Comparison of speciation sampler and PC-BOSS fine particulate matter organic material results obtained in Lindon, Utah, during winter 2001-2002.

ABSTRACT

The Particle Concentrator-Brigham Young University Organic Sampling System (PC-BOSS) has been previously verified as being capable of measuring total fine particulate matter ([PM.sub.2.5]), including semi-volatile species. The present study was conducted to determine if the simple modification of a commercial speciation sampler with a charcoal denuder followed by a filter pack containing a quartz filter and a charcoal-impregnated glass (CIG) fiber filter would allow for the measurement of total [PM.sub.2.5], including semi-volatile organic material. Data were collected using an R & P (Rupprecht and Pastasnik Co., Inc.) Partisol Model 2300 speciation sampler; an R & P Partisol speciation sampler modified with a BOSS denuder, followed by a filter pack with a quartz and a CIG filter; a Met One spiral aerosol speciation sampler (SASS); and the PC-BOSS from November 2001 to March 2002 at a U.S. Environmental Protection Agency (EPA) Science to Achieve Results (STAR) sampling site in Lindon, UT. Total [PM.sub.2.5] mass, ammonium nitrate (both nonvolatile and semi-volatile), ammonium sulfate, organic carbon (both nonvolatile and semi-volatile), and elemental carbon were determined on a 24-hr basis. Results obtained with the individual samplers were compared to determine the capability of the modified R & P speciation sampler for measuring total [PM.sub.2.5], including semi-volatile components. Data obtained with the modified speciation sampler agreed with the PC-BOSS results. Data obtained with the two unmodified speciation samplers were low by an average of 26% because of the loss of semi-volatile organic material from the quartz filter during sample collection.

INTRODUCTION

Human health endpoints associated with exposure to airborne particulate matter (PM) include increased mortality and morbidity from respiratory and cardiopulmonary disease. (1-3) These effects are observed with exposure to concentrations substantially below the U.S. coarse PM ([PM.sub.10]) ambient air quality standard. The observed exacerbation of health problems is believed to be associated more closely with exposure to fine particles (<2.5 [micro]m) than coarse particles (>2.5 [micro]m). As a result, the U.S. Environmental Protection Agency (EPA) has promulgated revised standards for PM, which establishes new annual and 24-hr fine particulate standards with fine PM ([PM.sub.2.5]), measured according to the Federal Reference Method (FRM) ([PM.sub.2.5] FRM), as the indicator. (4,5) This recognition of fine and coarse particles as different classes of PM pollutants is an advance in the understanding and control of PM. However, ambient [PM.sub.2.5] is not a single pollutant, but a mixture of many chemical species. Major components include: sulfate, nitrate, ammonium, and hydrogen ions; trace elements (including toxic and transition metals); organic material; elemental carbon (EC or soot); and crustal components. EPA has promulgated the use of commercial speciation samplers to determine the chemical composition of [PM.sub.2.5]. (4,5) Speciation samplers commonly use a variety of filter packs to collect stable species and a diffusion denuder system suitable for the determination of fine particulate nitrate. Stable species such as trace and crustal elements and sulfate can be measured by these filter samplers, and nitrate is measured with the diffusion denuder module in these systems. However, semi-volatile fine particulate organic material is not determined by these techniques. (6-8) Positive artifacts are seen in the collection of organic material, especially using quartz filters, due to adsoprtion of gas-phase organic compounds. (6,13) Negative artifacts occur because of the loss of particulate semi-volatile material during sampling. (6-9) Both of these artifact problems can be minimized using a diffusion denuder sampler to determine fine particulate organic material. (6-9) This study was initiated to compare speciation sampler and denuder measurement of fine particulate carbonaceous material and to see if a simple modification of a commercial speciation sampler would allow for the accurate determination of semi-volatile fine particulate organic material.

EXPERIMENTAL METHODS AND PROCEDURES

Sampling Site

The sampling site for this experiment is located next to the State of Utah Air Quality Monitoring site at Lindon Elementary School in Lindon, UT. Lindon is an urban community located approximately 20 km north of Provo, UT. During winter inversions the community is usually impacted by primary emissions and secondary components formed from combustion products from mobile sources and wood-burning stoves. (8) A nearby integrated steel mill was not operational during this study.

Sampling Period

A total of 24 sample sets were collected periodically during winter inversion conditions from November 15, 2001 until January 14, 2002. The 24-hr samples were collected each sampling day from 12:00 a.m. to 12:00 p.m. Samples were collected on days when stable inversions with no precipitation were meteorologically forecast. The composition and sources of the collected particles under these conditions have been previously described. (8,9)

Sampling Instruments

Three different speciation samplers were used for collection of [PM.sub.2.5]. These are shown schematically in Figure 1.

PC-BOSS. The combination of technology used in the High-Volume Brigham Young University Organic Sampling System (10) (BIG BOSS) and the Harvard particle concentrator (11) resulted in the Particle Concentrator-Brigham Young University Organic Sampling System (PC-BOSS). (6,7,12) The configuration and operation of the PC-BOSS as used in the Wasatch Front Environmental Monitoring for Public Awareness and Community Tracking (EMPACT) and Science to Achieve Results (STAR) programs, has been previously described. (8) The PC-BOSS was used for sample collection to determine fine particulate mass, sulfate, carbonaceous material (elemental and organic), nitrate, semi-volatile organic material, and semi-volatile nitrate. Inlet flow was 130 L/min. Samples for the chemical characterization of [PM.sub.2.5] in the minor flow following a particle concentrator and a BOSS diffusion denuder were collected in a filter pack containing a pre-fired 47-mm quartz filter (18 L/min, Pallflex) followed by 47-mm charcoal-impregnated glass (CIG) fiber filter (Schliecher and Schuell) to determine fine particulate carbonaceous material and nitrate, including semi-volatile species lost from the particles during sampling. A second parallel filter pack containing a 47-mm Teflon (12 L/min, Whatman) filter followed by a 47-mm nylon (Gelman, Nylasorb) filter was used to determine [PM.sub.2.5] filter-retained (nonvolatile) mass, sulfate, nitrate, and any nitrate lost from the particles during sample collection. A side-flow filter pack, before the particle concentrator, containing a 47-mm polycarbonate (10 L/min, Whatman, Nuclepore, 0.4-[micro]m pore size) filter followed by a 47-mm CIG fiber filter, collected particles (excluding semi-volatile species lost during sampling), and gas-phase organic material after the 2.5-[micro]m outlet cut. These data were compared with data from the minor flow filters to determine the particle concentrator efficiency. (7,12)

[FIGURE 1 OMITTED]

R & P Model 2300 Speciation Sampler. The R & P Partisol Model 2300 speciation sampler was used to determine fine particulate mass, sulfate, nitrate, and carbonaceous material. The sampler consists of three sampling channels, each subdivided into four sampling banks through which air can be pulled at the desired rate via internal pumps. The filter pack systems (ChemComb cartridges) include an anodized impactor plate followed by a vertical flow chamber. Samples were collected in four parallel ChemComb cartridges. The first cartridge was used to determine total [PM.sub.2.5] mass collected on a 47-mm Teflon (17 L/min, Whatman) filter. The second cartridge contained a 1% NaHC[O.sub.3]:1% glycerine-coated glass honeycomb denuder, which was used to determine concentrations of gas-phase nitric acid ([HNO.sub.3]) and sulfur dioxide (S[O.sub.2]). This was followed by a 47-mm nylon (17 L/min, Gelman, Nylasorb) filter used to determine concentrations of fine particulate nitrate and sulfate. The third cartridge used a Teflon-coated impactor plate followed by two pre-fired 47-mm quartz (10 L/min, Pallflex) filters to determine fine particulate sulfate and carbonaceous material after the tandem quartz filter protocols. (13) The fourth cartridge was modified from the original design of R & P by the addition of a short BOSS (14) diffusion denuder to the anodized inlet before the filters. The denuder consists of parallel strips of charcoal-impregnated cellulose fiber filters (CIF) (Schliecher and Schuell) with high capacity for interfering gas-phase organic compounds, ozone, [HNO.sub.3] and S[O.sub.2]. The denuder was followed by a 47-mm pre-fired quartz (10 L/min, Pallflex) filter followed by a 47-mm pre-fired CIG fiber (Schleicher and Schuell) filter to determine fine particulate sulfate and carbonaceous material, including any semi-volatile organic compounds (SVOCs) that may be removed from the particles during sampling.

Met One Speciation Sampler. The Met One spiral aerosol speciation sampler (SASS) consists of several individual channels through which air is sampled. These filter systems include a sharp cut cyclone followed by a short vertical flow chamber. The first chamber included a 47-mm Teflon (Whatman) filter to determine total [PM.sub.2.5] particulate mass and sulfate. A second parallel chamber consisted of an aluminum honeycomb denuder followed by a 47-mm nylon (Gelman, Nylasorb) filter used to determine fine [PM.sub.2.5] nitrate and sulfate material. The third parallel chamber used two pre-fired 47-mm quartz (Pallflex) filters to determine [PM.sub.2.5] fine particulate sulfate and carbonaceous material. Flow was 7 L/min in each of the systems channels.

Analytical Methods

Mettler UMT2 Microbalance. A Mettler UMT2 microbalance was used for the determination of collected fine particulate mass at 23 [degrees]C and 30% relative humidity on a 47-mm Teflon (Whatman) filter sampled from the PC-BOSS minor flow or the Teflon filters of the [PM.sub.2.5] speciation samplers.

Temperature-Programmed Volatilization Analysis. Temperature-programmed volatilization (TPV) (10,15) was used in the analysis of collected samples for total carbonaceous material. In this method, the various sampled filters are heated from ambient temperature at a predetermined ramp rate to a predetermined termination temperature. The ramp rate and termination temperatures are dependent on the type of filter being analyzed. Quartz filters are heated to 800 [degrees]C over a 20-min period in an [N.sub.2]/[O.sub.2] (80%/20%) atmosphere. The separation of the data into organic carbon (OC) and EC has been described. (10,15) Charcoal-impregnated filters are heated to 400 [degrees]C over a 12-min period in a [N.sub.2] atmosphere. Carbon in compounds desorbed from the filters during the heating process is catalytically converted to carbon dioxide ([CO.sub.2]) and detected by nondispersive infrared absorption. (6-8) The instrument was calibrated daily with three certified standards that span the concentration range seen in the samples. Regression analysis of the calibration data gave [R.sup.2] values greater than 0.99.

Dionex Ion Chromatograph. A Dionex Model 500 ion chromatograph with a separator column, anion fiber suppressor, and conductivity detector was used for the analysis of collected samples for nitrate and sulfate. Sample filters are ultrasonically extracted in water (or in the case of Nylasorb filters, in the ion chromatography eluent) and the concentrations of sulfate and nitrate in the various solutions were determined by peak area measurement and comparison to standards.

RESULTS AND DISCUSSION

The complete dataset that resulted from the speciation sampler comparison study at Lindon includes the comparison of concentrations of [PM.sub.2.5] mass and sulfate from November 2001 through January 2002 and of both nonvolatile and semi-volatile organic matter obtained over 24 sampling periods from November 2001 through March 2002.

Sulfate

The concentration of sulfate in fine particles was determined from a total of 10 different filters (see Figure 1): PC-BOSS samples included the Nuclepore filter (PCB Nu) in the PC-BOSS side-flow (used to determine the particle concentrator efficiency) and both the Teflon (PCB T) and quartz (PCB Q) filters in the two filter packs of the minor flow after the particle concentrator and BOSS denuder. Samples obtained using the R & P speciation sampler included the Teflon filter (RP T), the particle collecting quartz filter (RP Q) from the filter pack with two quartz filters, the Nylasorb filter after the glass honeycomb denuder (RP Ny) and the quartz filter after the BOSS denuder (RP [Q.sub.Den]). For the Met One system, samples included Teflon (MET T), Nylasorb (after the aluminum honeycomb denuder, MET Ny), and quartz (MET Q) filters. The various samples analyzed for sulfate are all compared with the RP T results in Figure 2. Linear regression and error analysis results for the comparison given in Figure 2 (for all but the Nylasorb filters) are given in Table 1. As indicated, the various results for sulfate are all in agreement within [sigma] = [+ or -]0.37 [micro]g sulfate/[m.sup.3] ([+ or -]15%), and a bias between the comparisons of only 0.04 [micro]g/[m.sup.3] compared with the average concentration of 2.4 [micro]g/[m.sup.3]. This result indicates that the dataset is robust with no significant difference in the sulfate results among the various samplers. Results are also given in Table 1 for the comparison of the RP T filter sulfate results to those obtained on the two Nylasorb filters. As indicated, the Nylasorb filter results are biased higher by 0.53 [micro]g/[m.sup.3]. This difference is less than twice the [sigma] value for the first comparison but may indicate that there is some adsorption of S[O.sub.2](g) not removed by the denuder in front of the Nylasorb filter. In general, concentrations of [PM.sub.2.5] sulfate were low. The two samples with the highest sulfate concentrations occurred on December 18 and 30, 2001. These were both days with strong inversions and high humidity, leading to higher [PM.sub.2.5] sulfate concentrations at the sampling site. (16)

[FIGURE 2 OMITTED]

Quartz Filter Carbon

Carbonaceous material in particles collected on a quartz filter was determined on two filters that were not preceded by a denuder, RP Q and MET Q. The same concentration was determined for particle-collecting filters that were preceded by a denuder, PCB Q and RP [Q.sub.Den]. The direct comparison of the various quartz filter results is complicated by the possible presence of several sampling artifacts. The positive artifact (13,18) is the portion of total carbon content on the front quartz filter (e.g., RP Q and MET Q) that is the result of absorbed organic gases, not collected particulate organic material. (6,13) Adsorbed organic gases are also expected to be present on a subsequent quartz filter. However, the positive organic compound artifact should not be present on a Teflon filter because a Teflon filter is not expected to absorb gas-phase organic compounds. (13) Likewise, the positive artifact is not expected to be present on the quartz filter after the particle concentrator and the BOSS denuder, for example, PCB Q and RP [Q.sub.Den], because the gas-phase compounds that can adsorb on a quartz filter are effectively removed in the particle concentrator-denuder combination of the PC-BOSS (7,12) or the short BOSS denuder at a low flow rate. (16) A second artifact that may be present in the collection of particulate organic material is the loss of SVOCs from the particles collected on the quartz filter during sampling. (6) This artifact may be present on any of the particle-collecting filters in the sampling system. If the particle-collecting filter is Teflon, these SVOCs lost from particles will pass through the Teflon filter. However, if the particle-collecting filter is a quartz filter, the quartz filter that is collecting the particles may reabsorb some of the SVOCs lost from the particles during sampling. It has been previously shown that the positive quartz filter artifact will result in a peak in the TPV analysis below 120 [degrees]C that is not present if the gas-phase material is removed. (19) The carbonaceous material found on the various undenuded second quartz filters was dominated by material in this low temperature region, and this peak was missing in the denuded quartz filters. Therefore, it was assumed that the material found on the second quartz filters in the various undenuded quartz filters originated from gas-phase organic compounds adsorbed by the quartz.

The carbonaceous material collected on the denuded quartz filter in the R & P speciation sampler, RP [Q.sub.Den], is compared with the other quartz filter results in Figure 3. In this comparison, the two undenuded quartz filter results, RP Q and MET Q, are corrected by subtracting the carbonaceous material found on the second quartz filter. This correction averaged 1.2 [micro]g C/[m.sup.3], and varied from negligible to 4 [micro]g C/[m.sup.3]. As indicated by the statistical analysis results given in Table 1, the concentrations of carbonaceous material, corrected for the presence of gas-phase carbonaceous material on the undenuded RP quartz filters, on any quartz filter for a given sample were comparable. For the comparison of the denuded quartz filters, RP [Q.sub.den] and PCB Q, the linear regression slope is 0.94 [+ or -] 0.09 with [sigma] = [+ or -]0.80 [micro]g C/[m.sup.3] (Table 1). The comparison of RP [Q.sub.den] with the undenuded quartz filters gave a linear regression slope of 0.97 [+ or -] 0.06 with a comparable [sigma] (Table 1). These [sigma] values are close to the estimated uncertainty of [+ or -]0.5 [micro]g C/[m.sup.3] in the data on the basis of the variability in blank filter values and replicate sample analysis. The various quartz filter results were in agreement.

[FIGURE 3 OMITTED]

SVOCs Lost from Particles during Sampling

The efficiency of removal of gas-phase organic compounds by the denuders of the PC-BOSS and the modified R & P speciation sampler are expected to be >99% for the PC-BOSS (7,12) and above 96% for the modified speciation sampler. (16,17) TPV analysis indicated that gas-phase organic material consistent with the expected denuder efficiency did penetrate the speciation sampler short BOSS denuder. SVOCs lost from particles are expected, based on past studies, to be evolved from the CIG filters during the TPV analysis from approximately 220 to 280 [degrees]C in a symmetrical peak centered around 250 [degrees]C. Gas-phase compounds collected without a denuder (e.g., the side flow of the PC-BOSS sampler) will be evolved from approximately 140 to 280 [degrees]C. However, past studies (6-8,10) with the BOSS have shown that the compounds that evolve at the highest temperature are removed with the greatest efficiency (20) so that gas-phase compounds that pass through the BOSS denuder will evolve from the CIG filter from around 140 to 230 [degrees]C in a symmetrical peak centered around 180-190 [degrees]C. Thus, compounds that were originally in the gas phase that pass through the denuder can be distinguished from gas-phase SVOCs lost from the particles collected after the denuder in a TPV analysis. Evidence was seen for the presence of both absorbed gas-phase compounds (evolving around 140-250 [degrees]C) and SVOCs lost from particles (evolving around 220-280 [degrees]C) in the TPV analysis of the RP [Q.sub.Den] filters. The data were corrected for that material that was originally gas phase by subtracting the second quartz filter results. This correction varied from approximately 10 to 40% of the total carbonaceous material evolved from the CIG filters during the TPV analysis. On the basis of comparison of the results obtained from the CIG filter in the side-flow of the PC-BOSS (that measures all gas-phase material that can be adsorbed by a CIG filter; e.g., see Figure 2) and the gas-phase material on a CIG filter after a denuder (estimated as described above), the efficiency of the short BOSS denuder for the removal of gas-phase organic compounds in the modified speciation sampler was 97-98%, in agreement with expectations. (16)

The SVOCs lost from particles after a denuder for the PC-BOSS and modified speciation sampler are shown in Figure 4. Statistical analysis of these data is given in Table 1. As indicated, the concentrations of SVOCs were the same for both samplers with negligible bias and a precision of [+ or -]0.27 [micro]g SVOC C/[m.sup.3]. This [sigma] value is comparable to the uncertainty in the SVOC data. Thus the speciation sampler, modified with a small BOSS denuder, can be used to determined total fine particulate organic material, including the SVOCs lost from particles during sampling.

Total Fine Particulate Carbonaceous Material

Total fine particulate carbonaceous material is the sum of the nonvolatile carbonaceous material retained by the quartz filter and the SVOCs lost from the particles and collected by the following CIG filter. The concentrations of total carbonaceous material determined by the modified R & P speciation sampler and the PC-BOSS are compared in Figure 5. As indicated by the results of the statistical analyses given in Table 1, the two results agree with a bias-corrected precision of [+ or -]0.75 [micro]g/[m.sup.3] ([+ or -]9.4%), and a bias of only 0.28 [micro]g/[m.sup.3] (3.5% of the average). A Student's t test also indicates there is no statistical difference between the two results. The modified R & P sampler is capable of determining total fine particulate carbonaceous material, including the semi-volatile material. Although the analysis of samples collected with the modified speciation sampler does require analysis of a CIG filter, two different groups (21,22) have shown that this analysis can be done using a commercial Sunset Carbon Aerosol Analysis lab instrument with a constant temperature ramp. Thus the analysis may be done with the instrumentation used in the national program for analysis of speciation sampler filters.

[FIGURE 4 OMITTED]

In contrast, analysis of the quartz filters only (Figure 5) results in a biased undermeasurement of the total fine particulate carbonaceous material by an average of 2.1 [micro]g/[m.sup.3], an average of 26% of the total fine particulate carbonaceous material for the samples collected here. Results obtained in other urban areas indicate that the undermeasurement of fine particulate carbonaceous material can be expected to vary from 20 to 50% using the conventional speciation sampler. (7,8,18,20,23)

Concentrations of [PM.sub.2.5] Components

The concentrations of sulfate and carbonaceous material collected during each sampling period, on the basis of the diffusion denuder results where complete data are available, are given in Figure 6. For the data given in Figure 6 it is assumed that the sulfate was present as ammonium sulfate and the carbonaceous material was 61% carbon. (24) The highest concentrations of [PM.sub.2.5] components were measured during the latter part of December. These high concentrations were associated with persistent inversion conditions during this time. The major component of the fine particles was the nonvolatile carbonaceous material. Ammonium sulfate and [PM.sub.2.5] SVOCs were of comparable average concentrations. However, the highest concentrations of ammonium sulfate are associated with humid inversion conditions, which contribute to the conversion of S[O.sub.2] to sulfate. (16) The highest concentrations of SVOCs are associated with clear day conditions, which result in lower sulfate concentrations but contribute to the formation of secondary SVOCs. (8)

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

CONCLUSIONS

The modification of an R & P Partisol Model 2300 speciation sampler by addition of a small BOSS denuder followed by a filter pack with a quartz filter and a CIG fiber filter allows for the facile determination of both nonvolatile OC (NVOC) and SVOCs in [PM.sub.2.5]. During the November 2001-March 2002 sampling period at the Lindon, UT, sampling site, the NVOC (particulate OC retained on a quartz filter) averaged 5.4 [micro]g/[m.sup.3] and the SVOCs averaged 2.2 [micro]g/[m.sup.3]. The observation of a SVOC component equal to approximately 40% of the NVOC suggests that additional measurements of the SVOC component of [PM.sub.2.5] are needed. The CIG filters can be analyzed by TPV, as was done in this study, or by EPA speciation carbon analysis techniques. However, if the EPA TOT protocols (25) are used, rather than the constant temperature ramp method used by us and others, (21,22) it will probably be required that an independent evaluation of the denuder breakthrough be made.

ACKNOWLEDGMENTS

EPA through its Office of Research and Development partially funded and collaborated in the research described here under Cooperative Agreement CR827364 with Brigham Young University. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of EPA. Input to the manuscript from William E. Wilson is appreciated. The technical assistance of Jordan Inouye and Michael Simpson is appreciated.

REFERENCES

1. Pope, C.A., III. Epidemiology of Fine Particulate Air Pollution and Human Health: Biologic Mechanisms and Who's at Risk; Environ. Health Perspect. 2000, 108, 713-723.

2. Schwartz, J.; Dockery, D.W.; Neas, L.M. Is Daily Mortality Associated Specifically with Fine Particles?; J. Air & Waste Manage. Assoc. 1996, 46, 927-939.

3. Air Quality Criteria for Particulate Matter; EPA/600/P-95/001aF; U.S. Environmental Protection Agency: Research Triangle Park, NC, 1996.

4. Musick, D. A Summary of the Ambient Air Program for [PM.sub.2.5]: the Standard, the Federal Reference Method, and the Quality Assurance Program; EM 1999, February, 17-20.

5. Schaefer G.; Hamilton W.; Mathai C.V. Implementing the Revised National Ambient Air Quality Standards and the Federal Advisory Committee Act Subcommittee for Ozone, Particulate Matter and Regional Haze; EM 1997, 22-28.

6. Eatough, D.J.; Long, R.W.; Modey, W.K.; Eatough, N.L. Semi-Volatile Secondary Organic Aerosol in Urban Atmospheres: Meeting a Measurement Challenge; Atmos. Environ. 2003, 37, 1277-1292.

7. Lewtas, L.; Booth, D.; Pang, Y.; Reimer, S.; Eatough, D.J.; Gundel, L.A. Comparison of Sampling Methods for Semi-Volatile Organic Carbon Associated with [PM.sub.2.5]; Aerosol Sci. Technol. 2001, 34, 9-22.

8. Long, R.W.; Eatough, N.L.; Mangelson, N.F. Thompson, W.; Fiet, K.; Smith, S.; Smith, R.; Eatough, D.J.; Pope, C.A.; Wilson, W.E. The Measurement of [PM.sub.2.5], Including Semi-Volatile Components, in the EMPACT Program: Results from the Salt Lake City Study; Atmos. Environ. 2003, 37, 4407-4417.

9. Grover, B.D.; Carter, C.B.; Kleinman, M.A.; Richards, J.S.; Eatough, N.L.; Eatough, D.J.; Dasgupta, P.K.; Al-Horr, R.; Ullah, S.M. Monitoring and Source Apportionment of Fine Particulate Matter at Lindon, Utah; Aerosol Sci. Technol. 2006, 40, 941-951.

10. Tang, H.; Lewis, E.A.; Eatough, D.J.; Burton, R.M.; Farber, R.J. Determination of the Particle Size Distribution and Chemical Composition of Semi-Volatile Organic Compounds in Atmospheric Fine Particles; Atmos. Environ. 1994, 28, 939-947.

11. Sioutas, C.; Koutrakis, P.; Burton, R.M. Development and Evaluation of a Low Cutpoint Virtual Impactor; Aerosol Sci. Technol. 1994, 21, 223-235.

12. Ding, Y.; Pang, Y.; Eatough, D.J. High Volume Diffusion Denuder Sampler for the Routine Monitoring of Fine Particulate Matter: I. Design and Optimization of the PB-BOSS; Aerosol Sci. Technol. 2002, 36, 369-382.

13. Turpin, B.J.; Huntzicker, J.J.; Hering, S.V. Investigation of the Organic Aerosol Sampling Artifacts in the Los Angeles Basin; Atmos. Environ. 1994, 28, 3061-3071.

14. Eatough, D.J.; Eatough, N.L.; Obeidi, F.; Pang, Y.; Modey, W.; Long, R. Continuous Determination of [PM.sub.2.5] Mass, Including Semi-Volatile Species; Aerosol Sci. Technol. 2001, 34, 1-8.

15. Ellis, E.C.; Novakov, T. Application of Thermal Analysis to the Characterization of Organic Aerosol Particles; Sci. Total Environ. 1982, 23, 227-238.

16. Ren, Y. Ph.D. Dissertation, Brigham Young University, Provo, UT, August 1999.

17. Grover, B.D.; Kleinman, M.; Eatough, N.L.; Eatough, D.J.; Cary R.A.; Hopke, P.K.; Wilson, W.E. Measurement of Fine Particulate Matter NonVolatile and Semi-Volatile Organic Material with the Sunset Laboratory Carbon Aerosol Monitor. J. Air & Waste Manage. Assoc. 2008, 58, 72-77.

18. Eatough D.J. In Gas and Particle Phase Measurements of Atmospheric Organic Compounds; Lane, D.A., Ed.; Gordon and Breach Science: Amsterdam, The Netherlands, 1999; Vol. 2, pp 233-285.

19. Eatough, D.J.; Eatough, N.L.; Pang, Y.; Sizemore, S.; Kirchstetter, T.W.; Novakov, T.; Hobbs, P.V. Semi-Volatile Particulate Organic Material in Southern Africa during SAFARI-2000; J. Geophys. Res. 2003, 108, 8479, doi: 10.1029/2002JD002296.

20. Cui, W.; Eatough, D.J.; Eatough, N.L. Fine Particulate Organic Material in the Los Angeles Basin 1: Assessment of the High-Volume Brigham Young University Organic Sampling System, Big BOSS; J. Air & Waste Manage. Assoc. 1998, 48, 1024-1037.

21. Huebert, B.; Bertram, T.; Howell, S.; Eatough, D.J.; Kline, J.; Blomquist, B. Measurements of Organic and Elemental Carbon in Asian Outflow during ACE-Asia from the SNF/NCAR C-130; J. Geophys. Res. 2004, 109, D19S11, doi: 10.1029/2004JD004700.

22. Cabada, J.C.; Pandis, S.N.; Subramanian, R.; Robinson, A.L.; Polidori, A.; Turpin, B. Estimating the Secondary Organic Aerosol Contribution to [PM.sub.2.5] Using the Organic Carbon to Elemental Carbon Ratio Method; Atmos. Environ. 2004, 38, 140-155.

23. Pang, Y.; Eatough, N.L.; Eatough, D.J. [PM.sub.2.5] Semi-Volatile Organic Material at Riverside, California: Implications for the [PM.sub.2.5] from Sampler; Aerosol Sci. Technol. 2002, 26, 1-12.

24. Turpin, B.J.; Lim, H.J. Species Contributions to [PM.sub.2.5] Mass Concentrations: Revisiting Common Assumptions for Estimating Organic Mass; Aerosol Sci. Technol. 2001, 35, 602-610.

25. National Institute for Occupational Safety and Health Manual of Analytical Methods (NMAM), Method 5040, 4th ed.; O'Conner, P.F., Ed.; 1994, DHHS (NIOSH) Publication No. 94-113.

About the Authors

Cory Carter, Norman L. Eatough, and Delbert J. Eatough are with the Department of Chemistry and Biochemistry at Brigham Young University in Provo, UT. Neal Olson is from the Air Monitoring Division of the Utah State Department of Environmental Quality in Salt Lake City, UT. Russel W. Long is with EPA in Research Triangle Park, NC. Please address correspondence to: Delbert J. Eatough, Department of Chemistry and Biochemistry, E114 Benson Building, P.O. Box 25700, Brigham Young University, Provo, UT 84602; phone: +1-801-422-6040; fax: e-mail: Delbert_eatough@byu.edu.

Cory Carter, Norman L. Eatough, and Delbert J. Eatough

Department of Chemistry and Biochemistry, Brigham Young University, Provo, UT

Neal Olson

Air Monitoring Division, Utah State Department of Environmental Quality, Salt Lake City, UT

Russell W. Long

U.S. Environmental Protection Agency, Research Triangle Park, NC

RELATED ARTICLE: IMPLICATIONS

EPA approved speciation samplers include a system for the measurement of semi-volatile ammonium nitrate, but do not include a measurement of semi-volatile organic material that may be lost from the particle collection filter. This semi-volatile fine particulate organic material can be determined with an appropriate diffusion denuder sampler. A simple modification of a commercial speciation sampler allows this measurement to be made. This modification allows for the more complete determination of fine particulate composition with a speciation sampler. Table 1. Results of the statistical analysis of [PM.sub.2.5] 24-hr samples collected in the Lindon study. x vs. y n [I.sup.2] Slope (a) RP T sulfate vs. other 48 0.95 1.02 [+ or -] 0.02

quartz or Teflon sulfate 0.95 1.07 [+ or -] 0.04 RP T sulfate vs. Nylasorb 20 0.88 1.11 [+ or -] 0.05

sulfate 0.91 0.97 [+ or -] 0.07 RP [Q.sub.Den] TC (c) vs. 14 0.90 0.98 [+ or -] 0.05

PCB Q TC (c) 0.90 0.94 [+ or -] 0.09 RP [Q.sub.Den] TC (c) vs. 33 0.90 0.99 [+ or -] 0.03

SS [Q.sub.Corr] TC (d) 0.90 0.97 [+ or -] 0.06 RP CIG SVOC (e) vs. PCB 15 0.82 1.03 [+ or -] 0.04

CIG SVOC (e) 0.83 0.95 [+ or -] 0.12 RP Den Q TC + SVOC (e) vs. 14 0.94 0.95 [+ or -] 0.02

PCB Q TC + SVOC (e) 0.94 0.91 [+ or -] 0.07 RP Den Q TC + CIG SVOC (e) 24 0.60 0.74 [+ or -] 0.06

vs. SS Q TC (d) 0.60 0.83 [+ or -] 0.14

Intercept x (Average x-y (Bias

([micro]g/ [micro]g/ [micro]g/ x vs. y [m.sup.3]) [m.sup.3]) [m.sup.3]) RP T sulfate vs. other 0.0 [+ or -] 0.5 2.41 0.04

quartz or Teflon sulfate -0.2 [+ or -] 0.5 RP T sulfate vs. Nylasorb 0.0 [+ or -] 0.7 2.60 -0.53

sulfate 0.6 [+ or -] 0.7 RP [Q.sub.Den] TC (c) vs. 0.0 [+ or -] 1.2 6.25 0.09

PCB Q TC (c) 0.2 [+ or -] 1.2 RP [Q.sub.Den] TC (c) vs. 0.0 [+ or -] 1.1 5.55 0.02

SS [Q.sub.Corr] TC (d) 0.1 [+ or -] 1.1 RP CIG SVOC (e) vs. PCB 0.0 [+ or -] 0.4 2.27 -0.09

CIG SVOC (e) 0.2 [+ or -] 0.4 RP Den Q TC + SVOC (e) vs. 0.0 [+ or -] 1.1 8.14 0.28

PCB Q TC + SVOC (e) 0.4 [+ or -] 1.1 RP Den Q TC + CIG SVOC (e) 0.0 [+ or -] 2.3 6.99 1.91

vs. SS Q TC (d) -0.7 [+ or -] 2.4

[sigma] (b)

([micro]g/ [sigma] x vs. y [m.sup.3]) (%) RP T sulfate vs. other 0.37 15.5

quartz or Teflon sulfate RP T sulfate vs. Nylasorb 0.28 9.8

sulfate RP [Q.sub.Den] TC (c) vs. 0.80 12.8

PCB Q TC (c) RP [Q.sub.Den] TC (c) vs. 0.77 13.9

SS [Q.sub.Corr] TC (d) RP CIG SVOC (e) vs. PCB 0.27 12.0

CIG SVOC (e) RP Den Q TC + SVOC (e) vs. 0.75 9.4

PCB Q TC + SVOC (e) RP Den Q TC + CIG SVOC (e) 2.10 34.8

vs. SS Q TC (d) Notes: (a) Nonweighted linear regression slopes are given for (1) zero intercept and (2) calculated intercept. (b) [sigma] is calculated as {1/2n[[SIGMA]([x.sub.1,[bar.i]] - [x.sub.2,i])[.sup.e] - ([bar.x.sub.1] - [bar.x.sub.2])[.sup.2]]}[.sup.1/2] where [x.sub.1] and [x.sub.2] are the two datasets being compared. (c) TC = total carbonaceous material. (d) SS = RP Q (corrected for second Q filter), and MET Q (corrected for second Q filter) filters. (e) SVOC is the semi- volatile organic material lost from particles during sampling.