Current Methods of Measurement of NO, Nitrogen Oxide (NO_2) and Ammonia Analysis in Olefins Plants

摘要

After the incident in the Port Arthur (1) cold box in 2008 that wasattributed to NOx gums, we believed good on-line measurement of NO in severalstreams in the cracker was necessary. A variety of technologies wereinvestigated and no off-the-shelf solution was found. In the 1993 Roy Hallepaper (2), the analytics section suggested Gas Chromatography with aPhotoionizaton detector (GC/PID) for use in this application. Several companiesput these instruments in their plants and it remains a viable technology.However, the detector suggested in this study is no longer available and otherPID’s do not show the sensitivity needed. Lab GC’s using the chemiluminescentdetector (C_LD), worked well but no on-line instrument is available. Questionsconcerning the on-line application of this technology were posed to several onlineGC manufacturers. Of these, none were willing to advance the technology toon-line instruments because of the size of the detector and electricalclassification required for on-line instrumentation. Other technologies such asGas Chromatograph/Discharge Ionization Detector (GC/DID), GasChromatography /Nitrogen Phosphorus Detector (GC/NPD) and GasChromatography/Helium Ionization Detector (GC/HID) were explored but a greatdeal of research would be necessary to bring these technologies to the marketfor NO analysis. Two basic problems are: 1.) The detection limit required is lessthan 5 ppb, 2.) Universal detectors such as DID that have the sensitivity wouldrequire very difficult chromatography to gain the necessary separation of NOfrom other components. CI Analytics very recently had introduced an on-lineversion of their lab tape analyzer to detect and quantify NO+NO2 in lighthydrocarbon streams. This technology has the advantage of good sensitivity andthe tape is selective for NO+NO2 only. C.I. Analytics claims that there are noknown interferences in the method.In addition, should a new technology show promise, one must assure thatthe technology measures NO accurately and precisely. Precision can bemeasured without a reference by measuring the same sample over and over.Accuracy requires calibration, as well as assuring that there are no matrix effectsfrom the actual sample. At Port Arthur, we are fortunate in having the TOTALResearch Labs near us in La Porte, Texas. TOTAL Research does have aproprietary GC/C_LD instrument. We felt that if we could find a suitabletechnology, we could compare with the TOTAL Research instrument and assurethat the accuracy of our technique is suitable. However, comparing with an offlinetechnology requires sampling, which adds another dimension of problemssomeof which are discussed here.This paper will deal with the evolution of the NO/NO2 tape analyzer fromCI Analytics and the comparison with the GC/C_LD technology at TOTAL Researchin La Port. Also, a short section concerning the technologies available for NH3analysis is included.