Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network

Optimized Novozym 435 (Candida antarctica lipase B immobilized on acrylic resin)- catalyzed esterification of adipic acid and various monohydric alcohols was successfully performed. Solvent-based synthesis of adipate esters was carried out in small scale reaction using 30 mL screw-capped vials. T...

Full description

Saved in:

Bibliographic Details
Main Author:	Langroodi, Naz Chaibakhsh
Format:	Thesis
Language:	English English
Published:	2010
Subjects:	Neural networks Surfaces (Physics) - Analysis Esters
Online Access:	http://psasir.upm.edu.my/id/eprint/12437/1/FS_2010_22A.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-upm-ir.12437
record_format	uketd_dc
spelling	my-upm-ir.124372013-05-27T07:52:14Z Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network 2010-03 Langroodi, Naz Chaibakhsh Optimized Novozym 435 (Candida antarctica lipase B immobilized on acrylic resin)- catalyzed esterification of adipic acid and various monohydric alcohols was successfully performed. Solvent-based synthesis of adipate esters was carried out in small scale reaction using 30 mL screw-capped vials. The synthetic reaction was optimized by Response Surface Methodology (RSM) based on central composite rotatable design (CCRD) to evaluate the interactive effects of reaction parameters including temperature, time, enzyme amount and alcohol/acid molar ratio. A high percentage yield (>96.0%) using optimum conditions was obtained using a minimum amount of enzyme, which matched well with the predicted values. Artificial Neural Network (ANN) approach was also employed for the estimation of esterification yield in enzymatic synthesis of adipate esters. Various feedforward neural networks were performed using different learning algorithms. The best algorithm was found to be Levenberg–Marquardt (LM) for a network composed of seven hidden nodes with hyperbolic tangent sigmoid transfer function. ANN showed better prediction ability compared to RSM. A high coefficient of determination (R2) (>0.9) and a low mean absolute error (MAE) and root mean squared error (RMSE) for training, validating and testing data implied the good generalization of the developed models for predicting the reaction yield. In order to develop an efficient enzyme catalyzed process, alcohol specificity of enzyme in terms of chain length and structure in the synthesis of adipate esters was determined. Methanol, n-butanol, octanol, dodecanol, octadecanol, isobutanol, sec-butanol and tertbutanol were the alcohols used for this study. The results demonstrated that alcohol chain length and structure were determining factors that affect the optimum condition of the reaction parameters for the enzymatic synthesis of adipate esters. Minimum reaction time for achieving maximum ester yield was obtained for isobutanol. The initial rates of synthesis of adipate esters for primary and secondary alcohols were nearly the same. Kinetic study of the lipase-catalyzed adipate ester synthesis in solvent-based system was carried out as a preliminary step for future industrial scale bioreactor design. The reaction was found to obey the ping-pong bi-bi mechanism with alcohol inhibition. The coefficient of determination (R2) and MAE values between the simulated and experimental initial rates were determined as 0.9904 and 2.4×10-4 which shows a good quality of fit between the simulated and experimental values.In order to make the synthesis process more environmentally friendly and improve the productivity of the reactor to the highest amount, the reaction was performed in a solvent-free system using 0.5-L batch and 4-L continuous stirred tank reactors. Due to low solubility of the substrate and high viscosity of the reaction mixture, a continuous stirred tank reactor was used for continuous production of the ester. A high percentage conversion was achieved (about 96%) using minimum amount of enzyme (2.5%w/w) indicating the high efficiency of solvent free-system for synthesis of adipate ester. Continuous production of adipate ester was successfully performed with an average yield of 92.7% and high operation stability of enzyme for 28 hours, which is indicative of performing a successful process for the ester synthesis. Neural networks Surfaces (Physics) - Analysis Esters 2010-03 Thesis http://psasir.upm.edu.my/id/eprint/12437/ http://psasir.upm.edu.my/id/eprint/12437/1/FS_2010_22A.pdf application/pdf en public phd doctoral Universiti Putra Malaysia Neural networks Surfaces (Physics) - Analysis Esters Faculty Of Science English
institution	Universiti Putra Malaysia
collection	PSAS Institutional Repository
language	English English
topic	Neural networks Surfaces (Physics) - Analysis Esters
spellingShingle	Neural networks Surfaces (Physics) - Analysis Esters Langroodi, Naz Chaibakhsh Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
description	Optimized Novozym 435 (Candida antarctica lipase B immobilized on acrylic resin)- catalyzed esterification of adipic acid and various monohydric alcohols was successfully performed. Solvent-based synthesis of adipate esters was carried out in small scale reaction using 30 mL screw-capped vials. The synthetic reaction was optimized by Response Surface Methodology (RSM) based on central composite rotatable design (CCRD) to evaluate the interactive effects of reaction parameters including temperature, time, enzyme amount and alcohol/acid molar ratio. A high percentage yield (>96.0%) using optimum conditions was obtained using a minimum amount of enzyme, which matched well with the predicted values. Artificial Neural Network (ANN) approach was also employed for the estimation of esterification yield in enzymatic synthesis of adipate esters. Various feedforward neural networks were performed using different learning algorithms. The best algorithm was found to be Levenberg–Marquardt (LM) for a network composed of seven hidden nodes with hyperbolic tangent sigmoid transfer function. ANN showed better prediction ability compared to RSM. A high coefficient of determination (R2) (>0.9) and a low mean absolute error (MAE) and root mean squared error (RMSE) for training, validating and testing data implied the good generalization of the developed models for predicting the reaction yield. In order to develop an efficient enzyme catalyzed process, alcohol specificity of enzyme in terms of chain length and structure in the synthesis of adipate esters was determined. Methanol, n-butanol, octanol, dodecanol, octadecanol, isobutanol, sec-butanol and tertbutanol were the alcohols used for this study. The results demonstrated that alcohol chain length and structure were determining factors that affect the optimum condition of the reaction parameters for the enzymatic synthesis of adipate esters. Minimum reaction time for achieving maximum ester yield was obtained for isobutanol. The initial rates of synthesis of adipate esters for primary and secondary alcohols were nearly the same. Kinetic study of the lipase-catalyzed adipate ester synthesis in solvent-based system was carried out as a preliminary step for future industrial scale bioreactor design. The reaction was found to obey the ping-pong bi-bi mechanism with alcohol inhibition. The coefficient of determination (R2) and MAE values between the simulated and experimental initial rates were determined as 0.9904 and 2.4×10-4 which shows a good quality of fit between the simulated and experimental values.In order to make the synthesis process more environmentally friendly and improve the productivity of the reactor to the highest amount, the reaction was performed in a solvent-free system using 0.5-L batch and 4-L continuous stirred tank reactors. Due to low solubility of the substrate and high viscosity of the reaction mixture, a continuous stirred tank reactor was used for continuous production of the ester. A high percentage conversion was achieved (about 96%) using minimum amount of enzyme (2.5%w/w) indicating the high efficiency of solvent free-system for synthesis of adipate ester. Continuous production of adipate ester was successfully performed with an average yield of 92.7% and high operation stability of enzyme for 28 hours, which is indicative of performing a successful process for the ester synthesis.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Langroodi, Naz Chaibakhsh
author_facet	Langroodi, Naz Chaibakhsh
author_sort	Langroodi, Naz Chaibakhsh
title	Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
title_short	Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
title_full	Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
title_fullStr	Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
title_full_unstemmed	Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network
title_sort	modeling and optimization of lipase-catalyzed synthesis of adipate esters using response surface methodology and artificial neural network
granting_institution	Universiti Putra Malaysia
granting_department	Faculty Of Science
publishDate	2010
url	http://psasir.upm.edu.my/id/eprint/12437/1/FS_2010_22A.pdf
_version_	1747811366212730880

Modeling And Optimization Of Lipase-Catalyzed Synthesis Of Adipate Esters Using Response Surface Methodology And Artificial Neural Network

Similar Items