In this thesis, I systematically investigated the manufacturing process of high-power lithium-ion batteries based on spinel LiMn₂O₄ cathodes and graphite anodes. The research was motivated by the increasing demand for high-performance, low-cost, and safe power sources for electric vehicles, where the EV battery pack remains the most critical and expensive subsystem. I focused on the key process parameters that determine the electrochemical performance, consistency, and safety of large-format cells suitable for EV battery pack integration. These parameters include the electrode formulation, capacity balance coefficient, compaction density of both positive and negative electrodes, electrolyte filling amount, and formation protocol. Through systematic experiments, I optimized these parameters and successfully fabricated a 60 Ah prismatic lithium-ion power cell. The cell exhibited excellent rate capability, long cycle life, high charge retention, and satisfactory safety characteristics. All results met or exceeded the requirements of the Chinese standard for lithium-ion traction batteries for electric vehicles. I believe that the manufacturing technology developed in this work can be directly scaled up for commercial EV battery pack production.
1. Introduction and Literature Review
1.1 Background and Motivation
The depletion of fossil fuels and the increasing severity of environmental pollution have accelerated the global transition toward clean and renewable energy. Electric vehicles are widely regarded as one of the most promising solutions to reduce greenhouse gas emissions and dependence on petroleum. However, the commercialization of electric vehicles largely depends on the development of advanced battery systems. Among all energy storage technologies, the lithium-ion battery has become the preferred power source for modern electric vehicles because of its high operating voltage, high energy density, low self-discharge rate, long cycle life, and environmental friendliness. A reliable EV battery pack must simultaneously satisfy the conflicting requirements of high specific energy, high specific power, long calendar life, and excellent safety under abusive conditions.
The global market for electric vehicles has expanded rapidly. By 2030, it is projected that the market for EV battery pack systems will reach several hundred gigawatt-hours annually. Major automobile manufacturers in the United States, Japan, Europe, and China have invested heavily in the research and development of lithium-ion traction batteries. In my opinion, the key technical bottleneck is no longer the basic chemistry but the engineering of cell manufacturing processes that can guarantee high yield, low cost, and consistent quality. This is especially true for large-format cells used in EV battery pack assemblies, where the failure of a single cell can compromise the entire pack.
1.2 Overview of Power Batteries for Electric Vehicles
Various electrochemical power sources have been considered for electric vehicles, including lead-acid, nickel-metal hydride, lithium-ion, and fuel cells. Each technology has its own advantages and limitations. In the early stage of electric vehicle development, lead-acid batteries were the most widely used due to their low cost and mature manufacturing technology. However, their low specific energy and relatively short cycle life severely limited vehicle range. Nickel-metal hydride batteries offered better energy density and power capability, and they were successfully commercialized in hybrid electric vehicles such as the Toyota Prius. Nevertheless, their high self-discharge rate, high-temperature performance degradation, and relatively high cost limit their further deployment in full electric vehicles.
Lithium-ion batteries, by contrast, provide the best combination of high specific energy, high specific power, long cycle life, and design flexibility. The following table compares the major battery chemistries used for EV battery pack applications.
| Parameter | Lead-acid | Ni-MH | Lithium-ion | Fuel cell |
|---|---|---|---|---|
| Specific energy (Wh kg⁻¹) | 30 – 50 | 60 – 120 | 150 – 250 | 300 – 600 (system) |
| Specific power (W kg⁻¹) | 150 – 400 | 200 – 1000 | 500 – 2000 | 100 – 300 |
| Cycle life | 300 – 500 | 500 – 1000 | 1000 – 3000 | 5000 + |
| Cost | Low | Medium | Medium to high | Very high |
| Environmental impact | Lead pollution | Mild | Mild | Hydrogen production |
| Maturity | Mature | Mature | Rapidly maturing | Pre-commercial |
From the perspective of EV battery pack design, lithium-ion batteries currently offer the most practical pathway to meet the demanding targets of driving range, acceleration, and fast charging. I therefore selected lithium-ion chemistry as the basis of my research.
1.3 Cathode Materials for Lithium-Ion Batteries
The cathode is the most important component in a lithium-ion cell because it determines the cell voltage, specific capacity, cost, and safety. The cathode active material typically accounts for 30% to 40% of the total cell cost. The most widely studied cathode materials are LiCoO₂, LiNiO₂, LiMn₂O₄, and LiFePO₄.
LiCoO₂ has a layered structure and offers excellent rate capability and cycling stability. It was the first successful commercial cathode material for lithium-ion batteries. However, cobalt is expensive and geographically concentrated. Moreover, LiCoO₂ suffers from structural instability at high voltage and severe safety concerns under overcharge conditions. Therefore, LiCoO₂ is not suitable for large-format electric vehicle batteries.
LiNiO₂ has a higher specific capacity than LiCoO₂ and lower cost. However, its synthesis is difficult because of the tendency to form non-stoichiometric compounds. The thermal stability of LiNiO₂ is also poor, and it readily decomposes at elevated temperatures, which poses serious safety risks in EV battery pack applications.
LiFePO₄ has attracted significant attention due to its excellent thermal stability, long cycle life, and low cost. The olivine structure of LiFePO₄ can accommodate lithium insertion and extraction with minimal volume change. However, its intrinsic electronic conductivity is very low, requiring elaborate carbon coating or nano-structuring strategies. The energy density of LiFePO₄ is also relatively low because of its low operating voltage.
Spinel LiMn₂O₄ is one of the most promising cathode materials for power batteries. It offers a three-dimensional lithium-ion diffusion channel, high operating voltage of about 4 V, low cost, abundant manganese resources, and good safety characteristics. The theoretical capacity of LiMn₂O₄ is 148 mAh g⁻¹, and its practical capacity is around 110 – 120 mAh g⁻¹. The following table summarizes the characteristics of these four cathode families.
| Material | Structure | Theoretical capacity (mAh g⁻¹) | Practical capacity (mAh g⁻¹) | Advantages | Limitations |
|---|---|---|---|---|---|
| LiCoO₂ | Layered | 274 | 140 – 150 | High voltage, mature technology | High cost, poor safety |
| LiNiO₂ | Layered | 274 | 180 – 200 | High capacity, lower cost | Difficult synthesis, thermal instability |
| LiMn₂O₄ | Spinel | 148 | 110 – 120 | Low cost, good safety, high power | Capacity fading at elevated temperature |
| LiFePO₄ | Olivine | 170 | 140 – 160 | Excellent safety, long life | Low conductivity, low energy density |
In my study, I selected spinel LiMn₂O₄ as the cathode material because it offers the best trade-off among cost, safety, power capability, and environmental friendliness for EV battery pack applications.
1.4 Anode Materials and Electrolytes
Graphite is the most common anode material for lithium-ion batteries. It has a low and flat lithium intercalation potential, high coulombic efficiency, and excellent cycle stability. During the first charge, the electrolyte decomposes at the graphite surface to form a solid electrolyte interphase (SEI) film. This film is electronically insulating but ionically conductive, and it prevents further electrolyte decomposition. However, the formation of the SEI film consumes a portion of the lithium inventory, which contributes to the initial irreversible capacity loss.
The electrolyte in a lithium-ion cell is typically a solution of lithium hexafluorophosphate (LiPF₆) in a mixture of organic carbonate solvents, such as ethylene carbonate, dimethyl carbonate, diethyl carbonate, and ethyl methyl carbonate. The electrolyte must have high ionic conductivity, wide electrochemical stability window, and good compatibility with both the cathode and the anode. Additives are often introduced to improve the SEI quality, suppress gas generation, and enhance overcharge protection.
1.5 Progress on Spinel LiMn₂O₄ Batteries
Spinel LiMn₂O₄ has been studied for decades as a candidate cathode material for electric vehicle batteries. Its three-dimensional lithium-ion diffusion pathways allow rapid lithium insertion and extraction, which is beneficial for high-rate discharge. The main challenge is the dissolution of manganese in the electrolyte, particularly at elevated temperatures, which leads to capacity fading. Doping with cations such as Al, Cr, Co, and Ni, as well as surface coating with metal oxides, has been shown to improve the structural stability and suppress manganese dissolution.
For EV battery packs, LiMn₂O₄ offers a distinct advantage in terms of safety. The material is thermally stable up to about 400 °C, much higher than the decomposition temperature of LiCoO₂. This characteristic reduces the risk of thermal runaway in a large-format EV battery pack. In addition, manganese is abundant in nature, which helps to reduce the material cost and ensure the sustainability of large-scale production.
1.6 Significance and Research Content of This Thesis
The objective of my research was to develop a complete manufacturing technology for high-power, large-format LiMn₂O₄/graphite lithium-ion batteries. The specific goals were as follows:
- To optimize the positive electrode formulation and determine the optimal active material content.
- To determine the appropriate capacity balance coefficient between the positive and negative electrodes.
- To identify the optimal compaction density for both electrodes and the appropriate assembly tightness.
- To establish the optimal electrolyte filling amount for large-format cells.
- To develop a suitable formation protocol that maximizes the utilization of active materials and ensures the stable formation of the SEI film.
- To design, fabricate, and evaluate a 60 Ah prismatic lithium-ion power cell for EV battery pack applications.
2. Manufacturing Process and Testing Methods
2.1 Raw Materials and Instruments
The raw materials used in my experiments included spinel LiMn₂O₄ from Hunan Shanshan, polyvinylidene fluoride (PVDF) from Shanghai San-Aifu, N-methyl-2-pyrrolidone (NMP), conductive carbon black, graphite from Shenzhen BTR, water-based binder, aluminum foil, copper foil, separator, steel case, aluminum tabs, nickel tabs, and electrolyte from Guangzhou Tinci. The main instruments are listed in the following table.
| Instrument | Model | Manufacturer |
|---|---|---|
| Vacuum mixer | JL-200 | Liuzhou Haojiete |
| Coating machine | DY-400 | Shaoyang Dali |
| Roller press | SG-400 | Shaoyang Dali |
| Slitting machine | QF-400 | Shaoyang Dali |
| Winding machine | JW-450 | Shaoyang Dali |
| Laser welding machine | HL-300 | Wuhan Chuyuan |
| Battery tester | CT-3008 | Wuhan Jinnuo |
| Internal resistance tester | BK-300 | Shenzhen Chaosisi |
| SEM | XL-30 | Philips |
| XRD | D/Max-3C | Rigaku |
2.2 General Manufacturing Flow
The manufacturing process for prismatic lithium-ion power cells consists of electrode preparation, cell assembly, electrolyte filling, formation, and grading. In my work, I paid special attention to the environmental control of the dry room and the vacuum baking process because moisture can severely degrade the electrochemical performance and safety of lithium-ion cells. The entire flow can be described as follows:
- Mixing of the positive slurry (LiMn₂O₄, conductive carbon black, PVDF, and NMP) and the negative slurry (graphite, conductive carbon black, water-based binder, and water).
- Coating of the slurries onto aluminum foil and copper foil, respectively.
- Drying, calendering, slitting, and cutting of the coated electrodes.
- Tab welding, winding, and insertion of the electrode assembly into the steel can.
- Laser welding of the cover, vacuum drying, and electrolyte filling.
- Formation, degassing, sealing, and final grading.
This route is compatible with large-scale manufacturing and ensures that the cells can be assembled into a reliable EV battery pack with high consistency.
2.3 Electrode Preparation
In the mixing step, I first dried the LiMn₂O₄ powder and conductive carbon black separately to remove adsorbed moisture. Then, PVDF was dissolved in NMP in a vacuum mixer for about 60 minutes to obtain a transparent viscous solution. The pre-mixed LiMn₂O₄ and carbon black powder was then added in batches under continuous stirring for about 3 hours. The resulting slurry was filtered through a 100-mesh screen before coating. For the negative electrode, the water-based binder and water were first mixed, followed by the addition of pre-mixed graphite and carbon black. The negative slurry was subjected to vacuum stirring for about 2 hours and then degassed to eliminate bubbles.
The coating process was carried out on a continuous coating machine. The thickness of the wet coating was controlled by adjusting the gap between the doctor blade and the coating roller. The coated electrodes were dried in a multi-zone hot-air drying oven. I monitored the coating thickness and surface density periodically to ensure uniformity. The calendering step was performed with a roller press to obtain the desired compaction density. I was careful to avoid over-compression, which could destroy the porous structure of the electrode and impede electrolyte penetration.
2.4 Cell Assembly
After calendering, the electrodes were cut into strips of the required width. Aluminum tabs were ultrasonically welded to the positive electrode, and nickel tabs were welded to the negative electrode. The electrode strips and separator were wound together on a semi-automatic winding machine to form a jelly-roll. The jelly-roll was then inserted into a laser-welded prismatic steel can. After welding the cover, the cells were vacuum-dried at high temperature to remove residual moisture. The electrolyte was filled in a glove box under an argon atmosphere. The amount of electrolyte was precisely controlled by weighing the cells before and after filling.
2.5 Formation and Grading
After electrolyte filling, the cells were allowed to rest for a certain period to ensure complete wetting of the electrodes and separator. The formation process was then carried out to activate the active materials and form a stable SEI film on the graphite anode. During formation, gas is generated, so I used an open-formation configuration followed by a final sealing step. Finally, the cells were charged and discharged to determine their capacity and internal resistance, and then sorted into different capacity grades.
2.6 Electrochemical and Safety Testing
All the electrochemical tests were performed at room temperature (25 ± 5 °C). The charge protocol was constant-current constant-voltage: the cell was charged at a specified constant current to 4.2 V, then held at 4.2 V until the current declined to a cut-off value. The discharge tests were carried out at various constant currents down to a cut-off voltage of 2.75 V. Internal resistance measurements were conducted with an AC internal resistance tester on cells in the fully charged state after a rest period. Safety tests included overcharge, external short circuit, heating, and nail penetration, and were performed in accordance with the standard QC/T 743-2006 for lithium-ion traction batteries for electric vehicles.
3. Studies on Key Manufacturing Parameters
3.1 Selection of Electrode Materials
I selected spinel LiMn₂O₄ as the positive active material. The material had a particle size distribution between 5 and 20 μm, which provided a good balance between high surface area for electrochemical reaction and manageable slurry handling. The purity of the LiMn₂O₄ powder, expressed as the manganese weight percentage, was greater than 99%. High purity is essential to minimize self-discharge and hydrogen evolution inside the cell.
For the negative electrode, I selected natural graphite from Shenzhen Beterui. The graphite had a reversible specific capacity of approximately 360 mAh g⁻¹ and a lamellar morphology that is favorable for lithium intercalation.
3.2 Effect of Active Material Content on Cell Performance
The electrode formulation strongly influences the performance of the final cell. I prepared cells with active material contents of 88%, 90%, 92%, and 94% in the positive electrode, while keeping the conductive agent and binder in the same relative proportion. The cells were of the 0630480 prismatic type with a nominal capacity of about 11 Ah.
3.2.1 Effect on Discharge Capacity
The first-cycle discharge curves at the 0.3C rate are compared in terms of the relative capacity, as summarized in the table below.
| Active material content (%) | Relative discharge capacity (%) |
|---|---|
| 88 | 93.5 |
| 90 | 97.0 |
| 92 | 100.0 |
| 94 | 100.8 |
It can be observed that increasing the active material content from 88% to 92% significantly improved the discharge capacity. However, the increment from 92% to 94% was relatively small. This behavior can be explained by the fact that a higher active material content reduces the proportion of the conductive agent and binder, which are necessary for maintaining electrical contact and mechanical integrity within the electrode.
3.2.2 Effect on Internal Resistance
The internal resistance of the cells was measured after full charging. The results are shown below.
| Active material content (%) | Internal resistance (mΩ) |
|---|---|
| 88 | 6.8 |
| 90 | 7.1 |
| 92 | 7.5 |
| 94 | 7.9 |
As expected, the internal resistance increased with decreasing conductive agent content. Although a lower internal resistance is generally desirable, the concurrent reduction in energy density and increase in material waste made the 88% and 90% formulations less attractive for practical production.
3.2.3 Effect on Cycle Performance
I also evaluated the capacity retention of the cells after 100 cycles at the 1C rate. The relative capacity retention values are shown below.
| Active material content (%) | Capacity retention after 100 cycles (%) |
|---|---|
| 88 | 98.2 |
| 90 | 96.5 |
| 92 | 95.3 |
| 94 | 91.0 |
The cycle stability deteriorated significantly at the highest active material content. This is because the reduced binder content weakens the adhesion between the active particles and the current collector, and the reduced conductive agent content increases the contact resistance between particles. Considering the discharge capacity, internal resistance, and cycle life together, I selected an active material content of 92% as the optimal value for the positive electrode.
3.3 Capacity Balance Coefficient
In a lithium-ion cell, the capacity of the negative electrode must be greater than that of the positive electrode to prevent lithium metal deposition on the graphite surface during charging. The capacity balance coefficient is defined as follows:
$$
K = \frac{C_N}{C_P} = \frac{S_N \, w_N \, q_N}{S_P \, w_P \, q_P}
$$
where \(C_N\) and \(C_P\) are the areal capacities of the negative and positive electrodes, \(S\) is the areal density, \(w\) is the mass fraction of the active material, and \(q\) is the reversible specific capacity of the active material.
3.3.1 Influence of Manufacturing Capability
In actual production, the areal density of the coated electrodes is subject to variation. In my coating process, the areal density deviation was controlled within ±3%. Therefore, the minimum design value of the capacity balance coefficient had to be greater than 1.06 to ensure that the worst-case combination of a heavy positive electrode and a light negative electrode still satisfied the condition \(K > 1\).
3.3.2 Effect on Discharge Capacity
I assembled cells with capacity balance coefficients of 1.04, 1.06, 1.08, and 1.10, and compared their discharge capacities. The results are presented below.
| Capacity balance coefficient | Relative discharge capacity (%) |
|---|---|
| 1.04 | 96.2 |
| 1.06 | 100.0 |
| 1.08 | 100.3 |
| 1.10 | 99.1 |
When the balance coefficient was 1.04, the capacity was the lowest because a portion of the SEI formation consumed part of the lithium inventory, and the limited graphite capacity restricted the degree of lithium intercalation. When the coefficient was 1.10, the cell capacity also decreased because the excessive negative electrode thickness increased the internal resistance and made the electrolyte wetting more difficult.
3.3.3 Effect on Safety Performance
I also performed safety tests on cells with different balance coefficients, including overcharge, short circuit, and nail penetration. The maximum surface temperature during the tests is recorded in the table below.
| Safety test | K = 1.04 | K = 1.06 | K = 1.08 | K = 1.10 |
|---|---|---|---|---|
| Overcharge | Vent activated | Stable | Stable | Stable |
| External short circuit (max. temperature) | 85 °C | 70 °C | 68 °C | 67 °C |
| Nail penetration (max. temperature) | 120 °C | 95 °C | 92 °C | 90 °C |
The safety risk increased significantly when the balance coefficient was only 1.04. Under overcharge conditions, lithium dendrites could form on the graphite surface, leading to internal short circuits and excessive heat generation. Based on these findings, I selected the capacity balance coefficient in the range of 1.06 – 1.10 for the production of EV battery pack cells.
3.4 Compaction Density of the Electrodes
The compaction density of an electrode determines its porosity, which in turn affects the penetration of the electrolyte, the electronic conductivity, and the mechanical strength of the electrode. I investigated the effect of positive electrode compaction density in the range of 2.8 to 3.6 g cm⁻³ with a fixed negative electrode compaction density of 1.5 g cm⁻³.
3.4.1 Assembly Situations at Different Compaction Densities
Cell assembly was attempted for each compaction density. The results are summarized below.
| Compaction density (g cm⁻³) | 2.8 | 3.0 | 3.2 | 3.4 | 3.6 |
|---|---|---|---|---|---|
| Number of cells assembled | 20 | 20 | 20 | 20 | 20 |
| Number of short-circuit cells | 6 | 3 | 0 | 0 | 2 |
At a low compaction density of 2.8 g cm⁻³, the electrodes were too thick to be inserted smoothly into the steel can, and the edges of the jelly-roll were often scratched, causing internal short circuits. When the compaction density was increased to 3.2 – 3.4 g cm⁻³, the assembly yield reached 100%. However, at 3.6 g cm⁻³, the electrodes became brittle, and cracking of the coating led to a slight increase in the defect rate.
3.4.2 Effect on Discharge Capacity
The normalized discharge capacity of the cells as a function of the positive electrode compaction density is shown below.
| Compaction density (g cm⁻³) | Relative discharge capacity (%) |
|---|---|
| 2.8 | 93.0 |
| 3.0 | 98.5 |
| 3.2 | 100.0 |
| 3.4 | 98.8 |
| 3.6 | 95.5 |
When the compaction density was too low, the large pores in the electrode reduced the effective contact area between the active particles, leading to poor utilization of the active mass. In contrast, when the compaction density was too high, the electrode became too dense for the electrolyte to penetrate, which increased the diffusion resistance of lithium ions. Therefore, the optimal positive electrode compaction density was 3.2 g cm⁻³.
3.4.3 Effect on Rate Capability
I further examined the rate capability of cells with different positive compaction densities. The capacity retention at various discharge rates is summarized below.
| Compaction density (g cm⁻³) | 0.3C (%) | 1C (%) | 3C (%) | 5C (%) |
|---|---|---|---|---|
| 2.8 | 100 | 93.5 | 82.0 | 71.5 |
| 3.0 | 100 | 96.2 | 88.5 | 80.2 |
| 3.2 | 100 | 97.0 | 90.5 | 83.8 |
| 3.4 | 100 | 95.8 | 86.0 | 76.5 |
| 3.6 | 100 | 94.0 | 81.5 | 70.0 |
The results confirm that an appropriate compaction density is essential for maintaining good rate capability. At 3.2 g cm⁻³, the electrode porosity provides a favorable balance between electronic conduction and ionic transport. I similarly optimized the negative electrode compaction density and selected 1.5 g cm⁻³. Based on these values, the assembly tightness of the cell was determined as follows:
$$
\text{Tightness} = \frac{t_{\text{electrode stack}} + t_{\text{separator}}}{D_{\text{inner}}} \times 100\%
$$
I found that the appropriate assembly tightness was in the range of 93% – 97%.
3.5 Electrolyte Filling Amount
The electrolyte amount has a profound influence on the cell performance. I used the 0630480 cell format with a nominal capacity of about 11 Ah as the experimental vehicle. The electrolyte filling amount was varied from 5.0 to 7.0 g Ah⁻¹.
3.5.1 Effect on Capacity and Internal Resistance
The relationships between the electrolyte amount, discharge capacity, and internal resistance are shown in the following table.
| Electrolyte amount (g Ah⁻¹) | Relative discharge capacity (%) | Internal resistance (mΩ) |
|---|---|---|
| 5.0 | 93.2 | 8.2 |
| 5.4 | 97.5 | 7.8 |
| 5.8 | 99.3 | 7.6 |
| 6.2 | 100.0 | 7.5 |
| 6.6 | 100.2 | 7.5 |
| 7.0 | 100.3 | 7.5 |
When the electrolyte amount was below 5.8 g Ah⁻¹, the capacity decreased significantly and the internal resistance increased. This is because insufficient electrolyte fails to wet all the pores of the electrode, leaving part of the active material electrochemically inactive. When the electrolyte amount reached 6.2 g Ah⁻¹, the capacity and internal resistance stabilized. Further increasing the electrolyte amount produced no additional benefit.
3.5.2 Further Confirmation of the Optimal Electrolyte Amount
To confirm the optimal value, I prepared cells with electrolyte amounts of 5.6, 6.0, 6.2, and 6.6 g Ah⁻¹ and examined the capacity distribution. The results are listed below.
| Electrolyte amount (g Ah⁻¹) | Cells with capacity > 11 Ah (%) | Cells with capacity 10 – 11 Ah (%) | Cells with capacity < 10 Ah (%) |
|---|---|---|---|
| 5.6 | 55 | 35 | 10 |
| 6.0 | 82 | 15 | 3 |
| 6.2 | 96 | 4 | 0 |
| 6.6 | 98 | 2 | 0 |
Although an electrolyte amount of 6.6 g Ah⁻¹ resulted in a slightly higher yield of cells with capacity above 11 Ah, I observed that some cells in this group experienced electrolyte leakage during open-formation. This can corrode the equipment and create safety issues in large-scale production. Therefore, I selected 6.2 g Ah⁻¹ as the optimal electrolyte filling amount for the 0630480 cell format.
3.6 Formation Protocol
The formation protocol is critical for the initial activation of the cell and the quality of the SEI film. I compared three formation protocols, as summarized below.
| Protocol | Formation procedure |
|---|---|
| Protocol A | Constant-current charging at 0.1C to 4.2 V |
| Protocol B | Constant-current charging at 0.2C to 4.2 V |
| Protocol C | Charging at 0.1C for 30 min, then charging at 0.2C to 4.2 V |
3.6.1 Effect on Discharge Capacity
I fabricated 30 cells with each protocol and measured the discharge capacity after formation. The capacity distributions are shown below.
| Protocol | Cells > 11 Ah (%) | Cells 10 – 11 Ah (%) | Cells < 10 Ah (%) |
|---|---|---|---|
| A | 60 | 35 | 5 |
| B | 50 | 40 | 10 |
| C | 65 | 30 | 5 |
Protocol C produced the largest proportion of high-capacity cells. The initial small-current charging step allows the SEI film to form gradually and uniformly on the graphite anode, which reduces the consumption of active lithium and improves the utilization of the cathode material.
3.6.2 Effect on Internal Resistance and Charge Efficiency
The internal resistance and charge-discharge efficiency of the cells are presented below.
| Protocol | Internal resistance < 7.5 mΩ (%) | First-cycle efficiency (%) | Second-cycle efficiency (%) |
|---|---|---|---|
| A | 75 | 88.5 | 96.8 |
| B | 65 | 86.0 | 96.2 |
| C | 82 | 89.5 | 97.5 |
The first-cycle efficiency is directly related to the irreversible capacity loss caused by SEI formation. Protocol C resulted in the highest first-cycle efficiency, indicating that the SEI film formed was stable and consumed the least amount of active lithium.
3.6.3 Effect on Cycle Life
The capacity retention of the cells after 200 cycles at the 1C rate is shown below.
| Protocol | Capacity retention after 200 cycles (%) |
|---|---|
| A | 94.5 |
| B | 93.8 |
| C | 95.2 |
Protocol C again yielded the best cycle stability. Based on these results, I selected Protocol C as the standard formation protocol: charging at 0.1C for 30 minutes, followed by charging at 0.2C to 4.2 V. The cell is then allowed to rest for degassing and sealing, after which it is subjected to capacity grading.
3.7 Performance of the Optimized 0630480 Cells
After applying all the optimized parameters, I fabricated 0630480 cells and evaluated their overall performance. The results are summarized in the following table.
| Parameter | Value |
|---|---|
| Discharge capacity at 0.3C | 11.5 Ah |
| Discharge capacity at 1C | 11.2 Ah |
| Discharge capacity at 3C | 10.6 Ah |
| Internal resistance | 7.5 mΩ |
| Self-discharge rate (per day) | 0.05% |
In addition, the cells passed all safety tests in the category of abusive use, including overcharge, external short circuit, nail penetration, and crushing. No explosion or fire was observed in any of the tests.
3.8 Summary of the Key Parameter Studies
Through the above systematic studies, I established the following optimal manufacturing parameters for LiMn₂O₄/graphite power cells:
- Positive active material content: 92%
- Capacity balance coefficient: 1.06 – 1.10
- Positive electrode compaction density: 3.2 g cm⁻³
- Negative electrode compaction density: 1.5 g cm⁻³
- Assembly tightness: 93% – 97%
- Electrolyte filling amount: 6.2 g Ah⁻¹
- Formation protocol: 0.1C charging for 30 min, then 0.2C charging to 4.2 V
4. Design and Fabrication of Large-Capacity LiMn₂O₄ Power Cells
4.1 Structure and Morphology of the Cathode Material
The XRD pattern of the LiMn₂O₄ powder used in my large-capacity cell showed the characteristic diffraction peaks of the cubic spinel phase. The strong peaks at 2θ values of approximately 18.6°, 36.1°, 44.0°, 58.1°, and 63.9° correspond to the (111), (311), (400), (511), and (440) planes, respectively. The sharpness of the peaks indicated good crystallinity. The SEM image revealed that the primary particles had an irregular polyhedral shape with a particle size of 5 – 15 μm. Some fine particles adhered to the surface of larger particles, which is beneficial for increasing the contact area between the active material and the conductive agent.
The photograph below shows a representative power battery that I evaluated during the course of this research.
4.2 Design Targets for the Large-Capacity Cell
I designed a prismatic cell with a nominal capacity of 60 Ah for electric vehicle applications. The design targets are listed below.
| Parameter | Design value |
|---|---|
| Cell dimensions (mm) | 51 × 133 × 242 |
| Positive active material | LiMn₂O₄ |
| Negative active material | Graphite |
| Nominal capacity | 60 Ah |
| Nominal voltage | 3.8 V |
| Internal resistance | ≤ 3.0 mΩ |
| Specific power | ≥ 800 W kg⁻¹ |
| Specific energy | ≥ 85 Wh kg⁻¹ |
| Safety | No explosion or fire |
4.3 Cell Design Calculations
The design capacity is usually set about 10% higher than the nominal capacity to compensate for possible losses during cycling:
$$
C_{\text{design}} = 1.1 \times C_{\text{rated}} = 1.1 \times 60 = 66 \text{ Ah}
$$
The mass of the positive active material was calculated from the design capacity and the specific capacity of LiMn₂O₄. Assuming a specific capacity of 105 mAh g⁻¹ for the full electrode formulation, the required mass of the positive active material is given by:
$$
m_{\text{positive}} = \frac{C_{\text{design}}}{q_{\text{positive}}} = \frac{66}{0.105} \approx 629 \text{ g}
$$
Based on the capacity balance coefficient \(K = 1.08\), the required mass of the negative active material was:
$$
m_{\text{negative}} = \frac{C_{\text{design}}}{q_{\text{negative}}} \times K = \frac{66}{0.360} \times 1.08 \approx 198 \text{ g}
$$
With the selected areal densities of the positive and negative electrodes, I calculated the area of each electrode and then determined the electrode length and the number of jelly-rolls required. I chose a multi-jelly-roll design to reduce the current path and improve heat dissipation in the large-format cell. This is particularly important for EV battery pack applications, where high current and rapid heat generation are common.
During the charging process, the graphite anode expands when lithium ions are intercalated. The lattice spacing of graphite increases from about 0.335 nm to 0.372 nm for the fully lithiated state:
$$
\epsilon_{\text{anode}} = \frac{0.372 – 0.335}{0.335} \times 100\% \approx 11\%
$$
Therefore, the assembly tightness was controlled carefully to leave sufficient space for the electrode expansion without creating excessive internal stress.
4.4 Electrode Preparation and Cell Assembly
For the positive electrode, I mixed LiMn₂O₄, conductive carbon black, and PVDF in a mass ratio of 92 : 4 : 4 in NMP. The slurry was coated on aluminum foil with a thickness of 20 μm, dried, and calendered to a compaction density of 3.2 g cm⁻³. For the negative electrode, graphite, carbon black, and water-based binder were mixed in a mass ratio of 95 : 1.5 : 3.5, coated on copper foil, and calendered to a compaction density of 1.5 g cm⁻³. The electrodes were cut, welded with tabs, and wound together with a porous polyolefin separator. The jelly-roll assembly was inserted into a prismatic steel can, laser-welded, vacuum-dried, and filled with electrolyte. After formation and degassing, the cells were sealed and graded.
4.5 Electrochemical Performance of the Large-Capacity Cell
4.5.1 Initial Charge-Discharge Characteristics
The first-cycle charge-discharge curve of the 60 Ah cell at a current of 18 A is shown in the table below. The cell was charged at 18 A to 4.2 V, then held at constant voltage until the current decreased to 0.9 A. The discharge was then conducted at 18 A to 2.75 V.
| Parameter | Value |
|---|---|
| Charge capacity | 61.8 Ah |
| Constant-current charge capacity | 59.3 Ah |
| Discharge capacity | 61.5 Ah |
| First-cycle coulombic efficiency | 99.5% |
| Mid-point discharge voltage | 3.80 V |
The high first-cycle efficiency indicated that the optimized formation protocol successfully minimized the irreversible capacity loss and that the electrode materials were well matched.
4.5.2 Internal Resistance
The internal resistance of the 60 Ah cell was measured to be 1.8 mΩ. This value is well below the design target of 3.0 mΩ. A low internal resistance is essential for maintaining a stable output voltage under high current and for reducing heat generation inside the EV battery pack.
4.5.3 Rate Capability
I evaluated the rate capability of the cell at discharge rates of 0.3C, 1C, 3C, and 5C. The results are presented in the table below.
| Discharge rate | Discharge capacity (Ah) | Capacity retention relative to 0.3C (%) | Mid-point voltage (V) |
|---|---|---|---|
| 0.3C | 61.5 | 100.0 | 3.80 |
| 1C | 60.5 | 98.4 | 3.72 |
| 3C | 58.6 | 95.3 | 3.58 |
| 5C | 55.5 | 90.2 | 3.46 |
Even at the 5C rate, the cell retained more than 90% of its 0.3C capacity. This demonstrates the excellent high-rate capability of the spinel LiMn₂O₄ cathode and the optimized cell design. The cell delivered a specific power of approximately 875 W kg⁻¹ and a specific energy of approximately 87 Wh kg⁻¹ at the 5C discharge rate, which meets the requirements for EV battery pack applications.
4.5.4 Charge Retention and Capacity Recovery
I tested the charge retention of the cell by fully charging it, storing it at room temperature for 28 days, and then discharging it at the 0.3C rate. The stored discharge capacity was 57.2 Ah, corresponding to 95.3% of the nominal capacity. The self-discharge rate was calculated as:
$$
\eta_{\text{self}} = \frac{61.5 – 57.2}{61.5 \times 28} \times 100\% \approx 0.25\% \text{ per day}
$$
After this test, the cell was recharged and discharged again at the 0.3C rate. The recovered capacity was 59.8 Ah, corresponding to a capacity recovery rate of 96.8%, which is well above the standard requirement of 95% for traction batteries.
4.5.5 Cycle Life
I tested the cycle life of the cell at a 1C discharge rate and a 0.5C charge rate. The capacity retention as a function of cycle number is summarized below.
| Cycle number | Discharge capacity (Ah) | Capacity retention (%) |
|---|---|---|
| 1 | 60.5 | 100.0 |
| 200 | 58.4 | 96.5 |
| 400 | 56.8 | 93.9 |
| 600 | 55.5 | 91.7 |
| 700 | 55.0 | 90.9 |
The cell retained more than 90% of its initial capacity after 700 cycles, demonstrating good cycle stability for an EV battery pack application.
4.5.6 Safety Evaluation
I conducted overcharge, external short circuit, nail penetration, and heating tests on the 60 Ah cell. The results are listed below.
| Safety test | Test condition | Requirement | Result |
|---|---|---|---|
| Overcharge | Charge at 1C to 10 V | No explosion, no fire | Pass |
| External short circuit | External resistance < 5 mΩ | No explosion, no fire | Pass |
| Nail penetration | Steel nail ≥ 3 mm | No explosion, no fire | Pass |
| Heating | 130 °C for 30 min | No explosion, no fire | Pass |
All safety tests were passed without any explosion or fire. The excellent safety characteristics are attributed to the intrinsic thermal stability of the spinel LiMn₂O₄ cathode, the optimized capacity balance coefficient, and the robust cell design. These features are essential for the reliable operation of a large-scale EV battery pack.
4.6 Summary of the Large-Capacity Cell Performance
The 60 Ah prismatic LiMn₂O₄/graphite cell exhibited outstanding electrochemical performance. The cell delivered an initial discharge capacity of 61.5 Ah at the 0.3C rate, and retained over 90% of this value at the 5C rate. The internal resistance was only 1.8 mΩ, and the cycle life exceeded 700 cycles with more than 90% capacity retention. The charge retention after 28 days of storage was 95.3%, and the capacity recovery reached 96.8%. All safety tests satisfied the Chinese national standard QC/T 743-2006. Therefore, the cell is well suited for integration into an EV battery pack for electric vehicles.
5. Conclusions and Perspectives
In this thesis, I systematically studied the manufacturing process of LiMn₂O₄-based lithium-ion power batteries and established a complete set of optimized process parameters. The main conclusions of my research are as follows:
-
Spinel LiMn₂O₄ with a particle size of 5 – 20 μm and a purity greater than 99% is a suitable cathode active material for lithium-ion power batteries.
-
The optimal active material content in the positive electrode is 92%. This composition provides the best compromise among discharge capacity, internal resistance, and cycle stability.
-
The optimal capacity balance coefficient is in the range of 1.06 – 1.10. This range ensures that the negative electrode has sufficient capacity to accommodate lithium ions from the positive electrode during charging, while minimizing the waste of anode material.
-
The optimal compaction density is 3.2 g cm⁻³ for the LiMn₂O₄ positive electrode and 1.5 g cm⁻³ for the graphite negative electrode. The corresponding assembly tightness should be maintained at 93% – 97%.
-
The optimal electrolyte filling amount is 6.2 g Ah⁻¹ for the 0630480 cell format. This amount saturates the electrode pores and provides stable ionic conduction without causing electrolyte leakage.
-
The recommended formation protocol is charging at 0.1C for 30 minutes followed by charging at 0.2C to 4.2 V. This protocol effectively activates the active materials, forms a stable SEI film, and reduces the gassing-related degradation of cell performance.
-
The 60 Ah prismatic cell fabricated with these optimized parameters demonstrates excellent rate capability, high specific power, long cycle life, and satisfactory safety performance. The cell is suitable for use in EV battery pack systems for electric vehicles.
In the future, I believe the following research directions will be important. First, further improvement of the high-temperature cycle life of LiMn₂O₄ cathodes is needed. This can be achieved through cation doping, surface coating, and the development of electrolyte additives that suppress manganese dissolution. Second, the cost reduction of cell manufacturing is critical for market expansion. This requires continuous optimization of the electrode design and the assembly process to increase throughput and reduce material waste. Third, the integration of cells into large-scale EV battery pack systems requires thermal management and cell balancing technologies to ensure consistent performance over the lifetime of the pack. Fourth, the design of intelligent battery management systems is essential to avoid overcharge, over-discharge, and thermal runaway, thereby ensuring the long-term safety and reliability of electric vehicles. With these advances, I am confident that LiMn₂O₄-based lithium-ion power batteries will play an increasingly important role in the electrification of transportation.