Lithium-ion batteries (LIBs) are broadly utilized in the energy storage market. The uneven distribution of Li source and flammable organic electrolytes bring about high industry cost and safety problems in practical applications [1–6]. The AZIB systems constructed using aqueous electrolytes and Zn metal anode, are viewed as a potential battery system owing to low price, high security, non-flammability, high energy density, and abundant Zn reserves [7–14]. The advancement of AZIBs requires the cathode materials with outstanding electrochemical behaviors (e.g., high energy density and cycling stability) [15].

Mn-based oxide cathode materials have been increasingly explored owing to large theoretical capacity, high working voltage, eco-friendliness, natural abundance and low cost [16–19]. MnO₂ has diverse crystallographic polymorphs (e.g., α-MnO₂ [20], β-MnO₂ [21], δ-MnO₂ [22] and γ-MnO₂ [23]), and is therefore considered as a prominent cathode material. However, the application of MnO₂ cathode still faces many technical problems. Low electronic conductivity seriously limits the energy storage ability of MnO₂ materials [24–26]. Strong Zn²⁺-MnO₂ columbic interaction leads to a sluggish Zn²⁺ diffusion kinetics in the structure [27,28], which could cause the structural collapse at high Zn²⁺ intercalation condition during the electrochemical reaction of Zn²⁺/H⁺ co-intercalation [29]. Furthermore, MnO₂ can be dissolve into the electrolyte in the electrochemical reactions, which also leads to structural breakdown and capacity attenuation during cycling [30].

To resolve these problems, various methods have been attempted to modify the MnO₂ cathode materials. Metal ion pre-intercalation into MnO₂ materials is able to effectively enhance their electrochemical performances. For example, Chen et al. reported that cobalt can be pre-intercalated into α-MnO₂, which effectively enhances the structural stability of MnO₂ and promotes the ion diffusion and zinc ion adsorption upon battery cycling [31]. The pre-intercalation of Bi³⁺ into α-MnO₂ can increase the electrical conductivity, weaken the Zn²⁺-O²⁻ chemical bond and enhance structural stability [32].

The intercalation of organic molecules into MnO₂ is also proved to enhance the electrochemical performances. For example, the intercalation of organic poly-vinylpyrrolidone (PVP) into the layered δ-MnO₂ can validly improve the electrochemical performances of oxide material, because the C=O groups in PVP can facilitate H⁺ ion absorption and transportation [33]. The introduction of poly(4,4’-oxybisbenzenamine) (PODA) into α-MnO₂ can effectively boost ion diffusion and control manganese dissolution, improving redox dynamics and structural integrity of materials. The PODA chains with C=N groups can give more surface/interface for ion/electron migration and Zn²⁺ storage in the material [34]. The polyaniline (PANI)-reinforced the layered δ-MnO₂ can validly inhibit the H⁺/Zn²⁺ insertion and the structural breakdown, which is considered to be beneficial for the long cycling life and high capacity [35].

Recently, the organic-inorganic co-modification strategy has been applied in some vanadate-based oxide materials. The Na⁺ and polyaniline co-intercalation into ammonium vanadate (NaNVO-PANI) shows high reversible capacity (610.7 mAh/g) and exceptional cycling stability (98% after 5000 cycles). The synergistic effect of organic-inorganic co-modification in NaNVO-PANI expands the interlayer spacing, improves the electronic conductivity and lowers zinc ion diffusion barrier [36]. The Zn²⁺ and C₅H₁₄ON⁺ are successfully inserted into layered V₂O₅. The inherent high conductivity of C₅H₁₄ON⁺ and the oxygen vacancies created via ion pre-intercalation promote the electrical transfer. The synergistic co-inserted method enlarges the layer spacing, boosts the electronic conductivity, gives more active sites for zinc absorption and reinforces the structural stability upon Zn intercalation [37].

In this work, we utilize organic-inorganic co-modification strategy in the tunnel α-MnO₂ structure. The α-MnO₂ is composed by a group of [MnO₆] octahedra with a tunnel 2 × 2 structure and the channel of ~4.6 Å × 4.6 Å. The structural stability and appropriate channel in α-MnO₂ facilitate fast and reversible insertion and extraction for Zn²⁺ in the electrochemical reactions [31]. The inorganic Al³⁺ ions and organic poly-vinylpyrrolidone (PVP) are successfully incorporated into the α-MnO₂ structure via a simple hydrothermal method. The C=O groups in PVP were reported to be able to facilitate cations absorption and transportation, which can catch cations with reduced diffusion barriers and faster proton diffusion [33]. Structural characterizations indicate that the Al³⁺/PVP co-intercalation results in the deformation of MnO₆ octahedral framework in the tunnel α-MnO₂ structure. DFT calculations displays that the Zn²⁺ adsorption energy in the Al³⁺/PVP co-intercalated material is effectively lowered when compared with the original α-MnO₂ structure. Electrochemical tests indicate that the PVP-Al-MnO₂ electrode exhibits excellent electrochemical performances, a capacity of 306.8 mAh/g at 0.3 A/g and 93.1% capacity retention over 2000 cycles at 1.0 A/g. In addition, the aqueous PVP-Al-MnO₂||ZnClO₄||Zn battery is able to work properly at low temperature (-45 ℃). This work introduces an effective method to modify and optimize the advanced materials for AZIBs.

The α-MnO₂, Al-MnO₂ and PVP-Al-MnO₂ samples were synthesized through a hydrothermal method as depicted in the supporting information. The XRD data were collected in order to get the information on the crystal structures of three samples. All diffraction peaks can correspond to the characteristic peaks of α-MnO₂ (JCPDS No. 044-0141) with tunnel structure (Figs. 1a and c). The peaks at 12.8^o, 18.1^o, 28.8^o and 37.5^o were indexed to the (110), (200), (310) and (211) facets, respectively. Previous studies demonstrate that the intercalation of metal atoms into the layered δ-MnO₂ commonly results in the expanded lattice space, displaying on the angle shift of diffraction peaks [38,39]. Nevertheless, there is no obvious angle shift in the Al-intercalated and PVP-Al-intercalated samples (Fig. 1a), which should be attributed to the stable three-dimensional framework of tunnel α-MnO₂ (Fig. 1c). However, the obvious changes of diffraction intensity can be clearly observed on the (211) and (310) peaks. As shown in Fig. 1b, the Al-intercalated (R = 2.24) and PVP-Al-intercalated (R = 2.51) samples show higher intensity ratios of the characteristic peaks [R = I(211)/I(310)] than the pristine α-MnO₂ sample (R = 1.6). Similar phenomena were also observed in the 3D tunnel-structured FeF₃·0.33H₂O with Nb doping [40,41]. In addition, the intensity of (002) peak in the layered δ-MnO₂ is significantly weakened after the intercalation of PVP into layered structure [33]. Therefore, the intensity variation of (211)/(310) peaks observed in PVP-Al-MnO₂ (Fig. 1b) should be attributed to the distortion of MnO₆ octahedral framework with Al ions and PVP intercalation.

The full survey of XPS was collected in order to determine the chemical composition of PVP-Al-MnO₂ (Fig. 1d), showing the signal peaks of Al 2p, C 1s, N 1s, O 1s, Mn 2p_1/2 and Mn 2p_3/2. Successful insertion of PVP and Al³⁺ ions into α-MnO₂ is evidenced by the existence of N-C=O (Fig. 1e), N (Fig. S1 in Supporting information) and Al (Fig. 1f) as shown in the XPS results, consistent with previous results [33,42,43]. FTIR spectra (Fig. 1g) indicate the typical vibrations of C=O, -CH₂- and C-N groups in PVP-Al-MnO₂. Raman spectra reveal the symmetric Mn-O vibrations for the PVP-Al-MnO₂ (Fig. 1h), which are associated with the bands centered at ~573 and 634 cm⁻¹. The results indicate that the 3D tunnel structure consisting of MnO₆ octahedra is well-maintained after the PVP and Al ions intercalation, and the blue shift observed in PVP-Al-MnO₂ compared with α-MnO₂ should be attributed to the structural distortion of 3D tunnel framework after the PVP and Al ions intercalation, consistent with previous report [44,45]. The Mn 2p spectra of MnO₂, Al-MnO₂ and PVP-Al-MnO₂ were collected in order to determine the Mn valence variation after the chemical modifications (Fig. 1i). The peaks at 654.5 and 643.5 eV are clearly observed in α-MnO₂, matching well to 2p_1/2 and 2p_3/2 signals of Mn⁴⁺. The oxidation state of Mn is reduced in PVP-Al-MnO₂. The presence of Jahn-Teller Mn³⁺ cations in PVP-Al-MnO₂ also leads to the structural distortion of 3D tunnel framework, consistent with the observations in XRD patterns (Figs. 1a and b) and Raman spectra (Fig. 1h). Overall, the Al ions and PVP are proved to be successfully incorporated into α-MnO₂ with 3D tunnel structure via a simple hydrothermal method.

The morphological information of all the samples was collected using SEM and TEM. PVP-Al-MnO₂, Al-MnO₂ and bare MnO₂ all exhibits nanofibre morphologies (Figs. 2a–c and Fig. S2 in Supporting information). SAED was performed in order to obtain the structural information on PVP-Al-MnO₂, and the result is shown in Fig. 2d. The spots can be well indexed to the typical reflections of tunnel structure along the [113] zone. The lattice spacings of 0.24 nm and 0.32 nm can be seen in HRTEM (Fig. 2e) image, which matches well to the (211) and (310) planes of tetragonal a-MnO₂ phase. The elemental mapping images (Fig. 2f) present a uniform distribution of C, N, Al, Mn and O elements, confirming the formation of PVP-Al-MnO₂ material.

The electrochemical behaviors of PVP-Al-MnO₂ were studied in coin batteries with Zn as the negative electrode. The electrochemical performances of PVP-Al-MnO₂ were investigated by charge/discharge in the voltage of 0.8-1.8 V at 0.3 A/g (Fig. 3a). The charge/discharge profiles of MnO₂, Al-MnO₂ and PVP-Al-MnO₂ cathodes in the 1^st cycle at 0.3 A/g (Fig. 3a). The PVP-Al-MnO₂ cathode provides a capacity of 306.8 mAh/g, which is higher than the values of Al-MnO₂ (228.7 mAh/g) or MnO₂ (137.3 mAh/g). The EIS was tested to acquire the information on the conductivity of PVP-Al-MnO₂ and MnO₂ (Fig. 3b). The R_ct value of PVP-Al-MnO₂ (7.5 Ω) is prominently lower than that of MnO₂ (14.3 Ω), revealing that the PVP-Al-MnO₂ material has faster charge transfer. Figs. 3c and d show the rate capabilities of PVP-Al-MnO₂, Al-MnO₂ (Figs. S3 and S4 in Supporting information) and MnO₂ from 0.3 A/g to 1.5 A/g. PVP-Al-MnO₂ shows the enhanced rate performances when compared with the original α-MnO₂, and delivers reversible capacities of 314.4, 275.5, 250.8, 218.4 and 178.4 mAh/g at 0.3, 0.5, 0.7, 1.0 and 1.5 A/g, respectively. The capacity of PVP-Al-MnO₂ returns to 307.9 mAh/g when the current density is reset to 0.3 A/g, demonstrating a rapid reaction kinetics and extraordinary reversibility upon fast Zn ion insertion/extraction. The cycling performances of PVP-Al-MnO₂ were explored at 0.3 A/g (Fig. 3e and Fig. S5 in Supporting information). The capacity of PVP-Al-MnO₂ is 306.8 mAh/g in the 1^st cycle, which is significantly higher than Al-MnO₂ (228.7 mAh/g) and MnO₂ (137.3 mAh/g). The battery shows capacity retention of 99.6% after 50 cycles and delivers reversible capacities of 300.9, 300.6, 302.9, 304.3 and 305.5 mAh/g at the 10^th, 20^th, 30^th, 40^th and 50^th cycles (Fig. 3f). The charge/discharge profiles are well maintained during the cycling, demonstrating excellent structural stability of PVP-Al-MnO₂ material upon repetitive Zn²⁺ insertion/extraction. To study the cycling behaviors of PVP-Al-MnO₂, the material was examined at 1.0 A/g (Fig. 3g and Fig. S6 in Supporting information). The PVP-Al-MnO₂ cathode provides an initial capacity of 189.9 mAh/g with the capacity retention of 93.1% after 2000 cycles. Overall, the high capacity and superior capacity retention of PVP-Al-MnO₂ implies that the kinetics and electrochemical stability of tunnel MnO₂ were effectively improved by the synergistic role of PVP and Al co-doping. As discussed in Fig. 1, the incorporation of PVP and Al into α-MnO₂ leads to the distortion of MnO₆ octahedral framework, which lowers the geometric symmetry of the ligand field and removes the 3d orbital degeneracy of the Mn center. The Jahn-Teller distortion of Mn³⁺ center is effectively inhibited and the cycling steadiness is remarkably enhanced. In addition, low-valence Mn results in a reduction in electrostatic repulsion, thereby effectively enhancing rate performances [46]. The electrochemical behaviors of our PVP-Al-MnO₂ material are also compared with other Mn-based materials (Table S1 in Supporting information and Fig. 3h). For example, the anion-deficient MnO_2-δ cathode shows a capacity of 147 mAh/g with capacity retention of only 78% after 1000 cycles [47]. The Zn pre-inserted ZnMn₂O₄ cathode shows a capacity of 166.6 mAh/g at 1.0 A/g, and the capacity can retain 143.9 mAh/g after 1500 cycles with 86.4% capacity retention [48]. It is demonstrated that the PVP and Al co-doping in tunnel α-MnO₂ can effectively promote the electrochemical performances and cycling stability.

To obtain the information on the charge storage and kinetic mechanism of PVP-Al-MnO₂, CV tests were applied to detect the energy storage behaviors of PVP-Al-MnO₂ (Fig. 4a). The curve shows 2 pairs of redox peaks clearly, matching well to the reversible insertion/extraction of hydrogen ion and zinc ion in the electrochemical processes as reported previously about the tunnel α-MnO₂ material [49,50]. The peak current (i) and sweep rate (v) can be referred by the formula [51,52]:

i=avb

(1)

logi=blogv+loga

(2)

where a and b are variants, and the b value is an important parameter to judge the dominant capacity contribution. The b values of the four main peaks are approaching to 1 (0.68, 0.79, 0.77 and 0.89) (Fig. 4b), demonstrating the charge storage behavior of PVP-Al-MnO₂ cathode is controlled by the capacitance process. The contribution of capacitive-like reaction in PVP-Al-MnO₂ cathode accounts for 89.3% at 1.0 mV/s (Fig. 4c). Additionally, the formula is used to elucidate the particular ratio of 2 various capacity contributions in PVP-Al-MnO₂ cathode [53]:

i(V)=k1v=k2v1/2

(3)

i is the current value. k₁ and k₂ denote the capacitive-like and diffusion-control ratio, respectively. The various contribution proportions at different scan rates are shown in Fig. S7 (Supporting information). To summarize capacitive contribution (Fig. S7), the pseudocapacitance contribution is gradually increased from 63.1% to 89.3% with the increased scan rate, manifesting that the pseudocapacitance-controlled behavior is favorable for the high-rate behaviors.

The GITT were further conducted to evaluate the Zn²⁺ diffusion kinetics within the material (Fig. 4d). The Zn²⁺ diffusion coefficient can be estimated by the equation [54]:

DZn2+=(4/πτ)(mBVM/MBA)2(ΔEs/ΔEτ)2

(4)

where D_Zn²⁺ is the diffusion coefficient of Zn²⁺, τ is the pulse time, m_B and M_B are the molecular weight/mass of the active substance, V_M is the molar volume, A is the area, and ΔE_s/ΔE_τ is the steady-state voltage/total change of the battery voltage E. The D_Zn²⁺ value of PVP-Al-MnO₂ estimated is ~10⁻⁹ cm²/s, which is noticeably higher than α-MnO₂ (Fig. 4d). This interprets why PVP-Al-MnO₂ exhibits improved rate performance and electrochemical stability compared with MnO₂ (Fig. 4d).

The charge compensation mechanism of PVP-Al-MnO₂ in the electrochemical processes was further explored by XPS measurements. The Mn 2p spectra of PVP-Al-MnO₂ show the variation of manganese valence during the discharge-charge process (Fig. 4e). When the material was discharged to 0.8 V, the ratio of Mn³⁺ is apparently increased in the Mn 2p spectra, proving that Mn⁴⁺ is partially decreased to Mn³⁺ (2MnO₂ + Zn²⁺ + 2e⁻ ↔ ZnMn₂O₄). When the battery was recharged to 1.8 V, the peak proportion of Mn⁴⁺ is increased prominently, implying that the extraction of zinc ions from the electrode is followed by the decline of Mn ions. The reversible valence variation of Mn 2p further proves the reversible charge storage in the PVP-Al-MnO₂ material. In addition, 2 peaks with binding energies of 1045.4 (Zn 2p_3/2) and 1022.4 eV (Zn 2p_1/2) are observed at fully charged and discharged states in the Zn 2p (Fig. 4f). This confirms the insertion of zinc ions into PVP-Al-MnO₂ framework in the discharge process. When the cell was charged to 1.8 V, Zn²⁺ cannot be fully extracted from the structure and there is some remaining Zn in the lattice.

The electrochemical energy storage mechanism for PVP-Al-MnO₂ was explored by ex-situ XRD results. The batteries with PVP-Al-MnO₂ cathodes, zinc anode and 3 mol/L Zn(ClO₄)₂/0.2 mol/L MnSO₄ electrolyte were cycled at 0.3 A/g, and the PVP-Al-MnO₂ electrodes were detached at different stages from A to G (Fig. 4g). The tunnel structure of PVP-Al-MnO₂ is well maintained in the whole electrochemical processes. From ex-situ XRD diffraction patterns (Fig. 4h), the formation of MnOOH (JPCDS No. 24-0713), ZnMn₂O₄ (JPCDS No. 71-2499), Zn₄SO₄(OH)₆·0.5H₂O (ZHS, JPCDS No. 44-0674), and Zn₄ClO₄(OH)₇ (PDF #41-0715) were simultaneously detected in the complete discharge product (G), consistent with previous results [55,56]. This suggests the electrochemical reaction mechanism of zinc ion and hydrogen ion intercalation into the PVP-Al-MnO₂ framework, which induces the generation of ZnMn₂O₄ and the formation of by-product hydroxides [57]. The products gradually disappear and regenerate in the charge and discharge processes, implying the reversible electrochemical reaction based on Zn²⁺ and H⁺ ions intercalation/de-intercalation. Fig. 4i shows the variation of the (310) peaks in the electrochemical processes. The characteristic peaks shift to the lower angle during the discharge process and shift back to the higher angle during the charging process. This implies that the reversible intercalation/de-intercalation of Zn²⁺/H⁺ ions leads to the shift of the characteristic peaks. This strongly confirms that the introduction of PVP molecule and Al ions into α-MnO₂ does not change the structural characteristics of tunnel structure in the electrochemical reaction, although the electrochemical performances of the material are effectively enhanced. The Zn storage abilities of different materials were evaluated by calculating the E_ads of tunnel MnO₂ with various intercalations. As shown in Figs. 4j and k, the E_ads of Zn on Al and PVP co-intercalated MnO₂ shows lower E_ads than the unmodified MnO₂, Al-MnO₂ or PVP-intercalated MnO₂, indicating that the co-intercalation of Al and PVP into the tunnel material plays a synergetic role in enhancing the adsorption capability of Zn²⁺. This could explain why the PVP-Al-MnO₂ material exhibits significantly enhanced capacity and long-life cycling stability (Fig. 3).

One of main technical challenges for aqueous battery systems is their poor low-temperature electrochemical performances, because the high liquid-solid transition temperature of aqueous electrolyte seriously limits the ion diffusion at low operation temperature. The 3 mol/L Zn(ClO₄)₂/0.2 mol/L MnSO₄ electrolyte is selected in this work. The Zn²⁺ storage capabilities of PVP-Al-MnO₂ electrode are therefore tested at a low working temperature of -45 ℃ (Fig. 5). The rate capabilities of PVP-Al-MnO₂ electrode and charge-discharge profiles are shown in Figs. 5a and b. It can be observed that the PVP-Al-MnO₂ electrode exhibits specific capacities of 79.9, 75.0, 73.5, 66.5, 65.3 and 60.9 mAh/g at 40, 60, 80, 100, 150 and 200 mA/g, respectively. The reversible capacity is restored to the original value when the current density is reduced to 40 mA/g, demonstrating the extremely high structural stability upon fast Zn²⁺ ion de(intercalation) processes on PVP-Al-MnO₂. The GITT measurement was also carried out to obtain the information on diffusion coefficient of PVP-Al-MnO₂ at -45 ℃. The material shows an ion diffusion coefficient of 10⁻¹²-10⁻⁹ cm²/s at -45 ℃, lower than the value at room temperature (Fig. 5c). The result confirms the low diffusion activity of PVP-Al-MnO₂ at -45 ℃.

Figs. 5d and e depict the cycling behaviors of the PVP-Al-MnO₂ electrode and the charge-discharge profiles at 100 mA/g at -45 ℃. The PVP-Al-MnO₂ electrode displays a capacity of 60.5 mAh/g at 100 mA/g with capacity retention of 98% after 1000 cycles (Fig. 5d). Furthermore, the charge-discharge curves at 0.1 A/g (Fig. 5e) also displays similar electrochemical curves, confirming the outstanding electrochemical reversibility and structural stability of PVP-Al-MnO₂ electrode at -45 ℃. The electrochemical behaviors of PVP-Al-MnO₂ material at various temperatures were also systematically examined. The electrochemical behaviors are effectively inhibited as the operation temperature reduces gradually from 25 ℃ to -45 ℃. As shown in Figs. 5f and g, the PVP-Al-MnO₂ material exhibits reversible capacities of 332.7, 186.1, 157.9, 94.4 and 62.7 mAh/g at room temperature, 0, -10, -20 and -45 ℃ at 100 mA/g. The pouch cells were also constructed in order to test if the battery can operate properly in practice (Figs. 5h and i). Two pouch cells in-series can power LED lights properly at room temperature (Fig. 5h). When the working temperature is reduced to -45 ℃, the LED lights can still be working as normal (Fig. 5i), confirming the excellent electrochemical performances of zinc-ion batteries at low temperature. Overall, the PVP-Al-MnO₂||ZnClO₄||Zn full cell described in this work is able to be working at low temperature.

In summary, PVP-Al-MnO₂ was synthesized through a hydrothermal method. The inorganic Al³⁺ ions and organic poly-vinylpyrrolidone (PVP) are successfully incorporated into the tunnel α-MnO₂ structure. Structural characterizations (XRD, FTIR, Raman and XPS) confirm the incorporation of Al³⁺/PVP into the tunnel α-MnO₂ structure, which leads to the distortion of MnO₆ octahedral framework. DFT calculations were conducted and the result reveals that the Zn²⁺ adsorption energy in the Al³⁺/PVP co-intercalated material is effectively lowered when compared with the original α-MnO₂ structure. Electrochemical tests indicate that the PVP-Al-MnO₂ electrode exhibits excellent electrochemical performances, a reversible capacity of 306.8 mAh/g at 0.3 A/g and 93.1% capacity retention over 2000 cycles at 1.0 A/g. Moreover, the aqueous PVP-Al-MnO₂||ZnClO₄||Zn battery can be working at low temperature (-45 ℃). This work serves an encouraging approach to the modification and optimization of electrode materials for aqueous multivalent ion cells.

Declaration of competing interests

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.

CRediT authorship contribution statement

Mengdan Tian: Investigation. Chuanzheng Zhu: Investigation. Kun Luo: Supervision.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110702.