LAUSR

Abstract Residual lipids with high fatty acid content represent an important source of raw material for renewable biofuels. They are not suitable for the production of biodiesel via alkaline transesterification but can be converted to biodiesel via esterification. For these feedstocks, hydrodeoxygenation is suitable for producing hydrocarbons. However to limit hydrogen consumption, thermal processes are also object of studies. In these thermal processes nickel has been used in the form of supported catalysts but has not been well studied in the form of mixed oxides. This work aims at studying the Ni/Al ratio influence on the properties of unsupported mixed oxides prepared from layered double hydroxides (LDH) precursors, to evaluate their performance in the fast pyrolysis of myristic acid (MA) as a model compound of residual acidic lipids. Prepared materials had molar ratios Ni/Al = 0.3, 1 and 3. The catalysts were obtained through calcination of the precursors, and were characterized by XRD, TGA, FTIR, N2 adsorption isotherms, TPD of NH3 and EDX analysis. Myristic acid pyrolysis, with catalyst:MA mass ratio of 5:1, was carried out in a micropyrolyzer at 550 °C, and the products were analyzed through online GC/MS. Results show that the obtained catalysts presented a rather high surface area, important mesoporosity and NiO and NiAl2O4 crystalline phases. The catalysts with the highest Al content favored cracking with the production of hydrocarbons in the gasoline range, while the sample with highest Ni content favored the production of hydrocarbons in the kerosene range. The catalyst obtained with the combination of Ni/Al = 1 presented the highest surface area and acidity and the lowest NiO crystallite size, resulting in the best MA conversion (∼84 %) and deoxygenation degree. It also produced the highest yield in aromatic compounds due to the higher acidity and hydrogen transfer, compared to other catalysts. The present results confirm that catalytic reaction under pyrolysis conditions is able to identify some intrinsic properties of catalysts using model molecules as starting feed.