Newsletter Issue 19

Chair Emeritus' Column

Storage Stability of Reformulated Gasolines (RFG)

Leo L. Stavinoha
San Antonio, TX

As discussed in Newsletter #18, the U. S. Department of Defense's continued use of some gasoline-consuming military materiel/equipment, has required prepositioning of the newer reformulated gasolines and has prompted a Defense Fuel Supply Center (DFSC) sponsored investigation to assess the storage stability of these oxygenated fuels. For DFSC's procurement of RFG, as defined in procurement clause C16.18-1, a storage period of up to four years after time of acceptance is anticipated, with an average storage temperature assumption of approximately 20EC. For this initial procurement, the oxidation stability of the RFG using the standard induction period method (ASTM D 525) has been tentatively specified at a minimum of 480 minutes. Additionally, use of oxidation inhibitors over the range of 5 to 15 pounds per 1,000 barrels (PTB) of gasoline, and an approved metal deactivator at 1 to 3 PTB of gasoline are being required. An approved corrosion inhibitor may be added but is not required. While these additives have been previously developed and used for many years in conventional gasoline, their effectiveness in RFG has not been established.

This investigation makes use of previous data developed by the U. S. Army to demonstrate utility of using a 6-hr test - ASTM D 873 AOxidation Stability of Aviation Fuels (Potential Residue Method) - as recommended for procurement of motor gasoline storable for four years in the NATO distribution system. The washed gum limit was set at 5 mg/100mL as a maximum for the D 873 6-hr test. This limit has been retained in the DFSC work; however, as the storage conditions are at higher temperatures than the NATO underground storage tanks, an 8-hr test time-period is being utilized.

The ASTM D 4814, "Standard Specification for Automotive Spark-Ignition Engine Fuel", provides a summary of U. S. Environmental Protection Agency (EPA) regulations controlling fuel composition. In their "Regulation of Fuels and Fuel Additives; Standards for Reformulated and Conventional Gasoline, Final Rule", published in the 16 February 1994 Federal Register, it is mandated that RFG must meet three compositional requirements: 2.0 mass % minimum oxygen, 1.0 vol% maximum benzene, and no heavy metals (such as lead or manganese). For fuels containing aliphatic ethers and/or alcohols (excluding methanol), the maximum oxygen content allowed is 2.7 mass % under the "Substantially Similar Rule", as summarized in Appendix X3 of ASTM D 4814. However, for fuels intended for long-term storage, based upon prior U.S. Army experience with GASOHOL, it has been recommended that only ether oxygenates be allowed, as the alcohols are more sensitive to moisture resulting in phase separation with the more dense water-alcohol mixture separating to the tank bottom. This is a serious matter when the alcohol represents as much as 10 % of the fuel, and the alcohol-water phase is present in a very poor-performing gasoline. Ethers such as methyl tertiary-butyl ether (MTBE), tertiary-amyl methyl ether (TAME), and ethyl tertiary-butyl ether (ETBE) do not undergo phase separation in the presence of excess water.

In the Clean Air Act Amendments of 1990, Congress specified that, beginning January 1995, all gasoline sold to the ultimate consumer in the U. S. must contain additives to prevent the accumulation of deposits in motor vehicle engines and fuel systems. For RFG procured for long-term storage, these detergent additives (i.e., deposit control additives) make predictive stability testing extremely difficult. Predicting the stability of gasolines fully formulated with detergents requires development of a new bench testing protocol which is not yet available.

A two-phase laboratory program to investigate the storage stability characteristics of both representative RFG samples being procured by DFSC and special laboratory formulated blends (containing either MTBE, TAME, or ETBE) to enable the optimum antioxidant and metal deactivator combinations to be determined for various ether-type oxygenates, has been completed. The objective of this program was to investigate (1) the storage stability characteristics of representative RFG samples being procured by DFSC, and special formulated blends to enable the optimum antioxidant and metal deactivator combinations to be determined for various ether-type oxygenates, and (2) the ultimate useability of the RFG's when "stored" outside of the continental U. S.

This project was accomplished in two phases. Phase 1 covered the storage stability assessment of DFSC-supplied RFG, both with additive package outlined in DFSC procurement clause C16.18-1 and without additive package. The U. S. Army TARDEC Fuels and Lubricants Research Facility (SwRI) formulated three RFG blends using a moderately stable gasoline blending stock, obtained by the use of a reference fuel ("J" fuel) which was made unstable by the addition of dimethylhexadiene (DMHD). The ethers (MTBE, TAME, and ethyl tertiary-butyl ether (ETBE)) were used at volume percents to provide the oxygen content of 2.7 mass %. Alcohols were not investigated as they are not recommended for inclusion in fuels intended for long-term storage due to moisture sensitivity.

DFSC-supplied RFG (procured for long-term storage) with and without additive package were evaluated for stability characteristics using test methods ASTM D 525 (Induction Test Method) and ASTM D 873 (Accelerated Gum Test Method). Similarly, SwRI-formulated RFG were evaluated using the following additives:

• Antioxidants (required by Clause 16.18-1)
• One phenylenediamine
• One hindered phenol
• 50/50 blend of above antioxidant additives
• Metal deactivator (required by Clause 16.18-1), both of the two approved
• Corrosion inhibitor (not mandatory)
• Detergent (mandatory for ultimate distribution of RFG but not required by clause 16.18-1 for long-term storage).

Phase I analyses support the suggestion that the additives which were evaluated were not antagonistically affected by the presence of any of the three ethers.

Fuel samples were formulated with varying concentrations of antioxidant. Using ASTM D 525 and ASTM D 873 8-hr gum data the following observations were made:

• The hindered phenol, and the phenylenediamine and a mixture of the two are essentially equal in stabilizing unstable "J" fuel (containing DMHD), using antioxidant treat levels of both 5 and 15 PTB of fuel.
• In the range of 10 to 30 % in unstable "J" fuel, MTBE tends to increase the induction period while hexane decreases the induction period, indicating that the ether can improve stability.
• The hindered phenol, and the phenylenediamine and a mixture of the two are essentially equal in stabilizing unstable J-RFG containing MTBE ("J" fuel with DMHD and ether to make unstable reformulated gasoline). These effects were noted for both induction periods and potential gum of unstable J-RFG.
• Both the hindered phenol and the phenylenediamine, as well as a mixture of the two are essentially equal in stabilizing unstable J-RFG containing ETBE, except that the hindered phenol appears more potent at the lower treatment concentration. The D 873 480-minute potential gum remained above 5 mg/100mL for the maximum treatment level which gave induction periods slightly above 500 minutes. These data also demonstrate that a minimum induction period of 480 minutes is not a guarantee of a low potential gum at 480 minutes. Similar data and observations were observed for unstable J-RFG (TAME)
• The hindered phenol, and the phenylenediamine and a mixture of the two are essentially equal in stabilizing unstable J-RFG containing TAME

The effectiveness of corrosion inhibitor was measured using ASTM D 130 "Detection of Copper Corrosion from Petroleum Products by the Copper Strip Tarnish Test" and ASTM D 665 "Rust-Preventing Characteristics of Inhibited Mineral Oil in the Presence of Water." The presence of 15 vol% MTBE, 15 vol% TAME, or 17 vol% ETBE in unstable J-RFG had no negative effect on the ability of corrosion inhibitor to prevent rust or copper corrosion. The effectiveness of the two metal deactivator additives (MD#2 and MD#75) were found to be effective in neat reference fuel, J-RFG, and DFSC-RFG when contaminated with copper and were not adversely affected by the presence of detergent (when added to test samples). Zinc (organically compounded) was found to have no effect on instability even at 2.4 mg/L or in the presence of 15 vol% MTBE in reference fuel. This was also substantiated by D 873 8-hr potential gum. Some storage tanks still have zinc rich protective coatings which are particularly deleterious to diesel fuel quality (mainly increased microparticulate levels).

As all RFG formulations require a deposit control additive if distributed to the ultimate fuel user, a limited number of samples were also made with two different types of detergents including the DFSC selected detergent, to determine effects on stability. The DFSC detergent was ineffective in reducing the induction period of J-RFG (MTBE), metal deactivator stabilized copper contaminated (0.2 mg/L) 'J" fuel, and unstable "J" fuel. The results with DFSC detergent in the fuel were not completely conclusive in that accelerated testing did not always produce low washed gums. It seems best to recommend against accelerated testing of marginally stable fuels containing detergent, if possible, and evaluating addition of detergent to either stored fuel or accelerated aged fuel for determining efficacy. This was addressed in phase II of this project.

In Phase II, the useability of RFG'S exposed to storage aboard military prepositioned ships (MPS), was to be addressed. As no RFG's have been previously stored in MPS, the first DFSC-supplied RFG (Phase I) and the base fuel from Phase I (limited to MTBE as the oxygenate) were used in this phase. Testing included gum and ISD (Intake System Deposits) type testing to identify usability.

In general, as a fuel ages, it develops higher intake valve deposit (IVD) depositing capabilities which are measured indirectly by ASTM D 381 washed gum values and FTM 791C, Method 500.1 ISD appearance and mass values.. When commercial deposit control additive was added to base fuels in this program, they gave relatively low D 381 washed gum and were somewhat ineffective at 80 PTB for fuels which were probably dirtier than the reference fuel used to obtain the initial EPA qualification for this additive.

Data was developed to establish the detergent quality and relative response in aged fuel (added both prior to ageing and after ageing). Testing to confirm adequacy of DFSC RFG's detergency requirement, for use in the continental U. S., suggested that use of DFSC detergent may require higher treat rates than the minimum EPA effective treat rate. In practice, the treat rate should be determined by D 381 testing for washed gum and ISD testing (both visual and mass of deposit) with neat and deposit control additive treated fuel.

Based on the data and discussions developed in this project, the following recommendations are made for use of additives in DFSC-RFG:

• Do not require the presence of deposit modifier in gasoline for long-term storage. Actually, it is recommend that procurement clause C16.18-1 state that the gasoline not contain deposit control additive. Addition of DFSC detergent is recommended at twice the minimum effective treatment during final distribution for use in the continental U. S., and possibly outside the continental U. S., if the fuel deteriorates sufficiently to warrant its use.
• Maintain D 525 limit at 480 minutes, minimum in procurement clause. Consider addition of D 873, 8-hr limit of 5 mg/100 mL increase in washed gum, maximum, to procurement clause.
• Add phenylenediamine or hindered phenol at the maximum treat rates.
• For stability testing, provide 1 each 1-quart samples (95 vol% full) of gasoline from shore tank and delivery nozzle/pipeline (after additive addition point) or from barge or from MPS tank.