Chemical Pressure Effect on the Stabilization of Rock-Salt ZnO—Lin−2MeOn−1 Solid Solutions Synthesized at High Pressure

Metastable ZnO—Lin−2MeOn−1 (Me = Sc3+, Ti4+, Ta5+) solid solutions with a rock-salt structure were synthesized through the solid-state reaction of ZnO with Lin−2Men+On−1 (n = 3, 4, 5) complex oxides at 7.7 GPa and 1300–1500 K. In all investigated systems, single-phase rock-salt solid solutions can be quenched down to ambient conditions in a wide (up to 80 mol% ZnO) concentration range. The phase composition, thermal stability, and thermal expansion of the recovered rock-salt solid solutions were studied by synchrotron powder X-ray diffraction. At ambient pressure, these solid solutions exhibit high thermal stability (up to 1000 K), with the decomposition temperature and decomposition products depending on the nature of the multiple charge cations.


Introduction
There are a number of crystal structures that have the potential to combine a wide variety of cations in a single compound over a wide range of concentrations. This is necessary in the design of catalytic materials or to control the properties of semiconductors and other functional materials. Examples of such universal structures are perovskites, garnets, spinels, etc. [1,2]. Also, as a universal platform for creating inorganic materials, we can consider the rock-salt structure, which is typical for alkali metal halides or binary ionic compounds such as Me 2+ O oxides. Rock-salt modification is also known for a semiconducting material such as zinc oxide. In this regard, zinc oxide and ZnObased solid solutions are very promising objects [3,4]. Zinc oxide belongs to the family of wide band gap semiconductors with strong ionic character of chemical bonds [3,4]. Under ambient conditions, ZnO has a hexagonal wurtzite structure (P6 3 mc, S.G. 186, w-ZnO), which changes to a cubic rock-salt structure (Fm-3m, S.G. 225, rs-ZnO) at pressures above 5 GPa [5][6][7][8][9].
The rock-salt structure, in contrast to the wurtzite one, allows the synthesis of ZnO solid solutions with a wide variety of cations. Previously, for Zn 2+ ← Me 2+ substitution, isovalent cations (Me 2+ = Mg 2+ , Ni 2+ , Fe 2+ , Co 2+ , and Mn 2+ ) with a tendency to octahedral environments were selected, which significantly affected the thermal stability of the formed solid solutions [8][9][10][11][12][13][14]. The most stable solid solutions with high ZnO content (up to 80 mol%) were those with magnesium and nickel oxides. For example, the cubic Ni 0.2 Zn 0.8 O is stable up to 970 K, whereas the cubic Co 0.3 Zn 0.7 O is stable only up to 670 K. The decomposition products are usually mixtures of two solid solutions, one with a cubic structure and another with a wurtzite structure [8][9][10].
In addition to binary solid solutions, ternary ZnO solid solutions were also synthesized by replacing two zinc cations with a combination of a single-charged lithium cation and a triple-charged metal cation, i.e., 2Zn 2+ ← Li + + Me 3+ . Extensive solid solutions have been reported for LiFeO 2 [9,15], LiCrO 2 [16] and LiTiO 2 [9,15,17]. Limited cubic solid solutions in the ZnO-Li 2 TiO 3 system have been described at ambient pressure [18]. As in the case of isovalent substitution, the concentration range and thermal stability of the synthesized solid solutions with the combination of Li + + Me 3+ essentially depend on the nature of the trivalent cation. Rock-salt (LiTiO 2 ) 0.2 (ZnO) 0.8 is stable up to 800 K, while (LiFeO 2 ) 0.2 (ZnO) 0.8 is stable only up to 670 K [15]. In fact, other examples of ternary cubic ZnO-based solid solutions are not found in the literature.
In the present work, in addition to the system with the trivalent cation scandium, a similar principle has been extended to the four-(3 Zn 2+ ← 2 Li + + Me 4+ ) and five-valent (4 Zn 2+ ← 3 Li + + Me 5+ ) cations, using titanium and tantalum as examples. All selected multivalent cations in the maximum oxidation state (Me n+ ) form individual phases with lithium (Li n−2 Me n+ O n−1 ) although not in a rock-salt crystal lattice.

Experiment
The precursors listed in Table 1 were prepared by solid-state reactions of lithium carbonate (Li 2 CO 3 , Merck, min. 99%) with the corresponding metal oxide (reactive grade; a full list of the starting reagents is given in the Supplementary Materials). Due to lithium evaporation at high temperatures, Li 2 CO 3 was taken in 5% excess. The fine powder mixtures were pressed into pellets and annealed in an alumina crucible in air at 970 K for 2 h. The product was reground and annealed again at 1070 K for 2 h. According to X-ray diffraction measurements (MZIII Seifert CuKα radiation), all of the as-synthesized precursors were single phase (see Figures S1 and S2 in the Supplementary Materials).  [25], which is close to the lattice parameter of rs-ZnO (4.28 Å [5,8,9]). § At ambient pressure, the phase transition between the two monoclinic modifications of Li 3 TaO 4 is observed at 1173 K [28)]. A metastable cubic (Fm-3m) Li 3 TaO 4 has been reported to be kinetically stable below 1093 K [29,31]; its lattice parameter at room temperature is 4.2207(4) Å [31] or 4.214(5) Å [32].
Mixtures of complex oxide and w-ZnO (Alfa Aesar, 99.99%) with a ZnO molar fraction (x) from 0.5 to 0.8 (with a 0.1 step) were thoroughly ground in an agate mortar with acetone, pressed into pellets and placed in gold capsules. High-pressure synthesis was performed in a toroid-type apparatus at the LSPM-CNRS. The experimental details and pressuretemperature calibration have been described elsewhere [8]. In general, the samples were gradually compressed up to 7.7 GPa, heated for 5 min at the desired temperature (1300-1500 K), then quenched by turning off the power supply and slowly decompressed. After disassembling the high-pressure cell, samples with high ZnO content (x > 0.7) usually crumbled into a white microcrystalline powder, while the recovered samples with lower ZnO content retained the tablet shape and looked like dense and well-sintered ceramics. The morphology of the recovered samples was studied by scanning electron microscopy Leo Supra 50VP (Carl Zeiss, Germany).
The high-pressure (up to 5 GPa)-high-temperature (up to 1400 K) behavior of the initial LiScO 2 complex oxide was studied in situ by synchrotron energy-dispersive X-ray diffraction using the MAX80 multi-anvil system at F2.1 beamline of the DORIS III storage ring (HASYLAB-DESY); the details have been described elsewhere [13,14]. Thermal stability in air at temperatures up to 1200 K was studied by simultaneous thermogravimetric/differential thermal analysis using a Perkin Elmer Diamond Thermal Analyzer (TG/DTA) at a heating rate of 10 K/min.
In situ high-temperature (up to 1100 K) powder X-ray diffraction measurements at ambient pressure were performed at the B2 beamline of the DORIS III storage ring (HASYLAB-DESY) [33]. Debye-Scherrer geometry with a rotating fused silica capillary was used. Diffraction patterns were collected in the 2-70 • 2θ-range for 2 min in real time using an OBI image plate detector [34]. A NIST powder LaB 6 (Pm-3m, a = 4.15695 Å) was used as the standard sample for detector calibration. The furnace temperature was kept constant within 1 K using a Eurotherm temperature controller and Pt10%Rh-Pt thermocouple. The temperature step was 50 or 100 K. Before each data acquisition step, the sample temperature was stabilized for 2-3 min; the acquisition time was 2 min. The collected powder diffraction patterns were analyzed by the Le Bail method [35] using Powder Cell 2.4 [36] and DatLab 1.34 [37] software.

ZnO-LiScO 2 System
The as-synthesized tetragonal (I4 1 /amd) LiScO 2 contained no impurities and had lattice parameters of a = 4.187(5) Å and c = 9.315(3) Å, which were in good agreement with the literature data [19,20]. It should be noted that the I4 1 /amd structure is the only crystal form of LiScO 2 that has been reported. This is a cation-ordered NaCl-like structure. The partial cation-ordering results in the primitive tetragonal body-centered unit cell of LiScO 2 , which can be described as a NaCl superstructure doubled along the c-axis. The coordination polyhedron for both lithium and scandium cations is a distorted octahedron, slightly elongated in the c-direction, making this structure similar to anatase [19,20].
In a special series of experiments at 5 GPa, the crystal structure of LiScO 2 was studied in situ up to 1400 K using a MAX80 multi-anvil press and energy-dispersive X-ray diffraction. No melting or phase transition(s) of the pristine I4 1 /amd tetragonal structure was observed in the investigated temperature range. The LiScO 2 quenched from 7.7 GPa after a 5-min hold at 1370 K had the initial I4 1 /amd structure with a small amount of cubic (Ia-3) Sc 2 O 3 impurity; no other cubic phase(s) was observed (see Figure S1a).
Four compositions with x varying from 0.8 to 0.5 (in steps of 0.1) were synthesized in the ZnO-LiScO 2 system. The results are summarized in Figures 1-3 and Table 2, where the observed ranges of solid solutions are given in terms of x, the fraction of Zn 2+ ions substituted by other cations. In all cases, the recovered samples were white insulating bulks. According to the SEM data, the synthesized solid solutions were dense, practically poreless ceramics with a grain size of about 10 microns.  Figure S3a-c shows the representative X-ray diffraction patterns of the synthesized rock-salt solid solution containing a small amount of Sc 2 O 3 as an impurity. The sharp symmetrical narrow lines indicate that the solid solution is well crystallized. The dependence of the lattice parameters on the composition of the ZnO-LiScO 2 system is shown in Figure 1. This dependence is nearly linear and follows Vegard's law, as in the case of the related ZnO-LiTiO 2 and ZnO-LiFeO 2 systems [15].      Table 2 Figure S3a-c shows the representative X-ray diffraction patterns of the synthesized rock-salt solid solution containing a small amount of Sc2O3 as an impurity. The sharp symmetrical narrow lines indicate that the solid solution is well crystallized. The dependence of the lattice parameters on the composition of the ZnO-LiScO2 system is shown in Figure 1. This dependence is nearly linear and follows Vegard's law, as in the case of the related ZnO-LiTiO2 and ZnO-LiFeO2 systems [15].
A sequence of powder X-ray diffraction patterns for the rs-(LiScO2)1−x(ZnO)x solid solution composition with high ZnO content (x = 0.7) collected during stepwise heating up to 1100 K is shown in Figure 2. The first signs of change become visible at 873 K, when weak and broad reflections of the wurtzite phase appear, indicating the beginning of the decomposition of the initial single-phase cubic solid solution into a mixture of wurtzite solid solutions and Sc2O3 according to the following scheme: where w-ZnO:Li is ZnO-Li2O solid solutions of different compositions. The intensities of the lines corresponding to the cubic phase gradually decrease, and these lines disappear completely only at the highest temperatures reached in the experiment (~1100 K). At the same time, according to the simultaneous TG/DTA analysis, the weight loss and any thermal effects are not observed over the whole temperature range studied.  Table 2 A sequence of powder X-ray diffraction patterns for the rs-(LiScO 2 ) 1−x (ZnO) x solid solution composition with high ZnO content (x = 0.7) collected during stepwise heating up to 1100 K is shown in Figure 2. The first signs of change become visible at 873 K, when weak and broad reflections of the wurtzite phase appear, indicating the beginning of the decomposition of the initial single-phase cubic solid solution into a mixture of wurtzite solid solutions and Sc 2 O 3 according to the following scheme: where w-ZnO:Li is ZnO-Li 2 O solid solutions of different compositions. The intensities of the lines corresponding to the cubic phase gradually decrease, and these lines disappear completely only at the highest temperatures reached in the experiment (~1100 K). At the same time, according to the simultaneous TG/DTA analysis, the weight loss and any thermal effects are not observed over the whole temperature range studied.
The temperature dependencies of the unit cell volumes of rs-(LiScO 2 ) 1−x (ZnO) x solid solutions are shown in Figure 3a. The experimental thermal expansion data were fitted to the following equation: the coefficients of which are given in Table 2. The unit for V(T) and V 0 is Å 3 ; the units for α 1 and α 2 are K −1 and K −2 , respectively. The quadratic dependence of V(T) has been previously observed for ZnO-rich cubic solid solutions in other Li-containing systems [9,10,15], and α 1 and α 2 are close to the previously reported values [15]. The temperature dependencies of full width at half maximum (FWHM) for the 200 line (100% intensity) of rs-(LiScO 2 ) 1−x (ZnO) x solid solutions are shown in Figure 3b. In comparison, the most intense line 110 of the NIST standard LaB 6 also has a similar (~0.05 • ) FWHM, indicating a high degree of crystallinity of the synthesized solid solutions. As the temperature increases, the FWHM of the 200 line of the cubic phase remains almost constant up to 700 K (Figure 3b). Then a sharp (2 to 3 times) increase in FWHM is observed, which begins 100-200 K below the temperature when the decomposition of the cubic phase becomes visible (i.e., the appearance of w-ZnO lines). Such FWHM behavior is associated with the segregation of microcrystalline nonequilibrium cubic solid solutions at the nanoscale with the formation of ZnO-enriched and -depleted solid solutions (pre-phase solid solution separation). Then the wurtzite zinc oxide and cubic scandium oxide phases crystallize from these intermediate solid solutions.
Thus, (LiScO 2 ) 1−x (ZnO) x cubic solid solutions of the composition 0.5 ≤ x ≤ 0.8 with high thermal stability at ambient pressure were synthesized for the first time at high pressures and high temperatures.

ZnO-Li 2 TiO 3 System
The as-synthesized monoclinic (C2/c) Li 2 TiO 3 contained no impurities ( Figure S2) and had lattice parameters of a = 5.049(2) Å; b = 8.807(1) Å; and c = 9.715(6) Å, which were in good agreement with the literature data [22][23][24][25]. This compound is the only precursor known to have a high-temperature phase with a rock-salt structure (Fm-3m), in which Li + and Ti 4+ cations are completely disordered in their sublattice. At ambient pressure, the cubic phase has its stability field in the phase diagram at temperatures of 1430-1820 K, i.e., up to the melting point [21][22][23], and can be quenched under normal conditions. Another feature of Li 2 TiO 3 is that the Ti 4+ ion has the minimum size (0.60 Å) for the whole studied series of Me n+ cations [8,9,12,15].
Eight single-phase cubic solid solutions with x varying from 0.8 to 0.1 (in steps of 0.1) were synthesized in the ZnO-Li 2 TiO 3 system. The powder X-ray diffraction pattern of the solid solution with the maximum ZnO content, (Li 2 TiO 3 ) 0.2 (ZnO) 0.8 , is shown in Figure S3d  At room temperature, the dependence of the cell parameter of cubic solid solutions on the composition is nonlinear (Figure 5), with a negative deviation from Vegard's law. This is indicative of the manifestation of the chemical pressure effect, when the multi-charged cation with a smaller radius produces an electrostatic lattice compression  At room temperature, the dependence of the cell parameter of cubic solid solutions on the composition is nonlinear (Figure 5), with a negative deviation from Vegard's law. This is indicative of the manifestation of the chemical pressure effect, when the multi-charged cation with a smaller radius produces an electrostatic lattice compression effect [38]. Since in this case, we know the endpoints of the concentration dependence of the a parameter, it is possible to quantify this effect from the difference in the lattice parameters and compressibility of rs-ZnO [39]. The chemical pressure calculated in this way was 3-5 GPa, depending on the composition of the solid solution. It should be noted that for the ZnO-LiScO 2 system, such an effect is not observed due to the larger radius of the Sc 3+ cation compared to Zn 2+ , and a smaller charge compared to Ti 4+ . At room temperature, the dependence of the cell parameter of cubic solid soluti on the composition is nonlinear (Figure 5), with a negative deviation from Vegard's l This is indicative of the manifestation of the chemical pressure effect, when the m ti-charged cation with a smaller radius produces an electrostatic lattice compress effect [38]. Since in this case, we know the endpoints of the concentration dependenc the a parameter, it is possible to quantify this effect from the difference in the lattice rameters and compressibility of rs-ZnO [39]. The chemical pressure calculated in this w was 3-5 GPa, depending on the composition of the solid solution. It should be noted for the ZnO-LiScO2 system, such an effect is not observed due to the larger radius of Sc 3+ cation compared to Zn 2+ , and a smaller charge compared to Ti 4+ .  [7][8][9] and solid sq [26,27] are the literature data. The dashed line represents Vegard's law.  [7][8][9] and solid square [26,27] are the literature data. The dashed line represents Vegard's law.
It was previously reported that Li 2 TiO 3 and LiFeO 2 form continuous NaCl-type solid solutions with MgO at ambient pressure above 1300 K, and a deviation from Vegard's law was observed [18,[40][41][42]. Due to the similarity of the ionic radii and chemical nature of the Zn 2+ and Mg 2+ cations, the related effects of incipient immiscibility and/or short-range ordering of cations [41] can occur in ZnO-based cubic solid solutions.
The mechanism of decomposition can be described by the following scheme: As the ZnO fraction in the cubic solid solution decreases, the onset temperature of decomposition (T d ) increases from 870 K for x = 0.8 ( Figure 6) to 1040 K for x = 0.5. Compositions with lower zinc oxide content are kinetically stable up to 1100 K, the maximum temperature available in our experiments. In some cases, small amounts of the C2/c monoclinic phase, characteristic of Li 2 TiO 3 in this temperature range, have been observed in the X-ray diffraction patterns of decomposition products [25]. The FWHM values of the diffraction lines of all cubic solid solutions studied remain almost constant with increasing temperature.
Compositions with lower zinc oxide content are kinetically stable up to 1100 K, the maximum temperature available in our experiments. In some cases, small amounts of the C2/c monoclinic phase, characteristic of Li2TiO3 in this temperature range, have been observed in the X-ray diffraction patterns of decomposition products [25]. The FWHM values of the diffraction lines of all cubic solid solutions studied remain almost constant with increasing temperature.    Table 3).
The experimental thermal expansion data were fitted to the following equati where the coefficients of which are given in Table 3. The unit of V(T) and V0 is Å 3 , unit of α is K −1 . For these solid solutions, V(T) is linear or quasi-linear. For x < dependence of the thermal expansion coefficient on the composition is practically  Table 3).
The experimental thermal expansion data were fitted to the following equation where the coefficients of which are given in Table 3. The unit of V(T) and V 0 is Å 3 , and the unit of α is K −1 . For these solid solutions, V(T) is linear or quasi-linear. For x < 0.6, the dependence of the thermal expansion coefficient on the composition is practically absent ( Table 3, Figure S4), whereas for x ≥ 0.6, a tendency for its decrease with increasing zinc oxide content is observed. Note that extrapolation to 100% ZnO content (rs-ZnO) gives a value close to that previously reported by us [10].  (1) 0.0 [25] 71.20(2) 6.52 (4) Thus, for the first time, we have synthesized cubic ZnO-Li 2 TiO 3 solid solutions in a wide concentration range (0.1 ≤ x ≤ 0.8), which are characterized by high thermal stability.

ZnO-Li 3 TaO 4 System
The initial Li 3 TaO 4 was metastable and had a cubic structure (Fm-3m) with lattice parameter a = 4.215(1) Å, which is in agreement with the literature data (a = 4.2207(4) Å [31] and a = 4.214(5) Å [32]). Quenched from 7.7 GPa and 1370 K, Li 3 TaO 4 lost its cubic structure; the X-ray diffraction patterns before and after the high-pressure-high-temperature treatment are shown in Figure S1b.
Seven compositions with x varying from 0.8 to 0.3 (in steps of 0.1) were synthesized in the ZnO-Li 3 TaO 4 system. The results are summarized in Figures 8-10 and Figure S5 and Table 4, where the observed ranges of solid solutions are given in terms of x, the fraction of Zn 2+ ions substituted by other cations. In all cases, the recovered samples were white insulating bulks. The powder X-ray diffraction pattern of the (Li 3 TaO 4 ) 0.5 (ZnO) 0.5 solid solution is shown in Figure S3f, which indicates that it is cubic and single phase. The profiles of the diffraction lines are perfectly symmetric, and the lines themselves are of high intensity. Note that in the related system, MgO-Li 3 TaO 4 solid solutions with a cubic structure do not form [43].
Seven compositions with x varying from 0.8 to 0.3 (in steps of 0.1) were sy in the ZnO-Li3TaO4 system. The results are summarized in Figures 8-10 and ble 4, where the observed ranges of solid solutions are given in terms of x, the Zn 2+ ions substituted by other cations. In all cases, the recovered samples wer sulating bulks.  [7][8][9], solid and diamond [32] are the literature data. The dashed line represents Vegard's law. Figure 9. Sequence of synchrotron powder X-ray diffraction patterns of (Li3TaO4)0.5(Z solution taken during stepwise heating to 1100 K (wavelength λ = 0.68805 Å). The init phase is stable up to 1000 K; the decomposition products are mainly a mixture of es-wurtzite and rock-salt.  [7][8][9], solid square [31], and diamond [32] are the literature data. The dashed line represents Vegard's law.
Seven compositions with x varying from 0.8 to 0.3 (in steps of 0.1) were syn in the ZnO-Li3TaO4 system. The results are summarized in Figures 8-10 and S5 ble 4, where the observed ranges of solid solutions are given in terms of x, the fra Zn 2+ ions substituted by other cations. In all cases, the recovered samples were w sulating bulks.  [7][8][9], solid sq and diamond [32] are the literature data. The dashed line represents Vegard's law. Figure 9. Sequence of synchrotron powder X-ray diffraction patterns of (Li3TaO4)0.5(ZnO solution taken during stepwise heating to 1100 K (wavelength λ = 0.68805 Å). The initial phase is stable up to 1000 K; the decomposition products are mainly a mixture of tw es-wurtzite and rock-salt.  Table 4).  (1) The powder X-ray diffraction pattern of the (Li3TaO4)0.5(ZnO)0.5 solid solution is shown in Figure S3f, which indicates that it is cubic and single phase. The profiles of the diffraction lines are perfectly symmetric, and the lines themselves are of high intensity. Note that in the related system, MgO-Li3TaO4 solid solutions with a cubic structure do not form [43].
The diffraction pattern of the solid solution with nominal composition (Li3TaO4)0.2(ZnO)0.8 shows a more complex picture ( Figure S3e): two groups of narrow and intense lines are seen, both of which can be described as structures belonging to the Fm-3m space group. The lattice parameter of the first (18 vol% content) is 4.278(2) Å, and the lattice parameter of the second is 4.2404(1) Å (82 vol% content) while other lines are absent. The presence of two cubic NaCl-like structures with slightly different lattice parameters may indicate the presence of micron-scale regions with different compositions. With the increasing temperature, the first group of lines gradually decreases in intensity and disappears completely at 620 K. The intensity of the second group of lines also decreases smoothly up to 1070 K. Previously, only rs-Ni1−xZnxO cubic solid solutions [10] have shown such high thermal stability.
For the initial composition (Li3TaO4)0.7(ZnO)0.3, the formation of a mixture of phases is observed, among which we can distinguish rock-salt and wurtzite phases. Also observed is a group of lines that cannot be described by any known structure but that are similar to the lines of the rs-Li3TaO4 decomposition products.
A negative deviation from Vegard's law is observed for the dependence of the cell parameter of cubic solid solution on the composition (Figure 8). The calculated values of chemical pressure vary in the range of 3-5 GPa, depending on the composition of the solid solution.  Table 4).
The diffraction pattern of the solid solution with nominal composition (Li 3 TaO  For the initial composition (Li 3 TaO 4 ) 0.7 (ZnO) 0.3 , the formation of a mixture of phases is observed, among which we can distinguish rock-salt and wurtzite phases. Also observed is a group of lines that cannot be described by any known structure but that are similar to the lines of the rs-Li 3 TaO 4 decomposition products.
A negative deviation from Vegard's law is observed for the dependence of the cell parameter of cubic solid solution on the composition (Figure 8). The calculated values of chemical pressure vary in the range of 3-5 GPa, depending on the composition of the solid solution.
The mechanism of decomposition of cubic solid solutions in this case is as follows: rs-(Li 3 TaO 4 ) 1−x (ZnO) x → w-ZnO + rs-(Li 3 TaO 4 ) 1−y (ZnO) y + phase(s) X The temperature dependencies of the unit cell volumes of rs-(Li 3 TaO 4 ) 1−x (ZnO) x solid solutions are shown in Figure 10. The experimental thermal expansion data were fitted to Equation (4), the coefficients of which are given in Table 4.
Thus, we have demonstrated for the first time that synthesis at high pressure and high temperature allows for single-phase cubic solid solutions in the ZnO-Li 3 TaO 4 system to be obtained in a relatively narrow concentration range (0.5 ≤ x ≤ 0.8). The synthesized solid solutions exhibit high thermal and phase stability.

General Remarks
The concentration dependence of the decomposition temperature of metastable ZnO-Li n−2 MeO n−1 solid solutions (Me n+ = Sc 3+ , Ti 4+ , and Ta 5+ ) at ambient pressure for all investigated systems is shown in Figure 11. It is easy to see that the thermal stability increases with the cation charge (n). Such behavior is apparently related to the difficulty of diffusion of multivalent ions in the cubic lattice.
The temperature dependencies of the unit cell volumes of rs-(Li3TaO4)1−x(ZnO)x solid solutions are shown in Figure 10. The experimental thermal expansion data were fitted to Equation (4), the coefficients of which are given in Table 4.
Thus, we have demonstrated for the first time that synthesis at high pressure and high temperature allows for single-phase cubic solid solutions in the ZnO-Li3TaO4 system to be obtained in a relatively narrow concentration range (0.5 ≤ x ≤ 0.8). The synthesized solid solutions exhibit high thermal and phase stability.

General Remarks
The concentration dependence of the decomposition temperature of metastable ZnO-Lin−2MeOn−1 solid solutions (Me n+ = Sc 3+ , Ti 4+ , and Ta 5+ ) at ambient pressure for all investigated systems is shown in Figure 11. It is easy to see that the thermal stability increases with the cation charge (n). Such behavior is apparently related to the difficulty of diffusion of multivalent ions in the cubic lattice. Figure 11. Comparison of the phase stability of rock-salt ZnO-Lin−1MeOn−2 (Me n+ = Sc 3+ , Ti 4+ , Ta 5+ ) solid solutions at ambient pressure. Below the dotted lines are the kinetic stability fields of the corresponding cubic phases; above, the decomposition with formation of different phases is observed, n = 3, 4, 5 is an oxidation state of metal cations.
It is known that there are several factors that determine the kinetic stability of metastable phases quenched from high pressure and/or high temperature. In the case of the pure cubic ZnO phase, it is the kinetic stabilization of the small grains of the high-pressure phase that prevents the appearance of wurtzite phase nuclei in a small volume [7]. High-entropy multicomponent solid solutions whose stabilization is determined by high values of configurational entropy have been described in the literature [44,45]. For example, an equimolar mixture of MgO, CoO, NiO, CuO, and ZnO oxides can be stabilized in the rock-salt structure, although two of these oxides (CuO and ZnO) do not form cubic polymorphs as individual oxides. The necessary condition for this type of Figure 11. Comparison of the phase stability of rock-salt ZnO-Li n−1 MeO n−2 (Me n+ = Sc 3+ , Ti 4+ , Ta 5+ ) solid solutions at ambient pressure. Below the dotted lines are the kinetic stability fields of the corresponding cubic phases; above, the decomposition with formation of different phases is observed, n = 3, 4, 5 is an oxidation state of metal cations.
It is known that there are several factors that determine the kinetic stability of metastable phases quenched from high pressure and/or high temperature. In the case of the pure cubic ZnO phase, it is the kinetic stabilization of the small grains of the high-pressure phase that prevents the appearance of wurtzite phase nuclei in a small volume [7]. Highentropy multicomponent solid solutions whose stabilization is determined by high values of configurational entropy have been described in the literature [44,45]. For example, an equimolar mixture of MgO, CoO, NiO, CuO, and ZnO oxides can be stabilized in the rocksalt structure, although two of these oxides (CuO and ZnO) do not form cubic polymorphs as individual oxides. The necessary condition for this type of stabilization is the same cation charge. In the case of the cubic solid solutions considered in the present work, these factors either do not work or their contribution can be estimated to be negligible [46].
In multiphase systems, one of the components can play the role of a "shell", preventing the reverse phase transition of the metastable phase when the pressure is released. Such examples are described for the MgO-ZnO [47] and NaCl-ZnO [48] systems. Stability in such systems is provided either by the similar structure (MgO) or mechanical properties (NaCl) of the second component.
The octahedral environment preference factor [49] previously reported in [9,10,15] cannot be a determining factor in the case of ZnO solid solutions with Li 2 TiO 3 and Li 3 TaO 4 . Although lithium tends to an octahedral environment, its radius (0.76 Å) is slightly larger than that of Zn 2+ (0.74 Å) and the charge is half as low, so the influence of Li + on the stability of the cubic solid solution cannot be significant. However, it should be noted that the strong ionicity of the Li-O bond (lithium has the lowest electronegativity value in the series of elements considered) is a favorable factor.
In the case of ternary systems, the stabilization of the metastable cubic ZnO could be explained by the presence of a multi-charged cation in the solid solution lattice, i.e., by the fact that a cation with a smaller radius but a higher charge (see Table 1) "compresses" the crystal lattice (the so-called "chemical pressure" effect), which is accompanied by a kinetic stabilization of the cubic structure in the form of solid solution. According to the deviation of the lattice parameters of solid solutions from Vegard's law, the chemical pressure for titanium Ti 4+ and tantalum Ta 5+ in the ZnO-Li 2 TiO 3 and ZnO-Li 3 TaO 4 systems was estimated to be 3-5 GPa.
Thus, the metastable ZnO-Li n−1 MeO n−2 solid solutions in Figure 11 are completely consistent with the charge of the Me n+ cation. Solid solutions with Sc 3+ , where the effect of chemical compression on the concentration dependence of the lattice parameter is not observed, exhibit the minimum thermal stability, and solid solutions with Ta 5+ show the maximum.
We believe that some solid solutions with a rock-salt crystal structure, synthesized for the first time in the present work, may be of interest as promising Li-containing materials [50,51] or as materials with high dielectric characteristics [52,53].
The described approach of stabilizing metastable high-pressure phases under normal conditions by forming extended substitution solid solutions may also be promising for a number of other oxide and non-oxide (ZnS-LiGaS 2 , ZnS-Li 2 TiS 3 and ZnS-Li 3 NbS 4 [54][55][56]) systems whose high-pressure phases may have interesting functional properties. However, this hypothesis certainly requires experimental verification.

Conclusions
For the synthesized rock-salt ZnO-Li n−2 MeO n−1 (Me = Sc 3+ , Ti 4+ , Ta 5+ ) solid solutions, the limits of concentration (x up to 0.8) and thermal (up to 1000 K) stabilities were established. Coefficients of thermal expansion were determined for all the systems studied, showing either linear or quadratic dependence on temperature. The decomposition temperature of metastable solid solutions depends on the composition and nature of the multivalent cation, which allows us to define it as a chemical pressure effect.

Data Availability Statement:
The data presented in this study are available upon request.