Introduction

Pure zirconia follows a monoclinic (P21/c) → tetragonal (P42/nmc) → cubic (Fm3m) phase transition sequence upon heating [1]. The monoclinic phase transformation is avoided by stabilizing the tetragonal or cubic phase by doping with yttria (Y2O3) depending on its molar fraction. Zirconia-based materials are used widely in solid oxide fuel cells, medicine, catalysts, etc. [2,3,4]. From a mechanical point of view, the tetragonal phase has more practical applications than the cubic phase [5]. Pure galia (Ga2O3) has five polymorphs: α, β, γ, δ and ε, of which the β polymorph β-Ga2O3 (monoclinic galia) is stable in its entire temperature range up to the melting point [6]. While galia is a poor thermal and electrical conductor, in reducing conditions, it becomes an n-type semiconductor due to the formation of oxygen vacancies [6, 7].

Recent literature has repeatedly mentioned that investigations into the quasi-binary zirconia–galia system are gravely lacking and the works regarding this system are insufficient [8, 9]. The studies that do exist are not updated [10, 11]. The current research is inspired by landmark studies on the zirconia–alumina system where it has been reported that alumina has a tendency to suppress crystal growth in zirconia–alumina powder, allowing a preservation of crystal size in the lower nano-range as compared to non-doped zirconia, which exhibits more extensive crystal growth [12, 13]. The slow grain growth in doped zirconia is attributed to solute ion segregation into the grain boundaries inhibiting the grain growth. For instance, it was shown that faster powder sintering and reduced sintering temperature was required for densification of zirconia-doped alumina solutions [14]. Similar trends have been reported [15, 16] upon the addition of small amounts of galia to samaria-(Sm2O3) doped ceria (CeO2) and to yttria-doped ceria. The preparation of ultra-fine powders with a smaller particle size is significant, among others, for decreasing sintering temperature and improving densification and mechanical properties. Gallium has similarities to aluminum and yet remains on the outskirts of the mixed oxides field, whereas zirconia–alumina system has been researched in depth [17,18,19,20,21]. In this study, the influence of galia addition to zirconia-rich powders was investigated regarding the crystallographic phase transformations and the connection between the concentration of galia and calcination temperature on the crystal size and surface area.

Materials and methods

Sample preparation

Powder batches were synthesized via the sol–gel method based on the one reported in Kuo et al. [22] using different precursor salts to form zirconia–galia powders: zirconium oxynitrate hydrate (ZrO(NO3)2·2H2O) (purity ~ 99%) and gallium nitrate octahydrate (Ga(NO3)3·8H2O). The sol–gel method is a robust and simple method allowing the preparation of zirconia powders with versatile contents of galia. The solution of the precursors was made by dissolving stoichiometric amounts of precursors in deionized water (18.3 MΩ cm resistance) at room temperature. Ammonium hydroxide solution (1.5 M) was dripped slowly into the mixture under continuous stirring. The dripping continued until the pH value was 9. The gel’s precipitates were rinsed with deionized water. Eventually, the wet precipitates were dried using freeze-drying technique [23]. The gel was indirectly frozen by placing it in a closed beaker, which was submerged into liquid nitrogen and then freeze-dried at 223 K in a vacuum chamber. Each zirconia–galia powder was then divided into different samples which were calcined in air using alumina crucibles for 2 h at different temperatures. The different compositions obtained were ZrO2 with: 0, 0.7, 2, 3, 5, 7, 14 and 100 mol% of Ga2O3.

Phase composition analysis was performed using X-ray diffraction (XRD). The crystal size of zirconia–galia powders was approximated using three different methods: Brunauer–Emmett–Teller (BET) analysis, for the lower nano-range, XRD analysis, using the Williamson–Hall approach, for the intermediate nano-range (below 60 nm) and high-resolution scanning electron microscopy (HR-SEM) using ImageJ (image processing software) in the higher nano-range (above 60 nm).

Sample characterization

The composition analysis of the sol–gel powder was investigated by inductively coupled plasma optic emission spectroscopy (ICP-OES; Spectro Arcos, Kleve, Germany). The crystalline phase composition was determined by X-ray diffraction analysis (XRD; DMAX 2100 powder diffractometer, Rigaku, Japan) upon applying the Rietveld refinement. The surface area of powder was measured by BET (Brunauer–Emmett–Teller) method (NOVAtouch™, surface area and pore size analyzer, TouchWin™ software version 1.0, Quantachrome Instruments, USA). Sample preparation for BET analysis included outgassing of samples in vacuum under heating to a maximum temperature of 573 K and using nitrogen at 77 K for obtaining adsorption isotherms. The morphology of the powder after calcination was examined using a high-resolution scanning electron microscope (HR-SEM; JSM-7400, JEOL, Tokyo, Japan). Crystal size in the higher nano-range (above 60 nm) was approximated based on the measurement of more than 150 well-defined crystals apparent in the HR-SEM images using ImageJ (image processing software, National Institutes of Health, Bethesda, MD, USA).

Results and discussion

ICP-OES results confirmed that following the sol–gel synthesis, the powder composition well agreed with the theoretical design. The results are shown in Table 1.

Table 1 ICP-OES results
Full size table

XRD analysis was the key tool for mapping the phase compositions as a function of the galia concentration and calcination temperature. The XRD patterns of zirconia with different molar concentrations of galia are shown in Fig. 1, and the representative Rietveld refinements are shown in Fig. 2.

Figure 1
figure 1

XRD patterns of: a pure ZrO2, b ZrO2 + 0.7 mol% Ga2O3, c ZrO2 + 2 mol% Ga2O3, d ZrO2 + 3 mol% Ga2O3, e ZrO2 + 5 mol% Ga2O3, f ZrO2 + 7 mol% Ga2O3, g ZrO2 + 14 mol% Ga2O3 and h pure Ga2O3 at different calcination temperatures calcined for 2 h

Full size image
Figure 2
figure 2

Rietveld refinements of a ZrO2 + 0.7 mol% Ga2O3 calcined at 873 K and b ZrO2 + 5 mol% Ga2O3 calcined at 1273 K for 2 h

Full size image

Figure 3 shows a mapping of the different phase compositions obtained from the Rietveld analysis on a diagram of the calcination temperature as a function of the molar percentage of galia in the range of 0–14 mol%, where “m” refers to monoclinic phase and “t” to tetragonal phase. The mass fraction of each phase for multi-phases is shown in parentheses.

Figure 3
figure 3

Rietveld analysis results of the crystallographic phases obtained after calcination at different temperatures for 2 h

Full size image

Investigation of the XRD patterns in Fig. 1 reveals that the addition of galia to zirconia resulted in a considerable peak broadening and had a major influence on the crystallization temperature. Such a broadening can be easily seen upon calcination at 673 K of 0.7 mol% galia (Fig. 1b) compared to pure zirconia (Fig. 1a) or upon calcination at 873 K of 5 mol% galia (Fig. 1e) compared to 3 mol% galia (Fig. 1d). It can be also seen that a quasi-amorphous structure is obtained after calcination at 673 K for 2 h when adding 0.7 mol% galia to zirconia. When adding 5 mol% galia to zirconia, the same structure is maintained after calcination at a higher temperature of 873 K.

For each given concentration of galia in zirconia, it can be observed that the calcination temperature influences the crystalline domain size as the diffraction peaks become sharper as temperature rises and larger crystals were grown. Figure 3 shows that the addition of galia to zirconia stabilizes completely or partially the tetragonal phase depending on its concentration and calcination temperature. This finding is in agreement with the inability of galia to stabilize the cubic polymorph in the zirconia–galia system as was also previously reported [9]. From Fig. 3, for 14 mol% galia calcined at temperatures of 1273 and 1473 K, a monoclinic gallium oxide phase (C2/m) was segregated. From analysis of Fig. 1h, the monoclinic crystallization of pure galia starts in the range of 873–923 K in accordance with previously reported data [24]. For calcination temperatures of 673 and 873 K, the cubic spinel-type γ-Ga2O3 was obtained, and for higher temperatures of 923, 1073, 1473 K, the stable monoclinic β-Ga2O3 polymorph was formed. The monoclinic structure was detected in the 0.7 and 2 mol% galia compositions calcined at 1073–1473 K and in the > 2 mol% galia compositions calcined at 1273–1473 K for 2 h. Therefore, as the concentration of galia increases, the appearance of the monocline phase is delayed and diminished. For lower calcination temperatures, the monoclinic phase appearance was avoided and a single tetragonal zirconia phase was detected (as shown in Fig. 3). Table 2 shows the tetragonal unit cell parameters and volume based on the Rietveld results.

Table 2 Refined unit cell parameters values of the t-ZrO2 solid solutions
Full size table

Table 2 shows that for a given calcination temperature, the unit cell’s volume decreased as the concentration of galia increased.

Regarding the powder morphology, when tetragonal crystals transformed into monoclinic, twins were started to form as demonstrated in Fig. 4. A closer investigation of this figure reveals that although some of the large crystalline domains were twinned, twinning was much more emphasized in the smaller crystalline domains. The transformation from tetragonal phase into monoclinic phase is followed by formation of hydrostatic pressure and strains since monoclinic zirconia has lower density and thus requires more volume [1]. Large crystals apply pressure on smaller crystals which transform into the monoclinic phase due to mechanical deformation, while the large crystals transformed into the monoclinic phase in thermal transformation.

Figure 4
figure 4

HR-SEM of powders calcined at 1473 K for 2 h: a ZrO2 + 5 mol% and b ZrO2 + 7 mol% Ga2O3

Full size image

The crystallite size was calculated via the Williamson–Hall plot:

$$ \beta \cdot \cos \left( \theta \right) = k \cdot \lambda /L + 4 \cdot \varepsilon \cdot \sin \left( \theta \right), $$
(1)

where β is the broadening of the diffraction line measured at half of the maximum intensity, λ is the wavelength (Cu-κα), θ is the Bragg angle for a given diffraction and k is a constant, which is in general equal to 0.94 for powders. The instrumental broadening effect was estimated by subtracting the full width at half maximum of a standard sample (LaB6) from β of the respective Bragg peaks. L represents the crystallite size, and ε represents the lattice strain. The characteristic Williamson–Hall plot corresponds to the graph of β·cos(θ) versus sin(θ). After data collection, a linear regression should give a linear fit. The crystallite size was extracted from the y-intercept of the (\( k \cdot \lambda /L \)) fit. The Williamson–Hall approach becomes less accurate when crystals are bigger than approximately 60 nm and therefore, HR-SEM images were analyzed based on measuring the crystal size distribution. The average crystal size obtained for zirconia with 14 mol% galia calcined at 1073 K for 2 h (as a representative case) was estimated as 20 nm and was also cross-checked by HR-SEM which provided a similar average value.

Following the same calcination temperature of 1073 K, the 0.7% mol galia composition exhibited larger crystals in the size range of 40–55 nm (as derived from XRD crystal size analysis and HR-SEM image processing), highlighting the effect of galia on inhibiting crystalline domains growth. Investigating the average crystal size at the highest calcination temperature tested of 1473 K as was measured from the HR-SEM images using ImageJ image software clearly indicated a reduction trend from approximately 400 nm for 0.7 mol% galia, to 300 and 200 nm for powders containing 7 and 14 mol% galia, respectively, highlighting again the domains growth effect of galia (Fig. 5).

Figure 5
figure 5

HR-SEM of zirconia with 14 mol% galia calcined at 1073 K for 2 h

Full size image

BET (Brunauer–Emmett–Teller) measurements were carried out to pre-calcined galia-doped powders in order to confirm the effects found by XRD. Powders showed surface area in the range of 110–150 m2/g. Figure 6 shows that surface area increased with increase of the concentration of galia. The increase in surface area is an indication to the decrease in powder particle size.

Figure 6
figure 6

BET analysis for different contents of galia in zirconia

Full size image

Following 1473 K calcination, nano-agglomerated particles with irregular shapes were observed by HR-SEM, as can be seen in Fig. 7a for the secondary electron (SE) and Fig. 7b the backscattered electrons (BSE) investigation modes of the 14 mol% galia composition. Energy-dispersive X-ray spectroscopy (EDS) of these micrographs indicated the presence of monoclinic zirconia (brighter phase in Fig. 7b) and monoclinic galia (darker phase in Fig. 7b), in agreement with the XRD results in Fig. 3, indicating some precipitation of monoclinic galia.

Figure 7
figure 7

HR-SEM images of ZrO2 + 14 mol% Ga2O3 calcined at 1473 K for 2 h: a SE image and b BSE image

Full size image

Conclusions

In the current research, zirconia–galia nano-powders were synthesized via the sol–gel method. Zirconia–galia-based compositions with galia content in range of 0.7–7 mol% were found to form tetragonal (P42/nmc) or mixed tetragonal (P42/nmc)–monoclinic (P21/c) solid solution phases depending on the composition and calcination temperature. The addition of galia suppressed the crystallization of zirconia resulting in an extended amorphous range. Galia also decreased the crystalline domains size and correspondingly the powder particle size as the content of galia increased up to 14 mol%, resulting in finer nano-powders with higher surface area, compared to those previously reported for non-doped zirconia.