OCR Text |
Show 1.6.7 NO reduction. The model of Sec. II cannot account for this shift. Instead, it predicts that as A decreases from 1.6 to 0, the optimum temperature will increase (cf. Fig. 3). Neither can the model account for the relatively small dependence on A of the reduction in NO observed at low temperatures. Instead, the model simply predicts that [NO]/[NO]0 « 1 - A s* 1 - AT(T) (low T). These deficiencies are not present in the results of the extended model presented in Sec. IV. The values obtained for k3[OH3 and [N03c can also be used to calculate [NO] as a function of time t for given temperature T. Curves for five different values of T are shown in Fig. 4, together with the experimental data of Fig. 5 in [4]. The curves are applicable to the same conditions as used to generate the square data points (A = 1.04, C02] = 4%). The latter points were taken at T = 1297 K, but fall along a curve for which the temperature is about 40 K higher. The circular data points were taken under slightly different conditions (A = i.08, [02] = 2%), at T = 1238 K. Except for the point at t = 190 ms, they, too, fall along a curve for which the temperature is some 40 K higher. Corresponding curves for [NH3] as a function of t are shown in Fig. 5. Again, the data generally fall along curves for which the temperature is about 40 K higher. Apparently, the OH-concentration in these experiments was larger than would follow from the measured temperatures together with the data shown in Fig. 3. A similar conclusion was reached by Salimian and Hansen [16], who suggested that the initial OH-concentration in the experiments of [4] may have been higher than the equilibrium level. Apart from this discrepancy, the agreement between data and curves is satisfactory. Using Eqs. (4) and (17) together with the definition tc = l/(k3[OH]), it is found that CNO]c P tc = P k2/(kik3) * 4.4 ppm atm s, (20) where P is the total pressure. This equation illustrates the trade-off that exists between [NO]c and tc: if one of these parameters is decreased, the other increases proportionally. Other numerical evaluations that can be made concern e = (k3/'k2)xc and COH1 Adopting the relation kj = 5.0 x 1013 cm^gmole'^s"1 listed on p. 40 of [25], and using Eq. (17) leads to e si 4.4 x 10~3 xc, so that indeed c 3 1. Using Fenimore's value k3 = 3 x 101* cn^gmole'^s"* [12], the present results (16), (18) and (19) yield [OH3 « 9.2 x 1013 exp(-3.95 x 104/T) ppm, where T is in kelvin. This relation can be compared with the equilibrium value [OH3eq, calculated from the relation [OH3eq = 10*aCoH/KH2O1/2^O2 1'l,pk2Ol/2/P) p p m' where K Q H a n d KH?0 *re equilibrium constants taken from the JANAF tables [28]. Furthermore, P Q ? is the partial pressure in atmosphere of O2, given in [4] as 0.04, and PH^O |