ELECTROSTATIC PHENOMENA IN GAS-SOLIDS FLUIDIZED BEDS

CHINA PARTICUOLOGY Vol 3. No 6. 395-399. 2005ELECTROSTATIC PHENOMENA IN GAS-SOLIDSFLUIDIZED BEDSHsiaotao T. biFluidization Research Centre, Department of Chemical and Biological Engineering, University of British Columbia, Vancouver, BC, CanadaE-mail: xbi@chm. ubc.caAbstract Electrostatic charges are generated by particle-wall, particle-particle and particle-gas contacts ingas-solids transport lines and fluidized bed reactors. High particle charge densities can lead to particle agglomerationparticle segregation, fouling of reactor walls and intemals, leading to undesirable by-product and premature shut-down ofrocessing equipment. in this paper, the charge generation, dissipation and segregation mechanisms are examinedbased on literature data and recent experimental findings in our laboratory. the particle-wall contact charging is found tobe the dominant charge generation mechanism for gas-solids pneumatic transport lines, while bipolar charging due tointimate particle-particle contact is believed to be the dominant charge generation mechanism in gas fluidized beds. Sucha bipolar charging mechanism is also supported by the segregation patterns of charged particles in fluidized beds inwhich highly charged particles tend to concentrate in the bubble wake and drift region behind rising bubblesKeywords electrostatics, charge generation, charge dissipation, charge segregation, gas-solids flow, fluidization1 Introductionparticle to the other, with charges dissipated from chargedparticles by electric conduction. Use of conductive or iorOne of the main problems in gas-solids fluidized beds,ized gases of opposite charges to the bed particles alsosuch as those used in drying processes and polyolefin enables the removal of charges from the particles. The netproduction, is particle agglomeration, with electrostatic change in particle charge level in a gas-solids fluidizedcharges as the key causal factor( Bailey, 1984; Cross, bed or transport line is thus a result of the balance of1987: Astbury Harper, 1999). Electrostatic forces incharge generation and dissipation, as shown in Fig. 1duced by the charges carried by particles can also change When the charge generation rate is much higher than thethe hydrodynamics of gas-solid fluidized beds. In additioncharge dissipation rate, a net charge could build up in aunintentional accumulation of electrostatic charges could fluidized bed or solids transport line, creating an electriccause hazardous electrical discharges, leading to sparks, field which imposes on individual charged particles. Con-fires and even explosions (Jones, 1999). The accumusequently, charged individual particles interact with eachlated electrostatic charges on powder particles or plastic other via the Coloumb forces, impacting on hydrodynam-films inside large commercial fluidized bed reactors can ics locally via particle-particle interaction and globally viainterfere with their performance. Despite these negative particle-electric field interactioneffects of electrostatic charging, the mechanisms of chargegeneration and dissipation are still poorly understoodagglomeration&Electrostatics is a complex phenomenon across multi-eactor foulingple scales. The charge generation at a microscale level isgenerally associated with the surface properties of parti-cles. Tribo-electrification associated with charge separation due to the contact of two dissimilar surfaces has beencommonly considered as the main cause for charge ger.wall contacteration in gas-solids handling and processing systemsBi-polar charging associated with charge separation be-contactween particles of similar materials but different sizes orsurface morphologies was also identified recently in fiuid- Fig. 1 Multiscale phenomena of electrostatic charging in gas-particleized systems with size distributions(Zhao et al., 2003Mehrani et aL., 2005). Frictional charging due to fluid-parle contact occurs when the gas adjacent to the particle The electrostatic phenomenon in gas-solids fluidizedrface is ionized due to friction-induced temperature beds is further complicated by the non-homogeneous flowrises in the near surface regionstructuredhubbles The electricCharge dissipation occurs simultaneously with charge forces中国煤化工 no longer be con-eneration. In a fluidized bed or transport line with sideCNMHGto the gas bubblesgrounded walls, charged particles may lose charges to the SimilarlyLulu UI LIyed particle bed canall during their contact with the wall Particle-particle col- no longer be treated as in a homogeneous medium. In-lision can also result in the transfer of electrons from one stead, heterogeneous flow structure at the bubble(mesoCHINA PARTICUOLOGY Vol 3. No 6. 2005scale has to be considered in order to properly understand plain why fine particle entrainment rate is reduced when thethe electrostatic forces and fields associated with charged fluidized particles are highly charged, because charged fineparticlesparticles may establish a strong attractive interaction withAt the unit(macro) scale, the reactor performance is coarse particles carrying opposite charges, preventingmpacted by electrostatics as reflected by particle ag- them from being entrained out of the fluidized bedglomeration, particle segregation, wall fouling andIf the particle-wall collision is the dominating mecha-non-uniform temperature distribution. A multiscale ap- nism for charge generation, the particle charge levelproach in which the proper understanding of charge gen- expected to decrease with increasing the column diameeration and dissipation mechanisms at the particle scale ter because the surtace area per unit of particle flux, Gs(or even the molecular or crystal scale) coupled with the (kg. m." ), is inversely proportional to the column di-distribution of charged particles at the bubble scale is re- ameter. This appears to be true in pneumatic transportquired to properly predict the impact of electrostatic lines(Napier, 1994), suggesting that the particle-wall col-charges on the overall reactor performance and to de- lision could be the dominant charge generation mechaelop effective tools for mitigating electrostatic chargesnism in pneumatic transport lines. In the gas-solids fluidBased on the information in the literature on electro- ized bed where the particle velocity in the near wall regionstatic charge generation and dissipation and the charge is much lower than in pneumatic transport lines, thesegregation in gas-solids fluidized beds and transport charge level appears to be independent of the columnlines, this paper attempts to elucidate the underlying diameter(Rojo et al., 1986), suggesting that particle-wallcharge generation and dissipation mechanisms in order to collision is not the dominant charging mechanism. Insteadidentify the dominant mechanisms under different oper- the bipolar charging due to particle-particle contact couldating conditions of gas-solids two-phase flowbe the dominant mechanism2. Charge Generation in Gas-Solids 3. Charge Segregation in Bubbling Flu-Transport Lines and Fluidized Bedsidized BedsTriboelectrification due to collision with wall surfaces ofOnce charges are built up on particles in the transporttransfer lines or reactors has been considered as the line or fluidized bed, they may not be uniformly distributedprimary source for particle charging. Extensive work has if particles are not charged uniformly or particle segrega-been reported in the literature on charge generation via tion occurs In gas-solids bubbling fluidized beds, chargecolliding single charged particles with charged surtaces distribution around a rising gas bubble was first studied byunder controlled conditions to investigate the effect of Boland and Geldart (1972)in a two-dimensional fluidizedparticle pre-charge level, surface pre-charge level, colli- bed using an induction probe. The positive and negativsion angle and collision speed on the degree of particle voltage/current signals associated with the passage of thetransfer and separation. The charge generated by colli- bubble were speculated to be caused by a positivelysion of a dielectric particle with a metal surface was found charged nose region in front of the bubble and a nega(Murata, 1994)to increase with the collision speed and to tively charged wake region behind the bubblebe at the maximum for a head-on collision with the platebut to decrease with increasing the pre-charge level of theparticle. Applying the results to pneumatic transport linesand fluidized beds the collision between the surface of thetransport lines and the reactor walls and particles is expected to generate electrostatic charges on the particlesAlthough charges can be generated on particle surfaces只A。410℃和Kdue to the friction between particles and gases, it is notexpected to play a significant role under ambient condi-tions, as shown in the recent experimental data of Me.hrani et aL. (2005). However, under high temperatures, gasionization may take place, resulting in charge transfer fromionized gases to particles during the gas-particle contact.Particle-particle collision in fluidized beds has been re- Fig. 2 Electrostatic charges registered by a colision probe duringcently found to result in charge separation, i.e., bipolarpassage of a sinale bubble in a two-dimensional fiuidizedcharging, between fine and large particles(zhao et a中国煤化工 ction model based on2003: Mehrani et aL., 2005), with fine particles charged-hen et aL 2003opposite to large particles. When fine particles are en-CNMHGtrained from the top of the fluidized bed, a net charge could Similar signals were obtained by Park et al. (2002)us-be built up in the fluidized bed. Such a non-uniform charge ing collision probes in two-dimensional fluidized beds. Thedistribution among particles of different sizes may also ex- interpretation of signals using a combined charge induc-with the experimental data could only be achieved when density distribution around rising gas bubbles in ga. 2oBi: Electrostatic Phenomena in Gas-Solids Fluidized Bedstion and charge transfer model revealed that it is not and the probe, signals received from the four probes werenecessary to have the bubble wake to be charged oppo- used, in conjunction with the bubble position capturedsitely to the bubble nose region. Instead, good agreement a synchronized digital camera, to reconstruct the chagher charge density of the same sign as the dense ids fluidized bedsemulsion phase is assigned to the bubble wake and driftFigure 3 shows the reconstructed charge distributionregion, see Fig. 2around a rising gas bubble. The charge inside the airVerification of the charge distribution surrounding rising bubble is seen to be almost zero, and the charge densitygas bubbles in gas-solids fluidized beds requires either increases gradually toward the dense phase region outdirect measurements of the specific charge density on side the bubble, with the sign of charges in the deenseparticles surrounding the bubble or indirect measurement phase remote from the bubble being negative. There is aof the electric field surrounding the bubble. Inductionmore negatively charged wake, confirming the postulateprobes, which have been widely used for the measure. based on measurements with a single collision probement of surface charge distribution with high spatial (Park et al., 2002; Chen et al., 2003). The charge densresolution, were used by Chen et al. (2005)to measure outside the bubble in the dense phase is approximatelythe electrical field induced by rising gas bubbles in. 3. 10 C kg. The charge in the wake is abtwo-dimensional Plexiglas column. With four induction6.8x10C kg. Fig 3(b)compares measured inducedprobes placed flush with the outer surface of the column charges with those calculated from the reconstructedto eliminate probe interference with the motion in the bed charge density in Fig. 3(a), confirming the success ofand charge transfer due to collisions between particles field-to-charge transformation by reconstructioninsde bubblescatter reconnectedaea=10mm,△dmM=13mm,△dch2=13mm,Ns2=4,N2=137.Nm=12×48,△da=5mm,2m=90mm列e2-90mmFig. 3 (a)Reconstructed charge distnbution and(b) companson of measured induced charge signals and simulated induced charge signals based onreconstructed charge distribution(H-0.7 m, de=0.08 m, ug=0.45 m-s. do=565 um, P, =2500 kg-m-2, Qar 1.78 m -s")(From Chen et al., 2005)The non-uniform charge density distribution surround- the surface. In large reactors only a small fraction of par-ing the gas bubble in gas-solids fluidized beds demon- ticles has the opportunity to collide with the wall to givestrated that charged particles could be distributed non- out charges while the majority of particles can only transuniformly in fluidized beds, with the highly charged parti- fer charges to neighbouring particles via particle-particlecles segregated into the bubble wake region. Such a collision. Due to the poor contact between particles in thesegregation pattern can be explained by the mechanism suspension, the electrical resistance of the suspendedthat chatre generated by particle-particle collision in particles is generally very high, even for highly conductivethe near-bubble region, and highly charged particles are particles(Pulsifer Wheelock, 1978). Therefore, groundsubsequently trapped into the bubble wake regioning of the reactor walls or transport lines is not effective indissipating electrostatic charges once charges are built up4. Charge Dissipation and Neutralizationon the particles because charges cannot be easily trans-ferred to the wall surfaces, although grounding is effectiveCharges carried by particles in gas-solids fluidization in eliminating the risk of sparkling and discharge of theand transport lines can be either dissipated by the transferical potential on theof charges to the grounded reactor wall surfaces, or neu- reacto山中国煤化工tralized by the injection of ionized gases/particles carryingCNMH Gases is effectivecharges of opposite polarity to the bed materials. Transfer removeUIlI u le pai LICle surfaces when ionizedof charges to the reactor wall depends on the level of par- gases of opposite polarity to the particle charges are usedticle charges and the collision frequency of particles with Revel et al., 2002). The application of ionized gases forCHINA PARTICUOLOGY Vol 3. No 6. 2005charge neutralization in commercial fluidized bed reactors ated with charged particles in gas-solids fluidized beds, onor pneumatic transport lines such as gas phase polym. the other hand, requires the coupling of charge generationerization reactors, however, is limited due to the potentia and dissipation mechanisms with the hydrodynamic beroduct contamination and/or catalyst poisoning by intro. haviour and particle flow and mixing patterns of the fluid-ducing the ionized gases. The increase of surface con- ized bed. As the first step in modeling and simulating suchductivity of particles and the fluidizing gases by water in- a complex phenomenon, we have identified that the trijection has been found effective in reducing charge boelectrification from particle-wall collision is the dominantbuildup in gas-solids transport lines(Guardiola et al., 1996; charging mechanism in pneumatic transport lines, whileYao et alL., 2002). The increase in surface conductivity due bipolar charging appears to be the dominant charge gento the adsorption of a thin layer of water molecules can eration mechanism in dense gas solids fluidized bedsincrease the rate of charges conducted through the sus- Highly charged particles due to particle-particle contactpended solids to the grounded reactor wall. On the other the near- bubble region tend to segregate and concentratehand, the increase in the humidity of the fluidizing gas into the bubble wake region, causing non-uniform chargemay also lower the break-down potential of the gases, density distribution. Dissipation of charges from particlehelping dissipation of particle surface chargessurfaces in fluidized beds is the rate limiting step becauseThe use of fine antistatic powders has been practised of the low conductivity of the dense particle mixture Infor decades. It is generally believed that the antistatic crease of particle surface conductivity by humidification,powders work only in the presence of certain moisture addition of ionized gases and improvement on particontent in the fluidizing or transport gases. the fine anti- cle-particle contact by the addition of fine particle are bothstatic powders first capture the moisture and then attach effective in increasing the dissipation of particle chargesemselves to the bed particles to build a moisture layeon the bed particle surface, helping the dissipation of Acknowledgementcharges by the increased effective conductivity of bedparticles. Other researches (Wolny Opalinski, 1983The support from Chinese Academy of Sciences for studies onGuardiola et a., 1992; Rowley, 2001)found that the addi- Sciences and Engineering Research Council (NSERC)of Cantion of fine particles, either dielectric or conductive, aisada for studies on electrostatic phenomena in gas-particle sys-resulted in significant reductions of charge buildup. The tems is gratefully acknowledged. The author also thanks Profpossible explanation is the lubrication of the fine powders John Grace, Dr. Aihua Chen and Dr. Poupak Mehrani for theiras"spacers"attached to bed particles so that charge dis- contributions to the studies on electrostaticssipation by particle-to-particle contact is enhanced due tothe increased particie contact, The difference between Nonmenclatureantistatic powders and fine powders is still unclear andrequires further investigation in the futureAndit ratio of charge density in the wake over charge den-sity in the dense phaseCharge generation increases with decreasing particle Dp, probe diameter of the static probe,msize for mono-sized particles due to the increase in sur- D,, da bubble diameter, mce areas( Guardiola et al., 1996; Rowley, 2001). Particle Ddiameter of the dnft following a bubble, msize distribution may also influence the electrostatic be. d, d, mean particle diameter, umhaviour of powders based on the observations of the ex. Hacked bed height, mtence of bipolar charging between fine and coarse partilength of the bubble wake and drift region, mcles and the effect of the addition of fines on the reduction K,K constants related to the rate of collision chargingof charge buildup. The implication from these findings is ninenumber of pixels used in reconstructionthat proper design of particle size and size distribution can probenumber of probes.potentially reduce electrostatic charge generation and Qairbuild-up in solids transport lines and fluidized bed reactorsspecific charge density of particles in dense phase, Ckgbased on the present exploration of dominant charge Feradius of the bubble. mbubble rise velocity, m-sgeneration and dissipation mechanisnxy,z Cartesian coordinates, z=vertical, y=horizontaWidth. m5. Concluding RemarksGreek lettersUnderstanding of the impact of electrostatics on gas- Asolids transport lines and fluidized bed reactors related toparticle agglomeration and reactor fouling requires the Refe中国煤化工xploration of electrostatAstbpation mechanisms and segregation patterns of clCNMHlargescale chemicalhazards, Electrostaticsparticles. The level of charge buildup on particles is de-1999: Proceedings of the foth international Conference ontermined by the balance of charge generation and dissiElectrostatics(pp 207-210). Cambridge, UK.pation rates. Prediction of the electrostatic forces assoBailey, A. G.(1984). Electrostatic phenomena during powderBi: Electrostatic Phenomena in Gas-Solids Fluidized Bedshandling Powder Technol., 37, 71-80Napier, D, H. (1994). Generation of static electricity in a fluidizedBoland, D.& Geldart, D.(1972). Electrostatic charging in gasbed and in powder conveying. Proceedings of 2nd Worldfluidized beds. Powder Technol. 5. 289-297.Congress on Particle Technology. Nuremberg, GermanyChen, A. H, Bi, H. T. Grace, J. R. (2003). Effects of charge Park, A H, Bi, H. T, Grace, J. R.& Chen A. H (2002). Modelingdistribution around bubbles on charge induction and transfer tocharge transter and induction in gas-solid fluidized beds. Ja ball probe in gas-solid fluidized beds. J. Electrostatics, 58,Electrostatics, 55, 135-168.91-115Pulsifer, A. H& Wheelock, T. D (1978). The electrical resistanceChen, A. H, van Willigen, F.K., Bi, H. T.Ommen, J. R.(2005). Measurementaround a rising bubble in a two-dimensof chargeJ. R. vanof gas fluidized beds. Fluidization(pp. 76-80. CambridgeCambridge University Pressidized bed. Revel, J. Gatumel, C, Dodds, J. A.& Taillet, JAlChE J, in presscharge elimination in a slugging fluidized bed. Proceedings ofCross, J.(1987). Electrostatics: Principles, Problems and Appli-the 4th World Congress on Particle Technology.cations. Bristol: Hilger.traliaGuardiola, J, Ramos, G& Romero, A(1992). Electrostatic be. Rojo, V, Guardiola, J&Vian, A(1986). A capacitor model tohavior in binary dielectric /conductor fluidized beds. Powderinterpret the electric behaviour of fluidized beds. Influence ofTechnol,73,11-19apparatus geometry. Chem. Eng. Sci., 41, 2171-2181Guardiola,J, Rajo, V& Ramos, G. (1996). Influence of particle Rowley, G.(2001). Quantifying electrostatic interactions insize, fluidization velocity and relative humidity on fluidized bedpharmaceutical solids systems. Int. J Pharm., 227, 47-55electrostatics, J. Electrostatics. 37. 1-20Wolny, A& Opalinski, I (1983). Electric charge neutralization byJones, T.B. (1999). Electrostatics and dust explosions in powderaddition of fines to a fluidized bed composed of coarse dielec-handling In Yang, W.C. (Ed ) Selected Topics on Fluidization, tric particles, J. Electrostatics, 14, 279-289Solids Handling, and Processing(pp. 817-871) Noyes Publi- Yao, L, Bi, H. T& Park, A H(2002). Electrostatic charges incations. Park Ridfreely bubbling fluidized beds with dielectric particles. J. Elec-Mehran, P. Bi, H T& Grace, J. A(2005 Electrostatic charge trostatics, 56, 183-197.generation in gas-solid fluidized beds. J. Electrostatics, 63, Zhao, H, Castle, G. S. P, Inculet, I. I.& Bailey, A.G.(2003)Bipolar charging of poly-disperse polymer powders in fluidizedMurata, Y.(1994). Mechanism of contact charging of powderbeds. IEEE Trans. Ind. App., 39, 612-618particles. NEPTIS-3: Electrostatic Problems in Powder Technology(pp 38-45), Kyoto, JapanManuscript received September 30, 2004 and accepted November fo, 2004中国煤化工CNMHG