- Email: [email protected]
Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics
+Model
ARTICLE IN PRESS
JECS-10003; No. of Pages 9
Available online at www.sciencedirect.com
ScienceDirect Journal of the European Ceramic Society xxx (2015) xxx–xxx
Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics Jiao He a , Xiaodong Li a,∗ , Shaohong Liu a , Qi Zhu a , Ji-Guang Li a,b , Xudong Sun a a
Key Laboratory for Anisotropy and Texture of Materials (Ministry of Education), School of Materials and Metallurgy, Northeastern University, Shenyang 110004, China b Nano Ceramics Center, National Institute for Materials Science, Tsukuba, Ibaraki 305-0044, Japan Received 1 December 2014; received in revised form 6 February 2015; accepted 7 February 2015
Abstract The effects of pre-treating a nano-sized Y2 O3 powder with sulfuric acid were studied. Both the structural evolution and sintering behavior as a function of SO4 2− -ion concentrations were investigated. The surface morphologies and aggregation state of particles were notably affected by the sulfuric acid-based treatment. The powder untreated revealed particle of flat surfaces and sharp edges, whereas the powders with sulfate dosage were composed of round particles of isotropic surfaces after calcination. Dispersive Y2 O3 nanopowder with excellent sintering property was obtained by calcinating the powders treated with 7 mol% sulfuric acid. The optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces, protecting particles from forming hard agglomerates by contributing to mass transportation mechanisms leading to particle coarsening upon calcination. All the ceramics prepared using treated-powders were transparent after vacuum sintering at 1700 ◦ C, whereas it was opaque for the sample without sulfate dosage. © 2015 Elsevier Ltd. All rights reserved. Keywords: Yttria; Transparent ceramics; SO4 2− -ion; Sintering; Powder
1. Introduction Recently, yttria transparent ceramic material has been extensively studied, because of its promise for a number of optical applications, such as luminous pipes for high-intensitydischarge lamps, heat-resistive windows, missile domes, host materials for scintillators and solid lasers, etc. [1–3]. In order to achieve a high degree transparency in yttria ceramics, internal porosity of samples should be completely removed [4]. One of the key technologies for fabricating highly dense ceramics is the synthesis of ceramic powders with excellent sintering activity, generally possessing such powder characteristics as low agglomeration state, ultrafine particle size (∼100 nm), narrow particle size distribution, and spherical particle shape [5,6]. Thus, the past two decades have witnessed tremendous efforts endeavored in the synthesis of high quality yttria nano powders through
∗
Corresponding author. Tel.: +86 24 83691580. E-mail address: [email protected] (X. Li).
different processing methods, such as precipitation methods [7–9], hydrothermal synthesis [10], emulsion synthesis [11], spray-drying [12], sol–gel synthesis [13,14], and combustion synthesis [15]. Among which, precipitation synthesis routes are attractive because of their accessibility. Most developed precipitation methods for synthesizing sinterable yttria powders always involve the addition of SO4 2− -ions [7,8,16–21]. It was found that the morphology, microstructure and sinterability of the calcined powders are highly affected by the [SO4 2− ]/[Y3+ ] molar ratios. Similar results have also been shown when the precipitation methods were used for the synthesis of Al2 O3 [22,23], Sc2 O3 [5,24,25] and YAG [6,26–31] powders by doping SO4 2− -ions. The doping by SO4 2− -ions can be realized in several different ways when ammonium sulfate is used as SO4 2− -ion sources. The ammonium sulfate can be added into reaction mixture during precipitation [19,27,30] during aging [7,17], or the resulting precipitant can be rinsed by ammonium sulfate solution [16]. Alternatively, SO4 2− -containing substances can be used as a starting material in the reaction [5,25,28,29].
http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008 0955-2219/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 2
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Previously, an amount of investigations have been performed to reveal the mechanisms of doped SO4 2− ions on precursor and calcined powders [5–9,25,26]. Ammonium sulfate was proved to be a regulator, mediating the nucleation and growth of precursors [8,9]. Among common inorganic anions, SO4 2− has a higher coordination capability than NO3 − and Cl− and shows stronger complexing effects [25]. The addition of the ammonium sulfate directly into reaction mixture gives rise to a local change of solution chemistry and consequently variations of chemical composition of resulted precursor powders [26]. It has been indicated that even a trace amount of SO4 2− decreased the monodispersity of the carbonate particles obtained via homogeneous precipitation with urea [8]. When ammonium sulfate was added into reaction mixture during aging, or the resulting precipitant was rinsed by ammonium sulfate solution, on the other hand, the morphological features of precursor particles were less frequently affected [7]. With respect to calcined powders, although it was generally accepted that doping a certain amount of SO4 2− -ions promotes formation of monodispersed spherical powders upon calcination, the influence mechanisms were not clearly verified. The reasons may be two-folds. On one hand, the high-temperature calcination process results in both the dissociation of precursor and the desulfurization of sulfate; the consecutive occurrence of two thermal decomposition procedures complicated the analysis aimed at understanding the mechanism of doped sulfate on calcined powders [24,26]. On the other hand, previous synthetic methods involving sulfate addition are not amenable to precise control of the amount of the doped sulfate in the precursors; it should be noted that the precursors produced by precipitation method are repeatedly washed to remove impurities such as NH4 + and NO3 − , resulting in a lot of sulfate added being washed away. As a consequence, the optimum sulfate dosage was significantly varied in different reaction systems [30,31]. And a correlation between powder property and the exact amount of doped sulfate was not always established. In this work, we investigate the effects of doped SO4 2− ions on a commercial Y2 O3 powder by pre-treating the starting material with sulfuric acid. The purpose of the work is to understand the underlying mechanism for the observed morphological changes and sinterability improvement achieved upon sulfate dosage. Both the structural evolution and sintering behaviors as a function of the amount of doped SO4 2− -ions were investigated. The direct application of SO4 2− to pure yttria, in deed, allows us to discriminate and analyze particle features contributed solely from the doped sulfate. In particular, we show that the optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces, protecting particles from forming hard aggregates by contributing mass transportation mechanisms leading to particle coarsening while suppressing the mechanisms giving rise to densification upon calcination. This investigation may provide an insight into more efficient application of anion dopant for manipulating particle features and improving the sinterability of available ceramic powders by a simple sulfate-based pre-treatment.
2. Experimental procedure 2.1. Preparation process Commercial Y2 O3 (99.99% pure, Huizhou Ruier Rare Chemical Hi-Tech Co. Ltd., Huizhou, China) powder was used as the starting material. The powder has a BET surface area of 31.5 m2 /g. X-ray diffractometric (XRD) analysis indicated that the Y2 O3 powder was a cubic-structured polymorph and the crystallite size calculated from Scherrer’s equation is 19 nm. In a typical processing, 60 g starting Y2 O3 powder was dispersed in ethyl alcohol with the addition of sulfuric acid and 2 wt% of dispersant (ammonium salt of poly(methacrylic acid)), and then the suspensions were ball milled in polyurethane jars filled with ZrO2 balls for 8 h. The amounts of added sulfuric acid were expressed as a molar percentage of the Y2 O3 powder basis (mol%). For comparison purpose, the starting Y2 O3 powder without sulfuric acid addition was also ball-milled identically. The resultant suspensions were then dried at 110 ◦ C, followed by sieving through a 200-mesh nylon sieve for pulverization. After sieving, the powder was calcined at 900–1200 ◦ C for 4 h in furnace in a stagnant air condition. The green compacts were prepared by dry pressing at 100 MPa, followed by hydrostatical isostatic pressing under a pressure of 200 MPa. 2.2. Characterization Powder morphologies were observed by transmission electron microscopy (TEM, Model JEM-2100F, JEOL, Tokyo, Japan). The average particle size (dTEM ) was determined from over 200 randomly selected particles with an image analysis software (WinRoof, Mitani Corp., Tokyo, Japan), by assuming a circular shape. The specific surface area of the calcined powders was measured by the BET method (Model TriStar II 3020, Micromeritics Instrument Corp., Norcross, GA) via nitrogen adsorption at 77 K. The equivalent particle size (dBET ) was estimated from BET surface area based on the following equation: dBET =
6 × 103 ρS
(1)
where ρ is the theoretical density (5.01 g cm−3 ) of Y2 O3 crystal; S is the specific surface area (m2 /g) determined by BET measurement. Thermal analysis of the powders doped with sulfate was performed using a TG/DSC analyzer (Model SETSYS Evolution-16, Setaram, Lyons, France) in flowing oxygen atmosphere (40 mL/min). The heating rate was 10 ◦ C/min. Fourier transform infrared spectroscopy (FT-IR) (Model Spectrum RXI, Perkin-Elmer, CT, USA) spectra of the yttria powders were recorded by the standard KBr method. The calcined Y2 O3 powder of 0.003 g was grinded and dispersed homogeneously in 0.25 g dried KBr powder using the agate mortar and pestle, and then uniaxially compacted into transparent pellets at 28 MPa. The resultant pellets were dried in the oven before investigation. The quantitative sulfur analysis
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
in calcined powder was performed using an infrared adsorption method with a simultaneous carbon and ultralow sulfur analyzer (Model CS-244, LECO, St. Joseph, MI). Densification behavior of the powder compacts was investigated in flowing O2 (20 mL/min) on a thermal mechanical analyzer (Model SETSYS Evolution 2400, Setaram, Lyons, France), using constant heating and cooling rates of 10 ◦ C/min and 40 ◦ C/min, respectively. Alternative vacuum sintering was performed for the pellets at temperatures between 1500 and 1700 ◦ C for 5 h, in a vacuum sintering furnace with molybdenum heating element (Model VSF-7, Shenyang, China). The density of a green compact was calculated from its weight and geometric dimensions, and densities of sintered bodies were measured by the Archimedes method, with distilled water as the immersion medium. Relative density was then obtained as a ratio of the actual density to the theoretical density of Y2 O3 . Samples (1.5 mm thick) that were polished on both surfaces were used to measure the optical transmittance over 200–800 nm wavelength region by using an ultraviolet/visible/near infrared spectrophotometer (Model Lambda 750S, Perkin Elmer, CT, USA). The microstructure of the sintered ceramics was observed using a field emission scanning electron microscope (FE-SEM, Model JSM-7001F, JEOL, Tokyo, Japan). The samples were cut into fractured slices, which were then examined after gold coating for better conductivity. 3. Results 3.1. Effects of sulfuric acid treatment on particle size Fig. 1 shows the effects of sulfuric acid treatment on the BET specific surface area (SBET ) for the powders calcined at various temperatures. The dopant concentration had significant influences on SBET for calcination temperatures ranged between 900 and 1100 ◦ C. Under these calcination conditions, SBET increased with increasing sulfate dosage up to a concentration of 7 mol%. Further increase the sulfate content to above 7 mol%, however,
3
Table 1 Particle diameters of different Y2 O3 powders. Sample
dBET (nm)
dTEM (nm)
n
Y11 Y11-4S Y11-7S Y11-10S Y11-14S Y12-7S Y10-7S Y09-7S
197.6 130.8 92.0 239.2 285.3 833.8 63.8 50.2
107.2 88.7 86.3 90.4 93.5 120.4 43.7 34.5
1.84 1.47 1.07 2.65 3.05 6.93 1.46 1.46
SBET decreased. The sulfate content only has minor influence on SBET when the powders were calcined at 1200 ◦ C. The particle diameter (dBET ) calculated using the BET data was summarized in Table 1 for selected samples. For comparison, particle size data determined from TEM images (dTEM ) was also listed. The samples in Table 1 are represented by both sulfate concentrations and calcination temperatures. Thus, Y11-7S indicates that the powder with 7 mol% sulfate dosage was calcined at 1100 ◦ C. For a calcination temperature of 1100 ◦ C, a minimum dBET = 92 nm was achieved for the 7 mol% sample. For dosages more than or less than 7 mol%, dBET increased significantly. The doped sulfate showed similar effects on dTEM , where the minimum dTEM = 86 nm was also achieved upon 7 mol% sulfate dosage. Clearly, an optimal sulfate content was required to obtain the finest particle size. Generally, the extent of difference between dBET and dTEM provides evidence revealing the particle agglomeration state. The difference between dBET and dTEM implied that some regions of the sample are not accessible to the gas as a consequence of close contact between particles [32]. For clarity, the ratio of dBET /dTEM (n) was calculated for the powders, as also listed in Table 1. The very similar value of dBET and dTEM (i.e., n ≈ 1) for the Y11-7S sample indicated a powder with well-dispersed particles. Dosage other than 7 mol%, on the other hand, may have induced aggregation of the particles, as evidenced by the increased n values. For a specific sulfate dosage (7 mol%), the effects of calcination temperature on particle size were also investigated, as revealed in Table 1. Powders calcined at temperatures of 900 and 1000 ◦ C showed moderate agglomeration and fine particle size around 40 nm (dTEM ). When the calcination temperature was raised to 1200 ◦ C, particle growth proceeds significantly, resulting in large particle size of ∼120 nm and severe particle aggregation (n = 6.93). 3.2. Effects of sulfuric acid treatment on particle morphology
Fig. 1. The BET specific surface areas of Y2 O3 powders treated with various concentrations of sulfuric acid.
The effects of sulfuric acid treatment on particle morphology were investigated, as shown in Fig. 2 for the TEM images for the powders calcined at 1100 ◦ C. The untreated Y2 O3 shows particles with relatively sharp edges and flat facets, revealing morphologies of regular-shaped grains (Fig. 2(a)). It is known that a pure crystal particle usually has flat surfaces and sharp edges, in contrast to the round surfaces of an amorphous
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 4
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Fig. 2. The TEM micrographs of Y2 O3 powders treated with various concentrations of sulfuric acid calcined at 1100 ◦ C: (a) untreated; (b) 4 mol%; (c) 7 mol%; (d) 10 mol%; (e) 14 mol%.
particle [7]. On the other hand, all the powders doped with sulfate are composed of round particles, as shown in Fig. 2(b)–(e), suggesting rather isotropic structure. Clearly, the doped sulfate ions can reduce crystal anisotropy among crystal planes of present Y2 O3 particles during calcination. This phenomenon has also been observed in previous powder synthesis research on calcinating Y2 O3 and YAG precursors doped with sulfate [7,17,19,28]. Moreover, Fig. 2(a) also revealed occurrence of sintering necks between two adjacent particles in untreated Y2 O3 , giving rise to relatively large-sized aggregates, in agreement with previous particle size measurement. It should be noted that the sintering necks are formed by diffusion mass transport and the resulted hard aggregates are difficult to break up in powder processing stage in ceramics processing. In contrast, although agglomeration of primary particles was also observed within the treated-powders, the sintering necks between particles were less frequently observed, suggesting a rather loosely agglomerated state. Besides, within the doped samples, the Y2 O3 powders with 10 mol% and 14 mol% dosage are seemingly composed of particles of two different size ranges, ∼100 nm for the larger particles and several tens nanometers for those of the smaller ones, as
shown in Fig. 2(d) and (e). This phenomenon could be attributed to different origins from which these particles evolved. Different decomposition behavior of precursors leads to a wider size distribution for the calcinate, as will be discussed later. 3.3. Evolution of the added sulfuric acid The doped sulfuric acid can react with yttria powder in the powder milling procedure, with the formation of Y2 (SO4 )3 . In subsequent calcination, Y2 (SO4 )3 decomposed into Y2 O3 via Y2 O2 SO4 , which is an intermediate phase frequently observed during the thermal decomposition of rare earth sulfate (Ln2 (SO4 )3 ) [33,34]. Detailed reactions are represented as follows: Y2 O3 + 3H2 SO4 → Y2 (SO4 )3 + 3H2 O
(2)
Y2 (SO4 )3 → Y2 O2 SO4 + 2SO3
(3)
Y2 O2 SO4 → Y2 O3 + SO3
(4)
Fig. 3 exhibits the TG–DSC result for the powder treated with 14 mol% sulfuric acid. Three steps can be distinguished
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
5
Fig. 3. TG/DSC curve for the powder treated with 14 mol% sulfuric acid.
20–300 ◦ C,
300–600 ◦ C
on the thermogravimetric curve in and 1000–1200 ◦ C temperature ranges. The weight loss observed up to 300 ◦ C is caused mainly by removal of adsorbed and hydrated water. In the temperature range of 300–600 ◦ C, the decomposition of the added dispersant and partial decomposition of sulfate occur. The final significant weight loss between 1000 and 1200 ◦ C is about 5% by mass and is associated with the final decomposition of sulfate, as also confirmed in subsequent quantitative sulfur measurement. Fig. 4 shows the FTIR spectra of powders calcined at varied temperatures. The broad-band peaking at ∼3500 cm−1 is associated with the effects of free hydroxyl groups. The wide absorption bands at 1534 and 1405 cm−1 were assigned to asymmetric stretch of C O band in CO3 2− . It is believed that the H2 O and CO3 2− groups observed in the FTIR data arise from H2 O and CO2 adsorption from the air during various stages of handling. The bands in the region ∼580 and ∼510 cm−1 were assigned to the bonding of Y O. It can be seen that the FTIR spectra of samples with and without sulfate ions calcined at 900 ◦ C are similar, except for the absorption peak around 1133 cm−1 , verifying
Fig. 4. Infrared spectra of Y2 O3 powders calcined at varied temperatures: (a) treated-Y2 O3 at 900 ◦ C; (b) untreated Y2 O3 at 900 ◦ C; (c)–(e) treated Y2 O3 at 1100 ◦ C, 1200 ◦ C and 1300 ◦ C, respectively.
Fig. 5. The sulfur content for the powder doped with 7 mol% sulfuric acid (corresponding to 0.95 wt% S), as a function of calcination temperature.
the existence of SO4 2− ions on the surface of the treated-Y2 O3 particles. Moreover, the peak at 1133 cm−1 becomes weaker for powder calcined at 1100 ◦ C, indicating that the amount of sulfate anion in Y2 O3 powder decreases as the increase of calcination temperatures. Disappearing of absorption band in 1133 cm−1 at 1200 ◦ C and 1300 ◦ C verified removal of sulfate ions, as shown with curves (d) and (e) in Fig. 4. The exact amount of sulfur remained in the calcined powders with 7 mol% sulfate dosage during calcination was further assessed by a simultaneous carbon and sulfur analyzer, as shown in Fig. 5. The samples showed slow loss of sulfur at temperature ≤1100 ◦ C. Significant decrease of sulfur content takes place at temperatures between 1100 and 1200 ◦ C, resulting in 0.05 wt% of sulfur for the powder calcined at 1200 ◦ C. Significant loss of sulfur between 1100 and 1200 ◦ C indicated that successive decomposition of Y2 (SO4 )3 mostly takes place at this temperature range, corresponding well with the FTIR data (Fig. 4). Above temperature range of desulfurization, however, was much higher than that deduced from TG–DSC data (1000–1200 ◦ C, as shown Fig. 3). This discrepancy can be ascribed to different heating conditions. It should be noted that the TG–DSC measurement was performed in a flowing O2 atmosphere, whereas the powders used for the elemental analysis were calcined in a stagnant air condition. Moreover, some residual sulfur can survive even after thermal decomposition. It is seen in Fig. 5 that 0.009 wt% sulfur remained in the powder calcined at 1300 ◦ C. Previously, 100 ppm sulfur residual has been found existed in the yttria powder derived from ammonium sulfate doped precursor calcined at 1200 ◦ C in flowing O2 atmosphere. For beryllium oxide obtained by thermal decomposition of beryllium sulfate, it was shown that some sulfate dissociates remained in the BeO powder even at >1000 ◦ C, although beryllium sulfate decomposes at ∼700–930 ◦ C [35]. However, the exact sulfur-containing species in the calcined powders at >1200 ◦ C could not be determined at this time due to limited data. It is worthy noting that the survival of SO4 2− ions in the YAG powder in 900–1300 ◦ C temperature
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 6
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Fig. 6. (a) Shrinkage and (b) shrinkage rate curves for the Y2 O3 powders calcined at 1100 ◦ C in the course of heating (10 ◦ C/min).
range has been verified using XPS data in a more recent study [6]. 3.4. Effects of sulfuric acid treatment on the sintering behaviors of the powders The effects of sulfuric acid treatment on the sintering behaviors were investigated for Y2 O3 powders calcined at 1100 ◦ C using a thermal mechanical analyzer (TMA), as shown in Fig. 6. Fig. 6(a) shows the change of the shrinkage in the course of 10 ◦ C/min heating. The onset temperature of shrinkage was ∼1150 ◦ C for the untreated Y2 O3 , and was slightly hindered for the 4 mol% sample. Further increasing sulfate concentration, the onset temperature was shifted to higher temperatures. Using the shrinkage curves in Fig. 6(a), the temperature changes of the densification rate were determined and shown in Fig. 6(b). All the samples showed a broad maximum peaked at temperatures between 1400 and 1500 ◦ C, representing the temperature of pronounced densification. It can be seen in Fig. 6(b) that the temperature of maximum peak shifted to lower temperatures with the doping of sulfate, indicating the increased sinterability for the treated-Y2 O3 powder. The 7 mol% sample revealed the lowest peaking temperature, ∼60 ◦ C lower than that of untreated Y2 O3 powder. The sintering behaviors for the powders were further assessed by performing sintering in vacuum. Fig. 7 shows the temperature dependent relative density for the powders untreated and treated with 7 mol% sulfuric acid. Significant densification was achieved for both samples at temperature <1500 ◦ C. The treated Y2 O3 showed a relative high density of 98.1% when sintered at 1500 ◦ C, much higher than that of 94.2% for the untreated Y2 O3 , indicating its excellent sinterability. This result is also in agreement with above sintering kinetic study. Sintered density of both samples increased with increasing sintering temperature, reaching >99% density at 1700 ◦ C. Fig. 8 compares the microstructures of fractured surfaces for the samples untreated and treated with 7 mol% sulfate sintered at 1550 ◦ C. Pores both in the grains and at the grain boundaries were observed in the sintered sample from untreated Y2 O3 , as shown in Fig. 8(a). The powder untreated presented some hard agglomeration (Fig. 2(a)) and the pores were difficult to remove during sintering and were entrapped in the sample. On the other hand, the ceramics
from treated-powder reveals a well-defined microstructure and pores were less frequently observed (Fig. 8(b)). The homogenous microstructure of the sintered sample can be attributed to the well-dispersed powders facilitated by the sulfuric acid-based pre-treatment. Fig. 9 shows the images for the ceramics vacuum-sintered at 1700 ◦ C. The pellets are of 1.5 mm thickness. The ceramics fabricated from treated powders are transparent, and the word under the samples can be clearly read. All the ceramics without sulfate dosage are opaque, irrespective of the calcination temperature of the powders. The in-line transmittance (at the wavelength of 800 nm) of the ceramics shown in Fig. 9 is compared in Fig. 10. All the samples untreated have an in-line transmittance <5%, whereas it was significantly higher for those of the treated samples. It is seen in Fig. 10 that both the calcination temperature and amount of sulfate dosage influence the transmittance; the highest transmittance was achieved for the sample treated with 7 mol% sulfuric acid and calcined at 1100 ◦ C. It is also should be noted that the in-line transmittance data achieved for the sample with the optimal sulfate dosage is still less than the theoretic value of single crystal Y2 O3 (∼81.9% at 800 nm [36]). Elaboration of powder processing process or optimization of subsequent consolidation and sintering procedures may required to further increase the optical transmittance of present Y2 O3 ceramics.
Fig. 7. Densification behaviors for the samples treated and untreated with 7 mol% sulfate sintered at various temperatures in vacuum.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
7
Fig. 8. Microstructure of the fractured surface of Y2 O3 ceramics (a) untreated and (b) treated with 7 mol% sulfate sintered at 1550 ◦ C for 5 h in vacuum.
4. Discussions 4.1. Effects of the added sulfuric acid on powder sintering ability
Fig. 9. Images for Y2 O3 ceramics vacuum-sintered at 1700 ◦ C for 5 h. All the samples were polished and are of 1.5 mm thickness. The samples are denoted by both sulfate concentrations and calcination temperatures. Y11-7S indicates that the powder with 7 mol% sulfate dosage was calcined at 1100 ◦ C.
The sintering behavior of present Y2 O3 powder could be greatly improved by pre-treating starting materials with an optimum amount of sulfuric acid, as shown in Fig. 7, where the treated-powder sintered at 1500 ◦ C has a relative density of 98.1%, being significantly higher than that of the sample untreated (∼94.2%). When sintered at 1700 ◦ C, it is seen that all the samples prepared using treated-powders were transparent, whereas it was totally opaque for the untreated sample, further confirmed the excellent sintering ability for the treated-yttria. The enhanced sintering ability for the treated-yttria can be correlated to several particle features facilitated by the sulfuric acid-based treatment. We can see that the powder untreated revealed particle morphologies of flat surfaces and sharp edges, whereas the powder with sulfate dosage are composed of round particles of isotropic surfaces, as shown in Fig. 2(b)–(e). It has been reported that isotropy in particle morphologies exceed anisotropic microstructural features, with respect to homogeneous packing of particles in consolidation and elimination of pores in the final stage of sintering [37]. Moreover, treating with sulfuric acid induced particles with finer particle size and lesser extent of aggregation, as revealed in Fig. 3 and Table 1. Sintering kinetics theories indicates that the shrinkage rate of a powder compact increases appreciably as the particle size decreases [38]. Besides, for powders showing severe agglomeration of the primary particles, increased relative content of agglomerates was found to hinder pore elimination in the sintering stage. 4.2. The mechanisms of the added sulfuric acid on powder properties
Fig. 10. The inline transmittance of 1.5-mm thick yttria ceramics at wavelength of 800 nm.
The underlying reasons for the observed morphological differences between the treated and untreated Y2 O3 powders can be elucidated using CRH sintering data between 1100 and 1300 ◦ C. It is seen in Fig. 6(a) that the temperature at which
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 8
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
densification begins was shifted to higher temperatures for the treated powder. The onset temperature of shrinkage was ∼1150 ◦ C for the starting Y2 O3 powder, whereas it was 1250 ◦ C for the powder treated with 7 mol% SO4 2− . However, Table 1 reveals that all the calcined powders had similar particle size at 1200 ◦ C, which is also the case for the green compacts of these powders in CRH sintering. Clearly, although similar trend of particle coarsening occurred in the powder compacts, densification was greatly suppressed in 1100–1250 ◦ C temperature range for the treated yttria, suggesting different mass transportation mechanisms between the treated and untreated powders. According to the sintering theory, five mass transport processes, namely evaporation–condensation, surface diffusion, volume diffusion, grain boundary diffusion and viscous or creep flow, contribute to particle coarsening and sintering of a powder compact. The processes of volume diffusion, grain boundary diffusion, and viscous or creep flow induce neck growth or particle growth, accompanied by shrinkage among particles. Evaporation–condensation and surface diffusion from the surface to the neck area, on the other hand, can only lead to particle growth, because these mechanisms do not allow the particle centers to move closer together. Based on the above analysis, it is then may be concluded that the doped sulfate contributes to mechanisms leading to particle coarsening at temperature <1250 ◦ C, whereas the mechanisms contributing to densification dominate in the untreated Y2 O3 . Thus the observed morphological differences between the treated and untreated Y2 O3 can be partially explained as follows. For the treated powder, the doped sulfate inhibits mass transfer processes related to densification, hence particles tend to coarsen rather than densify, producing a loosely agglomerated and sinterable powder. On the other hand, mass transport processes leading to densification dominate in the untreated yttria powder. As a result, particles coarsening proceeds accompanied by simultaneous sintering, producing hard agglomerates during the calcination stage. A powder compact containing such agglomerates is nonsinterable. At this point, however, it is still unclear which mechanism leads to particle coarsening in the sulfate-doped powders. In view of low vapor pressure of yttria, Ikegami et al. [39] suggested that surface diffusion is the most likely process for particle growth in sulfate-ion-doped yttria up to 950 ◦ C. The existence of SO4 2− -ions has been verified for the treated-yttria at temperature <1200 ◦ C. Considering low vapor pressure of sulfate, contributions from sulfate-contributed mass transportation processes cannot be fully eliminated. Detailed sintering study of green compacts for powders calcined at different temperatures has been performed to further elucidate the mechanism of doping sulfate, and will be reported in our subsequent paper. 4.3. Determination of the optimum sulfate dosage The direct application of a dopant in pure Y2 O3 powders allow us precisely establishing a correlation between powder property and the amount of doped sulfate. Both sintering data and powder characterization indicate an optimum dosage of 7 mol%. In view of strong electrostatic attraction between positive yttria particle surfaces and negative SO4 2− -ions [30], we postulated
that the optimum sulfate dosage may be interconnected with the full coverage of SO4 2− -ions on the yttria particle surfaces. The ionic radius of SO4 2− is 0.235 nm and the present yttria powder has a BET surface area of 31.5 m2 /g. Supposing a single layer absorption, the calculated sulfate required for the full coverage of yttria particle surfaces was 6.98 mol%. The good match of the calculated data and experimental result further confirmed our assumption. Thus, the different characteristics for the powders treated with varied amount of sulfate can be explained as follows. For the powder doped with <7 mol% of SO4 2− -ions, insufficient sulfate coverage results in mixed particles, with a great portion of which possessing features of untreated yttria (e.g. larger particle size, more severe aggregation). For the dosage >7 mol%, however, the un-adsorbed free sulfate may accumulate, forming single particles of sulfate. Upon calcination, the free sulfate decomposed into oxide simultaneously. As a result, the calcined products were composed of particles evolved from two distinct origins, leading to a wider size distribution in the calcinates. This assumption was confirmed in Fig. 2(d) and (e), where particles of two different features were clearly revealed. 5. Conclusions The effects of doped SO4 2− ions on structural evolution and sintering behaviors of a commercial Y2 O3 powder were investigated. The sulfate dosage was achieved by pre-treating the starting material with sulfuric acid. The surface morphology and size of the particles were notably affected by the applied sulfuric acid-based treatment. The powder untreated revealed particle of flat surfaces and sharp edges, whereas the powder with sulfate dosage are composed of round particles of isotropic surfaces. Besides, treating with sulfuric acid resulted in particles with finer size and lesser aggregates. The doped sulfate inhibits mass transfer processes related to densification, producing a loosely agglomerated and sinterable powder upon calcination. Mass transport processes leading to densification dominate in the untreated yttria powder, where particles coarsening proceeded accompanied by simultaneous sintering, producing hard agglomerates in the calcination stage. The sintering behavior of present Y2 O3 powder could be greatly improved by pre-treating starting materials with an optimum amount of sulfuric acid. Dispersive Y2 O3 nanopwder with excellent sintering property was obtained from the powders treated with 7 mol% sulfuric acid and calcined at 1100 ◦ C for 4 h. The optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces. Insufficient sulfate coverage results in powders containing a great portion of particles featuring those of untreated yttria (e.g. larger particle size, more severe aggregation). All the ceramics prepared using treated-powders were transparent after vacuumsintering at 1700 ◦ C, whereas it was totally opaque for those prepared with the untreated powder. Acknowledgements The work was supported in part by the Program for New Century Excellent Talents in University (NCET-11-0076), National
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Natural Science Foundation of China (51172037, 51102040 and 50990303), and the Fundamental Research Funds for the Central Universities (N130810003, N130610001, and N110802001). References [1]. Kong J, Lu J, Takaichi K, Uematsu T, Ueda K, Tang DY, et al. Diode-pumped Yb:Y2 O3 ceramic laser. Appl Phys Lett 2003;82:2556–8. [2]. Eilers H. Fabrication, optical transmittance, and hardness of IR-transparent ceramics made from nanophase yttria. J Eur Ceram Soc 2007;27:4711–7. [3]. Zhang HB, Kim B-N, Morita K, Yoshida H, Hiraga K, Sakka Y. Fabrication of transparent yttria by high-pressure spark plasma sintering. J Am Ceram Soc 2011;94:3206–10. [4]. Mouzon J, Maitre A, Frisk L, Lehto N, Oden M. Fabrication of transparent yttria by HIP and the glass-encapsulation method. J Eur Ceram Soc 2009;29:311–6. [5]. Li J-G, Ikegami T, Mori T, Yajima Y. Sc2 O3 nanopowders via hydroxyl precipitation: effects of sulfate ions on powder properties. J Am Ceram Soc 2004;87:1008–13. [6]. Deineka TG, Doroshenko AG, Mateychenko PV, Tolmachev AV, Vovk EA, Vovk OM, et al. Influence of sulfate ions on properties of co-precipitated Y3 Al5 O12 :Nd3+ nanopowders. J Alloys Compd 2010;508:200–5. [7]. Ikegami T, Li J-G, Mori T, Moriyoshi Y. Fabrication of transparent yttria ceramics by the low-temperature synthesis of yttrium hydroxide. J Am Ceram Soc 2002;85:1725–9. [8]. Qin HM, Liu H, Sang YH, Lv YH, Zhang XL, Zhang YY, et al. Ammonium sulfate regulation of morphology of Nd:Y2 O3 precursor via urea precipitation method and its effect on the sintering properties of Nd:Y2 O3 nanopowders. CrystEngComm 2012;14:1783–9. [9]. Qin HM, Zhang XH, Liu H, Sang YH, Wang JY. Mechanism of ammonium sulfate regulation effect on microstructure of Y2 O3 nanopowders via urea precipitation method. CrystEngComm 2013;15:5076–81. [10].Li N, Yanagisawa K. Controlling the morphology of yttrium oxide through different precursors synthesized by hydrothermal method. J Solid State Chem 2008;181:1738–43. [11].Hirai T, Orikoshi T, Komasawa I. Preparation of Y2 O3 :Yb,Er infraredto-visible conversion phosphor fine particles using an emulsion liquid membrane system. Chem Mater 2002;14:3576–83. [12].Rulison AJ, Flagan RC. Synthesis of yttria powders by electrospray pyrolysis. J Am Ceram Soc 1994;77:3244–50. [13].Subramanian R, Shankar P, Kavithaa S, Ramakrishnan SS, Angelo PC, Venkataraman H. Synthesis of nanocrystalline yttria by sol–gel method. Mater Lett 2001;48:342–6. [14].Dupont A, Parent C, Le Garrec B, Heintz JM. Size and morphology control of Y2 O3 nanopowders via a sol–gel route. J Solid State Chem 2003;171:152–60. [15].Mangalaraja RV, Mouzon J, Hedstrom P, Camurri CP, Ananthakumar S, Oden M. Microwave assisted combustion synthesis of nanocrystalline yttria and its powder characteristics. Powder Technol 2009;191:309–14. [16].Saito N, Matsuda S, Ikegami T. Fabrication of transparent yttria ceramics at low temperature using carbonate-derived powder. J Am Ceram Soc 1998;81:2023–8. [17].Ikegami T, Mori T, Yajima Y, Takenouchi S, Misawa T, Moriyoshi Y. Fabrication of transparent yttria ceramics through the synthesis of yttrium hydroxide at low temperature and doping by sulfate ions. J Ceram Soc Jpn 1999;107:297–9. [18].Onodera T, Ikegami T, Yajima Y, Kawamura M, Sakai M, Moriyoshi Y. Effects of the sulfate ion exerted on densification of yttria on precursor synthetic. J Ceram Soc Jpn 2003;111:664–8.
9
[19].Wen L, Sun XD, Xiu ZM, Chen SW, Tsai C-T. Synthesis of nanocrystalline yttria powder and fabrication of transparent YAG ceramics. J Eur Ceram Soc 2004;24:2681–8. [20].Wen L, Sun XD, Lu Q, Xu GX, Hu XZ. Synthesis of yttria nanopowders for transparent yttria ceramics. Opt Mater 2006;29:239–45. [21].Li J, Liu WB, Jiang BX, Zhou J, Zhang WX, Wang L, et al. Synthesis of nanocrystalline yttria powder and fabrication of Cr,Nd:YAG transparent ceramics. J Alloys Compd 2012;515:49–56. [22].Sacks MD, Tseng T-Y, Lee SY. Thermal decomposition of spherical hydrated basic aluminium sulfate. Ceram Bull 1984;63:301–10. [23].Kara F, Sahin G. Hydrated aluminium sulfate precipitation by enzymecatalysed urea decomposition. J Eur Ceram Soc 2000;20:689–94. [24].Li J-G, Ikegami T, Mori T. Solution-based processing of Sc2 O3 nanopowders yielding transparent ceramics. J Mater Res 2004;19:733–6. [25].Li J-G, Ikegami T, Mori T, Yajima Y. Monodispersed Sc2 O3 precursor particles via homogeneous precipitation: synthesis, thermal decomposition, and the effects of supporting anions on powder properties. J Mater Res 2003;18:1149–56. [26].Matsushita N, Tsuchiya N, Nakatsuka K, Yanagitani T. Precipitation and calcination process for yttrium aluminum garnet precursors synthesized by the urea method. J Am Ceram Soc 1999;82:1977–84. [27].Lv YH, Zhang W, Liu H, Sang YH, Qin HM, Tan J, et al. Synthesis of nanosized and highly sinterable Nd:YAG powders by the urea homogeneous precipitation method. Powder Technol 2012;217:140–7. [28].Xu XJ, Sun XD, Liu H, Li J-G, Li XD, Huo D, et al. Synthesis of monodispersed spherical yttrium aluminum garnet (YAG) powders by a homogeneous precipitation method. J Am Ceram Soc 2012;95:3821– 6. [29].Li XX, Wang WJ. Preparation of uniformly dispersed YAG ultrafine powders by co-precipitation method with SDS treatment. Powder Technol 2009;196:26–9. [30].Li J, Li JP, Chen Q, Wu WJ, Xiao DQ, Zhu JG. Effect of ammonium sulfate on the monodispersed Y3 Al5 O12 nanopowders synthesized by co-precipitant Method. Powder Technol 2012;218:46–50. [31].Wang JQ, Zheng SH, Zeng R, Dou SX, Sun XD. Microwave synthesis of homogeneous YAG nanopowder leading to a transparent ceramic. J Am Ceram Soc 2009;92:1217–23. [32].Caponetti E, Martino DC, Saladino ML, Leonelli C. Preparation of Nd:YAG nanopowder in a confined environment. Langmuir 2007;23:3947– 52. [33].Skrobian M, Sato N, Yamada K, Fujino T. Thermogravimetric study of reduction and sulfurization of Y2 (SO4 )3 using carbon disulfide. Thermochim Acta 1995;255:201–9. [34].Lian JB, Liu F, Wang XJ, Sun XD. Hydrothermal synthesis and photoluminescence properties of Gd2 O2 SO4 :Eu3+ spherical phosphor. Powder Technol 2014;253:187–92. [35].Ikegami T, Mori Y, Matsuda M, Suzuki H. Surface structure of BeO powders obtained by calcination of sulfate. Yogyo Kyokaishi 1973;81:379– 82. [36].Nigara Y. Measurement of the optical constants of yttrium oxide. Jpn J Appl Phys 1968;7:404–8. [37].Ikegami T, Eguchi K. Two kinds of roles of MgO in the densification and grain growth of alumina under various atmospheres: sensitive and insensitive roles to the experimental procedures. J Mater Res 1999;14:509–17. [38].Coble RL. Initial sintering of alumina and hematite. J Am Ceram Soc 1958;41:55–62. [39].Takayasu I, Li J-G, Sakaguchi I, Hirota K. Morphology change of undoped and sulfate-ion-doped yttria powder during firing. J Am Ceram Soc 2004;87:517–9.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
ARTICLE IN PRESS
JECS-10003; No. of Pages 9
Available online at www.sciencedirect.com
ScienceDirect Journal of the European Ceramic Society xxx (2015) xxx–xxx
Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics Jiao He a , Xiaodong Li a,∗ , Shaohong Liu a , Qi Zhu a , Ji-Guang Li a,b , Xudong Sun a a
Key Laboratory for Anisotropy and Texture of Materials (Ministry of Education), School of Materials and Metallurgy, Northeastern University, Shenyang 110004, China b Nano Ceramics Center, National Institute for Materials Science, Tsukuba, Ibaraki 305-0044, Japan Received 1 December 2014; received in revised form 6 February 2015; accepted 7 February 2015
Abstract The effects of pre-treating a nano-sized Y2 O3 powder with sulfuric acid were studied. Both the structural evolution and sintering behavior as a function of SO4 2− -ion concentrations were investigated. The surface morphologies and aggregation state of particles were notably affected by the sulfuric acid-based treatment. The powder untreated revealed particle of flat surfaces and sharp edges, whereas the powders with sulfate dosage were composed of round particles of isotropic surfaces after calcination. Dispersive Y2 O3 nanopowder with excellent sintering property was obtained by calcinating the powders treated with 7 mol% sulfuric acid. The optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces, protecting particles from forming hard agglomerates by contributing to mass transportation mechanisms leading to particle coarsening upon calcination. All the ceramics prepared using treated-powders were transparent after vacuum sintering at 1700 ◦ C, whereas it was opaque for the sample without sulfate dosage. © 2015 Elsevier Ltd. All rights reserved. Keywords: Yttria; Transparent ceramics; SO4 2− -ion; Sintering; Powder
1. Introduction Recently, yttria transparent ceramic material has been extensively studied, because of its promise for a number of optical applications, such as luminous pipes for high-intensitydischarge lamps, heat-resistive windows, missile domes, host materials for scintillators and solid lasers, etc. [1–3]. In order to achieve a high degree transparency in yttria ceramics, internal porosity of samples should be completely removed [4]. One of the key technologies for fabricating highly dense ceramics is the synthesis of ceramic powders with excellent sintering activity, generally possessing such powder characteristics as low agglomeration state, ultrafine particle size (∼100 nm), narrow particle size distribution, and spherical particle shape [5,6]. Thus, the past two decades have witnessed tremendous efforts endeavored in the synthesis of high quality yttria nano powders through
∗
Corresponding author. Tel.: +86 24 83691580. E-mail address: [email protected] (X. Li).
different processing methods, such as precipitation methods [7–9], hydrothermal synthesis [10], emulsion synthesis [11], spray-drying [12], sol–gel synthesis [13,14], and combustion synthesis [15]. Among which, precipitation synthesis routes are attractive because of their accessibility. Most developed precipitation methods for synthesizing sinterable yttria powders always involve the addition of SO4 2− -ions [7,8,16–21]. It was found that the morphology, microstructure and sinterability of the calcined powders are highly affected by the [SO4 2− ]/[Y3+ ] molar ratios. Similar results have also been shown when the precipitation methods were used for the synthesis of Al2 O3 [22,23], Sc2 O3 [5,24,25] and YAG [6,26–31] powders by doping SO4 2− -ions. The doping by SO4 2− -ions can be realized in several different ways when ammonium sulfate is used as SO4 2− -ion sources. The ammonium sulfate can be added into reaction mixture during precipitation [19,27,30] during aging [7,17], or the resulting precipitant can be rinsed by ammonium sulfate solution [16]. Alternatively, SO4 2− -containing substances can be used as a starting material in the reaction [5,25,28,29].
http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008 0955-2219/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 2
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Previously, an amount of investigations have been performed to reveal the mechanisms of doped SO4 2− ions on precursor and calcined powders [5–9,25,26]. Ammonium sulfate was proved to be a regulator, mediating the nucleation and growth of precursors [8,9]. Among common inorganic anions, SO4 2− has a higher coordination capability than NO3 − and Cl− and shows stronger complexing effects [25]. The addition of the ammonium sulfate directly into reaction mixture gives rise to a local change of solution chemistry and consequently variations of chemical composition of resulted precursor powders [26]. It has been indicated that even a trace amount of SO4 2− decreased the monodispersity of the carbonate particles obtained via homogeneous precipitation with urea [8]. When ammonium sulfate was added into reaction mixture during aging, or the resulting precipitant was rinsed by ammonium sulfate solution, on the other hand, the morphological features of precursor particles were less frequently affected [7]. With respect to calcined powders, although it was generally accepted that doping a certain amount of SO4 2− -ions promotes formation of monodispersed spherical powders upon calcination, the influence mechanisms were not clearly verified. The reasons may be two-folds. On one hand, the high-temperature calcination process results in both the dissociation of precursor and the desulfurization of sulfate; the consecutive occurrence of two thermal decomposition procedures complicated the analysis aimed at understanding the mechanism of doped sulfate on calcined powders [24,26]. On the other hand, previous synthetic methods involving sulfate addition are not amenable to precise control of the amount of the doped sulfate in the precursors; it should be noted that the precursors produced by precipitation method are repeatedly washed to remove impurities such as NH4 + and NO3 − , resulting in a lot of sulfate added being washed away. As a consequence, the optimum sulfate dosage was significantly varied in different reaction systems [30,31]. And a correlation between powder property and the exact amount of doped sulfate was not always established. In this work, we investigate the effects of doped SO4 2− ions on a commercial Y2 O3 powder by pre-treating the starting material with sulfuric acid. The purpose of the work is to understand the underlying mechanism for the observed morphological changes and sinterability improvement achieved upon sulfate dosage. Both the structural evolution and sintering behaviors as a function of the amount of doped SO4 2− -ions were investigated. The direct application of SO4 2− to pure yttria, in deed, allows us to discriminate and analyze particle features contributed solely from the doped sulfate. In particular, we show that the optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces, protecting particles from forming hard aggregates by contributing mass transportation mechanisms leading to particle coarsening while suppressing the mechanisms giving rise to densification upon calcination. This investigation may provide an insight into more efficient application of anion dopant for manipulating particle features and improving the sinterability of available ceramic powders by a simple sulfate-based pre-treatment.
2. Experimental procedure 2.1. Preparation process Commercial Y2 O3 (99.99% pure, Huizhou Ruier Rare Chemical Hi-Tech Co. Ltd., Huizhou, China) powder was used as the starting material. The powder has a BET surface area of 31.5 m2 /g. X-ray diffractometric (XRD) analysis indicated that the Y2 O3 powder was a cubic-structured polymorph and the crystallite size calculated from Scherrer’s equation is 19 nm. In a typical processing, 60 g starting Y2 O3 powder was dispersed in ethyl alcohol with the addition of sulfuric acid and 2 wt% of dispersant (ammonium salt of poly(methacrylic acid)), and then the suspensions were ball milled in polyurethane jars filled with ZrO2 balls for 8 h. The amounts of added sulfuric acid were expressed as a molar percentage of the Y2 O3 powder basis (mol%). For comparison purpose, the starting Y2 O3 powder without sulfuric acid addition was also ball-milled identically. The resultant suspensions were then dried at 110 ◦ C, followed by sieving through a 200-mesh nylon sieve for pulverization. After sieving, the powder was calcined at 900–1200 ◦ C for 4 h in furnace in a stagnant air condition. The green compacts were prepared by dry pressing at 100 MPa, followed by hydrostatical isostatic pressing under a pressure of 200 MPa. 2.2. Characterization Powder morphologies were observed by transmission electron microscopy (TEM, Model JEM-2100F, JEOL, Tokyo, Japan). The average particle size (dTEM ) was determined from over 200 randomly selected particles with an image analysis software (WinRoof, Mitani Corp., Tokyo, Japan), by assuming a circular shape. The specific surface area of the calcined powders was measured by the BET method (Model TriStar II 3020, Micromeritics Instrument Corp., Norcross, GA) via nitrogen adsorption at 77 K. The equivalent particle size (dBET ) was estimated from BET surface area based on the following equation: dBET =
6 × 103 ρS
(1)
where ρ is the theoretical density (5.01 g cm−3 ) of Y2 O3 crystal; S is the specific surface area (m2 /g) determined by BET measurement. Thermal analysis of the powders doped with sulfate was performed using a TG/DSC analyzer (Model SETSYS Evolution-16, Setaram, Lyons, France) in flowing oxygen atmosphere (40 mL/min). The heating rate was 10 ◦ C/min. Fourier transform infrared spectroscopy (FT-IR) (Model Spectrum RXI, Perkin-Elmer, CT, USA) spectra of the yttria powders were recorded by the standard KBr method. The calcined Y2 O3 powder of 0.003 g was grinded and dispersed homogeneously in 0.25 g dried KBr powder using the agate mortar and pestle, and then uniaxially compacted into transparent pellets at 28 MPa. The resultant pellets were dried in the oven before investigation. The quantitative sulfur analysis
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
in calcined powder was performed using an infrared adsorption method with a simultaneous carbon and ultralow sulfur analyzer (Model CS-244, LECO, St. Joseph, MI). Densification behavior of the powder compacts was investigated in flowing O2 (20 mL/min) on a thermal mechanical analyzer (Model SETSYS Evolution 2400, Setaram, Lyons, France), using constant heating and cooling rates of 10 ◦ C/min and 40 ◦ C/min, respectively. Alternative vacuum sintering was performed for the pellets at temperatures between 1500 and 1700 ◦ C for 5 h, in a vacuum sintering furnace with molybdenum heating element (Model VSF-7, Shenyang, China). The density of a green compact was calculated from its weight and geometric dimensions, and densities of sintered bodies were measured by the Archimedes method, with distilled water as the immersion medium. Relative density was then obtained as a ratio of the actual density to the theoretical density of Y2 O3 . Samples (1.5 mm thick) that were polished on both surfaces were used to measure the optical transmittance over 200–800 nm wavelength region by using an ultraviolet/visible/near infrared spectrophotometer (Model Lambda 750S, Perkin Elmer, CT, USA). The microstructure of the sintered ceramics was observed using a field emission scanning electron microscope (FE-SEM, Model JSM-7001F, JEOL, Tokyo, Japan). The samples were cut into fractured slices, which were then examined after gold coating for better conductivity. 3. Results 3.1. Effects of sulfuric acid treatment on particle size Fig. 1 shows the effects of sulfuric acid treatment on the BET specific surface area (SBET ) for the powders calcined at various temperatures. The dopant concentration had significant influences on SBET for calcination temperatures ranged between 900 and 1100 ◦ C. Under these calcination conditions, SBET increased with increasing sulfate dosage up to a concentration of 7 mol%. Further increase the sulfate content to above 7 mol%, however,
3
Table 1 Particle diameters of different Y2 O3 powders. Sample
dBET (nm)
dTEM (nm)
n
Y11 Y11-4S Y11-7S Y11-10S Y11-14S Y12-7S Y10-7S Y09-7S
197.6 130.8 92.0 239.2 285.3 833.8 63.8 50.2
107.2 88.7 86.3 90.4 93.5 120.4 43.7 34.5
1.84 1.47 1.07 2.65 3.05 6.93 1.46 1.46
SBET decreased. The sulfate content only has minor influence on SBET when the powders were calcined at 1200 ◦ C. The particle diameter (dBET ) calculated using the BET data was summarized in Table 1 for selected samples. For comparison, particle size data determined from TEM images (dTEM ) was also listed. The samples in Table 1 are represented by both sulfate concentrations and calcination temperatures. Thus, Y11-7S indicates that the powder with 7 mol% sulfate dosage was calcined at 1100 ◦ C. For a calcination temperature of 1100 ◦ C, a minimum dBET = 92 nm was achieved for the 7 mol% sample. For dosages more than or less than 7 mol%, dBET increased significantly. The doped sulfate showed similar effects on dTEM , where the minimum dTEM = 86 nm was also achieved upon 7 mol% sulfate dosage. Clearly, an optimal sulfate content was required to obtain the finest particle size. Generally, the extent of difference between dBET and dTEM provides evidence revealing the particle agglomeration state. The difference between dBET and dTEM implied that some regions of the sample are not accessible to the gas as a consequence of close contact between particles [32]. For clarity, the ratio of dBET /dTEM (n) was calculated for the powders, as also listed in Table 1. The very similar value of dBET and dTEM (i.e., n ≈ 1) for the Y11-7S sample indicated a powder with well-dispersed particles. Dosage other than 7 mol%, on the other hand, may have induced aggregation of the particles, as evidenced by the increased n values. For a specific sulfate dosage (7 mol%), the effects of calcination temperature on particle size were also investigated, as revealed in Table 1. Powders calcined at temperatures of 900 and 1000 ◦ C showed moderate agglomeration and fine particle size around 40 nm (dTEM ). When the calcination temperature was raised to 1200 ◦ C, particle growth proceeds significantly, resulting in large particle size of ∼120 nm and severe particle aggregation (n = 6.93). 3.2. Effects of sulfuric acid treatment on particle morphology
Fig. 1. The BET specific surface areas of Y2 O3 powders treated with various concentrations of sulfuric acid.
The effects of sulfuric acid treatment on particle morphology were investigated, as shown in Fig. 2 for the TEM images for the powders calcined at 1100 ◦ C. The untreated Y2 O3 shows particles with relatively sharp edges and flat facets, revealing morphologies of regular-shaped grains (Fig. 2(a)). It is known that a pure crystal particle usually has flat surfaces and sharp edges, in contrast to the round surfaces of an amorphous
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 4
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Fig. 2. The TEM micrographs of Y2 O3 powders treated with various concentrations of sulfuric acid calcined at 1100 ◦ C: (a) untreated; (b) 4 mol%; (c) 7 mol%; (d) 10 mol%; (e) 14 mol%.
particle [7]. On the other hand, all the powders doped with sulfate are composed of round particles, as shown in Fig. 2(b)–(e), suggesting rather isotropic structure. Clearly, the doped sulfate ions can reduce crystal anisotropy among crystal planes of present Y2 O3 particles during calcination. This phenomenon has also been observed in previous powder synthesis research on calcinating Y2 O3 and YAG precursors doped with sulfate [7,17,19,28]. Moreover, Fig. 2(a) also revealed occurrence of sintering necks between two adjacent particles in untreated Y2 O3 , giving rise to relatively large-sized aggregates, in agreement with previous particle size measurement. It should be noted that the sintering necks are formed by diffusion mass transport and the resulted hard aggregates are difficult to break up in powder processing stage in ceramics processing. In contrast, although agglomeration of primary particles was also observed within the treated-powders, the sintering necks between particles were less frequently observed, suggesting a rather loosely agglomerated state. Besides, within the doped samples, the Y2 O3 powders with 10 mol% and 14 mol% dosage are seemingly composed of particles of two different size ranges, ∼100 nm for the larger particles and several tens nanometers for those of the smaller ones, as
shown in Fig. 2(d) and (e). This phenomenon could be attributed to different origins from which these particles evolved. Different decomposition behavior of precursors leads to a wider size distribution for the calcinate, as will be discussed later. 3.3. Evolution of the added sulfuric acid The doped sulfuric acid can react with yttria powder in the powder milling procedure, with the formation of Y2 (SO4 )3 . In subsequent calcination, Y2 (SO4 )3 decomposed into Y2 O3 via Y2 O2 SO4 , which is an intermediate phase frequently observed during the thermal decomposition of rare earth sulfate (Ln2 (SO4 )3 ) [33,34]. Detailed reactions are represented as follows: Y2 O3 + 3H2 SO4 → Y2 (SO4 )3 + 3H2 O
(2)
Y2 (SO4 )3 → Y2 O2 SO4 + 2SO3
(3)
Y2 O2 SO4 → Y2 O3 + SO3
(4)
Fig. 3 exhibits the TG–DSC result for the powder treated with 14 mol% sulfuric acid. Three steps can be distinguished
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
5
Fig. 3. TG/DSC curve for the powder treated with 14 mol% sulfuric acid.
20–300 ◦ C,
300–600 ◦ C
on the thermogravimetric curve in and 1000–1200 ◦ C temperature ranges. The weight loss observed up to 300 ◦ C is caused mainly by removal of adsorbed and hydrated water. In the temperature range of 300–600 ◦ C, the decomposition of the added dispersant and partial decomposition of sulfate occur. The final significant weight loss between 1000 and 1200 ◦ C is about 5% by mass and is associated with the final decomposition of sulfate, as also confirmed in subsequent quantitative sulfur measurement. Fig. 4 shows the FTIR spectra of powders calcined at varied temperatures. The broad-band peaking at ∼3500 cm−1 is associated with the effects of free hydroxyl groups. The wide absorption bands at 1534 and 1405 cm−1 were assigned to asymmetric stretch of C O band in CO3 2− . It is believed that the H2 O and CO3 2− groups observed in the FTIR data arise from H2 O and CO2 adsorption from the air during various stages of handling. The bands in the region ∼580 and ∼510 cm−1 were assigned to the bonding of Y O. It can be seen that the FTIR spectra of samples with and without sulfate ions calcined at 900 ◦ C are similar, except for the absorption peak around 1133 cm−1 , verifying
Fig. 4. Infrared spectra of Y2 O3 powders calcined at varied temperatures: (a) treated-Y2 O3 at 900 ◦ C; (b) untreated Y2 O3 at 900 ◦ C; (c)–(e) treated Y2 O3 at 1100 ◦ C, 1200 ◦ C and 1300 ◦ C, respectively.
Fig. 5. The sulfur content for the powder doped with 7 mol% sulfuric acid (corresponding to 0.95 wt% S), as a function of calcination temperature.
the existence of SO4 2− ions on the surface of the treated-Y2 O3 particles. Moreover, the peak at 1133 cm−1 becomes weaker for powder calcined at 1100 ◦ C, indicating that the amount of sulfate anion in Y2 O3 powder decreases as the increase of calcination temperatures. Disappearing of absorption band in 1133 cm−1 at 1200 ◦ C and 1300 ◦ C verified removal of sulfate ions, as shown with curves (d) and (e) in Fig. 4. The exact amount of sulfur remained in the calcined powders with 7 mol% sulfate dosage during calcination was further assessed by a simultaneous carbon and sulfur analyzer, as shown in Fig. 5. The samples showed slow loss of sulfur at temperature ≤1100 ◦ C. Significant decrease of sulfur content takes place at temperatures between 1100 and 1200 ◦ C, resulting in 0.05 wt% of sulfur for the powder calcined at 1200 ◦ C. Significant loss of sulfur between 1100 and 1200 ◦ C indicated that successive decomposition of Y2 (SO4 )3 mostly takes place at this temperature range, corresponding well with the FTIR data (Fig. 4). Above temperature range of desulfurization, however, was much higher than that deduced from TG–DSC data (1000–1200 ◦ C, as shown Fig. 3). This discrepancy can be ascribed to different heating conditions. It should be noted that the TG–DSC measurement was performed in a flowing O2 atmosphere, whereas the powders used for the elemental analysis were calcined in a stagnant air condition. Moreover, some residual sulfur can survive even after thermal decomposition. It is seen in Fig. 5 that 0.009 wt% sulfur remained in the powder calcined at 1300 ◦ C. Previously, 100 ppm sulfur residual has been found existed in the yttria powder derived from ammonium sulfate doped precursor calcined at 1200 ◦ C in flowing O2 atmosphere. For beryllium oxide obtained by thermal decomposition of beryllium sulfate, it was shown that some sulfate dissociates remained in the BeO powder even at >1000 ◦ C, although beryllium sulfate decomposes at ∼700–930 ◦ C [35]. However, the exact sulfur-containing species in the calcined powders at >1200 ◦ C could not be determined at this time due to limited data. It is worthy noting that the survival of SO4 2− ions in the YAG powder in 900–1300 ◦ C temperature
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 6
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Fig. 6. (a) Shrinkage and (b) shrinkage rate curves for the Y2 O3 powders calcined at 1100 ◦ C in the course of heating (10 ◦ C/min).
range has been verified using XPS data in a more recent study [6]. 3.4. Effects of sulfuric acid treatment on the sintering behaviors of the powders The effects of sulfuric acid treatment on the sintering behaviors were investigated for Y2 O3 powders calcined at 1100 ◦ C using a thermal mechanical analyzer (TMA), as shown in Fig. 6. Fig. 6(a) shows the change of the shrinkage in the course of 10 ◦ C/min heating. The onset temperature of shrinkage was ∼1150 ◦ C for the untreated Y2 O3 , and was slightly hindered for the 4 mol% sample. Further increasing sulfate concentration, the onset temperature was shifted to higher temperatures. Using the shrinkage curves in Fig. 6(a), the temperature changes of the densification rate were determined and shown in Fig. 6(b). All the samples showed a broad maximum peaked at temperatures between 1400 and 1500 ◦ C, representing the temperature of pronounced densification. It can be seen in Fig. 6(b) that the temperature of maximum peak shifted to lower temperatures with the doping of sulfate, indicating the increased sinterability for the treated-Y2 O3 powder. The 7 mol% sample revealed the lowest peaking temperature, ∼60 ◦ C lower than that of untreated Y2 O3 powder. The sintering behaviors for the powders were further assessed by performing sintering in vacuum. Fig. 7 shows the temperature dependent relative density for the powders untreated and treated with 7 mol% sulfuric acid. Significant densification was achieved for both samples at temperature <1500 ◦ C. The treated Y2 O3 showed a relative high density of 98.1% when sintered at 1500 ◦ C, much higher than that of 94.2% for the untreated Y2 O3 , indicating its excellent sinterability. This result is also in agreement with above sintering kinetic study. Sintered density of both samples increased with increasing sintering temperature, reaching >99% density at 1700 ◦ C. Fig. 8 compares the microstructures of fractured surfaces for the samples untreated and treated with 7 mol% sulfate sintered at 1550 ◦ C. Pores both in the grains and at the grain boundaries were observed in the sintered sample from untreated Y2 O3 , as shown in Fig. 8(a). The powder untreated presented some hard agglomeration (Fig. 2(a)) and the pores were difficult to remove during sintering and were entrapped in the sample. On the other hand, the ceramics
from treated-powder reveals a well-defined microstructure and pores were less frequently observed (Fig. 8(b)). The homogenous microstructure of the sintered sample can be attributed to the well-dispersed powders facilitated by the sulfuric acid-based pre-treatment. Fig. 9 shows the images for the ceramics vacuum-sintered at 1700 ◦ C. The pellets are of 1.5 mm thickness. The ceramics fabricated from treated powders are transparent, and the word under the samples can be clearly read. All the ceramics without sulfate dosage are opaque, irrespective of the calcination temperature of the powders. The in-line transmittance (at the wavelength of 800 nm) of the ceramics shown in Fig. 9 is compared in Fig. 10. All the samples untreated have an in-line transmittance <5%, whereas it was significantly higher for those of the treated samples. It is seen in Fig. 10 that both the calcination temperature and amount of sulfate dosage influence the transmittance; the highest transmittance was achieved for the sample treated with 7 mol% sulfuric acid and calcined at 1100 ◦ C. It is also should be noted that the in-line transmittance data achieved for the sample with the optimal sulfate dosage is still less than the theoretic value of single crystal Y2 O3 (∼81.9% at 800 nm [36]). Elaboration of powder processing process or optimization of subsequent consolidation and sintering procedures may required to further increase the optical transmittance of present Y2 O3 ceramics.
Fig. 7. Densification behaviors for the samples treated and untreated with 7 mol% sulfate sintered at various temperatures in vacuum.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
7
Fig. 8. Microstructure of the fractured surface of Y2 O3 ceramics (a) untreated and (b) treated with 7 mol% sulfate sintered at 1550 ◦ C for 5 h in vacuum.
4. Discussions 4.1. Effects of the added sulfuric acid on powder sintering ability
Fig. 9. Images for Y2 O3 ceramics vacuum-sintered at 1700 ◦ C for 5 h. All the samples were polished and are of 1.5 mm thickness. The samples are denoted by both sulfate concentrations and calcination temperatures. Y11-7S indicates that the powder with 7 mol% sulfate dosage was calcined at 1100 ◦ C.
The sintering behavior of present Y2 O3 powder could be greatly improved by pre-treating starting materials with an optimum amount of sulfuric acid, as shown in Fig. 7, where the treated-powder sintered at 1500 ◦ C has a relative density of 98.1%, being significantly higher than that of the sample untreated (∼94.2%). When sintered at 1700 ◦ C, it is seen that all the samples prepared using treated-powders were transparent, whereas it was totally opaque for the untreated sample, further confirmed the excellent sintering ability for the treated-yttria. The enhanced sintering ability for the treated-yttria can be correlated to several particle features facilitated by the sulfuric acid-based treatment. We can see that the powder untreated revealed particle morphologies of flat surfaces and sharp edges, whereas the powder with sulfate dosage are composed of round particles of isotropic surfaces, as shown in Fig. 2(b)–(e). It has been reported that isotropy in particle morphologies exceed anisotropic microstructural features, with respect to homogeneous packing of particles in consolidation and elimination of pores in the final stage of sintering [37]. Moreover, treating with sulfuric acid induced particles with finer particle size and lesser extent of aggregation, as revealed in Fig. 3 and Table 1. Sintering kinetics theories indicates that the shrinkage rate of a powder compact increases appreciably as the particle size decreases [38]. Besides, for powders showing severe agglomeration of the primary particles, increased relative content of agglomerates was found to hinder pore elimination in the sintering stage. 4.2. The mechanisms of the added sulfuric acid on powder properties
Fig. 10. The inline transmittance of 1.5-mm thick yttria ceramics at wavelength of 800 nm.
The underlying reasons for the observed morphological differences between the treated and untreated Y2 O3 powders can be elucidated using CRH sintering data between 1100 and 1300 ◦ C. It is seen in Fig. 6(a) that the temperature at which
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9 8
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
densification begins was shifted to higher temperatures for the treated powder. The onset temperature of shrinkage was ∼1150 ◦ C for the starting Y2 O3 powder, whereas it was 1250 ◦ C for the powder treated with 7 mol% SO4 2− . However, Table 1 reveals that all the calcined powders had similar particle size at 1200 ◦ C, which is also the case for the green compacts of these powders in CRH sintering. Clearly, although similar trend of particle coarsening occurred in the powder compacts, densification was greatly suppressed in 1100–1250 ◦ C temperature range for the treated yttria, suggesting different mass transportation mechanisms between the treated and untreated powders. According to the sintering theory, five mass transport processes, namely evaporation–condensation, surface diffusion, volume diffusion, grain boundary diffusion and viscous or creep flow, contribute to particle coarsening and sintering of a powder compact. The processes of volume diffusion, grain boundary diffusion, and viscous or creep flow induce neck growth or particle growth, accompanied by shrinkage among particles. Evaporation–condensation and surface diffusion from the surface to the neck area, on the other hand, can only lead to particle growth, because these mechanisms do not allow the particle centers to move closer together. Based on the above analysis, it is then may be concluded that the doped sulfate contributes to mechanisms leading to particle coarsening at temperature <1250 ◦ C, whereas the mechanisms contributing to densification dominate in the untreated Y2 O3 . Thus the observed morphological differences between the treated and untreated Y2 O3 can be partially explained as follows. For the treated powder, the doped sulfate inhibits mass transfer processes related to densification, hence particles tend to coarsen rather than densify, producing a loosely agglomerated and sinterable powder. On the other hand, mass transport processes leading to densification dominate in the untreated yttria powder. As a result, particles coarsening proceeds accompanied by simultaneous sintering, producing hard agglomerates during the calcination stage. A powder compact containing such agglomerates is nonsinterable. At this point, however, it is still unclear which mechanism leads to particle coarsening in the sulfate-doped powders. In view of low vapor pressure of yttria, Ikegami et al. [39] suggested that surface diffusion is the most likely process for particle growth in sulfate-ion-doped yttria up to 950 ◦ C. The existence of SO4 2− -ions has been verified for the treated-yttria at temperature <1200 ◦ C. Considering low vapor pressure of sulfate, contributions from sulfate-contributed mass transportation processes cannot be fully eliminated. Detailed sintering study of green compacts for powders calcined at different temperatures has been performed to further elucidate the mechanism of doping sulfate, and will be reported in our subsequent paper. 4.3. Determination of the optimum sulfate dosage The direct application of a dopant in pure Y2 O3 powders allow us precisely establishing a correlation between powder property and the amount of doped sulfate. Both sintering data and powder characterization indicate an optimum dosage of 7 mol%. In view of strong electrostatic attraction between positive yttria particle surfaces and negative SO4 2− -ions [30], we postulated
that the optimum sulfate dosage may be interconnected with the full coverage of SO4 2− -ions on the yttria particle surfaces. The ionic radius of SO4 2− is 0.235 nm and the present yttria powder has a BET surface area of 31.5 m2 /g. Supposing a single layer absorption, the calculated sulfate required for the full coverage of yttria particle surfaces was 6.98 mol%. The good match of the calculated data and experimental result further confirmed our assumption. Thus, the different characteristics for the powders treated with varied amount of sulfate can be explained as follows. For the powder doped with <7 mol% of SO4 2− -ions, insufficient sulfate coverage results in mixed particles, with a great portion of which possessing features of untreated yttria (e.g. larger particle size, more severe aggregation). For the dosage >7 mol%, however, the un-adsorbed free sulfate may accumulate, forming single particles of sulfate. Upon calcination, the free sulfate decomposed into oxide simultaneously. As a result, the calcined products were composed of particles evolved from two distinct origins, leading to a wider size distribution in the calcinates. This assumption was confirmed in Fig. 2(d) and (e), where particles of two different features were clearly revealed. 5. Conclusions The effects of doped SO4 2− ions on structural evolution and sintering behaviors of a commercial Y2 O3 powder were investigated. The sulfate dosage was achieved by pre-treating the starting material with sulfuric acid. The surface morphology and size of the particles were notably affected by the applied sulfuric acid-based treatment. The powder untreated revealed particle of flat surfaces and sharp edges, whereas the powder with sulfate dosage are composed of round particles of isotropic surfaces. Besides, treating with sulfuric acid resulted in particles with finer size and lesser aggregates. The doped sulfate inhibits mass transfer processes related to densification, producing a loosely agglomerated and sinterable powder upon calcination. Mass transport processes leading to densification dominate in the untreated yttria powder, where particles coarsening proceeded accompanied by simultaneous sintering, producing hard agglomerates in the calcination stage. The sintering behavior of present Y2 O3 powder could be greatly improved by pre-treating starting materials with an optimum amount of sulfuric acid. Dispersive Y2 O3 nanopwder with excellent sintering property was obtained from the powders treated with 7 mol% sulfuric acid and calcined at 1100 ◦ C for 4 h. The optimum sulfate dosage can be well correlated with a single layer coverage of SO4 2− -ions over the particle surfaces. Insufficient sulfate coverage results in powders containing a great portion of particles featuring those of untreated yttria (e.g. larger particle size, more severe aggregation). All the ceramics prepared using treated-powders were transparent after vacuumsintering at 1700 ◦ C, whereas it was totally opaque for those prepared with the untreated powder. Acknowledgements The work was supported in part by the Program for New Century Excellent Talents in University (NCET-11-0076), National
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008
+Model JECS-10003; No. of Pages 9
ARTICLE IN PRESS J. He et al. / Journal of the European Ceramic Society xxx (2015) xxx–xxx
Natural Science Foundation of China (51172037, 51102040 and 50990303), and the Fundamental Research Funds for the Central Universities (N130810003, N130610001, and N110802001). References [1]. Kong J, Lu J, Takaichi K, Uematsu T, Ueda K, Tang DY, et al. Diode-pumped Yb:Y2 O3 ceramic laser. Appl Phys Lett 2003;82:2556–8. [2]. Eilers H. Fabrication, optical transmittance, and hardness of IR-transparent ceramics made from nanophase yttria. J Eur Ceram Soc 2007;27:4711–7. [3]. Zhang HB, Kim B-N, Morita K, Yoshida H, Hiraga K, Sakka Y. Fabrication of transparent yttria by high-pressure spark plasma sintering. J Am Ceram Soc 2011;94:3206–10. [4]. Mouzon J, Maitre A, Frisk L, Lehto N, Oden M. Fabrication of transparent yttria by HIP and the glass-encapsulation method. J Eur Ceram Soc 2009;29:311–6. [5]. Li J-G, Ikegami T, Mori T, Yajima Y. Sc2 O3 nanopowders via hydroxyl precipitation: effects of sulfate ions on powder properties. J Am Ceram Soc 2004;87:1008–13. [6]. Deineka TG, Doroshenko AG, Mateychenko PV, Tolmachev AV, Vovk EA, Vovk OM, et al. Influence of sulfate ions on properties of co-precipitated Y3 Al5 O12 :Nd3+ nanopowders. J Alloys Compd 2010;508:200–5. [7]. Ikegami T, Li J-G, Mori T, Moriyoshi Y. Fabrication of transparent yttria ceramics by the low-temperature synthesis of yttrium hydroxide. J Am Ceram Soc 2002;85:1725–9. [8]. Qin HM, Liu H, Sang YH, Lv YH, Zhang XL, Zhang YY, et al. Ammonium sulfate regulation of morphology of Nd:Y2 O3 precursor via urea precipitation method and its effect on the sintering properties of Nd:Y2 O3 nanopowders. CrystEngComm 2012;14:1783–9. [9]. Qin HM, Zhang XH, Liu H, Sang YH, Wang JY. Mechanism of ammonium sulfate regulation effect on microstructure of Y2 O3 nanopowders via urea precipitation method. CrystEngComm 2013;15:5076–81. [10].Li N, Yanagisawa K. Controlling the morphology of yttrium oxide through different precursors synthesized by hydrothermal method. J Solid State Chem 2008;181:1738–43. [11].Hirai T, Orikoshi T, Komasawa I. Preparation of Y2 O3 :Yb,Er infraredto-visible conversion phosphor fine particles using an emulsion liquid membrane system. Chem Mater 2002;14:3576–83. [12].Rulison AJ, Flagan RC. Synthesis of yttria powders by electrospray pyrolysis. J Am Ceram Soc 1994;77:3244–50. [13].Subramanian R, Shankar P, Kavithaa S, Ramakrishnan SS, Angelo PC, Venkataraman H. Synthesis of nanocrystalline yttria by sol–gel method. Mater Lett 2001;48:342–6. [14].Dupont A, Parent C, Le Garrec B, Heintz JM. Size and morphology control of Y2 O3 nanopowders via a sol–gel route. J Solid State Chem 2003;171:152–60. [15].Mangalaraja RV, Mouzon J, Hedstrom P, Camurri CP, Ananthakumar S, Oden M. Microwave assisted combustion synthesis of nanocrystalline yttria and its powder characteristics. Powder Technol 2009;191:309–14. [16].Saito N, Matsuda S, Ikegami T. Fabrication of transparent yttria ceramics at low temperature using carbonate-derived powder. J Am Ceram Soc 1998;81:2023–8. [17].Ikegami T, Mori T, Yajima Y, Takenouchi S, Misawa T, Moriyoshi Y. Fabrication of transparent yttria ceramics through the synthesis of yttrium hydroxide at low temperature and doping by sulfate ions. J Ceram Soc Jpn 1999;107:297–9. [18].Onodera T, Ikegami T, Yajima Y, Kawamura M, Sakai M, Moriyoshi Y. Effects of the sulfate ion exerted on densification of yttria on precursor synthetic. J Ceram Soc Jpn 2003;111:664–8.
9
[19].Wen L, Sun XD, Xiu ZM, Chen SW, Tsai C-T. Synthesis of nanocrystalline yttria powder and fabrication of transparent YAG ceramics. J Eur Ceram Soc 2004;24:2681–8. [20].Wen L, Sun XD, Lu Q, Xu GX, Hu XZ. Synthesis of yttria nanopowders for transparent yttria ceramics. Opt Mater 2006;29:239–45. [21].Li J, Liu WB, Jiang BX, Zhou J, Zhang WX, Wang L, et al. Synthesis of nanocrystalline yttria powder and fabrication of Cr,Nd:YAG transparent ceramics. J Alloys Compd 2012;515:49–56. [22].Sacks MD, Tseng T-Y, Lee SY. Thermal decomposition of spherical hydrated basic aluminium sulfate. Ceram Bull 1984;63:301–10. [23].Kara F, Sahin G. Hydrated aluminium sulfate precipitation by enzymecatalysed urea decomposition. J Eur Ceram Soc 2000;20:689–94. [24].Li J-G, Ikegami T, Mori T. Solution-based processing of Sc2 O3 nanopowders yielding transparent ceramics. J Mater Res 2004;19:733–6. [25].Li J-G, Ikegami T, Mori T, Yajima Y. Monodispersed Sc2 O3 precursor particles via homogeneous precipitation: synthesis, thermal decomposition, and the effects of supporting anions on powder properties. J Mater Res 2003;18:1149–56. [26].Matsushita N, Tsuchiya N, Nakatsuka K, Yanagitani T. Precipitation and calcination process for yttrium aluminum garnet precursors synthesized by the urea method. J Am Ceram Soc 1999;82:1977–84. [27].Lv YH, Zhang W, Liu H, Sang YH, Qin HM, Tan J, et al. Synthesis of nanosized and highly sinterable Nd:YAG powders by the urea homogeneous precipitation method. Powder Technol 2012;217:140–7. [28].Xu XJ, Sun XD, Liu H, Li J-G, Li XD, Huo D, et al. Synthesis of monodispersed spherical yttrium aluminum garnet (YAG) powders by a homogeneous precipitation method. J Am Ceram Soc 2012;95:3821– 6. [29].Li XX, Wang WJ. Preparation of uniformly dispersed YAG ultrafine powders by co-precipitation method with SDS treatment. Powder Technol 2009;196:26–9. [30].Li J, Li JP, Chen Q, Wu WJ, Xiao DQ, Zhu JG. Effect of ammonium sulfate on the monodispersed Y3 Al5 O12 nanopowders synthesized by co-precipitant Method. Powder Technol 2012;218:46–50. [31].Wang JQ, Zheng SH, Zeng R, Dou SX, Sun XD. Microwave synthesis of homogeneous YAG nanopowder leading to a transparent ceramic. J Am Ceram Soc 2009;92:1217–23. [32].Caponetti E, Martino DC, Saladino ML, Leonelli C. Preparation of Nd:YAG nanopowder in a confined environment. Langmuir 2007;23:3947– 52. [33].Skrobian M, Sato N, Yamada K, Fujino T. Thermogravimetric study of reduction and sulfurization of Y2 (SO4 )3 using carbon disulfide. Thermochim Acta 1995;255:201–9. [34].Lian JB, Liu F, Wang XJ, Sun XD. Hydrothermal synthesis and photoluminescence properties of Gd2 O2 SO4 :Eu3+ spherical phosphor. Powder Technol 2014;253:187–92. [35].Ikegami T, Mori Y, Matsuda M, Suzuki H. Surface structure of BeO powders obtained by calcination of sulfate. Yogyo Kyokaishi 1973;81:379– 82. [36].Nigara Y. Measurement of the optical constants of yttrium oxide. Jpn J Appl Phys 1968;7:404–8. [37].Ikegami T, Eguchi K. Two kinds of roles of MgO in the densification and grain growth of alumina under various atmospheres: sensitive and insensitive roles to the experimental procedures. J Mater Res 1999;14:509–17. [38].Coble RL. Initial sintering of alumina and hematite. J Am Ceram Soc 1958;41:55–62. [39].Takayasu I, Li J-G, Sakaguchi I, Hirota K. Morphology change of undoped and sulfate-ion-doped yttria powder during firing. J Am Ceram Soc 2004;87:517–9.
Please cite this article in press as: He J, et al. Effects of pre-treatment of starting powder with sulfuric acid on the fabrication of yttria transparent ceramics. J Eur Ceram Soc (2015), http://dx.doi.org/10.1016/j.jeurceramsoc.2015.02.008