Chemical Engineering Science 65 (2010) 66--73 Contents lists available at ScienceDirect Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces Bimetallic Co–Ni/Al 2 O 3 catalyst for propane dry reforming: Estimation of reaction metrics from longevity runs Faisal M. Althenayan a , Say Yei Foo a , Eric M. Kennedy b , Bogdan Z. Dlugogorski b , Adesoji A. Adesina a, ∗ a Reactor Engineering and Technology Group, School of Chemical Sciences and Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia b Priority Research Centre for Energy, University of Newcastle, Callaghan, New South Wales 2308, Australia ARTICLE INFO ABSTRACT Article history: Received 2 July 2008 Received in revised form 6 February 2009 Accepted 16 March 2009 Available online 2 April 2009 Keywords: Catalysis Carbon deactivation Propane dry reforming Mathematical modelling Parameter identification Reaction engineering Dry reforming of hydrocarbons is accompanied by carbon deposition making it difficult to unambiguously estimate the true reaction metrics (rate constant, yield and selectivity) without the masking effect of coke formation. This study employed a method originally proposed by Levenspiel [1999. Industrial & Engineering Chemistry Research 38, 4140–4143] to determine the intrinsic reaction rate simultaneously with the carbon-induced deactivation coefficient from transient rate data over an extended period of time (up to 72 h), for propane dry reforming over a Co–Ni catalyst at 823–973 K. The rate constant k ′ and deactivation coefficient, k d were determined from a fit of the concentration history data to the hyperbolic reaction–deactivation model for 1st-order kinetics in a plug flow reactor. However, the product H 2 : CO ratio was generally invariant with time over the 3-day period for different CO 2 :C 3 H 8 feed ratio values (4–7) but remained within a band of between 0.4 and 0.6. Both k ′ and k d exhibited a negative order dependency on the CO 2 :C 3 H 8 ratio at -0.575 and -2.39, respectively. Arrhenius treatment of these two reaction metrics also yielded activation energy estimates of 92.3 and 164.4 kJ mol -1 for the true reforming reaction and deactivation process, respectively. Catalyst characterization was carried out using XRF, liquid N 2 adsorption, XRD, H 2 chemisorption, tem- perature programmed desorption of NH 3 and CO 2 , temperature-programmed reduction (with H 2 ) and oxidation (with air) as well as solid TOC content analysis. © 2009 Elsevier Ltd. All rights reserved. 1. Introduction Hydrocarbon dry reforming has been receiving considerable at- tention in recent literature as an alternative route for synthesis gas (H 2 /CO mixture) production (Pena et al., 1996; Raberg et al., 2007; Castro-Luna and Iriarte, 2008) because it yields relatively low H 2 : CO product ratio ( < 2) which is well-suited to downstream usage in GTL fuels production and the manufacture of olefin-rich feedstock for the petrochemical industries (Xu et al., 1999; Wilhelm et al., 2001). The attendant reduction in overall green-house gas (GHG) emission from the natural gas processing plant due to CO 2 utilization is a fur- ther advantage of the dry reforming route. The reaction is conventionally carried out over a Ni catalyst at temperatures above 773 K. However, the accompanying car- bon deposition often leads to catalyst deactivation and patholog- ical reactor operation (such as bed clogging and excessively high pressure drop). Since both dry reforming and carbon deposition ∗ Corresponding author. Tel.: +61 2 93855268; fax: +61 2 93855966. E-mail address: a.adesina@unsw.edu.au (A.A. Adesina). 0009-2509/$ - see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.ces.2009.03.037 take place concurrently, the true reforming reaction rate is con- tinuously masked by on-line catalyst activity decay which is non- separable from the intrinsic kinetics. Levenspiel (1999) proposed a reaction–deactivation coupled rate analysis which permits ex- traction of the intrinsic rate constant as well as the deactivation coefficient from transient concentration profile of the reactant(s) for 1st order reaction kinetics taking place in a plug-flow or stirred- tank reactor for different activity decay laws. This analysis was subsequently generalised and extended to other schemes by Hardiman et al. (2005). In what follows, this method is used to anal- yse the CO 2 reforming of propane which satisfies the basic features present in coke-induced deactivation of hydrocarbon-mediated reactions. Experiments were conducted in fixed bed reactor containing par- ticles of 5Co:10Ni alumina-supported catalyst. The dry reforming of propane is given by 3CO 2 + C 3 H 8 → 6CO + 4H 2 (1) with carbon deposition proceeding via propane dehydrogenation C 3 H 8 → 2C + CH 4 + 2H 2 (2)