Received August 3, 2020, accepted September 2, 2020, date of publication September 9, 2020, date of current version September 21, 2020. Digital Object Identifier 10.1109/ACCESS.2020.3022929 Experimental Floating Electrode Electric Curtain Evaluation to Contain Airborne Particles Resulting From the Paper Industrial Manipulation JOAN J. GARCÍA-GARCÍA , (Member, IEEE) Grupo de Aplicaciones Electro-Magnéticas Aplicadas (GAEMI), Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelola, 08193 Barcelona, Spain e-mail: joan.garcia@uab.es This work has been funded by the Grupo de Aplicaciones Electro-Magnèticas Industriales (GAEMI) group of the Electronic Department of the Universitat Autònoma de Barcelona. ABSTRACT This paper describes the design, fabrication and characterization of a three phase floating electrodes electroportation system for dust confinement, compatible with industrial optical quality control systems. The proposed Floating Electrode Electronic Curtain (FE-EC) is composed of floating parallel wire electrodes framed by a two-layer PCB. The floating electrodes are connected to one of the synchronized three channels to generate an electric field traveling wave. The separation between the floating wire electrodes has been determined by minimizing the number of trajectories crossing the FE-EC using a numerical simulation tool. It is shown that the effect of the FE-EC in the luminosity of an industrial led luminaire can be compensated by an 1% increment in the led supply voltage. The performance of the fabricated FE-EC prototype is experimentally quantified, showing efficiencies higher than 95% in simulated relevant conditions. INDEX TERMS Airborne dust control, dielectrophoretic force, electric curtain. I. INTRODUCTION contain airborne suspended paper dust particles in industrial The electrodynamic screen concept was introduced by F.B. environments to protect delicate equipment, i.e. the elements Tatom [1] in 1967, and further developed and implemented of an optical quality control system. The FE-EC is based on by Masuda et al. at the Tokyo University in the 1970s the so called dielectrophoretic (DEP) force. The DEP force [2]–[4]. Since then, electroporation based techniques have term was introduced in 1951 by H. A. Pohl [24] and constitute been used in a very wide application range, from solar the seed of considerable research activity [25]. The utilization cell cleaning [5]–[10] to radioactive dust manipulation in of DEP forces to filter and precipitate airborne particles, has fusion reactors [11], [12], microfluidics [13], [14] , control been previously reported in the literature [26], [27]. However, of bubbles in dielectric liquids [15], sorting and manipu- this is the first time that the technique is applied to generate lating particles by size [16]–[19], cell transport in liquid a Floating Electrode Curtain to content suspended industrial medium [20], [21] or dust reduction in professional envi- paper dust. The dynamic of small particles interacting with ronments in the so called electrostatic precipitators [18], non-homogeneous electric fields is a major problem that [22], [23]. The electrostatic precipitator operation relays in depends in a complex way of parameters like the particle the natural or induced particle charge and has been con- mass, size, permittivity and charge of the target particles solidated as an excellent method for the removal of parti- as well as the field distribution [28]–[34]. The electropor- cles greater than 1 µm suspended in gases, constituting the tation design proposed in this work is based on the three most commonly technology to clean the emission in power phase interdigital wire electrodes scheme, common in the plants [23]. The present work has the practical objective to solar cell cleaning systems, with the particularity that the develop a Floating Electrodes-Electric Curtain (FE-EC) to wire electrodes are floating (and the air can freely circulate through the barrier). The applied voltage pulses in the FE-EC The associate editor coordinating the review of this manuscript and are low frequency asymmetric pulses up to 2.5 kV which approving it for publication was Shuaihu Li . generate a single front in the non-homogeneous electric field This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. 164114 For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ VOLUME 8, 2020 J. J. García-García: Experimental FE-EC Evaluation to Contain Airborne Particles traveling wave. A functional FE-EC prototype has been especially during the transient that lead to the establishment designed, fabricated and tested. To quantify the effect of of the electrostatic field distribution [39], [40]. The proposed the FE-EC as dust barrier, the number of particles crossing FE-EC takes advantage of this intense initial transient effect the FE-EC with and without the effect of the electric pulses in combination with the traveling wave effect to avoid that traveling across the wire electrodes has been compared. Dust the trajectories of the airborne particles cross the plane deter- samples used to characterize the FE-EC, are composed of mined by the FE-EC floating wire electrodes. cellulose agglomerates fragments, kaolin and other paper chemical additives mixed with arabic gum 30 µm average III. DUST PARTICLES DESCRIPTION size spheres formed in air suspension as a result of industrial The industrial paper manipulation utilizes processes in which processes. The developed prototype has been designed to the paper webs circulate at rates in the order of 100 m per safely operate up to 4.4 kV DC voltages. 3D CST-Particle minute or even higher, inevitably generating the expulsion Studio software has been used as numerical tool to set the of particles to the ambient. Furthermore, the utilization of parameters that minimize the number of particle trajectories volatile chemical products like ink, paper chemical additives crossing the FE-EC. This numerical optimization tool avoids or other substances produce suspended micro-precipitates dealing with the complexity of the details of an analytical that contribute to the airborne dust. Real time quality controls theoretical framework, simplifying the design process. The based on optical systems are commonly used to monitor the FE-EC prototype has shown efficiencies greater than 95% industrial paper processes requiring maintenance tasks due to acting as dust containing barrier. The light transparency of the dust accumulation that hinder the production. The objec- the FE-EC has been characterized to check the compatibility tive of this work is the design and fabrication of a dust con- of the prototype with industrial led luminaries for the optical tainment barrier that minimizes the dust accumulation in the quality control systems. luminaires of the optical quality control systems. The FE-EC to develop should be effective for air suspended non-charged II. THEORETICAL FRAMEWORK dielectric particles with sizes ranging from a few microns to The applied force on a homogeneous spherical particle in 30 µm. Fig. 1 shows two different examples of this target presence of an electric field has the four main contributions dust. Fig. 1 (a) sample is composed of small and irregular described in Equation (1) [19] particles mainly formed by agglomerations of paper loads and εp − ε0 cellulose fibers with an average density of 1.5 gr/cm3 with FE = mp · gE + qs EE + 6π ξ Rp U E + 2πR3p E |E|2 ∇ (1) average dimensions lower than 10 µm. εp +2 · ε0 Particles around 0.5 µm are expected to be formed by where mp is the particle mass, gE the gravity constant crystals and precipitators of chemistry substances used to (9.8 m/s2), qs, the particle charge, ξ is the kinematic air control color, flexibility and other properties of the paper. viscosity (between 1.71 · 10-5 Pa · s and 1.95 · 10-5 Pa · s), In this case, the mean size is below 10 µm, although larger UE is the particle velocity, Rp the particle radius, ep the agglomerates including cellulose fiber can be found. particle dielectric permittivity, eo the free space permittivity Fig. 1 (b) shows a completely different dust type mainly (8.9 · 10-12 F/m ) and EE is the applied electric field. formed by 30 µm average diameter spheres of arabic gum The consideration of all the terms in equation (1) which typical density is 0.64 gr/cm3 . The overall average in the analysis of particle’s dynamics when a spatially expected mass of the particles is 0.014 µgr. The dust to non-homogeneous electric field is applied, leads to complex, contain may be a mix between typically 2 µm crystals of and even chaotic trajectories [28], [32]. However, for practical chemical precipitates, 10 µm cellulose particles, and 30 µm applications only the conditions to establish an effective drift average diameter arabic gum spheres. velocity to the dust particles are necessary rather than the analytical details of the trajectories. Numerical methods are IV. PROPOSED FE-EC DESIGN DESCRIPTION commonly used to deal with the complexity of the prob- The structure of the proposed FE-EC is an interdigi- lem [35]–[37]. Considering the typical intrinsic dust particle tal three-phase square pulses scheme, which has already charge reported in the literature [10], [27], [30], it can be tested and successfully applied in other electronic curtain shown that the dominant term in the equation (1) for voltages designs [31], [32]. Unlike of the electric curtains in other in the range of kV and for electrode distances <1 mm is the works, the electrode wires of the proposed FE-EC are float- DEP force, being the rest of the terms orders of magnitude ing, in such a way that the air and light can flow though the lower. This dominant term is proportional to the electric electric curtain electrode wires. The effect of the electronic field gradient and the permittivity difference between the curtain is to modify the trajectory of the flying particles in particle and the medium [38]. In the case of air background, such a way that the plain containing the floating electrodes the particles tend to move towards the more intense electric becomes a barrier for the suspended dust particles. field regions from the lower electric field regions [18] mak- The physical structure of the curtains consists on a stan- ing possible the design of electric curtains for non-charged dard 2-layer FR4 PCB frame containing the interconnection dielectric particles, Moreover, it has been shown that the DEP matrix to connect each floating wire electrode with one of forces contribute significantly in the particle’s movement the three synchronized channels. The floating electrodes are VOLUME 8, 2020 164115 J. J. García-García: Experimental FE-EC Evaluation to Contain Airborne Particles FIGURE 1. Microscope images of two different dust samples. a) dust FIGURE 2. Details of the three-phase structure. (a) Layout of the TCE composed by cellulose fibers, kaolin and other chemical additives structure showing the top (green lines), bottom (red lines) metallic layers agglomerations with 10 µm average dimensions. b) dust mainly and the holes (blue) in the PCB, and the main dimensions. 20.124 mm composed by 30 µm average radius spheres of arabic gum. width of the floating electrodes windows, 1 mm minimal separation between electrode wires PCB holes. (b) picture showing the top of the FE-EC before soldering and passivate. implemented using 150 µm (38 SWG) copper wires with an enamel coating resistant up to 400 ◦ C and 1.5 kVRMS electric breakdown. This insulation level combined with the 3 kV/mm air breakdown ratio provides a safe operability up to 4.5 kV DC in the floating electrodes area when using an inter-electrode distance of 0.8 mm. The PCB interconnection matrix structure is passivated using the encapsulation Epoxy ER22193SL after the soldering process, guaranteeing an elec- trical strength of 12 kV per mm in the PCB frame area. The three channels of the proposed FE-EC are excited with square pulses out of phase to generate an electric field dis- tribution (see Fig.3), traveling perpendicularly to the parallel wire electrodes. After several tests, a low operating frequency FIGURE 3. Electric scheme of the electric curtain three channel exciting of 10 Hz has proved to be the most effective for the tested pulses. The rise, hold and down time is T/3 where T is a period of 0.1 s corresponding to 10 Hz. The nominal pulse value is set equal for the three dust samples. The FE-EC channels have been implemented channels in values between 0 kV and 2.4 kV. This value is surpassed in the by synchronizing three Advanced Energy Ultravolt 4HVA rising flank in a 15% overshoot (<1ms). precision high voltage amplifier which offers rising times for the pulses < 1 ms. As it has been said, to deal with the complexity of the 3D spatial field distribution generated by the electrodes the electroportation problem is common the utilization of (Fig. 4 (a)) and to design a proper insulation strategy to numerical tools to describe the interaction between the avoid spark discharges produced by electrical break-down. dielectric particles and the spatial inhomogeneous electric On the other hand, the CST-Particle Studio module has fields [32], [34], [35]. In our case, the full 3D electromagnetic been used in a parametric study that allows to explore the simulation software CST Studio has been used to visualize robustness of the system. Fig. 4 (b) shows the trajectories of 164116 VOLUME 8, 2020 J. J. García-García: Experimental FE-EC Evaluation to Contain Airborne Particles FIGURE 5. Effect of the electric curtain floating electrode wires in the luminosity of a led industrial illumination system for different led supply voltages. In the inside is depicted the reduction percentage of the luminosity by the effect of the FE-EC screening. The voltage increment to compensate the screening is <1% in all the points tested. FIGURE 4. (a) 3D absolute electric field isolines over a plane perpendicular to the plane defined by the parallel floating electrodes. (b) simulated particle’s trajectories after the optimization process in which the plane defined by the floating parallel wire electrodes (in grey) defines an effective barrier for the particles. particles interacting with the electric field distribution shown in Fig. 4 (a). The parametric analysis points out that for 150 µm wires with 0.8 mm separation distance and three phases with 2.4 kV the trajectories of for particles in the range between 10−12 gr and 10−11 gr does not cross the FE- EC plain (see Fig. 4 (b)). The particle charge parameter for the simulations, has been estimated using the dielectric satu- FIGURE 6. Picture of the experimental setup used to quantify the effect ration charge expression Qsat = 12πεo Eo R2P [25], obtaining of the FE-EC as dust barrier. The FE-EC acts as a vertical boundary of the confining the Airborne atmosphere. The FE-EC removed dust is deposited 3.0 · 10−11 C in the case of the 30 µm gum arabic particles, in the bottom of the volume. 3.3 · 10−12 for the 10 µm cellulose particles or 1.3 · 10−13 C for the 2 µm kaolin precipitates. To ensure the compatibility of the TE-EC prototype with atmosphere is generated by a vibrating 50 µm sieve in the top the industrial optic quality control system, the effect of the of the confined volume. The electric curtain has been fixed parallel electrode wires on the luminosity of an industrial led perpendicular to the ground, limiting the confined volume luminaire has been quantized. The luminosity of the lumi- containing the flying dust particles. The natural tendency of naire can be adjusted by the supply voltage being possible to the dust particle is to fall and be spreader in the volume operate in a range between 24 V and 28 V offering brightness due to the air turbulences. The number of particles crossing between and 10 lx. The results in Fig. 5 show that the effect the electric curtain are counted using a Fluke 985 Airborne luminosity reduction introduced by the FE-EC is between Counter. This device requires a small and continuous air flux 6.5% and 8%. trough the internal sample reservoir where the particles are As can be observed in Fig.5, the light attenuation intro- detected and counted using a laser sensor. As a consequence, duced by the FE-EC can be compensated with less than an the sensing process produces a small air flux through the 1% increment in the led supply voltage. This transparency electric curtain that favors the particle curtain permeability. level is important from a practical point of view to guarantee The effect of the electric curtain is quantified using the that the FE-EC compatibility with the industrial environment. efficiency defined in Eq. 2. This efficiency figure is obtained by normalizing the number of particles crossing the FE-EC V. EXPERIMENTAL SETUP AND MEASUREMENTS barrier in the case of non-active curtain (all channels set to The effectivity of the proposed FE-EC as airborne dust bar- 0 V pulses), with the number of particles detected when the rier has been tested using the experimental setup showed curtain is active applying pulses with different amplitudes. in Fig. 6. The experimental setup aims to reproduce the The airborne dust generated depends of the size of the dust operating conditions of an industrial led illumination system particles, the air turbulences or the vibrating frequency of the exposed to an atmosphere with dust in suspension. The dust sieve. In order to compare the FE-EC efficiency of different VOLUME 8, 2020 164117 J. J. García-García: Experimental FE-EC Evaluation to Contain Airborne Particles The optimum operating frequency depends on the mass and charge of the particles, as well as on the distance between the electrodes and the voltage difference between them. Fig. 8 illustrate the results of this experimental analysis in the case of the dust sample 1. Similar tendencies can be found for the rest of the analyzed samples. VI. CONCLUSION A dust container barrier based on a three Channels scheme FE-EC has been designed, fabricated and tested showing efficiencies up to 95% for 2.4 kV square pulses operating at 10 Hz. The parallel floating electrodes constitute dust barrier transparent and permeable to air. Numerical tools have been FIGURE 7. Effect of the amplitude of the pulses applied in the FL.EC used to set the minimal distance between electrodes to mini- barrier efficiency for three different industrial dust samples: sample 1: mize the dust trajectories traveling through the FE-EC barrier. average 30 µm arabic gum spheres; sample 2: dust containing arabic gum spheres and paper dust; sample 3: paper dust sample. The FE-EC prototype has demonstrated to be more effi- cient containing dielectric particles with dimensions between 10 and 30 µm. The increment of the square pulse ampli- tude produces a clear increment of the efficiency, which is accentuated for voltages greater than 2 kV in all the evaluated samples. This effect is attributed to the polarization of the particles due to the intense electric fields produced. Work is in progress to analyze in detail this effect and to apply the technique industrially. REFERENCES [1] F. B. Tatom, V. Srepel, R. D. Johnson, N. A. Contaxes, J. G. Adams, H. Seaman, and B. L. Cline, ‘‘Lunar dust degradation effects and removal/prevention concepts,’’ NASA, Washington, DC, USA, Tech. Rep. TR-792-207A, 1967. [2] S. Masuda, K. Fujibayashi, and K. Ishida, ‘‘Electric curtain for repulsion, confinement, and transport of charged dust clouds,’’ Trans. Inst. Electr. Eng. Jpn., vol. 92-B, 1972. FIGURE 8. Efficiency tendency with the frequency increment for the Dust [3] B. S. Masuda and Y. Matsumoto, ‘‘Theoretical characteristics of standing- Sample 1with the frequency increment. Similar behavior is found for wave electric curtains,’’ Electr. Eng. Jpn., vol. 1, pp. 7–71, Dec. 1973. samples 2 and 3. No significant differences has been found for [4] S. Masuda and T. Kamimura, ‘‘Approximate methods for calculating a frequencies between 10 and 1 Hz. non-uniform travelling field,’’ J. Electrostatics, vol. 1, no. 4, pp. 351–370, Nov. 1975. [5] C. I. Calle, C. R. Buhler, J. L. McFall, and S. J. Snyder, ‘‘Particle removal by electrostatic and dielectrophoretic forces for dust control during lunar samples and conditions, the measures are normalized, and the exploration missions,’’ J. Electrostatics, vol. 67, nos. 2–3, pp. 89–92, efficiency expressed in %. May 2009. [6] R. Sharma, C. A. Wyatt, J. Zhang, C. I. Calle, N. Mardesich, and Number of particles detected M. K. Mazumder, ‘‘Experimental evaluation and analysis of electrody- with the FE − EC ON namic screen as dust mitigation technology for future mars missions,’’ η =1− (2) IEEE Trans. Ind. Appl., vol. 45, no. 2, pp. 591–596, May 2009. Number of particles detected [7] M. K. Mazumder, R. Sharma, A. S. Biris, J. Zhang, C. Calle, and M. Zahn, with the FE − EC OFF ‘‘Self-cleaning transparent dust shields for protecting solar pan and other devices,’’ Particulate Sci. Technol., vol. 25, no. 1, pp. 5–20, 2007. As it can be observed in Fig. 7 graph, the effectivity of the FE- [8] L. K. Verma, M. Sakhuja, J. Son, A. J. Danner, H. Yang, H. C. Zeng, EC as dust particle barrier reaches values greater than 95% for and C. S. Bhatia, ‘‘Self-cleaning and antireflective packaging glass for solar modules,’’ Renewable Energy, vol. 36, no. 9, pp. 2489–2493, 2.6 kV amplitude pulses. 2011. The results shown in Fig. 7 point out that the FE-EC is more [9] A. Sayyah, D. R. Crowell, A. Raychowdhury, M. N. Horenstein, and efficient with the arabic gun 30 mm spherical particles (sam- M. K. Mazumder, ‘‘An experimental study on the characterization of electric charge in electrostatic dust removal,’’ J. Electrostatics, vol. 87, ple 1) for any voltages. The efficiency improvement beyond pp. 173–179, Jun. 2017. 2 kV observed in the tree analyzed samples, is attributable to [10] H. Kawamoto and T. Shibata, ‘‘Electrostatic cleaning system for removal the generation of real charge in the particles due to the effect of sand from solar panels,’’ J. Electrostatics, vol. 73, pp. 65–70, Feb. 2015. of the intense electric fields in the FE-EC. However, more [11] M. Onozuka, Y. Ueda, Y. Oda, K. Takahashi, Y. Seki, I. Aoki, S. Ueda, and R. Kurihara, ‘‘Development of dust removal system using static electricity tests should be done to clarify this interpretation which are for fusion experimental reactors,’’ J. Nucl. Sci. Technol., vol. 34, no. 11, out of the scope of this work. pp. 1031–1038, Nov. 1997. 164118 VOLUME 8, 2020 J. J. García-García: Experimental FE-EC Evaluation to Contain Airborne Particles [12] F. Q. Friesen, B. John, C. H. Skinner, A. L. Roquemore, and C. I. Calle, [26] J. D. Duff, ‘‘Dielectrophoretic precipitation of airborne particles,’’ Ph.D. ‘‘Evaluation of an electrostatic dust removal system with potential appli- dissertation, 2013, Paper 376, doi: 10.18297/etd/376. cation in next-step fusion devices,’’ Rev. Sci. Instrum., vol. 82, no. 5, [27] H. Kawamoto and M. Adachi, ‘‘Electrostatic particle-size classification of May 2011, Art. no. 053502, doi: 10.1063/1.3587619. lunar regolith for in-situ resource utilization,’’ in Proc. 7th Symp. Space [13] H. Yang, H. Jiang, D. Shang, A. Ramos, and P. Garcia-Sanchez, ‘‘Exper- Resource Utilization, National Harbor, MD, USA, 2014, pp. 1–8. iments on traveling-wave electroosmosis: Effect of electrolyte conduc- [28] G. Liu and J. S. Marshall, ‘‘Particle transport by standing waves on an tivity,’’ IEEE Trans. Dielectr. Electr. Insul., vol. 16, no. 2, pp. 417–423, electric curtain,’’ J. Electrostatics, vol. 68, no. 4, pp. 289–298, Aug. 2010. Apr. 2009. [29] C. E. P. K. Johnson Srirama, R. Sharma, K. Pruessner, J. Zhang, and [14] P.-W. Yen, S.-C. Lin, Y.-C. Huang, Y.-J. Huang, Y.-C. Tung, S.-S. Lu, and M. K. Mazumder, ‘‘Effect of particle size distribution on the performance C.-T. Lin, ‘‘A low-power CMOS microfluidic pump based on travelling- of electrodynamic screens,’’ in Proc. IEEE Conf. Ind. Appl., Oct. 2005, wave electroosmosis for diluted serum pumping,’’ Sci. Rep., vol. 9, no. 1, pp. 341–345. Dec. 2019, Art. no. 14794, doi: 10.1038/s41598-019-51464-7. [30] H. Kawamoto, K. Seki, and N. Kuromiya, ‘‘Mechanism of traveling-wave [15] M. Aoyama, T. Oda, M. Ogihara, Y. Ikegami, and S. Masuda, ‘‘Electro- transport of particles,’’ J. Phys., vol. 39, pp. 1249–1256, Apr. 2006. dynamical control of bubbles in dielectric liquids using a non-uniform [31] M. N. Horenstein, M. K. Mazumder, R. C. Sumner, J. Stark, T. Abuhamed, traveling field,’’ J. Electrostat., vol. 30, pp. 247–258, 1993. and R. Boxman, ‘‘Modeling of trajectories in an electrodynamic screen for [16] H. Kawamoto, ‘‘Some techniques on electrostatic separation of particle obtaining maximum particle removal efficiency,’’ IEEE Trans. Ind. Appl., size utilizing electrostatic traveling-wave field,’’ J. Electrostat, vol. 66, vol. 49, no. 2, pp. 707–713, Mar. 2013. pp. 220–222, 2008. [32] M. N. Horenstein, M. Mazumder, and R. C. Sumner, ‘‘Predicting parti- [17] Y. Zhao, U.-C. Yi, and S. K. Cho, ‘‘Microparticle concentration and sepa- cle trajectories on an electrodynamic screen—Theory and experiment,’’ ration by traveling-wave dielectrophoresis (twDEP) for digital microflu- J. Electrostatics, vol. 71, no. 3, pp. 185–188, Jun. 2013. idics,’’ J. Microelectromech. Syst., vol. 16, no. 6, pp. 1472–1481, [33] J. R. Melcher, E. P. Warren, and R. H. Kotwal, ‘‘Theory for finite-phase Dec. 2007. traveling-wave boundary-guided transport of triboelectrified particles,’’ [18] A. Belgacem, A. Tilmatine, Y. Bellebna, F. Miloua, N. Zouzou, and IEEE Trans. Ind. Appl., vol. 25, no. 5, pp. 949–955, Sep. 1989. L. Dascalescu, ‘‘Experimental analysis of the transport and the separation [34] J. R. Melcher, E. P. Warren, and R. H. Kotwal, ‘‘Traveling-wave delivery of plastic and metal micronized particles using travelling waves conveyor,’’ of single-component developer,’’ IEEE Trans. Ind. Appl., vol. 25, no. 5, IEEE Trans. Dielectr. Electr. Insul., vol. 25, no. 2, pp. 435–440, Apr. 2018. pp. 956–961, Sep. 1989. [19] M. Adachi, H. Moroka, H. Kawamoto, S. Wakabayashi, and T. Hoshino, [35] N. G. Green, A. Ramos, and H. Morgan, ‘‘Numerical solution of the ‘‘Particle-size sorting system of lunar regolith using electrostatic traveling dielectrophoretic and travelling wave forces for interdigitated electrode wave,’’ J. Electrostatics, vol. 89, pp. 69–76, Oct. 2017. arrays using the finite element method,’’ J. Electrostatics, vol. 56, no. 2, [20] S. Masuda, M. Washizu, and I. Kawabata, ‘‘Movement of blood cells in pp. 235–254, Sep. 2002. liquid by nonuniform traveling field,’’ IEEE Trans. Ind. Appl., vol. 24, [36] A. Zouaghi, N. Zouzou, and L. Dascalescu, ‘‘Effect of travelling wave no. 2, pp. 217–222, Mar. 1988. electric field on fine particles motion on an electrodynamic board,’’ in Proc. [21] W. M. Arnold and G. Fuhr, ‘‘Increasing the permittivity and conductivity of IEEE Ind. Appl. Soc. Annu. Meeting, Oct. 2017, pp. 1–9. cellular electromanipulation media,’’ in Proc. IEEE Ind. Appl. Soc. Annu. [37] E. Amah, M. Janjua, and P. Singh, ‘‘Direct numerical simulation of parti- Meeting, Dec. 1994, pp. 1470–1476. cles in spatially varying electric fields,’’ Fluids, vol. 3, p. 52, Sep. 2018, [22] A. Mizuno, ‘‘Electrostatic precipitation,’’ IEEE Trans. Dielectr. Electr. doi: 10.3390/fluids3030052. Insul., vol. 7, no. 5, pp. 615–624, Oct. 2000. [38] M. Abraham, The Classical Theory of Electricity and Magnetism. [23] A. Jaworek, A. Marchewicz, A. T. Sobczyk, A. Krupta, and T. Czech, New York, NY, USA: Hafner, 1932. ‘‘Two-stage precipitators for the reduction of PM2.5 particle emission,’’ [39] H. A. Pohl, Dielectrophoresis: The Behavior of Neutral Matter in Nonuni- Prog. Energy Combustion Sci., vol. 67, pp. 233, Apr. 2018. form Electric Fields, vol. 80. Cambridge, U.K.: Cambridge Univ. Press, [24] H. A. Pohl, ‘‘The motion and precipitation of suspensoids in divergent 1978. electric fields,’’ J. Appl. Phys., vol. 22, no. 7, pp. 869–871, Jul. 1951. [40] T. B. Jones, Electromechanics of Particles. Cambridge, U.K.: Cambridge [25] R. Pethig, ‘‘Dielectrophoresis: Status of the theory, technology, and appli- Univ. Press, 1995. cations,’’ Biomicrofluidics, vol. 4, Apr. 2010, Art. no. 022811. VOLUME 8, 2020 164119