1、Executive Summary Literature Review Impact of Gasoline Blended with Ethanol on the Long-Term Structural Integrity of Liquid Petroleum Storage Systems and Components JUNE 2002 EXECUTIVE SUMMARY Literature Review Impact of Gasoline Blended with Ethanol on the Long-Term Structural Integrity of Liquid P
2、etroleum Storage Systems and Components sponsored by American Petroleum Institute Washington, DC Fiberglass Tank and Pipe Institute Houston, Texas Steel Tank Institute Lake Zurich, Illinois Western States Petroleum Association Glendale, California prepared bySimpson Gumpertz however, the Institute m
3、akes no representation, warranty, or guarantee in connection with this publication and hereby 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. Su
4、ggested revisions are invited and should be submitted to API, Standards department, 1220 L Street, NW, Washington, DC 20005, standardsapi.org. Table of Contents CONTENTS Page 1. INTRODUCTION 1 2. LITERATURE REVIEW 1 3. FINDINGS.1 3.1 Overview of Ethanol in Gasoline Fuel.1 3.2 Phase Separation and Wa
5、ter Bottoms2 3.3 Fiberglass Reinforced Plastic (FRP) Tanks 4 3.4 Steel Tanks .6 3.5 Coatings/Linings8 3.6 Piping 9 3.7 Other Components10 4. CONCLUSIONS AND RECOMMENDATIONS.12 4.1 Summary of Findings 12 4.2 Research Needs .14 REFERENCES APPENDIX A Literature Review - 1 -1. INTRODUCTION This report s
6、ummarizes the results of a literature review conducted for the American Petroleum Institute on the impact of gasoline blended with ethanol on the long-term structural integrity of liquid petroleum storage systems and components. It is anticipated that the use of ethanol in motor fuels will continue
7、to increase. This has generated interest about the potential long-term structural effects of ethanol on liquid petroleum storage systems, including underground storage tanks (USTs), underground piping, and associated components. The objective of the literature review is to determine the state of ind
8、ustry knowledge and research on the effects of ethanol/gasoline blends on the long-term structural integrity of UST systems and components. This review is intended to assist decision-makers on further research requirements and needed changes or supplements to existing standards for underground stora
9、ge systems and components used for storing and dispensing gasoline blended with ethanol. 2. LITERATURE REVIEW Attached in Appendix A are the synopsis and bibliographic information for all articles reviewed for the project. The report is organized by article index numbers. Reference numbers cited in
10、this report refer to the article index number. 3. FINDINGS 3.1 Overview of Ethanol in Gasoline Fuel Ethanol is an alcohol produced through the fermentation of biomass, typically corn, although other sources may be used. Ethanol containing some water is called “hydrated”; ethanol that is further proc
11、essed to remove all water is called “anhydrous”. As a fuel oxygenate, ethanol is blended with gasoline to increase the oxygen content of the fuel. The resulting mixture is typically a blend of 10% ethanol and 90% gasoline, conventionally termed “gasohol,” although other mix proportions have been use
12、d. Ethanol is also used as a volume extender and octane enhancer in gasoline fuels. The term “ethanol fuel” refers to the use of ethanol as the primary - 2 -energy containing substance. Storage systems may be called upon to store any concentration of ethanol. Therefore, while focusing on gasohol, th
13、is literature search examines all ethanol blends. Furthermore, studies pertaining to higher concentrations of ethanol are relevant when examining effects of the lower layer of a phase-separated ethanol/gasoline blend. Methanol, an alcohol produced from natural gas, is also used as an oxygenate or fu
14、el substitute. It has been suggested that methanol studies may be indicative of worst case limits for ethanol compatibility; however, this has not been proven for all materials. Multiple sources indicate that methanol is much more aggressive than ethanol to materials used in fuel storage (49, 97, 13
15、0). Ethanol has been used as an oxygenate in the Midwest for over 20 years (239). Relatively few material compatibility problems have been reported with the use of 10% ethanol blends (56) and no recorded major leak or failure has been directly attributed to ethanol use (248). The most often reported
16、 problems are swelling, hardening, or minor leakage of elastomeric seals and o-rings (66). Notwithstanding this experience, there is interest to determine the long-term effects of exposure of storage system materials to ethanol/gasoline blends and to determine if there are legitimate concerns about
17、possible leaks in the system due to shrinkage and cracking of seals and o-rings in dispensing equipment in systems that are switched from ethanol service to non-ethanol service. 3.2 Phase Separation and Water Bottoms Water and gasoline are immiscible, and phase separation will result from almost any
18、 quantity of water in gasoline, separating into water and gasoline layers. The lower water phase is termed “water bottom” in fuel tanks. Unlike gasoline, alcohol is a polar molecule and will therefore mix readily with water, up to various levels of saturation depending on other soluble and suspended
19、 components in the mixture. Alcohol is also miscible with gasoline but has a higher affinity to water than to gasoline and will adhere preferentially to the water. If sufficient water is present, the ethanol and water will separate from the gasoline. This process is called phase separation and will
20、occur at approximately 0.5% water content for 10% ethanol / 90% gasoline blends (58, 106). The blend separates into an alcohol/water lower phase (water phase) and an upper phase consisting of gasoline with a slightly reduced alcohol concentration (hydrocarbon phase) (186). No study reports the exact
21、 composition of each phase, but one paper states that the - 3 -lower layer consists of 75% ethanol, cosolvents, and 25% water (124, 135). Phase separation is a concern because it creates an alcohol rich water bottom, which can increase the potential for localized corrosion of steel tank walls (50, 1
22、24). Higher corrosivity has been attributed to increased oxygen content and increased conductivity of the ethanol/water phase (50, 73, 135). No test data are available to quantify the change in the conductivity and associated corrosion rate in alcohol-rich water bottoms. Other effects include the di
23、ssolution of corrosion by-products by the alcohol (55, 56, 58), thus exposing the metal to continued corrosive activity, and the presence of other fuel components in the ethanol/water phases (such as acetic acid) that accelerate corrosion rates (161). Conversely, one study indicates that alcohol-ric
24、h water bottoms are no more corrosive than non-alcohol water bottoms (55). However, the study did not include the effects of impurities in the solution. The higher alcohol content in water bottoms may be locally detrimental to FRP tank walls that are qualified only for low alcohol blends; however, n
25、o such problems have been reported or evaluated in the literature. In order to reduce the likelihood of phase separation, efforts must be taken to avoid water infiltration into systems containing ethanol/gasoline blends (44). Tanks should be free of water prior to initial filling and all precautions
26、 should be taken to minimize the potential for water infiltration and condensation in the tank (58). Removal of the water layer is recommended by API (132). Manual and automatic methods are available to detect the presence of a water phase in buried tanks (177). One article suggests that the presenc
27、e of ethanol reduces the likelihood of water bottoms by maintaining the water in solution (295). The dissolved water is carried out of the system with fuel consumption. This mechanism relies on small quantities of water entry as required by current specifications such that the water quantity does no
28、t exceed the threshold for phase separation. Microbial growth in storage tanks can be fostered by water bottoms (177, 243). Microbial contamination can affect both certain metallic and non-metallic materials. Microbes, both bacterial and fungal, can enter a storage system through fuel transfers, ven
29、ts, and equipment and can grow wherever water is present, for example at the fuel/water interface in water-bottoms and on the upper walls and ceilings of tanks where condensation is prevalent. - 4 -Microbial growth can be controlled through cleaning and use of biocides (254, 243). Existing literatur
30、e suggests that ethanol may increase the occurrence of microbial attack, but provide no specifics on the extent of reported increase (243). 3.3 Fiberglass Reinforced Plastic (FRP) Tanks FRP tank laminates are composites of thermosetting resins and glass fibers. Some thermosetting resins (e.g. orthop
31、hthalics) do not perform well in alcohols and alcohol fuels (20, 201), and are not known to have been used for this service. Resins typically used in FRP tanks (e.g. isophthalics, terephthalics, and vinyl esters) exhibit some level of swelling (with weight gain) and softening when exposed to alcohol
32、s, including ethanol (20, 50, 56). Methanol produces more pronounced effects than ethanol (148), and ethanol/gasoline mixtures may be more aggressive than pure ethanol (20, 50). Softening can be measured as reduction in hardness and material stiffness, which can be measured by mechanical tests for B
33、arcol surface hardness and modulus of elasticity. These physical changes are a function of permeation, which can be measured by absorption and swelling tests, and are largely reversible if the laminate is allowed to dry (18, 22, 24). After desorption of alcohol, some permanent loss of mechanical pro
34、perties may occur, but the mechanism is not well defined (22). Reduced stiffness, along with creep, results in a lower resistance to buckling, a key design criterion for FRP tanks. This behavior may affect the adequacy of safety factors for buckling design of FRP tanks exposed to ethanol/gasoline bl
35、ends (50). The U.S. standard governing the manufacturer of FRP underground storage tanks (UST) is Underwriters Laboratories (UL) 1316 “Glass-Fiber-Reinforced Plastic Underground Storage Tanks for Petroleum Products, Alcohols, and Alcohol-Gasoline Mixtures” (107). The standard was first published in
36、1983. Since 1987 this standard has required long term (180 days) immersion tests on FRP tank laminates in ethanol and ethanol/gasoline blends for tanks intended for that service. Prior to the 1987 edition, the standard did not require testing for ethanol or other alcohol exposure. The current standa
37、rd does not require tanks intended for petroleum products only to be tested for alcohol immersion. For tanks intended for ethanol/fuel mixtures, after high-temperature, double-sided exposure of test specimens, the material must retain 50% of its short-term flexural strength and stiffness, and impact
38、 resistance when test results are extrapolated to 270 days (107). - 5 -UL 1316 does not examine creep, creep buckling or creep rupture. Creep buckling is a gradual increase in the amplitude of buckles with time, and is a function of both the stiffness of the FRP laminate and the resistance to deform
39、ation provided by the soil envelope around the tank. Currently, there are no published data on creep buckling of FRP underground tanks, and only limited data on creep rupture of FRP laminates (32). Major manufacturers of FRP USTs currently provide warranties for single-wall tanks used for storage of
40、 gasohol (10%ethanol/90%gasoline blends) and other blends up to 10% ethanol, and some manufacturers warrant double-wall tanks for higher ethanol concentrations (60, 61, 258, 259, 260, 261, 262). At least one tank manufacturer provided a similar gasohol warranty as early as 1981 (262), and has since
41、stated that pre-1981 tanks should perform equally as well as later tanks when used to store 10% ethanol blends (264, 109). Another manufacturer has stated that resin types have not changed since the inception of its tanks and that tanks were qualified for gasohol storage as early as 1980 (60). Howev
42、er, since ethanol immersion testing was not required by UL 1316 prior to 1987, the literature does not indicate how these earlier tank laminates were qualified by manufacturers for ethanol exposure. One recent state governmental advisory suggests that manufacturers began testing tank laminates for e
43、thanol exposure in 1984 and advises users to obtain compatibility performance information for tanks manufactured prior to this date before converting tanks to ethanol storage (52). Lack of documentation of industry experience has raised the issue about the susceptibility of FRP tanks to ethanol atta
44、ck (112) and led to the US Department of Energys recommendation to install a chemical-grade rubber lining in FRP tanks prior to ethanol fuel use (130). However, the Department of Energys basis for this requirement did not examine specific resins and therefore is too broad to apply to all FRP tanks.
45、Tests conducted in 1992 on certain resin types found that isophthalic resins tested did not meet UL 1316 requirements when immersed in 30% ethanol blends (24). A 1986 study, showed, based on retention of short-term strength and stiffness properties, that “older” tank laminates perform adequately in
46、10% ethanol blends, but not as well in 20% ethanol blends. “Newer” tanks were found to perform adequately in all ethanol blends (148). It is believed that most FRP tanks will perform satisfactorily for storage of gasohol (10% ethanol) (56). Field experience has indicated no adverse effects from gaso
47、hol storage (64, 248). However, few analytical or experimental data have been published to show how reduced properties due to ethanol exposure (e.g. softening and stiffness reduction) relate to long-term performance under - 6 -sustained (creep) loads such as caused by soil and hydrostatic pressure o
48、n underground tanks. Stress corrosion cracking (degradation of strength of glass fiber reinforced composites under sustained stress) has been studied in acidic aqueous environments, but no results have been reported for ethanol blended fuels (37). Creep rupture tests on FRP with E-glass fibers subje
49、cted to water indicated a rupture strength after 10,000 hrs. of sustained stress of about 30 to 50% of the short term strength (26). One study showed that ethanol exposure was less severe than water exposure in contributing to creep rupture failure of FRP piping (173). Note that FRP tanks containing ethanol/fuel mixture are exposed to water or brine on their exterior surface, depending on whether they are single or double walled tanks. Properties of FRP laminates change through permeation and swelling and accompanied by a slower process of resin relaxation (38). Caution must be us
