(547c) Development of Robust Models for the Prediction of Reid Vapor Pressure (RVP) in Fuel Blends and Their Application to Oxygenated Fuels

The growing demand for inexpensive, more environmentally friendly fuels has sparked research into fuel properties that affect fuel production, handling and engine efficiency. Reid Vapor Pressure (RVP) has been shown to be one such fuel property. Having models that can accurately predict the RVP of a fuel quickly and robustly can help in the development of novel fuels, increase engine efficiencies and value for refiners in their existing operations. Due to non-linear blending effects, especially when oxygenates are involved, current methods often fail to predict RVP correctly. New methods are required which can accurately take non-linear blending effects into account. To that end, models were developed that can predict the RVP of complex Blendstocks for Oxygenated Blending (BOB) using the SAFT-γ-mie Equation of State (EOS). This approach captures the underlying physics and has been shown to be reliable for the prediction of thermophysical properties of complex fluids. The models increase in fidelity and accuracy, with increasing complexity, i.e. number of compounds included, to construct. Three molecular based “surrogate” models are presented. Each model uses a set of simple rules to determine which components must be included, with the small surrogate model including the fewest number of components, and the large surrogate model including the largest number of components. In order to assess the accuracy of each model, seven BOBs with a wide range of known RVPs and representing a wide range of chemical compositions were evaluated. The present absolute average deviation %AAD, averaged over these seven BOBs, is calculated to be 15.09%, 9.49%, and 3.66%, for the small, moderate, and large surrogate models, respectively. To test the performance of oxygenates added to the system, two of the BOBs, a winter and summer, were blended with several oxygenates of varying molecular class. The %AAD for the summer BOB data is 19.72%, 14.90%, and 3.48% for the small, moderate, and large surrogate model, respectively. The %AAD for the winter BOB data is 15.50%, 12.16%, and 2.58% for the small, moderate, and large surrogate model, respectively. In addition, excess thermodynamic properties for several binary systems were examined. These were used to create a physical model of RVP, which can be used to inform future fuel research. Lastly, using the large surrogate model, the RVP was predicted for a suite of oxygenates for blends of 10, 20, and 30%, by weight. Although the majority of oxygenates tested decreased the RVP, the branched alcohols, and the ketones showed the largest decreases, suggesting possible refinery applications through the decrease of fugitive losses.