Technical Paper

Intl J. of Aeronautical & Space Sci. 12(3), 296304 (2011) DOI:10.5139/IJASS.2011.12.3.296

Effect of Particle Loading Ratio and Orifice Exit Velocity on a ParticleLaden Jet
Kyong-Yup Paik1, Jungsoo Yoon1, Jeongjae Hwang1, Jae Mook Chung1, Nicolas Bouvet2 and Youngbin Yoon*
School of Mechanical and Aerospace Engineering, Seoul National University, Seoul 151-742, Korea

Abstract
In order to design a shear coaxial injector of solid particles with water, basic experiments on a particle laden jet are necessary. The purpose of the present study is to understand the effect of particle loading ratio on the particle spray characteristics (i.e. spreading angle, distribution of particle number density, velocity profiles, and particle developing region length). Hydro-reactive Al2O3 particles with a primary particle diameter of 35~50 m are used in this experiment. An automated particle feeder was designed to supply constant particle mass flowrates. Air is used as the carrier gas. To determine the air velocity at the orifice exit, tracers (aluminum oxide, 0.5~2 m primary diameter) are also supplied by a tracer feeder. A plain orifice type injector with 3 mm diameter, and 20 mm length was adopted. Particle image velocimetry is used to measure the mean and fluctuating velocity components along the axial and radial directions. Key words:  Particle-laden jet, Two-phase jet, Particle loading ratio, Spreading angle, Particle number density, Particle developing region, Turbulent intensity

Nomenclature
D L Ldeveloping ma mp Nc Nr r r1/2 Uc U o Ur Vr u x a a orifice exit diameter orifice length length of the particle developing region air mass flow rate particle mass flow rate particle number density along the centerline particle number density along the radial direction radial distance radial distance at 50 % velocity of centerline velocity mean axial velocity of centerline mean axial air velocity of orifice exit mean axial velocity along the radial direction mean raial velocity along the radial direction rms velocity fluctuation distance from the orifice exit particle loading ratio, (= mp / ma) kinetic viscosity coefficient of air air density

1 Graduated student, School of Mechanical and Aerospace Engineering 2 Post doctoral fellow, School of Mechanical and Aerospace Engineering * Corresponding author. Professor, School of Mechanical and Aerospace Engineering * E-mail address: ybyoon@snu.ac.kr

Received: July 19, 2011 Accepted: September 14, 2011 Copyright The Korean Society for Aeronautical & Space Sciences

296

http://ijass.org pISSN: 2093-274x eISSN: 2093-2480

Kyong-Yup Paik1 Effect of Particle Loading Ratio and Orifice Exit Velocity on a Particle-Laden Jet

1. Introduction
Supercavitation underwater vehicles use rocket propulsion systems with combustion of hydro-reactive metal fuel unlike other conventional underwater vehicles using the propeller propulsion systems. When a vehicle moves under the water, cavitation occurs on the body surface due to velocity difference between the body and the water. The cavitation on the surface develops rapidly with increasing velocity, and if this cavitation becomes supercavitation it is enough to wrap the whole body, subsequently the underwater vehicle escapes from the high drag force of water, and can potentially reach high speeds over 400 km/h. Moreover, a supercavitation underwater vehicle using a rocket propulsion system has a higher propulsive efficiency per fuel unit volume as compared to conventional underwater vehicles, because the surrounding water can directly serve as the oxidizer. Aluminum (Al2O3) and magnesium (MgO2) are mostly used as hydro-reactive metal fuels. The combustion mechanism differs from other conventional particles (for example, coal or gunpowder). Indeed, oxide films covering the metal surface under normal state should be melted prior to the reaction with H2O. A uniform distribution of the metal particles in the jet helps their total surface to be larger, so the oxide films are able to melt easier. In the combustion reaction of water and hydro-reactive particles, Foote et al. (1996) reported that uniform mixing performance of water and the hydro-reactive particles causes a decrease of the ignition temperature and

increase of the flame propagation speed, which indicate an increase in the specific impulse. Therefore, in order to design shear coaxial injectors that are able to uniformly spray water and particles, basic experiments on spray characteristics of the solid particles with the fluidization medium are required. Investigations about particle injection have been actively carried out to develop coal combustors, which are part of the integrated gasification combined cycle plants since 1980. Modarress et al. (1984) investigated the effect of particle size, jet exit velocity, and loading ratio for an axisymmetric turbulent jet by measuring mean velocities, turbulent intensity, and shear stress. With increase of a particle loading ratio, a linear increase of jet center velocity at x/D = 20, a reduction of particle spreading angle, and a decrease of turbulent intensity are observed. Fleckhaus et al. (1987) examined the effect of the particle primary diameter on a particle-laden jet. They also fitted their experimental results with a set of analytical functions for improving the basic understanding of two phase flow phenomena. They concluded that the decrease of the particle primary diameter causes a reduction of the air velocity decay rate and the particle spreading angle, due to a decrease in the turbulent intensity. Gillandt et al. (2001) investigated the effects of influence turbulence modulation by analyzing the one-dimensional energy spectra of the axial velocity fluctuations for single and two phase jet flows. The addition of the particles to the jet causes turbulence intensity decrease because turbulence energy in the jet is dissipated by

Table 1. Experimental conditions of previous works

Author Popper et al. (1974) Levy and Lockwood (1981) (Modarress et al. (1984) Shuen et al. (1985) Fleckhaus et al. (1987) Tsuji et al. (1988) Barlow and Morrison (1990) Jou et al. (1993) Gillandt et al. (2001) Present Study Orifice Diameter 25 mm 15 mm 20 mm 10.9 mm 13 mm 20 mm 10 mm 15 mm 12 mm 3 mm Orifice length 600 mm 1,800 mm 981 mm 1,000 mm 2,058 mm 2,025 mm 1,080 mm 20 mm Orifice exit Re 16,450~49,450 20,000 13,300, 14,100 19,000 20,000 14,500~34,300 780 20,000 5,700 23,665~66,264 Particle mean Diameter <50 m 180~1,200 m 50, 200 m 79~207 m 64, 132 m 170~1,400 m 163 m 210, 780 m 110 m 35~50 m Particle density Oil Sand 2,990 kg/m3 glass beads 2,620 kg/m3 sand 2,590 kg/m3 glass beads 1,020 kg/m3 polystyrene 2,825 kg/m3 glass beads 1,020 kg/m3 polystyrene Glass beads 3,970 kg/m3 Aluminum oxide Loading ratio 0.001 1.14~3.5 0.8, 0.85 0.2, 0.66 0.3 0.39~1.85 6.5~54 0.49~3.06 1 0.116~0.39

297

http://ijass.org

Intl J. of Aeronautical & Space Sci. 12(3), 296304 (2011)

friction with the particle surface (Gillandt et al., 2001). Thus, it was found that the turbulent intensity is decreased with increasing total particle surface in the jet. Decreasing the turbulent intensity is related to diminishing the momentum exchange between the jet and the surrounding gas. As such, the decrease of the turbulent intensity causes the change in the jet velocity profile. The centerline velocity decay of the two phase jet is slower than that of a single phase jet, and the radial velocity profile of the two phase jet is steeper than that of the single phase jet (Laats, 1969). Additionally, the slip velocity is lessened by lowering decay rate of the gas phase due to diminishing the momentum exchange (Modarress et al., 1984). Experiment conditions of previous works are listed in Table 1. As shown in Table 1, all previous research adopted long orifice lengths from 600 mm to 2,025 mm, and large orifice diameters from 10 mm to 25 mm. Due to the limitation of space in the underwater vehicle, the orifice length and diameter should be downsized as compared with other investigations. In this study, with particle image velocimetry (PIV) measurement, the profiles of particles velocity, turbulent intensity, and number density of particles are investigated to know the effect of the loading ratio on the particle developing region and the particle spreading angle of the two phase jet using the plain orifice type injector.

2. Experimental Arrangement
The experiment setup is shown in Fig. 1. Air is used as the carrier gas, and the air flow rate is measured by a mass flow meter (M-1500SLPM-D; Alicat Scientific, Tucson, AZ, USA), and manually controlled with a gas regulator. A pressure sensor (PTAF0020KAA; Sensys, Ansan, Korea) is used for a secondary measurement to check the air flow

rate. Particles of primary size 35~50 m are supplied to the pipeline via a particle feeder using an automated feeding screw. To determine the orifice exit velocity, tracer particles with a primary size of 0.5~2 m are respectively provided in the pipeline by a tracer feeder. A damping chamber is installed on the experimental rig to eliminate the pulsation of the particle flow rate. The injector shown in Fig. 2 is placed on a vertical traverse to enable vertical adjustment. The experimental rig is sealed with transparent polycarbonate covers with the ceiling open, and the rig size designed in order to avoid the boundary effect of 1,000 1,000 1,500. A PIV measurement is used to measure the air and particles velocities, respectively. The laser system used in the experiment is a dual head PIV Nd-YAG laser (Surelite II-PIV). A pulsed laser sheet beam with a 10 Hz pulse frequency, made by a digital pulse generator (DG535; Stanford Research Systems, Sunnyvale, CA, USA) and optical lenses, passed vertically through the jet center, and particle scattering images are captured by a charge-coupled device (CCD) camera (MegaPlus ES2020; Princeton Instruments, Trenton, NJ, USA) with a 1,200 1,600 size. In each case, 300 pairs of CCD images are acquired to obtain averaged velocity vector fields. For the CCD image processing, in-house PIV code is used (Hart, 1998). Figure 3 shows an automated particle feeder designed to supply constant particle mass flow rates. A cone structure inside of the particle feeder is sealed with a polycarbonate cover and pressurized to the same pressure using a tube where the air passes through to prevent a backward flow of particles. A screw, which is put inside the particle feeder, is rotated by an AC variable speed motor attached to an end of the screw. When the screw spins, particles loaded between the screw`s pitch are transported to an air line. The particles fluidized by the air are then sprayed through the injector orifice. In order to get particle flow rates for each feeding

Fig. 1. Experimental apparatus.

DOI:10.5139/IJASS.2011.12.3.296

298

Kyong-Yup Paik1 Effect of Particle Loading Ratio and Orifice Exit Velocity on a Particle-Laden Jet

condition, particles sprayed through the orifice are collected into a flask filled with sticky syrup for a certain period of time, and then mass changes are measured by a load cell. Figure 4 shows averaged particle flow rates. In Fig. 4, the Direct condition means the particle supply rates of the particles coming out from the particle feeder exit are collected on the load cell and measured directly. Motor rpm levels, which represent gradations on the motor control panel, are arbitrary experimental conditions to get the constant particle flow rates. Direct points grow linearly with increasing motor rpm

levels in Fig. 4. As such, it can be seen that the rotational speed of the screw increases linearly with increasing the motor rpm level step by step. Comparing Direct with other particle mass flow rates measured from the orifice exit, all feeding conditions are well supplied without any accumulation in the line except for the condition of air flow rate 1.184 g/s and the highest rpm condition. In the condition of air flow rate 1.184 g/s, the air velocity is not enough to carry the particles, so particles are believed to deposit in the damping chamber. The particle mass loading ratio is the ratio of particle mass flow rate to the gas mass flow rate. The mass loading ratios obtained from the experimental conditions are shown in Table 2. More information about the experiment is listed in Table 3.

3. Results and Discussion

Fig. 2. Plain orifice type injector.

Fig. 2. Plain orifice injector. Fig. type 2. Plain orifice

type injector.

In this experiment, the particle-laden jet shows the particle developing region, which is defined as a zone from the orifice exit to the point where particle velocity reaches maximum, as shown in Fig. 5. If the orifice length is short, the particles accelerated by drag force from the orifice entrance are not able to reach the air exit velocity due to their inertia, and so the particle developing region appears. After the particle developing region, the particle velocity dwindles, and the turbulent intensity increases. It is considered that
Table 2. Loading ratios
Rpm levels Air flow rate (g/s) 1.184 1 0.116 0.216 0.178 0.138 2 0.139 0.287 0.227 0.179 3 0.195 0.340 0.248 0.232 4 0.227 0.386 0.275 0.280

Fig. 3. Automated particle feeder.

Fig. 3. Automated particle feeder.

1.579 1.973 2.368

Table 2. Loading ratios

Table 3. Experimental conditions
Air flow rate (g/s) Rpm levels

1 0.116Fig. 0.216 0.178 0.138

4 0.227 0.386 0.275 0.280

Air flow rate (g/s) Air temperature () Particle feeding rate (g/s) Orifice diameter (mm) Orifice length (mm) Nd-YAG Laser pulse freq. (Hz) Beam delay (s)
3

1.184 1.579 1.973 2.368

3. Automated 0.139 particle feeder. 0.195

0.287 0.227 0.179

1.184, 1.579, 1.973, 2.368 18.5 0.137~0.664 3.0 20 10 4 Al2O3, 3.97 35~50 0.5~2

0.340 0.248 0.232

Table 2. Loading ratios

Particle and Tracer density (g/cm ) Rpm levels Fig. 4. feeding rate test (data points represents an average of direct: particle 4 1Particlerate 2 represents Fig. 4. Particle feeding test (data points an average of3 10 runs; Particle primary diameter (m) runs; direct: particle collected at the feeder exit, other symcollected at10 the feeder exit, other symbols: particle collected at the orifice exit). rate (g/s)

1.184 1.579 1.973 2.368

0.116 0.139 0.195 0.216 0.287 0.340 Table 3. Experimental conditions 0.178 0.227 0.248 1.184, 1.579, 0.138 0.179 0.232 Air flow rate (g/s)
Air temperature () Particle feeding rate (g/s) Orifice diameter (mm) Orifice length (mm) Nd-YAG Laser pulse freq. (Hz) Beam delay (s) 1.973, 2.368 18.5 0.137~0.664 3.0 20 10 4

bols: particle collected at the orifice exit).

Tracer primary diameter (m)

299

0.227 0.386 0.275 0.280

http://ijass.org

Intl J. of Aeronautical & Space Sci. 12(3), 296304 (2011)

with particles at once to measure the air velocity, the tracer could have an effect on the behavior of particles. According to Modarress et al. (1984), the velocity of air at the orifice exit is the same regardless the presence of particles in the jet. For this experiment, although particles and tracer are sprayed separately, it can be considered that the air velocities of both, with and without particles at the orifice exit, are similar because of similar spray conditions. Therefore, the centerline velocity profiles of both are normalized by measured air velocity at the orifice exit. In Fig. 6, the velocity of particles is lower than that of air at the orifice exit. From the entrance of the orifice, with acceleration of air, particles are accelerated by the drag force, Fig. 5. The particle developing region. but the velocity of particles does not reach that of the air at Fig. 5. The particle developing region. the orifice exit due to collision between the particles and Fig. 5. The particle developing region. the orifice wall, inter-particle collision, inertia of particles, and insufficient orifice length to accelerate. After coming out of the orifice, the air momentum is transferred to the particles and surrounding air, and so its velocity rapidly decreases. Meanwhile, particles are accelerated by receiving momentum until they reach the end of a particle developing region, where particle velocity is at maximum (Park and Chen, 1989). After the particle developing region, air keeps the steep reduction of the velocity due to the momentum transfer to the surrounding air, while particles show a gradual reduction of the velocity because of their inertia. Therefore, the decay rates of the velocities of air and particles are different right after the particle developing region. As shown in Fig. 6, it has been observed that turbulent intensities rapidly increase from the point where velocities decrease for both air and particles. Especially, in the case of particles, Fig. 6.  Comparisons of profiles of velocity and turbulent intensity beFig. 6. Comparisons of profiles of velocity turbulent intensity between particle and air the turbulent intensity starts to increase after the particle tween particle and air (Uo and = 152.8 m/s, = 0.340). (Uo = 152.8 m/s, = 0.340). developing region. It seems that velocity decreases and Fig. 6. Comparisons of profiles of velocity and turbulent intensity between particle and air increases after the particle developing turbulent intensity gure 6 represents thethe velocity and turbulent intensity profiles of air and particles along the jet m/s, = with 0.340). particles (U exchange momentum surrounding gas region because the entrainment of surrounding air totally o = 152.8 terline. The turbulententrained intensity is defined as the ratio of the root-mean-square of the velocity in the jet center after the particle developing region. permeates the jet center after that region. gure 6 represents the velocity andvelocity. turbulent intensity of air andby particles along the jet velocity profiles of particle and air tuation, |u`|, to the mean stream profiles aredeveloping obtained theFigure injection ofnormal the From the orifice exit, the These longer the profiles particle 7 shows terline. The turbulent intensity is profiles defined as the ratio of actively the root-mean-square of the velocity icles and tracers respectively, of air and particle are independent of each other. The for different orifice exit velocities region, theso farther a spot where particles diffuse. along the radial direction tuation, |u`|, to the mean stream velocity. profiles are obtained by the of the Thus, it is believed that the These length of the particle and loading ratios. Normal velocities of radial direction on that particles and tracer are not sprayed simultaneously is developing to focus only on injection the behavior of region contains information of the particle diffusivity along are normalized by the icles and tracers respectively, so profiles of air and particle are independent of each other. The icles. If the tracer was sprayed with particles at once to measure the air velocity, the tracer could centerline axial velocity, and radial the axial are distance the jet on particles tracer not sprayed simultaneously is to focus onlythe on velocity the from behavior of e anthat effect on the and behavior ofdirection. particles. According to Modarress et al. (1984), of air atcenter is normalized by r1/2 , which Figure 6 represents the velocity and turbulent intensity is full width at half maximum of the velocity profile. In Fig. icles. If theis tracer was regardless sprayed with to measure theFor air this velocity, the tracer could orifice exit the same the particles presenceat ofonce particles in the jet. experiment, although profiles of air and particles along the jet centerline. The 7a, the velocity profiles of the air display self-similarity, e an effect on the behavior of particles. it According to Modarress et al. the of velocity of air at icles and tracer are sprayed separately, can be considered that the air(1984), velocities both, with and turbulent intensity is defined as the ratio of the root-meanregardless of the orifice exit velocity in the fully developed orifice exit is the same regardless the presence of particles in the jet. Forconditions. this experiment, although hout particles at the orifice exit, are similar because of spray square of the velocity fluctuation, |u`|, tosimilar the mean stream regionTherefore, where x/D the is higher than 10. The velocity profiles icles and tracer are sprayed separately, it can be considered that the air velocities of both, with andin Figs. 7b and c, are not self-similar terline velocity profiles of bothThese are normalized measured air injection velocity of at the orifice exit.as velocity. profiles are by obtained by the of particles, shown hout at theof orifice exit, are similar because similar spray conditions. Therefore, Fig. particles 6, the velocity particles isand lower than that of airof at profiles the orifice exit. From thedifferent entrance of the the particles tracers respectively, so of air and with loading ratios. In both graphs, the condition of terline velocity profiles both are normalized measured air reason velocity at the exit. particle are independent ofby each other. higher loading ratio shows a steeper velocity profile than the ice, with acceleration ofof air, particles are accelerated by theThe drag force,that but theorifice velocity of particles and are not sprayed is to exit. focus other. A steeper velocity Fig. the velocity ofparticles particles istracer lower than that to ofsimultaneously air at the between orifice the entrance of the profile shows that the momentum s not6, reach that of the air at the orifice exit due collision the From particles and the orifice only on the behavior of particles. If the tracer was sprayed exchange between the jet and the surrounding air is slow ice, with acceleration of air, particles are accelerated by the drag force, but the velocity of particles

s not reach that of the air at the orifice exit due to collision between the particles and the orifice
DOI:10.5139/IJASS.2011.12.3.296

300

ratios. In been both observed graphs, the condition of higher loading ratio shows a steeper velocity pr right after the particle developing region. As shown loading in Fig. 6, it has that turbulent intensities rapidly increase from the point where velocities decrease for both air and particles. than the other. A steeper velocity profile shows that the momentum exchange between the jet an intensities rapidly increase from the point where velocities decrease for both air and particles. Especially, in the case of particles, the turbulent intensity starts to increase after the particle surrounding air is compared theparticle less steep one. If the jet contains more particles, large ed Especially, in the case of particles, the turbulent intensity starts toslow increase after to the developing region. It seems that velocity decreases and turbulent intensity increases afterair, theeasily particle which are produced by entrained break up into small eddies (Gillandt et al., 2001). developing region. It seems that velocity decreases and turbulent intensity increases after the particle developing region because the entrainment of surrounding air totally permeates the jet center after that phenomenon that the the jet momentum exchange to become small, and therefore the velocity pr developing region because the entrainment of surrounding air totally causes permeates center after that region. of the particles becomes steep. region. Kyong-Yup Paik1 Effect of Particle Loading Ratio and Orifice Exit Velocity on a Particle-Laden Jet

(a)

(a) (a)

(a)

(a)

(b)

(b) (b)

(b)

(b)

Fig. 8. The effect of the loading ratio on the two phase jet. (a) Axial profiles of velocity and tur Fig. 8.  The effect of the loading ratio on the two phase jet. (a) Axial intensity. (b) Radial profiles of the particle number density. profiles of velocity and turbulent intensity. (b) Radial profiles of
the particle number density.

Figure 8a shows the changes in the profiles of velocity and turbulent intensity with different l ratios in the same orifice exit velocity. According to Biing-Hwang Jou et al. (1993), the particle mean velocity is almost independent of the particle loading ratio. However, in this experiment been found that the decay rate of the less axial velocity becomes slow asmore the loading ratio increase compared to the steep one. If the jet contains particles, large which are produced byparticle entrained is also related to the eddy break up.eddies, Due to the increase of the loading ratio, large edd air, easily break up into small eddies (Gillandt et is al., 2001). related to the eddy readily split into small eddies, thus the turbulent intensity which strongly This phenomenon causes that the momentum exchange damped, as shown in Fig. 8a. Damping of the turbulent intensity makes it difficult to exchan to become small, and therefore the velocity profile of the momentum between particles and surrounding air, therefore the slowing down of the velocity particles becomes steep. rate can be explained. Radial profiles of the particle number density with different loading rat Figure 8a shows the changes in the profiles of velocity shown in Fig. 8b. Comparing profiles of x/D = 30, the condition of higher loading ratio has a s and turbulent intensity with different loading ratios in the steeper distribution. From this result, it can be concluded that the increase exit velocity. According to Biing-Hwang Jou of the loading ratio (c) same orifice (c) the distribution of particles narrow, owing to the decrease of the momentum et al. (1993), the particles axial mean velocity is almost exchange. = Fig. 7. Axial velocity profiles of particle and air along the radial direction. (a)independent Air. (b) Particles U of the particle loading ratio. However, in this o Fig. 7.  Axial velocity profiles of particle and air along the radial direc114.6 m/s. (c) Particles Uo = 188.7 m/s. experiment, it has been found that the decay rate of the axial tion. (a) Air. (b) Particles Uo = 114.6 m/s. (c) Particles Uo = 188.7
m/s.

Figure 7 shows normal velocity profiles of particle and air along the radial direction for different orifice exit velocities and loading ratios. Normal velocities of radial direction are normalized by the centerline axial velocity, and radial distance from the jet center is normalized by r1/2 , which is full 301 width at half maximum of the velocity profile. In Fig. 7a, the velocity profiles of the air display selfsimilarity, regardless of the orifice exit velocity in the fully developed region where x/D is higher than 10. The velocity profiles of particles, as shown in Figs. 7b and c, are not self-similar with different loading ratios. In both graphs, the condition of higher loading ratio shows a steeper velocity profile than the other. A steeper velocity profile shows that the momentum exchange between the jet and the

velocity becomes slow as the loading ratio increases. This is

http://ijass.org

velocity makes is 167.6 m/s has to much more momentum than the other jet. It seems that the effect mped, as shown in Fig. 8a. Damping of the turbulent intensity it difficult exchange the velocity is greater that of the decay turbulent intensity on the velocity profiles. The effect omentum between particles and surrounding air, therefore the slowing downthan of the velocity turbulent intensity is revealed better in e can be explained. Radial profiles of the particle number density with different loading ratiosthe areprofiles of the particle number density. In Fig. 9b, th condition of exitloading velocity 167.6 m/s shows a much wider particle distribution than the spray co own in Fig. 8b. Comparing profiles of x/D = 30, the condition of higher ratio has a slightly of exit velocity 114.6 m/s at x/D = 30. The increase of the orifice exit velocity produces bigger eeper distribution. From this result, it can be concluded that the increase of the loading ratio makes and particles are widely distributed due to the increase of the momentum transferring. e distribution of particles narrow, owing to the decrease of theso momentum exchange. Intl J. of Aeronautical & Space Sci. 12(3), 296304 (2011)

(a)

(a)

(a)
(a)

(b)

(b)
(b)

(b)

Fig. 9. The effect of orifice exit velocity on the two phase jet injet in =0.227. (a) Axial of Fig. 10. Spreading angles. (a) profiles A comparison of different exit velocity conditions. (b) Consideri Fig. 9.  The effect of orifice exit velocity on the two phase = Fig. 10.  Spreading angles. (a) A comparison of different exit velocity velocity and turbulent intensity. (b) profiles of the particle conditions. 0.227. (a) Axial profiles ofRadial velocity and turbulent intensity. (b) number density. conditions. (b) Considering all conditions.
Radial profiles of the particle number density.

Turbulent intensity is proportional to jet velocity at the orifice exit, because if the velocity gap between the jet and the surrounding air becomes larger as the orifice exit velocity increases, bigger eddies are generated in the jet. Figure 9a shows the effect of the jet velocity at the orifice exit on exchange. also related to the eddy break up. Due to the increase the particles. In this figure, although the turbulent intensity increases, theof velocity profile of the centerline Turbulent is proportional to jet velocity at particle loading ratio, large eddies are readily split into small declines slowly, contrary to the effect of the loading ratio. This is because the jet for intensity which exit the orifice exit, because if the velocity gap between the jet thus the turbulent intensity which is strongly related velocity is 167.6 m/s eddies, has much more momentum than the other jet. It seems that the effect of the and the surrounding air becomes larger as the orifice exit to the eddy size, is damped, as shown in Fig. 8a. Damping velocity is greater than that of the turbulent intensity on the velocity profiles. The effect of the velocity increases, bigger eddies are generated in the jet. of the turbulent intensity makes it difficult to exchange the urbulent intensity is revealed better in the profiles of the particle number density. In Fig. 9b, the spray Figure 9a shows the effect of the jet velocity at the orifice exit momentum between particles and surrounding air, therefore condition of exit velocity 167.6 down m/s shows a much wider particle distribution than the spray on particles. In condition this figure, although the turbulent intensity the slowing of the velocity decay rate can be explained. of exit velocity 114.6 m/s at x/D = 30. The increase of the orifice exit velocity produces bigger eddies, increases, the velocity profile of the centerline declines Radial profiles of the particle number density with different and so particles are widely distributed due to in the increase of the momentum slowly, contrary to the effect of the loading ratio. This is loading ratios are shown Fig. 8b. Comparing profiles of transferring.
x/D = 30, the condition of higher loading ratio has a slightly steeper distribution. From this result, it can be concluded that the increase of the loading ratio makes the distribution of particles narrow, owing to the decrease of the momentum because the jet for which exit velocity is 167.6 m/s has much more momentum than the other jet. It seems that the effect of the velocity is greater than that of the turbulent intensity on the velocity profiles. The effect of the turbulent intensity is

Fig. 11. Velocity ratios with different loading ratios along the radial direction. (x/D = 30, Uo = m/s)
DOI:10.5139/IJASS.2011.12.3.296

Figure 10 shows spreading angles with different spray conditions. The spreading angles are ob from a distance between D/2 and r1/2 at x/D = 30. In Fig. 10a, spreading angles are reduce the increasing the orifice exit velocity, and are shown to be parabolic curves as the loadin increases. These trends correspond well to the trend of Fig. 10b. Decreasing of the spreading an the loading ratio increases can be explained by the reducing of the momentum exchange, b

302

(b)

ig. 10. Spreading angles. (a) A comparison of different exit velocity conditions. (b) Considering all conditions.

Kyong-Yup Paik1 Effect of Particle Loading Ratio and Orifice Exit Velocity on a Particle-Laden Jet

Fig. 12.  The length of the developing region considering all condi= the 188.7 Fig. 11. Velocity ratios with direction. different loading along the radial direction. (x/D =length 30, Uoof Fig. 12. The developing region considering all conditions. (x/D = 30, Uo ratios = 188.7 m/s) tions. m/s) The result of changes in the length of the developing region considering all spray conditio igure 10 shows spreading angles with different spray conditions. The spreading are obtained shown in Fig. 12. In the angles experimental conditions, the lengths of the developing region normaliz at x/D = 30. In Fig. 10a, spreading angles are reduced m a distance betweenrevealed D/2 better and in r1/2 the range ofwith 9 to 14. This result the effects the profiles of the particle numberdiameter density. are within the orifice distributed within the range of 9 means to 14. that This result means that the effects o e increasing the orifice exit velocity, and are of shown to be parabolic curves as the loading ratio of the loading ratio and the orifice exit velocity on the length In Fig. 9b, the spray condition exit velocity 167.6 m/s shows loading ratio and the orifice exit velocity on the length of the developing region are ambiguous creases. These trends correspond well to the trend ofthan Fig. 10b. Decreasing ofthe the spreading angle as are oflength the developing region ambiguous,region and so it isa expected a much wider particle distribution the condition so it spray is expected that of the particle developing is function of the orifice geom that the length of thethe particle developing region is a function of exit velocity 114.6 m/s by at x/D 30. The increase the e loading ratio increases can be explained the =reducing of the of momentum exchange, but of the orifice geometry. orifice exit velocity produces bigger particles ndency of the spreading angle to increase again iseddies, seen, and as so the radial velocity components of are widely distributed due to the increase of the momentum rticles increase as shown in Fig. 11. This is because, when the particle number density increases in transferring. 4. Conclusions e jet, inter-particle collisions become more intensified, making the radial velocity components 4. Conclusions Figure 10 shows spreading angles with different spray larged. It is considered that the increase of axial velocity An component and the decrease of the particle experiment conditions. The spreading angles are obtained from a of solid particle spray is conducted for designing a proper injector o mber density cause the diminution of the spreading angle with the increase of the orifice exit An vehicle experiment of solid particle spray type is conducted distance between D/2 and r1/2 at x/D = 30. In Fig. supercavitation underwater by using a plain orifice injector. In a condition of o locity. From Fig. 10b,10a, it isspreading predicted that the spreading angle increases again with a further increase for designing design a proper injector the supercavitation angles are reduced with the increasing length 20 mm, the which is a required criteria, theof effect of particle loading ratio and orific the orifice exit velocity. underwater by intensity, using a plain type injector.density are investigat orifice exit velocity, and are shown tovelocity be parabolic curves on the particles velocity, vehicle turbulent andorifice particle number

Fig. 11.  Velocity ratios with different loading ratios along the radial

In a condition of orifice length 20 mm, which is a required as the loading ratio increases. These trends correspond means of PIV measurement. Further, the particle spreading angle, and the length of pa design criteria, the effect of particle loading ratio and orifice well to the trend of Fig. 10b. Decreasing of the spreading developing region, which have information of particle distribution in the jet, are examined. exit velocity on the particles velocity, turbulent intensity, and angle as the loading ratio increases can be explained by the particle number density are investigated by means of PIV reducing of the momentum exchange, but the tendency of 1. Axial velocity profiles of particle along the radial direction are not self-similar with diff measurement. Further, the particle spreading angle, and the the spreading angle to increase again is seen, as the radial particle loading ratios, and the profiles of higher loading ratios become steeper. length of particle developing region, which have information velocity components of particles increase as shown in 2. As the particle loading ratio increases, decay rate of axial particle velocity pr of particle distribution in the jet,the are examined. Fig. 11. This is because, when the particle number density becomes turbulent intensity profiles are the reduced. Axial velocity profiles of particle along radial direction increases in the jet, inter-particle collisions become more slow, and are not self-similar with different particle loading and particle velocity pr intensified, making the radial velocity components enlarged. 3. As the orifice exit velocity increases, the decay rate ratios, of axial the profiles of higher loading ratios become steeper. It is considered that the increase of axial velocity component becomes slow, and turbulent intensity profiles are increased. As the particle loading ratio increases, the decay rate of and the decrease of the particle number density cause the Spreading reduced with the increasing the and orifice exit velocity, and are show axialare particle velocity profiles becomes slow, turbulent diminution of the spreading angle with the 4. increase of the angles be parabolic curves as the loading ratio increases. intensity profiles are reduced. orifice exit velocity. From Fig. 10b, it is predicted that the As the orifice exit velocity increases, the decay rate of spreading angle increases again with a further increase of 5. The lengths of particle developing regions are within the range of 9 to 14 in the exper the orifice exit velocity. conditions. axial particle velocity profiles becomes slow, and turbulent intensity profiles are increased. The result of changes in the length of the developing Spreading angles are reduced with the increasing the region considering all spray conditions is shown in Fig. 12. In orifice exit velocity, and are shown to be parabolic curves as the experimental conditions, the lengths of the developing the loading ratio increases. region normalized by the orifice diameter are distributed

303

http://ijass.org

Intl J. of Aeronautical & Space Sci. 12(3), 296304 (2011)

The lengths of particle developing regions are within the range of 9 to 14 in the experiment conditions.

Acknowledgements
This work has been supported by Agency for Defense Development (09-01-05-24), and Mid-career Researcher Program through NRF grant funded by the MEST (20110016455) contracted through the Institute of Advanced Aerospace Technology at Seoul National University.

References
Barlow, R. S. and Morrison, C. Q. (1990). Two-phase velocity measurements in dense particle-laden jets. Experiments in Fluids, 9, 93-104. Fleckhaus, D., Hishida, K., and Maeda, M. (1987). Effect of laden solid particles on the turbulent flow structure of a round free jet. Experiments in Fluids, 5, 323-333. Foote, J. P., Lineberry, J. T., Thompson, B. R., and Winkelman, B. C. (1996). Investigation of aluminum particle combustion for underwater propulsion applications. Proceedings of the 32nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Lake Buena Vista, FL. Gillandt, I., Fritsching, U., and Bauckhage, K. (2001). Measurement of phase interaction in dispersed gas/particle two-phase flow. International Journal of Multiphase Flow,

27, 1313-1332. Hart, D. P. (1998). The elimination of correlation errors in PIV processing. Proceedings of the 9th International Symposium on Applications of Laser Techniques to Fluid Mechanics, Lisbon, Portugal. Jou, B. H., Sheen, H. J., and Lee, Y. T. (1993). Particle mass loading effect on a two-phase turbulent downward jet flow. Particle and Particle Systems Characterization, 10, 173-181. Laats, M. K. (1969). Experimental study of the dynamics of an air-dust jet. Journal of Engineering Physics, 10, 6-8. Levy, Y. and Lockwood, F. C. (1981). Velocity measurements in a particle laden turbulent free jet. Combustion and Flame, 40, 333-339. Modarress, D., Tan, H., and Elghobashi, S. (1984). Twocomponent lda measurement in a two-phase turbulent jet. AIAA Journal, 22, 624-630. Park, C. J. and Chen, L. D. (1989). Experimental investigation of confined turbulent jets. Part II. Particleladen flow data. AIAA Journal, 27, 1511-1516. Popper, J., Abuaf, N., and Hetsroni, G. (1974). Velocity measurements in a two-phase turbulent jet. International Journal of Multiphase Flow, 1, 715-726. Shuen, J. S., Solomon, A. S. P., Zhang, Q. F., and Faeth, G. M. (1985). Structure of particle-laden jets: measurements and predictions. AIAA journal, 23, 396-404. Tsuji, Y., Morikawa, Y., Tanaka, T., Karimine, K., and Nishida, S. (1988). Measurement of an axisymmetric jet laden with coarse particles. International Journal of Multiphase Flow, 14, 565-574.

DOI:10.5139/IJASS.2011.12.3.296

304