The transition to a sustainable society is vital and requires electrification. Sodium and potassium ion-based electrolytes will likely play an important role in energy storage as these elements are very abundant. The latter cations and chloride are especially interesting since life on the planet is somehow based on biological transfers of these ions through cell membranes. K+ is the key charge carrier in plants. Here, we characterize electrochemically, electrostatically, and structurally novel electrolytes, K3ClO and K2.99Ba0.005ClO, and compare their performance with Na3ClO and Na2.99Ba0.005ClO in symmetric and asymmetric structural electrode-less cells, such as K/K2.99Ba0.005ClO in a cellulose membrane/K, Na/Na2.99Ba0.005ClO in a cellulose membrane/Na, Al/K2.99Ba0.005ClO composite/Cu, and Al/Na2.99Ba0.005ClO composite/Cu, at temperatures that range from −45 to 65 °C. An ab initio molecular dynamics structural study followed by band structure determination using density functional theory and hybrid simulations allowed us to compare the amorphous character of the structures, bandgap, and electron localization function for both K3ClO at 25 °C and Na3ClO at 37 °C, temperatures at which preliminary studies indicate that these compounds are already amorphous. As in Na+-based electrolytes, the ferroelectric character of the K+-based electrolytes is well recognizable, especially at −45 °C, where the relative real permittivity achieves 1013 in K/K2.99Ba0.005ClO in cellulose membrane/K symmetric cells for an ionic conductivity of ∼120 mS/cm. As in Na+-based electrodes-less structural battery cells, self-charge and self-cycling phenomena are also demonstrated reinforcing the ferroelectric nature of the A3ClO (A = Li, Na, and K) family of electrolytes. These studies may contribute to understanding the K+ and Na+ transfer behavior in energy harvesting and storage as well as the biologic world.

Over the years, climate change and global warming pose an increasing danger to ecosystems, human health, and growing economies around the world. These changes are occurring due to the release of large amounts of greenhouse gases into the atmosphere from human activities, particularly, by burning fossil fuels to generate huge amounts of electricity and heat, as well as for transportation purposes. The burning of fossil fuels emits air pollutants that are extremely harmful to the environment and human health. Therefore, there is a great need to fulfill the gigantic demand for energy in an eco-friendly and sustainable manner.1–6 The development, production, and use of environment-friendly batteries are, hence, the key to achieving a climate-neutral economy due to their important role in storing renewable energy and implementing zero-emission mobility.1–11 

Currently, the most popular rechargeable energy source is lithium-ion batteries (LIBs). They are mainly used in portable electronic devices and electric and hybrid electric vehicles due to their high energy density, long cycle lives, and high working electrical potential difference.10–15 However, to meet the development of large-scale high-capacity energy storage systems and reduce the cost of renewable energy storage, the development of alternative technologies with more economical advantage and even better performance than that achieved by LIBs is crucial.8,9,16 The raw materials required for the production of LIBs because of the lack of abundance of lithium resources (∼0.002 wt. % in crust abundance), high cost, and uneven distribution in the Earth’s crust preclude the use of current LIBs as low-cost energy storage devices capable of storing energy from renewable energy sources, such as solar or wind.16–19 Moreover, since commercial LIBs use liquid flammable electrolytes, these batteries cannot be freely operated at temperatures above 40 °C due to the risk of thermal runaway and thus battery explosion, resulting in the release of toxic electrolyte derivatives.20–23 

Therefore, the most promising alternatives to LIBs are sodium-ion batteries (NIBs) and perhaps potassium-ion batteries (KIBs).18,19,24–31 Due to the high content of Na and K in the Earth’s crust (∼2.3 and ∼1.5 wt. %, respectively) compared to Li, the cost of producing electrodes and electrolytes for NIBs and KIBs is much lower than for LIBs.18,24,27,28,31,32 Moreover, this type of battery can use aluminum current collectors in place of the copper employed in LIBs due to the lack of reaction of Na and K with Al, which reduces the manufacturing cost and the weight of these alternative energy sources themselves.27,33–35 However, KIBs have a slight advantage over NIBs.26–28,33 Potassium has a lower reduction potential than sodium [K+/K: −2.93 V and Na+/Na: −2.71 V vs standard hydrogen electrode (SHE) and −1.51 and −1.73 eV, respectively, vs electrons at rest at vacuum], which allows KIBs to operate at a higher voltage than NIBs or even LIBs (in non-aqueous electrolytes).27,30,33,35,36 Furthermore, it is believed that due to the rapid diffusion of K+ ions, potassium-ion batteries should have a higher power generation capacity.27,37,38 More importantly, graphite can be used as an electrode in KIBs, which can accommodate reversible K+ de/intercalation but is not capable of reversible storage of the Na+, thus eliminating graphite in NIB technology.27,38,39 It is also important to note that potassium-based liquid electrolytes show higher conductivity than both Li- and Na-based electrolytes. This is most likely due to the smaller Stokes radius of the dissolved K+ ions caused by their weak Lewis acidity and the low interfacial reaction resistance due to the low activation energy of desolvation.27,33,37,38 Consequently, K-based electrolytes have quickly attracted much interest, and various materials have been developed and evaluated as potential KIB electrodes.27,29

However, KIBs also have disadvantages such as the poor diffusion of K+ ions in solid electrolytes, which greatly slows down the reaction kinetics in solid-state KIBs. In addition, during the K+ ion de/intercalation process, the volume change in the electrode material due to the large K+ radius in KIBs is larger than that in NIBs and LIBs.27,28

Due to the lower electrochemical potential of potassium, a reduction in the solvent in the liquid electrolyte at the electrode surface in KIBs is possible, resulting in undesirable side reactions.27 Moreover, potassium itself has a lower melting point (63.5 °C) than Na (98 °C) and Li (180.5 °C) and has a much higher reactivity especially with air components (i.e., oxygen and water vapor), which strongly decreases the operational safety of such batteries.27,29

Therefore, to bring NIBs, or even KIBs, to the commercial market in place of the currently used LIBs will require intensive work on electrode materials for these batteries. It is also extremely important to obtain an efficient and safe electrolyte, which has a crucial part in forming protective layers on both the cathode (surface layer) and the anode [solid electrolyte interface (SEI) layer].29,31

With no conventional cathode and anode needed, potassium- and sodium-based solid electrolytes are ideal for structural battery applications.20,40–48 With collectors–electrodes such as aluminum or zinc and carbon or copper to fix the difference in chemical potentials and a Na+- or K+-based solid electrolyte that fixes the capacity of the cell by plating on the collectors, large extensions of, for example, coaxial-beam shaped cells, can be applied to the interior of a vehicle, on a wall, in industrial facilities, and in databanks with multiple functionalities of storing energy, harvesting wasted heat and thermal energy, and being protective against mechanical impacts.

Herein, we report new electrolytes, K3ClO and K2.99Ba0.005ClO, and compare their properties with Na3ClO and Na2.99Ba0.005ClO electrolytes in both symmetric and asymmetric structural electrode-less cells. Electrochemical properties of Na2.99Ba0.005ClO electrolytes measured in symmetric cells with blocking electrodes have been first reported by Braga et al. in Ref. 49.

Ab initio molecular dynamics (AMD) structural studies followed by the band structure determination by density functional theory (DFT) and hybrid simulations allowed for the comparison between the amorphous nature of the structures, bandgap, and electron localization functions of K3ClO at 25 °C and Na3ClO at 37 °C, temperatures at which preliminary studies indicate that the compounds are already amorphous.

Two different types of all-solid-state ferroelectric electrolytes with enhanced electrochemical and thermal properties were synthesized (A2.99Ba0.005ClO with A = Na, K), and their properties were compared in different cell configurations, including K/K2.99Ba0.005ClO in the cellulose membrane/K, Na/Na2.99Ba0.005ClO in the cellulose membrane/Na, Al/K2.99Ba0.005ClO composite/Cu, and Al/Na2.99Ba0.005ClO composite/Cu.

The synthesis of the dry glass Na+ or K+ based solid electrolytes was realized in compliance with the protocol presented by Braga et al.49 The precursors NaCl (>99%, Merck) or KCl (99.5%, PanReac AppliChem), Na(OH) (>99%, Merck) or K(OH) (85.7%, Alfa Aesar), and Ba(OH)2 (94%–98%, Alfa Aesar) were mixed with deionized water before letting them react and dry between 230 and 250 °C. The solid electrolytes A2.99Ba0.005OCl1−x(OH)x with A = Na or K were subsequently dried to 230 °C to eliminate the hydroxide phases and obtain the glassy A2.99Ba0.005ClO (A = Na, K). Once in its final configuration, the solid-state electrolytes were ground for 45 min at 350 rpm using a ball milling machine with a hermetically closed agate container and balls with a diameter of 20 mm.

The first type of cells that were prepared to characterize the electrochemical properties of these Li-free all-solid-state electrolytes were the symmetrical coin cells (with Na- or K-metal electrodes). The CR2032 coin-cells were selected to host symmetric cells manufactured with disks of pure A (A = Na or K) alkali-metal electrodes with a reference diameter of 8 mm. The alkali-metal disks were cut from the raw material chunks of Na (>99%, Sigma-Aldrich) or K (>98%, Sigma-Aldrich). The interleave electrolyte-separator was a non-woven cellulose layer impregnated with the correspondent A+-based electrolyte A2.99Ba0.005ClO (A = Na, K) with a diameter of 16 mm and thickness of ∼1 mm. The separators’ disks were soaked in a slurry composed of absolute ethanol (>99.5%, VWR chemicals) mixed with the electrolyte powders. The separators were then let to dry overnight in the Ar-dry glovebox at 70 °C before proceeding with the assembly of the cells.

As described previously, all materials were handled in an Ar-dry glovebox with O2 % < 1.0 ppm; extra attention was given to the metals by removing the oxide layer on the exposed surface before cutting.

An all-solid-state coaxial structural battery design was selected as an application for the proposed electrolytes. In this configuration, introduced by Danzi et al.,20 the electrolyte was mixed with the thermoplastic polyvinyl acetate (PVAc) (C4H6O2)n in a 4 A2.99Ba0.005OCl:1 PVAc ratio. This coaxial structural battery design is composed of a [90/0/+45/−45]S outer shell of carbon fiber reinforced plastic (CFRP) fabricated using T800-736LT 100 gsm. The tubular structure works as a host for the coaxial battery fabricated with a copper with a thickness of 0.127 mm from Alfa Aesar as a positive electrode/current collector co-cured to the CFRP outer shell, while in the axis of the circular beam, a 4 mm diameter rod of commercial aluminum is used as the negative electrode/current collector. The gap between the two was then filled with the electrolyte-based mixture previously described. The geometries and further details used in this work were presented in Ref. 20.

The electrochemical performance of the all-solid-state electrolytes and correspondent cells was then evaluated via a series of electrochemical tests.

Four types of tests were adopted: cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), chronopotentiometry (CP), and electrochemical discharge.

The CV, EIS, and CP tests with the symmetric coin-cells were performed using a Biologic SP-240 potentiostat/galvanostat/impedance spectroscope in an Ar-dry glovebox. A Biologic VMP-300 potentiostat/galvanostat/impedance spectroscope and a Gamry reference 3000 potentiostat/galvanostat/ZRA were used for testing the structural coaxial batteries.

The CV experiments were carried out to determine the capacitance of different cell configurations presented here and the relative real permittivity or dielectric constant of the electrolytes. All the experiments were carried out superimposing ±0.1 V to the initial differential potential corresponding to the open circuit voltage (OCV) of the cell. The CR2032 symmetric cells with the alkali-metal electrodes were tested in a temperature range of −45 to 70 °C for Na2.99Ba0.005ClO and −45 to 40 °C for K2.99Ba0.005ClO. The lower temperature is the minimum temperature achievable in the freezer unit integrated within the glovebox, while the upper range has been defined considering a safety margin from the melting point of the alkaline metals used here. Ten CV cycles at 0.1 mV/s correspond to each temperature. Temperature equilibration was achieved unequivocally. The CV analyses of the asymmetric cylindrical cells were instead carried out at three different temperatures only: 0 °C, room temperature (∼25 °C), and 40 °C. These tests were performed varying the scan rate from 0.1 up to 50 mV/s, and the values of the average current were measured at the measured OCV value of the cell.

The EIS analyses were performed to determine the internal resistance of the distinct types of cells. All the experiments were carried out with an alternate current AC with an amplitude of ±10 mV for an initial differential potential corresponding to the OCV of the cell, and the frequency range was from 3.0 down to 200 mHz.

The latter CV and EIS experiments were also performed with Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu (foil) coaxial cells.

Chronopotentiostatic cycles with a constant current of 2 mA/cm2 were performed at −40, −20, −10, 0, and 10 °C to assess the K/K2.99Ba0.005ClO and Na/Na2.99Ba0.005ClO interfaces’ behaviors and compare internal resistances obtained with this DC method vs obtained with EIS using AC. The cycles were carried on for (2 + 2) h.

Electrochemical discharges were performed with external physical resistors that were connected to the coaxial cells inside the sand or silicon-oil bath at ∼25 or 40 °C. The use of external physical resistors is intended to unequivocally control the discharge of the cell (and the self-charge and self-cycling), not making it dependent on the amplifier of the potentiostat or the variation in the internal resistance of the cell. The measuring instrument works solely as a voltmeter connected in parallel with the external resistor.

The capacitance of the cell was calculated at OCV using the 〈i〉 vs dV/dt slope and the following equation:

i=CdVdt,
(1)

where ⟨i⟩ is the average current at OCV, C is the capacitance of the cell, and dV/dt is the voltage rate.

From the capacitance in Eq. (1), the real relative permittivity or dielectric constant ɛr of the dielectric material in the double-layer coin-cells is determined. The thickness of the electrolyte-cellulose membrane ferroelectric-dielectric separator is d and the cross-sectional surface area is A, which obeys the condition required by Gauss’s law d2A,

C=εrε0Ad,
(2)

where ɛ0 is the permittivity of the vacuum. In the CR2032 coin-cells studied herein, A = 0.5 cm2 and d = 0.1 cm.

The conductivity is obtained from the resistance R using

R=1σdA.
(3)

The resistance was determined from the impedance spectroscopy (EIS) data using the equivalent circuit discussed in Ref. 49, when possible, or simply by identifying the real passive resistance on the Re(Z) axis, or using Ohm’s law V = RI on the chronopotentiometry data.

From the capacitance in Eq. (1), the dielectric constant in the coaxial cells is determined as

C=2πεrε0lnba,
(4)

where is the length of the cell, b is the external radius, and a is the internal radius of the ferroelectric-electrolyte.

The resistance toward the radial ionic movement is then attained,

R=12πσlnba,
(5)

which allows us to calculate the ionic conductivity in coaxial cells. In the cells studied here, = 10 cm, b = 0.6 cm, and a = 0.2 cm.

Ultimately, discharge curves of coaxial asymmetric cells were performed with physical resistors of 1.8 and 26.6 kΩ for Al/K2.99Ba0.005ClO composite/Cu (mesh) and Al/Na2.99Ba0.005ClO composite/Cu (foil), respectively.

Molecular Dynamics (MD) enables the development of a model in which the molecules are continuously moving. It simulates the behavior of a limited number of molecules, confined to a given volume and interacting with one another through a given pair potential.51 

In an Ab initio Molecular Dynamics (AMD) run, the forces calculated in a given geometry step are used to update the atomic positions. The system dynamics, i.e., the ionic movements, are subject to Newton mechanics, while the forces acting on the ions are calculated from ab initio using a self-consistent electronic density (Hellmann–Feynman forces) as implemented in the Vienna Ab initio Software Package (VASP).52 Periodic boundary conditions are assumed and allow a molecule, which leaves the volume across one face, to re-enters across the opposite face. In the statistical ensemble, the various states of the system differ in positions and velocities of the component’s particles. The space of all possible system states of dimension 6N for N particles is termed the phase space.

A system at imposed pressure P and temperature T is represented by an isothermal–isobaric ensemble. This ensemble plays an important role as chemical reactions are usually carried out under constant pressure conditions. The number of molecules and the pressure are identical for all system states belonging to the ensemble, but they differ in total energy, which is a fluctuating variable in this ensemble. Each state j of the isothermal–isobaric ensemble occurs with a probability proportional to eEj/kBT, where kB is the Boltzmann constant and Ej is the total energy (kinetic + potential) of the system in state j. Here, we use the symbol E for the total energy (kinetic and potential). The Boltzmann factor eEj/kBT expresses that low energy states are favored when compared to high-energy states. Increasing temperature broadens the energy distribution in the ensemble, and as a consequence, the average energy is increased.

Ab initio molecular dynamics (AMD) as implemented by VASP52 was used to simulate a closed system thermostatted to a heat bath at a constant temperature. The studied systems were (K3ClO)27 and (Na3ClO)27 that were left to relax. The simulations extended for at least 360 fs. The initial structures were the optimized structures at the correspondent temperatures after performing microcanonical simulation NV’E (with volume V′ and total energy E constant) from the crystalline optimized structure (antiperovskite, cubic Pm-3m). Isothermal–isobaric simulations NPT were performed to set the volume at the correspondent temperature. The AMD simulation with 4 fs time steps was used to determine each structure, and the velocities were rescaled every time step to maintain a constant temperature. The temperatures, 25 and 37 °C (298 and 310 K) for (K3ClO)27 and (Na3ClO)27, respectively, were chosen to assure that the system is in the amorphous state, immediately above its transition to amorphous.

The simulations were based on density functional theory (DFT) and the generalized gradient approximation (GGA)-Perdew–Burke-Ernzerhof (PBE)53,54 exchange–correlation functional and hybrid functional HSE06 for describing the interactions in both structures obtained with AMD. It is noteworthy that hybrid functional HSE06 is more precise in what concerns the computation of the band structure of a semiconductor.

Electronic structure calculations were conducted with a plane-wave cutoff energy of 400 or 500 eV. The electronic iteration convergence was 10−5 eV using the normal (blocked Davidson) algorithm and real and reciprocal space projection operators. The requested k-spacing for the non-local exchange was 0.5 Å−1, which leads to a 1 × 1 × 1 mesh, or 0.3 Å−1, which leads to a 2 × 2 × 2 mesh, using first-order Methfessel-Paxton smearing with a width of 0.2 eV.

The pair distribution functions (PDFs) g(r) vs the interatomic distances r and x-ray diffraction (XRD) patterns for CuKα source radiation were also obtained for controlling the non-periodic character of the amorphous structures. The electron localization functions were also established to determine the nature of the bonding, the dipole character, and the dynamics of the electrons favored by the possible dynamics of the ions in the present semiconductors at a certain temperature.

It is important to emphasize that if an event is observed during the AMD relaxation, the correspondent manifestation should be observed in a real environment. However, processes with slow dynamics will not be observed in AMD, even if occurring in nature.

One of the most striking features in the A2.99M0.005ClO (A = Li, Na, K and M = Mg, Ca, Sr, Ba) family of glassy electrolytes is their ferroelectric character. A ferroelectric material polarizes spontaneously, and its polarization can be reversed by the application of an electric field.55 The ferroelectric phenomenon is of a quantum mechanics nature, but its coherence is maintained at the classical scale. In face of a novel electrolyte of the same family, one of the most important properties to determine is the relative real permittivity or dielectric constant at different temperatures (Fig. 1). The permittivity was then determined by cyclic voltammetry (CV) using non-blocking alkali-metal electrodes. The permittivity is higher at all temperatures than ever found before,50,56 and for that, the following factors may contribute: (1) the presence of non-blocking electrodes for which the electrolytes of this family have a very good affinity, not exhibiting the formation of an SEI layer but forming a layer of electrolyte leading to plating; (2) the low voltage rate used in the CV experiments allowing for ferroelectric polarization of the cells; (3) the drying efforts conducted in the glovebox to avoid hydroxide crystalline phase formation that are detrimental to the fast ionic conduction necessary to the polarization optimization process; and (4) the affinity these electrolytes show toward the presence of a non-woven cellulose matrix.57,58 Cellulose is a dielectric (ɛr = 1.0–1.5) material with (OH) groups along the fibers that can attract the alkali mobile cations and facilitate both polarization and ionic conduction.57–59 It was previously highlighted that K+ is the key charge carrier in plants, and therefore, there might be a relationship between the presence of cellulose and the ionic enabled polarization in the K3ClO family of electrolytes and similarly in the Na3ClO. Cellulose is a very relevant structural component of the primary cell wall of green plants.

FIG. 1.

Cyclic voltammetry (CV) vs temperature for K2.99Ba0.005ClO and Na2.99Ba0.005ClO all-solid-state ferroelectric electrolytes at 0.1 mV/s in K/K2.99Ba0.005ClO in cellulose/K and Na/Na2.99Ba0.005ClO in cellulose/Na symmetric cells; the latter corresponding to Na+ (1) and Na+ (2) are similar. The measurements denoted in green dots corresponding to the Na+ (2) cell were performed to confirm the transition at ∼−20 °C and allow for estimating the temperature range associated with it. Data obtained at OCV: (a) average current, (b) capacitance, and (c) permittivity.

FIG. 1.

Cyclic voltammetry (CV) vs temperature for K2.99Ba0.005ClO and Na2.99Ba0.005ClO all-solid-state ferroelectric electrolytes at 0.1 mV/s in K/K2.99Ba0.005ClO in cellulose/K and Na/Na2.99Ba0.005ClO in cellulose/Na symmetric cells; the latter corresponding to Na+ (1) and Na+ (2) are similar. The measurements denoted in green dots corresponding to the Na+ (2) cell were performed to confirm the transition at ∼−20 °C and allow for estimating the temperature range associated with it. Data obtained at OCV: (a) average current, (b) capacitance, and (c) permittivity.

Close modal

It is clear that the symmetric cell with potassium shows a huge permittivity, which is higher at −45 °C, refuting the variation trend in the sodium-based permittivity. The sodium-based cell, on the other hand, shows a transition temperature at ∼−20 °C, which was observed in previous works of the group with different types of cells with Na2.99Ba0.005ClO and was not fully characterized yet. Another feature worth mentioning is the polarization current at OCV of K/K2.99Ba0.005ClO in the cellulose/K cell, which assumes the value of 3 mA/cm2 at −45 °C for a capacitance of ∼15 F (Fig. 1). From −45 to 40 °C, the permittivity of the K+-based ferroelectric-dielectric separator varies from 4 × 1013 to 1013 [Fig. 1(c)]. The permittivity of the Na+-based ferroelectric-dielectric separator in Na/Na2.99Ba0.005ClO in the cellulose/Na cell varies from 8 × 1011 to 2 × 1013, which is an equally elevated permittivity but not as outstanding as that obtained with K/K2.99Ba0.005ClO in the cellulose/K symmetric cell.

The conductance, and consequently, the ionic conductivity of the cells, should be related to their permittivity as the polarization of the ferroelectric-electrolyte is partially enabled by the hopping of the mobile cations; yet, the conductance of the symmetric cell might not be synchronized with its permittivity as not only other phenomena might influence the conduction process across the cell, but the relaxation phenomena might also play an important role as observed before in relaxation oscillators obtained with these families of electrolytes60 and analyzed hereafter.

From the EIS data, using Eq. (3), it is possible to determine the conductivity of the K/K2.99Ba0.005ClO in the cellulose/K cell, σ = 191 mS/cm (−20 °C) to 127 mS/cm (−40 °C) and 150 mS/cm (25 °C), which includes all resistances, such as the bulk ionic conductivity and interfaces, as it was impossible, in this case, due to such a high conductivity, to determine just the bulk ionic conductivity from the Nyquist plot (see the supplementary material, Table S1). The interfaces’ ohmic resistance, including the K-metal/stainless steel, is predominant.

For Na/Na2.99Ba0.005ClO in the cellulose/Na cell, the ionic conductivity varies from σ = 171 mS/cm (−40 °C) to 69 mS/cm (0 °C) and also includes all cell’s internal resistance. This latter analysis shows a possible lag between the transition that is observed at −20 °C in the CV measurements and at 0 °C in EIS measurements.

The impedances in Figs. 2(a) and 2(b) are positive and approximate to pure imaginaries corresponding to inductive components Z(Ω) = +jωL, where L is the inductance and ω is the frequency. This inductive component is not due to magnetism but due to the polarization of the ferroelectric material parallel to the applied electric field at a certain temperature.

FIG. 2.

Potentiostatic (EIS) and chronopotentiostatic (CP) measurements vs temperature for K2.99Ba0.005ClO and Na2.99Ba0.005ClO all-solid state ferroelectric electrolytes in K/K2.99Ba0.005ClO in cellulose/K and Na/Na2.99Ba0.005ClO in cellulose/Na symmetric cells. (a) EIS for the K+-based cell; (b) EIS for the Na+-based cell; (c) CP for the K+-based cell at 0 °C; and (d) CP for the K+-based cell at −40 °C; the full run is available in the supplementary material, Fig. S2. Note: The color defines the temperature of the run and the tone defines the order, being the darkest the last consecutive run and the lightest the first.

FIG. 2.

Potentiostatic (EIS) and chronopotentiostatic (CP) measurements vs temperature for K2.99Ba0.005ClO and Na2.99Ba0.005ClO all-solid state ferroelectric electrolytes in K/K2.99Ba0.005ClO in cellulose/K and Na/Na2.99Ba0.005ClO in cellulose/Na symmetric cells. (a) EIS for the K+-based cell; (b) EIS for the Na+-based cell; (c) CP for the K+-based cell at 0 °C; and (d) CP for the K+-based cell at −40 °C; the full run is available in the supplementary material, Fig. S2. Note: The color defines the temperature of the run and the tone defines the order, being the darkest the last consecutive run and the lightest the first.

Close modal

Chronopotentiometry (CP) measurements show results that agree with the EIS’s at 0 °C, the CP of K/K2.99Ba0.005ClO in the cellulose/K cell returns an ionic conductivity of σ = 130 mS/cm, while EIS returns σ = 181 mS/cm (see the supplementary material, Table S1). It is noteworthy that a DC method, such as CP, usually delivers lower conductivity than an AC method, such as EIS, due to the nature of the applied current and its effects on the material being a function of the current’s frequency.

Another feature worth highlighting on the CP measurements shown in Figs. 2(c) and 2(d) is the tendency to a preferential direction for the fastest conduction of the ions that was progressive from 10 to −40 °C and the self-oscillation that only arises at −40 °C (see the supplementary material, Figs. S1 and S2).

The asymmetric Al/K2.99Ba0.005ClO composite/Cu and Al/Na2.99Ba0.005ClO composite/Cu ferroelectric coaxial structural batteries have shown much higher resistance than the symmetric cells as reflected in Fig. 3 and the supplementary material, Tables S1 and S2. For the asymmetric cells, the theoretical OCV is given by the difference between the chemical potential of the electrodes, which, in this study, are commercial Al and pure Cu. The aluminum chemical potential is likely affected by the oxide layer that inevitably forms on its surface, and therefore, the OCV is 1.15–1.2 V for K+-based and 1.05–1.13 V for Na+-based asymmetric cells. The OCV of the asymmetric cell may also have a noticeable contribution from the polarization of the electrolyte according to Landau–Devonshire’s theory for a ferroelectric material.50,55 The highest conductivity σK+coaxialcell,Cumesh = 0.17 mS/cm is attained for the Al/K2.99Ba0.005ClO composite/Cu (mesh) coaxial battery at 40 °C, but it is very similar for all cells at the same temperature σK+coaxialcell,Cufoil = 0.14 mS/cm, although slightly lower for Al/Na2.99Ba0.005ClO composite/Cu σNa+coaxialcell,Cufoil = 0.10 mS/cm. Nonetheless, at 0 °C, the internal resistance of the Al/Na2.99Ba0.005ClO composite/Cu corresponding to a conductivity σNa+coaxialcell,Cufoil = 0.0014 mS/cm is similar to that of the Al/K2.99Ba0.005ClO composite/Cu coaxial (foil and mesh) batteries σK+coaxialcell,Cufoil = 0.0026 mS/cm and σK+coaxialcell,Cumesh = 0.0016 mS/cm (see the supplementary material, Table S2).

FIG. 3.

Potentiostatic EIS plots for Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu ferroelectric coaxial structural batteries at 0, 25, and 40 °C. Nyquist plots for (a) Al/K2.99Ba0.005ClO composite/Cu (foil), (b) zoomed-in view of (a), (c) Al/K2.99Ba0.005ClO composite/Cu (mesh), (d) zoomed-in view of (c), (e) Al/Na2.99Ba0.005ClO composite/Cu (foil), and (f) zoomed-in view of (e). Note: The color defines the temperature of the run and the tone defines the order, being the darkest the last consecutive run and the lightest the first.

FIG. 3.

Potentiostatic EIS plots for Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu ferroelectric coaxial structural batteries at 0, 25, and 40 °C. Nyquist plots for (a) Al/K2.99Ba0.005ClO composite/Cu (foil), (b) zoomed-in view of (a), (c) Al/K2.99Ba0.005ClO composite/Cu (mesh), (d) zoomed-in view of (c), (e) Al/Na2.99Ba0.005ClO composite/Cu (foil), and (f) zoomed-in view of (e). Note: The color defines the temperature of the run and the tone defines the order, being the darkest the last consecutive run and the lightest the first.

Close modal

The temperature dependency of the resistance for the Al/K2.99Ba0.005ClO composite/Cu asymmetric cell does not follow the permittivity dependency obtained with K/K2.99Ba0.005ClO in the cellulose/K symmetric cell and not even the resistance obtained with it. The conductivity σK+coaxialcell,Cufoil = 0.048 mS/cm, σK+coaxialcell,Cumesh = 0.018 mS/cm, and σNa+coaxialcell,Cufoil = 0.033 mS/cm was attained at 25 °C for Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu, respectively. Although a symmetric cell reflects the properties of the electrolyte (Table I), conversely to an asymmetric cell, for the differences between cells’ conductivities, several factors might influence: (a) the presence of moisture during fabrication, which is more difficult to avoid even after the cell’s treatment in the Ar-dry glovebox as the shape of the cell does not facilitate the release of moisture; (b) the presence of a thermoplastic to aggregate and facilitate contact among powders and between electrolyte and metals, instead of cellulose; (c) the less affinity to plate K and Na on Cu than to plate on the correspondent alkali metal; and finally, (d) the pressure deficiency not enough to keep the electrolyte in good contact with the collectors. In the Nyquist plot, the highest frequency Cole-Cole semicircle corresponds to the freer movement of the mobile ionic species, and the lowest to the more constraint movement due to Coulombic forces. Therefore, it is more accurate to compare the impedance in the symmetric cells with the impedance correspondent to the semicircles at the highest frequencies in EIS (Fig. 3 and Table I). Both electrolytes (K+- and Na+-based) in the coaxial asymmetric batteries show the same conductivity at 0 °C–0.02 mS/cm. At 25 and 40 °C, the K2.99Ba0.005ClO composite demonstrates a conductivity that is approximately one order of magnitude higher than the Na2.99Ba0.005ClO composite at the same temperatures (Table I). How the coaxial cells behave electrochemically and electrostatically at 25 and 40 °C seems to be much more related to the ionic conductivity than to the dielectric constant, as demonstrated hereafter.

TABLE I.

Resistances and conductivities for the electrolyte in coaxial structural asymmetric batteries at different temperatures. Boldface denotes the average resistance and ionic conductivity of the electrolyte at 0, 25, and 40 °C, respectively, calculated in thecorrespondent asymmetric coaxial cell.

Al/K2.99Ba0.005ClO composite/Cu (mesh)0 °C25 °C40 °C
Resistance, R(Ω) 523 14.1 2.49 
932 9.12 2.57 
1151 9.30 ⋯ 
Average resistance, R(Ω) 869 10.8 2.53 
Conductivity, σ (mS/cm) 0.020 1.61 6.91 
Al/K2.99Ba0.005ClO composite/Cu (foil) 
Resistance, R(Ω) 1057 15.0 1.20 
892 9.81 2.03 
⋯ 14.5 1.85 
Average resistance, R(Ω) 974 13.1 1.69 
Conductivity, σ (mS/cm) 0.018 1.33 10.3 
Al/Na2.99Ba0.005ClO composite/Cu (foil) 
Resistance, R(Ω) 1146 83.7 37.1 
1175 101.4 20.1 
1005 106.2 ⋯ 
Average resistance, R(Ω) 1109 97.1 28.6 
Conductivity, σ (mS/cm) 0.016 0.18 0.61 
Al/K2.99Ba0.005ClO composite/Cu (mesh)0 °C25 °C40 °C
Resistance, R(Ω) 523 14.1 2.49 
932 9.12 2.57 
1151 9.30 ⋯ 
Average resistance, R(Ω) 869 10.8 2.53 
Conductivity, σ (mS/cm) 0.020 1.61 6.91 
Al/K2.99Ba0.005ClO composite/Cu (foil) 
Resistance, R(Ω) 1057 15.0 1.20 
892 9.81 2.03 
⋯ 14.5 1.85 
Average resistance, R(Ω) 974 13.1 1.69 
Conductivity, σ (mS/cm) 0.018 1.33 10.3 
Al/Na2.99Ba0.005ClO composite/Cu (foil) 
Resistance, R(Ω) 1146 83.7 37.1 
1175 101.4 20.1 
1005 106.2 ⋯ 
Average resistance, R(Ω) 1109 97.1 28.6 
Conductivity, σ (mS/cm) 0.016 0.18 0.61 

The permittivity of the asymmetric cells shows a different trend with the temperature when compared with the symmetric cells. It is higher at higher temperatures, as shown in Fig. 4; at 40 °C, the permittivity is >1011 for the Al/K2.99Ba0.005ClO composite/Cu (foil), which is approximately two orders of magnitude lower than the permittivity of the symmetric cell K/K2.99Ba0.005ClO composite/K at the same temperature. The ratio between the ionic conductivities of the symmetric/asymmetric cells is >3 × 103 at 40 °C (see the supplementary material, Tables S1 and S2).

FIG. 4.

Properties obtained from cyclic voltammetry with Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu (foil) ferroelectric asymmetric coaxial structural batteries. (a) Permittivity for Al/K2.99Ba0.005ClO composite/Cu (foil); (b) Permittivity for Al/K2.99Ba0.005ClO composite/Cu (mesh); (c) Permittivity for Al/Na2.99Ba0.005ClO composite/Cu; (d) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (foil) at 40 °C and for the rate of 0.1 mV/s (second cycle); (e) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (foil) at 40 °C and for the rate of 0.1 mV/s (first cycle), and illustrative schematics of a coaxial cell and the direction of the currents while charging at increasing applied voltage (N – number of electrons, I – current; and ē – current of electrons); (f) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (mesh) at 40 °C and for the rate of 0.1 mV/s (first cycle); (g) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (mesh) at 40 °C and for the rate of 0.25 mV/s (second cycle); (h) I–V plot from cyclic voltammetry for Al/Na2.99Ba0.005ClO composite/Cu at 40 °C and for the rate of 0.1 mV/s (first cycle); and (i) I–V plot from cyclic voltammetry for Al/Na2.99Ba0.005ClO composite/Cu at 40 °C and for the rate of 0.1 mV/s (second cycle).

FIG. 4.

Properties obtained from cyclic voltammetry with Al/K2.99Ba0.005ClO composite/Cu (foil and mesh) and Al/Na2.99Ba0.005ClO composite/Cu (foil) ferroelectric asymmetric coaxial structural batteries. (a) Permittivity for Al/K2.99Ba0.005ClO composite/Cu (foil); (b) Permittivity for Al/K2.99Ba0.005ClO composite/Cu (mesh); (c) Permittivity for Al/Na2.99Ba0.005ClO composite/Cu; (d) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (foil) at 40 °C and for the rate of 0.1 mV/s (second cycle); (e) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (foil) at 40 °C and for the rate of 0.1 mV/s (first cycle), and illustrative schematics of a coaxial cell and the direction of the currents while charging at increasing applied voltage (N – number of electrons, I – current; and ē – current of electrons); (f) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (mesh) at 40 °C and for the rate of 0.1 mV/s (first cycle); (g) I–V plot from cyclic voltammetry for Al/K2.99Ba0.005ClO composite/Cu (mesh) at 40 °C and for the rate of 0.25 mV/s (second cycle); (h) I–V plot from cyclic voltammetry for Al/Na2.99Ba0.005ClO composite/Cu at 40 °C and for the rate of 0.1 mV/s (first cycle); and (i) I–V plot from cyclic voltammetry for Al/Na2.99Ba0.005ClO composite/Cu at 40 °C and for the rate of 0.1 mV/s (second cycle).

Close modal

The permittivity depends on the applied external electric field rate reflected on the applied voltage rate, as shown in Fig. 4, and as discussed in Ref. 20, this dependency is due to the possibility of polarizing more efficiently at lower frequencies, such as 0.1 mV/s, than at 50 mV/s.

The I–V plots, or CV curves, of Figs. 4(d) and 4(e) show what is likely to be a cathode’s oxidation reaction at VOC,hollowed circles + 0.15 V, which is displaced from the correspondent reduction reaction of 0.20 V [Fig. 4(c)] corresponding to the difference between the charge and discharge plateau voltage and reflecting the effect of the internal resistance.

Another interesting feature that is not observed in the symmetric cells and only shown in the Al/K2.99Ba0.005ClO composite/Cu is the negative differential resistance (NDR), given by r = dV/dI, as demonstrated in Figs. 4(d)4(g) and in the supplementary material, Fig. S3, and calculated as r = −154 Ω from the slope of the curve between 1.211 and 1.227 V in the CV of Fig. 4(e). The NDR attained while charging and corresponding to a negative current reflects the tunneling of electrons from the electrolyte surface to the negative electrode [Fig. 4(e)] as shown herein later when referring to Poincaré feedback, leading to an increase in the chemical potential of the negative electrode. The NDR is in agreement with the phenomena observed in Fig. 5 when a cell set to discharge connected to a 1.8 kΩ resistor is heated and starts to self-charge [Fig. 5(a)]. Besides these ferroelectric electrolyte-based cells, other single cells showing “real” (not circuit driven) negative differential resistances in I–V curves are the tunnel diodes.61 A tunnel diode or Esaki diode is a type of semiconductor diode that has a negative resistance due to electron tunneling, which is a quantum mechanical effect.

FIG. 5.

Typical curve for the all-solid-state Al/K2.99Ba0.005ClO composite/Cu (mesh) coaxial structural battery that was set to discharge at ∼25 °C with a 1.8 kΩ resistor after 22.8 h from the beginning of the run and then heated to ∼42.5 °C. (a) Full discharge and (b)–(e) details of the periodic potential and temperature vs time.

FIG. 5.

Typical curve for the all-solid-state Al/K2.99Ba0.005ClO composite/Cu (mesh) coaxial structural battery that was set to discharge at ∼25 °C with a 1.8 kΩ resistor after 22.8 h from the beginning of the run and then heated to ∼42.5 °C. (a) Full discharge and (b)–(e) details of the periodic potential and temperature vs time.

Close modal

Non-linear phenomena corresponding to self-oscillations with different periods may arise and vary with the bath’s temperature increase. Conversely, the oscillation of the potential of the cells seems to determine the oscillation of the temperature of the cell [Fig. 5(b)]. After 92.5 h, the lowest oscillation potential was in phase with the lowest oscillation temperature, but the highest potential spike corresponded to halfway up to the highest temperature of the matching peak [Figs. 5(c) and 5(d)]. After 270 h, once only two different periods for the pulsating potential are observed, the potential spike becomes in phase with the temperature peak. When four different periods corresponding to four different shaped peaks could be analyzed—a maximum ΔV = 0.16 V corresponded to a ΔT = 1.7 °C. After 488 h, the potential and temperature synchronized with just the same ∼1 h period observed for both. This latter synchronization is another phenomenon demonstrative of emergence arising in complex systems.62 

It emphasized that self-charging (corresponding to a voltage step-up) corresponds to a NDR, as shown in the CVs of Figs. 4(d)4(g). In Fig. 5(a), a self-charge of ΔV = 0.77 V is observed for the Al/K2.99Ba0.005ClO composite/Cu (mesh) cell after 24.1 h corresponding to ΔTbath/cell = 16 °C, while the cell is set to discharge connected to the 1.8 kΩ resistor. The resistor was connected to the cell after 22.8 h. The output current can be calculated using Ohm’s law V = RextI, where Rext = 1800 Ω. The average current was ∼0.5 mA.

A similar self-charging phenomenon as described in Fig. 5 is shown in Fig. 6 for an Al/Na2.99Ba0.005ClO composite/Cu coaxial cell that had been described in Ref. 20 that has overcome 5330 h self-charging uninterruptedly. The coaxial cell was set to discharge with a 26.6 kΩ resistor 6213 h (8.6 months) ago, and it still shows a potential of 1.43 V [Fig. 6(a)]. After 1104 h, the self-cycling potential amplitude was reduced spontaneously, which is likely to have contributed to the potential rise from 1.28 to 1.53 V, thereafter. With the reduced intensity of the oscillations, less energy is spent transforming a DC phenomenon into an AC, and hysteresis is avoided. For this cell set to discharge with a 26.6 kΩ external resistor, the internal resistance resulting in the Joule effect is overcome by a negative differential resistance due to the feedback electron-current phenomenon that is likely to take place at the surface of the semiconductor ferroelectric-electrolyte, leading to self-charge [Fig. 6(b)]. The latter topologic conduction does not screen the charge accumulation at the electric double-layer capacitors (EDLCs) formed at the interfaces to align the electrode/electrolyte electrochemical potentials or Fermi levels. In other words, the conduction of electrons does not annihilate the EDLCs as the potential, not only is not reduced but increases from 1.06 to 1.53 V. The phenomenon previously described is schematically represented in Fig. 6(b). The two map63 feedback Poincaré model is considered to be a suitable model to describe the processes taking place in the ferroelectric-electrolyte based cells when set to discharge with an electrical load.64 

FIG. 6.

Typical curve for Al/Na2.99Ba0.005ClO composite/Cu coaxial structural battery at ∼40 °C when the battery-cell was set to discharge with a resistor. (a) Na+-electrolyte based coaxial-cell connected to a 26.6 kΩ resistor at t = 0 and a correspondent equivalent circuit after 8.6 months of self-charge; (b) (left) photograph of a demonstrative coaxial cell where the Cu cylinder’s inner radius is 0.6 cm, the Al radius is 0.2 cm, and the electrolyte composite occupies the inner space between Cu and Al; the length of the cell is 10 cm, where 2 cm is thermal insulating glue; (middle) the possible schematic representation of the feedback current accountable for self-charge (blue pathway and arrows), and (right) self-cycle (green and blue arrows) observed when the cells are set to discharge with an electrical load; Poincaré two maps adapted from Ref. 63, a model used to understand the feedback leading to self-charge in the present ferroelectric-electrolyte based cells.64 

FIG. 6.

Typical curve for Al/Na2.99Ba0.005ClO composite/Cu coaxial structural battery at ∼40 °C when the battery-cell was set to discharge with a resistor. (a) Na+-electrolyte based coaxial-cell connected to a 26.6 kΩ resistor at t = 0 and a correspondent equivalent circuit after 8.6 months of self-charge; (b) (left) photograph of a demonstrative coaxial cell where the Cu cylinder’s inner radius is 0.6 cm, the Al radius is 0.2 cm, and the electrolyte composite occupies the inner space between Cu and Al; the length of the cell is 10 cm, where 2 cm is thermal insulating glue; (middle) the possible schematic representation of the feedback current accountable for self-charge (blue pathway and arrows), and (right) self-cycle (green and blue arrows) observed when the cells are set to discharge with an electrical load; Poincaré two maps adapted from Ref. 63, a model used to understand the feedback leading to self-charge in the present ferroelectric-electrolyte based cells.64 

Close modal

At 6213 h, the Al/Na2.99Ba0.005ClO composite/Cu coaxial cell’s [Fig. 6(a)] output capacity was 330 mA h (∼33 mA h/cm2) and the energy was 465 mW h (46.5 mW h/cm2) for a V = 1.41 V and a I = 53.1 µA.

Ab initio simulations allow us to access the structure of unknown materials and their properties. Standing on our previous experience of both simulations and experiments with ferroelectric electrolytes of the A3ClO family with A = Li, Na, the disordered structure of K3ClO at 25 °C (298 K) was simulated, and its PDF, XRD, electron localization function, band structure, and energy of formation were calculated. The structure was first simulated by AMD, as described previously. The same procedure was taken for Na3ClO at 37 °C (310 K) to be able to compare the structures and properties of these electrolytes thought to be amorphous at these temperatures.

By the analyses of Figs. 7(a) and 7(b), it is concluded that the K3ClO electrolyte at 25 °C (298 K) is much more disordered than the Na3ClO at 37 °C (310 K). The distance to the third neighbors in the PDF of K3ClO cannot be identified, but in Na3ClO, it is still distinguishable within a certain broad range of r.

FIG. 7.

Ab initio simulations for K3ClO at 25 °C (298 K) and Na3ClO at 37 °C (310 K). (a) PDF; (b) XRD for CuKα radiation; electron localization function of three (001) Miller plans for (c) and (d) K3ClO and (e) Na3ClO. (c) highlights how the pairs of electrons are conducted through adjacent oxygen anions with the mediating role of the K+ cations.

FIG. 7.

Ab initio simulations for K3ClO at 25 °C (298 K) and Na3ClO at 37 °C (310 K). (a) PDF; (b) XRD for CuKα radiation; electron localization function of three (001) Miller plans for (c) and (d) K3ClO and (e) Na3ClO. (c) highlights how the pairs of electrons are conducted through adjacent oxygen anions with the mediating role of the K+ cations.

Close modal

The electron localization functions may shed light on how the ferroelectric-electrolyte structures become disordered with temperature-forming polymer-like chains of (AO) aligned dipoles.

One of the most intriguing features of the simulated electron localization functions is that observed in the center of Fig. 7(c) and in Fig. 7(d), where a K+ may attract another K+ bending the lattice formed by the (KOK)n chains. This phenomenon is similar to what is thought to be the mechanism subjacent to the conduction of Cooper pairs yielded by a positively charged lattice. In Fig. 7(d), how the oxygen anion’s electron polarization changes extending toward another oxygen anion by the attraction action of the potassium cations is demonstrated. This mechanism may result in a topological current observed in these ferroelectric-electrolytes, leading to self-charge and self-cycling.

The same type of mechanisms should occur in Na3ClO, although this electrolyte might need to be doped with Ba2+ or to be at a higher temperature to have its structural disorder increased. Both electrolytes, K3ClO at 25 °C and Na3ClO at 37 °C, are semiconductors (Table II). It is worth mentioning that the semiconductor and thermoelectric Bi2Te3, which is also a topologic insulator, shows a bandgap of 0.390 eV (DFT).

TABLE II.

Ab initio simulation-based data for K3ClO and Na3ClO.

Energy ofBandgap energy ofBandgap energy of
formationdensity functionalhybrid functionalMaterial and
CompositionTemperature (K)(kJ/mol)Density (g/cm3)theory (DFT) (eV)HSE06 (eV)bandgap type
K3ClO 298 −574.8 2.515 1.601 2.358 Semiconductor direct 
Na3ClO 310 −709.9 2.431 2.292 2.565 Semiconductor direct 
Energy ofBandgap energy ofBandgap energy of
formationdensity functionalhybrid functionalMaterial and
CompositionTemperature (K)(kJ/mol)Density (g/cm3)theory (DFT) (eV)HSE06 (eV)bandgap type
K3ClO 298 −574.8 2.515 1.601 2.358 Semiconductor direct 
Na3ClO 310 −709.9 2.431 2.292 2.565 Semiconductor direct 

The A2.99M0.005ClO (A = Na, K and M = Ba) family of ferroelectric glassy electrolytes was studied employing symmetric coin-cells and asymmetric coaxial electrode-less batteries. The features that stand out are the huge permittivity associated with a very small resistance (high ionic conductivity) in the alkali-metal symmetric cell where the electrolyte is embedded in a cellulose matrix that may also contribute to the polarization of the electrolyte.

Another feature that stands out in this study is the self-charge and self-cycling behavior of the cell containing the K2.99Ba0.005ClO composite, especially for ∼40 °C. This cell can step increase 0.8 V by having the sand/silicone bath to increase its temperature by 17 °C from 25–26 to 42 °C. Moreover, as for self-charge (negative resistance), quantum signatures, such as those expressed on charge/discharge self-cyclings, may reflect emergent phenomena attributed to complex systems.

The symmetric cells’ optimal features set a goal for the optimization of the asymmetric cells. The strategy may pass by using collectors–electrodes that show higher affinity to K and/or Na plating, by synthesizing different composites, and by using pressure to obtain better contact between the electrolyte and the current collector. Another possibility is the use of traditional cathode materials reinforcing the battery character of the cell. A rectification strategy with diodes or capacitors or just 3 min charges will attenuate or avoid the less practical self-cycling phenomena while discharging for 24 h.

See the supplementary material for additional information on the resistance and ionic conductivity of Na/Na2.99Ba0.005ClO in the cellulose membrane/Na and K/K2.99Ba0.005ClO in the cellulose membrane/K symmetric cells at −40, −20, 0, and 25 °C obtained by EIS and CP. In addition, equivalent data on asymmetric cells, such as Al/Na2.99Ba0.005ClO composite/Cu (foil), Al/K2.99Ba0.005ClO composite/Cu (foil), and Al/K2.99Ba0.005ClO composite/Cu (mesh), at RT, 40, and 0 °C are also available. A chronopotentiostatic (CP) plot for the K2.99Ba0.005ClO all-solid state ferroelectric electrolyte in K/K2.99Ba0.005ClO in a cellulose membrane/K, showing self-oscillation only observed at −40 °C, is likewise available as well as 1400 h of the same run. Cyclic voltammetry plots for the Al/K2.99Ba0.005ClO composite/Cu (mesh) at 40 °C, for rates of 0.10 mV/s (five cycles) and 0.25 mV/s (five cycles), showing negative differential resistance and negative current, are also available as complementary information.

M.H.B. thanks Professor John B. Goodenough for endowment to the Mater Lab. F.D., M.V., and M.H.B. acknowledge FCT for the Project UIDP/50022/2020 Emerging Technologies – LAETA and PTDC/QUI-ELT/2593/2021. S.T. acknowledges the Erasmus Staff Mobility for the Training Programme and for funding her travel and stay at FEUP in July 2021.

The authors declare no conflict of interest.

F.D., M.V., and S.T. contributed to cell fabrication and experimental findings. M.H.B. helped in conceptualization, formal analysis, simulations, and overview. All authors contributed to writing, review, and editing and have read and agreed to the published version of the manuscript.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
A.
Crimmins
,
J.
Balbus
,
J. L.
Gamble
,
C. B.
Beard
,
J. E.
Bell
,
D.
Dodgen
,
R. J.
Eisen
,
N.
Fann
,
M. D.
Hawkins
,
S. C.
Herring
,
L.
Jantarasami
,
D. M.
Mills
,
S.
Saha
,
M. C.
Sarofim
,
J.
Trtanj
, and
L.
Ziska
,
The Impacts of Climate Change on Human Health in the United States: A Scientific Assessment
(
U.S. Global Change Research Program
,
Washington, DC
,
2016
).
2.
Y.
Kim
,
W. M.
Seong
, and
A.
Manthiram
, “
Cobalt-free, high-nickel layered oxide cathodes for lithium-ion batteries: Progress, challenges, and perspectives
,”
Energy Storage Mater.
34
,
250
259
(
2021
).
3.
B.
Dunn
,
H.
Kamath
, and
J.-M.
Tarascon
, “
Electrical energy storage for the grid: A battery of choices
,”
Science
334
(
6058
),
928
935
(
2011
).
4.
D.
Aarcher
and
J.-M.
Tarascon
, “
Towards greener and more sustainable batteries for electrical energy storage
,”
Nat. Chem.
7
,
19
29
(
2015
).
5.
K.
Zaghib
,
M.
Dontigny
,
A.
Guerfi
,
P.
Charest
,
I.
Rodrigues
,
A.
Mauger
, and
C. M.
Julien
, “
Safe and fast-charging Li-ion battery with long shelf life for power applications
,”
J. Power Sources
196
,
3949
3954
(
2011
).
6.
X.
Zhu
,
S. V.
Savilov
,
J.
Ni
, and
L.
Li
, “
Carbon nanoflakes as a promising anode for sodium-ion batteries
,”
Funct. Mater. Lett.
11
,
1840011
(
2018
).
7.
Z.
Yang
,
J.
Zhang
,
M. C. W.
Kintner-Meyer
,
X.
Lu
,
D.
Choi
,
J. P.
Lemmon
, and
J.
Liu
, “
Electrochemical energy storage for green grid
,”
Chem. Rev.
111
(
5
),
3577
3613
(
2011
).
8.
M.
Armand
and
J.-M.
Tarascon
, “
Building better batteries
,”
Nature
451
(
7179
),
652
657
(
2008
).
9.
J. M.
Carrasco
,
L. G.
Franquelo
,
J. T.
Bialasiewicz
,
E.
Galvan
,
R. C.
PortilloGuisado
,
M. A. M.
Prats
,
J. I.
Leon
, and
N.
Moreno-Alfonso
, “
Power-electronic systems for the grid integration of renewable energy sources: A survey
,”
IEEE Trans. Ind. Electron.
53
,
1002
1016
(
2006
).
10.
S.
Hameer
and
J. L.
van Niekerk
, “
A review of large-scale electrical energy storage
,”
Int. J. Energy Res.
39
,
1179
1195
(
2015
).
11.
M. S.
Whittingham
, “
Lithium batteries and cathode materials
,”
Chem. Rev.
104
(
10
),
4271
4302
(
2004
).
12.
B.
Scrosati
, “
Recent advances in lithium ion battery materials
,”
Electrochim. Acta
45
(
15–16
),
2461
2466
(
2000
).
13.
N.
Boaretto
,
I.
Garbayo
,
S.
Valiyaveettil-SobhanRaj
,
A.
Quintela
,
C.
Li
,
M.
Casas-Cabanas
, and
F.
Aguesse
, “
Lithium solid-state batteries: State-of-the-art and challenges for materials, interfaces and processing
,”
J. Power Sources
502
,
229919
(
2021
).
14.
Y.
Hamedi Jouybari
,
F.
Berkemeier
, and
G.
Schmitz
, “
High performance all-solid-state lithium battery: Assessment of the temperature dependence of Li diffusion
,”
J. Power Sources
517
,
230709
(
2022
).
15.
J.
Li
,
C.
Ma
,
M.
Chi
,
C.
Liang
, and
N. J.
Dudney
, “
Solid electrolyte: The key for high-voltage lithium batteries
,”
Adv. Energy Mater.
5
,
1401408
(
2015
).
16.
J.-M.
Tarascon
and
M.
Armand
, “
Issues and challenges facing rechargeable lithium batteries
,”
Nature
414
,
359
367
(
2001
).
17.
H.
Vikström
,
S.
Davidsson
, and
M.
Höök
, “
Lithium availability and future production outlooks
,”
Appl. Energy
110
,
252
266
(
2013
).
18.
S.-W.
Kim
,
D.-H.
Seo
,
X.
Ma
,
G.
Ceder
, and
K.
Kang
, “
Electrode materials for rechargeable sodium-ion batteries: Potential alternatives to current lithium-ion batteries
,”
Adv. Energy Mater.
2
,
710
721
(
2012
).
19.
H.
Kim
,
H.
Kim
,
Z.
Ding
,
M. H.
Lee
,
K.
Lim
,
G.
Yoon
, and
K.
Kang
, “
Recent progress in electrode materials for sodium-ion batteries
,”
Adv. Energy Mater.
6
,
1600943
(
2016
).
20.
F.
Danzi
,
P. P.
Camanho
, and
M. H.
Braga
, “
An all-solid-state coaxial structural battery using sodium-based electrolyte
,”
Molecules
26
,
5226
(
2021
).
21.
X.
Feng
,
M.
Ouyang
,
X.
Liu
,
L.
Lu
,
Y.
Xia
, and
X.
He
, “
Thermal runaway mechanism of lithium ion battery for electric vehicles: A review
,”
Energy Storage Mater.
10
,
246
267
(
2018
).
22.
Q.
Wang
,
P.
Ping
,
X.
Zhao
,
G.
Chu
,
J.
Sun
, and
C.
Chen
, “
Thermal runaway caused fire and explosion of lithium ion battery
,”
J. Power Sources
208
,
210
224
(
2012
).
23.
A.
Iturrondobeitia
,
F.
Aguesse
,
S.
Genies
,
T.
Waldmann
,
M.
Kasper
,
N.
Ghanbari
,
M.
Wohlfahrt-Mehrens
, and
E.
Bekaert
, “
Post-mortem analysis of calendar-aged 16 Ah NMC/graphite pouch cells for EV application
,”
J. Phys. Chem. C
121
(
40
),
21865
21876
(
2017
).
24.
H.
Pan
,
Y.-S.
Hu
, and
L.
Chen
, “
Room-temperature stationary sodium-ion batteries for large-scale electric energy storage
,”
Energy Environ. Sci.
6
(
8
),
2338
2360
(
2013
).
25.
G. L.
Xu
,
R.
Amine
,
A.
Abouimrane
,
H.
Che
,
M.
Dahbi
,
Z. F.
Ma
,
I.
Saadoune
,
J.
Alami
,
W. L.
Mattis
,
F.
Pan
,
Z.
Chen
, and
K.
Amine
, “
Challenges in developing electrodes, electrolytes, and diagnostics tools to understand and advance sodium-ion batteries
,”
Adv. Energy Mater.
8
(
14
),
1702403
(
2018
).
26.
Y.
Wang
,
S.
Song
,
C.
Xu
,
N.
Hu
,
J.
Molenda
, and
L.
Lu
, “
Development of solid-state electrolytes for sodium-ion battery—A short review
,”
Nano Mater. Sci.
1
,
91
100
(
2019
).
27.
X.
Zhang
,
J.
Meng
,
X.
Wang
,
Z.
Xiao
,
P.
Wu
, and
L.
Mai
, “
Comprehensive insights into electrolytes and solid electrolyte interfaces in potassium-ion batteries
,”
Energy Storage Mater.
38
,
30
49
(
2021
).
28.
H.
Sun
,
P.
Liang
,
G.
Zhu
,
W. H.
Hung
,
Y.-Y.
Li
,
H.-C.
Tai
,
C.-L.
Huang
,
J.
Li
,
Y.
Meng
,
M.
Angell
,
C.-A.
Wang
, and
H.
Dai
, “
A high-performance potassium metal battery using safe ionic liquid electrolyte
,”
Proc. Natl. Acad. Sci. U. S. A.
117
,
27847
27853
(
2020
).
29.
J.-Y.
Hwang
,
S.-T.
Myung
, and
Y.-K.
Sun
, “
Recent progress in rechargeable potassium batteries
,”
Adv. Funct. Mater.
28
,
1802938
(
2018
).
30.
H.
Kim
,
J. C.
Kim
,
M.
Bianchini
,
D. H.
Seo
,
J.
Rodriguez-Garcia
, and
G.
Ceder
, “
Recent progress and perspective in electrode materials for K-ion batteries
,”
Adv. Energy Mater.
8
,
1702384
(
2018
).
31.
T.
Masese
,
K.
Yoshii
,
Y.
Yamaguchi
,
T.
Okumura
,
Z.-D.
Huang
,
M.
Kato
,
K.
Kubota
,
J.
Furutani
,
Y.
Orikasa
,
H.
Senoh
,
H.
Sakaebe
, and
M.
Shikano
, “
Rechargeable potassium-ion batteries with honeycomb-layered tellurates as high voltage cathodes and fast potassium-ion conductors
,”
Nat. Commun.
9
,
3823
(
2018
).
32.
H.
Wang
,
D.
Zhai
, and
F.
Kang
, “
Solid electrolyte interphase (SEI) in potassium ion batteries
,”
Energy Environ. Sci.
13
(
12
),
4583
4608
(
2020
).
33.
W.
Zhang
,
Y.
Liu
, and
Z.
Guo
, “
Approaching high-performance potassium-ion batteries via advanced design strategies and engineering
,”
Sci. Adv.
5
(
5
),
eaav7412
(
2019
).
34.
S.
Komaba
,
T.
Hasegawa
,
M.
Dahbi
, and
K.
Kubota
, “
Potassium intercalation into graphite to realize high-voltage/high-power potassium-ion batteries and potassium-ion capacitors
,”
Electrochem. Commun.
60
,
172
175
(
2015
).
35.
T.
Hosaka
,
K.
Kubota
,
A. S.
Hameed
, and
S.
Komaba
, “
Research development on K-ion batteries
,”
Chem. Rev.
120
(
14
),
6358
6466
(
2020
).
36.
M.
Niro
,
U.
Kisaburo
, and
T.
Zen’ichi
, “
Standard potentials of alkali metals, silver, and thallium metal/ion couples in N,N′-dimethylformamide, dimethyl sulfoxide, and propylene carbonate
,”
Bull. Chem. Soc. Jpn.
47
(
4
),
813
817
(
1974
).
37.
M.
Okoshi
,
Y.
Yamada
,
S.
Komaba
,
A.
Yamada
, and
H.
Nakai
, “
Theoretical analysis of interactions between potassium ions and organic electrolyte solvents: A comparison with lithium, sodium, and magnesium ions
,”
J. Electrochem. Soc.
164
,
A54
A60
(
2017
).
38.
V.
Anoopkumar
,
J.
Bibin
, and
T. D.
Mercy
, “
Potassium-ion batteries: Key to future large-scale energy storage?
,”
ACS Appl. Energy Mater.
3
(
10
),
9478
9492
(
2020
).
39.
Y.
Mizutani
,
T.
Abe
,
K.
Ikeda
,
E.
Ihara
,
M.
Asano
,
T.
Harada
,
M.
Inaba
, and
Z.
Ogumi
, “
Graphite intercalation compounds prepared in solutions of alkali metals in 2-methyltetrahydrofuran and 2,5-dimethyltetrahydrofuran
,”
Carbon
35
(
1
),
61
65
(
1997
).
40.
F.
Danzi
,
R. M.
Salgado
,
J. E.
Oliveira
,
A.
Arteiro
,
P. P.
Camanho
, and
M. H.
Braga
, “
Structural batteries: A review
,”
Molecules
26
,
2203
(
2021
).
41.
R.
Harnden
,
D.
Zenkert
, and
G.
Lindbergh
, “
Potassium-insertion in polyacrylonitrile-based carbon fibres for multifunctional energy storage, morphing, and strain-sensing
,”
Carbon
171
,
671
680
(
2021
).
42.
R.
Harnden
,
K.
Peuvot
,
D.
Zenkert
, and
G.
Lindbergh
, “
Multifunctional performance of sodiated carbon fibers
,”
J. Electrochem. Soc.
165
,
B616
B622
(
2018
).
43.
L. E.
Asp
,
M.
Johansson
,
G.
Lindbergh
,
J.
Xu
, and
D.
Zenkert
, “
Structural battery composites: A review
,”
Funct. Compos. Struct.
1
(
4
),
042001
(
2019
).
44.
P.
Thomas
and
D.
Billaud
, “
Sodium electrochemical insertion mechanisms in various carbon fibres
,”
Electrochim. Acta
46
(
22
),
3359
3366
(
2001
).
45.
M.
Dubois
,
J.
Ghanbaja
, and
D.
Billaud
, “
Electrochemical intercalation of sodium ions into poly(para-phenylene) in carbonate-based electrolytes
,”
Synth. Met.
90
(
2
),
127
134
(
1997
).
46.
T.
Pereira
,
Z.
Guo
,
S.
Nieh
,
J.
Arias
, and
H. T.
Hahn
, “
Energy storage structural composites: A review
,”
J. Compos. Mater.
43
(
5
),
549
560
(
2009
).
47.
Q.
Pan
,
D.
Gong
, and
Y.
Tang
, “
Recent progress and perspective on electrolytes for sodium/potassium-based devices
,”
Energy Storage Mater.
31
,
328
343
(
2020
).
48.
Y.
Liu
,
C.
Gao
,
L.
Dai
,
Q.
Deng
,
L.
Wang
,
J.
Luo
,
S.
Liu
, and
N.
Hu
, “
The features and progress of electrolyte for potassium ion batteries
,”
Small
16
,
2004096
(
2020
).
49.
M. H.
Braga
,
A. J.
Murchison
,
J. A.
Ferreira
,
P.
Singh
, and
J. B.
Goodenough
, “
Glass-amorphous alkali-ion solid electrolytes and their performance in symmetrical cells
,”
Energy Environ. Sci.
9
,
948
954
(
2016
).
50.
M. H.
Braga
,
J. E.
Oliveira
,
T.
Kai
,
A. J.
Murchison
,
A. J.
Bard
, and
J. B.
Goodenough
, “
Extraordinary dielectric properties at heterojunctions of amorphous ferroelectrics
,”
J. Am. Chem. Soc.
140
(
51
),
17968
17976
(
2018
).
51.
T. E.
Faber
,
An Introduction to the Theory of Liquid Metals
(
Cambridge University Press
,
1972
).
52.
G.
Kresse
and
J.
Furthmüller
, “
Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set
,”
Phys. Rev. B
54
,
11169
11186
(
1996
).
53.
P. E.
Blochl
, “
Projector augmented-wave method
,”
Phys. Rev. B
50
,
17953
17979
(
1994
).
54.
J. P.
Perdew
and
Y.
Wang
, “
Accurate and simple analytic representation of the electron-gas correlation energy
,”
Phys. Rev. B
45
,
13244
13249
(
1992
).
55.
M.
Hoffmann
and
S.
Salahuddin
, “
Ferroelectric gate oxides for negative capacitance transistors
,”
MRS Bull.
46
,
930
(
2021
).
56.
M. H.
Braga
,
A. J.
Murchison
,
J. E.
Oliveira
, and
J. B.
Goodenough
, “
Low-temperature performance of a ferroelectric glass electrolyte rechargeable cell
,”
ACS Appl. Energy Mater.
2
(
7
),
4943
4953
(
2019
).
57.
M. H.
Braga
,
N. S.
Grundish
,
A. J.
Murchison
, and
J. B.
Goodenough
, “
Thermodynamic considerations of same-metal electrodes in an asymmetric cell
,”
Mater. Theory
3
(
1
),
1
(
2019
).
58.
M. H.
Braga
,
C. M.
Subramaniyam
,
A. J.
Murchison
, and
J. B.
Goodenough
, “
Nontraditional, safe, high voltage rechargeable cells of long cycle life
,”
J. Am. Chem. Soc.
140
(
20
),
6343
6352
(
2018
).
59.
C.
Yang
 et al, “
Copper-coordinated cellulose ion conductors for solid-state batteries
,”
Nature
598
,
590
596
(
2021
).
60.
M. H.
Braga
,
J. E.
Oliveira
,
A. J.
Murchison
, and
J. B.
Goodenough
, “
Performance of a ferroelectric glass electrolyte in a self-charging electrochemical cell with negative capacitance and resistance
,”
Appl. Phys. Rev.
7
(
1
),
011406
(
2020
).
61.
L.
Esaki
, “
New phenomenon in narrow germanium p–n junctions
,”
Phys. Rev.
109
(
2
),
603
604
(
1958
).
62.
S.
Kivelson
and
S. A.
Kivelson
, “
Defining emergence in physics
,”
npj Quantum Mater.
1
,
16024
(
2016
).
63.
S.
Varigonda
and
T. T.
Georgiou
, “
Dynamics of relay relaxation oscillators
,”
IEEE Trans. Autom. Control
46
(
1
),
65
(
2001
).
64.
M. H.
Braga
, “
Coherence in the ferroelectric A3ClO (A = Li, Na) family of electrolytes
,”
Materials
14
(
9
),
2398
(
2021
).

Supplementary Material