Coarse PM Methods Evaluation Study

(1)

Coarse PM

Methods Evaluation Study

Study Design and Preliminary Results

Thomas Ellestad, Robert Vanderpool, Thomas Ellestad, Robert Vanderpool,

Paul Solomon, and Mary Harmon Paul Solomon, and Mary Harmon

USEPA/ORD/NERL USEPA/ORD/NERL

Sanjay Natarajan, Chris Noble, and Bob Murdoch Sanjay Natarajan, Chris Noble, and Bob Murdoch

Research Triangle Institute Research Triangle Institute

[email protected] [email protected]

919

(2)

Background

§

1997 air regulations established NAAQS for

_{1997 air regulations established NAAQS for}

PM2.5 and PM10 as separate metrics

§

U.S. courts have reviewed subsequent

_{U.S. courts have reviewed subsequent}

litigation and ruled that the PM10 metric is a

“poorly matched indicator” because it

includes the PM2.5 fraction

§

EPA has since been considering the possibility

of vacating the PM10 regulation and

developing a separate standard for PMc

(3)

Study Objectives

§

Evaluate

_Evaluate

the field performance of

leading

methods for monitoring the coarse fraction of

PM10 (PMc = PM10

–

PM2.5)

§

Evaluate samplers which are either already

_{Evaluate samplers which are either already}

commercially available or in their final stages of

development

§

Include both filter

_{Include both filter}

-

based (time

-

integrated) and

semi

(4)

PM2.5 and PM10 FRM Samplers

§

§ Standard lowStandard low--vol PM10 inlets vol PM10 inlets aspirating at 16.7 lpm

aspirating at 16.7 lpm

(actual conditions)

§

§ PM2.5 aerosol fractionation PM2.5 aerosol fractionation using aWINS equipped with

using aWINS equipped with

DOS impaction oil

§

§ Filters were conditioned at Filters were conditioned at 22C and 35% RH, analyzed

22C and 35% RH, analyzed

gravimetrically. Post

gravimetrically. Post-

-sampling filters archived at

sampling filters archived at

--30C for subsequent chemical 30C for subsequent chemical analysis

analysis

§

§ 3 FRM pairs from BGI, 3 FRM pairs from BGI, R&P, and Thermo

R&P, and Thermo--Andersen Andersen equipped with teflon filters

equipped with teflon filters

(4

(4thth _{FRM pair equipped with}_{FRM pair equipped with}

quartz filters) quartz filters) P M 10 PM10 (WINS Removed) PM2.5 (with WINS) PMc = PM10 – PM2.5

(5)

R&P Partisol

-

Plus 2025 Dichot

§

§ Standard PM10 inlet Standard PM10 inlet aspirating at 16.7 lpm

aspirating at 16.7 lpm

(actual)

§

§ Aerosol fractionation by Aerosol fractionation by custom virtual impactor

custom virtual impactor

(15 lpm and 1.67 lpm)

§

§ PM2.5 and PMc mass PM2.5 and PMc mass

collected on 47 teflon filters

for gravimetric analysis

§

§ Sequential sampler with Sequential sampler with multi

multi--day capabilityday capability

§

§ 4 units used in our study 4 units used in our study (3 teflon and 1 quartz)

(6)

R&P Coarse Particle TEOM

§

§ Modified PM10 inlet Modified PM10 inlet

aspirating at 50 lpm (actual)

§

§ PM10 aerosol is fractionated PM10 aerosol is fractionated by a custom virtual impactor

by a custom virtual impactor

(2 lpm coarse flow and 48 lpm

fine flow)

§

§ PMc fraction is heated to 50 C PMc fraction is heated to 50 C to remove particle bound

to remove particle bound

water

§

§ Coarse aerosol is collected Coarse aerosol is collected and quantified by a standard

and quantified by a standard

TEOM sensor

§

(7)

Tisch SPM

-

613D Dichot Beta Gauge

§

§ Standard PM10 inlet Standard PM10 inlet

aspirating at 16.7 lpm (~std)

aspirating at 16.7 lpm (~std) §

§ Aerosol heated >25CAerosol heated >25C

§

§ Aerosol fractionation by Aerosol fractionation by custom virtual impactor

custom virtual impactor §

§ PM2.5 and PMc mass PM2.5 and PMc mass

collected on polyflon tape

roll

roll §

§ PM2.5 and PMc mass PM2.5 and PMc mass quantified hourly using

quantified hourly using

separate beta sources and

detectors

detectors §

(8)

TSI Model 3321 Aerodynamic Particle Sizer

§

§ Standard PM10 inlet aspirating Standard PM10 inlet aspirating at 16.7 lpm (actual)

at 16.7 lpm (actual)

§

§ Isokinetic fraction of PM10 Isokinetic fraction of PM10 aerosol removed at 5 lpm and

aerosol removed at 5 lpm and

enters the APS inlet

§

§ APS sizes individual particles APS sizes individual particles aerodynamically using time of

aerodynamically using time of

flight approach

§

§ Single particle volume converted Single particle volume converted to mass using mean density

to mass using mean density

provided by user

§

§ Total aerosol mass is sum of Total aerosol mass is sum of individual particle mass

individual particle mass

§

§ APS provide only PMc; not APS provide only PMc; not applicable for PM2.5 or PM10

applicable for PM2.5 or PM10

§

§ Only sampler in study which Only sampler in study which

provides detailed size distribution

information

§

(9)

Mobile Sampling Platform

(Side View)

(10)

Mobile Sampling Platform

(Front View)

(11)

Sampler Performance Issues

§

Relative bias versus collocated FRMs

_{Relative bias versus collocated FRMs}

§

Precision (2 or 3 samplers of each type)

_{Precision (2 or 3 samplers of each type)}

§

Field reliability

_{Field reliability}

§

Evaluation under a wide range of

_{Evaluation under a wide range of}

weather conditions and aerosol types

(12)

QA/QC Initiatives

§

§ QAPP was reviewed and approved by EPAQAPP was reviewed and approved by EPA

§

§ Study design and operation passed EPA’s systems Study design and operation passed EPA’s systems audit

audit

§

§ SOPs were reviewed by the sampler manufacturersSOPs were reviewed by the sampler manufacturers

§

§ Sampler manufacturers were allowed to verify the Sampler manufacturers were allowed to verify the

working condition of their respective samplers prior to

sampling at each site

§

§ Sampling and fractionation components cleaned prior Sampling and fractionation components cleaned prior to each study

to each study

§

§ NISTNIST--traceable sampler calibration equipment was traceable sampler calibration equipment was used for all sampler calibrations and audits

used for all sampler calibrations and audits

§

§ Three performance audits and three field blank tests Three performance audits and three field blank tests were conducted at each site

were conducted at each site

§

§ Replicate weighings were conducted at the site as well Replicate weighings were conducted at the site as well as at EPA’s RTP weighing facility

(13)

Study Details

§

§ Using 22 hour, daily sampling periods for comparisons Using 22 hour, daily sampling periods for comparisons (11 am to 9 am local time)

(11 am to 9 am local time)

§

§ Chemical analysis (XRF, IC, thermal optical) of Chemical analysis (XRF, IC, thermal optical) of archived filters will provide particle composition,

archived filters will provide particle composition,

which may explain observed sampler performance

§

§ Full data set from all three sites will be ready by Feb Full data set from all three sites will be ready by Feb 2004. Detailed results from Gary, IN will be presented

2004. Detailed results from Gary, IN will be presented

at the October AAAR meeting

(14)

Study Sites

§

§ RTP, NC (10 days of shakedown tests, Jan. 2003)RTP, NC (10 days of shakedown tests, Jan. 2003)

§

§ Gary, IN (30 days of tests under cold, snow/rain, Gary, IN (30 days of tests under cold, snow/rain, variable PM2.5/PM10 ratios, March

variable PM2.5/PM10 ratios, March--April, 2003)April, 2003)

§

§ Phoenix, AZ (30 days of tests under hot, dusty Phoenix, AZ (30 days of tests under hot, dusty conditions, consistently low PM2.5/PM10 ratios,

conditions, consistently low PM2.5/PM10 ratios,

May

May--June, 2003)June, 2003)

§

§ Riverside, CA (30 days of tests under warm Riverside, CA (30 days of tests under warm

conditions, higher PM2.5/PM10 ratios than Phoenix,

July

(15)

Gary, IN

(16)

GARY, IN SIZE DISTRIBUTION DATA March - April, 2003 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Sample Day PM2.5/PM10 Concentration Ratio GARY, IN FRM DATA PM2.5/PM10 Range = 0.32 to 0.83; Mean = 0.55

(17)

PMc FRM MEASUREMENTS - GARY vs RTP WEIGHING Gary, IN (March - April, 2003)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 0 5 10 15 20 25 30 35 SAMPLE DAY PMc Conc. (micrograms/m 3 ) GARY WEIGHING RTP WEIGHING RTP/Gary = 1.00

(18)

INTERMANUFACTURER PM2.5 FRM MEASUREMENTS (RTP WEIGHING)

Gary, IN (March - April, 2003)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY PM2.5 Conc. (micrograms/m 3 ) AND R&P BGI INTERMANUFACTURER CV = 1.5% Max PM2.5 = 46.9 ì g/m3 Min PM2.5 = 10.3 ì g/m3 Mean PM2.5 = 22.8 ì g/m3

(19)

INTERMANUFACTURER PM10 FRM MEASUREMENTS (RTP WEIGHING)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY PM10 Conc. (micrograms/m 3 ) AND R&P BGI INTERMANUFACTURER CV = 2.4% Max PM10 = 84.9 ì g/m3 Min PM10 = 22.6 ì g/m3 Mean PM10 = 42.6 ì g/m3

(20)

INTERMANUFACTURER PMc FRM MEASUREMENTS (RTP WEIGHING)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY PMc Conc. (micrograms/m 3 ) AND R&P BGI INTERMANUFACTURER CV = 5.7% Max PMc = 58.1 ì g/m3 Min PMc = 4.5 ì g/m3 Mean PMc = 19.9 ì g/m3

(21)

DICHOT AND FRM TIMELINE (PM_2.5) GARY, IN (MARCH - APRIL, 2003)

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PM 2.5 Conc. (micrograms/m 3 ) PM2.5 FRM R&P DICHOTS DICHOT/FRM = 0.99

(22)

Dichot versus FRM PM2.5 Concentrations Gary, IN (March - April, 2003)

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 FRM PM2.5 Concentrations (micrograms/m3)

Dichot PM2.5 Concentrations (micrograms/m3)

Dichot = 0.99*FRM + 0.0 R Square = 0.998

(23)

DICHOT AND FRM TIMELINE (PMc) Gary, IN (March - April, 2003)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PMc Conc. (micrograms/m 3 ) FRM PMc (PM10-PM2.5) R&P DICHOTS Dichot/FRM = 0.89 Dichot = 0.87*FRM + 0.4 R square = 0.969

(24)

DICHOT AND FRM TIMELINE (PM10) Gary, IN (March - April, 2003)

0 10 20 30 40 50 60 70 80 90 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PM10 Conc. (micrograms/m 3 ) PM10 FRM R&P DICHOTS Dichot/FRM = 0.94 Dichot = 0.95*FRM - 0.5 R square = 0.981

(25)

TISCH SPM-613D AND FRM TIMELINE (PM_2.5) Gary, IN (March - April, 2003)

0 10 20 30 40 50 60 70 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PM 2.5 Conc. (micrograms/m 3 ) PM2.5 FRM TISCH SPM-613D TISCH CV = 7.1% TISCH/FRM = 1.26 Tisch = 1.17*FRM + 1.6 R square = 0.948

(26)

TISCH SPM-613D AND FRM TIMELINE (PM_c) Gary, IN (March - April, 2003)

0 10 20 30 40 50 60 70 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PMc Conc. (micrograms/m 3 ) FRM PMc (PM10-PM2.5) TISCH SPM-613D TISCH/FRM = 0.91 TISCH = 0.89*FRM + 0.3 R square = 0.979 TISCH CV = 10.5%

(27)

R&P COARSE TEOM AND FRM TIMELINE (PMc) Gary, IN (March - April, 2003)

0 10 20 30 40 50 60 70 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PMc Conc. (micrograms/m 3 ) PMc (PM10-PM2.5)

R&P CONTINUOUS COARSE

TEOM PMc/FRM = 0.69

TEOM PMc CV = 4.4% TEOM = 0.68* FRM + 0.18

(28)

TSI APS vs FRM PMc Concentrations Gary, IN (March - April)

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY PMc Conc. (micrograms/m 3 ) APS 1 PMc APS 2 PMc Mean FRM PMc APS/FRM = 0.42 APS CV = 16.8% APS = 0.48*FRM -1.4 R square = 0.795

(29)

(30)

Phoenix versus RTP FRM Weighing May - June 2003 0.0 50.0 100.0 150.0 200.0 250.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Sampling Day

Mass Concentration (micrograms per

cubic meter) PM2.5 (RTP) PM10 (RTP) PMc (RTP) PM2.5 (Site) PM10 (Site) PMc (Site) PM10 PMc PM2.5 Inter-manufacturerer PM2.5 CV = 3.4% Inter-manufacturerer PMc CV = 3.3% Inter-manufacturerer PM10 CV = 3.6% PM2.5/PM10 Range = 0.10 to 0.28: Mean = 0.22

(31)

Dichot PM2.5 versus FRM PM2.5 Phoenix, AZ May - June, 2003 0.0 5.0 10.0 15.0 20.0 25.0 30.0 0.0 5.0 10.0 15.0 20.0 25.0

FRM PM2.5 (micrograms per cubic meter)

Dichot PM2.5 (micrograms per

cubic meter)

Dichot = 1.24 * FRM - 1.6 R Square = 0.97

(32)

DICHOT AND FRM TIMELINE (PMc) Phoenix, AZ (May - June, 2003)

0.0 50.0 100.0 150.0 200.0 250.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PMc Conc. (micrograms/m 3 ) PMc (PM10-PM2.5) R&P DICHOTS Mean Dichot/FRM = 0.80 Dichot = 0.70*FRM + 5.0 R square = 0.98

(33)

DICHOT AND FRM TIMELINE (PM10) Phoenix, AZ (May - June, 2003)

0.0 50.0 100.0 150.0 200.0 250.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PM10 Conc. (micrograms/m 3 ) PM10 FRM R&P DICHOTS Mean Dichot/FRM = 0.84 Dichot = 0.75*FRM + 5.9 R square = 0.98

(34)

Tisch & FRM PM2.5 Concentrations Phoenix AZ: May - Jun, 2003

0 10 20 30 40 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day PM2.5 concentrations (ug/m3) Tisch PM2.5-1 Tisch PM2.5-2 Tisch PM2.5-3 Mean FRM-PM2.5 Average Tisch CV = 5.9 % Mean Tisch/FRM = 1.70 Tisch = 2.03*FRM - 3.4 R square = 0.947

(35)

Tisch, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003

0 20 40 60 80 100 120 140 160 180 200 220 1 3 5 7 9 ₁₁ ₁₃ ₁₅ ₁₇ ₁₉ ₂₁ ₂₃ ₂₅ ₂₇ ₂₉ Sample Day PMc concentrations (ug/m3) Tisch PMc-1 Tisch PMc-2 Tisch PMc-3 Mean FRM-PMc Average Tisch PMc CV = 9.5 % Mean Tisch/FRM = 1.04 Tisch = 0.92*FRM + 6.0 R square = 0.996

(36)

TEOM, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003

0 20 40 60 80 100 120 140 160 180 200 220 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day PMc concentrations (ug/m3) TEOM PMc1 TEOM PMc2 TEOM PMc3 FRM-PMc Average TEOM PMc CV = 6.6 % Average TEOM/FRM Ratio = 1.05 TEOM = 0.79*FRM +12.8

(37)

APS, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003

0 20 40 60 80 100 120 140 160 180 200 220 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day PMc concentrations (ug/m3) APS 1-PMc APS 2-PMc FRM PMc Average APS PMc CV (Days 1-15) = 2.2 %

Average APS/FRM = 0.55 APS = 0.56*FRM -0.2 R square = 0.992

(38)

PM Size Distributions (TSI APS) Gary,IN and Phoenix,AZ

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 1 10 100 Log Dp

Delta Mass/Delta Log Dp

Gary Phoenix

(39)

Summary of Results

(independent of site)

§

§ FRMs show strong interFRMs show strong inter--manufacturer precision manufacturer precision (CV<4% for all three metrics) with no tendency for

(CV<4% for all three metrics) with no tendency for

producing negative PMc values

§

§ FilterFilter--based dichots show strong precision (CV<4% based dichots show strong precision (CV<4%

for all metrics)

§

§ Site weighing results agree closely with RTP resultsSite weighing results agree closely with RTP results

§

§ Precision of the semiPrecision of the semi--continuous samplers is continuous samplers is

considered to be acceptable

§

§ Correlation (RCorrelation (R22_{) of all continuous samplers is}_{) of all continuous samplers is}

typically strong versus the collocated FRMs

(40)

0.55 0.42 APS/FRM PMc TSI APS 1.04 0.91 Tisch/FRM PMc 1.16 1.09 Tisch/FRM PM10 1.70 1.26 Tisch/FRM PM2.5 TISCH 1.05 0.69 TEOM PMc/FRM TEOM PMc 0.79 0.90 Dichot/FRM PMc 0.84 0.94 Dichot/FRM PM10 1.09 1.00 Dichot/FRM PM2.5 DICHOTS 0.18 0.55 PM2.5/PM10 Ratio 0.10 - 0.28 0.32 - 0.83 PM2.5/PM10 Range 55.6 19.9 PMc Mean ( ìg/m3₎ SITE AEROSOL PHOENIX, AZ GARY, IN