L29. Destruction of Ethane in Corona Discharge: Experiment and Modeling

L29. Destruction of Ethane in Corona Discharge: Experiment and Modeling
Anatoli A. Chernov^a, Larisa G. Krishtopa, Oleg P. Korobeinichev^a and Lev N. Krasnoperov

Department of Chemical Engineering, Chemistry,
and Environmental Science
New Jersey Institute of Technology
University Heights
Newark, NJ 07102

E-mail: Krishtopa@adm.njit.edu
E-mail: Krasnoperov@adm.njit.edu

^a)Institute of Chemical Kinetics and Combustion RAS,
Novosibirsk, 630090, Russia

E-mail: Chernov@kinetics.nsc.ru
E-mail: Korobeinichev@kinetics.nsc.ru

This study was undertaken to clarify whether a detailed elementary free radical reaction model could explain the kinetics, energy efficiency and the products of the destruction of a simple hydrocarbon molecule in corona discharge. Ethane was chosen as a simplest hydrocarbon molecule next to methane. Methane was found to be an exception rather than a representative of hydrocarbons for the process of destruction in corona discharge.¹ This was explained based on the high ionization energy of this molecule compared to the ionization energy of molecular oxygen.¹ Ethane has ionization energy lower than that of molecular oxygen, therefore it was expected that the destruction of ethane would be similar to the destruction of the majority of hydrocarbons (this expectation was confirmed in the current experiments).

Destruction of ethane was studied using a flow corona discharge reactor coupled with on line GC/MS and quadrupole mass-spectrometry. A tubular coaxial wire AC high-voltage dielectric barrier discharge flow reactor connected to a Hewlett-Packard 5973 GC/MS and a Finnigan 4021 quadrupole MS was used. The experiments were performed at ambient temperature (298 ± 3 K) and pressure (1.00 ± 0.05 bar). Mixtures of ethane and Zero Air (12, 109, 1033 and 10000 ppm) were passed through the reactor with the flow rates of 0.17 – 4.8 sccs. The active discharge power was varied in the range 0.01 – 4.0 W. The reactants (ethane) and the products of the destruction were monitored on line using the electron impact MS as well as the GC/MS.

Ozone was measured using two independent methods: the UV absorption (253.7, absorption cross section = 1.15x10^-17 cm²/molecule) and the quadrupole mass spectrometry.

The degree of the destruction was measured at different ethane concentrations, flow rates, and discharge powers. The dependences of the ethane concentration on the specific power are double-exponential. The energy efficiencies and the G-values were determined. Methyl nitrate (CH₃ONO₂), ethyl nitrate (C₂H₅ONO₂) and acetic acid (CH₃COOH) were identified among the destruction products. A sequence of elementary reactions leading to these products is suggested.

The experimental data were compared with the modeling results. An elementary reaction model that consists of 795 reactions of 86 species and incorporates the low temperature oxidation reactions of hydrocarbons as well as of nitrogen oxides was built and used in the model calculations. The Konnov’s mechanism² was used as a basis for the model. This mechanism is applicable to the high temperature oxidation of ethane. More than 100 reactions of low temperature importance were added to the mechanism based on the NIST 97 kinetics database. Reactions involving nitrogen oxides, including excited nitrogen and oxygen atoms and some molecular metastable states, were reviewed and incorporated into the mechanism. The kinetics and mechanism of formation of ethyl nitrate and methyl nitrate were taken from the literature.^3,4 The initiation process was assigned to the dissociation of molecular oxygen and nitrogen in corona discharge.^5-7 In addition, the destruction of ozone (which can accumulate up to high concentrations) and water (which represents a source of very active hydroxyl radicals) were included. The model was calibrated based on the comparison of the predicted and the experimentally observed yield of ozone.

The G-values of the destruction determined from the experimental data at different concentrations of ethane, together with the modeling results are shown in Fig. 1. At the highest ethane concentration used (1%) the discrepancy between the predicted and observed efficiencies of destruction is ca. factor of 10. The discrepancy increases at low concentrations and reaches ca. factor of 10⁴ at the lowest concentration used (12 ppm).

Fig.1. G-values for the destruction of ethane in corona discharge: Experiment and modeling.

The final products of the destruction, water and carbon dioxide, were observed using the MS detection. No attempt of quantitative measurements of these products was made. A search for CH₄, C₂H₂, C₂H₄, CH₂O, CH₃OH was performed. These products were not detected. The following intermediate products were identified and quantified: ethanol (C₂H₅OH), nitromethane (CH₃NO₂), methyl nitrate (CH₃ONO₂), ethyl nitrate (C₂H₅ONO₂) and acetic acid (CH₃COOH). At low initial concentrations of ethane a broad peak in the GC/MS chromatogram was observed. The mass spectrum identification based on the NIST MS spectral library indicated 1,2-propanediol, 3,3’-oxydi-, tetranitrate (C₆H₁₀N₄O₁₃) as a product responsible for this peak.

At high concentrations the model qualitatively predicts formation of all observed compounds, however, no quantitative agreement can be reached. The quantitative discrepancy between the model and the experiment increases at low concentrations of ethane. The model predicts higher relative yields of the products of free-radical reaction compared to the experimental data. The relative yield of the species responsible for the broad peak tentatively identified as C₆H₁₀N₄O₁₃ increases at low ethane concentrations.

In this work a detailed experimental and modeling study of the destruction of ethane in zero air in a dielectric barrier corona discharge was performed. The model qualitatively predicts the destruction products with a factor of 10 discrepancy in the destruction efficiency for the concentration of ethane of 1%. The discrepancy in the predicted and the experimentally observed efficiency of the destruction increases at low concentrations and reaches a factor of 10⁴ at 12 ppm. Simultaneously, the relative yield of the free-radical products also decreases at low concentrations of ethane. The results unambiguously indicate a complete failure of the free-radical mechanism to account for the experimental observations. The results of the current study are in agreement with the kinetic characteristics of the destruction of other hydrocarbons in corona discharge.¹

1. Krasnoperov, L.N.; Krishtopa, L.G.; Bozzelli, J.W. J. Adv. Oxidation Technologies, 1997, 1, 243.

2. Konnov, A. Combustion Mechanism, Release 0.5,.http://homepages.vub.ac.be/~akonnov/science/mechanism/

3. Scholtens, K.W.; Messer, B.M.; Cappa, C.D.; Elrod, M.J. J. Phys. Chem. A. 1999, 103., 4378.

4. Ranschaert, D.L.; Schneider, N.J.; Elrod, M. J. J. Phys. Chem. A. 2000, 104, 5758.

5. Eliasson, H. M.; Kogelshatz, U. J. Phys. B: Appl. Phys. 1986, 22, 1241.

6. Peyrous, R.; Pignolet, P.; Held, B. J. Phys. B: Applied Physics. 1989, 22, 1658.

7. Penetrante, B.M.; Hsiao, M.C.; Bardsley, J.N.; Merritt, B.T.; Vogtlin, G.E.; Wallman, P.H.; Kuthi, A.; Burkhart, C.P.; Bayless, J.R. Physics Letters A, 1995, 209, 69.