Experimental Investigation into Gas Hydrate Formation in Sediments

Article pubs.acs.org/IECR

Experimental Investigation into Gas Hydrate Formation in Sediments with Cooling Method in Three-Dimensional Simulator Yu Zhang,†,‡ Xiao-Sen Li,*,†,‡ Zhao-Yang Chen,†,‡ Gang Li,†,‡ and Yi Wang†,‡ †

Key Laboratory of Gas Hydrate, Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, People’s Republic of China ‡ Guangzhou Center for Gas Hydrate Research, Chinese Academy of Sciences, Guangzhou 510640, People’s Republic of China ABSTRACT: Gas hydrate formation behaviors in sediments were investigated in a three-dimensional reactor using the cooling method. The characteristics of the temperature change, the heat transfer, the gas consumption, and the electrical resistance change in the hydrate formation process were studied. The results show that the temperature in the reactor gradually decreases from the near-wall to the center in the cooling process. The gas hydrate preferentially forms in the inner-wall regions of the reactor, which has a low temperature in the cooling process, and then the formation spreads to the surrounding area. On the basis of the balance of energy, it was found that the temperature change was roughly equal to the values calculated in the early stage of the hydrate formation. At the beginning of the hydrate formation, the hydrate formation rates at different places are very different from each other. The gas consumption rate first rises and then falls as the supercooling degree increases during the whole process. The changes of the temperature and resistance illustrate that the methane hydrate inhomogeneously distributes in the reactor.

1. INTRODUCTION Natural gas hydrate is a kind of nonstoichiometic, crystalline, inclusion compound formed by water and gas molecules.1,2 The natural gas hydrate reservoir vastly distributes naturally all over the earth,3 mostly in the permafrost on land and offshore areas.4−8 It is estimated that the potential reserves of natural gas hydrates are over 1.5 × 1016 m3. Therefore, the natural gas hydrates constitute a substantially unconventional source of energy, as well as a significant concern for their impact on climate change and sea-floor stability.9,10 Understanding hydrate formation in porous sediments is of importance for various energy and environmental issues, including drilling and seismic observations of naturally occurring hydrates point to preferred hydrate formation in sand-rich sediments.2 It is also important for the assessment of the hydrate reserve and distribution in the natural reservoir, and developing technology exploiting methane from naturally occurring methane hydrate-bearing sediments without geohazards. In the past, the investigations into the hydrate in porous sediments mainly focus on the thermodynamics and crystal structures of natural gas hydrates.11−13 Several researches focusing on the hydrate formation behaviors in porous sediments have been presented. Melnikov and Nesterov14 studied the equilibrium and kinetics of propane hydrate formation in porous media. They reported that water migration happens to the hydrate formation front, and the mineralization of water promotes hydrate formation in the clay. Kvamme15 presented nucleation theory for the kinetics of hydrate formation and growth based on the multicomponent diffusive interface theory, and considered that methane hydrate initiation from the gas side of the gas/water interface may be dominant in hydrate formation. Buffett and Zatsepina16 carried out the experiments in which the hydrate is crystallized from dissolved © 2014 American Chemical Society

CO2 in natural porous media, and they reported that the hydrate crystals can nucleate and grow directly from an aqueous solution in the absence of free gas. Chuvilin et al.17,18 reported that only a part of the groundwater in the sediments transforms to hydrate, and the rest of the water freezes at subzero temperatures. Katsuki et al.19 visually observed the nucleation, growth, and aging of methane hydrate crystals formed in the porous medium filled with the methane-saturated liquid water, and they found that the crystal morphology depends upon the magnitude of the mass transfer of the methane molecules in the liquid water, which is consistent with that in a bulk methanewater system. Kneafsey et al.20 studied hydrate formation in the methane-water system in porous media in a large X-ray transparent pressure vessel. The pressure, temperature were monitored, and the local density changes were determined through X-ray computed tomography in the experiments. They reported that the rate of hydrate formation is not always proportional to the driving force in the porous medium. They also concluded that the multiple measurement techniques are important to understand the hydrate behavior during hydrate formation/decomposition. Kono et al.21 studied the kinetics of methane hydrate formation within consolidated sediments. The methane gas hydrate was synthesized in laboratory at temperature of 273.5 K and at a pressure of 6.8−13.6 MPa. The synthesized methane gas hydrate is almost like that made by Mother Nature in the strata in the natural gas hydrate field. The reaction rate equation was developed in terms of reaction engineering from experimental data. Liu et al.22 investigated the characteristics and dissociation of methane hydrate by Raman Received: Revised: Accepted: Published: 14208

January 19, 2014 August 18, 2014 August 18, 2014 August 18, 2014 dx.doi.org/10.1021/ie500256z | Ind. Eng. Chem. Res. 2014, 53, 14208−14216

Industrial & Engineering Chemistry Research

Article

Figure 1. Schematic of the experimental apparatus.

Figure 2. Schematic of distributions of thermometer, electrodes, and well in the CHS.

spectroscopy. The methane hydrate was synthesized in silica sands in the range of 53−75, 90−106, 106−150, and 150−180 μm. It was found that the methane hydrate formed in silica sands has similar characteristics regarding cage occupancy and hydration number to bulk hydrate. There is no influence of the particle size on hydrate composition. Zhao et al.23 investigated

the formation characteristics of THF hydrate in various-sized quartz glass beads. The formation process of THF hydrate was observed using magnetic resonance imaging technology. The experimental result suggested that the third surface has an effect on hydrate formation and the hydrate leans to form on the glass beads and in their adjacent area. It was also found that the 14209

dx.doi.org/10.1021/ie500256z | Ind. Eng. Chem. Res. 2014, 53, 14208−14216


Article

reactor. Two pressure transducers are placed at the 13(B) and 13(C) in the reactor to measure the pressures in the middle and the bottom of the hydrate-bearing layer, respectively. 2.2. Experimental Procedure. During the experiment, the raw dry quartz sands with a size range between 300 and 450 μm, and porosity of approximaely 48%, were tightly packed in the reactor, and then the reactor was evacuated twice to remove air in it with a vacuum pump. The quartz sand in the reactor was wetted to saturation with deionized water at atmospheric pressure using a metering pump. The sand sediment was considered as saturation when the rate of water produced from the vessel is equal to that of the water injected. By measurement, the water injected into the reactor is 1537 mL. Then the methane gas was injected into the reactor until the pressure in the reactor reached approximately 20 MPa. As calculated, the water saturation is 55.2%, and the gas saturation is 44.8%, respectively. The inlet valve was closed to keep the system in a constant volume condition. After the pressure and temperature in the reactor became to be constant, the temperature of the water bath was set to the predetermined temperature required for the gas hydrate formation. The pressure inside the reactor decreased continuously as the temperature decreased and the hydrate formation proceeded. After the hydrate formation process lasted about 7 days, the formation experiment was ended and the reactor was heated by adjusting the water bath to a sufficiently high value, which ensures all the hydrate in the reactor can be dissociated. After that, the water bath was set to another temperature to start a new experiment. In this work, a total of 3 experiments were carried. The sequence of the experiments and their initial conditions are given in Table 1. During the hydrate formation

nucleation gets easier and the formation processes faster as the formation temperature decreases. Zhang et al.24 studied the formation behaviors of methane hydrate in the porous media with different pore and particle diameters. They reported that the formation rate of methane hydrate increases as the mean pore diameter increases and the particle range decreases. Linga et al.25 designed a new apparatus to investigate the potential dependency of the kinetics of hydrate formation on the size of the silica sand bed. The hydrate was formed in the water and occupied the interstitial space of the water-saturated silica sand bed. They concluded that the dependency of the results on the size of the bed should be taken into account when the rate of hydrate formation in a porous bed is modeled. Zhou et al.26 presented hydrate formation data in the water-saturated silica sand (particle size of the sand in the range of 100−500 μm) matrix conducted at 2.2 C and 4137 kPa in a 72 L vessel. The authors reported water conversion to hydrate of 11% for the formation experiment. So far, only scattered qualitative studies under limited conditions were published, while the characteristics of the heat transfer and the hydrate distribution in the hydrate reservoir are hardly investigated. The hydrate formation in porous sediments is a reaction process accompanied by heat and mass transfer in a three-dimensional space. It has not yet been clearly accomplished experimentally. In the experiments, the system was first cooled down before the gas injection. Therefore, the temperature change in the reactor could be caused by the gas compression, and the hydrate formation would be promoted by the gas injection. Besides, the gas injection may disturb the water and gas distribution in the reactor, resulting in the different gas and water distributions in different experiments. Meanwhile, the nucleation and the induction time cannot be detected in their works. Therefore, a systematic experimental study on the hydrate formation characteristics in porous sediments in a three-dimensional reactor should be taken. The purpose of this study is to investigate the nucleation, growth rate, the heat transfer and the hydrate distribution during the hydrate formation in sediments in a threedimensional reactor. In this work, the three-dimensional cubic hydrate simulator (CHS) with the effective volume of 5.8L has been developed for the investigation into the formation behaviors of the methane hydrate in sediments.

Table 1. Hydrate Formation Experimental Conditions along with Induction Time and Super-Cooling Degrees runs

P0 (MPa)

T0 (K)

Tf (K)

induction time (h)

supercooling degree (K)

1 2 3

19.24 21.83 20.55

293.10 298.71 292.94

280.15 282.15 285.15

4.17 2.23 5.53

5.31 7.11 4.41

experiment, the temperatures, resistances and pressures in the reactor are recorded at 20 s intervals. The gas consumptions in the hydrate formation were calculated using the method given by Li et al.29

2. EXPERIMENTAL SECTION 2.1. Experimental Apparatus. The details of the experimental apparatus have been reported in our previous work.27,28 In brief, the experimental apparatus for the hydrate formation involves a high-pressure reactor, the temperature control system, the gas and liquid injection equipment, a data acquisition system, and some measurement units distributed in the reactor. The schematic of the three-dimensional cubic hydrate simulator (CHS) work is shown in Figure 1. The CHS is cubic inside, with the edge length of 180 mm and effective volume of 5.8 L. The CHS can withstand pressures of up to 25 MPa and is placed in a water bath. Figure 2 gives the schematic plot of the inner CHS and the well design in the CHS. The reactor is equipped with multiple sets of the measuring points and wellheads at three layers, namely, top (A), middle (B), and bottom (C), with each layer having a total of 25 temperature measuring points and 12 electrical resistance measuring points. Panels a and b in Figure 2 show the distributions of the temperature and resistance measuring points in the different layers within the three-dimensional

3. RESULTS AND DISCUSSION Figure 3 shows the temperature changes versus pressure during the hydrate formation for runs 1, 2, and 3. The equilibrium hydrate formation pressures of the methane hydrate (Pe) are also given in Figure 3. The equilibrium hydrate formation pressures were calculated using the fugacity model of Li et al.30 We also calculated the methane pressure changes (PT) with the average temperature change without hydrate formation at the constant volume by SRK equation, (shown in Figure 3). As a typical example, point A in Figure 3 corresponds to the beginning of the experimental run 2, which is in the temperature−pressure region where no hydrate forms. Between points A and C, as the temperature decreases, only a small linear reduction of the pressure is observed. The pressure change agrees with that just caused by the temperature change, the little difference may be caused by that the pressures were calculated by the average temperature in the reactor and the 14210



Article

Figure 3. Temperature change versus pressure for runs 1, 2, and 3.

temperatures in the reactor in the cooling process were not exactly same. It illustrates that there is no hydrate formed between points A and C. It can be noted that the pressure is lower than the equilibrium formation pressure of methane hydrate from point B. However, no hydrate formation and obvious pressure reduction are observed. From point C, the pressure reduces quickly, and the average temperature begins to increase. These jumps of temperature and pressure in the curves verify that the hydrate starts to form. From point D, the temperature gradually decreases to the setting temperature because of the heat transfer and the decrease of the hydrate formation rate. The average temperature decreases to the setting temperature (point E) and keeps nearly constant. The pressure decreases continuously as the hydrate forms. After the hydrate formation, the temperature of the water bath was set to the to a sufficiently high value, the temperature and pressure in the reactor gradually increase until the hydrate dissociates completely. Figure 4a shows the temperature change profiles of 1(B), 7(B), 13(A), 13(B), and 13(C) for the experimental run 2. The average temperatures, the equilibrium hydrate formation temperatures corresponding to the pressures and the gas consumption are also given in Figure 4a. Figure 4b is a more detailed view of part a between the 2nd and 7th h. As shown in Figure 4a,b, the temperatures decrease quickly in the early stage of the experiment with the cooling of the water bath. The temperature at 1(B) drops most quickly, and the temperature from point 1(B) to point 13(B) successively decreases due to that the heat of the sediments is successively transferred from the center of the reactor to the surrounding. The decline rates of the temperatures at 13(A), 13(B), and 13(C) have little difference. B, B′, and B″ are the points when the average temperature and the temperatures at 1(B) and 7(B) decrease to the equilibrium hydrate formation temperatures, respectively. It can be noted that point B′ is much earlier than points B and B″. C, C′, and C″ are the points when the average temperature, and the temperatures at 1(B) and 7(B) begin to increases, respectively. From points C, C′, and C″, the hydrate begins to form. This time marks the nucleation point. It can be seen that the temperature at 1(B) begins to increase the earliest but the temperature increase are the lowest. It indicates that the hydrate forms first at 1(B). The lowest temperature increase may be because the heat released from the hydrate formation is removed fastest from the sediment to the water bath, and the

Figure 4. Curves of temperatures in the reactor, equilibrium hydrate formation temperature, and gas consumption for run 2.

temperature change is slight. The temperatures at other sites increase slightly later than that at 1(B) but the temperature increases are higher. It indicates that the temperature increases at other sites are due to the hydrate formation rather than the heat transferred from point 1(B). Through the different time when the temperature reaches the equilibrium hydrate formation temperature and the similar time for the hydrate formation, we suppose that when the hydrate begins to form at one site in the reactor, the hydrate clathrate will grow quickly to the surrounding place even there is not a large amount of hydrate formed in the reactor. Therefore, the hydrate formation at 7(B) is not due to the individual nucleation of hydrate at 7(B), and the time difference between C′ and C″ is the duration that the hydrate grows from point 1(B) to point 7(B). The induction time in this experiment should be determined by the point whose temperature first decreases to the hydrate formation temperature. For run 2, the time between points B′ and C′ is the induction time, which is 2.23 h, and the temperature difference between points B′ and C′ is the supercooling degree, which is 7.11 K. Using the same method, the induction times and the supercooling degrees for runs 1 and 3 are also determined. The induction time and supercooling degrees for different experiments are given in Table 1. Figure 5 gives the average temperature change versus time for runs 1, 2, and 3. The inset gives the more detailed view of the temperature changes from the average temperature 14211



Article

temperatures at measuring points around it. Figure 6a gives the temperature distribution at 4.32 h, when no hydrate forms in the reactor. It can be noted that the temperatures at various measuring points are similar. The temperatures in the near-wall region are lower than those in the center of the reactor owing to the heat transfer from the reactor to the water bath. Figure 6b gives the temperature distribution at 4.90 h, when the average temperature begins to increase. As shown in Figure 6b, the temperatures in the near-wall region are also lower than those in the center region of the reactor. However, the temperatures at some places have been obviously high due to the hydrate formation. Figure 6c gives the temperature spatial distribution at 5.21 h, when the temperature at 7(B) reached the highest value during the temperature increase process. As shown in Figure 6c, the temperatures in the center region of the reactor are much higher than those in Figure 6b, the hydrate formation occurs in the entire reactor. The temperatures in the near-wall region are lower than that in the center region of the reactor, but it does not mean that there is no hydrate formed in this region, on the contrary, it is due to the quick heat transfer from the hydrate reservoir to the water bath. Figure 6d gives the temperature spatial distribution at 5.63 h. At this time, the temperatures at different places decrease compared to those in Figure 6c. Figure 7a,b gives the electrical resistance ratio (the ratio of the electrical resistance in the hydrate formation to that at the beginning of the experiment) changes versus time during the hydrate formation process for the experimental run 2. As shown, from the beginning of the experiment to the starting time of the hydrate formation (point C), the resistances increase gradually. In this period, there is no phase transition, and the resistance change is caused by the temperature decrease. In this work, the deionized water containing a few ions was used to form the hydrate. As the temperature decreases, the electrical resistance of the water will increase because of the lower mobility of the ions in the water, which results in the increases of the electrical resistances. As shown in Figure 7b, at point C, when the hydrate begins to form, the resistance changes have sudden jumps, showing that the hydrate begins to form. However, the jumps are slight, thereby, it is hardly to determine the hydrate formation time just through the resistance changes. After point D, with the hydrate formation, the hydrate saturation in the reactor gradually increases, and the resistances increase continuously in general. However, the resistances significantly fluctuate during the whole formation process due to the temperature fluctuation. Consequently, it also can be found that the higher hydrate saturation, the greater resistance fluctuations. Figure 8a−d show the resistance ratio spatial distribution for different gas consumption percents in the experimental run 2. The resistance ratio spatial distribution is obtained using the same method of the temperature spatial distribution. It can be noted that as the hydrate forms, the resistances at different places gradually increase. However, the increases of the resistance are inhomogeneous. It illustrates the hydrate forms at different sites in the reactor with different rates, and the result is similar to the that obtained from Figure 4. Zhou et al.26 analyzed the hydrate distribution in the hydrate formation by the energy balance equation. In order to make the energy balance calculation more reliable, more temperature measuring points and the cooling method for the hydrate formation are adopted in this work. Equation 1 gives the energy balance in the hydrate formation in sediments:

Figure 5. Average temperature change versus time for runs 1, 2, and 3.

decreasing to the equilibrium hydrate formation temperature. Recall that time zero coincides with the time when the average temperature reaches the equilibrium hydrate formation temperature. As shown in Table 1, for the three experiments, the supercooling degree decreases as the induction time decreases. From Figure 5, it can be seen that the reduction rates of the average temperature are different among runs 1, 2, and 3. It is attributed to the different cooling rate of the water bath. The cooling rate of the water bath is affected by the temperature of the air because it is difficult to keep the air temperature constant in the different experiments. Therefore, the cooling rate of the water bath is different, resulting in the different temperature reduction rate for different experiments. As shown in Figure 5 and Table 1, it can be found that the induction time increases and the supercooling degree decreases with the decrease of the temperature reduction rate. Figure 6 shows the temperature spatial distributions over time for run 2. The temperature spatial distributions are drawn using the MATLAB software. The temperature at each site in the reactor is calculated according to the interpolation of the

Figure 6. Temperature spatial distributions over time for run 2. 14212



Article

representing sediments, water, methane gas, and methane hydrate, respectively. Here, because the specific heat of the gas phase is small and the amount of hydrate in the early stage of the hydrate formation is small, the gas phase and hydrate phase were ignored. The values of Cs and Cw are 745 and 4.2 × 103 J/ kg/K, respectively.31 Q is the total heat released from the hydrate formation, which can be calculated by the following: Q=

Vtotal × ΔH 22.4

(2)

where Vtotal refers to the total volume of the methane consumed in the hydrate formation. The latent heat of per mol hydrate formation ΔH is 54.19 kJ.2 From eq 1, the average temperature change in the reactor can be calculated by the following:

ΔT =

Q CSMS + CwM w

(3)

Figure 9 gives the calculated average temperature change compared to that at point C over time in the reactor (ΔT1 in

Figure 7. Resistance ratio changes versus time for run 2.

Figure 9. Temperature changes in the hydrate reservoir versus time for run 2.

Figure 10). The temperature changes at different measuring points and the average temperature change compared to those at point C in the reactor are also given in Figure 9. Compared the average temperature change and ΔT1, it is found that the average temperature change is lower than ΔT1, and ΔT1 continuously increases in the hydrate formation process. Clearly, it is incorrect because of the ignorance of the heat transfer from the hydrate reservoir to the ambient by the cooling of the water bath. Therefore, the heat transfer should be taken into account in the calculation. As shown in Figure 4, the reduction rates of the temperatures at different points and the average temperature in the reactor are almost stable from the 3rd h to the beginning of the hydrate formation. As calculated, the reduction rates is about 3.2 K/h. The heat transferred from the reactor to the water bath can be approximately calculated by the following:

Figure 8. Resistance ratio spatial distributions over time for run 2.

ΔT (CSMS + CwM w + CmM m + CMHMMH) = Q

Q t = (CSMS + CwM w ) × RT × Δt

(1)

(4)

where RT is the temperature reduction rate, Δt is the time from point C. Equation 3 should be rewritten as follows:

where Cx is specific heat capacity, Mx is sample mass, and the subscript x refers to the components s, w, m, and MH 14213



Article

ΔT2. This illustrates that the hydrate formation at different places is not uniform. The hydrate formation rates at 7(B), 13(A), 13(B), and 13(C) are very high, and the rapid hydrate formation results in the high temperature increase, which is obviously higher than the average temperature change. For the temperature change at 1(B), it is lower than the calculated temperature change. Because the lower temperature could be caused by the heat transfer from the reactor to the ambient, it cannot be considered that the hydrate formation rate at point 1(B) is lower than the average hydrate formation rate. As the hydrate formation proceeds, the calculated average temperature gradually comes to be higher than the actually average temperature. It is because that the heat transfer rate gradually decreases as the temperature in the reactor decreases, resulting in that the heat transfer rate to the ambient calculated in eq 6 is higher than the actually value. Figure 10a,b show the curves of the gas consumption, the fugacity corresponding to the pressure and the hydrate formation driving force for run 2. The hydrate formation driving force is the difference between the fugacity and the equilibrium hydrate formation fugacity. As shown, from point B, the hydrate formation driving force becomes higher than zero and increases quickly due to the temperature decrease. From point C, the hydrate formation driving force has a drop due to the temperature increase, and then increases again. When the temperature decreases to the stable value, the hydrate formation driving force decreases gradually resulted from the pressure decrease. As shown in Figure 10a,b, at the beginning of the hydrate formation, the gas consumption rate is significantly high due to the great hydrate formation driving force. During the hydrate formation, the gas consumption rate decreases gradually, even the hydrate formation driving force still increases after point D. It illustrates that the hydrate formation rate is not always proportional to the hydrate formation driving force. Figure 11 gives the cumulative gas consumption versus time during hydrate formation for runs 1, 2, and 3. Recall that time zero coincides with the time when the hydrate begins to form. As shown, in the starting stage of the hydrate formation, the gas consumption rate increase as the super-cooling degree increases. For example, the experimental run 2 has the highest supercooling degree and the quickest gas consumption rate in

Figure 10. Curves of the fugacity in the reactor, driving force for hydrate formation and gas consumption for run 2.

ΔT =

Q − Qt CSMS + CwM w

(5)

Combined eqs 4 and 5, the average temperature change of the reservoir in the reactor compared to that at point C can be calculated as follows: ΔT =

Q − RT × Δt CSMS + CwM w

(6)

The calculated results of the temperature changes considering the heat transfer (ΔT2) are also given in Figure 9. As shown, ΔT2 is almost equal to the average temperature change in the starting stage of the hydrate formation. This illustrates that the calculation of the temperature change based on eq 6 is reliable. It also illustrates that the average temperature change can represent the average hydrate formation rate, and that the local temperature increase higher than the average temperature change indicates the higher hydrate formation rate at the corresponding site. For the temperatures at 7(B), 13(A), 13(B), and 13(C), at the starting time of the hydrate formation, the temperatures are lower than the calculated temperature change. As shown in Figure 4, the temperatures at these points increase lags behind the temperature at 1(B). However, with the hydrate formation, the temperature changes at these points increase quickly to be higher than the average temperature and

Figure 11. Cumulative gas consumption during hydrate formation for runs 1, 2, and 3. 14214



Article

(5) Milkov, A. V.; Dickens, G. R.; Claypool, G. E.; Lee, Y. J.; Borowski, W. S.; Torres, M. E.; Xu, W. Y.; Tomaru, H.; Trehu, A. M.; Schultheiss, P. Co-existence of gas hydrate, free gas, and brine within the regional gas hydrate stability zone at hydrate ridge (Oregon margin): Evidence from prolonged degassing of a pressurized core. Earth Planet. Sci. Lett. 2004, 222 (3−4), 829−843. (6) Kvenvolden, K. A. Gas hydrates: Geological perspective and global change. Rev. Geophys. 1993, 31 (2), 173. (7) Makogon, Y. F.; Holditch, S. A.; Makogon, T. Y. Natural gas hydratesA potential energy source for the 21st century. J. Pet. Sci. Eng. 2007, 56, 14. (8) Klauda, J. B.; Sandler, S. I. Global distribution of methane hydrate in ocean sediment. Energy Fuels 2005, 19, 459. (9) Koh, C. A.; Sloan, E. D. Natural gas hydrates: Recent advances and challenges in energy and environmental applications. AIChE J. 2007, 53, 1636. (10) Makogon, Y. F.; Holditch, S. A.; Makogon, T. Y. Natural gashydrateA potential energy source for the 21st century. J. Pet. Sci. Eng. 2007, 56, 14. (11) Clarke, M. A.; Pooladi-Darvish, M.; Bishnoi, P. R. Amethod to predict equilibrium conditions of gas hydrate formation in porous media. Ind. Eng. Chem. Res. 1999, 38, 2485. (12) Anderson, R.; Llamedo, M.; Tohidi, B.; Burgass, R. W. Experimental measurement of methane and carbon dioxide clathrate hydrate equilibria in mesoporous silica. J. Phys. Chem. B 2003, 107, 3507. (13) Zhang, W.; Wilder, J. W.; Smith, D. H. Interpretation of ethane hydrate equilibrium data for porous media involving hydrate-ice equilibria. AIChE J. 2002, 48, 2324. (14) Melnikov, V.; Nesterov, A. Modelling of gas hydrates formation in porous media. Proceedings of the 2nd International Conference on Gas Hydrates, Toulouse, France, 1996; p 541. (15) Kvamme, B. Kinetics of hydrate formation from nucleation theory. Int. J. Offshore Polar Eng. 2002, 12, 256. (16) Buffett, B. A.; Zatsepina, O. Y. Formation of gas hydrate from dissolved gas in natural porous media. Marine Geology 2000, 164, 69. (17) Chuvilin, E. M.; Kozlova, E. V.; Makhonina, N. A.; Yakushev, V. S.; Dubinyak, D. V. Peculiarities of methane hydrate formation/ dissociation P/T conditions in sediments of different composition. Proceedings of the 4th International Conference on Gas Hydrates, Yokohama, Japan, 2002; p 433. (18) Chuvilin, E. M.; Makhonina, N. A.; Titenskaya, O. A.; Boldina, O. M. Petrophysical investigations of frozen sediments artificially saturated by hydrate. Proceeding of the 4th International Conference on Gas Hydrates, Yokohama, Japan, 2002; p 734. (19) Katsuki, D.; Ohmura, R.; Ebinuma, T.; Narita, H. Methane Hydrate crystal growth in a porous medium filled with methanesaturated liquid water. Philos. Magn. 2007, 87, 1057. (20) Kneafsey, T. J.; Tomutsa, L.; Moridis, G. J.; Seol, Y.; Freifeld, B. M.; Taylor, C. E.; Gupta, A. Methane hydrate formation and dissociation in a partially saturated core-scale sand sample. J. Pet. Sci. Eng. 2007, 56, 108. (21) Kono, H. O.; Narasimhan, S.; Song, F.; Smith, D. H. Synthesis of methane gas hydrate in porous sediments and its dissociation by depressurizing. Powder Technol. 2002, 122, 239−246. (22) Liu, C. L.; Lu, H. L.; Ye, Y. G.; Ripmeester, J. A.; Zhang, X. H. Raman spectroscopic observations on the structural characteristics and dissociation behavior of methane hydrate synthesized in silica sands with various sizes. Energy Fuels 2008, 22, 3986. (23) Zhao, J. F.; Yao, L.; Song, Y. C.; Xue, K. H.; Cheng, C. X.; Liu, Y.; Zhang, Y. In situ observations by magnetic resonance imaging for formation and dissociation of tetrahydrofuran hydrate in porous media. Magn. Reson. Imaging 2011, 29, 281. (24) Zhang, Y.; Wu, H. J.; Li, X. S.; Li, G.; Chen, Z. Y.; Zeng, Z. Y. Experimental study on formation behavior of methane hydrate in porous media. Acta Chim. Sin. 2011, 29, 2221 (In Chinese). (25) Linga, P.; Haligva, C.; Nam, S. C.; Ripmeester, J. A.; Englezos, P. Gas hydrate formation in a variable volume bed of silica sand particles. Energy Fuels 2009, 23, 5496.

the starting stage of the hydrate formation. This is due to the fact that the higher super-cooling degree means there is a higher hydrate formation driving force at the beginning of the hydrate formation. With the hydrate formation proceeding, the gas consumption rate of run 2 decreases gradually and becomes lower than that of run 1. At the later stage of the hydrate formation, the gas consumption rate increases as the last stable temperature decreases.

4. CONCLUSIONS In this work, the temperature changes, heat transfer, gas consumption rate, hydrate nucleation, and hydrate distribution in the methane hydrate formation process in sediments were investigated in the three-dimensional reactor with the cooling method. The following conclusions are drawn: The hydrate formation occurs earlier in the inner-wall regions of the reactor and the hydrate grows quickly from the first nucleation point where the hydrate grows to the other places in the reactor. The induction time increases, and the supercooling degree decreases as the temperature reduction rate decreases. The temperatures in the reactor begins to increase quickly when the hydrate starts to form and then decrease gradually afterward, reaching the highest values due to the heat transfer to the ambient. On the basis of the calculation balance of the energy, the temperature increase in sediments is equal to that caused by the latent heat from the hydrate formation in the starting period of the hydrate formation. In the starting stage of the hydrate formation, the gas consumption rate increases as the supercooling degree increases. In the later stage, the gas consumption rate increases as the hydrate formation temperature decreases. According to the discussion about the temperatures and electrical resistances in the hydrate reservoir, it is confirmed that the hydrate saturation and the hydrate formation rate at different places in the reactor are inhomogeneous.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +86-20-87057037. Fax: +86-20-87034664. E-mail: lixs@ ms.giec.ac.cn. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (51106160), the National Science Fund for Distinguished Young Scholars of China (51225603), Science & Technology Program of Guangzhou (2012J5100012) and Key Arrangement Programs of the Chinese Academy of Sciences (Grants KGZD-EW-301-2), which are gratefully acknowledged.

■

REFERENCES

(1) Davidson, D. W. Gas hydrates. In Water: A Comprehensive Treatise; Plenum Press: New York, 1973; Vol. 2, pp 115−234. (2) Sloan, E. D. Clathrate Hydrates of Natural Gases, 2nd ed.; Marcel Dekker, Inc.: New York, 1998. (3) Makogon, Y. F.; Trebin, F. A.; Trofimuk, A. A.; Tsarev, V. P.; Cherskii, N. Detection of a pool of natural gas in a solid (hydrated gas) state. Dokl. Akad. Nauk SSSR 1971, 196, 203. (4) Kvenvolden, K. Potential effects of gas hydrate on human welfare. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3420. 14215



Article

(26) Zhou, Y.; Castaldi, M. J.; Yegulalp, T. M. Experimental investigation of methane gas production from methane hydrate. Ind. Eng. Chem. Res. 2009, 48, 3142. (27) Li, X. S.; Wang, Y.; Li, G.; Zhang, Y.; Chen, Z. Y. Experimental investigation into methane hydrate decomposition during threedimensional thermal huff and puff. Energy Fuels 2011, 25, 1650. (28) Li, X. S.; Zhang, Y.; Li, G.; Chen, Z. Y.; Wu, H. J. Experimental investigation into the production behavior of methane hydrate in porous sediment by depressurization with a novel three-dimensional cubic hydrate simulator. Energy Fuels 2011, 25, 4497. (29) Li, G.; Li, X. S.; Tang, L. G.; Zhang, Y. Experimental investigation of production behavior of methane hydrate under ethylene glycol injection in unconsolidated sediment. Energy Fuels 2007, 21, 3388. (30) Li, X. S.; Zhang, Y.; Li, G.; Chen, Z. Y.; Yan, K. F.; Li, Q. P. Gas hydrate equilibrium dissociation conditions in porous media using two thermodynamic approaches. J. Chem. Thermodyn. 2008, 40, 1464. (31) Moridis, G. J.; Seol, Y.; Kneafsey, T. J. Studies of reaction kinetics of methane hydrate dissociation in porous media. Proceeding of the 5th International Conference on Gas Hydrates, Trondheim, Norway, 2005.

14216


Experimental Investigation into Gas Hydrate Formation in Sediments

Recommend Documents