Potassium-ion batteries (PIBS) are a promising low-temperature energy storage battery. However, due to the lack of feasible anode materials and compatible electrolytes, the current research on low-temperature potassium-ion batteries is limited to half-cells using metal potassium as the anode, and there are still many challenges in realizing rechargeable full batteries.
Brief introduction of the results
Recently,Professor Wang Hua of Beijing University of Aeronautics and Astronautics and othersFor the first time, a hard carbon (HC)-based low-temperature potassium-ion all-cell battery was reported. Through experimental evidence and theoretical analysis, the authors revealed the potassium storage behavior of HC anodes in matched low-temperature electrolytes, including defect adsorption, interlayer co-insertion, and nanopore filling. These unique potassium processes exhibit low interfacial resistance and low reaction activation energy, resulting in HC with 175 mAh G-1 capacity and excellent cycling performance at 40°C. At 40°C, the HC anode-based whole cell exhibits high energy density and rechargeability above 100 Wh kg-1.
**Reading guide
Figure 1Schematic diagram of the K storage behavior of graphite and (b) HC electrodes at low temperatures; (c) Schematic diagram of the K storage mechanism of the HC electrode.
By coupling the HC anode with a compatible electrolyte, the authors succeeded in constructing a low-temperature rechargeable potassium-ion full cell. The authors found that potassium storage behaviors in HC include defect adsorption, interlayer co-interpolation, and nanopore filling (Figure 1C). The whole potassium storage process exhibited low resistance values and reaction activation energies, indicating that temperature had the least effect on the electrochemical performance of the HC electrode. hc at 0High reversible capacity of 175 mA Hg-1 is provided at 2 C and 40 °C. In addition, K-ion whole cells assembled with iron (FeHCN) or perethylene-tetracarboxydianhydride (PTCDA) as the cathode and HC as the anode show excellent performance at 40°C.
Figure 225 and -40 cases (a) specific discharge capacity of graphite, soft carbon and HC electrodes in various electrolytes and (b) median discharge voltage; (c) Graphite, soft carbon, and HC electrodes at 0Constant current charge-discharge curves in 2C, 40°C formulated electrolytes.
First, the authors evaluated the use of graphite, soft carbon, and HC as anode materials in 25 and -40 M KPF6DME-LiNO3,0Potassium storage performance in three electrolytes: 8 M KPF6 EC dec and 3 M KFSI DME. The results show that the three carbon materials are only matched to the optimized electrolyte and have reversible capacities at 40°C and HC at 0The highest volume retention rate was exhibited at 2C (Figure 2A). The constant current charge-discharge curve in Figure 2c is shown at 0The discharge specific capacity of HC at 2 C and 40 °C is 175 mAh g-1, and the median discharge voltage is about 018 v (vs.k+k) (Figure 2b). This voltage value is higher than that of conventional graphite (< 01 V) to reduce the risk of potassium metal deposition during HC potation. In contrast, at 40°C, graphite and soft carbon have a much lower capacity and steeper curves.
Figure 3(a) the cyclic stability of the HC electrode and (b) the corresponding constant current charge-discharge curves were formed at different temperatures of 25 to 50°C; (c) the rate performance of the HC electrode and (d) the corresponding constant current charge-discharge curve at 40°C; (e) The HC electrode is at 40°C and 0After 5 times of activation at 5 C, at 0Cycling performance at 5 C; (f) at 001 and 2The scan rate between 0 v is 0CV curve of HC electrode at 1 mV-1; (g) Gitt curves, and (h) corresponding k-ion diffusion coefficients under different discharge charge states.
The authors further tested the potassium storage performance of HC and formulated electrolytes at low temperatures. As shown in Figures 3a and b, at 0At 5 C current density, the specific capacitance of HC at 0°C, 20°C, 40°C, and 50°C is maintained at % and 22% of 25°C, respectively. At the same time, the potassium-depotassium potential distribution remains very stable, indicating that there is a consistent electrochemical reaction in this temperature range.
Similarly, at 40°C, the charge-discharge curves of different current densities exhibit a similar shape at 0The stable capacity was 171 mAh g-1 at 2 C and the reversible capacity was 45 mAh g-1 even at 2C (Figures 3c and D), indicating good ion storage performance at low temperatures. When the current density returns to 0At 2 C, the capacity can be fully recovered, which indicates that the HC anode has good stability after high-rate discharge and charge tests.
In addition to this, long-term cycling performance measurements show a current density of 0After 400 cycles at 5 C and 40°C, the volume retention rate of HC was 73%, and the reversible capacity was 128 mAhg-1 (Figure 3E), indicating that HC has good low-temperature capacity stability in the optimized electrolyte. To study the storage behavior of potassium over the entire electrochemical range, the authors first started with 0A scan rate of 1 mV-1 was tested with cyclic voltammetry (CV) to reduce voltage hysteresis at 25°C.
In the formulated electrolyte, the CV curve of the HC electrode shows three pairs of redox peaks (Figure 3F), which correspond to surface adsorption (0.).93/0.79 V), interlayer insertion (0.).24/0.28 V), as well as nanopore filling of potassium metal ions (001/0.12 v）。Among them, the incomplete distribution of the cathode peaks around 0 V suggests that the potassium storage reaction at this stage did not end at the potential near the metal potassium deposition, which means that the ongoing process may be related to K-metal plating. At the same time, it can be observed from the magnified plot of the post-discharge curve of the constant current intermittent titration technique (GITT) at 25 °C that the slope of the discharge curve remains constant (Figure 3g), indicating that the K storage reaction remains constant in the near-zero voltage region.
In addition, in 2Around the cut-off voltage of 0 V, the diffusion coefficient of K+ decreases rapidly during charging, while from 2 V to 0During the discharge of 01 V, it remained stable in the tail voltage region (Figure 3h). These results suggest that the potassium storage behavior that occurs in the later stages of the discharge process (possibly nanopore filling) is similar to that of potassium deposition.
Figure 4Study on the mechanism and kinetics of potassium storage in HC electrodes.
The authors then performed qualitative spectroscopic measurements to reveal the storage mechanism of the HC electrode at 40°C. First, the authors used in-situ Raman spectroscopy to study the evolution of chemicals during the charge-discharge process of the HC electrode (Fig. 4A). During the potassiumization phase, the G peak gradually shifted from 1591 to 1577 cm-1, the Id Ig value decreased, and a new peak appeared at 1053 cm-1.
These phenomena can be explained as follows: First, the intercalation of k increases steric hindrance and electron density, resulting in the stretching and subsequent expansion of carbon-carbon bonds in the plane, which leads to the red shift of the g peak. The decrease in the Id Ig value indicates an increase in the degree of graphitization during the potassiumization process. The new peak at 1053 cm-1 is due to the co-insertion of K+- solvent compounds in the carbon layer.
To further understand the intercalation mechanism of solvated K+, the authors estimated the number of DME molecules co-inserted with K+ by detecting the mass changes of HC electrodes in different charge states (SoCs). By measuring the specific volume of 175 mAh G-1 at 40°C, the K-C ratio of HC under total potassium is approximately 1:13. As shown in Figure 4B, the weight change of the electrode during potassium-depotassiumation closely corresponds to a slope line of A=1, indicating that K+-DME is reversibly co-inserted in HC.
Subsequently, the authors performed small wide-field X-ray scattering (SAXA and WAXS) to fully understand the nanopore storage mechanism of HC at 40°C. As shown in Figure 4c, at Q=1A broad peak appears at 64 a-1, indicating the formation of k clusters, the intensity of which increases as the potassium process increases. When the potential drops below 0V, the K metal begins to be deposited, at Q=1A strong Bragg reflex from the k-eigens can be detected at 71 A-1 (Figures 3D and E).
These results show that the deposition of large K metals does not occur above 0 V, even at 40°C. In addition, the geometric size of the nanopores can be inferred from the scattering intensity of the electrodes at different voltage states. As shown in the in-situ SAXS mode in Figure 4F, the scattering intensity from the nanopores decreases gradually, indicating that there is more and more k in the nanopores as the depth of discharge increases. When charging the HC to 2At 0 V, the scattered signal of the nanopores returns to their initial intensity, suggesting that the nanopores tend to act as sites for reversible storage k.
Based on the above characterization analysis, three storage behaviors of HC at low temperatures are clearly illucated: 1) adsorption of K on surface defects; 2) K+-DME co-inserted into the graphene layer, and 3) K clusters filled with nanopores. To further elucidate the storage kinetics of potassium in HC at low temperatures, the authors performed in-situ electrochemical impedance spectroscopy (EIS) measurements at 40°C. The EIS profile typically shows an intercept at high frequencies corresponding to the bulk electrolyte and electrode resistance, two semicircles of high and medium frequencies, attributed to the solid inter-electrolyte phase (SEI) layer resistance (RSEI) and charge transfer resistance (RCT), respectively, a straight line representing the ion diffusion impedance within the host material. At the beginning of the discharge (Figure 4g), the Nyquist plot shows an incomplete high-frequency semicircle and a straight line with a slope of about 45° at low frequencies (Fig. 4h), while there is no semicircle at the mid-frequency.
This indicates that the K-ion adsorption at the surface-active site of HC is dominant in the voltage tilt region. Subsequently, a semicircle with a large radius appears at the mid-frequency and decreases rapidly with potassiumation in the plateau region (Figure 3i), suggesting a change in the storage behavior of K, which may be determined by interlayer co-intercalation. Finally, the high-frequency semicircle becomes the main part, and the mid-frequency is further reduced to 9 in the later stage of discharge.
The resistance values at different voltage states are summarized in Figure 4j and it can be seen that even at 40°C, the maximum value of resistance is only 98, which is significantly lower than the value reported in the literature. This explains why HC electrodes can exhibit high reversibility and rate performance in low temperature environments. In addition, the activation energies of three potassium storage behaviors were obtained from the Arrhenius plot (Figure 4K). The activation energy of the surface adsorption process and the nanopore filling process is smaller, which is 0., respectively61 kJ mol-1 and 184 kj mol-1。Due to the facilitating effect of the k-solvent co-intercalation, the estimated activation energy is only 316 kJ mol-1, so K+ does not need to completely remove its solvated sheath.
Figure 5Computational analysis of energy distribution and reaction pathways.
The authors further performed density functional theory calculations to elucidate the diffusion mechanism of the three K storage behaviors in HC (Fig. 5). As shown by the calculated charge density difference (Fig. 5a, c, e), part of the charge of the K+-DME complex can be compensated by DME molecules outside the graphite layer or the surface of the K(002) crystal during diffusion in the surface adsorption, insertion, and nanopore filling states.
This is due to the fact that the oxygen atom in the dimethyl ether molecule is an electron donor. The effect of partial charge compensation of the DME molecule weakens the interaction of the K+-DME complex with the graphite layer or the K(002) crystal surface, which contributes to a smaller diffusion energy barrier and thus provides a more favorable reaction kinetics environment compared to the charge compensation alone on the graphite layer or K(002) crystal surface.
As expected, the calculated diffusion energy barriers of solvated ions in HC from site 0 6 along the optimal path of adsorption, insertion, and archiving were respectively. 224 and 0163 EV (Figures 5b, d, and f). These low-energy barriers do not significantly impede the electromigration or free diffusion of solvated K ions, explaining the almost temperature-independent characteristics of the diffusion coefficient estimated from GITT. In general, the reduction of the resistance value, the activation energy of the reaction, and the ion diffusion barrier during the whole charge-discharge process is conducive to the efficient and rapid storage of K ions in HC at low temperatures.
Figure 6Experimental and theoretical analysis of different electrolytes.
In addition to the electrode material, electrolyte chemistry also plays a key role in determining the capacity of a rechargeable battery at low temperatures. In principle, a compatible low-temperature electrolyte should have high ionic conductivity and good wettability.
Therefore, the electrolyte composed of 1 M KPF6-DME-LiNO3 was chosen for the following reasons:
1) DME solvent has a low freezing point and a moderate dielectric constant, which guarantees smooth ion mobility and high ionic conductivity at low temperatures2) The relatively high LUMO energy level enhances its compatibility with K metals, reduces the formation of SEIs, and promotes the intercalation of solvent molecules3) PF6 was selected as an anion due to its low donor count, reducing energy consumption during desolvation;4) Add a small amount of LiNO3 to improve the oxidation stability of the electrolyte at high pressure.
First, the authors used differential scanning calorimetry (DSC) measurements to characterize the freezing point of electrolytes. As shown in Figure 6a, 3 M kfsi-dme and 0The freezing point of 8 M KPF6-EC dec was 376°C and 186°c。In comparison, the freezing point of KPF6-DME-LiNO3 is estimated to be as low as 619°C, it has the potential to hold fluids at low temperatures. In addition, KPF6-DME-LiNO3 exhibits low viscosity and high conductivity of the electrolyte at different temperatures (Figures 6b and c).
Raman spectroscopy was used to probe the solvation structure of the electrolyte (Fig. 6D). Specifically, the characteristic peak of the free DME molecule is centered at 828 cm-1 and corresponds to the tensile vibration of the C-O bond, but when KFSI is added to the mixture, the peak appears blue-shifted to 836 cm-1, which is due to the increased coordination between K+ and DME solvents, while the peak of KPF6-DME does not change.
Meanwhile, in the KPF6-based electrolyte, the blue-shifted Raman peak of the PF6 anion can be observed when ECDEC is used instead of DME, which can be attributed to the presence of the PF6 complex solvated by DME and its weak interaction with K+. In addition, the designed electrolyte has K-(DME)5., as shown in the molecular dynamics (MD) simulation snapshot and K+ radial distribution function (RDF) data of the formulated electrolyte (Figure 6E).4(pf6–)1.1. Characteristic solvent separation ion-pair structure.
Based on the above analysis, the authors can infer that K+ has a moderate affinity for solvents, while their interaction with anions is relatively weak. This property mediates solvation copolymerization, which inhibits energy expenditure during desolubilization, resulting in rapid kinetics at low temperatures.
Figure 7Electrochemical performance of the whole cell at 40°C based on HC anode.
Finally, the authors evaluated the electrochemical performance of the HC-based anode and PTCDA or FEHCN cathode at low temperatures in the whole cell (Figs. 7A and B). The whole battery with PTCDA cathode exhibits excellent reversible charge-discharge capacity and rate performance at 25°C. At 40°C, at 0The battery has a discharge capacity of 89 mAh g-1 and an energy density of 157 Wh kg-1 at 5 C, maintaining 76% of its capacity at room temperature (Figure 7C). In addition, at 0It exhibited stable cycling performance over 100 cycles at 5 C and retained 79% of its initial capacity well (Figure 7e).
At the same time, the whole cell using the FeHCN cathode also showed reversible capacity and cycling capacity at 25 and 40°C (Fig. 7D,E), with capacity retention rates of 80% and 72% after 100 cycles, respectively. The FehCN HC full cell achieves a high energy density of 102 Wh kg-1 at 40°C. These results show that the HC-based whole battery assembled with two different cathodes can achieve good energy density and stable cycling performance at low temperatures, indicating the great potential of HC-based KIBS under low temperature conditions.
Bibliographic information
jiangchun chen, dong an, sicong wang, han wang, yingyu wang, qiaonan zhu, dandan yu, mengyao tang, lin guo, and hua wang. angew. chem. int. ed. 2023, e202307122.