Azeotropic distillation, ethanol washing, and freeze

Available online at www.sciencedirect.com

CERAMICS INTERNATIONAL

Ceramics International 41 (2015) 14148–14156 www.elsevier.com/locate/ceramint

Azeotropic distillation, ethanol washing, and freeze drying on coprecipitated gels for production of high surface area 3Y–TZP and 8YSZ powders: A comparative study R.H. Pivaa,n, D.H. Pivaa, J.J. Pierrib, O.R.K. Montedoc, M.R. Morellia,b Laboratory of Ceramic Synthesis and Formulation, Department of Materials Engineering, Federal University of São Carlos – UFSCar, São Carlos, São Paulo, Brazil b Department of Materials Engineering, Federal University of São Carlos ‒ UFSCar, São Carlos, São Paulo, Brazil c Grupo de Pesquisa CERTEC, Universidade do Extremo Sul Catarinense ‒ UNESC, Criciúma, Santa Catarina, Brazil

a

Received 22 May 2015; received in revised form 24 June 2015; accepted 6 July 2015 Available online 14 July 2015

Abstract Coprecipitation is a well-established method for synthesizing yttria-stabilized zirconia (YSZ); however, the hard agglomerates formed during the drying stage make it difficult to obtain high surface area powders. This paper compares azeotropic distillation, ethanol washing, and freeze drying, the most common dehydration methods, and assesses their ability to yield non-agglomerated powders of coprecipitated 3 mol% Y2O3 (3Y–TZP) and 8 mol% Y2O3-stabilized zirconia (8YSZ). Azeotropic distillation most effectively increased the surface area of the powders, yielding 3Y–TZP and 8YSZ powders with specific surface areas of 94.73 and 109.37 m2 g 1, respectively, and crystallite sizes between 9 and 10 nm. Ethanol washing did not homogenously dry out the precipitate, whereas the freeze drying method promoted the formation of hard agglomerates, as evidenced by the peculiar powder characteristics derived from this technique. Interestingly, the surface area seems to be increased in methods that reduce the crystallization enthalpy on the hydrous zirconia precipitates. & 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Keywords: D. ZrO2; Azeotropic distillation; Ethanol washing; Freeze drying

1. Introduction Yttria-stabilized zirconia (YSZ) is a multifunctional material that is widely utilized in several technological fields. When 8 mol% Y2O3 is alloyed in the zirconia-rich portion of the ZrO2–Y2O3 system, fully-stabilized zirconia (8YSZ) is formed [1]. Because of its large concentration of oxygen vacancies, 8YSZ is currently the electrolyte material of choice in solid oxide fuel cells [2], and is also applied in membrane technology for oxygen separation [3], and in gas sensors [4,5]. Alternatively, addition of 3 mol% Y2O3 produces stabilized tetragonal n Correspondence to: Universidade Federal de São Carlos – UFSCar, Programa de Pós-Graduação em Ciência e Engenharia de Materiais, Rodovia Washington Luís, Km 235, Sao Carlos CEP 13565-905, São Paulo, Brazil. Tel.: þ 55 16 3351 8508. E-mail address: [email protected] (R.H. Piva).

http://dx.doi.org/10.1016/j.ceramint.2015.07.037 0272-8842/& 2015 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

zirconia polycrystals (3Y–TZP) [1,6]. Dense fine-grained microstructures of 3Y–TZP exhibit excellent fracture toughness, high strength, hardness and wear behavior; thus, 3Y–TZP is an important structural material [7,8]. Moreover, these features are coupled with good biocompatibility, supporting the use of 3Y– TZP ceramic in orthopedic and dental implants [9,10]. Powders of 8YSZ and 3Y–TZP are most commonly synthesized by coprecipitation. This technique yields high purity powders with the desired morphology, controlled particle size, and chemical homogeneity [11,12]. Furthermore, the whole process occurs at room temperature and ambient pressure and requires no sophisticated equipment, so is relatively low-cost [13]. Nonetheless, these advantages are largely offset by the formation of hard agglomerates, which persists as a serious drawback of coprecipitated YSZ powders [14]. Hard agglomerates are formed by strong intercrystallite bonding during removal of the surface hydroxyl groups at the

R.H. Piva et al. / Ceramics International 41 (2015) 14148–14156

drying stage, and are consolidated in the calcination step by the following reaction [15–17]: Zr-OH þ HO-Zr-Zr-O-Zrþ H2O

(1)

These hard agglomerates hinder the development of the initial stages of the sintering step and hence the densification of YSZ [11,18]. Moreover, in the presence of hard aggregates, highly reactive powders that might reduce the sintering temperature cannot be achieved [1,18]. In contrast, in the presence of non-agglomerated powders with high surface area, the temperature of sintering initiation can be reduced by more than 200 1C [19]. For example, dense YSZ samples have been reported in ultra-low sintering temperatures such as 1100 1C [20], 1070 1C [21,22], and even 1050 1C [23,24]. In all of these cases, non-agglomerated particles with high surface area were essential for lowering the sintering temperature [22]. As is well-known, agglomeration in YSZ powders synthesized by coprecipitation is often reduced by post-synthesis treatment of the precipitates. Among the several methods available for this purpose, azeotropic distillation [25], ethanol washing [26], and freeze drying [27] methods have been regarded as the most effective. All of these approaches remove the capillary water and the surface hydroxyl groups from the coprecipitated gels. The underlying phenomena of these three methods have been intensively investigated in the dehydration of hydrous zirconia. However, to our knowledge, a comparative study of these procedures in the production of 3Y–TZP and 8YSZ ceramics is not available in the current literature. In this paper, the three aforementioned dehydration methods are compared for their ability to yield coprecipitated 3Y–TZP and 8YSZ powders with high surface area. The chemical modifications on the precipitates and the powder morphologies resulting from each method are explored and discussed in detail. 2. Materials and methods 2.1. Synthesis by coprecipitation and dehydration methods Zirconyl chloride (Alfa Aesar, ZrOCl2 8H2O, 99.9%) and yttrium oxide (Alfa Aesar, Y2O3, 99.9%) were used as starting materials. The 3Y–TZP and 8YSZ powders were synthesized by coprecipitation in aqueous media. An initial solution of zirconyl chloride was prepared in distilled deionized water. A stoichiometric amount of yttrium oxide was first dissolved in hot nitric acid, then mixed into the zirconyl solution (0.14 M). This solution was homogenized by stirring for 10 min at 25 1C. For coprecipitation, the precursor solution was added dropwise at 0.4 mL min 1 into a vigorously stirred ammonia solution (6 M, pH 10-11). The resulting precipitates were left to settle for 21 h without stirring in their mother liquor at room temperature. The aged precipitates were filtered and washed five times with distilled and deionized water. In the first two water washes, cation loss was prevented by an ammonia solution (1 M). The following three washes were performed in water only until no Cl ions remained (assessed by reaction in 1 M AgNO3 aqueous solution). The coprecipitated gels were dehydrated by three methods; azeotropic distillation (AD), ethanol washing (EW), and freeze

14149

drying (FD). For comparison, the coprecipitated gels were also dried by the conventional oven drying (OD) method, which usually forms hard agglomerates. In the standard OD method, the coprecipitated gels were oven-dried at 80 1C for 24 h. For the FD procedure, a Liotop freeze dryer model L101 was used. Initially, the precipitates were placed in petri dishes and cooled in a liquid nitrogen bath (-196 1C) for 2 min. The frozen precipitates were then quickly placed in a freeze dryer to remove the frozen water by sublimation. The condensator and storage camera were kept at -52 and 20 1C, respectively. The storage camera was vacuumed at 5 Pa, and the entire freeze drying process took 48 h. In the EW method, the precipitates were washed twice with ethanol for 1 h with stirring, then left to settle for 30 min and filtered. The precipitate:ethanol mass ratio was 1:3, close to the optimum precipitate:ethanol ratio recently published for 3Y–TZP [28]. Note that we do not investigate a fourth method that is sometimes used, the acetone–toluene–acetone (ATA) washing method, because this technique is (at most) as effective as EW [29,30]. In the AD method, the precipitates were mixed with iso-butanol at a mass ratio of 1:3 (precipitate:iso-butanol) with vigorous stirring for 1 h. Water was removed from this mixture by coevaporation at 93 1C with stirring, followed by iso-butanol evaporation at 117 1C. iso-Butanol is the preferred solvent in AD, as it permits a higher water content in the azeotrope [25]. Finally, the dehydrated powders were calcined at 550 1C in air for 1 h with heating/cooling rates of 3 1C min 1. No milling step was used to preserve all information on the characteristics of the powders resulting from each method. The whole synthesis process and sample identifications are shown in Fig. 1. 2.2. Powder characterization The phase compositions of the samples were identified by Xray diffraction (XRD) using a diffractometer (Siemens D5005) in Bragg–Brentano geometry with CuKα radiation at a scan speed of 11 min 1. The fraction of monoclinic zirconia in the calcined powder was determined from Toraya et al.'s equations [31]. Chemical modifications in the precipitates resulting from the different dehydration methods were obtained by Fourier transform infrared (FTIR) spectroscopy and differential scanning calorimetry (DSC). FTIR spectra were recorded with a Thermo Scientific NICOLET-IR200 spectrophotometer with 4 cm 1 resolution in the spectral range 400–4000 cm 1. Powder samples were mixed with KBr for FTIR recording. DSC analyses (TA Instruments SDTQ 600) of the dehydrated precipitates were recorded up to 700 1C at a heating rate of 10 1C min 1 in dry air, using alumina as the reference material. The temperature and enthalpy of crystallization of the precipitates were obtained from the DSC curves. The specific surface areas (SBET) of the calcined powders were measured by the Brunauer–Emmett–Teller method based on N2 adsorption–desorption (Micrometrics Gemini 2370 V1.02). The measured SBET was converted to particle size (dBET) according to the formula dBET ¼ 6/(ρSBET), where ρ is the theoretical density of 6.049 g cm 3 for 3Y–TZP and 5.965 g cm 3 for 8YSZ, calculated on the basis of previously measured XRD lattice parameters of the tetragonal (a¼ 0.51003 nm, c¼ 0.51749 nm) and cubic (a¼ 0.51361 nm) phases. The particle

14150


Fig. 1. Flow chart of the coprecipitation synthesis route and investigated dehydration methods. The 3Y–TZP and 8YSZ powders are distinguished by the numbers 3 and 8, respectively, prefixing the dehydration codes.

sizes and morphologies of the samples were observed by transmission electron microscopy (TEM) with an FEI Tecnai G2 F20 instrument (acceleration voltage¼ 200 kV). For TEM observations, the powders were ultrasonically dispersed in acetone for 10 min, dripped onto carbon films placed on copper grids (200 mesh), then dried at room temperature. The powder morphologies were observed by scanning electron microscopy (SEM) using a Philips XL-30 FEG microscope. The calcined powders were also characterized by volume occupancy test, where 1.5 g of powder was added to an assay tube and accommodated on a vibratory laboratory table for 60 s. For sintering evaluation, the calcined powders were uniaxially diepressed under 150 MPa into pellets of diameter 5 mm and length 6 mm. These green samples were subjected to optical dilatometry measurements up to 1500 1C at a heating rate of 10 1C min 1 (Misuras HSM ODHT 1400). 3. Results and discussion 3.1. Precipitate modifications induced by the dehydration methods Powders dehydrated by the standard OD method were checked for retention of the metastable tetragonal and cubic phases at room temperature. The XRD patterns (Fig. 2) reveal only the cubic phase (Fm3m) in the 8OD sample, whereas the 3OD sample exhibits diffraction peaks (110)/(002) at 2θ 351, (112)/(200) at 2θ 501 and (103)/(211) at 2θ 591, which characterize the tetragonal phase (P42/nmc) [32,33]. This sample also contains 14 vol% monoclinic zirconia (P21/c), as evidenced by the reflections of the (-111)m and (111)m planes. Hence, neglecting the minor phases, the calcined 8YSZ and 3Y–TZP powders can be regarded as cubic and tetragonal phases, respectively. The same phases will dominate in powders formed by the other dehydration methods, because all powders originated from the same coprecipitated gel.

Fig. 2. XRD patterns of (a) 8OD and (b) 3OD samples heat-treated at 1000 1C for 1 h. At this temperature, the tetragonal and cubic phases are reliably distinguished.

The DSC curves are shown in Fig. 3. As can be observed, each dehydration method introduces some peculiarities in the precipitates. The DSC curves of the precipitates dehydrated by OD and FD are quite similar, presenting two major peaks. Initially, both samples exhibit an endothermic peak at 100 1C, attributed to evaporation of nonstructural adsorbed water [27,34,35]. The water removal progresses with some minor fluctuations through the thermal readings up to the temperature of zirconia crystallization, which is characterized by an exothermic peak. In fact, the DSC curves of OD and FD dehydrated precipitates are expected to be similar, because these methods differ only the nature of the water removal at the drying stage (evaporation in OD; sublimation in FD). Therefore, the changes in the precipitates induced by FD and standard OD are not significantly different. In contrast, dehydration using organic media significantly alters the DSC curves. In the precipitates dehydrated by EW and AD, the broad initial endothermic peak


14151

Fig. 3. DSC curves of powders dehydrated by different methods: (a) 3Y–TZP and (b) 8YSZ.

of OD and FD is replaced by a narrow peak. The EW method yields an additional exothermic peak at 294/300 1C (3-YTZP/ 8YSZ) associated with the replacement of hydroxyl groups by chemisorbed ethoxy groups [26,28]. The AD-dehydrated samples exhibit one exothermic peak at 263/265 1C and another at around 307/306 1C, corresponding to the removal of butoxyl groups that are chemisorbed when hydrous zirconia interacts with iso-butanol [25,36]. Table 1 shows other striking features of the DSC data. The temperature at which the hydrous zirconia exothermically crystallizes is always higher in the 8YSZ than in the 3Y–TZP samples. Reportedly, this reflects the influence of Y3þ ions on the Zr-OH bonds, which changes the atomic diffusion during crystallization [37,38]. Furthermore, the crystallization temperature of the 3Y–TZP and 8YSZ powders strongly depends on the dehydration method. However, no reasonable correlation can be drawn because our results for YSZ, which revealed a higher crystallization temperature for precipitates dehydrated by EW then by OD (similar to the MgSZ system [39]), opposes the reported behavior in the CeYSZ system [40]. More remarkably, the crystallization enthalpy is greatly affected by the dehydration method, as shown in Table 1. The 3AD and 8AD samples showed the smallest crystallization enthalpy, followed by the EW, OD and FD methods. As discussed later, this feature intimately affects the surface area of the powders. Chemical modifications of the precipitates caused by the different dehydration methods are shown in Fig. 4. Consistent with the DSC results, the FTIR spectra of the 8OD and 8FD samples reveal no chemical changes on the precipitates. All of the dehydration methods present a broad band between 3000 and 3750 cm 1, corresponding to the stretching vibration of structural -OH groups and adsorbed water [41]. The band at 1630 cm 1 accounts for the vibration of hydroxyl groups in zirconium hydroxide [35,42]. The band at 1370 cm 1 indicates the presence of carbonate species resulting from the incorporation of atmospheric CO2 into the precipitates [43]. The FTIR spectrum of the EW dehydrated sample confirms that ethanolic groups form on the surface of the hydrous zirconia precipitate.

Table 1 Thermal features, specific surface areas and particle sizes of the powders obtained by different dehydration methods. Dehydration procedure

Crystallization temperature (1C)

Crystallization enthalpy (J g 1)

SBET dBET (m2 g 1) (nm)

3AD 3EW 3OD 3FD 8AD 8EW 8OD 8FD

472 483 480 496 502 508 501 516

37.16 101.70 107.80 123.20 58.15 85.89 91.90 113.50

94.73 81.63 59.23 65.89 109.37 93.40 61.20 66.82

10.47 12.15 16.75 15.05 9.20 10.77 16.44 15.05

In this case, the band at 1046 cm 1 indicates the replacement of terminal hydroxyl groups by ethoxy groups (-OC2H5) via Reaction 2 [41]. In the same way, AD dehydration yields additional small bands at 2873 and 2928 cm 1, corresponding to C-H stretching vibrations. The peak at 1044 cm 1 corresponds to butoxy groups in -OC4H9 terminal species, which replace the hydroxyl groups (Reaction 3) [14,44]. These spectra provide further evidence that ethoxy and butoxy groups adsorb to the precipitate surface in the EW and AD methods, respectively, supporting the DSC data. Zr1-xYx(OH)y þ HOC2H5-H2O þ (OH)y 1…Zr-OC2H5

(2)

Zr1-xYx(OH)y þ HOC4H9-H2O þ (OH)y 1…Zr-OC4H9

(3)

3.2. Powder characteristics and their relations to the dehydration methods The benefits of using organic media in the dehydration methods are clearly visible in the surface area results (see Table 1). AD dehydration yielded the highest SBET values (94.73 and 109.37 m2 g 1 for 3AD and 8AD samples,

14152


respectively). These high surface areas are related to the substitution of -OH groups on the precipitates by -OC4H9 molecules, which are weakly adsorbed by van der Waals forces. This well-known effect is detected in the DSC and FTIR analyses (see Figs. 3 and 4). Thus, in the AD method, the residual water content is sufficiently small that agglomerates cannot form by hydroxide bridging (Reaction 1) during the drying stage and subsequent calcination stages [14,44,45]. Note that, regardless of dehydration method, the cubic phase shows greater surface area than the tetragonal phase (Table 1), consistent with a recent report [46]. Although the surface area is lower in powders dehydrated by EW than by AD, the mechanism responsible for reducing the agglomeration degree is reportedly similar in EW and in distillation with iso-butanol

Fig. 4. FTIR spectra of 8YSZ powders dehydrated by different methods. The spectra of the 3Y–TZP and 8YSZ samples are essentially identical.

[17,26,41]. The surface tensions of ethanol (21.82 mN m 1 at 25 1C [47]) and iso-butanol (24.18 mN m 1 at 25 1C [48]) are very close, suggesting that surface tension is not the main influencer of the degree of agglomeration. The surface areas in the EW and AD techniques likely differ because butoxy groups more effectively prevent the formation of surface Zr−O−Zr bonds than ethoxy groups. This hypothesis is supported by the significantly lower crystallization enthalpy of the ADdehydrated precipitates than of EW-dehydrated precipitates. Furthermore, one general tendency is clearly shown in Table 1: there is an increase of the surface area values for methods which induce lower crystallization enthalpy. The TEM images shown in Fig. 5 also demonstrate the advantage of the AD over other studied methods. The 8AD sample (Fig. 5a) exhibits the highest crystallite dispersion; such fine and non-agglomerated crystallites are responsible for the continuous ring pattern in the electron diffraction [49]. The SBET of the 8AD sample corresponds to a crystallite size of 9.20 nm, consistent with the TEM image and also with Yao et al. [36], who recently reported a crystallite size of 7.8–16.5 nm for 8YSZ distillated precipitates calcined between 400 and 600 1C. In contrast, the TEM image of the 8FD sample shows much agglomeration (Fig. 5c), concordant with its smaller surface area. Importantly, the dehydration method only prevents agglomerates from forming; therefore, it should not significantly change the crystallite size [17,50]. The larger particle sizes of the 8FD samples (see Table 1) can be attributed to overlap of adjacent crystallites via the Zr−O−Zr bonds, which would affect the crystallite sizes calculated from the SBET measurements. The potential benefit of the FD method is that water on the precipitates is removed by sublimation. Theoretically, this process should reduce the agglomeration degree by lowering the effect of the surface tension of liquid water on the precipitates [51–53]. However, this mechanism is not sufficient to increased surface area of the investigated YSZ powders. For example, the SBET values of the 8FD and 8OD samples are very similar (66.82 m2 g 1 versus 61.20 m2 g 1). Likewise, Nair et al. [54] reported that coprecipitated lanthanum

Fig. 5. TEM images and electron diffraction patterns of (a) 8AD, (b) 8OD, (c) 8FD, and (d) 8EW.


zirconate precipitates dehydrated by FD and OD present essentially identical surface areas. The higher agglomeration in powders dehydrated by FD than by other methods is visible in the SEM images (Fig. 6). The 8FD powder (Fig. 6a) comprises dense and highly agglomerated particles, as also observed in TEM (Fig. 5c). Agglomeration in the FD powders is likely promoted by a phenomenon called freeze pressing [53,55], in which aggregation is mechanically induced by the formation and growth of ice crystals during freezing in liquid nitrogen. Such aggregation promotes the approach and subsequent compression of the amorphous hydrous zirconia particles, leading to large and highly densified agglomerates. Moreover, some of the agglomerates in the FD-dehydrated sample exhibit a plate-like morphology. Such bidimensional plate-like growth was reported in a previous kinetic study on the crystallization and growth of FD-based 3Y–TZP powders [13]. The FD method induces this peculiar morphology by excluding hydrous zirconia particles from the advancing ice front, forcing them into aligned structures [53,56,57]. On the other hand, fluffy agglomerates appear on the precipitate dehydrated by AD (Fig. 6c). The homogenous distribution of these agglomerates is another advantage of the AD method over the EW method. The dense agglomerates in the SEM images of the 8EW powder (Fig. 6d) are sourced from the high residual water content in some parts of the precipitate, indicating that water removal is less complete during the EW procedure than during AD. The morphological characteristics and agglomeration degree that results from each dehydration method significantly affects

14153

the particle packing, as shown in Fig. 7. According to these volume occupancy tests, the FD method yields comparatively small volumes of the 8YSZ and 3Y–TZP powders, consistent with the high agglomeration degree of FD-dehydrated particles.

Fig. 7. Volume occupied by (a) 3Y–TZP and (b) 8YSZ calcined powders.

Fig. 6. SEM images of 8YSZ calcined powders dehydrated by different methods: (a) 8FD, (b) 8OD, (c) 8AD, and (d) 8EW. The 8FD sample is imaged at a lower magnification to accommodate its larger agglomerates.

14154


sintering test. This behavior was expected, as the previous characterizations revealed higher surface area and lower agglomeration in powders derived from AD than from other methods. These results imply that, by dehydrating the coprecipited 3Y– TZP and 8YSZ gels by the AD method, we can improve the sintering performance of the powders. The FD powders demonstrated poor sinterability on account of the hard agglomerates formed at the drying stage, which persisted even at high sintering temperatures. This highlights the importance of properly selecting the dehydration method to maximize the surface area, and thereby achieve proper sintering of the YSZ ceramics. 4. Conclusions

Fig. 8. Linear shrinkage and relative density as functions of temperature for (a) 8YSZ and (b) 3Y–TZP. Insets are images of samples sintered at 1500 1C, displayed for visual examination of the shrinkage behavior.

Although powders with greater surface areas values should occupy greater volumes, as reported in tap density experiments of the EW technique [28], this assumption may not hold when comparing different dehydration methods. As shown in Fig. 7, the powder products of AD-dehydrated precipitates occupied lower volume than the EW dehydration products, whereas the reverse relationship is expected if the volume is related only to the surface area. Since the agglomerate morphology largely determines the particle packing behavior [58], this unexpected lower volume occupancy after AD (despite the enlarged SBET) probably originates from the reduced friction generated between the homogeneous fluffy agglomerates formed by this method, relative to the irregular agglomerates yielded by the EW method (inefficient EW agglomerates in Fig. 6d). Fig. 8 shows the results of simple linear shrinkage (ΔL/L0) and relative density tests. From these results, we can elucidate how the dehydration method affects the sintering behavior. The linear shrinkage curve is converted to (percentage) relative density as follows [59]: 3 ρ ¼ 1= 1 ΔL=L0 ρ0

ð4Þ

where ρ and ρ0 are the sintered and green densities, respectively. The 8YSZ and 3Y–TZP powders dehydrated by AD demonstrated superior sinterability, approaching full density − higher temperature than normally reported for nanosized YSZ powders [20–24] probably due to the high heating rate − at the end of the

In the production of 3Y–TZP and 8YSZ powders with high surface area, azeotropic distillation is a more efficient dehydration method than ethanol washing and freeze drying. Moreover, this method obtains high-quality yttria-stabilized zirconia powders from coprecipitated gels synthesized in aqueous medium at room temperature and ambient pressure, which reduces the cost of synthesis. The major advantage of azeotropic distillation over ethanol washing is the homogeneity after the drying stage, apparently because the butoxy groups more effectively prevent agglomeration of the zirconia crystallites than ethoxy groups. Furthermore, the crystallization enthalpy is closely related to the surface area of the powders. In both 3Y–TZP and 8YSZ powders, lower crystallization enthalpy was associated with higher surface area. The results of freeze drying and standard oven drying were almost identical, demonstrating that freeze drying (with water as solvent) cannot easily yield powders with high surface area from yttria-stabilized zirconia precipitates. Acknowledgments This work is supported by the São Paulo Research Foundation (FAPESP) Grant no. 2013/14189-8 and the Brazilian National Council of Technological and Scientific Development (CNPq). References [1] O.C. Standard, C.C. Sorrell, Densification of zirconia – conventional methods, Key Eng. Mater. 153-154 (1998) 251–300. [2] H. Nomura, S. Parekh, J.R. Selman, S. Al-Hallaj, Fabrication of YSZ electrolyte for intermediate temperature solid oxide fuel cell using electrostatic spray deposition: II – cell performance, J. Appl. Electrochem. 35 (2005) 1121–1126. [3] F. Ciacchi, Tubular zirconia–yttria electrolyte membrane technology for oxygen separation, Solid State Ion. 152-153 (2002) 763–768. [4] J.W. Fergus, Doping and defect association in oxides for use in oxygen sensors, J. Mater. Sci. 38 (2003) 4259–4270. [5] W.C. Maskell, Progress in the development of zirconia gas sensors, Solid State Ion. 134 (2000) 43–50. [6] D. Kim, T.-Y. Tien, Phase stability and physical properties of cubic and tetragonal ZrO2 in the system ZrO2-Y2O3-Ta2O5, J. Am. Ceram. Soc. 74 (1991) 3061–3065. [7] R.H.J. Hannink, P.M. Kelly, B.C. Muddle, Transformation toughening in zirconia-containing ceramics, J. Am. Ceram. Soc. 83 (2004) 461–487. [8] X.-J. Jin, Martensitic transformation in zirconia containing ceramics and its applications, Curr. Opin. Solid State Mater. Sci. 9 (2005) 313–318.

R.H. Piva et al. / Ceramics International 41 (2015) 14148–14156 [9] C. Piconi, G. Maccauro, Zirconia as a ceramic biomaterial, Biomaterials 20 (1999) 1–25. [10] J. Chevalier, What future for zirconia as a biomaterial?, Biomaterials 27 (2006) 535–543 [11] P. Llewellyn, L. Montanaro, F. Rouquérol, Investigation of the dehydration of Y-TZP gels by controlled transformation rate thermal analysis, Solid State Ion. 95 (1997) 23–28. [12] M. Keshmiri, O. Kesler, Colloidal formation of monodisperse YSZ spheres: kinetics of nucleation and growth, Acta Mater. 54 (2006) 4149–4157. [13] Y.-W. Hsu, K.-H. Yang, K.-M. Chang, S.-W. Yeh, M.-C. Wang, Synthesis and crystallization behavior of 3 mol% yttria stabilized tetragonal zirconia polycrystals (3Y–TZP) nanosized powders prepared using a simple coprecipitation process, J. Alloy. Compd. 509 (2011) 6864–6870. [14] L. Chang-Wei, Q. Hai-Bo, F. Chu-De, X. Tong-Geng, Y. Xue-Hua, C. Yun-Xian, G. Jing-kun, Investigation on the mechanism of n-butanol interaction with zirconia powder, Thermochim. Acta 246 (1994) 99–105. [15] K. Haberko, A. Ciesla, A. Pron, Sintering behaviour of yttria-stabilized zirconia powders prepared from gels, Ceramurg. Int. 1 (1975) 111–116. [16] S.L. Jones, C.J. Norman, Dehydration of hydrous zirconia with methanol, J. Am. Ceram. Soc. 71 (1988) C-190–C-191. [17] M.J. Readey, R. Lee, J.W. Halloran, A.H. Heuer, Processing and sintering of ultrafine MgO–ZrO2 and (MgO, Y2O3)-ZrO2 powders, J. Am. Ceram. Soc. 73 (1990) 1499–1503. [18] W.H. Rhodes, Agglomerate and particle size effects on sintering yttriastabilized zirconia, J. Am. Ceram. Soc. 64 (1981) 19–22. [19] Z.Z. Fang, H. Wang, Densification and grain growth during sintering of nanosized particles, Int. Mater. Rev. 53 (2008) 326–352. [20] M. Trunec, K. Maca, Compaction and pressureless sintering of zirconia nanoparticles, J. Am. Ceram. Soc. 90 (2007) 2735–2740. [21] C.D. Sagel-Ransijn, A.J.A. Winnubst, A.J. Burggraaf, H. Verweij, Grain growth in ultrafine-grained Y-TZP ceramics, J. Eur. Ceram. Soc. 17 (1997) 1133–1141. [22] P. Duran, M. Villegas, F. Capel, J.F. Fernandez, C. Moure, Nanostructured and near defect-free ceramics by low-temperature pressureless sintering of nanosized Y-TZP powders, J. Mater. Sci. 32 (1997) 4507–4512. [23] G.S.A.M. Theunissen, A.J.A. Winnubst, A.J. Burggraaf, Sintering kinetics and microstructure development of nanoscale Y-TZP ceramics, J. Eur. Ceram. Soc. 11 (1993) 315–324. [24] M.M.R. Boutz, R.J.M. Olde Scholtenhuis, A.J.A. Winnubst, A. J. Burggraaf, A hydrothermal route for production of dense, nanostructured Y-TZP, Mater. Res. Bull. 29 (1994) 31–40. [25] H. Qiu, L. Gao, C. Feng, J. Guo, D. Yan, Preparation and characterization of nanoscale Y-TZP powder by heterogeneous azeotropic distillation, J. Mater. Sci. 30 (1995) 5508–5513. [26] P.D.L. Mercera, J.G. Van Ommen, E.B.M. Doesburg, A.J. Burggraaf, J.R.H. Ross, Influence of ethanol washing of the hydrous precursor on the textural and structural properties of zirconia, J. Mater. Sci. 27 (1992) 4890–4898. [27] C. Tallón, R. Moreno, M.I. Nieto, Synthesis of ZrO2 nanoparticles by freeze drying, Int. J. Appl. Ceram. Technol. 6 (2009) 324–334. [28] S.B. Patil, A.K. Jena, P. Bhargava, Influence of ethanol amount during washing on deagglomeration of Co-precipitated calcined nanocrystalline 3YSZ powders, Int. J. Appl. Ceram. Technol. 10 (2013) E247–E257. [29] B.-S. Chiou, W.Y. Hsu, J.G. Duh, Dehydration of synthesized calciastabilized zirconia from a coprecipitation process, J. Mater. Sci. Lett. 5 (1986) 931–934. [30] D.M. Ávila, E.N.S. Muccilo, Effects of synthesis parameters on electrical conductivity and microstructural development of fine zirconia powders, J. Mater. Sci. Lett. 16 (1997) 685–688. [31] H. Toraya, M. Yoshimura, S. Somiya, Calibration curve for quantitative analysis of the monoclinic-tetragonal ZrO2 system by X-ray diffraction, J. Am. Ceram. Soc. 67 (1984) C-119–C-121. [32] C.G. Kontoyannis, M. Orkoula, Quantitative determination of the cubic, tetragonal and monoclinic phases in partially stabilized zirconias by Raman spectroscopy, J. Mater. Sci. 29 (1994) 5316–5320.

14155

[33] A. Ghosh, A.K. Suri, M. Pandey, S. Thomas, T.R. Rama Mohan, B. T. Rao, Nanocrystalline zirconia-yttria system–a Raman study, Mater. Lett. 60 (2006) 1170–1173. [34] P. Jakubus, A. Adamski, M. Kurzawa, Z. Sojka, Texture of zirconia obtained by forced hydrolysis of ZrOCl2 solutions Influence of aging on thermal behavior, J. Thermal Anal Calorim. 72 (2003) 299–310. [35] M. Rezaei, S.M. Alavi, S. Sahebdelfar, Z.-F. Yan, H. Teunissen, J.H. Jacobsen, J. Sehested, Synthesis of pure tetragonal zirconium oxide with high surface area, J. Mater. Sci. 42 (2007) 1228–1237. [36] H.-C. Yao, X.-W. Wang, H. Dong, R.-R. Pei, J.-S. Wang, Z.-J. Li, Synthesis and characteristics of nanocrystalline YSZ powder by polyethylene glycol assisted coprecipitation combined with azeotropicdistillation process and its electrical conductivity, Ceram. Int. 37 (2011) 3153–3160. [37] S. Ramanathan, R.V. Muraleedharan, S.K. Roy, P.K.K. Nayar, Dehydration and crystallization kinetics of zirconia-yttria gels, J. Am. Ceram. Soc. 78 (1995) 429–432. [38] J.-D. Lin, J.-G. Duh, The effect of dopants and organic washing on crystallization temperature of yttria- and ceria-doped zirconia gels, J. Mater. Sci. Lett. 16 (1997) 843–845. [39] S. Wang, Y. Zhai, X. Li, Y. Li, K. Wang, Coprecipitation synthesis of MgO-doped ZrO2 nano powder, J. Am. Ceram. Soc. 89 (2006) 3577–3581. [40] J.G. Duh, J.U. Wan, Developments in highly toughened CeO2–Y2O3– ZrO2 ceramic system, J. Mater. Sci. 27 (1992) 6197–6203. [41] M.S. Kaliszewski, A.H. Heuer, Alcohol interaction with zirconia powders, J. Am. Ceram. Soc. 73 (1990) 1504–1509. [42] D. Sarkar, D. Mohapatra, S. Ray, S. Bhattacharyya, S. Adak, N. Mitra, Synthesis and characterization of sol–gel derived ZrO2 doped Al2O3 nanopowder, Ceram. Int. 33 (2007) 1275–1282. [43] H. Wang, G. Li, Y. Xue, L. Li, Hydrated surface structure and its impacts on the stabilization of t-ZrO2, J. Solid State Chem. 180 (2007) 2790–2797. [44] H. Shan, Z. Zhang, Preparation of nanometre-sized ZrO2/Al2O3 powders by heterogeneous azeotropic distillation, J. Eur. Ceram. Soc. 17 (1997) 713–717. [45] W. Luan, L. Gao, J. Guo, Study on drying stage of nanoscale powder preparation, Nanostruct. Mater. 10 (1998) 1119–1125. [46] M. Trunec, K. Castkova, P. Roupcova, Effect of phase structure on sintering behavior of zirconia nanopowders, J. Am. Ceram. Soc. 96 (2013) 3720–3727. [47] G. Vazquez, E. Alvarez, J.M. Navaza, Surface Tension of Alcohol WaterþWater from 20 to 50 1C, J. Chem. Eng. Data 40 (1995) 611–614. [48] G.C. Benson, V.T. Lam, Surface tensions of binary liquid systems. II. Mixtures of alcohols, J. Colloid Interface Sci. 38 (1972) 294–301. [49] N.A. Zaporina, O.A. Doynikova, A.P. Krumina, D.A. Bocharov, J.P. Grabis, Methods of electron microdiffraction and X-ray analysis in structure study of nanodisperse partially stabilized ZrO2 powders, J. Surf. Investig. X-Ray, Synchrotron Neutron Tech. 3 (2009) 464–467. [50] A.K. Pandey, K. Biswas, Effect of agglomeration and calcination temperature on the mechanical properties of yttria stabilized zirconia (YSZ), Ceram. Int. 40 (2014) 14111–14117. [51] Z.L. Wang, W.H. Finlay, M.S. Peppler, L.G. Sweeney, Powder formation by atmospheric spray-freeze-drying, Powder Technol. 170 (2006) 45–52. [52] L. Qian, H. Zhang, Controlled freezing and freeze drying: a versatile route for porous and micro-/nano-structured materials, J. Chem. Technol. Biotechnol. 86 (2011) 172–184. [53] O.Y. Kurapova, V.G. Konakov, S.N. Golubev, V.M. Ushakov, I. Y. Archakov, Cryochemical methods for manufacturing nanosized ceramics and ceramic precursor powders with low agglomeration degree : a review, Rev. Adv. Mater. Sci. 32 (2012) 112–132. [54] J. Nair, P. Nair, J.G. Ommen, J.R.H. Ross, A.J. Burggraaf, Effect of postprecipitation treatment on the pore-structure stability of sol-gel-derived lanthanum zirconate, J. Am. Ceram. Soc. 81 (2005) 1487–1492. [55] J. Zheng, L. Winnubst, S. Fang, D. Salamon, Manipulation of Sintering Behavior by Initial Freeze Pressing an Aqueous Alumina Suspension, Adv. Eng. Mater. 13 (2011) 77–81. [56] S. Deville, E. Saiz, A.P. Tomsia, Ice-templated porous alumina structures, Acta Mater. 55 (2007) 1965–1974.

14156


[57] E. Munch, E. Saiz, A.P. Tomsia, S. Deville, Architectural control of freeze-cast ceramics through additives and templating, J. Am. Ceram. Soc. 92 (2009) 1534–1539. [58] C.D. Sagel-Ransijn, A.J.A. Winnubst, B. Kerkwijk, A.J. Burggraaf, H. Verweij, Production of defect-poor nanostructured ceramics of yttria-zirconia, J. Eur. Ceram. Soc. 17 (1997) 831–841.

[59] T.R.G. Kutty, P.V. Hegde, K.B. Khan, U. Basak, S.N. Pillai, A. K. Sengupta, G.C. Jain, S. Majumdar, H.S. Kamath, D.S. C. Purushotham, Densification behaviour of UO2 in six different atmospheres, J. Nuclear Mater. 305 (2002) 159–168.