Numerical Validation and Optical Study of Injection of Different Oxymethylene Ether Fuels for Heavy-Duty Application

A reliable toolchain for the validation and evaluation of numerical spray break-up simulation for the potentially carbon-neutral fuels polyoxymethylene dimethylether (POMDME, or short OME) is developed and presented. The numerical investigation is based on three-dimensional computational fluid dynamics (3D-CFD) with the commercial code STAR-CD v2019.1 using a Reynolds-averaged Navier-Stokes (RANS) equations approach. Fuel properties of the representatives OME1 and OME3 are implemented into the software and with that the fuels are investigated numerically. For validation purposes, optical experimental results in a heated spray chamber with inert nitrogen-pressurized atmosphere are presented. The measurement data are based on Mie scattering of the liquid phase and Schlieren imaging of the vapor phase. Solely experimental results are shown for OME1b and OME3–6 to assess if the knowledge from the numerical modeling with OME1 and OME3 can also be transferred to the corresponding multicomponent fuels. While the results for a match between OME3 and OME3–6 are close, the measurement for OME1b exceeds the result of OME1 in the liquid penetration significantly. This is explained by the molecular structure of the low-volatile additive in OME1b based on long-chained polyglycol ethers. For the numerically modeled operating conditions, the fuel injection rate with the corresponding fuel is measured. Two atomization and spray break-up approaches are investigated in simulation, based on Reitz-Diwakar (RD) models and a combination using Huh’s atomization and the Kelvin-Helmholtz Rayleigh-Taylor (KHRT) spray break-up models. A holistic parameter study in a single operating point with the fuel OME1 helps to determine the sensitivities of the models. Adjustments to the spray momentum by a variation of the parameter for the nozzle hole diameter are used to get results closely aligned with measurement data. The transfer of the calibrated RD model to a validation study with OME3 at different operating conditions matches well to measurement with no further adjustments necessary.