European Combustion Meeting 2015, Budapest. F. Battin-Leclerc

Size: px

Start display at page:

Download "European Combustion Meeting 2015, Budapest. F. Battin-Leclerc"

Buck Lyons
5 years ago
Views:

1 European Combustion Meeting 2015, Budapest F. Battin-Leclerc

A fascinating subject of scientific interest for many years certainly because it describes intriguing experimental features Autoignition

Vanhove, RCM, Workshop, 2012 In order to heat the gas whose ignition temperature was to be determined, it was necessary to enclose it in

2 A fascinating subject of scientific interest for many years certainly because it describes intriguing experimental features Autoignition First rapid compression machines built in 1906 (G. Falk, JACS, ) G. Vanhove, RCM, Workshop, 2012 In order to heat the gas whose ignition temperature was to be determined, it was necessary to enclose it in a small vessel supplied with a device for allowing the gas to be compressed instantaneously To cause adiabatic compression, a weight of 25 kg was dropped on the piston from heights of48 to 86 cm. 2

A fascinating subject of scientific interest for more than 100 years which describes intriguing experimental features Cool flames First observed in 1817 by Sir Humphry

3 A fascinating subject of scientific interest for more than 100 years which describes intriguing experimental features Cool flames First observed in 1817 by Sir Humphry Davy J. Griffiths, New Scientist, 2004 In a flow reactor C 2 H 6 In a static reactor J.A. Gray, J. Chem. Soc., 1953 J.A. Knox and G.W. Norrish, Trans. Farad. Soc.,

4 A fascinating subject of scientific interest for about100 years which describes intriguing experimental features Negative temperature coefficient (NTC) First mentioned in 1929 (R.N. Pease, JACS) Measurements in static reactors C 3 H 8, f = 5 C 2 H 6, f = 3.5 D.M. Newitt and L.S. Thornes, J. Chem. Soc., 1937 J.A. Knox and G.W. Norrish, Trans. Farad. Soc.,

5 A subject with important practical applications Knocking in SI engines, anti-knock strategy (octane number) Low-temperature oxidation HCCI and LTC diesel engines Safety of hydrocarbon oxidation process Post-oxidation oxidation reactions of exhaust gases 5

Letters, 2013 CFD predicted spatial temperature distribution in a diesel engine fed by a n-heptane/

6 A subject with important practical applications: How LT kinetics can influence engine performances S. Som, M.J. Davis et al., J. Phys. Chem. Letters, 2013 CFD predicted spatial temperature distribution in a diesel engine fed by a n-heptane/ methyl butanoate mix. (869 reactions) Tunneling not considered in rate constant calculation Crank angle = 2 Tunneling considered in rate constant calculation

7 Autoignition of fuel representative of gasoline available since the end of the 90s Autoignition of n-heptane in a shock tube (Aachen) and a rapid compression machine (Lille) Symbols correspond to experiments and lines to simulations (f = 1, P from 3 to 42 bar) Livermore Milano Nancy NTC E. Ranzi, P. Dagaut et al. F. Buda, P.A. Glaude, V. Warth, H.J. Curran, C.K. Westbrook et al. Combust. Flame 1995 F. Battin-Leclerc et al. Combust. Flame 1998 Combust. Flame species and 2450 reactions 100 species and 2000 reactions 360 species and 1820 reactions Automatic generation Models lead to similar agreement for reproducing the available experimental results, but with very different LT kinetics

; Green Energy and Technology; Springer London, 2013 rléans i F j i Q P, T, V o F j o Q Perfect material mixing if based on the design of D.

8 Many detailed kinetic models of LT Model validation for product formation often using jet-stirred reactor (JSR) data. Herbinet and G. Dayma., chapter 8, in Cleaner Combustion; Battin-Leclerc, F.; Simmie, J. M.; Blurock, E., Eds.; Green Energy and Technology; Springer London, 2013 rléans i F j i Q P, T, V o F j o Q Perfect material mixing if based on the design of D. Matras and J. Villermaux Chem Eng Sci 1973 Nancy Reactor when locating heating resistance wires in Nancy Thermal homogeneity if preceeding by a preheating zone and used with high dilution 8

9 Many detailed kinetic models of LT Models can predict JSR major product formation xidation of 2-methylheptane Symbols correspond to experiments and lines to simulations (f = 1, P = 10 bar, t = 0.7 s, 0.1 % initial fuel, GC measurements) NTC S.M. Sarathy, C.K.Westbrook, P. Dagaut et al. Combust. Flame

10 Many detailed kinetic models of LT Models can predict JSR major product formation xidation of n-heptane Symbols correspond to experiments and lines to simulations (f = 1, P = 1 bar, t = 2 s, 0.5 % initial fuel, GC (red) and PI-MS (blue) measurements) NTC. Herbinet, Z. Serinyel, V. Warth, F. Battin-Leclerc, F. Qi et al. Combust. Flame C c = 3 3 C H H C 2 C H 1 H H H H H H

Relative Intensity Many detailed kinetic models of LT Models cannot predict all JSR minor product formation xidation of n-heptane Symbols correspond to experiments and lines to simulations (f = 1, P

11 Relative Intensity Many detailed kinetic models of LT Models cannot predict all JSR minor product formation xidation of n-heptane Symbols correspond to experiments and lines to simulations (f = 1, P = 1 bar, t = 2 s, 0.5 % initial fuel, GC measurements) Electron impact mass spectrum of 2,4-diethyl-oxetane m/z. Herbinet, F. Battin-Leclerc et al., Fuel, Herbinet, Z. Serinyel, V. Warth, F. Battin-Leclerc, F. Qi et al. Combust. Flame 2012 More deviations in non-stoichiometric mixtures Always need of kinetic parameter adjustments 11

Chemical knowledge used in alkane LT RH + models 2 or H initiation steps H

branching steps QH R + H 2 R +H 2 RH H + cyclic ethers, aldehydes or ketones QH F.

12 Chemical knowledge used in alkane LT RH + models 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H 2 + alkene H 2 R RH + 2 degenerate branching steps QH R + H 2 R +H 2 RH H + cyclic ethers, aldehydes or ketones QH F. Battin-Leclerc, Prog. Energ. Combust Sci J. Zádor, C.A. Taatjes, R.X. Fernandes, Prog. Energ. Combust Sci degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 12

13 Chemical knowledge used in alkane LT RH + models 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H 2 + alkene H 2 R RH + 2 degenerate branching steps QH R + H 2 R +H 2 F. Battin-Leclerc, Prog. Energ. Combust Sci J. Zádor, C.A. Taatjes, R.X. Fernandes, Prog. Energ. Combust Sci RH H + cyclic ethers, aldehydes or ketones degenerate branching steps R + H QH U(H) 2 ketohydroperoxides + H When temperature increases, reactivity decreases and the NTC zone appears Product including 2 carbonyl groups Acid + product including a carbonyl group 13

14 rigin of kinetics used in alkane LT models R +H 2 2 RH RH + 2 or H initiation steps H abstractions R R + alkene 2 2 H 2 + alkene H 2 R RH + 2 degenerate branching steps QH R + H 2 Many LT reactions are equilibrated, so accurate thermochemical data determinations are also needed R +H 2 RH H + cyclic ethers, aldehydes or ketones QH F. Battin-Leclerc, Prog. Energ. Combust Sci J. Zádor, C.A. Taatjes, R.X. Fernandes, Prog. Energ. Combust Sci degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 14

15 Hydrogen atom abstractions from fuel RH + 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 QH 2 degenerate branching steps R + H R +H 2 RH H + cyclic ethers, aldehydes or ketones QH degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 15

Many measurements of the rate constant of H + X = XH + H 2 0 (3159 records in NIST kinetics data base) Laser flash photolysis Reactant + radical precursor ( 3, HN 3, H 2 2 )

16 Many measurements of the rate constant of H + X = XH + H 2 0 (3159 records in NIST kinetics data base) Laser flash photolysis Reactant + radical precursor ( 3, HN 3, H 2 2 ) + carrier gas H radicals have a determinant role in atmospheric chemistry Detector ften by Laser Induced Fluorescence To pump but mostly below 400 K and for C 1 -C 3 alkanes 16

17 C 2 -C 7 alkane high temperature study using a shock tube D e t e c t o r +car rier Reactant + Reflected shock radical precursor (t-butyl hydroperoxide ) wave + dilutant By courtessy of R.S.tranter and J.V. Michael R. Sivaramakrishnan and J.V. Michael, JPCA 2009 H absorption profiles for k H + n-heptane Experimental measurements and rate constant fitting from a 49 step mechanism. The dashed lines are fits with changes in k H + n-heptane by +20%. 17

18 High temperature study using a shock tube for 9 linear and branched C 2 -C 7 alkanes R. Sivaramakrishnan and J.V. Michael, JPCA 2009 Sivaramakrishnan and Michael Koffend and Cohen, IJCK, 1996 Shock tube, 1186 K k H + Alkanes ( K) Experimentally studied, obtained from derived group additivity correlations K experimental work Arrhenius plot of k H + n-heptane More studies on large alkanes are needed 18

19 R 2 and H 2 radicals are also present in large amounts at low-temperatures Influence on CFD predictions in a diesel engine S. Som, M.J. Davis et al., J. Phys. Chem. Letters, 2013 Kinetic parameters only based on theoretical studies or complex mechanism fitting 19

20 Reactions involving alkylperoxy radicals RH + 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 degenerate branching steps QH 2 R + H R +H 2 RH H + cyclic ethers, aldehydes or ketones QH degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 20

21 Reactions involving R* radicals R Schematic potential energy surfaces for the reactions of n-propyl radicals with oxygen molecules J.D. DeSain, S.J. Klippenstein, J.A. Miller, C.A. Taatjes, JPCA, QH R* Cyclic ether + H R Alkene + H 2 Rate constants need to be parametrized as a function of temperature and pressure using master equation This should also be done for C 3+ alkanes 21

22 Reactions involving H 2 radicals (f = 1, P = 1 bar, t = 2 s, 12 % initial fuel) Flow rate analysis for JSR propane oxidation at 630 K Using an EXGAS model with the Desain et al. s kinetic values 40% H 3 C CH 3 51% H 2 C CH H C 3 3 CH CH H 2 C CH % + 2 CH 3 1% 8% H + 6% 21% M. Cord, R. Fournet, F. Battin-Leclerc, F. Qi et al., JPCA, % H 3 C CH 3 H H 1/3 of propane is consumed to produce propene and H 2 radicals Importance of the H 2 H 2 2 system 22

H 2 H 2 2 system H 2 2 and H 2 radicals probed during n-butane JSR oxidation H 2 2 and H 2 by Cavity Ring Down Spectroscopy (CRDS) H 2 2 C. Bahrini,. Herbinet, C. Fittschen, F. Battin-Leclerc, et al.

23 H 2 H 2 2 system H 2 2 and H 2 radicals probed during n-butane JSR oxidation H 2 2 and H 2 by Cavity Ring Down Spectroscopy (CRDS) H 2 2 C. Bahrini,. Herbinet, C. Fittschen, F. Battin-Leclerc, et al., JACS, 2012 H 2 2 H 2 2 and H 2 radicals M. Djehiche, G. Dayma, P. Dagaut et al., JACS, /cm -1 H and H 2 by Fluorescence Assay Gas Expansion (FAGE) M. Bloquet, C. Schoemaecker,. Herbinet, F. Battin-Leclerc, C. Fittschen et al., PNAS, 2013 Loss of radicals when probed from JSR to an external optical cell 23

24 t(butane, H 2, H 2 2 )/ s H 2 H 2 2 system Probing H 2 2 by CRDS in a complex combustion mixture JSR, P = 1 bar, F = 1, t = 6 s, 2.3% n-butane, 650 K H HCH C. Bahrini,. Herbinet, C. Fittschen, F. Battin-Leclerc, et al., JACS, Reactive mixture /cm H t(hch)/ s 24

$Mole fraction H 2 H 2 2 system Quantifying H 2 2 by CRDS during JSR oxidation of n-heptane Symbols correspond to GC experiments and lines to simulations (f = 1, P = 1 bar, t = 2 s, 0.$

25 Mole fraction H 2 H 2 2 system Quantifying H 2 2 by CRDS during JSR oxidation of n-heptane Symbols correspond to GC experiments and lines to simulations (f = 1, P = 1 bar, t = 2 s, 0.5 % initial fuel, A. Rodriguez,. Herbinet, C. Fittschen, F. Battin-Leclerc, unpublished results, x f=0.25 f=1 f= Temperature (K) 1400 CRDS can probe H 2 2 even for current fuel surrogates 25

26 log(k (s -1 )) Isomerizations of R radicals Direct experimental determination Laser flash photolysis of neo-c 5 H 11 I in 2 /He with H LIF detection, K 6 K.J. Hughes, T. Turányi, M.J. Pilling, et al. Proc. Combust. Inst Theoretical calculations H H H CH 3 k CH 2 H 3 C H 3 C CH H 3 CH 3 CH 2 H 3 H. + C H H 3 C CH3 4 2 CH 3 k = exp(14800/t) s M. Cord, R. Fournet, A. Tomlin, F. Battin-Leclerc et al., JPC A 2012 S. Sharma, W.H. Green et al., JPCA 2011 A. Miyoshi, JPCA 2011 S.M. Villano, H.H. Cartensen, A.M. Dean et al., JPCA / (T (K)) Deviations betwen values calculated by the different groups up to a factor of 4 2.0x10-3 Need to detect ephemeral QH radicals A first success from 1,3-cycloheptadiene J.D. Savee, C. Taatjes et al., Science,

27 Reactions of R radicals with other radicals t Flow rate analysis for JSR propane oxidation at 630 K Using an EXGAS model with the theoretical calculated rate constants for isomerizations (f = 1, P = 1 bar, t = 2 s, 12 % initial fuel) H 12% CH 3 M. Cord, R. Fournet, F. Battin-Leclerc, F. Qi et al., JPC A, 2012 H 3 C CH 40% 3 51% H 2 C CH H C 3 3 CH CH H 2 C CH 3 2 1% 8% 39% + 43% + 2 CH 3 20% QH 6% 21% + 2 H 3 C CH 3 0.9% +H 2 +H Q'H H 3 C CH 3 1/3 of propane is consumed by reactions of peroxy and H 2 radicals Very few rate constants available above room temperature and for C 3+ compounds 20% H T.J. Wallington, P. Dagaut, M.J. Kutylo, Chem. Rev J.J. rlando and G.S. Tyndall, Chem. Soc. Rev

28 Formation of cyclic ethers RH + 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 QH 2 degenerate branching steps R + H R +H 2 RH H + cyclic ethers, aldehydes or ketones QH degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 28

29 Formation of cyclic ethers RH + 2 or H initiation steps H abstractions R +H 2 2 R +H 2 RH RH R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 degenerate branching steps QH R + H 2 H + cyclic ethers, aldehydes or ketones degenerate branching steps R + H A single indirect experimental rate constant determination (neo-pentane) QH U(H) 2 ketohydroperoxides + H R.R. Baldwin, R.W. Walker et al., J. Chem. Soc. Faraday Trans Several theoretical studies, such as for R 2 isomerizations Product including 2 carbonyl groups Acid + product including a carbonyl group 29

Detection of cyclic ether in photolytically initiated

formation of Cl atoms which react with a RH/ 2 /He

30 Detection of cyclic ether in photolytically initiated alkane oxidation Mass spectrometry combined with tunable synchrotron vacuum ultraviolet photoionization C.A. Taatjes, N. Hansen, D. sborn, T. Cool et al. PCCP, 2008 Time of flight MS Laser flash photolysis formation of Cl atoms which react with a RH/ 2 /He mixture slowly flowing in a quartz tube of 1.05 cm diameter Berkeley Advanced light source By courtessy of C.A. Taatjes

Detection of cyclic ether in photolytically initiated alkane oxidation Mass spectrometry combined with tunable synchrotron vacuum ultraviolet photoionization to study n-butane oxidation A.J.

31 Detection of cyclic ether in photolytically initiated alkane oxidation Mass spectrometry combined with tunable synchrotron vacuum ultraviolet photoionization to study n-butane oxidation A.J. Eskola, D. sborn, C.A. Taatjes et al., JPCA 2008 Time behaviors and photoionization spectra of cyclic ethers at 575 K, 4 Torr Time (ms) Photon energy (ev) A good source of data to confirm the theoretically calculated rate constants

Modeling the formation of cyclic ethers during JSR alkane oxidation Inclusion of theoretically calculated rate constants in automatically generated models E, kcal 20 10 0 32.4 (31.9) [33.4] 11.9 (15.

32 Modeling the formation of cyclic ethers during JSR alkane oxidation Inclusion of theoretically calculated rate constants in automatically generated models E, kcal (31.9) [33.4] 11.9 (15.3) [14.3] 21.0 (20.6) [22.3] 17.1 (19.0) [30.3] 23.4 (22.9) [23.9] 12.8 (14.6) [14.0] M. Cord, R. Fournet, A. Tomlin, F. Battin-Leclerc et al., JPCA 2012 Thermodynamic data also revised from theoretical calculations Quantum calculations at the CBS-QB3 level of theory (0 K) using Gaussian (Green et al., JPC A, 2003, 2010) [Taatjes et al. Faraday Disc. (2001)] High pressure limit rate constants calculated using transition state theory for C 4 alkanes + New C 4+ kinetic correlations for isomerizations and cyclic ether formations Rate constants of the R+ 2 system taken from DeSain et al

33 Modeling the formation of cyclic ethers during JSR alkane oxidation xidation of n-hexane Symbols correspond to experiments and lines to simulations (f = 1, P = 1 bar, t = 2 s, 2 % initial fuel) Z. Serinyel,. Herbinet, V. Warth, F. Battin-Leclerc et al. 8 th meditereanean meeting 2013 The prediction of cyclic ethers is well improved 33

34 Reactions producing and consuming ketohydroperoxides RH + 2 or H initiation steps H abstractions R +H 2 2 RH R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 QH 2 degenerate branching steps R + H R +H 2 RH H + cyclic ethers, aldehydes or ketones QH degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 34

35 Reactions producing and consuming ketohydroperoxides R +H 2 2 RH RH + 2 or H initiation steps H abstractions The most obscure part of the mechanism often based on the assumption that QH has a reactivity toward 2 close to R radicals, but it could be 10 times higher R R + alkene 2 2 H H 2 + alkene R 2 RH + 2 degenerate branching steps QH R + H 2 J.D. Savee, C. Taatjes et al. Science 2015 R +H 2 RH H + cyclic ethers, aldehydes or ketones QH degenerate branching steps R + H U(H) 2 ketohydroperoxides + H Product including 2 carbonyl groups Acid + product including a carbonyl group 35

molecular-beam sampling system Isotherm quartz jet-stirred reactor at atmospheric

36 Ketohydroperoxide detection only recently Coupling of a mass spectrometer combined with tunable synchrotron vacuum ultraviolet photoionization (PI-MS) to a JSR through a molecular-beam sampling system Isotherm quartz jet-stirred reactor at atmospheric pressure (JSR) F. Battin-Leclerc,. Herbinet, P.A. Glaude, F. Qi et al. Angewandte Chemie Int. Ed

37 Signal Ketohydroperoxide detection only recently Coupling of PI-MS to a JSR through a molecular-beam sampling system Study of the low-temperature oxidation of alkanes: n-butane F. Battin-Leclerc,. Herbinet, P.A. Glaude, F. Qi et al. Angewandte Chemie Int. Ed btained mass spectrum P = 1 bar, F = 1, t = 6 s, 4% n-butane, 590 K, 10 ev 72: C 4 oxygenated compounds : Butene 58: Butane H C H 3 C 4 H 9 H C H CH 3 H C 2 H 5 H 90 H Mass

38 Ketohydroperoxide detection only recently Coupling of PI-MS to a JSR through a molecular-beam sampling system Study of the low-temperature oxidation of alkanes: n-butane F. Battin-Leclerc,. Herbinet, P.A. Glaude, F. Qi et al. Angewandte Chemie Int. Ed Ionization energy (IE) measurements P = 1 bar, F = 1, t = 6 s, 4% n-butane, 590 K, 10 ev IE of CH 3 H = 9.83 ev IE of C 2 H 5 H = 9.61 ev IE of C 4 H 9 H = ev IE of ketohydroperoxides = ev Zero-point energy corrected adiabatic IEs have been calculated from the CBS-QB3 method using Gaussian03. 38

Signal (Arbitrary unit) Ketohydroperoxide detection only recently Comparison between ketohydroperoxide photoionization spectrum obtained during n-butane oxidation in a JSR and

$0 800 Torr, stoichiometric mixture, residence time = 6 s, initial n-butane mole fraction = 0.04 9.5 10.0 10.5 Photon energy (ev) F. Battin-Leclerc,. Herbinet, P.A. Glaude, F.$

39 Signal (Arbitrary unit) Ketohydroperoxide detection only recently Comparison between ketohydroperoxide photoionization spectrum obtained during n-butane oxidation in a JSR and in a photolysis tube Jet-stirred reactor (this study) Pulsed photolysis experiments (RH + Cl. + 2 ) Torr, stoichiometric mixture, residence time = 6 s, initial n-butane mole fraction = Photon energy (ev) F. Battin-Leclerc,. Herbinet, P.A. Glaude, F. Qi et al. Angewandte Chemie Int. Ed A.J. Eskola, C.A. Taatjes et al., Proc; Combust. Inst Both spectra are similar, showing that the same isomer is formed in thermal or photolysis experiments whatever the pressure 39

40 Ketohydroperoxide detection only recently Formation of hydroperoxides according to alkanes studied in JSR Fuel Propane M. Cord, R. Fournet, F. Battin- Leclerc, F. Qi et al. JPCA 2012 n-butane F. Battin-Leclerc,. Herbinet, P.A. Glaude, F. Qi et al. Angewandte Chemie Int. Ed n-pentane. Herbinet, F. Battin-Leclerc, F. Qi et al. unpublished results,2015 Hexane isomers Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPCA 2014 n-heptane. Herbinet, Z. Serinyel, V. Warth, F. Battin-Leclerc, F. Qi et al. Combust. Flame 2012 Alkyl hydroperoxides Yes yes yes No No Keto hydroperoxides No yes Yes Yes Yes 40

41 Signal (arbitrary unit) Ketohydroperoxide detection only recently Formation of ketohydroperoxides from linear and branched hexanes in JSR Ketohydroperoxide signal evolution with temperature P = 1 bar, F = 1, t = 2 s, m/z = 132, ionization energy = 10.5 ev Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPCA n-hexane 2-methyl-pentane 3-methyl-pentane 2,2-dimethyl-butanex Temperature (K) The more branched the isomer, the higher the temperature of appearance of ketohydroperoxides 41

42 Mole fraction Mole fraction Hydroperoxide quantification. herbinet, F. Battin-leclerc et al. PCCP 2011 Quantification of C 4 hydroperoxides Symbols correspond to experiments and lines to simulations P = 1 bar, F = 1, t = 6 s, 4% n-butane C 4 H 9 H Ketohydroperoxides 2.0x x Simulated mole fractions / Temperature (K) Temperature (K) 800 Calibrated from butene GC measurements with s i taken as equal to that of tetrahydrofuran 42

$Mole Fraction Mole Fraction Ketohydroperoxide consumption not only yields alkoxy and H radicals Formation of carboxylic acids Acid mole fraction evolution with temperature during hexane isomer JSR$

43 Mole Fraction Mole Fraction Ketohydroperoxide consumption not only yields alkoxy and H radicals Formation of carboxylic acids Acid mole fraction evolution with temperature during hexane isomer JSR oxidation P = 1 bar, F = 1, t = 2 s, PI-MS Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPC A x10-3 Acetic Acid 1.6x10-3 Propanoic Acid n-hexane 3-methyl-pentane 2-methyl-pentane Temperature (K) Temperature (K)

Ketohydroperoxide consumption not only yields alkoxy and H radicals H Formation of acetic acid from ketohydroperoxides Modeling acetic acid formation from Nancy JSR alkane oxidation Symbols

44 Ketohydroperoxide consumption not only yields alkoxy and H radicals H Formation of acetic acid from ketohydroperoxides Modeling acetic acid formation from Nancy JSR alkane oxidation Symbols correspond to experiments and lines to simulations P = 1 bar, F = 1, t = 6 or 2 s, GC and PI-MS results Propane n-butane n-heptane H A. Jalan, D.G. Truhlar, W.H. Green et al. JACS 2013 C H 3 H + H 2 C E. Ranzi, T. Faravelli et al. M. Pelucchi, E. Ranzi et al. Combust. Flame 2015 Energy&Fuel

45 Signal (other isomers) Ketohydroperoxide consumption not only yields alkoxy and H radicals Formation of diones Evolution with temperature of the signal at m/z = 114 during hexane isomer JSR oxidation P = 1 bar, F = 1, t = 2 s, PI-MS at 10.5 ev Signal (n-hexane) Temperature (K) Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPC A

46 Signal Ketohydroperoxide consumption not only yields alkoxy and H radicals Formation of diones Evolution with temperature of the signal at m/z = 114 during hexane isomer JSR oxidation P = 1 bar, F = 1, t = 2 s, PI-MS at 10.5 ev Ionization energy measurement for 2-methylpentane Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPC A Name of most probable diones 4-methyl-3-oxopentanal Structure Calculated ionization energy (ev) (CBS-QB3 method) PIE (ev) methyl-3-oxopentanal

47 Signal (other isomers) Signal (arbitrary unit) Ketohydroperoxide consumption not only yields alkoxy and H radicals Formation of diones Evolution with temperature of the signal at m/z = 114 During hexane isomer JSR oxidation P = 1 bar, F = 1, t = 2 s, PI-MS at 10.5 ev Z. Wang,. Herbinet, F. Battin-Leclerc, F. Qi et al. JPC A Signal of sum of C 6 H 12 species Temperature (K) Signal (n-hexane) m/z 100 (10.5 ev) Temperature (K) Diones: an important type of products not yet satisfactorily modeled

Still many unknowns in detailed kinetic LT models at

unavoidable prerequisite in the development of

Singh, DE Vehicle Technologies ffice 2013 LT models

possibilities of coupling with CFD using high

48 Still many unknowns in detailed kinetic LT models at almost every step of the mechanism LT models are an unavoidable prerequisite in the development of cleaner engines B. Cuenot J3P 2014 G. Singh, DE Vehicle Technologies ffice 2013 LT models are of even more interest with expected increasing possibilities of coupling with CFD using high performance computing But need reliable models in extended ranges of T, P, f 48

Fuels do not only contain alkanes FUELS Saturated aldehydes via Waddington mechanism + H Addition of H 2 or H H H-abstractions Addition of H and to 2 Hydroxy cyclic ethers or ketohydroperoxides

49 Fuels do not only contain alkanes FUELS Saturated aldehydes via Waddington mechanism + H Addition of H 2 or H H H-abstractions Addition of H and to 2 Hydroxy cyclic ethers or ketohydroperoxides Addition of H or or Alkanes Alkenes Aromatics Esters Akohols Ethers + 2 or Addition to 2 1,3-hexadiene leading to + unsaturated H 2 cyclic ethers and ketohydroperoxides Isomerisation + 2 Combination with H 2 giving hexenyl hydroperoxide a source of unsaturated aldehydes Addition to 2 leading to saturated cyclic ethers and ketohydroperoxides 1,3-hexadiene + H 2 F. Battin-Leclerc,.Herbinet, F. Qi, et al. JPCA 2014 The chemistry of other fuel components (e.g. alkenes) can be still more complex This can be worse for biofuels (see posters) 49

Baptiste Sirjean Roda Bounaceur Pierre A.

Warth Also: Christa Fittschen, Elizeo Ranzi, Katharina Kohse-Höinghaus,

Bénédicte Cuenot for helpful discussions and all the co-authors of the

50 Baptiste Sirjean Roda Bounaceur Pierre A. Glaude Colleagues of KinCom team livier Herbinet René Fournet Valérie Warth Also: Christa Fittschen, Elizeo Ranzi, Katharina Kohse-Höinghaus, Charlie Westbrook, Rob Tranter, Guillaume Vanhove, Craig Taatjes, Bénédicte Cuenot for helpful discussions and all the co-authors of the mentioned Nancy papers Results from Nancy were supported by European Research Council: advanced researcher Grant Clean-ICE 50

What do we Really Know about the Low-Temperature Oxidation of Alkanes?

What do we Really Know about the Low-Temperature xidation of Alkanes? F. Battin-Leclerc * Laboratoire Réactions et Génie des Procédés, CNRS, Université de Lorraine, 1 rue Grandville, 54 Nancy, France Abstract