Production of Renewable Diesel (RD) and Sustainable Aviation Fuels (SAF) from biological sources (natural oils) has seen exponential growth in recent years, stemming from worldwide government mandated climate change initiatives alongside the need for carbon capture and sequestration. Significant, rapid investments have occurred in retrofitting / adapting existing refinery hydroprocessing infrastructure to process natural oils or coprocess natural oils blended with crudes to produce RD and SAF. This stems from the fact that natural oils have the hydrocarbon (HC) structures to fit within the mid-distillate fuel product such as diesel and aviation fuel as well as that these processes are optimized for removal of unwanted Sulfur and Oxygen removal.

In these modified hydroprocessing applications, high temperature decomposition of triglycerides (TRG) leads to production of RD and SAF through hydroprocessing of esters and free fatty acids (FFA). The resulting oxygen free-RD and SAF products are completely fungible with petroleum hydrocarbons and have the added benefit of producing propane as a byproduct. Hydro-processing of refined natural oils (degummed, bleached, neutralized, and deodorized) has its own unique corrosion problems. FFA formed in pre-heat trains and off-gases (H2O, CO2, CO, H2) differ from those traditionally found in refinery hydro-processing. While aqueous carbonic acid corrosion due to CO2 is well understood, high temperature corrosion by transient formation of FFA is not and forms the focus of this paper.

Over the last few years, the authors introduced a molecular mechanistic model to quantify simultaneous high temperature naphthenic acid and sulfidation corrosion (CorrExpert®-Crude) in refinery CDU/VDU operations. That model has been adapted to address high temperature FFA corrosion, given that FFA are carboxylic acids, akin to naphthenic acids found in conventional fossil-fuel based crude unit process streams. Above 540F (280C), TRGs generate mixtures of free fatty acids (FFA) that vary in corrosivity and stability as a function of molecular shape / structure and saturation. Saturated FFA are rigid, tend to align in clumps by self-association and are much less corrosive when compared to conventional HC nap acids. Unsaturated FFA, which leave access for a 2nd acid, are similar to HC nap acids and are equally corrosive. The proposed model facilitates differentiated treatment of unsaturated and saturated FFA to quantify surface reactions sensitive to acid shape and reactivity. Algorithms have been created to predict the “effective TAN” for TRG at temperature as a function of residence time and are designed to be incorporated into the existing corrosion prediction model.