Experimental study and detailed modeling of toluene degradation in a low pressure stoichiometric premixed ch4/o2/n2 flame

Experimental study and detailed modeling of toluene degradation
in a low pressure stoichiometric premixed CH4/O2/N2 Flame
L. Dupont1, A. El Bakali*1, J.F. Pauwels1, A. Rida2, P. Meunier2
1 UMR CNRS 8522 PC2A "Physicochimie des Processus de Combustion et de l’Atmosphère" Université des Sciences et Technologies de Lille, 59655 Villeneuve d’Ascq Cedex, France 2 GAZ DE FRANCE, Direction de la Recherche, B.P. 33, 93211 Saint Denis La Plaine Cedex, France Abstract
Temperature and mole fraction profiles have obtained in laminar stoichiometric premixed CH4/O2/N2 and CH4/1.5%C7H8/O2/N2 flames at low pressure (0.052 atm) by using thermocouple, MB/MS and GC/MS experimentaltechniques. The construction and the evaluation of a detailed reaction mechanism for toluene oxidation have beencarried out by using these experimental data. The proposed model correctly reproduces the main experimentalobservations, in particular the effect of toluene addition on the structure of the methane flame.
Introduction
Experiment
Understanding of detailed mechanisms of aromatic Temperature and species mole fraction profiles oxidation is of great interest for theoretical and practical have been measured in laminar stoichiometric premixed points of view. Aromatics are well known to be harmful CH4/O2/N2 and CH4/1.5%C7H8/O2/N2 flames. Flame for health and environment due to their toxicity and conditions were respectively as follows: pressure, 0.052 because their oxidation can form other toxic species.
atm.; cold gas velocity:1.8 cm.s-1; 11.1% and 6%CH4; Aromatics constitute a significant portion of most 22.2% and 23%O2; 66.7% and 69.5%N2. Details on the practical fuels like kerosene and diesel fuels. To experimental setup used in this work have been represent the aromatics in the models of these presented previously [1, 2] and only the main features commercial fuels, more information in a more complex are briefly reviewed here. The flames were stabilized aromatic compounds than benzene, is needed.
above a water–cooled porous plug flat flame burner.
This work is part of a research on the thermal Chemical species were sampled by a deactivated 60° degradation of aromatic VOCs in low-pressure laminar quartz cone with an orifice diameter of about 100µm premixed methane/air flames: the influence of benzene, and form a molecular beam passing through three toluene and para-xylene on the structure of a reference differentially pumped stages delimited by the sampling methane flame has been studied experimentally at cone, the skimmer and the collimator respectively. The molecular beam is ionized by electron impact and A detailed kinetic mechanism of oxidation of analyzed by a quadrupole mass spectrometer. After aromatic VOCs has been developed in flame conditions mass discrimination, the signal is amplified by a second based on different existing sub-mechanisms or by electron multiplier and an electrometer. The output is developing a new specific scheme in the case of para- then fed to a phase sensitive amplifier for background xylene. The complete detailed kinetic mechanism, signal substraction. When it is possible, the ion source including 166 species and 1130 reactions, reproduces parameters are set to avoid strong fragmentation effects correctly our experimental observations. Recently, we perturbing the flame composition profiles and/or to reported experimental and modeling study of benzene discriminate species of same m/e. The mass depletion in methane flame [1]. In the present paper, we spectrometer calibration is performed using the usual report experimental results obtained in methane / cold gas procedure for most stable species, the oxygen / nitrogen and methane /1.5%toluene / oxygen / conservation of the total number of atoms of H2O and nitrogen flames. The detailed chemical kinetic reaction the pseudo-equilibrium method in the burnt gases for H, mechanism developed for toluene oxidation to simulate O and OH. The stable species were also analyzed by gas these experimental data is also discussed.
chromatography (GC/FID/TCD) and by gaschromatography / mass spectrometry (GC/MS). Theexperimental error on the mole fraction of stable specieswas estimated to be about ±10 %. The mole fractions ofall stable species presented in this paper were obtained *Corresponding author: [email protected] Web site: http://www.univ-lille1.fr/umr8522Proceedings of the European Combustion Meeting 2003 by gas chromatography (GC-FID/TCD and GC/MS) methane flame. Modeling showed that methane except H2O and phenol which were measured by using depletion is similar in methane / 1.5%benzene / oxygen / nitrogen and methane / 1.5%toluene / oxygen / Temperature profiles were obtained by using a nitrogen flames. Therefore, only the depletion of coated Pt/Rh 6% - Pt/Rh 30% thermocouple of 100µm toluene and the aptitude of the model to reproduce the in diameter located 200µm upstream the cone tip.
main intermediate aromatic chemical species is Conduction heat losses were avoided by setting the discussed in the following. The reaction routes of the thermocouple in a plane perpendicular to the laminar most important intermediate aromatic species are flow. Radiative heat was corrected by using an electric interpreted through reaction path analyses. The most compensation method. Errors in the peak temperatures significant chemical reactions involved in the formation/consumption of a given species is presented.
Reaction mechanism and thermodynamics
The consumption of toluene is well predicted by the model (Fig. 1). However, the model overestimates the mole fraction of toluene near the burner surface.
simulation of the structure of a premixed laminar flame Modeling indicates that toluene is mainly consumed by have been used [3,4]. Species concentrations were H-abstraction reaction and by elimination of methyl, computed with the experimental temperature profiles taken as input data in the code. Thermodynamic data forthe computation of the rate constants of backward reactions have been taken from literature sources [4-6] or computed [7]. The mechanism used here derivedfrom previous studies conducted on the oxidation of The oxidation of toluene by O atom yielding several saturated and unsaturated hydrocarbons [8-10].
methylphenoxy radical OC6H5 CH3 also participates to The reaction mechanism for the oxidation of toluene Comparison of experiment and modelling : results
and discussion
Experimental temperature profiles measured are similar in both flames. The addition of toluene does not affect the temperature. The quantity of toluene (1.5%) added is too low to strongly modify the chemical structure of methane flame; the chemistry is largely determined by the CH4/O2 system in these conditions.
Mole fraction profiles of reactants, final products, reactive and stable intermediate species have been Fig. 1 : Comparison of the experimental (symbols) and analyzed. Oxygenated species detected in a significant computed (solid line) mole fraction profiles of toluene.
concentration were acetaldehyde, acrolein, propanal.
The main aromatic species analyzed are benzene, The benzyl radical was not detected in this work. The phenol, ethylbenzene, benzylalcohol, styrene and maximum computed mole fraction of this species is 1.7 10-4. It is formed principally by the H abstraction As reported for methane :benzene flame, the reactions from addition of H and OH on toluene. The experimental study showed the toluene additive changes consumption of benzyl radical proceeds via three the major products concentrations: H2 and H2O mole reactions yielding benzyl alcohol C6H5CH2OH, fractions decrease while CO and CO2 increase in the burnt gases. Concerning intermediate species, animportant increase of mole fraction species considered as soot precursors (acetylene, ethylene, propene, allene and propyne) was observed. Since these observations were discussed in details for methane/benzene flame ina recent paper [1], it will not be repeated here.
The formation of benzylalcohol is governed by thereaction 920. Its mole fraction profile is correctly A detailed reaction mechanism for toluene predicted by the model as shown in figure 2.
oxidation has been carried out by using these Benzylalcohol then reacts with H-atom producing experimental data. The proposed model correctly reproduces the main experimental observations, in decomposition reaction yielding benzaldehyde particular the effect of toluene on the structure of the are predicted by the model with a good satisfactory as Fig. 2 : Comparison of the experimental (symbols) and Fig. 4 : Comparison of the experimental (symbols) and computed (solid line) mole fraction profiles of computed (solid line) mole fraction profiles of The figure 3 compares the computed and measured mole fraction profile of benzaldehyde ; the agreement between the model and experiment is good.
Benzaldehyde is mainly formed by the reaction 941 and The consumption of benzaldehyde forms principally Fig. 5 : Comparison of the experimental (symbols) and computed (solid line) mole fraction profiles of styrene.
Ethylbenzene is principally formed by the recombination of benzyl and methyl radicals : It is consumed by the thermal decomposition of 1- Fig. 3 : Comparison of the experimental (symbols) and The thermal decomposition 1-phenylethyl radical is the computed (solid line) mole fraction profiles of major source of styrene in our conditions: The benzoyl radical decomposes to form phenyl andcarbone monoxide : Two other reactions contribute to the styrene formation : Ethylbenzene C6H5C2H5 and styrene C6H5C2H3 wereanalyzed in this work and their mole fraction profiles Styrene reacts with H-atom leading to the elimination of is dominated by CH4/O2 system. However, phenol was
analyzed and the model predicts its mole fraction profilewith a very good accuracy as shown in Fig. 7.
The major intermediate aromatic species analyzed inour conditions is benzene. Figure 6 compares the computed and exprimental mole fraction profiles for Fig. 7 : Comparison of the experimental (symbols) and computed (solid line) mole fraction profiles of phenol.
The formation of phenol occurs via reaction (676) and the recombination reaction of H-atom with phenoxy: Fig. 6 : Comparison of the experimental (symbols) andcomputed (solid line) mole fraction profiles of benzene.
Its consumption is dominated by H-atom abstractionreactions with active species H, O and OH: The modeling indicates that benzene formation isgoverned by the following reactions : The reactions (684 to 686) are the most importantsource of phenoxy radical. The phenoxy radical is also Benzene is also formed from benzyl radical and phenol formed by addition of O-atom or molecular oxygen on phenyl radical (678 and 680) and is consumed basicallyby the thermal decomposition reaction yielding cyclopendienyl radical C5H5 and CO (688). The reaction of recombination of phenoxy with H-atom to form phenol (690) is negligible in our conditions.
It is consumed by the H-transfert with H and OH The comparison of the predicted and experimental mole yielding phenyl or by the O-oxidation yielding phenol fractions for cyclopentadiene is presented at Fig. 8. and a very good agreement was obtained for this species.
The evolution of cyclopentadiene is connected to the cyclopendienyl radical via the following reactions: Phenyl radical mainly produced by the two previous reactions (672, 675), is then consumed byrecombination reaction with H-atom (682) and by its The formation of C5H5 predominantly proceeds via the reaction (693) and the thermal decomposition of C 2 (680) yielding respectively benzene and Cyclopentadienyl radical is significantly consumed by Phenyl and phenoxy were not detected in our cyclopentadiene and its isomerization yielding the experimental conditions. The methane flame was seeded by a small quantity of toluene (1.5%) and the chemistry increase in the burnt gases. Concerning intermediate species, an important increase of mole fraction speciesconsidered as soot precursors (acetylene, ethylene, propene, allene and propyne) was observed.
The construction and the evaluation of a detailed reaction mechanism for toluene oxidation have been carried out by using these new experimental results.
Globally, the model correctly reproduces the main experimental observations, in particular the mainaromatic species analyzed in this work. Reactions paths analyses showed that the kinetic scheme is largely References
[1] Dupont, L., El Bakali, A., Pauwels, J.F., Da Costa,I., Meunier, Ph. and Richter H. Combust. Flame. In Fig. 8 : Comparison of the experimental (symbols) and computed (solid line) mole fraction profiles of [2] Turbiez, A. Ph.D. Thesis, Université des Sciences et Technologies de Lille "Etude expérimentale etmodélisation de la combustion du gaz naturel dans des The reaction paths leading to oxygenated cyclic C5 flammes laminaires prémélangées" (1998).
[3] Kee, R.J., Grcar, J.F., Smooke, M.D., and hydroxycyclopentadienyl radical C5H4OH) participate Miller,J.A. SANDIA National Laboratories Report, Livermore, CA, SAND85-8220, 1985.
[4] Burcat, A., and McBride, B. TAE 675, Technion- Israel Institute of Technology, 1993. [5] Kee, R.J., Rupley, F.M., and Miller,J.A. SANDIANational Laboratories Report, Livermore, CA, C5H4OH undergoes a thermal decomposition yielding H-atom and cyclopentadienone. The consumption of [6] Turbiez, A., Pauwels, J.F., Sochet, L.R., Poitou S., cyclopentadienone is exclusively controlled by the Perrin M. Int. Gas Research Conference, San Diego (USA) 371 (1998).
[7] Muller, C., Michel, V., Scacchi G., and Côme, G.M., J. Chimi. Phys. Phys.-Chim. Biol., 92 :1154 (1995).
[8] Dagaut, P., and Cathonnet M., Combust. Sci. and The main routes that govern the consumption and formation of C4, C3, C2, C1 species can be found in [9] El Bakali, A., Braun-Unkhoff,M., Dagaut, P., Frank, P., and Cathonnet, M., Proc. Combust. Inst., 28:1631(2000).
Conclusions
[10] Ristori, A., Dagaut, P., El Bakali, A., andCathonnet, M., Combust. Sci. and Tech., 165 :197 Experimental mole fraction profiles of stable ans reactive species have been measured in laminar [11] Lindstedt, R.P., and Skevis, G., Combust. Flame, CH4/1.5%C7H8/O2/N2 flames at low pressure (0.052atm) by using MB/MS and GC/MS techniques.
Temperature profiles were measured with a coatedPt/Rh thermocouple in the sampling conditions. Molefraction profiles of reactants, final products (CO2, H2O,CO, H2), reactive (CH3, H, O, OH) and stableintermediate species (C2H2, C2H4, C2H6, C3H8, C3H4,C4H6, isomers of C4H8, cyclopentadiene, phenol,toluene) have been analyzed. Oxygenated speciesdetected in a significant concentration wereacetaldehyde CH3CHO, acrolein C2H3CHO andpropanal C2H5CHO. The experimental study showed thetoluene additive does not affect the temperature profile,but changes the major products concentrations: H2 andH2O mole fractions decrease while CO and CO2

Source: http://www.gfcombustion.asso.fr/ecm/2003/022_Elbakali.pdf