SPHERICAL CALCIA STABILIZED ZIRCONIA POWDERS OBTAINED BY SPRAY PYROLYSIS

H. E. Esparza-Ponce*, A. Reyes-Rojas, M. Miki Yoshida

Centro de Investigación en Materiales Avanzados S.C., Miguel de Cervantes 120,
Complejo Industrial Chihuahua, Chihuahua, Chih. C.P. 31109, México
hilda.esparza@cimav.edu.mx, armando.reyes@cimav.edu.mx, mario.miki@cimav.edu.mx

Abstract

Physical and chemical properties of ceramic powders are highly dependent on the morphology and chemical composition of the precursor powders. Therefore, properties of ceramic powders are expecting to be improved by controlling the most important parameters of morphology such as composition of precursors, temperature and airflow. Promising techniques to produce ceramic powders with controlled morphology are sol-gel, coprecipitation, and pyrolysis. Among these techniques, spray pyrolysis is one of the most suitable to control the morphology and the chemical composition of the particles, because the spray pyrolysis technique integrates the evaporation, precipitation, decomposition and sintering stage of powder synthesis into a single continuos process. Spherical powders were prepared from acetylacetonate of zirconium and acetate calcium hydrated solutions by pneumatic spray pyrolysis. The morphology and size of the powders change with variation of airflow, temperature, and solution concentration.

Keywords: Spray pyrolysis technique, ultrafine powders, precursor ceramics, zirconia and calcia.

Resumen 

   Las propiedades físicas y químicas de los polvos cerámicos dependen grandemente de la morfología y de la composición química de las soluciones precursoras. Por lo tanto, las propiedades de los polvos cerámicos están esperando ser mejoradas controlando los parámetros que más influyen en la morfología como son la composición de las soluciones precursoras, temperatura y flujo de aire. Algunas de las técnicas que ofrecen producir polvos cerámicos con morfología controlada son: sol gel, coprecipitación y pirólisis. Dentro de estas técnicas la aspersión pirolítica es una de las más adecuadas para el control de la morfología y la composición química de las partículas, ya que la aspersión pirolítica integra la evaporación, precipitación, descomposición y estado de sinterización para la síntesis de polvos en un sólo proceso continuo. En este trabajo, los polvos obtenidos son esféricos y han sido preparados a partir de acetilacetonato de circonio y acetato de calcio en solución; mediante la técnica de aspersión pirolítica utilizando un aspersor neumático. La morfología y el tamaño de partícula cambian con las variaciones del flujo de aire, temperatura y la concentración molar de la solución.

Palabras clave: Zirconia, calcia, aspersión pirolítica, polvos nanométricos.

1. Introduction

   In the past twenty years, a considerable interest has arisen in zirconia-based (ZrO2-based) solid electrolyte materials for numerous high-temperature technology applications. High oxygen-ion conductivity combined with high thermochemical stability are responsible for the use of stabilized zirconia materials, for example in oxygen sensor or fuel cells [1, 2]. More applications as: knives and scissors, precision ball, valve balls and seats, high-density ball and pebble mill-grinding media. Rollers and guides for metal tube forming, thread and wire guides, hot metal extrusion dies, deep well down-hole valves and seats, powder compacting dies, marine pump seals and shaft guides, oxygen sensors, high temperature induction furnace susceptors, fuel cell membranes, electric furnace heaters over 2000 °C in oxidizing atmospheres [3].

Zirconia exists in three crystal phases at different temperatures. At very high temperatures (>2370 °C) the material has a cubic structure. At intermediate temperatures (1170 to 2370°C) it has a tetragonal structure. At low temperatures (below 1170°C) the material transforms to the monoclinic structure. The transformation from tetragonal to monoclinic is rapid and is accompanied by a 3 to 5 percent volume increase that causes extensive cracking in the material. This behavior destroys the mechanical properties of fabricated components during cooling and makes pure zirconia useless for any structural or mechanical application. Several oxides dissolve in the zirconia crystal structure can eliminate these crystal structure changes. Commonly used effective additives are MgO, CaO, and Y₂O₃. With sufficient amounts added, the high temperature cubic structure can be maintained at room temperature. Cubic stabilized zirconia is a useful refractory and technical ceramic material because it does not go through destructive phase transitions during heating and cooling.

 2. Experimental procedure

2.1. Preparation of Calcia stabilised Zirconia powder.

The experimental system used in this work is shown in Figure 1. It mainly consists of a pneumatic atomizer (a), in which the precursor solution is converted in a fine mist, then it was carried out by the airflow and injected by a nozzle (b) into the quartz tube (c), which takes the place of a reaction chamber. The airflow was regulated with a mass flow controller (d). The solution temperature was controlled precisely (± 1 °C) by partially immersing the solution container in a deionized water bath (e). The quartz tube was located inside of a cylindrical furnace (f), which has a very precise temperature control (± 1 °C). Finally, at the outermost side of the quartz tube, the electrostatic precipitator (g) is placed. This is composed of a recollection tube (h) and anode wire (i) connected to a high voltage source (j).

Figure 1. Schematic diagram of the pyrolytic system

Calcia stabilized zirconia (CaSZ) Zr 0.85 Ca0.15 O 1.85 powders were prepared from a solution of Zr[CH3COCH=C-O-CH3]4 and Ca(CH3COO)2 H2O, in a mixture of 60% deionized water and 40% absolute ethanol. A Zr:Ca molar proportion of 0.86:0.14 was chosen and total precursor concentrations of 0.01, 0.05 and 0.1 mol dm-3 were tested. The furnace temperature was varied between 800 to 1000 °C; different carrier gas flows was tried among 33 to 167 cm3.s-1.

2.2. Powder characterization.

The crystalline structure of the zirconia-based powders was analysed by X ray diffraction (XRD) in a D5000 X-ray diffractometer, using Cu Ka radiation, at 40kV and 30 mA. The 2q angle were swept in a range from 5 to 80°, with a 2q step size of 0.1 ° and step time of 5 s.

Particle composition was analysed by electron microscopy, using a JSM 5800LV scanning and a CM200 transmission electron microscopes, both equipped with a DX prime energy dispersive X ray spectrometer (EDS). In addition, crystalline structures were studied by selected area electron diffraction patterns (SAED), and compared with XRD results.
Additionally, particle size, morphology and microstructure were evaluated as functions of the experimental conditions. For the determination of the average particle sizes (APS) of each sample, around 200 particles were measured directly from several bright field transmission electron micrographs, using a digital imaging analyser IMAGE-PRO PLUS 4.1.

3. Results

3.1 Parameters in powder preparation by spray pyrolysis.

It is well established, for conventional pneumatic spraying, that the droplet sizes decrease when the airflow increases [2]. If the sprayed droplets are focussed directly towards the reaction chamber, the droplets size corresponds to that mentioned above. However, if the atomizer sprays down to the solution container and the mist of solution is carried out in an upward direction by the airflow; the droplet size selection occurs due to the combined effect of gravitational and Stokes' force [1]. When these forces reach equilibrium the droplet travels with a constant terminal velocity, which is function of the particle size. Therefore, only the droplets smaller than certain critical sizes can be carried out to the reaction chamber. This critical size (d_e) is airflow dependent, and for low and medium airflow (0 < Φ_A < 8 dm³ min^-1) in fact reverse the reliance between droplet size and airflow, i.e. average droplet size increase with the increase of airflow just to certain limit flow. Only if the gas carrier flow increases further (Φ_A Å 8 dm³ min^-1) the average droplet size can decrease. The real residence time (τ) of the droplets is in the region of high the temperatures of the cylindrical furnace, it also was estimated as a function of the gas carrier flow rate (Φ_A), taking into account the geometry of the reactor. Gas expansion correction was estimated by this factor, since the work of Milosevic et al. [2]. Subsequently, at a fixed furnace temperature the residence time decreased when airflow increased. In addition, the rate of droplet heating (R_H) has been estimated considering furnace temperature distribution, airflow rate and geometry of the reaction chamber. It can be concluded that, under the experimental conditions utilized in this work, very high increases of the heating rate (several thousand of °C s-1) happens when the airflow increases.

3.2 Composition and crystalline structure of ZrO₂-CaO Fine Powders.

Figure 2 shows characteristic EDS spectrum of CaSZ powders. The average Zr:Ca molar proportion in the powders is almost the same as in the precursor solution; i.e. Zr:Ca and is approximately equal to 0.86:0.14.

Figure 2. Characteristic EDS spectra of CaSZ powder.

Figure 3 shows the XRD spectrum of the calcia stabilized zirconia (CaSZ) powder. It is observed that diffraction peaks corresponding to the cubic and tetragonal phases, indicating that a partially stabilized structure has been formed. SAED results also confirm this finding. Table 1 shows interplanar distance calculated from SAED patterns of CaSZ powders. It can be concluded, the presence of Cubic and Tetragonal phases. A preliminary estimation using the Rietveld method reports a proportion of 5 weight percent for the tetragonal phase. An exhaustive study is in course.

Figure 3. XRD spectrum of CaSZ powder obtained at 900 °C.

Table 1. Interplanar distance calculated from SAED patterns. "T" Tetragonal and "C" Cubic phases.

3.3 Particle size of CaSZ powders.

In general, the growth and characteristics of the powders prepared by spray pyrolysis are determined by the combined effect of the droplet size generated in the sprayer, the size of those effectively transported (d_e) to the reaction chamber, the residence time (τ) and the rate of droplet heating (R_H) in the chamber. These parameters are, as well, dependent on the geometry of the system, principally spraying and reaction chambers, solution composition and concentration, airflow and to the reactor temperature.

First of all, it is worthwhile to consider that the average particle size (APS) measured in all the samples of this work are nano-metric (40 to 350 nm), they were considerably smaller than those calculated from the average droplet size using a well known approach [2]. This result can be explained by two possible mechanisms considering that each droplet did not convert to a unique particle. The first one, due to the occurrence of an important droplet broke down [2] or explosive boiling [6]. These events probably take place when the droplets travel into the nozzle, before they have been injected into the reaction chamber. At this stage the residence time (~10-3 s) is shorter than the drying time (~0.1 s [5]) and the heating rate is very high (>105 °C s-1), in these conditions a vigorous boil of the droplets can happen, given as a consequence, their broke down.

The second additional mechanism can occur after dehydration of the droplet/particle, at the beginning of their travel in the quartz tube. Due to the high temperature and very large surface area of the precursor salt particles, the total evaporation rate can be extremely high, given as a consequence, the loose of the precursor material or the formation of many ultra fine particles. This mechanism was also suggested for ZnO powders by Liu et al. [7]. It is expected that precursor evaporation was more intense at low airflow, i.e. when the residence time in the reaction chamber was high. Moreover, at high solution concentration the droplets tended to evaporate the solvent more quickly, thus, they formed precursor salt particles more easily and consequently the precursor evaporation rate increased. The experimental results are consistent with these two considerations. The APS was a bimodal distribution of sizes, small sizes (» 40 nm) at low 6 dm³ min^-1 and large size distribution (350 nm) at higher flows (see figure 4); additionally, the APS increased with the increase of solution concentration (see figure 5).

Figure 4. APS as a function of the airflow.

Figure 5. APS as a function of the solution concentration.

3.4. Morphology and inner structure of the CaSZ particles.

Secondary electron SEM micrograph representative of CaSZ powder is shown in figure 6. For this sample obtained at 900 ºC, 0.05 mol dm^-3, 4.6 kV, 5 dm³ min^-1, it can be observed that spherical particles are the predominant morphology; only a small number of them show a hollow or irregular shape. 

Figure 6. Secondary electron SEM micrograph of powders CaSZ.

Airflow has also a noticeable effect on the morphology of the particles. It is shown in figure 7, that the number of irregular hollow shaped particles increases as the airflow increases. The formation of hollow particles is related to the localized super-saturation of the solute with the consequent formation of a crust on the droplet surface [1]; which depends on the evaporation rate of the solvent, and on the strength and permeability of the crust. When the concentration of precursor solution is high enough, the crust is formed earlier on the droplet surface. The internal solvent, encapsulated by the forming crust, expands when heated and increases the particle size of the powder; rose after the complete evaporation of solvent of hollow particles that can break into irregular particles. It is expected that the influence of the residence time is higher when it is shorter than the time needed for complete evaporation of the solvent. In this case, with high enough airflow, the particle size tends to grow, because the formation of a crust occurs earlier by the cooperative effect of low τ and high R_H. This proposal is in concordance with the particle size (as discussed before) and porosity of the particles (see below), and with the results of other works [2, 7, 8, 1].                      

a)

b)

c)

d)

 

Figure 7. Bright field TEM micrographs of calcia stabilized zirconia obtained at airflow: a) 4, b) 6, c) 8 and d) 10 dm³ min^-1. The solution concentration was 0.05 mol dm^-3, furnace temperature 900 ºC and voltage of electrostatic precipitator 4.6 kV.

Figure 8 shows the variation of particles porosity as a function of the airflow. It is observed as a decrease of the porosity, hence the permeability, of the particles as the airflow increases. The logical consequences are the increase of irregular hollow particles and particle sizes with the increase of airflow, as described above.

In addition, TEM analysis demonstrates that the spherical particles contain inner small crystallites. Figure 9 shows that inner crystallite size growth as furnace temperature increases and airflow decreases. Figures 9a and 9b show bright field TEM micrographs of powders obtained at 900 and 1000 °C respectively. (6 dm³ min^-1, 4.6 kV, 0.1 mol dm^-3). In this case, it can be noted as a slight increment of crystallite size, from 2 - 6 nm to 3 - 8 nm approximately. In contrast, for samples obtained at 900 °C and an airflow of 2 dm³ min^-1 the crystallite size increases to 24 - 41 nm (see figure 9c).

a)

b)

c)

d)

Figure 8. Bright field TEM micrographs of calcia stabilized zirconia obtained at airflow: a) 4, b) 6, c) 8 and d) 10 dm³ min^-1. The solution concentration was 0.05 mol dm^-3, furnace temperature 900 ºC and voltage of electrostatic precipitator 4.6 kV.

a)

b)

c)

Figure 9. Bright field TEM micrographs of calcia stabilized zirconia obtained at: a) 900 °C, airflow 6 dm³ min^-1; b) 1000 °C, airflow 6 dm³ min^-1; and c) 900 °C and airflow 2 dm³ min^-1. The solution concentration was 0.1 mol dm^-3.

 4. Conclusions

The feasibility of produces Zirconia and Calcia stabilized zirconia nanopowders without post-heat treatment by spray pyrolysis has been demonstrated. Compared with the traditional method the powders are obtained in a step, saving time and energy.

The size, morphology, and structural characteristics of zirconia and CaSZ nanopowders depend principally on the source solution concentration, reaction chamber temperature, and airflow. Average particle size range between 40 to 350 nm.

Each particle is composed by many small crystallites of sizes varying between 2 to 40 nm.

In spite of porous or dense ceramic powder production, an appropriate precursor solution concentration is needed to fit the requirements.

Acknowledgment

The authors particularly acknowledge the experimental assistance of W. Antúnez-Flores, K. Gómez-Fierro, D. Lardizabal, and F. Paraguay D.

 References

[1 ] E.C. Subbarao, Zirconia-An Overview; pp 1-24 in Advances in Ceramics, Vol.3, Science and Technology of Zirconia I. Edited by A.H. Heurer and L.W. Hobbs. American Ceramic Society, Columbus, OH. 1981.         [ Links ]

[2] K.Okada, A. Tanaka, S. Hayashi, N. Otsuka, Journal of Materials Science Letters, 12 (1993) 854-857.        [ Links ]

[3] http://www.accuratus.com/Zirconium_Dioxide.htm.        [ Links ]

[4] M. Miki-Yoshida and E. Andrade, Thin Solid Films, 224, (1993) 87-96.         [ Links ]

[5] O. Milosevic et al., Thin Solid Films, 296 (1997) 44 - 48.        [ Links ]

[6] M. Shusser, T. Ytrehus, D. Weihs, Fluid Dynamics Research 27 (2000) 353.        [ Links ]

[7]T-Q. Liu, O. Sakurai, N. Mizutani, M. Kato, Journal of Materials Science (1986) 3698-3703.        [ Links ]

[8] S. Y. Cho, J. H. Lee, S. J. Park, J. of Materials Science 30 (1995) 3274 - 3278        [ Links ]

[9] B. Zhao, R. Zhang, L. Lu, H. Xie, Material Science and Engineering B49 (1997) 36 - 41.        [ Links ]

* Corresponding author. Fax.: (52 1) 439 11 12