Bipolar Neutralization using Radioactive, X-ray, and AC Corona Methods

Transcription

1 Bipolar Neutralization using Radioactive, X-ray, and AC Corona Methods Jacob Swanson, Jean de La Verpillière, and Adam Boies University of Cambridge 2013 Cambridge Particle Meeting 1

2 Outline Motivation Background charge fraction Experimental charging conditions Neutralizers Carrier gas Results Size distributions Ion mobility Charged fraction Summary and conclusions 2

3 CPC counts Motivation particle size distribution measurements DMA CPC Series1 neutralizer Time, s 1.E+05 dn/dlogdp, part/cm 3 8.E+04 6.E+04 4.E+04 2.E+04 0.E Dp, nm 3

4 CPC counts Motivation particle size distribution measurements DMA CPC Series1 neutralizer Time, s 1.E+05 dn/dlogdp, part/cm 3 8.E+04 6.E+04 4.E+04 2.E+04 0.E Dp, nm Prevailing conditions Delay time Residence time Transfer function (Ω) Transport losses CPC counting efficiency Fraction of particles with +1 charge 4

5 Objectives and methodology Objective: determine sources of uncertainty in the charge-toconcentration inversion required for size distribution measurement Measure charging characteristics in diverse systems using different neutralization techniques Particle charging (+1 fraction) depends on ion mobility and mass Ion mobility and mass depends on carrier gas properties Quantify experimentally o Particle size distributions o Ion distributions o Particle charge Calculate sensitivity (Fuchs theory) 5

6 Outline Motivation Background charge fraction Experimental charging conditions Neutralizers Carrier gas Results Size distributions Ion mobility Charged fraction Summary and conclusions 6

7 Charge probability +1 fraction stationary charge distribution Fuchs limited sphere model to calculate charge distribution Temperature Ion mobility Ion mass Wiedensohler approximation of Fuchs (Implemented in SMPS TM software) Ion mobility measured (radioactive source) Ion mass fitted result (Hussin et al. 1983) nm 100 nm Number of elementary charges Our calculations Ion mobility measured Ion mass calculated from Kilpatrick (1972) relationship Hussin, A., Scheibel, H. G., Becker, K. H. and Porstendorfer (1983) J. Aerosol Sci. 14, 671. Kilpatrick, W. D. (1971) Proc. Annual Conf on Mass Spectroscopy 19th., Wiedensohler, A. (1988) Journal of Aerosol Science 19(3), Fuchs, N. A. (1963) Geofis. pura appl. 56, 185.

8 Outline Motivation Background charge fraction Experimental charging conditions Neutralizers Carrier gas Results Size distributions Ion mobility Charged fraction Summary and conclusions 8

9 Experimental apparatus Aerosol neutralizers TSI 3077 (2 mci 85 Kr with est. current activity = 0.84 mci) TSI 3077A (10 mci 85 Kr with est. current activity = 8.3 mci) MSP M1090 Electrical Ionizer (AC corona discharge) TSI 3087 Advanced Aerosol Neutralizer (soft X-ray) Neutralizing conditions Dry nitrogen (N 2 ) Humidified air (various H 2 O) Humidified air with 20 ppb sulfur dioxide (SO 2 ) 9

10 Outline Motivation Background charge fraction Experimental charging conditions Neutralizers Carrier gas Results Particle size distributions Ion mobility distributions Charged fraction Summary and conclusions 10

11 Apparatus particle size distributions Test carrier gas Particle generation Silver Olive oil Soot Vent Test neutralizer Size distribution inversion from Wiedensohler, 1988 (standard method) Nano - DMA 3025 CPC Flowrate = 1.5 L/min 11

12 Silver size distributions (high concentration) dn/dlogdp, part/cm 3 5.E+08 4.E+08 3.E+08 2.E+08 1.E+08 2 mci 85Kr 10 mci 85Kr X-ray AC corona Carrier gas: air with 28% RH 0.E Dp, nm 12

13 Normalized total concentration Silver size distributions (high concentration) 5.E+08 2 mci 85Kr Carrier gas: air with 28% RH dn/dlogdp, part/cm 3 4.E+08 3.E+08 2.E+08 1.E mci 85Kr X-ray AC corona E Dp, nm mci 85Kr2 mci 85Kr Soft X-ray AC Corona 13

14 Silver size distributions (low concentration) dn/dlogdp, part/cm 3 1.5E E E+04 2 mci 85Kr 10 mci 85Kr X-ray AC Corona Carrier gas = air with 50% RH 0.0E Dp, nm 14

15 Normalized total concentration Silver size distributions (low concentration) 1.5E+05 2 mci 85Kr Carrier gas = air with 50% RH dn/dlogdp, part/cm 3 1.0E E mci 85Kr X-ray AC Corona E Dp, nm 10 mci 85Kr2 mci 85Kr Soft X-ray AC Corona 15

16 Oil droplet size distributions dn/dlogdp, part/cm 3 1.E+04 8.E+03 6.E+03 4.E+03 2.E+03 2 mci 85Kr 10 mci 85Kr X-ray AC Corona Challenge gas = air with 50% RH 0.E Dp, nm 16

17 Normalized total concentration Oil droplet size distributions 1.E+04 2 mci 85Kr Challenge gas = air with 50% RH dn/dlogdp, part/cm 3 8.E+03 6.E+03 4.E+03 2.E mci 85Kr X-ray AC Corona E Dp, nm mci 85Kr2 mci 85Kr Soft X-ray AC Corona 17

18 Soot size distributions 2.5E+06 dn/dlogdp, part/cm 3 2.0E E E+06 2 mci 85Kr X-ray AC corona 5.0E E Dp, nm MARC STETTLER, Jacob Swanson, Adam M Boies Evaluation of Uncertainties in Aircraft Engine Soot Emissions Derived from Engine Smoke Number, 2012 AAAR conference

19 Normalized total concentration Soot size distributions 2.5E+06 2 mci 85Kr dn/dlogdp, part/cm 3 2.0E E E X-ray AC corona E E Dp, nm 2 mci 85Kr Soft X-ray AC corona MARC STETTLER, Jacob Swanson, Adam M Boies Evaluation of Uncertainties in Aircraft Engine Soot Emissions Derived from Engine Smoke Number, 2012 AAAR conference

20 Apparatus ion mobility distributions Test carrier gas HEPA filter Silver tube furnace Vent Test neutralizer Nano - DMA Electrometer Flowrate = 1.5 L/min 20

21 Normalized current Normalized current Measured mobilities from radioactive source Positive ions Negative ions 1.0 Dry N2 1.0 Dry N2 0.8 Air 20% RH Air 50% RH 0.8 Air 20% RH Air 50% RH 0.6 Distributions for 10 mci 85 Kr Ion mobility, cm 2 /V-s 0.2 Distributions for 10 mci 85 Kr Ion mobility, cm 2 /V-s Ion mobility depends on carrier gas properties 21

22 Mean mobility, cm 2 /V-s Measured ion mobilities for air, 50% RH Positive ions = solid; negative ions = hashed Ion mobility depends on neutralizer Wiedensohler, A. (1988) Journal of Aerosol Science 19(3),

23 +1 charged fraction Normalized concentration Effect of mobility on inverted size distribution Calculated +1 charge fractions Inverted arbitrary distribution mci 85Kr AC Corona X-ray Lognormal distribution 10 mci 85Kr AC corona Concentration ratio = Wiedensohler, X-ray Carrier gas: air, 50 %RH Dp, nm 30% higher charge fraction = 30% concentration of particles Dp, nm Arbitrary particles are charged according to the corresponding ion mobility but all distributions are inverted using the same Wiedensohler (1988) charge fraction approximation 23

24 Apparatus particle charged fractions Test carrier gas Silver tube furnace Nano - DMA with 85 Kr Vent Test neutralizer HV supply ESP 3025 CPC Flowrate = 1.5 L/min 24

25 Charged fraction Charged fraction Charged fractions theory and measurements 10 mci 85 Kr AC X-ray corona Dry N2 - exp. Air 50% RH - exp. Dry N2 - theory Air 50% RH - theory Wiedensohler, Dry N2 - exp. Air 50% RH - exp. Dry N2 - theory Air Air 50% RH - theory Wiedensohler, Wiedensohler, Dp, nm Dp, nm Measurements are compared with the Wiedensohler approximation and with calculations using Fuchs theory (as adapted by Wiedensohler) with measured ion mobilities but calculated masses as input parameters 25

26 Summary and conclusions Measurements of size distributions of diverse aerosols revealed large differences, even for low particle concentrations The incorrect +1 fraction is being used to invert data Why is this? Measurements showed ion mobility (thus, charging) depends on: o Carrier gas composition o Relative humidity o Neutralizer type These parameters are different for every measurement 26

27 Summary and conclusions cont. If ultimate absolute concentration accuracy is of utmost importance to a project, it is recommended that a CPC be used as a concentration reference in addition to a [sizing spectrometer]. 1 YES! Hypothesis: Fuchs theory alone not sufficient to predict differences in neutralizers, even if all else is known. 1 TSI soft x-ray literature (2012) 27

28 Thank you for your attention Acknowledgements UK EPSRC MSP Corp and Copley Scientific for loan of M1090 Electrical Ionizer Dr Francisco Romay and Aaron Collins (MSP Corp) for technical assistance Cambustion Ltd. for loan of TSI 3077A 28