PMP Comparison Study of Particle Measurement Systems Martin Mohr & Urs Lehmann EMPA Swiss Federal Laboratories for Materials Research and Testing Laboratory for Internal Combustion Engines and Furnaces Dübendorf, Switzerland www.empa.ch contact: martin.mohr@empa.ch Abstract The international programme PMP under the auspices of the UNECE WP29/GRPE is focused on a new approach to the measurement of particles in vehicle exhaust emissions, which may be used to complement or to replace the existing regulated mass based measuring system. This abstract deals with a comparative study of particle instruments within the collaborative programme PMP. The objective was to gain comparable information about repeatability, linearity, detection limit, response time and correlation of about 2 particle measurement instruments. Simultaneous measurements with all instruments were carried out on a real diesel exhaust of a modern heavy-duty engine. The tests were performed on two different emission levels: downstream of a particle trap representing the technical lowest available level and a level about two third of the future emission limit (Euro 4). In a second phase an aerosol generator (CAST) was applied as an emission source to obtain engine independent results of the performance of the instruments. The group of investigated measurement systems comprises a wide range of measurement principles that apply optical, electrical and filter methods and combination of these. The instruments to quantify the particle concentration use different metrics (number, length, surface, mass).
MOTIVATION Demands on a future measurement method: Quantification of particle emissions on sub Euro 4 and post-trap level in a reliable way taking into account the present knowledge of health impact CONTENTS Investigated instruments Experimental set-up Results > Repeatability > Sensitivity > Detection limit > Time response > Size measurement > Absolute values > Correlations > Linearity Conclusions
MEASUREMENT TECHNIQUES Instrument No Manufacturer Dilution Principle Diffusion Charger 2 Matter, TSI hot dil DMS 1 Cambustion CVS SMPS/CPC 3 Grimm, TSI hot dil ELPI 1 Dekati hot dil MASMO 1 Dekati hot dil El. Diffusion Battery 1 Matter hot dil Photoelectric Sensor 1 Matter hot dil TEOM 1 R&P CVS PM gas analysis 1 Horiba CVS Coulometry 1 PFD Grav. Filter Method 1 CVS Photoacoustic Sensor 1 TU Munich CVS LII 1 Esytec CVS Light Scattering 2 Grimm, Sensors hot dil Light Extinction 3 Hartridge, Wizard, AVL hot dil El. Charging Impactor Filter Optical det. EMISSION SOURCES HD Engine 7 l, 6 Cyl., TC, IC, EURO 3, (provided by Volvo) equipped with particle filter (CRT-System) Diesel fuel, <1 ppm sulphur Emission levels post-trap partial bypass of particle trap 4% below Euro 4 Combustion aerosol generator (CAST) Settings (as determined by SMPS, TSI) 4 concentrations: 5-15 cm -3 (total number) 2 sizes: 4 nm and 14 nm (mode)
REPEATABILITY OF ENGINE PARAMETERS relative standard deviation [%] 12% 1% 8% 6% 4% 2% % 1% exhaust flow engine power ETC total high emission all measurements CO T.HC NOx NO (r) CO2 Temp after CRT ETC HIGH.53%.1% 5.2% 1.3% 1.2% 2.%.5%.9% REPEATABILITY High emission level relative standard deviation [%] 2% 18% 16% 14% 12% 1% 8% 6% 4% 2% % DMS 1% Gra LII Mexa PASS Mas -Mo Coulometry Light ext ETC total high emission all measurements C3 C4 C5 C6 C6E C7 C8 R1 R11 R11E R12 R14 R15 R16 R17 R18 R19 R2 R21 R22 R23 R24 total 12% 7.7% 3.7% 1.8% 4.8% 2.% 7.9% 19% 23% 144% 7.3% 1.4% 5.% 7.6% 1.9% 1.3% 3.4% 1.1% 1% 7.5% DC DB CPC ELPI
REPEATABILITY Post-trap emission level 8% relative standard deviation [%] 7% 6% 5% 4% 3% 2% 1% % DMS 3% Gra LII Mexa PASS Mas -Mo ETC total low emission all measurements C3 C4 C5 C6 C6E C7 C8 R1 R11 R11E R12 R13 R14 R15 R16 R17 R18 R19 R2 R21 R22 R23 R24 total 121% 2% 7.6% 6.1% 7.4% 13% 24% 41% 783% 34% 1% 1% 4% 24% 53% 33% 29% 16% 7.1% DC DB CPC DC ELPI POST-TRAP EMISSIONS 2. Ratio post-trap to background concentration Ratio ETC low / Background 1.5 1..5 - Mas -Mo Coulometry Coulometry C3 C4 C5 C6 C6E C7 C8 R1 R11 R11E R12 R13 R14 R15 R16 R17 R18 R19 R2 R21 R22 R23 R24 ETClow/bg 15 1.4 1.2 1.1.9 1.45.34.22.36 1.3 1.5 275 1. 1.6 1.3 1.7
SENSITIVITY Ratio high to low emission level Ratio ETC high / ETC low [-] 16 14 12 1 8 6 4 2 Filter weighing LI2SA MEXA PM137 PASS CPC ELPI EAD PAS C3 C4 C5 C6 C6E C7 C8 R1 R11 R11E R12 R13 R14 R15 R16 R17 R18 R19 R2 R21 R22 R23 R24 ratio 51.8 7.2 1.3 5.6 52.3 28. 521 8.17 259 438 165 1339 37.8 143 2164 665 1527 9.88.94 DETECTION LIMIT Number Instruments 1.E+5 8.E+4 Number [cm-3] 6.E+4 4.E+4 2.E+4.E+ DMS Dust SMPS+/ 5 Monitor CPC EDB2 PM3 CPC SMPS/ ELPI 322 CPC LOD 1.6E+6 n.a. n.a. 1.7E+4 8.9E+4 4.E+3 4.E+3 5.3E+3
DETECTION LIMIT Mass Instruments.4.3 Mass [mg/m3].2.1 LI2SA PASS PAS LOD.37.29.2.357.139.189.61.4 n.a..36 n.a. 1.22 Coul Coul EC Gravi metry Mexa Mexa EC Mass TEOM Mo DPSO AVL 439 TIME RESPONSE Particle concentration [a.u.] 1.4 1.2 1.8.6.4.2 LII DC CPC Engine Power [kw] -5 1 16 Time [s] 22 28
SIZE MEASUREMENTS Normalised value 1.2 1.8.6.4.2 C3 R12 R13 R16 R18 R21 R22 C5 R1 1 1 Particle size [nm] 1 1 ABSOLUTE VALUES Number - high emission level Number concentration [1/Ncm3] 3.E+7 2.5E+7 2.E+7 1.5E+7 1.E+7 5.E+6.E+ DMS R15, R2 & R23 - different metrics, transferred to number with the assumptions: spheres, avg. size 6 nm, density 1 g/cm3 R21 fixed to small size channel at 8 nm CPC ETC total high emission absolute number values all measurements EDB CPC ELPI PM3 C3 R12 R13 R15 R16 R18 R19 R2 R21 R22 R23 2.3E+7 5.3E+3 5.1E+6 1.2E+7 9.4E+6 2.8E+6 7.9E+6 4.1E+6 2.7E+6 9.6E+6 2.2E+6
ABSOLUTE VALUES Mass - high emission level Mass concentration [ng/ncm3] 1 9 8 7 6 5 4 3 2 1 Netscape 7..lnk Gra LII ETC total high emission absolute mass values all measurements Mexa Pass MasMo Coulometry opacity C4 C5 C6 C6E C7 C8 R1 R11 R11E R14 R17 R24 2.16 1.68 1.86.99 2.19 2.56 2.38 1.53 1.6 8.68 5.43 PAS CORRELATION 3.E+7 4 Number conc. [1/Ncm3] R19 Number conc. [1/Ncm3] R19 2.5E+7 2.E+7 1.5E+7 1.E+7 5.E+6.E+ 3.E+7 2.5E+7 2.E+7 1.5E+7 1.E+7 5.E+6 CPC CPC R 2 R 2 =.96 without SM idle H LI2SA..5 1. 1.5 2. 2.5 3. Mass conc. [ng/ncm3] C5 Gravim R 2 =.77 w/o SM idle H w/o SM B1 H R 2 =.77.E+ 1 2 3 4 5 6 Mass conc. [ng/ncm3] C4 Mass conc. [ng/ncm3] C8 Mass conc. [ng/ncm3] C5 3.5 3 2.5 LI2SA PASS 2 1.5 1.5 3 2.5 2 1.5 1.5 R 2 = =.99.99.5 1 1.5 2 2.5 3 3.5 Mass conc. [ng/ncm3] C5 R 2 =.78 LI2SA R 2 =.78 Gravim 1 2 3 4 5 6 Mass conc. [ng/ncm3] C4
CAST - REPEATABILITY rel. deviation of CAST meas. [%] C5 C8 METAS R1 R15 R16 R17 R19 R2 R21 R22 8% 6% 4% 2% % -2% -4% -6% CAST size setting B -8% 1% 3% 6% 9% Conc. level. Figure 1: Relative Relative deviation deviation CAST measurements of concentration before & after within the engine two test weeks period / CAST size B LINEARITY 12% C3 / before C3 / after C5 / before C5 / after C8 / before C8 / after R1 / before R1 / after R15 / before R15 / after R16 / before R16 / after R17 / before R17 / after R19 / before R19 / after R2 / before R2 / after R21 / before R21 / after R22 / before R22 / after 1% 8% 6% 4% 2% % 1% 3% 6% 9%
CONCLUSIONS SENSITIVITY Number-based methods higher sensitivity than mass-based methods better distinction of post-trap & GDI emissions DETECTION LIMIT Advanced mass-based methods lower detection limit than regulated filter method REPEATABILITY Several instruments similar or better repeatability than regulated filter method Key position: volatile material Data on Reproducibility needed CORRELATION Regulated filter method poor correlation to most other instruments CALIBRATION In contrast to mass-based methods a traceable standard does not yet exist for particle number, length or surface. Report www.empa.ch/pmp Thanks to my EMPA colleagues Sepp Rütter Roland Graf Rolf Ziegler Urs Lehmann Daniel Schreiber Hans-Christian Opstad Uzi Saghi Funded by ASTRA EMPA BUWAL Manufacturers (instruments operation) METAS (CAST operation) R&P (cyclones & filter holders) SUVA (coulometry) VOLVO (engine)
back to index