Abstract

Neutron diffraction is one of the best methods for structural analysis of a complex, layered perovskite material with low symmetry by accurately detecting the oxygen positions through octahedral tilting. In this research, the crystal structure of NdSrMn2O5+δ was identified through X-ray diffraction (XRD) and neutron powder diffraction (NPD) at room temperature (RT), which indicated the formation of a layered structure in orthorhombic symmetry in the Pmmm (no. 47) space group. Rietveld refinement of the neutron diffraction data has confirmed the orthorhombic symmetry with unit cell parameters (a = 3.8367 (1) Å, b = 3.8643 (2) Å, and c = 7.7126 (1) Å), atomic positions, and oxygen occupancy. Thermogravimetric analysis revealed the total weight loss of about 0.10% for 20–950°C temperature, which occurred mainly to create oxygen vacancies at high temperatures. Rietveld analyses concurred with the XRD and neutron data allowing correlation of occupancy factors of the oxygen sites.

1. Introduction

The perovskite materials are used widely in solid oxide fuel cells (SOFCs) due to its diversity in chemical compositions. Ideal cubic symmetrical perovskite oxides have the general formula ABO3 [1], where A and B indicate A-site and B-site cations and O is the anion [2]. Perovskite oxides containing excess oxygen due to interstitial oxygen atoms are unstable thermodynamically [3, 4]. Since oxygen has a high electronegativity, it will always attract electrons from heated-site and B-site cations and make them mixed-valence state for stability. As a result, research is being concentrated on perovskite oxides which have oxygen deficiency, and this deficiency can be created by manipulating the cationic and anionic stoichiometry of ABO3 [5]. In recent times, the layered perovskites have attracted researchers because of their promising properties in energy sectors [68]. Rare-earth perovskites, such as PrBaMn2O5 and NdBaMn2O5, exhibited excellent redox stability (this implies that more easily reduced perovskite exhibits higher catalytic activity) and tolerance to coking and sulfur contamination from fuels [9]. Besides, some manganese-based layered perovskite can be used as oxygen storage materials in solid oxide fuel cells (SOFCs) [10] and solid oxide electrolyzer cells (SOECs) due to their electron-conductive nature [1113]. Some layered-type perovskites used as electrodes in SOFCs [1418] fueled through hydrogen or other syngas [1922] have also shown promising results.

The structural distortion is our core consideration as it affects the physical and electrochemical properties of the perovskite-type oxides [2325]. Neutron diffraction is a robust technique that can determine complex crystal structure, oxygen stoichiometry, and oxygen vacancy ordering. It is noteworthy that the neutron is scattered from the nuclei of the atoms allowing for the formation of different isotopes of the same atom and could detect light atoms masked by the heavy atoms [26]. The Bragg reflections of the powder pattern with a long Q-range can easily be detected. Many efforts have been dedicated to enriching the performance of layered perovskite by substituting various cations, especially Mn-doped rare-earth perovskite. As an anode for SOFCs, (PrBa)0.95(Fe0.9Mo0.1)2O5+δ (PBFM) demonstrated a high power density of 1.72 W·cm−2 at 800°C (as reported in [27]), whereas the composition SmBaCo0.5Mn1.5O5+δ demonstrated a power density of 377 mW/cm2 and SmBaMn2O5+δ exhibited high power density of 782 mW·cm−2 at 900°C as an electrode [16, 28].

The synthesis of a novel material was elaborately discussed, and the results of high-resolution neutron powder diffraction (NPD) studies were observed on the crystallized sample in this work. We report the complete structural data of these materials and describe the thermal properties from thermogravimetric analysis.

2. Materials and Methods

The solid-state synthesis technique [18, 2933] was applied to developing NdSrMn2O5+δ. Oxide powders of Nd2O3 (≥99.90%, Sigma-Aldrich), SrCO3 (≥99.90%, Aldrich), and MnO (≥99.50%, Aldrich) were weighed according to their stoichiometric ratios and ground with the aid of a mortar and pestle using ethanol as a reagent [34]. The powders were calcined at 1000°C for 10 hours after drying. The powders were pressed into pellets and sintered at 1200°C for 12 hours with 5°C·min−1 heating and cooling rate with intermediate grinding. Subsequently, the pellets were reground and resintered at 1400°C for another 12 hours. The whole synthesis process was operated under an Argon (Ar) atmosphere with a gas flow rate of 40 ml·min−1. X-ray powder diffraction (XRD) and neutron powder diffraction (NPD) were used to analyze the crystal structure of NdSrMn2O5+δ material.

The phase structure was first determined by XRD using a Bruker axs-D8 Advance diffractometer. Data were collected in the 2θ range from 10° to 79.995° with increments of 0.02° per second. The Rietveld refinement procedure was used to analyze the XRD data [35] using the Fullprof software [36]. A polynomial function (6-coefficient) was set for the background, and the pseudo-Voigt function was used to model the peak shapes.

Neutron powder diffraction data were collected at room temperature (RT) with the Polaris diffractometer (medium-resolution powder diffractometer at a high intensity) at the ISIS Neutron & Muon Source, UK [37, 38]. The time-of-flight (TOF) powder diffraction data were analyzed using GSAS-II [39] software. This material is debuted under the Pmmm space group through the Rietveld analysis of the high-resolution NPD data; a layered perovskite structure was formed. The Rietveld analysis used standard parameters for the refinement: a shifted Chebyshev series as background as instigated in GSAS software, zero shift, scale factor, profile parameters (type 3 in GSAS), cell parameters, atomic coordinates, site-occupancy factor (SOF), and atomic displacement factors (ADP).

To perform thermogravimetric analysis (TGA), a Netzsch-Gerätebau GmbH-STA 409 PC Luxx Simultaneous Thermal Analyzer was used to perceive the weight change with increasing temperature under flowing nitrogen. 99.51 mg of NdSrMn2O5+δ powder was placed in a ceramic crucible (Al2O3 DSC/TG pan) and heated from 20 to 950°C at a rate of 5°C·min−1 under 20 ml·min−1 of N2 flow. An isothermal hold for 1 hour removed absorbed species before cooling. The process was then repeated to ensure complete desorption of any contaminants. Upon complete desorption, N2 flow was substituted for airflow, and the mass change were recorded until equilibrium was reached.

3. Results and Discussion

Solid-state reaction methods have been used to prepare the layered perovskite NdSrMn2O5+δ. This layered perovskite is challenging to develop in a pure form, but the single-phase was obtained. Our synthesis method was also different from the method used to obtain NdBaMn2O5+δ [40] but similar to the synthesis process for PrSrMn2O5+δ [30]. Figure 1 shows the XRD pattern for NdSrMn2O5+δ sintered at 1400°C under Ar atmosphere. Some extra small peaks could not be indexed with the basic unit cell pattern. But, most of the peaks in Figure 1 can be indexed to an orthorhombic unit cell. The crystalline structure of this material was determined as ceramic through the XRD pattern. The XRD for the sample was measured at room temperature.

Fundamental understanding of the structure of the NdSrMn2O5+δ sample was investigated by neutron powder diffraction at room temperature. Oxygen vacancies are created in the material, which can balance the total charge. The single-phase orthorhombic structure was obtained from neutron diffraction with the space group, Pmmm. Rietveld refinement of room-temperature NPD data (Figure 2) revealed that NdSrMn2O5+δ achieved cell parameters a = 3.8367 (1) Å, b = 3.8643 (2) Å, and c = 7.7126 (1) Å with dimensions ap × ap × 2ap as observed in NdBaCo2−xMnxO5+δ [12]. Bank 2 (up to 7 Å) NPD data were analyzed via Rietveld refinement. The space groups, refinement factors (R-factors), and cell parameters are listed in Table 1, and atomic positions, Wyckoff positions, and isotropic temperature factors are listed Table 2, respectively. In Table 1, the other layered perovskite structures were compared with the present work.

The oxygen occupation was fixed at 1 at three oxygen sites for the space group Pmmm, O1, O2, and O3, repectively. These three sites remained locked in the Rietveld model refinement to detect the significant eccentricity from unity. No significant changes were found for three sites. Uiso as the thermal vibration parameters for Nd, Sr, and Mn were refined isotropically. These sites were set isotropically to get a standard deviation. Atomic displacement parameters (ADP) and the site-occupancy factors (SOF) correlated with each other. As a result, they were unable to be refined simultaneously. DIFA (a small correction in GSAS software to allow a reflection in the expected time-of-flight peak shifts due to sampling absorption), absorption, and scaling parameters were constrained in this case. The isotropic thermal vibration parameters (Uiso) also remained constrained in each phase. The average B-O bond lengths at room temperature can be compared to the calculated ionic radii by Shannon [43], where <Mn-O> is 1.9218 (6) Å (calc. 2.08 Å). Main bond distances and their average distances are tabulated in Table 3.

The surface morphology exhibits well-connected, large grains showing visible grain growth with an orthorhombic form. There were no secondary phases found at the grain boundary region in the NdSrMn2O5+δ sample. The grains were approximately 10 μm in size for the sample. The porous morphology (Figure 3) depicted that this material can be used as an electrode for fuel cells, the pores will assist the conduction of electrons, as well as allow fuel to pass easily through the structure [44].

An inert atmosphere is needed during the thermogravimetric experiment to prevent oxidation of the sample during thermal treatment. A vacuum environment was created inside the TGA-differential scanning calorimetry (DSC) chamber to ensure an utterly anoxic environment for the analysis. A small amount of the sample NdSrMn2O5+δ was taken for thermogravimetric analysis- (TGA-) differential scanning calorimetry (DSC) under a nitrogen environment. TGA showed that oxidation occurred at 200°C while heating in a single gradation (Figure 3) including weight loss with 1 oxygen atom in the formula unit. A small amount of weight loss was observed from 20°C to 950°C in the TGA-DSC curve in N2 atmosphere. But, there no phase transition occurred which is seen in the DSC curve as there is no exo- or endothermic peaks observed in Figure 3 [45]. The weight loss occurred due to evaporation of the moisture [46], formation of oxygen vacancy, and valence state of cations [3]. In the first step (200°C–500°C), the weight loss was high due to the moisture evaporation [47], and the decline in this region was about 0.084%. Above 500°C, the weight loss was less because all the organic compounds and all other elements end up in this step and the sample behaves as a thermally stable material [4853]; from 500°C to 950°C, the weight loss was approximately 0.016%. The total weight loss observed was about 0.10% for a temperature range between 20 and 950°C which is comparable with other perovskite materials; SmBaMn2O5+δ (0.036%) [54] and PrSrMn2O5+δ [30]. The oxygen content of the equilibrium stage decreases with temperature leading to oxygen vacancy formation during TGA-DSC. Table 4 shows that the calculated oxygen occupancy values from TGA are very close to the calculated values from Rietveld refinement. From Table 4, we can see that due to the fewer changes in oxygen occupancy in the NdSrMn2O5+δ crystal powder, a minimal weight loss has been observed which is close to or less than other layered structures from the literature that are comparable.

The crystal structure of NdSrMn2O5+δ is an example of layered orthorhombic symmetry, where both B-cations occupy the perovskite-like corner-shared octahedral (MnO1 and MnO2) sites. In this study, we evaluated the structural and thermal characteristics. According to these characterizations, we obtained a good result due to its high porosity, stable structure, and sufficient oxygen deficiency in comparison to similar types of layered perovskites. For NdSrMn2O5+δ, the Rietveld analysis indicates that all oxygen vacancies occur in the O1 and O2 sites. The volume of this NdSrMn2O5+ δ material is not so large without any long-range ordering in B-site which indicates an oxygen-deficient layered perovskite oxide. During TGA-DSC, when the heating starts, the weight loss was mainly observed from 200 to 950°C on account of increasing oxygen deficiency and thermally stable [55] due to its minimal amount of weight loss. Also, similar rare-earth layered perovskite has already given a promising result with the same space group for IT-SOFC [56]. These types of layered perovskites have recently gained a great deal of attention for SOFC anode materials because of their unusually high oxygen transport kinetics rate [57, 58]. It is established that the layered perovskite performs well for fuel cells reported by Abdalla et al. [40, 54].

4. Conclusions

Terminally, the solid-state synthesis method was used to prepare this single-phase novel, layered perovskite NdSrMn2O5+δ. XRD, NPD, and TGA-DSC analyses were used to determine structural and thermal properties. Both XRD and NPD data confirmed that the sample crystallizes in orthorhombic symmetry with the space group, Pmmm via Rietveld analysis. The structural features of this orthorhombic structure were measured by the action of flowing nitrogen (N2) with temperature and time, evidenced by a minimal weight loss (0.1%), which may be the weight loss attributed to the oxygen vacancy formation or a decrease in oxygen content. The minimal weight loss occurred in the TGA-DSC results, mainly for the fewer variations in oxygen occupancy in the NdSrMn2O5+δ crystal, whose value is closer to the results of oxygen occupancy in neutron refinement analysis. The development of layered perovskite remains an appealing research topic, and promising technology has emerged to improve SOFCs with further electrochemical experiments.

Nomenclature

δ:Oxygen nonstoichiometry
Uiso:Thermal vibrational parameters
Rp, Rwp, and Rf:Residual factors or R-factors
SOFC:Solid oxide fuel cell
SOEC:Solid oxide electrolyzer cell
XRD:X-ray powder diffraction
NPD:Neutron powder diffraction
RT:Room temperature
TOF:Time of flight
SOF:Site-occupancy factor
ADP:Atomic displacement factors
TGA:Thermogravimetric analysis
DSC:Differential scanning calorimetry.

Data Availability

The data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they do not have any conflicts of interest.

Acknowledgments

The author Shammya Afroze is grateful to Universiti Brunei Darussalam for awarding her by UBD Graduate Scholarship (UGS). The author is especially indebted to late Professor Sten G. Eriksson from Chalmers University of Technology, Sweden. The ISIS Neutron and Muon Facility, UK, is greatly acknowledged for its scheduled beam-time (RB1810638, DOI: https://doi.org/10.5286/ISIS.E.RB1810638).