NSChE Journal Articles
*The article download page is restricted to registered members only.
MODELLING AND PROCESS PARAMETERS OPTIMIZATION OF LUCKY NUT SEED OIL EXTRACTION BY ARTIFICIAL NEURAL NETWORK
Ajala S. O. and Betiku E*. Biochemical Engineering Laboratory, Department of Chemical Engineering, ObafemiAwolowo University, Ile-Ife 220005, Osun State, Nigeria
Article Abstract
Lucky nut (Thevetiaperuviana) seed oil extraction process was modelled and parameters involved were optimized. The effects of three independent factors (sample weight, extraction time and solvent type) on the seed oil yield and their interactions were evaluated. The DOptimal design of response surface methodology was used to generate 24 experimental runs, which were subsequently carried out. The model that best described the extraction process was achieved by multilayer normal feedforward incremental back propagation network with sigmoid function. The predicted optimal condition wassample weight of 20 g, extraction time of 3 h and solvent type, petroleum ether. The observedlucky nutseed oil yield was 77.63% (w/w), which compared well with the value of 77.42% (w/w) predicted by Artificial Neural Network (ANN). The physiochemical properties of the extracted lucky nut seed oil (% FFA of 7.05, acid value of 14.030.11 mg KOH/g oil, iodine value of 98.98 g I2/100g oil and peroxide value of 4.60 meq O2/kg oil)indicated that itwas non edible and the fatty acid profile also showed that the oil was highly unsaturated (76.13%) with oleic acid as the highest (43.81%). Thus, the seed oil could serve as a feedstock in oleiochemicals and biodiesel industries.
Click here to download full article *Access restricted to members only
<<<<<<<Back to journal listing
 
(c) 2010 Nigerian Society of Chemical Engineers ---------------- Contact Webmaster