Premium
Performance and evaluation of calcined limestone as catalyst in biodiesel production from high viscous nonedible oil
Author(s) -
Bharadwaj A. V. S. L. Sai,
Niju S.,
Meera Sheriffa Begum K. M.,
Narayanan Anantharaman
Publication year - 2019
Publication title -
environmental progress and sustainable energy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.495
H-Index - 66
eISSN - 1944-7450
pISSN - 1944-7442
DOI - 10.1002/ep.13342
Subject(s) - response surface methodology , biodiesel , transesterification , biodiesel production , methanol , coefficient of determination , mean squared error , calcium oxide , materials science , calcination , central composite design , catalysis , correlation coefficient , pulp and paper industry , design of experiments , chemical engineering , process engineering , environmental science , mathematics , chemistry , chromatography , engineering , organic chemistry , statistics
Biodiesel production by transesterification of rubber seed oil (RSO) using calcium oxide (CaO) derived from calcined limestone as a heterogeneous catalyst is presented in this study. Optimization of process parameters affecting the conversion of RSO to biodiesel is done by design of experiments (DOE) and an effective comparison of two different optimization methods, namely, response surface methodology (RSM) and artificial neural networks (ANN) is presented. A high conversion of 95.2% was obtained at 12:1 methanol: Oil molar ratio, 4 (wt%) catalyst and 5 hr of reaction time. The proposed design model of RSM is found to fit well with the predicted conversion and with molar ratio and reaction time as the significant process parameters affecting the conversion. Best validation performance of 8.8991 occurred at epoch 4 with a mean square error (MSE) of 1.55 in ANN model trained with Levenberg–Marquardt algorithm. By comparing the predicted coefficient of determination, R 2 , values of 0.8452 obtained by using RSM, and 0.9939 obtained by using ANN for biodiesel conversion, it is concluded that ANN model is the best model for predicting the percentage conversion of RSO to biodiesel with minimum error between experimental and predicted values.