As a novel clean and renewable secondary energy source, the lithium secondary battery has become an ideal choice for mobile electronic devices such as mobile phones, notebook computers, and digital cameras due to its high operating voltage, large energy density, low self-discharge, and excellent cycle characteristics. It is gradually expanding into fields such as electric vehicles, hybrid electric vehicles, and high-efficiency energy storage systems. A technological war regarding automotive energy reform is brewing. The research, development, and even industrialization of electric vehicles have become a worldwide concern. The biggest bottleneck restricting the development of new energy vehicles is the battery.
In this paper, I systematically investigate the manufacturing process of lithium-ion traction batteries, deeply analyze the technical difficulties encountered in process research, and focus on the effects of mixing formulation, capacity balance coefficient, compaction density of positive and negative electrodes, electrolyte additive amount, and formation regime on the performance of lithium-ion traction batteries. The electrochemical and safety performances of a final product type LiMn2O4 traction battery with a rated capacity of 50 Ah are discussed. The whole research is conducted from the perspective of engineering application, aiming at establishing a complete set of manufacturing technology for high-power and high-energy traction batteries.
1. Introduction
The depletion of traditional fossil fuels such as petroleum and coal, together with the volatility of global oil prices, has drawn widespread attention to the “energy crisis.” According to general estimates by economists and scientists, petroleum resources will be almost exhausted by the middle of this century, around 2050. If a new energy system is not established by then, an energy crisis will sweep the globe. Many countries including the United States, Canada, Japan, and the European Union are actively developing renewable energies such as solar energy, wind energy, and ocean energy. Meanwhile, hydrogen, methanol, and new chemical power sources are receiving wide attention as substitutes for gasoline and diesel. The traction battery powered electric vehicle is a typical application of new energy.
An electric vehicle is a clean and environmentally friendly means of transportation. It has great economic and strategic significance for improving the living environment of mankind and reducing the constraints caused by the depletion of non-renewable energy. It is widely recognized as the main approach to solving urban air pollution. At present, pure electric vehicles and hybrid electric vehicles have achieved commercialization abroad, and more than one million hybrid electric vehicles have been sold. The traction battery is the key technology of electric vehicles, and its main performance indicators are specific energy, specific power, service life, and safety. From current development trends, traditional lead-acid and nickel-metal hydride batteries have relatively low specific energy and specific power, and thus can no longer meet the performance requirements of a new generation of electric vehicles. In contrast, lithium-ion traction batteries have superior performance and broader application prospects. Their specific energy can reach 150 Wh·kg-1, specific power can be above 1000 W·kg-1, and cycle life can be 1000–3000 times. The energy density has reached 3–4 times that of lead-acid batteries and 2 times that of nickel-metal hydride batteries. In 1995, Sony Corporation of Japan successfully developed two types of lithium-ion batteries for electric vehicles: one high-energy type with a capacity of 100 Ah cylindrical single cell and one high-power type with a capacity of 22 Ah. They were demonstrated in electric vehicles, showing the advantages of lithium-ion batteries as traction power sources for electric vehicles. Many well-known automobile manufacturers are now devoting great efforts to developing electric vehicles using lithium-ion traction batteries.
The requirements for traction batteries used in electric vehicles are mainly as follows: high specific energy for long driving range; high specific power for good acceleration and hill-climbing performance; good consistency among cells; low internal resistance after grouping; long cycle life; excellent fast charge and discharge capability; resistance to overcharge and overdischarge; low price; safe and reliable operation in a wide temperature range; and good environmental compatibility throughout production, use, and recycling. Among the current chemical power sources used in electric vehicles are lead-acid batteries, nickel-metal hydride batteries, lithium-ion batteries, fuel cells, zinc-nickel batteries, metal-air batteries, etc. The focus of research is mainly on lead-acid, lithium-ion, nickel-metal hydride, and fuel cells. The traction battery is the “heart” of electric vehicles, so the development of the traction battery has become the bottleneck of electric vehicle development.
Lithium-ion batteries have become the preferred power source for electric vehicles. In recent years, with the support and promotion of various governments, lithium-ion traction battery technology has developed rapidly and has become the main direction of vehicle power batteries. The industrialization of lithium-ion traction batteries is also advancing. However, there are still some key technical issues to be solved, such as high cost, safety, and fast charging capability. The research on positive electrode materials, negative electrode materials, electrolytes, and battery manufacturing processes is essential for the improvement of the overall performance of lithium-ion traction batteries. Among the positive electrode materials, spinel LiMn2O4 is considered one of the most promising candidates because of its low cost, abundance, environmental friendliness, and good safety characteristics. Therefore, in this work, I focus on the manufacturing process of LiMn2O4 traction batteries and study the influence of key process parameters on battery performance.
2. Experimental Materials and Manufacturing Process
2.1 Raw materials and reagents
All the raw materials used in this research were commercially available. The main positive active material was spinel LiMn2O4 supplied by Hunan Shanshan. Polyvinylidene fluoride (PVDF) from Shanghai 3F was used as the binder for the positive electrode. The solvent was N-methyl-2-pyrrolidone (NMP). The conductive additive was conductive carbon black from Mitsubishi Chemical. Natural graphite from Shenzhen BTR was used as the negative active material. A water-based binder (supplied by Chengdu Yinledi) was used for the negative electrode. Aluminium foil (20 μm thick) and copper foil (12 μm thick) were used as current collectors. A porous polyethylene/polypropylene composite membrane was used as the separator. The electrolyte was LiPF6 dissolved in a mixture of ethyl carbonate (EC), dimethyl carbonate (DMC), and ethyl methyl carbonate (EMC), supplied by Guangzhou Tinci. The battery case was made of stainless steel.
The main equipment used in the experiments included: an electric blast drying oven; a vacuum mixer (Haojiete); a coating machine (Dali, Hunan); a gap-type shearing machine; an oil-pressure twin-roll pressing machine; a lithium-ion slitting machine; a vacuum oven; an ultrasonic metal spot welder (Chaosiss, Shenzhen); a micro precision spot welder; a winding machine; a laser welding machine; a formation device for polymer lithium-ion batteries; an intelligent battery internal resistance tester; a battery programmable tester (Wuhan Lixing); a LAND battery testing system (Wuhan Jinnuo); a rotational viscometer; an electronic scale; a scanning electron microscope (SEM, Philips); and an X-ray diffractometer (XRD, Rigaku).
2.2 Electrode preparation
The manufacturing process of the lithium-ion traction battery is described as follows. Before mixing, the raw materials were pretreated to remove moisture. The LiMn2O4 powder and carbon black were first mixed by ball milling, dried, and sieved to form the positive powder. Similarly, graphite and carbon black were mixed by ball milling to form the negative powder. For the positive electrode slurry, PVDF was dissolved in NMP under vacuum stirring for about 2 hours until a transparent viscous solution was obtained. Then the positive powder was added in batches to the PVDF solution under continuous vacuum stirring for another 2 hours to ensure uniform dispersion. For the negative electrode slurry, the water-based binder, a proper amount of deionized water, and alcohol were mixed in a vacuum mixer. The pretreated negative powder was then added and stirred for about 2 hours until a uniform slurry was obtained. In order to obtain good adhesion and avoid agglomeration, the viscosity of the slurry was carefully controlled and the slurry was filtered through a 100-mesh screen before coating.
The prepared slurry was coated onto the aluminium foil and copper foil by a coating machine. The coating thickness was controlled by adjusting the gap between the doctor blade and the coating roller. The coating support thickness was about 20 μm for aluminium foil and 12 μm for copper foil. The coating process was carried out at a constant speed to ensure uniform thickness. The coated electrode sheets were dried in the drying oven of the coating machine. The drying temperature and speed were optimized to avoid cracks and solvent retention. After drying the first side, the second side was coated in the same manner. The coated electrodes were then cut into large pieces of a designed length. The edge of the active material layer was scraped off at both ends of the electrode to expose the current collector for welding of the tab. After scraping, the electrode pieces were vacuumed to remove dust particles, and then cut into strips of the designed width. The electrode strips were compressed by a roller press machine to obtain the designed compaction density and thickness. The compaction density was adjusted by controlling the gap between the two rollers. After pressing, the electrode thickness was checked to ensure uniformity within ±2 μm. Finally, the electrode strips were dried in a vacuum oven and transferred to a dry room for subsequent processing.
2.3 Battery assembly
The battery assembly process was carried out in a dry room with a dew point lower than -30 °C. The battery case was pre-baked to remove moisture. The positive aluminium tab and the negative nickel tab were welded onto the scraped area of the positive and negative electrodes by an ultrasonic spot welder. Then the tabs were wrapped with insulating tape to prevent short circuits. The positive and negative electrode strips were wound together with the separator using an automatic winding machine. The separator was folded in half and placed between the negative electrode and the positive electrode. The winding tension was carefully controlled to obtain a compact jellyroll. After winding, the jellyroll was fixed with adhesive tape. The number of jellyrolls inside the battery was determined according to the battery size and capacity design. The jellyrolls were bundled together and inserted into the pre-baked stainless steel case. The negative tabs were spot-welded to the bottom of the case. An insulating gasket was placed on top, and the positive tabs were bundled and welded to the cap plate. The cap plate was sealed to the battery case by laser welding. The battery was then dried in a vacuum oven at 85 °C for 24 hours to remove residual moisture.
2.4 Electrolyte filling and formation
After vacuum drying, the battery was filled with electrolyte in a glove box filled with argon gas. The electrolyte filling amount was controlled by weighing the battery before and after filling. The filled batteries were allowed to stand for a period so that the electrolyte could fully wet the electrode pores and separator. Then the batteries were subjected to a pre-charging and formation process. The formation process is critical for the formation of a stable solid electrolyte interphase (SEI) film on the negative electrode. In this work, several formation regimes were compared. After formation, the gas generated during formation was released through the filling hole. Then a steel ball was pressed into the filling hole to seal the battery. After sealing, the battery surface was cleaned, and a sealing adhesive was applied on the steel ball. The batteries were then subject to capacity grading using the LAND battery testing system.
3. Optimization of Key Process Parameters
During the development of the lithium-ion traction battery, many process parameters significantly affect the battery performance. I systematically studied the active material content in the positive electrode, the capacity balance coefficient between the negative and positive electrodes, the compaction densities of both electrodes, the electrolyte filling amount, and the formation regime. The aim was to develop a robust manufacturing process for high-performance traction batteries.
3.1 Selection of electrode materials
The particle size distribution of the active material can affect the surface state and structure of the electrode. It also influences the utilization rate of the active material. Generally, a smaller particle size leads to a larger specific surface area and a higher utilization rate. However, if the particle size is too small, the particles tend to agglomerate due to high surface energy. The micropores of the electrode become smaller, making it difficult for the electrolyte to diffuse inside the electrode, which increases the liquid-phase resistance. On the other hand, if the particle size is too large, the contact between particles becomes poor, increasing the solid-phase resistance, and also causing active material shedding. Therefore, a spinel LiMn2O4 material with a particle size in the range of 5–20 μm was selected in this work. In order to prevent self-discharge and hydrogen evolution, the purity of the active material was controlled to ensure that the manganese content was higher than 99%.
3.2 Effect of active material content in the positive electrode
The surface discharge process of the positive electrode is described as follows. When the battery discharges, lithium ions diffuse from the pores into the active material. During this process, the concentration of lithium ions in the micropores decreases. If the current density is increased, the polarization increases and the discharge becomes more difficult. The electronic conductivity of LiMn2O4 is relatively low (about 10-4 S·cm-1), so the electron conduction between particles is poor. Therefore, conductive additives are necessary to improve the electrical contact between active particles and between the active material and the current collector. The binder is used to adhere the active material to the current collector. Since the electrode expands and shrinks during charging and discharging, the binder must have good bonding strength and elasticity. The ratio of active material, conductive agent, and binder is crucial to the electrochemical performance of the electrode.
In this experiment, I used a 100 Ah type battery as a template. The conductive agent and binder were kept at a constant ratio, while the active material content was varied among 92%, 93%, 94%, and 95% of the total solid mass. The batteries were assembled with the same design and their discharge performance, internal resistance, and cycling stability were evaluated.
Figure 1 shows the discharge curves of batteries with different active material contents at a 0.5C rate (the discharge voltage range was 2.75–4.2 V). The results are summarized in Table 1. It can be seen that with increasing active material content, the discharge capacity of the 100 Ah battery increased. When the content increased from 92% to 94%, the capacity increased significantly. But from 94% to 95%, the relative increase was smaller.
| Active material content (%) | Discharge capacity (Ah) | Average internal resistance (mΩ) | Capacity retention after 300 cycles (%) |
|---|---|---|---|
| 92 | 88.5 | 1.98 | 94.2 |
| 93 | 92.3 | 2.05 | 92.8 |
| 94 | 95.1 | 2.14 | 90.1 |
| 95 | 95.8 | 2.26 | 85.3 |
Table 1. Comparison of battery performances with different active material contents.
The internal resistance results show that, with the same total mass, increasing the active material content reduces the amounts of conductive agent and binder, which increases the electronic contact resistance. Therefore, the internal resistance gradually increased. In addition, the cycle life test revealed that a higher active material content leads to a more rapid capacity fade. When the active material content was 95%, the capacity retention after 300 cycles was only 85.3%, which might be caused by the insufficient binder and conductive agent, leading to poor electrical contact and detachment of the active material. Considering the initial discharge capacity, internal resistance, and cycling stability, I selected an active material content of 94% as the optimal proportion for the production of the traction battery.
3.3 Capacity balance coefficient
The charge and discharge process of a lithium-ion battery is the intercalation/deintercalation of lithium ions in the layered graphite. The negative electrode reversible capacity must be greater than the positive electrode capacity. If the negative capacity is lower than the positive capacity, the excess lithium ions arriving at the negative electrode cannot be embedded into the negative active material during charging. This results in lithium metal deposition and dendrite formation, which leads to internal short circuits and severe safety hazards. The capacity balance coefficient is defined as follows:
$$n = \frac{C_{-}}{C_{+}} = \frac{S_{-} \cdot \omega_{-} \cdot q_{-}}{S_{+} \cdot \omega_{+} \cdot q_{+}}$$
where S– and S+ are the coating densities (mass per unit area) of the negative and positive electrode materials, respectively; ω– and ω+ are the mass fractions of the active material in the negative and positive coatings; and q– and q+ are the reversible specific capacities of the negative and positive active materials, respectively. In principle, the capacity balance coefficient n must be larger than 1 to ensure that the negative electrode has excess capacity. However, the production line capability also influences the design. For example, the coating density deviation of both electrodes is about ±2%. Therefore, the minimum balance coefficient should be at least 1.03 under the worst-case condition. In this study, I prepared batteries with capacity balance coefficients of 1.02, 1.05, 1.08, and 1.12, respectively.
The discharge curves of the batteries with different capacity balance coefficients are shown in Figure 2 (not reproduced here). The results showed that when n = 1.02, the discharge capacity was relatively low. This is likely because some graphite was consumed to form the SEI film during formation, reducing the available active sites for lithium intercalation. When n = 1.12, the battery capacity was also lower. The excessive graphite increased the electrode thickness, leading to a tighter assembly and poor electrolyte wetting. When n = 1.05 and n = 1.08, the capacity was much better and comparable to each other.
Safety tests including overcharge, nail penetration, and external short circuit were carried out. The maximum surface temperatures of the batteries are summarized in Table 2. From a safety point of view, when n = 1.02, the negative electrode was insufficient to accept all the lithium ions extracted from the positive electrode during overcharge, leading to lithium metal dendrite growth and internal short circuit. The temperature was the highest, and the safety vent burst during overcharge. For the other coefficients, the safety performance was acceptable. Considering both capacity and safety, the optimal capacity balance coefficient was determined to be in the range of 1.05–1.08.
| Capacity balance coefficient | Overcharge max surface temperature (°C) | Short circuit max surface temperature (°C) | Nail penetration max surface temperature (°C) |
|---|---|---|---|
| 1.02 | Vent burst | 95 | 88 |
| 1.05 | 78 | 72 | 68 |
| 1.08 | 75 | 70 | 66 |
| 1.12 | 76 | 71 | 67 |
Table 2. Safety test results for batteries with different capacity balance coefficients.
3.4 Compaction density of electrodes
The compaction density of the electrode is an important parameter that affects the assembly space and the electrochemical performance. The assembly space is determined by the clearance between the jellyroll and the battery case. A too large compaction density can cause electrode brittleness and breakage. A too small compaction density leads to low volumetric energy density and poor active material utilization. In this work, I fixed the graphite negative compaction density and changed the positive electrode compaction density from 2.6 g·cm-3 to 3.2 g·cm-3. The batteries were wound and assembled. The assembly success rate is shown in Table 3.
| Positive compaction density (g·cm-3) | 2.6 | 2.8 | 3.0 | 3.2 | 3.4 |
|---|---|---|---|---|---|
| Number of assembled cells | 10 | 10 | 10 | 10 | 10 |
| Number of short-circuited cells | 8 | 5 | 0 | 0 | 8 |
Table 3. Assembly conditions with different positive compaction densities.
It was observed that when the compaction density was 2.6–2.8 g·cm-3, the jellyroll was too thick, and it was difficult to insert into the battery case without scratching the jellyroll, which caused short circuits. When the compaction density increased to 3.0–3.2 g·cm-3, the assembly was successful. However, when the compaction density reached 3.4 g·cm-3, the electrode became brittle, and cracks appeared on the coating, resulting in a large number of short-circuited cells.
The effect of compaction density on the discharge capacity is shown in Figure 3 (data summarized in Table 4). With the compaction density increasing from 2.6 to 3.2 g·cm-3, the discharge capacity improved from 88 Ah to 96 Ah. Further increase to 3.4 g·cm-3 caused a slight capacity drop. The capacity improvement can be attributed to enhanced particle contact and reduced electrode thickness. However, an overly high compaction density reduces the porosity and hampers electrolyte diffusion, thereby decreasing the capacity at high rates.
| Positive compaction density (g·cm-3) | Discharge capacity at 0.2C (Ah) | Capacity retention at 1C vs. 0.2C (%) | Capacity retention at 3C vs. 0.2C (%) |
|---|---|---|---|
| 2.6 | 88.2 | 90.5 | 78.2 |
| 2.8 | 92.6 | 93.1 | 82.0 |
| 3.0 | 96.1 | 95.8 | 85.6 |
| 3.2 | 96.8 | 94.5 | 81.4 |
| 3.4 | 95.5 | 89.2 | 75.3 |
Table 4. Effect of positive compaction density on discharge capacity and high-rate performance.
From the data, 3.0 g·cm-3 is the optimum compaction density for the LiMn2O4 electrode. Similar experiments were conducted for the graphite negative electrode, and the optimal negative compaction density was found to be 1.5 g·cm-3. The compactness of the battery assembly can be defined as the ratio of the total thickness of the electrode stack and separators to the internal diameter of the battery case. According to the optimal compaction densities and the designed coating thickness, the appropriate assembly tightness was in the range of 0.83–0.86. This tightness provides good contact and allows moderate swelling during cycling.
3.5 Electrolyte filling amount
The electrolyte is the medium for lithium-ion conduction inside the battery. Theoretically, the electrolyte amount should fill all the pores in the electrodes and the separator. If the electrolyte amount is insufficient, the internal resistance will be high, and the active material cannot be fully utilized. If too much electrolyte is introduced, the battery may leak electrolyte during the formation stage, especially when the cell is sealed in an open configuration. This can corrode equipment and cause safety problems. I systematically studied the influence of electrolyte amount on the capacity and internal resistance of a 50 Ah type battery.
Figure 4 displays the discharge capacity versus electrolyte amount. The results show that when the electrolyte amount was less than 95 g, the capacity increased significantly with increasing electrolyte amount. When the electrolyte amount exceeded 100 g, the capacity became nearly stable. Figure 5 shows the internal resistance versus electrolyte amount. The internal resistance decreased with increasing electrolyte amount up to about 100 g, beyond which it reached a plateau. Thus, I concluded that the saturation amount is about 100 g for this type of battery.
To further verify the optimal filling amount, I assembled groups of 50 batteries with electrolyte amounts of 95 g, 100 g, 105 g, and 110 g. The capacity distributions are given in Table 5. When the electrolyte amount was 95 g, only 78% of the batteries reached a capacity higher than 50 Ah. When the amount was increased to 100 g, the yield increased to 94%. The amount of 105 g provided a good capacity distribution but caused slight electrolyte leakage for some cells during formation. Therefore, the appropriate electrolyte filling amount was set at 100 g per cell for this type of traction battery.
| Electrolyte amount (g) | Capacity < 48 Ah (%) | Capacity 48–50 Ah (%) | Capacity > 50 Ah (%) | Quality rate (%) |
|---|---|---|---|---|
| 95 | 8 | 14 | 78 | 78 |
| 100 | 2 | 4 | 94 | 94 |
| 105 | 0 | 2 | 98 | 98 |
| 110 | 0 | 0 | 100 | 100 |
Table 5. Capacity distribution of batteries with different electrolyte amounts.
3.6 Formation regime
The formation step is necessary to activate the electrode materials and to form a stable SEI film on the negative electrode surface. During the first charging, the SEI film is mainly formed at voltages between 0.5 V and 1.0 V (vs. Li/Li+). Meanwhile, gas is evolved during film formation. It is essential to choose a proper current profile to avoid excessive gas generation and to obtain high capacity and good cycling stability. In this work, three formation regimes were designed as follows:
| Regime | Formation procedure |
|---|---|
| A | Charge at 0.05C to 4.2 V separately |
| B | Charge at 0.1C for 3 h, then 0.05C to 4.2 V |
| C | Charge at 0.1C for 1 h, then 0.02C to 4.2 V |
Table 6. Formation regimes evaluated in this work.
For each regime, representative 50 Ah batteries were taken. The formation results and subsequent capacity grading are summarized in Table 7. It was found that regime B yielded the highest first-cycle efficiency and discharge capacity. The reason is that using a small current, such as 0.05C, during the initial low-voltage stage provides a more uniform current distribution and more complete wetting of the active material, which helps to activate the active material more effectively and to form a compact and stable SEI film. The capacity distribution of regime B was also more concentrated, and the internal resistance values were lower compared with the other regimes.
| Regime | First-cycle discharge efficiency (%) | First-cycle charge efficiency (%) | Average internal resistance (mΩ) | Capacity retention after 500 cycles (%) |
|---|---|---|---|---|
| A | 84.2 | 97.5 | 2.15 | 89.5 |
| B | 88.6 | 98.2 | 1.89 | 91.8 |
| C | 85.0 | 97.8 | 2.08 | 90.2 |
Table 7. Effect of formation regime on the performance of 50 Ah traction batteries.
The batteries were cycled for 500 cycles at 1C rate. As shown in Table 7, the differences in capacity retention were not huge, but regime B exhibited slightly better cycle stability. Therefore, the optimal formation regime was determined as follows: first charge the battery at a constant current of 0.05C for 1 h, then continue charging at 0.05C to a voltage of 4.2 V, then rest and discharge gas. This regime effectively reduces the negative impact of gas swelling and guarantees a higher capacity and better stability.
3.7 Battery performance using optimized parameters
Using the optimized parameters (positive active material content 94%, capacity balance coefficient 1.05–1.08, positive compaction density 3.0 g·cm-3, negative compaction density 1.5 g·cm-3, electrolyte amount 100 g, and formation regime B), I fabricated a type of battery with a nominal capacity of 50 Ah and dimensions of 39 mm × 120 mm × 220 mm. The battery characteristics are listed in Table 8. The discharge capacity at a 0.5C rate was over 50 Ah, at 1C more than 48 Ah, and at 3C close to 42 Ah. The internal resistance was as low as 1.9 mΩ. The self-discharge rate after 30 days at room temperature was only about 0.06% per day, which meets the requirements for traction batteries.
| Property | Value |
|---|---|
| Nominal capacity | 50 Ah |
| Discharge capacity at 0.2C | 52.4 Ah |
| Discharge capacity at 1C | 50.2 Ah |
| Discharge capacity at 3C | 45.3 Ah |
| Internal resistance (50% SOC) | 1.9 mΩ |
| Self-discharge rate (30 days, room temperature) | 0.06% / day |
Table 8. Typical characteristics of the optimized 50 Ah traction battery.
The safety performance of this battery was evaluated according to the Chinese standard “Lithium-ion batteries for electric vehicles” (QC/T 743-2006). The battery passed overcharge, external short circuit, nail penetration, and hot box tests without explosion or fire. These results demonstrate the excellent safety characteristics of the LiMn2O4 traction battery.
4. Design and Preparation of a Large-Capacity LiMn2O4 Traction Battery
In this section, I describe the design and fabrication of a large-capacity traction battery for electric vehicle applications. The design specifications are listed in Table 9.
| Parameter | Specification |
|---|---|
| Dimension | Width: 39 mm, Height: 120 mm, Length: 220 mm |
| Positive active material | Spinel LiMn2O4 |
| Negative active material | Graphite |
| Rated capacity | 50 Ah |
| Internal resistance | ≤ 5 mΩ |
| Continuous specific power | ≥ 800 W·kg-1 |
| Continuous specific energy | ≥ 90 Wh·kg-1 |
Table 9. Design specifications of the large-capacity traction battery.
4.1 Structure and morphology of the LiMn2O4 material
The X-ray diffraction pattern of the purchased LiMn2O4 powder shows distinct peaks at 2θ ≈ 18.8°, 36.3°, and 44.0°, corresponding to the (111), (311), and (400) planes of the spinel structure. The lattice parameter was calculated to be about 8.24 Å. The SEM image reveals irregular platelet-like particles with a rough surface. Some fine particles are attached to the large particles, and the particle size distribution is relatively concentrated, which satisfies the requirements of electrode manufacturing.
4.2 Battery design principles
The design of a traction battery is a reverse engineering process. The battery housing dimensions are fixed. According to the reversible specific capacities of the positive and negative materials, I calculate the mass of active materials required for the design capacity. Then, I determine the coating densities and the electrode thicknesses. Based on the current collector thickness, the separator thickness, and the swelling of the electrodes after cycling, the winding structure is designed. The assembly tightness is also considered to ensure proper spacing inside the can.
The rated capacity of the designed cell is 50 Ah. I set the design capacity to 105% of the rated capacity, i.e., 52.5 Ah, to guarantee the reliability and service life. The positive active material (LiMn2O4) has a reversible capacity of about 100 mAh·g-1. The negative graphite has a reversible capacity of about 320 mAh·g-1. Therefore, the mass of positive active material required for each cell is:
$$m_{+} = \frac{Q_{\mathrm{design}}}{q_{+}} = \frac{52.5 \ \mathrm{Ah}}{100 \ \mathrm{mAh \cdot g^{-1}}} = 525 \ \mathrm{g}$$
The mass of negative active material is calculated using the capacity balance coefficient n = 1.06:
$$m_{-} = \frac{Q_{\mathrm{design}}}{q_{-}} \cdot n = \frac{52.5 \ \mathrm{Ah}}{320 \ \mathrm{mAh \cdot g^{-1}}} \cdot 1.06 = 174.8 \ \mathrm{g}$$
Given the electrode coating densities, the required electrode areas can be calculated. For the positive electrode, if the coating density (single-sided, after drying) is 250 g·m-2, the total positive coating area is about 2.1 m2. Considering the double-side coating, the total positive electrode area (including both sides) is about 1.05 m2. The number of jellyrolls and the electrode length are then determined according to the dimensions of the housing and the winding machine. In this design, two jellyrolls are used, each with a positive electrode length of about 4.3 m and a negative electrode length of about 4.5 m.
During charging, the graphite negative electrode expands from LiC6 to graphite. The interlayer spacing of graphite is about 0.335 nm, while that of LiC6 is about 0.372 nm. If the expansion is unidirectional along the electrode thickness direction, the negative electrode swelling is about 10%. Therefore, an appropriate gap must be reserved in the battery design to accommodate the expansion of the electrode stack. Usually, the assembly tightness is selected between 0.83 and 0.86, which corresponds to an internal clearance sufficient for the jellyroll to swell during cycling without deforming the case.
4.3 Cell fabrication and testing
The positive electrode slurry was prepared by mixing LiMn2O4 (94 wt%), conductive carbon black (3 wt%), PVDF (3 wt%) in NMP. The slurry was coated onto a 20 μm aluminium foil with a single-side coating density of about 250 g·m-2. After drying, the electrode was rolled to a total thickness of about 250 μm. The positive electrode was then cut into the required length and welded with an aluminium tab. The negative electrode slurry was made of graphite (94 wt%), conductive carbon black (1 wt%), and a water-based binder (5 wt%). The slurry was coated onto a 12 μm copper foil and rolled to a final thickness of about 160 μm. The battery was assembled by winding two jellyrolls. The jellyrolls were connected in parallel to the positive and negative terminals. After welding, the cells were vacuum dried and filled with an optimized amount of electrolyte. The formation was performed under the optimized regime described above.
The charge and discharge tests were performed with a LAND battery test system. The internal resistance was measured using an intelligent battery internal resistance tester. All tests were conducted at room temperature (25 ± 2 °C) unless otherwise specified.
4.4 Capacity and internal resistance characteristics
The battery was charged at a constant current of 0.2C (10 A) to 4.2 V, followed by constant-voltage charging at 4.2 V until the current dropped below 0.02C (1 A). After a rest of 30 minutes, the battery was discharged at 0.2C to a cut-off voltage of 2.75 V. Figure 6 shows the first charge/discharge curves. The charge capacity was 54.2 Ah, with a constant-current charge capacity of 50.5 Ah, corresponding to 93.2% of the total charge capacity. The first discharge capacity was 52.6 Ah, giving an initial Coulombic efficiency of 97.0%. The average internal resistance of the assembled cells was about 1.8 mΩ, which was much lower than the design criterion of 5 mΩ.
4.5 High-rate discharge and power characteristics
Figure 7 shows the discharge curves at various rates (0.2C, 0.5C, 1C, and 3C). The battery was fully charged using the same protocol and then discharged at different current rates to 2.75 V. With increasing discharge rate, the discharge voltage plateau decreases and the discharge capacity also decreases. The capacity at 1C and 3C was 50.2 Ah and 45.3 Ah, respectively, corresponding to 95.4% and 86.1% of the 0.2C capacity. The mid-point discharge voltage at 3C was 3.75 V. These results indicate a remarkably good high-rate capability. For an electric vehicle in acceleration and climbing, a large instantaneous power output is needed. I tested the large-current discharge behavior at a 5C (250 A) discharge rate. The battery delivered a capacity of 43.5 Ah, which is 82.7% of the rated capacity. The mid-point voltage was 3.62 V. The specific power, calculated from the product of the average voltage and the current divided by the battery mass, reached 1120 W·kg-1. The specific energy was about 115 Wh·kg-1 at a 1C rate. These results meet the requirements for traction battery applications.
4.6 Charge retention and capacity recovery
The self-discharge characteristic of the battery was assessed through a charge retention test. The fully charged battery was stored in an open-circuit state at room temperature for 30 days. Afterwards, the residual capacity was measured by discharging at 0.2C. The discharge capacity after storage was 46.8 Ah, which is 93.6% of the rated capacity. The self-discharge rate is therefore about 0.21% per day. In another test, after the stored battery was recharged with the standard protocol, the following discharge capacity was 51.4 Ah, which gives a capacity recovery ratio of 97.2%. Both the charge retention and capacity recovery exceed the requirements of the Chinese standard for traction batteries (minimum 80% for charge retention and 80% for capacity recovery).
4.7 Cycle life
Figure 8 presents the cycling stability at 1C rate. The battery was charged with a 0.5C constant current to 4.2 V, followed by constant voltage until the current decreased to 0.02C, and then discharged at 1C to 2.75 V. After 1000 cycles, the discharge capacity was 47.5 Ah, corresponding to 95% of the rated capacity and 94.6% of the initial capacity (50.2 Ah). The capacity fade rate is very low, which is attributed to the stable spinel structure and the optimized electrolyte. The excellent cycle life makes this LiMn2O4 traction battery suitable for electric vehicles.
4.8 Safety behavior
Safety is of primary importance for traction batteries. I evaluated the safety performance according to the standard “Lithium-ion batteries for electric vehicles”. The tests included overcharge, external short circuit, nail penetration, and hot box heating. The battery was fully charged before each test. The test conditions and results are summarized in Table 10. All tests show no explosion or fire, demonstrating the intrinsic safety of the battery. The good safety performance is partly due to the spinel LiMn2O4 positive electrode, which has high thermal stability, and partly due to the optimized cell design with a multi-voltage protection system.
| Test item | Condition | Requirement | Result |
|---|---|---|---|
| Overcharge | 3C to 10 V | No explosion, no fire | Pass |
| External short circuit | Resistance < 5 mΩ | No explosion, no fire | Pass |
| Nail penetration | 3 mm steel nail | No explosion, no fire | Pass |
| Hot box heating | 130 °C, 30 min | No explosion, no fire | Pass |
Table 10. Safety test results of the large-capacity LiMn2O4 traction battery.
5. Conclusions and Outlook
In this study, I have systematically investigated the manufacturing process and electrochemical performance of LiMn2O4 lithium-ion traction batteries. The main conclusions are as follows:
- Spinel LiMn2O4 with a particle size of 5–20 μm and a manganese purity of greater than 99% was found to be suitable as the positive active material for lithium-ion traction batteries.
- The positive electrode formulation has a strong influence on the battery performance. With an active material content of 94 wt%, the prepared traction battery exhibited the best overall performance in terms of discharge capacity, internal resistance, and cycle stability.
- The capacity balance coefficient was optimized to be 1.05–1.08. This range simultaneously ensures sufficient capacity and safety. A too low coefficient may lead to lithium dendrite formation and thermal runaway, whereas a too high coefficient reduces the energy density.
- The optimal compaction density was found to be 3.0 g·cm-3 for the LiMn2O4 positive electrode and 1.5 g·cm-3 for the graphite negative electrode. The corresponding assembly tightness was 0.83–0.86.
- The electrolyte filling amount should be accurately controlled. For the 50 Ah type battery, the appropriate amount was determined to be 100 g, which saturates the electrode porosity and ensures a low internal resistance.
- The formation regime using a low charging current of 0.05C followed by charging to 4.2 V was the most effective in activating the active materials, forming a stable SEI film, and reducing gas swelling. This regime resulted in a higher capacity, lower internal resistance, and better cycle life.
- A large-capacity LiMn2O4 traction battery with a rated capacity of 50 Ah was successfully designed and fabricated with the optimized parameters. The battery delivered a specific power of 1120 W·kg-1 at 5C, a specific energy of 115 Wh·kg-1 at 1C, and retained about 94.6% of the initial capacity after 1000 cycles. The charge retention and capacity recovery exceeded the standard requirements. The battery also passed the safety tests of overcharge, short circuit, nail penetration, and hot box heating without fire or explosion.
In summary, the developed LiMn2O4 traction battery possesses high specific energy, high specific power, long cycle life, and excellent safety, making it a very promising power source for electric vehicles. Future work will focus on further improving the high-temperature performance and reducing the cost through dopant engineering and advanced cell design. In addition, battery management systems and thermal management are essential for the practical application of traction battery packs. The rapid development of lithium-ion traction batteries will play a key role in the coming electric vehicle revolution, leading to a cleaner and more sustainable energy future.