SciELO - Scientific Electronic Library Online

 
vol.17 issue1Liquid-liquid extraction by reversed micelles in biotechnological processesNew robust stable MPC using linear matrix inequalities author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Brazilian Journal of Chemical Engineering

Print version ISSN 0104-6632On-line version ISSN 1678-4383

Braz. J. Chem. Eng. vol.17 n.1 São Paulo Mar. 2000

http://dx.doi.org/10.1590/S0104-66322000000100004 

Measurement of the electrostatic charge in airborne particles: I - Development of the equipment and preliminary results

 

W.D. Marra Jr. and J.R. Coury
Departamento de Engenharia Química, Universidade Federal de São Carlos, C.P. 676,
CEP 13565-905, São Carlos - SP, Brazil

 

(Received: April 20, 1999; Accepted: August 12, 1999)

 

 

Abstract - The design and construction of a equipment capable of measuring the electrostatic charges in aerosols, named the electrostatic charge classifier, were carried out. They were based on the concept of particle electromobility and the charge classifier was intended to classify the nature and the distribution of electrostatic charges as a function of particle size. The resulting piece of equipment is easy to dismount, which facilitates its cleaning and transport, and easy to operate. Early results indicate that the values of electrostatic charge measured on test particles are inside the range reported in the literature, indicating the adequacy of the technique utilized.
Keywords: electrostatic charges, charge measurement, electromobility, aerosol particles.

 

 

INTRODUCTION

In a number of industrial processes, in which the formation, dispersion and transport of particles occurs (drying, filtration, milling or storage of powder or granules), the appearance of electrostatic charges can occur spontaneously due to friction or shock between particles (triboelectrification or impact) (Coury, 1983).

The presence of charges is sometimes desirable, as in electrostatic precipitators and electrostatic coating, while in other operations it can be undesirable, as in storage silos and in the packing of very fine powders, causing difficulty in the flow and in the manipulation of the product. The presence of charges can also be dangerous, as in the transport and storage of cereal grains or dry powders, provoking ignition and explosion of the container. Whichever is the case, information on the size distribution of the particles and of their electrostatic charges is of paramount importance for the adequate treatment of the process.

Although widely recognized as an important phenomenon in particle technology, the measurement of electrostatic charges in particles has not received a great deal of attention (see, for example, the recent review by Cross, 1987). In particular, for particles in the 1x10-6 to 100x10-6 m size range, there is, to the best of our knowledge, no established technique for charge measurement.

Our work on gas cleaning, which involves the removal of particles from gas streams in the above - mentioned size range, suggests the need to quantify the electrostatic charges on airborne particles and then to develop a piece of equipment capable of measuring them.

 

THE ELECTROSTATIC CHARGE CLASSIFIER

Theoretical background

The charge classifier relies on the concept of electromobility of the particle, Zp, defined as the relation between the particle velocity due to electric forces and the intensity of the electric field to which it is submitted (Coury et al., 1991; Guang, 1991). Electromobility is based on the fact that the trajectory of a charged particle can be altered by the presence of an electric field. A particle with charge Q in an electric field, E, is subjected to an electric force, Fe = QE. As a result of this electric force, the particle moves with velocity Ux in the direction of the force lines, as observed in Figure 1.

 

a04f01.gif (3077 bytes)

 

As the Reynolds number of the particle is usually small and the fluid and particle relaxation times are also sufficiently small, the validity of the Stokes law can be assumed in calculating the drag force, Fd, that counterbalances the electric force:

a04i01.gif (415 bytes)

where m is the gas viscosity, dp the particles' Stokes diameter and Fs the Cunningham slip factor that corrects the drag force of particles whose size approaches the mean free path of the gas molecules (Clift et al., 1978).

Assuming that Fe = Fd, one can write:

a04i02.gif (486 bytes)

For the case of parallel plane plates, placed at distance L and with a difference in electrical potential of D V and the electric field in the direction x (see Figure 1), the field strength is given by:

a04i03.gif (305 bytes)

The particle velocity in the x direction, Ux, can be determined as a function of the gas velocity, U0, the length of the deflection plates, Z, and the distance traveled by the particle in the x direction, starting from the central position, X:

a04i04.gif (325 bytes)

By substituting Equations 3 and 4 into Equation 2, one obtains:

a04i05.gif (380 bytes)

The particle charge, Q, can be obtained by substituting Equation 5 into Equation 2 and rearranging as follows:

a04i06.gif (497 bytes)

Equation 6 is strictly valid for the particles that cross all the lines of force of the electric field and for a stable velocity profile in direction z.

The equipment

The body of the charge classifier is made of acrylic of a thickness of 10 mm. The construction details and dimensions can be seen in Figures 2 and 3. The equipment is made in such a way that it can be disassembled easily, thereby facilitating its cleaning and transport.

 

a04f02.gif (20076 bytes)

 

 

a04f03.gif (5608 bytes)

 

The experimental apparatus necessary for the measurement of the gas velocity profile and the evaluation of particle behavior is shown in the Figure 4. It can be noticed that the entrance of the classifier was connected to an aerosol generator (TSI-3450) capable of generating monodisperse particles of uniform diameter from a solution of dissolved solids. A solution of methylene blue in distilled water and alcohol was used. This aerosol was injected into the classifier through the central slit.

 

a04f04.gif (4321 bytes)

 

The air was sucked through the classifier by a vacuum cleaner connected to its exit. The air velocity profile was measured with a hot wire anemometer Cole-Parmer, Tri-Sense 37000-00, whose probe was introduced into the classifier at three positions along the z axis, as illustrated in Figure 5.

 

a04f05.gif (3384 bytes)

 

A source of high voltage (Spellman, model SL-300) was connected to one of the copper plates of the classifier to promote the necessary deflecting electric field. The other copper plate was connected to the ground. The high voltage plate was located at the position x = 4.25x10-2 m and the zero voltage plate at x = - 4.25x10-2 m (see Figures 2 and 3).

The deflection of the particles was detected by counting them with a particle counter (Hiac/Royco, model 5230) fed by an isokinetic probe located after the deflecting plates. The position of the probe could be varied along the x axis, so that the particle deflection profile could be measured.

An optical Olympus BX-60 microscope, coupled to an image analyzer, was used for the measurement of particle diameter.

 

RESULTS AND DISCUSSION

Figures 6 to 8 show the velocity profiles of the air as it moved through the equipment. At Positions 1, 2 and 3 (see Figure 5), the predominant velocity values lie between 0.28 and 0.33 m/s, with averages of 0.30± 0.022 m/s, 0.30± 0.043 m/s and 0.30± 0.040 m/s, respectively. It can be observed that in the lower central area of the classifier the air flows faster than in the other areas, this being more evident at Positions 1 and 2. This is probably due to the influence of the position of the aerosol inlet, which affects the air flow. Figure 2 shows that aerosol feeding occurs at the upper side of the classifier and the aerosol entrance cone would not be long enough to even out the air flow.

 

a04f06.gif (6392 bytes)

 

 

a04f07.gif (6607 bytes)

 

 

a04f08.gif (6578 bytes)

 

The aerosol entrance was then modified for a position parallel to the longitudinal axis of the classifier, and velocity profiles were again measured for this new situation, as shown in Figures 9 to 11. In this case, the predominant velocities remain between 0.28 and 0.33 m/s at Positions 1, 2 and 3 but with averages of 0.31± 0.018 m/s, 0.30± 0.039 m/s and 0.32± 0.036 m/s, respectively. A decrease in the standard deviations, a small variation in the mean velocity throughout the classifier of around 6%, and an alteration in the behavior of the air flow are observed, indicating that the position of the aerosol entrance influences the velocity profile. The position parallel to the longitudinal axis for aerosol feeding was adopted in order to minimize any possible deposition of particles on the lower wall of the classifier.

 

a04f09.gif (5684 bytes)

 

 

a04f10.gif (6285 bytes)

 

 

a04f11.gif (6911 bytes)

 

Figure 12 exhibits the results obtained in a test performed with the use of the source of high tension. The particle count as a function of its deflection with 0 kV applied to the deflecting plates was used to determine the initial position of the particles in relation to the x axis. After this, a DV of - 5 kV was applied between the deflecting plates and new particle counting as a function of x was performed. The results are shown in Figure 12 (a). Another test was carried out with a DV of – 10 kV and compared to a counting with zero voltage (0 kV) in Figure 12 (b).

 

a04f12a.gif (3361 bytes)

a04f12b.gif (4632 bytes)

 

The x positions corresponding to the maximum particle count in the curves obtained with the application of high voltage were named xC, except in the case of zero voltage, in which the maximum was denominated x0. With the determination of positions x0 and xC, it was possible to calculate the displacement, X, of the particles for each applied tension and, with this, to determine the charge of the particle, as given by Equation 6. Table 1 lists the charge of the particle obtained for the test performed. It is worth noting that the particle charge values obtained for - 5 kV and - 10 kV are similar. The xC values were in agreement with the behavior predicted by Equation 6, since when the value of applied tension was doubled with the other variables held constant, the xC value also doubled.

 

a04t01.gif (3192 bytes)

 

The next test was performed with the use of a radioactive source, Kr-85 (TSI 3054), for elimination of the particle charge. This source was placed inside the drying column of the aerosol generator for the purpose of evaluating whether the classifier would be capable of detecting this possible decrease in the particle charge. The results obtained are shown in Figure 13.

 

a04f13a.gif (3416 bytes)

a04f13b.gif (5129 bytes)

 

A difference in the behavior of the particles can be clearly observed when the neutralizing source is not used (Fig.13a) and when the equipment is operated with the source (Fig.13b). Such evidence shows that the classifier was sensitive to variations in particle charge.

The remaining tests were performed with the same procedure described above without the use of the charge neutralizing source.

Six other tests were carried out with the classifier to determine particle charge and the results are presented in Figure 14, in which particle charge is plotted as a function of diameter. A quite interesting behavior can be observed. Initially the absolute value of the particle charge increases (i.e., it becomes more negative) as its diameter increases, until a diameter of approximately 9x10-6 m is attained; then the absolute value of the particle charge decreases as its diameter increases with a change in the sign of the particle charge, when the average diameter reaches a value of around 11x10-6 m. The observed behavior is probably connected to the particle generation process. The complex phenomena involving dissolution, evaporation and solidification of the aerosol can certainly cause the generation of residual electrostatic charges in the particles in a way that is not yet understood. In this case, in spite of the fact that the operational conditions in the aerosol generator remain seemingly constant, the particles generated would acquire different electrostatic charges, depending on their size.

 

a04f14.gif (3656 bytes)

 

The results so far reported were calculated with the use of Equation 3 and need an independent confirmation. The intention is to use a Faraday cage to accomplish these comparisons in the next stage of this work. However, it is interesting to observe that the values for the electrostatic charges found in this work are of same magnitude as those reported in the literature. For example, in the works of Coury (1983) and Guang (1991), in which the charge of fly ash particles was measured, the mean values obtained for particles with diameters between 7x10-6 and 12x10-6 m, values of -1.5x10-16 C and of -1.7x10-16 C, respectively, were reported.

It is important to note that, from the theory, the particle charge is assumed to be independent of the applied voltage (see Equations 1 to 6). This was experimentally verified as seen in Figure 14, where the measured particle charge distribution are approximately the same for -5 kV and -10 kV.

 

CONCLUSIONS

The velocity profiles for the classifier were relatively uniform, and the mean air velocity remained practically constant throughout the classifier. In principle, this value can be used in the calculation of particle charge. However, in the case of large variations in this mean value, the use of numeric profiles will probably be needed for a more accurate determination of particle charge.

The behavior of the particles was stable in relation to deflection and to location relative to the center of the cross section of the classifier. The presence of the particle charges was evident and their response to the applied electric field was as expected.

The values of the charge measured in the particles are of the same order of magnitude as the ones reported in the literature, indicating the adequacy of the equipment and of the method employed. However, a comparison of the results with an independent method (charge measurement utilizing a Faraday cage, for instance) is necessary.

 

ACKNOWLEDGEMENTS

The authors would like to thank FAPESP (proc. 96/1956-3) and PRONEX-FINEP for the financial support that make this work possible.

 

NOMENCLATURE

 

m Gas viscosity, kg.s-1.m-1

D V Difference in electrical potential, V

dp Particle diameter, m

E Electric field, V.m-1

Fd Drag force, N

Fe Electric force, N

Fs Slip factor

L Parallel plane plates distance, m

Q Particle charge, C

U0 Gas velocity, m.s-1

X Particle deflection, m

Z Length of the deflection plates, m

Zp Electromobility of the particle

 

REFERENCES

Clift, R., Grace, J. R. and Weber, M. E., Bubbles, Drops, and Particles. Academic Press (1978).

Coury, J. R., Electrostatic Effects in Granular Bed Filtration of Gases. Ph.D. diss., University of Cambridge (1983).

Coury, J. R., Raper, J. A., Guang, D. and Clift, R., Measurement of Electrostatic Charge on Gas-terminal Particles and the Effect of Charges on Fabric Filtration, TransIChemE, vol. 69, part B, pp. 97 - 106 (1991).

Cross, J. A., Electrostatics: Principles, Problems and Applications. Adam Hilger, Bristol (1987).

Guang, D., In-Situ Measurement of Electrostatic Charge and Distribution on Fly Ash Particles in Power Station Exhaust Stream. Ph.D. diss., University of New South Wales (1991).

Marra Jr., W. D. and Coury, J. R., Estudo da Fluidodinâmica de um Equipamento para Medida da Eletromobilidade de Aerossóis: Resultados Preliminares , Proceedings of the XXV Congresso Brasileiro de Sistemas Particulados-ENEMP, São Carlos (SP), pp. 179-184 (1997).

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License