Phase equilibria of CO2+N2 and CO2+CH4 clathrate hydrates: Experimental measurements and thermodynamic modelling
Highlights
▸ A thermodynamic model based on the CPA EoS and vdW-P theory is applied. ▸ The dissociation conditions of hydrates of CO2/N2 and CO2/CH4 are modeled. ▸ New experimental dissociation data for CO2/N2 hydrates are reported. ▸ The developed model gives satisfactory results compared with available models.
Introduction
Clathrate hydrates (or gas hydrates) are crystalline structures consisting of H-bonded water molecules surrounding small molecules (Sloan, 1998). There is a wide variety of small molecules that can stabilize the water lattice, including some hydrocarbons (methane, ethane, propane, etc), carbon dioxide, nitrogen and hydrogen sulfide. Clathrate hydrates form three typical crystalline structures, namely structure I (sI), structure II (sII) and structure H (sH) (Sloan, 1998). These structures are stabilized by high pressures and low temperatures and they are a well known issue in the oil and gas industry. Indeed, they cause serious flow-assurance problems due to the blockage of oil and gas pipelines. One of the main techniques used to avoid hydrate formation involves using additives called thermodynamic inhibitors that shift the hydrate stability zone towards higher pressures or lower temperatures. This causes the hydrates to become less stable in the conditions relevant to oil and gas production. Sodium chloride, methanol and ethylene glycol are the most well known thermodynamic inhibitors (Jager et al., 2002; Jager, 2001). Most of the phase equilibrium data found in the literature involve systems containing hydrocarbon hydrate formers and thermodynamic inhibitors.
Hydrates have also been recognized as having benefits in several industrial applications. In particular, natural hydrates are expected to contain significant amounts of methane (Chatti et al., 2005, Kvenvolden, 1988). The potential reserves of natural hydrates are such that they could become a major unconventional source of energy in the future. The use of hydrate slurries for secondary-refrigerant loops taking benefit of the latent heat content to increase the energy density of the fluid has been suggested elsewhere (Martinez et al., 2008). This application may require the use of additives known as thermodynamic promoters to lower the pressures required for hydrate stability.
Several studies aiming at evaluating the use of hydrates as a means for separation purposes have been done. The use of hydrates can lead to a major reduction of the number of theoretical trays compared to a classic distillation where only the relative volatilities of the compounds are involved. Recently, several authors have suggested the use of hydrates in CO2 capture processes (Chatti et al., 2005, Ballard, 2002, Linga et al., 2008, Kang et al., 2001, Seo et al., 2001). This would be an interesting alternative to the expensive and energy-consuming process based on the use of amine solutions for scrubbing the CO2 from the flue/industrial gases. In the hydrate-based process, flue/industrial gases and an aqueous solution contact each other at the temperature and pressure conditions required for hydrates formation. Since CO2 hydrates are more stable than hydrates of other components of flue/industrial gases, it is expected that CO2 will preferentially enter into the hydrate phase. The CO2-rich phase is then sent into a second reactor where hydrates are dissociated and a CO2-rich gas is obtained for further transport and storage.
In a hydrate-based process, the flue/industrial gases must be compressed and cooled down to stabilize hydrates (Linga et al., 2007). This is a major drawback of this process that reduces its energy efficiency. The use of thermodynamic promoters is recommended to solve this problem. One of the most studied thermodynamic promoters for CO2 capture applications is tetrahydrofuran (THF) (Kang et al., 2001, Seo et al., 2001). Other promoters (dioxane, acetone etc.) (Seo et al., 2001) have also been studied, but the thermodynamic effect of THF is recognized as leading to an important improvement of the stability conditions of hydrates. More recently, several papers (Strobel et al., 2007, Duc et al., 2007, Lin et al., 2008) have investigated the use of tetra-butyl ammonium bromide and tetra-butyl ammonium fluoride as hydrate promoters.
As mentioned by Sloan and Koh (2008), thermodynamic models for hydrates of hydrocarbon systems have now reached a near-experimental accuracy, and permit good understanding of the thermodynamic phenomena involved in oil and gas applications. Classically, these models are based on the use of: (1) the van der Waals and Platteeuw model for the hydrate phase; (van der Waals and Platteeuw, 1959) (2) a cubic equation of state (EoS) for the oil and gas phases; and (3) an activity model to describe the behavior of the aqueous phase (Vidal, 1997).
Despite the availability of these models for oil and gas systems, describing the thermodynamic behavior of a hydrate-based CO2 capture process requires further work. A first reason is the limited amount of experimental data available in the literature, with respect to numerous data published for hydrocarbon systems. In Table 1, Table 2, we summarize the main references providing dissociation data of mixed-clathrate hydrates of carbon dioxide+nitrogen and carbon dioxide+methane, respectively.
We report new experimental hydrate dissociation data for various CO2/N2 gas mixtures in the presence of pure water.
The hydrate dissociation conditions were investigated at different composition ratios of CO2 in the feed gas. Capillary gas phase sampling was used to determine the initial composition of the different CO2/N2 gas mixtures. An isochoric pressure search method (Tohidi et al., 2000, Mohammadi et al., 2005) allowed measurement of the hydrate dissociation conditions of the mixed-gas hydrate systems. The dissociation data generated in this work are compared with the corresponding literature data. Thermodynamic model based on the use of the Cubic Plus Association Equation of State (CPA-EoS) (Kontogeorgis et al., 2008) for fluid phases combined with the van der Waals and Platteeuw (1959) model for the hydrate phase is proposed for modeling purpose. Its major advantage is that it leads to a homogeneous thermodynamic description of all fluid phases, i.e., all fluid phases are modeled through the same EoS. In addition, the CPA-EoS is able to take into account the associating behaviour of aqueous solution. The model in this work is used to predict the hydrate dissociation conditions of the CO2/N2 gas mixtures measured in this work and of those reported in the open literature. A similar comparison is presented for the hydrate dissociation conditions of CO2/CH4 gas mixtures reported in the literature. Predictions through CSMGem (Ballard, 2001) and HWHYD (Heriot-Watt University, hydrate model, 1993) models are reported herein for comparison purposes. The reliability of these two last models is compared to that of our model in terms of deviations between experimental and calculated dissociation temperatures. It should be noted that almost a similar approach has been recently applied by Herslund et al. (2012).
Section snippets
Description of the applied thermodynamic model
Equality of fugacities defines the thermodynamic equilibrium of the system (Poling et al. 2001; Wong and Sandler 1992; Folas et al., 2005). The model proposed in this work relies on the equality of fugacity of water in the hydrate phase and of water in the fluid phases. As mentioned earlier, the hydrate phase is modeled by the solid solution theory of van der Waals and Platteeuw. The CPA-EoS is used to model the thermodynamic behavior of both vapor and liquid (aqueous) phases.
Experimental setup and method
A schematic diagram of the experimental setup is shown in Fig. 1. Details of the equipment and experimental procedure have been described previously (Belandria et al., 2011a,b). The apparatus is based on the “static-analytic” technique with capillary gas phase sampling (ROLSI™) (Guilbot et al., 2000). This equipment is suitable for measurements at temperatures ranging from 233 to 373 K and compatible with corrosive gases. The main part of the apparatus is a cylindrical equilibrium cell which
Regression of thermodynamic parameters
The critical properties and acentric factors of carbon dioxide, nitrogen, methane and water are listed in Table 4. The CPA-EoS parameters of water, given in Table 5, are from Kontogeorgis et al. (2008). Experimental VLE data for each of the gas components (CO2, N2 and CH4) with water have been used to optimize binary interaction parameters with respect to the following objective function (OBF):where xi is the gas solubility in water. The superscripts c and exp stand for
Results and discussion
Hydrate dissociation (temperature and pressure) conditions were measured for various CO2/N2 compositions in the presence of water. The compositions of the CO2/N2 gas mixtures are given in the first column of Table 9. Average value of at least seven analyses is reported as the composition of the feed for each gas mixture. The measured hydrate dissociation data are reported in this table and plotted in Fig. 5. The hydrate phase boundaries of simple carbon dioxide and nitrogen hydrates are also
Conclusion
We have reported dissociation data for clathrate hydrates of various nitrogen+carbon dioxide mixtures (Table 9). The gas mixtures were prepared directly in the equilibrium cell and their compositions were determined using a ROLSI™ sampler (Guilbot et al., 2000) connected to a gas chromatograph. Isochoric pressure-search method (Tohidi et al., 2000, Mohammadi et al., 2005). was applied to measure the dissociation conditions. Our experimental data were compared with the literature data and
Nomenclature
- AAD
Average absolute deviation
- a
Radius of the spherical core (Å) or attraction term of the EoS
- BIP
Binary interaction parameter
- b
Covolume (m3)
- CPA
Cubic Plus Association
Langmuir coefficient of component j in cavity i.
- c1
Parameter in Eq. 9
- EoS
Equation of state
Fugacity of component j
Fugacity of water in the empty hydrate lattice
Fugacity of water in a hypothetical state ice or pure liquid
Radial distribution function
- H–Lw–V
Hydrate–Liquid water–Vapour
Constant of Boltzmann
- Mw
Molecular weight
Acknowledgement
Veronica Belandria is grateful to Fundayacucho of Venezuela for providing her a Ph.D. scholarship. Financial support by the Agence Nationale de la Recherche (ANR) as part of the SECOHYA project is gratefully acknowledged. The authors would also like to thank Pascal Théveneau and the CEP/TEP workshop for technical support during the experiments. Ali Eslamimanesh is acknowledged for his fruitful comments on the manuscript.
References (56)
- et al.
Fluid Phase Equilib.
(2010) - et al.
Energy Convers. Manage.
(2005) - et al.
Energy Convers. Manage.
(2007) - et al.
Fluid Phase Equilib.
(2005) - et al.
Fluid Phase Equilib.
(2006) - et al.
J. Chem. Thermodyn.
(2012) - et al.
Fluid Phase Equilib.
(2011) - et al.
Fluid Phase Equilib.
(2002) Chem. Geol.
(1988)- et al.
Fluid Phase Equilib.
(2001)