Experimental and simulative analysis of the NO2 formation in Diesel engines M. Rößler, C. Janzer, D. Notheis, U. Wagner, A. Velji, T. Koch, M. Olzmann (KIT) F. P. Zimmermann (Daimler AG) Institute of Internal Combustion Engines Prof. Dr. sc. techn. Thomas Koch Institute of Physical Chemistry, Molecular Physical Chemistry Prof. Dr. rer. nat. Matthias Olzmann Please insert a figure in the slide master KIT The Research University in the Helmholtz Association
Relevance of Nitrogen Dioxide Atmosphere Chemistry: Formation of ground ozone (O 3 ): Photolysis of NO 2 caused by solar irradiation Impacts of NO 2 Human Health and environment: Harmful for respiratory system. Offensive smell and yellow / brown color. NO 2 + hv NO + O O 2 +O O 3 Cause of acid rain: Reaction with H 2 O: 2 NO 2 + H 2 O HNO 2 + HNO 3 Exhaust Gas Aftertreatment Enhanced SCR performance: Fast SCR-Reaction : NO + NO 2 + 2NH 3 2N 2 + 3H 2 0 Promotion of passive particle filter regeneration. 2 7/5/2018
Formation Paths of Nitrogen Dioxide Atmosphere: Emitted NO reacts with ambient air to NO 2 corresponding to the thermodynamic equilibrium. Catalytic Components: Oxidation NO to NO 2 with excess air (λ > 1) at Diesel oxidation catalysts (DOC). Combustion Process: Engine-internal formation of NO 2 during combustion and expansion. 3 7/5/2018
Methodology #1 Design of Experiments #2 0D Chemical Kinetic #3 In cylinder fast gas sampling Identification of sources on the formation of in cylinder NO 2 4 7/5/2018
Agenda Introduction and Methodology Setup Results DoEs Chemical Kineetic Fast in Cylinder Gas Sampling Conclusion 5 7/5/2018
Engine and Test Bench Data Single cylinder research engine Based on EURO 6 medium-duty engine (Mercedes Benz OM 936). Common Rail Injection (up to 240 MPa) Cooled high pressure EGR Automated Test Bench (D2T Morphée) at IFKM: Conditioning units for fuel, coolant, oil and air (incl. humidity) High + low pressure indication Continuous fuel measurement and high pressure fuel indication Various gas analyzers Engine Specifications: Bore x Stroke 110 x 135 Displacement 1283 cc Compression Ratio 17,4:1 - Max. Charge Pressure 0,45 MPa Max. Torque 220 Nm Max. Engine Speed 2200 rpm 6 7/5/2018
Experimental Setup: Gas Analysis AVL AMA4000 with CLD4000hhd: 2-Channel-CLD with thermo-catalytic NO 2 -converter Measures NO and NO X NO 2 = NO X NO Concentrations of CO, CO 2, O 2, HC FTIR MKS MG2030: Measurement of various species, sampling rate 5 Hz Spectral Analysis of NO 2 : [1571,9 1633,7] cm -1 DEGAS Mid-IR-Analyzer (cooperation with Fraunhofer IPM): Fast 200 Hz Sampling Spectral Analysis of NO 2 with QCL: 1579,8 cm-1 (6,13 µm) ABB URAS AGR: Measuring of CO 2 and EGR rate AVL 415s filter-type smoke meter Soot: Filter Smoke Number 7 7/5/2018 NO, NO X CO, CO 2, HC, O 2 NO, NO 2, N 2 O CO, CO 2, NO 2 EGR / CO 2 FSN
Chemical Kinetic Mechanism n-heptane (CN 54 [1] ) as a model fuel n-heptane oxidation mechanism + submechanism for NO X formation n-heptane mechanism Mehl et al. [2] 633 species 5478 reactions Coupling and other reactions [4-7] + 4 species 666 reactions Detailed mechanism for n-heptane oxidation in air and NO X formation 668 species 6654 reactions NO X submechanism Faravelli et al. [3] + 31 species 510 reactions [1] J. Yanowitz, M. A. Ratcliff, R. L. McCormick, J. D. Taylor, M. J. Murphy, Technical Report. National Renewable Energy Laboratory, 2014. [2] M. Mehl, W.J. Pitz, C.K.Westbrook, H.J. Curran, Proc. Combust. Inst. 2011, 33, 193-200. [3] T. Faravelli, A. Frassoldati, E. Ranzi, Combust. Flame 2003, 132, 188 207. [4] P.A. Glaude, N. Marinov, Y. Koshiishi, N. Matsunaga, M. Hori, Energy Fuels 2008, 19, 1839 1849. [5] F. Contino, F. Foucher, P. Dagaut, T. Lucchini, G. D Errico, C. Mounaïm-Rousselle, Combust. Flame 2013, 160, 1476 1483. [6] G. Dayma, K.H. Ali, P. Dagaut, Proc. Combust. Inst. 2007, 31, 411-418. [7] National Institute of Standards and Technology (NIST), NIST Chemical Kinetics Database (http://kinetics.nist.gov/kinetics), 2014. 8 7/5/2018
Agenda Introduction and Methodology Setup Results DoEs Chemical Kinetic Simulation Fast in Cylinder Gas Sampling Conclusion 9 7/5/2018
Design of Experiments analysis Model Generation Identification of most important impacts on NO 2 and NO X Definition of boundary constraints and test procedure Data Acquisition at engine test bench Model building, optimization and validation Rail Pressure Start of Injection Rel. Air-/Fuel ratio EGR rate Rel. humidity Charge Temperature 10 15 7/5/2018
Specification DOE#1 DoE#1: Detail Parameter Variation on one operating point (A 25%) Operating Parameters Variation Unit Evaluation Value/Range EGR Rate % 0-45 Air mass g/cycle 0.9 1.8 T ChargeAir C 60 φ charge % 50 SoI CAD b TDC 9 Rail pressure MPa 135 Model Parameters and Quality Variation Unit Model Type Parameters RMSE R² NO2 g/kwh RBF (multiquatratic) 27 0.024 0.997 NO X g/kwh RBF (multiquatratic) 35 0.129 0.999 NO 2 /NO X % RBF (multiquatratic) 35 0.623 0.995 11 7/5/2018
Results DOE#1 NO 2 and NOx plotted in g/kwh x N = 1400 rpm IMEP_n = 0,65 MPa EGR = 18 % p_rail = 135 MPa Air mass = 1.4 g/cycle SOI = 7,9 CAD x x x Combined Analysis of NO 2 and NO X model: NO 2 is majorly influenced by the oxygen content, NO X is majorly influenced by EGR and the oxygen content. NO 2 /NO X can be increased: without change of NOX, NO2/NOX from 6% 11%, immense demand of charge pressure 12 7/5/2018
Specification DOE#2 DoE#2: Parameter Variation in the low speed low torque area Operating Parameters Variation Unit Evaluation Value/Range EGR Rate % 24 MFB 50% CAD a TDCF 5 Burn duration t90 CAD 35 Engine speed rpm 800-1500 Engine load bar 2-6 Model Parameters and Quality Variation Unit Model Type Parameters RMSE R² NO2 mg/cycle RBF (gaussian) 27 0.004 0.998 NO X mg/cycle RBF (multiquadratic) 33 0.021 0.999 NO 2 /NO X % RBF (multiquadratic) 33 0.942 0.996 13 7/5/2018
Results DOE#2 (1/2) NO 2 and NOx plotted in mg/cylce EGR Rate = 24 % MFB 50% = 5 CAD atdcf Burn x duration = 35 CAD NO X is majorly influenced by Engine Load (Temperature) and excess oxygen NO 2 needs oxygen and NO The influence of available oxygen is much more important The NO 2 /NO X amount runs contrary to the amount of NO X 14 7/5/2018
Results DOE#2 (2/2) NO 2 and NOx plotted in mg/cylce EGR Rate = 24 % MFB 50% = 5 CAD atdcf Burn duration = 35 CAD THE DoE allow the treatment of the engine speed with an explicit switch off of the burning parameters. NO 2, NO X and NO 2 /NO X shows no significantly influence on engine speed. 15 7/5/2018
Agenda Introduction and Methodology Setup Results DoEs Chemical Kinetic Simulation Fast in Cylinder Gas Sampling Conclusion 16 7/5/2018
Chemical 20 Kinetic simulations 100 15 80 10 60 5 40 0 0 2000 2500 3000 x(no 2 ) [ppm] ppm] x(no 2 ) [ppm] 20 0 0 500 1000 1500 2000 2500 3000 100 Chemical kinetic modeling 20 80 60 40 Engine measurement NO 2 /NO X [%] x(no) [ppm] x(no 2 ) NO 2 /NO X 20 0 0 500 1000 1500 2000 2500 3000 x(no) [ppm] 17 7/5/2018 20 15 10 5 15 10 5 NO 2 /NO X [%] NO 2 /NO X [%] Very good quantitative agreement between measurements and simulations. NO 2 increases nearly linearly with NO. x(no 2 ) NO 2 /NO X NO 2 /NOX decreases non-linearly with NO. Chemical Kinetic Simulation shows: NO 2 is directly formed during combustion A direct correlation between formed amounts of NO and NO 2 can be arrived Highest NO 2 ration with high temperatures and with a slight excess of air High NO 2 /NO X ratio are achieved at low temperature and high excess of air
Agenda Introduction and Methodology Setup Results DoEs Chemical Kinetic Simulation Fast in Cylinder Gas Sampling Conclusion 18 7/5/2018
Setup In Cylinder Fast Gas Sampling Fast Gas Sampling directly from the combustion chamber: Analysis of the NO 2 formation during the load cycle. Sampling conditions Sampling duration of 1.5 ms Time correction for the middle opening time GSV CAD model NO2 measurement: DEGAS Mid-IR-Analyzer Fast 200 Hz Sampling Spectral Analysis of NO 2 with QCL: 1579,8 cm-1 (6,13 µm) 19 7/5/2018
Interpretation In Cylinder Fast Gas Sampling I II III Evaluation of different engine conditions Before Combustion (I) homogenous mixture During combustion (II) heterogeneous mixture Only qualitative Statements possible Adequate distance after combustion (III) homogenous mixture 20 7/5/2018
Results In Cylinder Fast Gas Sampling NO 2 [ppm] 100 80 60 40 20 0 I II III Exhaust Emissions -40-20 0 20 40 60 Sampling Time [CAD] NO 2 concentration before / after combustion matches with measured values in the exhaust gas. NO 2 directly increases right after the start of combustion. Highest NO 2 concentrations measured during combustion. No further formation of NO 2 during later expansion and in the exhaust tract suspected. 21 7/5/2018
Agenda Introduction and Methodology Setup Results DoEs Chemical Kinetic Simulation Fast in Cylinder Gas Sampling Conclusion 22 7/5/2018
Conclusion NO 2 is produced always from NO (direct or indirect pathways). NO 2 emission always increases with increasing NO X emissions. The proportion of NO 2 usually decreases with increasing NO X. Highest NO 2 /NO X ratios at lowest NO X concentrations With an increase of the relative air-fuel ratio the NO 2 can be increased also at constant concentration of NO GSV measurements show early formation of NO 2 during combustion process: NO 2 formation takes place in combustion chamber and NO 2 is a combustion product 23 7/5/2018
Thanks for your kind attention Institute of Internal Combustion Engines Prof. Dr. sc. techn. Thomas Koch Institute of Physical Chemistry, Molecular Physical Chemistry Prof. Dr. rer. nat. Matthias Olzmann Please insert a figure in the slide master This research work was funded by the Bundesministerium für Wirtschaft und Energie (BMWi) and supervised by the Forschungsvereinigung Verbrennungskraftmaschinen e.v. (FVV Frankfurt, Germany). The authors express their thanks to BMWi, FVV, AiF and industrial partners involved. KIT The Research University in the Helmholtz Association