Frequency comparative study of coal-fired fly ash acoustic agglomeration

Availableonlineatwww.sciencedirect.comJOURNAL OFScienceDirectSN1001-0742CN11-2629JESJournal of Environmental Sciences 2011, 23(11)1845-1851wwJesc,ac cnFrequency comparative study of coal-fired fly ash acoustic agglomerationJianzhong Liu,, Jie Wang, Guangxue Zhang, Junhu Zhou, Kefa Cen'1. State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027, China. E-mail: jzliu@zju.educInstitute for Energy Engineering, China Jiliang University, Hangzhou 310018, chinaReceived 18 December 2010: revised 20 May 2011; accepted 30 May 2011AbstractParticulate pollution is main kind of atmospheric pollution. The fine particles are seriously harmful to human health and environment.Acoustic agglomeration is considered as a promising pretreatment technology for fine particle agglomeration. The mechanisms ofacoustic agglomeration are very complex and the agglomeration efficiency is affected by many factors. The most important andcontroversial factor is frequency. Comparative studies between high-frequency and low-frequency sound source to agglomerate coalfired fly ash were carried out to investigate the influence of frequency on agglomeration efficiency. Acoustic agglomeration theoreticalanalysis, experimental particle size distributions(PSDs) and orthogonal design were examined. The results showed that the 20 kHzhigh-frequency sound source was not suitable to agglomerate coal-fired fy ash. Only within the size ranging from 0. 2 to 0. 25 um theparticles agglomerated to adhere together, and the agglomerated particles were smaller than 2.5 um. The application of low-frequency(1000-1800 Hz) sound source was proved as an advisable pretreatment with the highest agglomeration efficiency of 75.3%, and all thenumber concentrations within the measuring range decreased. Orthogonal design L16 (4 was introduced to determine the optimumfrequency and optimize acoustic agglomeration condition. According to the results of orthogonal analysis, frequency was the dominantfactor of coal-fired fly ash acoustic agglomeration and the optimum frequency was 1400 HzKey words: coal-fired fly ash; acoustic agglomeration; frequency; agglomeration kernel; orthogonal designDOl:10.1016s1001-0742(10)606523Citation: Liu J Z, Wang J, Zhang G X, Zhou J H, Cen K F, 2011. Frequency comparative study of coal-fired fly ash acousticagglomeration Joumal of Environmental Sciences, 23(11): 1845-1851Introductionasthma(Chow et al., 1994). The current conventionalparticle filtering devices, such as electrostatic precipitatorsAir pollution has become increasingly concerns as the cyclone separators, bag houses and wet scrubbers, haveeconomic development(Bi et al., 2007; Fenger, 2009). higher total dust removal efficiency for the particles largerParticulate pollution as a main type of atmospheric pollu- than 5 um, but fall short of retaining very fine particulatetion is seriously harmful to human health and environment matter. The pretreatment of flue gas before entering fil-( u et al., 2010; Lee et al, 2007: Pope et al., 2002). tering devices is required to enlarge the particle averageIt has raised a great concern in international community size to the point where the conventional filter operates(Sandstrom et al., 2005; Seaton et al., 1995). Particulate In this way, it can substantially improve the fine particlepollution is mainly caused by the burning of fossil fuels removal efficiency of the conventional device. Among(Liu et al., 2006; Zhang et al., 2008), vehicle exhaust other innovative approaches for small-particle filtratioremissions(Clarke et al, 1996; Guo et al., 2010; Zhang et techniques, acoustic agglomeration has recently evolvedal., 2007), industrial processes( Gu et al., 2010)and other an efficient method for controlling small particle emissionshuman activities(Lewtas, 2007; Li et al., 2009a). Total prior to the actual filtration stage( boulaud et al, 1984suspended particulates(TSP)and particulate matter(PM) Gallego-Juaezet al., 1999; Reethof, 1986). The applicationwith the aerodynamic diameter less than 10 um(PM1o) of acoustic waves to the aerosol forces particles to movecontain many toxic substances, such as polycyclic aromatic toward each other and causes collisions between them.hydrocarbons and heavy metals(Liet al., 2009b). They are Once they collide, the particles are likely to adhere andrecognized as a serious public health concern, especially form larger ones due to the meshing of their irregularfor the smaller particles with aerodynamic diameter less surface and thetraction forcethan 2.5 um(PM25). These particles can travel deeper into The newly bor中国煤化alveolus even into the capillaries and have been linked particlesCNMHGprocessto premature deaths, chronic bronchitis and aggravated place simultaneously among all particles. The acousticsCorrespondingauthor.E-mail:jzliu@zju.edu.cnagglomeration results in a significant shift towards larger846Journal of Emvironmental Sciences 2011, 23(11)1845-1851/ Jianzhong Liu et al.vol. 23sizes in a short while of the particle size distributions reached a consensus on the value of fopt(PSDs).In this article, we utilized the acoustic agglomeration toA lot of theoretical and experimental work has been deal with the coal-fired fly ash, and comparative studiesdone on acoustic agglomeration process(boulaud et al. between high-frequency and low-frequency sound source1984: Hoffmann et al, 1993; Lee et al 1982; Liu et al., were carried out to investigate the influence of frequency2009; Shaw and Tu, 1979; Song et al., 1994; Tiwary and on agglomeration efficiency. Acoustic agglomeration the-Reethof, 1987; Wang et al., 2011), as the mechanisms oretical analyses, experimental particle size distributionspeculiar to such process are very complex and their (PSDs)and orthogonal design were examineddescriptions are not completely established. Orthokineticinteraction is based on the entrainment of different-size 1 Experimental setupparticles by the intensive sound wave in polydisperseaerosol(Cheng et al., 1983), and it is widely recognized as The acoustic agglomeration system( Fig. 1)consists ofa primary acoustic mechanism. Hydrodynamic interactions agglomeration chamber, sound source fly ash and seedresults from the mutual infuence of particles due to the particle feeding, aerosol sampling and measurement.Thenonlinear interaction of scattered waves (lee et al., 1982), agglomeration chamber is made of a vertical tube withand it becomes particularly important to the monodisperse an inside diameter of 99 mm and a length of 1500 mm.aerosol with the similar-size particles wherein orthokinetic The sound source is installed on the top of the chamber.interaction nearly vanishes due to the absence of relative Two sets of sound sources with high-frequency and lowoscillatory motion. In the acoustic agglomeration process, frequency are studied. The high-frequency sound sourcethe agglomeration efficiency is affected by many factors, consists of high-power piezoelectric ceramic transducersuch as frequency, sound pressure level(SPL), aerosol (ZJS-2000, Hangzhou Success Ultrasonic Equipment Co.,initial concentration, resistant time. The agglomeration Ltd, China)and ultrasonic control. The ultrasonic emis-effect is a result of mutual constraint, mutual infuence and sion area is 50 cm2 and SPL can be up to 150 dB incoordinated development of these factors.the atmosphere. The low-frequency sound source is mades. The most important and controversial factor is frequen- up of a horn and compression driver(YF-513, Yong FaAccording to the orthokinetic mechanism there is an Electronics Co, Ltd, China), with the maximum inputoptimal frequency, fopt. If the frequency is much lower power of 80 w and the frequency ranging from 180 tothan fopt, the suspended particles are totally vibrated with 5500 Hz. The compression driver is combined with ansound at the same amplitude; if the frequency is much amplifier(QSC RMX2450, QSC Audio Products, LLC,higher than fopt, the particles remain stationary and no USA), which is powered by a signal generator(Goodwillagglomeration takes place. Based on the mechanism of SFG-1013, Good will Instrument Co.. Ltd, Taiwan ).Ahydrodynamic interaction, the forces between the particles foam rubber is placed at the bottom of the agglomerationcan be enlarged as the frequency increases. The infiuence chamber to prevent the sound wave reflection to ensureof frequency has been investigated by many researchers. a relatively homogeneous travelling wave field in theCheng et al. 1983)performed an experimental investiga- chamber Three acoustic measurement points are set alongtions of acoustic agglomeration with the frequencies from the chamber wall at regular intervals.600 to 3000 Hz to deal with NHa CI aerosol and the results To simulate the testing aerosol similar to the coal-firedproved that the agglomeration effect was very sensitive flue gas, the coal-fired fly ash particles are collected fromto the frequency and the optimum frequency was 3 kHz. an electrostatic precipitator of a coal-fired boiler.TheseTiwary(1985)utilized the acoustic agglomeration to deal particles are continuously given by the micro feeder,andwith coal-fired boiler flue gas and found that the optimum then they are mixed with clean air-stream in the Venturifrequency was 2 kHz. Hoffman et al.(1993)combined the mixer. Coarse particles in the initial aerosol are removed by44 Hz low-frequency sound with the limestone with mean a cyclone with a cut-diameter of 10 um. Then the aerosoldiameter of 88 um as absorbent to agglomerate the coal-fired fy ash and observed a good agglomeration effect.CompressorCaperan et al. (1995)used 10 and 21 kHz sound sourcesto agglomerate the glycol fog aerosol and achieved a betterefficiency with the frequency of 21 kHz than that of 10amplifiergeneratorkHz. Gallego-Juarez et al. 1999)developed an agglom-feedereration chamber driven by four high-power and highlyAgglomerationBlowerdirectional acoustic transducers of 10 and/or 20 khz to dealwith the fume generated by a fluidized bed coal combustor,mixerand found that using the 20 kHz sound source could get aELPIbetter agglomeration effect. Riera-Franco de et al. (2000)utilized the 10 and 20 kHz sound sources to agglomeratethe submicron particles in the diesel exhaust and found thatV凵中国煤化工_ Bag houseusing the agglomeration effect of 20 kHz was better thanCNMHGmthat of 10 kHz. Overall the investigated frequency variedfrom 44 to 21000 Hz; however the conclusion has not yetFig. 1 Schematic diagram of acoustic agglomerationNo.11Frequency comparative study of coal-fired fy ash acoustic agglomeration1847enters the agglomeration chamber. Residence time ranging0.5from 3 to 7 sec are controlled by the flow ranging from 6 to1.0pm14 m/hr. Electrical low pressure impactor(ELPI; Dekati0.8-25pmtype, Finland) is installed at the outlet of the chambermeasure the aerosol real-time PSDs. it measures the0.6particle size distribution in the size range 0. um with +t12 channels. Before entering the ELPl, the aerosols arediluted by a two-stage diluter(DEKATI DI-1000; Dekatitype, Finland) at the ratio of 1/64.All the experiments are carried out at ambient temture(25C), and all the measurements are performedthe system is operated steadily for more than 10 min102 Results and discussionf(Hz)Fig 2 Influence of frequency on entrainment factor (HR)with2.1 Theoretical analysesdifferent particle sizes ranging from 0.05 to 10 um. up is particleoscillating velocity.The interaction of a single spherical particle suspendedEg.(2)can find that the entrainment factor is related tofreely in aerosol through which a plane acoustic wave many factors. While in the acoustic agglomeration theISpropagating leads two interesting results: one is calledparticle entrainment, which means the particle gains mo-density and the dynamic viscosity of the gas medium arementum from the wave and attempts to oscillate with the considered as constants, and the particle density is alsogas medium; the other is called wave scattering, which considered as constant, therefore, the entrainment factor ismeans the incident wave is partially scattered from themainly related to the frequency and particle size.surface of the particle.In this experimental condition, the particle density is2500 kg/m, and the gas dynamic viscosity is 1. x 10-7The local acoustic velocity u can be calculated by pa-sec, thus the influences of frequency on entrainmentEq.(1)factor with different particle sizes ranging from 0.05 to 10Re(voe o)(1) um are calculated(Fig. 2)The entrainment factor becomes lower as sound fre-where, Re means the real quantity of an unreliable figure, quency becomes higher. It means the particle vibrationUo is the velocity amplitude of the gas medium, w is the speed is smaller. There are two threshold values fsmin andangular frequency, t is the time, and j is the imaginary fsmax for the frequency infuence on entrainment. Whennumber. Assuming Uo is much smaller than the sound the frequency is lower than fsmin, the entrainment factor isspeed, and the particle diameter, dp, is much smaller than close to 1 and the particle oscillates fully with the mediumthe acoustic wavelength, the particle oscillating velocity, When the frequency is larger than fsmax, the entrainmentup,can be calculated by Eq(2)(Song et al., 1994).factor is close to 0 and the particle is almost at rest. Itup=Re(HUge"juryalso can be found that the frequency thresholds become(2)larger for smaller particle. When the particle size is smallerwhere,H called the complex entrainment function is than 0.05 um, the entrainment factor is very close to 1defined as the ratio of the complex velocity amplitude of It means no matter how much the frequency is, the finethe particle entrainment. The magnitude of H is zero when held. For the particle with the size ranging from I to 2.5the particle remains motionless and the magnitude of H is um, the entrainment factor will decrease rapidly when theunity when the particle oscillates fully with the mediumfrequency is larger than 1000 Hz. If the particle size isThe acoustic entrainment factorhich is defined as larger than 5um, the downward trend of entrainment factorthe ratio of the particle amplitude to gas amplitude in the is more obvioussound wave, can be calculated by Eq (3)(Temkin, 1994).In the aerosol dynamics calculation, agglomeration kernel K is used to describe the collision and agglomerationv=0(3 probability of two particles. It is defined as the number ofagglomeration times of two particles, a and b, in the unithere, Ta is the particle dynamic relaxation time, given by time, unit volume and unit particle number concentration,which can be expressed by Eq. (5)Eq(4)(Tiwary and Reethof, 1987)K(a, b)=dNnanp(4) whereN isv凵中国煤化工and nb is thenumber conceCNMHGparticle b.Towhere, Pp is the particle density, and u is the dynamic simulate the actual agglomeration three assumptions areviscosity of the gas medium. Combining Eq.(1)with introduced. First, assuming the particles in the aerosol can1848Journal of Environmental Sciences 2011, 23(11)1845-1851/ Jianzhong Liu et alVol. 23da and ds(da>do). The larger particle plays a role as the ne compared K(a, b)with the modified agglomeration ker-be seen as spheres and also can be divided into two sizes,k(a, b), it has some slight changes, but the trends arecore. The cylindrical area around the core is considered as similar. To simplify the analysis of the frequency influence,the agglomeration volume. Second, assuming the particle the simplified model is still viablenumber concentration in the agglomeration volume is the After the application of agglomeration, the initial parsame as in the gas medium. It ensures that the particles ticle number concentration is expected to decrease. Thecan fill into the agglomeration volume in an instant. Third, agglomeration efficiencies, m, is calculated by Eq (10)assuming all the smaller particles can collide with the coreparticle and all the collisions lead to the agglomeration. nux100%Therefore, the agglomeration volume can be calculated byEq(6)(Sheng and Shen, 2006)where, N is the particle number concentration, subscripts+db)2(60 and I represent before and after acoustic agglomeration,And the number of small particles within the agglomer2.2 Influence of high-frequency sound on agglomeraation volume can be calculated by Eq (7):tion effect(7)The application of 20 kHz high-frequency sound sourceNa= vabnachanges the particles size distribution immediately(Fig 4)According to the definition of agglomeration kermel K,The initial PSD shows as bimodal distribution and thecombining of Eqs. (3), (6)and(7), K(a, b)can be expressed peak particle diameters are 0.071 and 0. 76 um respectivelyby Eq ( 8)The application of high-frequency sound with the SPLof 150 dB, for the particles with the size smaller than0.201 um the number concentration decreases, and for theK(a, b=ab7-4(a+4b)2particle with the size from 0. 201 to 1.945 um the numberVu+(ar a)(1+(wra concentration increases. The submicron particles presum(8) ably agglomerate to larger ones in the high-frequencysound feld. The calculation results show that the numberTwo sets of simulation results of the agglomerationkemels for micro with sub-micro particles are compared initial of 3.37x 105 cm-3 to the final of 3.02x 105 cm-3The peak frequency can be considered as fopt. FigureFigure 4b gives another example of the high-frequency3a shows that fopt of sub-micron and micron particles is acoustic agglomeration with the different initial PSDshigher than 10 Hz, and fopt becomes higher as the particles When the SPL of the high-frequency sound field is smaller,become smaller. Compared with Fig 3b, it can find that foptthe agglomeration efficiency is a little lower, but theof the larger particles is much lower than that of submicron anation tendency of PSDs is similarand micron particles. Actually not all the smaller particlesFurther study finds that the application of highnear the core particle move straightly until hitting thefrequency sound decreases the total aerosol concentrationcore particle, the airflow detours the core particles and the but the agglomeration effect is slight; besides only thetrajectories of fine particles will deflect from the straight particles with the size smaller than 0.25 um adhere tolines. Collision efficiency e is introduced, and defined as agglomerate, and the sizes of the agglomerated particlesthe proportion of fine particles colliding with core particle are still smaller than 2.5 um. It can be explained by(Zhang et al., 2009). The agglomeration kemel K can be entrainment factor as shown in Fig. 2. When the soundamended as K(Eq (9))frequency is over 10 Hz, for the particle with size largerthan 2.5 um the entrainment factor is very close to 0.K(a, b)=Ek(a,b)(9) It proves that these larger particles are motionless in theo-d2=05μm;db-2μm2.51.0中国煤化工101021030102CNMHGIOf(Hz)/(Hz)Fig-3 Influence of frequency on agglomeration kernel for particles with different sizesNo.11Frequency comparative study of coal-fired fly ash acoustic agglomeration1849No soundNo sound20kHz150-20 kHz, 145 dB当8100.01Fig 4 Particle size distribution(PSDs)without and with high-frequency sound application.20 kHz sound field and no relative motion can cause the 75.3% and 62.7% respectively. It proves that the low-collision or agglomeration. Nevertheless for micron or frequency sound is suitable to agglomerate the coal-firedsubmicron particles with different sizes, the differences fly ash particles.of entrainment factors are obvious. It means that in the Based on previous experimental and theoretical analy-20 kHz sound field the submicron particles collide with ses, the influence of frequency is non-linear and the valueeach other to adhere together. It can also be proved by of the fop is the most important and complex factor forthe agglomeration kernel. As shown in Fig 3, the fopt for the operating condition of acoustic agglomeration. If thesubmicron and micron particles is around 20 kHz, while frequency deviates from fopt, the agglomeration efficiencythe fopt for larger particles is much lower than 20 kHz. will decrease(Shuster et al., 2002). Other factors, suchIn the industrial application, the fine particles are required SPL and residence time, also affect the acoustic agglomerto agglomerate larger than PM2.5 for the conventional ation efficiency. The infuences of these factors mentionedequipments to remove from the flue gas. The application above are not necessarily linear and may interact uponof 20 kHz high-frequency sound source has some slight each other. The orthogonal design form L16(4) withgglomeration effects on micro or submicron particles, but total of sixteen tests is introduced to find out the optimumthe agglomerated particles are still smaller than PM2.5. frequency and optimize acoustic agglomeration condition,Therefore high-frequency sound source is not an advisable including frequency, SPL and residence time. If a tradition-pretreatment for coal-fired fly ash.al full factorial design is employed to examine the effects2.3 Influence of low-frequency sound on agglomeration of three factors, each at four levels, a total of 64(4)testseffecthave to be conducted. It is obvious that the orthogonaldesign can significantly reduce the experimental workchange.application of low-frequency sound source also load.changes the particles size distribution immediately, but According to our previous study(Liu et al., 2009), thecompared with the application of high-frequency sound optimum frequencies for coal-fired Ay ash were in thesource, the PSDs varies wildly(Fig. 5)narrow range of 1400-1700 Hz. To achieve an efficientCompared with Fig 5a and b, the initial aerosol PSDs agglomeration, an SPL higher than 140 dB was requiredare different, but after the applications of low-frequency As higher SPL consumed more energy, SPL of 140-150dBsound the particle concentrations decrease for all the size was suitable. Residence time with the order of 1 sec wasrange 0.03-10 um with 12 channels. According to exper- practicable for commercial industrial applications. In thisimental results the total number concentrations decrease study, these three factors are at the four levels as follows:No sound30-160z5ldB富0.500L0.0中国煤化工CNMHGJourmal of Environmental Sciences 2011, 23(11)1845-1851/ Jianzhong Liu et alvl,23frequency of 1000, 1400, 1600 and 1800 Hz, SPL of 135, eration alone are as follows: f= 1400 Hz, SPL 150 dB,140. 145 and 150 dB. residence time of 4, 5, 6 and 7 sec. =4 sec.Table 1 shows the experimental and difference analysis Besides high-frequency sound approach has some drawresults of the orthogonal design.backs,such as the high energy consumption of theBased on the values of R, it proves that the influences of transducers, difficulties to penetrate into larger gas vol-requency, SPL and residence time on the agglomeration umes, and costly sound source technology. Low-frequencyefficiency decreases in the order: f>SPL >t. The fre- sound field are much cheaper to design and easier to oper-quency is the dominant factor on acoustic agglomeration. ate at high power than that of high-frequency counterpartsAccording to the level effects of influence, SPL and Also at lower frequency the acoustic penetration depth isresidence time( ig. 6), the influence of frequency is non- much higher. Therefore the application of low-frequencylinear within the range from 1000 to 1800 Hz, and the peak an advisable pretreatment for the conventional equipmentsfrequency is 1400 Hz. and the agglomeration efficiency to agglomerate the coal-fired fly ash.is higher as the SPL became larger. High SPL favors theagglomeration effect because larger displacement of the 3 Conclusionsparticles and stronger acoustically induced turbulence canbe expected. The influence of residence time is relatively In this article, comparative studies between high-small. It should be noted that too long residence time does frequency and low-frequency acoustic agglomeration arenot facilitate the improvement of agglomeration efficiency, carried out to investigate the frequency influence onas it ofers the opportunities for the agglomerated particle acoustic agglomeration. Theoretical analysis, experimentalto break up because of mutual collisions. Based on the PSDs and orthogonal design are examinedabove analyses, the optimal conditions of acoustic agglom- The results prove that although the application of 20 kHzhigh-frequency sound source decreases the total numberTable 1 Experimental and analysis results of the orthogonal designconcentration of the aerosol about 10%. the Psds showFactorsAgglomeration that only the particles within the size ranging from 0.2(A) (B)SPL (C)t efficiency(9) to 0.25 um agglomerate to adhere together, and the ag-(sec)glomerate particles are still smaller than PM2.5. It means28.l5high-frequency sound source is not suitable to removal100025.82coal-fired fly ash particles24.74567547667423.77The application of low-frequency(1000-1800 Hz)ge1400a better agglomeration efficiency for the coal-fired flyash acoustic agglomeration and all the number concen-trations within the measuring range decrease. The highest1400514116013529.44obtainable agglomeration efficiency is up to 75.3%. The3753comparative study proves that low-frequency acoustic41.68agglomeration is an advisable pretreatment for the conven-123456koval coal-fired fly ash parti28.183093Orthogonal design is introduced to investigate thefluences of frequency and other factors on agglomeration180015040.21efficiency. According to the orthogonal analysis results, the1024841234814937frequency is the dominant factor. The optimal conditions17331133.62138693kR148.57146.52136.53are as follows: f= 1400 Hz, SPL 150 dB, t=4 sec.134.57Acknowledgments70.8331.821503Optimal level 140This work was supported by the National Basic Re-aw=agglomeration efficiency at A, RA=max()-minleh. search Program (973)of China(No. 2010CB227001),the National Natural Science Foundation of China (No50576083), the Program New Century Excellent TalentsUniversity(No. NCET-04-0533and the Zhejiang Provincial Natural Science Foundation of China(No. Y1100299 ).ReferencesBi X H, Feng Y C, Wu J H, Wang Y Q, Zhu T, 2007apportionment of PMio in six cities of northBoulaud D. Fr中国煤化工 herbe,1984CNMHation and pref(Hz)SPL(dB)cipitationwin u acron Science, 15(3):Fig 6 Level effects of frequency () SPL and residence time(r)47252Frequency comparative study of coal-fired fly ash acoustic agglomeration1851Capean P. Somers J, Richter K, Fourcaudot S, 1995. Acoustic Liu Z, Zhang GX, Zhou J H, Wang J, Zhao WD, Cen K F, 2009agglomeration of a glycol fog aerosol: influence of particleExperimental study of acoustic agglomeration of coal-firedconcentration and intensity of the sound field at two fre-fly ash particles at low frequencies. Powder Technologyquencies. Journal of Aerosol Science, 26(4): 595-612.193(1):20-25Cheng MT, Lee PS, Berner A, Shaw DT, 1983. Orthokinetic ag- Pope C A Ill, Burnett R T, Thun M J, Calle EE, Krewski D,glomeration in an intense acoustic field. Journal of ColloidIto K et al., 2002. Lung cancer, cardiopulmonary mortality,and interface Science, 91(1): 176-187.and long-term exposure to fine particulate air pollutionChow J C, Watson JG, Fujita E M, Lu Z Q, Lawson D R,The Journal of the American Medical Association, 287(9)Ashbaugh L L, 1994. Temporal and spatial variations of1132-1141PM2s and PM1o aerosol in the Southern Califomia air Reethof G, 1986. Acoustic agglomeration of power plant flyquality study. Atmospheric Emvironment, 28(12): 2061sh for environmental and hot gas cleanup. Acs Symposium2080.Series,319:234252.Clarke A G. Chen J M, Pipitsangchand S, Azadi-BougarG Riera-Franco de s e, Gallego-Juaez J A, Acosta-Aparicio V M,A, 1996. Vehicular particulate emissions and roadside airRodriuez-Maroto JJ, Dorronsoro J L, Sanz-Rivera etpollution. Science of the Total Environment, 189-190: 417-al, 2000. Acoustic agglomeration of submicron particles indiesel exhausts: First results of the influence of humidityDu X, Kong Q, Ge W H, Zhang SJ, Fu L x, 2010. Charat two acoustic frequencies. Journal of Aerosol Science,acterization of personal exposure concentration of fine3l(Supl.l):827-828.particles for adults and children exposed to high ambient Sandstro T, Cassee FR, Salonen R, Dybing E, 2005.Recentconcentrations in Beijing, China. Journal of Environmentaloutcomes in European multicentre projects on ambientsciences,2(l):1757-1764.particulate air pollution. Toxicology and Applied PharmaFenger J, 2009. Air pollution in the last 50 years-from local tocology207(2, Suppl..1):261-268.globaL. Atmospheric Emvironment, 43(1): 13-22.Seaton A, Godden D, MacNee W, Donaldson K, 1995. Partic-Gallego-Juaez J A, de Sarabia E R-F, Rodriuez-Cormal G, Hoff-ulate air pollution and acute health effects. The lancermann T L, Gavez-Moraleda J C, 1999. Application of345(8943):176178.acoustic agglomeration to reduce fine particle emissions Shaw dT, Tu K w, 1979. Acoustic particle agglomeration due tofrom coal combustion plants. Environmental Science Ghydrodynamic interaction between monodisperse aerosols.Technology,33(21):3843-3849Joumal of Aerosol Science, 10(3): 317-328Gu Z P, Feng J L, Han W L, Li L, Wu M H, Fu J M, Sheng C D, Shen X L, 2006. Modelling of acoustic agglomSheng G Y, 2010. Diurnal variations of polycyclic aromaticeration processes using the direct simulation Monte Carlohydrocarbons associated with PM2.s in Shanghai, Chinamethod, Journal of Aerosol Science, 37(1): 16-36Journal of Environmental Sciences, 22(3): 389-396Shuster K, Fichman M, Goldshtein A, Gutfinger C, 2002. AgGuo X R, Cheng s Y, Chen D S, Zhou Y, Wang H Y, 2010glomeration of submicrometer particles in weak periodicEstimation of economic costs of particulate air pollutionshock waves. Physics of fluids, 14(5): 1802-1805from road transport in China. Atmospheric Environmenong L, Koopmann G H, Hoffmann T L, 1994. An improved44(28):3369-3377theoretical model of acoustic agglomeration. Journal ofHofimann T L, Chen W, Koopmann G H, Scaroni Avibration and Acoustics-Transactions of the Asme, 116(2)ong L, 1993. Experimental and numerical analysis of208-214.bimodal acoustic agglomeration. Journal of vibration and Temkin S, 1994. Gasdynamic agglomeration of aerosols. IAcoustics-Transactions of the Asme, 115(3): 232-240.Acoustic waves. Physics of fluids, 67): 2294-2303Lee J T, Son J Y, Cho Y S, 2007. The adverse effects of fine Tiwary R, 1985. Acoustic agglomeration of micron and submiparticle air pollution on respiratory function in the elderlycron fly-ash aerosols. Ph. D. Thesis. The Pennsylvania StateScience of The Total Environment, 385(1-3): 28-36UniversityLee P S, Cheng M T, Shaw D T, 1982. The influence of hydro- Tiwary R, Reethof G, 1987. Numerical-simulation of acous-dynamic turbulence on acoustic turbulent agglomeration.tic agglomeration and experimental-verification. JoumaAerosol Science and Technology, 1(1): 47-58of Vibration Acoustics Stress and Reliability in DesignLewtas J, 2007. Air pollution combustion emissions: characteri-Transactions of the Asme, 109(2): 185-191zation of causative agents and mechanisms associated with Wang J, Liu JZ, Zhang GX, Zhou J H, Cen K F, 2011. Orthogo-cancer, reproductive, and cardiovascular effects. Mutationnal design process optimization and single factor analysisResearch/Reviews in Mutation Research, 636(1-3):95-for bimodal acoustic agglomeration. Powder Technology210(3):315-322.Li XR, Guo X Q, Liu X R, Liu C S, Zhang SS, Wang Y Zhang G X, Liu J Z, Zhou J H, Cen K F, 2009. Simulation andS, 2009a. Distribution and sources of solvent extractableanalysis of collision efficiency in acoustic agglomerationorganic compounds in PM2s during 2007 Chinese SpringJournal of Chemical Industry and Engineering, 60(1):42-Festival in Beijing. Jounal of Environmental Sciences,21(2):142-149Zhang G X, Liu J Z, Zhou J H, Ye L, Zhang B S, Cen K F,Li X L, Zhang Y X, Tan M G, Liu J F, Bao L M, Zhang2008. Size distribution and emission characteristics of in-G L, Li Y, lida A, 2009b. Atmospheric lead pollutionhalable particulate from coal water slurry boiler. Journal ofin fine particulate matter in Shanghai, China. Joumal ofZhejiang Universiry( Engineering Science), 42(10): 1817-Environmental Sciences, 21(8): 1118-1124.Liu J Z, Zhang G X, Zhou J H, Fan H Y, Cen K F, 2006. Forma- Zhang R, Cao中国煤化工,200.Cartion and micromorphology characteristics of fine particlesbonaceousCN MH Ggases in wintergenerated during coal combustion. Journal of Chemicalin Beijing.tences,19(5):564Industry and Engineering(China), 57(12): 2976-2980.