Characteristics of Particulates from Gasoline Combustion Strategies

Transcription

1 Characteristics of Particulates from Gasoline Combustion Strategies David Rothamer Engine Research Center slide 1

2 Overview Background Health effects PM Regulations Particulate morphology definitions Spark Ignition Operating condition influence and fuel influence on PSDs, morphology, composition Compression Ignition Operating condtion impact on PSDs, composition, and morphology Conclusions slide 2

3 slide 3 BACKGROUND

4 Deposited Fraction of Ambient Aerosol [%] Health Implications of Particle Matter slide 4 SI PM size range Diameter [nm] Particulate matter deposition in respiratory system as a function of particle diameter [1] Reductions in PM mass do not imply reduced PM # PM is potentially linked to increased Mortality rate and cancer rate Occurrences of respiratory illnesses Particulates from gasoline engines have potential for deep penetration due to size [1] Witschger, O. and Fabries, J.-F. (2005), Ultrafine Particles and Occupational, Health, Paris, INRS,

5 Euro 6 # Euro 6 mass LEV II LEV III EPA Tier II EPA Tier III Partcle Number [10 12 / mile] Particulate Matter (PM) Regulations slide Comparison of light-duty particle standards on an estimated number basis** Light-duty Particulate Matter Standards Standard PM Mass PM Number Euro mg/km (7.2 mg/mi) 6 x /km* (9.7 x /mi) LEV II (CA) 10 mg/mi No regulation LEV III (CA, proposed) 3 mg/mi (2017) 1 mg/mi (2025) No regulation EPA tier II 10 mg/mi No regulation EPA tier III (proposed) 3 mg/mi No regulation * Standard is 6x10 12 /km until Currently only measures solid particles >23 nm in mobility diameter **Assumes 50 nm spherical particles and PM density of 1.86 g/cm 3 Future regulations will be challenging to meet requires understanding of particulate characteristics

6 Particulate Morphology Definitions Fractal Dimension R g d p Linear Agglomerate D f 1 Diffusion-limited Cluster-Cluster Agglomerate D f 1.8 Compact Spherical Agglomerate D f 3.0 slide nm R g = Radius of gyration d p = Primary Particle diameter D f = Fractal dimension

7 SPARK IGNITION PARTICULATE CHARACTERISTICS slide 7

8 Experimental Setup: Engine Single cylinder research engine Spark-Ignition Direct-Injection (SIDI) Tier II EEE fuel (certification gasoline) Fuel Injector 45 Stroke: 94.6 mm 35 Compression Ratio: Displacement: 549 cm 3 Much of the data to be shown taken in this engine slide 8 Bore: mm

9 Influence of Engine Operating Condition - Equivalence Ratio Indicated Specific Particulate Number [#/kw-hr] ~5X increase Increasing Φ Typical range for A/F dithering PMP* PSDs DI operation (EOI 220, 2100 rpm, IMEP = 334 kpa, CA50 = 8 CAD) [2] Equivalence Ratio Particle number DI operation (EOI 220, 2100 rpm, IMEP = 334 kpa, CA50 = 8 CAD) Strong influence of equivalence ratio Wall/piston films Rich regions in vaporized mixture? slide 9 [2] S. Sakai, et al., SAE

10 Influence of Engine Operating Condition - Equivalence Ratio dn/dlog(d p ) [#/cm 3 ] slide 10 PMP* Increasing Φ PSDs DI operation (EOI 220, 2100 rpm, IMEP = 334 kpa, CA50 = 8 CAD) [2] PSDs prevaporized and premixed [3] (2100 rpm, IMEP = 334 kpa, CA50 = 8) Strong influence of equivalence ratio Wall/piston films Rich regions in vaporized mixture? Threshold appears to be = Increasing Φ Mobility Diameter [nm] [3] M. Hageman and D. A. Rothamer,, 8th US National Combustion Meeting, Park City, UT,

11 Total Concentration [1/cm 3 ] 1.E+08 Operating Condition/Fuel Influence EEE E20 E85 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 E85 resulted in substantially lower particulate # slide 11

12 Total Concentration [1/cm 3 ] 1.E+08 Operating Condition/Fuel Influence Results for E20 are more complex EEE E20 E85 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 slide 12

13 dn/d(logd p ) [#/cm^3] Comparison to Diesel PSDs EEE EOI 220 EEE Rich EEE EOI Particle size distribution for three single-injection Diesel operating conditions [4]: Mode 3 = 1800 rpm, 7.3 bar, Mode 4 = 1200 rpm, 5.7 bar, Mode 5 = 1200 rpm, 14.2 bar Note: Particulate concentration magnitudes should not be compared directly between diesel and SIDI data due to significant differences in fueling and air flow rates. Near stoichiometric cases show much larger fraction of small particles compared to diesel distributions Rich condition shows similarity in shape to diesel distributions potentially indicating similarity in formation conditions slide 13 [4] Wirosakunchai, et al., SAE Midpoint Diameter, D p [nm] Particle size distribution for EEE at three operating conditions: EOI 220, EOI 280, Rich

14 SI Particulate Morphology 200 nm 200 nm DI operation EOI 220, 2100 rpm, IMEP = 330 kpa, CA50 = 8 slide 14

15 Particulate Morphology - Primary Particle Size Distribution of average primary particle size in aggregates, 1500 rpm, 280 BTDC injection timing, 8 bar load [5] Data from ORNL not engine shown earlier Wide range of primary particle sizes compared to CDC Data indicate two particle types at some conditions Agglomerates with larger ave. primary particle dia. Agglomerates with smaller ave. primary particle dia. Some particles are a mix Indicates formation under differing conditions Also seen in TEM images shown (last slide) and other investigations [Zelenyuk][Lee] slide 15 [5] T.L. Barone, Atmospheric Environment, 49 (2012)

16 Particulate Morphology- Aggregate Size Aggregate Radius of Gyration histograms for aggregates with (a) small d p and (b) bigger d p EOI 310, 2100 rpm, IMEP 650 kpa[] Aggregate size tied to primary particle diameter make some sense slide 16 [6] K. Lee, et al., SAE-ICE, 2013.

17 Particulate Morphology- Fractal Dimension Measurements of vacuum aerodynamic diameter at fixed mobility diameter give eff Effective density versus moblity diameter- EEE, =0.98, EOI 220, 2100 rpm, IMEP=334 kpa, CA50=8 CAD [7] slide 17 [7] Zelenyuk, et al., DEER conference, log-log plot- Fractal dimension D fa = slope + 3 D f generally between 1.9 and 2.2 (dependent on operating condition) Larger than typical for diesel soot More compact aggregates Difference in y-offset due to primary particle size difference

18 Particulate Composition - Fuel Effects Large fraction of non-volatile (bound) organic carbon Not removable with Volatile Particle Removers Large PAH fraction (white bars) Less ordered nano-structure than diesel soot, more amorphous (HRTEM measurements) Nature of organic fraction is still up for debate Organic/Elemental carbon ratio for various SIDI operating conditions [7] slide 18 [7] Zelenyuk, et al., DEER Conference, 2012

19 GASOLINE COMPRESSION IGNITION (GCI) PARTICULATE CHARACTERISTICS slide 19

20 GCI Engine and Operating Conditions GM 1.9 L Single-Cylinder Engine Specifications Operating conditions tested Compression Ratio [-] 16.5 Bore [mm] 82 Stroke [mm] 90.4 Connecting Rod Length [mm] 161 Intake Valve Open [ ATDC] 344 Intake Valve Close [ ATDC] -132 Exhaust Valve Open [ ATDC] 112 Exhaust Valve Close [ ATDC] 388 Condition slide 20 net IMEP (bar) Engine Speed (RPM) EGR (y/n) Inlet O2 conc (%) Inlet Pressure (kpa) Inlet Temp ( C) # Inject. SOI 1 SOI 2 SOI 3 fuel flow (kg/hr) n N/A n N/A n N/A n N/A y N/A y

21 dn/dlog(d p ) [1/cm 3 ] GCI PSDs Multiple injection bar IMEP 2 inj. Conditions LL LL2 LL3 LL4 5.5 bar IMEP HL5 2 inj. HL Mobility Diameter [nm] 13.4 bar IMEP 3 inj. Gasoline compression-ignition size distributions at low load and high load Condition LL1 LL2 LL3 LL4 HL1 HL2 IMEP [bar] Speed [rpm] SOI [ btdc] SOI [ btdc] SOI3 [ btdc] Major influence of load on PM PSD Median diameter shifts from ~10 nm to nm from low load to high load Minimal influence of injection timing at light load High load distributions resembles diesel distributions in shape slide 21

22 [7] Zelenyuk, et al., DEER conference, GCI Morphology and Composition Condition LL1 LL2 LL3 LL4 HL1 HL2 EC OC PAHs

23 Conclusions SI particulates Large variation in PSDs between operating conditions Little evidence of volatile nucleation mode particles Two primary particle sizes often present Large fraction of bound organics GCI particulates Spherical organic particles at low load combined with fractal aggregates Fractal like soot at high load Fractal dimension of 2.1 similar to SI particluates slide 23

24 References 1. Witschger, O. and Fabries, J.-F. (2005), Ultrafine Particles and Occupational, Health, Paris, INRS, S. Sakai, M. Hageman, and D. A. Rothamer, "Effect of Equivalence Ratio on the Particulate Emissions from a Spark-Ignited, Direct-Injected Gasoline," SAE Technical Paper , M. Hageman and D. A. Rothamer, "Sensitivity Analysis of Particle Formation in a Spark-Ignition Engine during Premixed Operation," presented at the 8th US National Combustion Meeting, Park City, UT, E. Wirojsakunchai, E. Schroeder, C. Kolodziej, D. E. Foster, N. Schmidt, T. Root, T. Kawai, T. Suga, T. Nevius, and T. Kusaka, "Detailed Diesel Exhaust Particulate Characterization and Real-Time DPF Filtration Efficiency Measurements During PM Filling Process," SAE Technical Paper , T. L. Barone, J. M. E. Storey, A. D. Youngquist, and J. P. Szybist, "An analysis of direct-injection spark-ignition (DISI) soot morphology," Atmospheric Environment, vol. 49, pp , Mar K. Lee, H. Seong, S. Sakai, M. Hageman, D. Rothamer, Detailed Morphological/Chemical Properties of Nanoparticles from Various Engine Combustion Sources, SAE-ICE, Zelenyuk, et al., Characterization of Pre-commercial Gasoline Engine Particulates Through Advanced Aerosol Methods, DEER conference, N. Matthias, C. Farron, D. Foster, M. Andrie, R. Krieger, SAE Technical Paper, (2011). slide 24

25 Acknowledgements Funding: General Motors through the Collaborative Research Lab. At UW-Madison Collaborators: Alla Zelenyuk (PNNL), Keyong Lee (ANL), ERC collaborators Students: Steve Sakai, Mitch Hageman, Cory Adams, Paul Loeper slide 25

26 slide 26 Questions?

27 Zelenyuk, et al., DEER conference, GCI - Composition Condition LL1 LL2 LL3 LL4 HL1 HL2 Condition LL1 LL2 LL3 LL4 HL1 HL2 IMEP [bar] IMEP [bar] Speed [rpm] Speed [rpm] SOI1 [ btdc] SOI1 [ btdc] SOI2 [ btdc] SOI2 [ btdc] SOI3 [ btdc] SOI3 [ btdc] 10 SPLAT II recorded MS: Av. mass spectrum for HL1 (a) Av. mass spectrum for LL2 (b) 2 types of particles were produced during LL runs Mass spectra of the fractal soot particles produced during LL1 LL4 runs (c) Mass spectra of the compact organic particles produced during LL1 LL4 runs (d)

28 Particulate Composition - Volatile Particle Removers Impact of thermodenuder and evaporative chamber (VPRs) on PSD for cold start condition [7] slide 28 [7] N. Matthias, et al., SAE Technical Paper (2011).

29 RCCI 75% Gasoline PFI 25% Diesel- DI 4.3 bar BMEP, 2300 rpm CDC and Diesel PCCI utilize EGR CDC 29% PCCI 40% Different CA 50 slide 29

30 slide 30 GDCI

31 Volume % Operating Conditions Baseline condition Parameter Unit Value Tol. Engine Speed [RPM] 2100 ± 5 Injection Timing [ btdc] IMEP (Gross) [kpa] 334 ± 6 Equivalence Ratio [-] 0.98 ± Spark Advance [ btdc] 25 - CA50 [ atdc] 8 ± 0.5 Injection Pressure [MPa] 11 ± 0.1 Intake Temperature [ C] 45 ± 2 Oil Temperature [ C] 90 ± 2 Coolant Temperature [ C] 90 ± 2 Fuel Flow [mg/cycle] 11 ± 0.15 Air Flow [mg/cycle] 162 ± 5 EGR % 0 ±0 All parameters from table above were held constant Fifteen equivalence ratios studied ranging from Stoichiometric condition verified experimentally by locating the intersection of the CO and O 2 exhaust concentrations Equivalence Ratio O2 CO 1.10 Verification of stoichiometric condition 1.15 slide 31

32 Equivalence Ratio Effect Equivalence Ratio Sweep Test Matrix C/O IMEP [kpa] COV of IMEP [%] CA50 [CAD] slide 32

33 Motivation Ultrafine particles negatively influence human health Euro 6 regulation sets a number-based limit of 6 x 1011 particles/km over the new European Drive Cycle for vehicles produced after September 2017 Current spark-ignition direct injection (SIDI) engine designs may have difficulty meeting future number-based standards Use of biofuels blended with gasoline (ethanol, butanol, etc.) complicates the particulates picture Euro *Piock, et al; SAE Tech. Paper slide 33

34 Operating Conditions Speed Injection Timing IMEP Φ Spark Timing CA-50 Injection Press. Intake Temp. Oil Temp. Coolant Temp. [RPM] [ btdc] [kpa] [-] [ btdc] [ atdc] [MPa] [ C] [ C] [ C] EOI EOI Heavy Load Rich Lean MBT Medium Press. Cold Start slide 34

35 Fuel-Neutral Basic Test Matrix Page 35 RED text highlights variations from the baseline condition (EOI 220) Collaborative Research Laboratory