Fast charging has become the defining requirement for next-generation electric vehicle lithium ion battery systems. While 4 C to 5 C fast charging is currently being commercialized, the development of 8 C and higher-rate charging technology must be pursued to maintain technological leadership. In a lithium ion battery using lithium iron phosphate as the cathode and graphite as the anode, the primary bottleneck for extreme fast charging is the sluggish lithium insertion kinetics of the graphite anode. When the charging rate exceeds the tolerance limit of graphite, lithium plating occurs on the graphite surface, leading to rapid capacity fade and severe safety hazards. The lithium insertion process in graphite involves several sequential steps: desolvation of lithium ions and transport through the solid electrolyte interphase, charge transfer and intercalation into the graphite interlayer, solid-state diffusion within the graphite layers, and finally diffusion through the porous anode film. To achieve 8 C fast charging, a synergistic design of coke raw material selection, grinding, granulation, and soft carbon coating is essential to enhance each kinetic step. In this work, I systematically investigated three graphite anode designs for 8 C fast-charging lithium ion battery applications. By selecting isotropic petroleum coke and anisotropic needle coke, grinding them to specific median particle sizes, granulating them into secondary particles, and applying soft carbon coatings, I aimed to balance lithium ion intercalation and interlayer diffusion. The resulting graphite materials were evaluated in full lithium ion battery cells for kinetic performance, high-temperature storage, and cycling stability.
Graphite is the dominant anode material for lithium ion battery systems due to its low operating potential, high theoretical capacity, and excellent cycling stability. However, its intrinsic kinetic limitations become pronounced at high charging rates. The lithium ion battery fast-charging capability is governed by the transport of lithium ions across the electrode-electrolyte interface and within the graphite particles. The diffusion coefficient of lithium ions in graphite is highly dependent on the crystallographic orientation and the degree of graphitization. Anisotropic graphite, such as needle coke-derived graphite, exhibits fast lithium ion transport along the graphene layers but slow transport perpendicular to the layers. In contrast, isotropic graphite, such as petroleum coke-derived graphite, provides more uniform lithium ion pathways in all directions. Therefore, the choice of coke precursor directly influences the rate capability of the final graphite anode.
To quantify the kinetic limitations, we can consider the characteristic time for lithium ion diffusion in a spherical particle. The diffusion time constant is given by:
$$ \tau_d = \frac{R_p^2}{D_s} $$
where $$R_p$$ is the particle radius and $$D_s$$ is the solid-state diffusion coefficient. For a given material, reducing the particle size decreases the diffusion time, thereby improving rate capability. However, smaller particles also increase the surface area, which can lead to more side reactions with the electrolyte, especially at high temperatures. Therefore, an optimal particle size must be identified to balance kinetics and stability. In this study, I ground isotropic petroleum coke to a median particle size (D50) of 7.5 μm and anisotropic needle coke to 6.5 μm. The smaller particle size for needle coke compensates for its anisotropic diffusion behavior, ensuring that lithium ion intercalation and interlayer diffusion are both sufficiently fast for 8 C charging.
Granulation into secondary particles is another effective strategy to improve the isotropy of graphite. By aggregating primary particles into spherical secondary particles, the random orientation of the primary particles increases the overall isotropy of the electrode. This reduces the tortuosity for lithium ion transport and enhances the rate performance. Furthermore, soft carbon coating is a classic and effective method for designing fast-charging graphite. The soft carbon layer increases the surface disorder, facilitates the formation of a uniform, robust, and lithium fluoride-rich solid electrolyte interphase, and lowers the charge transfer energy barrier. The soft carbon also acts as a binder between primary particles, providing electronic connections and mechanical integrity. The combination of granulation and soft carbon coating can synergistically improve the lithium ion battery fast-charging capability.
In this work, I designed three graphite materials labeled GX-1, GX-2, and GX-3. The raw coke materials were first characterized using polarized light microscopy. The petroleum coke exhibited a predominantly short-fiber and mosaic structure, indicating isotropic character. The needle coke exhibited a long-fiber structure, indicating anisotropic character. Because petroleum coke is more isotropic, I ground it to a larger D50 of 7.5 μm, whereas needle coke was ground to a smaller D50 of 6.5 μm. The primary particles were then granulated into secondary particles with a D50 of approximately 12–13 μm. For GX-1, the petroleum coke primary particles were first graphitized at 3000 °C, then granulated and coated with pitch, resulting in a soft carbon coating amount of 2.5%. For GX-2, the petroleum coke primary particles were granulated before graphitization, then graphitized at 3000 °C, and finally coated with pitch, resulting in a soft carbon coating amount of 1.8%. For GX-3, the needle coke primary particles were granulated before graphitization, then graphitized at 3000 °C, and finally coated with pitch, resulting in a soft carbon coating amount of 1.8%. The carbonization temperature for the pitch coating was 1150 °C in all cases.
The physical and chemical properties of the three graphites are summarized in Table 1. The particle size distributions were measured using laser diffraction. GX-1, GX-2, and GX-3 exhibited similar D50 values, although GX-3 showed slightly lower D50 and D90 due to the smaller grinding size of the needle coke. The degree of graphitization was calculated from X-ray diffraction patterns. GX-1 and GX-2 had similar degrees of graphitization, around 95.0% and 95.2%, respectively. GX-3 had a slightly lower degree of graphitization of 94.75%, but because needle coke inherently has a more ordered structure, its crystallite size (La and Lc) was the largest, and its La/Lc ratio was the highest. The OI value, defined as the intensity ratio of the 004 to 110 diffraction peaks, is a measure of the degree of anisotropy. All three graphites exhibited low OI values, indicating a high degree of isotropy, which is beneficial for fast charging. The tap density and specific surface area were also measured. GX-1 had the highest specific surface area due to its higher coating amount and less uniform coating. GX-2 and GX-3 had lower specific surface areas. The electrode rebound after calendering was measured after 24 hours. GX-2 exhibited the lowest rebound, which correlates with its highest powder compaction density. GX-1 and GX-3 had higher rebound values, which can be attributed to their higher specific surface areas and more porous structures.
| Graphite | D10 (μm) | D50 (μm) | D90 (μm) | Graphitization degree (%) | La (nm) | Lc (nm) | La/Lc | OI value | Tap density (g cm⁻³) | Specific surface area (m² g⁻¹) | Reversible capacity (mAh g⁻¹) | First efficiency (%) | Electrode rebound (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GX-1 | 7.24 | 12.88 | 20.72 | 95.00 | 106.22 | 35.22 | 3.02 | 1.86 | 0.93 | 1.17 | 351.34 | 94.48 | 7.27 |
| GX-2 | 7.75 | 13.10 | 21.34 | 95.20 | 112.71 | 38.96 | 2.89 | 2.57 | 1.00 | 0.81 | 350.41 | 94.48 | 5.35 |
| GX-3 | 7.29 | 11.92 | 19.80 | 94.75 | 151.39 | 43.70 | 3.46 | 2.15 | 1.01 | 0.96 | 349.40 | 94.33 | 9.22 |
The powder compaction density as a function of pressure was also evaluated. GX-2 exhibited the highest powder compaction density at all pressures, which can be attributed to its larger primary particle size and lower soft carbon coating amount. At low pressures, GX-3 showed a lower compaction density than GX-1, but at high pressures, the two became similar. The surface disorder of the soft carbon coating was analyzed using Raman spectroscopy. The intensity ratio of the D band to the G band (ID/IG) was calculated for each scanning point, and the distribution was statistically analyzed. GX-1 had the highest ID/IG values and the widest distribution, indicating the highest surface disorder but poor coating uniformity. This is because the graphitization-after-granulation-and-coating process results in less uniform coating and less pore filling. GX-2 and GX-3 had more uniform coatings but lower average ID/IG values due to the lower coating amount. Interestingly, although GX-2 and GX-3 had the same coating amount, GX-3 exhibited a lower surface disorder than GX-2. This is because GX-3 used a smaller primary particle size, and during coating, some pitch was consumed to bind the small particles together, leaving less soft carbon on the surface. The morphology of the three graphites was also observed, confirming the secondary particle structure with primary particles connected by soft carbon.
To evaluate the kinetic performance of the three graphites, full lithium ion battery cells were assembled. The cathode was lithium iron phosphate with a composition of 96.8% active material, 0.7% conductive carbon black, 0.5% graphene conductive slurry, and 2.0% polyvinylidene fluoride binder. The cathode coating density was 348 g m⁻² on both sides, and the compaction density was 2.30 g cm⁻³. The anode was composed of 96.5% graphite, 0.5% conductive carbon black, 1.2% sodium carboxymethyl cellulose, and 1.8% styrene-butadiene rubber binder. The anode coating density was 164 g m⁻² on both sides, and the compaction density was 1.65 g cm⁻³. A 12 μm polyethylene separator was used. The cells were assembled in an aluminum-plastic pouch, filled with electrolyte, formed, and aged. Electrochemical impedance spectroscopy was performed at 50% state of charge, and the Nyquist plots were fitted using an equivalent circuit model. The equivalent circuit consisted of an ohmic resistance (Rs) in series with a parallel combination of a constant phase element (CPE) and a charge transfer resistance (Rct) in series with a Warburg impedance (Zw), as shown in the following equation:
$$ Z = R_s + \frac{1}{j \omega C_{dl} + \frac{1}{R_{ct} + Z_W}} $$
where $$Z_W = \sigma (1-j) \omega^{-1/2}$$ and $$\sigma$$ is the Warburg coefficient. The fitted parameters Rs, Rct, and Rw (where Rw is the diffusion resistance related to Zw) were obtained. The exchange current density $$i_0$$ is related to Rct by:
$$ i_0 = \frac{RT}{n F R_{ct}} $$
where R is the gas constant, T is the absolute temperature, n is the number of electrons transferred, and F is the Faraday constant. A higher i0 indicates faster charge transfer kinetics. The diffusion coefficient D can be estimated from the Warburg coefficient using the following equation:
$$ D = \frac{1}{2} \left( \frac{RT}{n^2 F^2 A C \sigma} \right)^2 $$
where A is the electrode area and C is the concentration of lithium ions. The kinetic performance of the three graphites is summarized in Table 2. GX-1 exhibited the lowest Rs, which is consistent with its highest electrode rebound. The electrode rebound creates more porosity, which facilitates lithium ion transport in the electrolyte, reducing the ionic resistance. GX-2 had the highest Rs, likely due to its lower porosity and higher compaction. The Rct values followed the order GX-1 < GX-3 < GX-2. GX-1 had the highest surface disorder and the most uniform SEI, which reduced the charge transfer barrier. GX-3, despite having a lower surface disorder than GX-1, had a higher degree of isotropy and a more favorable crystallite orientation, which facilitated lithium ion intercalation. GX-2 had the highest Rct due to its lower surface disorder and higher anisotropy. The diffusion resistance Rw followed the order GX-1 < GX-3 < GX-2. GX-1 benefited from the soft carbon connections between primary particles, which provided fast lithium ion diffusion pathways. GX-3 benefited from its smaller particle size and more porous electrode structure, which shortened the diffusion length and reduced tortuosity. GX-2 suffered from larger primary particles and a denser electrode, which increased the diffusion resistance. The direct current resistance (DCR) at different depths of discharge is a comprehensive measure of the lithium ion battery kinetics. GX-2 had a DCR that was approximately 7% higher than that of GX-1 and GX-3. The DCR increase in GX-2 originated mainly from the higher Rct and Rw. Despite these differences in DCR, the constant-current charge acceptance at different rates was similar for all three graphites. This is because the higher DCR of GX-2 caused a larger temperature rise during charging, which accelerated the charge transfer and lithium ion transport, compensating for the higher resistance. At 8 C, the constant-current charge acceptance of all three graphites exceeded 87%. With further optimization of the electrode design, such as gradient electrode structures or fast-charging electrolytes, these graphites could potentially enable even higher charging rates in lithium ion battery cells.
| Graphite | Rs (mΩ) | Rct (mΩ) | Rw (mΩ) | DCR at 50% SOC (mΩ) | 8 C charge acceptance (%) |
|---|---|---|---|---|---|
| GX-1 | 18.2 | 12.5 | 8.3 | 42.1 | 88.5 |
| GX-2 | 20.5 | 18.7 | 12.1 | 45.0 | 87.2 |
| GX-3 | 19.1 | 14.2 | 9.0 | 42.5 | 88.1 |
The high-temperature storage performance of the lithium ion battery cells was evaluated at 55 °C and 100% state of charge. The cells were stored for 7, 14, 21, and 28 days. The capacity recovery rate and capacity retention rate were measured. The capacity recovery rate reflects the reversible capacity loss, while the capacity retention rate reflects both reversible and irreversible capacity losses. Therefore, the capacity retention rate is always lower than the capacity recovery rate. The storage performance followed the order GX-2 > GX-3 > GX-1. As the storage time increased, the storage performance of all three graphites deteriorated, and the degradation rate followed the same order: GX-2 > GX-3 > GX-1. This indicates that the graphite with better initial storage performance also degrades more slowly. Comparing the 28-day storage to the 7-day storage, the percentage decrease in both capacity recovery rate and capacity retention rate was similar. This suggests that the degradation during prolonged storage is mainly due to irreversible capacity loss. Previous studies have shown that a fully charged anode undergoes lithium metal dissolution upon heating, which then reacts with the electrolyte, resulting in irreversible capacity loss. For fast-charging graphite, a faster charge transfer process leads to faster lithium metal dissolution and worse storage performance. Therefore, the storage performance of the three graphites is inversely correlated with their Rct values. GX-1, with the lowest Rct, had the worst storage performance. GX-2, with the highest Rct, had the best storage performance. GX-3, with intermediate Rct, had intermediate storage performance. The storage performance data are summarized in Table 3.
| Graphite | Storage time (days) | Capacity recovery rate (%) | Capacity retention rate (%) |
|---|---|---|---|
| GX-1 | 7 | 96.5 | 94.2 |
| 28 | 92.1 | 89.5 | |
| GX-2 | 7 | 98.2 | 96.8 |
| 28 | 95.6 | 93.9 | |
| GX-3 | 7 | 97.3 | 95.5 |
| 28 | 94.0 | 92.1 |
The cycling performance of the lithium ion battery cells was evaluated under two conditions: fast-charging cycling at 25 °C with an 8 C stepped charging protocol, and high-temperature cycling at 45 °C with 1 C charge and 1 C discharge. The 8 C stepped charging protocol consisted of the following steps: 0.5 C constant current charge to 10% SOC, 4 C constant current charge to 20% SOC, 6 C constant current charge to 30% SOC, 8 C constant current charge to 50% SOC, 6 C constant current charge to 70% SOC, 4 C constant current charge to 80% SOC, and finally 0.5 C constant current-constant voltage charge to 100% SOC. The discharge was at 1 C. After 500 cycles at 25 °C, the capacity retention rates of GX-1, GX-2, and GX-3 were 95.10%, 95.09%, and 95.89%, respectively. GX-3 exhibited the best fast-charging cycling performance, with a projected cycle life of 4000 cycles to 80% capacity retention. GX-1 and GX-2 showed similar fast-charging cycling performance, with projected cycle lives of approximately 3500 cycles to 80% capacity retention. The high-temperature cycling performance at 45 °C after 600 cycles showed capacity retention rates of 87.57%, 88.22%, and 89.20% for GX-1, GX-2, and GX-3, respectively. GX-3 again exhibited the best high-temperature cycling performance, with a projected cycle life of 1500 cycles to 80% capacity retention. GX-2 was second, and GX-1 was the worst, with a projected cycle life of 1200 cycles to 80% capacity retention. The capacity loss during cycling is mainly attributed to side reactions and lithium plating. At low temperatures (25 °C), side reactions are slow, and lithium plating dominates the capacity loss. At high temperatures (45 °C), side reactions are more severe and contribute more to the capacity loss. Therefore, for both fast-charging and high-temperature cycling, GX-3, which has both good kinetics and good storage performance, performed the best. GX-1 had good kinetics but poor storage performance, while GX-2 had good storage performance but poor kinetics. Neither could match the overall performance of GX-3. The cycling performance data are summarized in Table 4.
| Graphite | Fast-charging cycling at 25 °C after 500 cycles (capacity retention, %) | Projected cycles to 80% retention (fast-charging) | High-temperature cycling at 45 °C after 600 cycles (capacity retention, %) | Projected cycles to 80% retention (high-temperature) |
|---|---|---|---|---|
| GX-1 | 95.10 | 3500 | 87.57 | 1200 |
| GX-2 | 95.09 | 3500 | 88.22 | 1300 |
| GX-3 | 95.89 | 4000 | 89.20 | 1500 |
The results of this study demonstrate that the design of graphite anodes for 8 C fast-charging lithium ion battery applications requires a careful balance between kinetics and stability. The use of isotropic petroleum coke allows for a larger particle size, which reduces the specific surface area and improves the high-temperature storage performance. However, the graphitization-after-granulation-and-coating process for petroleum coke (GX-1) results in a non-uniform coating and a higher surface area, which leads to severe side reactions and poor high-temperature storage and cycling performance. The alternative process of granulation before graphitization and coating after graphitization (GX-2) improves the coating uniformity and reduces the surface area, resulting in better high-temperature storage performance. However, the larger primary particle size and lower coating amount increase the charge transfer and diffusion resistances, leading to poorer kinetics. The use of anisotropic needle coke with a smaller particle size and granulation before graphitization followed by coating after graphitization (GX-3) provides the optimal combination. The smaller particle size compensates for the anisotropic diffusion, while the granulation and soft carbon coating improve the isotropy and charge transfer kinetics. The resulting graphite exhibits low charge transfer and diffusion resistances, good high-temperature storage performance, and excellent cycling stability. The projected cycle life of 4000 cycles for fast charging and 1500 cycles for high-temperature cycling to 80% capacity retention is promising for practical applications.
To further understand the kinetic limitations, we can analyze the lithium ion diffusion process using Fick’s second law. For a spherical particle with radius R, the concentration of lithium ions as a function of radial position r and time t is given by:
$$ \frac{\partial c}{\partial t} = D \left( \frac{\partial^2 c}{\partial r^2} + \frac{2}{r} \frac{\partial c}{\partial r} \right) $$
The solution for a constant surface concentration and initially uniform concentration is an infinite series. However, for fast charging, the characteristic diffusion time is more relevant. The dimensionless parameter that governs the diffusion limitation is the Fourier number:
$$ Fo = \frac{D t}{R^2} $$
When $$Fo \ll 1$$, the diffusion is limited by the particle size. For 8 C charging, the charging time is 450 seconds. Assuming a diffusion coefficient of $$10^{-10} \text{ cm}^2 \text{ s}^{-1}$$ for lithium ions in graphite, and a particle radius of 5 μm (for D50 = 10 μm), the Fourier number is approximately 0.18, which is not negligible. This indicates that lithium ion diffusion within the particle is a significant limitation. By reducing the particle size to 3.25 μm (for D50 = 6.5 μm), the Fourier number increases to 0.43, which is more favorable. However, reducing the particle size further increases the surface area and side reactions. Therefore, an optimal particle size must be found, which in this study is 6.5 μm for needle coke and 7.5 μm for petroleum coke.
The charge transfer kinetics can be described by the Butler-Volmer equation:
$$ i = i_0 \left[ \exp\left( \frac{\alpha_a n F \eta}{RT} \right) – \exp\left( -\frac{\alpha_c n F \eta}{RT} \right) \right] $$
where i is the current density, i0 is the exchange current density, αa and αc are the anodic and cathodic transfer coefficients, n is the number of electrons, F is the Faraday constant, η is the overpotential, R is the gas constant, and T is the temperature. The exchange current density is related to the charge transfer resistance by:
$$ i_0 = \frac{RT}{n F R_{ct}} $$
Therefore, a lower Rct corresponds to a higher i0 and faster charge transfer kinetics. The soft carbon coating reduces Rct by modifying the surface chemistry and promoting the formation of a favorable SEI. The presence of soft carbon also increases the electronic conductivity and provides more active sites for lithium ion intercalation. The uniform and robust SEI rich in lithium fluoride reduces the desolvation energy barrier and facilitates lithium ion transport across the interface. These factors contribute to the improved rate capability of the graphite anodes.
The high-temperature storage performance is strongly influenced by the reactivity of the lithiated graphite surface with the electrolyte. At high temperatures, the SEI can decompose or dissolve, exposing fresh graphite to the electrolyte and causing continuous side reactions. The soft carbon coating can mitigate this by providing a more stable and uniform SEI. However, if the coating is too thin or non-uniform, the protection is insufficient. In GX-1, the high coating amount (2.5%) and the post-graphitization coating process resulted in a high surface area and a non-uniform coating, which led to severe side reactions and poor storage performance. In GX-2 and GX-3, the lower coating amount (1.8%) and the pre-graphitization granulation process resulted in a more uniform coating and a lower surface area, which improved the storage performance. GX-3, with its smaller particle size and more isotropic structure, exhibited the best storage performance among the three, because its lower Rct reduced the driving force for lithium metal dissolution and side reactions.
The cycling performance is a complex interplay of lithium plating, SEI growth, and particle cracking. During fast charging, lithium plating can occur if the local overpotential at the graphite surface exceeds the threshold for lithium deposition. The overpotential is the sum of the charge transfer overpotential, the diffusion overpotential, and the ohmic overpotential. By reducing the particle size, improving the isotropy, and applying a soft carbon coating, the overpotential can be reduced, thereby suppressing lithium plating and improving the fast-charging cycle life. During high-temperature cycling, the SEI growth and side reactions dominate the capacity fade. A stable and uniform SEI, as achieved in GX-3, can significantly improve the high-temperature cycle life. The combination of good kinetics and good stability makes GX-3 the optimal design for 8 C fast-charging lithium ion battery applications.
The design principles established in this study can be extended to even higher charging rates. For 10 C or higher, further reductions in particle size and more advanced coating strategies may be required. However, the trade-off between kinetics and stability must always be considered. The use of isotropic cokes with well-controlled particle size distributions, granulation into secondary particles, and optimized soft carbon coatings can provide a pathway to achieve extreme fast charging without sacrificing cycle life and safety. The lithium ion battery industry is moving towards higher charging rates, and the graphite anode design will play a critical role in enabling this transition. The findings of this study provide a comprehensive framework for the development of next-generation fast-charging graphite anodes.
In summary, I have designed and evaluated three graphite anode materials for 8 C fast-charging lithium ion battery applications. The use of isotropic petroleum coke with graphitization after granulation and coating (GX-1) resulted in low charge transfer and diffusion resistances, but poor high-temperature storage and cycling performance due to severe side reactions. The use of isotropic petroleum coke with granulation before graphitization and coating after graphitization (GX-2) improved the high-temperature storage performance but increased the charge transfer and diffusion resistances, leading to suboptimal kinetics. The use of anisotropic needle coke with a smaller particle size, granulation before graphitization, and coating after graphitization (GX-3) provided the best overall performance, with excellent kinetics, high-temperature storage, and cycling stability. The projected cycle life of 4000 cycles for fast charging and 1500 cycles for high-temperature cycling to 80% capacity retention demonstrates the potential of this design for practical lithium ion battery applications. The synergistic design of coke raw material selection, grinding, granulation, and soft carbon coating is essential for achieving 8 C and higher fast charging in lithium ion battery systems. This work provides a valuable design strategy for the development of advanced graphite anodes for extreme fast-charging lithium ion batteries.
To further optimize the performance, future work should focus on the development of advanced coating materials and processes that can provide even lower charge transfer resistance while maintaining high stability. The use of gradient electrode designs, where the porosity and particle size vary across the electrode thickness, can also enhance lithium ion transport and reduce lithium plating. Additionally, the combination of fast-charging graphite with advanced electrolytes that form stable and conductive SEI layers can further improve the rate capability and cycle life. The lithium ion battery is a complex system, and the anode design must be integrated with the cathode, electrolyte, and cell design to achieve the best performance. The results of this study provide a solid foundation for the continued development of extreme fast-charging lithium ion battery technology.