New correlations predict aqueous solubility and density of carbon dioxide Alireza Bahadori a, *, Hari B. Vuthaluru a , Saeid Mokhatab b a Department of Chemical Engineering, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia b Gas Engineering Consultant, Vancouver, BC, Canada 1. Introduction The sequestration of anthropogenic carbon dioxide (CO 2 ) into geological formations has been considered as a potential method to mitigate climate change. Accurate evaluation of the capacity of a saline aquifer for CO 2 sequestration, and the fate of the injected fluids in sedimentary basins needs precise representation of brine and CO 2 PVT data (Hassanzadeh et al., 2008). In the petroleum industry, compositional reservoir simulators use equation of state (EOS) thermodynamic models to calculate the phase equilibrium properties of fluid mixtures (Chang et al., 1998). Although these kinds of thermodynamic models are well suited for compositional modeling of enhanced oil recovery (EOR) processes, the disadvan- tage of such models for the specific case of large-scale flow simulation of geological CO 2 storage in aquifers is that they represent computational ‘‘overkill’’ and are inappropriately expensive (Hassanzadeh et al., 2008). Indeed, there are a large number of experimental equilibrium data on the CO 2 –brine system that have been used to tune the equations of state for CO 2 and water under subsurface conditions. Accurate prediction of CO 2 solubility over a wide range of temperature, pressure and ionic strength (T–P–m) is important to the studies of the carbonate precipitation and to the tracing of the global carbon cycle (Butcher et al., 1992). There has also been extensive theoretical effort on modeling the CO 2 solubility in aqueous solutions. Although several models on CO 2 solubility have been published, few can predict CO 2 solubility in wide T–P–m range with accuracy close to experiment. For many temperature–pressure–composition conditions, CO 2 solubilities are still unknown. Based on the Peng–Robinson EOS, Henry’s law and the scaled-particle theory, Li and Ngheim (1986) presented a model (L–N model) to predict phase equilibrium of oil, gas and water/ brine mixtures. This model is also intended to calculate the solubility of CO 2 in pure water up to 473 K and in sodium chloride (NaCl) solutions with NaCl up to 4 molar. However, it is in general not accurate to predict the solubility of CO 2 in aqueous NaCl solutions. Harvey and Prausnitz (1989) developed an EOS (H–P model) to predict CO 2 solubility in pure water and in aqueous NaCl solutions at elevated pressures. However, it overestimates CO 2 solubility by more than 10–20% in NaCl solutions. Zuo and Guo (1991) extended the Patel–Teja EOS to phase equilibrium calculations for electrolyte solutions (Z–G model). The Z–G model is no more accurate than the H–P model for CO 2 solubility. For example, it underestimates CO 2 solubility by more than 12% in 20 wt.% NaCl solutions at high pressures (>1000 bar) and overestimated CO 2 solubility by more than 10% in 6 wt.% NaCl solution at moderate pressures. The precise knowledge of CO 2 solubility in water is essential prior to any modelling activity. Several thermodynamic models are available to analyse the solubility of CO 2 in aqueous solutions of alkanolamines and to correlate the equilibrium CO 2 loading. These models involve a large number of parameters and require more complicated and longer computations. Largely for environmental reasons, a great amount of work has been performed in order to determine the solubility of carbon International Journal of Greenhouse Gas Control 3 (2009) 474–480 ARTICLE INFO Article history: Received 26 July 2008 Received in revised form 30 October 2008 Accepted 27 January 2009 Available online 23 February 2009 Keywords: Carbon dioxide Density Aqueous solubility Greenhouse gas Correlation ABSTRACT In this paper, new correlations for predicting density and the solubility of carbon dioxide in pure water as well as the aqueous sodium chloride solutions are developed, where using the generated interaction parameters, the solubility model is applied to correlate the carbon dioxide solubilities in aqueous solutions for temperatures between 300 and 400 K and pressures from 50 to 700 bar. The correlation developed for predicting density of carbon dioxide accurately works for pressures between 25 and 700 bar and temperatures between 293 and 433 K. The results have been compared with the reported data and it was found that there is a good agreement between the prediction results and observed values. Crown Copyright ß 2009 Published by Elsevier Ltd. All rights reserved. * Corresponding author. Tel.: +61 8 9266 1782; fax: +61 8 9266 2681. E-mail address: Alireza.bahadori@postgrad.curtin.edu.au (A. Bahadori). Contents lists available at ScienceDirect International Journal of Greenhouse Gas Control journal homepage: www.elsevier.com/locate/ijggc 1750-5836/$ – see front matter . Crown Copyright ß 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijggc.2009.01.003