Feasibility of Solar Fuels for Sustainable Transportation: A Numerical Investigation

Abstract

Solar fuel is the fuel type that can be produced and stored using solar energy. Among these fuels is the OxyMethylene dimethyl Ethers (OMEl) or also known as PolyOxymethylene Dimethyl Ether (PODEl). OMEl is a promising alternative for conventional fuel sources, mainly because they relate to reduced (i) soot production and (ii) toxic emissions. However, the high oxygen content of OMEl is a significant limitation, as the high O2 content results in relatively low calorific value (energy content). The current work explores the ignition dynamics of OMEl using the Computational Singular Perturbation (CSP) algorithm. In particular, the manners by which the calorific value of OMEl ignition can be enhanced will be explored in order to control ignition. The current work is based on three reactive processes (constant volume and pressure homogeneous autoignition and laminar premixed flame) and three OMEl (OME2, OME3, OME4). The reactions and species that control OME2−4 autoignition and laminar premixed flame will be identified. This knowledge will lead to suitable clean and renewable additives that control the reactive process under consideration.

The work on OMEl is mainly based on the explosive mode that develops during intense chemical activity. The time scale of this mode, τe, provides a measure of the action. This feature of τe formed the basis of two distinct investigations. First, the possibility to use τe(0) in order to assess the sensitivity of IDT with respect to the fuel to air equivalence ratio ϕ was explored. The analysis is based on the homogeneous autoignition of six different fuels and a wide range of operating conditions. Then, considering homogeneous autoignition, the relation of this time scale at the start of the process τe(0) to the ignition delay time IDT, was investigated using Machine Learning (ML). It is concluded that τe is the best input to ML for accurate IDT prediction, when compared to all other independent variables that define the physical process.

Date of Award	Apr 2023
Original language	American English
Supervisor	Dimitrios Gkousis (Supervisor)