1、 STD-APIIPETRO PUBL DR53-ENGL L99b 0732290 05b4035 9LiO 9 -* TN American Petroleum Institute Characterization of Exploration and Production Associated Waste Health and Environmental Sciences Department Publication Number DR53 November 1996 STD.API/PETRO PUBL DR53-ENGL L79b 0732270 05b403b 887 = One
2、of ithie mmsi significant long-term trends affecting the future vitality of the petroleum industry is the publics about the environment. Recognizing this trend, API member companies have developed a positiva, mwmkboking strategy called STEP: Strategies for Todays Environmental Partnership. This ,pro
3、gram aims a0 address public concerns by improving our industrys environmental, health and safety perfomantx% documenting performance improvements; and communicating them to the public. The faindatbn d STEP W the API Environmental Mission and Guiding Environmental Principles. .W#RONMENTAL MISSION AND
4、 GUIDING ENVIRONMENTAL PRINCIPLES The mmlms af the American Petroleum Institute are dedicated to continuous efforts to improve the compatiMQ d ou operations with the environment while economically developing energy resources and supplying tqjh quality products and services to consumers. The members
5、recognize the importance of efficientiy mmhg societys needs and our responsibility to work with the public, the government, and others Eo $emlap and to use natural resources in an environmentally sound manner while protecting the health and sabty of our employees and the public. To meet these respon
6、sibilities, API members pledge to manage our ves according to these principles: To remw and to respond to community concerns about our raw materials, products and + To operate our plants and facilities, and to handle our raw materials and products in a manner that pmtects the environment, and the sa
7、fety and health of our employees and the public. + To make safety, health and environmental considerations a priority in our planning, and our d 3. 4. Recommendations for future studies. An initial constituent database for associated wastes; An assessment of potential impacts to groundwater posed by
8、 land-managed associated wastes; and SAMPLING AND ANALYSIS Sample collection and analysis were conducted in two phases. In Phase I (1989), 31 samples were collected and analyzed for a comprehensive list of organic and inorganic constituents. During Phase II (1990-1991), 89 additional samples were co
9、llected and analyzed for a more narrowly focused set of constituents and characteristics. In all, samples representing 12 different associated waste categories were collected from on-shore E however, under existing EPA regulations, used oil may be reintroduced into the crude stream for recycling if
10、the used oils are from normal operations and are to be refined with normal process streams at a petroleum refinery facility (see 40 CFR Section 279). A conservative approach was taken when collecting samples. A conscious effort was made to sample materials in a manner to capture the highest concentr
11、ations of constituents of potential environmental concern. Materials sampled ranged from freshly contaminated soil to a host of potential wastes from various process streams. Care was taken to address all significant wastes and potential wastes, obtain representative samples, and employ appropriate
12、quality assurance/quality control (QNQC). Some of the sampling difficulties encountered could be minimized in future efforts by following an established plan for associated waste sampling. Many associated waste samples contained percent levels of oil and parts per million (ppm) levels of volatile or
13、ganic compounds (benzene, toluene, ethyl benzene, and xylenes; “BTEX). A few samples were found to contain ppm levels of the semi-volatile compounds 1-methyl naphthalene and phenanthrene. A number of metals were detected: ppm levels of lead, ES-2 STD-API/PETRO PUBL DR53-ENGL L77b 0732290 0564049 435
14、 = chromium, copper, nickel, vanadium and zinc were found in many samples. Calcium, sodium, and potassium were found along with barium, a common drilling fluid additive. Since BTEX and semi-volatile compounds are naturally occurring constituents of crude oil and natural gas liquids, and the metals d
15、etected are ubiquitous in the environment, these results are not unexpected. Therefore, the transport and fate of these constituents in various media, such as soil and groundwater, must be evaluated before any significance can be placed upon the magnitude of the concentrations found. This study reve
16、aled several practical problems with the sampling and analysis of associated wastes. The two primary, and often related, sampling problems were: 1) obtaining a representative sample, and 2) scheduling the sampling event. For example, sampling methods must be carefully selected to obtain samples that
17、 are representative of much larger volumes of generated materials that are typically quite heterogeneous. Care must be taken to schedule sampling so that a true waste can be captured during an actual maintenance procedure (e.g., cleaning out a storage tank or removing waste glycol from a gas plant).
18、 The infrequency of certain maintenance events sometimes necessitated the sampling of materials which were still part of the process stream and would not normally be considered wastes. Many of the samples caused severe matrix interference problems with the EPA SW-846 methods used in this study. Matr
19、ix interference issues have been previously addressed in SW-846 and in comments on SW-846 in regard to petroleum matrices (USEPA, 1986). Low concentrations of organic constituents within an organic matrix would not have been detected, if present. These findings clearly show that associated wastes, e
20、specially those containing high levels of organic materials, require specialized analytical methods. This study generated a useful set of analytical data to serve as an initial, but limited, database describing the characteristics of associated wastes. When comparing the data in this study with data
21、 in future studies, the data quality elements of precision and accuracy should be evaluated. The RCRA Characteristics data collected in this study should be compared to other predictive tools (.e., alternative leaching protocols and fate and transport modeling) to determine the validity of continuin
22、g to use the EPA methods, such as the Toxicity Characteristic Leaching Procedure (TCLP), for associated wastes. ES-3 STD*API/PETRO PUBL DR53-ENGL 199b 0732290 05b11050 157 FATE AND TRANSPORT MODELING The composition and constituent concentration data generated in the sampling and analysis phase of t
23、his study were used to assess the potential impact on groundwater posed by land- managed associated waste. These data were entered into the API-developed Vadose and Saturated Zone Exposure (VADSAT) model which simulates the fate and transport of constituents from a land-management unit to a user-des
24、ignated drinking water well, directly downgradient. VADSAT accounts for most of the major processes that affect organic constituents moving through the shallow subsurface including: adsorption, dilution and biodegradation. Industry practices of land spreading, road spreading, and burial were modeled
25、 with VADSAT. The modeling investigation considered six associated waste types, four chemical species, two infiltration scenarios, eleven hydrogeological environments, and two hypothetical receptor locations. A total of 1,144 VADSAT Monte Carlo simulations were performed, each simulation involving i
26、teration of 1,000 sets of parameter values. This analysis produced a statistical distribution of possible receptor well concentrations for a wide range of hydrogeologic conditions. Associated waste management scenarios were converted to input understood by VADSAT using data from a range of sources.
27、Physical and chemical properties data not available from the sampling and analysis portion of this study were obtained from reference works. Hydrogeological settings were described by statistics compiled by API (Newell et al., 1989). Representative volumes of associated waste managed per disposal in
28、cident were compiled from information provided by API member companies. SUMMARY OF RESULTS AND FINDINGS 1. An initial constituent database for associated wastes was established. The data presented throughout this report indicate that the sampled associated wastes contain few Petroleum Refinery List
29、semi-volatile organic compounds, varying types and concentrations of metals, and a number of volatile organic compounds (VOCs), primarily benzene, toluene, ethylbenzene, and xylene (BTEX). These analytical results were not unexpected because BTEX and semi-volatile compounds are naturally occurring c
30、onstituents of crude oil and natural gas, and the metals detected are ubiquitous in the environment. ES-4 STD.API/PETRO PUBL DR53-ENGL L79b 0732270 05b4051 O73 W 2. Samplinq and analytical difficulties were identified. Proper schedulinq of sampling is a kev component. Due to scheduling problems, it
31、was often extremely difficult to collect actual waste samples. All of the categories sampled, except pisump waste, crude oil impacted soil, and dehydration condensate water, are generated intermittently from specific operations such as workovers or tank cleaning. These operations or maintenance proc
32、edures are infrequent and are usually scheduled only a few days in advance. They are subject to cancellation due to higher priority work, making it difficult for a sampling team to be present when an actual waste is generated. For some samples, such as tank bottoms and waste glycol, process fluids w
33、ere the bulk of the sample collected since true wastes were not available. In all cases, the collected sample was expected to contain equivalent or higher concentrations of constituents of possible environmental concern than contained in a true waste. Obtaininq representative samples was diff icult.
34、 Adding to scheduling problems is the extreme difficulty of obtaining representative samples. The sampling team collected samples of up to one liter from waste volumes ranging from one barrel (208 liters) to a maximum of 14,000 barrels. The materials were sometimes stratified and, in some cases, the
35、 composition changed from hour to hour. It is very difficult to obtain a representative sample from large volume heterogeneous materials. The EPA analytical methods were ineffective with many of the samples which contained hicih levels of orqanic constituents. Matrix interferences frequently interfe
36、red with test results. Matrix interference problems were frequently encountered when trying to analyze certain samples by EPA analytical methods (e.g., the TCLP method). Matrix interference involves problems created from substances in the samples that cause either a chemical or physical interference
37、 during the analysis of the sample. Approximately 60 percent of the samples in this study indicated a matrix interference problem with at least one constituent analyzed. The large concentrations of n-alkanes can mask the presence of other hydrocarbons and raise the detection limits for compounds of
38、interest. Comparison of analytical results from two different laboratories was limited due to the number of “non-detect results. Where positive analytical results could be compared, the agreement was limited. The analytical results for four duplicate samples collected by API and the Gas Research Ins
39、titute (GRI) were comparable for those analyses which did not experience matrix interference problems. API and GRI agreed to collect and analyze four split samples to better understand the variance in analytical results from two different labs, ENSECO and ENSR. The four samples used for this compari
40、son were molecular sieve from a dehydrator, spent molecular sieve from an isobutane sweetener, waste glycol, and glycol dehydrator condensate water. Whereas some differences were encountered in sulfide measurements, the TCLP constituent data were similar. Matrix interference problems created high ES
41、-5 STD-API/PETRO PUBL DR53-ENGL L77b H 0732290 05b4052 TZT 3. detection limits which made it difficult to compare the majority of the TCLP volatile organics data. E general chemical constituents such as pH, reactive sulfide, oil/water/solids, oil and greasehotal petroleum hydrocarbons (TPH) and chlo
42、ride. Other site specific constituents known to be present should also be analyzed. - ES-7 STD-API/PETRO PUBL DR53-ENGL L99b 6 0732270 0561i054 AT2 6 Section 1 BACKGROUND The oil and gas industry generates a number of wastes that are uniquely associated with its operations. These wastes include prod
43、uced water, drilling wastes, and “associated wastes.“ Associated wastes, which include crude oil impacted soil, tank bottoms, and workover fluids, comprise approximately 0.1 percent of the total volume of E therefore, sampling was accomplished by skimming emulsion and hydrocarbon from the surface an
44、d/or if solids were present, thieving solids from the bottom of the vessel. Rici Wash - Samples were collected in the sump to ensure the highest expected level of constituents of potential environmental concern would be present. Tank Bottoms - Tank bottom samples were thieved from tanks still in ser
45、vice. Samples were collected in this manner for two reasons: to ensure that the highest expected level of constituents of potential environmental concern would be present and, principally, because it was very difficult to schedule tank cleanouts and sample collection to coincide. Refer to the Tank B
46、ottoms Sampling Section of this report for more detail. cooler piping of engines in service. holding tanks where the waste was held prior to disposal. Used Enqine Lube Oil - Samples were taken directly from the engine crankcase or engine oil Workover Wastes - Samples were taken at the wellhead throu
47、gh valves, or from pits or ANALYTICAL METHODS This section discusses the general analytical test procedures used, and their limitations for E&P waste characterization. All analyses followed standard EPA methodologies and protocols and full quality assurance/quality control (QNQC) procedures. Appendi
48、x B contains additional details on the analytical methods utilized, QNQC, and constituents analyzed. During Phase I, conducted in 1989, it was requested that all samples be analyzed for oil/water/solids MODT: modified oven drying technique (API, 1987) Appendix IX volatile organic compounds, Petroleu
49、m Refinery List semi-volatile organic compounds, Appendix IX metals and RCRA Characteristics - ignitability, corrosivity and reactivity, TCLP volatile organic compounds, TCLP semi-volatile organic compounds, and TCLP metals (USEPA, 1986). (See Appendix B.) The Petroleum Refinery List, a subset of the Appendix IX list, contains compounds of concern to the petroleum industry, such as PNAs and phenols. To improve the reliability of the 2-6 STD.API/PETRO PUBL DR53-ENGL 177b 0732290 05b40b2 979 9 analysis of these compounds, cleanup procedures (Methods 3650 and 361 1)
