Abstract Safety specifications for jet fuel require a minimum flash point of 38°C. Synthetic kerosenes produced from non‐petroleum sources often contain different proportions of n ‐alkanes, iso‐alkanes, cycloalkanes, and aromatics than petroleum‐derived jet fuels, raising questions about the applicability of established flash point correlations. In this study, linear, branched, cyclic, and aromatic C 8 –C 10 hydrocarbons were blended at 5 wt.% and 10 wt.% into a model synthetic kerosene (60 wt.% n ‐dodecane, 20 wt.% isopropylcyclohexane, and 20 wt.% p ‐cymene). Selected C 8 –C 9 hydrocarbons were also blended into three actual synthetic kerosenes (olefinic from alcohol‐to‐jet conversion, olefinic from 1‐hexene and cyclohexene oligomerization, and paraffinic from wax hydrocracking) and two petroleum‐derived kerosenes (hydrotreated kerosene and Jet A‐1). Flash points were determined using closed‐cup methods, and densities and refractive indices were measured to assess deviations from ideal mixing. The dataset was used to evaluate three empirical flash point correlations (Wickey–Chittenden, Hu–Burns, and Gary–Handwerk). Blending C 8 hydrocarbons reduced the flash point of the model synthetic kerosene from 52°C to 32–49°C, whereas C 9 –C 10 hydrocarbons caused moderate to negligible reductions. Density measurements revealed small but statistically significant deviations from ideal behaviour for most blends. Flash point prediction errors were influenced primarily by carbon number rather than hydrocarbon class. Synthetic blends containing olefinic additives showed the largest underprediction. The Wickey–Chittenden and Gary–Handwerk correlations resulted in 0.6%–9.8% (AARD) and negative bias, indicating conservative predictions. In conclusion, established petroleum‐based flash point correlations were found to remain suitable for synthetic kerosene systems.
Bassane et al. (Tue,) studied this question.