oa Experimental and Statistical Studies for the Design of Novel Synthetic Jet Fuels Derived from Natural Gas

This paper highlights student research experience in the fuel characterization lab at Texas A&M University at Qatar. The research was carried out over one academic year with a multidisciplinary undergraduate team. A professional environment was applied regarding time management, safety and communication between team members and instructors. All research was carried out with strict adherence to safety standards such as waste disposal and solvent handling. This project allowed us to apply in-class concepts to hands-on lab experiments. Furthermore, this project gave us the opportunity to work with global consortium of leading scientists from industry (Shell and Rolls Royce) and academia (University of Sheffield-UK and German Aerospace Institution-DLR). This consortium is funded by Qatar Science & Technology Park with support from Qatar Airways as part of its initiatives to become the world-leading airline in clean synthetic fuels.

The project goal was to develop new synthetic jet fuels with the student focus being on the experimental aspect. Physical properties, such as freezing points, flash points and heat content, were tested for various blends. Experiments followed American Society for Testing and Materials (ASTM) standards and data documentation adhered to industry practices. The goal was achieved by creating a number of synthetic fuel blends which were subsequently tested multiple times. The results provided an extensive database, linking properties with compositions and provided an understanding of the influences of a fuel's molecular structure on its physical properties, which is essential for fuel certification.

Mathematical models were used to link the chemical composition and physical properties of blends. The purpose of these models was to predict intrinsic properties of the blends through analysis of paraffinic composition. These models were made using MATLAB, a software for simulation and programming. A neural network simulation approach was employed, with the experimental data used to train this model. This student-developed model will be used to take this project further.