In recent years, the development of high-altitude long-endurance unmanned aerial vehicles powered by solar energy has accelerated. These vehicles typically require secondary batteries for energy storage during operation. Lithium-ion batteries, as mature secondary batteries, have been widely used in this field. However, in more demanding environments, such as near-space flight (20–100 km), where temperatures are extremely low, traditional lithium-ion batteries suffer from reduced energy density and cycle life, hindering long-duration flight. While aerospace-specific lithium-ion batteries can mitigate low-temperature concerns, lithium resources are limited and often controlled by foreign countries, potentially constraining the development of near-space vehicles.
Sodium resources, in contrast, are abundant and virtually inexhaustible. With the booming development of sodium-ion battery industries, there is no impending anxiety over sodium resource availability. Sodium-ion batteries offer high power density, excellent low-temperature performance, and enhanced safety, enabling reliable operation in cold environments. For instance, studies have shown that wide-temperature sodium-ion batteries can retain about 70% of their room-temperature capacity even at −70°C. Compared to lithium-ion batteries, sodium-ion batteries demonstrate significant advantages for secondary battery applications in vehicles operating above 20 km altitude. In Japan, sodium-ion batteries have already been selected as backup power sources for rockets and are being demonstrated for high-altitude flight applications. Thus, sodium-ion batteries are strong candidates for specialized batteries paired with solar panels in near-space vehicles.
However, under abusive conditions such as fast charging or overcharging, sodium plating is an inevitable phenomenon in sodium-ion batteries. The deposition of sodium metal on electrode surfaces increases polarization, and in severe cases, sodium dendrites can pierce separators, leading to short circuits and explosions. In flight vehicle power systems, which consist of multiple battery cells, sodium plating in a single cell can pose safety risks to the entire module. Repeated plating events may eventually cause fires or explosions, threatening aerospace missions. Since sodium plating primarily occurs on the anode in sodium-ion batteries, understanding sodium plating on hard carbon anodes is crucial for safe battery operation. Currently, reports on sodium plating in hard carbon anodes are scarce, often focusing only on ex situ characterization of sodium metal signals, with limited attention to cycling performance post-plating and the underlying mechanisms.
In this study, we investigate sodium plating in hard carbon anodes for sodium-ion batteries by establishing a series of plating gradients. Using differential capacity (dQ/dV) curves, combined with scanning electron microscopy (SEM), galvanostatic intermittent titration technique (GITT), and electrochemical impedance spectroscopy (EIS), we monitor sodium plating signals, analyze failure behaviors at different plating levels, and elucidate the plating mechanisms. Our goal is to provide a comprehensive understanding of sodium plating to guide the management and usage of sodium-ion batteries, thereby preventing failures due to plating.
We begin by characterizing the commercial hard carbon used in this study. The pore structure and crystalline properties were analyzed using nitrogen adsorption, Raman spectroscopy, and X-ray diffraction (XRD). The hard carbon exhibits a specific surface area of 4.56 m²/g and a pore volume of 0.0116 cm³/g, with a pseudo-graphitic domain interlayer spacing of 0.38 nm. The Raman spectrum shows a D-band to G-band intensity ratio (I_D/I_G) of 1.10, indicating a disordered carbon structure typical for hard carbons. These structural parameters are summarized in the table below.
| Parameter | Value |
|---|---|
| Specific Surface Area (m²/g) | 4.56 |
| Pore Volume (cm³/g) | 0.0116 |
| Interlayer Spacing (nm) | 0.38 |
| I_D/I_G Ratio | 1.10 |
Electrodes were prepared by mixing commercial hard carbon powder (85 wt%), conductive carbon black (8 wt%), styrene-butadiene rubber (2 wt%), and sodium carboxymethyl cellulose (2 wt%). The slurry was coated onto copper current collectors and dried at 100°C. After cutting and rolling, the hard carbon electrodes had a mass loading of 5.2 mg/cm² and a compaction density of 7.9 g/cm³. Sodium metal sheets served as counter electrodes, glass fiber membranes as separators, and 1 M NaClO₄ in ethylene carbonate/diethyl carbonate (1:1 by volume) as the electrolyte. All cells were assembled in an argon-filled glovebox with H₂O and O₂ levels below 0.1 ppm.
Electrochemical tests were conducted to minimize solid electrolyte interface (SEI) interference. All cells underwent pre-cycling at a current density of 20 mA/g, discharging to 0 V and charging to 2 V, with a 1-minute rest, repeated once. The voltage range was 0–2 V. To explore sodium plating, cells were discharged to specific capacities (300, 350, 400 mAh/g) or voltages (0 V) at 20 mA/g, denoted as C-x (where x is discharge capacity, e.g., C-300, C-350, C-400) and V-y (where y is discharge voltage, e.g., V-0). After discharge, cells were charged to 2 V at the same current density, followed by a 1-minute rest. This cycling was repeated 60 times. All charge-discharge tests were performed on a battery tester. EIS measurements were taken using an electrochemical workstation over a frequency range of 0.01–100,000 Hz.
For electrode characterization, cells were disassembled in the glovebox after the first discharge in pre-cycling. SEM was used to observe morphology changes. Density functional theory (DFT) calculations were performed using VASP 5.4 software with the GGA-PBE functional, PAW pseudopotentials, and DFT-D3 dispersion correction. A 6×6 graphene supercell with lattice constants of 14.7 Å × 14.7 Å was used as the carbon substrate. The formation energy of sodium clusters was calculated using the formula:
$$ E_{\text{form}} = E_{\text{Na-C}} – E_{\text{C}} – n_i \mu_i $$
where \( E_{\text{Na-C}} \) is the total energy of the sodium cluster and carbon substrate, \( E_{\text{C}} \) is the energy of the carbon substrate, \( n_i \) is the number of atoms of element i, and \( \mu_i \) is the chemical potential of element i (i = Na or C).
To monitor sodium plating signals, we analyzed electrochemical curves at different plating levels. Discharge curves from V-0 to C-400 show two plateau regions during sodiation. The first plateau occurs at 0.1–0 V, corresponding to sodium intercalation into hard carbon. The second plateau appears near −30 mV, identified as the voltage inflection point where surface ion concentration peaks. For C-300, C-350, and C-400, the voltage reaches this inflection point. Upon desodiation, compared to V-0, C-300, C-350, and C-400 exhibit an additional desodiation plateau. Differential capacity (dQ/dV) curves reveal two desodiation peaks for these samples: peak “①” at around 0.1 V, associated with sodium extraction from hard carbon, and peak “②” at 0.03 V, which intensifies with deeper plating. This suggests that peak “②” corresponds to sodium metal stripping, enabling in situ monitoring of sodium metal.
SEM images confirm the morphological evolution. At the initial state (2 V), the hard carbon surface is smooth. After sodiation to V-0 (12.5 hours), clusters appear on the surface, indicative of initial sodium nucleation. With further sodiation to C-300 (15 hours), these clusters transform into block-like structures, covering the hard carbon. At C-350 (17.5 hours) and C-400 (20 hours), more blocks form, identified as sodium metal. The progression from clusters to blocks aligns with the enhancement of peak “②” in dQ/dV curves, validating the correlation between peak “②” and sodium metal presence.
Cycling performance was evaluated by comparing cells with different sodiation depths. The second-cycle reversible capacities for V-0.001, V-0, C-300, C-350, and C-400 are 231.4, 248.2, 254, 263, and 274 mAh/g, respectively, showing a slight increase with deeper plating. However, after 60 cycles, C-350 and C-400 experience rapid capacity decay. Capacity retention after 60 cycles is 84.7% for V-0.001, 79% for V-0, 90.2% for C-300, 74.5% for C-350, and 72% for C-400. Initial Coulombic efficiencies drop from 98.96% for V-0.001 to 68.59% for C-400. V-0.001 and V-0 maintain Coulombic efficiencies above 99.5% after the third cycle, while C-300, C-350, and C-400 show efficiencies below 80% at cycle 60. This indicates that sodium clusters (as in V-0) have minimal impact on cycling, but sodium metal blocks (as in C-400) degrade performance due to irreversibility and increased polarization.
Voltage and dQ/dV curves during cycling further illustrate this. For V-0, discharge curves remain stable over cycles, and peak “①” shows slight shifts due to SEI growth. For C-400, discharge curves shift downward with cycling, indicating increased polarization. After 60 cycles, peak “①” disappears, and peak “②” diminishes, reflecting reduced sodium intercalation and stripping capacity due to persistent sodium metal.
Kinetic properties were analyzed using GITT to calculate solid-state sodium ion diffusion coefficients (\( D_{\text{Na}^+} \)). The diffusion coefficient is derived from:
$$ D_{\text{Na}^+} = \frac{4}{\pi \tau} \left( \frac{m_B V_M}{M_B S} \right)^2 \left( \frac{\Delta U_\tau}{\Delta U_s} \right)^2 $$
where \( \tau \) is pulse duration, \( m_B \) is active material mass, \( V_M \) is molar volume, \( M_B \) is molar mass, \( S \) is surface area, \( \Delta U_\tau \) is voltage change during discharge, and \( \Delta U_s \) is voltage change during relaxation. \( D_{\text{Na}^+} \) values are high initially (adsorption behavior), decrease from 90–200 mAh/g (intercalation), rebound at 200–250 mAh/g (filling of closed pores), and drop to near zero beyond 270 mAh/g, indicating saturation. This suggests that after V-0 (12.5 hours), further sodiation beyond 270 mAh/g leads to sodium metal formation on the surface.
EIS results show that interfacial resistance (\( R_{\text{SEI}} \)) and charge transfer resistance (\( R_{\text{ct}} \)) increase with plating depth. At 250 mAh/g (near 0 V), \( R_{\text{SEI}} \) rises due to SEI thickening induced by sodium deposition. For C-300, \( R_{\text{SEI}} \) decreases slightly as plating penetrates the SEI, but then increases again with thicker sodium metal layers, elevating \( R_{\text{ct}} \). Compared to V-0, C-300 shows 14% and 16% increases in \( R_{\text{SEI}} \) and \( R_{\text{ct}} \), respectively. In severe plating, continuous SEI breakdown and repair during cycling exacerbate polarization, accelerating capacity fade.
DFT calculations provide insights into sodium cluster reversibility. Formation energies for Na, Na₄, and Na₁₀ clusters on a carbon substrate are 0.25 eV, 0.8 eV, and 0.99 eV, respectively. Lower formation energies for smaller clusters imply easier desodiation, explaining the reversibility of sodium clusters in V-0. In contrast, larger sodium metal blocks have higher desodiation barriers, leading to poor reversibility.
In summary, our experimental investigation of sodium plating in hard carbon anodes for sodium-ion batteries reveals key findings. First, sodium clusters form after 12.5 hours of sodiation at 20 mA/g, transitioning to sodium metal after an additional 2.5 hours. Sodium metal stripping can be monitored in situ via a differential capacity peak at 0.03 V. Second, sodium clusters have minimal impact on cycling performance due to low formation energies, whereas sodium metal blocks increase interfacial and charge transfer resistances, causing accelerated capacity decay. Third, kinetic analysis confirms that sodium metal forms once hard carbon reaches saturation, with initial plating increasing impedance and facilitating further plating.
The in situ monitoring of sodium metal enables early detection to prevent battery failure under abusive conditions. This study enhances the understanding of sodium plating mechanisms in sodium-ion batteries, offering valuable insights for battery management and safety in applications such as near-space vehicles. Future work could focus on optimizing hard carbon structures or electrolytes to mitigate plating, ensuring reliable performance of sodium-ion batteries in extreme environments.
To further elaborate, the significance of sodium-ion batteries extends beyond aerospace. Their low cost and abundance make them promising for grid storage, electric vehicles, and portable electronics. However, safety remains a paramount concern, and sodium plating poses similar risks as lithium plating in lithium-ion batteries. By studying plating behaviors, we can develop strategies to suppress dendrite growth, such as using electrolyte additives, modifying electrode architectures, or implementing smart charging algorithms.
In our experiments, the choice of 20 mA/g as a low current density minimizes polarization effects, allowing clear observation of plating phenomena. In practical applications, higher currents are common, which may exacerbate plating. Thus, future studies should investigate plating under fast-charging conditions. Additionally, the role of temperature should be explored, as low temperatures can increase plating tendency due to reduced ion mobility.
The table below summarizes the cycling performance data for different sodiation depths after 60 cycles.
| Sample | Reversible Capacity at Cycle 2 (mAh/g) | Capacity at Cycle 60 (mAh/g) | Capacity Retention (%) | Coulombic Efficiency at Cycle 60 (%) |
|---|---|---|---|---|
| V-0.001 | 231.4 | 196.0 | 84.7 | 99.5+ |
| V-0 | 248.2 | 196.0 | 79.0 | 99.5+ |
| C-300 | 254.0 | 229.0 | 90.2 | ~80 |
| C-350 | 263.0 | 196.0 | 74.5 | ~80 |
| C-400 | 274.0 | 196.0 | 72.0 | ~80 |
From a mechanistic perspective, sodium plating on hard carbon involves multiple steps. Initially, sodium ions adsorb on surface sites, then intercalate into pseudo-graphitic domains. Once these domains are filled, sodium ions accumulate at defects or pore openings, forming clusters. With continued sodiation, clusters coalesce into metallic sodium. The differential capacity peak at 0.03 V serves as a fingerprint for this metallic sodium, providing a non-destructive diagnostic tool.
Moreover, the diffusion coefficient data from GITT can be modeled using Fick’s laws to estimate plating thresholds. For instance, when \( D_{\text{Na}^+} \) approaches zero, the flux of sodium ions to the surface exceeds the intercalation rate, leading to plating. This can be expressed as:
$$ J_{\text{Na}^+} = -D_{\text{Na}^+} \frac{\partial C}{\partial x} $$
where \( J_{\text{Na}^+} \) is the ion flux, \( C \) is concentration, and \( x \) is distance. As \( D_{\text{Na}^+} \) decreases, the gradient \( \frac{\partial C}{\partial x} \) increases, promoting surface deposition.
In terms of battery management systems (BMS), our findings suggest that monitoring the differential capacity curve during charging can detect sodium metal formation early. If peak “②” appears, the BMS could reduce charging current or terminate charging to prevent damage. This is particularly relevant for sodium-ion batteries in critical applications like aerospace, where failure is not an option.
To conclude, sodium-ion batteries represent a transformative technology for energy storage, and understanding sodium plating is essential for their safe deployment. Our work provides a framework for studying plating phenomena, combining electrochemical techniques, microscopy, and computational models. By advancing this knowledge, we can accelerate the adoption of sodium-ion batteries across various sectors, contributing to a sustainable energy future.
Future research directions include in situ TEM observation of plating dynamics, development of plating-resistant hard carbons through heteroatom doping or nanostructuring, and exploration of novel electrolytes with high sodium ion transference numbers. Additionally, machine learning could be employed to predict plating based on operational data, enabling proactive battery management.
In the context of near-space vehicles, where weight and reliability are crucial, sodium-ion batteries offer a compelling alternative to lithium-ion systems. Their performance at low temperatures, coupled with inherent safety, makes them ideal for extreme environments. However, as with any battery technology, addressing failure modes like sodium plating is key to unlocking their full potential. Through continued investigation and innovation, we can ensure that sodium-ion batteries meet the rigorous demands of next-generation aerospace applications.
Ultimately, the insights gained from this study not only apply to sodium-ion batteries but also inform broader research on metal plating in electrochemical systems. By deepening our understanding of nucleation, growth, and stripping processes, we pave the way for safer and more efficient energy storage solutions. As the world transitions to renewable energy, batteries will play a central role, and sodium-ion batteries, with their unique advantages, are poised to make a significant impact.