Synergistic solvation-surface engineering for high-performance aqueous zinc metal batteries†

Kuhang Liu^a, Rui Huo^a, Dongming Xu*^b, Chengjun Liu^a, Lishun Bai^a, Yue Liu^a, Ying He^a, Feiyan Yu^a, Keli Bu^a and Zhi Chang*^a
aSchool of Materials Science and Engineering, Key Laboratory of Electronic Packaging and Advanced Functional Materials of Hunan Province, Central South University, Changsha, Hunan 410083, China. E-mail: zhichang@csu.edu.cn
^bCollege of Chemistry, Huazhong Agricultural University, Wuhan, 430070, P. R. China. E-mail: XuDM@mail.hzau.edu.cn

First published on 18th August 2025

---

---

Introduction

The rapid proliferation of electric vehicles (EVs), portable electronics, and grid-scale energy storage systems has spurred an urgent demand for sustainable, safe, and high-performance energy storage technologies.^1–6 Among the emerging alternatives to lithium-ion batteries (LIBs), aqueous zinc metal batteries (AZMBs) stand out as promising next-generation energy storage systems due to their abundant resources, low cost, and high safety.^7–9 Meanwhile, Zn metal boasts a high theoretical specific capacity (820 mAh g⁻¹ and 5845 mAh cm⁻³) and a low redox potential (−0.76 V vs. SHE).^10,11 However, conventional AZMBs often suffer from severe Zn corrosion, hydrogen evolution, and Zn dendrite formation. These issues are intrinsically linked to the reactive aqueous environment. Specifically, as shown in Fig. 1, the aqueous electrolyte contains abundant hydrogen-bonded free H₂O molecules, while solvated Zn²⁺ ions undergo partial desolvation of H₂O molecules near the Zn metal surface. These highly reactive desolvated and free H₂O molecules, severely corrode the Zn metal, exacerbate the hydrogen evolution reaction (HER), and lead to the formation of insulating by-products like Zn₄SO₄(OH)₆·xH₂O (ZSH).^12,13 Additionally, irregular Zn²⁺ flux results in disordered Zn deposition, promoting dendrite growth through the tip effect. These detrimental problems lead to low Coulombic efficiency (CE), poor cycling stability, rapid capacity decay, and even potential short circuits in AZMBs, particularly under high areal capacity, current density, and deep discharge conditions.^14,15 Addressing these challenges is critical to improve Zn metal reversibility and unlocking the full potential of AZMBs for energy storage.

To address these challenges, various strategies have been proposed. For instance, constructing Zn electrodes with preferred (002) crystal plane orientation¹⁶ or three-dimensional porous structures¹⁷ enables uniform Zn deposition through crystal orientation regulation and spatial confinement effects. Coating electrode surfaces with MOFs,¹⁸ conductive polymers,^19–21 or metal protective layers^22–24 can reduce Zn nucleation barriers. Functional separator designs focus on incorporating nanochannel ion sieves to optimize charge distribution via selective ion transport.^25,26 Additionally, employing high-concentration electrolytes,²⁷ water-in-salt systems,²⁸ or deep eutectic solvents^29–31 significantly suppresses hydrogen evolution by reconstructing Zn²⁺ solvation structures. However, these strategies often rely on complex fabrication processes or high-cost materials, hindering their large-scale application. In contrast, electrolyte additive engineering exhibits unique advantages of simplicity and cost-effectiveness.³² The introduction of additives into conventional electrolytes can simultaneously regulate Zn²⁺ solvation structures and improve interfacial chemistry, enabling homogeneous Zn deposition while suppressing side reactions.^33,34

Research on electrolyte additives has mainly focused on three key mechanisms: regulating hydrogen bond (HB) networks, optimizing solvation structures, and modulating the electrical double layer (EDL). For instance, additives such as 1,2-dimethoxyethane (DME),³⁵ pentaerythritol (PTT),³⁶ and polyhydroxy hexitol³⁷ have been proven to be able to form stronger hydrogen bonds with H₂O molecules. This increases the energy barrier of hydrogen bond breaking and reconstruction, and prevents protons from jumping between H₂O molecules, thereby suppressing the HER. Similarly, additives like carboxymethyl chitosan (CMCHS),³⁸ glycine,³⁹ and dimethyl sulfoxide (DMSO)⁴⁰ were reported to reconstruct the solvation shell of Zn²⁺ ions, reducing the contact between active H₂O molecules and the Zn negative electrode during the deposition process and suppressing unnecessary side reactions. At the same time, the regulation of the Zn²⁺ solvation shell can also affect the desolvation and nucleation processes of Zn²⁺, promoting the uniform deposition of Zn²⁺ ions. Additionally, saccharin (Sac),⁴¹ NH₄H₂PO₄,⁴² and lactobionic acid (LA)⁴³ have been employed to regulate the EDL at the anode/electrolyte interface, ensuring uniform Zn²⁺ flux and dendrite-free deposition by adsorbing onto the Zn metal surface. However, most existing additives are limited to targeting only one of these aspects—the hydrogen bond network, solvation structure, or EDL—and none can simultaneously achieve all three effects. Therefore, developing multifunctional additives capable of simultaneously regulating hydrogen bond networks, solvation structures, and EDLs is of crucial significance for achieving the long-cycle stability of Zn metal and advancing the practical applications of AZMBs.

In this study, we introduce L-threonine (L-Thr) as a groundbreaking multifunctional electrolyte additive for AZMBs and systematically elucidate its regulatory mechanisms. Rich in polar groups (–COOH, –OH, –NH₂), L-Thr exhibits high chemical activity with unique charge distributions and spatial orientations. As shown in Fig. 1, the amino group (–NH₂) in L-Thr can coordinate with Zn²⁺, displacing H₂O molecules within the Zn²⁺ solvation structure and disrupting the original HB network, thereby reducing both the amount and reactivity of free water. This minimizes highly reactive desolvated H₂O molecules and hydrogen-bonded free H₂O molecules, effectively eliminating severe water-related side reactions and the HER on the Zn metal surface. Additionally, the carboxyl (–COOH) group and –NH₂ in L-Thr preferentially adsorb onto the Zn(101) crystal plane, forming a hydrophobic layer that reduces direct contact between free H₂O molecules and the Zn metal, guiding uniform Zn²⁺ deposition. This unique physical interface barrier significantly suppresses water-related side reactions and dendrite formation, thereby enhancing the stability of the Zn metal. As a result, a Zn//Cu half-cell assembled with L-Thr additive achieves an exceptionally high CE of 99.71% over 2000 cycles at 2 mA cm⁻² (areal capacity of 1 mAh cm⁻²). A Zn//Zn symmetric cell with L-Thr operates stably for over 2500 hours at 0.5 mA cm⁻² and 0.5 mAh cm⁻². Even under harsh conditions of 10.0 mA cm⁻² and 10.0 mAh cm⁻², it retains stable cycling for 600 hours. Furthermore, a Zn//NH₄V₄O₁₀ (Zn//NVO) full cell with L-Thr exhibits excellent cycling stability, retaining 80.9% capacity after 1000 cycles and 77.9% after 2000 cycles at 5 A g⁻¹. An advanced Zn//NVO pouch cell with low negative capacity/positive capacity (N/P ratio = 2.94) and high cathode mass loading (12.03 mg cm⁻²) is achieved and retains 85.4% capacity after 200 cycles, showcasing its practical potential.

Experimental

Preparation of the electrolytes

First, exactly 57.512 g of ZnSO₄·7H₂O (Aladdin, 99.5%) was weighed and dissolved in deionized water, and then diluted to 100 mL in a volumetric flask to obtain the 2 M ZnSO₄ (ZSO) base electrolyte. Subsequently, precisely measured quantities of L-Thr (Aladdin, 99.0%) −0.238 g (0.1 M), 0.476 g (0.2 M), 0.953 g (0.4 M), 1.429 g (0.6 M), 1.905 g (0.8 M), 2.382 g (1.0 M), and 2.858 g (1.2 M) – were, respectively, added to 20 mL of the ZSO electrolyte to prepare a series of modified electrolytes with L-Thr concentration gradients.

Preparation of the NH₄V₄O₁₀ (NVO) cathode

The cathode material NH₄V₄O₁₀ (NVO) was synthesized according to a previous report.⁴⁴ NH₄VO₃ (2.74 mmol, Aladdin, 99%) was added to 40 mL deionized water and stirred continuously at 60 °C until completely dissolved. Subsequently, H₂C₂O₄·2H₂O (4.6 mmol, Aladdin, 98%) solid powder was directly added to the above-mentioned solution, followed by magnetic stirring for 30 minutes until a black-green solution was obtained. The obtained solution was then transferred into a 60 mL Teflon-lined stainless autoclave and maintained at 180 °C for 6 hours, yielding a blue-black solid product. The obtained precipitate was centrifugally washed three times with deionized water and once with absolute ethanol, then dried at 60 °C for 12 hours to produce the final NVO powder.

The cathode electrodes for full cells were composed of NVO, Super-P and polyvinylidene fluoride (PVDF) at a weight ratio of 7:2:1. After grinding the above powder evenly, N-methyl-2-pyrrolidone (NMP, Aladdin, 99%) was added as the solvent and stirred quickly to obtain a viscous slurry. The obtained slurry was coated on disc-shaped stainless steel meshes (12 mm), followed by drying in a vacuum oven at 80 °C for 12 h. The mass loading of the prepared NVO electrode was ∼1.0–2.0 mg cm⁻².

Material characterization

X-ray diffraction (XRD) patterns of all samples were recorded with a Rigaku diffractometer (Mini Flex 600) using Cu-Kα1 radiation (λ = 1.5406 Å) from 5° to 80° (scan rate: 10° min⁻¹). The SEM pictures of the Zn anode and the 3D reconstructed morphology of the Zn metal surface after cycling were confirmed using a field emission scanning electron microscope (Phenom Pharos G2) All the electrode samples were collected by disassembling the Zn//Zn or Zn//NVO cells after a certain number of cycles at given currents. The solvation structures of Zn²⁺ in various solutions were studied by Fourier-transform infrared spectroscopy (FTIR, Thermo Fisher Scientific Nicolet iS20), Raman microscopy (Horiba LabRam HR Evolution) and liquid-state nuclear magnetic resonance (NMR, Bruker Avance NEO 400 MHz). X-ray photoelectron spectroscopy (XPS) analysis was performed on the electrodeposited Zn foil using a Thermo Scientific ESCALAB Xi⁺. Contact angle measurement between Zn foil and electrolytes was performed on a Dataphysics OCA 20.

Electrochemical measurements

Zn metal plates (0.10 mm) were cut into suitable discs for direct use as electrodes for symmetrical, asymmetrical and standard full batteries, and glass fibers (Waterman, 19 mm in diameter) were used as diaphragms for the cells. All cells were assembled in CR2032-type cells containing 100 μL of electrolyte. The practicality of the L-Thr/ZSO electrolyte was tested by assembling Zn//NVO cells based on different electrolytes with Zn metal as the anode and NVO as the cathode. The electrochemical performance of Zn//NVO cells was tested in the voltage range of 0.3–1.4 V. The pouch cell consisted of a 2.2 cm × 1.7 cm sized Zn anode (10 μm), a 2.7 cm × 2.2 cm glass fiber separator, a 2.2 cm × 1.7 cm sized Ti foil and 2 cm × 1.5 cm sized NH₄V₄O₁₀ cathode. All electrochemical tests were performed on a Neware battery tester. The CE test was performed by plating a quantity of Zn (1/2 mA cm⁻², 1 mAh cm⁻²) on a copper electrode and charging it to 0.5 V. The amount of Zn that can be stripped per cycle was divided by the amount of Zn deposited to obtain the value of CEs. The Tafel test was carried out in a three-electrode measuring system with bare Zn, Pt foil and Ag/AgCl electrodes as the working, counter, and reference electrodes, respectively, in different electrolytes. The cyclic voltammetry (CV, sweep rate of 0.2 mV s⁻¹) and chronoamperometry (CA) (at overpotentials of −200 mV) tests were performed on a CHI660E electrochemical workstation, whereas the linear sweep voltammetry (LSV, sweep rate of 10 mV s⁻¹) and electrochemical impedance spectroscopy (EIS, from 10 mHz to 100 kHz) tests were performed on a CORRTEST CS300M electrochemical workstation.

Cyclic voltammetry was performed on Zn//Zn symmetric cells within a non-faradaic potential window of −50 mV to 50 mV at systematically varied scan rates (40–120 mV s⁻¹). The capacitive currents were extracted at 0 mV polarization potential through baseline correction and plotted versus the corresponding scan rates. The electric double-layer capacitance (EDLC) was determined from the linear regression slope according to:

	(1)

where C_EDL is the capacitance and Δi represents the current difference between half of the scanning voltage range at each scanning rate.

The activation energy (E_a) is defined as the minimum energy required to transition reactants from the ground state to an activated complex enabling chemical transformation. The EIS of an electrochemical system is measured at different temperatures (20–60 °C) and brought into the Arrhenius formula for fitting. The formula is given below:

	(2)

where R_ct is the charge-transfer resistance, R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), T is the absolute temperature (K), and A is the pre-exponential factor. Taking 1000/T as the horizontal coordinate and −ln(R_ct) as the vertical coordinate, the opposite of the slope of the straight line obtained by plotting the fit is the value of E_a.

Ionic conductivity (σ), quantifying the bulk ion transport capability within the electrolyte, is defined as the electrical conductance per unit length through a cross-sectional area. It was determined via the following formula:

	(3)

where σ denotes ionic conductivity in Ω cm⁻¹, l represents the electrolyte thickness (cm), S corresponds to the effective conductive area (cm²), and R signifies the bulk resistance (Ω) derived from the low-frequency intercept in EIS Nyquist plots. SS//SS symmetric cells were assembled for testing.

The transference number (t⁺) evaluates the mobility of a particular ion in the system and is generally expressed as the ratio of that ion to the total number of all ions in the system. Based on assembled Zn//Zn symmetric cells, a potential of 10 mV is applied while the change in current over time is recorded. t⁺ can be expressed using the following equation:

	(4)

where I₀ and I_s are the initial and steady state currents, respectively, and R₀ and R_s are the impedance between the electrode and electrolyte interface before and after polarization, respectively.

The depth of discharge (DOD), quantifying the utilization efficiency of metallic zinc electrodes, is defined as the percentage ratio between the electrochemically active zinc capacity and its theoretical maximum value. The area-specific DOD was calculated via:

	(5)

where C_a represents the capacity condition of the actual cell test, the unit is mAh cm⁻². C_s represents the theoretical capacity of Zn metal, usually 5855 mAh cm⁻³. L represents the thickness of the Zn foil (cm).

Calculations

All molecular calculations were performed using Gaussian 16 (Revision C.01) with the hybrid PBE0 functional.⁴⁵ Grimme's D3 dispersion correction with Becke–Johnson damping (DFT-D3BJ) was systematically applied to account for van der Waals interactions.⁴⁶ Geometries were optimized at the 6-311+G(d,p) basis set level incorporating the SMD implicit solvation model to simulate aqueous environments.⁴⁷ Frequency analyses confirmed all optimized structures as local minima (no imaginary frequencies). Electrostatic potential (ESP) distributions were quantitatively analyzed using Multiwfn and visualized with VMD interaction energies and were computed via:

Ebind = Ecomplex − (EpartA + EpartB)	(6)

where subscripts denote the respective molecular systems.⁴⁸

First-principles calculations based on density functional theory (DFT) were executed using the Vienna Ab initio Simulation Package (VASP).⁴⁹ The Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation (GGA) framework was employed with projected augmented wave (PAW) pseudopotentials.^50,51 A plane-wave cutoff energy of 450 eV ensured basis set completeness, while Methfessel–Paxton first-order smearing (σ = 0.2 eV) addressed partial orbital occupancy. Structural optimizations enforced convergence criteria of 10⁻⁵ eV for electronic self-consistency and 0.02 eV Å⁻¹ for atomic forces. For the optimization of both geometry and lattice size, the Brillouin zone integration was performed with a 1 × 1 × 1 Gamma centered sampling. To mitigate periodic boundary artifacts, a 17 Å vacuum layer was introduced perpendicular to surface planes. Dispersion corrections were incorporated via Grimme's DFT-D3 empirical scheme.^46,52 Adsorption energetics were evaluated as:

	(7)

where , E*, and E_adsorbent correspond to the adsorbed system, clean surface, and isolated adsorbate energies, respectively.

Results and discussion

L-Thr-mediated solvation and interfacial regulation

L-Threonine (L-Thr), an α-amino acid with the chemical formula C₄H₉NO₃, features both an amino (–NH₂) group and a carboxyl (–COOH) group, enabling potential Zn²⁺ chelation. The presence of a polar hydroxyl group (–OH) on its side chain confers high water solubility, as evidenced by the clear and transparent appearance of aqueous solutions of ZnSO₄ containing varying concentrations of L-Thr concentrations (Fig. S1a†). Within the selected concentration range, the pH of the electrolyte remains significantly below the isoelectric point (pI = 5.87) of L-Thr (Fig. S2†), likely shifting its structure from HOCH(CH₃)–CH(NH₃⁺)–COO⁻ to HOCH(CH₃)–CH(NH₃⁺)–COOH.⁵³ To elucidate L-Thr's regulatory mechanism on the solvation structure and interface, we investigated Zn²⁺ solvation behavior in different electrolytes through density functional theory (DFT) calculations and spectroscopic analysis. In conventional ZnSO₄ electrolyte (ZSO), Zn²⁺ primarily coordinates with six water molecules to form [Zn²⁺-6H₂O] (Fig. S3a†). However, upon L-Thr addition (L-Thr/ZSO), L-Thr's higher binding energy with Zn²⁺ (−2.71 eV vs. −2.24 eV for [Zn²⁺-6H₂O]) drives the formation of a new solvation structure, [L-Thr-Zn²⁺-5H₂O] (Fig. S3b and S4†). This binding energy difference confirms L-Thr's capacity to displace water molecules from the solvation shell.⁵⁴ Electrostatic potential (ESP) mapping (Fig. 2a) reveals significantly reduced electrostatic potential in [L-Thr-Zn²⁺-5H₂O] compared to [Zn²⁺-6H₂O], indicating effective charge repulsion mitigation and solvation structure reconstruction.⁵⁵ DFT calculations (Fig. 2b) further show stronger L-Thr-H₂O binding energy (−0.31 eV) than H₂O–H₂O (−0.16 eV), suggesting L-Thr disrupts hydrogen bonds in water networks, suppressing water activity and the HER. The optical images at −10 °C shown in Fig. S1b† further confirm this result. The –NH₂ groups in L-Thr molecules form stronger interactions with water molecules, effectively disrupting the original hydrogen-bond network of water and thereby lowering the freezing point of the electrolyte. Notably, the binding energy of Zn²⁺-L-Thr (−1.16 eV) exceeds that of both Zn²⁺-H₂O (−0.59 eV) and Zn²⁺-SO₄²⁻ (−1.04 eV), confirming preferential Zn²⁺ coordination with L-Thr. DFT calculations (Fig. 2c and d) also highlight L-Thr's preferential adsorption on the Zn metal surface. The –OH and –NH₂/–COOH groups of L-Thr exhibit adsorption energies of −1.300 eV and −1.473 eV on Zn(101), respectively, significantly surpassing that of H₂O, −0.85 eV. The presence of the characteristic N 1s peak in the X-ray photoelectron spectroscopy (XPS) results of the electroplated Zn anode (30 min) in the ZSO + 0.8 M L-Thr electrolyte further confirms the adsorption behavior of L-Thr on the Zn metal surface (Fig. S5†). This strong preferential adsorption establishes a water-deficient interface, physically isolating the Zn metal from water molecules while facilitating homogeneous Zn²⁺ deposition, thereby suppressing hydrogen evolution corrosion and dendrite growth.

Nuclear magnetic resonance (NMR) spectroscopy (Fig. 2e and S6†) shows downfield shifts in D₂O ¹H peaks from 4.7198 ppm (ZSO) to 4.7168 ppm (L-Thr/ZSO), indicating L-Thr's partial replacement of water molecules in the solvation shell of Zn²⁺.³⁹ Fourier-transform infrared spectroscopy (FTIR) (Fig. 2f and S7†) further supports this, showing a higher wavenumber shift in the –OH stretching vibration peak, indicating that L-Thr disrupts the original hydrogen-bond network by forming intermolecular hydrogen bonds, thereby reducing the activity of free water. In addition, the red-shift in ν(SO₄²⁻) stretching confirms that L-Thr weakens the interaction between Zn²⁺ and SO₄²⁻, thereby reconstructing the Zn²⁺ solvation structure.⁵⁶ A similar trend can also be observed in the Raman spectrum (Fig. 2g and S8†), with the –OH stretching vibration peak shifting to a higher frequency, revealing the coordination of L-Thr with Zn²⁺. The proportion of ν(C–H) and strong hydrogen bonds gradually increases (from 34.26% to 35.11%) with the rise of L-Thr concentration, while the proportion of weak hydrogen bonds gradually decreases, quantitatively verifying the restricting effect of L-Thr on free water.⁵⁷ The ν(SO₄²⁻) band shows enhanced solvent-separated ion pair (SSIP, [Zn²⁺(H₂O)₆ SO₄²⁻]) intensity and suppressed contact ion pair (CIP, [Zn²⁺(H₂O)₅ OSO₃²⁻]) intensity with increasing L-Thr concentration, demonstrating restricted SO₄²⁻ access to Zn²⁺ solvation shells due to strong Zn²⁺-L-Thr interaction (consistent with the results obtained in Fig. 2b).⁵⁸ Collectively, these results demonstrate L-Thr's synergistic enhancement of the electrolyte and Zn metal stability through solvation structure reconstruction, water-deficient interface formation, and hydrogen-bond network disruption.

Corrosion-resistant and kinetics-boosted Zn electrodes via L-Thr

The corrosion-inhibiting effect of the L-Thr additive on Zn metals was quantitatively evaluated through Tafel curves (Fig. 3a). Compared to the ZSO electrolyte (−0.995 V vs. Ag/AgCl, 3.360 mA cm⁻²), Zn metal in the L-Thr/ZSO electrolyte exhibited a higher corrosion potential (−0.989 V vs. Ag/AgCl) and lower corrosion current density (0.168 mA cm⁻²). Additionally, linear sweep voltammetry (LSV) test results (Fig. 3b and S9†) further indicated that the battery with the L-Thr additive lowered the HER potential and increased the oxygen evolution reaction (OER) potential. At a current density of −10.0 mA cm⁻², the HER potential shifted from −1.154 V (ZSO) to −1.249 V (ZSO + 0.8 M L-Thr), demonstrating L-Thr's effectiveness in suppressing hydrogen evolution corrosion and significantly expanding the electrochemical window. To investigate the adsorption behavior of the L-Thr additive on Zn metal, electric double-layer capacitance (EDLC) tests were conducted. With the addition of L-Thr, the EDLC value decreased significantly from 86.2 μF cm⁻² to 53.4 μF cm⁻² (Fig. 3c and S10†). The reduction was attributed to the adsorption of L-Thr on the Zn metal, which increased the thickness of the EDL. The preferential adsorption of L-Thr also slightly increased the Zn metal's surface energy, increasing the electrolyte contact angle on the Zn metal surface at low concentrations (Fig. S11†). Furthermore, the deposition mechanism of Zn²⁺ in two different electrolytes was further examined using chronoamperometry (CA) tests. At an overpotential of −200 mV, L-Thr significantly slowed the current increase rate (Fig. 3d), as its adsorption on the Zn metal surface restricted the 2D diffusion of Zn²⁺, promoting a more stable 3D diffusion mode. The cyclic voltammetry (CV) curves of the Zn//Cu asymmetric battery (Fig. S12†) revealed that the addition of L-Thr increased the nucleation overpotential (ΔV = 23 mV) compared to the ZSO electrolyte. This increase was probably due to the adsorption of L-Thr on the electrochemical active sites of the Zn metal, which hinders Zn²⁺ charge transfer and alters Zn²⁺ deposition behavior—consistent with the above-mentioned DFT calculations.

The kinetics of Zn²⁺ deposition was evaluated by quantitatively analyzing the desolvation activation energy (E_a) of Zn²⁺. According to the Arrhenius equation, the E_a was calculated based on the impedance of the Zn//Zn symmetric cells at temperatures ranging from 20 to 60 °C. As the temperature increased, the reaction kinetics accelerated, and the impedance values gradually decreased (Fig. S13†). When L-Thr was added to the ZSO electrolyte, the E_a decreased from 25.04 kJ mol⁻¹ to 18.49 kJ mol⁻¹ (Fig. 3e), indicating that L-Thr significantly lowers the desolvation energy barrier of Zn²⁺, thereby enhancing Zn²⁺ deposition kinetics. Moreover, L-Thr improves Zn²⁺ mass transfer in the liquid phase, thus reducing the electrolyte impedance and increasing electrical conductivity (Fig. 3f and S14†). The migration rate of Zn²⁺ in different electrolytes was further tested through potentiostatic polarization and EIS tests (Fig. 3g–i). In the ZSO + 0.8 M L-Thr electrolyte, the calculated Zn²⁺ transference number (t_Zn²⁺) increased to 0.53, significantly higher than the 0.38 in the ZSO electrolyte, confirming L-Thr's role in weakening Zn²⁺ solvation and enabling rapid Zn²⁺ migration.

Dendrite and byproduct suppression enabled by L-Thr

To investigate the influence of the L-Thr additive on Zn deposition behavior, the components and morphology of the cycled Zn anode were characterized. According to the X-ray diffraction (XRD) patterns (Fig. 4a), obvious strong diffraction peaks for the by-product ZSH were detected on the surface of the Zn metal cycled in a typical ZSO electrolyte. In sharp contrast, only weak ZSH diffraction peaks were detected on the Zn metal cycled in the ZSO + 0.8 M L-Thr electrolyte. Additionally, Zn foil immersed in the ZSO and ZSO + 0.8 M L-Thr electrolytes for 5 days showed stark differences. The Zn metal immersed in the ZSO electrolyte exhibited strong ZSH diffraction peaks (Fig. S15†), and significant by-product formation (Fig. S16b†). In sharp contrast, the Zn metal immersed in the ZSO + 0.8 M L-Thr electrolyte displayed minimal ZSH diffraction peaks and retained a smooth, dense surface (Fig. S16a†). These results indicate that the L-Thr additive simultaneously reduces solvated water and establishes a hydrophobic layer through synergistic coordination and adsorption, thereby effectively inhibiting the formation of by-products on the Zn anode.

SEM images of Zn metal cycled in the two electrolytes are shown in Fig. 4b–g. In the ZSO + 0.8 M L-Thr electrolyte, after the Zn//Zn symmetric cells were cycled for 100 h at a low current density and deposition capacity (0.2 mA cm⁻² and 0.2 mAh cm⁻²), the Zn metal exhibited a flatter and denser morphology (Fig. 4b). In contrast, the Zn metal cycled in the ZSO electrolyte showed an irregular and rough surface, indicating non-uniform Zn²⁺ deposition (Fig. 4d). Even at higher current density and capacity (1 mA cm⁻² and 1 mAh cm⁻²), the Zn metal cycled in the ZSO + 0.8 M L-Thr electrolyte retained a flat surface with minimal by-product formation (Fig. 4c), while Zn metal cycled in the ZSO electrolyte demonstrated uncontrolled dendrites and substantial by-products (Fig. 4e). Similar results can be observed under varying conditions (Fig. S17 and S18†). Cross-sectional SEM images further demonstrated that the Zn metal cycled in the ZSO electrolyte had an uneven, rough deposition layer (Fig. 4g), whereas the Zn metal cycled in the ZSO + 0.8 M L-Thr electrolyte formed a smoother and denser deposition layer (Fig. 4f), highlighting the positive role of the L-Thr additive in promoting uniform Zn deposition and suppressing side reactions. Furthermore, the SEM images of Cu foil with varying amounts of deposited Zn in Zn//Cu cells (Fig. S19†) revealed that the Cu foil in the ZSO + 0.8 M L-Thr electrolyte exhibited exceptionally uniform and compact Zn deposition, whereas the deposition in the ZSO electrolyte appeared highly localized and disordered. The three-dimensional laser confocal scanning microscopy (3D-LCSM) images showed that after 100 hours of cycling at 1 mA cm⁻² and 1 mAh cm⁻², the Zn metal cycled in the ZSO + 0.8 M L-Thr electrolyte displayed a remarkably smooth surface (Fig. 4h and S20a†), whereas the Zn metal cycled in the ZSO electrolyte exhibited concentrated thick Zn dendrites and hollows (Fig. 4i and S20b†), a trend further confirmed by the 3D contour maps (Fig. S20c and S20d†). Additionally, we used an in situ optical microscope to observe the Zn plating behavior at a high current density of 5 mA cm⁻² in two different electrolytes. In the ZSO electrolyte, non-uniform and loose Zn deposits were quickly formed within the first 5 minutes, followed by severe hydrogen evolution corrosion after 10 minutes (bubbles). In stark contrast, a uniform and compact Zn deposition layer was formed in the ZSO + 0.8 M L-Thr electrolyte, with no observable dendrites or bubbles throughout the plating process, further demonstrating the critical role of the L-Thr additive in suppressing hydrogen evolution corrosion.

Cycling performance of cells assembled with the ZSO/L-Thr electrolyte

The reversibility of Zn²⁺ deposition/stripping was evaluated by measuring the Coulombic efficiency (CE) of Zn//Cu asymmetric cells. At a current density of 2 mA cm⁻² and a deposition capacity of 1 mAh cm⁻², the Zn//Cu asymmetric cell using the typical ZSO electrolyte experienced sudden CE drops after only 165 cycles. However, with the addition of the L-Thr additive, the reversibility of Zn²⁺ deposition/stripping significantly enhanced. At an optimal L-Thr concentration of 0.8 mol L⁻¹ (ZSO + 0.8 M L-Thr), the cell achieved an exceptionally high average CE of 99.71% over 2000 cycles (Fig. 5a). The voltage–capacity profiles (Fig. 5b) revealed consistent capacity retention for the cell using the ZSO + 0.8 M L-Thr electrolyte, with an overpotential difference of ∼224 mV (compared to 55 mV in the ZSO electrolyte), likely due to the adsorption of L-Thr on the surface of Zn.⁵⁹ Additionally, at a current density of 1 mA cm⁻² and a deposition capacity of 1 mAh cm⁻², the Zn//Cu asymmetric cell assembled with the ZSO + 0.8 M L-Thr electrolyte demonstrated an average CE of 98.98% over 450 cycles, while the cell based on the ZSO electrolyte showed a lower average CE of 97.57% and failed within 80 cycles (Fig. S21†). Under the test conditions of the pre-activation mode (Aurbach method), the Zn//Cu asymmetric cell with the ZSO + 0.8 M L-Thr electrolyte also outperformed the cell based on the ZSO electrolyte, achieving a higher average CE (Fig. S22†). Additionally, we also assembled Zn//Zn symmetric cells to evaluate the stability of Zn metals with/without the L-Thr additive. Under the conditions of 0.25 mA cm⁻² and 0.25 mAh cm⁻², the Zn//Zn symmetric cell using the ZSO electrolyte experienced rapid short-circuiting after only 285 hours, marked by a sudden voltage drop to near 0 V, indicating severe Zn dendrite penetration (Fig. 5c). In contrast, the symmetric cell using the ZSO + 0.8 M L-Thr electrolyte (optimal concentration) achieved the longest cycle life of 2500 hours with stable voltage fluctuations, extending the cycle life by more than 8 times (Fig. 5c and S23a†). When the current density and areal capacity increased to 1 mA cm⁻² and 1 mAh cm⁻², the symmetric cell with the ZSO + 0.8 M L-Thr electrolyte still demonstrated a significantly improved lifespan (2100 hours) compared to the cell using the ZSO electrolyte (Fig. S23b†). This trend persisted under high current density (5 mA cm⁻²) and high areal capacity (5 mAh cm⁻²) conditions (Fig. S24b†) and even under harsh conditions of 10 mA cm⁻² and 10 mAh cm⁻², the symmetric cell with the ZSO + 0.8 M L-Thr electrolyte retained stable operation for 600 hours, achieving an exceptionally high cumulative plated capacity (CPC) of 6 Ah cm⁻², far surpassing that of the cell using the ZSO electrolyte (Fig. 5d, S24a and S25†).

The depth of discharge (DOD) of the Zn anode is an important indicator for practical applications. The symmetric cell using the ZSO + 0.8 M L-Thr electrolyte successfully achieved deep charge–discharge cycling under limited Zn (10 μm) conditions, operating for 240 hours at 68.6% Zn utilization (Fig. 5e and S26†). To verify the high-rate stability, symmetric cells were tested at different current densities (from 0.1 to 10 mA cm⁻²), with 10 cycles per condition. Obviously, the symmetric cell using the ZSO electrolyte exhibited severe voltage fluctuations and eventually short-circuited when cycled at 5 mA cm⁻² and 5 mAh cm⁻² (Fig. 5f). In stark contrast, the symmetric cell with the ZSO + 0.8 M L-Thr electrolyte demonstrated more stable cycling voltages. When the cycling conditions returned to 1 mA cm⁻² and 1 mAh cm⁻², the symmetric cell ultimately achieved a cycle life of 1300 hours (Fig. 5f, S27 and S28†), demonstrating the remarkable high-rate capability and cycling stability. As shown in Tables S1 and S2,† in previously reported studies involving electrolyte modification strategies, Zn//Zn symmetric cells and Zn//Cu half-cells under the synergistic solvation-surface engineering strategy using L-Thr exhibited excellent comprehensive performance, including a long cycle life (2500 hours), high CPC (6 mAh cm⁻²), and high Coulombic efficiency (99.71%). These results highlight the effectiveness of L-Thr as an electrolyte additive in improving the reversibility and stability of AZMBs.

Practical Zn//NVO full cells and pouch cells

To assess the practical feasibility of the L-Thr additive, we assembled Zn//NH₄V₄O₁₀ (Zn//NVO) full cells using ZSO and ZSO + 0.8 M L-Thr electrolytes with NVO as the cathode (Fig. S29†). At a current density of 5 A g⁻¹, the charge–discharge curves and cycling stability of Zn//NVO cells using different electrolytes revealed significant improvements with the L-Thr additive (Fig. 6a and b). The Zn//NVO cells with the ZSO + 0.8 M L-Thr electrolyte demonstrated higher capacity and superior cycling stability (Fig. 6b), retaining 80.9% and 77.9% of their capacity after 1000 and 2000 cycles, respectively. In stark contrast, the cells using the ZSO electrolyte retained only 18.8% of their capacity after 2000 cycles. The CV curves of the Zn//NVO cell displayed multiple redox peaks, with the L-Thr based cell showing enhanced current density and positive shifted redox potentials, indicating improved reaction kinetics (Fig. S30†).⁶⁰ Additionally, the Zn//NVO cells using the ZSO + 0.8 M L-Thr electrolyte also exhibited lower charge transfer resistance before and after cycling (Fig. 6c), highlighting the L-Thr's role in boosting specific capacity and charge transfer efficiency.

Additionally, the morphology of Zn anodes cycled for 1000 cycles in different electrolytes was examined using SEM. As shown in Fig. 6f, the Zn anode cycled in the ZSO + 0.8 M L-Thr electrolyte displayed a flat and dense Zn deposition, while the Zn anode cycled in the ZSO electrolyte was covered with vigorous dendrites and by-products. The rate performance of the Zn//NVO cells (Fig. 6d and S31†) revealed that the cell assembled with the ZSO + 0.8 M L-Thr electrolyte exhibited significantly higher reversible capacities across all current densities compared to the cells using the ZSO electrolyte. Specifically, capacities of 354 mAh g⁻¹ and 144 mAh g⁻¹ were achieved at 1 A g⁻¹ and 10 A g⁻¹, respectively. When the current density was restored to 1 A g⁻¹, the capacity recovered to 312 mAh g⁻¹, demonstrating excellent reversibility. Furthermore, Zn//NVO cells with the ZSO + 0.8 M L-Thr electrolyte and a high mass loading of 6.097 mg cm⁻² retained 80.9% (171 mAh g⁻¹) of their capacity after 600 cycles at 1 A g⁻¹ (Fig. 6e and S32†).To further explore the potential of L-Thr as an electrolyte additive, a pouch cell was assembled with a high-mass-loading NVO cathode (12.03 mg cm⁻²) and a limited Zn anode (10 μm, corresponding to a low N/P ratio of 2.94:1). The pouch cell with the ZSO + 0.8 M L-Thr electrolyte exhibited stable cycling over 200 cycles, retaining 85.4% (129.3 mAh g⁻¹) of its capacity at a current density of 0.8 A g⁻¹ (Fig. 6g and S33†). These results collectively demonstrate the feasibility of the L-Thr additive in enhancing the performance of AZMBs and also highlight its potential for commercial applications.

Conclusions

In summary, we reported a holistic solvation engineering and surface adsorption strategy by using multi-functional L-threonine (L-Thr) to significantly enhance the reversibility and stability of Zn metal anodes. It was revealed that L-Thr can not only preferentially adsorb on the Zn(101) crystal plane to form a hydrophobic layer, but also partially replace coordinated water molecules to reshape the Zn²⁺ solvation structure and reconstruct the internal hydrogen bond network between free water molecules, thereby effectively inhibiting water-related side reactions and guiding dendrite-free Zn deposition. Benefitting from the multifunctional nature of L-Thr, the Zn//Zn symmetric cell exhibited an impressive lifespan, operating for over 2500 hours at 0.5 mA cm⁻² and 0.5 mAh cm⁻². Even under harsh conditions of 10.0 mA cm⁻² and 10.0 mAh cm⁻², the Zn//Zn symmetric cell retained stable cycling for 600 hours, equivalent to a high cumulative plated capacity of 6 Ah cm⁻². Surprisingly, the Zn anode with 68.3% Zn utilization also can be operated over 240 h. Additionally, the Zn stripping/plating CE reached as high as 99.71% at a current density of 2 mA cm⁻² for over 2000 cycles. Considering practical applications, the Zn//NVO full cell demonstrated excellent cycling stability, retaining 80.9% capacity after 1000 cycles and 77.9% after 2000 cycles at 5 A g⁻¹. Even under harsh conditions—a low N/P ratio of 2.94:1 and a high mass loading of 12.03 mg cm⁻²—the Zn//NVO pouch cell delivered a high specific capacity of 129.3 mAh g⁻¹, retaining 85.4% capacity after 200 cycles. Our findings demonstrate the effectiveness of L-Thr as a multifunctional additive, providing crucial insights for advancing high-performance AZMBs toward commercial viability.

Data availability

Source data are provided with this paper and are also available from the corresponding authors upon request.

Author contributions

K. Liu, D. Xu and Z. Chang conceived the idea. K. Liu conducted the experiments and prepared the manuscript. L. Bai, D. Xu and Z. Chang supervised the project and revised the manuscript. The other authors contributed to data collection, analyses, and final manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant no. 92372201 and 22209208), National Key R&D Program of China (Program No. 2024YFB3814200), and High Performance Computing Center of Central South University.

Notes and references

1. M. Song, H. Tan, D. L. Chao and H. J. Fan, Adv. Funct. Mater., 2018, 28, 1802564 CrossRef .
2. G. J. Li, L. Sun, S. L. Zhang, C. F. Zhang, H. Y. Jin, K. Davey, G. M. Liang, S. L. Liu, J. F. Mao and Z. P. Guo, Adv. Funct. Mater., 2024, 34, 2214538 CrossRef .
3. J. N. Hao, X. L. Li, X. H. Zeng, D. Li, J. F. Mao and Z. P. Guo, Energy Environ. Sci., 2020, 13, 3917–3949 RSC .
4. Q. Zhang, D. Lei, N. Liu, Z. Liu, Z. Ren, J. Yin, P. Jia, W. Lu and Y. Gao, Adv. Mater., 2022, 34, 220569 Search PubMed .
5. J. Guan, H. W. Fu, A. M. Rao, J. Zhou, J. Q. Yu, Z. X. Tang, X. M. Yuan, X. Z. Yu and B. A. Lu, Interdiscip. Mater., 2025, 4, 162–174 CAS .
6. S. Xu, X. H. Yi, L. Fan and B. A. Lu, Batteries Supercaps, 2025, 8, e202400641 CrossRef CAS .
7. L. T. Hu, P. Xiao, L. L. Xue, H. Q. Li and T. Y. Zhai, EnergyChem, 2021, 3, 100052 CrossRef CAS .
8. D. L. Chao, W. H. Zhou, F. X. Xie, C. Ye, H. Li, M. Jaroniec and S. Z. Qiao, Sci. Adv., 2020, 6, eaba4098 CrossRef CAS PubMed .
9. W. H. Yang, Y. Yang, H. J. Yang and H. S. Zhou, ACS Energy Lett., 2022, 7, 2515–2530 CrossRef CAS .
10. Y. Q. Lv, Y. Xiao, L. T. Ma, C. Y. Zhi and S. M. Chen, Adv. Mater., 2022, 34, 2106409 CrossRef CAS .
11. P. C. Liang, J. Yi, X. Y. Liu, K. Wu, Z. Wang, J. Cui, Y. Y. Liu, Y. G. Wang, Y. Y. Xia and J. J. Zhang, Adv. Funct. Mater., 2020, 30, 1908528 CrossRef CAS .
12. Z. Li, M. Y. Su, Z. D. Guo, Y. Y. Mao, J. Y. Li, Z. Gong, Y. O. Wang, K. Zhu and D. X. Cao, Chem. Eng. J., 2024, 502, 157601 CrossRef CAS .
13. J. W. Zhao, J. Zhang, W. H. Yang, B. B. Chen, Z. M. Zhao, H. Y. Qiu, S. M. Dong, X. H. Zhou, G. L. Cui and L. Q. Chen, Nano Energy, 2019, 57, 625–634 CrossRef CAS .
14. C. Huang, X. Zhao, Y. S. Hao, Y. J. Yang, Y. Qian, G. Chang, Y. Zhang, Q. L. Tang, A. P. Hu and X. H. Chen, Energy Environ. Sci., 2023, 16, 1721–1731 RSC .
15. B. Li, Y. Zeng, W. S. Zhang, B. A. Lu, Q. Yang, J. Zhou and Z. X. He, Sci. Bull., 2024, 69, 688–703 CrossRef CAS PubMed .
16. J. D. Wang, B. Zhang, Z. Cai, R. M. Zhan, W. Y. Wang, L. Fu, M. T. Wan, R. Xiao, Y. T. Ou, L. Wang, J. J. Jiang, Z. W. Seh, H. Li and Y. M. Sun, Sci. Bull., 2022, 67, 716–724 CrossRef CAS .
17. R. Song, M. Q. Zhao, S. Wang, Y. Lu, X. B. Bao, Q. M. Luo, L. Gou, X. Y. Fan and D. L. Li, Acta. Chim. Sin., 2024, 82, 426–434 CrossRef CAS .
18. H. L. Shen, M. Dai, Z. Li, Y. N. Tian, T. W. Li and X. W. Hu, Prog. Chem., 2024, 36, 724–740 CAS .
19. Z. M. Zhao, J. W. Zhao, Z. L. Hu, J. D. Li, J. J. Li, Y. J. Zhang, C. Wang and G. L. Cui, Energy Environ. Sci., 2019, 12, 1938–1949 RSC .
20. K. Wu, J. H. Huang, J. Yi, X. Y. Liu, Y. Y. Liu, Y. G. Wang, J. J. Zhang and Y. Y. Xia, Adv. Energy Mater., 2020, 10, 1903977 CrossRef CAS .
21. H. Y. Qin, W. Kuang, D. Huang, X. Y. Zhang, J. Liu, L. Yi, F. Shen, Z. W. Wei, Y. P. Huang, J. Xu and H. B. He, J. Mater. Chem. A, 2022, 10, 17440–17451 RSC .
22. Z. Cai, Y. T. Ou, J. D. Wang, R. Xiao, L. Fu, Z. Yuan, R. M. Zhan and Y. M. Sun, Energy Storage Mater., 2020, 27, 205–211 CrossRef .
23. P. Xiao, H. B. Li, J. Z. Fu, C. Zeng, Y. H. Zhao, T. Y. Zhai and H. Q. Li, Energy Environ. Sci., 2022, 15, 1638–1646 RSC .
24. J. Chen, A. M. Rao, C. T. Gao, J. Zhou, L. M. Cha, X. M. Yuan and B. A. Lu, Nano Res., 2024, 17, 9671–9678 CrossRef CAS .
25. G. Wu, R. J. Zhu, W. H. Yang, Y. Yang, J. Okagaki, Z. Y. Lu, J. M. Sun, H. J. Yang and E. Yoo, Adv. Funct. Mater., 2024, 34, 2316619 CrossRef CAS .
26. F. L. Zhang, F. Y. Huang, R. Z. Huang, N. Dong, S. H. Jiao, R. G. Cao and H. L. Pan, Sci. China Mater., 2023, 66, 982–991 CrossRef CAS .
27. F. Wang, O. Borodin, T. Gao, X. L. Fan, W. Sun, F. D. Han, A. Faraone, J. A. Dura, K. Xu and C. S. Wang, Nat. Mater., 2018, 17, 543–549 CrossRef CAS PubMed .
28. J. Wang, J. X. Tian, G. X. Liu, Z. Z. Shen and R. Wen, Small Methods, 2023, 7, 2300392 CrossRef CAS .
29. P. Cui, T. T. Wang, Z. H. Wang, H. S. Geng, P. C. Song, F. Hu, J. H. You and K. Zhu, Chem. Eng. J., 2024, 500, 156971 CrossRef CAS .
30. W. H. Yang, X. F. Du, J. W. Zhao, Z. Chen, J. J. Li, J. Xie, Y. J. Zhang, Z. L. Cui, Q. Y. Kong, Z. M. Zhao, C. G. Wang, Q. C. Zhang and G. L. Cui, Joule, 2020, 4, 1557–1574 CrossRef CAS .
31. W. S. Lv, Y. Tan, C. Y. Guo, X. He, L. X. Zeng, J. L. Zhu, L. Yang, Z. J. Chen, X. C. Yin, J. Xu and H. B. He, Adv. Energy Mater., 2025, 15, 2403689 CrossRef CAS .
32. T. Li, A. Naveed, J. Zheng, B. Chen, M. Jiang, B. Liu, Y. Zhou, X. Li, M. Su, R. Guo, J. Sumner, C. C. Li and Y. Liu, Angew. Chem., Int. Ed., 2025, 64, e202424095 CrossRef CAS .
33. B. G. Ye, F. Wu, R. Zhao, H. H. Zhu, M. G. Lv, X. M. Han, T. D. Chen, X. R. Wang, Y. Bai and C. Wu, Adv. Mater., 2025, 37, 2501538 CrossRef CAS PubMed .
34. X. Guo, S. C. Zhang, H. Hong, S. X. Wang, J. X. Zhu and C. Y. Zhi, iScience, 2025, 28, 111751 CrossRef CAS PubMed .
35. J. Cui, X. Y. Liu, Y. H. Xie, K. Wu, Y. Q. Wang, Y. Y. Liu, J. J. Zhang, J. Yi and Y. Y. Xia, Mater. Today Energy, 2020, 18, 100563 CrossRef CAS .
36. Y. L. Zhu, Q. R. Chen, J. N. Hao and Y. Jiao, J. Mater. Chem. A, 2024, 12, 20097–20106 RSC .
37. P. J. Wang, T. C. Li, Y. F. Liu, C. J. Lin, Y. F. Cui, H. B. Song, B. G. Lu, S. Q. Liang, H. Y. Yang and J. Zhou, Angew. Chem., Int. Ed., 2025, 64, e202422547 CrossRef CAS .
38. M. Yang, S. S. Bi, X. Wang, Y. F. Zhang, Y. Hu, J. L. Tian, L. L. Liu and Z. Q. Niu, J. Am. Chem. Soc., 2025, 147, 9294–9303 CrossRef CAS PubMed .
39. Y. Liu, Y. K. An, L. Wu, J. G. Sun, F. Y. Xiong, H. Tang, S. L. Chen, Y. Guo, L. Zhang, Q. Y. An and L. Q. Mai, ACS Nano, 2023, 17, 552–560 CrossRef CAS .
40. H. Ren, S. Li, L. Xu, L. Wang, X. Liu, L. Wang, Y. Liu, L. Zhang, H. Zhang, Y. Gong, C. Lv, D. Chen, J. Wang, Q. Lv, Y. Li, H. Liu, D. Wang, T. Cheng, B. Wang, D. Chao and S. Dou, Angew. Chem., Int. Ed., 2025, 64, e202423302 CrossRef CAS .
41. C. Huang, X. Zhao, S. Liu, Y. S. Hao, Q. L. Tang, A. P. Hu, Z. X. Liu and X. H. Chen, Adv. Mater., 2021, 33, 2100445 CrossRef CAS .
42. W. H. Yang, R. J. Zhu, G. Wu, Y. Yang, H. J. Yang and E. Yoo, Energy Storage Mater., 2024, 65, 103153 CrossRef .
43. M. Shi, C. J. Lei, H. J. Wang, P. J. Jiang, C. Xu, W. Yang, X. He and X. Liang, Angew. Chem., Int. Ed., 2024, 63, e202407261 CrossRef CAS .
44. F. H. Cui, D. S. Wang, F. Hu, X. Yu, C. Guan, G. H. Song, F. Xu and K. Zhu, Energy Storage Mater., 2022, 44, 197–205 Search PubMed .
45. C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158–6170 CrossRef CAS .
46. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104 CrossRef .
47. T. Clark, J. Chandrasekhar, G. W. Spitznagel and P. V. R. Schleyer, J. Comput. Chem., 1983, 4, 294–301 CrossRef .
48. T. Lu and F. W. Chen, J. Comput. Chem., 2012, 33, 580–592 CrossRef CAS .
49. G. Kresse and J. Furthmuller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186 CrossRef CAS PubMed .
50. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775 Search PubMed .
51. P. E. Blochl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979 CrossRef PubMed .
52. S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456–1465 CrossRef CAS PubMed .
53. J. Sandoval-Lira, G. Mondragón-Solórzano, L. Lugo-Fuentes and J. Barroso-Flores, J. Chem. Inf. Model., 2020, 60, 1445–1452 Search PubMed .
54. N. Hu, J. Tao, Y. Tan, H. W. Song, D. Huang, P. G. Liu, Z. J. Chen, X. C. Yin, J. L. Zhu, J. Xu and H. B. He, Adv. Energy Mater., 2024, 14, 2404018 CrossRef CAS .
55. T. C. Li, Y. Lim, X. L. Li, S. Z. Luo, C. J. Lin, D. L. Fang, S. W. Xia, Y. Wang and H. Y. Yang, Adv. Energy Mater., 2022, 12, 2103231 CrossRef CAS .
56. G. F. Zeng, Q. Sun, S. Horta, P. R. Martínez-Alanis, P. Wu, J. Li, S. Wang, M. Ibáñez, Y. H. Tian, L. J. Ci and A. Cabot, Energy Environ. Sci., 2025, 18, 1683–1695 RSC .
57. H. B. Bi, X. S. Wang, H. L. Liu, Y. L. He, W. J. Wang, W. J. Deng, X. L. Ma, Y. S. Wang, W. Rao, Y. Q. Chai, H. Ma, R. Li, J. T. Chen, Y. P. Wang and M. Q. Xue, Adv. Mater., 2020, 32, 2000074 CrossRef CAS .
58. B. J. Wang, R. Zheng, W. Yang, X. Han, C. Y. Hou, Q. H. Zhang, Y. G. Li, K. R. Li and H. Z. Wang, Adv. Funct. Mater., 2022, 32, 2112693 CrossRef CAS .
59. Q. Kang, H. Y. Liang, Z. M. Zhu, F. X. Xing, B. Z. Zhu, Y. M. Li, S. H. Wang, S. D. Tang, H. Li, L. Pan, L. J. Yang, T. M. Mo and Y. F. Bu, Angew. Chem., Int. Ed., 2025, 64, e202420183 CrossRef CAS .
60. N. Hu, W. S. Lv, H. Tang, H. Y. Qin, Y. H. Zhou, L. Yi, D. Huang, Z. R. Wu, J. Liu, Z. J. Chen, J. Xu and H. B. He, J. Mater. Chem. A, 2023, 11, 14921–14932 RSC .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta05047b

This journal is © The Royal Society of Chemistry 2025