The sorption-enhanced reaction process (SERP) concept, developed by Air Products and Chemicals Inc. (Hufton et al., 1999), offers potential for process simplification as well as prospective energy conservation. Hereby, numerous theoretical and experimental studies on this concept have been recently published, investigating different process conditions and/or configurations; see for example Ding and Alpay (2000), Waldron et al. (2001) and Johnsen et al. (2006).
In the present work, the flow of pneumatically conveyed CO2 adsorbent particles within a stationary SMR catalyst phase is proposed, with adsorbent regeneration performed in a separate unit. Hence, the reaction and regeneration stages are decoupled, resolving the problem of possible disparity between the adsorption and desorption kinetics. Furthermore, using pneumatic transport of a CO2 scavenger enables a steady-state (non–periodic) operation, characterised by additional flexibility towards optimisation, including potential for heat integration i.e. use of regenerated adsorbent as energy carrier into the adsorptive reactor. A detailed fundamental analysis and parametric study of this process has been recently presented (Koumpouras et al., 2006). A significant degree of conversion enhancement (up to 75%) has been shown at moderate temperatures (750K). Simulations results have also pointed out the importance of the interaction between the reaction and regeneration stages of the process.
Further theoretical studies on this newly proposed process are presented in this paper. Particularly, a model-based optimisation approach is adopted to identify the optimum process scheme in terms of energy utilisation. Alternative process schemes and designs are examined and compared. For example, the use of pre-reactor in order to approach equilibrium conversion before adsorption enhanced reaction or the use of monolith channel arrangements in which some channels are used for reactant processing, and others for heat supply purposes. A set of scaling rules for the proposed process is also developed. As a result, the optimal pilot-scale process system can be scaled to meet industrial production, while a satisfactory basic performance, e.g. conversion, recovery or product purity, is maintained. Finally, the derived scaling rules are computationally validated.
References
Y. Ding, E. Alpay, Chem. Eng. Sci. 55 (2000) 3929.
J.R. Hufton, S. Mayorga, S.Sircar, A.I.Ch.E Journal 45 (1999) 248.
K. Johnsen, H.J. Ryu, J.R. Grace, C.J. Lim, Chem. Eng. Sci. 61 (2006) 1191.
G.C. Koumpouras, E. Alpay, F. Stepanek, Chem. Eng. Sci. submitted (2006).
A. Stankiewicz, Chem. Eng. Process. 42 (2003) 137.