|FR | ENGL|
|Analysis of hydrothermal precipitation. Comparison between PZT, Al2O3, ZrO2|
Laboratoire des matériaux Avancés Céramiques
Université de Valenciennes et du Hainaut-Cambrésis
ZI. Champ de l’abbesse
59600 Maubeuge – France
Hydrothermal synthesis at low temperature (T<300°C) is an alternative method by comparing with calcination route for powder synthesized through chemical process like sol-gel, co-precipitation and so on. Due to the low temperature used, it is of economical interest. Moreover, diffusion mechanism at solid state do not occur and prevent any formation of hard agglomerates, as the reactive medium is maintained under vigorous agitation during the hydrothermal treatment, the obtained powder is fine and well dispersed. This method leads to the synthesis of anhydrous, fine, crystalline and non agglomerated powders. A thermodynamically stable phase is usually obtained in oxide form(PZT, PT, ZT, PZ) or some semi-hydrate states (AlOOH is obtained previously to Al2O3 only stable upper 600°C). Generally, two hydrothermal precipitation mechanism are pointed out : 1) an in-situ mechanism for which the gel is simply dehydrated, resulting in a collapse of the gel structure; the final composition is only linked to the composition of the gel. 2) a dissolution-precipitation mechanism for which a species re-distribution can occur.
The presence of one of these limiting step greatly modifies the manner in which the precipitation is conducted. In this paper, we present the results we have obtained concerning the hydrothermal precipitation of various systems (PZT, Al2O3, ZrO2).
II Hydrothermal precipitation of lead titanium zirconate
On a thermodynamic point of view, the PZT solid solution is stable in hydrothermal environment for temperature as low as 160°C and for slight basic pH value. Lencka and Riman [LENC1995] have focused their work on the links between basicity, temperature of the thermal treatment and reactive concentrations. For example, the theoretical diagram established for T = 160°C is presented on figure 1. The abscise axis corresponds to the pH value of the reactive media determined at 160°C. Y axis corresponds to the logarithm of lead concentration expressed in molality. As the precursors are introduced in stoichiometric amount, this concentration refers the theoretical amount of PZT production per liter of mixture.
These theoretical results are compared to experimental ones. The reactive mixture is prepared in two steps. In a first step, an homogeneous precipitate consisting of titanium and zirconium is obtained by hydrolysis of a closely mixture of titanium tetraisopropoxide Ti(iPrO)4, and zirconium tetrapropoxide, Zr(nPrO)4. The amorphous powder is dispersed in water and mixed with an aqueous solution of lead nitrate or lead acetate. The basicity is adjusted by addition of KOH. The final mixture contains as more as 0,1 mole per liter of equivalent PZT. The hydrothermal synthesis is conducted in small autoclave (110 ml) and in a temperature range 100 to 220°C.
The evolution the product composition versus the pH and the temperature values is presented on figure 2. On this figure, the y axis correspond to the pH values calculated for each temperature plot in x axis. The area pointed out corresponds to the theoretical conditions leading on a PZT precipitation yield over than 99 %.
By increasing the reactive concentration, the hydrothermal precipitation occurs in a larger area of basicity. Notably, it is shown that the addition of KOH allows the decrease of temperature as far as 150°C, that represent a huge gain by comparing the works of Kutty & al. [KUTT1984] and Beal [BEAL1987]. For T≥150°C, the optimum basicity condition for PZT corresponds to 7≤pH(T)≤11,5. As pH(T) decreases with temperature, it is necessary to fix a higher value of pH (at room temperature) in the starting feedstock. In addition, these authors have shown that it is helpful to start from a closely mixture of titanium and zirconium precursors to obtain the ternary PZT solid solution. This assumption is the starting point of numerous developments in the synthesis of more or less complex structures: it is whispered that the quality of the cationic dispersion in the clear solution is maintained up to the final oxide throughout all the synthesis steps. If not, for example in a case of an heterogeneous precipitation of zirconium and titanium hydrates, then it is not possible to obtain PZT; PT and PZ precipitate preferentially.
The differential precipitation of these hydroxides is easily made as it can be seen in Figure 3 [CHOY1995]. The pH precipitation areas of the three hydrates are for the most cases separated, they are overlaid in a sharp pH basic area.
It was shown that, as lead hydroxide is not a network former, the hydroxide mixture consist of Zr(OH)4 and Ti(OH)4 on which lead hydroxide is adsorbed. If differential precipitation is made, the subsequent hydrothermal treatment at temperatures below 300°C promotes the synthesis of lead titanate, lead zirconium even lead oxide. If this mixture is simply calcined, [CHOY1995], the PZT structure is formed for temperature upper 500°C. Such a high value is necessary to improve diffusion mechanism at long distance. The involved atomic motions show clearly that an important re-arrangement has occurred between the amorphous mixture and the final oxide PZT powder.
In our classical hydrothermal conditions, the involved temperatures are not high enough to promote such a mechanism. Lead zirconate and lead titanate precipitate preferentially if no cationic re-organization mechanism (by dissolution/precipitation for example) interfere to smooth zirconium and titanium heterogeneity before. It is then necessary to start from a very homogenous mixture of precursors. Most of the hydrothermal processes developed fort PZT are based on this assumption. The goal consists in freezing the cationic distribution during all the hydrothermal treatment up to the final oxide. This can be done by complexation, resin making, micro emulsion…
We have also developped an original two-step synthesis route starting from non-homogeneous precursors mixture. These precursors are TiO2 and freshly precipitated Zr(OH)2. A first hydrothermal treatment is realized without any addition of lead but with [KOH]o = 5 mol/l. The obtained product is washed several times until neutrality. Lead is added as nitrate or oxide form and KOH is fixed at [KOH]o = 1 mol/l. The schedule is presented in figure 4.
By using this protocol, we have obtained crystalline PZT powders which exhibit the same characteristics compared to powders synthesis using an alkoxide based homogeneous mixture of precursors and synthesis following the step 2 of the previous protocol ([KOH]=1 mol.l-1, T=265°C, t=2 hours). X-ray diffraction diagram of the powders are presented in figure 5.
The two powders are both composed of tetragonal and trigonal PZT forms. The chemical dispersion is estimated as a contribution to the broadening of X-ray diffraction peaks. [COUR2000, TRAI2000]. For the two processed powders, the peak broadening is in the same order of magnitude. The chemical fluctuation do not seems to differ from one processed powder to the other. These diagrams present only different relative intensity between tetragonal and trigonal peaks. The powder morphology obtained following the second protocol is presented on figure 6.
Cubic shape particles are observed, they are submicronical sized and slightly agglomerated. This corresponds also to the one step processed powder (i.e. starting from homogeneous precursors).
III Hydrothermal precipitation of alumina/zirconia mixture and of alumina on SiC
For these materials the problem consists in obtaining after synthesis an homogeneous and fine mixture of zirconia and alumina or alumina and SiC. As the synthesis temperatures are low (<300°C), a long range homogenization by solid state diffusion cannot occur. If an homogenization is observed, it is only by using ionic species demonstrating a dissolution/precipitation mechanism.
Aluminum is an amphoteric species: it is then possible to obtain a gel by destabilizing acid or basic sols. We have tested the two routes. For alumina/zirconia mixture the starting sol is acid. After mixing, the mixed sol can be gelled and its hydrothermal treatment leads to a mixture of bohemite and zirconia. If the gelling is conducted under inadequate stirring or addition of concentrated base, it leads on the appearance of chemical fluctuations in the gel. These can be as more than a few hundred microns length and made of almost pure alumina or zirconia areas. This because of the difference between the pH range stability of Al(OH)3 and Zr(OH)4 species. If the base is not instantaneously distributed through all the sol then, locally, pH wavering occurs leading on preferential precipitation of aluminum or zirconium hydroxide. After hydrothermal treatment, these long range wavering remains and enriched zirconia and alumina cluster are observed. Such an example is presented on figure 7. There is no more homogenization during the hydrothermal treatment. It must be obtained firstly a very close mixing of the precursors. This was done with of a low speed gelification protocol by means of the addition of a low concentrated base under vigorous agitation. The progressive variation of the pH value in all the suspension favors the simultaneously gelification of both the two precursors. In the same way, for the alumina/SiC composite, if a bad conducted gelling leads on the agglomeration of SiC dispersion and the floculation of almost pure aluminum specie, then the hydrothermal treatment can only stabilize these heterogeneities.
For these materials a pure in situ mechanism is observed.
The precipitation of powders through hydrothermal process are based on several mechanisms, the predominance of one mechanisms depend on the nature of the powder and on the hydrothermal parameters (temperature, nature of the mineralizer and concentrations of the reactive). The examples discussed above present some cases it is possible to observe several limiting steps process. The predominance of one limiting step depends on the systems involved. In a first approach it is of interest to promote a dissolution/precipitation mechanism facilitating the re-homogenization of the different species. Unfortunately, in most cases some particular species can remain in ionic soluble state at a high level concentration (lead for PZT, barium for barium titanate). Adding initial excess and washing steps are then necessary. Due to experimental limitations, fixing these parameters constitutes the main difficulties in optimizing an hydrothermal powder synthesis process.
On an another hand, a pure in situ mechanism is not so often observed, demonstrating the role of the liquid media. If it occurs, then attention has to be made on the gelling process and the obtainment of the most homogeneous as possible precursors.
|Issues | Table of Content|