Comprehensive Passive Thermal Management Systems for Electric Vehicles

: Lithium-ion (Li-ion) batteries have emerged as a promising energy source for electric vehicle (EV) applications owing to the solution offered by their high power, high speciﬁc energy, no memory effect, and their excellent durability. However, they generate a large amount of heat, particularly during the fast discharge process. Therefore, a suitable thermal management system (TMS) is necessary to guarantee their performance, efﬁciency, capacity, safety, and lifetime. This study investigates the thermal performance of different passive cooling systems for the LTO Li-ion battery cell/module with the application of natural convection, aluminum (Al) mesh, copper (Cu) mesh, phase change material (PCM), and PCM-graphite. Experimental results show the average temperature of the cell, due to natural convection, Al mesh, Cu mesh, PCM, and PCM-graphite compared with the lack of natural convection decrease by 6.4%, 7.4%, 8.8%, 30%, and 39.3%, respectively. In addition, some numerical simulations and investigations are solved by COMSOL Multiphysics ® , for the battery module consisting of 30 cells, which is cooled by PCM and PCM-graphite. The maximum temperature of the battery module compared with the natural convection case study is reduced by 15.1% and 17.3%, respectively. Moreover, increasing the cell spacing in the battery module has a direct effect on temperature reduction.


Introduction
Global warming and air pollution have pushed researchers to replace a clean alternative source for fossil fuels [1,2]. Transportation is one of the main consumers which is related to fossil fuels. Electric vehicles (EVs) and hybrid electric vehicles (HEVs) with low CO 2 emissions are the most appropriate alternatives for conventional vehicles. Lithiumion (Li-ion) batteries are the most promising energy source for EVs and HEVs owing to their features comprising high specific energy, high capacity, high power, and no memory effect [3][4][5][6]. Nonetheless, Li-ion batteries produce a noticeable amount of heat, particularly during the fast charging/discharge process. Consequently, the design and build of a suitable thermal management system (TMS) are vital to preserving the battery temperature in a safe temperature range (25-40 • C) [7]. TMSs are generally divided into active and passive cooling systems. Active cooling systems like forced air and liquid and refrigerant cooling systems [8] need an external source of energy. However, passive cooling systems like phase change material (PCM) [9,10], natural convection, heat sink [11], heat exchangers [12], fin, and heat pipe [13][14][15] do not consume any energy [16].
The air cooling [16][17][18][19][20] and liquid cooling [21][22][23] systems are common active cooling systems that can afford effective cooling for the Li-ion batteries. However, air cooling In the experimental section, the performance of natural convection, Al and Cu mesh, PCM, and PCM-graphite has been considered on the cell level. The image of the test setup and the location of thermocouples are exposed in Figure 1. The experimental test setup included a tester, a cell, a Pico USB TC-08 data logger, four K-type thermocouples, and a personal computer. The accuracy of the thermocouples is around ±0.2 • C which are connected to the surface of the cell. It is important to note that the ambient temperature is set at 22 • C for all the tests, and T 4 is responsible for measuring it.  In the experimental section, the performance of natural convection, Al and Cu mesh, PCM, and PCM-graphite has been considered on the cell level. The image of the test setup and the location of thermocouples are exposed in Figure 1. The experimental test setup included a tester, a cell, a Pico USB TC-08 data logger, four K-type thermocouples, and a personal computer. The accuracy of the thermocouples is around ±0.2 °C which are connected to the surface of the cell. It is important to note that the ambient temperature is set at 22 °C for all the tests, and T4 is responsible for measuring it. The battery tester (PEC) is used to start cycling. The PEC tester controls the cycling and records the voltage and current of the cell. The temperature of the cell is recorded by the external thermocouples. The discharging process is done under the current rate of 8C (184 A) at 446 s. The heat production of the cell can be calculated based on the following equation: According to Equation (1) , cp, T, k, and signify the mass, heat capacity, temperature, thermal conductivity, and heat production of the cell, respectively. The battery tester (PEC) is used to start cycling. The PEC tester controls the cycling and records the voltage and current of the cell. The temperature of the cell is recorded by the external thermocouples. The discharging process is done under the current rate of 8C (184 A) at 446 s. The heat production of the cell can be calculated based on the following equation: According to Equation (1) m, cp, T, k, and q g signify the mass, heat capacity, temperature, thermal conductivity, and heat production of the cell, respectively.

Lack and Presence of Natural Convection
The influence of the natural convection cooling method is the initial phase to consider for thermal management. The natural convection cooling condition refers to a passive cooling system that does not consume any external energy. The goal of this section is to find out the consequence of the natural convection on the heat production of the LTO cell at the 8C discharging process, which is capable of preparing specific guidance for battery cooling to compare with other cooling methods. The temperature of the cell in the discharging process and pictures of the tested battery in the presence and lack of natural convection are illustrated in Figure 2. As it is obvious the natural convection has a minor effect on the temperature trends of the thermocouples. According to the calculation, the heat transfer coefficient is considered 6.87 W/m 2 K [18]. The average temperature of the cell at the end of the discharging process in the lack and presence of the natural convection reaches 57.2 • C and 53.5 • C, which show a 6.4% reduction. The influence of the natural convection cooling method is the initial phase to consider for thermal management. The natural convection cooling condition refers to a passive cooling system that does not consume any external energy. The goal of this section is to find out the consequence of the natural convection on the heat production of the LTO cell at the 8C discharging process, which is capable of preparing specific guidance for battery cooling to compare with other cooling methods. The temperature of the cell in the discharging process and pictures of the tested battery in the presence and lack of natural convection are illustrated in Figure 2. As it is obvious the natural convection has a minor effect on the temperature trends of the thermocouples. According to the calculation, the heat transfer coefficient is considered 6.87 W/m 2 K [18]. The average temperature of the cell at the end of the discharging process in the lack and presence of the natural convection reaches 57.2 °C and 53.5 °C, which show a 6.4% reduction.

Natural Convection Effect on Al and Cu Mesh
In this section, the effect of Al and Cu mesh on the temperature of the LTO cell in the presence of the natural convection has been considered. The metal mesh is a kind of heat sink classified as a passive heat exchanger that transfers heat generated by the battery to air as a fluid medium. The Al and Cu mesh is classified as a passive heat sink that benefits high reliability and low-cost characters. It can be seen in Figure 3 that the temperature of the cell has been uniformed, due to Al and Cu mesh. Moreover, the average temperature of the cell wrapped with Al and Cu mesh at the end of the discharging process reaches 52.9 • C and 52.1 • C, which show a 7.4% and 8.8% reduction compared with the lack of natural convection. In this section, the effect of Al and Cu mesh on the temperature of the LTO cell in the presence of the natural convection has been considered. The metal mesh is a kind of heat sink classified as a passive heat exchanger that transfers heat generated by the battery to air as a fluid medium. The Al and Cu mesh is classified as a passive heat sink that benefits high reliability and low-cost characters. It can be seen in Figure 3 that the temperature of the cell has been uniformed, due to Al and Cu mesh. Moreover, the average temperature of the cell wrapped with Al and Cu mesh at the end of the discharging process reaches 52.9 °C and 52.1 °C, which show a 7.4% and 8.8% reduction compared with the lack of natural convection.

PCM and PCM-Graphite Cooling
PCMs are using as an operational means of passive cooling in battery thermal management applications. Generally, PCM is named as a material that can store/release a huge amount of energy at a constant temperature or in a negligible temperature range during the phase change process. The usage of the PCMs is growing, owing to the effective features in temperature control and free energy consumption. In this study, for

PCM and PCM-Graphite Cooling
PCMs are using as an operational means of passive cooling in battery thermal management applications. Generally, PCM is named as a material that can store/release a huge amount of energy at a constant temperature or in a negligible temperature range during the phase change process. The usage of the PCMs is growing, owing to the effective features in temperature control and free energy consumption. In this study, for thermal management of the LTO cell, a PCM with a phase change temperature of 30 • C is chosen. The key data of the PCM and container are shown in Table 2. The utilized PCM in this study is organic paraffin wax, which is appropriate and operative for the suggested operating temperature range of the Li-ion battery cells. According to Figure 4, the temperature of the cell is effectively controlled by the PCM in a safe temperature range (25-40 • C). The average temperature of the cell at the end of the discharging process in the presence of the PCM reaches 39.98 • C, which shows a 30% reduction compare with the lack of natural convection. thermal management of the LTO cell, a PCM with a phase change temperature of 30 °C is chosen. The key data of the PCM and container are shown in Table 2. The utilized PCM in this study is organic paraffin wax, which is appropriate and operative for the suggested operating temperature range of the Li-ion battery cells. According to Figure 4, the temperature of the cell is effectively controlled by the PCM in a safe temperature range (25-40 °C). The average temperature of the cell at the end of the discharging process in the presence of the PCM reaches 39.98 °C, which shows a 30% reduction compare with the lack of natural convection.  Nevertheless, most PCMs suffer from low thermal conductivity in energy storage and cooling applications [13,[39][40][41]. Therefore, several approaches like adding nanoparticles, graphite, nanocarbon tube, and heat pipes are used to reimburse for this problem. In order to increase the cooling capability and thermal conductivity, porous graphite is added to the PCM. The main parameters of the PCM-graphite are mentioned in Table 3. Nevertheless, most PCMs suffer from low thermal conductivity in energy storage and cooling applications [13,[39][40][41]. Therefore, several approaches like adding nanoparticles, graphite, nanocarbon tube, and heat pipes are used to reimburse for this problem. In order to increase the cooling capability and thermal conductivity, porous graphite is added to the PCM. The main parameters of the PCM-graphite are mentioned in Table 3. To consider the cooling efficiency of the PCM-graphite, the Li-ion battery cell is directly submerged in the composite. The thermal contact resistance is extremely decreased via the submerged method, leading to higher cooling performance [42]. As is expected, the PCM-graphite displays a better cooling performance, due to higher thermal conductivity. The PCM composite gets benefitted from the high thermal conductivity of graphite and with a worthy heat storage capacity of the PCM. Figure 4d displays the performance of the PCM-graphite. The average cell temperature reaches 34.71 • C, which experiences 39.3% compared with the lack of natural convection.

Comparison Results
The value of the different passive cooling systems on the average temperature of the LTO cell is shown in Figure 5. In the same initial conditions, the average temperature of the cell in lack of natural convection, natural convection, Al mesh, Cu mesh, PCM, and PCM-graphite reaches the 57.24 • C, 53.52 • C, 52.96 • C, 52.18 • C, 39.98 • C, and 34.71 • C, respectively. As can be seen, PCM and PCM-graphite preserve the cell in the safe operating temperature range, which is shown by the dashed line.

Battery Thermal Modeling
Matlab/Simulink (MathWorks, Natick, MA, USA) with COMSOL Multiphysics ® (COMSOL , Stockholm, Sweden) has been used to build the 3D thermal behavior of the cell. Energy balance (Equation (1)) is utilized to describe the transient thermal generation inside the cell. In the current study, the heat production of the cell is calculated by Matlab/Simulink from the ohmic resistance of the cell. Therefore, an electrical model of the cell is built and validated with the dual-polarization electric-equivalent-circuit (ECM) approach [43]. Figure 6 displays the recommended impedance model where Voc, R0, and Vbatt present the open-circuit voltage, the series-connected ohmic resistance, and battery terminal voltage, respectively. It also shows two parallel R//C branches, which represent the time-dependent polarization processes.

Battery Thermal Modeling
Matlab/Simulink (MathWorks, Natick, MA, USA) with COMSOL Multiphysics ® (COMSOL , Stockholm, Sweden) has been used to build the 3D thermal behavior of the cell. Energy balance (Equation (1)) is utilized to describe the transient thermal generation inside the cell. In the current study, the heat production of the cell is calculated by Matlab/Simulink from the ohmic resistance of the cell. Therefore, an electrical model of the cell is built and validated with the dual-polarization electric-equivalent-circuit (ECM) approach [43]. Figure 6 displays the recommended impedance model where Voc, R0, and Vbatt present the open-circuit voltage, the series-connected ohmic resistance, and battery terminal voltage, respectively. It also shows two parallel R//C branches, which represent the time-dependent polarization processes.  Moreover, heat production of the tab domain is presented by Equation (3) [44]. Based on the above equations, I and R bt signify the current and ohmic resistance of the cell. In addition, for the tab domains R tab , ρ , l, and S are the electrical resistance, resistivity, length, and cross-section of the corresponding tab, respectively. The convective heat transfer to the ambient is also calculated as follows [45]: wherein, h and S signify the heat transfer coefficient and cross-section area of the cell. Additionally, T bt and T amb determine the cell and ambient temperature.

Illustrative Equations for PCM
The governing equations for PCM comprising continuity, momentum, and energy equations can be written as follows [13]: where in u , T 1 , T m , T 2 and k present the velocity, initial temperature, melting temperature, the final temperature (T 2 = T m + ∆T) and thermal conductivity of the PCM, respectively. Between the solid and liquid phases, a transition happens within the interval of ∆T which is named the mushy phase. The different phases of the PCM are as follows [37].
The heat capacity and the thermal conductivity of the PCM can be mentioned as follows (s, solid; t, transition; l, liquid): The total energy that can be stored in the PCM is calculated using the below equation: where m represents the mass of the PCM. Totally, COMSOL by the revealed equations simulates the melting of the PCM allowing for conductive and convective heat transfer [46].

Validation of the Thermal Model for Natural Convection, PCM and PCM-Graphite
The experimental results for the 23Ah LTO cell at 8C discharging rate and initial temperature of 22 • C are validated under the effect of natural convection, PCM, and PCMgraphite with the COMSOL Multiphysics ® . Four thermocouples are used experimentally to record the battery surface and ambient temperature, respectively. For validation, the T 1 thermocouple (Figure 1b) is selected for comparison with the experimental results. According to Figure 7, the difference between the simulation and experimental results for T 1 is in an acceptable range. The average relative errors for T 1 thermocouple, for states of a, b, and c are 4.6%, 1%, and 4.1%, respectively, within a standard error range less than 5% [47].
PCM-graphite with the COMSOL Multiphysics ® . Four thermocouples are used experimentally to record the battery surface and ambient temperature, respectively. For validation, the T1 thermocouple (Figure 1b) is selected for comparison with the experimental results. According to Figure 7, the difference between the simulation and experimental results for T1 is in an acceptable range. The average relative errors for T1 thermocouple, for states of a, b, and c are 4.6%, 1%, and 4.1%, respectively, within a standard error range less than 5% [47].

Configuration Design of the Module
In the current section, a battery module comprising 30 cells is simulated to describe the thermal effectiveness of the passive cooling strategies comprising natural convection, PCM, and PCM-graphite. Module thermal management needs special attention since cells are affected by each other heat generation during the charging/discharging process. In the current simulation, the module is discharged in a high current profile at 446 s. Active cooling systems are the most common types of cooling systems for the battery module/pack, especially in high current applications. However, in this study, the passive phase change cooling method is used for the thermal management of battery module temperature. The key data of the module are presented in Table 4. Table 4. The key data of the module adapted from [18].

Number of cells in series 30
Nominal voltage of the module (V) 69 Weight (kg) 16.5 Volume (L) 7.8 Stored energy in the module (KWh) 1.6 Figure 8a,b shows the schematic geometry components, dimensions, and mesh distribution of the module, respectively. In the following design, the module consisted of 30 cells which are submerged in PCM and PCM-graphite. The initial temperature of the module, PCM, and the ambient are set at 22 • C. The mesh type is unstructured tetrahedral, which is generated by default physics-controlled mesh in COMSOL Multiphysics ® .
PCM, and PCM-graphite. Module thermal management needs special attention since cells are affected by each other heat generation during the charging/discharging process. In the current simulation, the module is discharged in a high current profile at 446 s. Active cooling systems are the most common types of cooling systems for the battery module/pack, especially in high current applications. However, in this study, the passive phase change cooling method is used for the thermal management of battery module temperature. The key data of the module are presented in Table 4.  Figure 8a,b shows the schematic geometry components, dimensions, and mesh distribution of the module, respectively. In the following design, the module consisted of 30 cells which are submerged in PCM and PCM-graphite. The initial temperature of the module, PCM, and the ambient are set at 22 °C. The mesh type is unstructured tetrahedral, which is generated by default physics-controlled mesh in COMSOL Multiphysics ® .

Cooling Effect of Natural Convection, PCM and PCM-Graphite
The initial cooling phase of the battery module is natural convection condition, which can be used as a source to compare with the cooling efficiency of the PCM and PCM-graphite methods. The initial boundary condition for module testing is the same as the cell level. Figure 9 displays the battery module, which cooled through natural convection, PCM, and PCM-graphite that experiences a maximum temperature of 58.8 °C, 49.9 °C, and 48.6 °C, respectively. The limited cooling surface area and the low heat transfer coefficient of the air are the main inabilities of the natural convection cooling

Cooling Effect of Natural Convection, PCM and PCM-Graphite
The initial cooling phase of the battery module is natural convection condition, which can be used as a source to compare with the cooling efficiency of the PCM and PCMgraphite methods. The initial boundary condition for module testing is the same as the cell level. Figure 9 displays the battery module, which cooled through natural convection, PCM, and PCM-graphite that experiences a maximum temperature of 58.8 • C, 49.9 • C, and 48.6 • C, respectively. The limited cooling surface area and the low heat transfer coefficient of the air are the main inabilities of the natural convection cooling method. Moreover, the temperature uniformity is low, with more concentration in the center and top regions of the cells [14]. It is found that the maximum temperature of the module experiences a 15.1% and 17.3% reduction by PCM and PCM-graphite methods, respectively. Moreover, the temperature distribution uniformity has an excellent improvement. Nevertheless, the module temperature can be further improved using more spacing between the cells. method. Moreover, the temperature uniformity is low, with more concentration in the center and top regions of the cells [14]. It is found that the maximum temperature of the module experiences a 15.1% and 17.3% reduction by PCM and PCM-graphite methods, respectively. Moreover, the temperature distribution uniformity has an excellent improvement. Nevertheless, the module temperature can be further improved using more spacing between the cells.

Cooling Effect of Cell Spacing Using PCM and PCM-Graphite Methods
The maximum temperature of the cells is a critical problem for any battery module, which can effectively affect the battery's life and capacity. The spacing of the cells is an important element to increase the efficiency of the passive thermal management methods. In the current section, the influence of cell spacing from 0 to 8 mm is considered on the maximum temperature of the module. The appropriate structure and spacing reduce the possibility of battery thermal runaway. Table 5 summarizes the simulation results about the effects of the spacing on the maximum temperature of the battery module. It can be seen the maximum temperature of the module for 8 mm spacing reaches 43.4 °C and 41.9 °C, which results in 13% and 13.7% temperature reduction for PCM and PCM-graphite cooling methods, respectively.

Cooling Effect of Cell Spacing Using PCM and PCM-Graphite Methods
The maximum temperature of the cells is a critical problem for any battery module, which can effectively affect the battery's life and capacity. The spacing of the cells is an important element to increase the efficiency of the passive thermal management methods. In the current section, the influence of cell spacing from 0 to 8 mm is considered on the maximum temperature of the module. The appropriate structure and spacing reduce the possibility of battery thermal runaway. Table 5 summarizes the simulation results about the effects of the spacing on the maximum temperature of the battery module. It can be seen the maximum temperature of the module for 8 mm spacing reaches 43.4 • C and 41.9 • C, which results in 13% and 13.7% temperature reduction for PCM and PCM-graphite cooling methods, respectively.

Conclusions
The experimental and numerical studies were performed to investigate the cooling performance of different passive TMSs for the LTO cell/module in a high current discharging process. In the same initial conditions, the average temperature of the cell in lack of natural convection, natural convection, Al mesh, Cu mesh, PCM, and PCM-graphite reaches the 57.2 • C, 53.5 • C, 52.9 • C, 52.2 • C, 39.9 • C, and 34.71 • C, respectively. It is found experimentally that PCM and PCM-graphite cooling methods are preserved the battery temperature in a safe temperature zone. In the simulation section, the cooling effect of natural convection, PCM, and PCM-graphite are considered at the module level. The numerical results are validated with the experimental results in an acceptable agreement.
According to the results, there is a 15.1% and 17.3% reduction in maximum module temperature by PCM and PCM-graphite methods, respectively. Moreover, the results exhibit that cell spacing has an enormous impact on the temperature reduction of the module. The maximum temperature of the module for 8 mm spacing reaches 43.4 • C and 41.9 • C, which results in 13% and 13.7% temperature reduction for PCM and PCM-graphite cooling methods, respectively.

Future Work
Utilizing passive cooling methods like PCM cooling systems can be an effective way for battery thermal management applications. However, most PCMs suffer from low thermal conductivity. Therefore, using more additive materials like nanoparticles, carbon nanotube, heat pipe, fin in different percentages and configurations can be studied for the next step in pack level.

Data Availability Statement:
The supporting data will be made available on request.