American Journal of Environmental Engineering and Science 2015; 2(1): 1-8 Published online February 20, 2015 (http://www.aascit.org/journal/ajees) Keywords Orange Tree Pruning, Ethanol, Cellulose, Hemicelluloses, Polynomial Models Received: January 27, 2015 Revised: February 10, 2015 Accepted: February 11, 2015 Valorization of the Orange Tree Pruning by Ethanol Process Zoilo González 1 , Alberto Vega 2 , Pablo Ligero 2 , Alejandro Rodríguez 1, * 1 Chemical Engineering Department, Campus of Rabanales, Building Marie Curie (C-3), University of Córdoba, Córdoba, Spain 2 Department of Physical Chemistry and Chemical Engineering, University of A Coruña, A Coruña, Spain Email address a.rodriguez@uco.es (A. Rodríguez) Citation Zoilo González, Alberto Vega, Pablo Ligero, Alejandro Rodríguez. Valorization of the Orange Tree Pruning by Ethanol Process. American Journal of Environmental Engineering and Science. Vol. 2, No. 1, 2015, pp. 1-8. Abstract Spain is the country with the largest citrus production in Europe, with about 5 million tons per year (2007), 3.5 million tons corresponding to oranges. Worldwide, only Brazil, China and the US have a higher annual production, with 21, 19, and 10 million tons, respectively. This pruned fraction of orange tree is an important lignocellulosic resource and consists of leaves, bark and young branches. The most obvious application is in combustion processes, but it is also feasible to use in conversion processes to products with high added value, including the production of cellulose pulp by alternative methods. The aim of this work is the systematic study of valorisation of orange pruning by applying an organosolv method (ethanol) for the production of cellulose pulps by assessing the effect of the most influential treatment variables on the characteristics of the pulps by means of factorial designs and statistical analysis of the results. The values estimated by the polynomial models reproduce the experimental results of the different dependent variables, with errors less than 22%, while by the neurofuzzy models the error are less than 20%. Although neurofuzzy models use a greater number of parameters than second order polynomial models, both types of models are equally effective for a typical experimental design of three variables of operation, with a total of 15 experiments. 1. Introduction In addition to other treatments for lignocellulosic materials used to produce pulp, in recent decades organosolv processes have attracted significant interest. Both for legal reasons such as social pressure, the implementation of production processes pulp tends increasingly towards production methods that minimize environmental impact. At the same time, global demand for cellulosic products has increased significantly. Therefore, research on new methods and materials becomes urgent, in some scenarios (Muurinen, 2000). In the scientific literature there is a wide variety of processes applicable to plant materials to leverage its components that try to avoid the environmental problems associated with current methods, particularly those associated with the Kraft process to be the most widely established throughout the world (Jiménez and Rodríguez, 2010). Within these methods are organosolv, producing the delignification of the raw material with the aid of organic solvents, often in softer conditions than the Kraft process itself, and that, moreover, do not use sulphur compounds in cooking with subsequent environmental benefits.