Executive SummaryIn support of the U.S. Department of Energy Fuels Technologies Program Multiyear Program Plan goal of identifying fuels that can displace 5% of petroleum diesel by 2010, the National Renewable Energy Laboratory (NREL), in collaboration with the National Biodiesel Board (NBB) and with subcontractor Southwest Research Institute, performed a study of biodiesel oxidation stability. The objective of this work was to develop a database that supports specific proposals for a stability test and specification for biodiesel and biodiesel blends. B100 samples from 19 biodiesel producers were obtained in December of 2005 and January of 2006 and tested for stability. Eight of these samples were then selected for additional study, including long-term storage tests and blending at 5% and 20% with a number of ultra-low sulfur diesel (ULSD) fuels. These blends were also tested for stability. The study used accelerated tests as well as tests that were intended to simulate three real-world aging scenarios: (1) storage and handling, (2) vehicle fuel tank, and (3) high-temperature engine fuel system. Several tests were also performed with two commercial antioxidant additives to determine whether these additives improve stability. This report documents completion of NREL's Fiscal Year 2007 Annual Operating Plan Milestone 10.1.The B100 samples examined show a broad distribution of stability on accelerated tests, with oil stability index (OSI) or Rancimat induction time results ranging from less than 1 hour to more than 9 hours and ASTM D2274 total insolubles ranging from less than 2 mg/100 ml to nearly 18 mg/100 ml. The accelerated test data indicate that if the B100 stability is above roughly a 3-hour induction time, blends prepared from that B100 appear to be stable on the OSI and D2274 tests.The D4625 long-term storage results for B100 indicate that most biodiesel samples, regardless of initial induction time, will begin to oxidize immediately during storage. If induction time is near or below the 3-hour limit, the B100 will most likely go out of specification for either stability or acid value within 4 months. Even B100 with induction times longer than 7 hours will be out of specification for oxidation stability at only 4 months, although these samples may not have shown a significant increase in acidity or in deposit formation. The 3-hour B100 induction time limit appears to be adequate to prevent oxidative degradation for both B5 and B20 blends in storage for up to 12 months.For tests that simulated fuel tank aging and high temperature stability, we conclude that stable B100 (longer than 3 hours induction time) leads to stable B5 blends. For B20, the results are less definitive, but provide considerable evidence that B100 with induction time of at least 3 hours produces stable B20 blends, but the test cannot differentiate between intermediate and highly stable samples for acid number increase or sediment formation under these worst-case test conditions. Additional work is required to confirm this finding and to...