|
Title
Ethanol and soda pulping of tagasaste wood: Neural fuzzy modeling
Authors
Ascension Alfaro Antonio Perez Juan C. Garcia Francisco Lopez Minerva A. M. Zamudio and Alejandro Rodriguez
Received
July 29, 2009
Published
Volume 43 Issue 7-8 July-August
Keywords
ethanol pulping, neural fuzzy modeling, soda pulping, tagasaste
Abstract
A central composite experimental design, in conjunction with the ANFIS Edit Matlab 6.5 software (to develop a neural fuzzy model), was used in the study of the influence of ethanol and alkaline soda pulp process variables on the properties of pulps and paper sheets obtained from tagasaste (Chamaecytisus proliferus L.F. ssp palmensis) trimming residues. A variation range, involving three independent variables at three different (variation) levels, was established for each process variable (viz. temperature, cooking time and soda or ethanol concentration). The obtained models reproduced the experimental results of the dependent variables with errors below 8%. Under optimum experimental conditions, the yield of the ethanol process is 4% higher than that of the soda process. On the other hand, soda pulp has a higher ?-cellulose content and a lower lignin content than ethanol pulp. The optimum conditions for the soda process involve a temperature of 180 ºC, a cooking time of 60 min and an active alkali concentration of 16%, while those of the ethanol process are: 185 ºC, 90 min and a 60% v/v ethanol concentration. The results obtained recommend the tagasaste trimmings as an effective raw material for obtaining paper pulp by ethanol and alkaline soda pulp processes.
Link
|
