| OCR Text |
Show 1.7.2 I. INTRODUCTION In a companion paper C1D» two simple models are presented for the noncatalytic reduction of NO with NH3 . The experimental data available for testing these models are rather limited, and differ among themselves in several important respects. The data of Muzio, Arand and Teixeira C2] indicate a relatively small dependence of the NO reduction at low temperatures on the initial amount of N H 3 . This can be accounted for by postulating the formation of N H 2 from reactions between N H 3 and H-atoms (reaction 3a of [1]). However, other experimental results C3,4] do not show the same behavior. There also are large differences in the rate of NO reduction in the various available experimental results. These differences are likely to be the result of different values of the OH-concentration [4,1]. Finally, there are questions about the influence of wall reactions in experimental results obtained in tubes of small diameter. The present investigation was undertaken in an attempt to clarify the origin of these various differences, and also to assess the possible influence of wall reactions. The introduction of known initial concentrations of H 2 and 0 2 in the gas sample upstream of the N H 3 injection point allowed computation of the equilibrium concentration of OH. The influence of the wall reactions was investigated by using both argon and helium as a carrier gas. II. EXPERIMENTAL SET-UP The experiments were carried out using the high temperature flow reactor previously described by Roby [5]. A sketch of the reactor is shown in Fig. 1. The reactor has a length of 100 cm, and an inner diameter of 2.5 cm. The innermost tube of the reactor consists of McDanel 998 alumina tubing (99.8% A1203). The outer diameter of this tube is 3.2 cm. A groove in the outer surface of the tube accomodates a Pt-Rh wire used for resistance heating. The temperature of the outer surface can be monitored with several thermocouples. The tube is thermally insulated from the outside by concentric tubes of zirconia and alumina. The innermost tube is held in place by zirconia supports (see Fig. 1). The entire assembly is mounted in a vacuum housing consisting of a brass tube with a length of 110 cm and an inner diameter of 20 cm. The brass tube is provided at both ends with demountable flanges, and is cooled by water flowing through a copper tube coiled around the outer surface. A port on the brass housing is connected to the vacuum pump which maintains a constant gas flow through the reactor. Another port contains electrical feed-throughs for the heating cables and for the thermocouple wires. The flange at the upstream end of the reactor is provided with three ports through which the sample gases enter. These ports are used for a mixture of NO, helium and argon, a mixture of H 2 and 02 , and for NH3 , respectively. The ammonia is injected into the other gases 20 cm downstream of the flange. The flange at the downstream end can accomodate either |
