Solvation‐Pore Coupling Governs Fast Na Storage in Hard Carbon Anodes

ABSTRACT

Hard carbon (HC) is widely regarded as the most promising anode for sodium (Na)‐ion batteries (NIBs), yet its fast‐charging capability is intrinsically limited by sluggish Na⁺ insertion and transport within disordered micropores. Here, we elucidate how Na⁺ solvation structures regulate pore‐entry barriers and intra‐pore transport, thereby dictating Na‐storage chemistry and kinetics in HC. Using hard carbon spheres (HCSs) with well‐defined micropore architectures, we show that ether‐based electrolytes enable a micropore‐mediated partial desolvation pathway that accelerates Na⁺ insertion and promotes the formation of quasi‐metallic Na clusters/layers within HC featuring a broad pore‐size distribution. In contrast, carbonate electrolytes induce deeper desolvation and sluggish Na⁺ diffusion, suppressing quasi‐metallic Na formation and severely limiting high‐rate performance. Comprehensive ²³Na MAS solid‐state NMR (ssNMR) reveals a discontinuous, phase‐transition‐like transformation from ionic to quasi‐metallic Na at deep sodiation, driven by enhanced Na‐carbon electronic coupling. These results uncover a solvation‐pore‐coupled kinetic mechanism governing fast Na storage in HC, providing rational design principles for high‐rate and long‐life HC anodes via coordinated regulation of pore structure and interfacial desolvation chemistry.

This work elucidates how Na⁺ solvation governs pore‐entry barriers and intra‐pore transport in HC. Ether electrolytes enable partial desolvation, lower energy barriers, accelerate Na⁺ filling, and promote the formation of quasi‐metallic Na clusters (captured by ^{2 3}Na MAS ssNMR), thereby achieving markedly improved reaction kinetics and high‐rate sodium storage capability in HC.

1. Introduction

Sodium (Na)‐ion batteries (NIBs) are regarded as promising candidates for large‐scale energy storage owing to their natural abundance and cost advantages [1, 2, 3, 4]. However, fast‐charging NIBs remain difficult to commercialize because Na⁺ transport within the most widely adopted hard carbon (HC) anode remains intrinsically sluggish [5, 6, 7]. The complex, disordered structure and high porosity of HC result in slow Na⁺ (de)intercalation, severe polarization, and rapid degraded performance under elevated charging rates [8, 9, 10]. The Na storage mechanism in HC, such as the adsorption‐intercalation/pore‐filling model, has been extensively explored in the past decade [11, 12, 13, 14, 15]. However, it remains challenging to effectively guide the design of fast‐charging HC anodes in practice due to the intrinsic complex structure [16, 17, 18]. Moreover, the outstanding kinetic performance of HC anodes in ether‐based electrolytes cannot be fully explained by the thin, conductive SEI film alone [19, 20, 21, 22]. Therefore, Na⁺ transport and storage behavior within HC micropores is likely a more critical contributing factor [23, 24, 25].

In recent years, research focus has progressively shifted toward the solvation structure of Na⁺ and its coupling effects with HC anodes [26, 27, 28, 29, 30]. For instance, Zhou et al. observed that solvent molecules can penetrate pores at low potentials, influencing Na storage reversibility [31]. Hu et al. further revealed that efficient redistribution of local charges between Na⁺ and carbon layers of wedge nanopores HC supports the origin fast charging in HC materials [32]. These studies suggest that Na⁺ insertion and transport kinetics in HC can be fundamentally governed by both microstructure and dynamic behavior of Na⁺ solvation. However, how solvation‐sheath properties regulate intra‐pore ion transport kinetics and its relationship with irreversible Na loss under high‐rate conditions, remains insufficiently understood [33]. It is therefore imperative to clarify the influence of Na⁺‐solvent structures on Na⁺ transport kinetics in micropores of HC to establish the mechanisms that enable fast and efficient Na⁺ storage in HC anodes [34, 35, 36].

In this work, we elucidate how the solvation structure of Na⁺ regulates its insertion and migration kinetics and Na storage chemistries in micropores of HC. Comparative electrolyte studies reveal that the superior high‐rate performance of HC in ether‐based electrolytes originates from the unique partial‐desolvation pathway that facilitates rapid pore entry while enabling the efficient Na storage as quasi‐metallic Na within micropores across a wide pore size distribution. While slow pore‐entry kinetics of solvent Na⁺ significantly limits fast‐charging performance of HC like in carbonate electrolyte (Figure 1). This work deepens the understanding of the fast‐charging limitations in HC anodes, offering guidance for coupling the microstructural and interfacial (de)solvation chemistries for high‐rate and long‐life NIBs in practical applications.

FIGURE 1.

Diagram of pore‐entry kinetics determined and sodium cluster formation by different solvation structures.

2. Results and Discussion

To clarify the microstructure‐dependent Na⁺ storage kinetics of HC anodes, we synthesized uniform hard carbon spheres (HCSs) from sucrose (Figure 2a and Figure S1), thereby minimizing size and shape heterogeneity and isolating the intrinsic roles of graphitic ordering and pore structure [37, 38]. Controlled carbonization at 1000°C, 1200°C, and 1400°C produced a systematic microstructural gradient. X‐ray diffraction (XRD) patterns exhibit a progressive right‐shift of the (002) peak with temperature, indicating reduced d₀₀₂ and enhanced graphitic stacking (Figure 2b and Figure S2). A continuous decrease in I_D/I_G ratio confirms a decline in defect density and the enhanced graphitized domains with higher temperature (Figure S3). Small‐angle x‐ray scattering (SAXS) further reveals temperature‐driven coalescence of micropores, increasing the average pore size from 8.8 to 13.0 Å (Figure 2d and Figure S4). These tailored HCS samples provide a well‐defined model system to correlate interfacial solvation structures and Na⁺ transport kinetics with the electrochemical performance of HC, thereby offering mechanistic insights into fast‐charging behavior.

FIGURE 2.

(a) Schematic of the material synthesis and scanning electron microscopy (SEM) for HCSs. (b) X‐ray diffraction (XRD) of HCS‐1000, HCS‐1200, HCS‐1400, and HCS‐1500. (c) Integrated intensity ratio of the D‐band to the G‐band (I_D/I_G) for different HCSs from Raman spectra. (d) Pore size distribution of different HCSs characterized by small‐angle x‐ray scattering (SAXS).

To elucidate the intrinsic limitations of HC anodes under fast‐charging, we systematically investigated the Na‐storage kinetics of HCSs with well‐defined microstructure. Rate performance tests in Figure 3a and Figure S5 reveal a pronounced kinetic advantage of the ether‐based electrolyte (1 m NaPF₆‐DEGDME) over the carbonate counterpart (1 m NaPF₆‐EC/DEC), in agreement with previous reports [19, 27]. Notably, the HCS‐1200 with 10.6 Å pores still delivers 257 and 225 mAh g⁻¹ at 5 C and 10 C, respectively, in the ether electrolyte, whereas the capacity in the carbonate system declines to nearly zero (Figure 3b). The C‐rate‐dependent retentions of the slope and plateau capacities are summarized in Figure 3c and Figure S6. Surprisingly, in the ether electrolyte, the plateau capacity retention remains nearly constant up to 5 C and even exceeds that of the slope region, indicating highly robust plateau‐region kinetics. In contrast, the carbonate electrolyte exhibits a much more rapid decay of both slope and plateau capacities at elevated C rates [39]. The markedly larger disparity in plateau‐region capacity retention with increasing C rates highlights a strongly electrolyte‐dependent Na‐storage mechanism in this region. Moreover, this electrolyte‐dependent behavior is largely insensitive to HCS micropore size (Figure 3d,e); only samples with intermediate micropores (∼1.06 nm) show a slightly slower decay than those with smaller (0.88 nm) or larger (1.2 nm) pores in the ether‐based electrolyte.

FIGURE 3.

(a) Rate performance, (b) Charge‐discharge curves, and (c) capacity degradation behaviors across different regions of HCS‐1200 in 1 m NaPF₆‐DEGDME and 1 m NaPF₆‐EC/DEC (HC||Na half‐cells). (d) Capacity retention at 5 C and (e) rate performance of HCS with different pore sizes in 1 m NaPF₆‐DEGDME and 1 m NaPF₆‐EC/DEC electrolytes, respectively.

Previous studies typically attribute the superior rate‐performance of HC‐based anodes in ether electrolytes to the formation of thinner and more conductive SEI layers [22, 23]. However, the SEI‐exchange experiments challenge this interpretation (Figure S7). The results imply that the pore‐filling kinetics and chemical status of Na stored in HC could be fundamentally different in different electrolytes, thus governing the fast‐charging behavior and stability of HC anodes.

We employed NMR to probe the desolvation behavior of Na⁺ mediated by the micropores of HC in different electrolytes. In the bulk electrolytes, the sharp ²³Na resonance at −8.30 ppm corresponds to bulk solvated Na⁺ [21, 40]. Upon mixing HC with the ether‐based electrolyte, the NMR signal broadens significantly, which can be attributed to the restricted Na⁺ diffusion under micropore confinement of HC, implying the occurrence of interfacial desolvation of Na⁺ (Inset of Figure 4a). Under 15 kHz rotating, the ²³Na MAS ssNMR spectrum exhibits two main characteristic changes: (i) peak narrowing due to the averaging of anisotropic interactions from confined or partially desolvated Na⁺, and (ii) an up‐field shift that can be associated with a locally higher Na⁺ concentration due to the partial desolvation effect at the microporous HC material surface. ​The ²³Na NMR peaks in the carbonate electrolyte show similar changes but with markedly weaker signal intensity (Figure 4b), suggesting that carbonate solvent molecules are less prone to penetrate into the HC micropores, and instead undergo a more complete desolvation at the HC surface.

FIGURE 4.

Schematic diagram of the electrolyte systems with corresponding inserted ²³Na NMR spectra, showing bulk liquid electrolytes and HC‐electrolytes in (a) 1 m NaPF₆‐DEGDME and (b) 1 m NaPF₆‐EC/DEC. FT‐IR spectra of electrolytes with different NaPF₆ salt concentrations and (c) NaPF₆‐DEGDME, (d) NaPF₆‐EC/DEC. (e) Na⁺ diffusion coefficients calculated from the GITT profiles.

Infrared (IR) spectroscopy provides further evidence for interfacial desolvation of Na⁺ on HC. The CH₂ wagging vibration of DEGDME (840 cm⁻¹) typically red‐shifts with increasing Na salt concentration [31]. After mixing HC with the ether electrolyte with conventional 1 molar concentration, the CH₂ wagging vibration band of DEGDME shifts to a comparable wavenumber to the electrolyte with 2 molar salt (Figure 4c). This result is consistent with the partial desolvation behavior at the HC surface discussed above. A similar trend is observed for the carbonate electrolyte (Figure 4d), but with a larger redshift, implying a more dominant and complete interfacial desolvation process. Collectively, the spectroscopic results reveal that ether‐based electrolytes preferentially facilitate partial desolvation and subsequent Na⁺ insertion into the HC micropores.

These distinct interfacial desolvation behaviors result in pronounced differences in bulk Na⁺ diffusion kinetics. As shown by the GITT measurements in Figure 4e and Figure S8, the partial desolvation induced by ether molecules within HC micropores and on the HC surface significantly enhances Na⁺ diffusion in both the sloping and plateau voltage regions. In particular, the diffusion coefficient in the plateau region is improved by more than 2–5 times than that in carbonate system, accounting for the observed remarkable high‐rate capability. The partially desolvated Na⁺‐DEGDME species diffuse more favorably within confined micropores of HC material, enabling high reversible capacity retention even under fast‐charge conditions.

The above results illustrate how micropore‐induced desolvation governs both Na⁺ pore‐filling kinetics and reversible storage capacity. This effect is basically insensitive to pore size within the 0.8–1.1 nm range; however, for pores larger than 1.2 nm, the weakened desolvation reduces Na⁺ insertion efficiency and rate performance as discussed in Figure 3d. Consequently, the plateau storage behavior is determined by the coupling of Na⁺ solvation and pore‐entry kinetics at HC. These findings highlight that optimizing the interplay between HC micropore structure and interfacial solvation behavior is essential for achieving high‐performance HC anodes.

To clarify how pore‐entry barriers and diffusion kinetics regulate Na⁺ storage in HC micropores, ²³Na MAS ssNMR was employed to probe the local coordination and electronic states of stored Na. As shown in Figure S9, two distinct classes of ²³Na signals are observed. An antiferromagnetic doublet between −20 and 20 ppm is assigned to surface‐related Na species, where the −9.4 ppm peak originates from residual salts and the ∼6.0 ppm component mainly corresponds to SEI‐associated Na [21, 40, 41, 42]. While the broad resonance at 700–1000 ppm with a pronounced Knight shift indicates strong hyperfine coupling, characteristic of quasi‐metallic Na confined within HC micropores [43].

Using HCS‐1200 and HCS‐1400 as model systems, we compared Na⁺ environments under fully sodiated states (5 mV) across different electrolytes. Despite similar Na storage capacities (∼300 mAh g^{− 1} at 0.1 C), the quasi‐metallic Na feature in HC differs sharply. For HCS‐1200, the ether electrolyte yields a moderate Knight shift (802 ppm) in Figure 5a, whereas the carbonate electrolyte shows only a weak quasi‐metallic signal (774 ppm). Notably, enlarging the micropore size significantly amplifies the Knight‐shifted resonance in the ether system, yielding a sharper line‐shape and a shift approaching that of metallic Na of 1134 ppm (Figure 5b) [40]. This trend signifies more extensive Na clustering or layer‐like Na accumulation within micropores, enabled by enhanced Na⁺ mobility.

FIGURE 5.

²³Na MAS ssNMR spectra of the quasi‐metallic Na cluster region of (a) sodiated HCS‐1200 and (b) sodiated HCS‐1400. (c) ²³Na MAS ssNMR spectra of HCS‐1400 at different sodiation stages.

These observations suggest that fast Na⁺ transport in ether‐based electrolytes promotes cooperative Na‐Na interactions and electronic coupling between confined Na species and the carbon framework. Such coupling facilitates electron delocalization at the Na‐carbon interface of HC anode, thereby lowering charge‐transfer barriers and supporting rapid Na‐storage kinetics under fast‐charging conditions [32]. In contrast, sluggish Na⁺ diffusion in carbonate electrolytes kinetically inhibits the formation of quasi‐metallic Na. Consistently, a pronounced enhancement of the Knight‐shifted signal is observed only after resting fully sodiated HC electrodes at elevated temperature (Figure S12), indicating that quasi‐metallic Na formation in carbonate systems is diffusion‐limited at room temperature. This kinetic bottleneck fundamentally constrains the high‐rate Na‐storage capability of HC in carbonate electrolytes.

Finally, the evolution of quasi‐metallic Na during sodiation was tracked by ²³Na MAS ssNMR (Figure 5c). The Knight‐shifted resonance emerges abruptly only at deep discharge (≤5 mV), rather than increasing continuously with Na content. This discontinuous behavior indicates a phase‐transition‐like transformation from ionic to quasi‐metallic Na within HC micropores, likely driven by a critical threshold of Na concentration and electronic coupling with the carbon matrix. Collectively, these results highlight that regulating Na⁺ mobility and Na‐carbon electronic interactions is pivotal for activating quasi‐metallic Na storage, providing a mechanistic foundation for designing HC anodes with simultaneously high capacity and fast‐charging capability.

3. Conclusion

Combining electrochemical and comprehensive spectroscopic analysis, this work reveals that the fast‐charging performance of HC anodes is intrinsically governed by the interplay between Na⁺ solvation structures and HC micropore chemistry, which together determine the partial‐desolvation barrier at micropore entrances and the ensuing intrapore Na⁺ transport and ionic‐to‐quasi‐metallic Na evolution that ultimately dictates fast‐charging kinetics. Ether‐based electrolytes favor micropore mediated partial desolvation at pore entrances of HC, accelerate Na⁺ diffusion within confined pores, and promote the formation of quasi‐metallic Na clusters/layer that support rapid charge transfer and Na storage kinetics via a hyperconjugation‐like electron delocalization effect. Carbonate electrolytes, in contrast, hinder Na⁺ migration and suppress cluster formation, which thereby restricts the enhancement of rate‐performance of HC anodes. The identification of a phase‐transition‐like transition from ionic to quasi‐metallic Na status provides a new mechanistic framework for understanding plateau storage chemistries and kinetics behavior in HC. Collectively, these findings highlight that the rational coupling of electrolyte solvation engineering with micropore design is essential for advancing practical fast‐charging NIBs batteries for practical applications.

4. Experimental Section

4.1. Synthesis of HC Anode Materials

HCS was synthesized via the hydrothermal and pyrolysis methods. A 2 m sucrose aqueous solution (obtained from Alfa) was positioned in a Teflon‐lined stainless‐steel autoclave. The hydrothermal carbonization was executed at a temperature of 200°C for 12 h followed by repeated washing processes with dilute hydrochloric acid and deionized water. Further high‐temperature treatment was carried out at 1000°C, 1200°C, and 1400°C for 2 h in a tube furnace under argon flow. The ground samples were abbreviated as HCS‐1000, HCS‐1200, and HCS‐1400, respectively.

4.2. Electrode Preparation

The hard carbon (HC) anode was prepared from HC (90 wt.%), Super P (5 wt.%), and polyvinylidene fluoride (PVDF, 5 wt.%) in N‐methyl‐2‐pyrrolidone (NMP), coated on Cu foil, and calendared. Circular electrodes (ϕ12 mm) with a mass loading of ∼1.2 mg cm⁻² were punched for cell assembly. CR2032 coin cells were assembled using metal Na (ϕ15 mm), HC anodes (ϕ12 mm), glass fiber separators (ϕ19 mm), and 150 µL of 1 m NaPF₆‐DEGDME and 1 m NaPF6‐EC/DEC (V:V = 1:1). All coin cells were assembled in an argon‐filled glove box.

4.3. Electrochemical Methods

Galvanostatic charge‐discharge measurements and the Galvanostatic intermittent titration technique (GITT) was performed on coin cells using a LAND CT2001 battery tester. The HC electrode was tested using a current pulse of 0.05 C for 30 min, followed by a relaxation period of 2 h for each step. The HC||Na coin cells were cycled in the voltage range of 0.005–2 V at 25°C.

4.4. Characterizations

The morphology of the HCS was characterized using scanning electron microscopy (SEM, SU8010). The crystallographic information was characterized by x‐ray diffractometer (XRD, Ultima IV, Rigaku) with a Cu Ka x‐ray source. Atomic‐resolution high angle annular dark field scanning transmission electron microscopy (HAADF‐STEM) images were obtained at 200 kV using a JEOL‐ARM200F microscope equipped with a probe spherical aberration corrector. Before these tests, all electrode samples were washed with dimethyl carbonate (DMC) and thoroughly dried under vacuum. Solid‐state NMR experiments were conducted on Bruker Avance III HD 400 MHz spectrometers equipped with 3.2 mm magic‐angle spinning probes at a spinning rate of 15 kHz. The ²³Na MAS ssNMR measurements were performed at 9.4 T, corresponding to a resonance frequency of 105.8 MHz. The spectra were collected using a Hahn echo pulse sequence (90°‐τ‐180°‐τ‐acq) with τ = 66.7 µs. All sample preparation was carried out in a glove box. The ²³Na chemical shifts were referenced to 1 m NaCl aqueous solution at 0 ppm. Peak deconvolution was carried out using Topspin software. SAXS measurements were carried out on Xenocs Xeuss SAXS/WAXS System. The Fourier transform infrared (FT‐IR) spectra were recorded using an FT‐IR spectrometer (IS50, Nicolet iS10, Thermo Fisher Scientific).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll74104‐sup‐0001‐SuppMat.docx.

Data Availability Statement

Research data are not shared.