OCR Text |
Show 1.7.4 below the pressure in the reactor. Both of these pressures are kept just above atmospheric pressure in order to prevent air from leaking in. The probe is held in place by means of a brass sleeve screwed onto the flange. The gas samples were analyzed using a Varian Model 3700 gas chromatograph equipped with a Porapak Q column. This column has a very small elution time for NO (about 20 s), and is insensitive to other gases. A thermal conductivity detector is used to sense the presence of NO. The output of this detector is monitored by a strip chart recorder. A digital integrator provides the measure of the area under the NO signal. The magnitude of this area and the maximum value of the signal (the "peak height") both are measures for the amount of NO present in the gas sample. The system was calibrated by injecting various samples with known amounts of NO into the gas chromatograph, and by recording the resulting integrator reading and peak height. It was verified that the response was linear. The maximum amount of NO in the various samples corresponded to 5000 ppm. In the analysis of unknown samples it was found that the relative error in using peak height versus integrator reading is on the order of 5% provided the analysis and the calibration are done on the same day; otherwise the difference can increase to 10%. The possible reasons for these differences are many, and include carrier gas impurities and detector noise. In order to minimize errors, the system was recalibrated each time that measurements were made. III. EXPERIMENTAL PROCEDURES AND RESULTS. Upon turning on the heating circuits of the reactor, the initial current to each of them was set to 15 A. The reactor temperature was allowed to reach the lower end of the range of interest te 700 K), after which the control circuits were turned on. The gas chromatograph was turned on about 24 hr. earlier, and the various modules were set to the conditions required. These were as follows: carrier gas flow rate = 40 cc/min; column temperature 303 K; injector temperature 353 K; detector temperature 413 K; filament temperature 543 K; detector sensitivity 0.5 mV/full scale. It was verified that the detector zero was not drifting during the experiment. The total flow rate Q(t) through the reactor was set to about 12 standard litres/min. The mole fractions X(i) of the various gases l were selected using kinetic considerations, and were as follows: X(02) = 0.02; X(H2) = 0.026; X(NO * He) = 0.005'X(He) = 0.944; X(NH3 = A-X(NO), where A = 1.6, 1.3, 1.0, 0.5 or 0.3. The flow rates Q(i) follow from the relations Qd) = X(i)Q(t) for i = 02 , H2, N H 3 and He, and Q(NO + He) = X(He)Q(He) • X(NO)Q(NO). These flow rates were used to calculate the argon flow rate required to obtain an initial NO concentration of 5000 ppm. After turning on the gas flows, the temperature probe was replaced by the sample probe. The syringe was flushed a few times with air in order to remove any dust particles |