- Email: [email protected]
Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor
APT 2445
No. of Pages 7, Model 5G
11 October 2019 Advanced Powder Technology xxx (xxxx) xxx 1
Contents lists available at ScienceDirect
Advanced Powder Technology journal homepage: www.elsevier.com/locate/apt
2
Original Research Paper
6 4 7 5
Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor
8
Muhammad Zeeshan Zahir a, Ji-Eun Heo b, Se-Jin Yook b,⇑
9 10 12 11 13 1 2 5 9 16 17 18 19 20 21 22 23 24 25 26 27 28
a b
Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, Khyber Pakhtunkhwa, Pakistan School of Mechanical Engineering, Hanyang University, Seoul 04763, Republic of Korea
a r t i c l e
i n f o
Article history: Received 12 January 2019 Received in revised form 24 September 2019 Accepted 26 September 2019 Available online xxxx Keywords: Aerosol Inertial separation Virtual impactor Collection efficiency Wall loss
a b s t r a c t A two-partitioned horizontal inlet was developed for improving the collection efficiency and minimizing the wall loss problem in slit virtual impactor. The two-partitions were provided to simultaneously supply both aerosol and clean air to the virtual impactor. Both numerical and experimental investigations were carried out on the developed inlet configuration by considering different flowrate ratios of aerosol to clean air. The horizontal inlet was helpful in reducing the cutoff diameter, whereas the clean air prevented the particle deposition on the virtual impactor walls. The performance of two-partitioned horizontal inlet was compared with the conventional vertical inlet configuration for PM2.5, PM5 and PM10 virtual impactors. All the operating conditions and geometric parameters, such as the inlet flowrate; the width of collection nozzle; the width, length and span of acceleration nozzle; and the distance between collection and acceleration nozzles, were kept the same and only the inlet configuration was changed. The major-to-total flowrate ratio was kept at 0.9 and minor-to-total flowrate ratio at 0.1. It was observed that by using the two-partitioned horizontal inlet configuration, the cutoff diameters for PM2.5, PM5 and PM10 virtual impactors, were reduced by 16%, 10% and 11%, respectively, while the wall loss of particles near the cutoff size in all three cases were reduced from 16% to about 1%. Ó 2019 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
48 49
1. Introduction
50
Various devices have been developed for classification and sampling of aerosol particles [1–4]. One of those devices is an inertial impactor, that classifies aerosol particles according to their size. The inertial impactor is widely used as a particle collector because of its simple shape and operation. The particle size for which the impactors show 50% collection efficiency is termed as cutoff size. In ideal cases, particles larger than the cutoff size are supposed to be trapped by the impaction plate, whereas particles smaller than the cutoff size should escape without being captured on the impaction plate. Many researches have been conducted to improve the collection efficiency of the inertial impactor by changing its geometric parameters and operating conditions [5–11]. The effect of impaction plate material on the impactor collection efficiency has also been investigated [12,13]. To improve the collection efficiency of the inertial impactor, a punched impaction plate has been placed in the space between the acceleration nozzle and the impaction plate [14]. To lower the cutoff size of the inertial
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
⇑ Corresponding author. E-mail address: [email protected] (S.-J. Yook).
impactor, only the shape of aerosol inlet has been modified while keeping the same nozzle and the trapping part of the inertial impactor [15,16]. However, these inertial impactors are inherently associated with the problems of particle overloading and particle bounce. Efforts have been made by many researchers to solve these problems. Rao and Whitby [17], and Lee et al. [18] recommended the use of glass filter layer or oil over the impaction plate surface for minimizing the particle bounce. Kim et al. [19] used an impaction plate of elliptical concave curvature for reducing the particle bounce. However, due to the nature of the inertial impactor which accumulates particles on the impaction plate, particle overloading and particle bounce problems can still occur. To solve the problems of the inertial impactors, Hounam and Sherwood [20] invented a virtual impactor, which was further developed by many researchers [21–23]. In the virtual impactor, a collection nozzle is used instead of the impaction plate. The virtual impactors also work on the principle of inertia, by accelerating the incoming aerosol flow through an acceleration nozzle and dividing the incoming flow into two outflow sections; namely minor outflow which is in the direction of accelerated flow and goes through the collection nozzle, and the major outflow which is perpendicular to the acceleration direction. Typically, the
https://doi.org/10.1016/j.apt.2019.09.031 0921-8831/Ó 2019 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
APT 2445
No. of Pages 7, Model 5G
11 October 2019 2 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
138 139 140 141 142 143 144 145 146 147 148 149 150 151
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
flowrate of the major outflow is set larger than that of the minor outflow. The particles with small inertia move along the major outflow, whereas the particles with large inertia cannot follow the major outflow but flow with the minor outflow resulting in being collected at the collection nozzle section. Particles that get trapped on the walls in the virtual impactor cause wall loss, while particles collected at the minor outflow section determine the collection efficiency. Hassan et al. [24], and Marple and Chien [25] have confirmed the change in the performance of the virtual impactor by varying operating parameters such as major to minor flowrate ratio and Reynolds number for the aerosol flow through the virtual impactor. Wada et al. [26] have shown that the virtual impactor has higher sampling accuracy and better separation characteristics as compared to the conventional inertial impactor. Ding and Koutrakis [27] have verified through an experimental study, by varying the operating conditions and the geometric parameters of the virtual impactors, that most of the theoretical principles and laws used for the round nozzle virtual impactors are also applicable to the slit nozzle virtual impactors. Lee et al. [28] have improved the collection efficiency of the slit nozzle virtual impactor by placing an orifice upstream of the virtual impactor, but the addition of the orifice caused increase in wall loss for particles larger than the cutoff size. Most of the abovementioned studies on virtual impactors have focused on improving the collection efficiency by reducing the cutoff size. However, the major problem, that arises when using a virtual impactor to collect aerosol particles, is the loss of particles on the walls in the virtual impactor, especially for particle sizes close to the cutoff size. The particles deposited on the virtual impactor walls influence the sampling accuracy, and thus periodic cleaning is required to avoid the acceleration nozzle blockage. Not many studies have been performed to reduce the wall loss of the virtual impactor. Chen and Yeh [29], and Loo and Cork [30] performed parametric analyses of the virtual impactors and suggested a new design having lower wall loss than the previously studied virtual impactors. However, it is still required to develop virtual impactors having good separation characteristics and showing very little or no wall loss. If the problem of wall loss is solved, the accuracy of aerosol separation by particle size can be improved. Therefore, in this study, the conventional vertical inlet of the existing virtual impactor is replaced with a two-partitioned horizontal inlet. Aerosol is introduced into the virtual impactor through the upper partition, whereas the lower partition is used for supplying clean air. The clean air is provided to concentrate the aerosol particles toward the centerline, for reducing the cutoff size and avoiding particle deposition on the virtual impactor walls. Hence, the objective of this study is to increase the sampling accuracy of the virtual impactor by minimizing the wall loss and to investigate the performance of the proposed inlet configuration.
2. Numerical method A design of a commonly used slit nozzle virtual impactor having vertical inlet is shown in Fig. 1a. Wa and La were used for representing the width and length of the acceleration nozzle, Wc for the width of the collection nozzle, Lc for the distance between the acceleration nozzle and the collection nozzle, and S for the nozzle span. As per recommendations of previous literature [27,30], each geometric parameter of the virtual impactor used in this study was considered as a multiple of slit acceleration nozzle width, that is, La/Wa = 2.5, Lc/Wa = 1.5, Wc/Wa = 1.4, and I/Wa = 6. To ensure the application of two-dimensional analysis, the span-to-width ratio of the acceleration nozzle was set to 10, that is, S/Wa = 10. The flowrate of the aerosol flowing through the vertical inlet was denoted as QT, the flowrate through the major flow as QM, and
Fig. 1. Cross-sectional view of the slit virtual impactor with: (a) vertical inlet; (b) two-partitioned horizontal inlet.
the flowrate through the minor flow as Qm. Thus, QT was equal to the sum of QM and Qm. Fig. 1b shows a design of a slit nozzle virtual impactor having two-partitioned horizontal inlet proposed in this study. Here, I1 was used for representing the width of the inlet partition through which aerosol was introduced, I2 for the width of the inlet partition through which clean air was supplied, and D1 for the length of the inlet. These parameters were fixed at I1/Wa = I2/Wa = 5, and D1/Wa = 29. Beside the inlet shape, all other parameters (Wa, Wc, La, Lc, S) of both impactors were kept the same as shown in Fig. 1a and 1b. The upper partition of the horizontal inlet was designated for introducing aerosol into the impactor with a flowrate Qa, while the lower partition was provided for the clean air supply having flowrate Qca. The flowrate ratio of aerosol to clean air, that is, Qa:Qca was varied as 2:8, 3:7, 5:5, 7:3, and 9:1. The sum of both aerosol flowrate (Qa) and clean air flowrate (Qca) was equal to the total inlet flowrate (QT). This means that the total flowrates through the acceleration nozzles of both virtual impactors shown in Fig. 1 were the same. Similarly, the flowrate ratio of minor flow
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
189 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
to total flow in both impactors was kept at 0.1, that is, the major flowrate was set to 0.9QT and the minor flowrate to 0.1QT. The different values of inlet flowrate, along with the corresponding acceleration nozzle width required for obtaining the desired cutoff size in vertical inlet virtual impactor, were calculated and given in Table 1. The FLUENT software is known to be accurate enough to simulate impactor collection efficiency at nozzle Reynolds numbers lower than 3000 [31]. Therefore, in this study, ANSYS FLUENT Release 17.2 was used for flow simulation of the slit virtual impactors. The flow was assumed to be two-dimensional, laminar, incompressible, and steady. Due to the symmetric shape, half of the cross-section of the virtual impactor was considered as the computational domain. The number of computational cells used for the analysis was in the range of 10,000–50,000, that was determined based on the grid independence test performed on the impactor geometry. In all cases considered, the value of Reynolds number was set to be 1500 or less and was calculated as
q V Wa Re ¼ l
ð1Þ
Here, q was used for representing air density, V for the average flow velocity in the acceleration nozzle, and l for the air viscosity. The temperature of air was 20 °C, and the pressure was 101.3 kPa. The SIMPLE algorithm was used for solving the energy, continuity, and momentum equations iteratively, while setting the convergence criterion to 10–6. The boundary conditions used were velocity inlet for both aerosol and clean air inlets, outflow condition for major flow and minor flow outlets, no-slip condition at the impactor walls, and symmetry condition at the centerline of the virtual impactor. The discrete phase model (DPM), embedded in the FLUENT software, was used to calculate the particle trajectories. The particles were assumed to be of spherical shape with a density of 1 g/cm3, and were injected to the impactor inlet with a constant space among them. Stokes’ drag force, Brownian force, and gravitational force were considered as the forces affecting the particle behavior. Here, the Stokes’ drag force was corrected using the slip corrosion factor. The velocity profile at all inlets were assumed to be uniform. Any particle hitting the impactor walls was assumed to be trapped by the impactor walls without reentry or reflection. The collection efficiency (CE) and the wall loss (WL) were calculated from the following equations.
214 216
CE ¼
Nm 100ð%Þ NM þ Nm
217 219 220 221 222 223 224
WL ¼
ð2Þ
sizes listed in Table 1 were calculated and compared in terms of the square root of the Stokes number. The Stokes number was obtained from the following equation.
q
Stk ¼
9lW a
Cutoff Size with Vertical Inlet (lm)
QT (L/min)
Wa (mm)
1 2.5 5 8 9 10
7.26 8.37 17.28 54.61 47.71 14.82
0.5 1 2 4 5 3
226 227
228 230 231 232
233
A virtual impactor with two-partitioned horizontal inlet was manufactured for verifying the results obtained from simulation. The virtual impactor was manufactured from stainless steel material. The slit nozzle span and the width were kept at S = 10 mm and Wa = 1 mm, respectively. The other dimensions of the virtual impactor were Wc = 1.4 mm, I1 = I2 = 5 mm, D1 = 29 mm, La = 2.5 mm, and Lc = 1.5 mm. Fig. 2 shows the experimental schematic used for measuring the collection efficiency and wall loss of the virtual impactor having two-partitioned horizontal inlet. Polystyrene Latex (PSL, Thermo Scientific) particles of spherical shape were aerosolized with the help of an atomizer and were dried using a diffusion dryer. Only required number of aerosol particles were supplied to the virtual impactor and the excess aerosol particles were exhausted through a HEPA filter. Compressed air was passed through a HEPA filter, for ensuring the supply of clean air into the virtual impactor. The aerosol and clean air flowrates were maintained at the desired level by using flowmeters. Experiments were carried out for five different cases of Qa:Qca, that is, 2:8, 3:7, 5:5, 7:3, and 9:1. In all cases, the total flowrate (QT) introduced into the virtual impactor was kept constant at 8.37 L/min. The flowrate at the major flow section was 7.53 L/min and that at the minor flow section was 0.84 L/min. The particle number concentrations at different sections of the virtual impactor such as inlet, minor outflow, and major outflow were measured with the help of an optical particle counter (OPC, Model 1.109, GRIMM, Ainring, Bayern, Germany). The suction capability of the OPC was 1.2 L/min. Thus, 0.36 L/min of room air, after passing through a HEPA filter, was added to 0.84 L/min of sampled aerosol. The tube lengths and the number of bends both upstream and downstream of the virtual impactor were kept identical to minimize the effect of particle transport loss. The experiment for each case was repeated six times and the collection efficiency (CE) and wall loss (WL) by the experimental method were obtained by the following equations.
234
Cm Q m 100ð%Þ CM Q M þ Cm Q m
WL ¼
Table 1 Inlet flowrates and the corresponding acceleration nozzle widths for obtaining the desired cutoff diameters.
ð4Þ
225
3. Experimental method
ð3Þ
Here, Nm is the number of particles escaping through the minor flow outlet, NM is the number of particles going out through the major flow outlet, and NT is the total number of particles entering through the aerosol inlet. The collection efficiency and wall loss for different flowrate ratios of clean air to aerosol and different cutoff
2 p dp C c V
Here, qp denotes the particle density, dp represents the particle diameter, and Cc is the slip correction factor.
CE ¼
NM þ Nm 100ð%Þ 1 NT
3
C T Q T ðC M Q M þ C m Q m Þ 100ð%Þ CT Q T
ð5Þ
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
267 269 270
ð6Þ
272
Here, CT, Cm and CM are the particle number concentrations at inlet section, minor outflow section, and major outflow section, respectively. Similarly, QT, Qm and QM are the flowrates at inlet section, minor outflow section, and major outflow section, respectively.
273
4. Results and discussion
278
Numerical and experimental analyses were performed on the proposed two-partitioned horizontal inlet virtual impactor having a slit nozzle of width 1 mm. Various flowrate ratios of aerosol-toclean-air, that is, 2:8, 3:7, 5:5, 7:3, and 9:1, were considered. The corresponding collection efficiency and wall loss were determined and plotted against the aerodynamic particle diameter as shown in Fig. 3. The total inlet flowrate (QT) was 8.37 L/min. The flowrate
279
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
274 275 276 277
280 281 282 283 284 285
APT 2445
No. of Pages 7, Model 5G
11 October 2019 4
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
Fig. 2. Experimental schematic for finding collection efficiency and wall loss of two-partitioned horizontal inlet virtual impactor.
Fig. 3. Comparison of two-partitioned horizontal inlet virtual impactor performance for different flowrate ratios in terms of: (a) collection efficiency; (b) wall loss.
ratio of major-to-total-flow was 0.9, while that of minor-to-totalflow was 0.1. The virtual impactor dimensions were kept at Wc = 1.4 mm, Wa = 1 mm, S = 10 mm, I1 = I2 = 5 mm, D1 = 29 mm, La = 2.5 mm, and Lc = 1.5 mm. The lines in Fig. 3 indicate the numerical results, while the symbols with error bars were used for representing the experimental data. Overall, the experimental results were consistent with the numerically obtained results for all tested flowrate ratios of aerosol-to-clean-air flow. The results showed that the addition of clean air in the virtual impactor reduced the wall loss, especially near the collection nozzle. Moreover, by increasing the clean air flowrate, the particles in the virtual impactor concentrated towards the centerline, hence reducing the cutoff size. In all of the tested cases of aerosol-toclean-air ratio (Qa:Qca), the ratio 7:3 showed minimum wall loss, hence this ratio 7:3 was considered for further analyses of the two-partitioned horizontal inlet virtual impactor. The inlet flowrates and the slit nozzle widths required for the vertical inlet virtual impactor to obtain a desired cutoff size were calculated and given in Table 1. Simulations were performed on both vertical inlet and two-partitioned horizontal inlet virtual impactors for the inlet flowrates and nozzle widths listed in Table 1. For the two-partitioned horizontal inlet virtual impactor, the flowrate ratio of aerosol-to-clean-air was set to be 7:3. The collection efficiency and wall loss of both inlet configurations were plotted and compared in Fig. 4, for the PM2.5, PM5 and PM10 acceleration nozzles. Here, PM2.5, PM5 and PM10 imply the cutoff sizes of conventional vertical inlet virtual impactors, that is, 2.5 lm, 5 lm and 10 lm, respectively. By using the two-partitioned horizontal inlet, the cutoff size was reduced as the particles were forced towards the centerline by clean air. Hence, the cutoff size decreased to 2.1 lm for PM2.5 acceleration nozzle, 4.5 lm for PM5 acceleration nozzle, and 8.9 lm for PM10 acceleration nozzle. In all cases, the clean air acted as a shield and prevented the particles from depositing on the impactor walls, especially in the collection nozzle where the conventional vertical inlet virtual impactor showed the maximum wall loss. As a result, the wall loss decreased from 16% to about 1%. Fig. 5 shows the trajectories of 2.5-lm-particles in the virtual impactors having conventional vertical inlet and two-partitioned horizontal inlet, for PM2.5 acceleration nozzle case. Here, the total inlet flowrate (QT) was 8.37 L/min, while the aerosol flowrate (Qa) and the clean air flowrate (Qac) at the inlet sections of the two-
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
5
Fig. 5. Trajectories of 2.5 lm particles in: (a) a vertical inlet virtual impactor; (b) a two-partitioned horizontal inlet virtual impactor.
Fig.4. Comparison of simulation results of two-partitioned horizontal inlet and vertical inlet virtual impactors for: (a) PM2.5 acceleration nozzle; (b) PM5 acceleration nozzle; (c) PM10 acceleration nozzle.
328 329 330
partitioned horizontal inlet were 5.86 L/min and 2.51 L/min, respectively. The major-to-total flowrate ratio was 0.9, while the minor-to-total flowrate ratio was 0.1. The particle trajectories
confirmed that, in the vertical inlet configuration, particles which were injected into the impactor near the edges turned towards the major flow after leaving the acceleration nozzle, however, some of these particles collided with the walls of collection nozzle and caused wall loss. On the other hand, by using the twopartitioned horizontal inlet configuration, the clean air supplied in the lower inlet partition concentrated the particles towards the centerline, causing more particles to move towards the minor flow and prevented them from colliding with the walls of the collection nozzle. As a result, the collection efficiency improved and the wall loss reduced, due to the use of the two-partitioned horizontal inlet. Furthermore, for showing the dimensionless characteristics of the virtual impactor with the proposed two-partitioned horizontal inlet, or the validity of the proposed inlet design for various sizes of virtual impactors, simulations were performed by varying the
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
APT 2445
No. of Pages 7, Model 5G
11 October 2019 6 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
widths of slit acceleration nozzle from 0.5 to 5 mm as listed in Table 1. All other geometric dimensions were fixed at La/Wa = 2.5, Lc/Wa = 1.5, Wc/Wa = 1.4, S/Wa = 10, I1/Wa = I2/Wa = 5, and D1/ Wa = 29. The inlet flowrates used for different acceleration nozzle widths, were the same as enumerated in Table 1. The flowrate ratios of aerosol-to-clean-air (Qa:Qca), used for each nozzle width, were varied as 2:8, 3:7, 5:5, 7:3 and 9:1. The collection efficiency and wall loss for different widths of acceleration nozzle and for different values of flowrate ratio were plotted against the square root of the Stokes number, Stk1/2, as shown in Fig. 6. The collection efficiency and wall loss curves obtained for all nozzle widths at any given flowrate ratio appeared to overlap on a single curve, showing that the performance of the two-partitioned horizontal inlet virtual impactor is non-dimensional for all tested flowrate ratios of aerosol-to-clean-air and that the proposed inlet configuration can be used for performance improvement of various sizes of virtual impactors. It was observed that the collection efficiency of the virtual impactor increased by decreasing Qa:Qca ratio. This was due to the fact that the fraction of clean air coming into the virtual impactor increased by decreasing Qa:Qca ratio, which made more particles move towards the axis of the virtual impactor and thus prevented more particles from being exhausted through the major outflow.
5. Conclusion
370
An effort has been made to improve the collection efficiency and minimize the wall loss problem of the existing virtual impactor by replacing the existing conventional vertical inlet of slit virtual impactor with the two-partitioned horizontal inlet. Aerosol was supplied to the impactor through the upper inlet partition, while the lower inlet partition was used for clean air supply. Simulations and experiments were performed by changing the flowrate ratio of aerosol-to-clean-air for PM2.5 acceleration nozzle. The flowrate ratios (Qa:Qca) considered were 2:8, 3:7, 5:5, 7:3, and 9:1. Both experimental and numerical results agreed well with each other. It was found that increasing the clean air flowrate converged the particles towards the centerline and increased the number of particles moving towards the minor flow, as a result, the collection efficiency improved. A significant reduction in wall loss was also observed for all flowrate ratios of aerosol-to-clean-air. Especially, the flowrate ratio of 7:3 showed almost negligible wall loss, that is, less than 1.2%. After verification of the predicted results, further simulations were carried out for PM2.5, PM5 and PM10 acceleration nozzles by considering the same aerosol-toclean-air flowrate ratio and the results obtained were compared with the conventional vertical inlet configuration of virtual impactor. The simulation results showed that the cutoff size was decreased by 16%, 10%, and 11% in cases of PM2.5, PM5 and PM10 acceleration nozzles, respectively. Similarly, the wall loss near the cutoff sizes for all three acceleration nozzle cases were reduced from 16% to about 1%. Moreover, analysis was made on the proposed configuration of virtual impactor for different widths and total inlet flowrates by considering various aerosol-to-clean-air flowrate ratios for PM1, PM2.5, PM5, PM8, PM9 and PM10 acceleration nozzles. The results were plotted against the square root of the Stokes number and were shown to be non-dimensional for all given flowrate ratios of aerosol-to-clean-air. Hence it can be concluded that, by replacing the conventional vertical inlet of slit virtual impactor with the two-partitioned horizontal inlet, the aerosol particle sampling accuracy can be significantly improved by reducing the wall loss and increasing the collection efficiency.
371
Declaration of Competing Interest
407
The authors declare that there is no conflict of interest.
Fig. 6. Comparison of two-partitioned horizontal inlet virtual impactor performance for different nozzle widths and flowrate ratios in terms of: (a) collection efficiency; (b) wall loss.
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406
408
Acknowledgement
409
This research was supported by the Ministry of Environment as ‘‘Korea Environmental Industry & Technology Institute (KEITI) (No. 2018000120004)”.
410
References
413
[1] J.W. Choi, C.H. Ahn, S. Bhansali, H.T. Henderson, A new magnetic bead-based, filterless bio-separator with planar electromagnet surfaces for integrated biodetection systems, Sensor. Actuat. B: Chem. 68 (2000) 34–39. [2] J. Douwes, P. Thorne, N. Pearce, D. Heederik, Bioaerosol health effects and exposure assessment: progress and prospects, Ann. Occup. Hyg. 47 (2003) 187–200. [3] T. Han, H. Zhen, D.E. Fennell, G. Mainelis, Design and evaluation of the fielddeployable electrostatic precipitator with superhydrophobic surface (FDEPSS) with high concentration rate, Aerosol Air Qual. Res. 15 (2015) 2397–2408. [4] M. Wubulihairen, S.Y. Jiang, Z. Ning, Prototype development and laboratory evaluation of an aerosol to hydrosol sampler, Aerosol Air Qual. Res. 15 (2015) 776–785. [5] V.A. Marple, B.Y.H. Liu, Characteristics of laminar jet impactors, Environ. Sci. Technol. 8 (1974) 648–654. [6] B. Jurcik, H.C. Wang, On the shape of impactor efficiency curves, J. Aerosol Sci. 26 (1995) 1139–1147. [7] C.H. Huang, C.J. Tsai, Influence of impaction plate diameter and particle density on the collection efficiency of round-nozzle inertial impactors, Aerosol Sci. Technol. 36 (2002) 714–720.
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
411 412
414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
[8] S.A. Grinshpun, G. Mainelis, M. Trunov, R.L. Górny, S.K. Sivasubramani, A. Adhikari, T. Reponen, Collection of airborne spores by circular single-stage impactors with small jet-to-plate distance, J. Aerosol Sci. 36 (2005) 575–591. [9] F.J. Gómez-Moreno, J. Rosell-Llompart, J. Fernández de la Mora, Turbulent transition in impactor jets and its effects on impactor resolution, J. Aerosol Sci. 33 (2002) 459–476. [10] C.W. Park, G. Kim, S.J. Yook, K.H. Ahn, Investigation of collection efficiency of round-nozzle impactors at different atmospheric pressures and temperatures, Adv. Powder Technol. 26 (2015) 868–873. [11] M. Gensch, A.P. Weber, Fragmentation in oblique impaction for nanoparticleagglomerates with different structures, Adv. Powder Technol. 29 (2018) 3462– 3470. [12] C.H. Huang, C.J. Tsai, The particle collection efficiency curves by the porous substrate of an inertial impactor, Aerosol Air Qual. Res. 2 (2002) 1–8. [13] K. Fukui, H. Masuda, T. Yamada, T. Matsuzaki, Influence of wall properties on the morphological change of Sb2O3 fine particles induced by impaction against a wall, Adv. Powder Technol. 6 (1995) 177–189. [14] T.W. Cheon, J.Y. Lee, J.Y. Bae, S.J. Yook, Enhancement of collection efficiency of an inertial impactor using an additional punched impaction plate, Aerosol Air Qual. Res. 17 (2017) 2349–2357. [15] M.K. Kim, W.G. Kim, K.S. Lee, S.J. Yook, Collection efficiency of round-nozzle impactors with horizontal annular inlet, J. Aerosol Sci. 74 (2014) 63–69. [16] W.G. Kim, H. Lee, J.H. Myoung, S.K. Han, S.J. Yook, K.H. Ahn, Effect of a horizontal inlet on the collection efficiency of a rectangular-slit-nozzle impactor, Aerosol Sci. Technol. 48 (2014) 649–654. [17] A.K. Rao, K.T. Whitby, Non-ideal collection characteristics of inertial impactors–I. single-stage impactors and solid particles, J. Aerosol Sci. 9 (1978) 77–86. [18] S.J. Lee, P. Demokritou, P. Koutrakis, Performance evaluation of commonly used impaction substrates under various loading conditions, J. Aerosol Sci. 36 (2005) 881–895.
7
[19] W.G. Kim, S.J. Yook, K.H. Ahn, Collection efficiency of rectangular slit-nozzle inertial impactors with impaction plates of elliptical concave curvature, Aerosol Sci. Technol. 47 (2013) 99–105. [20] R.F. Hounam, R.J. Sherwood, The cascade centripeter: a device for determining the concentration and size distribution of aerosols, Am. Ind. Hyg. Assoc. J. 26 (1965) 122–131. [21] W.D. Conner, An inertial-type particle separator for collecting large samples, J. Air Pollut. Control Assoc. 16 (1966) 35–38. [22] T.G. Dzubay, R.K. Stevens, Ambient air analysis with dichotomous sampler and X-ray fluorescence spectrometer, Environ. Sci. Technol. 9 (1975) 663–668. [23] L.J. Forney, Aerosol fractionator for large-scale sampling, Rev. Sci. Instrum. 47 (1976) 1264–1269. [24] Y.A. Hassan, B.G. Jones, T.J. Yule, Analytical study of virtual impactor aerosol separators, T. Am. Nucl. Soc. 33 (1979) 182–184. [25] V.A. Marple, C.M. Chien, Virtual impactors: a theoretical study, Environ. Sci. Technol. 14 (1980) 976–985. [26] M. Wada, M. Tsukada, N. Namiki, W.W. Szymanski, N. Noda, H. Makino, C. Kanaoka, H. Kamiya, A two-stage virtual impactor for in-stack sampling of PM2.5 and PM10 in flue gas of stationary sources, Aerosol Air, Qual. Res. 16 (2016) 36–45. [27] Y. Ding, P. Koutrakis, Development of a dichotomous slit nozzle virtual impactor, J. Aerosol Sci. 31 (2000) 1421–1431. [28] H. Lee, D.H. Jo, W.G. Kim, S.J. Yook, K.H. Ahn, Effect of an orifice on collection efficiency and wall loss of a slit virtual impactor, Aerosol Sci. Technol. 48 (2014) 121–127. [29] B.T. Chen, H.C. Yeh, An improved virtual impactor: design and performance, J. Aerosol Sci. 18 (1987) 203–214. [30] B.W. Loo, C.P. Cork, Development of high efficiency virtual impactors, Aerosol Sci. Technol. 9 (1988) 167–176. [31] S. Leduc, C. Fredriksson, R. Hermansson, Particle-tracking option in Fluent validated by simulation of a low-pressure impactor, Adv. Powder Technol. 17 (2006) 99–111.
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
No. of Pages 7, Model 5G
11 October 2019 Advanced Powder Technology xxx (xxxx) xxx 1
Contents lists available at ScienceDirect
Advanced Powder Technology journal homepage: www.elsevier.com/locate/apt
2
Original Research Paper
6 4 7 5
Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor
8
Muhammad Zeeshan Zahir a, Ji-Eun Heo b, Se-Jin Yook b,⇑
9 10 12 11 13 1 2 5 9 16 17 18 19 20 21 22 23 24 25 26 27 28
a b
Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, Khyber Pakhtunkhwa, Pakistan School of Mechanical Engineering, Hanyang University, Seoul 04763, Republic of Korea
a r t i c l e
i n f o
Article history: Received 12 January 2019 Received in revised form 24 September 2019 Accepted 26 September 2019 Available online xxxx Keywords: Aerosol Inertial separation Virtual impactor Collection efficiency Wall loss
a b s t r a c t A two-partitioned horizontal inlet was developed for improving the collection efficiency and minimizing the wall loss problem in slit virtual impactor. The two-partitions were provided to simultaneously supply both aerosol and clean air to the virtual impactor. Both numerical and experimental investigations were carried out on the developed inlet configuration by considering different flowrate ratios of aerosol to clean air. The horizontal inlet was helpful in reducing the cutoff diameter, whereas the clean air prevented the particle deposition on the virtual impactor walls. The performance of two-partitioned horizontal inlet was compared with the conventional vertical inlet configuration for PM2.5, PM5 and PM10 virtual impactors. All the operating conditions and geometric parameters, such as the inlet flowrate; the width of collection nozzle; the width, length and span of acceleration nozzle; and the distance between collection and acceleration nozzles, were kept the same and only the inlet configuration was changed. The major-to-total flowrate ratio was kept at 0.9 and minor-to-total flowrate ratio at 0.1. It was observed that by using the two-partitioned horizontal inlet configuration, the cutoff diameters for PM2.5, PM5 and PM10 virtual impactors, were reduced by 16%, 10% and 11%, respectively, while the wall loss of particles near the cutoff size in all three cases were reduced from 16% to about 1%. Ó 2019 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
48 49
1. Introduction
50
Various devices have been developed for classification and sampling of aerosol particles [1–4]. One of those devices is an inertial impactor, that classifies aerosol particles according to their size. The inertial impactor is widely used as a particle collector because of its simple shape and operation. The particle size for which the impactors show 50% collection efficiency is termed as cutoff size. In ideal cases, particles larger than the cutoff size are supposed to be trapped by the impaction plate, whereas particles smaller than the cutoff size should escape without being captured on the impaction plate. Many researches have been conducted to improve the collection efficiency of the inertial impactor by changing its geometric parameters and operating conditions [5–11]. The effect of impaction plate material on the impactor collection efficiency has also been investigated [12,13]. To improve the collection efficiency of the inertial impactor, a punched impaction plate has been placed in the space between the acceleration nozzle and the impaction plate [14]. To lower the cutoff size of the inertial
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
⇑ Corresponding author. E-mail address: [email protected] (S.-J. Yook).
impactor, only the shape of aerosol inlet has been modified while keeping the same nozzle and the trapping part of the inertial impactor [15,16]. However, these inertial impactors are inherently associated with the problems of particle overloading and particle bounce. Efforts have been made by many researchers to solve these problems. Rao and Whitby [17], and Lee et al. [18] recommended the use of glass filter layer or oil over the impaction plate surface for minimizing the particle bounce. Kim et al. [19] used an impaction plate of elliptical concave curvature for reducing the particle bounce. However, due to the nature of the inertial impactor which accumulates particles on the impaction plate, particle overloading and particle bounce problems can still occur. To solve the problems of the inertial impactors, Hounam and Sherwood [20] invented a virtual impactor, which was further developed by many researchers [21–23]. In the virtual impactor, a collection nozzle is used instead of the impaction plate. The virtual impactors also work on the principle of inertia, by accelerating the incoming aerosol flow through an acceleration nozzle and dividing the incoming flow into two outflow sections; namely minor outflow which is in the direction of accelerated flow and goes through the collection nozzle, and the major outflow which is perpendicular to the acceleration direction. Typically, the
https://doi.org/10.1016/j.apt.2019.09.031 0921-8831/Ó 2019 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
APT 2445
No. of Pages 7, Model 5G
11 October 2019 2 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
138 139 140 141 142 143 144 145 146 147 148 149 150 151
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
flowrate of the major outflow is set larger than that of the minor outflow. The particles with small inertia move along the major outflow, whereas the particles with large inertia cannot follow the major outflow but flow with the minor outflow resulting in being collected at the collection nozzle section. Particles that get trapped on the walls in the virtual impactor cause wall loss, while particles collected at the minor outflow section determine the collection efficiency. Hassan et al. [24], and Marple and Chien [25] have confirmed the change in the performance of the virtual impactor by varying operating parameters such as major to minor flowrate ratio and Reynolds number for the aerosol flow through the virtual impactor. Wada et al. [26] have shown that the virtual impactor has higher sampling accuracy and better separation characteristics as compared to the conventional inertial impactor. Ding and Koutrakis [27] have verified through an experimental study, by varying the operating conditions and the geometric parameters of the virtual impactors, that most of the theoretical principles and laws used for the round nozzle virtual impactors are also applicable to the slit nozzle virtual impactors. Lee et al. [28] have improved the collection efficiency of the slit nozzle virtual impactor by placing an orifice upstream of the virtual impactor, but the addition of the orifice caused increase in wall loss for particles larger than the cutoff size. Most of the abovementioned studies on virtual impactors have focused on improving the collection efficiency by reducing the cutoff size. However, the major problem, that arises when using a virtual impactor to collect aerosol particles, is the loss of particles on the walls in the virtual impactor, especially for particle sizes close to the cutoff size. The particles deposited on the virtual impactor walls influence the sampling accuracy, and thus periodic cleaning is required to avoid the acceleration nozzle blockage. Not many studies have been performed to reduce the wall loss of the virtual impactor. Chen and Yeh [29], and Loo and Cork [30] performed parametric analyses of the virtual impactors and suggested a new design having lower wall loss than the previously studied virtual impactors. However, it is still required to develop virtual impactors having good separation characteristics and showing very little or no wall loss. If the problem of wall loss is solved, the accuracy of aerosol separation by particle size can be improved. Therefore, in this study, the conventional vertical inlet of the existing virtual impactor is replaced with a two-partitioned horizontal inlet. Aerosol is introduced into the virtual impactor through the upper partition, whereas the lower partition is used for supplying clean air. The clean air is provided to concentrate the aerosol particles toward the centerline, for reducing the cutoff size and avoiding particle deposition on the virtual impactor walls. Hence, the objective of this study is to increase the sampling accuracy of the virtual impactor by minimizing the wall loss and to investigate the performance of the proposed inlet configuration.
2. Numerical method A design of a commonly used slit nozzle virtual impactor having vertical inlet is shown in Fig. 1a. Wa and La were used for representing the width and length of the acceleration nozzle, Wc for the width of the collection nozzle, Lc for the distance between the acceleration nozzle and the collection nozzle, and S for the nozzle span. As per recommendations of previous literature [27,30], each geometric parameter of the virtual impactor used in this study was considered as a multiple of slit acceleration nozzle width, that is, La/Wa = 2.5, Lc/Wa = 1.5, Wc/Wa = 1.4, and I/Wa = 6. To ensure the application of two-dimensional analysis, the span-to-width ratio of the acceleration nozzle was set to 10, that is, S/Wa = 10. The flowrate of the aerosol flowing through the vertical inlet was denoted as QT, the flowrate through the major flow as QM, and
Fig. 1. Cross-sectional view of the slit virtual impactor with: (a) vertical inlet; (b) two-partitioned horizontal inlet.
the flowrate through the minor flow as Qm. Thus, QT was equal to the sum of QM and Qm. Fig. 1b shows a design of a slit nozzle virtual impactor having two-partitioned horizontal inlet proposed in this study. Here, I1 was used for representing the width of the inlet partition through which aerosol was introduced, I2 for the width of the inlet partition through which clean air was supplied, and D1 for the length of the inlet. These parameters were fixed at I1/Wa = I2/Wa = 5, and D1/Wa = 29. Beside the inlet shape, all other parameters (Wa, Wc, La, Lc, S) of both impactors were kept the same as shown in Fig. 1a and 1b. The upper partition of the horizontal inlet was designated for introducing aerosol into the impactor with a flowrate Qa, while the lower partition was provided for the clean air supply having flowrate Qca. The flowrate ratio of aerosol to clean air, that is, Qa:Qca was varied as 2:8, 3:7, 5:5, 7:3, and 9:1. The sum of both aerosol flowrate (Qa) and clean air flowrate (Qca) was equal to the total inlet flowrate (QT). This means that the total flowrates through the acceleration nozzles of both virtual impactors shown in Fig. 1 were the same. Similarly, the flowrate ratio of minor flow
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
189 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
to total flow in both impactors was kept at 0.1, that is, the major flowrate was set to 0.9QT and the minor flowrate to 0.1QT. The different values of inlet flowrate, along with the corresponding acceleration nozzle width required for obtaining the desired cutoff size in vertical inlet virtual impactor, were calculated and given in Table 1. The FLUENT software is known to be accurate enough to simulate impactor collection efficiency at nozzle Reynolds numbers lower than 3000 [31]. Therefore, in this study, ANSYS FLUENT Release 17.2 was used for flow simulation of the slit virtual impactors. The flow was assumed to be two-dimensional, laminar, incompressible, and steady. Due to the symmetric shape, half of the cross-section of the virtual impactor was considered as the computational domain. The number of computational cells used for the analysis was in the range of 10,000–50,000, that was determined based on the grid independence test performed on the impactor geometry. In all cases considered, the value of Reynolds number was set to be 1500 or less and was calculated as
q V Wa Re ¼ l
ð1Þ
Here, q was used for representing air density, V for the average flow velocity in the acceleration nozzle, and l for the air viscosity. The temperature of air was 20 °C, and the pressure was 101.3 kPa. The SIMPLE algorithm was used for solving the energy, continuity, and momentum equations iteratively, while setting the convergence criterion to 10–6. The boundary conditions used were velocity inlet for both aerosol and clean air inlets, outflow condition for major flow and minor flow outlets, no-slip condition at the impactor walls, and symmetry condition at the centerline of the virtual impactor. The discrete phase model (DPM), embedded in the FLUENT software, was used to calculate the particle trajectories. The particles were assumed to be of spherical shape with a density of 1 g/cm3, and were injected to the impactor inlet with a constant space among them. Stokes’ drag force, Brownian force, and gravitational force were considered as the forces affecting the particle behavior. Here, the Stokes’ drag force was corrected using the slip corrosion factor. The velocity profile at all inlets were assumed to be uniform. Any particle hitting the impactor walls was assumed to be trapped by the impactor walls without reentry or reflection. The collection efficiency (CE) and the wall loss (WL) were calculated from the following equations.
214 216
CE ¼
Nm 100ð%Þ NM þ Nm
217 219 220 221 222 223 224
WL ¼
ð2Þ
sizes listed in Table 1 were calculated and compared in terms of the square root of the Stokes number. The Stokes number was obtained from the following equation.
q
Stk ¼
9lW a
Cutoff Size with Vertical Inlet (lm)
QT (L/min)
Wa (mm)
1 2.5 5 8 9 10
7.26 8.37 17.28 54.61 47.71 14.82
0.5 1 2 4 5 3
226 227
228 230 231 232
233
A virtual impactor with two-partitioned horizontal inlet was manufactured for verifying the results obtained from simulation. The virtual impactor was manufactured from stainless steel material. The slit nozzle span and the width were kept at S = 10 mm and Wa = 1 mm, respectively. The other dimensions of the virtual impactor were Wc = 1.4 mm, I1 = I2 = 5 mm, D1 = 29 mm, La = 2.5 mm, and Lc = 1.5 mm. Fig. 2 shows the experimental schematic used for measuring the collection efficiency and wall loss of the virtual impactor having two-partitioned horizontal inlet. Polystyrene Latex (PSL, Thermo Scientific) particles of spherical shape were aerosolized with the help of an atomizer and were dried using a diffusion dryer. Only required number of aerosol particles were supplied to the virtual impactor and the excess aerosol particles were exhausted through a HEPA filter. Compressed air was passed through a HEPA filter, for ensuring the supply of clean air into the virtual impactor. The aerosol and clean air flowrates were maintained at the desired level by using flowmeters. Experiments were carried out for five different cases of Qa:Qca, that is, 2:8, 3:7, 5:5, 7:3, and 9:1. In all cases, the total flowrate (QT) introduced into the virtual impactor was kept constant at 8.37 L/min. The flowrate at the major flow section was 7.53 L/min and that at the minor flow section was 0.84 L/min. The particle number concentrations at different sections of the virtual impactor such as inlet, minor outflow, and major outflow were measured with the help of an optical particle counter (OPC, Model 1.109, GRIMM, Ainring, Bayern, Germany). The suction capability of the OPC was 1.2 L/min. Thus, 0.36 L/min of room air, after passing through a HEPA filter, was added to 0.84 L/min of sampled aerosol. The tube lengths and the number of bends both upstream and downstream of the virtual impactor were kept identical to minimize the effect of particle transport loss. The experiment for each case was repeated six times and the collection efficiency (CE) and wall loss (WL) by the experimental method were obtained by the following equations.
234
Cm Q m 100ð%Þ CM Q M þ Cm Q m
WL ¼
Table 1 Inlet flowrates and the corresponding acceleration nozzle widths for obtaining the desired cutoff diameters.
ð4Þ
225
3. Experimental method
ð3Þ
Here, Nm is the number of particles escaping through the minor flow outlet, NM is the number of particles going out through the major flow outlet, and NT is the total number of particles entering through the aerosol inlet. The collection efficiency and wall loss for different flowrate ratios of clean air to aerosol and different cutoff
2 p dp C c V
Here, qp denotes the particle density, dp represents the particle diameter, and Cc is the slip correction factor.
CE ¼
NM þ Nm 100ð%Þ 1 NT
3
C T Q T ðC M Q M þ C m Q m Þ 100ð%Þ CT Q T
ð5Þ
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
267 269 270
ð6Þ
272
Here, CT, Cm and CM are the particle number concentrations at inlet section, minor outflow section, and major outflow section, respectively. Similarly, QT, Qm and QM are the flowrates at inlet section, minor outflow section, and major outflow section, respectively.
273
4. Results and discussion
278
Numerical and experimental analyses were performed on the proposed two-partitioned horizontal inlet virtual impactor having a slit nozzle of width 1 mm. Various flowrate ratios of aerosol-toclean-air, that is, 2:8, 3:7, 5:5, 7:3, and 9:1, were considered. The corresponding collection efficiency and wall loss were determined and plotted against the aerodynamic particle diameter as shown in Fig. 3. The total inlet flowrate (QT) was 8.37 L/min. The flowrate
279
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
274 275 276 277
280 281 282 283 284 285
APT 2445
No. of Pages 7, Model 5G
11 October 2019 4
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
Fig. 2. Experimental schematic for finding collection efficiency and wall loss of two-partitioned horizontal inlet virtual impactor.
Fig. 3. Comparison of two-partitioned horizontal inlet virtual impactor performance for different flowrate ratios in terms of: (a) collection efficiency; (b) wall loss.
ratio of major-to-total-flow was 0.9, while that of minor-to-totalflow was 0.1. The virtual impactor dimensions were kept at Wc = 1.4 mm, Wa = 1 mm, S = 10 mm, I1 = I2 = 5 mm, D1 = 29 mm, La = 2.5 mm, and Lc = 1.5 mm. The lines in Fig. 3 indicate the numerical results, while the symbols with error bars were used for representing the experimental data. Overall, the experimental results were consistent with the numerically obtained results for all tested flowrate ratios of aerosol-to-clean-air flow. The results showed that the addition of clean air in the virtual impactor reduced the wall loss, especially near the collection nozzle. Moreover, by increasing the clean air flowrate, the particles in the virtual impactor concentrated towards the centerline, hence reducing the cutoff size. In all of the tested cases of aerosol-toclean-air ratio (Qa:Qca), the ratio 7:3 showed minimum wall loss, hence this ratio 7:3 was considered for further analyses of the two-partitioned horizontal inlet virtual impactor. The inlet flowrates and the slit nozzle widths required for the vertical inlet virtual impactor to obtain a desired cutoff size were calculated and given in Table 1. Simulations were performed on both vertical inlet and two-partitioned horizontal inlet virtual impactors for the inlet flowrates and nozzle widths listed in Table 1. For the two-partitioned horizontal inlet virtual impactor, the flowrate ratio of aerosol-to-clean-air was set to be 7:3. The collection efficiency and wall loss of both inlet configurations were plotted and compared in Fig. 4, for the PM2.5, PM5 and PM10 acceleration nozzles. Here, PM2.5, PM5 and PM10 imply the cutoff sizes of conventional vertical inlet virtual impactors, that is, 2.5 lm, 5 lm and 10 lm, respectively. By using the two-partitioned horizontal inlet, the cutoff size was reduced as the particles were forced towards the centerline by clean air. Hence, the cutoff size decreased to 2.1 lm for PM2.5 acceleration nozzle, 4.5 lm for PM5 acceleration nozzle, and 8.9 lm for PM10 acceleration nozzle. In all cases, the clean air acted as a shield and prevented the particles from depositing on the impactor walls, especially in the collection nozzle where the conventional vertical inlet virtual impactor showed the maximum wall loss. As a result, the wall loss decreased from 16% to about 1%. Fig. 5 shows the trajectories of 2.5-lm-particles in the virtual impactors having conventional vertical inlet and two-partitioned horizontal inlet, for PM2.5 acceleration nozzle case. Here, the total inlet flowrate (QT) was 8.37 L/min, while the aerosol flowrate (Qa) and the clean air flowrate (Qac) at the inlet sections of the two-
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
5
Fig. 5. Trajectories of 2.5 lm particles in: (a) a vertical inlet virtual impactor; (b) a two-partitioned horizontal inlet virtual impactor.
Fig.4. Comparison of simulation results of two-partitioned horizontal inlet and vertical inlet virtual impactors for: (a) PM2.5 acceleration nozzle; (b) PM5 acceleration nozzle; (c) PM10 acceleration nozzle.
328 329 330
partitioned horizontal inlet were 5.86 L/min and 2.51 L/min, respectively. The major-to-total flowrate ratio was 0.9, while the minor-to-total flowrate ratio was 0.1. The particle trajectories
confirmed that, in the vertical inlet configuration, particles which were injected into the impactor near the edges turned towards the major flow after leaving the acceleration nozzle, however, some of these particles collided with the walls of collection nozzle and caused wall loss. On the other hand, by using the twopartitioned horizontal inlet configuration, the clean air supplied in the lower inlet partition concentrated the particles towards the centerline, causing more particles to move towards the minor flow and prevented them from colliding with the walls of the collection nozzle. As a result, the collection efficiency improved and the wall loss reduced, due to the use of the two-partitioned horizontal inlet. Furthermore, for showing the dimensionless characteristics of the virtual impactor with the proposed two-partitioned horizontal inlet, or the validity of the proposed inlet design for various sizes of virtual impactors, simulations were performed by varying the
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
APT 2445
No. of Pages 7, Model 5G
11 October 2019 6 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx
widths of slit acceleration nozzle from 0.5 to 5 mm as listed in Table 1. All other geometric dimensions were fixed at La/Wa = 2.5, Lc/Wa = 1.5, Wc/Wa = 1.4, S/Wa = 10, I1/Wa = I2/Wa = 5, and D1/ Wa = 29. The inlet flowrates used for different acceleration nozzle widths, were the same as enumerated in Table 1. The flowrate ratios of aerosol-to-clean-air (Qa:Qca), used for each nozzle width, were varied as 2:8, 3:7, 5:5, 7:3 and 9:1. The collection efficiency and wall loss for different widths of acceleration nozzle and for different values of flowrate ratio were plotted against the square root of the Stokes number, Stk1/2, as shown in Fig. 6. The collection efficiency and wall loss curves obtained for all nozzle widths at any given flowrate ratio appeared to overlap on a single curve, showing that the performance of the two-partitioned horizontal inlet virtual impactor is non-dimensional for all tested flowrate ratios of aerosol-to-clean-air and that the proposed inlet configuration can be used for performance improvement of various sizes of virtual impactors. It was observed that the collection efficiency of the virtual impactor increased by decreasing Qa:Qca ratio. This was due to the fact that the fraction of clean air coming into the virtual impactor increased by decreasing Qa:Qca ratio, which made more particles move towards the axis of the virtual impactor and thus prevented more particles from being exhausted through the major outflow.
5. Conclusion
370
An effort has been made to improve the collection efficiency and minimize the wall loss problem of the existing virtual impactor by replacing the existing conventional vertical inlet of slit virtual impactor with the two-partitioned horizontal inlet. Aerosol was supplied to the impactor through the upper inlet partition, while the lower inlet partition was used for clean air supply. Simulations and experiments were performed by changing the flowrate ratio of aerosol-to-clean-air for PM2.5 acceleration nozzle. The flowrate ratios (Qa:Qca) considered were 2:8, 3:7, 5:5, 7:3, and 9:1. Both experimental and numerical results agreed well with each other. It was found that increasing the clean air flowrate converged the particles towards the centerline and increased the number of particles moving towards the minor flow, as a result, the collection efficiency improved. A significant reduction in wall loss was also observed for all flowrate ratios of aerosol-to-clean-air. Especially, the flowrate ratio of 7:3 showed almost negligible wall loss, that is, less than 1.2%. After verification of the predicted results, further simulations were carried out for PM2.5, PM5 and PM10 acceleration nozzles by considering the same aerosol-toclean-air flowrate ratio and the results obtained were compared with the conventional vertical inlet configuration of virtual impactor. The simulation results showed that the cutoff size was decreased by 16%, 10%, and 11% in cases of PM2.5, PM5 and PM10 acceleration nozzles, respectively. Similarly, the wall loss near the cutoff sizes for all three acceleration nozzle cases were reduced from 16% to about 1%. Moreover, analysis was made on the proposed configuration of virtual impactor for different widths and total inlet flowrates by considering various aerosol-to-clean-air flowrate ratios for PM1, PM2.5, PM5, PM8, PM9 and PM10 acceleration nozzles. The results were plotted against the square root of the Stokes number and were shown to be non-dimensional for all given flowrate ratios of aerosol-to-clean-air. Hence it can be concluded that, by replacing the conventional vertical inlet of slit virtual impactor with the two-partitioned horizontal inlet, the aerosol particle sampling accuracy can be significantly improved by reducing the wall loss and increasing the collection efficiency.
371
Declaration of Competing Interest
407
The authors declare that there is no conflict of interest.
Fig. 6. Comparison of two-partitioned horizontal inlet virtual impactor performance for different nozzle widths and flowrate ratios in terms of: (a) collection efficiency; (b) wall loss.
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406
408
Acknowledgement
409
This research was supported by the Ministry of Environment as ‘‘Korea Environmental Industry & Technology Institute (KEITI) (No. 2018000120004)”.
410
References
413
[1] J.W. Choi, C.H. Ahn, S. Bhansali, H.T. Henderson, A new magnetic bead-based, filterless bio-separator with planar electromagnet surfaces for integrated biodetection systems, Sensor. Actuat. B: Chem. 68 (2000) 34–39. [2] J. Douwes, P. Thorne, N. Pearce, D. Heederik, Bioaerosol health effects and exposure assessment: progress and prospects, Ann. Occup. Hyg. 47 (2003) 187–200. [3] T. Han, H. Zhen, D.E. Fennell, G. Mainelis, Design and evaluation of the fielddeployable electrostatic precipitator with superhydrophobic surface (FDEPSS) with high concentration rate, Aerosol Air Qual. Res. 15 (2015) 2397–2408. [4] M. Wubulihairen, S.Y. Jiang, Z. Ning, Prototype development and laboratory evaluation of an aerosol to hydrosol sampler, Aerosol Air Qual. Res. 15 (2015) 776–785. [5] V.A. Marple, B.Y.H. Liu, Characteristics of laminar jet impactors, Environ. Sci. Technol. 8 (1974) 648–654. [6] B. Jurcik, H.C. Wang, On the shape of impactor efficiency curves, J. Aerosol Sci. 26 (1995) 1139–1147. [7] C.H. Huang, C.J. Tsai, Influence of impaction plate diameter and particle density on the collection efficiency of round-nozzle inertial impactors, Aerosol Sci. Technol. 36 (2002) 714–720.
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031
411 412
414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
APT 2445
No. of Pages 7, Model 5G
11 October 2019 M.Z. Zahir et al. / Advanced Powder Technology xxx (xxxx) xxx 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
[8] S.A. Grinshpun, G. Mainelis, M. Trunov, R.L. Górny, S.K. Sivasubramani, A. Adhikari, T. Reponen, Collection of airborne spores by circular single-stage impactors with small jet-to-plate distance, J. Aerosol Sci. 36 (2005) 575–591. [9] F.J. Gómez-Moreno, J. Rosell-Llompart, J. Fernández de la Mora, Turbulent transition in impactor jets and its effects on impactor resolution, J. Aerosol Sci. 33 (2002) 459–476. [10] C.W. Park, G. Kim, S.J. Yook, K.H. Ahn, Investigation of collection efficiency of round-nozzle impactors at different atmospheric pressures and temperatures, Adv. Powder Technol. 26 (2015) 868–873. [11] M. Gensch, A.P. Weber, Fragmentation in oblique impaction for nanoparticleagglomerates with different structures, Adv. Powder Technol. 29 (2018) 3462– 3470. [12] C.H. Huang, C.J. Tsai, The particle collection efficiency curves by the porous substrate of an inertial impactor, Aerosol Air Qual. Res. 2 (2002) 1–8. [13] K. Fukui, H. Masuda, T. Yamada, T. Matsuzaki, Influence of wall properties on the morphological change of Sb2O3 fine particles induced by impaction against a wall, Adv. Powder Technol. 6 (1995) 177–189. [14] T.W. Cheon, J.Y. Lee, J.Y. Bae, S.J. Yook, Enhancement of collection efficiency of an inertial impactor using an additional punched impaction plate, Aerosol Air Qual. Res. 17 (2017) 2349–2357. [15] M.K. Kim, W.G. Kim, K.S. Lee, S.J. Yook, Collection efficiency of round-nozzle impactors with horizontal annular inlet, J. Aerosol Sci. 74 (2014) 63–69. [16] W.G. Kim, H. Lee, J.H. Myoung, S.K. Han, S.J. Yook, K.H. Ahn, Effect of a horizontal inlet on the collection efficiency of a rectangular-slit-nozzle impactor, Aerosol Sci. Technol. 48 (2014) 649–654. [17] A.K. Rao, K.T. Whitby, Non-ideal collection characteristics of inertial impactors–I. single-stage impactors and solid particles, J. Aerosol Sci. 9 (1978) 77–86. [18] S.J. Lee, P. Demokritou, P. Koutrakis, Performance evaluation of commonly used impaction substrates under various loading conditions, J. Aerosol Sci. 36 (2005) 881–895.
7
[19] W.G. Kim, S.J. Yook, K.H. Ahn, Collection efficiency of rectangular slit-nozzle inertial impactors with impaction plates of elliptical concave curvature, Aerosol Sci. Technol. 47 (2013) 99–105. [20] R.F. Hounam, R.J. Sherwood, The cascade centripeter: a device for determining the concentration and size distribution of aerosols, Am. Ind. Hyg. Assoc. J. 26 (1965) 122–131. [21] W.D. Conner, An inertial-type particle separator for collecting large samples, J. Air Pollut. Control Assoc. 16 (1966) 35–38. [22] T.G. Dzubay, R.K. Stevens, Ambient air analysis with dichotomous sampler and X-ray fluorescence spectrometer, Environ. Sci. Technol. 9 (1975) 663–668. [23] L.J. Forney, Aerosol fractionator for large-scale sampling, Rev. Sci. Instrum. 47 (1976) 1264–1269. [24] Y.A. Hassan, B.G. Jones, T.J. Yule, Analytical study of virtual impactor aerosol separators, T. Am. Nucl. Soc. 33 (1979) 182–184. [25] V.A. Marple, C.M. Chien, Virtual impactors: a theoretical study, Environ. Sci. Technol. 14 (1980) 976–985. [26] M. Wada, M. Tsukada, N. Namiki, W.W. Szymanski, N. Noda, H. Makino, C. Kanaoka, H. Kamiya, A two-stage virtual impactor for in-stack sampling of PM2.5 and PM10 in flue gas of stationary sources, Aerosol Air, Qual. Res. 16 (2016) 36–45. [27] Y. Ding, P. Koutrakis, Development of a dichotomous slit nozzle virtual impactor, J. Aerosol Sci. 31 (2000) 1421–1431. [28] H. Lee, D.H. Jo, W.G. Kim, S.J. Yook, K.H. Ahn, Effect of an orifice on collection efficiency and wall loss of a slit virtual impactor, Aerosol Sci. Technol. 48 (2014) 121–127. [29] B.T. Chen, H.C. Yeh, An improved virtual impactor: design and performance, J. Aerosol Sci. 18 (1987) 203–214. [30] B.W. Loo, C.P. Cork, Development of high efficiency virtual impactors, Aerosol Sci. Technol. 9 (1988) 167–176. [31] S. Leduc, C. Fredriksson, R. Hermansson, Particle-tracking option in Fluent validated by simulation of a low-pressure impactor, Adv. Powder Technol. 17 (2006) 99–111.
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
Please cite this article as: M. Z. Zahir, J. E. Heo and S. J. Yook, Influence of clean air and inlet configuration on the performance of slit nozzle virtual impactor, Advanced Powder Technology, https://doi.org/10.1016/j.apt.2019.09.031