Published: March 03, 2011 r2011 American Chemical Society 3818 dx.doi.org/10.1021/ie101946c | Ind. Eng. Chem. Res. 2011, 50, 3818–3828 ARTICLE pubs.acs.org/IECR Kinetics of Aqueous Extraction of Hemicelluloses from Spruce in an Intensified Reactor System Henrik Gr  enman, † Kari Er € anen, † Jens Krogell, ‡ Stefan Willf€ or, ‡ Tapio Salmi, † and Dmitry Yu. Murzin* ,† † Laboratory of Industrial Chemistry and Reaction Engineering, Process Chemistry Centre, Department of Chemical Engineering, Åbo Akademi University, Biskopsgatan 8, FI-20500 Åbo/Turku, Finland ‡ Laboratory of Wood and Paper Chemistry, Åbo Akademi Process Chemistry Centre, Porthansgatan 3, FI-20500 Åbo/Turku, Finland ABSTRACT: The aqueous extraction of hemicelluloses has gained increasing interest with new emerging applications for hemicelluloses in the modern forest-based bioreﬁnery concept. The extraction kinetics play a key role in their industrial utilization. The traditional kinetic studies and models for the selective dissolution of softwoods, however, always incorporate high concentrations of sodium hydroxide and sodium sulﬁde relevant to pulping, and the kinetics in pure water is left outside the scope of these investigations. Aqueous extraction of hemicelluloses from spruce sapwood was investigated with a new cascade reactor setup, which was developed for intensiﬁed investigation of solid-liquid dissolution kinetics. The experiments were performed at 150-170 °C with a particle size of 1.25-2 mm and solid loads of about 6.25 g of dry wood/L in the kinetic regime of intrinsic kinetics. The pH of the liquid phase was measured during the reaction. The selectivity of the dissolution and degradation of hemicelluloses was examined qualitatively. The total concentrations of arabinose (Ara), xylose (Xyl), galactose (Gal), glucose (Glc), mannose (Man), rhamnose (Rha), glucuronic acid (GlcA), 4-O-methylglucuronic acid (4-O-MeGlcA), and galacturonic acid (GalA) were analyzed from the liquid and solid phases during the reaction. The dissolution was observed to be highly temperature dependent, and degradation of the compounds was observed. A kinetic model was developed for the overall extraction of the hemicelluloses. The activation energy was determined to be 135 kJ/mol. No signiﬁcant inﬂuence of the pH on the actual extraction kinetics was observed, even though it inﬂuences strongly the degradation of the compounds through hydrolysis. 1. INTRODUCTION Solid-liquid reactions are involved to a large extent in the chemical industry of today. Many ﬁelds of industry, e.g., hydro- metallurgy, fuel production, agrochemicals, and pharmaceuticals, employ solid-liquid reactions at some point of production, which explains the vast research in the ﬁeld. The research of the reaction kinetics has long been a subject of extensive investigations due to its importance in the design and optimiza- tion of the production processes. A large sector of industry, which is dependent almost solely on a solid-liquid process in production, is the chemical pulping industry, where the ultimate aim is to remove lignin, but unfortunately for the production of pulp, a large part of the hemicelluloses are dissolved simultaneously, too. The basic production methods used commonly today, i.e., the kraft process and soda pulping, date back to the 19th century, to the days long before chemical engineering made its breakthrough in the modeling and optimization of industrial production. With the development of chemical engineering, these processes have been optimized and enhanced tremendously. However, some aspects still rely partly on empirically developed experimental optimiza- tion without a comprehensive understanding of the underlying chemical and physical phenomena. One of these aspects is the kinetics of the selective dissolution of wood chips in the cooking liquor (often called deligniﬁcation), which is the core process in pulp production. There are several reasons for this: the kinetics of solid-liquid reactions is more diﬃcult to investigate than the kinetics of homogeneous systems due to such factors as the determination of surface areas, combined control of diﬀusion and intrinsic kinetics, and the heterogeneity of the materials. Moreover, the chemistry of wood is very complex, comprising interlinked macromolecules which are cleaved into numerous smaller fragments during the process. The shear number of di ﬀerent compounds makes it impossible to obtain reaction models consisting of elementary steps only, but still, improvement can be made in the understanding of the overall process. However, some works have been published in which models have been proposed for the inﬂuence of a certain parameter on the extraction rate or with the aim of modeling the overall reactions. 1-11,14-28 Probably the best known one is the H-factor model, which describes the eﬀect of temperature on the extent of deligniﬁcation. 1 More complex models were developed later on to incorporate the inﬂuence of the concentrations of diﬀerent compounds. A common methodology has been to divide the dissolution process or the dissolved components at diﬀerent stages into categories, e.g., initial, bulk, and residual. 2-7 Diﬀerent reaction orders have been determined for the diﬀerent phases in the diﬀerent regimes of the extraction. Some of the models assume that the diﬀerent phases (of lignin and carbohydrates) are present in the wood from the beginning of the reaction and that they react simultaneously, resulting in parallel reactions, while others assume that the phases are formed during the extraction, resulting in consecutive reaction kinetics. Another approach used to describe the extraction process is to use time- or conversion- dependent rate constants, which basically leads to the Received: September 21, 2010 Accepted: February 8, 2011 Revised: February 1, 2011