Sequestration In Aquifers Research Articles

Estimation of CO2-Brine Interfacial Tension Based on an Advanced Intelligent Algorithm Model: Application for Carbon Saline Aquifer Sequestration.

The emission reduction of the main greenhouse gas, CO2, can be achieved via carbon capture, utilization, and storage (CCUS) technology. Geological carbon storage (GCS) projects, especially CO2 storage in deep saline aquifers, are the most promising methods for meeting the net zero emission goal. The safety and efficiency of CO2 saline aquifer storage are primarily controlled by structural and capillary trapping, which are significantly influenced by the interactions between fluid and solid phases in terms of the interfacial tension (IFT) between the injected CO2 and brine at the reservoir site. In this study, a model based on the random forest (RF) model and the Bayesian optimization (BO) algorithm was developed to estimate the IFT between the pure and impure gas-brine binary systems for application to CO2 saline aquifer sequestration. Then three heuristic algorithms were applied to validate the accuracy and efficiency of the established model. The results of this study indicate that among the four mixed models, the Bayesian optimized random forest model fits the experimental data with the smallest root-mean-square error (RMSE = 1.7705) and mean absolute percentage error (MAPE = 2.0687%) and a high coefficient of determination (R2 = 0.9729). Then the IFT values predicted via this model were used as an input parameter to estimate the CO2 sequestration capacity of saline aquifers at different depths in the Tarim Basin of Xinjiang, China. The burial depth had a limited influence on the CO2 storage capacity.

Reciprocal cross‐correlation analysis of two‐phase seepage processes and reservoir heterogeneities in CO2 saline aquifer sequestration

AbstractWhen CO2 saline aquifer storage is carried out, the heterogeneity of reservoir rock is an important factor affecting CO2 transport, and the reservoir heterogeneity in numerical simulations is mainly manifested as the heterogeneity of the parameter field. Since the parameter distributions across the reservoir are not available with the existing probes, the stochastic finite element method is combined with a two‐phase flow model to establish an unconditional random field of permeability, and computations are performed using the Monte Carlo method. The permeability, CO2 maximum migration distance (Md) and CO2 sweep area (Sa) were analyzed for mutual correlation. The permeability correlation area affecting Md and Sa was obtained, and the changes in the correlation area under the coefficient of variation (Cv) and correlation length (λx) of the permeability field in the different reservoirs were analyzed. The kriging superposition approach (KSA) was subsequently used to estimate both the Md and Sa of the target reservoir by establishing conditional random fields based on the sampling parameters in regions with different correlations, resulting in errors of 0.66% for Md and 0.96% for Sa in the high correlation region and 4.86% and 3.12% for Md and Sa in the low correlation region, which suggested that the sampling results from the high correlation region were less biased in the estimation. Under limited sampling conditions, it is recommended that samples be collected in regions with high correlations to reduce the uncertainty of CO2 transport analysis due to unknown heterogeneity. © 2024 Society of Chemical Industry and John Wiley & Sons, Ltd.

Re-evaluation of CO<sub>2</sub> storage capacity of depleted fractured-vuggy carbonate reservoir

<p>Confronting the dual crises of energy supply-demand imbalances and climate change, carbon neutrality emerges as a vital strategy for China in mitigating resource and environmental constraints, while fostering technological advancement and sustainable growth. In the context of extensive hydrocarbon exploitation, the CO<sub>2</sub> storage capacity within depleted oil fields could be significantly underestimated in comparison to the prevalent practice of saline aquifer sequestration. In this study, we employ both theoretical and computational models to investigate the temporal (from microseconds to millennia) and spatial (spanning pore, Darcy, and hybrid scales) dynamics of CO<sub>2</sub> trapping mechanisms in post-depletion carbonate reservoir with fractured-vuggy systems. The multiscale storage efficiency factor is obtained from simulation results and substituted into the existing analytical models for calculating CO<sub>2</sub> storage volume in field cases, reappraising the carbon sequestration potential of fracture-vuggy carbonate. Drawing from comparative results, we discern that depleted carbonate can dissolve and mineralize more CO<sub>2</sub> than saline layer, despite the storage volume can be considerably less. The annual storage capacity per well of two geological systems are comparable. Under unfavorable geological conditions, the minimum unit storage capacity of carbonate reservoir exceeds that of saline aquifer. The study's discoveries offer fresh perspectives on reliable and efficient CO<sub>2</sub> geological storage, contributing to the reduction of atmospheric carbon emissions and advancing the utilization of underground resources and global energy transformation.</p>

Effects of Pore Geometry and Saturation on the Behavior of Multiscale Waves in Tight Sandstone Layers

AbstractGeometric heterogeneities in tight reservoir rocks saturated with a fluid mixture may exhibit different scale distribution characteristics. Conventional models of rock physics based on poroelasticity, which usually consider single‐scale pore structure and fluid patches, are inadequate for describing elastic wave responses. A major challenge is to establish the relationship between the wave response at different spatial scales and frequencies. To address this problem, three sets of observational data over a wide frequency range were obtained from a tight oil reservoir in the Ordos Basin, China. Ultrasonic measurements were made on eight sandstone samples at partial oil‐water saturation at 0.55 MHz. Data from six borehole measurements and seismic profiles were acquired and analyzed at about 10 kHz and 30 Hz, respectively. Analysis of the cast thin sections shows that dissolution pores and microcracks generally develop, with fractal dimensions of the pores ranging from 2.45 to 2.67 for the samples with porosities between 5.1% and 10.2%. Compressional wave velocity and attenuation were estimated from the observed data. The results show that the velocity dispersion from seismic to ultrasonic frequencies is 10.02%, mostly occurring between sonic and ultrasonic frequencies. The attenuation is stronger at higher oil saturation. The relationships between velocity, attenuation, and wavelength were established and can be used for further forward modeling and seismic interpretation studies. A partial saturation model has been derived based on effective differential medium theory and a double double‐porosity model, assuming that the medium contains fractal cracks and fluid patches. The effects of scale and saturation on wave responses are prevalent. Modeling results consistent with observed data show that the radii of cracks and fluid patches range from 0.1 μm to 2.8 mm, affecting ultrasonic, acoustic, and seismic attenuation. The multiscale data and proposed model quantify the relationship between fracture and fluid distributions and attenuation and could be useful for upscaling to the reservoir scale. The study helps improve the understanding of seismic wave propagation in partially saturated rocks, which has potential applications in seismic exploration, hydrocarbon production in reservoirs, and CO2 sequestration in aquifers.

Isothermal CO2 injection into water-saturated porous media: Lattice-Boltzmann modelling of pulsatile flow with porosity, tortuosity, and optimal frequency characterization

The Lattice Boltzmann Method (LBM) is used to simulate the isothermal injection of CO2 into a water-saturated, homogeneous porous medium. The complex multiphase flow is modeled using the 2D-pseudopotential Shan-Chen multiphase LBM. This work proposes a unique fluid-pulse technology similar to the one applied to enhance oil recovery with CO2 sequestration. This technology is different from classical CO2 sequestration in aquifers and, to our knowledge, has never been used before. The primary purpose is to investigate whether our LBM simulation allows the introduction of fluid-pulse CO2 injection patterns similar to those reported in the literature on hydrocarbon reservoirs and to evaluate the benefit of using such a technique compared to classical uniform injection. First, our simulations consider a transient homogeneous flow with uniform inlet velocity to determine the natural frequency of the instabilities triggered downstream of the interstitial pore array. Then, CO2 is injected at the inlet in a pulsating regime, varying the forcing frequency near and far from the natural frequency to determine its effect on CO2 penetration. Results demonstrate that a pulsatile injection effectively enhances CO2 sequestration in aquifers when forced with a characteristic frequency. However, this frequency resulting after CO2 is injected to displace the water differed from the natural frequency preliminarily obtained from the homogeneous water flow. Injecting at the characteristic frequency has proven an effective technique for boosting CO2 penetration into a water-saturated porous medium, with a significant enhancement estimated between 1.5% and 16% for a range of pulsatile amplitudes between 1% and 10%.

Study on influence factors of saline aquifer CO2 sequestration under the constraint of cap rock

In view of the practical problems faced by the CO2 sequestration in aquifers, this paper considers the influence of injection rate, water production rate, and other factors on CO2 injection amount, residual sequestration amount, dissolution sequestration amount, and pressure evolution under the constraint of cap rock breakthrough pressure is studied by using a numerical simulation method, and the sequestration potential is evaluated. The results show that: (1) Water production measures can improve CO2 injection, dissolution sequestration, and residual sequestration. (2) Low ratio of vertical permeability to horizontal permeability is conducive to the security and stability of CO2 sequestration in the long term. The research content provides support for the evaluation of CO2 saline aquifer sequestration potential and field practice.

Optimization of carbon dioxide dissolution in an injection tubing for geologic sequestration in aquifers

Dissolution of liquefied carbon dioxide in a turbulent tubing flow for geologic sequestration in aquifers is simulated. The problem is solved by a two-fluid approach in a three dimensional formulation. For accurate calculation of the droplet dissolution rate, an evolution of droplet size distribution along a tubing is modelled by the population balance equation accounting for droplet breakup, coalescence and interphase mass transfer. The dissolution rate in a horizontal tubing is rather slow due to gravity-induced droplet stratification. The dissolution process in a horizontal tubing is compared with that in a coiled tubing wound on a horizontal reel. In a coiled tubing flow, droplet stratification significantly decreases due to the gravity force causing periodical motion of droplets across the tubing. At relatively high flow velocities, CO2 droplets are well-dispersed even across a horizontal tubing. Droplet dissolution is rather fast in this case, and a notable difference in the dissolution rates in straight and coiled tubing is not observed. An effect of the flow rate on the dissolution process at different tubing diameters is illustrated. Thus, numerical studies show that a coiled tubing can be efficiently used for intensification of liquefied carbon dioxide dissolution for relatively low and moderate flow rates. • Dissolution of CO 2 droplets in a tubing flow in 3-D formulation is modelled. • Dissolution rate in horizontal tubing flow is low due to droplet stratification. • Using coiled tubing wound on a reel strongly enhances dissolution rate. • 3-D simulations allow optimal selection of coiled tubing parameters.

Modeling Immiscible Fluid Displacement in a Porous Medium Using Lattice Boltzmann Method

The numerical investigation of the interpenetrating flow dynamics of a gas injected into a homogeneous porous media saturated with liquid is presented. The analysis is undertaken as a function of the inlet velocity, liquid–gas viscosity ratio (D) and physical properties of the porous medium, such as porous geometry and surface wettability. The study aims to improve understanding of the interaction between the physical parameters involved in complex multiphase flow in porous media (e.g., CO2 sequestration in aquifers). The numerical simulation of a gaseous phase being introduced through a 2D porous medium constructed using seven staggered columns of either circular- or square-shaped micro-obstacles mimicking the solid walls of the pores is performed using the multiphase Lattice Boltzmann Method (LBM). The gas–liquid fingering phenomenon is triggered by a small geometrical asymmetry deliberately introduced in the first column of obstacles. Our study shows that the amount of gas penetration into the porous medium depends on surface wettability and on a set of parameters such as capillary number (Ca), liquid–gas viscosity ratio (D), pore geometry and surface wettability. The results demonstrate that increasing the capillary number and the surface wettability leads to an increase in the effective gas penetration rate, disregarding porous medium configuration, while increasing the viscosity ratio decreases the penetration rate, again disregarding porous medium configuration.

Modeling of carbon dioxide dissolution in an injection well for geologic sequestration in aquifers

Carbon dioxide (CO2) sequestration is considered to be one of the most effective technologies of mitigating greenhouse gas emissions. In this technology, single phase supercritical CO2 is injected into an underground geological formation such as a deep saline aquifer. Existing sequestration projects demonstrate that successful implementations are possible; however, significant uncertainties associated with the risks of leakage remain an obstacle for broader use of this technology. The security of underground disposal could be considerably increased by dissolving the CO2 in a brine produced from the aquifer, then re-injecting the mixture underground. The dissolution process occurs before the mixture reaches the aquifer; this significantly reduces or completely eliminates the risks of CO2 leakage. This technique can drastically extend the amount of worldwide aquifers available for carbon sequestration. As was previously shown, complete dissolution could be achieved in a surface pipeline operating under the pressure of a target aquifer, where CO2 is injected. In this paper, a comprehensive model of CO2 droplet dissolution in a vertical injection well is presented. The model accounts for droplet breakup, coalescence, and dissolution processes as well as temperature and pressure variations over well depth. Feasibility and results are discussed and compared with surface dissolution options.

An experimental study of density-driven convection of fluid pairs with viscosity contrast in porous media

CO2 dissolution in saline water during aquifer sequestration causes natural convection because of the density difference between CO2-saturated brine and native brine. In this study, the convective fingers behavior during convection in porous media was visualized by magnetic resonance imaging (MRI). Four kinds of miscible fluid pairs were used as the equivalents of CO2-saturated brine and native brine. Images of finger development were obtained. There are two types of convective fingers that form vertically symmetrical shapes: downward fingers of the dense fluid and upward fingers of the light fluid. The typical stages of natural convection, finger appearance, finger propagation, new finger generation, and finger coalescence were observed in this study. We analyzed the effects of viscosity contrast variation on convective onset. It was found that the viscosity contrast affects not only the convection time but also the finger growth velocity. The dimensionless mass transfer coefficient is characterized by the Sherwood number. The Sherwood number has a power law relationship with the Rayleigh number, with the exponent being larger than 1 for fluid pairs with apparent viscosity contrast. Experimental results showed that a higher viscosity contrast of the fluid pair causes a viscosity increase in the transition zone, resulting in lower mobility of the fluids and retarding convective mixing. For fluid pairs with similar densities, a high viscosity contrast is not conducive to efficient mass transfer in the density-driven convection process.

Study of Density Driven Convection in a Hele-Shaw Cell with Application to the Carbon Sequestration in Aquifers

Faster dissolution of CO2 within aquifers plays a crucial role in the optimal storage of carbon dioxide injected into reservoirs. The mixture of CO2 and brine triggers convection currents in the aquifer and creates a remarkable increase of the dissolution rate of carbon dioxide. It is very vital for the risk assessment of leakage to obtain accurate description of the CO2 dissolution rate in the aquifers. In our experiments a series of CO2 solute-driven convection laboratory tests were performed in transparent vertical Hele-Shaw cells with a 1.2 mm gap that contained brine water overlain by gas phase and visualization of convective phenomena was achieved with the help of bromocresol purple completely solved into the brine at first. After CO2-brine film come to existence, small phase fingers caused by density contrast would form and they merge, widen, and penetrate deep into the cell. Much series of images recording the dynamic convection were analyzed. Binarization process can help calculate timescale changes of average/all rich- CO2-area which present mass transfer rate of CO2 into the brine. Effects of such different formation parameters as brine concentration, temperature, high pressure and heterogeneity on the onset and development of convection are investigated in our study. Two time intervals, including onset time and developing time took our most attention. Increasing system Rayleigh number, the formation parameters can greatly strengthen the convective mixing and accelerate the growth of CO2 solution into the test cell. The sodium chloride solution concentration gradient has an obvious influence on the form and development of convection phenomenon in our experiments. Here, only part results are exhibited in this paper and much further work are under way.

Pore-scale study based on lattice Boltzmann method of density driven natural convection during CO2 injection project

Saline aquifers are chosen for geological storage of greenhouse gas CO2 because of their storage potential. In almost all cases of practical interest, CO2 is present on top of the liquid and CO2 dissolution leads to a small increase in the density of the aqueous phase. This situation results in the creation of negative buoyancy force for downward density-driven natural convection and consequently enhances CO2 sequestration. In order to study CO2 injection at pore-level, an isothermal Lattice Boltzmann Model (LBM) with two distribution functions is adopted to simulate density-driven natural convection in porous media with irregular geometry obtained by image treatment. The present analysis showed that after the onset of natural convection instability, the brine with a high CO2 concentration infringed into the underlying unaffected brine, in favor of the migration of CO2 into the pore structure. With low Rayleigh numbers, the instantaneous mass flux and total dissolved CO2 mass are very close to that derived from penetration theory (diffusion only), but the fluxes are significantly enhanced with high Ra number. The simulated results show that as the time increases, some chaotic and recirculation zones in the flow appear obviously, which promotes the renewal of interfacial liquid, and hence enhances dissolution of CO2 into brine. This study is focused on the scale of a few pores, but shows implications in enhanced oil/gas recovery with CO2 sequestration in aquifers.

Assessment of the recovery and front contrast of CO2 EOR and sequestration in a new gas condensate reservoir by compositional simulation and seismic modeling

While mature oil reservoirs and aquifers are considered to be good potential candidates for CO2 sequestration, we proposed and investigated an alternative which combines CO2 sequestration and CO2 EOR at the very start of production in a gas condensate reservoir. First, we established a co-simulation workflow with a combination of compositional reservoir simulation and synthetic seismic simulation. Next, we conducted compositional reservoir simulation and synthetic seismic simulation in a five-spot well pattern to investigate whether seismic data can monitor the CO2 front and gas condensate bank with CO2 injected from the very beginning of well production. Then, we compared the compositional simulation results with the seismic simulation results. Although the density contrast among reservoir gas, injected CO2, and condensate is lower than the density contrast in the case of CO2 sequestration in aquifers, the seismic signal may have the potential to capture this smaller difference, monitor the CO2 injection front, and locate the condensate zone, depending on the temperature, pressure and phase properties. When no adequate data are available for reservoir characterization at an early period of production, the seismic data is the only direct measurement of inter-well properties. It may be very valuable for reservoir characterization and for the evaluation of the condensate block. It can serve as the basis for time-lapse monitoring. We compare production by natural depletion with production by CO2 EOR and sequestration started at the very beginning of well production. It shows that the latter will speed up the recovery process and increase the recovery rate while simultaneously storing a large amount of CO2 in the reservoir.

Temperature influence on permeability of Sioux quartzite containing mixtures of water and carbon dioxide

CO2 sequestration in aquifers and depleted hydrocarbon reservoirs is a promising option for reducing atmospheric CO2. However, its implementation requires understanding how CO2 propagates and how it is stored in the reservoir. CO2 mixed with water exists in two forms, an immiscible phase (gas or liquid depending on the conditions of temperature and pressure) and a dissolved phase (a minor part of which reacts with H2O to form carbonic acid, H2CO3). The motion of CO2 in the reservoir is therefore a two-phase flow problem. Fluid/rock chemical reactions may also be important, but they are not considered here. When several immiscible fluids occupy the pore space, each phase may act as an obstacle to the motion of the others. Indeed, experimental evidence from oil production, vadose zone or CO2 sequestration studies [1] demonstrates that permeability is affected by the presence of a second phase in the pore space. The motion and spatial distribution of pore fluids (like water and carbon dioxide) are controlled not only by pore geometry and fluid pressure, but also by the viscosity, solubility, surface tension and wetting characteristics of each fluid phase [2], properties that usually vary with temperature. In the case of carbon dioxide and water, the solubility of CO2 is particularly sensitive to temperature. Thus, temperature variations may also affect the distribution and flow of the phases in the pore space. In this work, we measured the effect of temperature variations on the permeability of a rock saturated with a mixture of carbon dioxide and water. We used the oscillating pore pressure method (OPPM) developed by Kranz et al. [3] and Fischer et al. [4], because it minimizes fluid flux and, therefore, is not expected to disturb the phase distribution in the pore space as much as a steadily flowing fluid would. Although the method was devised to measure permeability of rocks with a single fluid phase, we can assess the effect of the second phase by comparing the permeability measurements made using a single fluid to those with a two-phase mixture.

20502 CO_2海底下貯留条件における含水多孔質内への液体CO_2の溶解特性

20502 CO_2海底下貯留条件における含水多孔質内への液体CO_2の溶解特性

S0820202 充填層内におけるハイドレート生成を伴う液体CO_2の溶解特性([S082-02]高効率火力発電およびCCS技術(2),動力エネルギーシステム部門)

Ocean sequestration is one of the technologies for mitigation of the global warming. Seabed aquifer sequestration is that stored CO_2 in the aquifer approximately 500m depth. CO_2 hydrate generated in the aquifer acts as Cap rock and it makes a possible by storing CO_2 under the hydrate layer. It is supposed that the injected CO_2 under the hydrate layer is dissolved in pore water and stored long term. But geologic properties of the aquifer influence the hydrate generation. Namely, to research on the influence of the geologic properties, such as porosity and particle diameter, on the hydrate generation and dissolution properties of the liquid CO_2 is indispensable in the estimation of the CO_2 storage capacity. However, in previous study, it is focused on the process of CO_2 diffusion in the aquifer. Therefore, knowledge of the geologic properties and solubility properties of the CO_2 is poor. In this study, influence of geologic properties on the hydrate is researched by the hydrate generation experiment. And, the thickness of the CO_2 dissolved water is observed by using ultrasonic testing.

Influence of capillary-pressure models on CO2 solubility trapping

The typical shape of a capillary-pressure curve is either convex (e.g., Brooks–Corey model) or S-shaped (e.g., van Genuchten model). It is not universally agreed which model reflects natural rocks better. The difference between the two models lies in the representation of the capillary entry pressure. This difference does not lead to significantly different simulation results for modeling CO2 sequestration in aquifers without considering CO2 dissolution. However, we observe that the van-Genuchten-type capillary-pressure model accelerates CO2 solubility trapping significantly compared with the Brooks–Corey-type model. We also show that the simulation results are very sensitive to the slope of the van-Genuchten-type curve around the entry-pressure region. For the representative examples we study, the differences can be so large as to have complete dissolution of the CO2 plume versus persistence of over 50% of the plume over a 5000-year period.The cause of such sensitivity to the capillary-pressure model is studied. Particularly, we focus on how the entry pressure is represented in each model. We examine the mass-transfer processes under gravity-capillary equilibrium, molecular diffusion, convective mixing, and in the presence of small-scale heterogeneities. Laboratory measurement of capillary-pressure curves and some important implementation issues of capillary-pressure models in numerical simulators are also discussed. Most CO2 sequestration simulations in the literature employ one of the two capillary-pressure models. It is important to recognize that these two representations lead to very different predictions of long-term CO2 sequestration.

The effect of interface movement and viscosity variation on the stability of a diffusive interface between aqueous and gaseous CO2

Carbon dioxide injected in an aquifer rises quickly to the top of the reservoir and forms a gas cap from where it diffuses into the underlying water layer. Transfer of the CO2 to the aqueous phase below is enhanced due to the high density of the carbon dioxide containing aqueous phase. This paper investigates the behavior of the diffusive interface in an enclosed space in which initially the upper part is filled with pure carbon dioxide and the lower part with liquid. Our analysis differs from a conventional analysis as we take the movement of the diffusive interface due to mass transfer and the composition dependent viscosity in the aqueous phase into account. The same formalism can also be used to describe the situation when an oil layer is underlying the gas cap. Therefore we prefer to call the lower phase the liquid phase. In this paper we include these two effects into the stability analysis of a diffusive interface between CO2 and a liquid in the gravity field. We identify the relevant bifurcation parameter as q = εRa, where ε is the width of the interface. This implies the (well known) scaling of the critical time ∼Ra−2 and wavelength ∼Ra−1(The critical time tc and critical wavelength kc are defined as follows: σ(k) ⩽ 0 ∀t ⩽ tc; equality only holds for t = tc and k = kc). Inclusion of the interface upward movement leads to earlier destabilization of the system. Increasing viscosity for increasing CO2 concentration stabilizes the system. The theoretical results are compared to bulk flow visual experiments using the Schlieren technique to follow finger development in aquifer sequestration of CO2. In the appendix, we include a detailed derivation of the dispersion relation σ(k) in the Hele-Shaw case [C. T. Tan and G. M. Homsy, Phys. Fluids 29, 3549–3556 (1986)]10.1063/1.865832 which is nowhere explicitly given.

A numerical procedure to model and monitor CO2 sequestration in aquifers

Carbon Dioxide (CO2) sequestration into geologic formations is a means of mitigating greenhouse effect. In this work we present a new numerical simulation technique to model and monitor CO2 sequestration in aquifers. For that purpose we integrate numerical simulators of CO2-brine flow and seismic wave propagation (time-lapse seismics). The simultaneous flow of brine and CO2 is modeled applying the Black-Oil formulation for two phase flow in porous media, which uses the Pressure-Volume-Temperature (PVT) behavior as a simplified thermodynamic model. Seismic wave propagation uses a simulator based on a space-frequency domain formulation of the viscoelastic wave equation. In this formulation, the complex and frequency dependent coefficients represent the attenuation and dispersion effect suffered by seismic waves travelling in fluid-saturated heterogeneous porous formations. The spatial discretization is achieved employing a nonconforming finite element space to represent the displacement vector. Numerical examples of CO2 injection and time-lapse seismics in the Utsira formation at the Sleipner field are analyzed. The Utsira formation is represented using a new petrophysical model that allows a realistic inclusion of shale seals and fractures. The results of the simulations show the capability of the proposed methodology to monitor the spatial distribution of CO2 after injection.

Three‐phase compositional modeling of CO2 injection by higher‐order finite element methods with CPA equation of state for aqueous phase

Most simulators for subsurface flow of water, gas, and oil phases use empirical correlations, such as Henry's law, for the CO2 composition in the aqueous phase, and equations of state (EOS) that do not represent the polar interactions between CO2and water. Widely used simulators are also based on lowest‐order finite difference methods and suffer from numerical dispersion and grid sensitivity. They may not capture the viscous and gravitational fingering that can negatively affect hydrocarbon (HC) recovery, or aid carbon sequestration in aquifers. We present a three‐phase compositional model based on higher‐order finite element methods and incorporate rigorous and efficient three‐phase‐split computations for either three HC phases or water‐oil‐gas systems. For HC phases, we use the Peng‐Robinson EOS. We allow solubility of CO2in water and adopt a new cubic‐plus‐association (CPA) EOS, which accounts for cross association between H2O and CO2 molecules, and association between H2O molecules. The CPA‐EOS is highly accurate over a broad range of pressures and temperatures. The main novelty of this work is the formulation of a reservoir simulator with new EOS‐based unique three‐phase‐split computations, which satisfy both the equalities of fugacities in all three phases and the global minimum of Gibbs free energy. We provide five examples that demonstrate twice the convergence rate of our method compared with a finite difference approach, and compare with experimental data and other simulators. The examples consider gravitational fingering during CO2sequestration in aquifers, viscous fingering in water‐alternating‐gas injection, and full compositional modeling of three HC phases.