Modeling of the devolatilization kinetics during pyrolysis of grape residues Luca Fiori a,⇑ , Michele Valbusa a , Denis Lorenzi b , Luca Fambri b a Department of Civil and Environmental Engineering, Trento University, Via Mesiano 77, 38123 Trento (TN), Italy b Department of Materials Engineering and Industrial Technologies, Trento University, Via Mesiano 77, 38123 Trento (TN), Italy article info Article history: Received 7 June 2011 Received in revised form 8 September 2011 Accepted 26 September 2011 Available online 2 October 2011 Keywords: Thermo-gravimetric analysis (TGA) Distributed activation energy model (DAEM) Dynamic pyrolysis Devolatilization kinetics Grape marc abstract Thermo-gravimetric analysis (TGA) was performed on grape seeds, skins, stalks, marc, vine-branches, grape seed oil and grape seeds depleted of their oil. The TGA data was modeled through Gaussian, logistic and Miura–Maki distributed activation energy models (DAEMs) and a simpler two-parameter model. All DAEMs allowed an accurate prediction of the TGA data; however, the Miura–Maki model could not account for the complete range of conversion for some substrates, while the Gaussian and logistic DAEMs suffered from the interrelation between the pre-exponential factor k 0 and the mean activation energy E 0 – an obstacle that can be overcome by fixing the value of k 0 a priori. The results confirmed the capabilities of DAEMs but also highlighted some drawbacks in their application to certain thermodegradation exper- imental data. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction About seven million tons of grape marc is produced annually worldwide (OIV, 2008). This residue originating from vine-making consists of grape seeds, skins and stalks. Grape marc exploitation for energy purposes has been studied (Encinar et al., 1996; Celma et al., 2007; Özçimen and Ersoy-Meriçboyu, 2008; Xu et al., 2009; Fiori and Florio, 2010; Demiral and Ayan, 2011), but many of those studies dealt with the pyrolysis of grape marc or its con- stituents and not with thermo-gravimetric analyses (TGA). TGA data in a non-oxidizing atmosphere can be of interest for the de- sign of thermochemical conversion processes (slow pyrolysis, gas- ification in downdraft or updraft gasifiers) where grape residues are used. Therefore, the present work focuses on TGA under inert atmosphere for grape marc and its constituents. Tests with vine- branches, grape seed oil and defatted grape seed powder were also performed. The experimental data was modeled by means of a simple first- order single reaction model and three different distributed activa- tion energy models (DAEMs). DAEMs have already been successfully applied to agricultural residues (Sonobe and Worasuwannarak, 2008) and wood (Shen et al., 2011). The current study is thus a com- prehensive comparison among the models’ characteristics and performances. 2. Experimental method 2.1. Biomass samples and their preparation for TGA experiments Grape marc was collected immediately after wine-pressing, rapidly transferred to the laboratory, and stored at 20 °C. Before TGA experiments, the grape marc was dried in a ventilated oven at 65 °C for 12 h. Dried grape skins, seeds and stalks were obtained by separating the dried grape marc into its constituents utilizing vibrating sieves. Branches were collected in vineyards and dried in the laboratory by using a ventilated oven. Before TGA experiments, dried grape marc (substrate 1), grape skins (substrate 2) and stalks (substrate 3) were manually ground with mortar and pestle into small particles. Branches (substrate 4) were chopped in an electrical miller to a size of about 0.4–0.8 mm. Grape seeds (substrate 5) were provided dried by an alcohol producing company. A single whole seed was used in TGA experi- ments. After being milled with a blender, some seeds were ex- tracted with supercritical CO 2 as described by Fiori (2007). The extraction process resulted in grape seed oil (substrate 6) and oil-free grape seeds (substrate 7, ‘extracted seeds’). 2.2. Equipment and procedures The pyrolysis experiments were performed in a thermobalance Mettler TG50 at heating rates of 3, 10 and 30 °C/min, in the tem- perature range 40–700 °C and under a nitrogen flow of 200 ml/ min. The thermobalance had a sensitivity of 0.001 mg and 0.1 °C. 0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2011.09.113 ⇑ Corresponding author. Tel.: +39 0461 282692; fax: +39 0461 282672. E-mail address: luca.fiori@ing.unitn.it (L. Fiori). Bioresource Technology 103 (2012) 389–397 Contents lists available at SciVerse ScienceDirect Bioresource Technology journal homepage: www.elsevier.com/locate/biortech