Zürich Testing on Fuel Effects and Future Work Programme

Transcription

1 Zürich Testing on Fuel Effects and Future Work Programme Benjamin Brem 1,2, Lukas Durdina 1,2 and Jing Wang 1,2 1 Empa 2 ETH Zürich FORUM on Aviation and Emissions Workshop Amsterdam

2 Context: fuel chemistry and soot emissions Fuel composition- soot emissions link has been the focus of research for more than half a century (e.g. smoke point) Fuel rich pockets within the flame promote reactions that form heavy PAHs which subsequently pyrolyze and form soot PM Fuel aromatic content is critical Aliphatic species can also form ring structures and subsequent PAHs, but reaction rates are typically slower than reactions on already present aromatic species PAHs and soot have typically short lifetimes and most of them are oxidized in fuel lean zones Soot formation mechanism in premixed flames (Bockhorn 1983) 1

3 Compositional parameters associated with soot emissions in standard jet fuel Combustion stability, safety and performance have been the criteria for the development of jet fuels Property Unit ASTM Method Jet A-1 (ASTM D1655) Annex 16 Vol. II Appendix 4 Typical Value (ZRH) Total Aromatics % v/v D 1319 < /-0.34 Smoke point mm D 1322 > /-1.3 Naphthalenes % v/v D 1840 < /-0.11 Hydrogen % m/m D 5291 N.A /-0.25 Compositional variations allowed, but large variations are not observed in practice (some exceptions: Canada higher aromatics and South Africa lower aromatics) Annex 16 specifications for engine certification are narrow enough that visible smoke emissions are not affected by compositional variations. What about nvpm mass and number emissions? 2

4 Objectives 1. Investigate the effect of Jet A-1 total aromatics content on nvpm mass and number emissions in a engine certification-like setting 2. Evaluate the influence of poly- vs. mono-aromatics content on the emissions 3. Develop a suitable parameterization of the measured results that can be used to correct engine nvpm emission data for fuel aromatics effects 3

5 Experimental background Majority of measurements performed during the A-PRIDE 7 campaign (2014) on an in-production turbofan Two aromatic solvents (Solvesso 150 and 150ND) injected into fuel supply line to engine Solvent injection at low thrust was not successful, due to inaccurate fuel flow measurement Low thrust experiments repeated at low thrust during A-PRIDE 8 (2015) on the same engine model but different serial number, and service status Only one aromatic solvent (Solvesso 150ND) Property Unit ASTM Method A-PRIDE 7 Base A-PRIDE 8 Base Solvesso 150ND Solvesso 150 Total Aromatics % v/v D > Smoke point mm D <5 Naphthalenes % v/v D Hydrogen % m/m D 7171 (mod)

6 Experimental setup More Information: Brem et al. Environ. Sci. Technol. 2015, 49,

7 Detail: aromatic solvent injection hardware 6

8 Change in nvpm emissions vs. change in aromatics (A-Pride 7) Ei nvpm Mass / Ei nvpm Total Aromatics = 17.8 % v/v Thrust Solvesso 150ND Solvesso Total Aromatics [% v/v] Mass Ei nvpm Number / Ei nvpm Total Aromatics = 17.8 % v/v Thrust Solvesso 150ND Solvesso Total Aromatics [% v/v] Number Aromatics effect is most profound at 30 and 65% thrust nvpm mass changes up to 60% with the 6% increase in total aromatics content Mass is slightly more affected than the number emissions Solvent/ naphthalenes effect at 30% and 65% thrust More Information: Brem et al. Environ. Sci. Technol. 2015, 49,

9 Change in nvpm emissions vs. change in hydrogen (A-PRIDE 7) Ei nvpm Mass / Ei nvpm H = % m/m Mass Solvesso 150ND Solvesso H [% m/m] Thrust [%] b) Number Solvesso 150ND Solvesso Except for 100% thrust, fuel hydrogen mass content shows a better correlation with emissions than the total aromatics content Complex aromatics chemistry is lumped into one predictor Ei nvpm Number / Ei nvpm H = 14.31% m/m H [% m/m] Thrust [%] More Information: Brem et al. Environ. Sci. Technol. 2015, 49,

10 Preliminary: Change in nvpm emissions vs. change in hydrogen A-PRIDE 7 & A-PRIDE 8 Ei nvpm Mass / Ei nvpm Mass, Neat Fuel Mass Thrust [%] 0.0 A-PRIDE 7 Solvesso 150 A-PRIDE 7 Solvesso 150ND 20 A-PRIDE 8 Solvesso 150ND H [% m/m] Ei nvpm Number / Ei nvpm Number, Neat Fuel Number H [% m/m] A-PRIDE 8 data at 65% and 30% thrust fall in line with A-PRIDE 7 data even though base fuel and engine were different Trend of increasing emissions at thrust levels < 30% confirmed with A- PRIDE 8 data 9

11 Results parameterization Multiple linear regression models (e.g. Moore et al. 2015) were attempted but results were indecisive The best fit was obtained with a simple model (similar to Speth et al. 2015) using change in hydrogen mass content and percentage of engine thrust as predictors: EE x = α 0 + α 1 F H Variable α 0 α 1 Adjusted R 2 Ei BC Mass ± ± Ei nvpm Number ± ± Ei Combined ± ± Modeling will be repeated as soon as all A-PRIDE 8 data are available and checked More Information: Brem et al. Environ. Sci. Technol. 2015, 49,

12 Summary and conclusions of the fuel studies A 7% increase in fuel aromatics content increases the nvpm mass and number emissions by up to 100% Increase is dependent on engine power and the highest effect was found at idle thrust Fuel hydrogen mass content correlated best with nvpm emissions and seems to be able to account for naphthalenes in the range studied A simple emissions correction parameterization that uses engine thrust and change in hydrogen mass content was developed that explains the variability in the data Further research is needed on engines with different combustor technologies and pressure ratio Standard fuel hydrogen analysis method based on combustion (ASTM D 5291) lacks the precision to resolve small changes in H content; recommended are methods based on NMR spectroscopy (e.g. ASTM D 7171) as used in this work 11

13 Future work programme Funded by the Swiss Federal Office for Civil Aviation (FOCA) Builds on the experience, relationships, and infrastructure built in the previous FOCA-funded project from 2012 to 2015 Led by Jing Wang and me; three laboratories from Empa and SR Technics involved Total budget 5.8 million CHF, FOCA contribution 4.4 million CHF Project is expected to run for the next three years ( ) BAZL Bundesamt für Zivilluftfa hrt 12

14 EMPAIREX key topics nvpm standard Emissions variability Plume evolution Alternative fuels 13

15 EMPAIREX key topics and tasks Validate nvpm standard methodology nvpm standard Emissions variability SN-nvPM mass correlation Verify sampling system loss correction nvpm emissions from mixed flow engines Fuel doping with 8% aromatics drop-in fuel Engine deterioration effects on emissions Plume evolution TEM/NEXAFS analysis of the nvpm and total PM Metal PM in the exhaust Emissions sampling in the exhaust silencer Alternative fuels Genotoxic organic compounds nvpm / total PM optical properties Particle morphology / effective density, size 14

16 Time schedule One major measurement campaign each year + piggyback measurements 2016 campaign scheduled 22 Oct 14 Nov fuel doping with 8% aromatics drop in fuel 15

17 Prospective collaborations We welcome other teams nvpm standard related work (one of the four main objectives of this project) Accommodation of additional equipment (nvpm, total PM) and experiments that are in line with our objectives and can advance the state of scientific knowledge Needs careful planning and collaboration agreements 16

18 Acknowledgements SR Technics Swiss Federal Office of Civil Aviation (FOCA) SNECMA GE Aviation Instrument loans and calibrations Cardiff GTRC TSI Inc. Transport Canada National Research Council of Canada 17

19 Thank you for your attention 18

20 Questions? 19

21 Appendix Fuel and aromatic solvent properties Property Unit ASTM Method A-PRIDE 7 Fuel A-PRIDE 8 Fuel Solvesso 150ND Solvesso 150 Total Aromatics % v/v D > Sulfur total % m/m D <0.001 <0.001 Initial boiling point C D End point C D Density at 15 C Viscosity at -20 C Specific energy, net kg/m³ mm²/s D MJ/kg D Smoke point mm D <5 Naphthalenes % v/v D Hydrogen % m/m D 7171*mod H/C ratio (calculated)

22 Appendix: fuel total aromatics vs. fuel hydrogen content 26 Naphthalenes: 1.18 % v/v Total Aromatics [% v/v] Base Fuel: Naphthalenes: 0.78 % v/v 18 Solvesso 150ND Solvesso H [% m/m] Higher naphthalenes content lowers the hydrogen mass Fuel hydrogen mass content has been previously identified as the best correlation parameter for emitted carbon mass (Sampath 1986) and for smoke (Bowden 1984 and 1988) 12

23 Appendix: nvpm emission indices (A-Pride 7) Ei Mass (MSS) [g kg -1 ] Ei Mass (MSS) [g kg -1 ] Total Aromatics Level (%v/v) Solvesso 150ND Relative Thrust [%] Solvesso Relative Thrust [%] Ei Number [kg -1 ] Ei Number [kg -1 ] 5E14 4E14 3E14 2E14 1E14 Solvesso 150ND Relative Thrust [%] 5E14 4E14 3E14 2E14 1E14 Solvesso Relative Thrust [%] Solvesso injection at low thrust was not successful, due to inaccurate fuel flow measurement 10

24 Appendix: Particle size distribution change with aromatic content 2.0E6 65% T 3 ref. thrust 1.5E6 dn/dlogd p /cm 3 1.0E6 5.0E % d g = nm 21.1% d g = nm 22.4% d g = nm 23.6% d g = nm mobility diameter [nm]