Novel Proton-Conducting Layered Perovskites Based on BaLa₂In₂O₇ Produced by Cationic Co-Doping

Abstract

Proton conducting materials are used in electrochemical devices such as proton conducting fuel cells and proton conducting electrolyzers. These devices belong to the hydrogen energy field and serve the goals of clean energy and sustainable environmental development. Layered perovskites are a promising class of proton conducting electrolytes. Cationic co-doping is a well-known method to improve the transport properties of classical perovskite ABO₃. However, data on the application of this method to layered perovskites are limited. In this work, the bilayered perovskites BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x have been prepared and studied for the first time. The possibility of oxygen-ionic and proton transport was demonstrated. Cationic co-doping was shown to increase the proton conductivity values by up to 1.5 orders of magnitude.

1. Introduction

Proton conductivity in solid state materials is a required property for the electrolytic components of electrochemical devices such as proton conducting fuel cells (PCFCs) and proton conducting electrolyzers (PCECs) [1,2,3,4,5,6,7,8]. These devices belong to the hydrogen energy field and serve the goals of clean energy and sustainable environmental development [9,10,11,12,13,14,15,16]. The efficiency of PCFCs and PCECs depends on many factors, including parameters such as the magnitude of the proton conductivity of the electrolyte material. There are several strategies to improve conductivity, one of which is doping, including cationic co-doping. The co-doping (double doping) method worked well for the most studied proton conductors such as barium cerate [2,17,18,19,20,21] and lanthanum scandate [22,23,24,25,26,27,28,29,30,31]. However, data on the application of this method to improve the transport properties of proton conducting layered perovskites are limited.

The layered perovskites AA′_nB_nO_{3n + 1} are a structural class derived from the classical perovskite ABO₃, where the perovskite blocks (A′BO₃)_n are separated by the rock salt layers AO [32]. The ionic (oxygen-ionic) conductivity of this type of material was first opened up about 10 years ago by Fujii et al. and by Troncoso et al. for compositions based on BaNdInO₄ [33,34,35,36,37] and SrLaInO₄ [38,39,40], respectively. The possibility of proton conductivity in layered perovskites was opened in 2019 for the compositions based on BaLaInO₄ [41]. A few years later it was demonstrated for many materials based on Ba(Sr)La(Nd)In(Sc)O₄ [42,43,44,45,46]. These compositions can be described by the general formula AA′BO₄ and are called monolayerd perovskites (n = 1). Last year (2022), bilayerd perovskites AA′₂B₂O₇ (n = 2) based on BaLa₂In₂O₇ [47,48], BaNd₂In₂O₇ [49] and SrLa₂Sc₂O₇ [50] were investigated in terms of oxygen ionic and proton conductivity. Cationic doping was found to increase the proton conductivity values by up to 1.5 orders of magnitude [12]. However, the possibility of a co-doping strategy applied to bilayerd perovskites has not yet been investigated. In this work we have carried out double isovalent Gd³⁺→La³⁺ and acceptor Sr²⁺→La³⁺ doping in the cationic sublattice of the bilayerd perovskite BaLa₂In₂O₇. The influence of doping on the crystal lattice, the possibility of water uptake and the proton conductivity were revealed.

2. Experimental Procedure

The compositions BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x were prepared by the solid state method. The powders of the starting reagents BaCO₃, SrCO₃, La₂O₃, Gd₂O₃, In₂O₃ were dried and used in stoichiometric amounts. An agate mortar was used for grinding. The compositions were heated after each grinding. The annealing was carried out in the temperature range of 800, 900, 1100, 1200 and 1300 °C.

The phase identification of the obtained compositions was carried out using the Bruker Advance D8 Cu Kα diffractometer. The scanning electron microscope VEGA3 TESCAN equipped with the system for energy dispersive X-ray spectroscopy was used to define the morphology and chemical composition of the samples. The thermogravimetric and mass spectrometric investigations were carried out using the NETZSCH STA 409 PC analyzer equipped with the NETZSCH QMS 403C Aëolos mass spectrometer. Initially hydrated samples were used. The hydrated samples were obtained by slow cooling from 1100 to 150 °C (1 °C/min) under a flow of wet Ar (pH₂O = 2 × 10⁻² atm).

The electrical conductivity was measured with an impedance spectrometer Z-1000P, Elins, RF. The ceramic pellets with a 10 mm diameter and 2 mm thickness were pressed for the investigations. Pt-electrodes were applied on the surfaces of the samples. The investigations were carried out from 1000 to 200 °C with a cooling rate of 1^°/min under dry air or dry Ar. The dry gas (air or Ar) was prepared by circulating the gas through P₂O₅ (pH₂O = 3.5 × 10⁻⁵ atm). The wet gas (air or Ar) was obtained by bubbling the gas at room temperature first through distilled water and then through a saturated solution of KBr (pH₂O = 2 × 10⁻² atm).

3. Results and Discussions

The homogeneity range of the solid solution BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x was defined by XRD analysis. The full profile Le Bail refinement of the obtained data showed that compositions with x ≤ 0.15 are single phase and have orthorhombic symmetry, Pbca space group (Figure 1). All compositions were isostructural to the matrix composition BaLa₂In₂O₇, which has the structure of bilayer perovskite [51]. The higher dopant content x > 0.15 led to the appearance of impurities.

Table 1. Lattice parameters, unit cell volume and water uptake for the solid solution BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x.

Sample	a, b (Å)	c (Å)	Vcell (Å3)	Water Uptake (mol)
0	5.914(9)	20.846(5)	729.3365	0.17
0.05	5.918(5)	20.848(9)	730.3086	0.16
0.10	5.919(2)	20.853(6)	730.6461	0.17
0.15	5.920(3)	20.861(0)	731.1770	0.18

shows the lattice parameters of the compositions studied. As shown, doping led to an increase in lattice parameters and unit cell volume. The comparison of these data with the results for the related compositions obtained by doping the La³⁺ sublattice with Sr²⁺ ions BaLa_2−xSr_xIn₂O_7−0.5x [47] and Gd³⁺ ions BaLa_2−xGd_xIn₂O₇ [48] (Figure 1e) shows that the largest increase in parameters during doping is achieved for the acceptor-doped solid solution BaLa_2−xSr_xIn₂O_7−0.5x, where the ionic radii of the dopants are larger ($r_{{\mathrm{La}}^{3+}}$ = 1.216 Å; ${\mathrm{r}}_{{\mathrm{Sr}}^{2+}}$ = 1.31 Å; $r_{{\mathrm{Gd}}^{3+}}$ = 1.107 Å [52]). However, there is no direct correlation between the dopant radius and the change in the lattice parameters. The introduction of Gd³⁺ ions into the La³⁺ sublattice led to an increase in the interlayer space (lattice parameter c) due to the appearance of local distortions caused by the effects of mutual repulsion of ions of different electronegativity in the same sublattice [37]. It is obvious that the presence of three different cations also leads to the local distortion of the crystal lattice, and this is the reason for the significant decrease in the lattice parameter c for the solid solution BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x compared to BaLa_2−xSr_xIn₂O_7−0.5x. Therefore, the increasing in the lattice parameter c is observed for all three solid solutions, but this increasing is different.

Morphological analysis of the compositions was carried out using scanning electron microscopy (SEM). The diameter of the grains was about ~2–5 μm (Figure 2). Energy dispersive X-ray spectroscopy (EDS) analysis confirmed the presence of all cations in the samples and the agreement of their ratio with the theoretical values (

Table 2. The average element ratios determined by EDS analysis for the samples BaLa₂In₂O₇ (1), BaLa_1.85Sr_0.05Gd_0.1In₂O_6.975 (2), BaLa_1.8Sr_0.1Gd_0.1In₂O_6.95 (3), BaLa_1.75Sr_0.15Gd_0.1In₂O_6.925 (4) (theoretical values are given in the brackets).

Composition	Content of Element, Atomic %
	Ba	La	In	Sr	Gd
1	20.2 (20.0)	40.1 (40.0)	39.7 (40.0)	-	-
2	20.1 (20.0)	37.3 (37.0)	39.8 (40.0)	0.9 (1.0)	1.9 (2.0)
3	20.3 (20.0)	36.0 (36.0)	39.8 (40.0)	1.9 (2.0)	2.0 (2.0)
4	20.2 (20.0)	35.3 (35.0)	39.7 (40.0)	2.8 (3.0)	2.0 (2.0)

).

The electrical conductivities were obtained using the impedance spectroscopy method. Figure 3 shows the typical EIS plots for the sample BaLa_1.75Sr_0.15Gd_0.1In₂O_6.925. The semicircle starting from zero coordinates (high frequency semicircle) corresponds to the resistivity of the grain volume of polycrystalline sample. This is proved by the small capacitance value ~ 10⁻¹² F/cm. The conductivities were calculated from the resistivity values taken at the intersection of the high-frequency semicircle with the abscissa axis. The temperature dependencies of the conductivity obtained under dry (pH₂O = 3.5·10⁻⁵ atm) and wet (pH₂O = 2·10⁻²) air (pO₂ = 0.21 atm) and Ar (pO₂ ~10⁻⁵ atm) are shown in Figure 4.

As can be seen, the changes in atmospheric humidity (pH₂O variation) and oxygen partial pressure do not affect the general shape and regularities of the conductivity values changes. The increase in the dopant concentration leads to the increase of the electrical conductivity values (Figure 5a). Figure 5b shows the concentration dependence of the conductivity in dry Ar, i.e., at lower oxygen partial pressure, where the conductivity nature is predominantly oxygen-ionic. The conductivity increases with increasing strontium content, which is explained by an increase in the concentration of oxygen vacancies:

$$2\mathrm{SrO} \stackrel{{\mathrm{La}}_2{\mathrm{O}}_3}{\to }2{\mathrm{Sr}}_{\mathrm{La}}^{\prime }+2{\mathrm{O}}_{\mathrm{o}}^{\times }+{\mathrm{V}}_{\mathrm{o}}^{••}$$

(1)

where ${\mathrm{Sr}}_{\mathrm{La}}^{\prime }$ is strontium ion in lanthanum sites; ${\mathrm{V}}_{\mathrm{o}}^{••}$ is the oxygen vacancy; ${\mathrm{O}}_{\mathrm{o}}^{\times }$ is the oxygen atom in the regular position.

The comparison of the oxygen-ionic conductivity of the solid solutions BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x, BaLa_2−xSr_xIn₂O_7−0.5x and BaLa_2−xGd_xIn₂O₇ at 500 °C is shown in Figure 5c. As can be seen, the conductivity value for the co-doped composition BaLa_1.8Sr_0.1Gd_0.1In₂O_6.95 is higher than for the isovalent-doped composition BaLa_1.9Gd_0.1In₂O₇ with the same gadolinium concentration. The reason for this is the higher concentration of ionic charge carriers (oxygen ions) due to the formation of oxygen vacancies during acceptor doping. At the same time, the conductivity values for the acceptor-doped BaLa_2−xSr_xIn₂O_7−0.5x compositions are close to the values for the co-doped BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x compositions, except for the compositions with x = 0.1. The conductivity values for the BaLa_1.9Sr_0.1In₂O_6.95 composition are higher than for the BaLa_1.8Sr_0.1Gd_0.1In₂O_6.95 composition at the same oxygen vacancy concentration. It is clear that the conductivity values depend on both geometric and concentration factors. Because the lattice parameters and interlayer space are larger for the acceptor-doped composition than for the co-doped composition, the mobility of oxygen ions may be higher for the BaLa_1.9Sr_0.1In₂O_6.95 composition than for the BaLa_1.8Sr_0.1Gd_0.1In₂O_6.95 composition.

The proton conductivity values were obtained as the difference between the ionic conductivity under wet conditions (wet Ar) and dry conditions (dry Ar). The concentration dependence of the protonic conductivity for the compositions BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x is shown in Figure 6a. As can be seen, the protonic conductivity increases with increasing dopant concentration. For the correct analysis of these dependencies, the values of the proton concentrations are required. Figure 6b shows the results of thermogravimetry (TG), mass spectrometry (MS) and differential scanning calorimetry (DSC) analyses for BaLa_1.8Sr_0.1Gd_0.1In₂O_6.95 as an example. Mass loss occurs at the temperatures below 400 °C (TG curve) and is solely due to the release of water (MS(H₂O) curve). The values of water uptake were close for all doped BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x and undoped BaLa₂In₂O₇ compositions and are about ~0.17 mol of water per formula unit (Table 1). In other words, the proton concentration ($c_{H^{+}}$) was close for all compositions. Consequently, the increase in proton conductivity (${\sigma }_{H^{+}}$) with increasing x in the formula BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x is due to the increase in proton mobility (${\mu }_{H^{+}}$):

$${\sigma }_{H^{+}}=z\times e·{\mu }_{H^{+}}\times c_{H^{+}}$$

(2)

The proton conductivity value for the most conductive co-doped composition BaLa_1.75Sr_0.15Gd_0.1In₂O_6.925 is 8∙10⁻⁶ S/cm at 400 °C. The conductivity values for the undoped BaLa₂In₂O₇ composition, the isovalent-doped BaLa_1.9Gd_0.1In₂O₇ composition and the acceptor-doped BaLa_1.75Sr_0.15In₂O_6.925 composition were 0.3∙10⁻⁶, S/cm, 2∙10⁻⁶ S/cm and 15∙10⁻⁶ S/cm, respectively. Thus, the acceptor doped composition was more conductive than the co-doped one. However, co-doping increases the proton conductivity values by up to 1.5 orders of magnitude.

4. Conclusions

The bilayered perovskites BaLa_1.9−xSr_xGd_0.1In₂O_7−0.5x have been prepared and studied for the first time. The phase attestation, morphology, element content, possibility of water uptake and electrical conductivity were investigated and discussed. The possibility of oxygen-ionic and proton transport was demonstrated. It was shown that cationic co-doping led to an increase in proton conductivity values of up to 1.5 orders of magnitude. The cationic co-doping strategy is a promising way to improve the transport properties of bilayer perovskites. The selection of optimal dopants may be the task of future studies.