Abstract
Pulsed positive corona discharges are used to remove NO from the flue gas of a methane burner. At low power input this leads to an increase in NO2, which shows that the process is oxidative. Removal efficiency is greatest when discharges are produced with high-voltage pulses, which are shorter in duration than the time required by the primary streamers to cross the discharge gap, in combination with a dc bias. Other important parameters are input power density and residence time. The best result obtained so far is an energy consumption of 20 eV per NO molecule removed, at 50% deNOx i.e., a removal of 150 ppm NOx, using a residence time of 15 s and an input power density, of 3.5 Wh/Nm3. [Wh/Nm3 stands for watt-hour per normal cubic meter, i.e., at normal conditions (273 K and 1 bar). This implies that 1 Nm3 contains 2.505 1025 molecules.] There appears to be room for improvement by the addition of gaseous and particulate chemicals or the use of multiple corona treatment. It is argued front comparison between results from models and experiments that the direct production of OH by the discharge is only the initiation of the cleaning process.
Original language | English |
---|---|
Pages (from-to) | 227-247 |
Number of pages | 21 |
Journal | Plasma Chemistry and Plasma Processing |
Volume | 16 |
Issue number | 2 |
DOIs | |
Publication status | Published - 1996 |