Ni-doped δ-MnO2 as a cathode for Zn-ion batteries

Among the cathode materials for zinc-ion batteries, manganese oxides have been extensively studied. Thermal decomposition method was adopted to synthesize the Ni-doped δ-MnO2 material, called as NixMn1-xO2. The electrochemical performance and energy storage mechanism of NixMn1-xO2 were studied in 2M ZnSO4 + 0.2M MnSO4 electrolyte. Nickel doping can significantly improve the electrochemical activity of δ-MnO2. Therefore, Ni-doped δ-MnO2 shows better electrochemical cycling performance than δ-MnO2. At 100 mA g−1, NixMn1-xO2 (x=0.2) the specific capacity of 280mAh g−1 in the second circle, and the corresponding voltage platform is 1.36V and 1.23 two discharge voltage platforms. After 100 cycles, the specific capacity of NixMn1-xO2 (x=0.2) still maintain 167 mAh g−1 at a current density of 100 mA g−1. In addition, we also proved that Zn2+ and H+ are co-intercalated into NixMn1-xO2 during the process of charge and discharge and the electrochemical reaction is highly reversible. This research broadens the thinking of exploring ZIBs cathode materials with good specific capacity and cycle performance.


Introduction
As we all know, lithium ion batteries (LIBs) have excellent characteristics in terms of energy density and technological maturity, however, as a commercial energy storage system, LIBs have huge obstacles such as high cost and the existence of dendrites [1]. In the development of multivalent batteries, waterbased zinc ion batteries (ZIBs) have been extensively studied due to their low cost and high safety. The main body has the characteristics of high theoretical capacity (820 mAh g-1), low redox potential (-0.76 V), rich crust content, and low cost [2].
Manganese oxides are promising cathodes for zinc-ion batteries due to their high specific capacity, rich resources and low cost. Among them, δ-MnO2 has a larger spacing of approximately 0.7 nm, which is theoretically more suitable for Zn 2+ storage [3]. However, δ-MnO2//Zn with excellent electrochemical performance has not been obtained up to now. Defect engineering is currently one of the means to improve the electrochemical performance of materials [4]. In particular, defect measures are widely used in manganese-based oxides. Currently, various types of defect methods have been used in manganesebased oxides as cathode materials for zinc-ion batteries to enhance electrochemical performance.
In this article, we synthesized layered δ-MnO2 by thermal decomposition and the nickel doping strategy modified δ-MnO2 to obtain NixMn1-xO2. The electrochemical performance and energy storage mechanism of NixMn1-xO2 were studied in 2M ZnSO4 + 0.2M MnSO4 electrolyte. Nickel doping can significantly improve the electrochemical activity of the material. Therefore, the Ni-doped δ-MnO2 samples show better electrochemical cycling performance than δ-MnO2. At 100 mA g -1 , NixMn1-xO2 2) the specific capacity of 280mAh g -1 in the second circle, and the corresponding voltage platform is 1.36V and 1.23 two discharge voltage platforms. After 100 cycles at a current density of 100 mA g -1 , the specific capacity of NixMn1-xO2 (x=0.2) is still 167 mAh g -1 . Besides, we also proved that Zn 2+ and H + are co-intercalated into NixMn1-xO2 upon cycling and the electrochemical reaction is highly reversible.

Material preparation and Characterization
The raw materials used in this experiment are all analytically pure and have not been further purified. δ-MnO2 is synthesized by thermal decomposition method [5]. NiSO4 and KMnO4 were dissolved in 40 mL of distilled water in the proportions of (0:1; 1:10; 1:5; 2:5), followed by stirring at room temperature for 2 hours, and then heating to obtain the precursor. The precursor was put into a muffle furnace and calcined at 350°C for 5 hours with a heating rate of 5°C/min. Then, the resultant was washed by the suction filtration to remove the unreacted potassium permanganate and nickel sulfate in the product, and then dried in a vacuum drying oven at 60 ℃ overnight. The collected product is NixMn1-xO2 doped with different content of nickel (x=0, 0.1, 0.2, 0.4). Figure 1 is the X-ray diffraction (XRD) characterization diagram of NixMn1-xO2 (x = 0, 0.1, 0.2, 0.4). Figure 1 shows that the XRD diffraction peaks of NixMn1-xO2 are almost similar. All the XRD diffraction patterns of the as-prepared materials correspond to the standard card of birnessite (JCPDS 13-0105). This phenomenon indicates that Ni has successfully entered the crystal lattice of δ-MnO2 to form a single-phase NixMn1-xO2 (0≤x≤ 0.4). Obviously, there is no impurity peak existing in these Ni-doped MnO2 materials. δ-MnO2 possesses a crystal structure shared by the edges of MnO6 on the (001) crystal plane. The diffraction peaks point to different crystal planes of δ-MnO2 [6], respectively. In brief, δ-MnO2 and Ni doped derivatives show high crystallinity and high purity. In order to confirm the element and valence composition of the material before and after nickel doping, the material NixMn1-xO2 (x = 0, 0.2) was tested by X-ray Photoelectron Spectroscopy (XPS). Figure 2(a) shows the XPS spectra of δ-MnO2 and NixMn1-xO2 (x =0.2). As can be seen in Figure 2(a), the XPS spectra of δ-MnO2 and NixMn1-xO2 (x =0.2) have the split peaks of Mn 2p and O 1s. In addition, the XPS spectrum of NixMn1-xO2 (x = 0.2) is composed of the Ni2p splitting peak, implying the successful Ni-doping of δ-MnO2. As can be also seen in Figure 2(a), the peak at 373.7 eV corresponds to the K 2s splitting peak, which is attributed to the adsorption of the K + of the raw material KMnO4. Ni 2p3/2, which is consistent with the previous literature reported [7]. This result demonstrates the existence of Ni 2+ in NixMn1-xO2. Figure 2(c) displays the Mn 2p spectra of δ-MnO2 and NixMn1-xO2 (x = 0.2). As can be seen in Figure 2(c), the spin splitting energies of Mn for δ-MnO2 and NixMn1-xO2 (x = 0.2) are 11.7 eV. According to the relevant literature [8], the corresponding Mn valence of the compound is +4. It is inferred that Ni 2+ enters the interlayer in the manganese dioxide, existing in the form of ions. Thus, the valence state of Mn in Ni-doped δ-MnO2 has not changed, similar to La 3+ intercalation into δ-MnO2 [9].

Micro-morphology
In order to explore the morphology changes of the material before and after Ni doping δ-MnO2, the material was tested by SEM. Figure 3 shows the morphology of NixMn1-xO2 (x=0, 0.2). It can be seen that both materials are in the form of random particles. In brief, the morphology of the material with or without Ni doped material is almost not much different, indicating that Ni doping will not have a significant impact on the morphology of the material.

Electrochemical performance
The cyclic voltammogram and galvanostatic charge-discharge measurements are adopted to evaluate the electrochemical performance of NixMn1-xO2 as the cathode material of ZIBs in the electrolyte of 2 M ZnSO4 + 0.2 M MnSO4. The cyclic voltammetry curves of NixMn1-xO2 (x =0.2) and δ-MnO2 are depicted in Figure 4(a-b), respectively. As can be shown in Figure 4(a-b), the shape of the CV curves for NixMn1-xO2 (x =0.2) is similar to that of the CV curve for δ-MnO2. From Figure 4(a), NixMn1-xO2 (x = 0.2) has a reduction peak at 1.17 V vs. Zn 2+ /Zn and an oxidation peak at 1.61 V vs. Zn 2+ /Zn in the first cycle. In the later loop, there are two obvious redox pairs of redox couple appearing at 1.60/1.36 V and 1.63/1.23 V vs. Zn 2+ /Zn. As for δ-MnO2, the CV curve in Figure 4(b) has stronger oxidation peak and reduction peak in the first cycle, but the peak intensity is decreased after 3 cycles. Figure 4(c) shows the cycle stability performance of NixMn1-xO2 (x = 0, 0.1, 0.2, 0.4) in different Ni doping amounts at 100 mA g -1 for 10 cycles activity, followed by cycling at 200 mA g -1 . As can be seen in Figure 4(c), the discharge specific capacities of NixMn1-xO2 (x = 0~0.4) are decreased with the increasing of cycle number. Obviously, the cycle stability of NixMn1-xO2 (x =0.2) is better than that of δ-MnO2 and other δ-MnO2 in different Ni doped amount. After 110 cycles, the specific capacity of NixMn1-xO2 (x = 0.2) still achieves 167 mA g -1 , which is significantly higher than δ-MnO2 (77 mAh g -1 ) and NixMn1-xO2 (x = 0.1, 151 mAh g -1 ) and NixMn1-xO2 (x =0.4, 145 mAh g -1 ). Therefore, Therefore, Ni doping plays a key role in improving the electrochemical performance of δ-MnO2 and x=0.2 is the optimal doping amount. Figure 4(d) shows the discharge capacity-voltage profiles of NixMn1-xO2 (x = 0.2) at 100 mA g -1 . In Figure 4(d), NixMn1-xO2 (x = 0.2) has obvious discharge plateaus at 1.36 V and 1.23 V vs. Zn 2+ /Zn, in accordance with in the previous CV curve. At 100 cycles, NixMn1-xO2 (x = 0.2) delivers the first discharge specific capacity of ~230mAh g -1 , then reaches up to ~260mAh g -1 .  Even at high density of 3.2 A g -1 , NixMn1-xO2 (x = 0.2) exhibits a higher specific capacity than δ-MnO2. In Figure 5(a), the specific capacity of NixMn1-xO2 (x = 0.2) can still reach 100 mAh g -1 at 3.2 A g -1 . When the current density is 0.1 A g -1 , the specific capacity of NixMn1-xO2 (x = 0.2) is risen to 218 mAh g -1 , which is much higher than that of δ-MnO2 (167 mAh g -1 ). This result indicates that the Ni doping can improve the rate performance of δ-MnO2. Figure  5(b) and (c) displays the specific capacity-voltage curves of δ-MnO2 and NixMn1-xO2 (x = 0.2) at different current densities. A Figure 5(b-c) shows that at 0.1 A g -1 , at 0.1 A g -1 , both δ-MnO2 and NixMn1-xO2 (x = 0.2) have two couples of charge-discharge platforms at 1.55/1.35V and 1.50/1.40V. Compared with δ-MnO2, NixMn1-xO2 (x = 0.2) yields the specific capacity of ~100 mAh g -1 at 3.2 A g -1 , higher than that of δ-MnO2 (~75 mAh g -1 ). Figure 5(d) shows long-term performance and coulombic efficiency of δ-MnO2 and NixMn1-xO2 (x = 0.2) at 1 A g -1 . In the early stage, their specific capacities undergo rapid capacity decay, and then tend to stability. The coulombic efficiencies of δ-MnO2 and NixMn1-xO2 (x = 0.2) are close to 100%. In contrast to δ-MnO2, NixMn1-xO2 (x = 0.2) has better long-cycle performance. After 600 cycles, NixMn1-xO2 (x = 0.2) still remains the specific capacity of 72 mAh g -1 , while the specific capacity of δ-MnO2 is only 46 mAh g -1 . In short, the Ni-doped NixMn1-xO2 (x = 0.2) shows better long-term and rate performance than δ-MnO2. The Ni doping strategy can effectively increase the performance of layered δ-MnO2 as the cathode material of ZIBs.    6 (x = 0.2) electrodes, leading to no obvious change of the corresponding XRD diffraction profile [12]. In the second cycle, when discharged to 1.34 V (L4) and 1.0 V (L5), the NixMn1-xO2 (x = 0.2) electrode appears the new peaks of Zn4SO4(OH)6•xH2O and MnOOH. In the second cycle, it is speculated that H + and Zn 2+ are co-intercalated into the NixMn1-xO2 (x = 0.2) electrode. When charged to 1.8 V, the XRD diffraction peaks of the NixMn1-xO2 (x = 0.2) electrode is returned to the initial state, which indicates that the NixMn1-xO2 (x = 0.2) material has good electrochemical reversibility. The discharge and charge reactions of the NixMn1-xO2 (x = 0.2) in 2M ZnSO4 + 0.2M MnSO4 electrolyte are listed as follows [10]:

Reaction kinetic analysis
In order to explore the ion diffusion kinetics of NixMn1-xO2 (x=0.2) in zinc-ion batteries, the cyclic voltammetry tests are conducted at the sweep rates of 0.2-1.0 mV/s. Figure 7(a) and (b) show the CV curves and its corresponding Log(i)-log(v) linear fitting diagrams, respectively. As shown in Figure 7(a) as the sweep rate is increased, the intensity of the oxidation peaks and the reduction peaks also are increased. Generally speaking, from the formula i = av b in the CV curves. According to the value of b, the electrochemical mechanism of surface capacitance and diffusion capacitance can be determined. As can be seen in Figure 7(b), the b values of Peak 1 and Peak 4 are ranged from 0.5 to 1, which demonstrates that electrochemical process is dominated by the diffusion capacitance and the surface capacitance. In addition, the b values of the Peak 2 and Peak 3 are close to 0.5, which are mainly controlled by the diffusion capacitance. Figure 7(c) shows the proportions of diffusion capacitance and surface capacitance for the NixMn1-xO2(x= 0.2) electrode at different scanning rates. As the scanning rate is increased, the proportion of surface capacitance contribution also is improved, from 43.3% at 0.2 mV/s to 63.3% at 1.0 mV/s. These results, indicate that the Ni-doped NixMn1-xO2(x= 0.2) electrode has the characteristics of fast ion diffusion kinetics in ZIBs.

Conclusion
All in all, we have developed a Ni-doped layered birnessite δ-MnO2 as an excellent cathode material for water-based ZIBs. promotes nickel Zn 2+ intercalation/deintercalation kinetics, advanced NixMn1-xO2 (x=0.2) has larger zinc storage capacity and better rate performance. The specific capacity of Ni-doped materials is obviously better than that of undoped materials. Among them, NixMn1-xO2 (x=0.2) has the most superior electrochemical performance. After 100 cycles of 200 mA g -1 , the specific capacity is still