Simple modelisation of ozone generation by positive corona discharge

Simple modelisation of ozone generation by positive corona discharge K. Yanallah1, F. Pontiga2, A. Castellanos3, S. Hadj-Ziane1, Y. Messlem1 and A. Belasri4 1Laboratoire de Génie Physique, Université de Tiaret, B.P.78, Tiaret, Algeria 2Dpt. Física Aplicada II, Universidad de Sevilla, Av. Reina Mercedes s/n, Spain 3Dpt. Electónica y Electromagnetismo, Universidad de Sevilla, Av. Reina Mercedes s/n, Spain 4Laboratoire de Physique des Plasmas, des Matériaux Conducteur et leurs Applications, Université d’Oran (USTO-MB) Ozone molecule has unique oxidizing properties that are in the base of its many industrial applications, such as air cleaners, water purification, odor control, gas treatment, etc. [1-4]. Therefore, many studies have been devoted to unveil the mechanisms of generation of ozone in electrical discharges [5-7], and to increase the efficiency of ozone production. In this work, we present a numerical study of ozone production by positive DC corona discharge. The electrical discharge occurs between a wire and a coaxial cylinder, and pure oxygen is feed into the discharge cell (Fig. 1). As it is well known, the corona discharge is initiated when the electric field near the wire is sufficiently high to ionize the gaseous species. The minimum electric field is a function of the wire radius, the surface roughness of the wire, air temperature, and pressure [8]. Corona discharge has been simulated by using a hydrodynamics model that combines the physical processes in the corona discharge with the chemistry of ozone formation and destruction in the oxygen stream. Basically, it consists in a set of continuity equations, coupled with Poisson’s equation. The continuity equations govern the transport and the gain/loss balance of every species due to the chemical reactions induced by the electrical discharge. The current-voltage characteristic (CV) measured in experiments is used as input data to the numerical simulation. Fig. 2 presents the radial distribution of charged species and the electric field for a positive DC corona discharge from the wire (anode) to the cylinder (cathode) at room temperature (T = 300 K) and atmospheric pressure. The anode and cathode radius are 0.00625 cm and 1.35 cm, respectively, the applied voltage is 8 kV, and the gas flow rate is 100 cm3/min. Fig.1. Schematic diagram of the positive corona discharge cell. 32 0,01 0,1 1 101 103 105 107 109 1011 103 104 105 Density (cm-3) Radial distance, r (cm) e O2 + O2 - E(V/cm) Fig. 2. Radial distribution of charged species (left axis) and electric field (right axis) in a positive corona discharge, for V = 8 kV and Q = 100 cm3/min. The distribution of species is the result of the competition between ionization and electron attachment. Near the wire, the electric field is very high, so ionization prevails over attachment and new electrons are produced. A few millimeters away from the wire, the ionization rate equals the attachment rate. All newly produced electrons attach to oxygen molecules to form negative ions, e + O2 + M → O2- + M (M = O3, O2, etc.). Theses negative ions move then towards the discharge wire. Outside the ionization region, the electric field strength is insufficient to produce electrons. Therefore, positive ions drift into this volume towards the grounded cylinder. The plasma is in a non-equilibrium state and has a low degree of ionization. 0,2 0,4 0,6 0,8 1,0 1,2 0,0 7,0x1015 1,4x1016 2,1x1016 2,8x1016 3,5x1016 O3(cm-3) Radial distance, r (cm) z=10 cm z=3,6 cm z=0,36 cm z=0,036 cm z=0,0 cm Fig. 3. Radial distribution of ozone density at five different locations along the cylinder, for V = 8 kV and Q = 100 cm3/min. The radial variation of the ozone density is represented in Fig. 3. As expected, ozone concentration increases progressively as the oxygen flow advances along the cylinder. Just after the entry of the oxygen gas into the discharge tube (z = 0.0 cm), the radial distribution of ozone is highly inhomogeneous. In that location, the density of ozone has a maximum in the vicinity of the wire, where the active region of the corona discharge is located, and then sharply declines. This decline reveals that ozone diffusion has not had enough time to act in the radial 33 direction. A crude estimate of the diffusion time of ozone gives td ~ 5 s, which is much shorter than elapsed time of flow. As oxygen gas advances along the cylinder, ozone diffuses towards the cathode and its concentration becomes more uniform. Finally, Fig. 4 shows the axial distribution of ozone at a constant radial distance and for two different values of the applied voltage. After a certain entrance length, ozone density is observed to increase linearly along the axial direction for low applied voltages. The predictions of the numerical simulation are compared with the experimental measurements in Fig. 5. In that figure, the averaged ozone density has been plotted as function of the applied voltage. The results of the simulation are in qualitative agreement with the experimental measured data. The numerical simulation tends to overestimate the ozone density, as the predicted ozone concentration is 50% higher than the actual measured concentration. 0246810 0 1x1016 2x1016 3x1016 4x1016 O3(cm-3) Axial distance, z(cm) V= 6 Kv V= 8 Kv Fig. 4. Axial variation of ozone density in the vicinity of the wire (r = 0.01668 cm) for two different applied voltages and Q = 100 cm3/min. 68 1x1015 2x1015 3x1015 4x1015 10 O3(cm-3) Applied voltage, V(Kv) Experimental Modelisation Fig. 5. Experimental and numerical results of ozone density as a function the applied voltage for Q = 100 cm3/min. REFERENCES [1] C. Gutiérrez-Tapia, E. Camps and O. Olea-Cardoso, “Perturbative method for ozone synthesis from oxygen in a single discharge”, IEEE Trans. Plasma Sci. 22, 979 (1994). [2] A. D. Moore, Electrostatics and Its Applications, John Wiley & Sons, New York (1973). 34 [3] J. A. Cross, Electrostatics: Principles, Problems and Applications, Adam Hilger, Bristol, England (1987). [4] M. L. Balmer, G. Fisher, J. Hoard., Society of Automotive Engineers, Inc., Warrendale, PA, 1999. [5] Skalny J D, “Mass spectrometry of low energy ions extracted from negative corona discharge in air at medium pressure”, Acta Physica Universitatis Comenianae, 28, 161 (1987). [6] M. Laan, J. Aarik, R. Josepson and V. Repän, J. Phys. D Appl. Phys. 36, 2667 (2003). [7] J. F. Loiseau, F. Lacassier, C. Monget, R. Peyrous, B. Held and C. Coste, “Numerical simulation of ozone axial and radial distribution in a cylindrical oxygen-fed ozonizer”, J. Phys. D Appl. Phys. 27, 63 (1994). [8] F. W. Peek, Dielectric Phenomena in High-Voltage Engineering, 3rd ed., New York, (1929). [9] C. Soria, F. Pontiga, and A. Castellanos, “Plasma chemical and electrical modelling of a negative DC corona in pure oxygen,” Plasma Sources Science and Technology, 13, 95 (2004). [10] P. Atten, M. McCluskey, and A. Lahjomri, “The electrohydrodynamic origin of turbulence in electsotatic precipitators,” IEEE Transactions on Industry Applications, 23, 705 (1987). [11] K.Yanallah, S.Hadj-Ziane, and A.Belasri, “Ozone decomposition on walls”, Plasma Devices and Operations, 14, 215 (2006). 35 View publication statsView publication stats