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

Abstract

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.

1. Introduction

Rechargeable lithium-ion batteries as promising energy storage devices have been widely applied in portable electronic devices and electric vehicles [1,2]. However, lithium-ion 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, aluminum-based 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 ($A{l}^{3+}+3e^{-}\leftrightarrow Al$), 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 Al³⁺/Al compared to that of H⁺/H₂ 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 TiO₂ anode and AlCl₃ electrolyte for the first time, evidencing the reversible intercalation/deintercalation of Al³⁺ ions into/from TiO₂ [15]. Subsequently, they used another AIB device using TiO₂ as the anode, while copper hexacyanoferrate and Al₂(SO₄)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, TiO₂ 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 H₂ evolution inhibitors for the AIB system [21].

Herein, we developed a novel aqueous AIB system consisting of a layered-structure K₂Ti₈O₁₇ anode obtained via a facile hydrothermal method. An AlCl₃ 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 plasma–optical 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⁺/Al³⁺ ion intercalation/deintercalation in aqueous AIB, inspiring a great potential to explore promising aqueous AIB anode materials.

2. Materials and Methods

2.1. Chemical Reagents

Titanium foil with a purity of 99% was purchased from Qingyuan Metal Materials Co., Ltd. (Xingtai, China) Potassium hydroxide (KOH), hydrochloric acid (HCl), hydrofluoric acid (HF), and concentrated nitric acid (HNO₃) were all purchased from Tianjin FengChuan Chemical Reagent Technology Corp. Ltd. (Tianjin, China) Aluminum chloride hexahydrate (AlCl₃·6H₂O) was purchased from Qinhuangdao Reagent Plant (Shanghai, China). Sodium acetate (NaAc) and activated carbon (AC) were purchased from Aladdin Corp., Ltd. (Shanghai, China) All these reagents were analytically pure (AR) and used without further purification.

2.2. Preparation of K₂Ti₈O₁₇ Electrode

The binder-free K₂Ti₈O₁₇ 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 cm². 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_xTi_yO₂ 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 K₂Ti₈O₁₇ electrode was approximately 1.0 mg cm⁻².

2.3. Preparation of AC Electrode

The AC electrode was prepared by coating AC slurry on a 2.0 × 2.0 cm² 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.

2.4. Physical Characterization of K₂Ti₈O₁₇ Electrode

The phase structure of the K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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.

2.5. Electrochemical Properties of the K₂Ti₈O₁₇ and AC Electrodes

The electrochemical properties of the K₂Ti₈O₁₇ 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 AlCl₃ solution and the mixed solution of 1 M of AlCl₃ and NaAc. The saturated calomel electrode (SCE, all the voltage values in this paper are herein referred to as SCE) and the 1 × 1 cm² Pt foil were used as the reference electrode and counter electrode, respectively. To further evaluate the practical application of the K₂Ti₈O₁₇ electrode in the aluminum ion electrolyte, an AIB battery was assembled using AC as the cathode electrode pasted onto the titanium mesh (K₂Ti₈O₁₇/Ti, 2 × 2 cm²) and K₂Ti₈O₁₇ as the anode electrode (1 × 2 cm²) in the hybrid electrolyte of 1 M of AlCl₃ and 3 M of NaAc (noted as AlCl₃/NaAc).

3. 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,g) indicate a single crystal structure of the prepared K₂Ti₈O₁₇ nanobelt. The growth direction of the nanobelt is along the [010] direction. The lattice plane distances of 0.19 nm and 0.27 nm correspond to (020) and ($\overline{4}0\overline{3}$) lattice planes of K₂Ti₈O₁₇ (Figure 1g), respectively. Even though a good contrast can be observed in the SAED pattern (Figure 1f), the FFT (fast Fourier transform) pattern of the HRTEM image inserted in Figure 1g reveals an elongation of the diffraction spot that may indicate the existence of defects in the single crystal.

The electrochemical properties of the as-prepared K₂Ti₈O₁₇ electrode were first investigated in 1 M of AlCl₃ 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 K₂Ti₈O₁₇ 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 Al³⁺ ions intercalation, indicating that the Al³⁺ ions intercalation is prior to the HER process. The ultra-high hydrogen evolution overpotential is ascribed to the low catalytic activity of both K₂Ti₈O₁₇ nanobelts and Ti substrate for HER [22]. Al³⁺ 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 Al³⁺ ions into/from the anatase TiO₂ electrode [15]. Similarly, it is possible that the Al³⁺ ions could insert/extract into/from the K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ electrode gradually increase with the scan rates. Under different sweep rates, the CV curves of the layered K₂Ti₈O₁₇ electrode exhibit similar quasi-symmetric shapes, illustrating the good reversibility of the Al³⁺ ions insertion/extraction. The electrochemical behavior of Al³⁺ ions in the K₂Ti₈O₁₇ electrode was evaluated by analyzing the collected CV data at various sweep rates according to the following equation:

$$i=a{v}^b$$

(1)

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 Al³⁺ ions in the K₂Ti₈O₁₇ electrode is a diffusion-controlled process. Figure 2d presents the cycling performance of the K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ electrode reach up to approximately 100% at the high discharge current density of 50 A g⁻¹ (Figure S1c). These results indicate the high-rate performance of the layered K₂Ti₈O₁₇ electrode, which should be ascribed to the layered structure for the fast diffusion of Al³⁺ ions. Additionally, the morphology evolution process of the precursors of the K₂Ti₈O₁₇ electrode with the hydrothermal reaction time was investigated via changing the hydrothermal time (0 h, 4 h, 8 h, 12 h, and 24 h). The SEM and XRD results are presented in Figure S2 in the supporting information. The compositional characterizations and electrochemical performance of the K₂Ti₈O₁₇ sample undergoing a 24 h hydrothermal process without the pyrolysis step are shown in Figure S3 in the supporting information. Compared with the K₂Ti₈O₁₇ 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 AlCl₃ electrolyte on the electrode, various concentrations of the NaAc was added into the AlCl₃ electrolyte as a buffer agent. The electrochemical performance of the K₂Ti₈O₁₇ electrode in the hybrid AlCl₃/NaAc aqueous electrolyte was evaluated by CV and GCD measurements (Figure 3 and Figure S4). The CV curves at different potential ranges show the symmetric shapes (Figure S4a), implying the good reversibility of the AIBs with the K₂Ti₈O₁₇ anode in the hybrid AlCl₃/NaAc aqueous electrolyte. Compared with the battery with pure AlCl₃ 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 Al³⁺ ions. The b value plots for log i against log v of the K₂Ti₈O₁₇ 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 AlCl₃/NaAc hybrid electrolyte (Figure 3b), suggesting a diffusion-controlled process in the K₂Ti₈O₁₇ electrode [11,24]. The cycling performance of the charge–discharge curves of the K₂Ti₈O₁₇ electrode in the aqueous AlCl₃/NaAc electrolyte shows a similar trend to that in the AlCl₃ electrolyte (Figure 3c and Figure S5a), exhibiting a decreasing trend with the potential ranges. In contrast, the cycling stability in the AlCl₃/NaAc hybrid electrolyte is improved compared with that in the single AlCl₃ electrolyte. To analyze the elemental composition of the cycled electrode, ICP-OES experiments for the cycled K₂Ti₈O₁₇ electrodes in both charge and discharge states, as well the tested electrolyte, were performed. The cycled K₂Ti₈O₁₇ 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 Al³⁺ ion and a trace of the Na⁺ ion are detected. Therefore, we can conclude that the ionic substitutions of K⁺ ions by Al³⁺/Na⁺ ions occur in the K₂Ti₈O₁₇ 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 Al³⁺ and Na⁺ ions. The ICP-OES measurement for the cycled K₂Ti₈O₁₇ electrode after washing shows the existence of Na⁺ and Al³⁺ ions. This indicates a possible mechanism of Na⁺ and Al³⁺ co-intercalation/deintercalation during the charge/discharge process. Although the specific capacitance of the K₂Ti₈O₁₇ electrode in the aqueous AlCl₃/NaAc electrolyte (159 mAh g⁻¹) is lower than that in the AlCl₃ 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 K₂Ti₈O₁₇ electrode in the AlCl₃/NaAc hybrid electrolyte (Figure 3e). The K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ electrode in the AlCl₃/NaAc electrolyte. It is interesting that the cycling performance and CEs of the K₂Ti₈O₁₇ electrode are improved with the increase in the charge–discharge current density (Figure S5b,c). The K₂Ti₈O₁₇ 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 well-defined 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⁺/Al³⁺ 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 the lattice planes in region 2 are (310) and (−310) planes of the K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ during the charge–discharge process. The irreversible phase transition-induced amorphization of the surface layer of K₂Ti₈O₁₇ primarily causes the capacity to fade. Analogously, the Al³⁺ ions intercalation and irreversible phase transition caused capacity degradation, which were also present in the transition-metal chalcogen for AIBs [12,27].

For comparison, the amorphous layer is not present in the K₂Ti₈O₁₇ electrode after cycling in the hybrid AlCl₃/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 K₂Ti₈O₁₇ phase. Selected-area electron diffraction (SAED) was employed to probe the phase structure of the K₂Ti₈O₁₇ nanobelt before and after cycling. The three SAED images in Figure 4f–h show the pristine K₂Ti₈O₁₇, as well as the K₂Ti₈O₁₇ after cycling in the AlCl₃ aqueous electrolyte and in the AlCl₃/NaAc hybrid electrolyte, respectively. The pristine K₂Ti₈O₁₇ nanobelt is a single crystal with high crystallinity (Figure 4f). After charge/discharge cycling, defects are introduced in the K₂Ti₈O₁₇, causing local structure distortions. As shown in Figure 4g,h, the diffraction patterns become blurred after cycling in both AlCl₃ and AlCl₃/NaAc electrolytes, confirming the structure distortions. The distortion extent of the K₂Ti₈O₁₇ electrode in the AlCl₃/NaAc electrolyte is lower than that in the AlCl₃ electrolyte. The trend of polycrystallization is also confirmed by the appearance of a diffraction ring. This could be attributed to the Na⁺/Al³⁺ intercalation into the layered K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ electrodes provide information on the homogenous distribution of Al into the K₂Ti₈O₁₇ phase (Figure S6). These results suggest that the Al³⁺ ions are successfully inserted into the layered K₂Ti₈O₁₇ structure after the charge/discharge behaviors. However, no amorphous layer is found in the reacted K₂Ti₈O₁₇ electrode in the hybrid AlCl₃/NaAc electrolyte, implying the contributions of the additive NaAc on maintaining the crystallinity of the K₂Ti₈O₁₇ electrode during the electrochemical process. Simultaneously, a significantly improved cycling performance of the K₂Ti₈O₁₇ electrode in the hybrid AlCl₃/NaAc electrolyte is obtained in comparison with that in the single AlCl₃ electrolyte (Figure 3d). It should be ascribed to the ability of alkaline NaAc to change the AlCl₃ aqueous electrolyte from neutral or acid conditions into alkaline conditions. As is known, the hydroxyl has a capacity to eliminate the formation of a Al₂O₃ passivation layer under alkaline conditions, while the formed Al₂O₃ 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 AlCl₃ aqueous electrolyte, thus improving the electrochemical characteristics of the K₂Ti₈O₁₇ electrode. Further evidence on the effect of the NaAc additive, however, is recommended in a study in the near future. As a comparison, the Na₂Ti₉O₁₉ sample was synthesized using the same experimental method with K₂Ti₈O₁₇. SEM, XRD, and electrochemical performance tests for the Na₂Ti₉O₁₉ sample as an electrode in the AlCl₃ electrolyte were conducted, as shown in Figure S7 in the supporting information. Compared to the K₂Ti₈O₁₇ sample, the Na₂Ti₉O₁₉ 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 K₂Ti₈O₁₇ 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 K₂Ti₈O₁₇ 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 TiO₂ nanosheet with respect to the reduced Ti⁴⁺ surface specie [30]. The peak e_g (L3) is related to the higher oxidation state of the cation, corresponding to the Ti⁴⁺ 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 Ti⁴⁺ during the discharge process. Moreover, differences between the pristine and the charged/discharged K₂Ti₈O₁₇ electrodes are observed in the O K-edge XANES spectra (Figure 5b), reflecting the effects of Na⁺/Al³⁺ 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 3d-related conduction band, splitting into t_2g and e_g subbands [32]. This result confirms the partial reduction of Ti⁴⁺ 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 V_{VS. SCE}, while the K₂Ti₈O₁₇ electrode demonstrates the capacitance in the potential of −1~0 V_{VS. 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 V_{VS. 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 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.

4. 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⁺/Al³⁺ 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.