Effects of biodiesel fuels on soot particle morphology and nanostructure

Abstract

This work investigates the effect of biodiesel fuels on the soot particle morphology and nanostructure in both a constant volume combustion chamber (CVCC) and a small-bore (0.498 l) single cylinder compression ignition engine. These parameters are essential for the understanding of the soot formation process itself and for optimization of the performance of exhaust aftertreatment systems. Moreover, measures, in order to lower the health impact and meet emission regulations, can be gained from these parameters. The experiments were conducted with three different kinds of biodiesel fuels. Namely, waste cooking oil (WCO), Jatropha oil and Karanja oil were used. All those fuels were derived by transesterification process. Even though the chemical principle to produce those fuels is the same they all exhibit, due to their different feedstock, dissimilar physical characteristics especially in kinematic viscosity, lower heating value and cetane number. In contrast, the elemental composition of the fuels (hydrogen, carbon, and oxygen) is comparable. For comparison, standard diesel fuel was selected. Background information about production and feedstocks of biodiesel fuels, soot formation mechanisms, the importance of the soot nanostructure itself, as well as commonly used parameters to describe the carbon layer arrangement inside the soot particles is provided in chapter 2. The theory behind the flame image analysis used for the CVCC experiment and the exhaust gas measurement later used in the experiment with the compression ignition engine is given in this chapter. To investigate the nanostructure parameters of the diesel soot high-resolution transmission electron microscopy (HRTEM) images were taken. The methodology for analyzing those images is described in chapter 3. Among them also in other works used principles for analyzing the primary particle size, fringe length, and tortuosity a newly developed code to measure the fringe spacing, based on a two-dimensional fast Fourier transform (2D-FFT), is described in this chapter. Also described in this chapter is the algorithm used for the flame image analysis. Chapter 4 continues with a detailed description of the tested fuels and their physical characteristics and chemical composition. Subsequently, the conducted experiments are described. For both experiments, the injection pressures were changed from 40 over 80 to 120 MPa. In case of the CVCC experiment, the soot sampling was conducted in the flame and in addition highspeed imaging of the combustion was conducted to visualize high temperature and soot formation regions by correlated color temperature and hue value analysis. The soot sampling conditions for the engine experiment were an indicated mean effective pressure (IMEP) of 0.5 MPa and an injection timing of -5$^\circ$ after top dead center (aTDC) at 1200 rpm. Further, also the exhaust gas composition was analyzed. Therefore, the injection timing was swept in six steps from -15$^\circ$ aTDC to -3$^\circ$ aTDC. The CVCC experiment unveiled a correlation between lower heating value, injection pressure, and primary particle size for the biodiesel fuels. It was shown that there is an approximately linear relationship between lower heating value and primary particle size, especially at 80 and 120 MPa injection pressure. The experiment has also shown, that there is a connection between the nanostructure parameters fringe length, tortuosity and spacing in this early stage of oxidation. In case of the experiment with the small-bore compression ignition engine, it could be shown, that there is a connection between the soot particle morphology, kinematic viscosity, and hydrocarbon emissions in case of the engine experiment. The analysis of the TEM images unveiled that the higher hydrocarbon emissions caused by wall wetting of the more viscous fuels led to an adsorption of the vapor hydrocarbon phase and the unburned fuel on the circumference of the soot particles. This organic fraction caused coalescence in case of the higher viscous fuels Jatropha and Karanja whereas for diesel and WCO agglomeration was dominating. The organic fraction as also found to cause a larger primary particle size in case of Karanja and Jatropha compared to WCO. The work closes with a summary of the findings in this work as well as recommendations for the application of biodiesel fuels in compression ignition engines and future works in this field in chapter 5. It was found, that WCO among the tested biodiesel fuels the best alternative to the standard diesel fuel. Moreover, Jatropha could be, if used in large bore engines, a promising alternative. However, therefore, modifications at the injection system, to lower the wall wetting, are needed. Suggested measures are smaller orifices of the injector to lower the liquid tip of the spray. Also preheating of the fuels prior injection might deliver the desired spray characteristics. Biodiesel derived from Karanja oil was observed to have due to its high viscosity and with that connected insufficient mixture preparation drawbacks in hydrocarbon emissions. Those are also causing, as the vapor hydrocarbon phase gets adsorbed by the particles, coalescence of the primary particles. This consequently leads to a decreased particle surface and therefore inferior oxidation characteristics of the soot.