin composite. Energy 2016;115:209–18. http://dx.doi.org/10.1016/j.
energy.2016.09.008.
[196] Jiang G, Huang J, Fu Y, Cao M, Liu M. Thermal optimization of composite phase
change material/expanded graphite for Li-ion battery thermal management. Appl
Therm Eng 2016;108:1119–25. http://dx.doi.org/10.1016/j.applthermaleng.
2016.07.197.
[197] Ling Z, Chen J, Fang X, Zhang Z, Xu T, Gao X, et al. Experimental and numerical
investigation of the application of phase change materials in a simulative power
batteries thermal management system. Appl Energy 2014;121:104–13. http://dx.
doi.org/10.1016/j.apenergy.2014.01.075.
[198] Yan J, Li K, Chen H, Wang Q, Sun J. Experimental study on the application of
phase change material in the dynamic cycling of battery pack system. Energy
Convers Manag 2016;128:12–9. http://dx.doi.org/10.1016/j.enconman.2016.09.
058.
[199] Lv Y, Yang X, Li X, Zhang G, Wang Z, Yang C. Experimental study on a novel
battery thermal management technology based on low density polyethylene-en-
hanced composite phase change materials coupled with low fins. Appl Energy
2016;178:376–82. http://dx.doi.org/10.1016/j.apenergy.2016.06.058.
H. Liu et al.
[200] Wu W, Yang X, Zhang G, Ke X, Wang Z, Situ W, et al. An experimental study of
thermal management system using copper mesh-enhanced composite phase
change materials for power battery pack. Energy 2016;113:909–16. http://dx.doi.
org/10.1016/j.energy.2016.07.119.
[201] Azizi Y, Sadrameli SM. Thermal management of a LiFePO4 battery pack at high
temperature environment using a composite of phase change materials and alu-
minum wire mesh plates. Energy Convers Manag 2016;128:294–302. http://dx.
doi.org/10.1016/j.enconman.2016.09.081.
[202] Lin C, Xu S, Chang G, Liu J. Experiment and simulation of a LiFePO4 battery pack
with a passive thermal management system using composite phase change mate-
rial and graphite sheets. J Power Sources 2015;275:742–9. http://dx.doi.org/10.
1016/j.jpowsour.2014.11.068.
[203] Balandin Aa. Thermal properties of graphene and nanostructured carbon mate-
rials. Nat Mater 2011;10:569–81. http://dx.doi.org/10.1038/nmat306.
[204] Wu W, Yang X, Zhang G, Chen K, Wang S. Experimental investigation on the
thermal performance of heat pipe-assisted phase change material based battery
thermal management system. Energy Convers Manag 2017;138:486–92. http://dx.
doi.org/10.1016/j.enconman.2017.02.022.
[205] Ling Z, Wang F, Fang X, Gao X, Zhang Z. A hybrid thermal management system for
lithium ion batteries combining phase change materials with forced-air cooling.
Appl Energy 2015;148:403–9. http://dx.doi.org/10.1016/j.apenergy.2015.03.
080.
[206] Hémery C-V, Pra F, Robin J-F, Marty P. Experimental performances of a battery
thermal management system using a phase change material. J Power Sources
2014;270:349–58. http://dx.doi.org/10.1016/j.jpowsour.2014.07.147.
[207] Rao Z, Wang Q, Huang C. Investigation of the thermal performance of phase
change material/mini-channel coupled battery thermal management system. Appl
Energy 2016;164:659–69. http://dx.doi.org/10.1016/j.apenergy.2015.12.021.
[208] Zhao J, Lv P, Rao Z. Experimental study on the thermal management performance
of phase change material coupled with heat pipe for cylindrical power battery
pack. Exp Therm Fluid Sci 2017;82:182–8. http://dx.doi.org/10.1016/j.
330
Energy Conversion and Management 150 (2017) 304–330
expthermflusci.2016.11.017.
[209] Wang Q, Rao Z, Huo Y, Wang S. Thermal performance of phase change material/
oscillating heat pipe-based battery thermal management system. Int J Therm Sci
2016;102:9–16. http://dx.doi.org/10.1016/j.ijthermalsci.2015.11.005.
[210] Zhao J, Rao Z, Liu C, Li Y. Experiment study of oscillating heat pipe and phase
change materials coupled for thermal energy storage and thermal management.
Int J Heat Mass Transf 2016;99:252–60. http://dx.doi.org/10.1016/j.
ijheatmasstransfer.2016.03.108.
[211] Vlahinos A, Pesaran AA. Energy efficient battery heating in cold climates. SAE
tech. pap., vol. 2002-01-19; 2002. http://dx.doi.org/10.4271/2002-01-1975.
[212] Ji Y, Wang CY. Heating strategies for Li-ion batteries operated from subzero
temperatures. Electrochim Acta 2013. http://dx.doi.org/10.1016/j.electacta.
2013.03.147.
[213] Stuart TA, Hande A. HEV battery heating using AC currents. J Power Sources
2004. http://dx.doi.org/10.1016/j.jpowsour.2003.10.014.
[214] Zhu J, Sun Z, Wei X, Dai H. An alternating current heating method for lithium-ion
batteries from subzero temperatures. Int J Energy Res 2016;40:1869–83. http://
dx.doi.org/10.1002/er.3576.
[215] Wang C-Y, Zhang G, Ge S, Xu T, Ji Y, Yang X-G, et al. Lithium-ion battery structure
that self-heats at low temperatures. Nature 2016;529:515–8. http://dx.doi.org/10.
1038/nature16502.
[216] Zhang G, Ge S, Xu T, Yang X-G, Tian H, Wang C-Y. Rapid self-heating and internal
temperature sensing of lithium-ion batteries at low temperatures. Electrochim
Acta 2016;218:149–55. http://dx.doi.org/10.1016/j.electacta.2016.09.117.
[217] Al Hallaj S, Selman JR. A novel thermal management system for electric vehicle
batteries using phase-change material. J Electrochem Soc 2000;147:3231. http://
dx.doi.org/10.1149/1.1393888.
[218] Huo Y, Rao Z. Investigation of phase change material based battery thermal
management at cold temperature using lattice Boltzmann method. Energy Convers
Manag 2017;133:204–15. http://dx.doi.org/10.1016/j.enconman.2016.12.009.