Journal of Energy Storage 35 (2021) 102301

Contents lists available at ScienceDirect

Journal of Energy Storage

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

An improved mini-channel based liquid cooling strategy of prismatic

LiFePO4 batteries for electric or hybrid vehicles
K. Monika a, Chanchal Chakraborty b, Sounak Roy b, Srikanta Dinda c, Satyapaul A. Singh c,
Santanu Prasad Datta a, *
a
Department of Mechanical Engineering, BITS Pilani, Hyderabad campus, 500078, India
b
Department of Chemistry, BITS Pilani, Hyderabad campus, 500078, India
c
Department of Chemical Engineering, BITS Pilani, Hyderabad campus, 500078, India

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

Keywords: Li-ion batteries are one of the most widely used energy storage devices owing to their relatively high energy
Prismatic li-ion battery density and power, yet they confront heating issues that lead to electrolyte fire and thermal runaway, especially
Rectangular mini channel in automotive applications. A well-designed thermal management system is necessary to mitigate the thermal
Liquid cooling
issues occurring in high charge/discharge conditions. Keeping this in view, an ingeniously designed rectangular
Thermal management
mini-channel cold plate is proposed to sandwich in between two consecutive 7Ah prismatic lithium iron phos­
Channel optimization
phate (LiFePO4) batteries with a provision of coolant flow through the mini-channels across the cold plate to
form a battery module. A numerical model for the varying channel number, channel width, coolant flow rate,
coolant and ambient temperature, etc. to uphold the battery module temperature within the range of 25 ◦ C-40 ◦ C
is developed in COMSOL Multiphysics 5.4. A detailed thermodynamic analysis suggests that a cold plate
comprising 5 mini channels of width 4 mm with parallel flow design, and water entry near to the charging port
with a flow rate of 0.003 kg.s − 1 and temperature of 25 ◦ C as the ideal trade-off between heat transfer and
pressure drop for better thermal management across the battery module. A uniform heat propagation in longi­
tudinal direction justifies the optimum design of the cold plate.

Till date, many mathematical models have been developed by the re­
1. Introduction searchers to estimate the heat generation rate and the C-rate as the ratio
of the discharge and the theoretic current at a rated nominal capacity of
With the impending environmental concern from traditional internal the battery, namely the equivalent circuit model [5], single-particle
combustion engine (ICE) based cars, the automobile industry is model [6], 1D, 2D and 3D electrothermal model [7,8]. Therefore, a
concentrating on the boulevard of electric vehicles (EVs) or hybrid profound understanding of the electrochemical and thermal character­
electric vehicles (HEVs) on a roll in the present decade. Batteries are one istics of a large capacity battery is required [9].
of the significant sources of the energy storage unit for EVs or HEVs [1]. From an experimental investigation on a Li-ion pouch cell of 10Ah
Presently, a series of batteries like lead-acid, NiMH, NiCad and Li-ion are for several sets of average cell temperature and edge-to-edge tempera­
incorporated in EVs and HEVs to empower the powertrains. Amongst ture difference under certain constant-current, pulse and drive cycle
these, the demand for Li-ion batteries is overgrowing because of its tests it is observed that the effects of these cycle tests on the battery are
lower self-discharging rate, long life, eco-friendly nature, higher power, larger at 15 ◦ C, while negligible at 35 ◦ C [10]. Conversely, extended
and energy density [2]. The operating temperature of the Li-ion battery exposure to higher environmental temperatures can lead to a loss in
module and the variation of temperature between the individual bat­ capacity and power of battery due to the dependency of electrochemical
teries significantly impact the functioning of the battery in terms of its reactions on temperature. For this, the maximum and minimum
life cycle, usable limit, and safety [3]. On the other hand, a considerable enduring temperature of a Li-ion battery pack is specified to be in the
quantity of heat is produced from the battery module throughout the range of 60 ◦ C to − 20 ◦ C, respectively [11]. However, the actual oper­
working cycle due to exothermic reactions, and internal resistance [4]. ating temperature is well below or above this range which may

* Corresponding author: Department of Mechanical Engineering, BITS Pilani- Hyderabad Campus, Jawahar Nagar, Kapra Mandal, Hyderabad 500 078, Telangana,
India.
E-mail address: [email protected] (S.P. Datta).

https://doi.org/10.1016/j.est.2021.102301
Received 27 August 2020; Received in revised form 4 December 2020; Accepted 4 January 2021
Available online 21 January 2021
2352-152X/© 2021 Elsevier Ltd. All rights reserved.
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Nomenclature gen generation

j joule heat
∆S entropy change (W ◦ C − 1) max maximum
∆P pressure drop across the plate (Pa) min minimum
A surface area (m2) p polarisation
cp specific heat at constant pressure (J kg− 1 ◦ C − 1
) r reaction
Dh hydraulic diameter (m) t total resistance
E open circuit voltage (V) w water
F Faraday constant w, in water inlet
f friction factor w, out water outlet
h heat transfer coefficient (W m − 2 ◦ C − 1)
I current per unit volume (A m − 3) Greek Symbols
j colburn factor µ dynamic viscosity (Pa s)
K thermal conductivity (W m − 1 ◦ C − 1) β cooling performance factor
L characteristic length (m) ρ density (kg m − 3)
Nu Nusselt number σ standard deviation ( ◦ C)
Ppump pump power (W) ω velocity (m s − 1)
P static pressure (Pa) ∇ gradient operator
Pr Prandtl number Acronyms
Re Reynolds number 1D one-dimensional
t time (s) 2D two-dimensional
T static temperature ( ◦ C) 3D three-dimensional
V operating voltage (V) BTM battery thermal management
Z standard value ECT electrochemical thermal model
ṁ mass flow rate (kg s − 1) EVs electric vehicles
Q̇ heat generation rate (W) HEVs hybrid electric vehicles
→v average velocity (m s − 1) ICE internal combustion engine
LFP lithium iron phosphate
Subscripts
LiFePO4 lithium iron phosphate
a ambient
MCMB mesocarbon micro beads
avg average temperature
NiCad nickel cadmium
b battery
NiMH nickel metal hydride
bm battery module
PCM phase change material
c channel
SOC state of charge
e real

contribute to thermal runaway in summer, and cold start in winter. In a Tesla Model S, BMW i-8, Ford Focus, whereas air cooling in Nissan leaf
battery pack, even if one cell gets degraded due to meagre thermal e-plus.
management, the overall efficiency of the pack gets reduced, and an Although the air cooling has the upsides of low expense, lesser
electrical imbalance arises across the cells [12]. weight, ample supply, and simplicity in maintenance, it is not suitable
As safety is of supreme importance while using Li-ion batteries in under extreme conditions such as higher discharge rates, cold and hot
automobiles, several research efforts are being devoted in developing an weather conditions due to limited heat transfer coefficient [19]. The
energy-efficient thermal control system for maintaining temperature provision of thermal runaway persists while the air attains the same
homogeneity amongst the batteries in a module or pack [13]. The temperature as surroundings, which again requires additional cooling.
thermal management techniques mainly include forced convection by Zhao et al. [20] investigated the temperature regulation of a cylindrical
air, water, mineral oil, and phase change materials (PCMs). The air battery module using air as a cooling medium on various parameters
cooling and liquid cooling techniques are classified into active and such as a gap in between batteries, type of ventilation, air velocity,
passive systems, whereas the phase change materials usually come ambient conditions, flow direction, number of cells present in a row and
under the passive system. A comparative investigation between three diameter of battery. Incidentally, they have observed that with the in­
cooling methods mentioned above suggests that liquid cooling is effec­ crease in air inlet velocity, there is a rise in maximum temperature. They
tive in reducing the peak temperature, while PCM cooling provides have also recommended keeping moderate spacing in between batteries
uniform heat distribution [14]. Besides this, under sub-zero or low at­ to decrease the overall temperature rise across the battery module.
mospheric temperatures, air cooling is opted over the other cooling Therefore, the liquid is preferred as a cooling substance because of its
methods. Furthermore, the authors clarified that it is vital to choose an higher heat capacity and efficient heat removal rate at reduced pumping
appropriate cooling technique during low operating or environmental power compared to the air [21].
temperatures and elevated discharge rates. In addition to the conven­ In designing a liquid cooling system, several sizing and rating pa­
tional methods, nanofluids [15], refrigerants [16], boiling liquids [17], rameters such as size, expense, weight, the contact resistance between
and liquid metal [18] have gained further attention in cooling the bat­ plate and battery, variation in temperature between the batteries,
tery module effectively. Battery thermal management (BTM) by utilizing leakage, overall module efficiency, and pumping power are to be
heat pipes, cold plates, or a combination of both is also explored for both considered [21]. In a liquid cooling system, different types of mini or
the active or passive cooling by several researchers. With the advent of microchannel cold plate [22], fin-plate [23] are sandwiched in between
new cooling technologies, most of the commercial EVs or HEVs are two consecutive batteries of a battery module. They are allowed to pass
implementing thermo-regulation strategies such as liquid cooling for different coolants like water, thermal oil, glycol solution, refrigerant,

2
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Table 1
Summary of the available literature to design an energy-efficient BTMs. The references cited in this figure are [36–41]

etc. to study the performance of the battery module for a wide range of operate at any diverse climatic conditions. An ideal design of a cold plate
operating conditions. Zhao et al. [24] suggests a mini-channel water includes many constraints such as the number of plates, width of the
cooling scheme to analyse the thermal behaviour of a cylindrical battery plate, coolant mass flow rate, flow direction, coolant temperature, etc.
module for different channel number, inlet size, the direction of airflow, [27,28]. An optimization technique such as orthogonal array is utilized
and optimized the sizing and rating parameters to maintain the battery by Patil et al. [29] to identify the primary and secondary factors in
module surface temperature under 40 ◦ C. Also, heat dissipation poten­ designing the double cold plate along with flow patterns and channel
tial is improved at the beginning and followed by reduced at a steady geometry. Sheng et al. [30] numerically identifies that the inlet and the
coolant mass flow rate along with an increase in entrance level channel outlet positions of the channel for a unique serpentine cold plate along
size. with the pathway of liquid flow could effectively reduce the rising
PCM based thermal management strategy is another popular tech­ temperature range of the battery module at a desirable range. Patil et al.
nique for temperature regulation owing to their greater heat capacity [31] numerically investigates the cooling efficiency of a U-turn cold
and corresponding thermal conductivity. To increase the thermal con­ plate at higher discharge rates under turbulent flow along with other
ductivity and thereby to enhance the heat transfer rate, a combined PCM design and operating parameters. Results indicate that there is an
module with various fin structures such as trapezoidal, rectangular, and enhancement in cooling of batteries for a coverage area ratio and hy­
I-shape, etc. are compared with standalone PCM module [25]. draulic diameter of 0.750 and 1.54 mm, respectively.
Comparatively, the optimized module achieves superior performance Although there is significant progress in enhancing the thermal ho­
and reduces the peak temperature by 2.38% and 9.28% at 2C and 3C mogeneity of pouch type [32] or cylindrical battery [33] modules, as
discharge rates, respectively. It may be noted that the PCM cooling has indicated by the above-cited literature and Table 1, thermal in­
the possibility of available heat capacity being exhausted at a high homogeneity still remains, implying the need to conduct more research
discharge rate and extreme temperatures, which lead to thermal insta­ in determining suitable strategies of temperature control of prismatic
bility of the battery module and provision of PCM leakage. To overcome Li-ion battery module. A series of cold plates with various designs like
these complications, Lopez et al. [26] assesses the combination of serpentine channel, u-bend channel, etc. are explored for battery ther­
different PCM structures such as lauric acid and paraffin wax encapsu­ mal management. However, they lead to a more significant pressure
lated in graphite with chilled cold plates to improve the thermal control drop and competitive temperature reduction [34]. Hence, it is essential
of batteries. Authors observe that the cell spacing and the PCM material to optimally design a cold plate with simpler geometry for minimum
properties have a significant influence on reducing the temperature pressure drop and to uphold the prismatic battery module temperature
gradient across the module. Moreover, incorporating an active cooling within 25 ◦ C-40 ◦ C. The primary role of a cold plate is to control un­
method along with PCM, contributes to an increased mass and volume of desirable temperature distribution across the module when operating
the cooling system by 55% and 60% respectively. Hence, further under higher currents and temperatures.
enhancement of PCM structures is required to commercialize them in Therefore, an ingeniously designed rectangular mini channel cold
automotive applications. plate is proposed to sandwich in between two consecutive 7Ah lithium
Finally, the above literature survey concludes that the liquid cooling iron phosphate (LiFePO4) batteries with a provision of liquid coolant
is substantially better compared to the air or PCM based BTM techniques flow through the mini-channels of the cold plate, and the absorption of
due to higher heat transfer rate, better temperature stability, ease of internal heat generated by the batteries. For this, a 3D electrochemical-
operation, absence of forced convection arrangement, and efficacy to thermal model of a scaled-down battery module comprising five

3
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Table 2 of a carbonate-based electrolyte, Graphite electrode LixC6 mesocarbon

The geometry and dimension of the cold plate placed between two microbeads (MCMB) as anode and LFP (Lithium iron phosphate) elec­
consecutive batteries. trode LiFePO4 as the cathode, with a capacity of 7 Ah and a maximum
Parameter Value discharge current of 35A. The heat produced from the batteries is
Channel type Rectangular
assumed as uniform, along with another assumption of an anisotropic
Material Aluminium variation of thermal conductivity. Whereas, the density and the heat
Dimensions 63 mm× 118 mm × 2 mm capacity across the battery are assumed as homogeneous [35]. A battery
Depth of flow 1 mm module containing five batteries and a cold plate sandwiched in between
Channel width 2–6 mm
two consecutive batteries with coolant flow passages is indigenously
No. of channels 2–7
designed (Table 2), and developed in COMSOL Multiphysics 5.4 (Fig. 1).
For the ease of the simulation and to minimize the computational time,
prismatic Li-ion batteries is developed numerically by using COMSOL only half of the module is simulated, and the rest is extended due to the
Multiphysics 5.4. Design parameters in the cold plate, such as the width symmetric nature of the proposed conjugate heat transfer model.
of the channel, the number of channels, location of inlet/outlet and flow
orientation in a cold plate, are varied to investigate their impact on the 2.2. Governing equations
variation of temperature across the battery module. In addition, the
effect of operating parameters on the overall performance of the battery The energy balance equation comprising the batteries along with the
module, such as coolant inlet temperature, mass flow rate, battery cold plates is expressed as
discharge rate, and atmospheric temperature, are also explored. The
main aim of the present work is to develop a general guideline in ana­
∂( )
ρ c T = ∇.(Kb ∇Tb ) + Q̇g (1)
lysing a cold plate design and its effect on the performance of the battery ∂t b p,b b
module for varying conditions. A large amount of heat is generated inside the battery due to
The framework of present study is outlined as follows: Section 2 exothermic chemical reactions during continuous charging or dis­
defines the equations governing the battery and coolant flow and nu­ charging cycles, and it can be divided into three parts namely, reaction
merical procedure, Section 3 elucidates the results obtained by varying heat, Q̇r , polarisation heat, Q̇p , and joule heat, Q̇j [42,43].
design and operating parameters, while the significant findings are
abridged in Section 4. Q̇g = Q̇r + Q̇p + Q̇j (2)

Reaction heat is generated due to internal chemical reactions and is

2. Methodology expressed as:

2.1. Physical problem I

Q̇r = − Tb ΔS (3)
nF
A prismatic battery module structure that is widely used in BMW i3, Polarization refers to the effect that reduces the performance of
Volkswagen, is selected for the present investigation. Each cell with a batteries. This occurs due to the deviation of electrode potential from its
dimension of 63 mm width, 118 mm height, and 13 mm thick comprises equilibrium value and is expressed as follows:

Fig. 1. Schematic representation of (a) a prismatic LiFePO4 battery, (b) cold plate with flow passages and (c) arrangement of the cold plate in between two
consecutive batteries.

4
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Table 3 water are [45]:

Thermodynamic properties of the materials used in simulation [44].
1 ◦ 1◦
∇.→
v =0 (7)
Material ρw (kg m − cw (J kg− C− Kw (W m − C− μw (Pa s)
3 1 1
) ) ) [ → ]
∂v
Coolant 998.2 4182 0.6 0.0001003 ρw + (→
v .∇)→
v = − ∇P + μ∇2 →
v (8)
∂t
(Water)
Cold plate (Al) 2719 871 202 –
Battery 2500 1000 3 ∂Tw
– ρw cw + ∇.(ρw cw →
v Tw ) = ∇.(Kw ∇Tw ) (9)
∂t

Table 4 2.3. Boundary conditions

Selected boundary conditions and operating ranges.
Table 4 represents the boundary conditions utilized for the present
Parameter Value
numerical investigation. The coolant mass flow rate at channel inlet is
Coolant Water considered as one of the boundary conditions, whereas pressure of 0 Pa
Coolant temperature ( ◦ C) 5–35
Ambient temperature ( ◦ C) 10–50
is defined as another boundary condition at the outlet. At the inner wall
Discharge rate (C) 1–5 of mini-channels, the no-slip shear condition is applied. A constant
Inlet coolant mass flow rate (kg s − 1) 0.001–0.005 current charge/discharge process is utilized in the present study. As an
Pressure outlet (Pa) 0 initial condition, it is assumed that both the coolant and the ambient are
Heat transfer coefficient (Wm2 ◦ C − 1) 5
at a constant temperature of 25 ◦ C. The battery module used for the
Flow type Laminar (< 2300)
present investigation is discharged at different C-rates (i.e., 1C to 5C).
The resistance due to contact between battery and aluminium plate is
Q̇p = I 2 Rp = I 2 (Rt − Re ) (4) assumed to be negligible. As per the theory of heat transfer, the
convective type boundary condition is considered on the outer walls of
Joule heat occurs due to transfer of current across the internal the module.
resistance of a cell and is expressed as:

Q̇j = I(E − V) = I 2 Re (5) 2.4. Numerical method

Therefore, the heat generated by the battery per unit volume can be Fig. 2 represents the numerical procedure of the present model. Most
calculated as follows: of the thermo-physical parameters utilized in the present paper are
I adopted from existing literature and COMSOL material library [44,46].
Q̇g = − TΔS + I 2 Rt (6) The electrochemical thermal (ECT) modelling of the battery module is
nF
performed by integrating the two models simultaneously. Initially, the
In the 3D thermal model, the heat generated from the active battery
heat produced by the cell is determined through the 1D ECT model by
material is assumed as homogenous. Here, water is considered as the
computing the electrochemical reactions taking place within the battery
coolant which travels across the inlet/outlet of the aluminium mini-
at regular intervals depending on the average temperature obtained
channel cold plate. The thermophysical properties [44] of the active
from the 3D model. Further, the heat generated per unit volume by the
battery material, cold plate, and water are summarized in Table 3. The
1D model is used as input to the 3D thermal model. The energy balance
mass, momentum and corresponding energy conservation equations for
on the cells is solved to evaluate the temperature variations of the 3D

Fig. 2. The flow chart representing the simulation steps followed in the present investigation.

5
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Table 5 data of a single battery without any active cooling from Ghalkhani et al.
Relative error (%) between consecutive grid sizes of a 5- channelled cold plate. [47] is used to validate the developed algorithm of 3D electrothermal
Grid Average Relative Pressure drop Relative numerical model. Ghalkhani et al. [47] conducts the experiment on a
size Temperature ( ◦ C) error (%) (Pa) error (%) single battery at a fixed ambient temperature of 14 ◦ C, and an infrared
4.36 × 132.1 7.9 27.16 0.29 (IR) camera is used to measure the temperature distribution across the
105 battery surface. Their findings showed that the temperature is higher
1.38 × 136.3 5.09 27.15 0.24 around the tab region compared to the remaining surfaces of the battery.
106 Fig. 3 reveals that the present simulation and experimental results are in
2.38 × 137.4 4.31 27.13 0.18
106
good agreement, and the deviation is less than 10%. However, minor
3.08 × 139.6 2.77 27.12 0.11 differences are observed at the beginning and end of the discharge
106 process between the numerical and experimental data. This occurs due
3.33 £ 143.1 0.33 27.10 0.07 to (a) the parameters used in the present paper are mostly obtained from
106
literature, which may vary from actual parameters of the experimental
3.72 × 143.6 NA 27.08 NA
106 battery, and (b) the assumption of uniform heat generation along the
active battery material does not entirely hold in reality.

2.6. Data reduction

The pressure drop of water across the inlet and outlet of mini-
channel cold plate is estimated as
/
Δpw = pout − pin = fDh ρw 2 Ach 2 2Lt ṁw 2 (10)

Here, the frictional coefficient for laminar flow across a mini-channel

of length, Lt (m), the hydraulic diameter of Dh (m) for a cross-sectional
area Ach (m2), Perimeter P (m) and at a mass flow rate of ṁ(kg.s − 1)
can be estimated as f = 64 /Re. The maximum Reynolds number for a
cold plate with 5 number of channels and 4 mm width at a mass flow rate
of 0.005 kg. s − 1 is observed to be 1941.7, which is less than 2300.
Hence, the laminar flow assumption is valid for all the cases considered.
Here, Re = ρw ωDh /μw ,Dh = 4Ach /P
The amount of heat removed per unit volume from the prismatic Li-
ion cell is calculated with Eq. (11)
( )
Fig. 3. Comparison of temperature difference for simulation and experimental Q̇w = ṁw cp,w Tw,out − Tw,in (11)
[47] test results.
The average heat transfer coefficient (W.m − 2. ◦ C − 1) of the coolant
as a function of the cooling capacity (W) is calculated for a surface area
battery module by defining the coolant boundary conditions. In the 3D of Ac as
conjugate model, the heat transfer in solids and laminar flow modules
are united through multiphysics non-isothermal flow. The Q̇
hw = [ ( w )/ ] (12)
Paradisotime-dependant solver is used with a relative tolerance of 0.001 Ac Tbm,max − Tw,out + Tw,in 2
for all the parameters considered, whereas the time step is set to 1 s so
A non-dimensional factor j/f is considered as it is widely used in
that the simulations can run faster.
evaluating the performance of plate in the heat exchanger and is
calculated as a function of Nusselt number (Nu = hw Lt /Kw )
2.5. Mesh independence test and validation Nu
j= (13)
Re.Pr0.33
The free tetrahedral mesh is used for constructing mesh. To ascertain
the reliability of results, mesh independence test is conducted for all where, Lt is the characteristic length (m), Kw (W.m2. ◦ C − 1) is the
channel designs considered in the present work. However, the results of thermal conductivity of coolant, and Pr is the Prandtl number, which is
a specific case of grid sensitivity test where a 5-channel cold plate is 6.22 for water at 25 ◦ C.
sandwiched between two consecutive batteries with a width of 3 mm The pumping power needed to circulate coolant through the mini-
and at a discharge rate of 5C are reported in Table 5. For this, the mass channel is estimated as
flow rate of coolant is set as 0.001 kg s − 1, whereas the initial coolant
Ppump = ΔPw .V̇ w (14)
and ambient temperature are kept at 25 ◦ C. Six grid numbers varying
from 4.36 × 105 to 3.72 × 106 are generated for the present model. The
where, volume flow rate (m3.s-1) V̇ w = ṁw /ρw (15)
overall deviation in pressure drop and average temperature across the
The thermal behaviour of the battery module is monitored through
module is observed to be less than 0.33% as the grid number exceeds
average temperature (Tavg), the standard deviation of temperature (Tσ ),
3.33 × 106 relative to the maximum grid size 3.72 × 106 (Table 5).
loss in pressure across the inlet and outlet ports of the mini-channel
Therefore, grid number 3.33 × 106 is chosen for further numerical in­
(∆pw).
vestigations to attain accurate results irrespective of the mesh size. It
∫
may be noted that the temperature gradient across the thickness of the TdAc
mini-channel is ignored due to the narrow size and ease of the simula­ A
Tavg = ∫c (15)
tion. However, in a practical situation, there can be a minor effect of the dAc
Ac
heat transfer distribution across the thickness of the cold plate.
Due to limitation of detailed thermo-physical properties about pris­
matic battery and their proprietary nature, the available experimental

6
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 4. Variation of average temperature across the battery module with

different fluids. Fig. 5. Schematic representation of different flow arrangements across
the channels.
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
√∫ ( )2
√ T − Tavg dAc
Tσ = √ Ac
∫ (16)
Ac
dAc

where, Ac represents the surface area (m2), and T is the static temper­
ature ( ◦ C) of the cold plate.

3. Results

The goal of the present investigation is to analyse the cooling

effectiveness of rectangular mini channel cold plate inscribed in a 7Ah
battery module, thereby maintaining the temperature of the battery
module within a safe range of 25 ◦ C-40 ◦ C. The pertaining factors like
the number of channels in the cold plate, channel width, coolant flow
rate and flow direction are estimated based on a detailed heat transfer
analysis. Further, the effect of variation of coolant inlet temperature,
ambient temperature, and battery discharge rate on the overall perfor­
mance of the battery module is also explored.
Fig. 6. Average temperature in the battery module for different flow arrange­
ments across the cold plate.
3.1. Selection of ideal coolant

Before analysing the effect of design and functioning parameters, the limit. It may be noted that the ethylene glycol acts as an antifreeze at
cooling performance of the different types of fluids such as air, engine lower temperatures which is suitable mostly for cold climatic regions.
oil, water, and water: ethylene glycol (EG) of volume ratio 50:50 is Hence for the considered boundary conditions, water is found as an
compared with battery module without any active cooling. A 5-channel excellent coolant to control the temperature rise in the module and is
cold plate with 3 mm channel width at a continuous flow rate of 0.001 chosen for further investigations.
kg s − 1 is chosen for the present investigation. Both the coolant tem­
perature at the channel inlet and ambient temperature are kept at 25 ◦ C. 3.2. Influence of flow direction
The variations of the average battery surface temperature for different
fluids at a discharge rate of 5C is reported in Fig. 4. As the proposed channel is working with a principle of a heat
In the case of the battery without active cooling, the temperature is exchanger, the flow directions of the coolant have some evident
observed to be 34.47 ◦ C, whereas, with air as a cooling medium, the consequence on the overall heat transfer performance of the system.
average surface temperature reaches up to 31.78 ◦ C compared to other Hence, four different flow arrangements of parallel and counter flow
cooling mediums at a fully discharged condition. With oil cooling, the directions with reverse entry points are considered for the present
average temperature reaches 28.64 ◦ C by the end of the discharge cycle. investigation (Fig. 5).
Although oil cooling has more heat transfer capability than air, it is For a fixed geometry of a cold plate, at a constant discharge rate of
acceptable where the effects of pumping power and viscosity are not 5C, the parallel flow arrangements show better heat transfer capability
predominantly considered. Compared to the above cooling mediums, over the counter flow arrangements for a battery module (Fig. 6).
liquid coolants have a high heat transfer coefficient and low viscosity, Alongside, the change in the direction of flow arrangement for parallel
which makes them more suitable for the indirect cooling of batteries. channels does not have any significant effect on the overall performance
Under the constant flow rate, the difference in the average surface of the battery module. Although there is a minimal change in the
temperature of water and water: EG is found to be minimal, i.e. 27.10 ◦ C average surface temperature and standard deviation due to modification
and 27.46 ◦ C, respectively. Although the average temperature of the in the flow direction, their influence is observed on the thermal distri­
battery module is well below the usual operating temperature, i.e. 25◦ C bution (Fig. 7). Incidentally, for the proposed four types of flow
to 40◦ C for present scaled-down investigation, in practice for an actual arrangement, the region near to the location of the inlet experiences
automobile under external influencing factors, it may reach beyond this lesser temperatures while it is higher at the region near to an outlet. This

7
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 7. Temperature distribution across the battery module for different flow directions through the cold plate.

Table 6
Performance of a battery module for different flow directions across the cold
plate.
Parameter Direction 1 Direction 2 Direction 3 Direction 4

Tσ ( ◦ C) 0.65 0.65 0.69 0.69

Tbm, max ( ◦ C) 28.53 28.84 28.77 28.5

Fig. 9. Temperature distribution across the battery module for different loca­
tions of inlet and outlet ports across the cold plate.

increased average temperature of 0.84 ◦ C, 0.07 ◦ C and 0.23 ◦ C compared

to design 1, 3 and 4, respectively (Fig. 9). This phenomenon occurs due
to the following reasons: (a) longer flow travel path, (b) poor flow dis­
tribution along the central channels of the cold plate, and (c) loss of
Fig. 8. Schematic representation of different inlet and outlet port orientations enthalpy of the coolant before reaching the hotspot due to complexity in
across the channels. design (Fig. 10). Further, a detail heat transfer analysis is entirely
against the selection of design 2 (∆P = 300.82 Pa, Tσ = 0.90 ◦ C) and
is due to the fact that the coolant carries away heat generated from the design 3 (∆P = 182.51 Pa, Tσ = 0.85 ◦ C) (Table 7). Therefore, it is wise to
battery module and becomes hot while passing from inlet to outlet decide the inlet and outlet ports of a cold plate amongst design 1 and
concerning the flow direction. In the case of direction 1, hotspots are design 4. For design 1, the Tbm, max and Tσ are noticed to be 2 ◦ C and
observed to be less when compared to all other directions, which can be 0.2 ◦ C lower compared to that of design 4, respectively. Moreover, a
taken as the best choice. Out of all the cases, directions 3 and 4 are found higher pressure drop across the mini channels up to 60.17 Pa discards
to be the worst as most of the modules’ surface area is experiencing the utility of design 4 and recommends design 1 for further assessment.
multiple hotspots. Therefore, to avoid high temperatures near the tab
region, direction 1 is considered as the best choice over direction 2 3.4. Influence of number of channels in a cold plate
despite the Tσ values being same and hence is chosen for further in­
vestigations (Table 6). Fig. 11 exhibits the influence of channel number on the Tavg of the
battery module for a constant width of 3 mm. The number of mini-
channels in a cold plate is varied from 2 to 7 in the cold plate with a
3.3. Influence of location of inlet and outlet port constant water flow rate of 0.001 kg s − 1 and a corresponding Re =
485.4. For a cold plate design with 2 number of channels, the Tavg is
The ideal location of inlet and outlet ports of mini channels are observed to be 27.35 ◦ C, whereas it decreases to 27.05 ◦ C as the number
finalized by randomly selecting four different port orientations without of channels in a cold plate increases to 7 at the end of the discharging
hindering the total volume of a 5-channel cold plate with a constant cycle. It is clearly visible from the figure that the Tavg is reduced with an
width of 3 mm (Fig. 8). At a constant discharge rate of 5C, the design 2 is increase in the number of channels in a single cold plate owing to the
showing inferior performance over the other three designs with an enhancement of the heat transfer area. Fig. 12 represents the velocity

8
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 10. Temperature distribution across the mid-plane of a cold plate for different locations of inlet and outlet ports.

channels increases. During this process, flow resistance increases mainly

Table 7 when the fluid changes its direction around the corners of mini channels,
Performance of a battery module for different locations of inlet and outlet ports leading to a faster reduction in pressure. This resistance to flow reduces
across the cold plate. once the fluid flows into the entire length of the channel, and as the
Parameter Design 1 Design 2 Design 3 Design 4 outlet pressure becomes constant. Eventually, the pressure drop of the
∆P (Pa) 143.1 300.82 182.51 203.27 liquid attains an equilibrium value in the mini channels. The pressure
Tσ ( ◦ C) 0.65 0.93 0.90 0.85 drop with the increasing number of channels across the cold plate is
Tbm, max ( ◦ C) 28.5 29.96 30.27 30.02 reported in Fig. 14. The pressure drop with 3 channelled cold plate is
observed to be lower than that of 4 channelled cold plate. A similar
phenomenon is noticed for 5 and 6 channelled cold plate. These results
indicate that the pressure drop in the even number of channels is more
significant than that of the odd number of channels due to higher flow
resistance instigated by a difference in flow distribution after entrance
point [34]. However, when the number of mini-channels is changed
from 5 to 7, the difference in Tavg is found to be minimal. Besides this, a
higher channel number would contribute to higher manufacturing cost.
It is seen from Fig. 14 that with an increase in the number of channels
from 2 to 7 in cold plate, heat transfer coefficient (hw) reduces by 6.66%,
whereas j/f factor increases due toan increase in surface area for coolant
flowing in the channels. Besides this, a difference in the varying trend of
j/f factor and pressure drop is observed between the even and odd
number of channels due to a difference in their flow resistance. Table 8
exhibits that the difference in maximum temperature (Tbm, max) and Tσ
with the increasing channel number from 4 to 7 is minimal. Whereas, the
drop-in temperature with 2 to 3 number of channels is large compared to
the same with 4 to 7 number of channels due to a difference in the heat
transfer area. However, increasing the channel number in cold plate
Fig. 11. Average temperature across the module for a varying number decreases hw and pumping power supplied. A trade-off must be main­
of channels. tained between the pressure drop and convective heat transfer to attain
better cooling performance of the battery module. Comparatively, 5-
distribution of water across the cold plate for varying channel number. It channel cold plate provides a lower pressure drop resulting in lower
is noticed that when the channel number is odd, the central channel near pumping power which is around 143.1 Pa and 1.43 £ 10− 4 W, respec­
to the location of the inlet carries away more than half of the cooling tively. Hence to be cost-effective and attain a desirable temperature
liquid. Whereas in case of even channel design, higher velocity is drop, a 5-channelled cold plate is chosen for further study.
observed along with the header after the entrance point. Owing to this,
turbulence and flow resistance is generated along the direction of liquid 3.5. Influence of channel width
flow in even channel designs. This implies that the heat carried away by
cooling liquid along the channels of extreme corners is less, which leads The variation of the battery module average temperature for change
to improper velocity distribution and thermal non-uniformity. in the channel width from 2 mm to 6 mm with corresponding Re varying
As the channel number increases, the distribution of fluid along each from 647.25 to 277.40, for a depth of flow and discharge rate fixed to 1
channel decreases, resulting in a pressure drop across the channel length mm and 5C respectively is reported in Fig. 15. It is observed that as the
(Fig. 13). It is noticed that at the beginning of the discharging process, channel width is varied, the variation in Tavg and Tσ at the end of the
the water pressure at the inlet region is more, and it reaches zero at the discharge cycle is only 0.28 ◦ C and 0.10 ◦ C, respectively. This indicates
extreme outlet since the channels are not entirely filled with the liquid. that the channel width has less significance on the temperature rise of
Pressure drop reaches to maximum when the coolant flows along the the module. Notably, for a channel width of 2 mm and 6 mm, throughout
channels, simultaneously the contact surface between the coolant and the discharging process, the Tavg shows the maximum and minimum

9
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 12. Coolant flow distribution across the cold plate for different channel designs.

Fig. 13. Effect of channel design on the pressure distribution across a cold plate.

Table 8
Performance of a battery module for varying channel number in a cold plate.
Number of channels Tbm,max ( ◦ C) Ppump (W) Tσ ( ◦ C)
− 4
2 28.64 2.67 £ 10 0.74
3 28.57 1.81 £ 10− 4 0.67
4 28.53 1.77 £ 10− 4 0.66
5 28.53 1.43 £ 10¡4 0.65
6 28.51 1.51 £ 10− 4 0.63
7 28.53 1.29 £ 10− 4 0.63

83.40% due to lesser turbulence and flow restriction inside the channel.
It may be noted that a narrow channel of 2 mm width leads to a higher
pressure drop due to a lower heat transfer area. In line with this, the hw
decreases up to 55.4% with the increment of the channel width from 2
mm to 6 mm owing to the non-uniform distribution of fluid despite the
reduced energy consumption [48]. The results mentioned above suggest
Fig. 14. Effect of channel design on the heat transfer performance of the that a channel width of 4 mm and corresponding Re of 388.35 as an ideal
cold plate. trade-off between heat transfer and pressure drop with a satisfactory
thermal and statistical index, like Tbm, max = 28.49 ◦ C, Tσ = 0.63 ◦ C and
deviation of around 0.70 ◦ C and 0.60 ◦ C, respectively. It may be noted ∆P = 100.3 Pa. Hence, to enhance overall heat transfer performance
that for the channel width of 5 mm and 6 mm, a similar Tσ is observed across the battery module, a 5-channel cold plate with a channel width
0.62 ◦ C and 0.60 ◦ C respectively. of 4 mm is selected for further investigations.
Pressure drop and pumping power reduce invariably as the channel
width enlarges from 2 mm to 6 mm, as shown in Fig.16 and Table 9. As
the heat transfer surface area increases, the loss in pressure decreases by

10
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 15. Variation of average temperature in the battery module for varying Fig. 17. Average temperature in the battery module for varying coolant mass
channel width across a cold plate. flow rate.

mass flow rate of 0.002 kg.s − 1 may not always be adequate to maintain
the temperature gradient of less than 5 ◦ C in between individual bat­
teries. It may be noted that the mass flow rate has to be adjusted ac­
cording to the atmospheric conditions and cooling requirements of the
module or pack [45]. Increasing the mass flow rate decreases the
channel number in the cold plate resulting in the lower material
requirement, and cost to manufacture.
The concentration of hot regions is noticed to be more near the
bottom of the battery at lower mass flow rates. However, with an in­
crease in mass flow rate, the distribution becomes homogenous with
fewer hot regions as shown in Figure 18. Besides this, the central battery,
which is surrounded by two cold plates experiences effective cooling
compared to the cell exposed to heat flux conditions and covered by a
single cold plate. The Tbm, max and Tσ exhibits a significant decrease with
mass flow rate after a certain extent (Table 10). With the increasing
water flow rate, the pressure drop is noticed to be increasing from 100.3
Pa to 1145 Pa, whereas hw is enhanced by 61.97%. Higher the mass flow
Fig. 16. Effect of variation of channel width on the heat transfer and pressure rate, higher is the pressure loss leading to additional energy consump­
drop across the cold plate. tion and operating cost, whereas it varies inversely with j/f factor due to
reduced contact area. Therefore, a trade-off must be maintained be­
tween pressure drop and operational cost. It is evident from Figure 19
Table 9 and Table 10 that the change in the overall reduction of Tbm, max and j/f
Performance of a battery module for increasing channel width. factor is insignificant beyond the flow rate of 0.003 kg.s − 1. Therefore, a
water flow rate of 0.003 kg.s − 1 and a corresponding Re of 1165 is found
Width of channels (mm) Re Tbm,max ( ◦ C) Ppump (W) Tσ ( ◦ C)
to be adequate and advantageous in maintaining a proper temperature
2 647.25 28.64 2.96 £ 10− 4 0.70 distribution despite an increase in the power and pumping cost, hence is
3 485.44 28.53 1.43 × 10− 4 0.65
considered as the optimal choice for further simulations. However, an
4 388.35 28.49 1.00 £ 10¡4 0.63
5 323.63 28.49 6.95 £ 10− 5 0.62 appropriate flow rate has to be maintained to reduce the pressure drop
6 277.40 28.48 4.91 £ 10− 5 0.60 and enhance hw for better cooling of batteries in actual operating
conditions.

3.6. Influence of mass flow rate

3.7. Influence of coolant temperature
A suitable flow rate is necessary to operate the module under desired
The coolant temperature has a considerable influence on battery
temperature range to meet the safety requirements of the considered
thermal behaviour [49,50]. The integration of cooling systems under
boundary conditions. This section varies mass flow rate from 0.001 kg.s
− 1 high environmental temperatures and operating conditions improve
to 0.005 kg.s − 1 to determine the cooling efficiency of a 5-channelled
battery performance. However, lower coolant temperatures have an
cold plate each with a width of 4 mm and the results are reported in
undesirable effect on the battery thermal performance. In this section,
Fig. 17. The coolant flow in the present work is maintained to be under
the coolant temperature is varied from 5 ◦ C to 35 ◦ C with an incremental
the laminar regime. The continuous increase in coolant mass flow rate
difference of 5 ◦ C to explore the overall heat transfer performance across
decreases the outlet temperature of coolant linearly as per Eq. (11). After
the battery module with a 5-channelled cold plate (Fig. 20). The initial
the discharge process, while the flow rate increases from 0.001 kg.s − 1
temperature of the battery and the ambient temperature are maintained
to 0.005 kg.s − 1, the drop in Tavg and Tσ are observed up to 1.03 ◦ C and
at 25 ◦ C.
0.34 ◦ C, respectively. A minimum flow rate must be maintained to
As the coolant temperature increases from 5 ◦ C to 35 ◦ C, Tavg and Tσ
control the temperature of the battery module within a desirable range.
reaches to 35.94 ◦ C and 3.12 ◦ C, respectively. For a water inlet tem­
For example, during high accelerating conditions in an EV or HEV, a
perature of 5 ◦ C, the average temperature across the battery modules

11
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 18. Comparison of temperature distribution in the module for varying coolant mass flow rate.

Table 10
Performance of battery module for different coolant mass flow rates.
Mass flow rate (kg. s Re Tbm,max ( Tbm,min ( Tw,out ( Tσ (
− 1
) ◦
C) ◦
C) ◦
C) ◦
C)

0.001 388.35 28.49 25.04 27.63 0.63

0.002 776.69 27.60 25.02 26.65 0.42
0.003 1165.0 27.28 25.02 26.28 0.34
0.004 1553.4 27.11 25.02 26.09 0.31
0.005 1941.7 27.00 25.02 25.97 0.29

Fig. 20. Variation of average battery module temperature for varying coolant
inlet temperature.

Table 11
Performance of battery module for different coolant inlet temperatures across
cold plate.
Coolant 5 10 15 20 25 30 35
Temperature (
◦
C)

Tσ ( ◦ C) 5.24 3.85 2.47 1.08 0.34 1.73 3.12

Tbm, ( ◦ C) 9.98 14.17 18.43 22.71 27.28 31.94 36.65
Fig. 19. Effect of varying coolant mass flow rate on performance of bat­ max

tery module.
becomes heated, which is clearly discerned from Fig. 21. A similar
reaches 6.91 ◦ C after discharging for 720s. Similar phenomena are pattern is observed for other operating conditions with a varying coolant
observed in the case of coolant inlet temperature of 10 ◦ C, 15 ◦ C and inlet temperature of 30 ◦ C and 35 ◦ C. Based on the above rigorous
20 ◦ C. It indicates that the Tavg is not within the optimal range (25 ◦ C- investigation, it can be concluded that a coolant inlet temperature of
40 ◦ C) for coolant inlet temperatures of 5 ◦ C, 10 ◦ C, 15 ◦ C and 20 ◦ C. 25 ◦ C is sufficient to maintain the temperature of a module within safer
Besides this, the possibility of these temperatures contributing to a range, resulting as a right choice for the considered boundary
reduction in capability and life cycle of the battery module is high due to conditions.
cold frosting, as shown in Table 11. This kind of phenomenon is
observed during winter, which can be easily avoided by placing a pre­ 3.8. Influence of ambient temperature
heater at the cold plate inlet. The difference in temperature between the
battery and coolant is found to be high along the discharge process, In real-world automobile applications, battery thermal efficiency is
which may be due to the low thermal conductivity of active battery greatly affected due to continuous operation, fast charging, climatic
material. As the discharge process progresses, the difference in tem­ conditions and the load of the vehicle. Higher environmental tempera­
perature reaches stability. tures can contribute to increased heat generation leading to thermal
Fig. 21 presents the transient variation of the surface temperature runaway [51], whereas in colder climates, batteries may suffer severe
across a 5-channel cold plate at a constant battery discharge rate of 5C degradation along with loss in energy and capacity [52]. In the present
while both the ambient condition and the coolant inlet temperature are work to further explore the cooling ability of cold plate, ambient tem­
assumed as 25 ◦ C. As time progresses, the lower part of the battery peratures are varied from 10 ◦ C to 50 ◦ C with an incremental difference

12
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 21. Transient variation of cold plate temperature for a discharge rate of 5C, ambient and coolant inlet temperature 25 ◦ C.

in a real driving condition, the vehicle may encounter varying discharge

rates. Due to this, thermal inhomogeneity occurs across the battery
module, which would contribute to a reduction in the lifetime of the
battery. A proper cooling arrangement is required to retain the module
within the optimal temperature range. Fig. 23 represents the variations
in average temperature for different discharge rates (1C-5C). For this,
the optimal configurations obtained from the above simulations, such as
a 5-channel cold plate with a 4 mm width is selected with a constant
water flow rate of 0.003 kg.s − 1 at the inlet. As the discharge rate in­
creases from 1C to 5C, the Tavg of the battery module rises by 1.13 ◦ C. In
the case of 4C discharging, the Tavg of the battery rises gradually after
300s of discharging and reaches up to 25.79 ◦ C. A similar pattern can be
visualized for the remaining discharge process. However, the Tavg of the
battery slightly reaches stability as the cooling takes place simulta­
neously with the discharge process.
The overall temperature of the battery module can be determined by
considering three factors: heat produced by the battery, heat extracted
Fig. 22. Comparison of the average battery module temperature for varying by the cooling liquid, and heat accumulated inside the battery. The heat
ambient temperature. produced by the battery will eventually be removed by the cooling
liquid during the discharge process. Referring to Fig. 24, the part of the
of 5 ◦ C for a 5-channelled cold plate with a constant mass flow rate of battery that is closer to the inlet of the cold plate (the cooler area)
0.003 kg.s − 1 (Fig. 22). For this, the coolant inlet temperature is fixed at
25 ◦ C, and the battery is allowed to discharge at 5C. It is observed that
with the change in ambient temperature from 10 ◦ C to 50 ◦ C, Tavg is
increased by 0.64 ◦ C, and Tσ reaches up to 0.46 ◦ C. It is seen from
Table 12 that the possibility of higher ambient temperatures lowering
the performance of the battery module is significant concerning the
values of Tσ. Hence, it may be noted that an atmospheric temperature
lesser than 20 ◦ C or greater than 35 ◦ C may trigger a safety issue and
decrease the battery life. As explained earlier, the consequence of
increasing the mass flow rate improves the cooling efficiency of battery
modules at the cost of higher energy usage. Furthermore, at extreme
temperatures, different methods can be utilized for safe running and
active cooling of the battery. Some of them are as follows: by using
nanofluids, thermal conductivity and heat capacity can be enhanced
[53], and by maintaining an appropriate coolant temperature according
to atmospheric conditions.

3.9. Influence of battery discharge rate

Fig. 23. Comparison of the average module temperature for varying battery
discharge rate.
In the previous subsections, a discharge rate of 5C is considered, but

Table 12
Performance of battery module for different ambient temperatures.
Ambient Temperature ( ◦ C) 10 15 20 25 30 35 40 45 50

Tσ ( ◦ C) 0.28 0.30 0.33 0.34 0.37 0.39 0.42 0.44 0.46

Tmax ( ◦ C) 26.68 26.85 27.04 27.28 27.59 28.18 28.76 29.35 29.94

13
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

Fig. 24. Comparison of temperature distribution across the battery module for different discharge rates.

experiences lesser temperature than the one near the outlet of the cold maximum temperature of the battery module is retained within the
plate (hotter area). The temperature of water increases in line with the optimal range, which justifies the use of straight channel design in
flow direction, deteriorating the cooling effect for a part of the battery ensuring better thermal management of batteries.
closer to the downstream of the fluid. The average temperature can be
uniform throughout the module if the heat generated and heat absorbed CRediT authorship contribution statement
is in equilibrium. However, this is difficult to maintain due to an uneven
heat generation by the battery. On adjusting the water flow rate and K. Monika: Methodology, Software, Investigation, Validation,
entry temperature at higher discharge rates, the temperature of the Formal analysis, Writing - original draft, Writing - review & editing.
module can be kept within the desirable range at the expense of higher Chanchal Chakraborty: Project administration. Sounak Roy: Project
pressure drop and operating costs. administration. Srikanta Dinda: Project administration. Satyapaul A.
Singh: Project administration. Santanu Prasad Datta: Conceptualiza­
4. Conclusions tion, Investigation, Writing - original draft, Writing - review & editing,
Supervision, Resources, Project administration, Funding acquisition.
An ingeniously designed BTMS is proposed to analyse the thermal
behaviour of LiFePO4 batteries. A prismatic battery module comprising Declaration of Competing Interest
5 cells each of dimension 63 mm × 118 mm × 13 mm is considered. An
aluminium based cold plate of size 63 mm × 118 mm × 2 mm is sand­ None
wiched between two consecutive cells. Different parameters effecting
the cooling effectiveness of cold plate is explored in the present inves­
Acknowledgement
tigation with a discharge rate of the battery module at 5C lasting for
720s. Regarding this, an electrochemical-thermal conjugate heat trans­
S. P. Datta acknowledges the support from the science and engi­
fer model of the battery module is developed and solved by using
neering research board (SERB), India under the scheme of early-career
COMSOL Multiphysics 5.4 software to optimize the channel design and
research (ECR) grant (ECR/2018/002291).
operating parameters. The critical insights based on the present inves­
tigation are as follows:
References

• A parallel flow arrangement with a coolant inlet port near to the tab [1] C.C. Chan, The state of the art of electric, hybrid, and fuel cell vehicles, Proc. IEEE.
region is found to be the best choice with Tavg = 27.10 ◦ C and Tσ = 95 (2007) 704–718.
0.65 ◦ C. [2] G. Zubi, R. Dufo-López, M. Carvalho, G. Pasaoglu, The lithium-ion battery: state of
the art and future perspectives, Renew. Sustain. Energy Rev. 89 (2018) 292–308.
• A 5-channelled cold plate of width 4 mm is observed as the ideal [3] X.M. Xu, R. He, Review on the heat dissipation performance of battery pack with
trade-off between heat transfer and pressure drop for thermal man­ different structures and operation conditions, Renew. Sustain. Energy Rev. 29
agement of the battery module with Tσ of 0.63 ◦ C, and a pressure (2014) 301–315.
[4] S. Al Hallaj, J. Prakash, J.R. Selman, Characterization of commercial Li-ion
drop of 100.1 Pa. batteries using electrochemical–calorimetric measurements, J. Power Sources. 87
• A coolant flow rate of 0.003 kg.s − 1 and the corresponding Re= (2000) 186–194.
1165.04 is found adequate to enhance the temperature homogeneity [5] M.K. Tran, A. Mevawala, S. Panchal, K. Raahemifar, M. Fowler, R. Fraser, Effect of
integrating the hysteresis component to the equivalent circuit model of Lithium-ion
across the battery module due to a moderate pressure drop of 512.7 battery for dynamic and non-dynamic applications, J. Energy Storage. 32 (2020),
Pa, and heat transfer coefficient of 153.9 W.m − 2. ◦ C − 1, along with a 101785.
j/f factor of 0.328. [6] H.E. Perez, S. Dey, X. Hu, S.J. Moura, Optimal Charging of Li-Ion Batteries via a
Single Particle Model with Electrolyte and Thermal Dynamics, J. Electrochem. Soc.
• A coolant inlet temperature of 25 ◦ C is found suitable to uphold the
164 (2017) A1679–A1687.
average temperature of the battery module within the desired range [7] E. Hosseinzadeh, R. Genieser, D. Worwood, A. Barai, J. Marco, P. Jennings,
of 25 ◦ C-40 ◦ C, at this instant Tavg = 26.20 ◦ C and Tσ = 0.34 ◦ C. A systematic approach for electrochemical-thermal modelling of a large format
• An ambient temperature less than 20 ◦ C and greater than 35 ◦ C is lithium-ion battery for electric vehicle application, J. Power Sources. 382 (2018)
77–94.
found to reduce the cooling effectiveness of cold plate. However, in [8] Y. Tang, L. Wu, W. Wei, D. Wen, Q. Guo, W. Liang, L. Xiao, Study of the thermal
real-time applications, an appropriate mass flow rate or coolant properties during the cyclic process of lithium ion power batteries using the
temperature is to be maintained to reduce the peak temperature rise electrochemical-thermal coupling model, Appl. Therm. Eng. 137 (2018) 11–22.
[9] A.N. Mistry, H.R. Palle, P.P. Mukherjee, In operando thermal signature probe for
and thereby increase the lifespan of batteries. lithium-ion batteries, Appl. Phys. Lett. 114 (2019).
[10] M. Klein, S. Tong, J.W. Park, In-plane nonuniform temperature effects on the
Based on the current investigation, it can be concluded that the performance of a large-format lithium-ion pouch cell, Appl. Energy. 165 (2016)
639–647.

14
K. Monika et al. Journal of Energy Storage 35 (2021) 102301

[11] A. Väyrynen, J. Salminen, Lithium ion battery production, J. Chem. Thermodyn. 46 batteries with mini-channel cold plates, Int. J. Heat Mass Transf. 164 (2021),
(2012) 80–85. 120568.
[12] T.M. Bandhauer, S. Garimella, T.F. Fuller, A critical review of thermal issues in [33] Y. Wang, G. Zhang, X. Yang, Optimization of liquid cooling technology for
lithium-ion batteries, J. Electrochem. Soc. 158 (2011). cylindrical power battery module, Appl. Therm. Eng. 162 (2019), 114200.
[13] M. Lu, X. Zhang, J. Ji, X. Xu, Y. Zhang, Research progress on power battery cooling [34] Z. Qian, Y. Li, Z. Rao, Thermal performance of lithium-ion battery thermal
technology for electric vehicles, J. Energy Storage. 27 (2020), 101155. management system by using mini-channel cooling, Energy Convers. Manag. 126
[14] R. Liu, J. Chen, J. Xun, K. Jiao, Q. Du, Numerical investigation of thermal (2016) 622–631.
behaviors in lithium-ion battery stack discharge, Appl. Energy. 132 (2014) [35] C. Lan, J. Xu, Y. Qiao, Y. Ma, Thermal management for high power lithium-ion
288–297. battery by minichannel aluminum tubes, Appl. Therm. Eng. 101 (2016) 284–292.
[15] B. Mondal, C.F. Lopez, P.P. Mukherjee, Exploring the efficacy of nanofluids for [36] L.H. Saw, H.M. Poon, H.S. Thiam, Z. Cai, W.T. Chong, N.A. Pambudi, Y.J. King,
lithium-ion battery thermal management, Int. J. Heat Mass Transf. 112 (2017) Novel thermal management system using mist cooling for lithium-ion battery
779–794. packs, Appl. Energy. 223 (2018) 146–158.
[16] S. Park, D.S. Jang, D.C. Lee, S.H. Hong, Y. Kim, Simulation on cooling performance [37] Z. Rao, Z. Qian, Y. Kuang, Y. Li, Thermal performance of liquid cooling based
characteristics of a refrigerant-cooled active thermal management system for thermal management system for cylindrical lithium-ion battery module with
lithium ion batteries, Int. J. Heat Mass Transf. 135 (2019) 131–141. variable contact surface, Appl. Therm. Eng. 123 (2017) 1514–1522.
[17] Z. An, L. Jia, X. Li, Y. Ding, Experimental investigation on lithium-ion battery [38] Y. Lv, G. Liu, G. Zhang, X. Yang, A novel thermal management structure using
thermal management based on flow boiling in mini-channel, Appl. Therm. Eng. serpentine phase change material coupled with forced air convection for
117 (2017) 534–543. cylindrical battery modules, J. Power Sources. 468 (2020), 228398.
[18] H.L. Liu, X.K. An, C.S. Wang, Heat transfer performance of T-Y type micro-channel [39] K. Li, J. Yan, H. Chen, Q. Wang, Water cooling based strategy for lithium ion
heat sink with liquid GaInSn coolant, Int. J. Therm. Sci. 120 (2017) 203–219. battery pack dynamic cycling for thermal management system, Appl. Therm. Eng.
[19] Q. Wang, B. Jiang, B. Li, Y. Yan, A critical review of thermal management models 132 (2018) 575–585.
and solutions of lithium-ion batteries for the development of pure electric vehicles, [40] F. Dong, D. Song, J. Ni, Investigation of the effect of U-shaped mini-channel
Renew. Sustain. Energy Rev. 64 (2016) 106–128. structure on the thermal performance of liquid-cooled prismatic batteries, Numer.
[20] J. Zhao, Z. Rao, Y. Huo, X. Liu, Y. Li, Thermal management of cylindrical power Heat Transf. Part A Appl. 77 (2020) 105–120.
battery module for extending the life of new energy electric vehicles, Appl. Therm. [41] Z. Rao, Investigation on thermal management performance of wedge - shaped
Eng. 85 (2015) 33–43. microchannels for rectangular Li - ion batteries, (2019) 3876–3890.
[21] W. Wu, S. Wang, W. Wu, K. Chen, S. Hong, Y. Lai, A critical review of battery [42] K. Smith, C.Y. Wang, Power and thermal characterization of a lithium-ion battery
thermal performance and liquid based battery thermal management, Energy pack for hybrid-electric vehicles, J. Power Sources. 160 (2006) 662–673.
Convers. Manag. 182 (2019) 262–281. [43] K. Onda, T. Ohshima, M. Nakayama, K. Fukuda, T. Araki, Thermal behavior of
[22] N.H. Naqiuddin, L.H. Saw, M.C. Yew, F. Yusof, T.C. Ng, M.K. Yew, Overview of small lithium-ion battery during rapid charge and discharge cycles, J. Power
micro-channel design for high heat flux application, Renew. Sustain. Energy Rev. Sources. 158 (2006) 535–542.
82 (2018) 901–914. [44] Y. Huo, Z. Rao, X. Liu, J. Zhao, Investigation of power battery thermal management
[23] D. Chen, J. Jiang, G.H. Kim, C. Yang, A. Pesaran, Comparison of different cooling by using mini-channel cold plate, Energy Convers. Manag. 89 (2015) 387–395.
methods for lithium ion battery cells, Appl. Therm. Eng. 94 (2016) 846–854. [45] Z. An, K. Shah, L. Jia, Y. Ma, A parametric study for optimization of minichannel
[24] J. Zhao, Z. Rao, Y. Li, Thermal performance of mini-channel liquid cooled cylinder based battery thermal management system, Appl. Therm. Eng. 154 (2019)
based battery thermal management for cylindrical lithium-ion power battery, 593–601.
Energy Convers. Manag. 103 (2015) 157–165. [46] Z. Rao, S. Wang, G. Zhang, Simulation and experiment of thermal energy
[25] V.G. Choudhari, A.S. Dhoble, S. Panchal, Numerical analysis of different fin management with phase change material for ageing LiFePO4 power battery,
structures in phase change material module for battery thermal management Energy Convers. Manag. 52 (2011) 3408–3414.
system and its optimization, Int. J. Heat Mass Transf. 163 (2020), 120434. [47] M. Ghalkhani, F. Bahiraei, G.A. Nazri, M. Saif, Electrochemical–Thermal Model of
[26] C.F. Lopez, J.A. Jeevarajan, P.P. Mukherjee, Evaluation of Combined Active and Pouch-type Lithium-ion Batteries, Electrochim. Acta. 247 (2017) 569–587.
Passive Thermal Management Strategies for Lithium-Ion Batteries, J. Electrochem. [48] Z. Rao, S. Wang, M. Wu, Z. Lin, F. Li, Experimental investigation on thermal
Energy Convers. Storage. 13 (2016) 1–10. management of electric vehicle battery with heat pipe, Energy Convers. Manag. 65
[27] A. Jarrett, I.Y. Kim, In fl uence of operating conditions on the optimum design of (2013) 92–97.
electric vehicle battery cooling plates, J. Power Sources. 245 (2014) 644–655. [49] M. Malik, I. Dincer, M.A. Rosen, M. Mathew, M. Fowler, Thermal and electrical
[28] A. Jarrett, I.Y. Kim, Design optimization of electric vehicle battery cooling plates performance evaluations of series connected Li-ion batteries in a pack with liquid
for thermal performance, J. Power Sources. 196 (2011) 10359–10368. cooling, Appl. Therm. Eng. 129 (2018) 472–481.
[29] M.S. Patil, J.H. Seo, S. Panchal, M.Y. Lee, Numerical study on sensitivity analysis of [50] A. Tang, J. Li, L. Lou, C. Shan, X. Yuan, Optimization design and numerical study
factors influencing liquid cooling with double cold-plate for lithium-ion pouch cell, on water cooling structure for power lithium battery pack, Appl. Therm. Eng. 159
Int. J. Energy Res. (2020) 1–27. (2019), 113760.
[30] L. Sheng, L. Su, H. Zhang, K. Li, Y. Fang, W. Ye, Y. Fang, Numerical investigation on [51] T.M. Bandhauer, S. Garimella, T.F. Fuller, Temperature-dependent electrochemical
a lithium ion battery thermal management utilizing a serpentine-channel liquid heat generation in a commercial lithium-ion battery, J. Power Sources. 247 (2014)
cooling plate exchanger, Int. J. Heat Mass Transf. 141 (2019) 658–668. 618–628.
[31] M.S. Patil, J.H. Seo, S. Panchal, S.W. Jee, M.Y. Lee, Investigation on thermal [52] J. Jaguemont, L. Boulon, Y. Dubé, A comprehensive review of lithium-ion batteries
performance of water-cooled Li-ion pouch cell and pack at high discharge rate with used in hybrid and electric vehicles at cold temperatures, Appl. Energy. 164 (2016)
U-turn type microchannel cold plate, Int. J. Heat Mass Transf. 155 (2020), 119728. 99–114.
[32] K. Monika, C. Chakraborty, S. Roy, S. Dinda, S.A. Singh, S.P. Datta, Parametric [53] R. Nimmagadda, K. Venkatasubbaiah, Conjugate heat transfer analysis of micro-
investigation to optimize the thermal management of pouch type lithium-ion channel using novel hybrid nanofluids (Al2O3 + Ag / Water), Eur. J. Mech. B/
Fluids. 52 (2015) 19–27.

15