Computers & Geosciences 30 (2004) 543–552 Adapting TOUGH2 for general equations of state with application to geological storage of CO 2 $ S.J. Talman a, *, J.J. Adams b,1 , R.J. Chalaturnyk a a Department of Civil Engineering, University of Alberta, Edmonton, Alta., Canada T6G 2E3 b Alberta Geological Survey, 4th floor 4999 98 Avenue, Edmonton, Alta., Canada Received 28 January 2003; received in revised form 14 January 2004; accepted 29 January 2004 Abstract The currently distributed version of the heat and mass flow model, TOUGH2, requires that pressure be a primary variable. This may limit the functionality and efficiency of the program, both in terms of convergence properties and ease of programming. TOUGH2 was adapted to allow for a general choice of independent (primary) variables. The changes to the code include increasing the size of the secondary parameter array for storage of pressure and changing the references to pressure in the flow equation module. Testing showed that the new code accurately replicates results of the original code. The program was applied to problems of relevance to geological storage of CO 2 using the modified code (D-EOS), and the original code (P-EOS), as primary variables. Two scenarios, addition of heat and addition of mass, were simulated using both versions of TOUGH2. The simulations produced very similar results; however, in most instances, the density-based routine solved the problems faster and satisfied analytical constraints placed on the problems more accurately than the pressure-based runs. It was generally the case that the greater the modelled pressure gradients the better D-EOS performed. The new formulation adds flexibility to TOUGH2, both in terms of improved convergence behaviour and ease of programming. The new flexibility to choose any primary variables which uniquely specify the equilibrium state of a system allows for increased efficiencies in coding EOS routines, and may have important consequences in optimization of the solution of the flow equations for a given problem. r 2004 Elsevier Ltd. All rights reserved. Keywords: Numerical modelling; TOUGH2; Equations of state; Primary variables; Span and Wagner EOS 1. Introduction Storage of anthropogenic CO 2 is becoming accepted (Freund, 2002) as a potential means of making significant reductions in greenhouse gas emissions. The main storage options are: storage in geological reser- voirs, storage in the deep ocean, and storage by conversion into a different material or purpose-built storage. Geological storage, such as enhanced oil and gas recovery, enhanced coal bed methane production, and disposal into saline aquifers and depleted oil/gas reservoirs, is currently the most technologically feasible option (Bachu and Stewart, 2002). Recently, in response to greenhouse gas emission issues, TOUGH2 has been adapted for modelling the fate of disposed CO 2 in the subsurface (Cole, 2000; Weir et al., 1995). This is a well-known three-dimensional simulator of heat and mass transport. Previous applica- tions have demonstrated the program’s versatility; it has ARTICLE IN PRESS $ Code available from server at http://www.iamg.org/CGE- ditor/index.htm. *Corresponding author. Tel.: +1-780-492-8099; fax: +1- 780-492-8198. E-mail address: stalman@ecn.ab.ca (S.J. Talman). 1 Current address: Department of Geology and Geophysics, University of Calgary, 2500 University Dr. NW, Calgary, Alta., Canada. 0098-3004/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.cageo.2004.01.004