Phosphorus-doped protonic conductors based on BaLa n In n O 3n+1 (n = 1, 2): applying oxyanion doping strategy to the layered perovskite structures

The creation of highly efficient and eco-friendly energy sources such as hydrogen energy systems is one of main vectors for the sustainable development of human society. Proton-conducting ceramic materials can be applied as one of the main components of such hydrogen-fueled electrochemical devices, including protonic ceramic fuel cells. The oxyanion doping strategy is a promising approach for improving transport properties of proton-conducting complex oxides. In this paper, this strategy was applied to proton-conducting layered perovskites for the first time. The phosphorus-doped protonic conductors based on BaLa n In n O 3n+1 (n = 1, 2) were obtained, and their electrical conductivity was thoroughly investigated. It was found that the phosphorous doping leads to an increase in the electrical conductivity values by ~0.7 orders of magnitude.


Introduction
The creation of high-efficiency and eco-friendly energy source is one of main objectives for the sustainable global development of human society [1−8]. Hydrogen energy belongs to the renewable energy industry and includes the systems for storage, transport and using of hydrogen for power generation [9−12]. Proton-conducting ceramic materials can be applied as the one of main component of such hydrogen-based electrochemical devices for various purposes, including electricity generation in protonic ceramic fuel cells, PCFCs [13−25]. The most studied protonic conductors have perovskite or perovskite-related structures [26−30]. Doping of cationic sublattices is a common way for improving their transport properties. However, the anion [31−35] and oxyanion [36,37] doping methods can increase proton conductivity in the complex oxides as well. The oxyanion doping strategy is based on the displacement of the [BO6] octahedra to the [B'O4] tetrahedra such as phosphate, sulphate and silicate ( Figure 1). Slater et al. proved the validity of this strategy for the protonconducting materials, studying barium indate, Ba2In2O5, as an example [36]. This confirms that substitution [PO4] → [InO6] is fundamentally possible, and the proton conductivity in such compositions can be improved by phosphorus doping. Barium lanthanum indates, BaLaInO4 and BaLa2In2O7, have a layered perovskite structure and can be written using a general formula, BaLanInnO3n+1 (n = 1, 2). They belong to the newly opened class of proton-conducting solid oxide materials [38−49]. It was proved that they are nearly pure (~95-98 %) protonic conductors under wet air below 350-400 °C [50]. Different ways of cationic (iso-and heterovalent) doping lead to increasing the protonic conductivity up to ~1.5 orders of magnitude (from 2•10 -7 S cm -1 for BaLaInO4 to 8•10 -6 S cm -1 for Ba1.1LaInO3.95 at 400 °C) [51−56]. Based on this fact, the other doping strategies, such as oxyanion (phosphorus) doping, can be applied to these materials. The reason of this materials search is necessity to create highconductive proton conductors with the layered perovskite structure because the promising cathode materials based on nickelates lanthanides [57−60] belong to the layered perovskites as well.
In the present study, the oxyanion doping strategy was applied to the proton-conducting layered perovskites for the first time. The phosphorus-doped protonic conductors based on BaLanInnO3n+1 (n = 1, 2) were obtained, and electrical conductivity of ceramic samples was investigated.

Experimental
The complex oxides of BaLaIn0.9P0.1O4.1 and BaLa2In1.9P0.1O7.1 were obtained by a solid state method. Firstly, high-purity starting powder materials were dried and the stoichiometric amounts of the reagents were weighed on a Sartorius analytical balances (Goettingen, Germany). The chemical reactions can be presented in as: Further, the milling of all reagents in an agate mortar followed by calcination of the obtained mixtures was made. The calcination was performed in a temperature range from 800 to 1300 °С with a step of 100 °С and 24 h of time treatments.
The X-ray diffraction (XRD) studies were performed by a Bruker Advance D8 diffractometer (Rheinstetten, Germany) with a Cu Kα radiation with a step of 0.01 o and at a scanning rate of 0.5 o min -1 . The morphology and chemical composition of the samples were studied using a Phenom ProX Desktop scanning electron microscope (Waltham, MA, USA) (SEM) integrated with an energy-dispersive Xray diffraction (EDS) detector.
For the investigations of the electrical properties, the pressed cylindrical pellets (1300 °C, 24 h, dry air) were obtained. The samples had a relative density of ~90% (density of the sintered samples was determined by the Archimedes method). The AC conductivity measurements were performed by a Z-1000P (Elins, RF) impedance spectrometer within a frequency range of 1-10 6 Hz. Electrical measurements were performed using Pt paste electrodes (sintering at 1000 °C for 2 h). The temperature dependencies of electrical conductivity were obtained in a temperature range 200-1000 °C (step 10-20 °C, 1 °C min -1 cooling rate). These investigations were performed under "dry" and "wet" air atmospheres. The dry air was produced by circulating the gas through P2O5 (pH2O = 3.5·10 −5 atm). The wet air 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). The humidity of the gas was controlled by a Honeywell HIH-3610 H2O-sensor (Freeport, USA).

Results and discussions
The XRD analysis of the powder samples BaLaIn0.  Table 1.   As can be seen, phosphorous-doping leads to a change in these characteristics for both doped compositions compared with undoped. The oxyanion doping for the monolayer composition of BaLaInO4 leads to a decrease in the a parameter and to an increase in the b and c parameters.
The applying of this doping strategy to the two-layered compositions of BaLa2In2O7 leads to an increase of all (a, b and c) lattice parameters. As it is known [61], the ionic radius of phosphorous is smaller than ionic radius of indium (r(P 5+ ) = 0.38 Å, r(In 3+ ) = 0.8 Å). However, the displacement of [InO6] octahedra to the [PO4] tetrahedra should inevitably lead to the appearance of local distortions and to a redistribution of bond lengths in the crystal structure. The microphotography (SEM-image) of the BaLa2In1.9P0.1O7.1 powder sample is presented in the inset of Figure 3. This composition consists of grains ~5 μm, forming agglomerates of ~15−30 μm.
The electrical conductivity was measured by the impedance spectroscopy method. The Nyquist-plots for BaLaIn0.9P0.1O4.1 composition obtained under dry air are presented in the Figure 4a, b. The fitting of the spectra was made using ZView software, and the obtained results are presented in the Table 2. According to the fitting of the spectra (red line) with using the equivalent circuit presented in the Figure 4c, three different electrochemical processes can be defined. As it was shown earlier [51], the Nyquist-plots for undoped BaLaInO4 composition were represented by one visible semicircle with a capacitance of around 10 -11 F. For the calculation of electrical conductivity, the bulk resistance values (R1) were used and discussed below. It can be noted, that due to a small depression of the semicircles, the constant phase element (CPE) was used during the analysis of Nyquist plots.
The results of the electrical conductivity investigations are presented in the Figure 5. As can be seen, phosphorous-doping leads to an increase in the conductivity values for both monolayer (BaLaInO4) and two-layered (BaLa2In2O7) compositions and. The conductivity growth is about 0.7 orders of magnitude for both compositions. We can assume that such increasing electrical conductivity is due to two factors. Firstly, an increase in the lattice parameters for the layered perovskites of BaLanInnO3n+1 results in a higher conductivity due to facilitating ionic transport [50]. Secondly, the phosphorous-doping can be considered as a donor doping (P 5+ → In 3+ ) that causes the appearance of interstitial ("additional") oxygen in the structure. It is obvious that an increase in the concentration of charge carriers (oxygen ions) should lead to the corresponding increase in the conductivity as well.
The change in atmospheric humidity also affects the electrical conductivity values ( Figure 5). The air humudification leads to an increase in the conductivity values at low temperatures (450 °C).   Because layered perovskites BaLaInO4 and BaLa2In2O7 are capable for the dissociative absorption of water from the gas phase [50], the reason of better conductivity is the appearance of proton contribution of conductivity. It can be concluded that the oxyanion doping strategy can be applied for layered perovskites for improving their transport properties.

Conclusions
In this paper, the oxyanion doping strategy was purposefully applied to the proton-conducting layered perovskites for the first time. The phosphorus-doped protonic conductors based on BaLanInnO3n+1 (n = 1, 2) were obtained, and their electrical properties were investigated. The BaLaIn0.9P0.1O4.1 and BaLa2In1.9P0.1O7.1 oxides were obtained for the first time. It was found that the phosphorousdoping leads to an increase in the electrical conductivity values by ~0.7 orders of magnitude. The oxyanion doping strategy is a promising method for improving transport properties of proton-conducting layered perovskites.

Supplementary materials
No supplementary materials are available.

Funding
This research was performed according to the budgetary plan of the Institute of High Temperature Electrochemistry and funded by the Budget of Russian Federation.