OCR Text |
Show 1.7.8 diluent, the input parameters are K = 0, D = 5.1 x 10"3, K* = 1.8 x 10"3; it now follows that K* = 9.1 x 10"1*, which is in good agreement with the helium result. The corresponding result for the surface reaction rate constant is k w = 0.12 cm/s. It can be concluded that under these conditions the overall reaction rate is limited by the rate of reaction at the wall, rather than by the rate of diffusion toward the wall. An estimate of the relative importance of wall reactions at higher temperatures can be made by using the deduced values of k3[OH3 to find a value for K*, and by using the value of k3 measured by Silver and Kolb [83 and Fujii [93 together with the calculated value of COH3e q to find a value for K. At T = 1100 K (the temperature of maximum reduction), the resulting values are K* = 3.5 x 10"2 and K = 1.4 x 10"2. Together with the value D = 6.1 x 10"3 found for argon at T = 1100 K, this yields K w = 2.5 x 10"2, and hence K^, - 4.5 cm/s. The corresponding rate of wall reaction at the axial position of the sample probe is 47% of the total rate. Similar values are found with helium as the diluent; the corresponding wall reaction rate in this case is about 60% of the total rate. It follows that at these temperatures the rate of wall reactions depends measurably on the rate of diffusion toward the wall. As a general conclusion, these estimates indicate that the occurrence of wall reactions is a possible explanation of the data obtained. It should be pointed out that an alternative explanation for the relatively large reduction rate at low temperatures is the possible presence of a superequilibrium concentration of OH radicals. This presence could be associated with continuing reactions between hydrogen and oxygen, caused by incomplete mixing and heating of these gases upstream of the ammonia injection station. Similar effects arising from the addition of hydrogen and hydrogen peroxide have been reported in [15-173. In [173, results are also reported for the combined heterogeneous and homogeneous NO reduction process. However, this alternative explanation would not readily account for the difference in maximum reduction observed with argon and helium as the carrier gas. It may be that the reduction rate observed in the present work arises from a combination of the two phenomena mentioned. The fit that could be obtained for the data with argon as the diluent with the extended model of [13 is shown in Fig. 8. The values used for the various parameters are TQHk 3 = 2.8 x 10~5 ppm"1, T^k4 = 9.9 x 10""H ppm"1, oc = 0.75. Compared with Fig. 7, the fit is improved as far as the locations of the minima are concerned. It should be kept in mind that the values listed are partly representative of wall reactions. A corresponding result for the data with helium as the diluent is shown in Fig. 9. It may be concluded that appreciable reduction of NO at relatively low temperatures may be obtained under the conditions prevalent in the thermal reactor. The simple models of [13 can be used to generate analytical results that fit the experimental data quite closely. The |