Peroxide-based route for the synthesis of zinc titanate powder

2.1 Zinc titanate preparation

0.1 M solutions of equimolar mixtures of TiOCl₂, Zn(NO₃)₂ and H₂O₂ and 0.1 M KOH solution was used. The precipitation was carried out by gradual addition of small portion of potassium hydroxide and mixture of zinc and titanium salts with hydrogen peroxide into 500 ml of distilled water. The pH of reacting solution was constantly kept in 8.5 ± 0.25 range. The suspension was continuously stirred by the propeller mixer at 300 rpm. The precipitates were decanted and washed multiple times with 3 · 10⁻⁶ M KOH solution (pH ∼ 8,5) to remove chloride, nitrate and excess potassium ions, until washed off water didn’t react with silver nitrate, diphenylamine and sodium hexanitritocobaltate(III). Afterward, all precipitates were vacuum filtered using Büchner funnel, washed with isopropyl alcohol and vacuum dried at 50 °C. All used reagents are pure and analytical grade (“Reachim” Ltd., Sigma Aldrich).

2.2 Zinc ions coprecipitation

The experiments of zinc ions coprecipitation were carried out as follows: potassium hydroxide was gradually added to the mixture of 0.01 M solutions of hydrogen peroxide, zinc and titanium salts to form precipitate of titanium peroxohydroxide. The final pH of reacting solution was in range of 5–6.5. Afterward, the solution was centrifuged to remove the precipitate, and zinc content in solution was determined.

2.3 Zinc titanate characterization

Zn–Ti solid materials were characterized by X-ray diffraction (XRD), thermogravimetric analyses (TGA) and differential thermal analyses (DTA). The XRD was performed on a PANalytical X’Pert Pro diffractometer (monochromatic Cu K_α1 radiation with wavelength K_α1 = 1.54056 Å and linear correction with wavelength K_α2 = 1.54433 Å). The precipitates were studied also by thermal diffraction method with programmed cycle of heating and registering of diffraction patterns stages in temperature range of 40–700 °C with 20 °C step, heating rate 12 °C/min and isothermal aging during 55 min to record each diffraction pattern.

The thermal analysis was performed on a differential scanning calorimeter Thermal Analysis Instruments SDT 2960 in the temperature range 20–600 °C in nitrogen flow (50 ml/min) with a heating rate of 10 °C/min. After heating, the sample was cooled at the same rate of 10 °C/min.

3.1 Theoretical analysis of chemical precipitation of mixed Zn(II) and Ti(IV) hydroxides

The possible transformations in Zn(II)−Ti(IV)−H₂O₂ system were viewed as combination of precipitation, complexing¹ and neutralization reactions: $$\begin{array}{cc}\hfill & {\text{Zn}}^{2+}+2{\text{OH}}^{-}\to \text{Zn}{(\text{OH})}_2\downarrow \text{,}\hfill \\ \hfill & {\text{TiO}}^{2+}+2{\text{OH}}^{-}\to \text{TiO}{(\text{OH})}_2\downarrow \text{,}\hfill \\ \hfill & {\text{TiO}}^{2+}+{\text{H}}_2{\text{O}}_2+{\text{H}}_2\text{O}\to {[\text{TiO}({\text{O}}_2\text{H})]}^{+}+{\text{H}}_3{\text{O}}^{+}\text{,}\hfill \\ \hfill & {[\text{TiO}({\text{O}}_2\text{H})]}^{+}+{\text{OH}}^{-}\to \text{TiO}({\text{O}}_2\text{H})(\text{OH})\downarrow \text{,}\hfill \\ \hfill & {\text{H}}_3{\text{O}}^{+}+{\text{OH}}^{-}\to 2{\text{H}}_2\text{O.}\hfill \end{array}$$

It is known, that major precipitation mechanisms of salts with low solubility at constant temperature and pressure are defined by concentration of reagents and acidity of solutions. It is convenient to represent the diagrams of such systems using lgS(pH) coordinates, where S – precipitate solubility (concentration of saturated solution) (Nikolenko et al., 2012). Fig. 1 shows the solubility diagrams for zinc and titanium hydroxides that we have calculated using experimental and literature data. Since there was no data on stability constant of titanium complexes in literature it was not possible to calculate solubility curve for titanium oxyhydroxide in presence of hydrogen peroxide. Because of that, we have studied the dependency of titanium oxyhydroxide on acidity of solution, experimentally. The acquired data are shown in Fig. 1.

Close

Figure 1

Solubility diagrams of precipitates: titanium oxyhydroxide (1), titanium peroxohydroxide (2) and zinc hydroxide (2). Curve (4) shows zinc ion content in solution after precipitation of titanium peroxohydroxide from solution with molar ratio Ti:Zn:H₂O₂ = 1:1:1.

Export to PPT

The conducted calculations showed that observed shift of titanium peroxohydroxide solubility curve cannot be attributed only to the change of solubility equilibrium. The complexing processes are the main reason for such drastic changes in solubility. From this point of view, it is possible to explain the observed occurrence of titanium peroxohydroxide solubility in basic medium at pH > 8 (Fig. 1, curve 2). Evidently, such solubility is attributed to increased stability of titanium peroxocomplexes because of HO₂⁻ ligands ionization in Ti(IV) coordination sphere, with formation, for instance, of complexes with composition ${[\text{TiO}({\text{O}}_2){(\text{OH})}_2]}_2^{2-}$ or ${[\text{Ti}{({\text{O}}_2)}_2{(\text{OH})}_2]}_2^{2-}$ (Fox et al., 1990).

We have not established the influence of hydrogen peroxide on the solubility of zinc hydroxide. In addition, the experiments of zinc ions coprecipitation showed that concentration of zinc ions in solution after precipitation of titanium peroxohydroxide decreases significantly. It was determined that at chosen experimental conditions (0.01 M solutions of hydrogen peroxide, zinc and titanium salts) the concentration of zinc ions after precipitation of titanium peroxohydroxide decreases almost by a degree. Since at pH < 6.5 of 0.01 M solution the Zn(OH)₂ precipitate does not form, the acquired data should be explained as occurrence of coprecipitation process. Obviously, the reason of such occurrence is attributed to kinetic properties of the system, when the system, because of the diffusion resistance, the condition of thermodynamic equilibrium is virtually unachievable.

Thus, taking into account the occurrence of coprecipitation process of zinc and titanium ions mixture, the precipitation process should be viewed as a chain of three stages:

• starting at pH₁ a solid phase of titanium peroxohydroxide is form. With further increase of pH the equilibrium concentration of titanium ions changes according to the curve 2 in Fig. 1;

• zinc ions are simultaneously coprecipitated with titanium ions (curve 4);

• the solid phase of zinc hydroxide starts to form pH₂ (curve 3).

Thus, when pH_fin is reached the precipitate should include two separate phase: a mixture of titanium peroxohydroxide with coprecipitated zinc ions and Zn(OH)₂ phase. In order to determine the precipitation conditions of mixed hydroxides at which highest degree of their mixing is achieved, let’s review known methods of hydroxide precipitation:

3.2 The influence of hydrogen peroxide on synthesis of zinc titanate

To determine the influence of H₂O₂ on the synthesis process of zinc titanate the samples of mixed zinc and titanium hydroxides with and without addition of hydrogen peroxide were prepared. In both cases the precipitation was carried out at the same conditions (pH = 8.5 ± 0.25) by simultaneous addition of solutions of base and mixed zinc and titanium salts to reacting solution. The prepared precipitates after calcinations for 2 h at temperatures of 500, 700 and 900 °C and their X-ray diffraction patterns were studied. Fig. 2 shows the comparison of diffraction patterns taken from mixed zinc and titanium hydroxides samples with and without addition of H₂O₂, after calcination at same conditions for 2 h at 900 °C. For comparison, Fig. 2 also shows the diffraction pattern taken from mixed zinc and titanium oxides that was calcinated for 6 h, that matches with references data for cubic ZnTiO₃ (JCPDS 39-0190). The formation of crystal phases occurs almost similarly: at first the structure with interplanar spacing 2.5 Å is form (peak at 2θ = 35.3°). However, the intensity of this peak for studied samples differs significantly: the sample synthesized with addition of hydrogen peroxides has higher intensity of the peak.

Close

Figure 2

XRD patterns for mixed titanium and zinc oxides prepared without addition (1) and with addition of H₂O₂. (2 and 3), after calcination at 900 °C for 2 h (1 and 2) and for 6 h (3).

Export to PPT

For quantitative comparison of XRD analysis, the intensities of dominant peaks of studied samples are compared using ln I(1/T) coordinates and shown in Fig. 3. Because samples were calcinated at the same condition, and the peak intensities characterize the crystallinity degree of studied substance, it was assumed that in this case the intensity value can be directly proportional to constant of crystal growth rate. Because of that, we have used coordinates of Arrhenius equation, to determine the influence of hydrogen peroxide on formation rate of ZnTiO₃ structure. According to Fig. 3, linear dependencies can be observed for both samples support what is described above.

Close

Figure 3

Comparison of peak intensity logarithm at 2θ = 35.30° of mixed titanium and zinc oxides, prepared without (1) and with addition of H₂O₂ (2), vs. reciprocal calcination temperature.

Export to PPT

Thus, the obtained data show that samples prepared with addition of hydrogen peroxide has a higher rate of zinc titanate formation than the samples prepared at the same conditions but without addition of hydrogen peroxide. This pattern proves the hypothesis described earlier about necessity of depolymerization of titanium hydroxo complexes prior to coprecipitation with zinc hydroxide.

3.3 TGA and DTA study of Zn(II) and Ti(IV) hydroxides and activation energy of thermal treatment stage

The carried out experiments showed that when base solution is mixed with solution of TiOCl₂, Zn(NO₃)₂ and H₂O₂ the yellow X-ray amorphous precipitates are formed. During heating they gradually lose their yellow color and turn white, which can be explain by decomposition of colored peroxocompound of titanium.

To determine the kinetic parameters of thermal treatment process of coprecipitated zinc hydroxide and titanium peroxocomplexes TGA and DTA methods were used. An example of obtained data for the air-dry sample with equimolar ratio Zn/Ti is presented in Fig. 4.

Close

Figure 4

Analysis results of thermal decomposition of Zn(II) and Ti(IV) hydroxide mixture, coprecipitated from solution with molar ratio Ti:Zn:H₂O₂ = 1:1:1.

Export to PPT

On the DTA cooling curve (dashed line on Fig. 4) no exo- or endothermic effects were observed which allowed using this curve as a base line when interpreting the DTA curve for heating of hydroxide mixture. It was determined, that thermal transformations are accompanied by mass loss until 600 °C. The six segments are observed on TG curve, that are characterized by different rate of mass loss: first segment – 20–120 °C (mass loss 7.8%), second – 120–280 °C (6%), third – 280–350 °C (1.2%), fourth – 350–450 °C (0.8%), fifth – 450–500 °C (mass of the sample didn’t change) and sixth – 500–600 °C (0.9%).

The DTA curve shows, that sample mass change occurs with absorption of heat (first endothermic effect with minimum at 86 °C, second – in the 150–280 °C range), but also with its release (four exothermic effects with maximums at 312, 396, 595 °C and wide plateau in the 450–500 °C range).

It is known that heat absorption and mass loss at the initial segments of the DTA and TGA curves of the air-dry oxide and hydroxide sample are attributed to a dehydration process. It should be noted, that in order to prevent particle agglomeration in studied sample resulting from capillary compression force, an isopropyl alcohol was used during drying. Its boiling temperature under air conditions is 82.4 °C. Because of that, the first endothermic effect on the DTA curve should be explained as evaporation of leftover alcohol that was present in pores of sample after vacuum drying at 50 °C.

The change in mass loss rate at 120 °C and sharp turn of the DTA curve at 150 °C, obviously, should be explained by evaporation of adsorbed and hydration water. Considering that this segment of the TG curve stretches up to 280 °C, it should be concluded that in this temperature range, in addition to removal of water from crystals, decomposition of titanium peroxide and full dehydration of zinc and titanium hydroxides also occur. The thermal decomposition of peroxo-groups as part of solid oxide phase was studied by authors (Ivanova et al., 2002). According to their data the peroxo-groups as part of complexes are relatively stable and during heating of oxides are preserved up to 300 °C.

Upon further heating relatively small exothermal peaks at 312, 396 and 595 °C can be observed on the DTA curve. According to the TG curve mass loss at these stages of thermal treatment is 0.8–1.2%, which allowed viewing these effects as a result of decomposition of impurities. It is known, that when relatively concentrated solution is used for chemical precipitation, the formed precipitate usually contains ion traces of accompanying electrolytes. In our case titanium oxychloride, zinc nitrate and potassium hydroxide should be viewed as impurities. From analysis of thermal-chemical properties of these compounds follows that nitrate ions have the lowest decomposition temperature. The decomposition of nitrates occurs with consequential formation of nitrites, oxynitrates and nitrogen oxides. Oxygen and nitrogen are also found among decomposition products. Thus, exothermic effects at 312 and 396 °C can be attributed to decomposition of nitrates to nitrites ( ${\text{NO}}_3^{-}\to {\text{NO}}_2^{-}+0.5{\text{O}}_2\uparrow$ ) with consequential decomposition of nitrites with the release of gaseous products.

The high temperature of the exothermal effect at 595 °C and mass loss of 0.9% in temperature range 500–600 °C can be explained by decomposition of titanium oxychloride ( $2{\text{TiOCl}}_2\to {\text{TiCl}}_4\uparrow +{\text{TiO}}_2$ ) with release of gaseous TiCl₄.

The data of the TG curve allow determining kinetic parameter of processes that are accompanied by mass loss. For instance, for processes that are described by first-order kinetic equations, the activation energy can be calculated by using, Kissinger’s formula:

The order of thermal decomposition reactions can be determined by using data about peak symmetry of the DTA curve (Horowitz and Metzger, 1963). According to the obtained DTA data, the symmetry factors of observed endo- and exothermal effects are close to 1, which indicates that these processes can be describe by first-order equations.

The activation energies for stages of dehydration and decomposition of precursor impurities that are accompanied by mass loss were calculated by using formula (1) and results are presented in Table 1. According to the obtained data, the activation energies decomposition of the zinc and titanium precursors impurities are approximately 5–6 times higher than that of titanium peroxide decomposition. Such pattern looks logical considering low stability of hydrogen peroxide upon heating.

3.4 XRD study and activation energy of grain growth of ZnTiO₃ powders

The precipitates were studied by thermal diffraction method with programmed cycle of heating. An example of data acquired for temperature range 100–700 °C with step 100 °C is presented in Fig. 5. It was determined, the formation of highly crystalline cubic zinc titanate (JCPDS No. 39-0190) only occurs at temperature higher than 500 °C.

Close

Figure 5

Analysis result for precipitate with equimolar Zn/Ti ratio using thermal diffraction method. The X-ray pattern for temperature range of 100–700 °C with step 100 °C.

Export to PPT

A more detailed investigation with calculation of interplanar distances showed that transformation of amorphous structure of mixed oxide into crystalline occurs at 580 °C. Fig. 6 shows results of calculating interplanar distance alongside a axis of cubic crystalline structure of zinc titanate during sample calcination. According to this data, the formation of crystalline zinc titanate starts only at 580 °C.

Close

Figure 6

Evolution of interplanar distance alongside a axis of crystalline ZnTiO₃ structure during calcination.

Export to PPT

To determine activation energy of ZnTiO₃ grain growth stage a dependency of grain size on temperature was used (Chang et al., 2002):

Grain size can be determined from XRD pattern. It’s known that (Płóciennik et al., 2016) the width of the diffraction peak can be influenced by physical factors characterizing materials such as the crystallite sizes and lattice strains. The source of lattice strains can derive from displacements of the unit cells around their normal positions as well as dislocations, domain boundaries, surfaces, etc. To describe the experimental intensity distribution of the diffraction line derived from these physical factors, different theoretical functions (Gauss- or Cauchy-like function) can be used for the calculations. It is assumed that the diffraction line broadening due to the small size of the crystallites β can be expressed by the relation known as the Scherrer’s equation:

Distortion resulting from the lattice strains can be expressed by a dependence known as the Taylor equation:

This above equation is known as Williamson−Hall equations. Plotting of the dependence βcos θ as a function of the sin θ for diffraction peaks allows determining the average value of crystallite size (Płóciennik et al., 2016). The calculation using (2) and (5) for 600–700 °C range, showed that the activation energy of grain growth stage is equal to 23.5 kJ/mol. The activation energy for samples that were prepared without addition of hydrogen peroxide is 46.7 kJ/mol.

It should be noted, that according to the data obtained by Chang et al. (2004, 2002), when zinc titanate is prepared by a modified solgel route including the Pechini process, the activation energy for ZnTiO₃ grain growth is equal to 20.8 kJ/mol, while during the calcination of zinc and titanium oxides grinded in the ball mill, the activation energy is 48.8 kJ/mol.

In our opinion the decrease of activation energy for ZnTiO₃ cubic structure formation stage, prepared by calcination of highly homogeneous mixtures of zinc and titanium oxides, can be explained as follows. It is known, that the condition for thermodynamic equilibrium of solid-state transformation, as any other chemical process, is an equality of chemical potentials of components in the initial and resulting medium. During interaction of two solid phases the mention equality of chemical potential can be realized through the rearrangement of component in the initial phases with formation of solid solutions, and as a result of formation of new phases with crystalline structure. The component transfer in reacting medium is required in both cases. In thermodynamics it is not required to take the mechanism of such transfer into account, but for chemical kinetics the way of its realization defines the process effectiveness. It is known, that the crystallization process must be viewed as a sequence of at least two stages – formation of crystal nucleuses of new phase and its growing. Obviously, the lower the “local” stoichiometric ratio of zinc and titanium oxide mixture (i.e. stoichiometric ratio in the formation region of nucleus of new crystalline phase), the higher time required for mixing reacting ions, and as a result the lower probability of formation of such nucleuses per time unit.

The connection between mixing degree of components and activation energy can also be explained from point of view of theory of absolute rates of reactions. In the approximate imagination of active complex the reactant particles must not only get close but also to take the configuration of active complex, upon decomposition of which, the final product of reaction is formed. In addition, the free activation energy is defined as a change of enthalpy of activation (defined by chemical nature of the compound), and as a change of entropy of activation. The value of latter component in free activation energy depends not only on the nature of compound (for instance, composition and configuration of intermediate state), but also on entropy value of initial reactants. In our opinion this value is influenced by mixing method of reactants, which defines the “local” stoichiometric ratio of reactant mixture.