1. Introduction
With the development of wearable devices and electric vehicles, there is a high demand for sustainable energy storage systems. Lithium-ion batteries (LIBs) are currently being extensively studied for high energy-density, wide electrochemical window and long cycle life [
1]. However, because of the limited resources of lithium, the high cost and the toxicity and flammability of organic electrolytes, LIBs assembled with organic electrolytes result in environmental pollution and poor safety performance. Also, the use of organic liquid electrolytes have the inevitable leakage, poor flexibility, etc. [
2]. Compared to organic electrolyte systems, aqueous rechargeable batteries are expected to be used on a large scale in the energy storage field due to their low production cost, environmental friendliness and high safety performance. In recent years, various aqueous batteries with multivalent metal ions (Zn
2+, Mg
2+, Ca
2+ and Al
3+) have been widely reported [
3]. Among them, the reserves of metallic Zn is 300 times than that of Li in the lithosphere. Aqueous zinc ion batteries (ZIBs) possess high theoretical specific capacity (820 mAh g
‒1 or 5854 mAh cm
‒3 )[
4], low toxicity, low fabrication cost, low redox potential (‒0.76 V) [
5], and inherent safety. However, current aqueous zinc ion batteries have a series of problems during long cyclic life, such as the corrosion of Zn negative electrode [
6] and the formation of Zn dendrites [
7]. Even, Zn dendrites might cause the separator to be punctured [
8] and then easily induce short circuit inside batteries [
9]. There are extensive research efforts on improving theirs cycling stabilities, for example, through interfacial modification of the zinc negative electrode, electrolyte formulation optimization and hydrogel electrolytes [
10]. Hydrogels are composed of polymers with hydrophilic functional groups, which promote water storage and structural integrity through weak physical hydrogen bonding. Meanwhile, polymer gel electrolytes act both as an electrolyte and as a separator, which can effectively avoid electrolyte leaking. Besides, hydrogel gel electrolytes ensures close contact between electrode and electrolyte, maintaining the integrity of the flexible ZIBs under external strain [
11].
Presently, some hydrogel gel electrolytes reported so far are based on synthetic polymers, such as polyacrylamide (PAM) [
12], polyvinyl alcohol (PVA) [
13], and polyacrylic acid (PAA) [
14], etc. Unfortunately, these hydrogels’ either poor mechanical strength or low ionic conductivity remains challenging. On the other hand, polymer hydrogels derived from nature matrices, such as sodium alginate (SA) [
15], guar gum (GG) [
16], xanthan gum (XG) [
17] and gelatin [
18], are cheap, biocompatible and hydrophilic, and widely used in flexible Zn-ion batteries. They are multiply cross-linked internally by hydrogen, ionic and covalent bonds to obtain a stable three-dimensional network structure, which results in improving their mechanical properties and ionic conductivities. A classical development in ZIBs hydrogel gel electrolytes is the use of sodium alginate (SA) as a polymer matrix to form a hierarchically three-dimensional Zn
2+-conductor gel electrolyte. SA consists of two monomer units, 𝛽-D-mannuronic acid (M-block) and 𝛼-L-guluronic acid (G-block) [
19]. Because of SA possessing high concentration of polar groups, high modulus and easy cross-linking with Zn
2+, many strategies based on natural polysaccharide-SA, such as guar gum/SA/glycol [
20], SA-polyacrylamide [
21] and gelatin/SA [
22], has been proposed to construct high performance ZIBs hydrogel gel electrolytes. These SA-based hydrogels exhibit high mechanical strength but they suffer from toxic raw materials [
23], invoking inert crosslinker initiators [
24] and complex preparation process [
25].
Hence, in this work, we prepare a flexible and stable N-isopropylacrylamide (NIPAM)/sodium alginate ZIBs hydrogel gel electrolytes (noted as N-SA) via a simple chain entanglement method. We directly incorporate NIPAM into the SA hydrogel to further enhance its mechanical strength and ionic conductivity. By soaking in 4 mol L‒1 (M) ZnSO4 and 0.1 M MnSO4 aqueous solution, the chain entanglements of short-chain NIPAM were formed. During soaking process, Zn2+, Mn2+ and SO42‒ ions could penetrate into the N-SA hydrogel matrix and give the high conductivity of the N-SA hydrogels. Thus, due to NIPAM chain entanglements, the N-SA hydrogels show a high conductivity of 2.96×10‒2 S cm‒1 at room temperature, and this hydrogel can effectively form a uniform Zn deposition and suppress side reaction. Specially, the assembled Zn/N-SA/MnO2 batteries can deliver the highest capacity of 182 mAh g‒1 (~98% retention) at a current density of 0.5 A g‒1 after 650 cycles. Therefore, this work provides a facile method to fabricate high-performance SA-based hydrogel electrolytes that illustrates their potential for flexible batteries for wearable electronics.
3. Results and Discussion
The preparation process of the N-SA gel electrolyte is shown in
Figure 1a. The N-SA aqueous solution is a transparent and homogeneous viscous liquid with disordered molecular chains. After adding 4 M ZnSO
4 + 0.1 M MnSO
4 solution into N-SA aqueous liquid solution for 30 min, the N-SA gel electrolyte with cross-linked structures is online formed through Zn
2+ coordinating function [
20].
Figure 1b-c shows an optical image of a uniform N-SA electrolyte, which is transparent and has excellent flexibility. The N-SA gel electrolyte can be bent at any angles and has a thickness of 325 μm as shown in
Figure 1d. The FTIR spectra of the SA and N-SA hydrogel electrolyte as well as pure SA and NIPAM powder were collected as shown in
Figure 1e. For the pure SA, the absorption peaks at 1590 cm
‒1 and 1401 cm
‒1 are attributed to the asymmetric stretching vibrations and symmetric stretching vibrations of -COO
‒ groups on the SA chains, respectively. In contrast, for the SA and N-SA hydrogel electrolyte, these two peaks shift to higher values because of coordinate bonds between -COO
‒ groups and Zn
2+ [
28]. Peaks were observed in the spectra of NIPAM at 532 cm
‒1 and 1517 cm
‒1, which assigned to the -NH wagging vibration and the C=O stretching vibration, respectively [
29]. For the N-SA hydrogel, these similar characteristic peaks also appear, which suggests the presence of NIPAM. The surface morphology of the N-SA gel electrolyte has uniformity and even (
Figure 1f). As determined by the cross-sectional morphology (
Figure 1d), the thickness of the N-SA gel electrolyte is 60
µm, which is lower than the thickness measured by micrometer. This reason is that the sample of SEM images is dried through freezing treatment.
Figure 2a shows EIS spectra of stainless steel (SS)/SS symmetrical batteries with the liquid, SA and N-SA gel electrolyte at room temperature. The corresponding ionic conductivity was calculated by the reported method [
30]. Notably, the N-SA gel electrolyte shows a high ionic conductivity of 2.96×10
‒2 S cm
-1, which is higher than that of the SA gel electrolyte and liquid electrolyte. LSV tests were conducted to evaluate the electrochemical stability of the electrolyte. As shown in
Figure 2b, the oxidation potential of the N-SA gel electrolyte is 2.51V (vs. Zn
2+/Zn). The LSV curve of the cells with the N-SA gel electrolyte also exhibits a reduction potential of (‒0.15 V vs. Zn
2+/Zn), which is lower than that of cells with liquid electrolyte (‒0.094 V vs. Zn
2+/Zn). These LSV results suggest that the N-SA gel electrolyte has a wide electrochemical window. In addition, we tested the CV curves of Zn/Zn symmetric cells at 1 mV s
‒1 as shown in
Figure S1. The nucleation overpotential of Zn
2+ in the N-SA hydrogel electrolyte is greater than that in the liquid electrolyte. The larger nucleation overpotential indicates a smaller nucleation radius, implying an easier homogeneous deposition. This demonstrates that the N-SA gel electrolyte is beneficial for small, denser and homogeneous of Zn deposition [
31]. Furthermore, the Tafel curve was used to measure the corrosion of Zn foil in the electrolyte as shown in
Figure 2c. Compared to the liquid electrolyte, the corrosion potential of zinc with the N-SA gel electrolyte is increased from ‒0.0084 to 0.0177 V, and corrosion current of zinc with the N-SA gel electrolyte is decreased from ‒5.984 to ‒5.286 A. Higher corrosion potentials and lower corrosion currents indicate a smaller tendency for corrosion reactions and lower corrosion rates, respectively[
32].
The Zn/Cu asymmetric cells with different electrolytes were used to study the cycling stability of zinc plating/stripping in this work. As shown in
Figure 2d-f, we compare the voltage/capacity curves of Zn/Cu of the N-SA, SA hydrogel electrolyte and liquid electrolyte for different numbers of cycles. For the N-SA cells cycled at 1 mA cm
‒2, the discharge/charge behavior kept the same over 500 cycles (see
Figure S2). However, the voltage profile of cells with the liquid electrolyte is increased during both the discharge and charge process. This means that the polarization potential of the N-SA hydrogel electrolyte was considerably lower than those of the liquid electrolyte and SA hydrogel electrolyte. The low polarization potential of zinc plating/stripping in the N-SA hydrogel electrolyte is beneficial for the uniformity of zinc deposition [
33]. Notably, the coulomb efficiency (CE) is a crucial parameter for investigating the reversibility and stability of the zinc negative electrode [
34]. As shown in
Figure 2e, the coulombic efficiency of the N-SA hydrogel cell increases from 86% in the first cycle to 98% after 20 cycles and remains stable at 99% for subsequent cycles. Meanwhile, the Zn/Cu cell with the N-SA hydrogel electrolyte can be stably cycled at 1 mA cm
‒2 for more than 500 cycles. In contrast, the CE of Zn/Cu cells with the liquid electrolyte and SA hydrogel electrolyte quickly decreases to 0 after 83 and 200 cycles, respectively. The high and stable coulombic efficiency of the N-SA cell is probably because the side reactions and dendrite growth are significantly restrained.
Figure 2h-i shows SEM images of Cu foils after Zn deposition for 5 h at 2 mA cm
‒2 in the Zn/Cu asymmetric cells with the N-SA hydrogel electrolyte and liquid electrolyte. For the liquid electrolyte, a large numbers of micron-grade dead Zn particles with irregular shapes appear on the surface of Cu, which is adverse for a high plating/stripping efficiency. By contrast, the deposited Zn is uniform and smoothly in the N-SA hydrogel electrolyte as shown in
Figure 2h. Therefore, this N-SA hydrogel electrolyte indicates a homogeneous nucleation process of zinc deposition.
Long-term cycling property of the symmetric Zn/Zn cells with the N-SA hydrogel electrolyte and liquid electrolyte at a current density of 1 mA cm
‒2 is shown in
Figure 3a-c. The cell with N-SA hydrogel electrolyte shows a stable voltage profile, which maintains a low level of about 100 mV over 2600 h without an obvious short circuit or any overpotential increasing. In contrast, the cell with the liquid electrolyte exhibits a sudden increase in polarization during the cycling process. This indicates that the N-SA hydrogel electrolyte possesses a stable Zn stripping/plating process. Furthermore, the voltage profiles of the Zn/Zn symmetrical cell with the N-SA gel electrolyte and liquid electrolyte were investigated for different current densities from 0.1 to 2.0 mA cm
‒2, as shown in
Figure 3d. When the current density increased from 0.1 mA cm
‒2 to 2.0 mA cm
‒2, the polarization voltages of the N-SA cells are always below 200 mV. However, the symmetric Zn cell with the liquid electrolyte suffer from shorts circuit. This suggests that the N-SA gel electrolyte is a promising candidate for ZIBs at wide current densities. The XRD patterns of the Zn negative electrode collected from the N-SA cells after 30 cycles displays similar signals compared to that of fresh Zn foils (
Figure 3e). However, the XRD patterns of Zn negative electrode cycled in the liquid electrolyte appear several new obvious peaks located at 8.2º, 16.3º and 24.5º (2θ). These peaks suggest the formation of Zn
4SO
4(OH)
6· 5H
2O (PDF#69-0688). This indicates that the N-SA hydrogel significantly inhibits interfacial side reactions, reduces the production of "dead" zinc, and results in a highly reversible and efficient plating/stripping of zinc ions.
Electrochemical performance of Zn/MnO
2 cells with the N-SA, SA hydrogel electrolyte and liquid electrolyte at 25℃ was shown in
Figure 4.
Figure 4a shows CV curves of the cells with the N-SA hydrogel electrolyte at 0.1 mV s
‒1 in the voltage range of 0.9~1.9 V at room temperature. The first cycle shows a single reduction peak at approximately ~1.17 V, while two reduction peaks at ~1.34 and ~1.22 V appear in the following cycles. The change in the peak number and position is attributed to the phase transition and morphology evolution during the first cycle [
35]. In the oxidation reaction, two overlapped peaks at 1.59 and 1.62 V are observed. This indicates a Zn
2+ insertion and extraction process in the charge storage mechanism [
35]. Besides, the CV curves in the second and third cycles nearly overlap, suggesting the electrochemical reversibility of the cells. As displayed in
Figure 4b-c, the Zn/MnO
2 cells with the N-SA hydrogel electrolyte exhibit reversible capacities of 310, 266, 176, and 104 mAh g
‒1 at current densities of 0.1, 0.2, 0.5, and 1 A g
‒1, respectively. When the current density decreases back to 0.5 mA cm
‒2, the capacities immediately recover to 180 mAh g
‒1. Thus, the cell with the N-SA hydrogel electrolyte exhibits a good rate performance. Long-term cycling stabilities of the Zn/MnO
2 cells with the N-SA, SA hydrogel electrolyte and liquid electrolyte at 0.5 A g
‒1 are shown in
Figure 4d. The Zn/MnO
2 cells with the N-SA hydrogel electrolyte maintain superior cycling stability compared to the cells in the SA and liquid electrolyte. The reversible capacities are 169, 171, 188, 185, and 182 mAh g
‒1 after 10, 200, 400, 500, 600 cycles (
Figure S2), respectively. Even, the N-SA cell delivers a highly reversible capacity of 182 mAh g
‒1 after 650 cycles, indicating a superior capacity retention of 98% and near 100% coulombic efficiency. This indicates that the N-SA hydrogel electrolyte possesses ultra-stable and highly reversible electrochemical performance during the long cycling, which maybe ascribed to the uniform Zn reversible deposition.
After charged back to 1.9 V after 300 cycles, the corresponding SEM images of two electrodes collected from Zn/MnO
2 cells are shown in
Figure S3 and
Figure 4e-f. As depicted in
Figure S3, some flake-like deposits are observed on the surface of the MnO
2 electrode in the liquid electrolyte after 300 cycles. In contrast to this flake-like surface, the MnO
2 surface collected from the N-SA hydrogel electrolytes remains relatively clean without any obvious flaked after cycling. Furthermore, as shown in
Figure 4e-f, the SEM images of Zn electrodes are collected from Zn/MnO
2 cells with the N-SA hydrogel electrolyte and liquid electrolyte after 300 cycles. The corresponding SEM images of Zn electrodes cycled after 500 cycles are also displayed in
Figure S4. It can be found that after cycling in the liquid electrolyte, the uneven Zn surface with a large number of sheet-like dendrites and byproducts was observed, which may be due to the interfacial degradation of Zn with liquid electrolyte. Conversely, after cycling with the S-NA hydrogel electrolyte, the surface of Zn electrode keeps very flat and uniform. Therefore, for the N-SA hydrogel electrolyte, the parasitic reaction between electrodes and electrolyte are effectively suppressed, which results in satisfactory long-term cycling stabilities of the assembled Zn/MnO
2 cells.