Features of the Preparation of Ni-Doped Bismuth Tantalate Pyrochlore

Abstract

The possibility of high-temperature solid-phase synthesis of Ni-doped bismuth tantalate pyrochlore, not only from the oxide compounds BiTaO₄ and NiO but also using nickel chloride NiCl₂ as a precursor, was shown for the first time. The concentration range of the formation of nickel pyrochlore Bi₂Ni_xTa₂O_9−δ (0.85 ≤ x ≤ 1.0) at an equal molar ratio of bismuth(III)/tantalum(V) ions was determined. This indicates that the pyrochlore structure may be stable at 20–33% vacant bismuth sublattice. The influence of non-stoichiometric composition relative to bismuth and tantalum, by the example of Bi_2±xNiTa₂O_9−δ and Bi₂NiTa_2−xO_9−δ (x ≤ 0.5) compositions, on the phase composition of ceramics was studied. X-ray powder diffraction phase analysis of Bi₂Ni_1+yTa_1.75−yO_9−δ samples made it possible to determine the solubility limit of nickel ions in the tantalum sublattice. It was shown that the molar ratio of metal atoms (Ta, Ni) in the B₂O₆ (B-Ni(II), Ta(V)) octahedral sublattice of mixed pyrochlores is also limited.

1. Introduction

The crystal structure of pyrochlore is represented by the general formula A₂B₂O₇, in which two interpenetrating weakly interacting sublattices A₂O and B₂O₆ are distinguished [1]. The octahedra of the B₂O₆ basis structure, arranged around the diamond grid points, form a tetrahedral framework connected at the vertices. Two large A cations and one additional O atom per each formula unit of the (BO₃)₂ framework are placed in its cavities. Relatively small B cations (Ru(IV), Ta(V), Ti(IV), Zr(IV), Nb(V)) are arranged in octahedra, and large A cations (Bi(III), Pb(II), Sm(III)) are arranged in a distorted octahedron formed by oxygen atoms of A₂O and B₂O₆ sublattices. Due to the structural flexibility, pyrochlores correspond to a wide range of cation substitutions in the crystallographic positions A and B, which contributes to the formation of hundreds of compositions with a certain set of practically useful properties [2,3,4,5,6,7,8]. Materials based on pyrochlores are used in solid-state devices as thermistors, thick-film resistors, and communication elements, and also as components of ceramic forms for radioactive waste. Materials based on bismuth pyrochlores are promising for multilayer ceramic capacitors and tunable microwave dielectric components due to their excellent dielectric properties, possibility of dielectric properties regulation by electric field, low sintering temperature, and chemical compatibility with fusible Ag, Cu conductors [9,10,11,12].

Traditionally, the optimization of the physicochemical properties of ceramics is performed by modifying their chemical composition. Nickel doping of bismuth pyrochlores positively affects the dielectric tunability and thermal dielectric constant [13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30] (−57 ppm/°C for Bi_1.5NiNb_1.5O₇ [19]). Nickel pyrochlores based on bismuth niobate Bi_1.6Ni_2/3−yNb_4/3+yO_6.4+3y/2 exhibit high dielectric permittivity values of 127–168 while varying the y coefficient from −0.1 to 0.1 [14]. Bi₂Ni_2/3Nb_4/3O₇ ceramics show close values of dielectric permittivity 122 and dielectric loss tangent tanδ = 0.001 at 1 MHz [30]. Nickel-containing bismuth tantalate pyrochlores have lower dielectric permittivity compared to niobium ones, which is associated with a highly porous ceramic microstructure and decreased overall polarizability when replacing Nb(V) by Ta(V). For example, the dielectric permittivity value for Bi₃Ni_1.4Ta₃O_13.4 is 44.85 (RT, 10⁵ Hz) [18] and for Bi₂NiTa₂O₉ it is 32 for 30 °C and 1 MHz [16]. Meanwhile, nickel pyrochlores based on bismuth tantalate, which have a highly porous microstructure, are characterized by low values of the band gap width and are promising as photocatalysts [31]. Despite the prospects for use of nickel pyrochlores, the concentration stability region of bismuth tantalate-based nickel pyrochlores, in contrast to niobium pyrochlores [13], has not been studied in detail. In addition to the study of pyrochlore composition Bi₂NiTa₂O₉ [16], a number of compositions Bi₃Ni_2−xTa₃O_14−x (−0.2 ≤ x ≤ 1.0) were systematically investigated [18]. Only one composition at x = 0.6 was referred to phase-pure pyrochlores. In our study, we determined the concentration range of formation of the nickel pyrochlore Bi₂NixTa₂O_9−δ and the solubility limit of nickel ions in the tantalum sublattice. The samples with vacancies in the bismuth and tantalum sublattices were also studied. In addition, the possibility of solid-phase synthesis of Ni-doped bismuth tantalate pyrochlore, not only from the oxide compounds BiTaO₄ and NiO but also using nickel chloride NiCl₂ as a precursor, was demonstrated.

2. Experimental

Bi₂NiTa₂O₉ pyrochlore was synthesized by solid-state reaction from bismuth orthoniobate BiTaO₄ and nickel(II) oxide. The stoichiometric precursor mixture was homogenized in an agate mortar for one hour. Then, the obtained mixture was compacted in the form of disks for close contact of ceramic grains. The samples were heat-treated in two ways. In the first case, the sample was calcined once at 1050 °C for 15 h. In the second, calcination was carried out in two stages of 15 h at 950 and 1050 °C [32].

In another case, in addition to nickel(II) chloride, bismuth(III) and tantalum(V) oxides were used for the solid-state synthesis of Bi₂NiTa₂O₉. The stoichiometric mixture of precursors was thoroughly ground in an agate mortar for one hour and then pressed in the form of disks. The samples were calcined in four 15 h steps at 650, 850, 950, and 1050 °C [16,17]. After calcination, the samples were homogenized and pressed again. In the traditional version of the ceramic synthesis for samples including Bi₂Ni_xTa₂O_9−δ, the synthesis conditions remained unchanged, but nickel(II) oxide was used instead of nickel chloride.

X-ray powder diffraction data were collected using a Shimadzu 6000 X-ray diffractometer (CuKα radiation; 2θ = 10–80°; scanning speed 2.0 deg/min). A vertical type of goniometer was used. The scanning mode was ϴ/2ϴ; the scanning radius was 185 mm. A scintillation counter was used as a detector. The scintillator is NaI. The irradiation zone of the sample is 1 cm². Soller slots with parameters DS = 1, SS = 1, RS = 0.15 were used. The cell parameters were calculated using the CSD software package. Investigations of surface morphology and local quantitative elemental analysis were performed by scanning electron microscopy and energy dispersive X-ray spectroscopy (electron scanning microscope Tescan VEGA 3LMN, energy dispersion spectrometer INCA Energy 450). The optimum working distance was fixed within 15 mm with operating voltage at 10–15 kV. The spectral resolution was 0.01 keV. The content of chloride ions in the sample (Bi₂NiTa₂O₈Cl₂) was determined by atomic emission spectrometry using an optical emission ICP spectrometer SPECTRO CIROSCCD. The detection limit of chloride ions in the solution was 1.6 mg/L. For the chemical analysis, several sample weights (~7–8 mg) were taken. The volume of the solution prepared by dissolving the sample is 25 cm³.

3. Results and Discussion

3.1. Ceramic Synthesis

Previously, in a study of pyrochlore phase formation, it was found that bismuth orthotantalate is a precursor of pyrochlore phase. In this case, BiTaO₄ cannot transform into pyrochlore alone because it has a large ratio of ionic radii r(Bi³⁺)/r(Ta⁵⁺) ~1.83 at r(Bi³⁺) = 1.17 Å (c.n. = 8) and r(Ta⁵⁺) = 0.64 Å (c.n. = 6) [31]. According to [32], the pyrochlore phase is actively synthesized at temperatures above 850 °C during the interaction of orthorhombic bismuth orthotantalate (α-BiTaO₄) with nickel oxide. This gives a potential possibility for high-temperature synthesis of pyrochlore from these precursors. Initially, pyrochlore was synthesized by calcination at 1050 °C of the sample obtained from a homogeneous stoichiometric mixture of bismuth orthotantalate and nickel(II) oxide. According to X-ray powder diffraction phase analysis (Figure 1a), the sample contains the main pyrochlore phase and the impurity phase of nickel oxide. The pyrochlore unit cell parameter is 10.5547(4) Å. The presence of nickel oxide admixture indicates that the synthesis of nickel pyrochlore is a kinetically inhibited process and a single 15 h calcination was not enough. In the case of double 15 h calcination at 950 and 1050 °C (Figure 1b), the minimal amount of bismuth orthotantalate impurity was observed in X-ray powder diffraction patterns. The unit cell parameter of pyrochlore is 10.5224(8) Å. As can be seen, the unit cell parameter decreases during the formation of stoichiometric pyrochlore. It should be noted that the value of 10.5224(8) is comparable with the unit cell parameter of phase-pure Bi₂NiTa₂O₉. Meanwhile, this study shows that repeated calcination with intermediate homogenization of the sample is necessary to obtain phase-pure pyrochlore, even from BiTaO₄ and NiO.

Nickel(II) chloride was used as a precursor in order to clarify the possibility of synthesizing nickel pyrochlore with a mixed anionic sublattice consisting of chlorine and oxygen ions. Synthesis conditions were identical to those for pyrochlore synthesis from oxides. The sample was calcined four times at 650, 850, 950, and 1050 °C with intermediate homogenization. X-ray powder diffraction phase analysis showed the presence of bismuth orthotantalate triclinic impurity in the sample in the amount of 11 wt. % (Figure 1c). The unit cell parameter of pyrochlore is 10.4884(8) Å. The color of the sample is browner compared to the lilac color for nickel pyrochlore synthesized from oxides (Figure 2). Apparently, there are several reasons for the color change of the sample. Among these are the replacement of a part of oxygen ions in the octahedral environment of nickel(II) ions, change in the symmetry of the coordination environment or the oxidation degree of nickel ions, and influence of impurity substances. Since chloride ions are weak-field ligands, this substitution can lead to a decrease in the energy of splitting by the crystal field. A change in the splitting energy will affect the energy of the absorbed radiation and it will decrease. For this reason, the wavelength of the absorbed radiation will shift to the long wavelength region. As is known, the color of the compound is complementary to the color of the absorbed radiation. In this case, the color of the sample should be a shorter wavelength, which is not consistent with the experiment. Oxidation of nickel(II) ions to nickel(III) (Jahn–Teller ion) can lead to an increase in the energy of splitting by the crystal field and a decrease in the wavelength of absorbed radiation, which will lead to an increase in the wavelength of the complementary color. We observe a similar effect on our sample. The influence of the impurity on the color of nickel pyrochlore is unlikely because triclinic bismuth orthotantalate does not absorb visible radiation and its content in the sample is low. The effect of a change in the symmetry of the coordination of polyhedron of nickel ions on the color of the sample due to the imperfection of the anion sublattice caused by the partial replacement of oxygen ions by chloride ions or anion vacancies cannot be excluded.

Energy dispersion spectroscopy analysis did not reveal a reliable content of chloride ions in the sample, either due to trace amounts or due to overlapping of chloride absorption lines (L and K series in the region of 0.183 eV and 2.62, 2.82 eV) with bismuth M-series lines in the 2.42–2.74 eV region. Atomic emission spectroscopy analysis of the preparation showed that chloride ion content in the sample was less than the limit of device sensitivity (1.6 mg/L). If one conventionally assumes that the content of chloride ions in the prepared solution is equal to the detection limit, then the chemical composition of the preparation corresponds to the formula Bi₂NiTa₂O_8+δCl_≤0.15 instead of the expected Bi₂NiTa₂O₈Cl₂ composition. From this result, we can conclude that high-temperature calcination (1050 °C) of the chloride-containing sample in air leads to the chlorine loss and violation of the compound stoichiometry.

3.2. Study of Pyrochlore Stability Area

Previously, we reported [16,18] on the preparation of phase-pure nickel-containing pyrochlore of the composition Bi₂NiTa₂O₉ and Bi₃Ni_1.4Ta₃O_14−δ. Considering that nickel(II) ions occupy predominantly octahedral Ta(V) positions [16], it was assumed that the bismuth sublattice may be ~33% vacant compared to the tantalum one (Bi₂NiTa₂O₉), while the tantalum(V) sublattice is 1/3 filled with nickel(II) ions. Therefore, the aim was to determine experimentally the maximum and minimum vacancy levels of the bismuth(III) sublattice and the degree of substitution of tantalum(V) ions by nickel(II) ions at which the pyrochlore structure remains stable. In order to determine the concentration range of nickel pyrochlore formation in the Bi₂O₃-NiO-Ta₂O₅ system, several series of compositions were investigated.

First of all, the influence of bismuth and tantalum composition non-stoichiometry on the phase composition of ceramics was investigated on the examples of Bi_2−xNiTa₂O_9−δ, Bi_2+xNiTa₂O_9+δ, and Bi₂NiTa_2−xO_9−δ (x ≤ 0.5) compounds. X-ray powder diffraction patterns of the samples are shown in Figure 2. According to the X-ray powder diffraction phase analysis data, the Bi_2−xNiTa₂O_9−δ samples are non-phase pure even at x = 0.25. The X-ray powder diffraction patterns of the samples demonstrate reflexes of a single impurity NiTa₂O₆ (sp. gr. P-1). The formation of the nickel tantalate impurity can be explained by the fact that the bismuth sublattice is significantly unfilled, by 42% and 50%, respectively. These values, as will be shown later, are above the bismuth sublattice defectivity limit for pyrochlore stabilization. In the case of Bi_2+xNiTa₂O_9+δ compositions, the sample at x = 0.4 is single-phase. Based on this, one can conclude that the minimum vacancy limit of the bismuth sublattice is likely to be 20%. Thus, the allowable range of bismuth sublattice vacancy is about 20–33%.

X-ray powder diffraction phase analysis of Bi₂NiTa_2−xO_9−δ (x = 0.25; 0.5) samples showed that the Bi₂NiTa_2−xO_9−δ (x = 0.25) sample is single-phase. In the X-ray powder diffraction pattern of the composition at x = 0.5, the NiO impurity phase is determined in trace amounts. The degree of bismuth sublattice defectiveness for both compositions is acceptable and is equal to 27 and 20%, respectively. As will be shown below, the presence of a nickel oxide impurity is associated with an excess of the degree of substitution of Ta(V) ions. Apparently, for the stability of the octahedral tantalum sublattice, it is important to keep the strict molar ratio of n(Ni(II)/n(Ta(V)). In other words, there should be a limit for the substitution of sublattice B ions by heterovalent ions in mixed pyrochlores for the octahedral framework to remain stable. This limit should probably depend on the nature of the substituting atoms, as in the case of Ni(II) and Fe(III) (3d cations) and their ability to distribute over the bismuth(III) and tantalum(V) ion sublattices.

The degree of substitution of Ta(V) ions by Ni(II) ions for the considered compositions was calculated. For the Bi₂NiTa_2−xO_9−δ samples (x = 0.25; 0.5), it was 36.4% and 40%, respectively. Since the Bi₂NiTa_2−xO_9−δ sample (x = 0.5) is multiphase, the value of ~36% may be the limit in the case of nickel(II) ions. In order to estimate more precisely the limit of substitution of tantalum(V) ions by nickel(II) ions and to find out whether this number depends on the number of vacancies in the bismuth sublattice, two series of samples with different degrees of bismuth sublattice defects (indicated in percentages and given in parentheses) were considered. To solve this problem, Bi₂Ni_1+yTa_1.75−yO_9−δ (y = 0.1) (28%) and Bi₂Ni_0.85+yTa_1.75−yO_9−δ (y = 0.29) (30%) samples were prepared. The ratio of pyrochlore sublattice atoms A and B is variable (this ratio is in the range 27–33%), but the ratio of octahedral sublattice ions n(Ni(II)/n(Ta(V)) is fixed (degree of substitution 40%). The X-ray powder diffraction patterns of the studied samples show an impurity of NiO, regardless of the vacancy percentage of the bismuth sublattice. This result may indicate that the maximum degree of substitution of Ta(V) ions by nickel(II) ions is not affected by the degree of bismuth(III) sublattice defectiveness.

The X-ray powder diffraction phase analysis of the samples of Bi₂Ni_xTa₂O_9−δ series (0.4 ≤ x ≤ 1.2) with the equal molar ratio of bismuth(III)/tantalum(V) ions allowed us to establish the minimum and maximum degree of substitution of nickel(II) and Ta(V) ions to preserve the stability of the tantalum(V) octahedral sublattice. X-ray powder diffraction patterns of the synthesized samples are shown in Figure 3. The compositions at 0.85 ≤ x ≤ 1.0 are referred to as single-phase samples. The impurity composition of the multiphase samples is different. The samples at 0.4 ≤ x ≤ 0.8 exhibit β-BiTaO₄ as an impurity [33], and the nickel tantalate (NiTa₂O₆) and NiO as an impurity are found at x > 1.0. This is confirmed by the results of [18]. This work is devoted to the investigation of Bi₃Ni_2−tTa₃O_14−x samples (–0.2 ≤ t ≤ 1.0). Among these samples, the authors detected one phase-pure pyrochlore at t = 0.6, which fits into the specified interval 0.85 ≤ x ≤ 1.0. Apparently, the formation of the NiTa₂O₆, NiO impurity is caused by exceeding the substitution limit of the octahedral sublattice, and the formation of bismuth orthotantalate indicates an insufficient ratio of Ni(II)/Ta(V) ions. Assuming that nickel(II) ions predominantly occupy the octahedral positions [16], it may be concluded that the octahedral framework, including the pyrochlore structure, is potentially stable if the octahedral positions are replaced in the range of 30–36%. It should be noted that 33% corresponds to the substitution of 1/3 of all octahedral positions. If one assumes that 2/3 of the positions are occupied by Ta(V) ions and 1/3 of the positions are occupied by Ni(II) ions, then the average oxidation degree of the octahedral sublattice B ions is +4. This conditional charge of the B ions corresponds to the traditional stoichiometry of the oxide pyrochlore A⁺³₂B⁺⁴₂O₇, provided that the A ions are three-charged cations, such as Bi⁺³ [1,34].

Based on the results of the analysis, one can conclude that the molar ratio of metal atoms (Ta, Ni) in the octahedral sublattice of B₂O₆ (B-Ni(II),Ta(V)) mixed pyrochlores is also limited. To preserve the stability of the Ta₂O₆ octahedral sublattice, the degree of substitution of tantalum(V) ions by nickel(II) ions cannot be less than 30% or greater than 36% (33 ± 3%). According to X-ray powder diffraction phase analysis, these conditions are crucial for the formation of pyrochlore structure. As shown above, the ultimate degree of substitution of Ta(V) ions does not depend on the number of vacancies in the bismuth sublattice. It should be noted that the established stability criteria for mixed pyrochlores do not contradict the traditional pyrochlore stability coefficient [1] but are only an addition that is true for mixed pyrochlores containing divalent nickel cations. According to [1], to describe the stability of oxide pyrochlores, the notion of the “pyrochlore stability field” based on the ratio of cationic radii is used. The r(A)/r(B) = 1.46–1.80 and 1.40–2.20 values range limits the stability of pyrochlores having A³⁺₂B⁴⁺₂O₇ and A²⁺₂B⁵⁺₂O₇ composition, respectively. For instance, for nickel pyrochlore of Bi₂NiTa₂O₉ composition, it is equal to 1.78 and belongs to the specified intervals. Thus, this paper demonstrates that the degree of substitution of the octahedral coordination ions and the bismuth sublattice vacancy are important for the formation of phase-pure mixed bismuth pyrochlores.

4. Conclusions

The possibility of the ceramic high-temperature synthesis of the nickel pyrochlore based on the bismuth tantalate from BiTaO₄ and NiO precursors is shown. It has been established that to obtain phase-clean pyrochlore, repeated heat treatment with intermediate homogenization of the sample is required. Using ISP atomic emission spectroscopy and EDS analysis, it has been shown that in the course of high-temperature (1050 °C) solid-state synthesis of the nickel pyrochlore using the NiCl₂ precursor, the content of chloride ions in the synthesized pyrochlore is significantly lower than the target amount. In order to determine the concentration range of the stability of Ni-doped bismuth tantalate pyrochlores, the phase composition of several series of nickel-containing samples was studied. It was found that phase-pure nickel-doped bismuth tantalate pyrochlores are formed with a partially vacant bismuth sublattice and almost 1/3 of the Ta(V) positions are occupied by nickel(II) ions. It is shown that the most important condition for the formation of the pyrochlore structure is the requirement for an optimal degree of substitution of tantalum sites. The revealed conditions for the formation of nickel pyrochlores will make it possible to theoretically predict the stoichiometric composition of preparations for the synthesis of phase-pure pyrochlores.