Enhancing ammonium-ion storage in Mo-doped VO₂ (B) nanobelt-bundles anode for aqueous ammonium-ion batteries†

Long Chen , Jie Zhang , Zuoshu Wang and Dewei Wang *
College of Materials Science and Engineering, North Minzu University, Yinchuan 750021, People's Republic of China. E-mail: wangdewei@yeah.net; Tel: +86 951 2067378

First published on 13th June 2024

---

---

Introduction

The depletion of traditional fossil fuels, climate warming, and environmental pollution have made energy a pressing concern for mankind. Consequently, the exploitation of environmentally friendly, efficacious, and safe energy storage devices has become an urgent necessity.^1–4 Particularly, in the transition towards a renewable energy economy, the exploration of novel electrode materials for efficient and sustainable energy storage systems is of paramount importance. With the increasing demand for energy storage solutions that can complement intermittent renewable energy sources, research in this area has gained significant momentum. Lithium-ion batteries (LIBs) have been industrially successful owing to their high energy density and high-power density, but LIBs have been limited from further development in the status of large-scale commercialization due to cost issues and safety concerns.^5,6 Therefore, in recent years, aqueous rechargeable batteries have gradually become a hotspot for researchers because of their environmental sustainability, and high safety.^7–10 Currently, most aqueous batteries employ metal ions as carriers, including K⁺, Na⁺, Li⁺, Mg²⁺, Zn²⁺, Ca²⁺, and Al³⁺. Large metal ions can cause considerable expansion and contraction of electrode materials during repeated insertion and deinsertion.^11,12 This phenomenon can lead to structural damage, fatigue, and decreased stability of the electrode materials, ultimately affecting the overall battery lifespan.^13,14 Therefore, it is necessary to seek alternative charge carriers that are both environmentally friendly and cost-effective.

The non-metallic NH₄⁺ serves as a promising charge carrier, possessing unique characteristics that are unattainable by metal ions. This is because NH₄⁺ are abundant in nature, lightweight (minimum molar mass 18 g mol⁻¹), fast diffusing (smaller hydrated ion size of 3.31 Å), and less corrosive and hydrogen evolution than other cations.^15–19 These excellent properties enable it to be an ideal charge carrier. In view of this, a series of AAIBs have been gradually reported in the literature in recent years. However, previous studies on electrode materials for AAIBs mainly focused on cathode materials, only very limited anode materials have been developed. Since the pioneering work on the NH₄⁺ storage in organic solid, 3,4,9,10-perylene tetracarboxylic diimide (PTCDI),²⁰ significant endeavors have been devoted to developing high-capacity anodes for NH₄⁺ storage, including h-MoO₃,²¹h-WO₃,²² organic materials,^23,24 and other inorganic compounds.^25,26 For example, Long et al. introduced polar surface-terminated Ti₃C₂T_x MXene as a conductive backbone into perylene-3,4,9,10-tetracarboxylic acid dianhydride (PTCDA) to address its poor electrical conductivity and low capacitance during NH₄⁺ insertion/extraction as well as low conductivity and dissolution problems during the charge–discharge process.²⁷ Different from the layered Ti₃C₂T_x MXene, h-MoO₃ has a tunnel-like structure, in which the ammonification/deamidation mechanism is governed by non-diffusion pseudocapacitive behavior, was synthesized by Zhi and co-workers.²¹ It exhibits ultra-long and stable cycling performance up to 100000 cycles with 94% capacity retention. Nevertheless, the existing anode materials still encounter challenges such as limited NH₄⁺ storage capacity and the absence of a consistent voltage plateau. Consequently, the majority of anode materials reported for AAIBs have not been able to meet the simultaneous requirements for high capacity and stable working voltage.

Vanadium-based materials have become increasingly favored by researchers compared to other anode materials due to their structurally stable and versatile properties, as well as their promising ability to store various cations.^28,29 Among these materials, VO₂ (B), which is widely recognized as one of the most popular vanadium-based materials in energy storage, exhibits significant potential as a NH₄⁺ host.³⁰ This is because it possesses desirable characteristics such as a suitable structure, applicable specific capacity, and a large lattice spacing that facilitates the insertion and extraction of NH₄⁺. For instance, Zhang and Dong's research groups independently demonstrated that the defective VO₂ (B) with abundant oxygen vacancy is a propitious anode material for NH₄⁺ storage, where the largest size of the tunnel-like structure (4.18 and 3.59 Å) will afford a smooth transport pathway for NH₄⁺. Consequently, it can present a large reversible specific capacity (200–240 mA h g⁻¹ at 0.1 A g⁻¹) and good long-term cycling durability (70–80% capacity retention capacity retention rate of 1000 cycles).^31,32 Usually, an additional electrolyte additive polyvinyl alcohol is required to slow down the dissolution of VO₂ (B) electrode. Moreover, sluggish kinetics and rapid capacity decay result from vanadium dissolution-induced structural deterioration, and poor conductivity still hinders its further application in aqueous NH₄⁺ storage. Thus, the direct application of VO₂ (B) in dilute aqueous electrolyte for NH₄⁺ storage is still a significant challenge. Recently, the application of VO₂ (B) electrodes in rechargeable aqueous magnesium-ion batteries and zinc-ion batteries has also been developed.^33,34 It is acknowledged that the designing of novel microstructures combined with defect engineering (e.g., oxygen vacancy, heteroatoms doping) can manage the poor structure stability and strengthen the conductivity as well as increase active sites, and thus the overall electrochemical performance has been improved.^29,35,36 Especially, heteroatom doping has gained significant attention for improving the intrinsic electron configuration and reestablishing the balance of charge distribution. By integrating heteroatoms into the structure of VO₂ (B), the formation of a local electrical field can be optimized, thus addressing the challenges of poor conductivity and inadequate active sites.^37–39 However, as far as we know, there is currently no report available on the utilization of heteroatoms doped VO₂ (B) in aqueous NH₄⁺ storage.

Based on the above discussion, in the present work, through a facile hydrothermal route, Mo-doped VO₂ (B) with unique nanobelt-bundles (denoted as MVO NBBs) structure was designed and produced with commercial V₂O₅ as a precursor. The energy storage mechanism and charge density distributions of the resulting MVO NBBs during NH₄⁺ insertion/extraction are investigated in detail. In particular, MVO delivers an ultrahigh reversible specific capacity of 283.5 mA h g⁻¹ at 0.3 A g⁻¹ within the potential window of −1–0 V versus SCE, as far as we know, which is much larger than those of the previously described anode materials for NH₄⁺ storage. Moreover, it also displays excellent long-term capacity retention and good rate performance with the delivered capacity of 95.8 mA h g⁻¹ at 5 A g⁻¹, and 86.7% of the specific capacity can be retained after 4500 cycles, respectively. Benefiting from these features, the as-obtained MVO NBBs can couple with CuHCF to assemble AAIBs, which can present a large energy density of 57.9 W h kg⁻¹ and a long cycle life of 81.5% capacity retention after 1000 cycles in 0.5 M (NH₄)₂SO₄ electrolyte.

Experimental

Reagents

Vanadium pentoxide (V₂O₅), oxalic acid (H₂C₂O₄), sodium molybdate (Na₂MoO₄·2H₂O), and ammonium sulfate ((NH₄)₂SO₄) were obtained from Sinopharm Chemical Reagent Co., Ltd. All the reagents were analyzed pure grade and could be used directly without additional purification.

Synthesis of MVO NBBs

MVO NBBs were synthesized by a simple one-step hydrothermal method. Typically, 2 mmol of V₂O₅ powder was dispersed in 15 ml H₂O to form suspension A, and 3 mmol H₂C₂O₄ was dissolved into 20 ml H₂O to form solution B. Afterwards, the solution C was collected by blending suspension A and solution B homogeneously under magnetically stirred for 15 min. Subsequently, 0.2 mmol of Na₂MoO₄·2H₂O (the corresponding Mo/V ratio is 5 mol%) was added into solution C and magnetically stirred for another 30 min. The resulting solution was transferred to a 50 mL Teflon-lined autoclave, which was then maintained at 180 °C for 24 h. After cooling down to room temperature, the products were washed with deionized water and ethanol and further dried under vacuum at 80 °C for 12 h. For comparison, pristine VO₂ (B) (PVO) was also produced in the same conditions in the absence of Na₂MoO₄·2H₂O.

Characterization methods

The morphology of the samples was examined using scanning electron microscopy (SEM, Zeiss Sigma 500). Transmission electron microscopy (TEM, FEI Tecnai F20) coupled with energy dispersive spectroscopy (EDS) was employed to observe the microstructure and perform element analysis. The crystallographic characteristics were investigated using X-ray diffractometry (XRD, LabX XRD-6100). Raman spectra were measured using a spectrophotometer (Horiba LabRAM HR Evolution). X-ray photoelectron spectroscopy (XPS, ESCALAB Xi+) was utilized to determine the elemental composition and valence changes. Fourier transform infrared spectroscopy (FTIR) analysis was conducted on a WQF-520A spectrometer. Electron paramagnetic resonance (EPR) tests were carried out on a Bruker EMX PLUS, and specific surface area analysis (BET) was performed using an ASAP2460 instrument.

Electrochemical measurements

For the preparation of the anodes, the active substances (MVO and PVO), polytetrafluoroethylene (PTFE), and conductive carbon black were mingled in a weight fraction of 8:1:1. A slurry was formed by adding a small amount of anhydrous ethanol, which was then laminated on a titanium mesh and dried at 60 °C for 6 hours to obtain the working electrode. The mass loading of MVO or PVO is ∼2 mg cm⁻² in each electrode. The aqueous ammonium-ion rechargeable batteries (AAIBs) were constructed with MVO and PVO as the anodes, CuHCF as the cathode in 0.5 M (NH₄)₂SO₄ electrolyte. The electrochemical performance of the single electrodes and devices was evaluated using a CHI 660E electrochemical workstation, with a saturated calomel electrode as the reference electrode and a graphite rod as the counter electrode. This included galvanostatic charge/discharge (GCD), cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS) measurements. Cycle stability and galvanostatic intermittent titration technique (GITT) tests were conducted using a constant current tester (LAND CT3001, Wuhan).

Results and discussion

The overall synthetic procedure of MVO NBBs is demonstrated in Fig. 1a. During the hydrothermal conditions, the commercial V₂O₅ (V⁵⁺) was first partially reduced to VOC₂O₄ with H₂C₂O₄,³⁷ and then VOC₂O₄ was further converted to VO₂, while some of the V⁴⁺ was replaced by molybdenum ion to form Mo-doped VO₂. The microscopic morphologies of the as-synthesized MVO and PVO were observed by SEM and TEM. As presented in Fig. 1b and c, the PVO displays a typical nanobelts structure with a width of ∼70 nm and ranges in length from a few hundred nanometers to several micrometers. In the high-resolution TEM (HRTEM) image in Fig. 1d, PVO exhibits continuous and uninterrupted lattice fringes demonstrating that it has good crystallinity and few defects. In Fig. 1e, the HRTEM image indicates the lattice spacing of PVO is ∼0.35 nm, which matches with the (110) crystal plane of the monoclinic phase VO₂ (B). The selected area electron diffraction (SAED) pattern of the PVO exhibits regular bright spots denoting a single crystalline structure (Fig. 1f). As depicted in Fig. 1g and h, the as-produced MVO sample manifests the microstructure of nanobelt-bundles assembled by several thin nanobelts, implying that the topographic characteristics of the sample have obviously adjusted since the changing of surface energy after Mo substitution. From the HRTEM image shown in Fig. 1i and j, the lattice fringes become vague and dislocation, illustrating that the MVO has decreased crystalline and contains a certain number of structural defects as compared with PVO. In addition, the bright spots also exhibit deformed and twisted in the SAED pattern referring to the presence of considerable structural defects although it is still almost single crystalline in nature (Fig. 1k). The HAAD-STEM image and EDS mappings shown in Fig. 1l and m demonstrate the uniform scattering of V, Mo, and O in the MVO NBBs.

As unveiled by the XRD pattern in Fig. 2a, both PVO and MVO display similar diffraction peaks without other impurities, in line with the monoclinic VO₂ (B) phase (JCPDS no. 31-1438), implying that the doping of Mo atoms does not change the phase structure of PVO. Nevertheless, the MVO presents weaker characteristic peaks in comparison with the PVO, which reveals that the Mo substitution reduces the crystallinity of the PVO. The decreased crystallinity of MVO further confirms the presence of abundant structural defects resulting from the difference in ionic radius of Mo and V ions, which afford additional active sites for NH₄⁺ storage.^40,41 The Raman spectra provided additional confirmation of the internal structural characteristics of MVO and PVO (Fig. 2b). The primary peaks observed at ∼280, ∼405, and ∼995 cm⁻¹ are assignable to the bending and stretching vibrational of the VO bond.⁴² Conversely, the peaks detected at 483, 526, and 686 cm⁻¹ correspond to the bending vibrational of the V–O–V bonds and the two distinct stretching vibrational of the V3–O bonds.⁴³ XPS was conducted to elucidate the impact of Mo atom doping on the surface constitution and valence state information of the as-obtained PVO and MVO samples. As demonstrated in Fig. 2c, the elements of V and O were identified in the survey XPS spectra of PVO, while an additional molybdenum signal can be observed for the MVO sample (Fig. 2d), demonstrating the successful incorporation of Mo into VO₂(B) without impurities. In V 2p spectra, the peaks at 517.5 and 525.3 eV correlate to the 2p_1/2 and 2p_3/2 energy levels of V⁵⁺, while the peaks at 516.2 and 523.5 eV are ascribed to the 2p_1/2 and 2p_3/2 energy levels of V⁴⁺,^37,44 indicating a mixed oxidation state of V⁴⁺/V⁵⁺. Notably, the V⁴⁺ content shows an obvious increase after Mo doping, likely resulting from the formation of structural defects in the MVO. The measured V⁴⁺/V⁵⁺ proportion of MVO (0.74) is apparently smaller than that of PVO (0.81) through the integration peak area integral, signifying that the diminution of V⁵⁺ to V⁴⁺ attributed to the doping of Mo atoms. It is important to note that oxalic acid (H₂C₂O₄) functions as a weak reducing agent, which is incapable of fully reducing V⁵⁺ to a V⁴⁺ or lower valence state. Instead, it affects only a partial reduction of the pentavalent vanadium. Consequently, a residual amount of pentavalent vanadium must be present in the product. This observation aligns with earlier findings reported in some previous work,³⁸ yet it differs from certain studies that claim to have detected V⁴⁺ and V³⁺ in their products.³¹ The high-resolution O 1s spectra can be split into three peaks at 530, 531.5, and 532.9 eV (Fig. 2e), which are related to the V–O bond (lattice oxygen), oxygen vacancies, and absorbed oxygen, respectively.^40,45 The evidently larger integrated peak area associated with the oxygen vacancies in MVO proves that MVO comprises more oxygen vacancies than PVO. The high-resolution Mo 3d spectra of MVO clearly show two main peaks at 233.1 and 236.2 eV (Fig. 2f), which correspond to Mo 3d_5/2 and Mo 3d_3/2 of Mo⁶⁺, respectively. The minor peaks around 232.3 and 235.2 eV are assigned to Mo⁵⁺,⁴⁶ indicating that the majority of Mo ions remain in their original valence state (Mo⁶⁺), with some Mo⁶⁺ being reduced to Mo⁵⁺ under hydrothermal conditions. Furthermore, the presence of molybdenum signal was confirmed after argon ion etching for 30 seconds, suggesting that molybdenum is not only doped on the surface but also present in the bulk (Fig. S1, ESI†). The detailed atomic ratio of Mo and V in MVO is estimated to be 0.057:1 based on ICP-OES analysis. Furthermore, the presence of oxygen vacancies can be confirmed by the EPR spectrum, as the signal arises from the unpaired electrons located at the defect sites when subjected to a magnetic field. As displayed in Fig. 2g, the typical response signal at g = 2.003 in MVO exhibits a higher intensity compared to that of PVO, indicating increased oxygen vacancies in MVO, which is in line with the O 1s spectra results.⁴⁷ Next, the pore structure, crucial for the electrochemical performance with regard to ion transport and the interface of electrode/electrolyte, was investigated using nitrogen adsorption–desorption examinations. The specific surface areas of PVO and MVO were estimated to be 16.3 and 31.8 m² g⁻¹, respectively, with total pore volumes of 0.10 and 0.17 cm³ g⁻¹, respectively. Both samples exhibit type-III curves in the nitrogen adsorption–desorption isotherms (Fig. 2h). The corresponding pore size distribution curves are manifested in Fig. 2i, the content of small mesopore pores in the range of 3–5 nm was substantially enhanced for MVO in comparison with that for PVO. The expanded specific surface area and the existence of mesopores is favorable for building large electrode/electrolyte interfaces and rapid electrolyte transport and ensuring a high charge–discharge rate.⁴⁸

To evaluate the NH₄⁺ storage performance of the as-obtained PVO and MVO samples, a three-electrode system with a graphite rod, a saturated calomel electrode (SCE), and 0.5 M (NH₄)₂SO₄ solution as counter, reference electrodes, and electrolyte, respectively. The first five-cycle CV curves of MVO at 0.5 mV s⁻¹ within the potential window of −1–0 V are displayed in Fig. 3a, from which two pairs of redox peaks at around −0.1/-0.2 V and −0.6/−0.8 V can be associated with the transformation of V from V⁵⁺ to V⁴⁺. Additionally, the comparison between the first cycle and subsequent cycles showed that the CV curves maintained a similar shape, indicating the highly reversible NH₄⁺ insertion/extraction behavior. The ascendancy of MVO electrode is proved by the higher current density than that of PVO accorded with the larger capacity resulting from the increased activate sites through Mo⁶⁺ doping (Fig. 3b).

The rate performance of two electrodes was examined over a range of current densities from 0.3 to 5 A g⁻¹, with the findings presented in Fig. 3c and d. The specific capacities (derived from the last cycle at each current density) for the MVO electrode were measured as 283.5, 256.5, 219, 177.4, 145.6, 118.7, and 95.8 mA h g⁻¹ at 0.3, 0.5, 1, 2, 3, 4, and 5 A g⁻¹, respectively. For the PVO electrode, these values were 175.8, 152, 117.8, 85.2, 66.3, 54.1, and 41 mA h g⁻¹. Upon returning to a current density of 0.3 A g⁻¹, the MVO electrode retained a significant specific capacity of 229.6 mA h g⁻¹, which represents 81% of its initial value, whereas the PVO electrode retained 74%. Fig. 3e illustrates the cycling performance of PVO and MVO at a low current density of 0.5 A g⁻¹. After 300 cycles, MVO retains a specific capacity of 198.8 mA h g⁻¹ (80.6% retention), while PVO maintains a value of 45.7 mA h g⁻¹ (29.7% retention). From Fig. S2,† it is evident that MVO still retains its original nanobelt-bundles morphology, whereas the original nanobelt morphology of PVO is significantly damaged. This indicates that Mo doping can enhance the structural stability of the products. The electrochemical performance of the resulting samples is contingent on the sodium molybdate content. As depicted in Fig. S3,† even a nominal addition of sodium molybdate markedly enhances the specific capacity relative to the pure PVO, suggesting that molybdenum doping is conducive to improved electrochemical performance. Conversely, as the quantity of sodium molybdate augments, a decline in electrochemical performance is observed. As depicted in Fig. 3e, the MVO electrode demonstrates a superior capacity when compared at the same current density to previous anode materials for NH₄⁺ storage, including defective VO₂ (B),^31,32h-MoO₃,²¹h-WO₃,²² MoO_x,⁴⁶ barium vanadate,⁴⁹ PTCDA/Ti₃C₂T_x MXene,²⁷ and synthetic polymer materials.⁵⁰ As illustrated in Fig. 3f, the electrochemical performance of the MVO is considerably restrained after annealing at 500 °C for 2 hours (Fig. S4†), probably due to the elimination of crystalline water from the MVO, which hinders the diffusion of NH₄⁺. Apart from its remarkable specific capacity, the MVO electrode also demonstrates impressive long-term cyclability (Fig. 3g), exhibiting approximately 86.7% capacity preservation and ∼100% coulombic efficiency after 4500 cycles at 3 A g⁻¹. In sharp contrast, only 17.2% capacity retention is obtained for PVO, demonstrating that Mo doping can not only increase the specific capacities but also help maintain good structural stability of the MVO electrode. After cycling tests, the concentration of V ions in the electrolyte for the MVO electrode after 4500 cycles was 19.2 mg L⁻¹ (ppm), significantly lower than for the PVO electrode (51.0 mg L⁻¹) according to ICP-OES results (Fig. S5†). This indicates that Mo ions doping can improve structural stability and restrain the dissolution of V ions, and consequently a good cycle stability performance of the MVO electrode. In addition, from the XRD patterns obtained after cycling (Fig. S6†), it can be observed that MVO retains its original crystal structure, while the weak diffraction peaks of PVO indicate that its crystal structure has been disrupted.

To attain a deeper insight into the electrochemical reaction kinetics of MVO, the CV curves at various sweep rates, as shown in Fig. 4a. It is observed that the oxidation and reduction peaks experience only a slight increase and decrease in voltage, respectively, as the scan rate increases, suggesting a small polarization effect. The relationship between scan rate (v) and peak current (i) is linear and can be represented by the equation i = aν^b, where a and b are constants.^51–53 A b-value of 0.5 indicates a diffusion-controlled process for charge storage, while a b-value of 1.0 signifies capacitance-controlled behavior. Fig. 4b displays the b-values for MVO at different peak currents, which are 0.76, 0.57, 0.53, and 0.79, respectively. For PVO, the b-values are 0.74, 0.55, 0.50, and 0.73 (Fig. S7†). This suggests that both MVO and PVO are influenced by a combination of diffusion-controlled insertion and surface-capacitive processes. Furthermore, the relative contributions of the surface-capacitive and diffusion-controlled process can be determined using the equation i(v) = k₁v + k₂v^1/2, where k₁v depicts the capacitance contribution and k₂v^1/2 illustrates the diffusion contribution at a fixed voltage.^51,54,55 The results depicted in Fig. 4c and d demonstrate that at a sweep rate of 0.1 mV s⁻¹, the capacitive contribution corresponds to 35.9%. With an increase in the sweep rate from 0.1 to 0.5 mV s⁻¹, the surface capacitive processes contribute to 35.9–73.4% of the overall capacities. This finding suggests the prominent role of surface capacitive processes in the electrochemical reaction and their significant contribution to the superior rate performance of MVO. Electrochemical impedance spectroscopy (EIS) was applied to investigate the spread kinetics of NH₄⁺. The Nyquist plots in Fig. 4e and f demonstrate that the semicircular diameter of the mid-frequency part is smaller for the MVO compared to the PVO, with a charge-transfer resistance (R_ct) of 1.8 Ω. Additionally, the MVO exhibits a smaller intercept, indicating a relatively low internal resistance (R_s) of 2.0 Ω. This suggests that Mo doping increases the disorder of structural changes on the surface of VO₂ (B), leading to a bigger specific surface area and increased porosity. These factors contribute to faster ion diffusion kinetics in MVO. Moreover, the slope of the MVO electrode is smaller compared to that of the PVO electrode, indicating a larger value of Warburg coefficient (σ) for MVO (Fig. 4g).^56,57 This suggests that MVO exhibits a better ion diffusion rate. The ion diffusion rate of MVO and PVO samples in (NH₄)₂SO₄ electrolyte was studied further using the constant current intermittent titration technique (GITT).⁵⁸Fig. 4h and Fig. S8† illustrate the log (D cm⁻² s⁻¹) and E (V) variation of ion diffusion coefficients for both samples over the entire voltage range. The D value for MVO ranges from 5.4 × 10⁻¹⁰ cm⁻² s⁻¹ to 4.8 × 10⁻¹² cm⁻² s⁻¹, while the D value for PVO ranges from 2.3 × 10⁻¹⁰ cm⁻² s⁻¹ to 4.7 × 10⁻¹⁴ cm⁻² s⁻¹. This suggests that MVO demonstrates a significantly faster ion diffusion rate in (NH₄)₂SO₄ electrolyte. The excellent electrochemical performance of MVO can be related to the substitution of Mo ions, leading to a more robust tunneling structure, and the introduction of oxygen defects, promoting ion transfer rates and increasing active sites for ion storage. By examining the EIS of various samples at different temperatures (Fig. S9†), it can be observed from Fig. 4i that the activation energy (E_a) of the MVO electrode is 9.6 kJ mol⁻¹, which is lower than that of the PVO electrode (11.1 kJ mol⁻¹). This indicates that doping with Mo ions reduces the activation energy, resulting in lower energy for the NH₄⁺ insertion/extraction process.

To explore the energy storage mechanism of NH₄⁺ on the MVO anode, several ex situ characterization techniques were employed to analyze the structural changes during charging and discharging. Fig. S10† depicts the GCD curves of MVO in selected states. The XRD patterns displayed in Fig. S11† and the magnified peaks of the (110) and (−311) crystal planes depicted in Fig. 5a demonstrate that MVO retains its monoclinic structure during the NH₄⁺ insertion/extraction process. In the initial state, these peaks appear at 2θ = 25.0° and 2θ = 33.4°, respectively. During discharge, NH₄⁺ insertion causes a shift in the peaks due to hydrogen bonding formation between ammonium ions and O atoms,^31,59 leading to a contraction of the crystal structure. Subsequent charging facilitates the gradual recovery of the structural contraction as NH₄⁺ is extracted from MVO. Upon charging to 0 V, the characteristic peaks revert to their original positions, denoting the reversible NH₄⁺ insertion/deinsertion behavior of MVO. XPS measurements were executed to prove the change in valence state during the NH₄⁺ insertion/extraction process. Fig. 5b illustrates the V 2p spectrum, which indicates several sharp peaks categorized into V⁴⁺ (516.1 and 523.2 eV) and V⁵⁺ (517.5 and 525.3 eV). After full discharge, the V⁴⁺/V⁵⁺ ratio increased from 0.81 to 1.67, while V⁴⁺ increased significantly. However, after full charging, the ratio decreased again to 1.14 following NH₄⁺ extraction. This suggests a highly reversible charge transfer reaction, indicating that MVO is a suitable electrode material for NH₄⁺ storage. Furthermore, the O 1s spectra depicted three major components associated with V–O, oxygen vacancy, and H–O at 530 eV, 531.5 eV, and 532.9 eV, respectively (Fig. 5c). The oxygen vacancy content was estimated to be around 46.1%, which can act as binding sites for NH₄⁺ storage and provide to improving the specific capacity.^7,60 After discharge, NH₄⁺ insertion weakened the intensity of oxygen defects, which were restored during charging, confirming their interaction with NH₄⁺. Fig. 5d illustrates that no N 1s signal peaks were detected in the initial state. However, after full discharge, two peaks of N–H and N–H–O at 399.9 eV and 401.8 eV, respectively, were detected.⁶¹ The obvious increased intensity of N–H peak during discharge indicates NH₄⁺ insertion into MVO, and the formation of N–H–O signals suggests an interaction between NH₄⁺ and lattice oxygen of MVO, owing to the formation of hydrogen bond between NH₄⁺ and MVO. Fig. 5e illustrates the changes in the Mo 3d peak. Initially, the Mo ion in the MVO compound exists as Mo⁶⁺ and Mo⁵⁺. Upon full discharge, the Mo 3d_3/2 peak shifts to lower binding energy, accompanied by a small new peak, indicating the presence of Mo⁴⁺. When fully charged, the Mo⁴⁺ disappears, and some Mo⁵⁺ converts back to Mo⁶⁺. These findings suggest that molybdenum likely plays a role in the electrochemical reactions and contributes to the overall capacity. Furthermore, FT-IR analysis (Fig. 5f) verifies the presence of hydrogen bonding between NH₄⁺ and the V–O backbone. During the discharge process, H atoms combine with O atoms to establish hydrogen bonding after NH₄⁺ enters the tunneling structure of MVO, resulting in four new characteristic peaks. The peak at 3170 cm⁻¹ arises from the stretching vibration of the nitrogen–hydrogen bonds, whereas the band at 1394 cm⁻¹ corresponds to the bending vibration of these bonds.^59,60 Furthermore, the peaks observed at 2997 cm⁻¹ and 1007 cm⁻¹ can be attributed to the hydrogen atoms of NH₄⁺ bonded to oxygen atoms on the MVO backbone.^32,62 The reversible insertion and extraction process of NH₄⁺ in MVO is outlined in Fig. S12.† These findings indicate the efficient and reversible storage of non-metallic NH₄⁺ in MVO, thereby enabling excellent energy storage properties.

Density functional theory (DFT) calculations were engaged to further understanding of the NH₄⁺ storage mechanism in PVO and MVO. Compared with the PVO, the introduction of Mo-dopant does not distort the original structure (Fig. 5g and Fig. S13†). As presented in Fig. 5h and Fig. S14,† hydrogen bond can be formed between the H atom in NH₄⁺ and the O atom in PVO/MVO. The differential charge density analysis, comparing the two electrodes when a single NH₄⁺ is inserted into the tunnel structure, is presented in Fig. 5i and S14† respectively. In PVO, electrons accumulate around the NH₄⁺ (indicated by yellow regions), which signifies a strong interaction between NH₄⁺ and O²⁻. In MVO, a notable rearrangement of charge around the NH₄⁺ is observed, with more electrons accumulating compared to PVO. This could be a contributing factor to the increased NH₄⁺ storage capacity in MVO. The calculation results indicate that the insertion of a single NH₄⁺ in the PVO tunnel structure has strong adsorption of −1.41 eV, which decreases to −1.82 eV upon Mo doping (Fig. 5j). This suggests that Mo doping significantly enhances the adsorption capacity of the tunnel structure for NH₄⁺, thereby improving the exceptional NH₄⁺ storage performance of MVO. Additionally, there is more electron transfer between NH₄⁺ and MVO compared to PVO, further confirming the strong interaction between NH₄⁺ and O²⁻. This is consistent with the findings from the differential charge density analysis.

To evaluate the practical application, AAIB full cell was garnered with MVO as an anode and CuHCF as a cathode in 0.5 M (NH₄)₂SO₄ electrolyte. The structural characterization and electrochemical performance of CuHCF are presented in Fig. S15.† According to the CV curves of CuHCF cathode and MVO anode, as displayed in Fig. 6a, the voltage window of CuHCF//MVO AAIB can be theoretically extended to 2.0 V. As demonstrated in Fig. S16,† the working voltage window can be steadily bordered to 2.0 V without obvious polarization, and the coulombic efficiency above 95% at various current densities. The CV curves of CuHCF//MVO and CuHCF//PVO AAIBs at a scan rate of 1 mV s⁻¹ are compared in Fig. 6b. It can be observed that CuHCF//MVO demonstrates a larger area, indicating a higher specific capacity compared to CuHCF//PVO. Additionally, Fig. 6c displays the CV curves of CuHCF//MVO devices at different scan rates ranging from 0.2 to 1 mV s⁻¹, which all exhibit obvious redox peaks, indicating battery characteristics. Moreover, Fig. 6d and e show the GCD curves of the CuHCF//MVO devices, revealing specific capacities at various current densities. At current densities of 0.3, 0.5, 1, 3, and 5 A g⁻¹, the specific capacities for CuHCF//MVO devices are recorded as 70.6, 64.6, 57.5, 41.2, and 31.7 mA h g⁻¹, respectively. It is noteworthy that the Ragone plots in Fig. 6f display the power and energy densities of the CuHCF//MVO and CuHCF//PVO AAIBs. The CuHCF//MVO device reaches a maximum energy density of 57.9 W h kg⁻¹ and a maximum power density of 3884 W kg⁻¹. Additionally, after 1000 cycles at 3 A g⁻¹, the CuHCF//MVO device retains 81.5% of its specific capacity (Fig. 6g), demonstrating superior cycling stability compared to the CuHCF//PVO AAIB (only 2.7%). Lastly, a practical application is demonstrated in Fig. 6h, where a “VO₂”-shaped LED panel consisting of 60 LEDs connected in series is lit up for over five minutes after charging CuHCF//MVO device to 2.0 V (Video S1†), further validating its suitability for portable/wearable electronics. surpassing other previously reported NH₄⁺ storage anode materials.

Conclusions

In conclusion, our study has successfully demonstrated the effectiveness of doping VO₂ with Mo⁶⁺ ions as a means to enhance the NH₄⁺ storage capacity. Through a one-step hydrothermal reaction, Mo substituted VO₂ with a thin nanobelt-bundles morphology was synthesized, thereby creating more active sites and facilitating fast NH₄⁺ transport. This microstructure design resulted in excellent rate capabilities and cycling properties. The MVO anode displayed outstanding electrochemical performance, showing a remarkable specific capacity of 283.5 mA h g⁻¹ at 0.3 A g⁻¹. Furthermore, it exhibited notable cycling stability at 3 A g⁻¹, with a capacity retention of 86.7% after 4500 cycles. These results highlight the potential of efficiently designing microstructures and modulating electronic structures to improve the NH₄⁺ storage performance in vanadium oxides. Overall, our findings contribute to the field of NH₄⁺ storage materials and give assistance to further advancements in the development of high-performance rechargeable batteries. The strategy of doping VO₂ with transition metal ions opens new possibilities for enhancing the capabilities of energy storage devices.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors are grateful to the financial supports from the National Natural Science Foundation of China (Grant No. 22365001) and Jiajia Ye from SCI-GO (https://www.sci-go.com) for DFT theoretical calculations.

References

1. T. J. Quill, G. LeCroy, D. M. Halat, R. Sheelamanthula, A. Marks, L. S. Grundy, I. McCulloch, J. A. Reimer, N. P. Balsara, A. Giovannitti, A. Salleo and C. J. Takacs, Nat. Mater., 2023, 22, 362 CrossRef CAS PubMed .
2. H. W. Park and K. C. Roh, J. Power Sources, 2023, 557, 232558 CrossRef CAS .
3. M. S. Javed, A. Mateen, I. Hussain, S. Ali, S. Asim, A. Ahmad, E. T. Eldin, M. A. Bajaber, T. Najam and W. H. Han, Chem. Eng. J., 2023, 452, 139455 CrossRef CAS .
4. S. F. Kuchena and Y. Wang, Chem. – Eur. J., 2021, 27, 15450–15459 CrossRef PubMed .
5. S. S. Shah, S. M. Abu Nayem, N. Sultana, A. J. S. Ahammad and M. A. Aziz, ChemSusChem, 2022, 15.1, e202101282 CrossRef PubMed .
6. V. Sedajova, A. Bakandritsos, P. Blonski, M. Medved, R. Langer, D. Zaoralova, J. Ugolotti, J. Dzibelova, P. Jakubec, V. Kupka and M. Otyepka, Energy Environ. Sci., 2022, 15, 740–748 RSC .
7. D. Yang, Y. Song, M. Y. Zhang, Z. M. Qin, J. Liu and X. X. Liu, Angew. Chem., 2022, 134, e202207711 CrossRef .
8. D. Lin and Y. Li, Adv. Mater., 2022, 34, 2108856 CrossRef CAS PubMed .
9. Z. Bao, C. Lu, Q. Liu, F. Ye, W. Li, Y. Zhou, L. Pan, L. Duan, H. Tang, Y. Wu, L. Hu and Z. Sun, Nat. Commun., 2024, 15, 1934 CrossRef CAS PubMed .
10. D. Ling, Q. Wang, G. Tian, H. Yu, D. Zhang and Q. Wang, ACS Nano, 2023, 17, 25222–25233 CrossRef CAS PubMed .
11. M. Huang, X. P. Wang, X. Liu and L. Q. Mai, Adv. Mater., 2022, 34, 2105611 CrossRef CAS PubMed .
12. J. Shin, D. S. Choi, H. J. Lee, Y. Jung and J. W. Choi, Adv. Energy Mater., 2019, 9, 1900083 CrossRef .
13. M. M. Han, J. W. Huang, S. Q. Liang, L. T. Shan, X. S. Xie, Z. Y. Yi, Y. R. Wang, S. Guo and J. Zhou, iScience, 2020, 23, 1 Search PubMed .
14. X. P. Zhang, Y. G. Zhang, J. L. Qian, Y. H. Zhang, L. Sun and Q. Wang, Nanoscale, 2022, 14, 2004–2012 RSC .
15. S. Y. Dong, W. Shin, H. Jiang, X. Y. Wu, Z. F. Li, J. Holoubek, W. F. Stickle, B. Key, C. Liu, J. Lu, P. A. Greaney, X. G. Zhang and X. L. Ji, Chem, 2019, 5, 1537–1551 CAS .
16. J. Han, A. Varzi and S. Passerini, Angew. Chem., 2022, 134, e202115046 CrossRef .
17. R. D. Zhang, S. Wang, S. L. Chou and H. L. Jin, Adv. Funct. Mater., 2022, 32, 2112179 CrossRef CAS .
18. J. J. Holoubek, H. Jiang, D. Leonard, Y. T. Qi, G. C. Bustamante and X. L. Ji, Chem. Commun., 2018, 54, 9805–9808 RSC .
19. X. Y. Chen, Z. Y. Feng, X. Y. Dong, H. M. Jiang, C. G. Meng and Y. F. Zhang, SusMat, 2023, 3, 263–275 CrossRef CAS .
20. X. Y. Wu, Y. T. Qi, J. J. Hong, Z. F. Li, A. S. Hernandez and X. L. Ji, Angew. Chem., Int. Ed., 2017, 56, 13026–13030 CrossRef CAS PubMed .
21. G. J. Liang, Y. L. Wang, Z. D. Huang, F. N. Mo, X. L. Li, Q. Yang, D. H. Wang, H. F. Li, S. M. Chen and C. Y. Zhi, Adv. Mater., 2020, 32, 1907802 CrossRef CAS PubMed .
22. Y. Z. Zhang, J. Liang, Z. H. Huang, Q. Wang, G. Y. Zhu, S. Y. Dong, H. F. Liang and X. C. Dong, Adv. Sci., 2022, 9, 2105158 CrossRef CAS PubMed .
23. Z. N. Tian, V. S. Kale, Y. Z. Wang, S. Kandambeth, J. Czaban-Jozwiak, O. Shekhah, M. Eddaoudi and H. N. Alshareef, J. Am. Chem. Soc., 2021, 143, 19178–19186 CrossRef CAS PubMed .
24. G. Y. Zhou, X. Y. An, C. Y. Zhou, Y. Wu, Y. E. Miao and T. X. Liu, Compos. Commun., 2020, 22, 100519 CrossRef .
25. M. R. Lukatskaya, O. Mashtalir, C. E. Ren, Y. Dall'Agnese, P. Rozier, P. L. Taberna, M. Naguib, P. Simon, M. W. Barsoum and Y. Gogotsi, Science, 2013, 341, 1502–1505 CrossRef CAS PubMed .
26. M. S. Javed, X. F. Zhang, S. Ali, S. S. A. Shah, A. Ahmad, I. Hussain, S. Hussain, S. Khan, M. Ouladsmane, S. M. T. ElDin, W. Ul Arifeen and W. H. Han, Chem. Eng. J., 2023, 471, 144486 CrossRef .
27. K. Niu, J. J. Shi, L. Zhang, Y. Yue, S. Y. Mo, S. F. Li, W. B. Li, L. Wen, Y. X. Hou, L. Sun, S. W. Yan, F. Long and Y. H. Gao, Adv. Sci., 2024, 11, 2305524 CrossRef CAS PubMed .
28. S. Liu, L. Kang, J. M. Kim, Y. T. Chun, J. Zhang and S. C. Jun, Adv. Energy Mater., 2020, 10, 2000477 CrossRef CAS .
29. W. W. Zhang, W. H. Zhong, J. J. Wang, Q. Y. An and L. Q. Mai, Adv. Energy Sustainability Res., 2023, 4, 2300139 CrossRef CAS .
30. M. Huang, Q. He, J. J. Wang, X. Liu, F. Y. Xiong, Y. Liu, R. T. Guo, Y. Zhao, J. L. Yang and L. Q. Mai, Angew. Chem., Int. Ed., 2023, 62, e202218922 CrossRef CAS PubMed .
31. Y. L. Wu, S. Y. Dong, N. Lv, Z. K. Xu, R. Q. Ren, G. Y. Zhu, B. L. Huang, Y. Z. Zhang and X. C. Dong, Small, 2022, 18, 2204888 CrossRef CAS PubMed .
32. J. N. Gong, P. F. Bai, J. J. Sun, Y. Y. Liu, X. Y. Dong, T. Hu, C. G. Meng and Y. F. Zhang, Nano Res., 2024, 17, 4 Search PubMed .
33. J. W. Ding, Z. G. Du, L. Q. Gu, B. Li, L. Z. Wang, S. W. Wang, Y. J. Gong and S. B. Yang, Adv. Mater., 2018, 30, 1800762 CrossRef PubMed .
34. R. X. Yang, Q. C. Zhang, J. S. Cai, X. Cao, Q. Yang, Y. Z. Song and W. Zhang, Rare Met., 2024, 43, 2842–2850 CrossRef CAS .
35. W. W. Xu and Y. Wang, Nano-Micro Lett., 2019, 11, 1–30 CrossRef PubMed .
36. T. Zhou, Q. Han, L. L. Xie, X. L. Yang, L. M. Zhu and X. Y. Cao, Chem. Rec., 2022, 22, e202100275 CrossRef CAS PubMed .
37. W. W. Zhang, J. F. Liu, W. Y. Cai, M. Zhou, W. H. Zhong, G. F. Xiao, P. Luo, Y. Zhao and Q. Y. An, Chem. Eng. J., 2023, 464, 142711 CrossRef CAS .
38. S. Y. Deng, H. Li, B. H. Chen, Z. J. Xu, Y. Jiang, C. H. Li, W. Xiao and X. M. Yan, Chem. Eng. J., 2023, 452, 139115 CrossRef CAS .
39. M. Li, J. R. Mou, L. Zhong, T. Liu, Y. T. Xu, W. H. Pan, J. L. Huang, G. M. Wang and M. L. Liu, ACS Sustainable Chem. Eng., 2021, 9, 14193–14201 CrossRef CAS .
40. Z. L. Li, S. Ganapathy, Y. L. Xu, Z. Zhou, M. Sarilar and M. Wagemaker, Adv. Energy Mater., 2019, 9, 1900237 CrossRef .
41. Y. Q. Yang, Y. Tang, S. Q. Liang, Z. X. Wu, G. Z. Fang, X. X. Cao, C. Wang, T. Q. Lin, A. Q. Pan and J. Zhou, Nano Energy, 2019, 61, 617–625 CrossRef CAS .
42. X. Liu, G. B. Xu, Q. Zhang, S. J. Huang, L. Li, X. L. Wei, J. X. Cao, L. W. Yang and P. K. Chu, J. Power Sources, 2020, 463, 228223 CrossRef CAS .
43. Y. Y. Liu, T. T. Lv, H. Wang, X. T. Guo, C. S. Liu and H. Pang, Chem. Eng. J., 2021, 417, 128408 CrossRef CAS .
44. Y. B. Liu, Y. N. Zou, M. Y. Guo, Z. X. Hui and L. J. Zhao, Chem. Eng. J., 2022, 433, 133528 CrossRef CAS .
45. N. Zhang, M. Jia, Y. Dong, Y. Y. Wang, J. Z. Xu, Y. C. Liu, L. F. Jiao and F. Y. Cheng, Adv. Funct. Mater., 2019, 29, 1807331 CrossRef .
46. Y. L. Lin, L. Ta, J. M. Meng, Y. Song and X. X. Liu, Chem. Commun., 2023, 59, 1481–1484 RSC .
47. Z. Q. Li, Y. K. Ren, L. Mo, C. F. Liu, K. Hsu, Y. C. Ding, X. X. Zhang, X. L. Li, L. H. Hu, D. H. Ji and G. Z. Cao, ACS Nano, 2020, 14, 5581–5589 CrossRef CAS PubMed .
48. A. Bhatia, J. H. Xu, J. P. Pereira-Ramos, G. Rousse and R. Baddour-Hadjean, Chem. Mater., 2021, 34, 1203–1212 CrossRef .
49. T. K. Wang, X. J. Li, S. S. Zhao, H. X. Bu, C. L. Li, N. Li, X. X. Zhang and X. J. Xu, Chin. Chem. Lett., 2024, 35, 108449 CrossRef CAS .
50. S. N. Zhang, K. Zhu, Y. Y. Gao and D. X. Cao, ACS Energy Lett., 2023, 8, 889–897 CrossRef CAS .
51. V. Augustyn, P. Simon and B. Dunn, Energy Environ. Sci., 2014, 7, 1597–1614 RSC .
52. D. Chen, H. Wang, L. Ren, M. Zhu, Z. Bai, C. Li, C. Shi, H. Wang, Y. Tang and Y. Zhang, Sci. China Mater., 2023, 66, 4605–4614 CrossRef CAS .
53. C. Lin, X. Yang, P. Xiong, H. Lin, L. He, Q. Yao, M. Wei, Q. Qian, Q. Chen and L. Zeng, Adv. Sci., 2022, 9, 2201433 CrossRef CAS PubMed .
54. H. Lin, C. Lin, F. Xiao, L. He, P. Xiong, Y. Luo, X. Hu, Q. Qian, Q. Chen, Z. Wen and L. Zeng, Adv. Funct. Mater., 2024, 34, 2310486 CrossRef CAS .
55. J. Luo, M. Yang, D. Wang, J. Zhang, K. Song, G. Tang, Z. Xie, X. Guo, Y. Shi and W. Chen, Angew. Chem., Int. Ed., 2023, 62, e202315076 CrossRef CAS PubMed .
56. S. Wang, Z. Hu, Z. Pan and D. Wang, J. Alloys Compd., 2021, 876, 160203 CrossRef CAS .
57. J. M. Ge, Y. Y. Zhang, Z. K. Xie, H. B. Xie, W. H. Chen and B. A. Lu, Nano Res., 2023, 16, 4996–5005 CrossRef CAS .
58. C. Lin, L. He, P. Xiong, H. Lin, W. Lai, X. Yang, F. Xiao, X.-L. Sun, Q. Qian, S. Liu, Q. Chen, S. Kaskel and L. Zeng, ACS Nano, 2023, 17, 23181–23193 CrossRef CAS PubMed .
59. Q. Chen, J. L. Jin, M. D. Song, X. Y. Zhang, H. Li, J. L. Zhang, G. Y. Hou, Y. P. Tang, L. Q. Mai and L. Zhou, Adv. Mater., 2022, 34, 2107992 CrossRef CAS PubMed .
60. Y. Song, Q. Pan, H. Z. Lv, D. Yang, Z. M. Qin, M. Y. Zhang, X. Q. Sun and X. X. Liu, Angew. Chem., Int. Ed., 2021, 60, 5718–5722 CrossRef CAS PubMed .
61. D. W. Wang, J. Q. Sun and L. Chen, ChemSusChem, 2023, 16, e202300207 CrossRef CAS PubMed .
62. X. Chen, P. Wang, Z. Feng, C. Meng and Y. Zhang, Chem. Eng. J., 2022, 445, 136747 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4nr02149e

This journal is © The Royal Society of Chemistry 2024