Conventional leak detectors based on liquids

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: es bypassing the reactor and through the reactor was used to check whether the reactor was leak free before starting each experiment. Conventional leak detectors based on liquids producing bubbles in case of a leak had proven not to be suitable for the detection of small leaks at atmospheric pressure. A portable thermal conductivity detector (TCD) was also not sufficient. The reactor was assumed leak free only if the difference in the flow rate between bypassing the reactor and through the reactor was ≤ 0.1% (≤ 0.04 ml/min at a total flow rate of 40ml/min). Only if this criterion was fulfilled the catalytic tests were reproducible. The peak area of the TCD signal was calibrated for all reactants and products by onepoint calibration except for H2, C3H4 and C3H6. O2 and C3H8 were calibrated using the not reacted substrate feed, for C3H4 and C3H6 the same calibration coefficients were assumed as for C3H8. CO, CO2, CH4, C2H2, C2H4, C2H6 were calibrated from a calibration mixture each compound at ~1% in N2 (Scott). Due to the closeness of the retention times of CO and N2 and the large amount of N2 in the calibration mixture (~94%), it was not possible to achieve a sufficient separation of the CO and N2 peak using the Carboxen 1004 column. Therefore CO and N2 were separated using a freshly activated Molsieve 5A column. The peak area of CO for the Carboxen column was calculated from the ratio of the well separated CO and N2 peak area using the molsieve fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff 64 2. Experimental fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffftfffffffffffff 5A column and the total area of the not sufficiently resolved N2-CO peak over the carboxen column according to: Area CO Molsieve 5 A Area (CO + N2 ) Molsieve 5 A = Area CO Carboxen 1004 Area (CO + N2 ) Carboxen 1004 As N2 was never present in the catalytic tests the problem of the separation of N2 and CO occurred only during the calibration. Since H2 has a non-linear thermal conductivity...
View Full Document

Ask a homework question - tutors are online