1、GPA TP-21 9b 3824699 0038402 TbL Technical Publication TP-21 Trace Contaminants in Natural Gas Liquids GPA Technical Section C July 24, 1996 6526 East 60th Street Tulsa, Oklahoma 74145 Phone: 918/493-3872 Fax: 9181493-3875 GPA TP-21 Yb 3824699 OOI18403 T8 D Disclaimer GPA publications necessarily ad
2、dress problems of a general nature and may be used by anyone desiring to do so. Every effort has been made by GPA to assure accuracy and reliability of the information contained in its publications. With respect to particular circumstances, local, state, and federal laws and regulations should be re
3、viewed. It is not the intent of GPA to assume the duties of employers, manufacturers, or suppliers to warn and properly train employees, or others exposed, concerning health and safety risks or precautions. GPA makes no representation, warranty, or guarantee in connection with this publication and h
4、ereby expressly disclaims any liability or responsibility for loss or damage resulting from its use or for the violation of any federal, state, or municipal regulation with which this publication may conflict, or for any infringement of letters of patent regarding apparatus, equipment, or method so
5、covered Foreword This GPA Technical Publication was drafted in response to a recommendation from the Trace Contaminants Workgroup composed of members of GPA Technical Section C, Specifications, and GPA Technical Section B, Analysis and Test Methods. It is intended to summarize and bring to closure t
6、he work that GPA has conducted relative to trace contaminants in NGLs. This recommendation was accepted by GPA Technical Section C and the GPA Technical Committee membership. Kevin Miller, Chairman Technical Section C Bri, Don Todd, Chairman Technical Committee Background Trace contaminants as discu
7、ssed below are those contaminants that can be present in natural gas liquids (NGLs) as a natural occurrence or from processing. Most contaminants, e.g. H2S, total sulfur, and moisture that affect the orderly commerce of NGLs are covered in current GPA Standard 2140 - Liquefied Petroleum Gas Specific
8、ations and Test Methods, but others are not addressed. These other contaminants are not normally considered troublesome in product used as residential or industrial fuel and gasoline blending, but can create problems when encountered in further processing operations, e.g. ethylene manufacture and bu
9、tane isomerization. This issue has been a topic of discussion for some time at meetings of GPA Technical Section C. Regarding petrochemical operations, this subject was first discussed in Section C at a 1985 meeting when a member company reported catalyst problems possibly caused by methanol in etha
10、ne feedstock. After considerable discussion as to whether or not a GPA ethane specification should be developed, Section C conducted an opinion letter survey of its members to determine if there was sufficient reason to develop specifications for a merchant ethane product. The majority of respondent
11、s to the survey were opposed to GPA developing an ethane specification because the membership felt this issue was best addressed between the two contracting parties for the product. A historical point should be made here. Throughout GPAs history of developing NGL specifications, the Association has
12、always adhered to an early directive from the GPA Board of Directors to develop specifications for LP-gas “. broad enough to cover all marketable products and restrictive enough to provide protection to the consumer”. Section C became interested in the issue of fluorides in LP-gas in the spring of 1
13、986. Four contamination incidents involving both HF and organic fluorides had been reported the previous year. Most of the incidents were attributed to upsets in refinery HF alkylation units. General discussion indicated that there was an apparent lack of knowledge concerning the maximum safe level
14、of organic fluorides and lack of data on the products of combustion from burning LP-gas containing fluorides. At the Fall 1986 meeting, Section C decided to revise GPA Standard 2140. The Foreword was to be revised to emphasize the fact that GPA 2140 does not provide specifications for various contam
15、inants not normally found in LP-gas extracted from natural gas. The committee also decided to conduct preliminary research to determine if there were any existing standards for fluorides and other contaminants, and to determine if there were existing test methods for these contaminants. In the Sprin
16、g 1987 meeting, Section C requested assistance from Section B (Analysis and Test Methods) to develop test methods for determination of fluorides and methanol in LP-gas. At this meeting Section C voted to recommend to the Technical Committee that research be conducted to establish a maximum fluoride
17、limit in LP-gas that corresponds to maximum exposure limits set by Federal regulation. The Technical Committee requested a written scope of work and cost estimates for doing the research. Section C established a workgroup to develop the scope at its next meeting. Work was also initiated at this time
18、 to develop a scope of work for a research project addressing possible specification limits for ammonia as a contaminant in LP-gas. Over the last few years, there had been several reports of ammonia contamination incidents which were allegedly caused by inadequate cleaning of transport truck tanks a
19、nd rail cars utilized in both propane and ammonia service. A scope of work was developed but the ammonia research project (GPA Research Project No. 918) did not receive funding by GPA membership. No work was actually performed so this activity will not be discussed further. In the Fall of 1989, Sect
20、ion C received a letter from a member company requesting a butylene specification in butane product because of catalyst life reduction in isomerization processes. The Committee voted to take no action on this issue. Also at this meeting, the scope of the fluorides research project was reviewed. The
21、preliminary scope of the project included: Determine the products of combustion of fluoride-contaminated LP-gas. 1 GPA TP-21 96 m 3824699 OOL8405 770 Recommend field test procedures and laboratory procedures for detecting fluorides (both HF and organic fluorides) in LP-gas in the ppm range. Determin
22、e the accuracy of the above field and lab procedures. Determine the maximum level of contamination permissible considering effects on materials, human safety and Federal and state regulations. Requests for proposals for this work were distributed in November 1989 to 25 research organizations. Four p
23、roposals were received. Three of the proposals exceeded $100,000; the fourth excluded the experimental combustion tests. Schedules for completing the work were estimated at one year. The proposals were discussed at Section C and Technical Committee meetings in March 1990. In view of the cutbacks tha
24、t had recently occurred in the GPA Research program, the Technical Committee felt it was unlikely that a new single-year project of this magnitude would be approved by the membership. An ad hoc workgroup was established to further review the proposals and to pursue clarifications and cost savings. A
25、 recommended plan of action for 1991 was then balloted in Section C. A reduced project (GPA Research Project No. 919) costing $40,000 was approved by GPA membership in October 1990. MSA Research Corporation in Pittsburgh, PA was selected to do the work in the Spring of 1991. MSAs final report on thi
26、s project was submitted in early 1994. The major results can be summarized as follows: Laboratory procedures exist for the determination of organic fluorides and HF but these require extensive equipment and skilled personnel, Results are usually reported as total fluorides. No simple field test evol
27、ved from the study. Propane contaminated with 47 ppmw isopropyl fluoride was burned in an experiment. This resulted in a concentration of 2 mg/m3 hydrogen fluoride in the combustion gas. This would be further diluted by room air. The threshold limit value set by the American Conference of Government
28、al Industrial Hygienists is 2.5 mg/rn3. The final report on this study was published as GPA Research Report No. 146. In a letter to GPA dated August 13, 1991 (Attachment i), a representative of the American Institute of Chemical Engineers (AIChE) Ethylene Producers Committee addressed concerns of th
29、at organization with feedstock quality. A subcommittee had been established to deal specifically with feedstock quality issues. There had been increasing emphasis on product quality assurance within the olefins manufacturing industry as a result of demands from their customers, primarily automobile
30、manufacturers. Other points noted in the letter included: Suppliers of NGLs will need to participate in the quality process. Transportation of NGLs in common carrier pipelines and commingling in underground storage can lead to uncontrolled variations in NGL quality and thus cause unanticipated chang
31、es in olefin plant operations. Widely accepted pipeline and storage specifications are”nominal” at best. Wide variations and departures from these specifications have been accepted without a great deal of mutual agreement between the NGL supplier and olefins manufacturer. The AIChE committee identif
32、ied the following NGL contaminants that needed to be reviewed: methanol, carbon dioxide, mercury, fluorides, chlorides, NOx, sulfides, sodium and radon. The AIChE letter was discussed at the October 3, 1991 Section C meeting. Several of those present volunteered to meet with the AIChE group. An AICh
33、E representative would be invited to attend the March 1992 Section C meeting to discuss feedstock quality issues as they affect the petrochemical producer. On October 31, 1991 two members of Section C and a member of GPA staff attended a meeting of the AIChE Ethylene Producers Committee in Kingwood,
34、 Texas. Information was exchanged regarding the GPAs historic role in developing product specifications. The issues of trace contaminants and possible petrochemical grade specifications were again discussed at the March 1992 Section C and Technical Committee meetings. The Technical Committee request
35、ed Section C to continue communication with the AIChE group and to obtain information on current pipeline NGL and product specifications. Section C was also directed to review gas plant NGL quality capabilities, existing 2 GPA TP-21 96 W 3824b%9 0038406 bD7 pipeline specifications, and olefin manufa
36、cturers requirements to determine if there was a real need for GPA to develop new specifications on trace contaminants. The results of the review would determine what action, if any, would be taken. A relevant paper presented at the 1992 Annual Convention (Attachment 2) discussed the need for more s
37、tringent specifications for is0 and normal butane. At the Fall 1992 Section C meeting, the activities of the AIChE Ethylene Producers Subcommittee were again discussed. This group had surveyed their member companies to evaluate the level of concern and consequences (on equipment, catalyst life, safe
38、ty, environment etc.) for various feedstock contaminants including methanol, chlorides, mercury, NOx, arsine, C02, sulfur, sodium, radon, fluorides, water, phosphorus, ammonia, etc. The AIChE survey also addressed analytical methods used. Section C decided to conduct a survey of major pipeline compa
39、nies from different regions of the country to analyze raw NGL feed streams for trace contaminants. A comparison of existing pipeline product specifications was presented and discussed. There were significant differences in the various companies specifications. The AIChE Ethylene group released their
40、 survey results in December 1992. The results were also presented at the Ethylene Plant Safety Symposium at the Spring 1993 AIChE meeting in Houston (Attachment 3). The trace contaminants workgroup consisting of members of Sections B and C and GPA staff met in Tulsa on January 14, 1993 to formulate
41、the methodology for the GPA survey. The survey should provide credible data that could be used to respond to the AIChE Ethylene Producers Subcommittee and GPA members when discussing the need for more stringent specifications for LP-gas. It was decided to analyze NGL product from two pipeline system
42、s prior to the March meeting. An extended hydrocarbon analysis plus a “wish list” of trace contaminants would be determined. The two samples were obtained and analyzed during February 1993. Results of the two pipeline sample analyses (Attachment 4) were discussed at the March 1993 Section C meeting.
43、 Although generally low in trace contaminants, detectable levels of such contaminants as mercury, arsenic, methanol and organic fluorides were found in the samples and this indicated the need for additional follow-up work. A workgroup from Section C was formed to develop a surveyhest program to dete
44、rmine concentration levels of a limited number of contaminants in both NGL and fractionated products. Several major fractionators and pipeline companies agreed to participate. Section B was requested to recommend a laboratory to perform the analyses. Another paper calling for tighter butane specs (A
45、ttachment 5) was presented at the 1993 GPA Annual meeting. During this time period a problem was reported at Mont Belvieu and other locations with butane contaminated with fluorides. Much of the n-butane delivered to Mont Belvieu was isomerized to produce isobutane which was then used as a feedstock
46、 in the manufacture of methyl tertiary butyl ether (MTBE) a major constituent in oxygenated motor fuel. It had been reported that fluorides contamination can poison catalysts used in these processes. It became common practice to segregate contaminated and non-contaminated butane in the Mont Belvieu
47、storage wells. To address the fluorides issue, a special meeting of Section C was held on April 23, 1993 at Mont Belvieu. Present were 11 members of Section C plus 22 guests which included a number of members of Section B, representatives of pipeline companies, consumers of butane, and personnel fro
48、m testing labs. Background of the problem and the context within the GPA trace contaminants project were discussed. A member company presented results of their ongoing butane analytical testing program. Test methods and possible sources of contamination were discussed. The pros and cons of more stri
49、ngent product specifications were discussed. It is noteworthy that since the Section C meeting in Mont Belvieu, reported butanes containing fluoride contamination have been greatly reduced. Based on the meeting at Mont Belvieu and feedback from the ethylene producers it was decided the GPA contaminant survey should focus on a fairly short list of contaminants limited to mercury, arsenic, olefins, fluorides and oxygenates (including methanol, isopropyl alcohol, etc.). A trace contaminants survey soliciting participation from member companies was distributed through Section C and the Tech
