Printer Friendly

Effect of reactor ambient pressure on the morphology of spray dried magnesium sulphate powders.

The effect of reducing the reactor air pressure on the morphology of spray dried magnesium sulphate powders is investigated, experimentally. A reactor, capable of drying and pyrolyzing solution sprays at low pressures, is designed and manufactured. A vibrating mesh nebulizer is employed to generate the spray. Four different pressures, starting from 60 Torr to the atmospheric pressure, and two different reactor air temperatures of 130[degrees]C and 420[degrees]C, are considered. In addition, two different concentrations of magnesium sulphate solutions are tested. The results are explained based on the effect of reactor air pressure on the droplet evaporation rate.

On a etudie de maniere experimentale l'effet de la reduction de la pression d'air dans le reacteur sur la morphologie de poudres de sulfate de magnesium sechees par atomisation. Un reacteur capable de secher et de pyrolyser des vaporisations de solutions a ete concu et fabrique. Les vaporisations sont effectuees a l'aide d'un orifice d'atomisation a mailles vibrantes. Quatre pressions differentes, partant de 60 Torr a la pression atmospherique, et deux temperatures d'air de reacteur differentes de 130[degrees]C et 420[degrees]C, sont considerees. De meme, deux concentrations de sulfate de magnesium differentes sont testees. Les resultats sont expliques d'apres l'effet de la pression d'air dans le reacteur sur la vitesse d'evaporation des gouttelettes.

Keywords: spray pyrolysis, spray drying, powder morphology, reduced pressure effects

INTRODUCTION

Spray drying and spray pyrolysis are two similar methods of powder production widely used in recent years. In these methods, a solution is sprayed into a tubular reactor where the solution droplets dry to form the final particles. In spray pyrolysis an additional decomposition process may occur after partial drying of the particles (droplets) due to substantial temperature rise in the particles. In these methods, depending on the operating conditions, either solid or hollow particles may form.

In recent years many researchers have focused on the production and characterization of various fine and ultra fine powders produced by spray drying and spray pyrolysis. Janackovic et al. (1996) and (1997) synthesized cordierite and mullite particles, respectively, using ultrasonic atomization. They studied the powder formation mechanism. Murugavel et al. (1997) prepared titania and zirconia nanopowders using ultrasonic spray pyrolysis. They studied the effect of process temperature on the crystallinity of the powders. They concluded that the use of metal organic precursors ensures complete decomposition at low temperatures. In addition they observed that some of the as-prepared particles are hollow and disrupted. Miao et al. (2001) prepared and characterized Zr[O.sub.2] and SiC ceramic thin films using electrostatic atomization. Nimmo et al. (2002) produced ultra fine zirconia powders using spray pyrolysis. In order to produce powders, they designed and employed a novel twin fluid atomizer, which generates small droplets. Mueller et al. 2004 used flame spray pyrolysis to produce zirconia powders at high rates. They investigated the effect of precursor concentration and dispersion gas flow rate on powder morphology. Tsai et al. (2004) presented their findings regarding the effects of precursor droplet size and precursor concentration on size and morphology of the powders. Lenggoro et al. (2004) synthesized nanoparticles of a doped oxide phosphor using conventional spray pyrolysis. However, in spite of numerous studies on powder synthesis using spray drying and pyrolysis, there virtually is no systematic investigation on the effect of reactor air pressure on the morphology of powders. The only available study on low pressure spray pyrolysis is that of Kang and Park (1995). However, their main purpose was to investigate a particular aerosol generator and not the details of the reactor air pressure effect on the particle formation. The morphology of the powders (ZnO and Z[n.sub.2]Si[O.sub.4]) produced by their aerosol generator was different from those produced by previous aerosols formed by ultrasonic atomizers (e.g. Lenggoro et al., 2000). In addition, in order to study the effect of process parameters on nanopowder formation, one can reduce the reactor pressure and employ micron-sized droplets to simulate nanodroplet evaporation and nanopowder formation in the atmospheric pressures (Eslamian et al. 2006). As such, the present study is aimed at understanding the effect of pressures below atmospheric on the morphology of magnesium sulphate powders. The experimental set-up is described first, followed by the presentation and discussion of the results for powder production at various pressures and temperatures.

EXPERIMENTAL SET-UP AND PROCEDURE

A schematic of the experimental apparatus used in this study is shown in Figure 1. It consists of: (i) a vacuum chamber and its accessories including feedthroughs for fluid, thermocouple and power; (ii) a tubular reactor; (iii) a vibrating mesh nebulizer; (iv) a vacuum pump; and (v) three thermocouples and a pressure gauge. The stainless steel vacuum chamber is 2 m in height and 1 m in diameter. The full cylindrical reactor, which is placed inside the vacuum chamber, is comprised of two semi-cylindrical electrical radiant heaters. The inner diameter and the length of the reactor are 8 cm and 71 cm, respectively. Three thermocouples are mounted at the top, in the middle, and at the bottom of the reactor to measure the air temperature during each experiment at 1 cm away from the reactor internal wall. Experiments showed that no major temperature gradient was present in the radial direction. The vacuum pump is a two stage mechanical pump (BOC Edwards, England, Model: E2M40/EH250). Before performing an experiment the chamber pressure was reduced gradually to a desired level using the vacuum pump, and then the pump valve was kept closed for the rest of the experiment.

A vibrating mesh nebulizer (Omron Healthcare Co., Model NE-U22) was employed for spray generation. This commercial nebulizer is designed for drug delivery as an inhaler. The nebulization rate of this inhaler is 0.4 to 0.5 ml/min. Figure 2a shows a schematic of the vibrating mesh nebulizer, and Figure 2b shows the resulting spray image that was taken using a laser sheet lighting. A liquid solution bottle is connected to a transducer horn. A high frequency (1.8 kHz) vibration of a piezoelectric crystal (PZT) is transmitted to the transducer horn. The vibration of the horn pushes the liquid solution through apertures in the mesh plate placed above it (Dhand, 2002). The spray was characterized using a two component Phase Doppler Particle Analyzer (PDPA) (TSI Inc.). The PDPA consists of a laser-based optical transmitter, an optical receiver, an electronic signal processor and software for data acquisitions and analysis. The droplet size and the vertical velocity distribution were obtained at 2 cm from the mesh plate, and at the centre line of the spray. The nebulizer performance and its spray characteristics depend on the solution properties, such as solution type and concentration. In the experiments, 0.7 M (M is the molarity of the solution) and 1.0 M solutions of magnesium sulphate were used. For spray characterization, a 1.0 M solution was employed as the test liquid (Figures 2c and 2d show droplet size and velocity distribution). The measurements show that the droplet average velocity is 1.5 m/s and the droplet mean diameter ([d.sub.10]) is 7.4 [micro]m, with a size range between 1 [micro]m to 13 [micro]m. The results also showed that larger droplets had higher velocities. The spray was not characterized for a 0.7 M solution, however, it was characterized for pure water, for which the droplet mean diameter (d10) was 6.0 [micro]m, with a size range between 1 [micro]m to 13 [micro]m, and the average droplet velocity was 1.9 m/s, which was close to the measurements for C = 1.0 M. Since the experiments were conducted in low pressures, no carrier gas was used for carrying the droplets. Instead, the natural upward convection of the air induced by the buoyancy force was utilized to carry the droplets upward. Therefore, the nebulizer was placed at about 5 cm from the bottom end of the hot reactor and the droplets were introduced to the centre of the reactor via an 8 cm copper tube (see Figure 1). The chamber temperature outside the reactor was much lower than the temperature inside the reactor and, therefore, we did not need to cool the nebulizer body. During the experiments, the sprayed droplets flowed upward into the reactor, and after drying fell back to the powder collector located at the bottom of the reactor. The droplet number density of the spray was low and, therefore, interaction between the droplets and the particles was small. For each experiment, the chamber pressure was reduced to a desired value. Once the steady state condition was recovered in the system (in 5 - 10 min) the nebulizer was turned on. The nebulizer was operating for about 30 min. Then the reactor was shut down and the collected powders were used for Scanning Electron Microscope (SEM) sample preparation. Our calculations show that the total time of evaporation of a 7.4 [micro]m water droplet at 130[degrees]C and 760 Torr is 38 ms. Therefore, the solution droplets dry at a distance less than 10 mm from the nebulizer outlet. However, dried particles may travel different distances from the nebulizer outlet, depending on the reactor pressure and temperature. At high temperatures (greater buoyancy force), powders travel to higher locations in the reactor and less amount of powder is collected at the bottom of the reactor. Reducing the pressure (lower buoyancy force), however, results in a decrease in the particle flight time and length. In order to have a more uniform temperature distribution in the reactor and to prevent the particles from escaping the reactor, the top end of the reactor was nearly closed (a small opening allowed a small upward drift). The minimum chamber pressure used in this study was 60 Torr. At lower pressures, the liquid inside the nebulizer bottle started to boil, and the nebulizer stopped working. At 41[degrees]C the water vapour pressure is 60 Torr, which implies that at chamber pressure of 60 Torr, if the nebulizer surrounding air temperature is 41[degrees]C or more, the solution in the nebulizer bottle boils and no spray is generated.

[FIGURES 1-2 OMITTED]

We considered four different reduced pressures of 760, 400, 250 and 60 Torr. The temperature at the bottom of the reactor was not controlled and it was changing depending on the chamber pressure. This configuration gave us the opportunity to investigate the effect of the pressure change on the temperature distribution along the reactor. The high reactor air temperature at the top of the reactor was 420[degrees]C, and the low one was 130[degrees]C. During each experiment, the temperature at the top of the reactor was continuously measured and controlled by a temperature controller (Watlow, series 965). The solution concentrations used in the powder production experiments were 0.7 M and 1.0 M. We attempted to perform experiments at higher solution concentrations than 1 M, but the nebulizer was not able to atomize high concentration liquid solutions. Also, at low solution concentrations, the powder production rate decreases.

For each experiment and at steady state conditions, air temperatures at 1 cm away from the reactor wall were measured at the middle of the reactor, ([T.sub.MIDDLE]) and at the bottom of the reactor, ([T.sub.BOTTOM]), while as noted earlier, the temperature at the top of the reactor was controlled and set to 130[+ or -]10[degrees]C and 420[+ or -]20[degrees]C. Figure 3 shows the variation of the temperature along the reactor for different pressures, when [T.sub.TOP] = 420[degrees]C. This figure shows that as the pressure decreases, the temperatures in the middle and at the bottom of the reactor increase. At the chamber pressures of 760 and 400 Torr, the temperature versus the distance from the reactor top end decreases. At pressures of 250 and 60 Torr, the temperature in the middle of the reactor is greater than the temperature at the top of the reactor. This is attributed to the low air currents at low pressures

[FIGURE 3 OMITTED]

and, therefore, accumulation of heat in the middle of the reactor. Figure 4 shows the variation of the temperature along the reactor for different pressures, when [T.sub.TOP] = 130[degrees]C. According to this figure, the reactor temperature decreases versus the distance from the reactor top end at chamber or reactor pressures of 760, 400 and 250 Torr. However, when the pressure is 60 Torr, the temperature in the middle of the reactor is greater than the temperature at the top of the reactor, [T.sub.MIDDLE] > [T.sub.TOP]. Again, this is attributed to the lack of air and convective heat transfer in reduced pressures.

[FIGURE 4 OMITTED]

EFFECT OF PRESSURE ON DROPLET EVAPORATION

In order to understand how the pressure can affect the particle morphology, we need to know how the pressure affects the evaporation rate of a solution droplet. Reducing the pressure, results in an increase in the mean free path of the surrounding gas molecules. The ratio of the mean free path of the surrounding gas molecules to the droplet radius is defined as the Knudsen number, Kn. When Kn number is greater than [10.sup.-3], the transport equations based on fluid continuum assumptions are no longer valid. At the continuum regime, the Kn number is smaller than [10.sup.-3]. When [10.sup.-3] < Kn < 0.1, the flow regime is called the slip regime. In the case of 0.1 < Kn < 10, the flow regime is called the transition regime. There is no simple equation to calculate the evaporation rate, at any of the above mentioned non-continuum regimes. Accurate results may be obtained by solving the simplified versions of the Boltzmann equation. Here, we make use of an interpolation formula based on the numerical solution of Boltzmann equation for the ratio of the evaporation rate in the slip and transition regime ([10.sup.-3] < Kn < 10) to the continuum regime (Loyalka, 1988):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where

[(dm/dt).sub.T] is the evaporation rate in the transition regime, [(dm/dt).sub.C] is the evaporation rate in the continuum regime, Kn is the Knudsen number, [lambda] is the mean free path of the air molecules, a is the droplet radius, and [xi] and [zeta] are matching parameters. Based on a molecular collision approximation, known as BGK approximation (Gombosi, 1994), [xi] = 1.0161 and [zeta] = 1.3330. Using Equation (1), we have calculated the ratio of the non-continuum to continuum evaporation rates, due to the non-continuum effects only, at the four operating pressures used in this study and for droplet diameters of 7.4, 0.74, and 0.074 [micro]m at 130[degrees]C. Results are shown in Figure 5. This figure shows that for the range of diameters considered here, the non-continuum effect is considerable at all pressures. In particular, when the pressure is as low as 60 Torr, the evaporation rate decreases to about 60% of that of the continuum-based evaporation rate (the effect of ambient pressure on the water vapour pressure at the droplet surface is not considered). Several of other studies (e.g. Eslamian et al., 2006) have shown that the low evaporation rate is favourable to the production of solid particles. On the other hand, depending on the nature of the precursor in spray pyrolysis, relatively high reactor temperature is essential to have a chemical decomposition in the precursor. Figure 5b indicates that reducing the operating pressure at relatively high temperatures, which is required for the chemical decomposition, decreases the evaporation rate, and therefore, increases the possibility of forming solid particles.

[FIGURE 5 OMITTED]

Assuming a spherically symmetric evaporation in a stationary bulk gas, Fick's law results in the following equation for the continuum evaporation rate (Li and Davis, 1996):

[(dm/dt).sub.C] = -4[pi]R[D.sub.AB[rho]g] ([W.sub.AR] [W.sub.A[infinity]] (3)

where m is the mass of the droplet at any time t, R is the droplet radius, [[rho].sub.g] is the gas density, [D.sub.AB] is the diffusion coefficient of the solvent vapour in air at the atmospheric pressure, [W.sub.AR] and [W.sub.A[infinity]] are the mass fraction of the vapour at the droplet surface and in infinity, respectively.

Using Equations (1) to (3), the evaporation rate of a droplet at the transition regime, and at different chamber pressures were calculated for [T.sub.TOP] = 420[degrees]C and [T.sub.TOP] = 130[degrees]C. In order to calculate the continuum evaporation rate from Equation (3), the vapour pressure at the surface of the droplet corresponding to its temperature has to be determined a priori. We here assumed that the droplet temperature reaches the wet-bulb temperature. The calculated values of the droplet life-time are in the order of several milliseconds. Considering the small droplet velocities in this case (1.5 m/s) all solution droplets dry after they have travelled less than 1 cm in the reactor. Therefore, the effective surrounding air temperature during the drying process is, [T.sub.BOTTOM]. In order to calculate the droplet wet-bulb temperature, one can write the heat balance equation around a droplet that is evaporating in stagnant air, in a spherically symmetric manner:

m[C.sub.p] d[T.sub.i]/dt 4[pi]Rk[T.sub.[infinity]] - [T.sub.l]) - L dm/dt = 0 (4)

where [C.sub.p] is the liquid heat capacity, k is the liquid thermal conductivity, [T.sub.[infinity]] is the droplet temperature far from the droplet (reactor air temperature), [T.sub.l] is the uniform droplet temperature, and L is the latent heat of vaporization. Droplet temperature reaches a constant value, after a short period of time. Since the objective is to calculate the wet-bulb temperature, the first term in Equation (4) is neglected. A standard procedure is used to calculate [T.sub.l]. Normally an initial guess is made for this temperature. Then by using the Clausius-Clapeyron equation, the vapour pressure corresponding to this temperature is calculated. Then, the evaporation rate, dm/dt, is calculated, employing Equations (1) to (3). Next, Equation (4) is used to find the corrected [T.sub.l]. This iterative procedure is repeated, until [T.sub.l] does not change any longer. The humidity of the chamber at the beginning of each experiment was measured to be about 45%. Therefore, for the wet-bulb temperature calculations, an air humidity of 45% was considered as the average humidity. Droplet mass evaporation rate, dm/dt, changes significantly with droplet diameter (Eslamian et al., 2006). Therefore, we need to use an average value for the mass evaporation rate. Based on our calculations, we used the mass evaporation rate of a 3 [micro]m droplet as the representative of the droplet average evaporation rate.

RESULTS AND DISCUSSION

Magnesium sulphate powders were produced at four different pressures of 760, 400, 250 and 60 Torr and at two different reactor top temperatures of 130[degrees]C and 420[degrees]C. Two initial solution concentrations of 0.7 M and 1.0 M were used. Scanning electron microscopy (SEM) tests were done on each sample to study the effect of pressure, temperature and evaporation rate on the morphology of the powders.

Figure 6 shows several SEM images of the powders produced from aqueous solution of magnesium sulphate at the reactor top temperature of 420[degrees]C and at four different pressures of 760, 400, 250, and 60 Torr. The evaporation rate for each chamber pressure is also shown below the images. Figures 6i and 6ii correspond to an initial solution concentration of 1.0 M, and Figures 6iii and 6iv correspond to an initial solution concentration of 0.7 M. No apparent difference exists between the morphology of the powders produced from a 0.7 M solution with those produced from a 1.0 M solution. This implies that the powder morphology and wall thickness is a weak function of the initial concentration, supporting what was obtained from the modelling of nanoparticle formation conducted by Eslamian et al. (2006). According to Figure 6a, regardless of the initial solution concentration, the particles produced at 760 Torr, are either solid or hollow but thick-walled. This is evident by the lack of any disrupted particle. The results also indicate that larger particles may not necessarily be solid. This is deduced by the existence of some large particles with a hole on their surface (Figure 6a-iii). Figure 6b shows the SEM images of the powders produced at 400 Torr. These particles have thinner crust, as compared to the atmospheric case. This is evident by the increased number of broken particles and direct imaging of the particle wall thickness. The particle wall thickness is measured to be about 0.4 [micro]m. Also, the particles seem to have both a larger mean size, and a broader size distribution. Figure 6c shows SEM images of the powders produced at 250 Torr. Several disrupted particles are observed. In addition, some of the particles are deformed and contracted. This is attributed to the condensation of the entrapped vapour inside the particle, which happens at the time the particle reaches the relatively cool zone of the reactor. Vapour condensation results in a substantial decrease in the particle internal pressure; therefore, the particle shrinks and contracts (Sydel et al., 2004; Eslamian and Ashgriz, 2005). Figure 6d shows the SEM images of the powders produced at 60 Torr. It is observed that regardless of the initial solution concentration, all of the particles are spherical and non-disrupted. This indicates that the particles are either solid or hollow but have a thick and strong wall. These results indicate that for the atmospheric pressure of 760 Torr and also for the lowest pressure of 60 Torr, the number of disrupted particles is lower and the particles seem to be more uniform in size and have better quality. By comparing the evaporation rates at different conditions as shown in Figure 6, we can explain the difference between the particle morphologies. The evaporation rate depends on the operating air pressure and the surrounding air temperature. For large droplets evaporating in the atmospheric pressures, for which Kn [congruent to] 0 and at constant ambient temperature reducing the pressure results in increasing the evaporation rate. However, for small droplets evaporating in the atmospheric pressures or larger droplets evaporating in reduced pressures, for which Kn > [10.sup.-3], the evaporation rate may increase or decrease. The surrounding temperature for each case is [T.sub.BOTTOM], as the droplets dry right after coming to the reactor bottom. Figure 6 shows that the evaporation rate at 400 and 250 Torr increases with a factor of about 2, and then decreases, as the pressure is reduced further to 60 Torr. When the evaporation rate is low, there is enough time for the solute to diffuse from the droplet centre to the droplet surface (or there is enough time for the water to diffuse from the droplet centre to the droplet surface). If at the onset of precipitation on the droplet surface, the solute concentration everywhere is higher than the equilibrium saturation, a volumetric precipitation occurs and the final particle is solid (Jayanthi et al., 1993). Low evaporation rates favour a volumetric precipitation within the droplet. For the intermediate pressures of 250 Torr and 400 Torr, the evaporation rate is relatively high and, therefore, a rapid evaporation takes place and some of the particles burst as it is evident from Figures 6b and 6c. At 60 and 760 Torr, the evaporation rate is low and solid or thick-walled particles form (Figures 6a and 6d).

[FIGURE 6 OMITTED]

Figure 7 shows SEM images of the powders produced from an aqueous solution of magnesium sulphate at the reactor top temperature of 130[degrees]C and at four different pressures of 760, 400, 250, and 60 Torr. Figures 6i and 6ii correspond to an initial solution concentration of 1.0 M, and Figures 6iii and 6iv correspond to an initial solution concentration of 0.7 M. In these cases the operating temperature was reduced substantially, in order to reduce the solvent evaporation rate. Figure 7 also provides the droplet evaporation rates at different conditions. Obviously, the evaporation rate at this case ([T.sub.TOP] = 130[degrees]C) is lower than the case in which [T.sub.TOP] = 420[degrees]C. No substantial change in the evaporation rate of droplets is observed with varying the pressures ([T.sub.BOTTOM] also changes). At this condition, the evaporation process is slow, which is favourable to volumetric precipitation and solid particle formation. In addition, the phenomenon of pressure build-up inside the droplet is absent and the particles do not burst during the process. Low evaporation rate results in developing thicker-walled particles or even the formation of solid particles. In Figure 7a, since the evaporation rate is low, the particles are largely solid. In Figure 7b, some of the particles are deformed due to the condensation of the entrapped vapour and particle contraction. In Figures 7c and 7d, again no disrupted particle is observed.

[FIGURE 7 OMITTED]

Comparison of images shown in Figures 6 and 7, indicate that regardless of the pressure and temperature, the powders can be ranked with respect to their corresponding evaporation rates. Table 1 provides the range of the evaporation rates in which the powders are solid, thick-walled shells or thin-walled shells.

CONCLUSIONS

The effect of reduced pressures on the morphology of spray dried magnesium sulphate powders was investigated experimentally. A vibrating mesh nebulizer was employed to generate a spray of small size droplets with an average size of 7.4 [micro]m. Aqueous solutions of magnesium sulphate at 0.7 M and 1.0 M were used. The following conclusions are made:

1. Reducing the reactor air pressure results in an increase in the evaporation rate of large droplets (~100[micro]m). However, due to the non-continuum effects, evaporation rate of small droplets (~1[micro]m) decreases as the pressure is reduced.

2. Whether the powders to be obtained from spray drying or pyrolysis will be hollow or solid, depends on the solute concentration distribution within the droplet at the onset of precipitation. Solute concentration distribution is a function of droplet evaporation rate.

3. Regardless of the reactor pressure and temperature, at low evaporation rates (~4 x [10.sup.-12] kg/s) the magnesium sulphate powders produced from 7.4 [micro]m droplets were solid. Increasing the evaporation rate results in hollow powder formation. At evaporation rate of 13 x [10.sup.-12] kg/s the powders were hollow with thin walls.

4. The particles may burst easier in low pressures than in the atmospheric pressure, as the pressure difference between the vapour entrapped in the drying particle and the surrounding gas is higher in low pressures. However, no extraordinary particle disruption due to pressure build-up was observed.

5. For powder synthesis by spray pyrolysis in which a relatively high temperature for the precursor decomposition is needed, reducing the operating pressure and, consequently the evaporation rate is a good means to produce solid spray pyrolyzed powders.

ACKNOWLEDGEMENTS

The authors are grateful to the Natural Sciences and Engineering Council of Canada for partial support of this research. The first author acknowledges the financial support of the Ministry of Science, Research and Technology of Iran.

REFERENCES

Dhand, R., "Nebulizers that Use a Vibrating Mesh or Plate with Multiple Apertures to Generate Aerosol," Respiration Care 47, 406-416 (2002).

Eslamian, M., M. Ahmed and N. Ashgriz, "Modeling of Nanoparticle Formation during Spray Pyrolysis," Nanotechnology 17, 1674-1685 (2006).

Eslamian, M. and N. Ashgriz, "The Effect of Atomization Method on the Morphology of Spray Dried Particles," Proc. ASME, 2005 Int. Mechanical Engineering Congr. and Exhibition (Orlando, FL, Nov. 2005) paper number IMECE2005-81896 (2005).

Gombosi, T. I., "Gaskinetic Theory," Cambridge University Press (1994), Chap. 5.

Janackovic, D., V. Jokanovic, L. Kostic-Gvozdenovic, S. Zec and D. Uskokovic, "Synthesis and Formation Mechanism of Submicrometer Spherical Cordierite Powders by Ultrasonic Spray Pyrolysis," J. Mater. Sci. 32, 163-168 (1997).

Janackovic, D., V. Jokanovic, L. Kostic-Gvozdenovic and D. Uskokovic, "Synthesis, Morphology, and Formation Mechanism of Mullite Particles Produced by Ultrasonic Spray Pyrolysis," J. Mater. Res. 11(7), 1706-1716 (1996).

Jayanthi, G. V., S. C. Zhang and G. L. Messing, "Modeling of Solid Particle Formation During Solution Aerosol Thermolysis," Aerosol Science and Technology 19, 478-490 (1993).

Kang, Y. C. and S. B. Park, "A High Volume Spray Aerosol Generator Producing Small Droplets for Low Pressure Applications," J. Aerosol Sci. 26, 1131-1138 (1995).

Lenggoro, I. W., T. T. Hata, F. Iskandar, M. M. Lunden and K. Okuyama, "An Experimental and Modeling Investigation of Particle Production by Spray Pyrolysis Using a Laminar Flow Aerosol Reactor," J. Mater. Res. 15 (3), 733-743 (2000).

Lenggoro, I. W., Y. Itoh, K. Okuyama and T. O. Kim, "Nanoparticles of a Doped Oxide Phosphor Prepared by Direct-Spray Pyrolysis," J. Mater. Res. 19(12), 3534-3539 (2004).

Li, W. and E. J. Davis, "Aerosol Evaporation in the Transition Regime," Aerosol Sci. Technol. 25, 11-21 (1996).

Loyalka, S. K., "Mechanics of Aaerosols in Nuclear Reactor Safety: A Review," Prog. Nucl. Energy 12(1), 1-56 (1983).

Mueller, R., R. Jossen, S. E. Pratsinis, M. Watson and M. K. Akhtar, "Zirconia Nanoparticles Made in Spray Flames at High Production Rates," J. Am. Ceram. Soc. 87(2), 197-202 (2004).

Murugavel, P., M. Kalaiselvam, A. R. Raju and C. N. R. Rao, "Sub-Micrometer Spherical Particles of Ti[O.sub.2], Zr[O.sub.2], and PZT by Nebulized Spray Pyrolysis of Metal-Organic Precursors," J. Mater. Chem. 7(8), 1433-1438 (1997).

Miao, O., W. Balachandran and P. Xia, "Characterization of Zr[O.sub.2] and SiC Ceramic Thin Films Prepared by Electrostatic Atomization," J. Mater. Sci. 36, 2925-2930 (2001).

Nimmo, W., D. Hind, N. J. Ali, S. Hampartsoumian and S. J. Milne, "The Production of Ultrafine Zirconium Oxide Powders by Spray Pyrolysis," J. Mater. Sci. 37, 3381-3387 (2002).

Sydel, P., A. Sengespeick, J. Blomer and J. Bertling, "Experimental and Mathematical of Solid Formation at Spray Drying," Chem. Eng. Technol. 27(5), 505-510, (2004)

Tsai, S. C., Y. L. Song, C. S. Tsai, C. C. Yang, W. Y. Chiu and H. M. Lin, "Ultrasonic Spray Pyrolysis for Nanoparticles Synthesis," J. Mater. Sci. 39, 3647-3657 (2004).

* Author to whom correspondence may be addressed. E-mail address: ashgriz@mie.toronto.edu

Manuscript received October 26, 2005; revised manuscript received May 11, 2006; accepted for publication May 12, 2006.

Morteza Eslamian and Nasser Ashgriz *

Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON, Canada M5S 3G8
Table 1. Mg[SO.sub.4] powder morphology versus droplet evaporation rate

Evaporation rate x [10.sup.12] (kg/s/) Powder morphology

3.54-5.78 Solid
8.46 Solid or thick-walled shell
13.3-16.5 Thin-walled shell
COPYRIGHT 2006 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Eslamian, Morteza; Ashgriz, Nasser
Publication:Canadian Journal of Chemical Engineering
Date:Oct 1, 2006
Words:5134
Previous Article:Direct numerical simulations of three-layer viscosity-stratified flow.
Next Article:Analytic model of laminar-turbulent transition for Bingham plastics.
Topics:

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters