Experimental Assessment on the Enzymatic Hydrolysis of Hydrothermally Pretreated Eucalyptus globulus Wood Aloia Romanı ´, Gil Garrote,* Jose ´ Luis Alonso, and Juan Carlos Parajo ´ Department of Chemical Engineering, Faculty of Science, UniVersity of Vigo (Campus Ourense), As Lagoas, 32004 Ourense, Spain, and CITI (Centro de InVestigacio ´n, Transferencia e InnoVacio ´n), UniVersity of Vigo, Tecnopole, San Cibrao das Vin ˜as, Ourense, Spain Hydrothermal processing of Eucalyptus globulus wood was evaluated as a pretreatment for bioethanol manufacture, to assess the effects caused by the severity of wood solubilization, composition of liquors, and spent solid and suceptibility of substrates toward enzymatic hydrolysis. A factorial design was employed to assess the hydrolysis kinetics of spent solids. The liquor to solid ratio (6-20 g/g) and the enzyme to substrate ratio (8-28 FPU/g) were considered as independent variables. Generalized empirical models suitable for predicting glucose concentrations were developed. Using selected substrates and hydrolysis conditions (LSR ) 6 g/g, ESR ) 28 FPU/g), media containing more than 82 g glucose/L were obtained (with cellulose-to- glucose conversions of 75-80%), whereas cellulose-to-glucose conversions of 80-100% were obtained under a variety of experimental conditions. 1. Introduction The use of second-generation bioethanol, derived from lignocellulosic materials (LCM), is one of the possible responses to the challenge posed by greenhouse gas emissions. Considered as raw materials for manufacturing biofuels, LCM is the only feedstock available in amounts enough to enable the replacement of a significant amount of fossil fuels. 1–3 The structural components of LCM are cellulose (a crystalline, linear polymer consisting of D-glucose molecules bound together by -1,4-glycosidic linkages), hemicelluloses (an amorphous, heterogeneous group of branched polymers, with building blocks such as glucose, galactose, mannose, xylose, or arabinose, which can be substituted by uronic acids, acetyl groups, or esterified phenolic acids), and lignin (a highly complex, three-dimensional polymer made up of different phenyl-propane units). Obtaining glucose solutions from LCM requires one first to alter the cell wall structure by chemical or physical pretreat- ments, which can be costly; 4–7 in fact, pretreatment has been considered as the most expensive individual stage in bioethanol production. 8 The development of an effective and cost-efficient pretreatment is one of the main challenges involved in the large- scale production of second-generation bioethanol production. 9 An ideal pretreatment could be defined by the following features: 6,10–13 (i) simple and economical structure (ii) limited requirements of energy, water, and process chemicals (iii) ability to break the structure of the LCM (iv) reduced polysaccharide losses (v) maximal production of valuable hemicellulose-derived products, with limited generation of undesired degradation compounds (vi) generation of spent, celullose-containing solids with high susceptibility toward enzymatic hydrolysis (vii) minimal generation of processing wastes In this context, an efficient utilization of LCM can be achieved using the “biorefinery” 14 or “forest refinery” approach: 15 the LCM is subjected to sequential treatments enabling the separation of their major components (as cellulose, hemicelluloses, and lignin) in different streams, which can be individually processed to obtain commercial products with minimal waste generation. Treatments with hot, compressed water (autohydrolysis or hydrotermal processing) 16 have been proposed for LCM frac- tionation. In these treatments, an aqueous suspension of LCM, without the addition of external chemicals, is heated to cause a controlled hydrolytic degradation of hemicelluloses (by means of reactions catalyzed first by the hydronium ions from water autoionization and, in later reaction stages, by hydronium ions coming from organic acids generated in situ). When autohy- drolysis is carried out under suitable conditions, the liquid phase is rich in hemicellulose-derived products, 17 including sugar oligomers and sugars, whereas the solid phase (mainly composed of cellulose and lignin) is suitable for further conversion according to the biorefinery concept (for example, by enzymatic hydrolysis, 18–20 leading to glucose solutions suitable for bioet- hanol production 9,21–23 and to a solid fraction mainly made up of lignin). The objective of this work is to provide an experimental assessment of the suitability of nonisothermal autohydrolysis Eucalyptus globulus wood for causing the selective solubiliza- tion of hemicelluloses, and for providing treated solids with enhanced susceptibility toward enzymatic hydrolysis. The autohydrolysis-enzymatic hydrolysis approach results in the separation of the three major components of wood (as soluble hemicellulose-derived products from hydrothermal processing, glucose solutions derived from the cellulose contained in spent solids and exhausted solids from the enzymatic hydrolysis step mainly made up of lignin). The experimental plan provides an insightful assessment of both the effects of pretreatments and the enzymatic hydrolysis of autohydolyzed solids, with a quantitative assessment of the effect of the major operational variables (enzyme loading and liquid-to-solid ratio). 2. Materials and Methods 2.1. Raw Material. Eucalyptus globulus wood samples were obtained from a local pulp factory (ENCE, Galicia, NW Spain), milled to pass an 8 mm screen, air-dried, homogenized in a single lot to avoid differences in composition among aliquots, and stored in a dark and dry place until use. * To whom correspondence should be addressed. Phone: +34988387075. Fax: +34988387001. E-mail: gil@uvigo.es. Ind. Eng. Chem. Res. 2010, 49, 4653–4663 4653 10.1021/ie100154m  2010 American Chemical Society Published on Web 04/23/2010