Pore-Structure Modulation of Macadamia Shell-Derived Hard Carbon for High-Performance Sodium-Ion Battery Anodes

Abstract

Hard carbon is widely recognized as a viable anode candidate for sodium-ion batteries (SIBs) owing to its electrochemical advantages, yet simultaneously enhancing specific capacity and rate capability, arising from insufficient plateau capacity, remains a long-standing challenge. Herein, we present a strategy for fabricating ZnCl₂-modified hard carbon (HCMZ-X) using waste macadamia shells and ZnCl₂ as a multifunctional structural modifier through a facile high-temperature carbonization. This approach effectively expands the graphite interlayer spacing to 0.394 nm, reduces microcrystalline size, and induces abundant closed pores, synergistically improving sodium-ion storage kinetics within the hard carbon framework. Mechanistic investigations confirm an “adsorption-intercalation-filling” storage mechanism. Hence, the optimized HCMZ-3 delivers a high reversible capacity of 382.05 mAh g⁻¹ at 0.05 A g⁻¹, with the plateau region contributing approximately 70%, significantly outperforming that of unmodified hard carbon (262.64 mAh g⁻¹). Remarkably, it achieves stable rate performance, delivering 190 mAh g⁻¹ at 1 A g⁻¹, along with excellent cycling stability, retaining over 90% after 500 cycles. By rational pore-structure modulation rather than excessive surface activation, this cost-effective method utilizing agricultural waste and ZnCl₂ dual-functional modification partially alleviates the intrinsic energy-density limitation of hard carbon anodes, advancing the development of high-performance, eco-friendly anodes for scalable energy storage systems.

1. Introduction

The escalating demand for sustainable and cost-effective energy storage solutions has propelled significant attempts at alternatives to conventional lithium-ion batteries (LIBs) [1,2,3]. Owing to the natural abundance and low cost of sodium resources, sodium-ion batteries (SIBs) demonstrate potential as a highly compelling candidate. Sharing similar electrochemical working principles with LIBs, SIBs hold substantial potential for application in large-scale energy storage and electric devices [4,5,6]. However, a critical bottleneck hindering their widespread commercialization is the development of high-performance anode materials capable of efficiently accommodating the larger ionic radius of Na⁺ compared to Li⁺ [7,8,9].

Hard carbon (HC), classified as a non-graphitizable carbon material, is commonly considered an attractive candidate for SIBs because of its favorable structural characteristics and cost advantages. Typically, HC exhibits a highly disordered microstructure with enlarged interlayer spacing (d₀₀₂ > 0.40 nm), which is beneficial for Na⁺ intercalation and storage [10,11,12]. As an example, Zhou et al. [13] reported coal-derived HC by eliminating SiO₂ impurities during carbonization, which demonstrated an outstanding Na⁺ storage capacity of 341 mAh g⁻¹ along with an initial Coulombic efficiency of 85%. Moreover, HC can be synthesized from inexpensive and readily available precursors, including biomass residues, resins [14], and pitch [15], through relatively simple pyrolysis processes. Biomass-derived HC, in particular, offers additional advantages, including environmental benignity, tunable porosity, and alignment with waste valorization and carbon neutrality goals [16]. Coconut shell, prized for its low impurity and robust structure, has been a primary industrial precursor, yielding HC with good structural stability and electrochemical performance under optimized carbonization conditions [17]. However, the large-scale application of coconut-shell-based HC is restricted by raw material supply limitations. Consequently, research has expanded to include alternative shell biomass products like walnut shells [18], hazelnut shells [19], peanut shells [20], and olive shells [21], demonstrating stable cycling performance after suitable structural regulation. Other agricultural waste such as corncob [22] and bamboo [23] have also shown promising performance. Importantly, carbonization temperature plays a decisive role in the electrochemical properties of the resulting hard carbon anodes, with optimal capacities over 230 mAh g⁻¹ generally achieved at 1300 °C [24]. These results collectively highlight the broad potential of HC derived from diverse biomass precursors in developing high-performance and sustainable anodes for SIBs.

Although the variety of biomass-derived HC applications has been continuously broadened, their electrochemical performance still requires further enhancement. Structural optimization of HC is crucial for improving electrochemical metrics, including initial Coulombic efficiency (ICE), reversible capacity, and rate performance. To date, research indicates that systematically modulating the pore structure (particularly within the 0.3 to 2 nm size range) facilitates additional active Na⁺ storage sites within the HC, thereby enhancing the reversible capacity and kinetics of SIBs [25]. Hence, various modification approaches have been investigated. Heteroatom doping (such as N, S, P, or O) enhances sodium-ion storage by introducing additional active sites, improving electronic conductivity, optimizing defect structures, and expanding the interlayer spacing [26,27]. Studies have demonstrated that tailored heteroatom doping (such as N or O), achieved either through precise control of carbonization temperature or via emulsion-based synthetic design, effectively enhances the capacitive performance of carbon materials by synergistically combining high surface area with electrochemically active sites [28,29]. Recently, zinc-based modification has shown remarkable promise. For instance, Zn-doped HC synthesized via Zn(AC)₂ polymerization exhibited expanded graphitic domains (d₀₀₂ = 0.408 nm) and optimized nanopores [30]. Recent work has demonstrated that ZnBr₂-assisted biomass fractionation can efficiently separate lignocellulose components, achieving an 88.0% yield of glucan oligomers and 92.4% lignin purity [31]. The resulting lignin, enriched with polar functional groups, exhibited excellent electrochemical performance as HC anodes in sodium-ion batteries, delivering a high specific capacity of 358.1 mAh g⁻¹ and an ICE of 88.4%. Wen et al. [32] reported that zinc salt-templated synthesis enables the formation of size-controlled ZnO microcrystals along with abundant closed pores in HC, enabling reversible capacities above 340 mAh g⁻¹ and ICE exceeding 90%. These structural enhancements synergistically boost Na⁺ storage kinetics and capacity. Despite these advances, the sodium storage capacity and rate performance of lower-cost biomass-derived HC still require substantial improvement to satisfy the requirements of practical applications.

Herein, we utilize waste macadamia shells, an abundant agricultural byproduct, as the precursor material and ZnCl₂ as a cost-effective and efficient structural modifier. Through a straightforward mixing and high-temperature carbonization procedure, the zinc-modified hard carbon (denoted as HCMZ-X) was synthesized. Systematic characterization confirms that ZnCl₂ introduction successfully expands the graphite interlayer spacing (up to 0.394 nm), reduces microcrystalline size, and generates abundant closed pores. These synergistic structural modifications lead to significantly optimized sodium storage performance. The optimal sample (HCMZ-3) delivers a high reversible capacity of 382.05 mAh g⁻¹ at 0.05 A g⁻¹, with an impressive plateau capacity contribution of 262.64 mAh g⁻¹, substantially outperforming the unmodified counterpart. Furthermore, the modified sample exhibits outstanding rate capability, maintaining 190 mAh g⁻¹ at 1.0 A g⁻¹, along with exceptional long-term cycling stability over 1000 cycles at the same current density. Mechanistic investigations, including electrochemical analysis and post-mortem characterization, robustly confirm that the sodium storage behavior in ZnCl₂-modified HC follows the “adsorption-intercalation-filling” model. This work not only presents a sustainable and economical strategy for synthesizing high-performance HC anodes from underutilized biomass waste but also provides fundamental insights into structure–property relationships and sodium storage mechanisms, laying a solid foundation for their application in sustainable SIBs.

2. Materials and Methods

2.1. Synthesis of Hard Carbon

Macadamia nutshells (MSs) (sourced from Yunnan Province, China) were pulverized and then acid-washed with 18 vol% hydrochloric acid (HCl) aqueous solution for 24 h to remove inorganic impurities and soluble organic components. The residue was filtered, rinsed with distilled water to neutrality, and dried at 80 °C for 24 h to yield a dark brown powder. This powder was subsequently blended with ZnCl₂ (98.0%) at mass ratios of 100:1, 100:3, and 100:5. The mixtures were transferred to a graphite crucible and carbonized in a tube furnace under an argon atmosphere. The temperature was increased at a rate of 5 °C min⁻¹ until reaching 1300 °C, where it was held constant for 2 h. The resulting hard carbon products were labeled HCMZ-X, where X denotes the ZnCl₂ content. For comparison, parallel experiments were carried out in the absence of ZnCl₂, and the resulting material was named HCMZ.

2.2. Materials Characterization

The pyrolytic behavior of the precursor was investigated using thermogravimetric analysis (TGA, TGA2, METTLER TOLEDO, Shanghai, China) at a heating rate of 10 °C min⁻¹. The surface morphology of the samples was examined via field emission scanning electron microscopy (SEM, Apreo 2, Thermo Fisher Scientific, Shanghai, China), while their detailed internal structures were characterized using high-resolution transmission electron microscopy (HR-TEM, Talos F200S, Thermo Fisher Scientific, Shanghai, China). Phase composition was studied via X-ray diffraction measurements (XRD, λ = 0.15418 nm) at 2θ of 10–60°, along with Fourier transform infrared spectroscopy (FTIR, Nicolet iS50, Thermo Fisher Scientific, Shanghai, China). Defect features were detected via Raman spectroscopy (Raman, DXRxi spectrometer, Thermo Fisher Scientific, Shanghai, China). Surface chemical composition was determined by X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo Fisher Scientific, Shanghai, China). The specific surface area and pore structure were determined from N₂ adsorption–desorption testing using a Brunauer–Emmett–Teller (BET, ASAP 2460, Micromeritics, Norcross, GA, USA) analyzer, along with the sample’s porosity.

2.3. Electrochemical Tests

HCMZ-X was coated with super P and polytetrafluoroethylene (PVDF) in a weight ratio of 8:1:1, then uniformly cast onto copper foil. The resulting electrodes were dried in a vacuum oven at 80 °C for 12 h to prepare anode sheets. CR2032 coin cells were assembled using sodium metal as the counter electrode and glass fiber as the separator, with 1 M NaPF₆ in diethyl carbonate (DEC) as the electrolyte (Figure S1). All cell assembly procedures were carried out in an argon-filled glovebox, where the concentrations of oxygen and moisture were maintained below 0.1 ppm. Galvanostatic charge–discharge (GCD) tests were conducted on a Neware battery testing system within a potential window of 0.01–2.0 V. Cyclic voltammetry (CV) and electrochemical impedance (EIS) spectroscopy measurements were performed using a CHI660E workstation, with CV scan rates varying from 0.1 to 1 mV s⁻¹ and an AC impedance frequency range of 0.01–10⁵ Hz. For full-cell evaluations, sodium-ion cells were constructed using Na₃V₂(PO₄)₃ (NVP) as the cathode and the HCMZ-X as the anode (Figure S1). The NVP half-cell was tested between 2.2 and 3.8 V (vs. Na⁺/Na), while the electrochemical performance of the full cell was assessed over a voltage range of 1.5–3.8 V.

3. Results and Discussion

3.1. Morphological and Structural Properties

MSs are abundant agriculture waste. The conventional synthesis of typical ZnCl₂-modified hard carbon (i.e., HCMZ-X) derived from MSs is schematically illustrated in Figure 1a. Initially the MS shells were pulverized into powder and subjected to HCl acid treatment to eliminate inorganic impurities. Subsequently, the resulting precursors were blended with ZnCl₂ and thermally treated at 1300 °C under an argon atmosphere, yielding a black hard carbon product. During heating, ZnCl₂ melts to form ZnO, which contributes to the modulation of the carbon framework [33]. As temperature rises, ZnO reacts with carbon [34], as in the following Equation (1):

$$Zn{Cl}_2+C=Zn\uparrow +CO\uparrow$$

(1)

and the resulting products evaporate. This process optimizes the pore structure of the carbon framework, thereby enhancing sodium-ion storage characteristics. For comparison, the pristine material (HCMZ) was prepared by carbonizing the acid-treated MS precursor under identical conditions but without the incorporation of ZnCl₂.

Figure 1. (a) Schematic representation of the preparation routes for HCMZ and HCMZ-X; SEM images of (b) HCMZ and (e) HCMZ-3; TEM images of (c) HCMZ and (f) HCMZ-3; HR-TEM images of (d) HCMZ and (g) HCMZ-3 (inset: corresponding SAED patterns); (h) XRD patterns of HCMZ-X and HCMZ; (i) Interlayer distance of all samples based on XRD calculations; (j) FITR spectra of HCMZ-X and HCMZ.

Thermogravimetric analysis (TGA) was employed to investigate the thermal decomposition processes of different precursors (Figure S2). A slight weight loss occurs around 80 °C, consistent with the evaporation of physically adsorbed moisture form the surface of the MS precursor [35]. Subsequently, a substantial weight reduction is observed around 350 °C, mainly associated with the decomposition of cellulose and hemicellulose in the MSs, along with the thermal decomposition of polymer chains [36]. Additionally, lignin decomposes into smaller molecular compounds such as eugenol due to the cleavage of ether bonds between 150 and 400 °C [18]. Furthermore, the carbon yield of HCMZ-3 after high-temperature carbonization at 1300 °C is around 30%, exceeding that of the unmodified HCMZ sample (27%). Scanning electron microscopy (SEM) observations reveal that HCMZ-3 exhibits slightly smaller micro-dimensions compared with HCMZ, though both of the samples display distinct irregular shapes (Figure 1b,e). Moreover, HR-TEM analysis depicted in Figure 1f indicates a cross-linked carbon layer arrangement in HCMZ-3, implying that ZnCl₂ incorporation alters the carbon framework compared with HCMZ (Figure 1c). As illustrated in Figure 1d,g, the HC anode materials obtained through high-temperature carbonization of the precursors exhibit a pseudo-graphite structure, which is defined as locally ordered sp² carbon domains with graphitic interlayer spacing but lacking long-range crystallinity. Moreover, based on the selected area electron diffraction (SAED) patterns of the two samples, it was determined that HCMZ-3 possesses a wider interlayer spacing compared to HCMZ. Specifically, HCMZ-3 contains more closed pores in contrast to HCMZ.

The X-ray diffraction (XRD) reveals the crystalline structure of HCMZ and HCMZ-X. As demonstrated in Figure 1h, XRD profiles of all samples display distinctive diffraction peaks at approximately 23° and 43°, which can be indexed to the (002) and (100) crystallographic planes of graphite carbon, characteristic of hard carbon structures. A systematic shift of the (002) diffraction peaks to lower angles from 22.66° to 22.47° was observed in HCMZ-X samples with increasing ZnCl₂ precursor content, compared to the sharp peak at 22.68° for HCMZ. This broadening phenomenon, combined with the absence of distinct diffraction peaks, indicates reduced crystallinity and expanded interlayer spacing in HCMZ-X, which can be attributed to structural disorders induced by ZnCl₂ incorporation. The increased interlayer spacing significantly enhances Na⁺ intercalation kinetics and ionic transport efficiency. According to the Bragg equation, the layer spacing (d₀₀₂) of HCMZ-3 is 0.394 nm, whereas the layer spacing of HCMZ without ZnCl₂ is only 0.391 nm (Figure 1i). Furthermore, an increase in ZnCl₂ content facilitates the enlargement of the interlayer spacing. Based on these diffraction features, the average graphite domain length $\left(La\right)$ and thickness $\left(Lc\right)$ using the Scherrer relationship is given in Equation (2) [37]:

$$L\left(nm\right)=k\lambda /\beta \cos \theta$$

(2)

where $\lambda$ denotes the wavelength of the X-ray radiation (0.15418 nm), $\beta$ indicates the full width at half maximum of the (100) and (002) diffraction peaks, and $\theta$ is the position of the corresponding peaks. $K$ is the Scherrer’s value, which is set to 1.84 and 0.89 [38]. The results (Figure S3) show that the $La$ and $Lc$ values of HCMZ-3 are the smallest, indicating that the addition of ZnCl₂ effectively limits graphitization. As shown in Figure 1j, functional groups in different samples were analyzed via FITR spectroscopy. The absorption band observed at 3500 cm⁻¹ is assigned to -OH functional groups within the materials [38,39]. The absorption band located near 1600 cm⁻¹ is assigned to C=O vibrations, whereas the feature appearing around 780 cm⁻¹ is mainly related to C-H functional groups associated with aromatic rings [40]. Additionally, a slight C-N stretching vibration is observed at 1320 cm⁻¹. The FTIR spectra of all samples present similar characteristic bands, indicating that the fundamental functional groups are preserved after carbonization. However, noticeable differences in peak intensity and transmittance are observed with increasing ZnCl₂ content, suggesting variations in the relative abundance and chemical environment of these functional groups. This variation is likely associated with ZnCl₂-assisted dehydration and condensation reactions during carbonization, which regulate surface chemistry rather than completely removing functional groups [41].

Raman spectroscopy (Figure 2a) analysis reveals the presence of distinct D and G bands in the samples, which are attributed to disordered defect and graphite-like sp²-bonded carbon domains, respectively [39]. The ratio of I_D/I_G is indicative of the defect density. The significantly elevated I_D/I_G values of three HCMZ-X samples compared to HCMZ demonstrate enhanced surface defect density and structural disorder induced by ZnCl₂ incorporation. Compared to HCMZ, HCMZ-X exhibits lower I_D/I_G values, indicating a higher degree of sp² carbon ordering, consistent with HR-TEM results. The Raman spectra could be deconvoluted into four peaks of D₄ peaks, D₁ peaks, D₃ peaks, and G peaks (Figure S4) [38]. Notably, the D₃ peak area (A_D3) of HCMZ-3 is larger, suggesting an increase in sp³ carbon within the disordered structure. This indicates an increase in interlayer spacing, being consistent with the XRD results [38].

Figure 2. (a) Raman spectra of HCMZ-X and HCMZ; (b) XPS spectra of HCMZ-X and HCMZ; (c) N₂ adsorption–desorption isotherms; (d) C1s spectra of samples; (e) O1s spectra of individual samples; (f) corresponding pore size distribution calculated from N₂ sorption isotherms; (g) BET specific surface area and pore volume of HCMZ and HCMZ-X.

The surface element composition was analyzed by X-ray photoelectron spectroscopy (XPS). As illustrated in Figure 2b, all spectra display distinct signals corresponding to C1s at 284.80 eV and O1s at 532.40 eV. Importantly, HCMZ-X do not exhibit significant Zn-related signals, suggesting that there are no Zn residues in the samples during carbonization at 1300 °C. The C1s peaks can be deconvoluted into four peaks at 284.55 eV, 284.91 eV, 286.40 eV, and 289.72 eV, which are attributed to sp²C, sp³C, C-O, and C=O, respectively (Figure 2d) [38]. The XPS fitting results (Table S3 and Figure S5a) show that HCMZ exhibits the highest sp²C content (44.51%), indicating a predominantly graphite-like structure. In contrast, the Zn-modified samples exhibit an increased fraction of sp³ carbon, with HCMZ-3 showing a maximum sp³C content of 37.68%. Combined with the HRTEM, XRD, and Raman results, these findings suggest that ZnCl₂ introduction generates abundant structure defects within the carbon layers and promotes the transformation of sp²C to sp³C [38,42]. Additionally, high-resolution O1s XPS spectrum (Figure 2e) reveals two distinct components centered at 531.8 eV and 533.2 eV with a binding energy separation of 1.4 eV, which are characteristic of oxygen-containing functional groups such as C-O and C=O in polymeric carbonyl structures. The results verify that, in HCMZ and HCMZ-X, oxygen is predominantly present in the form of C–O and C=O functional groups. Furthermore, incorporation of ZnCl₂ leads to a pronounced increase in the C=O component contribution in HCMZ-X compared to HCMZ, with the fitted C=O content of HCMZ-X increasing sequentially from 49.80% to 72.17% (Figure S5b).

Pore structure features were further characterized by N₂ adsorption–desorption measurements (Figure 2c,f,g) to reveal the specific surface areas and pore size distributions of samples. The specific data based on Brunauer–Emmett–Teller (BET) model measurements are illustrated in Table S4. As shown in Figure 2c, both HCMZ and HCMZ-X display type IV isotherms with a distinct H4 hysteresis loop (0.1 < P/P₀ < 1), indicative of a pore structure comprising both micropores and mesopores [38]. Specifically, both the BET surface area and total pore volume of the HCMZ-X samples are markedly reduced compared with those of HCMZ. Among them, HCMZ-3 exhibits the lowest values, with a specific surface area of 31.18 m² g⁻¹ and a pore volume of 0.0093 cm³ g⁻¹ (Figure 2g). Specifically, ZnCl₂, as an activator, alters the carbon structure through dehydration condensation and the melt template effect, promotes the formation of a denser carbon framework, and suppresses the development of excessive open micropores [43,44]. Therefore, the BET surface area varies nonlinearly with the content of ZnCl₂. In addition, the pore size distribution (Figure 2f) reveals a significant decrease in open pores, further indicating that ZnCl₂ modification introduces more disordered microcrystalline structures. This facilitates the conversion of accessible mesopores into closed pores, which in turn leads to a decrease in the surface area [38,45]. The reduced specific surface area provides increased reversible sodium-ion storage sites, leading to potential enhancement of the electrochemical Na⁺ storage performance.

3.2. Electrochemical Properties

To investigate and compare the sodium storage performance of HCMZ and HCMZ-X, half coin cells were assembled for electrochemical performance measurements employing metallic sodium as the counter electrode. The first three charge/discharge profiles of the four electrodes are presented in Figure 3a and Figure S6. The results indicate that, under a current density of 0.05 A g⁻¹, HCMZ-3 delivers a maximum reversible capacity of 382.05 mAh g⁻¹ with an ICE of 70%, whereas HCMZ shows a minimum value of 220.67 mAh g⁻¹ and an ICE of 59.1%. Specifically, the second charge/discharge profiles can be separated into two characteristic regions: the plateau region (<0.1 V), which is associated with Na⁺ intercalation between graphitic layers as well as the occupation of closed pore structures, and the slope region (>0.1 V) related to Na⁺ adsorption on the electrode surface. It is evident that the introduction of ZnCl₂ predominantly enhances the capacity associated with the low-voltage plateau region. In particular, the high specific capacity of HCMZ-3 originates mainly from the pronounced expansion of this plateau contribution. Relative to HCMZ, the plateau capacity of HCMZ-3 increases from 124.56 to 262.64 mAh g⁻¹, with a 12.26% increase in plateau region contribution. This enhancement is attributed to ZnCl₂ promoting graphite domain formation and expanding interlayer spacing, which synergistically facilitates Na⁺ intercalation and migration. Concurrently, ZnCl₂ modification generates abundant closed pores, significantly augmenting Na⁺ storage capacity within these confined spaces. Notably, a relatively low BET surface area is not necessarily detrimental to sodium-ion storage. On the contrary, excessive surface area in hard carbon anodes often leads to severe electrolyte decomposition and increased irreversible Na⁺ consumption due to extensive SEI formation, resulting in low ICE [44]. More importantly, sodium storage in hard carbon is not dominated by surface area but by specific pore types. The closed pores and pseudo-graphitic domains are critical for reversible Na⁺ storage via pore filling and insertion mechanisms. Therefore, the reduced surface area observed in this work is beneficial for improving ICE while maintaining effective Na⁺ storage through regulated pore structure and carbon microdomains, which is consistent with BET results.

Figure 3. (a) The first GCD curves at 0.05 A g⁻¹; (b) slope and plateau contribution of discharge capacity during the second cycle; (c) rate performance; (d) GCD profiles of HCMZ-3 recorded under various current densities; (e) cycling stabilities at 0.1 A g⁻¹ of HCMZ and HCMZ-3; (f) comparison of specific capacity and ICE with previously reported biomass-derived HCs; (g) long-term cycling stabilities of HCMZ and HCMZ-3 at 1.0 A g⁻¹.

As depicted in Figure 3c, the rate performance of HCMZ and HCMZ-X is quantified through galvanostatic charge/discharge cycling at different current densities. The HCMZ-X anode demonstrates superior rate capability, maintaining high specific capacities across current densities from 0.05 to 2 A g⁻¹. Notably, HCMZ-3 delivers 357.94 mAh g⁻¹ at 0.05 A g⁻¹, significantly exceeding HCMZ (220.8 mAh g⁻¹). Even at an elevated current density of 1 A g⁻¹, HCMZ-3 still retains 190 mAh g⁻¹ while HCMZ decreases to 50 mAh g⁻¹. Additionally, when reverting to 0.05 A g⁻¹, HCMZ-3 exhibited over 95% capacity recovery, confirming exceptional structural stability. Furthermore, HCMZ-3 delivers outstanding cycling stability with capacities of 287.38 mAh g⁻¹ (200 cycles, 0.1 A g⁻¹, Figure 3e) and 145.52 mAh g⁻¹ (1000 cycles, 1.0 A g⁻¹, Figure 3f), significantly surpassing HCMZ. This excellent cycling performance demonstrates that ZnCl₂ modification is beneficial to enhancement of sodium storage kinetics. Notably, it is also observed that the second cycle reversible capacity of both electrodes is lower than that of the initial cycle, which can be attributed to the formation of a solid electrolyte interphase (SEI) arising from irreversible reactions between Na⁺ and the abundant active sites. As illustrated in Figure 3f and Table S5, compared with other reported biomass-derived hard carbons, HCMC-3 exhibits considerable specific capacity and ICE. Unmodified hard carbon anodes generally exhibit limited energy density due to their moderate plateau capacity. Recent studies indicate that microstructural regulation, including closed-pore control and heteroatom-assisted optimization, is an effective strategy to enhance sodium storage efficiency without excessive surface area increase. In this work, the regulated pore structure contributes to improved reversible capacity and Coulombic efficiency, while further enhancement of energy density may rely on microstructure–electrolyte matching and full-cell optimization.

To gain deeper insight into how ZnCl₂ incorporation influences the electrochemical performance of HC, cyclic voltammetry (CV) analyses were carried out on HCMZ and HCMZ-X samples. Figure 4a,d and Figure S7 depict the initial three CV cycles of HCMZ and HCMZ-X at a scan rate of 0.1 mV s⁻¹. During the discharge process, all samples display pronounced cathodic current peaks at 0.01 V, while distinct anodic peaks appear around 0.1 V upon charging, together constituting typical redox couples associated with Na⁺ intercalation and deintercalation. Notably, ZnCl₂-modified HCMZ-X electrodes show overlapping profiles in subsequent cycles, demonstrating enhanced electrochemical reversibility during charge/discharge processes. To further elucidate the sodium storage kinetics and capacitive contribution in both HCMZ-X and HCMZ electrodes, CV measurements of HCMZ and HCMZ-3 were subsequently recorded at different scan rates from 0.1 to 1.0 mV s⁻¹, as shown in Figure 4b,e. The anodic/cathodic peak positions demonstrate negligible shift across different scan rates, while the peak current density increases proportionally, suggesting enhanced electrochemical reversibility and diffusion-controlled charge transfer kinetics in HCMZ-3.

Figure 4. The first three CV curves of (a) HCMZ-3 and (d) HCMZ at 0.1 mV s⁻¹; CV profiles of (b) HCMZ-3 and (e) HCMZ at different scan rates; diffusion and capacitive contribution for (c) HCMZ-3 and (f) HCMZ; pseudo-capacitive contribution of (g) HCMZ-3 and (h) HCMZ at 0.1 mV s⁻¹; (i) relationship between peak current ($i$) and scan rate ($v$).

The Na⁺ storage mechanism was analyzed by distinguishing capacitive and diffusion-controlled contributions across various scan rates, based on the relationship described in Equation (3) [37]:

$$i\left(v\right)=k_{1}v+k_2{v}^{0.5}$$

(3)

where $k_1$ and $k_2$ are constants fitted by linearity and $k_{1}v$ and $k_2{v}^{0.5}$ denote surface-controlled capacitive behavior and diffusion-induced battery behavior, respectively. Figure 4g,h demonstrate that, at a scan rate of 0.1 mV s⁻¹, HCMZ-3 exhibits a lower capacitive contribution than HCMZ. In addition, Figure 4c,f illustrate the capacitive contributions of HCMZ-3 and HCMZ across different scan rates, respectively. As the scan rate increases, the capacitance contribution of HCMZ-3 gradually increases from 32% to 78%, suggesting a shift in the dominant Na⁺ storage mechanism from graphite intercalation and closed-hole filling to surface adsorption. In addition, the dependence between peak current ($i$) and scan rate ($v$) can be derived from the following formula [37]:

$$i=a{v}^b$$

(4)

The $b$ parameter, obtained from the linear relationship between log($i$) and log($v$), reflects the relative contribution of different Na⁺ storage processes. A $b$ value approaching 1.0 signifies predominantly capacitive-controlled behavior, while a value nearing 0.5 corresponds to a diffusion-controlled mechanism. As depicted in Figure 4i, the calculated $b$ values are 0.79 for HCMZ and 0.42 for HCMZ-3. This indicates that Na⁺ storage in HCMZ is mainly capacitive, whereas in HCMZ-3, Na⁺ transport is predominantly diffusion-controlled.

Nyquist plots from electrochemical impedance spectroscopy (Figure 5a) show semicircles in high and medium frequencies corresponding to charge transfer resistance (Rct). The activated HCMZ-X exhibits lower total impedance and enhanced charge transfer capability versus unmodified HCMZ. Specifically, HCMZ-3 achieves an Rct of 40.19 Ω at the electrode–electrolyte interface, with a 70% reduction relative to HCMZ (165.3 Ω, Table S6). To clarify the differences in Na⁺ storage kinetic among samples, the Na⁺ diffusion coefficient was assessed using the galvanostatic intermittent titration technique (GITT). As shown in Figure 5b, the Na⁺ diffusion coefficient ($D_{{Na}^{+}}$) is determined from Equation (4) based on Fick’s second law [46]:

$$D_{{Na}^{+}}={\displaystyle \frac{4}{\pi \tau }}{\left({\displaystyle \frac{m_B{V}_M}{M_{B}S}}\right)}^2{\left({\displaystyle \frac{\Delta E_S}{\Delta E_{\tau }}}\right)}^2$$

(5)

where $\tau$ denotes the duration of the applied current pulse; $m_B$ and $M_B$ represent the mass and molar mass of the active material, respectively; $V_M$ corresponds to the molar volume; $S$ refers to the active surface area of HC electrodes; and $\Delta E_S$ and $\Delta E_{\tau }$ indicate the steady-state and pulse-induced voltage variation extracted from the GITT profiles. GITT profiles of HCMZ and HCMZ-3 (Figure 5c,d) exhibit analogous variation trends during sodiation/de-sodiation, indicating comparable Na⁺ storage kinetics. During the discharge process (Figure 5c), both samples display a consistent decrease in $D_{{Na}^{+}}$. The elevated $D_{{Na}^{+}}$ values persisting down to 0.1 V (Figure 5d) originate from rapid Na⁺ adsorption at active sites on the carbon surface defects. Upon saturation of these surface adsorption sites, Na⁺ must overcome repulsion barriers to intercalate into graphitic interlayers, causing a sharp decline in $D_{{Na}^{+}}$ between 0.1~0.06 V. The subsequent rapid $D_{{Na}^{+}}$ increase (0.06~0.01 V) corresponds to Na⁺ filling closed pores. This three-stage process aligns with the “adsorption-intercalation-filling” storage mechanism. For HCMZ-3, the synergistic interplay of elevated surface defects, enlarged interlayer spacing (0.394 nm), and plentiful closed pores collectively accelerates Na⁺ surface adsorption while enhancing interlayer diffusion kinetics. Consequently, HCMZ-3 maintains the highest $D_{{Na}^{+}}$ throughout the progress, suggesting that ZnCl₂ promoted sodium ion diffusion kinetics, consistent with its electrochemical performance and CV results.

Figure 5. (a) Nyquist plots for HCMZ-3 and HCMZ; (b) GITT profiles of HCMZ-3 and HCMZ; (c,d) calculated Na⁺ diffusion coefficient during the discharge and charge process derived from GITT; (e) GDC curves of HCs at 0.5 A g⁻¹ (inset: optical images of the electrodes immersed in a phenolphthalein–ethanol solution at two different states of discharge); (f) schematic illustration depicting the Na⁺ storage stages in the HCMZ-X anode.

Quasi-metallic sodium reacts with ethanol to generate H₂ and CH₃CH₂ONa, inducing red coloration in phenolphthalein solutions. As verified in Figure 5e, HCMZ and HCMZ-3 electrodes, discharged to 0.5 V and 0.01 V, respectively, were immersed in 1.0 wt% phenolphthalein/ethanol solution. Upon discharging to 0.5 V, the HCMZ electrode produced a few bubbles with light pink coloration, confirming that the Na⁺ is mainly stored through surface adsorption in this stage. Due to weak adsorption, the Na⁺ easily desorbed into the solution, forming CH₃CH₂ONa and causing the color change. Conversely, HCMZ-3 exhibited vigorous H₂ evolution upon immersion. When discharged to 0.01 V, the solution of HCMZ-3 developed a dark purple color compared to the HCMZ’s pale pink (Figure 5e inset), which demonstrates much more quasi-metallic sodium presence in HCMZ-3 as well as a closed-pore filling Na⁺ storage mechanism. As illustrated in Figure 5f, the Na⁺ storage mechanism of HCMZ-3 can be interpreted as a three-step process. Within the slope region above 0.1 V, Na⁺ primarily adsorbs on the surface and defect active sites of hard carbon. In the plateau region below 0.1 V, Na⁺ intercalates into graphitic interlayers with progressive filling of interlayer spaces. With further discharge to lower potentials (from 0.06 to 0.01 V), nucleation of sodium species takes place, resulting in the generation of quasi-metallic sodium clusters that subsequently filled the closed pores.

Given the outstanding electrochemical performance of HCMZ-3 in half-cells, a full cell was assembled using HCMZ-3 as the anode and commercial Na₃V₂(PO₄)₃ (NVP) as the cathode, with the corresponding schematic structure shown in Figure 6a. Electrochemical evaluation was carried out over a voltage window of 2.0 to 2.8 V at current densities ranging from 0.05 to 1.0 A g⁻¹. Initially, the electrochemical behavior of NVP was assessed in a half-cell over a potential window of 1.5–3.8 V, where it delivered a specific charge capacity of 163.08 mAh g⁻¹ at 0.1 A g⁻¹ (Figure S8a). Distinct voltage plateaus appearing around 3.38 V and 3.4 V can be attributed to the V³⁺/V⁴⁺ redox reaction and local redox environment changes induced by Na⁺ relocation, respectively. At 0.05 A g⁻¹, the assembled HCMZ-3//NVP full cell exhibited initial charge and discharge capacities of 216.57 and 193.14 mAh g⁻¹ (Figure S8b). Irreversible capacity loss may result from anode-side SEI film formation and Na⁺ depletion during SEI reactions. After 50 cycles at 0.1 A g⁻¹, the GCD curve of the HCMZ-3//NVP full cell maintained stable morphology (Figure 6b). The rate performance of the full cell is summarized in Figure 6c. At rates of 0.05, 0.1, 0.5, 0.8, 1.0, and 2.0 A g⁻¹, the corresponding specific capacities obtained were 216.57, 172.59, 143.21, 97.55, 59.05, and 41.38 mAh g⁻¹, respectively. Notably, as shown in Figure 6d, the full battery exhibited an exceptional cycling stability, retaining approximately 82% of its capacity after 100 cycles.

Figure 6. (a) Schematic of HCMZ-3//NVP full cell; (b) initial charge/discharge profiles of HCMZ-3//NVP full cell at 0.1 A g⁻¹; (c) GCD curves of HCMZ-3//NVP full cell at various current densities; (d) cycling performance of HCMZ-3//NVP full cell at 0.1 A g⁻¹.

4. Conclusions

In summary, hard carbon anode materials were prepared using renewable and readily available macadamia shells as precursors, along with ZnCl₂ introduced as a structural modifier. The electrochemical results revealed that the HCMZ-3 material modified with ZnCl₂ delivers a reversible specific capacity of 382.05 mAh g⁻¹, with 262.64 mAh g⁻¹ contributed by the plateau region. This performance markedly exceeds that of the pristine sample, which shows a total reversible capacity of 220.67 mAh g⁻¹ and a corresponding plateau contribution of 124.56 mAh g⁻¹. By systematically correlating pore-structure and sodium storage behavior, it is clarified that ZnCl₂-modified HC stores Na⁺ through a sequential “adsorption-intercalation-filling” mechanism, which is closely associated with the enlarged interlayer spacing, regulated microcrystalline domains, and generation of closed pores. Such rational pore-structure modulation enhances sodium storage efficiency without excessive surface activation, thereby partially alleviating the intrinsic energy-density limitation of unmodified hard carbon anodes. Overall, this work highlights a feasible way of ZnCl₂-assisted microstructure regulation in biomass-derived hard carbon and provides a sustainable and scalable pathway toward the design of high-performance anode materials for sodium-ion batteries. Future efforts may further improve energy density through electrode densification and full-cell configuration optimization.