1、GPA TITLE*SYMP 90 D 3824699 0002584 636 W October 10-1 1 , 1990 Dallas, Texas SYMPOSIUM ON LP-GAS ODORILATION TECHNOLOGY Sponsored jointly by Gas Processors Association National Propane Gas Association Propane Gas Association of Canada GPA TITLErSYMP 90 3824b 0002585 572 W Table of Contents Foreward
2、 . ii Summary Report of Odorant Investigations William B. McHenry and H. M. Faulconer 1 Trained Panel Determinations of Odorant Detection Thresholds D. L. Ripley, J. W. Goetzinger and M. L. Whisman . 5 Untrained Panel Determinations of Odorant Warning Level D. L. Ripley, J. W. Goetzinger and M. L. W
3、hisman 12 Field Testing for Ethyl Mercaptan in LP-Gas Storage Tanks in Four Marketing Areas M. D. Marshall and C. A. Palladino 22 Field Test Program of Measuring odorant In Continuous Use Tank LP-Gas Tanks William J. Hines and Carl G. Hefley 32 Ethyl Mercaptan Stability after Refill of New and Air-E
4、xposed Used LP-Gas Storage Tanks M. D. Marshall and C. A. Palladino 54 LP-Gas Odorant Stability Under Simulated Residential Conditions Michael R. Kuhlman . 58 Laboratory and Field Experience with Ethyl Mercaptan Odorant in Propane S. A. Holmes, P. B. VanBenthuysen, D. C. Lancaster and M. A. Tiller 6
5、8 A Review of Canadian Research on Propane Odorizatlon I. D. Campbell, R. J. Falkiner and A. L. Pickard . 72 Odorant Vapor-Liquid Equilbria-A Generalized Treatment John Jambus and David Swienton 79 Prediction of Odorant Distribution in LPG Tanks Mahmood Moshfeghian, Edgardo Lopez and Robert N. Maddo
6、x 87 LP-Gas Detection: Odorization and Electronic Methods - Complementary Rather Than Competitive Technologies W. Alan Bullerdiek . 94 Low Pressure Application of Stain Tube Test for Mercaptan Odorant Daniel W. Kemp 100 Author Biographies 107 i GPA TITLE*SYMP 90 M 3824699 000258b 409 Foreward Since
7、late 1986, a joint task force composed of representatives of the Gas Processors Association (GPA), the National Propane Gas Association (NFGA) and the Propane Gas Association of Canada (FGAC) has conducted an intensive study of LP-Gas odorants, odorization practices and related equipment in the LP-G
8、as fuel system. The task force investigation began with an extensive review of the published literature and available data, which revealed very little documented information on the behavior of odorants in LP-Gs. However, a large body of conjecture, unsupported by fact or data, demanded a thorough sc
9、ientific investigation of the odorant warning system. Early investigations included, among other things, experimental determination of vapor- liquid equilibria for current and possible alternate odorants, development of a simple and rapid field test to measure odorant concentrations, extensive resea
10、rch to determine the occurence and extent of odorant depletion in new and existing systems, and a critical review of industry practices. The first Symposium on LP-Gas Odorization Technology, held April 18-19,1989, documented the results of these and other early studies. This second Symposium on LP-G
11、as Odorization Technology presents resulta of subsequent research by sponsoring organizations and others. Papers and discussion presented at the symposia reflect only the views of the authors and do not necessarily reflect views or opinions of the sponsoring organizations, their member companies or
12、the Joint Odorization Task Force. Papers for this Symposium were solicited from any interested party through an open “Call for Papers“ with the intent of stimulating maximum discussion and exchange of information on the subject of LP-Gas Odorization technology. In some cases, specific invitations we
13、re issued; however, no individual, group of individuals or organization acted as a review body to screen or review papers prior to their presentation. 11 GPA TITLErSYMP 90 m 3824b99 0002587 345 m Summary Report of Odorant Investigations William B. McHenry Pro-Gas Sales andSeruce Company Mikegon, Mic
14、h no commercial odorant available totally satisfies these ideal conditions. Inasmuch as odorized LP-Gas is a fungible product which may end up in more than one distribution system, a single odorant for the entire industry is desirable, if not absolutely necessary. Odorization is commonly accomplishe
15、d at the point of custody transfer into the retail distribution system, namely at the plant loading rack or pipeline terminal. With this background, let us examine what was happening in 1986. Earlier in the mid-l980s, there appeared to be an increasing number of questions and allegations raised rela
16、ting to the effectiveness of the odorants used in LP-Gas. There was a growing concern within the industry to determine the validity of these allegations and, if valid, what could be done about them. be a broad problem impacting all industry segments (producers, transporters, terminal operators, and
17、marketers), the industry formed a Joint Task Force on LP- Gas Odorization in late 1986 composed of representatives of the National Propane Gas Association, Gas Processors Association, and Propane Gas Association of Canada to share information and develop a coordinated approach to evaluate, research
18、and recommend solutions to any problems that might exist. Initial Joint Task Force efforts envisioned multiple studies of all identifiable facets of the alleged fade problem. These planned studies included Compilation and review of all known and pertinent data; Development of a field test method for
19、emeasurement of mercaptans in consumer tanks in continuous service; Inasmuch as the alleged odorant problem was deemed to 1 GPA TITLEbSYMP 90 m 3824699 0002588 281 m Experimental work to determine physical properties of alternate or supplemental odorants; Investigation of odorant practices in other
20、counties; and; Investigation of the so called “fade” mechanism and effects on tank surfaces. As these studies progressed, the Task Force agenda was broadened to include the following: Increased hazardous labeling and dissemination of factual information to all industry segments; A symposium on LP-Ga
21、s Odorization Technology, intended to document all known data on this subject. A review and revision, if deemed necessary, of NFPA 58 relative to odorization requirements; Theoretical and human response research into alternate or supplemental odorants; and, Cooperative efforts and coordination of re
22、search with those of the U.S. Consumer Product Safety Commission. One of the first jobs for the Task Force was to examine the results of an exhaustive literature search that had been conducted by the Institute of Ggs Technology in early 1986. The search failed to reveal any published research or dat
23、a relative to “fade” of odorant in LP-Gas systems. The search reviewed IGT Gas Abstract (1945-1986), Chemical Abstracts (1967-1986), and the Department of Energy Data Base (1974-1986). There did not appear to be any relevant previously published information on the “odor fade” subject. The Task Force
24、 identified the following categories of study and has either engaged in research in these areas or has monitored the research of others: 1. Odorant oxidation and/or adsorption in continuous service containers. 2. Odorant oxidation and/or adsorption in new steel containers or reinstalled used contain
25、ers after prolonged exposure to the atmosphere which involves two situations: a. Initial operation and first refill b. Effect of manufacturing processes and post manufacturing treatments of the interior tank surfaces. 3. Loss of odorant in the vapor phase due to adsorption onto porous interior room
26、surfaces. 4. Alternate odorants. Supplementing the industry concerns was the publication A review of the adequacy of ethyl mercaptan, and of a CPSC report in July 1986 which recommended research to determine the availability of improved odorants or mixture of odorant that are not as susceptible to o
27、dorant fading and are detectable to a wider segment of the population. Force has either administered research projects or has During the ensuing four years to the present, the Task monitored work done by others that address these four categories, all of which have been reported in last years or this
28、 Symposium. Let us examine what has been done. STUDY CATEGORY ONE The first category addresses the so-called “fading” of ethyl mercaptan in existing LP-Gas systems in continuous service. The industry did not believe this to be a problem area, because the LP-Gas is handled in closed pressurized syste
29、ms with minimal sources of oxidizing agents. However, since no significant data was available, the Task Force authorized a series of tests of the odorant concentration in field containers. Before any objective, quantitative evaluation of odorants in LP-Gas systems could be made, a simple, inexpensiv
30、e field test method was necessary. Conventional laboratory analysis by gas chromatography, while more precise, requires expensive laboratory equipment, highly trained personnel, and extreme care in obtaining and transporting the samples for laboratory analysis. Although portable chromatography equip
31、ment is available, it is also expensive and also requires highly trained personnel to obtain meaningful data. developed two test methods, both of which are based on commercial length of stain detector tube technology, commonly used to detect trace contaminants in air and other gases. investigated us
32、e of the method for determining ethyl mercaptan concentrations in propane vapors. Evaluation of the procedure established that the associated results are well within the stated accuracy of plus or minus 25%. In actual experience accuracy is often within plus or minus 10% of the amount measured. The
33、vapor phase method was published in 1988 as GPA Standard 2188 titled Tentative Method for the Determination of Ethyl Mercaptan in LP-Gas Using Length of Stain Tubes”. The Propane Gas Association of Canada developed and demonstrated the reliability of a similar method to measure mercaptan concentrati
34、on in liquid propane. The method was published by the Canadian General Standards Board as CANKGSB-3.0 Test for Ethyl Mercaptan Odorant in Propane, Field Method”. With the development of the GPA Standard 2188 Test Method, it was then possible to initiate a massive field test program in mid 1988 to me
35、asure odorant in the vapor phase from LP-Gas tanks that had been in continuous use. Nine testers, typical of propane marketer distribution personnel, were provided instructions on the use of equipment and procedures described in GPA 2188. Of606 valid tests of tanks located in 14 states, not a single
36、 test indicated depletion of ethyl mercaptan. In additon, confirming sniff tests were positive for all tested tanks. The test program confirmed the absence of serious problems in tanks larger than 100 lb capacity that have been in continuous service. To confirm findings of these field tests conducte
37、d by non- technical employees of pro$ane marketing companies, the Mine Safety Appliance Research Corporation was engaged to do similar testing by technically qualified personnel. The test program was designed to conduct stain tube testing of the vapor phase of 400 LP-Gas consumer tanks in four separ
38、ate marketing areas, and to perform gas chromatography analysis for confirmation on the liquid and vapor phases of approximately 20 tanks. The results Working concurrently, the U.S. and Canadian industries The Gas Processors Association Technical Section B 2 GPA TITLEUSYMP 90 3824699 O002589 118 i p
39、roduced no evidence that tanks in continuous service undergo “Odorant Fade“. Product Safety Commission, MSA Research Corporation completed a study of the effects of time and temperature on odorant concentration. The concentration of ethyl mercaptan in the vapor phase of a commercial supply of LP- Ga
40、s was monitored for nine months in two continuous service tanks and one new 50.0 gallon tank all under static conditions and exposed to western Pennsylvania ambient weather. The odorant concentrations in the used tanks varied with the temperature of the tank liquid, measuring above 10 ppm at the rec
41、orded high (90 deg. F), and over 5 ppm at the lowest temperature (10 deg. F) recorded and showed no indication of odorant loss. The results from the new tank will be discussed when I cover testing of new steel tanks under category 2. In a separate temperature controlled study, as part of the MSA res
42、earch, concentration of ethyl mercaptan in the vapor phase of a commercial supply of LP-gas was monitored between the temperature limits of -30 to 120 deg. F. The ethyl mercaptan concentration in the used tanks at no time fell below limits specified in NFPA 58. Following completion of the MSA Resear
43、ch work for CPSC, MSA was then contracted by industry to perform an additional study on two of the tanks to study ethyl mercaptan concentration levels in simulated actual use conditions. The new tank, which had exhibited complete odorant depletion in four weeks, was emptied slowly by burning off pro
44、pane vapor at about 2 gallonshr. to approximately 2-3% of fill. This was done to simulate a real life consumer use pattern. The tank was then refilled with odorized propane. The concentration of ethyl mercaptan in the vapor phase remained steady and within acceptable limits throughout a test period
45、of over three months, indicating that the tank walls had become passivated and no further reaction was occurring. and depressured to atmospheric pressure. The tank was allowed to breath for a period of 72 hours through piping and an orifice to simulate an out-of-gas situation as might be experienced
46、 in real life. At the end of this exposure, the air concentration in the tank was 20%. The tanks was then refilled and monitored for ethyl mercaptan in the propane vapors. There was an immediate and rapid drop in ethyl mercaptan concentration in the vapor phase, followed by a recovery of odorant lev
47、els in 4-6 days to normal levels. The minimum ethyl mercaptan concentration observed even at the lower levels was well within safe limits for detection in the event of a leak. Also, in 1988, under a contract with the U.S. Consumer One of the continuous service tanks was likewise emptied STUDY CATEGO
48、RY TWO The second category of odorant investigations deals with new steel containers or reinstalled used containers after prolonged exposure to the atmosphere. of ethyl mercaptan in two new 500 gallon tanks after the initial filling of the tanks. prior to filling one tank was purged as per recommend
49、ations in the National Propane Gas Association Bulletin #133 and the other was not purged. With the tanks remaining in a static condition, measurement of odorant concentrations taken over a period of over 2 years revealed that the rate of loss in the purged tank was slower than in the unpurged tank. It was also observed that the rate of diethyl disulfide formation was In 1985 Phillips Petroleum Company tested the stability equal to the rate of loss of ethyl mercaptan, indicating that an oxidation type reaction was causing the loss. cylinders, it was observed that for samples stored in
