High-Performance Potassium-Ion-Based Full Battery Enabled by an Ionic-Drill Strategy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Sep 2021High-Performance Potassium-Ion-Based Full Battery Enabled by an Ionic-Drill Strategy Chunlei Jiang†, Xing Meng†, Yongping Zheng†, Jiaxiao Yan, Zhiming Zhou and Yongbing Tang Chunlei Jiang† Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 School of Materials Science and Engineering, Chongqing University of Technology, Chongqing 400054 , Xing Meng† Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 School of Materials Science and Engineering, Chongqing University of Technology, Chongqing 400054 , Yongping Zheng† Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 , Jiaxiao Yan Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 Nano Science and Technology Institute, University of Science and Technology of China, Suzhou 215123 , Zhiming Zhou Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 School of Materials Science and Engineering, Chongqing University of Technology, Chongqing 400054 and Yongbing Tang *Corresponding author: E-mail Address: [email protected] Functional Thin Films Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055 School of Materials Science and Engineering, Chongqing University of Technology, Chongqing 400054 Key Laboratory of Advanced Materials Processing & Mold, Ministry of Education, Zhengzhou University, Zhengzhou 450002 https://doi.org/10.31635/ccschem.021.202000463 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Potassium-ion (K+) batteries (PIBs) show great potential in large-scale energy storage applications owing to their low-cost and high-operating voltage. However, the development of practical PIBs remains hindered by their poor performance rate due to the large ionic radius of the K+ (1.38 Å). Herein, we report an ionic-drill strategy using smaller lithium ions (Li+) to open diffusion channels for K+ in the host electrode and enable much faster kinetics of K-ions. To realize this goal, a K+/Li+/PF6− hybrid configuration was rationally designed, and the K-based electrolyte hybridized with 20 atom % Li showed significantly improved diffusion kinetics and discharge capacity. The optimized K+-based full battery exhibited excellent rate performance with a specific capacity of 106 mAh g−1 at a current rate of 5 C and maintained 84.7 mAh g−1 up to 25 C. It also showed long-term cycling stability with capacity retention ∼100% over 500 cycles at 5 C. These results are the best known K+ full batteries. Thus, this work provides an effective strategy to enhance the electrochemical performance of PIBs for potential practical applications. Download figure Download PowerPoint Introduction High-performance and low-cost energy storage systems are urgently needed to utilize intermittent power sources such as solar and wind.1–9 Although lithium-ion (Li+) batteries (LIBs) have been applied successfully in portable electronic devices and electric vehicles, their application in grid-scale energy storage systems is still hindered by the high cost and limited resources of lithium.10–14 Alternatively, energy storage devices based on earth-abundant elements such as sodium,15–24 potassium,25–31 calcium,32,33 and magnesium,34–38 are attracting increasing attention. In particular, potassium-ion (K+) batteries (PIBs) are more promising due to the following merits: (1) the relatively low standard reduction potential of potassium (only 110 mV higher than lithium), which is beneficial to achieve higher energy density39; (2) the abundance of potassium (2.09 wt %) in the earth's crust (>1200 times higher than that of lithium) that offers a remarkable cost advantage.40 However, the electrochemical performance of PIBs is still far from the practical requirements due to the sluggish kinetics of K-ions with relatively larger ionic radius (1.38 Å).41–43 Here, we developed an ionic-drill strategy for PIBs by hybridizing a small part of Li-ions with fast kinetics, which functions as ionic drills to open diffusion channels for K-ions in the host electrode. To realize this aim, a hybrid K+/Li+/PF6− electrolyte was rationally designed with Sn and expanded graphite (EG) as the anode and cathode, respectively. The as-optimized full battery exhibited significantly enhanced rate performance up to 25 C with the capacity maintained as high as 84.7 mAh g−1. Moreover, long cycling stability over 500 cycles at 5 C was achieved with the capacity retention close to 100%. These results demonstrated that the highly effective ionic-drill strategy improved the reaction kinetics of PIBs and can also be applied to other energy storage devices suffering from sluggish kinetics. Experimental Methods Materials and synthesis Sn foil was used as the anode and punched into circular sheets with a diameter of 12 mm. Glass fiber separator was punched into circular sheets with a diameter of 16 mm. After that, the fabricated Sn foil and glass fiber sheets were vacuum dried at 80 °C for 24 h. The cathode was prepared by mixing EG (80 wt %), conductive carbon black (10 wt %), polyvinylidene fluoride (PVDF) (10 wt %), and N-methyl-2-pyrrolidone (NMP) into a uniform slurry, and then the slurry was coated onto the carbon-coated Al foil. Subsequently, the as-prepared positive electrodes were vacuum dried at 80 °C for 24 h. Thereafter, they were punched into circular sheets with a diameter of 10 mm and weighed with an electronic analytical balance (Sartorius-BT25S, resolution of 0.01 mg). Hybrid electrolytes (1 M) with different Li/K ratios in ethylene carbonate (EC) + ethyl methyl carbonate (EMC) + dimethyl carbonate (DMC) + propylene carbonate (PC) (2∶2∶1∶1 v/v/v/v) were prepared by swirling the indicated proportions of dissolved lithium salt LiPF6 and potassium salt KPF6. To optimize the electrolyte, the concentration of lithium in the electrolyte was set as 0, 10, 20, 30, 40, 50, and 100 atom %, respectively. CR2032 coin-type cells were assembled with the electrolyte amount controlled ∼0.2 mL cell−1. The fabrication of the electrolyte and cells was conducted in a glovebox (Etelux Lab2000) with water and oxygen content <0.1 ppm. Characterization and electrochemical measurements Scanning electron microscopy (SEM; ZEISS Supra 55-VP) equipped with energy-dispersive X-ray spectroscopy (EDX; OXFORD MAX 20) was carried out to determine the morphology and elemental composition of the electrode materials. Diffractometry (using the Rigaku Miniflex 600 diffractometer equipped with Cu Kα radiation operated at 40 kV and 15 mA) was performed for an ex situ X-ray diffraction (XRD) analysis. The step-length and scanning rate was set at 0.02° and 4° min−1, respectively. Before the above analysis, the electrode samples were washed with DMC. Linear sweep voltammetry (LSV) measurements were conducted to investigate the electrochemical stability of the hybrid electrolyte using the Autolab electrochemical workstation (Autolab PGSTAT302N). Stainless steel plates were used as both working and counter electrodes in coin-type cells. Hybrid electrolyte with different Li contents (0, 10, 20, 30, 40, 50, and 100 atom %) was tested in the potential range of 3.0–5.0 V. Polarization tests were carried out for symmetric Sn|Sn paired cells (CR2032 coin-type cells) using the battery test system (NEWARE CT-4008) at room temperature. Symmetric Sn|Sn cells with different Li contents (0, 10, 20, 30, 40, 50, and 100 atom %) in the hybrid electrolyte were tested at varied current densities (0.2, 0.5, and 1.0 mA cm−2). Also, galvanostatic charge/discharge measurements of the K+-based full batteries were performed on the NEWARE battery testing system (NEWARE CT-4008) at room temperature. The capacities of the K+-based full batteries were calculated based on the mass of the cathode material. Electrochemical impedance spectroscopy (EIS) was conducted with CR2032 coin-type cells using the Autolab electrochemical workstation (Autolab PGSTAT302N). The frequency range was set from 100 kHz to 10 MHz with an alternating current (AC) amplitude of 10 mV. Density functional theory calculation In this work, the geometries and total energies were calculated using the density functional theory (DFT) method implemented in the Vienna ab initio simulation package (VASP).44 The projector-augmented wave (PAW) method was adopted.45 The generalized gradient approximation (GGA) with Perdew–Burke–Ernzerhof (PBE) function was used to describe the exchange-correlation interactions.46 The cut-off energy was 450 eV, and the criteria of convergence for electronic energy and ionic force were 10−5 eV and 10−2 eV Å−1, respectively. For solvated Li- and K-ions in various solvents, the leading errors from neutralizing background charge were corrected by the following expression47: e 2 q 2 α L ɛ (1) where q is the net charge of the supercell, α is the Madelung constant of a point charge q in a homogeneous background charge −q, L is lattice constant, and ɛ is the dielectric constant. The solvation energy was calculated using the following formula: Δ E = E ( A + n B ) − E ( A + ) − n E ( B )(2)where E(A+nB), E(A+), and E(B) denote the total energy calculated from DFT of the solvated ion, naked ion, and solvent molecule, respectively, and n is the solvation number. The lithium and potassium migration barriers in the anode were calculated using a climbing image nudged elastic band (CI-NEB) method48 with force convergence criteria on each atom of 0.05 eV Å−1. Results and Discussion Main factors limiting the diffusion kinetics of K-ions include the transport of solvated ions through the electrolyte, desolvation, and diffusion through the solid-electrolyte interphase (SEI) film, and diffusion in the host materials.49,50 Based on our DFT calculations, the desolvation energies of K-ions in various solvents, including EC, PC, DMC, EMC, 1,2-dimethoxyethane (DME), dimethyl sulfoxide (DMSO), N,N-dimethylformamide (DMF), and acetonitrile (AN) were much lower than that of Li-ions (Figure 1a) due to the weaker Lewis acidity of K+, indicating higher mobility of K+ through the electrolyte. However, the diffusion kinetics of K+ were still restricted by the higher diffusion energy barriers in the solid host materials (Figure 1b). By contrast, the diffusion energy barrier of Li+ was relatively lower in solid host materials benefiting from its smaller ionic radius (0.76 Å), which might have functioned as an ionic drill to initiate the diffusion process by opening a way for the following K-ions. This ionic-drill strategy could be realized through a hybrid configuration or preinsertion of ions with a smaller ionic radius. In this work, a rationally designed K+/Li+ hybrid chemistry was employed to demonstrate its feasibility. Figure 1 | (a) Desolvation energies of K+ and Li+ in different solvents (EC, PC, DMC, EMC, DME, DMSO, DMF, and AN). (b) Schematic drawing of K- and Li-ions (atoms) energy barriers during the desolvation and diffusion steps. (c) Diffusion energy barriers of K in Sn, Li2Sn5, and LiSn3 crystals, and Li in the Sn crystal. (d) The configuration of the K+-based full battery with a K+/Li+/PF6− hybrid electrolyte, while Sn is used as both the anode and current collector, and EG is used as the cathode. Download figure Download PowerPoint In principle, the K+/Li+ hybrid configuration could be realized via the discovery of suitable anode materials capable of reacting with both K+ and Li+, such as carbon or alloying materials.51–58 In this work, we selected Sn as the anode because of the reversible K–Sn and Li–Sn alloying reactions.59,60 Besides, Sn also exhibits a high theoretical capacity (452 mAh g−1 for K2Sn and 990 mAh g−1 for Li22Sn5)61,62 and good electrical conductivity.63 To check the functional feasibility, the DFT method was used to calculate the diffusion energy barriers of K and Li atoms in the Sn crystal. The calculated diffusion energy barrier of the Li atom was 0.69 eV, which was lower than that of the K atom (Figure 1c). Owing to the much lower barrier, we envisioned the following favorable possibilities: (1) The formation of Li–Sn alloys would be favored energetically. (2) By mixing a small quantity of lithium, Sn-rich phases such as Li2Sn5 and LiSn3 would be formed according to the Li–Sn system.64,65 (3) The preferential formation of Li–Sn phases (e.g., Li2Sn5 and LiSn3) would boost the diffusivity of K since the diffusion energy barriers of K in the Li2Sn5, and LiSn3 crystals were 0.42–0.38 eV lower than that of K in the Sn crystal (Figure 1c). Therefore, the diffusion kinetics of K-ions in the Sn anode could be improved significantly by the ionic-drill effect of Li-ions. To realize the ionic-drill strategy in a full cell, searching for suitable cathode materials was also critical. Although various cathode materials have been reported for K- and Li-ions, it is still challenging to synthesize cathode materials enabling reversible insertion/desertion of both K+ and Li+ simultaneously. Recently, anions (e.g., PF6−) intercalation into graphite cathodes has been reported with high reversibility.66–71 Thus, we predicted that if the insertion/desertion reactions of K+ and Li+ were replaced by the intercalation/deintercalation reactions of anions, it would make the ionic-drill strategy possible through hybrid chemistry. To verify this idea, a K+/Li+/PF6− hybrid configuration was proposed with Sn and EG as the anode and cathode, respectively (Figure 1d). In this hybrid configuration, a 1 M K+/Li+/PF6− was mixed with EC + EMC + DMC + PC (2∶2∶1∶1 v/v/v/v) as the solvent to generate an electrolyte solution. Upon charging, the K+ and Li+ diffused through the electrolyte and alloy with the Sn anode; meanwhile, the intercalation of PF6− anions into the EG interlayers occurred on the cathode side by transporting through the electrolyte. During the discharging process, dealloying and deintercalation reactions occurred correspondingly, with cations and anions diffusing back into the electrolyte. Due to the need for high cut-off voltage (>4.0 V) to ensure the intercalation of PF6− into the EG cathode,72 the electrochemical stability of the hybrid electrolyte was first investigated within the voltage range of 3–5 V using the LSV method. Oxidative current densities of the hybrid electrolyte with different Li concentrations (0–100 atom %) are in the range of 0–0.1 mA cm−2 ( Supporting Information Figure S1), which were far below the standard cut-off value of 0.5 mA cm−2,73 indicating that ignorable oxidation reactions occurred within the voltage window. The K+/Li+ ratio was optimized in terms of diffusion kinetics and capacities. Polarization tests were conducted employing Sn|Sn symmetric cells at current densities of 0.2, 0.5, and 1 mA cm−2 using hybrid electrolytes with different Li ratios. With the increase in Li concentration, the polarization potential at 0.2 mA cm−2 decreases from 0.11 to 0.05 V (Figure 2a). Similarly, decreasing trends were observed at higher current densities of 0.5 and 1 mA cm−2 ( Supporting Information Figure S2), indicating that the diffusion kinetics could be improved significantly by the lithium ionic-drill effect. Further, EIS measurements were conducted for full batteries. When lithium was excluded, the charge-transfer resistance (Rct) value was ∼245 Ω ( Supporting Information Figure S3) owing to the poor diffusion kinetics of K+. By adding a small amount of lithium (10 atom %), the Rct value decreased almost a half to ∼120 Ω ( Supporting Information Figure S3). When the lithium content was increased to 20 atom %, a much lower Rct value (∼35 Ω) was obtained (Figure 2b), suggesting significantly enhanced diffusion kinetics. Although the Rct values could be reduced further by adding more (>20 atom %) lithium (Figure 2b), the maximum discharge capacity was achieved at 20 atom % Li ( Supporting Information Figure S4). Therefore, the Li ratio was optimized to be 20 atom % to achieve a good balance between the reaction kinetics and discharge capacity. Figure 2 | (a) Polarization curves of symmetric Sn|Sn cell using hybrid electrolytes with different Li concentrations (0–100 atom %). (b) EIS of K+-based full batteries with varied Li concentrations (20–100 atom %) in the electrolyte. (c) Charging/discharging curves of a K+-based full battery and the inset show the corresponding dQ/dV profile. (d) The XRD contour map of the EG cathode at different charging/discharging voltages. (e) XRD pattern of the fully charged Sn anode. (f) Different voltages are represented by color dots in a typical charging/discharging profile. (g) Ex situ XRD contour pattern of the Sn electrode at various charging/discharging states within the 2θ range from 16° to 50°. (h) The local enlarged image is highlighted by the yellow square indicated in the figure (g). Note that figures (c–h) refer to the K+-based full battery with an optimized electrolyte (20 atom % Li). Download figure Download PowerPoint After the optimization of the hybrid electrolyte, the electrochemical reaction mechanism was investigated. The intercalation/deintercalation of PF6− anions into/from the EG cathode was identified by the stage character observed in the charging/discharging curve and the corresponding dQ/dV profile within a voltage range of 3.0–4.95 V (Figure 2c and inset). Three stages in the charging (Stages I, II, and III) and discharging processes (Stages III′, II′, and I′) were observed, referring to different intercalation or deintercalation stages of PF6− anions into or from the EG cathode. The XRD contour map of the EG cathode at different charge/discharge states is shown in Figure 2d. As the charging proceeded, the (002) peak of the EG cathode shifted gradually toward the lower angle direction, suggesting an expansion of the interlayer spacing along with the PF6− intercalation. During the discharging process, the (002) peak shifts back to its initial position, indicating good reversibility. At the anode side, the silver-gray Sn foil turned dark after being fully charged ( Supporting Information Figure S5). Uniformly distributed K element was observed on the Sn foil surface by the EDX ( Supporting Information Figure S6c), suggesting the formation of K–Sn alloys. Note that F, P, and O elements were also detected with a homogenous distribution ( Supporting Information Figures S6d–S6f), representing the formation of continuous SEI films. XRD was further conducted for the charged Sn foil to investigate the alloying reaction mechanism. As illustrated in Figure 2e, K–Sn phase (K4Sn9), Li–Sn phase (Li2Sn5), and K–Li–Sn phase (K2LiSn4) were observed, demonstrating the alloying reactions of both K+ and Li+ with the Sn anode. To investigate the reversibility of the alloying reactions, ex situ XRD was performed at different charging/discharging voltages, as illustrated by color dots in Figure 2f. The corresponding XRD contour map was plotted with 2θ values ranging from 16° to 50° (Figure 2g). The (310) peak of Li2Sn5 at 27.43° and (015) peak of K4Sn9 at 28.796° were analyzed in detail (Figure 2h). During the charging process, all the peaks emerged at 4.0 V and became more intense with the increased charging voltage. At the end of the charging (4.95 V), the maximum intensity was achieved. With the discharging proceeding, the intensities of these characteristic peaks in Figure 2h decreased gradually until they vanished at the end of discharging (3.0 V), indicating that the alloying reactions of the Sn anode with both K and Li exhibited high reversibility. As discussed earlier, significantly improved electrochemical performance was expected due to the increased reaction reversibility of both cathode and anode. Coin-type cells were assembled to evaluate the rate and cycling performances of K+-based full batteries. The charging/discharging curves almost maintained the same shape even at a high current rate of 25 C (Figure 3a). As shown in Figure 3b, the discharge capacity remained as high as 84.7 mAh g−1 as the current rate increased from 5 to 25 C showing capacity retention of ∼80%, and it almost recovered its original value when the current rate returned, indicating high reversibility. Owing to the high anion (PF6−) intercalation potential of the EG cathode, high working voltage ∼4.1 V was achieved at 5 C, and only 0.2 V was lost at 25 C (Figure 3c) with smooth charging/discharging voltage profiles (Figure 3c and inset), suggesting high stability of the working voltage, especially at high current rates. Superior rate capability (Figure 3d) and working voltage (Figure 3e) were achieved by the ionic-drill strategy, compared with other previously reported results of K+ full batteries. Outstanding cycling stability >500 cycles at 5 C was also achieved with a capacity retention of ∼100% (Figure 4a). By contrast, the K-based batteries without lithium showed low capacity retention of ∼42% >300 cycles (Figure 4a). The charging/discharging curves (5 C) at varied cycles from 50th to 500th, and almost overlapped with one another (Figure 4b), showing the best result among the reported K+ full batteries (Figure 4c). Figure 3 | Charge/discharge curves (a), rate capacities (b), and working voltage (c) of the K+-based full battery with optimized electrolyte (20 atom % Li) with current rates varied from 5 to 25 C, the inset in figure (c) are the corresponding voltage curves. (d) Rate capability and (e) working voltage comparison of the full battery with other previously reported K+ full batteries with detailed literature data is listed in Supporting Information Table S1. Download figure Download PowerPoint Conclusion An ionic-drill strategy has been developed using smaller ions (Li+) to significantly promote the diffusion kinetics of K-ions in the anode material. High feasibility has been demonstrated via a rationally designed K+/Li+/PF6− hybrid configuration with Sn and EG as the anode and cathode, respectively. After optimization in terms of reaction kinetics and discharge capacity, full batteries delivered high rate capability up to 25 C with the capacity being maintained as high as 84.7 mAh g−1 and superior cycling performance of >500 cycles at a high current rate of 5 C and capacity retention of ∼100%. Moreover, high working voltage (∼4.1 V at 5 C and ∼3.9 V at 25 C) was achieved, making our current findings so far the best reported among K+ full batteries. The present results show that the highly effective hybrid configuration realizes the ionic-drill strategy, which could also be applied to other energy storage devices suffering from sluggish kinetics. Notably, other approaches, such as preinsertion of ions in host materials with smaller ionic radii are also feasible. Figure 4 | (a) Cycling stability comparison (5 C) of full batteries with the optimized electrolyte (20 atom % Li) and pure K in the electrolyte, and (b) charging/discharging curves of the optimized K+-based full battery at varied cycles form 50th–500th. (c) Cycling performance comparison of the optimized full battery with other previously reported K+ full batteries with detailed literature data is listed in Supporting Information Table S1. Download figure Download PowerPoint Supporting Information Supporting Information is available. Conflict of Interest There is no conflict of interest to report. Acknowledgments The authors gratefully acknowledge financial supports from the Key-Area Research and Development Program of Guangdong Province Science of and Shenzhen Science and Technology and Guangdong and Research and Science and Technology of Guangdong Province for Enabled by a High on and in Li on in and with for Zhou Based on with an C. Tang with High for Advanced Zhou Zhou with Li Based on the of and 10, via Advanced and Zhou Li for Zhou Li Battery Enabled by Li C. of for Yan in with for for C. Tang Strategy Full with High and Rate the Rate and by and

Topics & Concepts