Prediction of Flash-Point Temperature of Alcohol/Biodiesel/Diesel Fuel Blends

Significant advances have been achieved in the last decades for predicting flash point, not only for pure substances but also for multicomponent fuels, even if they are complex nonideal mixtures. This is the case of blends of alcohols with diesel or diesel/biodiesel fuels. Prevention of hazards is necessary when a light and volatile alcohol such as ethanol or n-butanol (with increasing interest as diesel fuel components) is included in the fuel composition. However, an accurate modeling requires detailed knowledge of the diesel fuel components, which is not always available. In this work, a tailored two-option model is proposed, which can be use depending on the information available. If detailed information about the diesel fuel composition is available, a first method based on the Liaw equation is recommended. If not, an extended method based on the combination of the Liaw equation and the Gibbs-Duhem equation is proposed instead, with diesel fuel being represented by a three-component surrogate (one component belonging to each of the main diesel-component families). These two methods were compared with experimental flash-point results, and it was shown that the simplified method provides results as accurate as the complete method, and only if results for ethanol blends are considered unreliable, the complete method would be slightly more accurate. Both experimental and modeled results show that the rate of decrease of flash point is sharper for ethanol than for n-butanol, but the latter presents a more noticeable minimum flash-point behavior. However, the presence of biodiesel in the blends has the potential to neutralize both hazardous behaviors.