Ag-Doping Effect on MnO₂ Cathodes for Flexible Quasi-Solid-State Zinc-Ion Batteries

Abstract

Rechargeable aqueous Zn/MnO₂ batteries are very potential for large-scale energy storage applications owing to their low cost, inherent safety, and high theoretical capacity. However, the MnO₂ cathode delivers unsatisfactory cycling performance owing to its low intrinsic electronic conductivity and dissolution issue. Herein, we design and synthesize a Ag-doped sea-urchin-like MnO₂ material for rechargeable zinc-ion batteries (ZIBs). Doping Ag was found to reduce charge transfer resistance, increase the redox activity, and improve the cycling stability of MnO₂. The unique sea-urchin-like structure maintains rich active sites for charge storage. As a result, the Ag-doped MnO₂-based ZIB presents a high reversible specific capacity to 315 mA h g⁻¹ at 50 mA g⁻¹, excellent rate performance, and a capacity retention of 94.4% when cycling over 500 cycles. An ex situ TEM test demonstrates the low-dissolution property of Ag-doped MnO₂. A flexible quasi-solid-state ZIB is successfully assembled using Ag-doped MnO₂ on graphite paper, which shows a stable specific capacity of 171 mA h g⁻¹ at 1 A g⁻¹ when cycled over 600 cycles. Our investigation demonstrates the significant role played by Ag doping in enhancing the ZIB performance of MnO₂, and gives some insight into developing advanced active materials by heteroatom doping.

1. Introduction

Growing concerns on environmental issues and increasing demands for renewable energy trigger more research attentions on advanced electrical energy storage systems [1,2]. Rechargeable aqueous zinc-ion batteries (ZIBs) have been actively pursued recently by virtue of their high theoretical capacity, good safety, low cost of raw materials, and low redox potential intrinsic to the zinc metal anode [3,4]. So far, a few kinds of materials have been studied as the cathode materials for ZIBs, such as manganese-based oxides, vanadium-based materials, Prussian blue analogues, and some active organic compounds [5,6,7,8]. Among them, manganese-based oxides, especially MnO₂, are perhaps the most attractive candidate for practical applications, considering their combined merits in terms of low cost, environmental benignity, high operation voltage (∼1.4 V vs. Zn/Zn²⁺), and high theoretical capacity (∼308 mA h g⁻¹) [9,10,11]. Liu et al. [12] reported α-MnO₂ as the cathode for ZIBs, exhibiting an operating voltage of 1.44 V along with a maximal capacity to 285 mA h g⁻¹. Chen et al. [13] reported that β-MnO₂ also possesses a high reversible capacity of 225 mA h g⁻¹, while 94% of capacity retention could be achieved when cycled over 2000 cycles.

However, in theory, all MnO₂ polymorphs experience a Jahn–Teller distortion under discharged state, leading to Mn dissolution into the electrolyte followed by rapid capacity fading [14]. Meanwhile, the MnO₂ cathode suffers from low intrinsic electronic conductivity and strong electrostatic repulsion with Zn²⁺, resulting in limited Zn ion-storage performance [15]. Previous research efforts have actually sought diverse strategies to premeditatively overcome the instability issues of MnO₂, such as pre-addition of Mn²⁺ into electrolyte [12], surface coating [16,17], defect engineering, and heteroatom doping [18,19,20,21]. Heteroatom doping is an effective method to modify the intrinsic physicochemical properties of MnO₂ [22]. Kim et al. reported V-doped MnO₂ via a simple ambient redox reaction [23]. Doping with V atoms not only increased the specific surface area of MnO₂, but also improved its electronic conductivity. Pan et al. found that Ni dopants enabled a higher degree of tetragonal orthorhombic distortion for improved Grotthuss proton transport in the tunnels of α-MnO₂, and the Ni-doped α-MnO₂ delivered a capacity of 303 mA h g⁻¹ and impressive specific energy density of 421 W h kg⁻¹ [24]. Recently, N and F doping of MnO₂ was pursued, which created additional oxygen vacancies; as a result, improved ion insertion and transport kinetics, as well as enhanced electrical conductivity could be achieved [25,26]. In despite of the fact that heteroatom doping could effectively overcome the intrinsic shortcomings of MnO₂, the performance improvement has been restricted by the limited choice of the heteroatom types that can be well distributed in the MnO₂ bulk phase. Thus, it is desirable to develop new and efficient dopants for better performance of MnO₂. Ag has an excellent conduction nature, and doping Ag has been proved to enhance the electronic conductivity, increase the formation of more oxygen vacancies, and modify the band gap of electrode materials, which has been successfully applied in supercapacitors [27], catalysts [28], solar cells [29]. In addition, the morphologies and particle sizes of MnO₂ also have effects on electrochemical performance of ZIBs [22]. It is reported that iron-doped hollow sea-urchin-like MnO₂ with many nanowires extended from a common center process larger surface area which can provide more active sites for electrochemical reaction, showing good electrochemical performance for supercapacitors [30].

Here, Ag-doped MnO₂ with a sea-urchin-like structure was synthesized via a simple hydrothermal method, which was first utilized as a cathode material for ZIBs. A series of characterizations indicate that not only a unique sea-urchin-like structure provides more active sites, but also doping Ag heteroatoms improves the ionic conductivity of MnO₂. Besides, the Ag redox in charge/discharge has contributed to increased capacity., leading to improved electrochemical Zn²⁺ insertion and transport kinetics. As a result, by tuning the Ag-doping content, the Ag-doped MnO₂-based ZIB delivers high reversible specific capacity (315 mA h g⁻¹ at 50 mA g⁻¹), enhanced cyclic stability (500 cycles with the capacity retention of 94.4%), and excellent rate performance. In addition, flexible quasi-solid-state ZIBs are successfully assembled based on graphite papers supported by Ag-doped MnO₂, which exhibits a stable capacity of 171 mA h g⁻¹ at 1 A g⁻¹ over 600 cycles.

2. Experimental

2.1. Synthesis of Ag-Doped Sea-Urchin-like MnO₂ and Pure MnO₂

First, MnSO₄ (0.48 g, AR), (NH₄)₂S₂O₈ (0.73 g, AR) and 80 mL of deionized water were put into a beaker and stirred to make the reactant totally dissolved. Then, 3.2 mL of concentrated H₂SO₄ (98%) was slowly added into the beaker. Then, 1, 3, 5, 7, 9 wt.% of AgNO₃ (AR) was dissolved into 1.6 mL of deionized water and added into above solution. The solution was put into a Teflon contained autoclave and heated at 120 °C for 6 h. The precipitates were collected, washed with deionized water and absolute ethanol, respectively, several times to remove impurities, and then dried at 60 °C overnight. Pure MnO₂ was synthesized following a same procedure without adding AgNO₃. The products designated as 1%Ag-MnO₂, 3%Ag-MnO₂, 5%Ag-MnO₂, 7%Ag-MnO₂, 9%Ag-MnO₂ and MnO₂, respectively. In addition, 5%Ag-MnO₂ samples were also synthesized using different reaction time of 1, 2, and 4 h.

2.2. Synthesis of the Polyacrylamide (PAM) Hydrogel Electrolyte

Typically, acrylamide (15 g, 99%), N,N′-methylenebisacrylamide (10 mg, AR) and potassium persulfate (75 mg, AR) were dissolved in 50 mL of deionized water and stirred to make the reactant totally dissolved. PAM was obtained after polymerization at 60 °C for 60 min.

2.3. Synthesis of the Flexible Zn Anode

The flexible Zn anode was prepared by an electrodeposition method [31]. ZnSO₄·7H₂O (10.23 g, AR) and Na₃C₆H₅O₇·2H₂O (14.7 g, AR) were put into 100 mL of deionized water with magnetic stirring. Chronoamperometry was applied to electrodeposit Zn on carbon cloth substrates at −1.4 V (vs. saturated calomel electrode) for 15 min.

2.4. Materials Characterizations

The crystal structures of the electrodes were characterized by the X-ray diffraction (XRD; Rigaku, Ultima IV, Tokyo, Japan; Cu Kα radiation λ = 0.15418 nm) technique. The morphology of samples was investigated by field-emission scanning electron microscope (FESEM; JEOL, JSM-7800F, Tokyo, Japan) and transmission electron microscope (TEM; JEOL, JEM-2100 plus). The element compositions of samples were tested by X-ray photoelectron spectroscopy (XPS; PHI5000 Versaprobe III). The Raman spectra were tested using a Raman spectroscopy (Renishaw inVia Raman spectrometer) at a laser wavelength of 532 nm.

2.5. Electrochemical Measurements

The cathodes were prepared by mixing active materials, acetylene black and polyvinylidene fluoride (PVDF) at an appropriate weight ratio of 7:2:1. The mixture was uniformly coated on titanium mesh and dried, at 80 °C, overnight. The mass loading of active material per electrode was about 2 mg cm⁻². The electrochemical properties were measured in 2032-type coin cells using Zn metal plate as anode, glass fiber as separator and 2 M ZnSO₄ with 0.1 M MnSO₄ additive as electrolyte. The discharge and charge performance was measured on a multi-channel battery test system (Neware CT-4008 W, Shenzhen, China) in the voltage range of 0.8–1.8 V vs. Zn/Zn²⁺. Cyclic voltammetry (CV) and Electrochemical Impedance Spectroscopy (EIS) measurements were taken on CHI660E. All these tests were carried out at room temperature.

3. Results and Discussion

The Ag-doped MnO₂ and pure MnO₂ were synthesized by a simple hydrothermal method, and the synthesis route was illustrated in Figure 1a. The XRD patterns of Ag-doped MnO₂ and pure MnO₂ are shown in Figure S1, the MnO₂ exhibits apparent diffraction peaks at 12.4°, 17.8°, 28.4°, 36.3°,37.3°, 41.6°,49.4°, 55.1°, 59.8°, 65.1°, 69.0°, and 72.6°, which can be assigned to the (110), (200), (310), (400), (211), (301), (411), (600), (521), (002), (541), and (312) planes of α-MnO₂ (space group: Tetragonal I4/m, JCPDS: 44–0141). The Ag-doped MnO₂ samples exhibit same XRD patterns compared with pure MnO₂ without any additional peaks, indicating the crystal structure of MnO₂ can be maintained when doped with Ag. However, the diffraction peaks from (110), (200), (400), (301), (600) planes weakens and become much broader when increased Ag dopant was used, indicating decreased grain size and increased amount of lattice defects [32]. The diffraction peak located at 37.3° intensifies after doping with Ag, indicating preferred orientation growth of (211) plane. Moreover, the diffraction peaks from (411), (600), and (521) planes gradually shift to a high diffraction angle when doped with increased amounts of Ag, indicating increased lattice distortion of MnO₂ unit cell [33,34]. To identify lattice distortion degree and location of the Ag dopant in the unit cell, the refinement XRD patterns of pure MnO₂ and 5%Ag-MnO₂ are presented in Figure 1b,c. The refinement crystal structures of pure MnO₂ and 5%Ag-MnO₂ can be found in the inset of Figure 1b,c, respectively. The Rietveld-refined fractional atomic parameters of 5%Ag-MnO₂ are recorded in Table S1. It is found that 10% of Mn sites are substituted by Ag. Refined parameters for pure MnO₂ are calculated to be a = 9.94 Å, b = 9.94 Å, c = 2.87 Å, α = 90°, β = 90°, and γ = 90°. However, the lattice parameters are transformed to a = 9.91 Å, b = 9.91 Å, c = 2.87 Å, α = 90°, β = 90°, and γ = 90°, when doped with Ag. Raman spectra of Ag-doped MnO₂ and pure MnO₂ are shown in Figure 1d. The peak at 640 cm⁻¹ belongs to the stretching vibration of Mn-O, which exhibits an obvious blueshift for MnO₂ doped with increased amounts of Ag owing to increased Mn³⁺ content [35,36]. Chemical composition and their surface electronic states were analyzed by XPS spectra. Photoelectron emission peak from Ag element can be positively observed in the wide-scan spectrum of 5%Ag-MnO₂ (Figure S2), confirming successful doping of Ag into MnO₂. Figure 1e presents the high-resolution XPS spectrum from Ag 3d electrons of 5%Ag-MnO₂. The peaks at 367.6 and 373.6 eV belong to 3d_5/2 and 3d_3/2 photoelectrons of Ag⁺, respectively [37]. Meanwhile, the peaks at 368.3 and 374.3 eV correspond to the standard values of metallic Ag [38]. As shown in Figure 1f, the high-resolution Mn 3s spectrum of 5%Ag-MnO₂ exhibits a spin-energy separation of 4.9 eV, which is higher than that of pure MnO₂. Therefore, Mn elements in 5%Ag-MnO₂ has a lower chemical valence than that in pure MnO₂ owing to Ag doping [39]. The high-resolution O 1s spectrum can be fitted into three peaks (Figure 1g). The peaks of 529.6, 531.1, and 532.7 eV are related to Mn-O, oxygen defects, and adsorbed water [40]. The calculated area ratios of oxygen defects increase from 19% to 23% after introducing Ag into MnO₂. Increased amount of oxygen defect will give rise to narrower band gap and faster electron migration rates, and thus higher reaction activity and fast reaction kinetics can be anticipated for 5%Ag-MnO₂ [41].

Figure 2a,b show the FESEM images of pure MnO₂. Pure MnO₂ exhibits a sea-urchin-like structure which is composed of many nanowires. The 5%Ag-MnO₂ also displays a sea-urchin-like structure; however, increased amounts of nanowires extended from a common center (Figure 2d,e). It is important to say that the weight ratio of Ag dopant has a great influence on the overall morphology of Ag-doped MnO₂, and only the 5%Ag-MnO₂ sample can form a complete sea-urchin-like structure. Note to say, the morphology does not change with reaction time, we also synthesized the 5%Ag-MnO₂ sample at reaction time of 1, 2, and 4 h, and they exhibit a same sea-urchin-like structure (Figure S3). The reaction time of 6 h is chosen for sample synthesis so as to ensure full precipitation of Mn²⁺. The sea-urchin-like structure of other Ag-doped MnO₂ all cracked into different segments (Figure S4). Energy-dispersive X-ray spectroscopy (EDS) pattern of 5%Ag-MnO₂ verifies the coexistence of Ag, Mn, and O (Figure S5), and the corresponding EDS mapping images (Figure 2h) display uniform distribution of these elements. TEM images (Figure 2c,f) further verify a sea-urchin-like overall structure of pure MnO₂ and 5%Ag-MnO₂, while more nanowires can be observed on the sea-urchin-like structure for 5%Ag-MnO₂. High-resolution TEM image in Figure 2g shows clear lattice fringes with the space of 0.49 nm, which is corresponding to (200) plane of α-MnO₂. In addition, N₂ adsorption–desorption isotherms (Figure S6) of 5%Ag-MnO₂ and pure MnO₂ samples are both exhibit a type IV isotherm with a H3-type hysteresis loop. The 5%Ag-MnO₂ has a comparable Brunauer–Emmett–Teller (BET) area to pure MnO₂ with a surface area of 58.4 m² g⁻¹ delivered. However, the 5%Ag-MnO₂ has more mesopores in pore size range of 2–35 nm compared to pure MnO₂. As Zn²⁺ is bivalent with a large radius when the ion is hydrated, relative large pores may promote electrochemical reactions and efficient penetration of the electrolyte into active material [42].

These Ag-doped MnO₂ and pure MnO₂ were used as the zinc-ion host materials in CR2032 coin cells to investigate their energy storage performance in 2 M ZnSO₄ electrolyte with 0.1 M MnSO₄ additive. Figure 3a shows initial two cycles of CV curves at a scan rate of 0.3 mV s⁻¹ for 5%Ag-MnO₂ and pure MnO₂-based ZIBs. The first cycle of CV curve both exhibit two pairs of redox peaks, indicating a two-step electrochemical reaction of above two cathode materials [43]. Compared with pure MnO₂-based ZIB, the Zn/5%Ag-MnO₂-based ZIB exhibits higher potentials for cathodic peaks, while lower potentials for anodic peaks. In addition, there are a couple of weak redox peaks at 0.92 V and 1.16 V exist in the CV curve of 5%Ag-MnO₂-based ZIB, manifesting the successful doping of Ag into MnO₂ (Figure S7). The redox couple of Ag⁺/Ag from the Ag dopant provides an additional reaction for energy storage. Meanwhile, the CV curves of 5%Ag-MnO₂-based ZIB shows intensified redox peaks and larger enclosed area than that of ZIB based on pure MnO₂, indicating that reaction activity is enhanced through Ag doping. Figure 3b shows the galvanostatic charge/discharge (GCD) curves of ZIBs based on pure MnO₂ and Zn/5%Ag-MnO₂ at a specific current of 50 mA g⁻¹. The Zn/5%Ag-MnO₂ cell shows a discharge capacity of 315 mA h g⁻¹ which is about 22% higher than that of Zn/MnO₂ cell (259 mA h g⁻¹). Furthermore, the polarization value between the discharge and charge plateau is lower after Ag doping owing to improved ion insertion and transport kinetics. Compared to the GCD curves of Zn/MnO₂ cell, the additional discharge plateau at around 0.93 V and charge plateau at around 1.15 V (Figure S8) in the GCD curve of Zn/5%Ag-MnO₂ cell correspond to the reversible redox reaction of Ag dopant. Electrochemical impedance spectroscopy (EIS) measurements (Figure 3c) are used to investigate the charge transfer rate and ion diffusion kinetics, and the charge-transfer resistance (R_ct) of 5%Ag-MnO₂ (R_ct = 16 Ω) is lower than that of pure MnO₂ (R_ct = 26 Ω), indicating improved reaction kinetics for MnO₂ when doped with Ag. The rate performance of ZIBs based on 5%Ag-MnO₂ and pure MnO₂ are shown in Figure 3d. The specific capacity of both two ZIBs increase in initial cycles because of the active progress of cathodes. The 5%Ag-MnO₂-based ZIB delivers specific capacities of 315, 290, 250, 215, 177, 128, and 85 mA h g⁻¹ at specific currents of 0.05, 0.1, 0.2, 0.3, 0.5, 1.0, and 2.0 A g⁻¹, respectively; this ZIB shows a specific capacity of 341 mA h g⁻¹ when the specific current returns back to 0.05 A g⁻¹. Above capacity values are much higher than the performance achieved by pure MnO₂. Doping Ag increases the redox activity of Mn ions in MnO₂, resulting in a high specific capacity. In addition, inevitably, the deposition of pre-added Mn²⁺ ions may contribute to extra capacity, resulting in a capacity of Ag-doped electrodes higher than the theoretical capacity of MnO₂. Figure 3e shows the long-term cycling performance at 0.5 A g^–1 for Ag-doped MnO₂ and pure MnO₂. MnO₂ with different Ag-doped amounts all exhibit better cycling performance than the pure MnO₂, especially for the ZIB based on 5%Ag-MnO₂. The 5%Ag-MnO₂-based ZIB presents a specific capacity around 150 mA h g⁻¹ over 500 cycles with a capacity retention of 94.4% can be achieved, which is better than most Mn-based and heteroatom doping cathodes (Table S2), while a capacity retention of 76.1% is achieved by pure MnO₂. Better cycling performance of 5%Ag-MnO₂ was further verified by the cycling performance under large specific current of 0.8 A g⁻¹ (Figure S9). The morphology of 5%Ag-MnO₂ was also characterized by TEM test when cycling for different cycles. Owing to the Jahn–Teller distortion under discharged state, the original MnO₂ will dissolve into electrolyte when cycling within 50 cycles [14]. However, many nanowires still can be found when the electrode cycled for 50 to 500 cycle, as shown in Figure 3f–i, indicating stable cycling stability of 5%Ag-MnO₂.

To understand the energy storage mechanism of the 5%Ag-MnO₂ electrode, ex situ XRD analysis was conducted during the initial two cycles. As shown in Figure 4a, a set of diffraction peaks appear when discharged to 1.25 V during the first discharge cycle, and their intensity increases continuously until the ZIB is discharged to 0.8 V. In the meantime, the diffraction peaks from α-MnO₂ weakens during the discharge process. The set of newly appeared diffraction peaks can be attributed to Zn₄SO₄(OH)₆·5H₂O (PDF#78-0246 and PDF#39-0688), as shown in Figure 4b, which is a typical by-product in ZIBs [44]. The presence of Zn₄SO₄(OH)₆·5H₂O indicates relative alkaline environment near the electrode surface created by insertion of H⁺, and this reaction mechanism can be evidenced by the newly formed peak at 21.2° during discharge process, which can be assigned to α-MnOOH (PDF#24-0713). In the first charge process, the diffraction peaks from Zn₄SO₄(OH)₆·5H₂O weakens as the voltage increases, and those from α-MnOOH disappears when the voltage charged to 1.6 V. In the voltage region of 0.8–1.8 V, the diffraction peaks from 5%Ag-MnO₂ gradually intensifies as the voltage is increasing. Meanwhile, as shown in Figure 4c, a new peak located at 32.0° is formed at the charge voltage of 1.6 V, which can be attributed to the AgO phase (PDF#76-1489). Therefore, the Ag dopant participates in the redox reaction like a zinc-silver battery [45], and its reaction equation can be written as 2Ag + H₂O–2e⁻ ↔ Ag₂O+2H⁺. The signal of Ag cannot be found at full discharge state, which is the possible result of its low content and the strong diffraction peaks of Zn₄SO₄(OH)₆·5H₂O. In the second cycle, the charge–discharge process is highly reversible. The diffraction peaks of α-MnOOH are observed when the discharge and charge voltages are lower than 1.3 and 1.6 V, respectively. Considering the Zn²⁺ in the electrolyte is prerequisite for energy storage, the discharge/charge plateaus at high voltages are attributed to Zn²⁺ intercalation/extraction, while these at low voltages belong to H⁺ intercalation/extraction [15,46].

In addition, a flexible quasi-solid-state ZIB is successfully assembled using 5%Ag-MnO₂@graphite paper as cathode, Zn@cabon cloth as anode, and polyacrylamide as electrolyte (Figure 5a). As shown in Figure S10, the Zn particles are uniformly deposited on carbon cloth, and the XRD pattern of the electrode can be assigned to pure Zn metal (PDF#87-0713), indicating successful electrodeposition of Zn metal on carbon cloth without any impurities. The quasi-solid-state ZIB shows good flexiblity, which can be blended into different shapes, as shown in Figure 5b. The flexible ZIB has a high open circuit potential to 1.57 V (Figure 5c). As shown in Figure 5d, our quasi-solid-state ZIB can light a red light-emitting diode (LED) indicator under bending. Meanwhile, the flexible quasi-solid-state ZIB exhibits excellent cycling performance with a specific capacity of 171 mA h g⁻¹ and nearly 100% of coulombic efficiency maintained at 1 A g⁻¹ over 600 cycles (Figure 5e), which is better than most of other flexible quasi-solid-state aqueous ZIBs as shown in Table S3.

4. Conclusions

Ag-doped MnO₂ with a sea-urchin-like structure was designed by a hydrothermal method, which shows improved electrochemical behaviours as the cathode material for ZIBs. Upon Ag doping, the electron distribution of MnO₂ was finely tuned, resulting in reduced electrostatic repulsion for Zn²⁺ intercalation and lower charge transfer resistance. As a result, the Zn/5%Ag-MnO₂ ZIB presents highly reversible specific capacity (315 mA h g⁻¹ at 50 mA g⁻¹), enhanced cyclic stability (500 cycles with the capacity retention of 94.4%), and excellent rate performance. Moreover, the 5%Ag-MnO₂ is used to assemble flexible quasi-solid-state ZIBs, which exhibit outstanding mechanical flexibility and cyclability with a stable specific capacity of 171 mA h g⁻¹ achieved when cycled over 600 cycles.