Development of Key Manufacturing Parameters and Electrochemical Performance for Spinel LiMn₂O₄ Traction Battery Pack

1. Introduction and Literature Review

The depletion of traditional fossil energy sources, such as petroleum and coal, in conjunction with significant fluctuations in global oil prices, has rendered the “energy crisis” a universally recognized concern. Economists and scientists broadly estimate that petroleum resources will be nearly exhausted by the middle of this century, around the year 2050. Concurrently, the severe environmental pollution caused by conventional internal combustion engine vehicles has necessitated the exploration of clean, renewable alternatives. Electric vehicles (EVs) represent a quintessential application of new energy technology, lauded as a clean and environmentally benign mode of transport. The development of EVs is of great economic and strategic significance, mitigating the constraints imposed by the depletion of non-renewable energy and addressing urban air pollution. However, the performance of the power battery, the heart of an EV, serves as the primary bottleneck limiting the widespread commercialization of electric transportation. Key performance indicators for EV batteries include specific energy, specific power, cycle life, and safety.

Traditional lead-acid and nickel-metal hydride (Ni-MH) batteries, while mature in technology, suffer from lower specific energy and power, which curtails the driving range of EVs. As a result, they are increasingly inadequate for the demands of next-generation electric vehicles. In stark contrast, lithium-ion batteries have emerged as the most promising power source for EVs, attributed to their superior characteristics such as high operating voltage, high energy density, low self-discharge rate, long cycle life, and environmental friendliness. The specific energy of lithium-ion batteries can exceed 200 Wh·kg⁻¹, with a specific power of over 1000 W·kg⁻¹ and a cycle life ranging from 500 to over 2000 cycles. The energy density of lithium-ion batteries is approximately three to four times that of lead-acid batteries and two times that of nickel-hydrogen batteries. To meet the rigorous demands of the automotive industry, significant research has focused on developing large-capacity lithium-ion cells and safe, cost-effective materials, particularly for constructing a reliable **traction battery pack**. Consequently, a global effort is underway to develop high-performance active materials suitable for these demanding applications.

The advancement of electric vehicles directly dictates the future of the **traction battery pack** market. Projections suggest that the market for pure electric vehicles will surpass one million units, and hybrid electric vehicles will exceed four million units by 2030, with the global battery market expanding to an estimated 60 billion USD. In China, driven by the “Energy Saving and New Energy Vehicles” development plan, the demand for power batteries is predicted to reach 40 million units. This immense market potential underscores the urgency of perfecting lithium-ion battery technology for traction applications.

1.1 Overview of Lithium-ion Batteries for Traction Battery Pack

A lithium-ion battery operates on the principle of lithium-ion shuttling between the anode and cathode during charge and discharge cycles. The primary advantages of the lithium-ion chemistry over other conventional rechargeable systems can be summarized as follows:
1. High Voltage: A single cell can deliver a voltage of 3.6-4.2 V, which is two to three times higher than that of Ni-Cd or Ni-MH batteries.
2. High Specific Energy: The energy density can reach 200 Wh·kg⁻¹.
3. Long Cycle Life: A cycle life of more than 1000 cycles can be achieved.
4. Wide Operating Temperature Range: Good performance can be maintained across a wide temperature window.
5. Low Self-Discharge: The self-discharge rate is extremely low, improving shelf life.
6. No Memory Effect: The battery does not suffer from the “memory effect” common to Ni-Cd batteries, simplifying maintenance.

The performance and cost of a lithium-ion battery are inextricably linked to its constituent materials, particularly the electrode active materials. The cathode material is a major cost driver and is critical for determining the cell’s potential, capacity, and safety. The most widely used cathode materials are compared:

LiCoO₂: Possesses excellent electrochemical performance and is widely used in portable electronics. However, its high cost, limited cobalt resources, and poor thermal stability, which can lead to safety hazards under overcharge or high-temperature conditions, make it unsuitable for large-scale applications like a **traction battery pack**.

LiNiO₂: Offers a higher specific capacity and is less expensive than LiCoO₂. Nevertheless, its synthesis is difficult due to the instability of Ni³⁺, and its poor thermal stability and cycling performance have hindered its commercialization.

LiMn₂O₄: As a spinel-structured material, LiMn₂O₄ has become the focus of intense investigation for use in **traction battery pack** systems. It offers a three-dimensional tunnel structure that facilitates fast lithium-ion diffusion. Its advantages are multifold: abundant and inexpensive manganese resources, good electronic and ionic conductivity, high thermal stability (decomposition temperature above 400⁰C), excellent rate capability, and environmental friendliness. The theoretical capacity is 148 mAh·g⁻¹, and the actual capacity can reach about 120 mAh·g⁻¹. The material’s inherent safety, stemming from its high structural stability, even in the fully delithiated state (λ-MnO₂), makes it an optimal choice for EV applications where safety is paramount.

LiFePO₄: The olivine-structured LiFePO₄ is renowned for its outstanding safety and long cycle life. However, its intrinsically low electronic conductivity and low lithium-ion diffusion coefficient require complex carbon coating or nano-sizing strategies, which can complicate manufacturing and increase cost.

Given the balance of cost, safety, electrochemical performance, and resource availability, this research selects spinel LiMn₂O₄ as the cathode material for the development of a high-performance **traction battery pack**. The primary challenge associated with LiMn₂O₄ is the gradual capacity fade upon cycling, particularly at elevated temperatures. This fade is primarily attributed to manganese dissolution (disproportionation reaction 2Mn³⁺ → Mn⁴⁺ + Mn²⁺), Jahn-Teller distortion, and electrolyte decomposition. Careful engineering of the material composition and manufacturing process can mitigate these issues. Therefore, this study focuses on the influence of critical manufacturing parameters—from mixing and coating to formation—on the final performance of LiMn₂O₄-based lithium-ion cells designed for use in a **traction battery pack**.

2. Experimental Materials, Methods, and Manufacturing Process

2.1 Raw Materials and Reagents

The development of the high-power lithium-ion cells for the **traction battery pack** necessitated the use of high-purity raw materials. The table below lists the primary materials used in this study:

Table 2.1: Experimental Materials and Reagents

2.2 Equipment for Cell Assembly and Testing

The manufacturing of precision cells requires specialized equipment, particularly for the processing of electrodes and the assembly in a controlled environment.

Table 2.2: Key Experimental Instruments

2.3 Manufacturing Process of the Lithium-ion Cells

The overall process flow for fabricating the LiMn₂O₄ cells designated for the **traction battery pack** is detailed below. The complete flow, from slurry preparation to final cell sorting, was systematically optimized to ensure cell quality and consistency.

Step 1: Slurry Preparation (Mixing)

1. Pre-treatment: LiMn₂O₄ and carbon black powders were dried to remove absorbed moisture. The high hygroscopic nature of materials like NMP can introduce water into the system, which is detrimental to cell performance as it reacts with the LiPF₆ salt. The LiMn₂O₄, conductive carbon, and a portion of the binder were pre-mixed to ensure homogeneity.
2. Cathode Slurry: A specific amount of NMP solvent was added to a vacuum mixer. PVDF binder was then slowly added to the solvent. The mixture was stirred until the PVDF was completely dissolved to form a viscous, clear solution, which took approximately 2 hours. Afterward, the pre-mixed solid powders (LiMn₂O₄ + carbon black) were added in batches to the PVDF solution. The entire mixture was stirred under vacuum for an additional 4 hours to form a homogeneous cathode slurry.
3. Anode Slurry: A water-based system was employed for the anode. The process involved mixing the aqueous binder with a calculated amount of deionized water. Once homogeneous, the anode active material (graphite) and conductive carbon black were added. The slurry was stirred for approximately 4 hours. A defoaming agent, such as a small amount of alcohol, was added when necessary to prevent pinholes in the dry electrode coating.

Step 2: Coating and Drying

The prepared slurries were coated onto their respective current collectors using a continuous roll-coating machine. The cathode slurry was coated onto 20 μm thick aluminum foil, while the anode slurry was coated onto 15 μm thick copper foil. The coating process was meticulously controlled to ensure uniform thickness without surface defects like cracks or pinholes. The coated foils passed through a multi-zone drying oven with optimized airflow and temperature to facilitate uniform, rapid solvent evaporation without surface cracking. After drying, the electrode coils were removed. For different cell designs, the coating weight (active material loading per unit area) is a primary factor dictating cell capacity.

Step 3: Calendering (Roller Pressing)

The dried electrode sheets were cut into larger pieces and then subjected to calendering to achieve the target electrode density and thickness. The rollers’ gap was carefully adjusted to compress the electrode to the designed thickness, which is critical for ensuring proper electrode alignment, optimizing volumetric energy density, and maintaining good contact between active material particles and current collector. The compaction density parameter was correlated with the electrode performance, which will be discussed in detail in later sections.

Step 4: Slitting and Notching

The calendered electrodes are then cut into predetermined widths using a slitting machine. After slitting, the electrodes undergo a process of notching, where a portion of the active material is removed from one edge or the middle of the foil. This exposes the bare metal current collector to act as a location for tab welding.

Step 5: Tab Welding and Vacuum Drying

An aluminum tab was ultrasonically welded to the exposed area of the cathode foil, while a nickel tab was welded to the exposed area of the anode foil. The electrode strips were then rolled and placed into a vacuum oven to remove any residual moisture and solvent.

Step 6: Cell Winding and Assembly

The dry cathode, anode, and separator (polyethylene/polypropylene films) are assembled into a “jelly roll” or “electrode core” using a semi-automatic winding machine. The anode typically has a slightly larger width than the cathode to ensure that lithium ions are always intercalated within the anode material, preventing lithium plating. The finished jelly roll was then inserted into a pre-dried steel can, which had been cleaned and baked to remove surface-adsorbed contaminants.

Step 7: Welding and Electrolyte Filling

The nickel tab was first resistance-welded to the internal surface of the steel can. After placing an insulating ring, the aluminum tab was welded to the cover plate. The can assembly was then sealed using a laser welding machine. The cells were subjected to a final vacuum baking step at 90⁰C to reduce the water content to below the acceptable limit. Subsequently, the cells were moved into a dry argon-filled glove box where a precisely weighed amount of electrolyte was injected into each cell.

Step 8: Formation and Sorting

After injection, the cells were allowed to rest to ensure complete wetting of the electrode pores and separator. They were then subjected to a pre-charging step, known as formation. During formation, the initial capacity is evaluated, and more importantly, a stable solid electrolyte interface (SEI) layer is formed on the anode surface. The cells were then sealed and degassed, followed by aging and final electrical testing (capacity and internal resistance). Finally, the cells were graded based on their measured capacities and sorted accordingly.

3. Optimization of Critical Manufacturing Parameters

This section systematically investigates the influence of process parameters on the electrochemical performance (discharge capacity, internal resistance, cycle life, and rate capability) of the resulting cells. A consistent baseline cell type, designated as the 053048-type pouch cell, was used to validate the parameter optimization.

3.1 Active Material Content and Electrode Formulation

The ratio of active material, conductive agent, and binder, i.e., the slurry formulation, is critical. The conductive carbon black provides the necessary electronic network, while the binder ensures the adhesion of the coating to the foil. The percentage of active material directly affects the cell’s capacity but must be balanced against the need for adequate electronic conductivity and mechanical strength.

Figure 3.1 Influence of Active Material Content on Discharge Capacity (0.2C rate)

We compared four different formulations with active material contents of 80%, 85%, 90%, and 94%, in the positive electrode. The discharge curves (Figure 3.1) clearly indicate that the initial discharge capacity increased with active material content. In particular, increasing the content from 80% to 90% resulted in a noticeable capacity increase from ~850 mAh to ~950 mAh. Further increases to 94% yielded only a marginal increase in capacity (~965 mAh).

Table 3.1: Effect of Active Material Content on Battery Internal Resistance

| Cathode Active Material Content (%) | Internal Resistance (mΩ) | Average (mΩ) |
| :———————————– | :———————– | :———– |
| 80 | 34.5 | 24.3 |
| 85 | 24.5 | 18.4 |
| 90 | 16.8 | 12.9 |
| 94 | 18.2 | 14.1 |

The above results show that the internal resistance (IR) reaches a minimum at 90% active content. Initially, an increase in active material content decreases the amount of electrochemically inert binder and carbon black, which contributes to reducing internal resistance by reducing the thickness and improving the contact points within the electrode. However, beyond this optimum, the lack of sufficient conductive agent creates a high-resistance pathway between active particles, actually causing the internal resistance (IR) to increase. The cycle performance test, as shown in Figure 3.2, demonstrates that high active material content (94%) dramatically compromises the cycle stability. After 300 cycles, the capacity retention for 90% active material was ~80%, while for 94% it was significantly lower.

Figure 3.2 Cycle capacity retention for different active material contents

The key takeaway is to find a trade-off between initial capacity and cycle stability. The final formulation selected for manufacturing the **traction battery pack** cells contains 90% active material, 5% conductive agent, and 5% binder. This mix achieves a high initial capacity of 950 mAh, low internal resistance of 12.9 mΩ, and a stable cycle life.

3.2 Balance of Capacity between the Anode and Cathode

The performance and safety of any lithium-ion cell, particularly in a large-format **traction battery pack**, are fundamentally dependent on the proper balance between anode and cathode capacity. If the negative electrode (graphite) has less reversible capacity than the positive electrode (LiMn₂O₄), lithium metal will plate on the graphite surface during charge, creating a severe safety risk. To guarantee that the anode capacity is always in excess, a parameter called capacity balance coefficient (CBC) is defined:

$$ CBC = \frac{C_{anode}}{C_{cathode}} = \frac{S_{anode} \cdot m_{anode} \cdot \eta_{anode}}{S_{cathode} \cdot m_{cathode} \cdot \eta_{cathode}} $$

In this equation, S is the specific capacity, m is the active material mass, and η is the active material content. Although theoretical chemistry requires a CBC of >1, the manufacturing process introduces tolerances. The coating weights of both anode and cathode exhibited a tolerance of ±2%. Therefore, the minimum design value in a real production line must ensure that even under worst-case scenarios (thickest cathode + thinnest anode), plating does not occur.

$$ CBC_{design} > \frac{1}{\left(1 – \text{manufacturing tolerance}\right)^2} $$

Figure 3.3 Discharge curves of cells with different capacity balance coefficient (CBC)

We evaluated cells with various capacity balance coefficients to identify the optimal value. Table 3.2 shows the data. A CBC of 1.06 resulted in a low discharge capacity due to excessive consumption of lithium in the SEI layer during the first cycle. With very high CBC values (1.18), the anode is heavily loaded, increasing the electrode’s density and making the jelly roll too thick for the case. This hinders the wetting of the electrolyte, leading to a lower cell capacity.

Table 3.2: Capacity Balance Coefficient and Battery Safety Test

| CBC | Overcharge / Puncture Behavior | Max Surface Temp (⁰C) | Safety Assessment |
| :—- | :—————————————- | :——————– | :—————- |
| 1.06 | Venting/Fire Risk | >100 | Poor |
| 1.10 | Stable, No venting | 75 | Good |
| 1.14 | Stable, No venting | 65 | Good |
| 1.18 | Stable, Delayed venting | 85 | Fair |

From an electrochemical and safety perspective, designing the cell within a capacity balance coefficient range of 1.10 to 1.13 appears optimal. Achieving a capacity balance coefficient of 1.14 provides excellent capacity while ensuring that the anode can accept all incoming lithium ions during high-rate charging. The final large-format cells in this study were designed using a ratio of 1.10—safeguarding long cycle life without a significant penalty in available capacity. This ratio controls the irreversible capacity and heat generation during the overcharge test, as seen in Table 3.2.

3.3 Determination of Electrode Compaction Density

After coating and drying, electrodes contain a significant amount of void volume. Roller pressing (calendering) increases the tap density of the electrode, improving the volumetric energy density of the **traction battery pack**. However, there is an optimum. Too much pressure destroys the electrode porosity and fractures the particles, whereas too little pressure does not produce the required thickness or sufficient inter-particle contact.

The assembly of the jelly roll into the steel can is also influenced by the electrode thickness. We assessed the assembly yield based on variations in the cathode compaction density, as shown in the table below.

Table 3.3: Effect of Cathode Compaction Density on Winding and Assembly

| Compaction Density (g·cm⁻³) | Assembly Success Rate (%) | Short Circuit Rate (%) |
| :————————- | :———————— | :——————— |
| 2.6 (Low) | 80 | 20 |
| 2.8 (Medium) | 90 | 10 |
| 3.0 (High) | 100 | 0 |
| 3.2 (Very High) | 70 | 30 |

At low compaction densities, the jelly roll is too thick to be inserted into the can without scraping off the electrode material, creating debris and short circuits. At very high compaction densities, electrodes become excessively brittle, causing cracking and delamination. Regarding capacity, the discharge curves of the batteries, as a function of cathode compaction density, are presented in Figure 3.4.

Figure 3.4 Discharge curves of cells with different LiMn₂O₄ pressing densities
Figure 3.5 The rate discharge capacity retention of cathode materials under different compaction densities

The figure shows that cells prepared with a lower compaction density have a lower initial discharge capacity due to the poor contact between particles and a reduced amount of active material per unit volume. Raising the density improves conductivity and increases capacity. This trend, however, reverses at very high values. For a compaction density of 3.2 g·cm⁻³, the electrolyte infiltration is severely inhibited. A critical observation from Figure 3.5 is the behavior at high C-rates. Electrodes pressed to a medium density of 3.0 g·cm⁻³ performed best in terms of high-rate capacity retention (90% at a 2C rate). However, at the highest density of 3.2 g·cm⁻³, the highly tortuous pore structure prevented the diffusion of lithium ions at high rates. Thus, the optimal compaction density for the LiMn₂O₄ cathode was determined to be 3.0 g·cm⁻³. Through a parallel set of experiments for the graphite anode, the optimal density was judged to be 1.5 g·cm⁻³.

3.4 Assembly Tightness of the Wound Core

The concept of assembly tightness is crucial for the design of the jelly roll within the can. Assembly tightness ($ \epsilon $) is defined mathematically as:

$$ \epsilon = \frac{T_{electrode} + T_{separator}}{D_{case}} $$

In this equation, the numerator is the total thickness of the electrode stack (anode + cathode) plus the separator, and the denominator is the internal diameter/thickness of the steel case. During charging, graphite anodes expand by approximately 10%. To avoid deforming the steel can (or generating high internal pressures), the assembly must allow for some expansion. The appropriate assembly tightness for this battery design is 0.88–0.92. Within this range, the jelly roll has acceptable contact with the case wall (reducing internal resistance) while not compromising the safety of the can over the lifetime of the battery.

3.5 Electrolyte Quantity

The electrolyte quantity is an often-underestimated variable in cell design. The electrolyte serves to connect the cathode and anode ionically. If the amount of electrolyte is insufficient, the active material in the core area of the jelly roll remains dry, leading to incomplete utilization of the active material and a reduced capacity. Conversely, excess electrolyte can pose leakage risks and complicate the formation process. In the experiments to determine the optimal electrolyte amount, cells (rated at ~1000 mAh) were assembled with varied electrolyte amounts from 2.5 to 5.0 grams.

Figure 3.6 Relation between battery capacity and electrolyte amount
Figure 3.7 Relation between the internal resistance and electrolyte amount

The results shown in Figure 3.6 indicate that capacity plateaus for electrolyte contents above 4.0 g. Figure 3.7 demonstrates that internal resistance is higher for insufficient electrolyte, but it also reaches a plateau after the 4.0 g threshold. This confirm that a minimum amount of electrolyte is required to fill the pores of the electrode and separator thoroughly. To further verify this, we produced five cells at various electrolyte levels to check the stability of the capacity distribution:
– At levels below 3.5 g, most cells resulted in capacities below 900 mAh, failing to meet the target.
– When the electrolyte amount was raised to 4.0 g, the pass rate surpassed 95%, with a narrow distribution. Increasing to 4.5 g gave a slightly better capacity distribution; however, some cells tended to leak during open formation, which corrodes equipment. Thus, to balance high productivity and equipment safety, the optimized amount chosen was 4.0 g per cell.

3.6 Formation Process

The formation step is one of the most critical in the manufacturing of a high-performance **traction battery pack**. It is a slow charging step that aims to react the lithium ions with the graphite anode to form the Solid Electrolyte Interface (SEI). The quality of the SEI layer is the primary determinant of the battery’s lifespan, safety, and self-discharge characteristics. We compared three distinct formation regimes, as shown in Table 3.4.

Table 3.4: Different Formation Systems

| Process | Formation Regime | Constant Current (mA) | Description |
| :—— | :—————————————————— | :——————– | :————————————— |
| A | 0.2C charge to 4.1V, or | 200 | Single step process |
| B | 0.1C charge for 30 min, then 0.2C to 4.1V | 100→200 | Two-step current with duration control |
| C | 0.05C charge for 30 min, then 0.2C to 4.1V | 50→200 | Longer low current treatment |

The reaction $ 2LiMn_2O_4 \rightarrow Li_{2-x}Mn_2O_4 + Li_xMn_2O_4 $ is benign; however, electrolyte is always decomposed on the surface of the anode. An optimized formation regime helps passivate this reaction.

Table 3.5: Cell capacity distribution, IR and charging efficiencies for various formation conditions

| Process | First Charge Efficiency (%) | Internal Resistance Range (mΩ) | Average Capacity (mAh) |
| :———– | :————————– | :—————————– | :——————— |
| A (0.2C) | 84.2 | 15–20 | 935 |
| B (0.1C/0.2C) | 86.0 | 14–18 | 948 |
| C (0.05C/0.2C) | 88.5 | 10–14 | 962 |

The low-current formation process (C) showed the highest first-cycle Coulombic efficiency of 88.5% and the lowest internal resistance. The formation process is responsible for minimizing lithium consumption in the formation of SEI. If we form using a high first step current, the SEI becomes inhomogeneous and results in lower battery performance. Because the huge gas generation process primarily occurs between 3.0 and 3.5V, a low initial current allows the gas to be generated slowly and escape through the open case, preventing swelling. It was determined that the cells should be charged from 50 mA (0.05 C) for 30 minutes, and then increased to a 0.2C current until the voltage reaches 4.1 V (Cut-off), followed by a rest period. This occurs because the complete formation process with long duration provides more opportunity for the formation of a uniform SEI, which has a crucial impact on cycling lifespan. The graph in Figure 3.8 confirms that process (C) not only yields a higher initial capacity but also yields a stable cycle life over 300 cycles.

Figure 3.8 Cycle retention of different formation methods

4. Development of the Large Capacity Traction Battery Pack

Based on the fully optimized manufacturing parameter ranges derived in previous sections, we scaled up the process to design and fabricate a large-format cell intended for use in an electric vehicle **traction battery pack**. The target was a nominal capacity of 10 Ah with a high power output. Table 4.1 details the specifications.

Table 4.1: Designed Parameters of Large Capacity Lithium-ion Power Battery

4.1 Cell Design and Calculations

The power output of the cell directly relates to the electrode area. The design process began with the required capacity of 10 Ah, considering a negative/positive capacity balance coefficient of +10% for optimal cycle life.

$$ \text{Design Capacity} = \frac{\text{Rated Capacity}}{0.95} = \frac{10 \text{ Ah}}{0.95} \approx 10.5 \text{ Ah} $$

4.1.1 Mass of Cathode and Anode Active Material

The cathode is the limiting electrode. The required mass of active material is determined as follows:

$$ m_{LiMn_2O_4} = \frac{ E_{design} }{ Q_{spec, cathode} } = \frac{10.5 \text{ Ah}}{0.12 \text{ Ah·g}^{-1}} = 87.5 \text{ g} $$

where $ E_{design} $ is the design capacity, and $ Q_{spec} $ is the specific capacity of LiMn₂O₄ at a low rate (~120 mAh/g). Taking the capacity balance coefficient (CBC) into account for the anode, the theoretical anode requirement becomes:

$$ m_{graphite} = \frac{ E_{design} \times CBC }{ Q_{spec, anode} } = \frac{10.5 \text{ Ah} \times 1.1}{0.33 \text{ Ah·g}^{-1}} = 35 \text{ g} $$

Here, the anode specific capacity $ Q_{spec, anode} $ is 330 mAh·g⁻¹. Since the active material content is ca. 90%, the actual mass of slurry applied onto the collectors is ~20% higher.

4.1.2 Determination of Single-Side Loading Density and Electrode Dimensions

To achieve the target high discharge rate and power output (5C, meaning 50A drain), we chose to use a “thick substrate, thin electrode” architecture. The slurry was coated on both sides. We decided on an areal loading density ($ \rho_{areal} $) for the cathode active material. This is a key design variable. Assuming a binder, conductive carbon, and active material mix totaling 100% mass:

$$ L_{cathode} = \frac{m_{total, mixed}}{\rho_{areal} \times W} $$

If the total required cathode slurry dry mass required is 97 g (87.5 g active / 0.90 active content) and we choose a double-side areal loading density of 25 mg·cm⁻² for the active material, the required surface area would be:

$$ A_{electrode} = \frac{m_{active}}{\rho_{areal}} = \frac{87.5 \text{ g}}{25 \text{ mg·cm}^{-2}} = 3500 \text{ cm}^2 $$

This area is used to determine the number of jelly rolls and the electrode length. For a cell width of 110 mm, the total physical electrode length is calculated to be ~150 cm. This necessitates using a multi-roll winding technique to create the cell.

4.1.3 Assembly Tightness and Case Calculations for the Battery

To calculate the jelly roll’s total thickness, we have to use a cell thickness of 55 mm and an internal height of 110 mm. The negative electrode width will be slightly smaller than the separator. The thickness of each layer is:

1. Cathode foil: 20 μm (current collector)
2. Cathode coating (single side): ~170 μm
3. Anode foil: 15 μm
4. Anode coating: ~180 μm

Respecting the assembly tightness (0.88–0.92) obtained previously, the calculated thickness of the jelly-roll should equal the internal diameter of the case minus ~10 % of the total electrode thickness. The anode experiences significant expansion during lithium intercalation. The calculations based on the cell dimensions indicated that a minimum level of stacking is required.

4.2 Structural Design and Performance of Cells for Battery Pack

We successfully manufactured 10 Ah class cells. A critical evaluation was then conducted using a variety of electrical tests. The details of our testing illustrate the performance of the prepared LiMn₂O₄ cells.

Table 4.2: Testing standards for traction battery packs

4.3 Rate Discharge Capability and Internal Resistance

We examined the performance of the large-format cells under a variety of discharge currents. In the first experiment, batteries were charged at a constant current at 0.3C to a voltage of 4.2V, followed by a constant voltage step until the current dropped below 0.05C. The cells were allowed to rest for 1 hour, and then they were discharged at different C-rates (0.2C, 0.5C, 1C, 2C, and 5C).

$$ \text{Specific Power} = \frac{V_{avg} \times I_{discharge}}{m_{cell}} \quad \text{and} \quad \text{Specific Energy} = \frac{V_{avg} \times I_{discharge} \times t}{m_{cell}} $$

As part of the performance evaluation, the 10 Ah pristine cells were examined using the BMS-technology-adapted equipment. At 0.2C, the cell discharged 10.1 Ah, showing a first-cycle columbic efficiency of 89%. When repeating the test using a 5C rate, the cell yielded 9.0 Ah, ~90% of the capacity at the 1C rate. The cell’s internal resistance was measured to be about 4.8 mΩ. The above values satisfy the needs for a **traction battery pack**.

Table 4.3: Rate Capability Data of the large format cell

| Discharge Current | Discharge Capacity (Ah) | Midpoint Voltage (V) | Capacity Retention vs. 0.2C (%) |
| :—————- | :———————- | :——————- | :—————————— |
| 1C (10A) | 9.85 | 3.65 | 97.5 |
| 2C (20A) | 9.50 | 3.52 | 94.1 |
| 5C (50A) | 8.90 | 3.35 | 88.1 |

4.4 Low and High Temperature Performance for EV Application

An essential feature of storage and transit in different weather conditions is the temperature. The traction battery pack must work in a wide temperature window. We evaluated the discharge capacity at different temperatures. The discharge capacity of the LiMn₂O₄ cell at -20⁰C was about 70% of the room-temperature capacity. When the cell was discharged at 55⁰C, the capacity was slightly higher. However, the cycle life at high temperature is less stable. Nonetheless, the performance of the prepared cells was judged to be adequate.

4.5 Self-Discharge and Storage Characteristics

The self-discharge and subsequent capacity recovery after storage are critical parameters, especially for market readiness. For EV applications, the battery may be parked for long durations. A full charge was followed by a 28-day storage test. After storage, the cell was discharged to 2.7 V at a 0.2C discharge rate. The capacity was ~9.15 Ah after storage. This indicates a self-discharge rate calculated as follows:

$$ \text{Self-discharge rate} = \frac{10.10\text{ Ah} – 9.15\text{ Ah}}{10.10 \text{ Ah}} \times 100\% \approx 9.4\% \text{ per 28 days} $$

After being recharged at a 0.2C discharge, the cell showed a capacity recovery of >97%. This is an excellent result for high-power spinel LiMn₂O₄ cells, indicating minimal structural damage and side reactions over the storage period.

4.6 Cycling Stability of the Traction Battery Pack

The cycling performance is one of the most critical attributes for a **traction battery pack**. The prepared large cell was subjected to a 500-cycle test at a 0.5C charge and 0.5C discharge rate. The data from experiments (Figure 4.1) suggests that the discharge capacity of the cell underwent a gradual decline. The battery retained about 81% of its initial capacity after 500 cycles. The initial capacity was more than 10 Ah, which decreased to ~8.1 Ah, a high value. The excellent capacity retention can be attributed to the optimized manufacturing parameters, such as compaction density, which maintain a good electrode morphology and compensate for particle cracking and manganese dissolution. The capacity fade was observed to occur in two distinct phases: an initial gradual decline during the first 100 cycles, followed by a more stable period thereafter.

Figure 4.1 Cycling performance of the large-format LiMn₂O₄ cell at 0.5C rate

4.7 Safety Performance Evaluation

The safety of a **traction battery pack** is the most important criterion for automakers. The 10 Ah large cells were subjected to a series of abuse tests, and the results were compared against the Chinese national standard, “QC/T 743-2006, Lithium-ion batteries for electric vehicles”. The cells must not explode or catch fire under these conditions, and in our case, all cells passed.

Table 4.4: Safety Evaluation of the LiMn₂O₄ Traction Battery Pack

These safety results are closely tied to the intrinsic stability of the spinel LiMn₂O₄. The output of the X-ray diffraction (XRD) analysis confirmed that the experimental material synthesized was pure spinel LiMn₂O₄, which has significantly higher thermal stability and better tolerance to overcharge than conventional layered LiCoO₂. This material combined with a well-designed assembly process results in a very safe design.

5. Conclusions and Future Prospects

A systematic study was conducted on the manufacturing parameters of a spinel lithium manganese oxide (LiMn₂O₄) cathode and a graphite anode for applications in a high-voltage **traction battery pack**. In this work, key manufacturing steps such as coating, electrode formulation, jelly-roll assembly, and formation have been thoroughly examined.

Table 5.1: Final optimal manufacturing parameters

The study revealed that a 90% cathode active content provided the ideal trade-off between capacity and internal resistance (12.9 mΩ). The capacity balance coefficient was found to be critical for safety. Specifically, a balance ratio of less than 1.1 was shown to lead to lithium plating at high rates, leading to overheating and venting. Meanwhile, sequential and robust testing of compaction density proved that an excess of density in electrodes does not directly translate into improved performance; on the contrary, high densities hinder Li⁺ transport and negatively impact very high-rate performance. The optimum densities to be used in cell manufacturing were determined as 3.0 g·cm⁻³ for LiMn₂O₄ and 1.5 g·cm⁻³ for graphite. The electrolyte content and formation cycles were optimized to produce a stable SEI, which ensures long cycle life and safety.

By combining these parameters, a high-capacity (10 Ah) prismatic cell was fabricated and evaluated. This cell succeeded in delivering a capacity of 10.1 Ah at 0.2C. A high-rate measurement allowed for 90% of this capacity at a 5C discharge rate, which constitutes a high specific power scenario needed for accelerating an electric vehicle. It also had a low internal resistance of 4.2 mΩ and showed excellent safety robustness, passing nail penetration, overcharge, and short-circuit abuse tests. The electrochemical evaluation demonstrates that the **traction battery pack** developed in this research possesses the high specific energy, specific power, long cycle life, and inherent safety required to meet or exceed the stringent requirements of the electric vehicle market.

This research contributes to the development of high-power **traction battery pack** technology, using safely available materials. The pivotal conclusions and prospects for future research are as follows:
1. Material Processability: This study confirms that valuable spinel LiMn₂O₄ cells are good candidates for the **traction battery pack** given the optimized conditions and at reduced processing costs.
2. Cycle Life: The cycle life may be further improved through surface doping or coating of the cathode material. The addition of elements such as aluminum and chromium can, to some extent, moderate manganese dissolution.
3. Battery Management System (BMS): The final safety and endurance of the **traction battery pack** are not solely dependent on cells. BMS development is needed to ensure cell-to-cell balancing and to prevent abusive conditions.
4. Scaling up: Future work will involve the integration of these cells into modular blocks and, ultimately, a high-voltage **traction battery pack**. This requires careful attention to thermal management to minimize temperature gradients within the pack.
5. Cost: With the increase in usage of the cell, we expect a reduction in material and manufacturing costs.

In future work, we will expand our production line from 10 Ah cells to larger capacity cells by increasing electrode thickness or reducing dead spaces in the jelly roll, as well as continue to work with the cell engineering aspects to push the boundaries of energy density. The primary goal of replacing old Cobalt-based materials for the electric **traction battery pack** is quite feasible from this study. Thus, this work lays a strong foundation for future optimization.