Understanding chemistry-specific fuel differences at a constant RON in a boosted SI engine

Abstract

The goal of the US Department of Energy Co-Optimization of Fuels and Engines (Co-Optima) initiative is to accelerate the development of advanced fuels and engines for higher efficiency and lower emissions. A guiding principle of this initiative is the central fuel properties hypothesis (CFPH), which states that fuel properties provide an indication of a fuel’s performance, regardless of its chemical composition. This is an important consideration for Co-Optima because many of the fuels under consideration are from bio-derived sources with chemical compositions that are unconventional relative to petroleum-derived gasoline or ethanol. In this study, we investigated a total of seven fuels in a spark ignition engine under boosted operating conditions to determine whether knock propensity is predicted by fuel antiknock metrics: antiknock index (AKI), research octane number (RON), and octane index (OI). Six of these fuels have a constant RON value but otherwise represent a wide range of fuel properties and chemistry. Consistent with previous studies, we found that OI was a much better predictor of knock propensity that either AKI or RON. However, we also found that there were significant fuel-specific deviations from the OI predictions. Combustion analysis provided insight that fuel kinetic complexities, including the presence of pre-spark heat release, likely limits the ability of standardized tests and metrics to accurately predict knocking tendency at all operating conditions. While limitations of OI were revealed in this study, we found that fuels with unconventional chemistry, in particular esters and ethers, behaved in accordance with CFPH as well as petroleum-derived fuels.