Advertisement

Cathode Materials for Potassium-Ion Batteries: Current Status and Perspective

  • Qing Zhang
  • Zhijie Wang
  • Shilin Zhang
  • Tengfei Zhou
  • Jianfeng Mao
  • Zaiping Guo
Review article
  • 372 Downloads

Abstract

Potassium-ion batteries (PIBs) have recently attracted considerable attention in electrochemical energy storage applications due to abundant and widely distributed potassium resources and encouraging intercalation chemistries with graphite, the commercial anode of lithium-ion batteries. One main challenge in PIBs, however, is to develop suitable cathode materials to accommodate the large size of K+ ions with reasonable capacity, voltage, kinetics, cycle life, cost, etc. In this review, recent advancements of cathode materials for PIBs are reviewed, covering various fundamental aspects of PIBs, and various cathode materials in terms of synthesis, structure, and electrochemical performance, such as capacity, working potential, and K-storage mechanisms. Furthermore, strategies to improve the electrochemical performance of cathode materials through increasing crystallinity, using buffering and conducting matrixes, designing nanostructures, optimizing electrolytes, and selecting binders are summarized and discussed. Finally, challenges and prospects of these materials are provided to guide future development of cathode materials in PIBs.

Graphical Abstract

Keywords

Cathode materials Redox couples Potassium-ion batteries Energy storage 

PACS

82.45.Fk 

1 Introduction

Due to the depletion of fossil fuels and associated environmental issues arising from their use, there is growing interest in the exploration and application of renewable energy resources such as hydro, solar and wind. In conjunction with this, there is also a strong demand for eco-friendly and sustainable energy storage systems for the efficient storage and utilization of these renewable but intermittent energy resources. Here, rechargeable batteries are regarded as one of the best energy storage methods due to pollution-free operations, adjustable shapes and sizes (ranging from mobile phone batteries to megawatt-scale systems), high efficiencies, and long cycle lifespans. And of all the rechargeable batteries currently available, lithium-ion batteries (LIBs), first commercialization in 1991 by Sony, dominate the consumer electronics market such as mobile phones, cameras, and laptops due to high energy densities and efficiencies. To meet the ever-growing demand for cheaper, smaller and lighter portable electronics, however, LIB technologies need to be further improved in terms of energy/power density, efficiency, costs, etc. [1]. In addition, this demand is further fuelled by increasing developments in stationary electric energy storage systems and electric vehicles, inevitably leading to increased LIB prices due to the limited and remote lithium resources in the Earth’s crust (less than 20 ppm, with most being restricted to South America, Australia, China, and the USA) [2].

These concerns over the price and depletion of lithium resources have subsequently led to the development of sodium-ion batteries (SIBs) and potassium-ion batteries (PIBs) because sodium and potassium are not only abundantly available but are also evenly distributed around the world (Table 1) [3, 4, 5, 6, 7, 8, 9, 10, 11]. Moreover, Na and K do not form alloys with Al electrochemically, meaning that low-cost Al foil can be used as current collectors for both anodes and cathodes in SIBs and PIBs, whereas more expensive Cu foil is normally required in the anode side of LIBs, leading to lower expected costs for SIBs and PIBs as compared with LIBs. In addition, Na and K are in the alkali series next to Li and share similar chemical properties with Li, allowing for the knowledge from LIBs to be transferred to SIBs and PIBs. Exploration of SIBs started earlier than PIBs and has been boosted since 2011 [3]. Graphite, the commercial anode material in LIBs, produces poor capacities in SIBs however, because Na cannot intercalate into graphite to form graphite intercalated compounds [12]. Alternatively, hard carbon is currently regarded to be the most suitable anode material for SIBs due to its high specific capacity (300 mAh g−1), but suffers from high production costs and a low volumetric capacity of 450 mAh cm−3, which is around half that of graphite in LIBs [13]. In addition, the operational voltage of SIBs is lower compared with LIBs due to the relatively high standard potential ((Li/Li+): − 3.04 V; (Na/Na+): − 2.71 V). In contrast, graphite can intercalate and de-intercalate K+ ions reversibly [13], and therefore, well-established systems for LIBs can be more smoothly transferred to low-cost PIBs than other rechargeable batteries. In addition, the standard electrochemical potential of K/K+ (− 2.93 V vs. E°) is lower than that of Na/Na+ and close to that of Li/Li+ and in some carbonate solvents is even lower than that of lithium, suggesting potentially higher working voltages for PIBs compared with SIBs and LIBs [14, 15]. Therefore, based on its material abundance, standard electrode potential, and potassium intercalation chemistry, PIBs are suitable alternatives to LIBs [16].
Table 1

Comparison of the physical properties of “lithium”, “sodium”, and “potassium” for rechargeable batteries

 

Li+

Na+

K+

Atomic mass

6.94

22.99

39.09

Ionic radii (Å)

0.76

1.02

1.38

Melting point (°C)

180.5

97.7

63.4

Abundance in the Earth’s crust (mass %)

0.0017

2.3

1.5

vs. SHE (V)

− 3.04

− 2.71

− 2.93

E°vs. Li+/Li in PC (V)

0

0.23

− 0.09

Desolvation energy in PC (kJ mol−1)

215.8

158.2

119.2

Ionic conductivity of the electrolyte of 1 M MFSI in EC/DEC (mS cm−1)

9.3

9.7

10.7

Theoretical capacity of graphite (mAh g−1)

372

 

279

Reaction voltage of graphite vs. M+/M (V)

~ 0.1

0.01 for hard carbon

~0.2

Theoretical capacity of M3V2(PO4)3 cathodes (mAh g−1)

132

118

106

Reaction voltage of M3V2(PO4)3 cathodes vs. M+/M (V)

3.7

3.4

3.8

Studies on the electrochemical behaviours of K+ ions started in the 1980s [17, 18, 19, 20], and although this is close to the exploration timelines of Li+ ions and Na+ ions, research has been limited over the years due to concerns over the relatively heavy atomic mass and larger ionic radius of potassium as compared with lithium and sodium. And although it is true that heavier K can lead to decreased theoretical capacities, these differences become smaller if the comparisons of theoretical gravimetric capacity were made between host materials instead of metal electrodes because capacity is primarily determined by host materials. For example, in the case in which two-electron reactions of a V3+/V4+ redox couple are involved, theoretical capacities can reach 132, 118, and 106 mAh g−1 for Li3V2(PO4)3, Na3V2(PO4)3, and K3V2(PO4)3, respectively, with the calculated capacity of the PIB cathode being only 19.7% less than that of the LIB cathode. This difference becomes even smaller in other materials. In addition, this problem of decreased theoretical capacity can be compensated by increased energy densities because PIBs produce higher voltages than SIBs and LIBs in various organic electrolytes. For example, Komaba et al. [16] reported that the plating/stripping potential of K/K+ is 0.15 V lower than that of Li+/Li in EC/DEC-based electrolytes, and that this difference originates from different solvation energies due to the different sizes of K+ and Li+ ions, in which compared with K+ ions, smaller Li+ ions possess a larger desolvation energy and therefore a higher redox potential for Li+/Li. In addition, larger K+ ions with a smaller desolvation energy and solvated cations in liquid electrolytes will have higher transference numbers and ionic conductivities than smaller Li+ and Na+ ions, and therefore better kinetics and rate performances in PIBs [21]. And although it is undeniable that larger K+ ions may affect the structural stability of host materials and possible interphase formations during charge/discharge which can result in limited cycles, advances in material research will potentially resolve these issues.

The research into PIBs was further accelerated in 2015 when graphite was found to be capable of accommodating potassium electrochemically and reversibly at room temperature [14]. Starting from this important finding, numerous studies have been conducted, pointing to potential application in electric vehicles (EVs) and large-scale energy storage systems [11]. In addition, various anode materials for PIBs have also been reported, and other carbonaceous materials, such as hard carbon, soft carbon, and high-capacity metal-based anode materials have also been investigated and have produced excellent results [22, 23, 24]. In particular, the insertion potentials of K+ ions in most carbon anodes (~ 0.2 V vs. K/K+) and some non-carbon anodes such as Sn4P3 (~ 0.1 V vs. K/K+) are slightly higher than that in SIBs [24] and are well above the plating potential of K, but are low enough for anodes, suggesting that PIBs are safer operationally than SIBs. However, compared with the relatively successful research into anode materials, research into cathode materials for PIBs is much more challenging. Like in LIBs, cathode materials play a key role in determining the energy, power, cycling life, safety, and costs of PIBs. And despite the various reviews that are currently published concerning PIBs, a review into the progress of cathode materials is critical [6, 7, 8, 9, 10, 11]. Because of this, this review will summarize the current status and prospective of cathode materials for PIBs, starting with the importance, configuration, and the principle of PIBs, followed by a detailed focus on the redox reactions of cathodes along with their compositions, structures, and electrochemical properties, as well as strategies for enhancing their performance. Furthermore, the effects of electrolytes and binders on PIBs will be reviewed and discussed. Finally, a brief overview of the research on full cells will be presented along with prospects. Overall, this review aims to help researchers better design advanced cathode materials with high capacities, high voltages, and improved electrochemical performances for PIBs.

2 Configuration and Principle of PIBs

PIBs operate through a similar rocking-chair principle to that of LIBs, in which K+ ions are shuttled between the anode and cathode using an intercalation mechanism through an electrolyte (Fig. 1a). The battery structure and working mechanisms of PIBs are also similar to LIBs in which during charging, the cathode (e.g. K0.5MnO2) experiences an oxidation reaction with K de-intercalation and electron loss, and the anode simultaneously experiences a reduction reaction (e.g. graphite) with K intercalation and electron acquisition, as K+ ions and electrons move to the anode through the internal electrolyte and external conduction, respectively. During discharge, the opposite process occurs. Here, the electrochemical potential of the anode (µA) and the cathode (µC) needs to match the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) of the electrolyte, respectively, in which if the µA of the anode is above the LUMO or the µC of the cathode is below the HOMO, the electrolyte will reduce on the anode or oxidize on the cathode as a result of the thermodynamically favourable electron transfer from the anode to the electrolyte or from the electrolyte to the cathode. Therefore, passivation layers are required to block this electron transfer (Fig. 1b) [25].
Fig. 1

a Schematic of the operational principle and the charge/discharge mechanism of PIBs, in which cathode and anode materials are represented by layered K0.5MnO2 and graphite, respectively. b Schematic of the open-circuit energy diagram of the electrolyte window (Eg) and µA and µC of the anode and cathode. µA > LUMO and/or µC < HOMO requires kinetic stability through the formation of an SEI layer. Reprinted with permission from Ref. [25]. Copyright 2010 American Chemical Society

The energy density of PIBs is related to the capacity of the anode and cathode as well as the cell potential. Therefore, the theoretical capacity of electrode materials can be calculated by using the equation: Q = nF/3.6 M, in which n is the electron-transfer number, F is the Faraday constant, and M is the molar mass of the materials. Based on this, small molar masses and large electron-transfer numbers can lead to high capacities, suggesting that favourable elements for high-capacity electrode materials are mostly found in the first four rows of the periodic table. In addition, because anode materials normally possess higher capacities than cathode materials, K-ion storage capacities of PIBs are mostly determined by cathode materials. Here, cathode materials are in general transition metal-based compounds because the variable valence states of transition metals allow for the storage of more electrons and, therefore, can potentially offer higher capacities.

The potential of a cell is determined by the difference between the electrochemical potential of the anode (µA) and the cathode (µC), which is related to the partial molar quantity of the Gibbs free energy and follows the equation: E = − ∆G/nF, in which ∆G is the Gibbs free energy difference. Here, the µA values of carbon-based anodes are similar, whereas the µC values are intrinsically determined by the redox energy of a transition metal, as is the case for LiCO2 (vs. the Fermi level of metallic Li) (Fig. 2a) [25]. Therefore, the design of higher µC values is vital to increasing working voltages for cells. Figure 2b compares the redox potentials and capacities of phosphates in LIBs [26], and from this, it is clear that with increasing atomic numbers in the fourth row of the periodic table, the redox potentials of phosphates increase as well, following the electronegativity rule in which greater electronegativity leads to greater electrochemical potentials [27]. Additionally, the atomic radii of atoms decrease going from the bottom to the top and from the left to the right of the periodic table, in which smaller elemental radii lead to increased electronegativity as a result of the stronger attraction of valence electrons to positively charged nuclei. Furthermore, anionic groups with different elements or configurations also possess different electronegativity, which affects the type of bonds between metal cations and anionic groups. Figure 2c reveals that polyanionic groups such as phosphates, silicates, and sulphates exhibit higher voltages than oxides for some transition metal-based redox couples due to higher electronegativities in LIBs [28]. And because different polyanionic groups possess different electronegativities, they also possess different electrochemical potentials, which also follow the same electronegativity rule. Therefore, the tailoring of redox potentials depends not only on the valence state of cations, but also on the anion type and structure.
Fig. 2

a Schematic of the corresponding energy vs. density of state showing the relative positions of the Fermi energy in an itinerant electron band for LixC6, and the Co4+/Co3+ redox couple for LixCoO2. Reprinted with permission from Ref. [25]. Copyright 2010 American Chemical Society. b Mean voltage in phosphates vs. achievable maximum gravimetric capacity. Reprinted with permission from Ref. [26]. Copyright 2011 American Chemical Society. c Respective positions of Fe3+/Fe2+ redox couples vs. Li+/Li in polyanionic groups. Reprinted with permission from Ref. [28]. Copyright 2013 American Chemical Society

Currently, laboratory studies of electrode materials for alkali-ion secondary batteries mainly rely on half cells, in which alkali metals are used as counter electrodes. Here, the assembly procedure is similar, regardless of whether LIBs, SIBs, or PIBs are being constructed. Unlike lithium cells, however, whose Li foils are commercially available and can be directly used as electrodes, commercially available sodium or potassium metals are not available in foil forms and therefore require pre-processing before assembly in Na and K cells. In addition, more rigorous conditions (low O2 and moisture levels in the glove box) are required for assembling K cells because K metal is more active than Li and Na metals.

3 Cathode Materials

Generally, the charge/discharge potentials of cathode materials mainly depend on the redox potential of the element that participates in the electrochemical reaction. By analysing and studying redox reactions, these electrochemical potentials can be better understood and manipulated, allowing for the design of advanced cathode materials for PIBs. Until now, most reported cathodes for PIBs have been based on the redox couple of transition metals such as Fe, Mn, Co, and V; and in this review, the redox couple of these metals and their complexes will be discussed. In addition, various cases such as organic cathodes will also be discussed to present an integrated view of cathode materials for PIBs.

3.1 Single-Metal Redox Couple Cathodes

3.1.1 Iron Redox Couple

Single-electron transferred Fe2+/Fe3+ redox couple cathode materials have been widely used in LIBs and SIBs as a result of their desirable performances. For example, Fe2+ in LiFePO4 (a popular LIB cathode) can be oxidized to Fe3+ during the charge process to deliver a stable plateau at ~ 3.7 V and a capacity of ~ 160 mAh g−1 [29]. Other SIB cathode materials such as \( {\text{Na}}_{1.4} {\text{Cu}}^{\text{II}}_{1.3} {\text{Fe}}^{\text{II}} \left( {\text{CN}} \right)_{6} \) [30], \( {\text{Na}}_{1.94} {\text{Ni}}^{\text{II}}_{1.03} {\text{Fe}}^{\text{II}} \left( {\text{CN}} \right)_{6} \) [31], and \( {\text{Na}}_{2} {\text{Zn}}^{\text{II}}_{3} \left[ {{\text{Fe}}^{\text{II}} \left( {\text{CN}} \right)_{6} } \right]_{2} \) ·9H2O [32] can also store energy based on the Fe2+/Fe3+ redox couple. And apart from these examples, oxides such as NaFeO2 [33] can also store energy as SIB cathodes based on the redox couple of Fe3+/Fe4+. In the cases of PIBs, most Fe-based cathodes are comprised of Prussian blue (PB) and Prussian blue analogues (PBAs), which possess energy storage mechanisms based on the Fe2+/Fe3+ redox couple. The chemical formula of classic PB is KFeIII[FeII (CN)6], and it possesses a metal organic framework (MOF) structure in which Fe3+ and Fe2+ are bridged by cyanide groups (–C≡N–) (Fig. 3a) [10], leading to a three-dimensional (3D) network that is able to provide channels for rapid intercalation and de-intercalation, indicating that PBAs are promising for K+ storage [10]. Eftekhari et al. [34] directly utilized pure PB as the cathode for PIBs, and in this study, the researchers reported that during the charge process, Fe2+ oxidizes to Fe3+ based on Eq. (1):
$$ {\text{KFe}}^{\text{III}} \left[ {{\text{Fe}}^{\text{II}} \left( {\text{CN}} \right)_{6} } \right] \, \to {\text{Fe}}^{\text{III}} \left[ {{\text{Fe}}^{\text{III}} \left( {\text{CN}} \right)_{6} } \right] \, + {\text{ K}}^{ + } + {\text{ e}}^{ - } $$
(1)
Fig. 3

a Structural illustration of PB. Reprinted with permission from Ref. [10]. Copyright 2017 American Chemical Society. b Charge–discharge curves of PB at 0.1 C. Reprinted with permission from Ref. [34]. Copyright 2004 Elsevier B.V. c CV curves of K0.22Fe[Fe(CN)6]0.805·□0.195·4.01H2O at a scan rate of 0.1 mV s−1; d indication of different charge and discharge states; and e ex situ Raman spectra at different states of K0.22Fe[Fe(CN)6]0.805·□0.195·4.01H2O. Reprinted with permission from Ref. [35]. Copyright 2017 John Wiley and Sons. f CV curves at 0.1 mV s−1, g reaction mechanisms and h voltage profiles of K4Fe(CN)6/C. Reprinted with permission from Ref. [36]. Copyright 2018 John Wiley and Sons

Here, this one-electron-transfer reaction reportedly delivered a capacity of ~ 80 mAh g−1 with a sloping plateau between ~ 3.7–3.9 V (Fig. 3b) and, after 500 cycles at 0.1 C, still possessed a charge capacity of ~ 68 mAh g−1. These high-voltage profiles and good capacity retention rates demonstrate that PB is a promising cathode material for PIBs. In another example, Lei et al. [35] synthesized K0.22Fe[Fe(CN)6]0.805·□0.195·4.01H2O (□ represents the (CN)6 vacancy) and applied the resulting material as a cathode in PIBs. Here, a pair of redox peaks at 3.53/3.18 V during cyclic voltammetry(CV) testing (Fig. 3c) were observed and the researchers suggested that the CV peaks be related to the Fe2+/Fe3+ redox couple of C-coordinated Fe along with the K+ de-intercalation/intercalation process. To verify this mechanism, these researchers further conducted ex situ Raman analysis and reported that during charging, the intensity ratio of the C-coordinated Fe2+-peak (~ 2080 cm−1) to N-coordinated Fe3+-peak (~ 2130 cm−1) decreased as K+ was extracted from the lattice and increased during discharging (Fig. 3d, e), suggesting that the Fe2+/Fe3+ redox couple of the C-coordinated Fe be the redox-active site and be therefore electrochemically responsible for K+ storage.

Furthermore, Li et al. [36] applied low-cost potassium hexacyanoferrate(II) trihydrate (K4Fe(CN)6) as a cathode and studied its electrochemical behaviours. In this study, to fully exploit the electrochemical behaviours, the researchers mixed K4Fe(CN)6 with super P to enhance conductivity and obtained CV curves which displayed a pair of sharp peaks that located at 3.67/3.5 V, corresponding to the K+ extraction/insertion process (Fig. 3f). By calculating the ratio of the first charge capacity to the theoretical capacity, the researchers claimed that the K4Fe(CN)6/C composite can accommodate the extraction/insertion of 0.9 K+ per formula unit. In addition, in the X-ray diffraction (XRD) pattern of the charged cathode, the obtained diffraction peaks were attributed to a new phase of K3Fe(CN)6, indicating a Fe2+/Fe3+ redox reaction in the K4Fe(CN)6, corresponding to a one-electron-transfer reaction (Fig. 3g). Based on this mechanism, this study reported that the K4Fe(CN)6 cathode material was capable of delivering a discharge capacity of ~ 70 mAh g−1 (Fig. 3h).

Although Fe2+/Fe3+ redox couple-based cathodes with one-electron transfer show promising prospects in PIBs, their capacities are relatively low. However, by increasing the electron number involved in the redox reaction, cathode capacities can be increased. Therefore, enhancing the K content in electrodes (developing potassium-rich materials) is an effective method to increase capacities. For example, Chen et al. [37]. synthesized K1.92Fe[Fe(CN)6]0.94·0.5H2O as a cathode for PIBs and reported charge and discharge capacities of 170 mAh g−1 and 128 mAh g−1, respectively, which were higher than PB. Here, the researchers reported that in the structure of K1.92Fe[Fe(CN)6]0.94·0.5H2O, N-coordinated Fe exists as bivalent-Fe rather than trivalent-Fe (Fig. 4a), meaning that more than one type of Fe2+/Fe3+ redox couple is involved in the charge/discharge process. As a result, two redox couples were observed during CV testing: a redox peak for low-spin Fe2+/Fe3+ at ~ 4.3/3.95 V and a redox peak for high-spin Fe2+/Fe3+ at ~ 3.6/3.4 V (in DME-based electrolytes) (Fig. 4b).
Fig. 4

a CV curves for the second cycle and b voltage profiles of K1.92Fe[Fe(CN)6]0.94·0.5H2O. Reprinted with permission from Ref. [37]. Copyright 2017 Royal Society of Chemistry. c Structure of K2Fe[Fe(CN)6]·2H2O, d its CV curves at different scan rates in aqueous electrolytes, and e EELS analysis at different reaction stages. Reprinted with permission from Ref. [38]. Copyright 2017 John Wiley and Sons. f Structural evolution of KFe[Fe(CN)6]0.82·2.87H2O and g voltage profiles for the first five cycles. Reprinted with permission from Ref. [39]. Copyright 2017 Royal Society of Chemistry

Wang et al. [38] also studied K2Fe[Fe(CN)6]·2H2O nanocubes for aqueous PIBs and reported that all Fe possessed a valence state of +2 (Fig. 4c). Here, two pairs of well-separated redox processes appeared in the CV curves in which the researchers attributed the redox peak at ~ 0.95/0.8 V to the Fe2+/Fe3+ reaction of N-coordinated Fe, and the redox peak at ~ 0.4/0.1 V to the Fe2+/Fe3+ reaction of C-coordinated Fe (Fig. 4b). Furthermore, this study revealed two obvious plateaus during the discharge process at ~ 0.8 V and 0.2 V, and that both CV results and discharge curves indicate that this cathode stores energy through double one-electron processes. Accordingly, the K2Fe[Fe(CN)6]·2H2O nanocubes reportedly delivered a relatively high discharge capacity of 140 mAh g−1 at 200 mA g−1, and even at a high current density of 3 A g−1, discharge capacities reached 93 mAh g−1. Ex situ electron energy loss spectroscopy (EELS) spectra of the Fe cations were also obtained in this study and further verified that all Fe2+ ions oxidized to Fe3+ during charging and that Fe3+ reduced to Fe2+ stepwise during discharging (Fig. 4e). Thus, the reaction mechanism of K2Fe[Fe(CN)6] ·2H2O can be represented by Eq. (2) [38]:
$$ \begin{aligned} {\text{K}}_{2} {\text{Fe}}\left[ {{\text{Fe}}\left( {\text{CN}} \right)_{6} } \right] \cdot 2{\text{H}}_{2} {\text{O}} & \leftrightarrow {\text{KFe}}\left[ {{\text{Fe}}\left( {\text{CN}} \right)_{6} } \right] \cdot 2{\text{H}}_{2} {\text{O}} + {\text{K}}^{ + } + {\text{e}}^{ - 1} \\ & \quad \leftrightarrow {\text{ Fe}}\left[ {{\text{Fe}}\left( {\text{CN}} \right)_{6} } \right] \cdot 2{\text{H}}_{2} {\text{O}} + 2{\text{K}}^{ + } + 2{\text{e}}^{ - 1} \\ \end{aligned} $$
(2)

Liu et al. [39] further investigated another cathode (KFe[Fe(CN)6]0.82·2.87H2O) based on the development of Fe2+/Fe3+ redox couples. In this cathode, the researchers reported that N-coordinated Fe was bivalent and C-coordinated Fe was trivalent, which is different from the case in classical PB (Fig. 4f) and that during the first charging, Fe2+ oxidizes to Fe3+ in a one-electron process, whereas all Fe3+ ions reduce to Fe2+ during the first discharge through double one-electron processes. During the following cycles, the reaction mechanism involves two pairs of Fe2+/Fe3+ redox. Here, the researchers regarded the first discharging as an “activation” process because higher charge capacities can be achieved afterwards, in which galvanostatic testing was conducted and revealed that the charge capacity of the second cycle was ~ 135 mAh g−1, which was higher than that in the first charge (~ 95 mAh g−1) (Fig. 4g). In addition, two charge plateaus were observed in the voltage profile after the first cycle, and researchers assigned these plateaus at ~ 3.3–3.45 V and ~ 3.7–4.1 V to the oxidation of C-coordinated Fe2+ and N-coordinated Fe2+, respectively.

Apart from K-containing cathodes, K-free Prussian green (FeIII[FeIII(CN)6]) (PG) can also be utilized as an intercalation type cathode [40]. Similar to PB, Fe3+ is also bridged by cyanide groups (–C≡N–) in PG and forms an open 3D network for K+ intercalation [40]. Additionally, during the discharge process, both low-spin (C-coordinated) and high-spin Fe3+ (N-coordinated) reduces to Fe2+ successively, corresponding to CV peaks that locate at 3.6 and 3.4 V, respectively [41]. Benefiting from these two Fe3+/Fe2+ redox couples (two one-electron processes), PG can reportedly exhibit a stable discharge capacity of 120 mAh g−1 with only slight decays during cycling [41].

Overall, Fe2+/Fe3+ redox couple-based PBA cathodes show relatively high charge/discharge plateaus and high capacities, indicating promising application in PIBs. However, Fe2+/Fe3+ redox couple (low-spin, C-coordinated) with single-electron transfer cannot provide enough capacity for PIBs. To resolve this, transforming more Fe3+ (high-spin, N-coordinated) to Fe2+ per formula unit in the initial material is reportedly an effective approach to improve capacity. And based on this, K-rich PBA cathodes, providing Fe2+/Fe3+ redox couples (high-spin and low-spin) with two-electron-transfer capabilities, can provide clearly higher capacities than PB. In addition, Prussian green can also provide two types of Fe2+/Fe3+ redox couples (low-spin and high-spin) and demonstrates comparable capacities to K-rich PBAs cathodes. Furthermore, structural modifications of PB, especially the construction of multi-type redox couples, are viable prospects to improve cathode material performances as well.

3.1.2 Manganese Redox Couple

Mn-based cathode materials have been extensively studied in LIBs and SIBs due to desirable properties such as low costs, high safety, and high theoretical capacities, as well as eco-friendly synthesis approaches [42, 43, 44]. Analogous potassium manganese compounds are also possible candidates for PIBs, and Vaalma et al. [45] were the first to report that a layered birnessite potassium manganese compound with the formula K0.3MnO2 was a possible candidate. In their study, the layered K0.3MnO2 was composed of two layered orthorhombic unit-cells with a space group of Ccmm (this type of crystalline structure can also be described as a P2-type structure exhibiting ABBA oxygen packing with transition metal ions in the octahedral sites and alkali ions in the trigonal prismatic sites) and delivered a relatively high discharge capacity of 136 mAh g−1 (0.55 K+ per unit formula) in the voltage range of 1.5~4.0 V, despite a poor capacity retention of only ~ 58% after 50 cycles. In addition, the layered K0.3MnO2 delivered a capacity of 100 mAh g−1 between 2.0 and 4.0 V (associated with 0.39 K+ per unit formula) and 73% capacity retention after 50 cycles. Furthermore, the researchers reported that cycling stabilities were enhanced if voltage windows were narrowed down to 1.5~3.5 V, but correspondingly, capacities decreased to 70 mAh g−1 (0.26 K+ per unit formula). The researchers also reported that two oxidation plateaus at 3.7 and 3.9 V appeared in the voltage window of 1.5~4.0 V, indicating the occurrence of two-phase reactions. From their observation, it can be inferred that potassium reservoirs may be undermined during charging to high voltages (extracting more potassium). Here, the researchers provided two factors: (1) upon K extraction, adjacent oxygen layers induced by high Coulombic repulsion results in the gliding of the octahedra layer; and (2) the large volume change causes partial irreversibility. This study did not provide immediate proof to support this hypothesis, however.

In another study, Kim et al. [46] successfully synthesized a new layered K0.5MnO2 cathode and interpreted its mechanism of structural evolution upon intercalation/de-intercalation of potassium. Here, XRD results revealed two phases with the space groups R3 m and Cmcm (Fig. 5a), in which the R3 m space group is the major phase and is related to a P3-type layered structure that is defined as potassium prisms and MnO6 octahedra layers stacked in the order of ABBCCA (Fig. 5b). Furthermore, CV scans were conducted between 1.5 and 4.2 V and revealed two oxidation peaks at ~ 3.7/4.1 V (Fig. 6a). This study also reported that no obvious reduction peaks could be observed because the reaction was not reversed, but at the voltage window of 1.5–3.9 V (Fig. 6c), well-matched oxidation and reduction peaks were observed. These results were in accordance with charge/discharge profiles, in which the first charge/discharge capacity of 93/140 mAh g−1 demonstrated that ~ 0.39/0.57 K+ was transported between 1.5 and 4.2 V and that subsequent capacity degradations were unacceptable (47 mAh g−1 after only 20 cycles) (Fig. 6b). On the contrary, ~ 0.22/0.44 K+ was transported between 1.5 and 3.9 V in the first charging/discharging process, which corresponds to a capacity of 53/106 mAh g−1 (Fig. 6d), revealing that lowering the cut-off charge voltage from 4.2 to 3.9 V can result in much better cycling properties in which discharge capacity retentions can increase from 30% to 76%. To further eliminate kinetic limitations, comparisons were made between the cycling stabilities of K0.5MnO2 at room temperature by performing ex situ XRD at 45 °C to determine capacity fading at high-voltage regions as a result of material crystalline structure failure. Here, XRD patterns of K0.5MnO2 after 10 cycles revealed that the majority of the phases turned either into an amorphous form or layers of MnO6 octahedra that were no longer stacked as ABBCCA. This study also investigated the potassium storage mechanisms of K0.5MnO2 using in situ XRD (Fig. 6e), in which electrode XRD patterns were sequentially collected during the charging and discharging in a customized cell. Here, it was found that upon K extraction/insertion, the (003) and (006) peaks shifted to lower/higher angles, implying c-axis expansion/contraction, respectively. The evolution of the crystalline structure upon K extraction can be described briefly as P3 → two-phase reaction → O3 → two-phase reaction → X (O3, stacking in the order of ABCABC; X, a new phase at a high state of charge as the author denoted), which was reversible upon K insertion during discharging [46].
Fig. 5

a Riedtveld-refined XRD of P3-K0.5MnO2; b illustration of the P3- K0.5MnO2 structure. Reprinted with permission from Ref. [46]. Copyright 2017 John Wiley and Sons

Fig. 6

a, c CV curves of P3-K0.5MnO2 at a scanning rate of 0.03 mV s−1 from 1.5 to 4.2 V and 1.5~3.9 V, respectively; b, d charge/discharge profiles of P3-K0.5MnO2 at 5 mA g−1 from 1.5 to 4.2 V and 1.5–3.9 V, respectively; e in situ XRD pattern taken for 2-h scanning rate per pattern upon the typical charge/discharge profile of P3-K0.5MnO2 at a current rate of 2 mA g−1 with a voltage range between 1.5 and 3.9 V. Reprinted with permission from Ref. [46]. Copyright 2017 John Wiley and Sons

To date, the existing literature on Mn-based cathode materials for PIBs is mainly based on layered compounds related to the redox couple of Mn3+/Mn4+. Here, the valence states of Mn in KxMnO2 include both trivalent and tetravalent states, and their ratio is determined by the initial K concentration and distribution (K+/vacancies ordering) between octahedral manganese slabs. Despite the improvements in electrochemical performance that have already been made, both the capacity and cycling stability of PIBs are not yet competitive enough to that of LIBs and SIBs. Furthermore, although reaction mechanisms of Mn-based cathode materials in both LIBs and SIBs are complex, notable phenomena in LIBs and SIBs, such as layered-to-spinel phase transitions and manganese dissolution due to the Jahn–Teller effect [42, 44, 47], have been well studied. Alternatively, no research progress has been made on Mn-based cathode materials in PIBs, and researchers have not explicitly determined whether the same phenomena will also have substantive effects on PIBs. Therefore, more attention is required from researchers to better understand these effects.

3.1.3 Cobalt Redox Couple

Co-based cathode materials play a significant role in the development of alkali-ion batteries with LixCoO2 being the original cathode used in early commercialized LIBs. This is because LixCoO2 possesses strong oxidizing properties as result of the Co3+/Co4+ redox pair, achieving open-circuit voltages of 4~5 V and stable operating voltages at ~ 3.7 V [12]. Because of this, pioneering research has been conducted on Co-based cathode materials for LIBs and SIBs and has motivated researchers to apply these materials to applications in PIBs [48, 49].

In one example, Ceder et al. [50] investigated P2-type K0.6CoO2 as a cathode for PIBs and reported that the crystal structure of the material possessed a hexagonal symmetry with the space group P63/mmc (Fig. 7a, b) and that an average voltage of ~ 2.7 V with an initial charge capacity of 62 mAh g−1 and a discharge capacity of 80 mAh g−1 can be obtained at 2 mA g−1 from 1.7 to 4.0 V. Here, the researchers reported that during the first charge process, 0.27 K+ can be extracted, and 0.41 K+ can be inserted back into the host in the subsequent discharge process (Fig. 7c). These results were further confirmed by the shifting of peaks observed through in situ XRD during cycling in which the (008) peak clearly shifts to a lower angle during charging, indicating expansion along the c-axis, which can be attributed mainly to the increase in Coulomb repulsion between oxygen atoms as a result of K+ extraction [51, 52, 53]. In addition, several phase transitions were also detected from the in situ XRD patterns in this study, which the researchers attributed to K+/vacancies ordering. Furthermore, the researchers refreshed the potassium metal anode and electrolyte after the cathode had been cycled for over 120 times and found that capacities recovered, implying that capacity fading may be a result of the severe side reactions of the electrolyte with the cathode or potassium metal.
Fig. 7

a XRD pattern of P2-type K0.6CoO2 with the SEM image inserted; b schematic structure of P2-type K0.6CoO2; c charge/discharge profile of P2-type K0.6CoO2 at a current rate of 2 mA g−1, 1.7~4.0 V (insets: i, derivative curve of the second cycle and ii, enlarged charge/discharge curves with the K content between 0.5 and 0.65). Reprinted with permission from Ref. [50]. Copyright 2017 John Wiley and Sons

In another example, Komaba et al. [54] synthesized P2-K0.41CoO2 (P63/mmc) and accurately demarcated the oxidation/reduction peaks and related phase transitions. Here, the reversible capacity of the P2-K0.41CoO2 was reported to be 60 mAh g−1, which is lower than that previously reported by Ceder’s group. The researchers attributed this to the lower initial content of potassium in which the content of K in the P2-K0.41CoO2 varied between 0.23 and 0.47 during cycling. In addition, Operando XRD, which shares the same working principles with in situ XRD, showed the same changes along the c-axis as those reported by Ceder’s group, whereas during potassium insertion, the a-axis (Co–Co distance) increased due to the reduction of Co4+/Co3+. Here, the reduction plateaus at 3.10, 2.59, 2.55, 2.42, and 2.33 V were assigned to a two-phase reaction and the six single-phase regions were denoted as shown in Fig. 8a, b, in which both Komaba’s and Ceder’s results confirmed that the P2 structure persisted during K insertion/extraction.
Fig. 8

a Operando XRD patterns of P2-K0.41CoO2 upon potassium intercalation of the second cycle at a low rate of 0.80 mA g−1 from 2.0 to 3.9 V; b variations of dinterslab and neighbouring Co–Co distance as a function of x in KxCoO2. In the upper panel of (b), arrows denote six voltage jumps without any apparent change of diffraction lines. Reprinted with permission from Ref. [54]. Copyright 2017 Royal Society of Chemistry

It is worth mentioning that the above studies by Ceder et al. and Komaba et al. were more focused on the K-storage mechanisms of the material itself, whereas actual better electrochemical performances of Co-based cathode materials were achieved by Wang et al. [55]. In this study, the researchers reported a similar P2-type layered hexagonal structure for K0.6CoO2, which possessed a space group P63/mmc and a unique hierarchical morphology constructed from primary nanoplates. At 40 mA g−1 within the voltage window of 1.7–4.0 V, the material produced an initial capacity of ~ 74 mAh g−1 and maintained ~ 64 mAh g−1 after 300 cycles, with only 0.04% capacity loss per cycle. The researchers attributed this to the unique structure, which significantly boosted potassium transport kinetics.

Co-based cathodes are widely used in alkali-ion batteries, but high costs and toxicity have forced researchers to find better alternatives. As a transition metal, however, Co possesses multiple valences [56] and can be used as a dopant to modify other potassium transition metal oxide cathodes, such as by replacing Jahn–Teller active Mn3+ ions with Co3+ [57].

3.1.4 Vanadium Redox Couple

Compared with other transition metal-based (such as Fe, Mn, Co etc.) materials, Vanadium-based PIB cathodes provide higher K+ insertion/extraction voltage plateaus [58, 59, 60] and are therefore promising candidates for PIBs. For example, Nikitina et al. [58] investigated K+ storage behaviours of KVPO4F and found that all charge and discharge peaks were higher than 4 V. In another study, Komaba et al. [61] synthesized KVPO4F nanoparticles and studied their electrochemical performance and found that the structure of KVPO4F, made up of VO6 octahedra and covalently bonded PO4 tetrahedra, is an open framework that is advantageous for K+ diffusion and that the VO4F2 units were connected to each other through F in VO4F2 (Fig. 9a). In this study, the researchers reported that during the charging process, K+ are extracted from the structure at voltage plateaus above 4 V and that 0.8 K+ can be extracted from KVPO4F if it was charged to 4.8 V, providing a capacity of around 105 mAh g−1. In addition, at an average voltage of 4.02 V, the material could only provide a capacity of ~ 75 mAh g−1 during discharging (Fig. 9b). Furthermore, the researchers reported that if KVPO4F was charged to higher voltages (5 V), more K+ were extracted from KVPO4F and a relatively high capacity can be achieved (Fig. 9c). Here, the researchers suggested that the charging reaction mechanism be an oxidation process, transforming V3+ to V4+. These researchers also stated that two types of K in the structure (K1 and K2) are simultaneously extracted during charging. In addition, Pyo et al. [59] selected KVP2O7 as a high-energy cathode for PIBs and reported satisfactory voltage plateaus. Here, the researchers reported that during charging, KVP2O7 oxidizes to K0.4VP2O7, and the structure changes from monoclinic to triclinic (Fig. 9d,e). The researchers here also reported that the charging mechanism involves the V3+/V4+ redox couple and that KVP2O7 is capable of delivering better electrochemical performances at 50 °C (Fig. 9f), with the average charging plateau and discharging plateau being as high as ~ 4.6 V and ~ 4.3 V, respectively (Fig. 9f). The discharge capacity was limited in this study, however, to only 60 mAh g−1 at 0.25 C.
Fig. 9

a XRD pattern and structure of KVPO4F; b and c charge–discharge curves of KVPO4F at different voltage windows. Reprinted with permission from Ref. [61]. Copyright 2017 Royal Society of Chemistry. d XRD pattern and monoclinic structure for KVP2O7; e XRD pattern and triclinic for K1-xVP2O7 (x ≈ 0.6); f charge–discharge curves of KVP2O7. Reprinted with permission from Ref. [59]. Copyright 2018 John Wiley and Sons. g K+ insertion and extraction along with phase transformation of K3V2(PO4)2F3, in which VO4F2 octahedra and PO4 tetrahedra are labelled in blue and purple, K atoms in cyan and white parts indicate vacancies; h rate performance of K3V2(PO4)2F3 and corresponding voltage profiles. Reprinted with permission from Ref. [62]. Copyright 2019 Elsevier B.V

Although the above V3+/V4+ redox couple cathodes exhibit high-voltage plateaus, limited capacities hinder further application. Therefore, to improve the capacity of V-based cathode materials, Zhang et al. [62] developed a potassium-rich active material, K3V2(PO4)2F3, in which during charging, two K+ ions are extracted from the structure, and K3V2(PO4)2F3 oxidizes to KV2(PO4)2F3, corresponding to two V3+→V4+ reactions (Fig. 9g). Based on these two redox reactions, an initial charge capacity of ~ 130 mAh g−1 and a discharge capacity of ~ 104 mAh g−1 at 10 mA g−1can be reportedly achieved (Fig. 9h). In addition, after the first cycle, the researchers reported that this material can deliver a stable discharge capacity of over 100 mAh g−1. Despite this, the discharge plateaus were still relatively low, with one being at ~ 4.3 V and the other at ~ 3.4 V.

Aside from V3+/V4+ redox couple, active materials that store energy based on V4+/V5+ redox couple have also been reported. For example, based on KVPO4F, Komaba et al. [61] synthesized KVOPO4 and applied it as a cathode material in PIBs. Here, the researchers reported that similar to the structure of KVPO4F, KVOPO4 was composed of VO6 octahedra and covalently bonded PO4 tetrahedra, in which VO6 units were connected to each other through O in VO6, and that there were two types of K in the structure (K1 and K2) (Fig. 10a). In the subsequent electrochemical testing, obtained charge/discharge curves demonstrated that the KVOPO4 cathode was capable of delivering more than one plateau during charging (Fig. 10b), and dQ/dV curves revealed two anodic/cathodic peaks at 4.21/4.13 and 4.34/4.26 V, suggesting a multi-step reaction during charging and discharging. The researchers here also found that unlike KVPO4F, K1 in KVOPO4 was likely to be extracted first from the structure, causing V4+ to oxidize to V5+. Furthermore, the results also revealed that if the charge cut-off voltage was changed from 4.8 to 5 V, both the reversible charge and the discharge are improved (Fig. 10c). In another example, Zhu et al. [63] applied K0.48V2O5, a layered oxide, as a cathode for PIBs and reported that the K+ ions in this material were located between the layers of V2O5, forming a sandwich structure (Fig. 10c), and that during the charging and discharging process, K+ intercalated and de-intercalated within this layered structure. Here, the researchers calculated that 0.16 K+ was extracted during the initial charging process, corresponding to a capacity of 20 mAh g−1. In addition, ex situ XPS analysis was carried out and further confirmed that parts of the V4+ in K0.48V2O5 were oxidized to V5+ during charging (Fig. 10d) and that this change was reversible in the following cycles. Despite these promising findings, this cathode produced a relatively low discharge plateau (a slop between 3 and 2.5 V) and a limited capacity (about 70 mAh g−1 after 80 cycles at 20 mA g−1).
Fig. 10

a XRD pattern and structure of KVOPO4; b and c charge–discharge curves of KVOPO4 at different potential windows; d structural illustration of K0.48V2O5; e ex situ XPS analysis of K0.48V2O5 at different reaction stages. Reprinted with permission from Ref. [63]. Copyright 2018 John Wiley and Sons

Overall, V-based PIB cathodes store energy based on V3+/V4+ or V4+/V5+ redox couples. Although various reported materials have shown promising charge–discharge voltage plateaus, even higher than 4 V, all these materials suffer from low capacity, making application in high-energy cathode materials difficult. However, modification of these high-voltage materials through methods such as doping with appropriate transition metals can be effective in increasing capacity.

3.2 Multi-metal Redox Couple Cathodes

Aside from the oxidizing properties of single-metal redox couples, multi-metal redox couples have also been investigated in PIBs. For example, in the PBA structure, KxMIII[FeII(CN)6], if MIII (trivalent transition ion) is replaced with MII (bivalent transition ion), the accommodation of potassium ions will increase because more charge carrier ions will be involved, resulting in higher theoretical capacities. Based on this, Ji et al. [64] replaced Fe3+ with various bivalent transition metals (Fe, Cu, Co and Ni) in PW (Prussian white) and reported that the resulting FeFe-PW (K2Fe[Fe(CN)6]) was capable of delivering more than one redox couple, at 3.55/3.26 V and 4.16/3.91 V, and that the resulting reversible capacity of ~ 110 mAh g−1 was higher than that of the classical PB value of ~ 78 mAh g−1. The researchers in this study also reported that the resulting CoFe-PW (K2Co[Fe(CN)6]), NiFe-PW (K2Ni[Fe(CN)6]), and CuFe-PW (K2Cu[Fe(CN)6]) only showed one redox couple, however, and that their electrochemical performances were far from adequate, and were even lower than that of classic PB, demonstrating the significance of selecting suitable bivalent transition metal ions towards better performance.

Recent studies have shown that Mn2+ can be a better alternative to Fe3+. For example, Lee et al. [65] synthesized a potassium-enriched Fe/Mn-based PBA with the formula of K1.6Mn[Fe(CN)6]0.96·0.27H2O and reported that two pairs of redox peaks at 4.12/3.69 V and 4.25/3.93 V appeared during CV tests, which the researchers attributed to Fe2+/Fe3+ and Mn2+/Mn3+, respectively. As a result, this Fe/Mn-based PBA reportedly delivered a higher initial capacity of ~ 125 mAh g−1 (Fig. 11a, b). In another example, Komaba et al. [66] further increased the content of potassium in their Fe/Mn-based PBA (the chemical formula of this material was determined to be K1.75Mn[Fe(CN)6]0.93·0.16 H2O) and reported that the resulting cathode exhibited a discharge capacity of ~ 130 mAh g−1 at 30 mA g−1 after 100 cycles (Fig. 11c, d). Here, the initial cathodic scan in the obtained CV curves revealed two reduction peaks at 3.92 and 3.81 V, which the researchers assigned to Mn3+/Mn2+ and Fe3+/Fe2+, respectively. The researchers also attributed the enhanced capacity to the K-rich content in the structure. To date, the highest potassium content in Fe/Mn-based PBAs was reported by Goodenough et al. [67], whose reported cathode (K1.89Mn[Fe(CN)6]0.92·0.75H2O), based on two pairs of redox couples, delivered a reversible capacity of 146.2 mAh g−1 at 0.2 C. Even at 1 C, this cathode still reportedly provided a discharge capacity of ~ 80 mAh g−1 after 100 cycles (Fig. 11e).
Fig. 11

a CV curves of the first five cycles and b charge and discharge profiles at different current densities of K1.6Mn[Fe(CN)6]0.96·0.27H2O. Reprinted with permission from Ref. [65]. Copyright 2017 John Wiley and Sons. c CV curves of the first three cycles and d cycle performance of K1.75Mn[Fe(CN)6]0.93·0.16H2O. Reprinted with permission from Ref. [66]. Copyright 2017 Royal Society of Chemistry. e Galvanostatic charge/discharge profile of a K1.89Mn[Fe(CN)6]0.92·0.75H2O cathode between 2.5 and 4.6 V at 0.2 C. Reprinted with permission from Ref. [67]. Copyright 2017 American Chemical Society

Besides Fe/Mn-based PBAs, Fe/Mn-based layered oxides have also been investigated for PIBs. For example, Mai et al. [68] investigated interconnected K0.7Mn0.5Fe0.5O2 nanowires as the cathode for PIBs, in which galvanostatic intermittent titration technique (GITT) was applied to measure the theoretical discharge capacity. This produced a capacity of 220 mAh g−1, corresponding to the insertion of ~ 1.06 K+ ions per formula unit. In addition, two discharge voltage plateaus at 2.20 and 1.88 V were observed in the study (Fig. 12a), corresponding to 0.12 and 0.11 K+ ion insertions, respectively. More detailed research on the reaction mechanisms of Fe/Mn-based layered oxides was recently published by Wang et al. [69] in which two oxidation peaks at 2.30/2.74 V and two reduction peaks at 1.87/2.45 V were observed in the lower voltage region (Fig. 12b, c), which the researchers attributed to the low-spin Mn3+/Mn4+ redox pair. As for the higher-voltage region, an oxidation peak at 4.02 V and a reduction peak at 3.80 V were observed, which were attributed to the high-spin Fe3+/Fe4+ redox pair. The researchers also noted that the charge/discharge profiles were well consistent with corresponding CV curves. An overpotential was clearly observed at the initial charge, however, which the researchers attributed to high internal stress/strain. In addition, a more sloping potential plateau was observed in the test PIB as compared with LIBs and SIBs.
Fig. 12

a CV curves of K0.7Fe0.5Mn0.5O2 nanowires. Reprinted with permission from Ref. [68]. Copyright 2017 American Chemical Society. b CV curves and c charge–discharge curves of K0.65Fe0.5Mn0.5O2 microspheres. Reprinted with permission from Ref. [62]. Copyright 2018 John Wiley and Sons. d CV curves of K0.67Ni0.17Co0.17Mn0.66O2. Reprinted with permission from Ref. [70]. Copyright 2017 Elsevier B.V

Inspired by ternary cathode materials for LIBs and SIBs, Wang et al. [70] developed a ternary cathode, K0.67Ni0.17Co0.17Mn0.66O2, and applied it in PIBs. Here, the ternary cathode material was synthesized through a solid-state reaction method and the researchers found that at calcination temperatures of 850 °C, the resulting material produced optimal electrochemical performances, presenting a capacity retention of 87% (~ 66 mAh g−1) after 100 cycles at 20 mA g−1, in which two redox couples appeared in the CV curve (Fig. 12d) at 2.31/2.53 and 3.45/3.81 V, representing the reactions of Mn3+/Mn4+ and Ni3+/Ni4+, respectively. This study did not, however, mention the redox reaction of Co, and more detailed investigations of the reaction mechanisms need to be conducted.

The introduction of multi-metal redox couples into cathode materials for LIBs and SIBs has been demonstrated to be effective in enhancing capacity. Based on existing research, however, it is still uncertain whether current cathode materials with multi-redox couples can meet PIB requirements. Despite this, they are still promising and deserve more exploration.

3.3 Other Cathodes

Besides the aforementioned K-containing materials, several K-free cathode materials have also been reported due to their unique potassium intercalation/de-intercalation mechanisms. In general, organic cathode materials demonstrate potential because they are inexpensive, eco-friendly, and possess impressive electrochemical performances. Another advantage of organic electrodes is that organic crystals usually possess larger interlayer spacings because they are held together by van der Waals forces instead of ionic or covalent bonding. The first investigation of organic cathode materials for PIBs was reported by Ji et al. [71] who investigated the K+ storage behaviours of perylene tetracarboxylic dianhydride (PTCDA). Here, the researchers reported that PTCDA was capable of delivering a reversible capacity of 131 mAh g−1 within 1.5–3.5 V and produced two discharge plateaus at around 2.4 and 2.2 V and three charge plateaus at around 2.7, 2.9, and 3.2 V in the first cycle. The researchers in this study proposed a possible reaction mechanism of PTCDA (Fig. 13a, b) in which if PTCDA was used as a cathode in the voltage window of 1.5–3.5 V, potassium enolates formed due to reactions between potassium ions and carbonyl groups (2 K+ per unit). In a subsequent study, Ji et al. [72] employed ex situ XRD and ex situ infrared spectroscopy to further understand the potassium storage mechanisms of PTCDA. Here, the researchers suggested that the organic crystals were formed by π–π aromatic stacking and therefore exhibit low solubility in non-aqueous electrolytes. Because solubility is an issue that leads to poor cyclability, Xu et al. [73] attempted to eliminate this problem by introducing a polar functional group (–SO3Na) into anthraquinone (AQ) to synthesize anthraquinone-1,5-disulphonic acid sodium salts (AQDS). In this study, the resulting active material produced two reversible redox pairs at 1.53/1.90 and 1.96/2.14 V, and provided a relatively stable capacity of 78 mAh g−1 after 100 cycles. Another notable feature of organic cathode materials is their tuneable molecular constitution, meaning that redox potentials can be tailored. For example, Ji et al. [74] artificially designed a poly (anthraquinonyl sulphide) (PAQS) and reported that the resulting cathode provided two reversible redox pairs at 1.87/2.37 and 1.45/1.94 V, with a high reversible capacity of 190 mAh g−1 (84% of PAQS’s theoretical capacity) being delivered at 20 mA g−1 between 1.5 and 3.4 V. In another example, Chen et al. [75] comprehensively investigated a class of oxocarbon salts with an adjustable organic framework (M2(CO)n, M = Li, Na, K, n = 4, 5, 6) for LIBs, SIBs, and PIBs (Fig. 13c), and in a study by Yoshida et al. [76], the operating voltages of oxocarbon salts were reported to increase from four- to five-, to six-membered ring salts. And based on the discharge/charge curves of M/M2C4O4, M/M2C5O5, and M/M2C6O6 (M = Li, Na, and K) (Fig. 13d, e, and f) obtained in the study by Chen et al. [75], it can clearly be seen that redox potentials can be tailored by adjusting structures and that in the case of the six-membered ring, four alkali metal ions instead of two can be reversibly inserted into the organic framework. Interestingly, Chen et al. also noted that if the alkali metal ion is K+, much better rate performances and cyclability can be achieved, with the resulting material exhibiting a high capacity of 164 mAh g−1 at a high current density of 10 C, which is close to 80% of the capacity at 0.2 C (212 mAh g−1). In addition, from the obtained CV curves (Fig. 13g), two redox pairs at 2.4/1.2 and 2.8/1.3 V were observed. Furthermore, Chen et al. also mentioned that organic cathode materials possess high potential in PIBs due to the generalizability of rapid potassium ion insertion kinetics. And based on this, organic cathode materials deserve more attention from researchers, despite existing challenges such as poor conductivity, thermal instability, and the possibility of dissolution in electrolytes.
Fig. 13

a Selected schematic for the proposed electrochemical reaction mechanism of PTCDA in PIBs; b CV curves of the PTCDA electrode scanned at 0.2 mV s−1 within the voltage of 1.5–3.5 V. Reprinted with permission from Ref. [71]. Copyright 2015 Elsevier B.V. c Preparations, structures, and theoretical reactions of designed oxocarbon salts; d, e, f charge/discharge profiles of K/K2C4O4, K/K2C5O5, and K/K2C6O6 at a current density of 40 mA g−1 within different voltage ranges, respectively; g CV curves of a K/K2C6O6 battery at different scan rates. Reprinted with permission from Ref. [75]. Copyright 2018 John Wiley and Sons

In addition to organic cathode materials, several interesting inorganic cathode materials have also been introduced. For example, Recham et al. [77] reported a polymorph (FeSO4F) with large and empty channels for alkali ion (Li, Na, and K) insertion reactions, in which the structure of the host consisted of chains of FeSO4F2 octahedra that were connected through their vertices by using F atoms. In this study, the researchers tested the ability of the polymorph (FeSO4F) to accommodate cations in LIBs, SIBs, and PIBs, and found that 0.9 Li+, 0.85 Na+, and 0.8 K+ can be reversibly taken in per unit formula. Furthermore, the researchers reported in this study that FeSO4F (Fig. 14a) exhibited the highest redox potential of 3.6 V in PIBs as compared with those in LIBs (3.3 V) and SIBs (3.4 V).
Fig. 14

a Electrochemical behaviour of FeSO4F:(1) vs. Li, (2) vs. Na, and (3) vs. K at a current rate of C/20. Reprinted with permission from Ref. [77]. Copyright 2012 American Chemical Society. b Schematic representation of alkali-ion insertion in amorphous electrode hosts (FeSO4). Reprinted with permission from Ref. [78]. Copyright 2014 Springer Nature

Kim et al. [78] also reported that amorphous iron phosphates are capable of storing electrical energy through insertion reactions. In their study, the researchers stated that because of short-range ordering, FePO4 amorphous hosts can accommodate various alkali ions regardless of size (Fig. 14b), resulting in well performing FePO4 cathodes in potassium half cells, delivering a capacity of 150 mAh g−1 at an operating voltage of 2.5 V. In another study, Pyo et al. [79] reported that P’3-Na0.52CrO2 is feasible as a cathode for PIBs. In this study, the researchers reported that the initial material, O3-NaCrO2, changed phase to O3-Na0.52CrO2 after the first charge, providing the material with potassium storage capabilities. Upon subsequent K insertion/extraction, the phase reversibly undergoes transitions between monophasic Na0.52CrO2, biphasic O3-Na0.52CrO2 and P3-K0.3Na0.17CrO2, delivering a capacity of 88 mAh g−1 with a discharge voltage of 2.95 V. In another example, Barpanda et al. [80] investigated P2-Na0.84CoO2, another sodium layered oxide, as a potassium ion host and reported a reversible capacity of 82 mA g−1. Here, the researchers reported that within the first 20 cycles, the initial Na0.84CoO2 fully converted to Na0.34K0.5CoO2 and demonstrated reversible potassium ion deinsertion/insertion behaviours.

TiS2, a traditional cathode material, has also been well studied in LIBs and SIBs [81, 82]. However, its electrochemical performance as cathodes for PIBs is not promising due to the large size of K+ ions, which can induce irreversible structural changes and poor kinetics. To resolve this, Tian et al. [83] reported that the performance of TiS2 can be enhanced through chemical pre-potassiation (K0.25TiS2) and that compared with bulk TiS2, pre-potassiated K0.25TiS2 possesses reduced crystal domain sizes, which modify phase transitions and allow for more facile ion insertion kinetics, leading to improved Coulombic efficiency, rate capability, and cycling stability. These improvements subsequently led to the resulting K0.25TiS2 delivering a reported initial discharge capacity of 145 mAh g−1 at 0.1 C from 1.0 to 3.0 V. Furthermore, Wang et al. [84] discovered another method of improving the electrochemical properties of TiS2 in PIBs by altering the electrolyte solvent. Here, by replacing ester-based electrolytes with ether-based electrolytes, the researchers reported that the TiS2 cathode was capable of delivering a capacity of 80 mAh g−1 at 20 C and a capacity retention of 79% after 600 cycles in a working voltage range of 1.5–3.0 V. The researchers in this study attributed this enhancement to the ether electrolyte, which can provide better kinetics for charge transfer rates and apparent K cation diffusion coefficients.

All of these materials with different potassium storage mechanisms have broadened our horizons for designing novel cathode materials, but high potassium content in initial anode materials should be developed as well.

3.4 Strategies to Improve Electrochemical Performances

For specific cathode materials, aside from the study of K-storage mechanisms, improvements to electrochemical performance is also important, and there are several strategies that have been reported with the aim of promoting cathode material performances in PIBs. One proven and effective strategy is to increase the degree of crystallinity in PBs (especially cyclability). For example, Cui et al. [85, 86] in their studies developed highly crystalline PBs for aqueous PIBs and reported enhanced performances. In one study, Cui et al. [85] replaced Fe2+ with Cu2+ to obtain a copper hexacyanoferrate (K0.71Cu[Fe(CN)6]0.72·3.7H2O) (Fig. 15a) which possessed a much higher degree of crystallinity than classic PB (Fig. 15b) and delivered an excellent cycling performance in which even after 40000 cycles, capacity retention remained at 83% with 99.7% Coulombic efficiency (Fig. 15c). In a subsequent study, Cui et al. [86] prepared nickel hexacyanoferrate (K0.6Ni1.2Fe(CN)6·3.6H2O) by replacing Fe2+ with Ni2+ (Fig. 15d) and reported that the resulting sample also demonstrated a higher degree of crystallinity than PB if the sample was prepared at 70 °C (Fig. 15e). During electrochemical testing in aqueous PIBs, this highly crystallized nickel hexacyanoferrate reportedly demonstrated no capacity loss during the first 1000 cycles, with a capacity retention of 70% over 5000 cycles. In both studies, the researchers reported that the redox couple for both copper hexacyanoferrate and nickel hexacyanoferrate was Fe3+/Fe2+.
Fig. 15

a Crystal structure of copper hexacyanoferrate and b its XRD patterns in comparisons with PB; c electrochemical performance of copper hexacyanoferrate in aqueous systems. Reprinted with permission from Ref. [85]. Copyright 2011 Springer Nature. d Crystal structure of nickel hexacyanoferrate and e XRD patterns of samples prepared at different temperatures in comparisons with PB; f electrochemical chemical performance of copper hexacyanoferrate in aqueous systems. Reprinted with permission from Ref. [86]. Copyright 2011 American Chemical Society

Another effective strategy for improving electrochemical performances in LIBs and SIBs is to use carbon materials as a coating layer or matrix for cathodes [87, 88, 89, 90]. For example, studies have shown that carbon layer coatings on the surfaces of LiFePO4 or Li4Ti5O12 can increase electron conductivity and thus improve cycling and rate performances [87, 91]. In addition, carbon can help prevent the aggregation of active materials and thus maintain structural stability. In the cases of PIB cathodes, carbon materials can also be used for the same functions [92]. For example, Zhang et al. [93] constructed free-standing reduced graphene oxides Prussian blue stainless-steel mesh electrodes (RGO@PB@SSM) for PIBs and reported that the PB nanocubes in the electrode were attached on the surface of the SSM and that the RGO film was tightly adhered to each nanocube. Here, as a result of the RGO film and SSM skeleton being in close and robust contact with the PB nanocubes, in which highly conductive SSM and RGO can enable the effective transport of electrons and avoid the agglomeration and detachment of PB nanocubes (Fig. 16a, b), the resulting RGO@PB@SSM electrode reportedly delivered superior electrochemical performances (Fig. 16c). In another example, Xu et al. [94] developed a 3D porous K3V2(PO4)3/carbon nanocomposite as a PIB cathode (Fig. 16d) and reported that in this composite, K3V2(PO4)3 nanoparticles were uniformly embedded into a carbon matrix (Fig. 16e) in which the carbon can enhance the electrical conductivity of the K3V2(PO4)3 particles and protect them from aggregation during charge/discharge. As a result, relatively good capacity retention was achieved (Fig. 16f). Furthermore, Zarbin et al. [95] synthesized flexible films of single-walled carbon nanotubes (SWCNTs) in a composite with PB (SWCNTs/PB) and multi-walled carbon nanotubes (MWCNTs)/PB composite (Fig. 16g), and utilized these composite films as the cathode for flexible PIBs. Here, the researchers reported reversible capacities of 8.3 mAh cm−3 and 2.7 mAh cm−3 for SWCNTs/PB and MWCNTs/PB, respectively.
Fig. 16

a Illustration of the structure of RGO@PB@SSM electrodes; b SEM image of RGO@PB@SSM; c rate performance of RGO@PB@SSM electrodes and reference samples. Reprinted with permission from Ref. [93]. Copyright 2017 John Wiley and Sons. d SEM image and e TEM image of K3V2(PO4)3/C nanocomposites; and f rate performance of K3V2(PO4)3/C. Reprinted with permission from Ref. [94]. Copyright 2017 Royal Society of Chemistry. g Illustration of carbon nanotube/PB thin films as cathodes

In addition to poor electronic conductivity, cathode materials also suffered from large volume expansion induced by K+ insertion/extraction. To resolve this, carbon materials, especially nanocarbons, can be utilized as buffer matrixes as well as conductivity layers. In addition, as in the cases of LIBs and SIBs, nanocarbon materials can also be used as conductive addictives in electrodes. To date, however, the function of carbon in PIB cathodes is not fully understood and will become a major research area in future.

Nanostructuring has always been considered as a preferable approach to optimize the battery properties of electrode materials. This is because at the nanoscale, distances for the transport of alkali ions within particles are shortened, facilitating faster alkali ion reaction kinetics. In addition, materials at the nanoscale possess larger surface areas which can provide more alkali ion storage sites [96, 97]. As a result, this strategy has been successfully employed to synthesize advanced and preformed cathode materials in LIBs and SIBs, and several studies have demonstrated that this strategy can be applied to cathode materials of PIBs as well. For example, Nazar et al. [98] synthesized Prussian white (K1.7Fe[Fe(CN)6]0.9) with particle sizes in the nano, submicron, and micron scales (Fig. 17a, b, and c) and found that particles of 20 nm in size delivered a near theoretical capacity of 140 mAh g−1, whereas particles of 160 nm in size delivered a lower capacity of ~ 120 mAh g−1. In contrast, micron-sized particles (1.8 μm) delivered only a limited capacity. In another example, Wang et al. [55] designed a unique hierarchical nanostructure in which microspheres were assembled from aggregated nanoparticles (Fig. 17d, e) and reported that the synthesized K0.6CoO2 demonstrated impressive cycling performances as compared with irregular K0.6CoO2, in which the capacity retention of s-K0.6CoO2 (hierarchical-structural K0.6CoO2) remained at 87% after 300 cycles at 40 mA g−1, whereas i-K0.6CoO2 (irregular K0.6CoO2) only preserved 50% of its capacity after 100 cycles (Fig. 17f). These researchers in a subsequent study [69] synthesized another cathode with the same hierarchical structure(K0.65Mn0.5Fe0.5O2) and again reported enhanced potassium storage capabilities. Mai et al. [68] in another example constructed a 3D network consisting of 1D nanowires, in which K0.65Mn0.5Fe0.5O2 nanoparticles were uniformly distributed (Fig. 17g) and reported that the stable framework built on the nanostructures provided fast potassium ion diffusion paths and high electronic conductivity, resulting in an initial capacity of 178 mAh g−1 and 125 mAh g−1 after 45 cycles at 20 mA g−1(Fig. 17h).
Fig. 17

a, b, and c TEM images of nano-, submicron-, and micron-sized K1.7Fe[Fe(CN)6]0.9, respectively. Reprinted with permission from Ref. [98]. Copyright 2017 American Chemical Society. d, e SEM images of P2-type K0.6CoO2 microspheres with high and low resolutions, respectively; f long cycle performance of s-K0.6CoO2 and i- K0.6CoO2 at a current density of 40 mA g−1. Reprinted with permission from Ref. [55]. Copyright 2018 American Chemical Society. g Schematic illustrations of interconnected K0.7Fe0.5Mn0.5O2 nanowires; h rate performance of interconnected K0.7Fe0.5Mn0.5O2 nanowires. Reprinted with permission from Ref. [68]. Copyright 2017 American Chemical Society

Despite the numerous advantages of nanostructured materials, significant drawbacks are present, however, such as: (1) higher contact areas of nanomaterials cause more severe side reactions with electrolytes, leading to the consumption of electrolytes and low Coulombic efficiencies; and (2) nanomaterials possess lower tap densities, which reduce the volumetric energy density of energy storage devices. Also, because of these drawbacks, the design of cathode materials with elaborate and functional nanostructures needs to take the above aspects into consideration.

3.5 Comparison of Main Cathode Materials

To comparatively study the electrochemical properties of main cathode materials, average working potentials, charge/discharge capacities, and cycling performances are summarized in this review and shown in Fig. 18 and Table 2. From these, it can be easily seen that V-based materials possess the highest redox potentials but relatively low capacities. Alternatively, Prussian blue and its analogues provide the best lifespans and acceptable capacities through the redox couple of Fe2+/Fe3+, but can easily contain crystal water due to low-temperature solution synthesis routes, which may decrease capacities and increase side reactions, ultimately affecting electrochemical performances. As for Mn- and Co-based materials with layered crystalline structures, although they have been widely used in LIBs and SIBs, available results in literature are unsatisfactory for their practical application in PIBs as cathode materials. And although Co-based cathode materials can exhibit higher redox potentials than Mn-based cathode materials, cobalt materials are expensive and cause environmental issues. Because of this, it is better to invest efforts into the research of Mn-based materials because of their higher theoretical capacities, easier synthesis routes, lower costs, and more eco-friendly usage. Here, it is also worth mentioning that the introduction of other redox couples such as Fe3+/Fe4+ into Mn-based redox reactions can provide better electrochemical properties, indicating ongoing positive effectiveness. Moreover, improvements in the physical properties of materials such as their conductivity and morphological structure will also be helpful in improving electrochemical performances. Most importantly, better ideas need to be taken from the research of LIBs/SIBs and applied to PIBs to demonstrate effectiveness.
Fig. 18

Comparison of reversible capacities and average voltages for reported cathode materials of potassium-ion half cells

Table 2

Summary of current materials in terms of their electrochemical performance, electrolytes, and binders

Redox

Cathode

Average discharge potential (V)

Operating voltage range (V)

Cycle number

Initial capacity/cycled capacity (discharge, mAh g−1)

Current density (mA g−1)

Electrolyte

Binder

Reference

 Single redox

         

  Fe2+/Fe3+

PB

3.75

2.6–4.2

500

78/69

8.7

1 M KBF4 in 3:7 EC/EMC

N/A

[34]

 

K0.22Fe[Fe(CN)6]0.805·4.01H2O

3.25

2–4

50

78/75

50

0.8 M KPF6 in EC/DEC

PVDF

[35]

 

K4Fe(CN)6

3.56

2–3.8

400

65.5/48.8

20

1 M KPF6 in EC/PC

PVDF

[36]

 

K1.92Fe[Fe(CN)6]0.94·0.5H2O

3.3

2–4.3

200

86/123

13

0.05 M KClO4/PC

PVDF

[37]

 

KFe[Fe(CN)6]0.82·2.87H2O

3.3

2–4.5

1000

90.7/73

100

1 M KPF6 in EC/DEC/PC

PVDF

[39]

 

K1.68Fe1.09Fe(CN)6·2.1H2O

3.25

2–4.5

100

110.5/90.4

20

0.8 M KPF6 in PC with FEC + 4% FEC

PVDF

[64]

 

K1.69Fe[Fe(CN)6]0.90·0.4H2O

3.2

2–4.6

300

118/71

100

0.5 M KPF6 in EC/DEC + 5% FEC

PVDF

[98]

 

RGO@PB@SSM

3.25

2–4

305

84/61.4

50

0.5 M KPF6 in EC/DMC + 5% FEC

Self-supported

[93]

 

FeFe(CN)6

3.2

1.5–4

100

121/117

62.5

1 M KPF6 in EC/DEC

PVDF

[41]

Mn3+/Mn4+

K0.3MnO2

2.5

1.5–3.5

685

70/40

27.9

1.5 M KFSI in EC/DMC

PVDF

[45]

 

K0.5MnO2

2.5

1.5–3.9

20

106/81

5

0.7 M KPF6 in EC/DEC

PTFE

[46]

Co3+/Co4+

K0.6CoO2

2.75

1.7–4

120

59/36

100

0.7 M KPF6 in EC/DEC

PTFE

[50]

 

K0.6CoO2

2.75

1.7–4

100

74/65

40

0.9 M KPF6 in EC/DEC

PVDF

[55]

V3+/V4+

KVPO4F

4

2–4.8

50

70(2nd cycle)/68

C/20

0.7 M KFSI in EC/DEC

PVDF

[61]

 

KVP2O7

4.4

2–5

100

60/48

0.25 C

0.5 M KPF6 in EC/DEC

PVDF

[59]

 

K3V2(PO4)2F3

3.5

2–4.6

100

101/97

10

1 M KFSI in EC/PC

CMC

[62]

V4+/V5+

KVOPO4

4.2

2–4.8

50

68/72

C/20

0.7 M KFSI in EC/DEC

PVDF

[61]

 

K0.5V2O5

2.5

1.5–3.8

80

86/72

20

1.5 M KFSI in EC/DEC

CMC

[63]

 

K3V2(PO4)3/C

3.6

2.5–4.3

100

54/52

20

0.8 KPF6 in EC/DEC

PVDF

[94]

 Multi-redox

         

  Fe and Mn

K1.6Mn[Fe(CN)6]0.96·0.27H2O

3.9

3.2–4.3

25

115/88

20

1 M KFSI PC/FEC

PVDF

[65]

 

K1.75Mn[Fe(CN)6]0.93·0.16H2O

3.75

2–4.5

100

120/130

30

0.7 M KPF6 in EC/DEC

CMC

[66]

 

K1.89Mn[Fe(CN)6]0.92·0.75H2O

3.6

2.5–4.6

100

59/85

156

saturated KClO4 in PC with 10% FEC

N/A

[67]

 

K0.7Fe0.5Mn0.5O2

1.9

1.5–4

70

114/101

100

0.8 KPF6 in EC/DMC

PVDF

[68]

 

K0.65Fe0.5Mn0.5O2

2

1.5–4.2

300

97/77

100

0.9 KPF6 in EC/DEC

PVDF

[69]

Ni, Co and Mn

K0.67Ni0.17Co0.17Mn0.66O2

2.8

2–4.3

100

76/66

20

0.8 KPF6 in EC/DEC

PVDF

[70]

 Others

         

  Co and Fe

K1.55Co0.88Fe(CN)6·3.2H2O

3.4

2–4.5

15

60/38.4

20

0.8 M KPF6 in PC with FEC + 4% FEC

PVDF

[64]

Ni and Fe

K1.51Ni1.05Fe(CN)6·3.3H2O

3.6

2–4.5

15

64.3/60

20

0.8 M KPF6 in PC with FEC + 4% FEC

PVDF

[64]

Cu and Fe

K1.40Cu0.93Fe(CN)6·4.5H2O

3.5

2–4.5

15

35.2/29

20

0.8 M KPF6 in PC with FEC + 4% FEC

PVDF

[64]

Cr4+/Cr5+

Na0.52CrO2

2.65

2–3.5

45

74/65

0.05 C

1.5 KFSI in EC/DEC

PVDF

[79]

Poly anion

Amorphous FePO4

2.1

1.5–3.5

50

160/124

4

1 M KPF6 in EC/EMC

Teflonized acetylene black

[78]

 

FeSO4F

3.5

2–5

N/A

N/A

C/20

1 M KPF6 in EC/PC/DMC

N/A

[77]

Organic

Anthraquinone-1,5-disulphonic acid sodium salt

1.7

1.4–3

100

114.9/78

13

0.8 M KPF6 in EC/EMC

Sodium alginate

[73]

 

Oxocarbon salts (K2C6O6)

1.7

1.5–3.2

100

213/101

1 C

1.25 M KPF6/DME

PVDF

[75]

 

Poly(anthraquinonyl sulphide)

1.8

1.5–3.4

50

212.5/142

20

0.5 KTFSI

PVDF

[74]

 

3,4,9,10-Perylene–tetracarboxylic acid–dianhydride-2015

2.4

1.5–3.5

200

124/90

50

0.5 M KPF6 in EC/EMC

PVDF

[71]

 

3,4,9,10-Perylene–tetracarboxylic acid–dianhydride-2016

2.3

1.2–3.2

300

130/63

10

0.8 M KPF6 in EC/EMC

PVDF

[72]

4 Other Components of PIB Electrolytes and Binders

4.1 Electrolyte

The electrolyte (salt in solvent) is a proper functioning component in electrochemical reactions and the emerging potassium ion technology is no exception as well. The electrolyte in PIBs not only plays a fundamental role in bringing about stable solid-electrolyte interface (SEI) layer formation and a suitable operation window for the depotassiation/potassiation, but also determines the PIB performance, in such aspects as specific capacity and energy density as well as cycle life. When we choose suitable electrolyte solutions for PIB application, the key features to be considered are as follows: (1) high conductivity with low viscosity; (2) fast K+ migration rates; (3) high stability with high flash points and high decomposition temperatures; (4) high electrochemical stability, (5) safety features, and (6) low costs. In terms of cathode materials, desirable electrolytes should not decompose under higher operational voltages either. In addition, K metal possesses high reactivity with electrolytes (and other cell components), affecting the performance of batteries and is regarded as the dominant reason for the observed increase in polarization for electrodes in recent studies [10, 16]. Obtaining optimized electrolyte compositions that satisfy all the above-mentioned requirements is extremely difficult; however, selective electrode optimizations based on different applications are possible. And currently, development for electrolytes of PIB cathodes has followed the footsteps of LIB and SIB development and can be divided into water-based and organic-based electrolytes, respectively [8].

4.1.1 Aqueous electrolytes in Cathode Materials

Aqueous electrolytes have been extensively investigated by researchers due to advantages such as low costs, non-flammability, and high conductivity with low internal resistances. And as early as the 1980s, Neff et al. [17, 99] found that electrodeposited PB thin films can intercalate with potassium reversibly in KCl solutions. After this, Neff and others had tried to promote the results of this study for utilization in battery electrodes, but the limited loading amounts of active material had led to limited cycling [18, 19, 20]. Recently, Cui et al. [85, 86] discovered a new KNO3 electrolyte system in which copper and nickel hexacyanoferrate (PBAs) electrodes could be cycled for thousands of times even under high current densities. Similar results have also been observed for Prussian green, a type of PBAs, demonstrating a high reversible capacity of 121.4 mAh g−1 with a stable Coulombic efficiency of 98.7% in 1 M KNO3 in a deionized (DI) water solution. Furthermore, a stable cycle life of 1000 cycles within a voltage range of 0–0.5 V has also been demonstrated for CNTs/Prussian blue in 0.1 M KCl electrolyte [95]. In another study, Park et al. [100] used TiO2/α-MoO3 as an anode and potassiated PB as a cathode to determine the kinetic parameters of potassium intercalation during the charge and discharge process in 0.5 M KPF6 in deionized water. Notably, Wang et al. [38] reported an even higher specific energy of ~ 65 W h kg−1 and a specific power of 1250 W kg−1over hundreds of cycles for PB analogues in a 0.5 M K2SO4 electrolyte. Despite these promising figures, however, considering the output energy density, aqueous electrolytes are poor choices for practical PIBs due to narrow working voltage windows (about 1 V), which are restricted by the decomposition of water because hydrogen/oxygen evolution occurs at a negative/positive electrode potential of ~ 0/1.23 V vs. the standard hydrogen electrode (SHE) [101]. In addition, gas generation can cause serious safety issues and degradation of performance. These are the main reasons why most commercial rechargeable lithium batteries use organic-based (ester or ether-type) electrolytes instead of aqueous electrolytes. Overall, water-based electrolytes can be used in aqueous PIBs if issues over water oxidation/reduction can be solved or voltage limitations become acceptable [6]. More recently, emerging “water-in-salt” electrolytes (such as potassium acetate (KAc)-based electrolytes) have been found to expand working potential windows to 3.2 V, which allows for more materials to be selected and opens more possibilities, but this requires more detailed investigations [102].

4.1.2 Non-aqueous Electrolytes in Cathode Materials

Although aqueous electrolyte systems possess several advantages, non-aqueous-based electrolytes (aprotic electrolytes) are more promising for PIBs because their operational potential windows are typically in the upper range of 2.5–4.0 V. These enlarged voltage windows not only provide significant energy/power output densities, but also allow for the reconsideration of various materials that have been intensively investigated for LIBs and SIBs. Similar to water-based electrolytes, organic electrolytes typically consist of a conducting salt dissolved in various organic solvents and the general composition of most potassium electrolytes is based on a solution of potassium salts in a mixture of two or more solvents. This is because individual solvents do not possess diverse enough configurations to satisfy the different requirements of a battery. We illustrate some parameters of commonly used solutes and solvents in organic electrolytes in Tables 3 and 4, along with corresponding abbreviations for better understanding. Unlike water-based electrolytes, however, concerns such as higher costs, capacity degradation, lower conductivity, and safety issues related to volatility, toxicity, and flammability need to be considered in the use of organic-based electrolytes for commercial PIBs. Moreover, the purity of salts and solvents, the self-discharge problems and the water sensitive characteristics of organic electrolytes need to be considered in the fundament research of PIBs [101]. In this section, various types of common non-aqueous electrolytes will be carefully reviewed.
Table 3

Critical parameters of current organic solvents in battery systems

Type

Status

Solvent

Melt point Tm (°C)

Boiling point Tb (°C)

Flash point

Tf (°C)

Viscosity η, 25 °C (cP)

Ester-based

Cyclic

Ethylene carbonate (EC)

36.4

248

160

1.90 (40 °C)

Propylene carbonate (PC)

− 48.8

242

132

2.53

Chain

Diethyl carbonate (DEC)

− 43

125–126

31

0.75

Dimethyl carbonate (DMC)

2–4

86–89

16

0.59 (20 °C)

Ethyl methyl carbonate (EMC)

− 55

107

23

0.65

Ether-based

Cyclic

1,3-Dioxolane (DOL)

− 95

75–76

− 3

0.59

Chain

Dimethoxyethane (DME)

− 58

85

− 2

0.46

Diglyme (G2)

− 64

162

57

1.88

Tetraglyme (G4)

− 30

275–276

141

3.69

Table 4

Chemical and physical properties of various potassium salts in battery systems

Potassium salt

Molecular weight (g mol−1)

Al foil corrosion

Water insensitive

Solubility, 20 °C

Potassium tetrafluoroborate (KBF4)

125.91

No

Yes

Soluble in water (negligible), ester, and ether

Potassium perchlorate (KClO4)

138.55

N/A

Yes

Insoluble in water and ether (25 °C)

Potassium hexafluorophosphate (KPF6)

184.06

No

Yes

Soluble in water (negligible, 25 °C), ester, and ether

Potassium trifluoromethanesulphonate (KCF3SO3)

188.17

N/A

No

Soluble in water

potassium bis(fluorosluphonyl)imide

(KFSI, KF2NO4S2)

219.22

Yes

(high voltage)

No

Soluble in ether and ester

Potassium bis(trifluoromethane sulphonyl) imide (KTFSI, KC2F6NO4S2)

319.14

Yes

(high voltage)

No

Soluble in water, ether, and ester

4.1.3 KPF6 in Ester-Based Solution

As indicated in Table 2, aprotic electrolytes contain EC/DEC as the solvent and KPF6 as the salt are being extensively used as cathode materials for PIBs. This is mainly due to the sophisticated awareness and technology previously developed in LIBs and SIBs, along with superior dielectric ingredients, high flash and boiling points of EC, as well as low melting point and viscosity of DEC. For layered transition metal oxide materials, 0.7 M KPF6 in EC/DEC have been found to be electrochemically active for P2-type K0.6CoO2, and multiple phase transitions have been successfully observed in this type of electrolytes [50]. In addition, similar P2-type layered K0.6CoO2 microspheres have also been reported with better capacities and reversibility, which the researchers attributed to optimized structures [55]. Furthermore, researchers have also reported that these electrolytes (0.9 M KPF6 in EC/DEC) can enhance performances because ionic conductivities increase with salt concentrations (when concentration under 1 M) [103], and similar observations can be found on organic cathode materials in PIBs [71, 72]. However, Komaba et al. [16] and Passerini et al. [45] discovered that 1 M KPF6 was not able to completely dissolve into EC/DEC, resulting in white-coloured dispersion with concentrations below 1 M. Here, the researchers attributed this to the large ionic size and weak Lewis acidity of potassium ions as compared with Li and Na ions. This is also the cause of lower interaction effects and desolvation energies in EC/DEC solvents, leading to insufficient solubilities of K-salts and limited ionic conductivities [8, 21]. And although the concentrations of KPF6 in EC/DEC can reach 1 M or higher at particular temperatures, physical properties (viscosity, conductivity, etc.) are affected, and these high temperatures are beyond the range of practical interest. Researchers have, however, reported that KPF6 can completely dissolve in EC/PC or EC/DEC/PC to form 1 M solution, which might be the result of the high dielectric constant of PC [36, 39, 62, 104].

Working potential windows are critical for the composition of electrolytes and regardless of the concentration or the type of solvent, KPF6 is able to work under a broader working potential window. For example, Komaba et al. [66] demonstrated that two high-voltage PBAs, K1.75Mn[FeII(CN)6]0.93·0.16H2O (K-MnHCFe) and K1.64Fe[FeII(CN)6]0.89·0.15H2O (K-FeHCFe), can work well in KPF6-based electrolytes in which K-MnHCFe provided an oxidation peak near 4 V, whereas the corresponding reduction peaks located at 3.81 and 3.92 V in 0.7 M KPF6 in EC/DEC electrolyte. The researchers also reported that the oxidation peak of K-FeHCFe reached 4.13 V in the identical electrolyte. Furthermore, two successive cathodic peaks that located at 3.85 and 4.38 V were observed in the KVP2O7 (polyanion type) cathode. And if the high cut-off potential was limited to 5.0 V vs. K/K+, the researchers found that these types of materials can deliver stable and continuous capacities in 0.5 M KPF6/EC/DEC [59]. These results are interesting because EC/DEC solvents are reported to undergo oxidative decomposition at more than 4.9 V vs. Li/Li+. More recently, Zhang et al. [62] reported that K3V2(PO4)2F3 cathode possesses a high average voltage of 3.7 V with a capacity of ~ 100 mAh g−1 in 1 M KPF6 in an EC/PC electrolyte and that even in K-ion full cells, K3V2(PO4)2F3 matched well with graphite anodes in this type of electrolyte and provided an assembled 3.4 V-class PIB with decent performance.

4.1.4 KFSI in Ester-Based Solution

Since the low-solubility of KPF6 has negative influences on ionic conductivity, the identification of suitable electrolytes that enable sufficient concentrations of potassium to deliver high ionic conductivity is required. Compared with KPF6, KFSI shows high solubility (1.0–1.5 M) in commonly used EC/DEC [63, 79] and even in PC/FEC solvents [65]. In the formula of FSI anions, the strong electron-withdrawing feature of the fluorosulphonyl group along with the conjugation between them and the lone electron pair on the N can allow FSI anions to be well delocalized [103]. As a result, KFSI salts dissolve exceptionally well, even in low dielectric solvents, indicating good ionic conductivity. For example, a type of layered potassium vanadate of K0.5V2O5 was recently reported by Zhu et al. [63] which exhibited a fast rate capability, excellent Coulombic efficiency and good cycling stability over 200 cycles in 1.5 M KFSI in EC/DEC electrolytes. Here, the researchers suggested that the 1st cycle Coulombic efficiency can be improved by raising the high cut-off voltage. In another example, Lee et al. [65] reported that 1 M KFSI PC/FEC electrolyte possessed the lowest overpotential and improved redox kinetics as compared with 1 M KFSI in EC/DEC and 1 M KFSI in EC/DEC/FEC. Despite these promising findings, FSI anion can corrode Al current collectors at potentials above ~ 4.0 V vs. K/K+, which is an issue in application as a high-voltage electrolyte for PIBs [105].

4.1.5 Ether-Based Electrolyte

Unlike ester-based solvents, ether-based electrolytes have received much attention, especially for sulphides and two-dimensional (2D) materials in SIBs [106]. This is mainly because of their characteristics such as low viscosity and corresponding high ionic conductivity. In addition, the formation of dendrites appears to be effectively suppressed in these solvents, even at higher current densities [103]. As a result, KTFSI/DOL/DME electrolyte exhibits much better cyclability and lower polarization than KPF6/EC/DMC [74]. Chen et al. [75] also discovered that their 1.25 M KPF6/DME electrolyte produced the highest ionic conductivity, even exceeding the conductivity of 1 M LiPF6/EC/DEC and 1.25 M NaPF6/DME electrolytes. However, these ether-based electrolytes in PIBs are still troubled by issues such as inferior capacity retention, especially during prolonged cycling under high current densities. Another issue for ether-based electrolyte is their oxidative decomposition on cathode surfaces under high voltages (about 4.0 V), whereas EC/DEC remains stable up to 4.9 V [103]. Therefore, ether-based electrolytes are not the best choice for high-voltage cathode materials.

4.1.6 Other Electrolytes and Additives

Other K-conducting salts (KClO4 and KBF4) have also been reported to possess lower solubilities in carbonated electrolytes such as EC/DEC and PC solvents [16], but there has been little reported on these salts. A recent study reported that the empirical molarity of KClO4 in PC is only ~ 0.1 M, an order of magnitude lower in concentration than those of LiClO4 and NaClO4 in PC [107], and Goodenough et al. [67] recently employed saturated KClO4 in PC with a K1.70Mn[Fe(CN)6]0.90·1.10H2O cathode material, but reported a large polarization potential range from 3.56 to 4.26 V, which the researchers attributed to the low K+ concentration in the electrolyte. For previous LIB systems, although the working temperature range of LiBF4 is higher than that of LiPF6, applications of BF4 anion salts (LiBF4) in LIBs are rarely reported because of inferior ionic conductivities. However, the superior thermal stability of KBF4 in EC/EMC systems compared with others may attract more attention to KBF4 in PIBs [34].

The widespread use of FEC additives in SIB electrolytes can be generalized to PIBs because FEC is able to form a stable SEI layer that inhibits further decomposition reactions in electrolytes [8, 108]. For example, Komba et al. [66] employed 2 Vol% FEC and reported that the initial Coulombic efficiency of K-MnHCFe electrodes in a half-cell can drastically increase to 90%. Nevertheless, regardless of whether the introduction of FEC into PIB electrolytes is sufficient enough to support stable electrochemical performances or whether it can finally lead to higher polarization of cathode electrodes, more researches are required. And in general, the main factors that are important in the search for potential additives for practical PIBs are as follows: (1) facilitate the formation of the stable SEI/CEI on the surface of both the anode and cathode; (2) improve capacity and reduces gas generation, not only in the initial cycles, but also for long-term cycle life; (3) expand the thermal stability of aprotic electrolytes; (4) reduce the polarization and overcharge of cathode materials, and (5) improve the physical properties of the electrolyte such as wettability, viscosity, ionic conductivity. In addition, additives should have similar properties to electrolytes as mentioned above and excellent additives for LIBs and SIBs are good starting points for the early stage exploration of additives for PIBs.

4.1.7 Cathode Electrolyte Interface (CEI) Composites and Structure

Similar to LIBs and SIBs, the properties of the surface or interface between the electrolyte and the cathode at high voltages remain an issue for emerging PIBs. For example, the cathode electrolyte interphase (CEI) layers of LiCoO2 in LIBs are similar to SEI layers (formed at the anode surface side) and contain both inorganic components such as lithium fluoride and organic species such as carbonates and oligomers/polymers [109, 110]. For PIBs, however, the components of the CEI layer are truly dependent on the composition of the materials and the electrolyte. And although the electrochemical performance of cathodes is influenced by CEI compositions, few reports in literature have paid much attention to this important area. Therefore, it is vital to fully understand the fundamental facets of potassium systems before any potential PIBs can be achieved. Notably, with increasing cycles or voltage increases, the thickness of CEIs will gradually change, as observed in LIBs [111, 112], and because of this, various in situ facilities, such as neutron diffraction, Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS) are worthwhile for proper investigation in future.

In short, several recent and dominant trends in the progress on electrolytes for cathode materials in PIBs have been discussed in this review, and based on these discussions, KFP6 in EC/DEC solvents can be regarded as the most promising electrolyte system for high-voltage cathode materials (> 4.0 V) in PIBs. Difficulties also exist, however, such as the stabilization of CEI layers under high operating potentials which require the application of higher-voltage electrolytes, and the substantial corrosion of other solvents during exposure to Al foil which greatly restricts possible application in high-voltage PIBs. This latter issue is especially difficult because the role of Al as a current collector is very hard to replace on a practical scale. Therefore, tremendous efforts need to be made to explore various electrolytes which can decrease reactivity towards Al during battery operations through the selection of optimal anionic structures and through the composition of different salts or solvents.

4.1.8 Electrolytes in Anode Materials

The electrochemical stability of electrolyte in batteries is usually achieved in a kinetic rather than thermodynamic manner. This is not only important to the cathode side, but is also challenged by the heavily reducing nature of the anode side. Salt dissolves in aprotic solutions through strong ion–solvent interactions, and thus, the reduction voltage of electrolytes is always related to the concentration, solvent polarity, and ion–solvent coordination. Because of this, electrolyte compositions and anode materials have strong impacts on SEI layer properties of PIBs. As a result, deeper understanding of SEIs in PIBs is required.

In one study, Ji et al. [14] investigated initial electrochemical K intercalation into commercial graphite using 0.8 M KPF6 in EC/DEC electrolytes and reported that the low initial Coulombic efficiency (ICE, 57.4%) can be attributed to the formation of the SEI on the graphite surface. The researchers stated that this is because the sloping region from 1 to 0.4 V only exists in the first charge–discharge cycle. In addition, rising Coulombic efficiency (CE) values in subsequent cycles also demonstrated that SEI layer formation occurs during the initial cycling. In another study, Hu et al. [23] conducted CV analysis of coin cells consisting of graphite and potassium metal using 0.5 M KPF6 in EC/DEC electrolytes and reported a very broad peak at around 0.75 V during the first potassiation process, which they attributed to electrolyte decomposition and SEI formation. Subsequently, Guo et al. [117] conducted a series of studies on anode materials in which they applied 0.8 M KPF6 in EC/DEC electrolytes to find different anode candidates for possible application in PIBs [113, 114, 115, 116]. Recently, our group [117] revealed for the first time the SEI composition of carbon nanocages in 1 M KFSI EC/PC electrolytes, in which well-dispersed K, F, S, and O elemental mappings coming from the SEI component layers indicate the stable nature of the electrode/electrolyte interface during cycling. Here, it was speculated that the stable SEI layer may be responsible for the excellent performance of the carbon nanocages in PIBs. Additionally, researchers have also shown that EC solvents in electrolytes have critical influences on SEI layers because EC can protect the highly crystalline structure of graphite [103]. And unlike LIBs or SIBs which use DEC as the solvent in electrolytes, Kang et al. [118] found that linear DEC is unstable for carbon-based anodes in K-ion systems, possibly due to the stronger reducibility of linear DEC against anodes in PIBs, the decomposition of DEC is initiated by the breaking of C(H2)–O bonds in the solvent molecules.

As a result of further investigations into SEI manipulation in electrode materials, continuous improvements in the electrochemical performance of alloy-based anodes through the tuning of electrolyte salts (1 M KFSI and 0.8 M KPF6) in EC/DEC solvents have been made [119]. For example, researchers have found that FSI anions from KFSI can reduce to form stable SEI to effectively protect electrolytes against decomposition and modify surface passivation, resulting in better electrochemical performances. Furthermore, studies have shown that 1 M KFSI in EC/DEC electrolytes can effectively suppress potassium dendrite growth, whereas FEC additives can accelerate potassium dendrite growth [120]. Additionally, uniform and robust SEI layers on potassium surfaces were also reported to be capable of enabling reversible potassium plating/stripping with high efficiency in KFSI DME electrolytes [121].

And as a result of these promising findings, KFSI is one of the most promising potassium salts for anode electrolytes for the fundamental research of PIBs, in which higher ionic conductivity, solubility, ionic mobility, and thermal stability can provide uniform SEI layer morphologies and stable electrochemical performances in different types of anode materials.

4.1.9 Brief Summary

For LIBs, electrolytes and additives have been intensively studied for many years, most of which have been by commercial battery manufacturers. For PIBs, however, research has not been as prolific. Therefore, systematic studies into PIB electrolytes need to be intensified. Because electrolytes act as ionic charge carriers which are necessary for electrochemical reactions to occur, electrolytes with low ionic conductivity or high viscosity tend to produce inferior electrochemical performances, especially on a practical utilization scale. Furthermore, ideal electrolytes should be resistant to degradation into side products, which will result in uncontrolled formation of interface layers that can lead to inferior transport, higher polarization, and loss of charge carriers, all of which are factors contributing to rapid cell failure. Lastly, the function of additives in electrolytes needs to be investigated closer, not only for current anode materials, but also for various promising cathode materials.

4.2 Binders

Several recent studies have demonstrated that binders can play crucial roles in maintaining electrode integrity, allowing for LIBs with excellent and stable electrochemical performances [122, 123]. This is also true for PIBs, in which aside from electrolytes, binders are an important external component that are used to paste electrode materials onto current collectors (except for free-standing materials) and must be considered as more than just something with which to hold electrode particles onto current collector. For example, PVDF is a main binder for cathode materials because of its excellent chemical and electrochemical stability for long-time use (Table 2) [3] in which it is speculated that this is mainly because of the negligible volume change of the cathode electrode. However, most recent studies have focused on water-based binders such as Na-CMC [63] and sodium alginate [73] because the processing of slurries by using PVDF requires expensive and toxic N-methyl pyrrolidone (NMP). In addition, the observed defluorination from PVDF if polymers are attacked by alkali hydroxides has negative influences on the capacity and the battery lifetime. Therefore, in considering binder utilization for electrodes, certain factors must be taken into account: (1) binders should have good adhesion to electrodes and current collectors; (2) binders should be able to entirely coat the surface of electrodes to inhibit the formation of SEI layers and simultaneously facilitate K+ ion conduction; (3) binders should be mechanically strong to tolerate huge volume changes; and (4) binders should be stable in electrolytes. Therefore, based on these factors, carboxylic group containing binders (CMC, PAA) should be good choices, because they can efficiently improve adhesion between electrodes and current collectors in water solutions [124]. For example, Komaba et al. [16] recently evaluated the electrochemical performance of graphite anodes with different types of binders and reported that the introduction of Na-PAA and Na-CMC can reduce the consumption of potassium for the formation of SEI layers, leading to high initial Coulombic efficiencies.

From our own perspective, in the cases of cathode materials, the use of binders such as PVDF or PTFE is ideal, whereas CMC, PAA, and sodium alginate can be further considered for anode-type materials. However, the field of polymer chemistry is abundant with different polymers, some of which conductive, that have yet to be tested as binders for electrode materials, making this an interesting field for researchers to explore in future.

5 Potassium-ion Full Cells

Although there has been much progress in the research of cathode materials for PIBs, it is vital to assess their practicabilities. For this purpose, materials should be assembled and tested in full cells for evaluation. However, the field of potassium-ion full cells is still in its infancy, and until now, there have only been a few reports in the literature on full cell performances of developed cathode materials (performance summaries and conclusions shown in Fig. 19 and Table 5). As for anode materials, the most promising materials are carbon-based materials, especially graphite, which have been extensively applied as commercial anodes in LIBs and can be used in PIBs. As for binders and electrolytes in PIBs, optimizations are required, with KPF6/carbonate-based solvent electrolytes being the most used for potassium-ion full cells. Furthermore, Yan et al. [125] reported excellent cycling stabilities in potassium-ion full cells using Prussian blue as the cathode, in which their designed soft package cells reportedly delivered an initial discharge capacity of 80 mAh g−1 (based on the cathode) with no capacity fading after 60 cycles at a current density of 50 mA g−1. To date, the highest energy density reported in PIBs is ~ 385 Wh kg−1(calculated based on the active cathode material) [66]. And although these results are promising, they are far from adequate to meet the basic requirements of practical application. Therefore, further exploration is required to develop cathode materials with higher capacities, higher redox potentials, and longer lifespans for application in practical PIBs.
Fig. 19

Comparison of the electrochemical performances of reported full cells and energy density curves; calculations were based on the mass of active cathode materials only

Table 5

Summary of reported full cells in the literature

Cathode

Anode

Potential window (V)

Average discharge potential (V)

Cycle number

Initial capacity/cycled capacity (discharge, mAh g−1)

Current density (mA g−1)

Electrolyte

Reference

K0.3MnO2

Hard carbon/carbon black

0.5–3.5

2.05

100

90/45

32

1.5 M KFSI in EC/DMC

[45]

K2C6O6

K4C6O6

0.5–2

1.15

10

70

25

1.25 M KPF6 in DME

[75]

K0.7Fe0.5Mn0.5O2

Soft carbon

0.5–3.5

1.6

50

82/73

40

0.8 M KPF6 in EC/DMC

[68]

K0.22Fe[Fe(CN)6]0.805·4.01H2O

Super P

1–3.8

2.2

50

73/65

100

0.8 M KPF6 in EC/DMC

[35]

K1.75Mn[Fe(CN)6]0.93·0.16H2O

Graphite

1.5–4.5

3.5

60

110/80

30

0.7 M KPF6 in EC/DMC

[66]

K0.6CoO2

Graphite

0.5–3.8

2.5

5

53/25

3

0.7 M KPF6 in EC/DMC

[50]

K1.92Fe[Fe(CN)6]0.94·0.5H2O

K2TP

1.5–3.8

3.3/2.7

60

110/99

65

1 M KPF6 in DME

[37]

K0.61Fe[Fe(CN)6]0.92·0.32H2O

Graphite

2.0–4.0

2.9

60

80/80

50

0.8 M KPF6 in PC

[125]

K0.65Fe0.5Mn0.5O2

Hard carbon

0.5–3.5

2

100

76/60

100

0.9 M KPF6 in EC/DMC

[69]

K3V2(PO4)2F3

Graphite

1.5–4.6

3.4

50

84/59

10

1 M KPF6 in EC/PC

[62]

K0.6CoO2

Hard carbon

0.5–3.8

1.9

100

72/57

30

0.9 M KPF6 in EC/DMC

[55]

6 Summary and Outlook

Interest in potassium-ion batteries has grown considerably in recent years, and although the PIBs are unlikely to completely replace LIBs in portable electronics or electric vehicles in which high energy density is a key requirement, they can compete with LIBs in stationary energy storage applications due to the low cost and abundance of K resources. In addition, due to similarities between PIBs and LIBs/SIBs, knowledge accumulated in LIB/SIB research can be applied to the investigation of PIB. Due to the large ionic radius of K+ ions, however, most straightforward and successful approaches applied to LIBs/SIBs may not be suitable for PIBs because many specific details such as structures and K+ intercalation mechanisms are different. Therefore, detailed investigations into PIBs are needed. Considering the successful intercalation of K+ into graphite, the main challenge is to develop suitable cathode materials that can accommodate complex electrochemical behaviours and structural evolutions which occur during reversible potassiation and depotassiation processes.

Energy density is another key parameter for developing PIBs and is determined by the working voltage of electrodes and their specific capacities. The voltage plateaus of cathode materials are mainly determined by cation redox couples. And in the case of materials based on the same redox couples, they possess similar voltage plateaus in which specific capacities vary due to different crystal structures and molecular weights. In this review, reported PIB cathodes based on different redox couples have been summarized, in which PBAs based on the Fe2+/Fe3+ redox couple produce similar charge/discharge plateaus at ~ 3.4 V, and most show low capacities. These specific capacities can be increased, however, by increasing the content of Fe2+ per formula unit, because more electrons can be involved in the charge/discharge process. Therefore, K-rich PBAs providing two types of Fe2+/Fe3+ redox couples (low spin and high spin) can offer higher capacities. The presence of crystal water in PBAs, however, can lead to occurrences of side reactions and negatively affect electrochemical performances. As for safety concerns, thermal stability during charge and discharge also need to be investigated. Furthermore, the density of PBAs is normally lower compared with layered oxides, which is another disadvantage in terms of volumetric energy density. In addition, the application of Co3+/Co4+ and Mn3+/Mn4+ in layered oxides leads to limited capacities and a relatively low voltage plateau, which is possibly due to the fact that all reported K-containing layered oxides possess K-deficient compositions (x < 0.7 in KxMO2). Because of this, developing fully potassiated or K-enriched oxides is highly desirable to increase capacities, but may be difficult because layered structures do not have enough sites or spaces to accommodate large numbers of large K+ ions. As for V-based PIB cathodes, these materials store energy based on V3+/V4+ or V4+/V5+ redox couples and have demonstrated attractive voltage plateaus, even higher than 4 V. Most V-based cathodes are polyanionic materials, however, and in general suffer from low capacities and poor electronic conductivities. Clearly, most reported cathode materials possess advantages and disadvantages and we should apply advantages to the needs of specific applications. As for disadvantages, they should be overcome through material innovations.

To achieve higher specific energies, operating potentials or specific capacities of cathode materials must be increased and one possible method is through cationic or anionic substitution. Therefore, introducing multi-metal redox reactions is an effective method to enhance capacities or increase working potentials. For example, substituting Fe3+ with various divalent transition metals (e.g. Fe, Cu, Co, Ni, and Mn) in PBAs can lead to higher specific capacities because more K+ ions can be accommodated. Furthermore, to enhance the electrochemical performance of cathode materials in terms of energy density, cycling life, and rate capability, material design strategies such as increasing crystallinity, designing hierarchical nanostructures, and using carbon as matrixes or for conductivity have all been employed. Here, increasing crystallinity may be not suitable in all cases, however, because amorphous FePO4 structures have been reported to accommodate K+ ions very well as well. In addition, the introduction of nanostructures can facilitate ionic transportation, but may cause low tap density and high reactivity with electrolytes. Furthermore, using carbon may increase conductivities and buffer aggregation of materials, but the content and distribution of carbon need to be optimized. Therefore, more detailed investigations are required to explore the relationship between electrochemical performance and structural evolution. Moreover, in addition to excellent electrochemical performances, material abundance, costs, and eco-friendliness are also important and “winning” materials will be those that can not only exhibit satisfactory performance, but also be produced in large-scale, cost-effective, and clean processes.

In addition to the investigation of active materials, other components such as electrolytes, electrolyte additives, and binders also need to be optimized towards the development of practical PIBs. Similar to existing LIB and SIB systems, ideal electrolytes for PIBs must offer certain characteristics, including low viscosity, high ionic conductivity, wide potential window, good thermal stability, and low toxicity. They should also be conducive to uniform interface layers between electrodes and electrolytes, and long and stable voltage outputs. In addition to experimental investigations, theoretical simulations and calculations need to be conducted to search for new types of electrolytes for high-voltage PIBs, and excellent binders need to be found for both electrodes to maintain integrity and promote long cycle lifespans.

An additional challenge to PIBs is whether strategies suitable for half cells will work in full cells. Although half-cell setups are suitable for initial research because they allow researchers to focus on intrinsic properties of materials, for practical application, materials should be demonstrated in full cells. This is because experiments with full cells can help collect knowledge not only for materials engineering and a choice of electrolytes and binders, but also for achieving better understandings of cost-effective and practical PIBs.

In this review, the most recent progress on cathode materials, electrolytes, and binders for PIBs have been carefully summarized and their potential, challenges, and opportunities have been discussed. And although the research on PIBs is in its initial stages, some prototypes have already exhibited good performance, indicating potentials for practical use. The development of PIBs relies on similar chemistry to LIBs and uses the same commercial production lines, translating to high manufacturability and minimal investment. Future research should target not only energy density and cycle life, but also safety and other parameters to meet the requirements for various applications, such as stationary energy storage systems.

Notes

Acknowledgements

Financial support provided by the Australian Research Council (ARC) (FT150100109 and DP170102406) is gratefully acknowledged. Q.Z. and Z.J.W. acknowledge the China Scholarship Council (CSC) for their scholarships (Grant Nos. 201508420150 and 201706340049). The authors would also like to thank Dr. Tania Silver for performing critical revisions of the manuscript.

Compliance with ethical standards

Conflict of interest

The authors declare no conflict of interest.

References

  1. 1.
    Tarascon, J.M., Armand, M.: Issues and challenges facing rechargeable lithium batteries. Nature 414, 359 (2001).  https://doi.org/10.1038/35104644 CrossRefGoogle Scholar
  2. 2.
    Carmichael, R.S.: Physical Properties of Rocks and Minerals. CRC Press, Boca Raton (1989)Google Scholar
  3. 3.
    Yabuuchi, N., Kubota, K., Dahbi, M., et al.: Research development on sodium-ion batteries. Chem. Rev. 114, 11636–11682 (2014).  https://doi.org/10.1021/cr500192f CrossRefGoogle Scholar
  4. 4.
    Hwang, J.Y., Myung, S.T., Sun, Y.K.: Sodium-ion batteries: present and future. Chem. Soc. Rev. 46, 3529–3614 (2017).  https://doi.org/10.1039/c6cs00776g CrossRefPubMedGoogle Scholar
  5. 5.
    Mao, J., Zhou, T., Zheng, Y., et al.: Two-dimensional nanostructures for sodium-ion battery anodes. J. Mater. Chem. A 6, 3284–3303 (2018).  https://doi.org/10.1039/c7ta10500b CrossRefGoogle Scholar
  6. 6.
    Pramudita, J.C., Sehrawat, D., Goonetilleke, D., et al.: An initial review of the status of electrode materials for potassium-ion batteries. Adv. Energy Mater. 7, 1602911 (2017).  https://doi.org/10.1002/aenm.201602911 CrossRefGoogle Scholar
  7. 7.
    Zou, X., Xiong, P., Zhao, J., et al.: Recent research progress in non-aqueous potassium-ion batteries. Phys. Chem. Chem. Phys. 19, 26495–26506 (2017).  https://doi.org/10.1039/c7cp03852f CrossRefPubMedGoogle Scholar
  8. 8.
    Wu, X., Leonard, D.P., Ji, X.: Emerging non-aqueous potassium-ion batteries: challenges and opportunities. Chem. Mater. 29, 5031–5042 (2017).  https://doi.org/10.1021/acs.chemmater.7b01764 CrossRefGoogle Scholar
  9. 9.
    Kubota, K., Dahbi, M., Hosaka, T., et al.: Towards K-ion and Na-ion batteries as “beyond Li-ion”. Chem. Rec. 18, 459–479 (2018).  https://doi.org/10.1002/tcr.201700057 CrossRefPubMedGoogle Scholar
  10. 10.
    Eftekhari, A., Jian, Z., Ji, X.: Potassium secondary batteries. ACS Appl. Mater. Interfaces 9, 4404–4419 (2017).  https://doi.org/10.1021/acsami.6b07989 CrossRefPubMedGoogle Scholar
  11. 11.
    Kim, H., Kim, J.C., Bianchini, M., et al.: Recent progress and perspective in electrode materials for K-ion batteries. Adv. Energy Mater. 8, 1702384 (2018).  https://doi.org/10.1002/aenm.201702384 CrossRefGoogle Scholar
  12. 12.
    Ge, P., Fouletier, M.: Electrochemical intercalation of sodium in graphite. Solid State Ionics 28–30, 1172–1175 (1988).  https://doi.org/10.1016/0167-2738(88)90351-7 CrossRefGoogle Scholar
  13. 13.
    Stevens, D.A., Dahn, J.R.: The mechanisms of lithium and sodium insertion in carbon materials. J. Electrochem. Soc. 148, A803–A811 (2001).  https://doi.org/10.1149/1.1379565 CrossRefGoogle Scholar
  14. 14.
    Jian, Z., Luo, W., Ji, X.: Carbon electrodes for K-ion batteries. J. Am. Chem. Soc. 137, 11566–11569 (2015).  https://doi.org/10.1021/jacs.5b06809 CrossRefPubMedGoogle Scholar
  15. 15.
    Marcus, Y.: Thermodynamic functions of transfer of single ions from water to nonaqueous and mixed solvents: part 3-standard potentials of selected electrodes. Pure Appl. Chem. 57, 1129–1132 (2009).  https://doi.org/10.1351/pac198557081129 CrossRefGoogle Scholar
  16. 16.
    Komaba, S., Hasegawa, T., Dahbi, M., et al.: Potassium intercalation into graphite to realize high-voltage/high-power potassium-ion batteries and potassium-ion capacitors. Electrochem. Commun. 60, 172–175 (2015).  https://doi.org/10.1016/j.elecom.2015.09.002 CrossRefGoogle Scholar
  17. 17.
    McCargar, J.W., Neff, V.D.: Thermodynamics of mixed-valence intercalation reactions: the electrochemical reduction of Prussian blue. J. Phys. Chem. 92, 3598–3604 (1988).  https://doi.org/10.1021/j100323a055 CrossRefGoogle Scholar
  18. 18.
    Neff, V.D.: Some performance characteristics of a Prussian blue battery. J. Electrochem. Soc. 132, 1382–1384 (1985).  https://doi.org/10.1149/1.2114121 CrossRefGoogle Scholar
  19. 19.
    Honda, K., Hayashi, H.: Prussian blue containing nafion composite film as rechargeable battery. J. Electrochem. Soc. 134, 1330–1334 (1987).  https://doi.org/10.1149/1.2100668 CrossRefGoogle Scholar
  20. 20.
    Kaneko, M., Okada, T.: A secondary battery composed of multilayer Prussian Blue and its reaction characteristics. J. Electroanal. Chem. Interfacial Electrochem. 255, 45–52 (1988).  https://doi.org/10.1016/0022-0728(88)80003-2 CrossRefGoogle Scholar
  21. 21.
    Okoshi, M., Yamada, Y., Komaba, S., et al.: Theoretical analysis of interactions between potassium ions and organic electrolyte solvents: acomparison with lithium, sodium, and magnesium ions. J. Electrochem. Soc. 164, A54–A60 (2017).  https://doi.org/10.1149/2.0211702jes CrossRefGoogle Scholar
  22. 22.
    Jian, Z., Xing, Z., Bommier, C., et al.: Hard carbon microspheres: potassium-ion anode versus sodium-ion anode. Adv. Energy Mater. 6, 1501874 (2016).  https://doi.org/10.1002/aenm.201501874 CrossRefGoogle Scholar
  23. 23.
    Luo, W., Wan, J., Ozdemir, B., et al.: Potassium ion batteries with graphitic materials. Nano Lett. 15, 7671–7677 (2015).  https://doi.org/10.1021/acs.nanolett.5b03667 CrossRefPubMedGoogle Scholar
  24. 24.
    Zhang, W., Mao, J., Li, S., et al.: Phosphorus-based alloy materials for advanced potassium-ion battery anode. J. Am. Chem. Soc. 139, 3316–3319 (2017).  https://doi.org/10.1021/jacs.6b12185 CrossRefPubMedGoogle Scholar
  25. 25.
    Goodenough, J.B., Kim, Y.: Challenges for rechargeable Li batteries. Chem. Mater. 22, 587–603 (2010).  https://doi.org/10.1021/cm901452z CrossRefGoogle Scholar
  26. 26.
    Hautier, G., Jain, A., Ong, S.P., et al.: Phosphates as lithium-ion battery cathodes: an evaluation based on high-throughputab initio calculations. Chem. Mater. 23, 3495–3508 (2011).  https://doi.org/10.1021/cm200949v CrossRefGoogle Scholar
  27. 27.
    Liu, C., Neale, Z.G., Cao, G.: Understanding electrochemical potentials of cathode materials in rechargeable batteries. Mater. Today 19, 109–123 (2016).  https://doi.org/10.1016/j.mattod.2015.10.009 CrossRefGoogle Scholar
  28. 28.
    Masquelier, C., Croguennec, L.: Polyanionic (phosphates, silicates, sulfates) frameworks as electrode materials for rechargeable Li (or Na) batteries. Chem. Rev. 113, 6552–6591 (2013).  https://doi.org/10.1021/cr3001862 CrossRefGoogle Scholar
  29. 29.
    Yamada, A., Chung, S.C., Hinokuma, K.: Optimized LiFePO4 for lithium battery cathodes. J. Electrochem. Soc. 148, A224–A229 (2001).  https://doi.org/10.1149/1.1348257 CrossRefGoogle Scholar
  30. 30.
    Wu, X.Y., Sun, M.Y., Shen, Y.F., et al.: Energetic aqueous rechargeable sodium-ion battery based on Na2CuFe(CN)6-NaTi2(PO4)3 intercalation chemistry. Chemsuschem 7, 407–411 (2014).  https://doi.org/10.1002/cssc.201301036 CrossRefPubMedGoogle Scholar
  31. 31.
    Wu, X., Cao, Y., Ai, X., et al.: A low-cost and environmentally benign aqueous rechargeable sodium-ion battery based on NaTi2(PO4)3-Na2NiFe(CN)6 intercalation chemistry. Electrochem. Commun. 31, 145–148 (2013).  https://doi.org/10.1016/j.elecom.2013.03.013 CrossRefGoogle Scholar
  32. 32.
    Lee, H., Kim, Y.I., Park, J.K., et al.: Sodium zinc hexacyanoferrate with a well-defined open framework as a positive electrode for sodium ion batteries. Chem. Commun. 48, 8416–8418 (2012).  https://doi.org/10.1039/c2cc33771a CrossRefGoogle Scholar
  33. 33.
    Yabuuchi, N., Yoshida, H., Komaba, S.: Crystal structures and electrode performance of alpha-NaFeO2 for rechargeable sodium batteries. Electrochemistry 80, 716–719 (2012).  https://doi.org/10.5796/electrochemistry.80.716 CrossRefGoogle Scholar
  34. 34.
    Eftekhari, A.: Potassium secondary cell based on Prussian blue cathode. J. Power Sour 126, 221–228 (2004).  https://doi.org/10.1016/j.jpowsour.2003.08.007 CrossRefGoogle Scholar
  35. 35.
    Zhang, C., Xu, Y., Zhou, M., et al.: Potassium prussian blue nanoparticles: a low-cost cathode material for potassium-ion batteries. Adv. Func. Mater. 27, 1604307 (2017).  https://doi.org/10.1002/adfm.201604307 CrossRefGoogle Scholar
  36. 36.
    Pei, Y., Mu, C., Li, H., et al.: Low-cost K4Fe(CN)6 as a high-voltage cathode for potassium-ion batteries. Chemsuschem 11, 1285–1289 (2018).  https://doi.org/10.1002/cssc.201800057 CrossRefPubMedGoogle Scholar
  37. 37.
    Liao, J., Hu, Q., Yu, Y., et al.: A potassium-rich iron hexacyanoferrate/dipotassium terephthalate@carbon nanotube composite used for K-ion full-cells with an optimized electrolyte. J. Mater. Chem. A 5, 19017–19024 (2017).  https://doi.org/10.1039/c7ta05460b CrossRefGoogle Scholar
  38. 38.
    Su, D., McDonagh, A., Qiao, S.Z., et al.: High-capacity aqueous potassium-ion batteries for Large-Scale Energy Storage. Adv. Mater. 29, 1604007 (2017).  https://doi.org/10.1002/adma.201604007 CrossRefGoogle Scholar
  39. 39.
    Chong, S., Chen, Y., Zheng, Y., et al.: Potassium ferrous ferricyanide nanoparticles as a high capacity and ultralong life cathode material for nonaqueous potassium-ion batteries. J. Mater. Chem. A 5, 22465–22471 (2017).  https://doi.org/10.1039/c7ta08139a CrossRefGoogle Scholar
  40. 40.
    Padigi, P., Thiebes, J., Swan, M., et al.: Prussian green: a high rate capacity cathode for potassium ion batteries. Electrochim. Acta 166, 32–39 (2015).  https://doi.org/10.1016/j.electacta.2015.03.084 CrossRefGoogle Scholar
  41. 41.
    Shadike, Z., Shi, D.R., Wang, T., et al.: Long life and high-rate Berlin green FeFe(CN)6 cathode material for a non-aqueous potassium-ion battery. J. Mater. Chem. A 5, 6393–6398 (2017).  https://doi.org/10.1039/c7ta00484b CrossRefGoogle Scholar
  42. 42.
    Bruce, P., Armstrong, A., Gitzendanner, R.: New intercalation compounds for lithium batteries: layered LiMnO2. J. Mater. Chem. 9, 193–198 (1999)CrossRefGoogle Scholar
  43. 43.
    Billaud, J., Clement, R.J., Armstrong, A.R., et al.: beta-NaMnO2: a high-performance cathode for sodium-ion batteries. J. Am. Chem. Soc. 136, 17243–17248 (2014).  https://doi.org/10.1021/ja509704t CrossRefPubMedGoogle Scholar
  44. 44.
    Ma, X., Chen, H., Ceder, G.: Electrochemical properties of monoclinic NaMnO2. J. Electro. Soc. 158, A1307–A1312 (2011).  https://doi.org/10.1149/2.035112jes CrossRefGoogle Scholar
  45. 45.
    Vaalma, C., Giffin, G.A., Buchholz, D., et al.: Non-aqueous K-ion battery based on layered K0.3MnO2 and hard carbon/carbon black. J. Electrochem. Soc. 163, A1295–A1299 (2016).  https://doi.org/10.1149/2.0921607jes CrossRefGoogle Scholar
  46. 46.
    Kim, H., Seo, D.H., Kim, J.C., et al.: Investigation of potassium storage in layered P3-Type K0.5MnO2 Cathode. Adv. Mater. 29, 1702480 (2017).  https://doi.org/10.1002/adma.201702480 CrossRefGoogle Scholar
  47. 47.
    Reed, J., Ceder, G., Van Der Ven, A.: Layered-to-spinel phase transition in LixMnO2 electrochem. Solid-State Lett. 4, A78–A81 (2001).  https://doi.org/10.1149/1.1368896 CrossRefGoogle Scholar
  48. 48.
    Ding, J.J., Zhou, Y.N., Sun, Q., et al.: Electrochemical properties of P2-phase Na0.74CoO2 compounds as cathode material for rechargeable sodium-ion batteries. Electrochim. Acta 87, 388–393 (2013).  https://doi.org/10.1016/j.electacta.2012.09.058 CrossRefGoogle Scholar
  49. 49.
    Rai, A.K., Anh, L.T., Gim, J., et al.: Electrochemical properties of NaxCoO2 (x ~ 0.71) cathode for rechargeable sodium-ion batteries. Ceram. Inter. 40, 2411–2417 (2014).  https://doi.org/10.1016/j.ceramint.2013.08.013 CrossRefGoogle Scholar
  50. 50.
    Kim, H., Kim, J.C., Bo, S.H., et al.: K-ion batteries based on a P2-type K0.6CoO2 cathode. Adv. Energy Mater. 7, 1700098 (2017).  https://doi.org/10.1002/aenm.201700098 CrossRefGoogle Scholar
  51. 51.
    Komaba, S., Yabuuchi, N., Nakayama, T., et al.: Study on the reversible electrode reaction of Na1-xNi0.5Mn0.5O2 for a rechargeable sodium-ion battery. Inorg. Chem. 51, 6211–6220 (2012).  https://doi.org/10.1021/ic300357d CrossRefPubMedGoogle Scholar
  52. 52.
    Berthelot, R., Carlier, D., Delmas, C.: Electrochemical investigation of the P2-NaxCoO2 phase diagram. Nat. Mater. 10, 74–80 (2011).  https://doi.org/10.1038/nmat2920 CrossRefPubMedGoogle Scholar
  53. 53.
    Croguennec, L., Pouillerie, C., Delmas, C.: NiO2 obtained by electrochemical lithium deintercalation from lithium nickelate: structural modifications. J. Electrochem. Soc. 147, 1314–1321 (2000).  https://doi.org/10.1149/1.1393356 CrossRefGoogle Scholar
  54. 54.
    Hironaka, Y., Kubota, K., Komaba, S.: P2- and P3-KxCoO2 as an electrochemical potassium intercalation host. Chem. Commun. 53, 3693–3696 (2017).  https://doi.org/10.1039/c7cc00806f CrossRefGoogle Scholar
  55. 55.
    Deng, T., Fan, X., Luo, C., et al.: Self-templated formation of P2-type K0.6CoO2 microspheres for high reversible potassium-ion batteries. Nano Lett. 18, 1522–1529 (2018).  https://doi.org/10.1021/acs.nanolett.7b05324 CrossRefPubMedGoogle Scholar
  56. 56.
    Mandal, S., Rojas, R.M., Amarilla, J.M., et al.: High temperature Co-doped LiMn2O4-based spinels. Structural, electrical, and electrochemical characterization. Chem. Mater. 14, 1598–1605 (2002).  https://doi.org/10.1021/cm011219v CrossRefGoogle Scholar
  57. 57.
    Armstrong, A.R., Robertson, A.D., Bruce, P.G.: Structural transformation on cycling layered Li (Mn1-yCoy)O2 cathode materials. Electrochim. Acta 45, 285–294 (1999).  https://doi.org/10.1016/S0013-4686(99)00211-X CrossRefGoogle Scholar
  58. 58.
    Nikitina, V.A., Kuzovchikov, S.M., Fedotov, S.S., et al.: Effect of the electrode/electrolyte interface structure on the potassium-ion diffusional and charge transfer rates: towards a high voltage potassium-ion battery. Electrochim. Acta 258, 814–824 (2017).  https://doi.org/10.1016/j.electacta.2017.11.131 CrossRefGoogle Scholar
  59. 59.
    Park, W.B., Han, S.C., Park, C., et al.: KVP2O7 as a robust high-energy cathode for potassium-ion batteries: pinpointed by a full screening of the inorganic registry under specific search conditions. Adv. Energy Mater. 8, 1703099 (2018).  https://doi.org/10.1002/aenm.201703099 CrossRefGoogle Scholar
  60. 60.
    Lian, R., Wang, D., Ming, X., et al.: Phase transformation, ionic diffusion, and charge transfer mechanisms of KVOPO4 in potassium ion batteries: first-principles calculations. J. Mater. Chem. A 6, 16228–16234 (2018).  https://doi.org/10.1039/c8ta06708b CrossRefGoogle Scholar
  61. 61.
    Chihara, K., Katogi, A., Kubota, K., et al.: KVPO4F and KVOPO4 toward 4 volt-class potassium-ion batteries. Chem. Commun. 53, 5208–5211 (2017).  https://doi.org/10.1039/c6cc10280h CrossRefGoogle Scholar
  62. 62.
    Lin, X., Huang, J., Tan, H., et al.: K3V2(PO4)2F3 as a robust cathode for potassium-ion batteries. Energy Storage Mater. 16, 97–101 (2019).  https://doi.org/10.1016/j.ensm.2018.04.026 CrossRefGoogle Scholar
  63. 63.
    Deng, L., Niu, X., Ma, G., et al.: Layered potassium vanadate K0.5V2O5 as a cathode material for nonaqueous potassium ion batteries. Adv. Fun. Mater 1800670 (2018).  https://doi.org/10.1002/adfm.201800670
  64. 64.
    Wu, X., Jian, Z., Li, Z., et al.: Prussian white analogues as promising cathode for non-aqueous potassium-ion batteries. Electrochem. Commun. 77, 54–57 (2017).  https://doi.org/10.1016/j.elecom.2017.02.012 CrossRefGoogle Scholar
  65. 65.
    Jiang, X., Zhang, T., Yang, L., et al.: A Fe/Mn-Based Prussian blue analogue as a K-rich cathode material for potassium-ion batteries. ChemElectroChem 4, 2237–2242 (2017).  https://doi.org/10.1002/celc.201700410 CrossRefGoogle Scholar
  66. 66.
    Bie, X., Kubota, K., Hosaka, T., et al.: A novel K-ion battery: hexacyanoferrate(ii)/graphite cell. J. Mater. Chem. A 5, 4325–4330 (2017).  https://doi.org/10.1039/c7ta00220c CrossRefGoogle Scholar
  67. 67.
    Xue, L., Li, Y., Gao, H., et al.: Low-cost high-energy potassium cathode. J. Am. Chem. Soc. 139(6), 2164–2167 (2017).  https://doi.org/10.1021/jacs.6b12598 CrossRefPubMedGoogle Scholar
  68. 68.
    Wang, X., Xu, X., Niu, C., et al.: Earth abundant Fe/Mn-Based layered oxide Interconnected nanowires for advanced K-ion full batteries. Nano Lett. 17(1), 544–550 (2017).  https://doi.org/10.1021/acs.nanolett.6b04611 CrossRefPubMedGoogle Scholar
  69. 69.
    Deng, T., Fan, X., Chen, J., et al.: Layered P2-Type K0.65Fe0.5Mn0.5O2 microspheres as superior cathode for high-energy potassium-ion batteries. Adv. Fun. Mater. 28, 1800219 (2018).  https://doi.org/10.1002/adfm.201800219 CrossRefGoogle Scholar
  70. 70.
    Liu, C., Luo, S., Huang, H., et al.: K0.67Ni0.17Co0.17Mn0.66O2: a cathode material for potassium-ion battery. Electrochem. Commun. 82, 150–154 (2017).  https://doi.org/10.1016/j.elecom.2017.08.008 CrossRefGoogle Scholar
  71. 71.
    Chen, Y., Luo, W., Carter, M., et al.: Organic electrode for non-aqueous potassium-ion batteries. Nano Energy 18, 205–211 (2015).  https://doi.org/10.1016/j.nanoen.2015.10.015 CrossRefGoogle Scholar
  72. 72.
    Xing, Z., Jian, Z., Luo, W., et al.: A perylene anhydride crystal as a reversible electrode for K-ion batteries. Energy Storage Mater. 2, 63–68 (2016).  https://doi.org/10.1016/j.ensm.2015.12.001 CrossRefGoogle Scholar
  73. 73.
    Zhao, J., Yang, J., Sun, P., et al.: Sodium sulfonate groups substituted anthraquinone as an organic cathode for potassium batteries. Electrochem. Commun. 86, 34–37 (2018).  https://doi.org/10.1016/j.elecom.2017.11.009 CrossRefGoogle Scholar
  74. 74.
    Jian, Z., Liang, Y., Rodríguez-Pérez, I.A., et al.: Poly(anthraquinonyl sulfide) cathode for potassium-ion batteries. Electrochem. Commun. 71, 5–8 (2016).  https://doi.org/10.1016/j.elecom.2016.07.011 CrossRefGoogle Scholar
  75. 75.
    Zhao, Q., Wang, J., Lu, Y., et al.: Oxocarbon salts for fast rechargeable batteries. Angew. Chem. Int. Ed. 55, 12528–12532 (2016).  https://doi.org/10.1002/anie.201607194 CrossRefGoogle Scholar
  76. 76.
    Nokami, T., Matsuo, T., Inatomi, Y., et al.: Polymer-bound pyrene-4,5,9,10-tetraone for fast-charge and -discharge lithium-ion batteries with high capacity. J. Am. Chem. Soc. 134, 19694–19700 (2012).  https://doi.org/10.1021/ja306663g CrossRefPubMedGoogle Scholar
  77. 77.
    Recham, N., Rousse, G., Sougrati, M.T., et al.: Preparation and characterization of a stable FeSO4F-based framework for alkali ion insertion electrodes. Chem. Mater. 24, 4363–4370 (2012).  https://doi.org/10.1021/cm302428w CrossRefGoogle Scholar
  78. 78.
    Mathew, V., Kim, S., Kang, J., et al.: Amorphous iron phosphate: potential host for various charge carrier ions. NPG Asia Mater. 6, e138–e138 (2014).  https://doi.org/10.1038/am.2014.98 CrossRefGoogle Scholar
  79. 79.
    Naveen, N., Park, W.B., Han, S.C., et al.: Reversible K+-insertion/deinsertion and concomitant Na+-redistribution in P’3-Na0.52CrO2 for high-performance potassium-ion battery cathodes. Chem. Mater. 30, 2049–2057 (2018).  https://doi.org/10.1021/acs.chemmater.7b05329 CrossRefGoogle Scholar
  80. 80.
    Sada, K., Senthilkumar, B., Barpanda, P.: Electrochemical potassium-ion intercalation in NaxCoO2: a novel cathode material for potassium-ion batteries. Chem. Commun. 53, 8588–8591 (2017).  https://doi.org/10.1039/c7cc02791e CrossRefGoogle Scholar
  81. 81.
    Silbernagel, B.G., Whittingham, M.S.: An NMR study of the alkali metal intercalation phase LixTiS2: relation to structure, thermodynamics, and iconicity. J. Chem. Phys. 64, 3670–3673 (1976).  https://doi.org/10.1063/1.432731 CrossRefGoogle Scholar
  82. 82.
    Silbernagel, B.G., Whittingham, M.S.: The physical properties of the NaxTiS2intercalation compounds: asynthetic and NMR study. Mater. Res. Bull. 11, 29–36 (1976).  https://doi.org/10.1016/0025-5408(76)90210-5 CrossRefGoogle Scholar
  83. 83.
    Tian, B., Tang, W., Leng, K., et al.: Phase transformations in TiS2 during K intercalation. ACS Energy Lett. 2, 1835–1840 (2017).  https://doi.org/10.1021/acsenergylett.7b00529 CrossRefGoogle Scholar
  84. 84.
    Wang, L., Zou, J., Chen, S., et al.: TiS2 as a high perfromance potasium ion battery cathode in ether-based electrolyte. Energy Storage Mater. 12, 216–222 (2018).  https://doi.org/10.1016/j.ensm.2017.12.018 CrossRefGoogle Scholar
  85. 85.
    Wessells, C.D., Huggins, R.A., Cui, Y.: Copper hexacyanoferrate battery electrodes with long cycle life and high power. Nat. Commun. 2, 550 (2011).  https://doi.org/10.1038/ncomms1563 CrossRefPubMedGoogle Scholar
  86. 86.
    Wessells, C.D., Peddada, S.V., Huggins, R.A., et al.: Nickel hexacyanoferrate nanoparticle electrodes for aqueous sodium and potassium ion batteries. Nano Lett. 11, 5421–5425 (2011).  https://doi.org/10.1021/nl203193q CrossRefPubMedGoogle Scholar
  87. 87.
    Wang, J., Sun, X.: Understanding and recent development of carbon coating on LiFePO4 cathode materials for lithium-ion batteries. Energy Environ. Sci. 5, 5163–5185 (2012).  https://doi.org/10.1039/c1ee01263k CrossRefGoogle Scholar
  88. 88.
    Kucinskis, G., Bajars, G., Kleperis, J.: Graphene in lithium ion battery cathode materials: a review. J. Power Sources 240, 66–79 (2013).  https://doi.org/10.1016/j.jpowsour.2013.03.160 CrossRefGoogle Scholar
  89. 89.
    Balogun, M.S., Luo, Y., Qiu, W., et al.: A review of carbon materials and their composites with alloy metals for sodium ion battery anodes. Carbon 98, 162–178 (2016).  https://doi.org/10.1016/j.carbon.2015.09.091 CrossRefGoogle Scholar
  90. 90.
    Li, X., Chen, Y., Huang, H., et al.: Electrospun carbon-based nanostructured electrodes for advanced energy storage-A review. Energy Storage Mater. 5, 58–92 (2016).  https://doi.org/10.1016/j.ensm.2016.06.002 CrossRefGoogle Scholar
  91. 91.
    Kang, E., Jung, Y.S., Kim, G.H., et al.: Highly improved rate capability for a lithium-ion battery nano-Li4Ti5O12 negative electrode viacarbon-coated mesoporous uniform pores with a simple self-assembly method. Adv. Fun. Mater. 21, 4349–4357 (2011).  https://doi.org/10.1002/adfm.201101123 CrossRefGoogle Scholar
  92. 92.
    Nossol, E., Zarbin, A.J.G.: Electrochromic properties of carbon nanotubes/Prussian blue nanocomposite films. Sol. Energy Mater. Sol. C. 109, 40–46 (2013).  https://doi.org/10.1016/j.solmat.2012.10.006 CrossRefGoogle Scholar
  93. 93.
    Zhu, Y.H., Yin, Y.B., Yang, X., et al.: Transformation of rusty stainless-steel meshes into stable, low-cost, and binder-free cathodes for high-performance potassium-ion batteries. Angew. Chem. Int. Ed. 56, 7881–7885 (2017).  https://doi.org/10.1002/anie.201702711 CrossRefGoogle Scholar
  94. 94.
    Han, J., Li, G.N., Liu, F., et al.: Investigation of K3V2(PO4)3/C nanocomposites as high-potential cathode materials for potassium-ion batteries. Chem. Commun. 53, 1805–1808 (2017).  https://doi.org/10.1039/c6cc10065a CrossRefGoogle Scholar
  95. 95.
    Nossol, E., Souza, V.H., Zarbin, A.J.: Carbon nanotube/Prussian blue thin films as cathodes for flexible, transparent and ITO-free potassium secondary battery. J. Colloid Interface Sci. 478, 107–116 (2016).  https://doi.org/10.1016/j.jcis.2016.05.056 CrossRefPubMedGoogle Scholar
  96. 96.
    Bruce, P.G., Scrosati, B., Tarascon, J.M.: Nanomaterials for rechargeable lithium batteries. Angew. Chem. Int. Ed. 47, 2930–2946 (2008).  https://doi.org/10.1002/anie.200702505 CrossRefGoogle Scholar
  97. 97.
    Ji, L., Lin, Z., Alcoutlabi, M., et al.: Recent developments in nanostructured anode materials for rechargeable lithium-ion batteries. Energy Environ. Sci. 4, 2682 (2011).  https://doi.org/10.1039/c0ee00699h CrossRefGoogle Scholar
  98. 98.
    He, G., Nazar, L.F.: Crystallite size control of Prussian White analogues for nonaqueous potassium-ion batteries. ACS Energy Lett. 2, 1122–1127 (2017).  https://doi.org/10.1021/acsenergylett.7b00179 CrossRefGoogle Scholar
  99. 99.
    Neff, V.D.: Electrochemical oxidation and reduction of thin films of Prussian Blue. J. Electrochem. Soc. 125, 886–887 (1978)CrossRefGoogle Scholar
  100. 100.
    Schuppert, N.D., Mukherjee, S., Bates, A.M., et al.: Ex-situ X-ray diffraction analysis of electrode strain at TiO2 atomic layer deposition/α-MoO3 interface in a novel aqueous potassium ion battery. J. Power Sour 316, 160–169 (2016).  https://doi.org/10.1016/j.jpowsour.2016.03.064 CrossRefGoogle Scholar
  101. 101.
    Zhong, C., Deng, Y., Hu, W., et al.: A review of electrolyte materials and compositions for electrochemical supercapacitors. Chem. Soc. Rev. 44, 7484–7539 (2015).  https://doi.org/10.1039/c5cs00303b CrossRefPubMedGoogle Scholar
  102. 102.
    Leonard, D.P., Wei, Z., Chen, G., et al.: Water-in-salt electrolyte for potassium-ion batteries. ACS Energy Lett. 3, 373–374 (2018).  https://doi.org/10.1021/acsenergylett.8b00009 CrossRefGoogle Scholar
  103. 103.
    Xu, K.: Nonaqueous liquid electrolytes for lithium-based rechargeable batteries. Chem. Rev. 104(10), 4303–4418 (2004)CrossRefGoogle Scholar
  104. 104.
    Ma, J., Zhou, E., Fan, C., et al.: Endowing CuTCNQ with a new role: a high-capacity cathode for K-ion batteries. Chem. Commun. 54, 5578–5581 (2018).  https://doi.org/10.1039/c8cc00802g CrossRefGoogle Scholar
  105. 105.
    Cho, E., Mun, J., Chae, O.B., et al.: Corrosion/passivation of aluminum current collector in bis(fluorosulfonyl)imide-based ionic liquid for lithium-ion batteries. Electrochem. Commun. 22, 1–3 (2012).  https://doi.org/10.1016/j.elecom.2012.05.018 CrossRefGoogle Scholar
  106. 106.
    Hu, Z., Liu, Q., Chou, S.L., et al.: Advances and challenges in metal sulfides/selenides for next-generation rechargeable sodium-ion batteries. Adv. Mater. 29, 1700606 (2017).  https://doi.org/10.1002/adma.201700606 CrossRefGoogle Scholar
  107. 107.
    Moshkovich, M., Gofer, Y., Aurbach, D.: Investigation of the electrochemical windows of aprotic alkali metal (Li, Na, K) salt solutions. J. Electrochem. Soc. 148, E155–E167 (2001).  https://doi.org/10.1149/1.1357316 CrossRefGoogle Scholar
  108. 108.
    Komaba, S., Ishikawa, T., Yabuuchi, N., et al.: Fluorinated ethylene carbonate as electrolyte additive for rechargeable Na batteries. ACS Appl. Mater. Interfaces 3, 4165–4168 (2011).  https://doi.org/10.1021/am200973k CrossRefPubMedGoogle Scholar
  109. 109.
    Xu, K.: Electrolytes and interphases in Li-ion batteries and beyond. Chem. Rev. 114, 11503–11618 (2014).  https://doi.org/10.1021/cr500003w CrossRefPubMedGoogle Scholar
  110. 110.
    Zhang, J.N., Li, Q., Wang, Y., et al.: Dynamic evolution of cathode electrolyte interphase (CEI) on high voltage LiCoO2 cathode and its interaction with Li anode. Energy Storage Mater. 14, 1–7 (2018).  https://doi.org/10.1016/j.ensm.2018.02.016 CrossRefGoogle Scholar
  111. 111.
    Minato, T., Kawaura, H., Hirayama, M., et al.: Dynamic behavior at the interface between lithium cobalt oxide and an organic electrolyte monitored by neutron reflectivity measurements. J. Phys. Chem. C 120, 20082–20088 (2016).  https://doi.org/10.1021/acs.jpcc.6b02523 CrossRefGoogle Scholar
  112. 112.
    Cherkashinin, G., Nikolowski, K., Ehrenberg, H., et al.: The stability of the SEI layer, surface composition and the oxidation state of transition metals at the electrolyte-cathode interface impacted by the electrochemical cycling: X-ray photoelectron spectroscopy investigation. Phys. Chem. Chem. Phys. 14, 12321–12331 (2012).  https://doi.org/10.1039/c2cp41134b CrossRefPubMedGoogle Scholar
  113. 113.
    Chen, M., Wang, W., Liang, X., et al.: Sulfur/oxygen codoped porous hard carbon microspheres for high-performance potassium-ion batteries. Adv. Energy. Mater. 8, 1800171 (2018).  https://doi.org/10.1002/aenm.201800171 CrossRefGoogle Scholar
  114. 114.
    Wang, W., Zhou, J., Wang, Z., et al.: Short-range order in mesoporous carbon boosts potassuim-ion battery performance. Adv. Energy. Mater. 8, 1701648 (2018).  https://doi.org/10.1002/aenm.201701648 CrossRefGoogle Scholar
  115. 115.
    Wang, W., Jiang, B., Qian, C., et al.: Pistachion-shuck-like MoSe2/C core/shell nanostructures for high-performance potassium-ion storage. Adv. Mater. 30, 1801812 (2018).  https://doi.org/10.1002/adma.201801812 CrossRefGoogle Scholar
  116. 116.
    Yang, C., Feng, J., Lv, F., et al.: Metallic graphene-Like VSe2 ultrathin nanosheets: superor potassium-ion storage and their working mechanism. Adv. Mater. 30, 1800036 (2018).  https://doi.org/10.1002/adma.201800036 CrossRefGoogle Scholar
  117. 117.
    Cao, B., Zhang, Q., Liu, H., et al.: Graphitic carbon nanocage as a stable and high power anode for pontassium-ion batteries. Adv. Energy Mater. 8, 1801149 (2018).  https://doi.org/10.1002/aenm.201801149 CrossRefGoogle Scholar
  118. 118.
    Lei, Y., Qin, L., Liu, R., et al.: Exploring stability of nonaqueous electrolytes for potassium-ion batteries. ACS Appl. Energy Mater. 1, 1828–1833 (2018).  https://doi.org/10.1021/acsaem.8b00214 CrossRefGoogle Scholar
  119. 119.
    Zhang, Q., Mao, J., Pang, W.K., et al.: Boosting the potassium storage performance of alloy-based anode materials viaelectrolyte salt chemistry. Adv. Energy Mater. 8, 1703288 (2018).  https://doi.org/10.1002/aenm.201703288 CrossRefGoogle Scholar
  120. 120.
    Zhang, W., Pang, W.K., Sencadas, V., et al.: Understanding high-energy-density Sn4P3anodes for potassium-ion batteries. Joule (2018).  https://doi.org/10.1016/j.joule.2018.04.022 CrossRefGoogle Scholar
  121. 121.
    Xiao, N., McCulloch, W.D., Wu, Y.: Reversible dendrite-free potassium plating and stripping electrochemistry for potassium secondary batteries. J. Am. Chem. Soc. 139, 9475–9478 (2017).  https://doi.org/10.1021/jacs.7b04945 CrossRefPubMedGoogle Scholar
  122. 122.
    Liu, Y., Tai, Z., Zhou, T., et al.: An all-Integrated anode viainterlinked chemical bonding between double-shelled-yolk-structured silicon and binder for lithium-ion batteries. Adv. Mater. 29, 1703028 (2017).  https://doi.org/10.1002/adma.201703028 CrossRefGoogle Scholar
  123. 123.
    Kovalenko, I., Zdyrko, B., Magasinski, A., et al.: A major constituent of brown algae for use in high-capacity Li-ion batteries. Science 334, 75–79 (2011).  https://doi.org/10.1126/science.1209150 CrossRefPubMedGoogle Scholar
  124. 124.
    Obrovac, M.N., Chevrier, V.L.: Alloy negative electrodes for Li-ion batteries. Chem. Rev. 114, 11444–11502 (2014).  https://doi.org/10.1021/cr500207g CrossRefPubMedGoogle Scholar
  125. 125.
    Zhu, Y.H., Yang, X., Bao, D., et al.: High-energy-density flexible potassium-ion battery based on patterned electrodes. Joule 2, 736–746 (2018).  https://doi.org/10.1016/j.joule.2018.01.010 CrossRefGoogle Scholar

Copyright information

© Shanghai University and Periodicals Agency of Shanghai University 2018

Authors and Affiliations

  1. 1.Institute for Superconducting and Electronic Materials (ISEM), Australian Institute for Innovative Materials (AIIM)University of WollongongWollongongAustralia
  2. 2.School of Mechanical, Materials, Mechatronics and Biomedical EngineeringUniversity of WollongongWollongongAustralia

Personalised recommendations