[american institute of aeronautics and astronautics caneus 2004 conference on...

American Institute of Aeronautics and Astronautics

Collection of Non-gas Phase Airborne Nanoparticles by Microfabricated Electrostatic Precipitator

B.L. Chua* M. Zhang†

Anthony S. Wexler‡ Norman C. Tien§

Debbie A. Niemeier** University of California, Davis, CA, 95616

and

Britt A. Holmén †† University of Connecticut, Storrs, CT, 06269-2037

Airborne non-gas phase nanoparticles have been successfully collected using a relatively low power (approximately 200mW at 80µA) and small footprint (approximately 1cm2) microfabricated electrostatic precipitator. Polydispersed liquid phase oleic acid nanoparticles with size distribution of 30nm to 300nm are electrically charged and precipitated from the carrier gas phase medium at atmospheric pressure and temperature. The change in resistivity between the electrodes resulting from the build-up of precipitating nanoparticles is experimentally measured. Experimental airborne particles flux to the collection grid due to precipitation is approximately 4x104 particles per second for a concentration of approximately 1.5x106 particles per cubic centimeter.

Nomenclature Cc = Cunningham slip correction factor dp = diameter of particle D = pin to plane separation gap λ = mean free path of air at 298K εo = electrical permittivity of free space εr = electrical permittivity of particle E = applied electric field jNE, jo, j = current densities m = curve fitting parameter θ = cone angle q = total charge per particle Qp

∞ = saturation charge after t → ∞ rlimit = current limiting radius τQ = charging time constant t = residence time V = electrical drift velocity µ = dynamic viscosity of air

* Ph.D Candidate, Department of Electrical and Computer Engineering, University of California at Davis. † Postdoctoral Staff, Department of Mechanical and Aeronautical Engineering, University of California at Davis. ‡ Professor, Department of Mechanical and Aeronautical Engineering, University of California at Davis. § Professor, Department of Electrical and Computer Engineering, University of California at Davis. ** Professor, Department of Civil and Environmental Engineering, University of California at Davis. †† Professor, Department of Civil and Environmental Engineering, University of Connecticut.

CANEUS 2004--Conference on Micro-Nano-Technologies1 - 5 November 2004, Monterey, California

AIAA 2004-6731

Copyright © 2004 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.


I. Introduction irborne nanoparticles can be generally classified into two distinct size ranges: the nucleation mode (1 nm to 100 nm) and the accumulation mode (100 nm to 1000nm). Nucleation mode nanoparticles are created by

chemical or phase transformation. The formation processes include nucleation and condensation of combustion gases from anthropogenic activities. Accumulation mode nanoparticles can be created through coagulation or condensational growth of smaller particles1. Nucleation mode nanoparticles accounts for the largest number fraction while the accumulation mode nanoparticles accounts for the largest mass fraction of the overall airborne particles profile. Nanoparticles of both modes are either in liquid or solid phase, or a combination of both.

The study of airborne nanoparticles is important because of its effect on both the human health and the climate. Nanoparticles smaller than 300 nm are mainly deposited in the tracheobronchial and the pulmonary-alveolar regions of the human respiratory system upon inhalation. Potential health effects include cardiovascular abnormalities such as bronchitis and fibrosis, and cancer. In the advent of nanotechnology, where new materials manufactured via nanoscale synthesis seeks to supplant the conventional materials used in consumer and indoor products, potential nanoparticles emission from such materials and its indoor health effects may become a critical subject of study. This paper relates the application of a microfabricated corona ionizer structure2 in two key areas of interest pertaining to airborne nanoparticles and they are namely: (i) Measurement/Monitoring and (ii) Filtration/Removal. Ionization of atmospheric gases has been previously demonstrated using the above-mentioned corona ionizer structure2. Similar capability has also been reported by several other microfabricated ionizers3,4,5. However they did not demonstrate the charging of airborne non-gas phase nanoparticles. In this paper, the above-mentioned microfabricated corona ionizer structure is shown, in addition to gas ionization, to function as an electrostatic precipitator where it is able to actively gather and collect airborne non-gas phase nanoparticles from its carrier gas medium. In electrostatic precipitation, an asymmetrical electrode pair configuration (commonly pin-to-plane or wire-to-plane) is used to incept and sustain a single or multiple corona discharges. The nanoparticles charging process takes place mainly in the drift region of the corona discharge, which is away from the active (plasma) region. Material degradation of the nanoparticles is therefore minimized. Since a strong electric field is required to sustain the corona discharge, electrical migration of the nanoparticles takes place simultaneously within the drift region. For a negative corona discharge electrical precipitator, the anode also serves as the collection grid. Collection of particles is a critical feature in various applications such as high efficiency air filtration and airborne particles analysis. Compared to particle collection methods such as thermophoresis and inertia impaction, electrostatic precipitation has the broadest particle collection size range (10 nm to 100 µm) with efficiency sometimes exceeding 99.9%6. Microfabricated electrostatic precipitators can be adapted singularly for low cost, small footprint and lightweight airborne particle monitoring instruments, or in an array as high efficiency, high flowrate, low pressure loss and disposable air filter for enclosed vessels such as aircrafts and submarines or personal nuclear, biological and chemical (NBC) protection equipment. It can also be extended to collection of airborne microorganisms6.

II. Design and Principle of Operation The microfabricated electrostatic precipitator uses a

pin-to-plane electrode pair configuration. The plane is approximated by the use of a planar grid. Figure 1 shows the schematic of the electrostatic precipitator. The discharge tip has similar width and thickness of approximately 20 µm. The spacing between the discharge tip and the collection grid varies from 1.8mm to 2.2mm. The width of the collection grid is 3.5mm and its height is 1.1mm. The device is fabricated using a low-cost single lithography mask copper electroplating process on a glass substrate. Details of the microfabrication process are described in Ref. 2. A glass substrate is used due to its low cost as well as its ability to provide electrical isolation for high voltages at several kilovolts.

Negative D.C. corona discharge is used for this operation due to its relative stability as compared to positive corona discharge. The primary electron replenishment process in a negative corona discharge is the γ effect, which

A

Discharge TipGlass

Collection Grid

Figure 1. Schematic drawing of microfabricated electrostatic precipitator.


depends only on the discharge tip material and the medium gas composition. The discharge tip thus functions as the cathode and the collection grid functions as the anode.

The impact ionization of gases occurs in the immediate vicinity of the discharge tip’s surface. This impact ionization zone is referred to as the active (plasma) region, and it consists of neutrals, high-energy positive ions and electrons. However as the electrons migrate towards the collection grid (anode), it become less energetic due to the gradual decrease in the electric field strength. Therefore the impact ionization ceases as the electrons move at lower velocities. This zone consisting of low energy electrons and neutrals is referred to as the drift region.

The charging of the particles takes place within the drift region. As the airborne particles pass through the drift region, these low-energy electrons will attach themselves to the neutral particles upon contact. These electrons are driven towards the particles by both diffusion and by the virtue of the electric field between the discharge tip and the collection grid. Thus the charge density per particle and the required residence time for acquiring saturation charge are determined by both diffusion and field charging. There are established mathematical models for both diffusion and field charging. For combined diffusion and field charging, there are currently several empirical and semi-empirical models that describes it 7. Cochet’s charging model is used in this case because it accommodates all particle sizes and provides a close form analytical equation for the particle saturation charge as shown in Equation (1). Figure 1 shows the plot of particle saturation charge versus particle size. Figure 2 shows the plot of number of elementary charge carried by particle at saturation charge versus the particle size using Equation (1).

(1)

Figure 2. Plot of number of elementary charge carried by a particle at saturationcharge versus its size calculated at εr = 10 and E = 2700V for a 2mm gap at temperature of 293K.


The particle saturation charge is determined by both the particle properties and the operating parameters of the electrostatic precipitator. The time dependence7 of the particle saturation charge is given by Equation (2).

(2)

The corona current density distribution on the collection grid can be approximated using Warburg’s cosine power law8 as given by Equation (3).

(3) The corona current density approaches zero as the cone angle θ increases to 60o. The current limiting radius on the collection grid is given by Equation (4).

(4) Using Equation (4), the upper and lower bounds of the current limiting radius are calculated as 3.8mm and 3.1mm for pin to plane separation gaps of 2.2mm and 1.8mm respectively. Therefore it is reasonable to divide the operating corona current of 20µA by the collection grid area of 3.5mm2 (neglecting the grid effect) to obtain a conservative estimate for jNE. Since the charging takes place in the drift region, the relevant electric field can be approximated as a uniform field between the precipitator electrodes. For an operating voltage of 2.7 kV, the maximum value of τQ is calculated to be in the order of tens of picoseconds. Therefore it can be assumed that most of the particles that are collected by the precipitator had obtained their respective saturation charge. Particle size approximately 250nm in diameter corresponds to the lowest electrical mobility attainable. In a given electric field, it travels at the lowest electrical drift velocity. The electrical drift velocity of a particle is given by Equation (5).

(5)

The electrical drift velocity calculated for a 250nm particle under a uniform electric field by 2.7kV across 1.8mm gap spacing is approximately 500mm/s. The maximum flow velocity of the aerosol used in the experiment is in the order of 40mm/s. Since both the gap spacing and the width of the collection grid is on the same order of magnitude, it can be assumed that the variation of flow velocities in the experiment does not have significant effect on the precipitator’s collection efficiency.


III. Experimental Setup and Results The experimental setup consists of TSI 3076 constant output atomizer serving as aerosol generator and an

environmental test chamber where the precipitator is mounted as shown in Figure 2. The aerosol generated is a suspension of liquid phase polydispersed oleic acid nanoparticles in ambient air. The particle size and number distribution of the aerosol output has been measured using TSI Scanning Mobility Particulate Sizer. The particle size ranges from 30nm to 300nm and the geometric mean of the particles is 70nm. The packaged microfabricated electrostatic precipitator (ESP) is mounted on a jig. It is positioned in the sidewall of the environmental chamber to minimize deposition via means other than electrical charging and precipitation. The aerosol is passed through the environmental chamber at 1L/min via a needle valve. A diluting flow line is used to adjust the concentration of the nanoparticles within the environmental chamber via a separate needle valve. The aerosol is passed into an exhaust after exiting from the environmental chamber. The environmental chamber inlet flowrate is monitored using a flow meter. The operation of the precipitator is monitored via its corona current. The footprint of the precipitator is approximately 1cm2 and it consumes approximately 200mW at 80µA. The experiment is carried out under ambient conditions. The validation of successful collection of nanoparticles by the microfabricated electrostatic precipitator is carried by both visual observation and monitoring the resistivity change of the corona discharge “circuit”.

At the onset of the experiment, the precipitator is switched on and the corona current is adjusted to 80µA. The needle valve is subsequently opened to allow the aerosol to pass into the environmental chamber. The corona current is observed to decrease after the aerosol is introduced, and recovers after the needle valve is closed. The experiment is carried out for precipitators with electrode gap spacing of 1.8mm and 2.0mm.

Figure 3 shows the microfabricated electrostatic precipitator in operation. The corona current is observed to decrease after the introduction of the aerosol into the environmental chamber. This is due to the increase in resistivity due to accumulating oleic acid nanoparticles on the collection grid. Precipitated nanoparticles agglomerate and flows from the anode grid to its base (driven by minimization of Gibbs Free Energy and surface tension forces) as shown in Figure 4, allowing corona current recovery after aerosol exposure cutoff. This driving force is possibly a function of the volume of oleic acid residing on the grid. Figures 5 and 6 show the plot of corona current versus time elapsed for both precipitators. Both plots show varying corona currents that correspond to multiple aerosol exposure and cutoff. Forced dilution is also experimentally shown to accelerate corona current recovery as shown in Figure 7 (as indicated by the steeper recovery curve). .

Aerosol Flow

Micro ESP

EnvironmentalChamber

Jig Electrical Routing

Figure 2. Top view schematic of experimental setup within the environmental chamber.

Precipitated Oleic Acid (colored) Nanoparticles

Discharge Tip

Collection Grid

Figure 4. Photoprint of airborne oleic acid nanoparticles that were successfully charged and precipitated by the microfabricated electrostatic precipitator.

Discharge Tip CollectionGrid

Corona Active Region

Drift Region

1mm

Figure 3. Top view photoprint of the microfabricated electrostatic precipitator.


Figure 5. Plot of experimental corona current versus time elapsed for 1.8 mm device under repeated toggled aerosol exposure and recovery without forced dilution.

Figure 6. Plot of experimental corona current versus time elapsed for 2.0mm device under repeated toggled aerosol exposure and recovery without forced dilution.


The dilution flow line is used to vary the concentration of the aerosol going into the environmental chamber to provide more discrete levels of nanoparticles concentration. The aerosol flow rate from the aerosol generator remains constant at 1L/min while the dilution flow varies from 0.12 L/min to 2 L/min to provide six discrete levels of nanoparticles concentration. Figure 8 shows the experimental plot of corona current versus the time elapsed as the nanoparticles concentration varies through the six discrete levels of concentrations for 2.2mm device.

Figure 7. Experimental corona I-t plot of 2.0mm device under single aerosol exposure and forced dilution recovery. Forced dilution reduces concentration more rapidly, therefore accelerated the corona current recovery.

Figure 8. Experimental corona I-t plot of 2.2mm device for various oleic acid nanoparticles concentrations.


Although the flow velocity in the environmental chamber changes with the nanoparticles concentration, it is shown earlier that it has no significant effect on the precipitator’s collection efficiency. Therefore it is not necessary to adjust the aerosol flow rate to ensure that the flow velocity in the environmental chamber stays constant.

IV. Analysis and Discussion The nanoparticles flux to the collection grid of the precipitator for each concentration can be estimated given the corresponding resistivity changes and the nanoparticles number and size profile measured using the TSI Scanning Mobility Particulate Sizer. To simplify the calculations, the oleic acid nanoparticles are assumed to precipitate uniformly on the collection grid and the rate of change of resistivity due to the precipitation is constant. The rate of change of the precipitated nanoparticles layer thickness on the collection grid with respect to the corona current ∂(thickness)/ ∂(corona current) can be calculated by relating the added resistance of the precipitated nanoparticles layer (hence its thickness) to the corresponding corona current. The rate of change of corona current with respect to time ∂(corona current)/ ∂(time) can be derived from Figure 8 for each transition between concentration levels. Therefore the rate of change of the precipitated nanoparticles layer thickness with respect to time ∂(thickness)/ ∂(time) can be made known as well. By relating the thickness of the precipitated nanoparticles layer to its volume, the total number of nanoparticles arriving at the collection grid per unit time can thus be calculated. Table 1 shows the calculated total nanoparticles flux to the collection grid.

To improve the accuracy of the calculated total nanoparticles flux to the collection grid, a configuration specific particle transport model will have to be developed. It will probably have to include a more accurate representation of the current limiting area and the particles agglomeration effect. Significant further analytical and experimental work is required to establish the grade efficiency of the microfabricated electrostatic precipitator.

One major limitation of the microfabricated electrostatic precipitator in practical applications is the rapid build-up of collected particles on its collection grid. The build-up will possibly cause back-corona and deteriorates its performance. However it will be inexpensive to replace and dispose due to its low cost microfabrication process and non-toxic nature of the materials used.

V. Conclusion A microfabricated electrostatic precipitator has been demonstrated to charge and precipitate non-gas phase nanoparticles from it carrier gas medium. The collection of the nanoparticles has been verified experimentally via both visual observation and the change in corona current due to the added resistivity by the precipitated nanoparticles layer. The potential applications of the microfabricated electrostatic precipitator includes low cost, small footprint and lightweight airborne particle monitoring instruments, and high efficiency, high flowrate, low pressure loss and disposable air filter for enclosed vessels such as aircrafts and submarines or personal NBC protection equipment. It can also be extended to collection and removal of airborne microorganisms.

Table 1: Total nanoparticles flux to the collection grid.

Diluting Flowrate L/min Particle

Concentration #/cm3

Added Resistance MΩ

Approximate Total Nanoparticles Flux to Collection Grid #/sec

1.43 1.23x106 9 270 1.02 1.49x106 19 41 000 0.63 1.84x106 34 130 000 0.31 2.29x106 85 690 000 0.12 2.68x106 278 6 700 000


Acknowledgments The authors would like to thank Dr Yongjing Zhao and Dr Daniel T. McCormick. for their assistance in the experimental setup, Jian Wen for the TSI Scanning Mobility Particulate Sizer measurements, Seong S. Park for the valuable discussion, UC Davis Microfabrication Facility and Berkeley Sensor and Actuator Center. This project supported by the California Air Resources Board. This report was prepared as a result of work sponsored by the California Energy Commission (Commission, Energy Commission). It does not necessarily represent the views of the Commission, its employees, or the State of California. The Commission, the State of California, its employees, contractors, and subcontractors make no warranty, express or implied, and assume no legal liability for the information in this report; nor does any party represent that the use of this information will not infringe upon privately owned rights. This report has not been approved or disapproved by the Commission nor has the Commission passed upon the accuracy or adequacy of this information in this report.

References

1Seinfeld, J. H., and Pandis, S. N., Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, 2nd ed., John Wiley & Sons Inc, New York, 1997, Chap. 2.

2Chua, B. L., et al “A Unipolar Corona Discharge Microfabricated Ionizer Structure for Gases at Atmospheric Pressure and Composition”, 17th IEEE International Conference on Micro Electro Mechanical Systems, IEEE, Maastricht, The Netherlands, 2004, pp. 261-264

3Longwitz, R. G., van Lintel, H., Carr, R., Hollenstein, C., Renaud, R., “Study of Gas Ionization Schemes for Micro Devices”, 11th International Conference on Solid State Sensors and Actuators, Munich, Germany, 2001, pp. 1258-1261

4Longwitz, R. G., van Lintel, H., Renaud, R., “Micro-discharge and Electric Breakdown in a Micro-gap”, 29th EPS Conference on Plasma Phys. and Contr. Fusion, ECA, Montreux, Switzerland, 2002, pp. 2.026

5Ono, T., Sim, D. Y., Esashi, M., “Study of Micro Glow Discharge as Ion Sources”, J. Micromech Microeng., Vol. 210, 2000, pp. 445-451

6Mainelis, G, et al “Collection of Airborne Microorganisms by a New Electrostatic Precipitator”, Journal of Electrostatics, Vol. 33, 2002, pp. 1417-1432

7Parker, K. R., Applied Electrostatic Precipitation, 1st ed., Blackie Academic & Professional, London, 1997, Chaps. 1,3 8Jones, J. E., Cohen, A. M.,“Chebyshev Comparisons of Current Laws for Point-Plane DC Coronae in Air”, Journal of

Electrostatics, Vol. 39, 1997, pp. 111-128

[american institute of aeronautics and astronautics caneus 2004 conference on...

Documents