Stanford University Research Program Shock Tube/Laser Absorption Studies of Chemical Kinetics Ronald K. Hanson Dept. of Mechanical Engineering, Stanford University Experimental Methods Butanol Kinetics Methyl Ester Kinetics Future Work CEFRC Second Annual Conference August 17 19, 2011 Some of the work presented here is unpublished. Please check with RKH/DFD before regarding the data as final or importing it into publications. We would also value feedback from team members regarding our data and how it might be modeled. rkhanson@stanford.edu dfd@stanford.edu 1
Please Note Some of the work presented here is unpublished. Please check with RKH/DFD before regarding the data as final or importing it into publications. We would also value feedback from team members regarding our data and how it might be modeled. rkhanson@stanford.edu dfd@stanford.edu 2
Experiment Types Goal: High quality databases to validate detailed mechanisms Ignition delay times provide global targets Shock tubes provide constant volume data over wide pressure range Species time histories provide strong constraints on mechanisms Laser absorption can provide species time histories for: OH, CO, CO 2, CH 2 O, H 2 O, CH 3, CH 4, C 2 H 4, fuel, Direct determination of elementary reaction rate constants For reactions where estimates/theory are not sufficient 3
Kinetics Shock Tube 1 Kinetics Shock Tube 2 Experimental Approach Aerosol Shock Tube High Pressure Shock Tube 4
Advances in Shock Tube Methodology Tailored driver gas/new driver geometry provide extended test time for access to low T Driver inserts provide highly uniform reflected shock conditions approaching constant U/V Aerosol shock tube provides access to low vapor pressure fuels Gasdynamic modeling of shock tube flows to account for facility effects and energy release 5
Access to Low Temperatures Longer driver length and tailored gas mixtures can provide longer test times (> 40 ms) 2x Driver Extension 7 Shock tubes now can overlap with RCMs 6
Improvement in Temperature Uniformity Problem: Ignition delay times are artificially shortened by non ideal facility effects!! dp/dt 0 Solution: Driver Inserts Results: Near ideal constant volume performance!! dp/dt 0 V RS P 5 T 5 7
Aerosol Shock Tube for Low Vapor Pressure Fuels Dump Tank Diagnostics: Pressure Droplet scattering Fuel time history OH* emission Aerosol Tank Driver Section Driven Section Evaporated Aerosol Ultrasonic Nebulizers Does not require heated shock tube Eliminates fuel cracking and partial distillation Provides access to low vapor pressure fuels: large methyl esters, bio diesel surrogates 8
Current Laser Capabilities for Species Detection for real time, in situ sensing Ultraviolet CH 3 216 nm NO 225 nm O 2 227 nm HO 2 230 nm CH 2 O 305 nm OH 306 nm NH 336 nm Visible CN CH NCO NO 2 NH 2 HCO 388 nm 431 nm 440 nm 472 nm 597 nm 614 nm Infrared CO 2.3 m H 2 O 2.5 m CO 2 2.7 m Fuel 3.4 m NO 5.2 m MeOH 9.2 m MF 9.2 m C 2 H 4 10.5 m Coherent MIRA Ti Sapphire Spectra Physics 380 Ring NovaWave Mid IR DFG
Kinetics Shock Tube 1 Kinetics Shock Tube 2 Butanol Kinetics 10
Overview of Butanol Studies Ignition delay times: 1.5 45 atm, 800 1600 K Butanol isomers, high and low pressure Species time histories: N Butanol pyrolysis: OH, H 2 O, CH 2 O, C 2 H 4, CH 4, CO N Butanol oxidation: OH, H 2 O, C 2 H 4 Direct determinations of elementary rxn. rate constants: Butanol+OH=products, all isomers Butene+OH= Products, all isomers 11
Survey of Ignition Delay Times: Butanol Isomers at Low Pressure 1538 K 1429 K 1333 K 1250 K Variation in ignition delay times - tert-butanol slowest - 1-butanol fastest 1000 Lines - MIT (2011) 1-butanol 2-butanol i-butanol t-butanol t-but 2-but i-but 1-but MIT (2011) mechanism t ign [us] - Fair agreement with t- & i-butanol data 4% O 2 /Argon - Poorer agreement with 1- & 2-butanol data 100 1.5 atm, =1.0 Data of sufficient quality to refine reaction mechanisms 0.60 0.65 0.70 0.75 0.80 0.85 1000/T 5 [1/K] What happens at high pressure? 12
Survey of Ignition Delay Times: 2 Butanol Variation with Pressure 1538 K 1333 K 1177 K 1053 K Very low scatter ( 5-10 %) consistent with uncertainty in T Lines - MIT (2011) 1.5 atm 3 atm 19atm 43atm Ignition delay times scale approximately as P -0.7 MIT (2011) model P-dependence consistent with 2-butanol data t ign [us] 1000 100 2-Butanol 4% O 2 /Argon ~1.0 Data of sufficient quality to refine reaction mechanisms 0.6 0.7 0.8 0.9 1.0 1000/T 5 [1/K] Next step: need for species time histories! 13
N Butanol Pyrolysis First shock tube/laser absorption speciation study of n butanol pyrolysis OH (306 nm) H 2 O (2.5 microns) CH 2 O (305 nm) CO (4.6 nm) C 2 H 4 (10.5 microns) CH 4 (3.4 microns) n Butanol 14
N Butanol Pyrolysis: Species Time Histories OH and H 2 O OH & H 2 O time histories reveal large variation in model performance Clear opportunity for model refinement 15
N Butanol Pyrolysis: Species Time Histories CH 2 O and CO Formaldehyde and CO uniformly underpredicted! Measured OH+H 2 O+CH 2 O+CO account for >90% of O atoms! Remaining O atoms likely in CH 3 CHO, CH 2 CO, 16
N Butanol Pyrolysis: Species Time Histories C 2 H 4 and CH 4 All models underpredict C 2 H 4 ; better agreement for CH 4 Measured C 2 H 4 +CH 4 +CO+CH 2 O account for >70% of C atoms! Remaining C atoms likely in C 3 H 6, C 2 H 6, C 2 H 2, 17
OH+Butanol Products First order removal of OH measured using laser absorption in 30 ppm TBHP/200ppm butanol/argon mixtures Rate Constant [cc/mole/s] 3E13 1E13 1E12 1111 K 1000 K 909 K 1E11 0.8 0.9 1.0 1.1 1.2 1000/T [1/K] 1-but iso-but 2-but tert-but Sarathy (2011) Strong dependence on isomer MIT (2011) Model: Trends not consistent with data Sarathy (2011) Model: Consistent with data Overall rate dependent on product channel chemistry 18
Kinetics Shock Tube 2 Methyl Ester Kinetics Aerosol Shock Tube 19
Overview of Methyl Ester Studies Ignition delay times: Aerosol Shock Tube Studies Methyl Decanoate Methyl Oleate Species time histories during pyrolysis: Methyl Formate: CO, OH, C 2 H 4, CH 2 O, CH 3, CH 4, Me OH, MF Methyl Acetate/Propanoate: CO, CH 3, C 2 H 4 Methyl Butanoate: CO, CO 2, C 2 H 4, OH Reflected shock conditions: 1.5 6 atm, 1000 1400 K 20
Methyl Decanoate Ignition Delay Times: Aerosol Shock Tube Studies E A = 29.2 kcal/mol Westbrook model: 3500 species 17000 reactions E A = 42.5 kcal/mol Westbrook model correctly predicts ignition delay times, activation energies, and [O 2 ] dependence Can we apply AST method to larger esters? 21
Methyl Oleate Ignition Delay Times: Aerosol Shock Tube Studies Data reveal weak dependence on equivalence ratio Westbrook model: underestimates ignition delay times by about 50% captures reveal weak dependence on equivalence ratio Next step: Need for species time histories! 22
Methyl Formate Pyrolysis: Time Histories and Rate Data Species time histories: CO, OH, C 2 H 4, CH 2 O, CH 3, CH 4, MeOH, MF Rate constant determination for all three major decomposition channels (Princeton/NUI 2010) 1: MF CO+MeOH using CO (4.6 m) using MeOH (9.23 m) 2: MF CH 4 +CO 2 using CH 4 (3.4 m) using CO 2 (2.7 m) 3: MF 2CH2O using CH 2 O (306 nm) 23
Methyl Formate Decomposition: CH 3 OH+CO channel CO Mole Fraction [%] 0.20 0.15 0.10 0.05 0.00 Measurement Dryer et al. 2010 0.1% MF/Ar 1.5 atm CO Time Histories 1488K 1376K 1285K 1202K 0 200 400 600 800 Time [ s] 1607K k 1 [cm -3 mol -1 s -1 ] 10 10 10 9 10 8 10 7 MF CH 3 OH+CO 1667K 1538 1429 1333 1250 1176 10 11 Current study Best fit Dryer et al. 2010 (1.6 atm) Curran et al. 2008 (1.6 atm) P = 1.48-1.72 atm 10 6 0.60 0.65 0.70 0.75 0.80 0.85 1000/T [K] Direct measurement of CH 3 OH+CO channel possible with CO laser Excellent agreement at with Dooley et al. (2010) particularly at lower T 24
Methyl Formate Decomposition: CH 4 +CO 2 channel CH 4 Time Histories 1667 K MF CH 4 +CO 2 1562 1470 1389 1316 1250 CH4 Mole Fraction [%] 0.4 0.3 0.2 0.1 3% MF/Ar 1.5 atm Current study Dryer et al. 2010 Updated k 1 1408K 0.0 0.0 0.5 1.0 1.5 2.0 2.5 Time [ms] 1289K k 2 [cm 3 mol -1 s -1 ] 10 9 10 8 10 7 10 6 Current study Best fit Dryer et al. 2010 (1.45 atm) 0.60 0.64 0.68 0.72 0.76 0.80 1000/T [K] P = 1.36-1.54 atm Direct measurement of CH 4 +CO 2 channel possible with CH 4 or CO 2 laser Excellent agreement at with Dooley et al. (2010) 25
Methyl Formate Decomposition: CH 2 O+CH 2 O channel CH2O Mole Fraction [%] 0.8 0.6 0.4 0.2 0.0 CH 2 O Time Histories Current study Dryer et al. 2010 Updated k 1, k 2 3% MF/Ar 1.4 atm 1462K 1347K 1257K k 3 [cm 3 mol -1 s -1 ] 1667 K 10 9 10 8 10 7 10 6 MF CH 2 O+CH 2 O 1538 1429 1333 1250 P = 1.37-1.56 atm Current study Best fit Dryer et al. 2010 (1.4 atm) 1176 0 200 400 600 800 1000 Time [ s] 0.60 0.65 0.70 0.75 0.80 0.85 1000/T [K] Direct measurement of CH 2 O+CH 2 O channel possible with CH 2 O laser Excellent agreement at with Dooley et al. (2010) 26
Oxygen Balance in Methyl Formate Pyrolysis Oxygen Atom Balance 1.0 0.8 0.6 0.4 0.2 CH 3 OCHO CH 4 (CO 2 ) T = 1420K, P = 1.5atm (0.2-3% MF/Ar) CO CH 3 OH CH 2 O @ t = 300 s % Oxygen balance: 5.5% in MF 34.8% in MeOH 44.9% in CO 5.8% in CO2 7.2% in CH2O 0.0 0 100 200 300 Time [ s] Total: 98.2% Laser data successfully tracks all major contributors to O atom balance Significant opportunity for mechanism validation 27
Future Work Butanol Kinetics Extend species time history database to all isomers Extend speciation database to oxidation Methyl Ester Kinetics Extend ignition delay time database to large bio diesel components Extend speciation studies to other methyl esters (small and large) Other Opportunities Collaborate with modelers Apply multi species methods to other fuels/surrogates 28