Master of Engineering Science
Civil and Environmental Engineering
Professor E.K. Yanful
The present research examines solely the transesterification stage of the biodiesel production process. Six process variables that affect the yield of biodiesel at this stage are examined using the 12 run Plackett-Burman Design. To study the effect of FFA and to replicate real life situations where oil (feedstock) contain varying amounts of FFA, linoleic acid is used as FFA and intentionally added to high oleic acid canola oil containing less than 0.07% FFA. The process is catalyzed with potassium carbonate and evaluated at varying temperatures, stirring rates, reaction times and methanol oil ratios. The yields at the end of these reactions are measured and the active factors determined with the PB design in Minitab software. At the end of the experimental design, it was found that the FFA amount affects the yield of biodiesel the most and methanol-oil ratio the least for the range of values studied.
The Box-Behnken method was then applied in optimising the amount of free fatty acid, the amount of catalyst and the stirring speed. The relationship between the yield and the three (3) factors was determined using analysis of variance (ANOVA) and subsequently optimised.
The optimised factor combination for a percentage yield greater than 98% was found for a 1 hour reaction to be 0.5 g FFA, 400 rpm stirring speed and 4 g catalyst at 60˚C temperature and 6:1 methanol-oil ratio.
 FFA refers to Free Fatty Acid
 PB = Plackett-Burman Design
Paintsil, Arnold, "Optimisation of the Transesterification Stage of Biodiesel Production using statistical methods" (2013). Electronic Thesis and Dissertation Repository. 1693.