THERMAL TESTING MODEL AND SIMULATION RESULTS USING A SUPERIOR LITHIUM ION POLYMER BATTERY

THERMAL TESTING MODEL AND SIMULATION RESULTS USING A SUPERIOR LITHIUM ION POLYMER BATTERY A. PRUTEANU, A. NIAGU, S. URSACHE, R. C. CIOBANU Gheorghe Asachi Technical University of Iasi, Romania E-mail: apruteanu@ee.tuiasi.ro Abstract. The battery thermal management is a very important factor which can affect the batteries performance and life span. This paper presents a thermal behavior analysis of a Li-ion polymer battery. Because Li-ion polymer battery is characterized by a high power and energy density values, its starts to be used in different domain applications. One of the main inconveniences is represented by the temperature increase during life cycle. Therefore, this paper presents a model based inputs of the battery to determine the temperature distribution on the battery surface. appear: lithium plating and overheating. A high temperature of a battery can increase its performance, but also could cause unwanted chemical reaction which can affect the battery life. Contrary, at low temperature the reaction rate is reduced, making more difficult to insert the lithium ions into the intercalation spaces [3]. Keywords: thermal behavior, Li-Ion polymer, temperature distribution 1. INTRODUCTION In the last few years, the society is facing two major problems of interest: the climate changes and the limited access to primary energy sources. Regarding these concerns, the interest in development of new vehicle technologies has increased [1]. This relatively new domain of vehicles refers to the following categories: Electric Vehicles (EVs), Plug-in Electric Vehicles (PHEVs) and Hybrid Electric Vehicles (HEVs). EV depends on the battery type [2] and the improvement of their life-time will reduce the runtime and the costs for the vehicle. It is very important to know and manage the battery status, for achieving a better level of reliability and safety. Taking into consideration these aspects, it can be said that the battery performance, cost and life affect directly the life and performance of the EV. Thus, the need to extend the battery lifetime and to use it at their full capacity is an important aspect. Figure 1 presents an overview regarding the actual performances of main parameters in case of electrical energy storage devices. It can be observed that some actual parameters such as: self-discharge rate, operating temperature interval, costs, calendar and cycle life, have to be improved in order to support all the requirements needed for the new design technologies. In present, the operating temperature interval is limited by the battery chemistry and the breakdown temperature of some of the components (e.g. in extreme environments, heating and cooling procedures may be required to keep the cells within their operating temperature range). Another factor that determines the heating of battery is represented by Joule heating due to I 2 R. For example, in case of lithium ion (Li-ion) batteries, the temperature and the voltage represent the main factors that can affect the battery life and performance. Thus if the voltage charging is increased beyond the upper limit, two problems will Figure 1. Actual performance and design goal in case of energy storage devices This paper, presents a study of the temperature influence over a lithium-ion polymer battery by explaining the thermal testing model and the simulation results obtained. 2. DESCRIPTION OF MAIN BATTERIES USED IN VEHICLE DOMAIN Among the batteries available on market, lithium-ion batteries and lithium-ion polymer batteries are the most used energy storage devices for plug-in and full battery electric vehicles, which and are expected to dominate the market by 2017 [4]. In Figure 2 are presented different types of energy storage devices. It can be observed that the batteries based lithium-ion presents the biggest values of energy and power densities comparing to NiCd (nickel cadmium), NiMH (nickel metal hydride) or Lead acid batteries. Power density is the amount of energy that can be provided in a time interval and energy density represents the capacity to store the energy. The difference between energy storage devices appears due to selected technology, the materials used for manufacturing the anode, cathode, electrolyte and the separator, specific to each battery. Nowadays, Li-ion polymer batteries are becoming among the most significant technologies for the next generation of EVs industry, with the mention

that, these batteries have to be carefully monitored and managed (electrically and thermally) in order to reduce the problems related to safety (inflammability) and performance [5]. that the lithium salt electrolyte is kept in a solid polymer composite such as polyacrylonitrile, not in an organic solvent. Their first appearance was in 1995 when engineers succeeded to replace the hard shell with flexible foils. Figure 2. Specific energy and specific power by type of battery including supercapacitors Lead acid batteries composition is made from a leaddioxide cathode, a sponge metallic lead anode and a sulphuric acid solution electrolyte, which make these batteries to be toxic for the environment. Drawbacks include also the fact that these batteries are very heavy, during charging presents the danger of overheating, their life cycle is between 300 and 500 cycles, etc. Their advantages refer to: low internal impedance, robustness, reliability, lower costs, existence of wide range of capacities and sizes etc. The major applications of leadacid batteries are starting, lighting and ignition (SLI), including small portable equipment and stationary applications [6] Nickel-metal hydride batteries (NiMH) were patented in 1986 by Stanford Ovshinsky. They are very similar with the nickel cadmium batteries (NiCd), except that their anode is made from a less toxic hydride alloy. NiMH can storage energy up to three times more than the same sized NiCd, but having the disadvantage that under high loads they suffer faster discharge rates. Their applications refer to: mobile phones, camcorders, cameras, toothbrushes, medical instruments and equipment, automotive batteries and high power static applications. They are most expensive that the lead acid batteries, but more friendly for the environment. Lithium-ion (Li-ion) batteries are the ideal solution for the HEVs applications, most off all because they can deliver an optimal combination of high power, long cycle life and also reducing the battery weight. Compared with NiMH batteries, Li-ion batteries provide: higher power density allowing longer fuel efficiency and greater acceleration capability; higher energy density (having more energy per unit weight) which can reduce the vehicle weight and can increase the space in the power train due to a compact design; and zero memory effect meaning a bigger operating life of the battery. In Figure 3 are indicated different configurations of Li-ion batteries. Lithium-ion polymer batteries (Li-poly) are rechargeable batteries and differ from the Li-ion batteries by the fact Figure 3. Different configuration of Li-ion battery: a) cylindrical; b) coin; c) prismatic and d) thin and flat. The major advantages refer to: a lower cost of manufacture, reliability, resistant to physical trauma, much lighter and the fact that can be adapted to a wide variety of packaging shapes. Also, Li-ion polymer batteries have over 20% higher energy density than Liion batteries and are more resistant to overcharge. They provide 500 charge/discharge cycles with no memory effect. Li-ion polymer battery has very low volatility and flammability, being environmentally acceptable (the risk of leakage is eliminated due to the use of the solid electrolyte) [7]. 3. DESCRIPTION OF HEAT TRANSFER According to Muratori [8], in order to develop a thermal model, it has to be considered two major types of heat transfer. The first refers to the propagation of heat inside the battery, which is characterized by a nonstationary problem with nonhomogeneous boundary conditions. The second concerns the problem of energy conservation for analyzing heat transfer between the battery and the environment in which they are located. Energy can be transferred to or from a system by heat or mechanical work. In the present case, the energy transfer is due to the difference in temperature or heat. The equation used to calculate the energy balance of the system of energy received or transferred (first law of thermodynamics) can be written as: Q in Q E de (1) out gen termic Heat is a form of energy due to the presence of thermal gradient. The main methods of heat transmission are divided into three categories: - conduction heat is transferred by direct contact of particles of a body; in the proposed model conduction is described by the following expression:

T Cp kx ky kz q t x x y y z z (2) where is the density, C p is the heat capacity at constant pressure, ki (i = x, y, z) and thermal conductivity noted with q. - convection is the transfer of heat between a surface and a moving fluid: q h ( T T ) (3) conv s where h is the specific heat transfer coefficient by convection, and is independent of fluid properties. - radiation is the heat transfer from one body to another through electromagnetic waves, in the case of the proposed model was used expression: Figure 4. Schematic representation of model grouping of battery cells 4 s 4 q ( T T ) (4) rad where is the emissivity value for the body, T s and T represents the surface and ambient temperature and σ is the Stefan Boltzmann constant 8 W 5,67 10 2 4. mk In case of heat transfer by conduction, the main sources of heat generation in battery are due to Gibbs free energy changes or chemical reactions and Joule effect. Thus, the Gibbs free energy change is given by following expression: G H T S (5) where ΔH is the variation of enthalpy, T is the temperature and ΔS is the variation of entropy. Usually, in order to analyze a battery, the cells are grouped in series/parallel to increase the current or the voltage values. In Figure 4 are represented three types of cells grouped in modules and packs. For cooling a battery pack can be used two simple solutions based on air flow among them. Therefore, the first method consists in cooling based on a series configuration. As can be seen in Figure 5a, the air volume enters at one end of the stack and exits on the opposite side using the same air flow. Using the second method, shown in Figure 5b, is observed that for the same set of battery modules, because the total air flow is distributed equally to each cooling each battery module. Depending on the total number and the size of modules, it s possible to combine both cooling mechanism (series/parallel), in order to obtain better results. a) b) Figure 5. Cooling system based on air flow of an battery module Thus, due to the fact that irreversible damage can be made during the life cycle of li-ion battery, a study based on thermal modelling was proposed using a prediction model method realised with dedicated software. The prediction model results were compared and validated with the experimental results. 4. DESCRIPTION OF THE THERMAL MODEL In order to describe the thermal behavior of an energy storage device, the technical and operational aspects of the model are developed using a specialized program from Comsol Multiphysics. This program consists in different software packages used to simulate physical processes. The schematic diagram from Figure 6 represents the general structure of the model. Figure 6. Schematic representation of the model Model inputs were chosen as they would have faced a real test bench. Setting a specific cycle, a function I(t)

representing the current variation requested/provided, considering the ambient temperature and the surface temperature of the object, for observing properly the battery temperature distribution on the surface, the distribution of current density, the voltage and the gamma of temperature measurements. The battery considered for modeling, is composed from a total of 30 elementary cells connected in parallel. Cathode of a cell is made of a compound of LiMnCoNiO 2 and graphite as active material. The first step in development of battery model is presented in Figure 7b, where it can be observed the battery as a whole, and Figure 7a, where is structure by the 30 elementary cells. Since, the distribution of heat between the battery cell layers doesn t concern in this study, it was considered a tridimensional type based on finite element analysis with linear variation. Thus, it s obtained a discretization with more than 40000 tridimensional prismatic elements. The number of elements per unit surface of external contacts is lower than in the rest of battery because these parts which are characterized by a specific homogeneity of their material have no need to exaggerate the number of elements to get good results as shown in Figure 8. Figure 9. Current density distribution Figure 10. Battery temperature distribution Therefore, it can be observed that the model results simulates properly the thermal behavior of battery. 5. DESCRIPTION OF EXPERIMENTAL TESTS CARRIED OUT ON A LI-POLY BATTERY a) b) Figure 7. Battery model structure In addition, some experiments were performed, in order to validate the theory and modelling data obtain with the software Comsol. These test experiments were conducted in order to characterize the functioning of the battery examined with a series of standard measures. In Figure 11 is presented the block diagram of the test bench consist in: a power supply, an electronic load, a data acquisition (DAQ) and the battery under test. Figure 8. Mesh applied to battery model structure Figure 9 presents the characteristics of current density on the surface of the battery the model, which is indicated in white color lines. It can be seen that the current density lines in contact with positive contact is nearly vertical. The Figure 10 represents the distribution of temperature on the outer surface of the battery, which is not a symmetric one, due to the fact that the hottest area is near to the positive contact. Figure 11. The block diagram used for the experimental results. The configuration with a large number of thermocouples, has allowed the evaluation of the temperature difference along the outer surface of the battery. The technical specifications of the battery are listed in Table 2. There were analyzed three configurations of thermocouples on the surface of the battery. The first arrangement is presented in Figure 12. It can be observed

Table 2. Technical specification of battery under test Parameters of battery Rated Capacity Nominal Voltage Max. Continuous Charge Current Max. Continuous Discharge Current Operation Temperature Range Cell Dimension Weight Values Tip. 41 Ah, Min. 40 Ah 3,7 V 120 A 320 A Charge: 11 45 C Discharge: -21 56 C Length: Max. 223 mm Width: Max. 213mm Thickness: Max. 10,6mm Max. 1030 g in the Figure 14 and Figure 15, that most significant heat peaks occurs at higher values of discharge current as in Figure 13. The highest values of temperature were recorded by thermocouple T14, which is located near the positive contact. Figure 15. Temperature evolution on the back side, depending by the thermocouples T14, T12 and T2. In order to find the optimal number of thermocouples that can be placed on the battery, next were analyzed different configurations that can provide significant results and can reveal the hottest zones of the battery, considering also the influence of the external contacts and power cables. In Figure 16 is presented the second configuration, where the results related to this arrangement are shown in Figure 18 based on evolution of the current discharge represented in the Figure 17. Figure 12. Distribution of thermocouples on faces of the cell: a) front side respectively b) back side. Figure 16. Distribution of thermocouples in second configuration, including heating based cables Figure 13. Evolution of discharge current in first arrangement. Figure 17. Evolution of discharge current for second arrangement Figure 14. Temperature evolution on the front side, depending by the thermocouples T10, T8 and T6. Figure 18. Temperature evolution depending by the second arrangement of thermocouples

The third configuration of thermocouples was realized by placing the sensors, mainly in the most significant thermal zones (Figure 19). Due to this simplified configuration, in Figure 20 is presented the temperature evolution and the current discharge characteristics. It can be observed that the discharge current is increased until it reaches the value of 320 A. Figure 19. Distribution of thermocouples in third configuration 6. CONCLUSIONS This paper presents a thermal testing model and the simulation results using a superior lithium ion polymer battery where the main concern it is represented by the difference of temperature distribution along the outer surface of the battery. The two methods permit to explore the strong relationship between the world of simulation and the experimental one, appreciate the mutual indispensability of the two approaches in order to obtain reliable results. The information obtained regarding the distribution of temperature on the surface of the battery results in a variety of considerations concerning the construction of cells, especially the location of the external contacts. From a thermal point of view, it would be better to be placed on opposite sides of the external contacts, just to make more uniform the level of temperature on the battery. Due to the fact that usually, the batteries are used being connected in series/parallel where besides the thermal requirements, other needs are taken into account, such as: feasibility, repeatability and simplicity of the layout of the connections between the various elements. Thus, a study regarding the heating module-level pack should be realized. 7. ACKNOWLEDGMENTS Figure 20. Evolution of discharge current (max. 320A) and battery temperature in third configuration Figure 21 shows the comparison between the development cycle in the model and experimental tests. It can be observed that the two most noticeable differences between the model and the experimental results are: after 200s and almost after 600s. This difference can be explained by the fact that for reducing the level of complexity was introduced some approximations but taking into account all the physical aspects and their interactions. Figure 21. Comparison between simulation and experimental values of temperature evolution, depending by thermocouple T8 This paper was realized with the support of EURODOC Doctoral Scholarships for research performance at European level project, financed by the European Social Found and Romanian Government, Department of Batteries from C.R.F. Turin and 4D-POSTDOC, contract no. POSDRU /89 /1.5 /S/ 52603, project co-funded by the European Social Fund through Sectoral Operational Programme Human Resources Development 2007-2013. 8. REFERENCES [1] Chan C.C., The State of the Art of Electric, Hybrid, and Fuel Cell Vehicles, Proceedings of the IEEE, Vol. 95, Issue 4, April 2007, pp. 704-718 [2] Kai S., Qifang S., Overview of the types of battery models, Control conference (CCC), 2011 [3] Markel T., Simpson A., Cost-Benefit Analysis of Plug-In Hybrid Electric Vehicle Technology, World Electric Vehicle Association (WEVA) Journal, Vol. 1, 2007. [4] Deutsche Bank (2009), Autos & auto parts Electric Cars: Plugged in 2. from http://www.db.com. [5] Yinjiao X., Qiang M., K. L. Tsui, M. Pecht, Prognostics and health monitoring for lithium-ion battery, IEEE Intelligence and Security Informatics, 2011. [6] Electropaedia web site: www.mpoweruk.com [7] Explanation of common battery technologies, in: www.windowsnetworking.com/articles_tutorials/batteries.html [8] Muratori M., Thermal characterization of lithium-ion battery cell, Thesis, Politecnico di Milano, Italy, 2008.