Dual-Ion Stabilized Layered Structure of OVO for Zero-Strain Potassium Insertion and Extraction.

Potassium-ion batteries (KIB) have similar energy storage mechanism with lithium-ion battery, but the potassium (K) resource is rich, which shows great potential for large-scale energy storage system. Recently, the anode materials of KIB studied mainly include carbon materials, transition metal oxides, and alloy materials. The amorphous hard carbon shows the best comprehensive performance, but its intercalation potential is close to 0 V (versus K+ /K), which is easy to cause K dendrite and brings security risks. The oxide materials have high capacity but high intercalation potential, low first cycle efficiency, and unstable cycle. Here, based on the understanding of the K intercalation mechanism of vanadium oxides, a novel zero strain anode material with layered structure of dual-ions (Na+ /K+ ) is designed (NaK(VO3 )2 V2 O5 ). The introduction of Na/K ion contributed to the transmission and further stabilized the structure. It has an excellent rate performance (10 A g-1 , up to 25 000th cycle), and its special K storage mechanism and zero-strain characteristics are revealed for the first time by ex situ scanning electron microscope, X-ray powder diffraction, X-ray photoelectron spectroscopy, and other test methods. Considering the excellent performance endowed by these unique inherent properties, NaK(VO3 )2 V2 O5 shows great potential for commercial anode materials and may promote the innovation of KIB.

Introduction

In response to the energy crisis and environmental pollution the renewable energy sources, such as wind, solar and hydrogen energy have attracted much attention as supplement to the fossil fuels[ ] While, alkali‐ion battery and alkali‐metal battery are by far the best developed and commercially most viable option for the traditional energy storage system for their high energy density and no memory effect.[ ] The potassium (K) ion battery (KIB) is also a promising technic, due to the low cost and abundance in nature.[ ] However, K has a larger radius and reach a lower electrochemical kinetics, thus, the experience in lithium/sodium‐ion battery could not directly be applied in the KIB.[ ] To implement highly efficient intercalation of K+, the best way is to use specifically designed zero‐strain material to prepare long‐life KIB.

Typical layered materials have strong in‐plane chemical bonds and weak van der Waals interactions between adjacent layers, and large interlayer space and 2D channel width result in high ionic conductivity and strong charge transfer ability, which are often used for metal ion storage.[ ] Compared with K+ conversion and alloy electrode materials, layered materials provide a wide 2D channel for reversible intercalation/de‐intercalation of cations, resulting in a relatively better cyclic stability of the lattice of layered materials without serious deformation.[ ]

The orthorhombic vanadium pentoxide (α‐V2O5) is a typical layered cathode material for lithium‐ion battery, and the embedded cations are usually stored in the layers.[ ] Because of its high basal spacing, vanadium pentoxide could be applied as the cathode material in other alkali‐ion battery in theory.[ ] However, the intercalation and de‐intercalation of K+ between α‐V2O5 layers are difficult, due to the limitation of the unevenness and compact layer space.[ ] The insertion of metal ions has a positive effect on stabilizing the layered structure of V2O5.[ ] However, it may change the structure of V2O5 and reduce the specific capacity of V2O5‐based cathode. 2D structures with large active surface area can increase the contact area between electrode and electrolyte, providing a large number of active sites, and promoting ion transport, which may be an effective way to solve the problem.[ ] Making 2D V2O5 on a large scale remains a huge challenge. In addition, heating and annealing V2CT (MXene) produces a layered array of V2O5. Compared with the original V2O5, the original lamellar structure is retained in the crystal structure, and the interlamellar space is increased in the microscale, theoretically providing a good storage site for K ions through 2D ion channels.[ ] It suggests that it is more cost‐effective to do the former (Pre‐intercalation) than the latter (in Nano structure).[ , ] Pre‐intercalation can improve the inherent energy storage properties of layered materials, including increasing the interlayer distance, stabilizing the crystal phase, and enhancing the electrical conductivity.[ ] However, pre‐intercalation will sacrifice the active site and inevitably lose capacity. Therefore, optimizing pre‐insertion can provide as many active sites as possible while maintaining the layered structure.

Herein, a heterojunction layered material of NaK(VO3)2‐V2O5 was introduced for anode in KIB. Which was synthesized by one‐step solid‐phase mixing NaK alloy with V2O5. NaK(VO3)2 (The V valance of NaK(VO3)2‐V2O5 is V5+.) material has a uniform large lamellar spacing structure (3.22 A) and better [−2 2 1] crystal structure. Through support of O—V—O with dual‐ion (Na+/K+) in between and could let the rapid storage and diffusion of K+ realized without the modification of other conductive substrates. In terms of crystalline structure, even if de‐intercalation of K+ between the layers, NaK(VO3)2 crystal remains stable by Na+. In the application of anode electrode of KIB, NaK(VO3)2‐V2O5 shows high specific capacity and ultra‐high rate performance. In addition, it is found that the material has no phase change during charge and discharge from the X‐ray powder diffraction (XRD) test, which proved that zero strain anode was obtained. The mechanism of K storage enhancement was discussed by electrochemical kinetic analysis. This work could provide a general method for making zero‐strain anode materials from vanadium‐based materials for the next generation of battery systems.

Results and Discussion

Layered V2O5 has a unique layered crystalline structure with a spacing of ≈0.437 nm, which is considered an active material. However, K+ insertion into V2O5 is problematic because the interlaminar space for ion diffusion cannot be fully occupied by K+ supporting the layered structure (Figure  ). In order to create favorable K+ storage sites within the V2O5 framework, the structural layers are either enlarged from the material or exchanged with other smaller cations to leave sufficient vacancies.[ , ] During the repeated insertion process of K+, weak bond is difficult to withstand severe structural stress. To solve the problem, the insertion of non‐active ions has a positive effect on stabilizing the layered structure of V2O5 in theory. NaK(VO3)2 has dual‐ion, which could induce a strong bond between two adjacent O—V—O layers to form a stable structure. In addition, the more stable [−221] orientation of NaK(VO3)2 has a larger and uniform tunnel (Figure 1a), theoretically allowing rapid and stable storage/diffusion of K ions. High specific capacity needs both layered structure and large interlayer spacing. Therefore, keeping the special structure of NaK(VO3)2‐V2O5 could provide a promising technique for the further study on KIB.

Figure 1

a) The formation mechanism of NaK(VO3)2 via the dual‐ion (Na+/K+) intercalation process in V2O5. b) HRTEM image of V2O5 and NaK(VO3)2. c) XRD patterns of NaK(VO3)2. e–g) O 1s, V 2p, K2p, and C 1s XPS spectra of the as‐prepared NaK(VO3)2.

Morphology and phase structure of the prepared NaK(VO3)2‐V2O5 were investigated by scanning electron microscope (SEM) and transmission electron microscope (TEM) and XRD. SEM images show that the NaK(VO3)2‐V2O5 structure consists of many coarse‐faced nanosheets and is slightly different for the V2O5 (Figure S1, Supporting Information). At the same time, the “pre‐intercalation effect” has no significant effect on the microtopography of V2O5. From TEM images, it can be found that the two components of the heterojunction are face to face (Figure 1b,c). While, Figure 1c showed that the atomic grain‐boundary structure had dislocations and had very similar arrangement, indicating that the grain‐boundary was formed by the direct intercalation of Na+/K+. The crystal structure was further confirmed by XRD results (Figure 1d). The crystallinity of NaK(VO3)2‐V2O5 has two Vanadium‐Based Oxides (NaK(VO3)2 : PDF#26‐1474 and V2O5: PDF#41‐1426) and is obviously lower than that in the literature,[ ] which is related to the increased disorder of crystal structure caused by some vacancies, it can further accelerate the rapid transfer of electrons in the charging and discharging process due to the V2O5 provide more space. X‐ray photoelectron spectroscopy (XPS) results confirmed that the electron cloud bias from electron‐rich —V═O‐ and —V—O—V‐ component to low binding energy field inducing the formation of built‐in electric field at the heterojunction interface (Figure 1e,f), which may accelerate the charge transfer dynamics at the heterogeneous interface. One peak of O1s spectra of V2O5 in XPS belong to V═O and other is O—V—O.[ ] It can be seen from Figure 1g that the sample contains K elements, among which the binding energy peak of C comes from the C element used for calibration during XPS test.[ ] A direct proof of the K+ is intercalation in the inner structure and the reaction mechanism of NaK(VO3)2 has been first studied by XPS analysis.

The electrochemical performance of NaK(VO3)2‐V2O5 for KIB was further investigated. The aluminum foil is cheaper and lighter than copper foil,[ ] which can not only significantly reduce the price of the KIB, but also reduce the weight of the collector and solve the overdischarge problem. Such technology has been applied first in KIB and the battery model now looks like Figure  . Figure 2b shows that NaK(VO3)2‐V2O5 exhibits excellent K storage properties compare to V2O5, for example, the KIB based on NaK(VO3)2‐V2O5 maintained a high capacity of 208 mAh g−1 at 0.1 A g−1 while the V2O5 had a low capacity of 105 mAh g−1 after 300 cycles. It is worth noting that drastic fluctuation of the specific capacity and coulombic efficiency is due to repeated activation of NaK(VO3)2‐V2O5 anode. Although specific capacity and coulombic efficiency move up and down violently, specific capacity has been more reliable, typically falling from the peak of a cycle to its trough by only 18 mAh g−1. Moreover, the electrochemical properties of the two anode materials were studied by charge‐discharge curves at different cycles (Figure 2c). KIB of V2O5 of irreversible capacity loss occurred during the first discharge and the curve of NaK(VO3)2‐V2O5 is relatively flat. It is important that the inclined region is attributed to surface‐driven K+ storage behavior, and the quasi‐elevation region is attributed to K+ intercalation behavior,[ ] in line with expectations. In addition, the initial discharge capacity of the electrode is close to 1800 mAh g−1 with a small initial coulombic efficiency due to high electrolyte (3M KFSI) content and more decomposition in anode for stable SEI in Figure 2c.[ ] To find out the voltage value of the redox‐active materials, the cyclic voltammetry (CV) test was performed. The results of the first three cycles of NaK(VO3)2‐V2O5 and V2O5 at 0.1 mV s−1 are shown in Figure 2d. The peak value at 0.75 V in the first lap corresponds to SEI film production.[ ] The peaks at 0.85 V correspond to the intercalation of K+ in the NaK(VO3)2‐V2O5, and peaks at 0.51 and 0.92 V are de‐intercalation of K+, which was in agreement with the charge–discharge curve.

Figure 2

K‐storage behavior of the NaK(VO3)2‐V2O5 and V2O5. a) Schematic illustration of battery model of KIB with Al foil as current collector. b) Cycling stability and Coulombic efficiencies of NaK(VO3)2‐V2O5 and V2O5 work at a current density of 0.1 A g−1. c) Galvanostatic charge and discharge curves of NaK(VO3)2‐V2O5 and V2O5 cycled at current density of 0.1 A g−1. d) CV of NaK(VO3)2‐V2O5 measured at 0.1 mV s−1. e) Cycling stability and Coulombic efficiencies of NaK(VO3)2‐V2O5 and V2O5 work at different current densities of 0.1, 0.2, 0.5, 1, 2, 5, 10, and 20 A g−1. f) Galvanostatic charge and discharge curves of NaK(VO3)2‐V2O5 and V2O5 cycled at different current densities of 0.2 and 0.5 A g−1.

Potassiation:

De‐potassiation:

(n can be calculated in Supporting Information)

In addition, the inset in Figure 2c is the KIB of V2O5, which shows a high irreversible capacity loss. Furthermore, the rate tests indicated that the NaK(VO3)2‐V2O5 showed a good activation property much better cycling stability, and a receivable discharge capacity at high‐rate (Figure 2e). At 0.2–20 A g−1 current density, the capacities of NaK(VO3)2‐V2O5 are 264, 141, 264, 115, 96, 53, 47, and 37 mAh g−1, respectively, which are much higher than those of V2O5 without pre‐intercalation of Na+/K+ (75, 46, 35, 20, and 13 mAh g−1). It is important that dramatically increase during the initial cycles is due to KIB systems might become unstable during the high current density change.[ ] When back to 0.2 A g−1, the NaK(VO3)2‐V2O5‐based KIB maintained a capacity of 241 mAh g−1. The results suggested that through lattice regulation, better rate performance and more stable cycle performance were obtained of the KIB. The effect of dual‐ion intercalation in KIB has been investigated by the different rate on the charge and discharge processes. In Figure 2f, the NaK(VO3)2‐V2O5 has less capacity loss than V2O5, resulting in better efficiency. Thorough structural design of dual‐ion pre‐intercalation in the V2O5, obtained a rigid crystalline framework, and the zero‐strain anode with excellent performance was realized.

The battery was disassembled after 50 cycles, and the different status of XRD of NaK(VO3)2‐V2O5 and V2O5 for KIB are shown in Figure  . The characteristic peak of NaK(VO3)2‐V2O5 is completely unchanged during the discharge and charge process, indicating that the crystal structure of NaK(VO3)2‐V2O5 has a rare zero strain characteristic. On the contrary, analyses were conducted on the raw data (Figure S2, Supporting Information) and XRD patterns of V2O5 show that the discharge and charge products are amorphous. These results show that the crystal structure of NaK(VO3)2‐V2O5 has high stability and zero strain characteristics in the process of intercalation/de‐intercalation of K+, which is the fundamental internal reason for its super stable K storage performance. The XPS analysis of NaK(VO3)2‐V2O5 shows that SEI formed oxygen ions in O 2p is higher and rich in inorganic component, while O═V in material is thinner (Figure 3b,c), which demonstrates that K+ could easily intercalate and de‐intercalation into/from NaK(VO3)2‐V2O5. The difference in valances state of O and V in Figure 3b,c demonstrate that K+ embedded in V2O5 layered structure leads to structural damage, accompanied by the formation of the second phase, which will lead to capacity fading. Due to the stability of NaK(VO3)2‐V2O5 larger structure and no phase transition occur during the reversible deintercalation of K+, which will deliver excellent cyclic stability. Interestingly, the XPS results of different state of cycled anode of V2O5 also demonstrate that the curves have similar characters with raw material of NaK(VO3)2‐V2O5, so this is an irreversible process. Importantly, after analyzing the morphology changes before and after cycling (SEM, Figure S3, Supporting Information), it can be clearly observed that the particle size becomes larger and the fiber becomes coarser after the cycle, which is attributed to the volume expansion and SEI formation during the cycle of KIB. Although the morphology of NaK(VO3)2‐V2O5 is slightly changed, it is obvious that there is no damage or pulverization of the material after charge and discharge, which proves that the material has excellent structural stability. The structures and properties of NaK(VO3)2‐V2O5 are analyzed and characterized by means of XRD, XPS, and SEM, the results indicating that the NaK(VO3)2‐V2O5 is zero‐strain structure.

Figure 3

Structural evolution of NaK(VO3)2‐V2O5 and V2O5 of K ion after insertion/extraction. a) XRD results of the NaK(VO3)2‐V2O5 and V2O5 be potassiation/depotassiation after cycle. b,c) XPS results of two products of discharge and charge state.

The relationship between the electrochemical property and the crystalline structure was further investigated by the CV test at different scan rates (0.2–1 mV s−1), Figure  shows that the storage process of KIB of NaK(VO3)2‐V2O5 is mainly controlled by capacitive and pseudocapacitance behavior (b = 0.91, largely controlled by pseudocapacitance).[ ] In addition, the results of CV test show that the peaks of NaK(VO3)2‐V2O5 at 0.51 to 0.53 V are significantly higher than those of V2O5, indicating that the heterogeneous interface promotes the intercalation and de‐ intercalation process of K+ and makes K+ more inclined stabilize the layered structure of O—V—O, which is related to the synergistic effect of controllable grain boundary of heterojunction. Figure 4d is Log (current peak)‐v 0.5 curves. This result can reveal the electrochemical kinetics of the two materials (the detail in Supporting Information). The diffusion coefficient of NaK(VO3)2‐V2O5 is 5.9 × 10−13 cm2 s−1 which is much higher than that of V2O5 (2.8 × 10−13 cm2 s−1).[ ] This further revealed that the dual‐ion pre‐intercalation with heterojunction interface can effectively reduce the charge transfer resistance and accelerate the migration of K+, thus obtaining superior rate performance. At 5 A g−1, the capacity of KIB based on NaK(VO3)2‐V2O5 remained 89 mAh g−1 after 2500 cycles. The charging and discharging curve is close to 100% from 500th to 2000th (Figure 4e), even when the rate was up to 10 A g−1 and the discharge capacity had no obvious change during the cycle, showing good reversibility and stability (up to 25 000, Figure 4f). Compared with the reported anode materials of vanadium oxide and its derivatives for KIB, the NaK(VO3)2‐V2O5 is the best KIB material with the best rate and cycling performance so far.

Figure 4

Kinetic behaviors of NaK(VO3)2‐V2O5 and V2O5: a) CV curves at multiple scan rates of NaK(VO3)2‐V2O5. b) Linear b values of the CV peak currents. c) CV curves at multiple scan rates of V2O5. d) The linear fits of the CV peak currents for KIB of NaK(V2O5)3‐V2O5 and V2O5. e,f) Long‐term cyclability of NaK(VO3)2‐V2O5 electrode at 5 and 10 A g−1.

It is worth noting that the electrode is cycled under large current densities of 5 and 10 A g−1, and it is observed the specific capacity in the initial stage is very low and then it increases after several cycles. Due to an activation process, which could release the potential of NaK(VO3)2‐V2O5 to a greater extent with the large volume of K+, they all have an activation process in the rapid insertion/extraction. After the capacity increases to the maximum value and the attenuation decreases slowly with an acceptable specific capacity, indicating that the cycle stability is good. As for low current density (0.1 A g−1) at the beginning and the electrochemical reaction with NaK(VO3)2‐V2O5 with K+ insertion/extraction is sufficient, the high specific capacity can maintain in the initial.

Furthermore, Figure S4, Supporting Information, shows that a highly concentrated salt electrolyte of potassium difluorosulfonyl imide (KFSI, 4 M) can achieve a more stable (0.1 A g−1) and long cycle (1 A g−1, 2000 cycles) of NaK(VO3)2‐V2O5. However, high‐concentration salt electrolytes are costly and KFSI‐based electrolytes corrode aluminum foil (current collector),[ ] compared to conventional low‐concentration electrolyte based on 3 M KFSI, which generally have the advantages of better chemical stability and lower cost. Besides, with a carbonate base of 1 M potassium hexafluoroarsenate (KPF6) in ethylene carbonate/diethyl carbonate (EC/DEC) and their performance tends to degrade significantly with cycles (Figure S5, Supporting Information), which demonstrates that electrolyte is not fit for NaK(VO3)2‐V2O5 in KIB. Furthermore, the XPS indicates that the 3 M KFSI in DME is rather stable in the cycle and similar to 3 M KFSI in DME. It is pointed out that F 1s spectrum showed two different fluorine species, S‐F (687 eV) and KF (684 eV), respectively,[ ] KF in KPF6 with low binding energy like cycled of V2O5 (Figure S6, Supporting Information). The experimental result shows that 3 M KFSI in DME electrolyte shows more economical and effective in KIB. In ether electrolyte, NaK(VO3)2‐V2O5 exhibits a record ultra‐fast and ultra‐stable K storage performance. Combined with XRD and XPS analysis, it is revealed that the crystal structure of NaK(VO3)2‐V2O5 has rare zero strain characteristics, and the thin and tough SEI formed in ether electrolyte is beneficial to the K storage, is conducive to better cycle and rate performance.

Conclusion

The crystals engineering, the diffusion kinetics of K+ is enhanced by the Pre‐intercalation of dual‐ions (Na+/K+) at the heterojunction interface of NaK(VO3)2‐V2O5, which is beneficial to the improvement of electrochemical performance of KIB. The crystalline engineering modification, abundant active sites are provided for the heterostructures with homogeneous and stable layered structure, ensuring a higher contribution of pseudo‐capacitance, fast rates of diffusion of K+, and accelerating the reaction kinetics. In addition, the regulation of morphology engineering, the interaction between vanadium oxide is strengthened by inorganic crystals modulated, and the overall structure of the electrode material is kept stable to the zero‐strain of K+ repeated insertion/extraction. NaK(VO3)2‐V2O5 has excellent performance due to its unique intrinsic properties, which is expected to become a commercial anode material and promote the innovation of KIB.

Conflict of Interest

The authors declare no conflict of interest.

Supporting Information

Click here for additional data file.