Since the first commercial production of lithium-ion batteries by Sony in 1991, these energy storage devices have revolutionized various sectors, including electric vehicles, consumer electronics, and grid storage, due to their high energy density, excellent power capability, and long cycle life. The performance of a lithium-ion battery is intrinsically tied to its components: the cathode, anode, separator, and electrolyte. Among these, the anode material, which hosts lithium ions during charging and discharging, plays a critical role in determining key metrics such as energy density, rate capability, low-temperature operation, and cycling stability. Graphite-based materials dominate the anode market, accounting for over 95% of commercial applications, owing to their favorable lithium intercalation properties, structural stability, and relatively low cost. Graphite consists of layered hexagonal planes of carbon atoms held together by strong in-plane covalent bonds and weak interplanar van der Waals forces, facilitating the reversible insertion and extraction of lithium ions.
Graphite anodes are primarily classified into two categories: artificial graphite and natural graphite. Artificial graphite is synthesized from petroleum coke or pitch through processes like crushing, shaping, and high-temperature graphitization. It typically offers good rate performance and long cycle life but suffers from lower specific capacity (usually below 360 mAh/g) and higher production costs. In contrast, natural graphite is derived from mined graphite ore, which is processed via milling and spheroidization. It exhibits a higher specific capacity (often exceeding 360 mAh/g), approaching the theoretical limit of 372 mAh/g for graphite, and benefits from simpler manufacturing and lower raw material costs. However, natural graphite has inherent drawbacks, including internal porosity within particles, which leads to uneven stress distribution during cycling. This porosity causes particle cracking, electrolyte infiltration, and continuous solid electrolyte interphase (SEI) growth, resulting in rapid capacity fade and limited cycle life (typically around 500 cycles). To address these issues, modifications such as surface coating, particle shaping, and internal densification have been explored. Densification, in particular, involves filling the internal pores of natural graphite with carbonaceous precursors like asphalt under high pressure and temperature, followed by graphitization. This process reduces porosity, enhances structural integrity, and improves electrochemical performance.
In this study, we investigate the impact of densification treatment on natural graphite by varying the asphalt addition levels. We prepare densified natural graphite samples and composite them with artificial graphite at different ratios. The physical properties of these materials are characterized, and their electrochemical performances are evaluated in full-cell configurations with lithium cobalt oxide (LCO) cathodes. Our goal is to elucidate the relationships between densification degree, composite formulation, and battery performance, providing insights for optimizing anode materials in lithium-ion batteries.
The fundamental operation of a lithium-ion battery relies on the shuttling of lithium ions between the cathode and anode during charge and discharge cycles. The anode’s ability to intercalate and deintercalate lithium ions efficiently is paramount. For graphite anodes, the intercalation process can be described by the following reaction: $$ \text{C} + x\text{Li}^+ + x\text{e}^- \leftrightarrow \text{Li}_x\text{C} $$ where $x$ represents the stoichiometric coefficient, with a maximum theoretical value of 1, corresponding to a capacity of 372 mAh/g. The practical capacity, however, is influenced by factors such as graphitization degree, particle morphology, and surface chemistry. The graphitization degree ($g$) can be estimated from X-ray diffraction (XRD) data using the formula: $$ g = \frac{0.3440 – d_{002}}{0.3440 – 0.3354} \times 100\% $$ where $d_{002}$ is the interlayer spacing of the (002) plane in nanometers. Higher $g$ values indicate better crystallinity, which generally correlates with higher specific capacity and electrical conductivity.
Natural graphite typically contains both hexagonal (2H) and rhombohedral (3R) phases, as detected by XRD. The presence of the 3R phase can affect lithium ion diffusion pathways and mechanical stability. The orientation index (OI), calculated from XRD peak intensities as: $$ \text{OI} = \frac{I_{(004)}}{I_{(110)}} $$ where $I_{(004)}$ and $I_{(110)}$ are the intensities of the (004) and (110) peaks, respectively, indicates the preferred orientation of graphite crystallites in the electrode coating. Lower OI values suggest better isotropy, which is beneficial for reducing anisotropic expansion and improving lithium ion diffusion kinetics.
The electrochemical impedance of a lithium-ion battery, particularly the direct current internal resistance (DCIR), is a critical parameter influencing rate performance and low-temperature behavior. DCIR comprises ohmic resistance, charge transfer resistance, and diffusion resistance. It can be modeled using equivalent circuit models, but for simplicity, we focus on its empirical impact on battery performance. The capacity retention during cycling is often modeled by a power-law decay: $$ C_n = C_0 \cdot n^{-\alpha} $$ where $C_n$ is the capacity at cycle $n$, $C_0$ is the initial capacity, and $\alpha$ is the decay coefficient. Lower $\alpha$ values indicate better cycling stability.
To systematically evaluate the densification effect, we prepared several graphite samples. Natural graphite microspheres (with a median particle size D50 of approximately 15–16 μm) were used as the starting material. For ordinary natural graphite (NG), the microspheres were mixed with 5.5 wt% asphalt, carbonized at 1150°C, and then graphitized at 2500°C. For densified natural graphite, the microspheres were mixed with varying amounts of petroleum asphalt (5.5 wt%, 8 wt%, and 11 wt%) and subjected to hot isostatic pressing at 250°C and 150 MPa for 90 minutes. After cooling, the products were disaggregated, sieved, and graphitized at 2500°C to obtain samples labeled NG-0 (5.5 wt% asphalt), NG-1 (8 wt% asphalt), and NG-2 (11 wt% asphalt). Composite graphite samples were prepared by blending NG-2 with artificial graphite (secondary particles with a D50 of 14 μm, derived from petroleum coke) at mass ratios of 70:30 and 50:50, denoted as CG-1 and CG-2, respectively.
The physical properties of the graphite materials were characterized using laser diffraction particle size analysis, nitrogen adsorption-desorption for specific surface area (SSA), XRD for graphitization degree and phase analysis, and scanning electron microscopy (SEM) for morphological examination. Electrodes were fabricated by mixing the graphite materials with carbon black conductive agent, carboxymethyl cellulose sodium (CMC), and styrene-butadiene rubber (SBR) in a mass ratio of 96.5:1:1:1.5 in deionized water to form a slurry. The slurry was coated onto copper foil, dried, calendared, and cut into electrodes. The cathode was prepared using LCO (lithium cobalt oxide) with a cutoff voltage of 4.4 V, mixed with carbon nanotube (CNT), carbon black, and polyvinylidene fluoride (PVDF) in a ratio of 98:0.5:0.5:1 in N-methyl-2-pyrrolidone (NMP). Pouch full-cells with a design capacity of 1500 mAh were assembled using these electrodes, a separator, and standard electrolyte. Electrochemical tests included DCIR measurement, rate capability tests at various C-rates (from 0.2C to 2C), low-temperature discharge tests (from 25°C to -20°C), and long-term cycling tests at 25°C with a charge/discharge protocol of 0.5C/1C.
The physical properties of the graphite samples are summarized in Table 1. The data reveal that densification treatment and composite formation significantly influence key parameters. All natural graphite samples exhibit similar particle size distributions, but the composites show slightly smaller D50 values due to the incorporation of finer artificial graphite particles. As the asphalt content increases in densified natural graphite, the SSA decreases, indicating reduced internal porosity. This leads to improved initial coulombic efficiency (ICE), as less lithium is consumed in forming SEI on fresh surfaces exposed by pores. However, the graphitization degree and specific capacity slightly decline with higher asphalt addition, as the asphalt-derived carbon has lower crystallinity. The composites show further reductions in graphitization degree and specific capacity due to the inclusion of artificial graphite, which typically has lower graphitization. The ICE of composites is also lower, attributed to the smaller particle size of artificial graphite increasing surface area.
| Sample | Specific Surface Area (m²/g) | Particle Size D10 (μm) | Particle Size D50 (μm) | Particle Size D90 (μm) | Graphitization Degree (%) | Specific Capacity (mAh/g) | Initial Coulombic Efficiency (%) |
|---|---|---|---|---|---|---|---|
| NG | 2.05 | 10.8 | 17.4 | 28.2 | 98.0 | 362.4 | 94.3 |
| NG-0 | 2.10 | 10.8 | 17.3 | 28.0 | 98.0 | 363.1 | 94.7 |
| NG-1 | 1.65 | 10.6 | 17.0 | 26.1 | 97.5 | 360.7 | 95.3 |
| NG-2 | 1.81 | 10.7 | 17.2 | 26.4 | 97.0 | 359.5 | 95.5 |
| CG-1 | 1.76 | 10.5 | 16.9 | 26.1 | 95.5 | 358.2 | 95.2 |
| CG-2 | 1.62 | 10.3 | 16.6 | 25.7 | 94.1 | 356.9 | 94.8 |
SEM images reveal the morphological changes. Ordinary natural graphite (NG) particles show smooth surfaces, while densified samples (NG-0, NG-1, NG-2) exhibit rougher surfaces due to asphalt coating and processing. Cross-sectional SEM confirms that NG has numerous internal pores, whereas densified samples display progressively fewer pores with increasing asphalt content, demonstrating effective pore filling. The composite samples (CG-1, CG-2) show a mixture of spherical natural graphite and flaky artificial graphite particles.
XRD patterns indicate that all natural graphite samples contain both 2H and 3R phases, with characteristic peaks around 26.5° for (002) and smaller peaks in the 40–48° range for the 3R phase. The composites show diminished 3R phase peaks as artificial graphite content increases, with CG-2 nearly lacking detectable 3R phase, highlighting the phase composition differences.
The electrode properties were evaluated through compaction tests, orientation index (OI) measurements, and adhesion strength tests. The compaction curves show that natural graphite samples have higher compaction densities due to their softer and more spherical particles, which facilitate particle rearrangement under pressure. Densified natural graphite maintains similar compaction behavior, while composites exhibit lower compaction densities because artificial graphite is harder and less deformable. The OI values, calculated from XRD data of coated electrodes, decrease with densification and further with composite formation, indicating improved isotropy. This is beneficial for reducing anisotropic expansion during cycling. Adhesion strength, measured via peel tests, is higher for densified natural graphite due to rougher surfaces enhancing binder adhesion, whereas composites show slightly lower adhesion due to the presence of artificial graphite.
The electrochemical performance in full lithium-ion battery cells is critically analyzed. DCIR measurements at 50% state of charge (SOC) reveal that densified natural graphite samples have lower resistance compared to NG, with NG-2 showing the lowest among natural graphite samples. This reduction is attributed to decreased porosity and improved electrical connectivity within particles. The composites exhibit even lower DCIR, with CG-2 having the smallest value, due to the synergistic effect of densified natural graphite and highly conductive artificial graphite. The relationship between DCIR and asphalt content can be approximated by: $$ \text{DCIR} \approx R_0 – k \cdot w $$ where $R_0$ is the baseline resistance, $k$ is a positive constant, and $w$ is the asphalt addition level, indicating that higher densification reduces impedance.
Rate capability tests assess the ability of the lithium-ion battery to deliver capacity under high current densities. The results are summarized in Table 2. At low rates (≤0.5C), all samples show similar capacity retention. However, at higher rates (1C and 2C), densified natural graphite outperforms NG, and composites perform best. For instance, at 2C discharge, CG-2 retains over 90% of its 0.2C capacity, while NG retains only about 85%. This enhancement is linked to lower DCIR and better lithium ion diffusion kinetics. The capacity retention at a given C-rate can be modeled as: $$ \text{Retention}(\%) = 100 \cdot e^{-\beta \cdot I} $$ where $I$ is the discharge current and $\beta$ is a coefficient inversely related to material conductivity and diffusion rate.
| Sample | 0.2C | 0.5C | 1C | 2C |
|---|---|---|---|---|
| NG | 100 | 98.5 | 95.2 | 85.1 |
| NG-1 | 100 | 98.8 | 96.5 | 88.3 |
| NG-2 | 100 | 99.0 | 97.1 | 89.7 |
| CG-1 | 100 | 99.2 | 97.8 | 91.5 |
| CG-2 | 100 | 99.4 | 98.3 | 92.8 |
Low-temperature performance is crucial for lithium-ion batteries operating in cold climates. Discharge capacity retention at various temperatures is shown in Table 3. At 0°C and above, differences are minimal. However, at -10°C and -20°C, densified natural graphite maintains higher capacity than NG, and composites excel further. For example, at -20°C, CG-2 retains about 75% of its room-temperature capacity, while NG retains only 65%. This improvement stems from reduced charge transfer resistance and enhanced ion transport in densified and composite materials. The temperature dependence of capacity can be described by the Arrhenius equation: $$ C(T) = C_0 \cdot e^{-E_a / (R T)} $$ where $C(T)$ is the capacity at temperature $T$, $C_0$ is a pre-exponential factor, $E_a$ is the activation energy for lithium ion intercalation, and $R$ is the gas constant. Lower $E_a$ values, associated with densified and composite graphites, lead to better low-temperature performance.
| Sample | 0°C | -10°C | -20°C |
|---|---|---|---|
| NG | 95.2 | 82.4 | 65.3 |
| NG-1 | 96.1 | 85.7 | 70.1 |
| NG-2 | 96.8 | 87.3 | 72.5 |
| CG-1 | 97.5 | 89.4 | 74.8 |
| CG-2 | 98.0 | 91.2 | 77.2 |
Long-term cycling stability is a key metric for lithium-ion battery longevity. The cells were cycled at 25°C with a 0.5C charge and 1C discharge for 1000 cycles. The capacity retention and expansion rates are plotted and analyzed. Ordinary natural graphite (NG) shows rapid capacity fade, retaining only 79.1% after 600 cycles, and exhibits high thickness expansion (over 15% after 600 cycles). In contrast, densified natural graphite samples demonstrate significantly improved cycling performance. NG-1 retains 81% after 1000 cycles, and NG-2 retains 82.7%, with expansion rates below 10%. The composites perform even better: CG-1 retains 86.1% and CG-2 retains 87.7% after 1000 cycles, with expansion rates under 8%. The capacity decay follows the power-law model mentioned earlier, with decay coefficients ($\alpha$) decreasing from 0.012 for NG to 0.008 for NG-2 and 0.005 for CG-2, indicating enhanced stability. The expansion behavior can be correlated with the porosity reduction and improved isotropy. The expansion rate ($\Delta L/L_0$) after $n$ cycles can be approximated by: $$ \frac{\Delta L}{L_0} = \gamma \cdot (1 – e^{-\delta n}) $$ where $\gamma$ is the maximum expansion and $\delta$ is a rate constant. Densified and composite graphites show lower $\gamma$ and $\delta$ values, implying slower and less severe swelling.
The enhancement in cycling performance for densified natural graphite is primarily due to the reduction of internal pores, which minimizes stress concentration during lithium intercalation and deintercalation. This prevents particle cracking, reduces electrolyte penetration, and curbs continuous SEI growth, thereby conserving lithium inventory and maintaining capacity. The addition of artificial graphite in composites further boosts performance because artificial graphite particles are inherently dense and exhibit excellent structural stability. Moreover, the blend of natural and artificial graphite balances capacity, rate capability, and cycle life, offering a tailored solution for specific lithium-ion battery applications.
In conclusion, this study demonstrates that densification treatment effectively improves the electrochemical properties of natural graphite for lithium-ion battery anodes. By increasing asphalt addition, internal porosity is reduced, leading to lower impedance, better rate capability, enhanced low-temperature performance, and superior cycling stability. Composites of densified natural graphite with artificial graphite further optimize these attributes, with performance scaling positively with artificial graphite content. These findings provide a pathway for developing high-performance, cost-effective anode materials, enabling next-generation lithium-ion batteries with higher energy density, longer life, and broader operational temperature ranges. Future work could explore alternative densification agents, optimize composite ratios for specific applications, and investigate the long-term aging mechanisms in full lithium-ion battery systems under varied conditions.