Experimental study and empirical correlation development of fuel properties of waste cooking palm biodiesel and its diesel blends at elevated temperatures

• Fuel properties of methyl ester palm biodiesel and its blend was studied.
• Experiments were carried out at varying temperature and blend ratio condition.
• Both density and dynamic viscosity decreased with increase in temperature.
• Higher heating value of fuel decreased with increase in blend ratio.
• Regression correlations were proposed for each fuel properties.

In this experimental work, the density, dynamic viscosity and higher heating value of methyl ester based waste cooking palm-biodiesel oil (WMEPB) was investigated under varying temperature and blend ratio condition with No. 2 diesel fuel. The transesterified fatty acid methyl ester of palm vegetable oil collected from local food and beverage shops was used as neat biodiesel. Four different fuel blends (20%, 40%, 60% and 80% by volume mixing with base diesel) were studied along with base No. 2 diesel fuel and pure biodiesel. Tests for dynamic viscosity and density were performed in the temperature range 0–130 °C for each fuel sample whereas the higher heating values were determined at 25 °C room temperature condition. It is found that pure biodiesel has the highest density and dynamic viscosity at a given temperature whereas it exhibits lowest combustion heating value among the six fuels. Moreover, the density for each fuel sample decreases linearly with the increase in temperature. On the other hand, the dynamic viscosity decreases exponentially with the temperature for each fuel sample. In addition, based on the experimental results, regression correlations have been proposed for the density, dynamic viscosity, and higher heating value of the fuels. Subsequently, comprehensive error analyses of these proposed correlations were performed. In particular, the correlation for density and dynamic viscosity were respectively compared with Kay's mixing rule and Grunberg-Nissan mixing rule theory in order to validate their applicability. It is found that density correlations predicted within ±0.3% average error band. And, as high as 72.2% of the dynamic viscosity data were in the range of ±5% average error while the remaining data fell within ±10% error range. And finally, through a comparative study with the available fuel property results of fresh methyl ester palm biodiesel, it is found that available existing correlations derived from fresh palm biodiesel studies can not accurately predict the fuel properties of same waste biodiesel and its blends with diesel.