Solid State Ionics 176 (2005) 2955 – 2956 www.elsevier.com/locate/ssi
Nanostructured phosphorous tungsten bronzes from ultrasonic spray pyrolysis V. Jokanovic´ a,*, U.B. Mio* b, Z.P. Nedic´ b b
a Institute of Nuclear Sciences ‘‘Vincˇa’’, P.O.Box 522, 11001 Belgrade, Serbia and Montenegro Faculty of Physical Chemistry, University of Belgrade, P.O.Box 137, 11001 Belgrade, Serbia and Montenegro
Abstract Ultrasonic spray pyrolysis has been used to obtain fine spherical powders of WPA bronzes from a 29-tungstenphosphoric acid (WPA-29) feed solution. The average grain size of the obtained powder particles is ca. 1000 nm; these consist of subparticles in the 30 – 80 nm range. The experimentally determined average WPA bronze particle size values are compared with theoretically calculated ones; fairly good agreement between them is found. It is also shown that the average grain size and grain size distribution of the structure and substructure of the obtained powder are predictable. D 2005 Elsevier B.V. All rights reserved. PACS: 40; 60; 80 Keywords: Phosphorous doped tungsten bronzes; Spray pyrolysis; Nanostrcture; Nanodesign
1. Introduction A recently developed procedure for the synthesis of phosphorous-doped tungsten bronzes is based on thermally induced phase transformation of WPA and its salts [1] in the temperature range 600 – 1150 -C. Keggin anions of WPA are transformed by solid – solid recrystallization at about 600 -C into new phosphorus-doped tungsten bronzes of PW8O26 type. On further increasing the temperature to 1150 -C, the synthesized cubic bronze passes through three polymorphic phase transformations. The spray pyrolysis method provides a homogeneous chemical reaction at the molecular level throughout the whole particle volume and, consequently, very fast phase transformation and production of bronze powders in only one step, with a well-determined structure and substructure [2 –7]. 2. Experimental Synthesis of submicron WPA bronze spheres was performed in a laboratory set-up for ultrasonic spray pyrolysis using 20%
* Corresponding author. E-mail addresses:
[email protected],
[email protected] (V. Jokanovic´). 0167-2738/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ssi.2005.09.029
29-WPA solution as a precursor. Briefly, the laboratory set-up consists of an ultrasonic atomizer for aerosol generation (Gapusol 9001, RBI) with three transducers operating at a frequency of 1.7 MHz, a reaction chamber-furnace (Heraeus ROF 7/50) with a quartz tube, and a vessel for particle collection. The carrier gas (N2) flow rate was 0.022 l/s, while the temperature was maintained at 800 -C. The flow rate of the aerosol droplets in the furnace was 0.107 m/s. The size distribution of WPA bronzes was analyzed by a semi-automatic image analyzer (Videoplan, Kontron) connected to a scanning electron microscope (SEM, JOEL 5300). 3. Results and discussion 3.1. Design of the powder particle structure Comparison is made of data from a previously developed theoretical model, given in Jokanovic´ et al. [2 –5], for the primary and secondary WPA powder particle sizes and the Table 1 Experimentally determined particle size distribution for WPA bronze particle diameters and the frequency of their appearance (I) d p (nm) 560 670 780 890 1000 1110 1220 1390 1670 % 9.2 9.2 7.5 11.0 24.5 13.0 13.5 9.0 3.0
V. Jokanovic´ et al. / Solid State Ionics 176 (2005) 2955 – 2956
2956
Fig. 2. SEM micrograph of WPA particles and subparticles. Fig. 1. SEM micrograph of WPA particles.
experimentally determined values for the same particles. The sizes of the particles and subparticles were analyzed; the distribution of secondary particle sizes for the WPA bronze spheres is presented in Table l and Fig. 1. Regardless of the distribution function used for the data fitting (Gaussian or Lorenzian), almost identical values for the mean particle size were obtained: d G = 999.5 T 9 (Gaussian) and d L = 989.7 T 10 (Lorenzian). Assuming ideal dense packing (theoretical packing density) within the particle itself, the calculated mean WPA bronze particle size is 1066 nm [2– 5]. The mean particle size obtained from the previously determined mean size of the aerosol droplets generated in the ultrasonic field (frequency = 1.7 MHz) was found to be 4.32 Am. Input data for this analysis: molecular mass of precursor WPA-29 (H3PW12O40I29H2O) M r = 3404 g/dm3 and WPA bronze (H3PW12O40) M B = 2882 g/ cm3; theoretical densities of WPA-29 and WPA bronzes: 0.436 g/cm3 and 11.283 g/cm3, respectively, calculated from the corresponding cubic lattices of WPA-29, V pr = 12.695 nm3, and WPA bronze, V B = 0.424 nm3; contraction of the WPA-29 precursor: 20%; and precursor surface tension: 72 I 10 3 mN/m. This average droplet diameter (4.32 Am) was then used together with the reduction factor for droplet volume transformation into the powder particles (0.247) to finally give the average particle size. Under conditions of the low coupling, i.e., weak interaction between the particle and sub-elements (no bridging and no penetration between the particles due to surface diffusion), no diffusion along the grain boundaries or any other transfer mechanism occur. Analysis of the packing arrangement and packing density shows double hierarchical packing in the WPA bronzes. It can be seen from Table 2 that the percentage of particles with diameters in the range of 853 – 1320 nm is 80%. Experimental results show that over 70% of particles are in Table 2 Theoretical estimate of the droplet and particle diameter distributions (d d and d p) d d (Am) d p (nm) %
3.5 853 52
5.7 1320 30
7.7 1780 10
9.6 2220 8
the same range of discrete values, suggesting fair agreement with the theory. There is a slight difference between the theoretically obtained values for the average droplet size (1066 nm) and the complete spectrum of the particle size distribution (1190 nm). One value is lower and the other higher than the experimental value (by 7% and 19%, respectively). The average theoretical diameter of subdroplets was calculated to 155 nm; after including a factor for subdroplet volume reduction during their transformation, the average subparticle size was found to be 38 nm. The subparticle size spectrum gave values in the following order: 31, 51, 65 and 79 nm. The experimentally obtained subparticle size (Fig. 2) is 55 nm, in fair agreement with the theoretically predicted one. 4. Conclusions Quantitative analysis of the structural and substructural elements of WPA bronze powder particles has been made in terms of their mean diameter and size distribution. Theoretical and experimental values were found to be in fair agreement. The experimentally obtained value for the average size of WPA bronze particles is ca. 1000 nm, in good agreement with the theoretical estimate of 1066 nm. The average size of the subparticles (experimental = 55 nm and theoretical = 31 –79 nm) are also in fair agreement. References [1] U.B. Mioc, R.Z. Dimitrijevic, M. Davidovic, Z.P. Nedic, M.M. Mitrovic, Ph. Colomban, Journal of Materials Science 29 (1994) 3705. [2] V. Jokanovic´, D. Janac´kovic´, A.M. Spasic´, D. Uskokovic´, Materials Transactions, JIM 37 (1996) 627. [3] V. Jokanovic´, D. Janac´kovic´, D. Uskokovic´, Ultrasonics Sonochemistry 6 (1999) 157. [4] V. Jokanovic´, D. Janac´kovic´, D. Uskokovic´, Journal of Nanostructured Materials 12 (1999) 349. [5] J. Nedeljkovic´, Z. Sˇaponjic´, Z. Rako*evic´, V. Jokanovic´, D. Uskokovic´, Journal of Nanostructured Materials 9 (1997) 125. [6] J.M. Reed, Introduction to the Principles of Ceramic Processing, John Wiley & Sons, Inc., New York, 1998, p. 185. [7] G.L. Messing, S.C. Zhang, G.V. Yaynathi, Journal of the American Ceramic Society 76 (1993) 2707.
