Onset of convection controlled by N2 impurity during CO2 storage in saline aquifers
Introduction
The high cost of CO2 capture and purification from the flue gas is a fundamental barrier to CO2 sequestration projects, especially for conventional power plants (Nicot et al., 2013; Wei et al., 2015). Flue gas from the power plants contains impurities along CO2 and in this regard, N2 is the dominant impurity (Wilkinson et al., 2001; Pipitone and Bolland, 2009; Kather and Kownatzki, 2011; Porter et al., 2015; Mahmoodpour et al., 2018); therefore, the possibility of co-injection of the CO2- N2 mixture can greatly reduce the cost of the project (Zhang et al., 2011). Dissolution of the injected gas into brine is an important trapping mechanism, which insures permanent CO2 trapping through mineralization (Javaheri et al., 2010). Diffusion of CO2 molecules into brine initiates the dissolution process (Soltanian et al., 2016, 2017). The resulting diffusive layer is denser than the formation brine (presence of the denser fluid over the lighter one), leading to density-driven instabilities (Soltanian et al., 2017). As these instabilities grow, they migrate toward the formation bottom through convection. The activation of density-driven convection is an important aspect of dissolution process, because it accelerates the process (Seyyedi et al., 2014). However, although N2 does not have the hazardous effect of SO2 and NO2 gases around the injection well (Talman, 2015), there is a concern that it impedes the onset of convection (Li et al., 2015). Some theoretical work has estimated delayed onset of convection for the N2 impurity case (Li and Jiang, 2014; Li et al., 2015). However, in spite of its importance, there is no reported experimental investigation to examine the effect of the N2 impurity on the onset of convection.
The aim of this paper is to examine the onset of instability and density-driven convection in pure CO2 and CO2-N2 cases. Analysis has been done through dimensionless parameters; therefore, the introduced scaling relations can be used in similar systems. To the best of our knowledge, a study of this kind in a pressurized Hele-Shaw cell has not been reported yet, especially for the CO2- N2 case, for which there is no experimental examination even in a low pressure Hele-Shaw. We report the first insights regarding the effect of N2 impurity on the diffusion coefficient of CO2, which may also exist for other types of impurities. The experimental and theoretical results show the considerable effect of this impact and call for more studies to examine the behavior of ternary systems.
The difficulties in attaining a high-pressure visual setting to study CO2-brine systems have motivated researchers to use analog fluids (such as propylene glycol- water system instead of the CO2-water or brine) systems (Ennis-King and Paterson, 2005; Neufeld et al., 2010; Backhaus et al., 2011; MacMinn et al., 2012; Taheri et al., 2012; MacMinn and Juanes, 2013; Slim et al., 2013; Tsai et al., 2013; Agartan et al., 2015). However, studies point out that the resulting non-monotonic density profile in analog fluids may mean that analog fluids do not mimic the dynamics of CO2-brine system (Jafari-Raad and Hassanzadeh, 2015). Other authors used a low-pressure Hele-Shaw set-up to examine CO2-water (or brine) systems (Kneafsey and Pruess, 2010, 2011; Faisal et al., 2013), in which there is no accurate quantitative measurement of the progress of the process. Other types of tests have been conducted in a blind PVT cell, in which pressure decay data of gas were gathered (Yang and Gu, 2006; Farajzadeh et al., 2007, 2009; Nazari Moghaddam et al., 2012, 2015; Seyyedi et al., 2014). It is obvious that visual examination is impossible in the high-pressure tests mentioned so far. Thus, Taheri et al. (2017) and Khosrokhavar et al. (2014) combined qualitative and quantitative measurements in their studies to obtain a more complete understanding of the CO2 dissolution in brine. However, Taheri et al. implemented Hele-Shaw’s experiment in atmospheric pressure, in which even small errors could change the overall behavior of the system, while Khosrokhavar et al. used Schlieren’s method in the absence of porous media; in this case therefore, the high permeability of bulk brine means that the onset of convection is not detectable in the pressure curve (Khosrokhavar et al., 2014; Taheri et al., 2017).
For the first time, we have gathered the synchronic qualitative (image) and quantitative data (pressure of system) of gas mixture dissolution into brine for pure and impure cases under pressure to obtain the onset of convection. The results were analyzed, and the effective diffusion coefficient of CO2 estimated. Based on insights from the experimental data, a linear stability analysis and high resolution numerical simulation were then conducted to gain a whole understanding of the process.
Linear stability analysis (LSA) is widely used to characterize the stability of the diffusive boundary layer in CO2 dissolution (Riaz et al., 2006; Emami-Meybodi and Hassanzadeh, 2013; Emami-Meybodi, 2017). As the presence of N2 would reduce the solubility of CO2 in brine (Ziabakhsh-Ganji and Kooi, 2012), previous studies have considered only these changes of solubility in the presence of impurities and neglected the diffusion coefficient variation (Li et al., 2015; Li and Jiang, 2014). In this condition, for the N2 case, they concluded that it postpones the onset of convection. Jafari-Raad and Hassanzadeh found that the non-monotonic density profile resulting from different diffusion coefficients of the components in the impure gas stream and used this insight in their next study to analyze the effect of the H2S or N2 impurity on the onset of convection mixing (Jafari-Raad and Hassanzadeh, 2015, 2016). Kim and Song proposed new stability limits in a similar examination of the non-monotonic density profile (Kim and Song, 2017). In these previous studies, N2 received less attention, due its negligible solubility into brine. Since there is no experimental work concerning the CO2- N2 mixture (or for other impure situations), researchers have been using the same diffusion coefficient for CO2 in both pure and impure situations. However, experimental and theoretical investigations on ternary systems suggest that the diffusion coefficient of a component is also affected by the mole fraction of other components and that this change is considerable (Leahy‐Dios and Firoozabadi, 2007; Liu et al., 2012; Legros et al., 2015; Allie-Ebrahim et al., 2017). Based on the available evidence, binary system fluxes may show different values from a multi-component mixture. Direct measurement and data analysis of diffusion coefficient values in ternary systems is very complex and it requires special equipment (only a limited amount of data for special mixtures is available in the literature). In this study, we have tried to make an estimation of effective diffusion coefficient based on the total flux (Cadogan, 2015). We followed the approach of Jafari-Raad and Hassanzadeh (2016) and Kim and Song (2017) by considering the effective diffusion coefficient from experiments and compared the LSA results with theirs (the cases which are highlighted with “i” as the superscript in the results section).
Direct numerical simulation was implemented to examine the instabilities after creation of them. An extensive body of published works is available for pure CO2 case (Hassanzadeh et al., 2005; Neufeld et al., 2010; Pau et al., 2010; Slim, 2014; Emami‐Meybodi et al., 2015; De Paoli, 2016, 2017; Islam et al., 2016; Soltanian et al., 2016, 2017). In the limited studies of impure cases (CO2+ N2 or H2S or SO2), most of the researchers considered the same diffusion coefficient for the impurity as the CO2 (Li and Jiang, 2014; Li et al., 2015; Mahmoodpour et al., 2017). We considered the effective diffusion coefficient during our simulation studies and compared the results with recent studies where effective diffusion coefficient was not considered.
This paper is organized as follows: In Section 2, the designed Hele-Shaw cell, the experimental set-up and methodology are presented. The results are examined in Section 3 and the scaling relations are introduced. Finally, the research findings are summarized, and conclusions are drawn in Section 4.
Section snippets
Methodology
From start of the experimental tests until a rapid decline in the pressure curves (resulting from the convective flow), we face with three different phenomena: onset of instability, first appearance of the convective fingers, and onset of convection. In this study, we use theoretical, direct numerical simulation, and experimental tests to capture these times. Also, we examine these times through non-dimensional relations to find a possible trend between them. CO2 (or N2−CO2) dissolution starts
Results and discussion
In this section we first, discuss the experimental pressure data for Experiment 1 and then provide results for all the experiments. Fig. 3 shows pressure vs. time for Experiment 1. Since we are interested in the onset of convection, the early time of the pressure data is examined in more detail. Because of pressure fluctuations at early times, we discarded pressure data for the first 6 min, point a in Fig. 3, of each experiment. The next point on the pressure curve, point b, represents the
Summary and conclusions
We have conducted a series of visual experiments at high-pressure conditions to examine the effects of N2 as an impurity along CO2 on the onset of convection. Experimental tests showed earlier onset of convection for 90%CO2- 10%N2 mixtures in comparison to pure CO2. On the other hand, the onset of convection for an impure CO2 mixture with 20% N2 was close to that of the pure CO2 system. Analysis of the diffusion-dominant regime of the experimental tests revealed that the presence of N2 reduces
References (72)
- et al.
Thermodynamic description of aqueous nonelectrolytes at infinite dilution over a wide range of state parameters
Geochim. Cosmochim. Acta
(2003) - et al.
An improved model calculating CO 2 solubility in pure water and aqueous NaCl solutions from 273 to 533 K and from 0 to 2000 bar
Chem. Geol.
(2003) - et al.
Does impure CO 2 impede or accelerate the onset of convective mixing in geological storage?
Int. J. Greenh. Gas Control.
(2016) Visualization and investigation of natural convection flow of CO 2 in aqueous and oleic systems
J. Pet. Sci. Eng.
(2014)- et al.
Effect of impurities on the onset and growth of gravitational instabilities in a geological CO2 storage process: linear and nonlinear analyses
Chem. Eng. Sci.
(2017) - et al.
Laboratory experiments and numerical simulation studies of convectively enhanced carbon dioxide dissolution
Energy Procedia
(2011) Numerical analyses of the effects of nitrogen on the dissolution trapping mechanism of carbon dioxide geological storage
Comput. Fluids
(2015)- et al.
A numerical study of the impurity effects of nitrogen and sulfur dioxide on the solubility trapping of carbon dioxide geological storage
Appl. Energy
(2014) - et al.
Design-of-experiment-based proxy models for the estimation of the amount of dissolved CO2 in brine: a tool for screening of candidate aquifers in geo-sequestration
Int. J. Greenh. Gas Control.
(2017) Potential subsurface impacts of CO 2 stream impurities on geologic carbon storage
Energy Procedia
(2013)
Assessment of the different parameters affecting the CO2 purity from coal fired oxyfuel process
Int. J. Greenh. Gas Control
High-resolution simulation and characterization of density-driven flow in CO 2 storage in saline aquifers
Adv. Water Resour.
Power generation with CO2 capture: technology for CO2 purification
Int. J. Greenh. Gas Control
The range and level of impurities in CO2 streams from different carbon capture sources
Int. J. Greenh. Gas Control
Qualitative and quantitative experimental study of convective mixing process during storage of CO 2 in homogeneous saline aquifers
Int. J. Greenh. Gas Control.
Subsurface geochemical fate and effects of impurities contained in a CO 2 stream injected into a deep saline aquifer: what is known
Int. J. Greenh. Gas Control.
Experimental study of CO2 convective dissolution: the effect of color indicators
Int. J. Greenh. Gas Control
Geochemical impact of aquifer storage for impure CO2 containing O2 and N2: tongliao field experiment
Appl. Energy
An equation of state for thermodynamic equilibrium of gas mixtures and brines to allow simulation of the effects of impurities in subsurface CO 2 storage
Int. J. Greenh. Gas Control.
Experimental study on effects of geologic heterogeneity in enhancing dissolution trapping of supercritical CO2
Water Resour. Res.
Maxwell-Stefan diffusion coefficient estimation for ternary systems: an ideal ternary alcohol system
J. Chem. Soc. Faraday Trans.
Convective instability and mass transport of diffusion layers in a Hele-Shaw geometry
Phys. Rev. Lett.
Diffusion of CO2 in Fluids Relevant to Carbon Capture, Utilisation and Storage
Dispersion-driven instability of mixed convective flow in porous media
Phys. Fluids
CO2 dissolution in the presence of background flow of deep saline aquifers
Water Resour. Res.
Stability analysis of two-phase buoyancy-driven flow in the presence of a capillary transition zone
Phys. Rev. E
Role of convective mixing in the long-term storage of carbon dioxide in deep saline formations
Spe J.
Enhanced mass transfer of CO2 into water and oil by natural convection
EUROPEC/EAGE Conference and Exhibition
Enhanced mass transfer of CO2 into water: experiment and modeling
Ind. Eng. Chem. Res.
Experimental and numerical studies of density driven natural convection in saturated porous media with application to CO2 geological storage
Energy Procedia
Molecular modeling of diffusion coefficient and ionic conductivity of CO2 in aqueous ionic solutions
J. Phys. Chem. B
Density of aqueous solutions of CO2
Lawrence Berkeley National Laboratory
Elementary Applied Partial Differential Equations
Modelling of convective mixing in CO storage
J. Can. Pet. Technol.
Scaling behavior of convective mixing, with application to geological storage of CO2
Aiche J.
Reactive transport modeling of the enhancement of density-driven CO2 convective mixing in carbonate aquifers and its potential implication on geological carbon sequestration
Sci. Rep.
Cited by (22)
Numerical modeling for CO<inf>2</inf> storage with impurities associated with enhanced gas recovery in depleted gas reservoirs
2022, Journal of Natural Gas Science and Engineering