Data prepared by Bernard A. Olson, University of Minnesota

Department of Mechanical Engineering

111 Church Street, SE, Minneapolis, MN 55455

Methodology: Calibrations of the 4 LPM impactors was first report in 1987 by Marple et al¹. Versions of the unit have compared favorably to PM 2.5 and PM 10 from Dichotomous Sampler.^2,3,4 Since the original calibration the 4 LPM nozzles, the method has been refined and conducted on nozzles designed to provide specific 2.5 and 10.0 micron (µm) cut sizes at both 10 and 20 LPM. The higher flow rates allow greater mass to be collected in shorter time periods and acceptable weight to be collected at low concentrations. All of these nozzles compare favorably to the new PM 2.5 reference method. Recently, the 2.5 µm cut size, 4 LPM MST impactor was calibrated at 23 LPM to demonstrate a 1.0 µm cut size.

The particle calibrations consisted of generating monodisperse liquid oleic acid particles tagged with a uranine dye tracer using a vibrating orifice monodisperse aerosol generator (VOMAG). The aerosol was then sampled by the impactor, followed by washing the impaction plate, interior surfaces and afterfilter in 25 ml of 0.001 N aqueous solution of sodium hydroxide. Flourometric analysis was used to measure the quantity of dye in the wash solutions to determine the collection efficiency and interstage losses of the impactor for that particular size. This process was repeated for various particle sizes to determine the particle collection efficiency curve and interstage losses. A detailed description of the calibration technique is given by Marple et al. (1987)¹.

The procedure which has been used in the past for determining the actual (i.e. singlet) collection efficiency curve consisted of sampling the calibration aerosols generated by the VOMAG with an aerodynamic particle sizer (APS) to determine the singlet and multiple (doublets with twice the volume, triplets, etc.) concentrations for each run and then "backing out" the effects of the multiples from the collection efficiency curve determined from the fluorometric analysis. This is an important procedure since the VOMAG generates high concentrations of particles and some multiplets are formed in the process. The multiplets cause the aerosol to have a mass mean diameter that is typically on the order of 10 percent greater than an aerosol comprised exclusively of singlets. These multiplets affect the experimentally determined cut size and sharpness of cut of the impactor.

An improved method for determining the actual collection efficiency curve has been developed and was used for the recent impactor calibrations. It consists of placing a multiplet reduction impactor between the VOMAG, thus allowing the singlets to penetrate through to the impactor being calibrated. (Siegford et al., 1994)⁵. The multiplet reduction impactor physically removes the multiplets generated by the VOMAG, thus allowing the singlets to penetrate through to the impactor being calibrated. This method allows for the actual particle collection efficiency curve to be determined directly by fluorometric techniques and therefore, no secondary procedures are required.

Results: The results of the particle calibrations using the aforementioned procedures are shown in Tables #1, #2, #3, #4, #5 for the 2.5 m m cut size 10 LPM, 10 m m cut size 10 LPM, 2.5 m m cut size 20 LPM, 10 m m cut size 20 LPM, and 1.0 m m 23 LPM MST impactors, respectively. In general, the impactors were found to have sharp cutoff characteristics. Particle interstage losses for all of the impactors were very low with the exception of the 10 m m cut size 20 LPM impactor which was greater due to the higher flow rate.

Table #6 shows calibration data for the 2.5 µm cut size, 10 LPM MST impactor performed in 1987 and rechecked in 1997. When plotted the data fall on the same curve with the exception of the 2.35 m m particles from the 1987 data which show a slightly higher efficiency probably being due to multiplets..

All of these 2.5 mm cut nozzle laboratory calibrations compare favorably to the US E.P.A. WINS PM 2.5 impactor calibration data⁷ . Figure 1 shows a comparison of the penetration efficiency (1-collection efficiency) curves of the 2.5 mm cut size impactors at 10 and 20 LPM, the final version of the WINS impactor from the PM2.5 reference sampler, and the predicted penetration curve when a stacked pair of 10 LPM 2.5 mm nozzles is used (the predicted curve is derived from the square of the penetration efficiencies of a single nozzles). The shape of the 20 LPM nozzle curve is somewhat broader than the 10 LPM nozzle due to the larger jet Reynolds number (Radar and Marple, 1985)⁸, and is very similar to the US E.P.A. WINS impactor. Compared to a single nozzle impactor, the stacked pair of nozzles has a slightly lower d₅₀ (2.46mm), a much sharper drop in penetration above the d₅₀, and a broader increase below the d₅₀. This configuration is used when it is desirable to minimize the amount of coarse particle mass (above 2.5 mm) in the sample, such as in areas with large coarse to fine mass ratios (Allen et al., 1997)⁹ . This stacked configuration can also be used when long sample duration is needed that would overload a single-stage impactor, or when there are concerns regarding possible particle bounce from overloading. Other stacked configurations have been shown to be useful, such as a 10 mm impactor followed by a 2.5 mm impactor for use in sampling PM 2.5 at sites with chronic fog conditions (Hill et al.)⁹ .

The MST Area Sampler is a widely used instrument for PM measurement in both indoor and outdoor environments. Its low cost, modular impactor design, and small size make it a versatile exposure assessment tool for situations where compliance monitoring methods are not required. The sampler's sharp cut-point characteristics can be desirable in situations with high coarse to fine-mass ratios. Since this sampler has not undergone the rigorous testing required of U.S. EPA designated class II equivalent PM2.5 samplers or PM10 reference samplers, it is suggested that intercomparisons with US E.P.A. approved samplers be done on an ongoing basis to demonstrate the comparability of the methods.

1. V. A. Marple, K.L. Rubow, W.A. Turner, J.D. Spengler, " Low Flow Rate Sharp Cut Impactors for Indoor Air Sampling: Design and Calibration", JAPCA, 37:1303(1987).
2. P.J. Lioy, T. Wainman, W.A. Turner, V.A. Marple, "An Intercomparison of the Indoor Air Sampling Impactor and the Dichotomous Sampler for a 10-m m Cut Size", JAPCA , 38: 668 (1988).
3. G. A. Allen, P. Babich, E Wang, M. Davey, P. Koutrakis "PM 2.5 Method Comparisons in Birmingham, Alabama, USA" in : Proceedings of Measurement of Toxic an Related Air Pollutants, an AWA Specialty Conference, April 29 - May 1, 1997, RTP, North Carolina, Air and Waste Management Association Publication #VIP-74 (Pittsburgh, PA). (1997)
4. M. Pitchford, "Prototype PLM 2.5 FRM Field Studies Report. US EPA Staff Report sumitted to US EPA, OAQPS, RTP, NC July 9, 1997.
5. K.L. Seigford, V.A. Marple, K.L. Rubow, " A Multiplet Reduction Impactor For the Vibrating Orifice Aerosol Generator", AAAR Fourth International Aerosol Conference, August 29-September 2, 1997, Los Angeles, CA.
6. D. J. Radar and V.A. Marple, "Effect of Ultra-Stokesian Drag and Particle Interception on Impaction Characteristics", Aerosol Sci. Technol. 4:141 (1985).
7. Peters, T.M., Vanderpool, R.W. "Modification and Evaluation of the WINS Impactor", Final Report, Research Triangle Institute, Research Triangle Park, NC (September 1996).
8. D. J. Radar and V.A. Marple, "Effect of Ultra-Stokesian Drag and Particle Interception on Impaction Characteristics", Aerosol Sci. Technol. 4:141 (1985).
9. L.B. Hill, G.A. Allen, J. Carlson, "Characterization of Acid Aerosols and Regional Haze-Related Visibility Impairment, Great Gulf and Presidential Dry River Class-I Air Sheds, New Hampshire", Paper # 96-MP1A.06 in: Proceedings of the AWMA 89^th Annual Meeting, June 23-28, 1996, Nashville, TN

List of tables:

1. Calibration of 2.5 m m Cut size at 10 LPM MST Impactor (% collection efficiency)
2. Calibration of 10.0 m m Cut size at 10 LPM MST Impactor (% collection efficiency)
3. Calibration of 2.5 m m Cut size at 20 LPM MST Impactor (% collection efficiency)
4. Calibration of 10.0 m m Cut size at 20 LPM MST Impactor (% collection efficiency)
5. Calibration of 1.0 m m Cut size at 23 LPM MST Impactor (% collection efficiency)
6. Comparison of 1987 & 1996 Calibration of 2.5 m m Cut size at 10 LPM MST Impactor (% collection efficiency)

Table 3. Calibration of 2.5 m m Cut size at 20 LPM MST Impactor.

Figure 1. Penetration Factors for 2.5 um impactors: 1-nozzle at 10 and 20 LPM, and 2-nozzle at 10 LPM (calculated), and WINS final calibration. (G.Allen)