3284
Energy & Fuels 2007, 21, 3284–3291
CO2 Hydrate: Formation and Dissociation Compared to Methane Hydrate Carlo Giavarini,*,† Filippo Maccioni,† Monia Politi,‡ and Maria Laura Santarelli† Chemical Engineering Department, UniVersity of Rome “La Sapienza”, Via Eudossiana 18, 00184, Rome, Italy, and ENEL—Italian Electricity Agency, Area Tecnica Ricerca, Litoranea Salentina, Brindisi, Casalabate, Italy ReceiVed February 13, 2007. ReVised Manuscript ReceiVed July 23, 2007
For CO2 disposal in the form of hydrate it is important to know the decomposition kinetics at moderate pressures and temperatures, similar to those that could be realized in the storage systems. This paper has considered the preservation of CO2 hydrate containing different quantities of CO2, at pressures between 0.1 and 0.3 MPa and temperatures between -3 and 0 °C. At the conditions (P, T) of this work, CO2 hydrate does not present any anomalous self-preservation effect, and its dissociation is not affected by subcooling before storage. More than pressure, which is very important for methane hydrate, temperature affects the preservation. The temperature of -3 °C assures a good stability at atmospheric pressure, providing that CO2 saturation into the hydrate is not too high. CO2 hydrate is generally more stable than CH4 hydrate due to the different activation energy of decomposition.
1. Introduction Carbon sequestration is defined as the removal of greenhouse gases from industrial or utility plant streams and their longterm storage in a way they cannot interact with the climate system. Various methods for selective CO2 removal have been suggested, and some of them are based on gas absorption, membrane process, and cryogenic fractionation and are in commercial use. Disposal of captured CO2 in the ocean and in geological reservoirs has been proposed by many investigators.1–3 Among others, the Norwegian company Statoil has applied an injection process since 1996.2 Another challenge is to take advantage of the properties of CO2 hydrates for carbon sequestration. Gas hydrates are inclusion compounds (clathrates) formed by a lattice of hydrogenbonded water molecules which encage small gaseous molecules. One volume of CO2 hydrate can contain as much as 160 volumes of gaseous CO2.4,5 The formation of CO2 hydrates has been studied by a number of research groups.6–14 A hydratebased gas separation process has been proposed for CO2 recovery from flue gases.15 Carbon dioxide and water form a stable system when pressure and temperature fall within the hydrate formation region (P > 4.5 MPa and T < 283 K); in the ocean, a long-lived system is formed (at higher P and lower T) only when seawater is * Corresponding author: e-mail
[email protected]; Tel +390644585565; Fax +390644585416. † University of Rome “La Sapienza”. ‡ ENEL—Italian Electricity Agency. (1) Mertz, B.; Davidson O.; de Coninck, H.; Loss, M.; Meyer, L. IPCC Special Report on Carbon Dioxide Capture and Storage; Cambridge University Press: New York, 2005. (2) Herzog, H. J. EnViron. Sci. Technol. 2001, 35, 148–153. (3) Teng, H.; Yamasaki, A.; Chun, M. K.; Lee, H. Energy 1997, 22, 1111–1117. (4) Sloan, E. D., Jr. Clathrate Hydrates of Natural Gases; Marcel Dekker: New York, 1998. (5) Kimuro, H.; Yamaguchi, F.; Ohtsubo, K.; Kusanayagi, T.; Morishita, M. Energy ConVers. Manage. 1993, 34, 1089–1094.
Table 1. Ratios of Molecular Diameter to Cavity Diameters for Methane, Carbon Dioxide, and Ethane4 a molecular diameter/cavity diameter structure I cavity type
structure II cavity type
species
guest diameter (Å)
small
large
small
large
CH4 CO2 C2H6
4.36 5.12 5.5
0.855 1 1.08
0.744 0.834 0.939
0.868 1.02 1.1
0.652 0.769 0.826
a Small cavity for structures I and II: 512; large cavity for structure I: 51262; large cavity for structure II: 51264.
saturated with respect to CO2.3,16 The capacity and effectiveness of ocean sequestration can be increased by injecting CO2 into deep sediments: a few hundred meters of sediments provides permanent geologic storage; additionally, CO2 hydrate formation will impede the flow of liquid CO2.17 (6) Saji, A.; Yoshida, H.; Sakai, M.; Tanii, T.; Kamata, T.; Kitamura, H. Energy ConVers. Manage. 1992, 33, 643–649. (7) Nishikawa, N.; Morishita, M.; Uchiyama, M.; Yamaguchi, F.; Ohtsubo, K.; Kimuro, H.; Hiraoka, R. Energy ConVers. Manage. 1992, 33, 651–657. (8) Austvik, T.; Loken, K. P. Energy ConVers. Manage. 1992, 33, 659– 666. (9) Golomb, D. S.; Zemba, S. G.; Dacey, J. W. H.; Michaels, A. F. Energy ConVers. Manage. 1992, 33, 675–683. (10) Golomb, D. S. Energy ConVers. Manage. 1993, 34, 967–976. (11) Loken, K. P.; Austvik, T. Energy ConVers. Manage. 1993, 34, 1081–1087. (12) Circone, S.; Stern, L. A.; Kirby, S. H.; Durham, W. B.; Chakoumakos, B. C.; Rawn, C. J.; Rondinone, A. J.; Ishii, Y. J. Phys. Chem. B 2003, 107, 5529–5539. (13) Noda, H.; Saji, A.; Sakai, M.; Tanii, T.; Kamata, T.; Kitamura, H. In Carbon Dioxide Chemistry: EnVironment Issues; Paul, J., Pradier, C. M., Eds.; The Royal Society of Chemistry: London, 1994; pp 338–341. (14) A Research Needs Assessment for the Capture, Utilization and Disposal of Carbon Dioxide From Fossil-Fuel-Fired Power Plants. DOE Report DOE/ER-30194, Washington, DC, 1993. (15) Kang, S. P.; Lee, H. EnViron. Sci. Technol. 2000, 34, 4397–4400. (16) Giavarini, C.; Maccioni, F. La Termotecnica 2001, 10, 29–31. (17) House, K. Z.; Schrag, D. P.; Harvey, C. F.; Lackner, K. S. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 12291–12295.
10.1021/ef070080t CCC: $37.00 2007 American Chemical Society Published on Web 09/20/2007
CO2 Hydrates
Energy & Fuels, Vol. 21, No. 6, 2007 3285
Figure 1. Equilibrium P–T diagram of the water–CO2 system.20
Figure 2. Schematic of the experimental apparatus.
The potential of CO2 sequestration is also offered by disposal into caves or depleted natural gas reservoirs.18 In this case a large amount of CO2 could be potentially fixed by artificial rock
weathering: the calcium silicate component of wollastonite reacts with carbonate ions and becomes CaCO3 and silicate.19 However, the reaction kinetics is very slow, and CO2 should be
3286 Energy & Fuels, Vol. 21, No. 6, 2007
GiaVarini et al.
Figure 3. CO2 hydrate formation from bulk water (570 mL). The upper curve refers to the pressure; the lower curve refers to the temperature. The various repressurization steps are clearly visible in both curves.
Figure 4. CO2 hydrate formation from spray water. Typical pressure trend during water injection.
released slowly from a storage system so that it could gradually react with the rock. The hydrate could possibly provide such gradual release of the gas. Carbon dioxide hydrates, CO2 · nH2O (n ) 5.75), belong to a family of nonstoichiometric compounds in which hydrogenbonded water molecules are arranged in an icelike framework, forming polyhedral cavities occupied by guest molecules. CO2 (18) Oldenburg, C. M.; Pruess, K.; Benson, S. M. Energy Fuels 2001, 15, 293–298. (19) Wu, J. C. S.; Sheen, J.; Chen, S. Y.; Fan, Y. C. Ind. Eng. Chem. Res. 2001, 40, 3902–3905.
hydrate crystallizes in the same structure (sI) of the methane hydrate5 at relatively high pressure and low temperature, as shown by the equilibrium P–T diagram in Figure 1.20 In general, gas hydrates have three structural forms: sI, sII, and sH. Structure I is a body-centered-cubic structure. The cubic cell contains 46 H2O molecules, including two small cages, 12-hedra (512), and six large cages, 14-hedra (51262), where the notation 512 is used to indicate that the (20) CSMHYD.EXE software enclosed in Sloan, E. D., Jr. Clathrate Hydrates of Natural Gases; Marcel Dekker: New York, 1998.
CO2 Hydrates
Energy & Fuels, Vol. 21, No. 6, 2007 3287
Figure 5. Typical MDSC curve (reversing heat flow) of CO2 hydrate containing about 40% CO2. The first peak is due to ice melting while the second peak is referred to hydrate dissociation.
Figure 6. Typical MDSC curve (reversing heat flow) of CH4 hydrate containing about 40% CH4. The first peak is due to ice melting, and the second peak is referred to hydrate dissociation.
polyhedron contains 12 five-membered-ring faces and 51262 is used to indicate that the polyhedron also contains two sixmembered-ring faces.4 Table 1 shows the size ratio of methane, ethane, and CO2 within each of the four cavities of structures I and II. At ratios of molecule to cage size lower than about 0.76, the molecular attractive forces cannot contribute to cavity stability.4 Above the ratio of about 1.0, the guest molecule cannot fit into cavity.
As simple hydrates, methane can stabilize the smaller 512 cavities of sI, while CO2 and ethane can stabilize the larger 51262 cavities. The molecules which stabilize the smaller cavities will also occupy the 51262 cavities. Depending on pressure and formation conditions, CO2 molecules fit also into the 512 small cavities of sI;21–24 this is not the case for ethane molecules.20 In the form of hydrate, CO2 could have a number of uses (chemical production, greenhouse gas, etc.) or can be disposed
3288 Energy & Fuels, Vol. 21, No. 6, 2007
GiaVarini et al.
Figure 7. Typical dissociation curve showing the four temperature plateaus (at -3 °C, -2 °C, -1 °C, and 0 °C respectively, dotted line) and the corresponding pressure variations. The pressure increase tends to achieve a constant value (equilibrium pressure) due to the fact that the dissociation is carried out in a sealed vessel.
The present work is part of a project on the separation of carbon dioxide from the flue gases of powers plants in the form of hydrate. The project also includes the storage, use, and disposal of the hydrate. The main purpose of the present work is the evaluation of the decomposition kinetics of CO2 hydrate containing different quantities of ice, at low pressures (0.1–0.3 MPa) and temperatures between -3 and 0 °C. The formation conditions include moderate pressures, more economical in the hypothesis of future industrial application. A further purpose of the work is to compare these two chemical species (CH4 and CO2) during the hydrate formation and decomposition. 2. Experimental Section Figure 8. Correlation between the dissociation rate and CO2 concentration in the hydrate at different storing temperatures (P ) 0.1 MPa).
in suitable geological environments. In any case it is important to understand the hydrate decomposition kinetics during storage, transportation, and disposal. Depending on such kinetics the most suitable storage and disposal system could be studied. Previous works24–28 reported that methane hydrate remains metastable at 0.1 MPa and temperatures slightly below 0 °C (i.e., outside its stability region) and preserves in bulk as much as 80 K above its nominal equilibrium temperature for a certain time. The preservation is sensibly improved if the pressure is kept at 0.2 or 0.3 MPa and depends on the concentration of the CH4 gaseous molecules (former) in the hydrate.28 (21) Lee, H.; Yongwon, S.; Yu-Taek, S.; Moudrakovski, I. L.; Reepmester, J. A. Angew. Chem., Int. Ed. 2003, 42, 5048–5041. (22) Udachin, K. A.; Ratclife, C. I.; Ripmeester, J. A. J. Phys. Chem. B 2001, 105, 4200–4204. (23) Yu-Taek, S.; Moudrakovski, I. L.; Ripmeester, J. A.; Jong-won, L.; Huen, L. EnViron. Sci. Technol. 2005, 39, 2315–2319. (24) Stern, L. A.; Circone, S.; Kirby, S. H.; Durham, W. B. J. Phys. Chem. B 2001, 105, 1756–1762.
2.1. Apparatus. The experimental apparatus for the formation and dissociation tests consists of a jacketed stainless steel reactor (RC-Mettler Toledo) with an internal volume of about 2 L, equipped with a stirrer. A schematic view is shown in Figure 2. Pressure, temperature, and stirrer revolutions are transmitted to a computer through a data acquisition board and are recorded at 3 s intervals; the instrument accuracy is P ( 0.001 MPa and T ( 1 °C. For hydrate formation, CO2 gas (99.9%) is supplied from a cylinder bottle in the reactor containing the required amount of distilled water. The preparation of hydrates containing higher CO2 concentrations (>50%, with respect to the near ideal stoichiometry CO2 · 5.75H2O) is performed with a special nozzle installed at the top of the reactor chamber to spray water into the guest-gas (CO2) phase in the reactor in order to increase the gas/water interfaces. (25) Stern, L. A.; Circone, S.; Kirby, S. H.; Durham, W. B. New Insights into the Phenomenon of Anomalous “Self” Preservation of Gas Hydrates. In Proceedings of the 4th International Conference on Gas Hydrates, Yokohama, Japan, May 19–23, 2002, pp 673–677. (26) Stern, L. A.; Circone, S.; Kirby, S. H.; Durham, W. B. Can. J. Phys. 2003, 81, 271–283. (27) Circone, S.; Stern, L. A.; Kirby, S. H. Am. Mineral. 2004, 89, 1192– 1201. (28) Giavarini, C.; Maccioni, F. Ind. Eng. Chem. Res. 2004, 43, 6616– 6621.
CO2 Hydrates
Energy & Fuels, Vol. 21, No. 6, 2007 3289
Table 2. Dissociation Rates and Days Required for Complete Dissociation of Some Hydrate Samples (30% Concentration) dissociation rate (mol/s) pressure (MPa) 0.1 0.2 0.3
-3 °C 10-5
4.37 × 4.37 × 10-5 4.0 × 10-5
-2 °C 10-4
1.70 × 9.46 × 10-5 7.92 × 10-5
days to complete dissociation
-1 °C
0 °C
10-4
5.15 × 4.12 × 10-4 2.56 × 10-4
An external water circulation loop is attached to the chamber; the rate at which cold water flows through this loop is controlled by a plunger pump. 2.2. Hydrate Formation Procedure. Two different procedures are considered to produce the hydrate samples from water and carbon dioxide. The first procedure is used for the production of medium and low concentration hydrates (30–50% conversion with respect to the near ideal stoichiometry). Each experimental run starts by supplying about 570 g of water to the evacuated reactor chamber; the reactor is pressurized with carbon dioxide to 2 MPa at 10 °C. Stirring is then started at 100 rpm to dissolve CO2 into water. The solubility is detected by a pressure decrease to about 1.8 MPa and a small temperature increase. After two repressurization steps to 2.0 MPa, the pressure remains stable and the water is saturated. CO2 is a hydrophilic molecule more soluble into water than methane; this procedure is necessary to separate (during the following phases of the experimental run) the phenomena due to CO2 solubilization and to hydrate formation, respectively. Moreover, the dissolved CO2 provides the starting phase for the hydrate nucleation. The temperature is then lowered to 2 °C to enter the hydrate stability zone in the equilibrium diagram, and stirring is increased to 400 rpm. Hydrate formation begins immediately and is detected by a pressure decrease and a temperature increase from 2 to 4 °C (Figure 3). To speed the formation rate, the reactor is repressurized to 20 MPa before reaching the equilibrium pressure at that temperature. The pressure curve in Figure 3 shows a series of repressurization steps whose amplitude (pressure drop) is correlated to the amount of CO2 moles encaged by the hydrate lattice; the same amount can be calculated on the basis of the pressure decrease during the decomposition tests. During the various repressurization steps, the pressure drop is progressively decreased due to the solid hydrate formation. Figure 3 shows that after five steps and about 20 min from the starting of the experiment, the pressure decrease tends to be much slower, and from a practical point of view the reaction can be considered quite complete: the corresponding experimental pressure and temperature values are respectively 1.95 MPa and 4.8 °C, which is slightly outside the stability region of the equilibrium diagram (Figure 1). The second procedure (water spraying) is used to produce more concentrated samples (e.g., in the range 70–90%). Each experimental run starts by pressurizing to 2.0 MPa with CO2 the evacuated reactor chamber kept at 2 °C. The sprayed water comes from a
Figure 9. Correlation between the dissociation rate and pressure, at different storing temperatures (30% hydrate).
10-3
1.12 × 6.39 × 10-4 3.81 × 10-4
-3 °C
-2 °C
-1 °C
0 °C
26 26 29
7 12 15
2 3 5
1 2 3
reservoir maintained at the same temperature of the reactor, and it is introduced by the plunger pump at the pressure of 2.0 MPa. Each spraying cycle consists of 10 pump deliveries (totally 3 g of water), leaving a time for the reaction between two following cycles. About 30 cycles lasting about 18 h totally are performed during each experiment. In each cycle pressure and temperature are about 1.9 MPa and 1.9 °C, respectively, as shown in Figure 4, which includes two spraying cycles. The hydrate obtained with this procedure is granular, while the previous one is more coherent. 2.3. Hydrate Characterization. When the amount of added water is known, the percentage of hydrate in the sample can be calculated from the pressure drop during the hydrate formation (or/ and from the pressure increase during decomposition). Modulated differential scanning calorimetry (MDSC) can be used for hydrate characterization.29,30 With respect to conventional DSC, one of the major contributions of this technique is that the total heat flow rate can be separated into two additional signals.31 The first one is directly correlated to the heat capacity of the material, while the second one is a function of temperature and time characteristics of the tests. These terms can be plotted separately in two curves. In the present work, MDSC tests were performed at atmospheric pressure following the procedure described in previous papers.29,30 The reactor temperature was lowered to -20 °C; the reactor was then depressurized and opened. A sample was then quickly transferred in the MDSC pan kept a container at -20 °C. For the test, the pan was again transferred in the MDSC apparatus and kept at -20 °C. A typical MDSC reversing (heat capacity) curve of a sample containing about 40% of CO2 hydrate (respect to the near ideal stoichiometry) is shown in Figure 5 and can be compared to the methane hydrate curve in Figure 6. The curve in Figure 5 includes two peaks with apexes at -0.56 and 10.20 °C, respectively. The first peak is due to the melting of the ice and the second to the decomposition of the hydrate. The corresponding peaks of the CH4 hydrate (Figure 6) are at -0.23 and 4.47 °C, respectively. The higher temperature of the decomposition peak of CO2 hydrate confirms its higher stability with respect to CH4 hydrate. Methane hydrate requires less energy (about 54 kJ/mol) than CO2 hydrate to dissociate and, therefore, is less stable.4 Two chemically different molecules such as carbon dioxide and ethane (polar and not polar, respectively) show similar heats of dissociation (∆H ) 73 and 72 kJ/mol, respectively32,33). The activation energies for the decomposition of CH4, C2H6, and CO2 hydrates decomposition are respectively 81, 104, and about 103 kJ/mol;34 these values confirm the higher stability of CO2 and C2H6 hydrates with respect to CH4 hydrate. Kuhs et al. have found difference in the degree of perfection for ice produced from decomposing CH4 and CO2 hydrate, with the latter showing less imperfection and, therefore, better preservation below 0 °C.35,36 At (29) Giavarini, C.; Maccioni, F.; Santarelli, M. L. Ind. Eng. Chem. Res. 2003, 42, 1517–1521. (30) Giavarini, C.; Maccioni, F.; Santarelli, M. L. J. Therm. Anal. Calorim. 2006, 84, 419–424. (31) TA Instruments DSC 2920 operator’s manual appendix C: modulated DSC option, 1998. (32) Long, J. P.; Lederhos, J. P.; Sum, A.; Christiansen, R. L.; Sloan, E. D., Jr. In Proceedings of the 73rd Gas Process Association Annual Convention, New Orleans, LA, March 7–9, 1994, pp 85–89. (33) Handa, Y. P. J. Phys. Chem. 1986, 90, 5497–5498. (34) Clarke, M. A.; Bishnoi, P. R. Chem. Eng. Sci. 2004, 59, 2983– 2993.
3290 Energy & Fuels, Vol. 21, No. 6, 2007
GiaVarini et al.
Figure 10. Correlation of dissociation rates between subcooled and not-subcooled CO2 hydrate samples at pressure of 3 MPa.
temperatures g0 °C, the solubility of carbon dioxide in liquid water probably influences the decomposition rate. 2.4. Hydrate Dissociation Tests. The hydrate dissociation tests are conducted immediately after hydrate formation in the same (closed) reactor by increasing the temperature and following the procedure explained previously.28 In order to check how such a procedure affects the preservation, some samples are subcooled at -20 °C before the dissociation tests. During the tests, temperature and pressure are set at its nonequilibrium conditions, i.e., at pressures between 0.1 and 0.3 MPa and temperatures between -3 and 0 °C. The experimental apparatus permits rapid control of temperature and pressure. After the desired temperature is reached (e.g., -3 °C), the reactor is depressurized to the required pressure; the rate of dissociation is then followed by measurement of any pressure increase. About after 1 h the temperature is increased to the next value (e.g., -2 °C) and again to -1 and 0 °C, in the following steps, respectively. The typical dissociation curve of CO2 hydrate in Figure 7 shows the four temperature plateaus considered for the conservation tests. The dissociation rate can be calculated from the pressure curve.
3. Results and Discussion A previous work has demonstrated that methane hydrate dissociation depends, besides on storage temperatures and pressure, on the gas saturation and on the previous thermal history (subcooling) of the samples.28 Methane hydrate can be preserved metastably in bulk at 0.1 MPa (or better at 0.3 MPa) at temperatures slightly below 0 °C.24,28 Our experimental results have confirmed that, at the conditions (P, T) of this work, the dissociation of CO2 hydrate does not show any anomalous selfpreservation effect. In the following discussion, the dissociation rate is expressed as percent of gas moles lost with respect to the gas moles in the hydrate, over a time of 1 s. 3.1. Correlation between Hydrate Dissociation and Concentration. For a given pressure, the dissociation rate at a fixed temperature depends on the hydrate concentration in the (35) Kuhs, W. F.; Genov, G.; Staykova, D. K.; Hansen, T. Phys. Chem. Chem. Phys. 2004, 6, 4917–4920. (36) Kuhs, W. F.; Genov G.; Staykova, D. K. In Proceedings of the 5th International Conference on Gas Hydrates, Trondheim, Norway, June 13– 16, 2005, pp 14–20.
sample, as shown in Figure 8, which reports the dissociation rates vs temperature, in a range of hydrate conversion between 30% and 90%. Hydrates containing lower quantities of gas (i.e., richer in ice) are more stable than the more concentrated ones. In this case the dissociation behavior of CO2 hydrate is similar to that of methane hydrates. The fact that the particles of hydrates produced with the spray system (90% samples) have different sizes could influence the dissociation rate.37 However, the relative position of the 30%, 50%, and 90% curves in Figure 8 seems to confirm (at least qualitatively) the preservation effect of the ice: more ice in the sample is responsible for a higher preservation effect. The mixtures richer in ice are probably responsible for an “ice shielding” or/and “ice encapsulation” effect which could influence the dissociation rate.26 In other words, the gas diffusion can be hindered by ice. 3.2. Correlation between Hydrate Dissociation and P/T Parameters. Table 2 reports, as an example, the dissociation rates of low conversion (30%) CO2 hydrate at different temperatures and pressures. For a more immediate comprehension, the table also reports the time necessary to fully dissociate the samples. As expected, the dissociation rates increase with the temperature and decrease at higher pressures. The best preservation is reached at -3 °C, at any pressure. The effect of P and T is clearly visible in Figure 9, which shows that at -3 °C the influence of the pressure is quite negligible and becomes more important at higher temperatures. 3.3. Effect of Thermal History. Some hydrate samples were further cooled (subcooled) to -8 °C after preparation and kept at this temperature for 2 h. The temperature was then raised to the test value (at the planned pressure). Subcooling had no apparent effect on CO2 hydrate preservation. It has been suggested that further cooling of methane hydrates hardens the thincubicicecages(ice“perfection”),thushelpingself-preservation.35,36 However, differences have been found in the degree of perfection for ice produced from decomposing CH4 and CO2 hydrate, with the latter showing less imperfection and, therefore, better preservation.36 Figure 10 shows the behavior of subcooled and not-subcooled hydrate samples stored at pressure of 0.3 MPa.
CO2 Hydrates
Figure 11. Comparison between the dissociation rate of 50% CO2 and CH4 hydrates stored at 0.1 MPa in the temperature range -4 to 0 °C.
3.4. Comparison to CH4 Hydrate Preservation. The present experimental work has been conducted following the same procedure of a previous work on methane hydrate,28 and hydrates with similar concentrations were obtained in both cases. Therefore, it seems possible to compare the preservation behavior of the two hydrates. Figure 11 refers to CO2 and CH4 hydrate with the same concentration (50% hydrate) stored at 0.1 MPa and different temperatures. In the case of methane, both subcooled (to -8 °C) and not-cooled samples are considered; this is not necessary for CO2 because the behavior of subcooled and not-subcooled samples is similar. The CH4 hydrate decomposition at 0 °C is too fast to be detected and is not reported. In the considered temperature range, CO2 hydrate is more stable than CH4 hydrate, at both 0.1 and 0.2 MPa (not reported), and its decomposition can be detected up to 0 °C. At the temperature of 0 °C the hydrate dissociation gives CO2 and liquid water, instead of ice: the formation of liquid water is the key permitting rapid complete breakdown of the hydrate.12 The dissociation rate of the hydrate is influenced by CO2 concentration in bulk water,38 which is much higher than the methane concentration. Therefore, the CH4 hydrate dissociation is faster than CO2 hydrate. Methane hydrate becomes more stable than CO2 hydrate (50% concentration) when the pressure is increased to 0.3 MPa (Figure 12). The preservation pressure is very important for methane hydrate and has little effect on CO2 hydrate. By increasing to 90%, the hydrate concentration in the sample (i.e., decreasing the ice content) CO2 hydrate becomes again more stable, also at the pressure of 0.3 MPa (Figure 13). The ice is the key of the different behavior of methane and CO2 hydrate during dissociation at the same pressure; this is probably due to the complex self-preservation effect of methane hydrate, more linked to the ice content. With respect to methane, molecules such as CO2 form more stable hydrates structures due to the different cavity occupation4 and to their higher activation energies of decomposition.34 The preservation of CO2 hydrates is less influenced by pressure, which is more important in “keeping” the methane molecules inside the cages. Subcooling introduces “perfection” in the ice (37) Tanaka, S.; Maruyama, F.; Takano, O.; Uchida, K.; Oya, N. In Proceedings of the 5th International Conference on Gas Hydrates, Trondheim, Norway, June 13–16, 2005, pp 1314–1317. (38) Gabitto, J.; Tsouris, C. Energy ConVers. Manage. 2006, 47, 494– 508.
Energy & Fuels, Vol. 21, No. 6, 2007 3291
Figure 12. Comparison between the dissociation rate of 50% CO2 and CH4 hydrates stored at 0.3 MPa and different temperatures.
Figure 13. Dissociation rate of the 90% CO2 and CH4 hydrates stored at 0.3 MPa and different temperatures.
cages and contributes to stabilize the CH4 hydrate more than the CO2 hydrate.35,36 4. Conclusion This study on CO2 hydrate formation and dissociation was conducted to examine its potential for application to lowpressure carbon sequestration and to compare the preservation characteristic of CO2 and CH4 hydrates. In the studied P, T range, CO2 hydrate does not present any anomalous selfpreservation effect, and its dissociation is not affected by subcooling before storage. More than pressure (which is very important for methane hydrate), temperature affects the preservation of CO2 hydrate, at least in the studied pressure range. Among the explored conditions, the temperature of -2 °C, or better of -3 °C, can assure a good stability to hydrate at atmospheric pressure, providing that the degree of CO2 saturation is not to high. In fact, the mixtures richer in ice show a slower dissociation rate; “ice shielding” or “ice encapsulation” probably influences gas diffusion.26 Methane hydrate requires less energy to dissociate than CO2 hydrate and, therefore, is less stable. Kuhs et al.36 have found differences in the degree of perfection for ice produced from decomposing CH4 and CO2 hydrate, with the latter showing less imperfection and, therefore, better preservation below 0 °C; at temperatures higher than 0 °C, the solubility of CO2 in liquid water probably influences the decomposition rate. Subcooling after formation is important for methane hydrate preservation, but it does not substantially affect CO2 hydrate stability. EF070080T
