Novel K2Ti8O17 Anode via Na+/Al3+ Co-Intercalation Mechanism for Rechargeable Aqueous Al-Ion Battery with Superior Rate Capability

A promising aqueous aluminum ion battery (AIB) was assembled using a novel layered K2Ti8O17 anode against an activated carbon coated on a Ti mesh cathode in an AlCl3-based aqueous electrolyte. The intercalation/deintercalation mechanism endowed the layered K2Ti8O17 as a promising anode for rechargeable aqueous AIBs. NaAc was introduced into the AlCl3 aqueous electrolyte to enhance the cycling stability of the assembled aqueous AIB. The as-designed AIB displayed a high discharge voltage near 1.6 V, and a discharge capacity of up to 189.6 mAh g−1. The assembled AIB lit up a commercial light-emitting diode (LED) lasting more than one hour. Inductively coupled plasma–optical emission spectroscopy (ICP-OES), high-resolution transmission electron microscopy (HRTEM), and X-ray absorption near-edge spectroscopy (XANES) were employed to investigate the intercalation/deintercalation mechanism of Na+/Al3+ ions in the aqueous AIB. The results indicated that the layered structure facilitated the intercalation/deintercalation of Na+/Al3+ ions, thus providing a high-rate performance of the K2Ti8O17 anode. The diffusion-controlled electrochemical characteristics and the reduction of Ti4+ species during the discharge process illustrated the intercalation/deintercalation mechanism of the K2Ti8O17 anode. This study provides not only insight into the charge–discharge mechanism of the K2Ti8O17 anode but also a novel strategy to design rechargeable aqueous AIBs.


Introduction
Rechargeable lithium-ion batteries as promising energy storage devices have been widely applied in portable electronic devices and electric vehicles [1,2]. However, lithiumion batteries may be not the best choice for large-scale energy storage applications and practical power grids, concerning its uneven distribution and the long-term unavailability of lithium resources. Consequently, 'beyond-Li-ion' batteries, including Mg-based batteries [3], Zn-based batteries [4,5], Na-based batteries [6], and Al-based batteries [7][8][9], have emerged as promising alternatives for electrochemical energy storage. Therein, aluminumbased batteries (ABs) exhibit considerable advantages in terms of cost-effectiveness, safety, and high theoretical specific capacity due to the abundant aluminum reserves, nonflammability, light weight, and three-electron redox properties [10][11][12]. The trivalent aluminum ion, in principle, provides three reactive electrons involved in the electrochemical processes ( Al 3+ + 3e − ↔ Al ), endowing a high volumetric capacity of 8040 mAh cm −3 and a gravimetric capacity of 2980 mAh g −1 [13]. Furthermore, aluminum metal can be handled in ambient atmosphere, thus facilitating cell fabrication and resulting in a high level of safety of electrochemical storage systems [14].
According to the electrolytes, ABs are classified into aqueous and nonaqueous systems. In nonaqueous ABs, the aluminum metal can be employed directly as the anode [12,15]. However, high-cost ionic liquid electrolytes or complicated molten salt electrolytes limit the wide practical application of nonaqueous ABs [16,17]. Therefore, a growing number of researchers have turned their attention to the aqueous ABs due to the reduced cost and convenient operation of aqueous electrolytes. Unfortunately, the reversible charging process of ABs in aqueous electrolytes is hindered by the competitive hydrogen evolution reaction (HER). The lower redox potential of Al 3+ /Al compared to that of H + /H 2 implies that the Al (aluminum) metal anode can only be used in primary ABs rather than rechargeable ABs for aqueous electrolytes. Due to thermodynamics, the development of appropriate intercalation/deintercalation-type anode materials plays a critical role in the assembly of rechargeable aqueous ABs [17][18][19], which can be called aluminum-ion batteries (AIBs). Liu et al. proposed an aqueous AIB prototype using a TiO 2 anode and AlCl 3 electrolyte for the first time, evidencing the reversible intercalation/deintercalation of Al 3+ ions into/from TiO 2 [15]. Subsequently, they used another AIB device using TiO 2 as the anode, while copper hexacyanoferrate and Al 2 (SO 4 ) 3 were employed as the cathode and electrolyte, respectively. It delivered an acceptable discharge capacity of 21 mAh g −1 with a discharge voltage of 1.6 V [20]. Thus, TiO 2 is considered a promising candidate as an aqueous AIB anode. An enhanced rate capability or better cycling performance was required via selecting a superior electrolyte or adding H 2 evolution inhibitors for the AIB system [21].
Herein, we developed a novel aqueous AIB system consisting of a layered-structure K 2 Ti 8 O 17 anode obtained via a facile hydrothermal method. An AlCl 3 aqueous solution with NaAc additive was employed as the electrolyte, and activated carbon (AC) was used as the cathode. The aqueous AIB delivered an acceptable discharge capacity of 189.6 mAh g −1 based on the anode materials. An enhanced rate performance and high cycling stability were achieved for the assembled AIB. Inductively coupled plasmaoptical emission spectroscopy (ICP-OES), high-resolution transmission electron microscopy (HRTEM), and K-edge X-ray absorption near-edge spectroscopy (XANES) techniques were conducted to provide evidence of the mechanism descriptions of Na + /Al 3+ ion intercalation/deintercalation in aqueous AIB, inspiring a great potential to explore promising aqueous AIB anode materials.

Preparation of K 2 Ti 8 O 17 Electrode
The binder-free K 2 Ti 8 O 17 electrode material was prepared via a two-step method, including a hydrothermal reaction combined with a post-pyrolysis process. First, titanium foil with a thickness of 0.1 mm was cut into a patch with an area of 2.0 × 0.5 cm 2 . Then, ultrasonic treatments were sequentially carried out in acetone, ethanol, and deionized water for 20 min. The clean titanium foil was immersed in a mixture of hydrofluoric acid, nitric acid, and deionized water with a volume ratio of 1:2:7 for 2 min, and then repeatedly washed using deionized water. Subsequently, the treated titanium foil was put into a stainless-steel autoclave lined with polytetrafluoroethylene containing 9.08 g of KOH and 50 mL of deionized water. The hydrothermal reaction was performed at 220 • C for 24 h. After cooling to room temperature, the excess alkali solution and loose sample were washed repeatedly with deionized water, and then soaked in 30 mL of 0.6 M of hydrochloric acid for 30 min, and then the excess hydrochloric acid solution was washed again. The final titanium-supported K x Ti y O 2 sample was obtained after a pyrolysis process in a muffle furnace at 400 • C for 1 h. The mass of the loaded active substance as the K 2 Ti 8 O 17 electrode was approximately 1.0 mg cm −2 .

Preparation of AC Electrode
The AC electrode was prepared by coating AC slurry on a 2.0 × 2.0 cm 2 titanium mesh. The slurry was composed of AC, polytetrafluoroethylene (PTFE, 1 wt. %), and acetylene black (AB) with the mass ratio of AC:PTFE:AB = 8:1:1, uniformly dispersed in anhydrous ethanol. After the alcohol volatilized, the AC slurry was evenly coated on the titanium mesh, and the titanium mesh coated with AC was dried at 80 • C for 12 h to obtain the final AC cathode electrode. The mass of loaded active substance of the AC electrode was about 8 mg cm −2 .

Physical Characterization of K 2 Ti 8 O 17 Electrode
The phase structure of the K 2 Ti 8 O 17 electrode was characterized by X-ray diffraction (XRD, Bruker AXS D8 (Akishima, Tokyo, Japan) diffractometer with Cu K α radiation with λ of 0.15418 nm). To obtain a clear diffraction peak, it was necessary to scrape the sample off the Ti substrate and test the electron diffraction patterns of the powder sample. The morphology of the K 2 Ti 8 O 17 electrode was characterized by field-emission scanning electron microscopy (FESEM, SUPRA55, Carl Zeiss Corp., Jena, Germany), transmission electron microscopy (TEM, HT7700, Hitachi Corp., Tokyo, Japan, 100 kV, 10 µA), and high-resolution TEM (HRTEM, FEI Titan 80~300 kV, FEI Corp., Hillsboro, OR, USA). The K 2 Ti 8 O 17 electrode was cut into small pieces to stick onto the conductive tape for SEM measurement. A drop of slurry, uniformly suspended in ethanol, was dispersed onto the amorphous carbon film supported on a Cu grid for TEM and HRTEM analysis. Elemental analysis of the cycled K 2 Ti 8 O 17 electrode was carried out on the ICP-OES system (730, Agilent, Palo Alto, CA, USA). The obtained sample (20 mg) was dissolved into 5 mL of concentrated nitric acid and then 25 mL of deionized water was added for the ICP-OES test. K-edge X-ray absorption near-edge spectroscopy (XANES) measurements (Beijing, China) of Ti in L-edge and O K-edges were performed at the 4B7B soft X-ray experiment station of the Beijing Electron Positron Collider (BEPC) (Beijing, China). Powder samples were mounted on electrically conducting carbon tape.

Electrochemical Properties of the K 2 Ti 8 O 17 and AC Electrodes
The electrochemical properties of the K 2 Ti 8 O 17 and AC electrodes were evaluated by cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD) based on the CHI 604E electrochemical station (CH Instruments Corp., Shanghai, China) and Land CT 2001A instruments (Land Instruments Corp., Wuhan, China), respectively. All the measurements were performed by the three-electrode system both in the 1 M AlCl 3 solution and the mixed solution of 1 M of AlCl 3 and NaAc. The saturated calomel electrode (SCE, all the voltage values in this paper are herein referred to as SCE) and the 1 × 1 cm 2 Pt foil were used as the reference electrode and counter electrode, respectively. To further evaluate the practical application of the K 2 Ti 8 O 17 electrode in the aluminum ion electrolyte, an AIB battery was assembled using AC as the cathode electrode pasted onto the titanium mesh (K 2 Ti 8 O 17 /Ti, 2 × 2 cm 2 ) and K 2 Ti 8 O 17 as the anode electrode (1 × 2 cm 2 ) in the hybrid electrolyte of 1 M of AlCl 3 and 3 M of NaAc (noted as AlCl 3 /NaAc).

Results and Discussion
As shown in the schematic (Figure 1a), the nanobelt K 2 Ti 8 O 17 is formed on the surface of the Ti foil via the facile hydrothermal reaction combined with a post-pyrolysis process. Figure 1b displays the XRD patterns of the as-synthesized K 2 Ti 8 O 17 nanobelts. All diffraction peaks are indexed as a layered K 2 Ti 8 O 17 phase (JCPDF NO. 84-2057) with the space group of C2/m (12). The SEM image shows a network distribution of whiskers-like K 2 Ti 8 O 17 (Figure 1c). The magnified image shows the nanoclusters aggregated with the intertwined nanobelts (Figure 1d). The formed pore structure inside the nanoclusters facilitates the electrolyte to reach the crystal structure. The TEM image in Figure 1e

Results and Discussion
As shown in the schematic (Figure 1a), the nanobelt K2Ti8O17 is formed on the surface of the Ti foil via the facile hydrothermal reaction combined with a post-pyrolysis process. Figure 1b displays the XRD patterns of the as-synthesized K2Ti8O17 nanobelts. All diffraction peaks are indexed as a layered K2Ti8O17 phase (JCPDF NO. 84-2057) with the space group of C2/m (12). The SEM image shows a network distribution of whiskers-like K2Ti8O17 (Figure 1c). The magnified image shows the nanoclusters aggregated with the intertwined nanobelts (Figure 1d). The formed pore structure inside the nanoclusters facilitates the electrolyte to reach the crystal structure. The TEM image in Figure 1e reveals an individual K2Ti8O17 nanobelt with a length of several microns. The SAED pattern and HRTEM image (Figure 1f   The electrochemical properties of the as-prepared K 2 Ti 8 O 17 electrode were first investigated in 1 M of AlCl 3 aqueous electrolyte via cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD) curves at room temperature (25 • C). Figure 2a displays the CV curves at different potential ranges with a sweep rate of 25 mV s −1 , showing that the K 2 Ti 8 O 17 electrode can be operated at the negative potential range from 0.0 V to −1.4 V VS. SCE . An ultra-high overpotential for the hydrogen evolution reaction (HER) is observed after the Al 3+ ions intercalation, indicating that the Al 3+ ions intercalation is prior to the HER process. The ultra-high hydrogen evolution overpotential is ascribed to the low catalytic activity of both K 2 Ti 8 O 17 nanobelts and Ti substrate for HER [22]. Al 3+ ions with a smaller radius than that of Li + ions facilitate the intercalation/deintercalation reactions for the charge-discharge process, inducing the high reversibility of the insertion/extraction of Al 3+ ions into/from the anatase TiO 2 electrode [15]. Similarly, it is possible that the Al 3+ ions could insert/extract into/from the K 2 Ti 8 O 17 electrode. Charge and discharge plateaus are observed in GCD curves at different potential ranges ( Figure S1a), in good agreement with the CV curves. CV measurements were performed at different scan rates, as shown in Figure 2b. The peak currents of the K 2 Ti 8 O 17 electrode gradually increase with the scan rates. Under different sweep rates, the CV curves of the layered K 2 Ti 8 O 17 electrode exhibit similar quasi-symmetric shapes, illustrating the good reversibility of the Al 3+ ions insertion/extraction. The electrochemical behavior of Al 3+ ions in the K 2 Ti 8 O 17 electrode was evaluated by analyzing the collected CV data at various sweep rates according to the following equation: where i is the measured current and v represents the sweep rate. Both a and b refer to the adjustable parameters. The b-values are determined from the slope by plotting log i vs. log v. Well-defined b values approaching 1 indicate capacitive characteristics, while b values close to 0.5 represent diffusion-controlled characteristics of the electrode [23]. Here, the b values were determined as 0.527 and 0.572 from the redox peaks (Figure 2c), indicating that the intercalation/deintercalation of Al 3+ ions in the K 2 Ti 8 O 17 electrode is a diffusioncontrolled process. Figure 2d presents the cycling performance of the K 2 Ti 8 O 17 electrode at different potential ranges with a current density of 10 A g −1 . As shown, the specific capacity increases with the potential ranges, while the stability decreases. The similar shape of the charge and discharge curves also reveals the good reversibility of the K 2 Ti 8 O 17 electrode ( Figure 2e). The coulombic efficiencies (CEs) of the electrode at the current density of 10 A g −1 decrease with the cycling number ( Figure S1b), implying the degradation of the reversibility with the increase in cycles. The cycling stability under various current densities at −1.0~0 V vs. SCE is shown in Figure 2f. The K 2 Ti 8 O 17 electrode exhibits a specific capacity of 224 mAh g −1 at the current density of 2 A g −1 with the expense of cycling stability. The specific capacity of 107 mAh g −1 is maintained at the current density of 20 A g −1 . It is surprising that the specific capacity is still 43 mAh g −1 at the current density of 50 A g −1 with a superior cycling performance. The CEs of the K 2 Ti 8 O 17 electrode reach up to approximately 100% at the high discharge current density of 50 A g −1 ( Figure S1c).  Figure S2 in the supporting information. The compositional characterizations and electrochemical performance of the K 2 Ti 8 O 17 sample undergoing a 24 h hydrothermal process without the pyrolysis step are shown in Figure S3 in the supporting information. Compared with the K 2 Ti 8 O 17 sample before and after undergoing the pyrolysis step, the obviously enhanced electrochemical performance is achieved for the one undergoing pyrolysis.  To alleviate the effect of the acidic AlCl 3 electrolyte on the electrode, various concentrations of the NaAc was added into the AlCl 3 electrolyte as a buffer agent. The electrochemical performance of the K 2 Ti 8 O 17 electrode in the hybrid AlCl 3 /NaAc aqueous electrolyte was evaluated by CV and GCD measurements (Figures 3 and S4). The CV curves at different potential ranges show the symmetric shapes ( Figure S4a), implying the good reversibility of the AIBs with the K 2 Ti 8 O 17 anode in the hybrid AlCl 3 /NaAc aqueous electrolyte. Compared with the battery with pure AlCl 3 electrolyte, the cycling stability of the battery under a current density of 2 A g −1 improves via the addition of NaAc with the concentrations of 2 M and 3 M (Figure 3a). The corresponding CEs based on the GCD tests also improve with the addition of NaAc solution ( Figure S4b), although the discharge capacity of the battery slightly decreases ( Figure S4c). This demonstrates that the NaAc additive is a benefit for the rechargeable intercalation/deintercalation of Al 3+ ions. The b value plots for log i against log v of the K 2 Ti 8 O 17 anode were calculated on the basis of the CV curves at different scan rates ( Figure S4d) to obtain the b values via Equation 1. The calculated b values are 0.56 and 0.505 in the AlCl 3 /NaAc hybrid electrolyte (Figure 3b), suggesting a diffusion-controlled process in the K 2 Ti 8 O 17 electrode [11,24]. The cycling performance of the charge-discharge curves of the K 2 Ti 8 O 17 electrode in the aqueous AlCl 3 /NaAc electrolyte shows a similar trend to that in the AlCl 3 electrolyte (Figures 3c and S5a), exhibiting a decreasing trend with the potential ranges. In contrast, the cycling stability in the AlCl 3 /NaAc hybrid electrolyte is improved compared with that in the single AlCl 3 electrolyte. To analyze the elemental composition of the cycled electrode, ICP-OES experiments for the cycled K 2 Ti 8 O 17 electrodes in both charge and discharge states, as well the tested electrolyte, were performed. The cycled K 2 Ti 8 O 17 electrode in the charge state was attained when it was charged to −1 V after charging/discharging for 10 cycles. The electrode in the discharge state was obtained when it was discharged to 0 V after charging/discharging for 10 cycles, and the tested electrolyte was collected simultaneously. All the samples were immersed into deionized water overnight and then repetitively washed before the ICP-OES test. From the experimental results, the molar ratio of Ti:K:Al:Na is 1:0.084:0.011:0.003 with the total cation charge of K, Al, and Na of around 0.96 in the charge state, while it is 1:0.087:0.016:0.002 with the total cation charge of K, Al, and Na of around 1.096 under the discharge state. The difference in K + ions under charge and discharge conditions indicates the contributions from K + insertion/deintercalation. Additionally, the difference of 0.136 in terms of the total cation charge of K, Al, and Na implies that the proton seems to also contribute to the charge/discharge process. Obviously, a decrease in the K + ion is detected in the cycled electrode, while the Al 3+ ion and a trace of the Na + ion are detected. Therefore, we can conclude that the ionic substitutions of K + ions by Al 3+ /Na + ions occur in the K 2 Ti 8 O 17 electrode during the charge/discharge process. Simultaneously, the K + ions are detected in the cycled electrolyte, corresponding to the dissolution of K + from the electrode to electrolyte during the intercalation of Al 3+ and Na + ions. The ICP-OES measurement for the cycled K 2 Ti 8 O 17 electrode after washing shows the existence of Na + and Al 3+ ions. This indicates a possible mechanism of Na + and Al 3+ co-intercalation/deintercalation during the charge/discharge process. Although the specific capacitance of the K 2 Ti 8 O 17 electrode in the aqueous AlCl 3 /NaAc electrolyte (159 mAh g −1 ) is lower than that in the AlCl 3 electrolyte (234 mAh g −1 ), the cycling stability and CEs during the cycling are significantly improved (Figure 3d). Similar charge and discharge curve shapes reveal the excellent reversibility of the K 2 Ti 8 O 17 electrode in the AlCl 3 /NaAc hybrid electrolyte (Figure 3e). The K 2 Ti 8 O 17 electrode exhibits a superior rate capability and delivers good discharge capacities of 159, 141, 120, and 93 mAh g −1 with increasing current densities of 2, 5, 10, and 20 A g −1 , respectively (Figure 3f). The corresponding CEs at the current densities of 2, 5, 10, and 20 A g −1 are approximately 90%, 93%, 94%, and 95%, respectively. These results demonstrate excellent rate-performance of the K 2 Ti 8 O 17 electrode in the AlCl 3 /NaAc electrolyte. It is interesting that the cycling performance and CEs of the K 2 Ti 8 O 17 electrode are improved with the increase in the charge-discharge current density ( Figure S5b,c). The K 2 Ti 8 O 17 electrode exhibited good cycling stability when the electrode performed in the potential range of −1.0~0 V at a current density of 10 A g −1 (Figure S5d), implying the promising practical application of the electrode at high current densities. Definitely, the electrolyte also has a disparate influence on different anodes [25,26]. To further probe the microstructure evolution of the K2Ti8O17 electrode during the charge-discharge cycling process in both the pure AlCl3 and the AlCl3/NaAc hybrid electrolytes, HRTEM measurements were performed and are shown in Figure 4. Three welldefined regions are presented in the K2Ti8O17 electrode after cycling in the aqueous AlCl3 electrolyte (Figure 4a). After electrochemical activation during the charge-discharge process, the phase transition occurs in the surface region due to the Na + /Al 3+ intercalation/deintercalation in the K2Ti8O17 lattice, forming an amorphous phase (region 1 in Figure 4a). The corresponding fast Fourier transform (FFT) pattern confirms the amorphous structure (inset 1 of Figure 4a). Meanwhile, the well-defined crystalline lattice is still observed in the bulk of the K2Ti8O17 (region 3 in Figure 4a). The calculated lattice fringe distance of 0.37 nm via FFT corresponds to the (110) plane of the layered K2Ti8O17 phase. Between region 1 and 3, there is a crystallized layer that reveals a different FFT pattern from that of the bulk (region 2 in Figure 4a). The difference could be the orientation change of the K2Ti8O17 phase during the charge-discharge cycling process. The lattice plane distances of 0.31 nm, as well as the 73° cross angle of the lattice plane shown in Figure 4b, indicate that  Figure S4b. (c) GCD curves with different potential ranges at the current density of 10 A g −1 .
(d) Cycling performance in areal specific capacitances with current density of 2 A g −1 . (e) GCD curves with different current densities in the potential range from −1 to 0 V VS. SCE . (f) Discharge capacitance and corresponding coulombic efficiencies at current densities ranging from 2 to 20 A g −1 in a potential range from −1 to 0 V VS. SCE .
To further probe the microstructure evolution of the K 2 Ti 8 O 17 electrode during the charge-discharge cycling process in both the pure AlCl 3 and the AlCl 3 /NaAc hybrid electrolytes, HRTEM measurements were performed and are shown in Figure 4. Three well-defined regions are presented in the K 2 Ti 8 O 17 electrode after cycling in the aqueous AlCl 3 electrolyte (Figure 4a). After electrochemical activation during the charge-discharge process, the phase transition occurs in the surface region due to the Na + /Al 3+ intercalation/deintercalation in the K 2 Ti 8 O 17 lattice, forming an amorphous phase (region 1 in Figure 4a). The corresponding fast Fourier transform (FFT) pattern confirms the amorphous structure (inset 1 of Figure 4a). Meanwhile, the well-defined crystalline lattice is still observed in the bulk of the K 2 Ti 8 O 17 (region 3 in Figure 4a). The calculated lattice fringe distance of 0.37 nm via FFT corresponds to the (110) plane of the layered K 2 Ti 8 O 17 phase. Between region 1 and 3, there is a crystallized layer that reveals a different FFT pattern from that of the bulk (region 2 in Figure 4a). The difference could be the orientation change of the K 2 Ti 8 O 17 phase during the charge-discharge cycling process. The lattice plane distances of 0.31 nm, as well as the 73 • cross angle of the lattice plane shown in Figure 4b, indicate that the lattice planes in region 2 are (310) and (−310) planes of the K 2 Ti 8 O 17 phase. In the sample without the charge-discharge cycling process, no such layered structure is observed (see Figure 1f). The amorphization in region 1 and the reorientation in region 2 are ascribed to the irreversible phase transition of K 2 Ti 8 O 17 during the charge-discharge process. The irreversible phase transition-induced amorphization of the surface layer of K 2 Ti 8 O 17 primarily causes the capacity to fade. Analogously, the Al 3+ ions intercalation and irreversible phase transition caused capacity degradation, which were also present in the transition-metal chalcogen for AIBs [12,27]. Nanomaterials 2021, 11, x FOR PEER REVIEW 9 o the lattice planes in region 2 are (310) and (−310) planes of the K2Ti8O17 phase. In the sa ple without the charge-discharge cycling process, no such layered structure is observ (see Figure 1f). The amorphization in region 1 and the reorientation in region 2 are cribed to the irreversible phase transition of K2Ti8O17 during the charge-discharge proce The irreversible phase transition-induced amorphization of the surface layer of K2Ti8 primarily causes the capacity to fade. Analogously, the Al 3+ ions intercalation and ir versible phase transition caused capacity degradation, which were also present in transition-metal chalcogen for AIBs [12,27].  For comparison, the amorphous layer is not present in the K 2 Ti 8 O 17 electrode after cycling in the hybrid AlCl 3 /NaAc electrolyte. Accordingly, the clear lattice fringe in the HRTEM image and the corresponding FFT plot (Figure 4g) indicate the relative high crystallinity in the sample, referring to the layered K 2 Ti 8 O 17 phase. Selected-area electron diffraction (SAED) was employed to probe the phase structure of the K 2 Ti 8 O 17 nanobelt before and after cycling. The three SAED images in Figure 4f-h show the pristine K 2 Ti 8 O 17 , as well as the K 2 Ti 8 O 17 after cycling in the AlCl 3 aqueous electrolyte and in the AlCl 3 /NaAc hybrid electrolyte, respectively. The pristine K 2 Ti 8 O 17 nanobelt is a single crystal with high crystallinity (Figure 4f). After charge/discharge cycling, defects are introduced in the K 2 Ti 8 O 17 , causing local structure distortions. As shown in Figure 4g,h, the diffraction patterns become blurred after cycling in both AlCl 3 and AlCl 3 /NaAc electrolytes, confirming the structure distortions. The distortion extent of the K 2 Ti 8 O 17 electrode in the AlCl 3 /NaAc electrolyte is lower than that in the AlCl 3 electrolyte. The trend of polycrystallization is also confirmed by the appearance of a diffraction ring. This could be attributed to the Na + /Al 3+ intercalation into the layered K 2 Ti 8 O 17 crystal structure during the charge-discharge process. In addition, the low-magnification SEM, energy-dispersive X-ray spectroscopy (EDS), and element mapping images of the pristine and reacted K 2 Ti 8 O 17 electrodes provide information on the homogenous distribution of Al into the K 2 Ti 8 O 17 phase ( Figure S6). These results suggest that the Al 3+ ions are successfully inserted into the layered K 2 Ti 8 O 17 structure after the charge/discharge behaviors. However, no amorphous layer is found in the reacted K 2 Ti 8 O 17 electrode in the hybrid AlCl 3 /NaAc electrolyte, implying the contributions of the additive NaAc on maintaining the crystallinity of the K 2 Ti 8 O 17 electrode during the electrochemical process. Simultaneously, a significantly improved cycling performance of the K 2 Ti 8 O 17 electrode in the hybrid AlCl 3 /NaAc electrolyte is obtained in comparison with that in the single AlCl 3 electrolyte (Figure 3d). It should be ascribed to the ability of alkaline NaAc to change the AlCl 3 aqueous electrolyte from neutral or acid conditions into alkaline conditions. As is known, the hydroxyl has a capacity to eliminate the formation of a Al 2 O 3 passivation layer under alkaline conditions, while the formed Al 2 O 3 passivation layer in neutral solutions hinders the electrode reactions, and a severe HER and self-discharge possibly occur under acidic conditions [12]. Therefore, we speculate that the NaAc additive eliminates the adverse effects of low pH values from the solo AlCl 3 aqueous electrolyte, thus improving the electrochemical characteristics of the K 2 Ti 8 O 17 electrode. Further evidence on the effect of the NaAc additive, however, is recommended in a study in the near future. As a comparison, the Na 2 Ti 9 O 19 sample was synthesized using the same experimental method with K 2 Ti 8 O 17 . SEM, XRD, and electrochemical performance tests for the Na 2 Ti 9 O 19 sample as an electrode in the AlCl 3 electrolyte were conducted, as shown in Figure S7 in the supporting information. Compared to the K 2 Ti 8 O 17 sample, the Na 2 Ti 9 O 19 sample exhibits worse electrochemical performance from the CV curves and rate-capacity curves.
To gain insight into the reversibility of the electrode, an ex situ X-ray absorption near-edge structure (XANES) technique was employed here to identify the structural evolution of the K 2 Ti 8 O 17 electrode materials during charge/discharge. Figure 5a displays the Ti L-edge XANES features undergoing the background subtraction and normalization procedures, for the pristine and tested K 2 Ti 8 O 17 under various charge/discharge states. Four observed peaks located at 458.4 eV (a'), 460.7 eV (b'), 463.9 eV (c'), and 462.1 eV (d') are ascribed to the excitation of Ti 2p 3/2 (peaks t 2g (L3) and e g (L3)) and Ti 2p 1/2 (peaks t 2g (L2) and e g (L2)) core levels into empty Ti 3d states, respectively [28][29][30]. As reported, the peak t 2g (L3) is related to the lower oxidation state of the cation, which has been detected on the TiO 2 nanosheet with respect to the reduced Ti 4+ surface specie [30]. The peak e g (L3) is related to the higher oxidation state of the cation, corresponding to the Ti 4+ species herein. The decreased intensity ratio of e g (L3)/t 2g (L3) peaks (Table S1) after the discharge of the electrode indicate the partial reduction of Ti 4+ during the discharge process. Moreover, differences between the pristine and the charged/discharged K 2 Ti 8 O 17 electrodes are observed in the O K-edge XANES spectra (Figure 5b), reflecting the effects of Na + /Al 3+ intercalation/deintercalation in the layered K 2 Ti 8 O 17 electrode in the charge/discharge process. According to molecular orbital theory, the O K-edge features originate from a transition of the O 1s electron to the various partially occupied and unoccupied molecular orbitals of the oxides, considering the crystal-field splitting effects [31]. As shown in the O K-edge XANES spectrum of the pristine K 2 Ti 8 O 17 sample, the major peaks located at about 532.6 eV and 534.1 eV are assigned to excitations varying from the O 1s core level to the Ti 3d-related conduction band, splitting into t 2g and e g subbands [32]. This result confirms the partial reduction of Ti 4+ during the discharge of the K 2 Ti 8 O 17 electrode. served in the O K-edge XANES spectra (Figure 5b), reflecting the effects of Na + /Al 3+ intercalation/deintercalation in the layered K2Ti8O17 electrode in the charge/discharge process. According to molecular orbital theory, the O K-edge features originate from a transition of the O 1s electron to the various partially occupied and unoccupied molecular orbitals of the oxides, considering the crystal-field splitting effects [31]. As shown in the O K-edge XANES spectrum of the pristine K2Ti8O17 sample, the major peaks located at about 532.6 eV and 534.1 eV are assigned to excitations varying from the O 1s core level to the Ti 3drelated conduction band, splitting into t2g and eg subbands [32]. This result confirms the partial reduction of Ti 4+ during the discharge of the K2Ti8O17 electrode. To further explore the potential practical application of the K2Ti8O17 electrode, a full cell of AIB was assembled using the carbon coated on the Ti mesh (AC-Ti), K2Ti8O17 electrode, and the aqueous AlCl3/NaAc hybrid solution as the cathode, anode, and electrolyte, respectively. The working potential range of the AC-Ti and K2Ti8O17 electrodes was measured by CV and GCD curves. The electrochemical properties of the AC-Ti electrode are given in Figure S8, demonstrating that AC-Ti is a suitable cathode for the aqueous AIB. Figure 6a presents the CV curves of both the cathode and anode. As shown, the AC-Ti electrode displays typical capacitance characteristics in the voltage range of 0~0.7 VVS. SCE, while the K2Ti8O17 electrode demonstrates the capacitance in the potential of −1~0 VVS. SCE in the hybrid AlCl3/NaAc electrolyte. The AC-Ti cathode and K2Ti8O17 anode were also evaluated by GCD tests (Figure 6b). The potential ranges for the K2Ti8O17 and AC-Ti electrodes are consistent with the CV measurements. The maximal voltage of the as-assembled AIB device reaches up to 1.6 V, as shown in Figure 6c. The CV curves without any obvious shape distortions at various sweeping rates, even at a high sweeping rate of 200 mV s −1 , indicate its superior rate capability. No gas evolution peaks are observed in the CV curves even at the potential range of 0~2.0 VVS. SCE (Figure 6d). The specific capacities of the K2Ti8O17// AlCl3/NaAc//AC-Ti battery are 189.6, 134.3, 97.3, and 72.4 mAh g −1 at high scan rates of 8, 12, 16, and 20 A g −1 , respectively (Figure 6d,f). Additionally, the AIB exhibits a capacity retention of approximately 66.8% after 500 cycles at the current density of 20 A g −1 (Figure 6g), demonstrating its superior cycling stability. Noticeably, a slightly larger discharge capacity is observed in the first 10 cycles and then decreases. Irreversible changes likely occurred in the electrode at the early stage [33,34]. Furthermore, the asassembled AIB has the power to light-up a commercial light-emitting diode (LED) lasting more than one hour (Figure 6h), confirming its promising practical applications for energy storage. To further explore the potential practical application of the K 2 Ti 8 O 17 electrode, a full cell of AIB was assembled using the carbon coated on the Ti mesh (AC-Ti), K 2 Ti 8 O 17 electrode, and the aqueous AlCl 3 /NaAc hybrid solution as the cathode, anode, and electrolyte, respectively. The working potential range of the AC-Ti and K 2 Ti 8 O 17 electrodes was measured by CV and GCD curves. The electrochemical properties of the AC-Ti electrode are given in Figure S8, demonstrating that AC-Ti is a suitable cathode for the aqueous AIB. Figure 6a presents the CV curves of both the cathode and anode. As shown, the AC-Ti electrode displays typical capacitance characteristics in the voltage range of 0~0.7 V VS. SCE , while the K 2 Ti 8 O 17 electrode demonstrates the capacitance in the potential of −1~0 V VS. SCE in the hybrid AlCl 3 /NaAc electrolyte. The AC-Ti cathode and K 2 Ti 8 O 17 anode were also evaluated by GCD tests (Figure 6b). The potential ranges for the K 2 Ti 8 O 17 and AC-Ti electrodes are consistent with the CV measurements. The maximal voltage of the as-assembled AIB device reaches up to 1.6 V, as shown in Figure 6c. The CV curves without any obvious shape distortions at various sweeping rates, even at a high sweeping rate of 200 mV s −1 , indicate its superior rate capability. No gas evolution peaks are observed in the CV curves even at the potential range of 0~2.0 V VS. SCE (Figure 6d). The specific capacities of the K 2 Ti 8 O 17 // AlCl 3 /NaAc//AC-Ti battery are 189.6, 134.3, 97.3, and 72.4 mAh g −1 at high scan rates of 8, 12, 16, and 20 A g −1 , respectively (Figure 6d,f). Additionally, the AIB exhibits a capacity retention of approximately 66.8% after 500 cycles at the current density of 20 A g −1 (Figure 6g), demonstrating its superior cycling stability. Noticeably, a slightly larger discharge capacity is observed in the first 10 cycles and then decreases. Irreversible changes likely occurred in the electrode at the early stage [33,34]. Furthermore, the as-assembled AIB has the power to light-up a commercial light-emitting diode (LED) lasting more than one hour (Figure 6h), confirming its promising practical applications for energy storage.

Conclusions
In conclusion, we developed a novel layered K2Ti8O17 anode material for an aqueous AIB via a facile hydrothermal process. The AIB was assembled with the as-prepared K2Ti8O17 electrode, AC-Ti, and AlCl3/NaAc hybrid aqueous solution as the anode, cathode, and electrolyte, respectively. The NaAc additive in the AlCl3 aqueous electrolyte played important roles in the enhanced electrochemical performance, improved rate performance, and optimized cycling stability. HRTEM and XANES techniques were conducted to gain deep insight into the Na + /Al 3+ ion intercalation/deintercalation mechanism during charge and discharge processes. ICP-OES results indicate the multi-ions insertion/deintercalation contributions to the charge/discharge process in this AIB device. The designed aqueous AIB with the layered K2Ti8O17 anode and hybrid AlCl3/NaAc electrolyte exhibited a superior rate capability. Particularly, an excellent discharge capacity for this AIB was achieved up to 189.6 mAh g −1 , even at a current density of 8 A g −1 . This work therefore provides promising prospects for developing high-capacity and super rate-capability aqueous AIBs.

Conclusions
In conclusion, we developed a novel layered K 2 Ti 8 O 17 anode material for an aqueous AIB via a facile hydrothermal process. The AIB was assembled with the as-prepared K 2 Ti 8 O 17 electrode, AC-Ti, and AlCl 3 /NaAc hybrid aqueous solution as the anode, cathode, and electrolyte, respectively. The NaAc additive in the AlCl 3 aqueous electrolyte played important roles in the enhanced electrochemical performance, improved rate performance, and optimized cycling stability. HRTEM and XANES techniques were conducted to gain deep insight into the Na + /Al 3+ ion intercalation/deintercalation mechanism during charge and discharge processes. ICP-OES results indicate the multi-ions insertion/deintercalation contributions to the charge/discharge process in this AIB device. The designed aqueous AIB with the layered K 2 Ti 8 O 17 anode and hybrid AlCl 3 /NaAc electrolyte exhibited a superior rate capability. Particularly, an excellent discharge capacity for this AIB was achieved up to 189.6 mAh g −1 , even at a current density of 8 A g −1 . This work therefore provides promising prospects for developing high-capacity and super rate-capability aqueous AIBs.