Kinetics of Eucalyptus globulus Delignification in a Methanol-Water Medium Miguel A. Gilarranz, Francisco Rodrı ´guez,* Aurora Santos, Mercedes Oliet, Fe ´ lix Garcı ´a-Ochoa, and Julio Tijero Departamento de Ingenierı ´a Quı ´mica, Facultad de Quı ´mica, Universidad Complutense, 28040 Madrid, Spain The kinetics of Eucalyptus globulus delignification in methanol-water pulping has been studied. A total of 17 isothermal runs at a liquor-to-wood ratio of 50 L kg -1 were carried out to develop the kinetic model describing the system. In a first series of experiments, eight models were considered to study the influence of temperature on the delignification rate. The most suitable model, which was discriminated according to statistical criteria, describes delignification as the consecutive dissolution of three lignin species: initial, bulk, and residual lignin, their content in wood being 10, 69, and 21%, respectively. Initial and residual delignification were considered as irreversible reactions and bulk delignification as reversible. The influence of hydrogen ion concentration was taken into account by means of a general power-law expression. The model proposed was validated by reproducing the experimental data from four runs carried out under nonisothermal conditions and a liquor-to-wood ratio of 7 L kg -1 , which are closer to industrial operating conditions. Introduction The pioneering studies of Kleinert 1-3 on organosolv pulping generated a great interest in delignification with organic solvents. In the last years, some organosolv processes such as Organocell, 4 ASAM, 5 ALCELL, 6 and MILOX 7 have been developed at pilot plant and indus- trial scale. However, more research is needed to achieve fully competing processes and a better understanding of organosolv pulping. Organosolv processes will not solve all the problems of existing processes, but they could complement the production from certain raw materials and could provide some advantages such as easy bleachability, lesser odorous emissions, and lower mill size. 8 Information about organosolv pulping can be found reviewed elsewhere. 9 Among the variety of solvents employed in organosolv pulping, methanol meets some interesting characteris- tics. Thus, it has a low cost when compared to other pulping solvents and its volatility enables easy recovery by distillation. Besides, the methanol generated during the pulping can compensate the losses. 10 So far, the scope of methanol has been the alkaline organosolv pulping, 8 although it was shown that pulps with low lignin content and acceptable viscosity can be obtained by the methanol acidic pulping. 11 The studies on the kinetics of delignification in methanol-water can provide a better understanding of delignification in this medium and can be a valuable tool for process control and optimization. The aim of this work is the proposal of a kinetic model for acid- autocatalyzed pulping of Eucalyptus globulus wood in a methanol-water medium. In a first series of experi- ments, several models are proposed to study the influ- ence of cooking temperature. Once the most suitable model according to statistical criteria, i.e., lower value of sum-of-squares residuals, is discriminated, the influ- ence of hydrogen ion concentration on the rate constant is established. Finally, the developed model is validated by reproducing experimental data obtained under noniso- thermal conditions. Delignification Kinetics Wood delignification was found to take place accord- ing to several mechanisms. 12 At the beginning, delig- nification is rapid, but after certain portions of lignin have been removed from wood the delignification pro- ceeds at a slower rate. Thus, for the sake of simplicity, it can be divided into three stages: initial, bulk, and residual. In the bulk stage most of the lignin is removed. It has been observed that in organosolv-autocatalyzed pulping the transition from the bulk to residual period takes place when 80-85% of the total lignin has been dissolved, whereas in acid-catalyzed systems the amount of lignin removed at this transition point is higher. 13 The residual stage is the slowest, since at this point the most reticulated fraction of lignin is removed. The dissolution of lignin is usually described by means of the following differential equation: 12 The studies on delignification usually employ two dif- ferent approaches. In the first of them, it is assumed that wood only contains one type of lignin. In this case, the different phases observed in delignification cor- respond to changes in the mechanism that controls the reaction rate of the overall process. 14,15 The activation energy obtained with this assumption varies between 66.5 and 78.2 kJ mol -1 . In the second hypothesis, the lignin in wood is supposed to be composed of several species (usually initial, bulk, and residual lignin) dis- solving at different rates. 3,16-23 The most usual approach is to assume that the different lignin species react consecutively according to the first-order kinetic model expressed by eq 1. The ranges reported for the activation * To whom correspondence should be addressed. Phone: 34- 91-3944246. Fax: 34-91-3944243. E-mail: teresap@eucmax.sim. ucm.es. dL dt )-kL )-k 0 e -E/(RT) L (1) 3324 Ind. Eng. Chem. Res. 1999, 38, 3324-3332 10.1021/ie990161f CCC: $18.00 © 1999 American Chemical Society Published on Web 07/30/1999