J. Phys. Chem. B 2010, 114, 6973–6978
6973
Formation Characteristics of Synthesized Natural Gas Hydrates in Meso- and Macroporous Silica Gels Seong-Pil Kang*,† and Jong-Won Lee‡ Clean Fossil Energy Research Center, Korea Institute of Energy Research, 102 Gajeong-ro, Yuseong-gu, Daejeon 305-343, Republic of Korea, and Department of EnVironmental Engineering, Kongju National UniVersity, 275 Budae-dong, Cheonan, Chungnam 331-717, Republic of Korea ReceiVed: January 27, 2010; ReVised Manuscript ReceiVed: March 10, 2010
Phase equilibria and formation kinetics of the natural gas hydrate in porous silica gels were investigated using the natural gas composition in the Korean domestic natural gas grid. The hydrate-phase equilibria in the porous media are found to shift to the inhibition area than that in the bulk phase. The measured phase equilibrium data, combined with the Gibbs-Thomson equation, were used to calculate the hydrate-water interfacial tension. The value was estimated to be 59.74 ( 2 mJ/m2 for the natural gas hydrate. In addition, the inhibition effect is observed to be more significant in the meso-sized pore than the macro-sized one. In the formation kinetics, it was found that the hydrate formation reached the steady-state in a short period of time without mechanical stirring. Furthermore, the formation rate was found to be faster at 275.2 K than 273.2 K even though the driving force at 273.2 K is larger than that of 275.2 K. Even though the porous silica gels have smaller volume than other methods for gas storage, the gas consumption was found to be significantly enhanced in this study (for example, 120 vol/vol for the silica gels and 97 vol/vol for wet activated carbon). In this regard, using porous silica gels can be a potential alternative for gas storage and transportation as a nonmechanical stirring method. Although this investigation was performed with the natural gas composition in the Korean domestic grid, the results can also be expanded for designing or operating any hydrate-based process using various gas compositions. Introduction In the engineering and industrial fields, gas hydrates were first known as hazardous materials that can cause blocking problems in oil and gas pipelines. When water vapor contained in natural gas reacts with “guest” species in pipelines, gas hydrate is formed and expands to block flows through the pipelines.1 Therefore, initial studies were mainly focused on various parameters such as temperature, pressure, and gas composition affecting hydrate formation so as to inhibit or remove hydrate formation in transmission lines.2 From early investigations, it has been found that the gas hydrate can be categorized into three distinct crystal structures, which are structure-I (sI), structure-II (sII) and structure-H (sH), and that types and numbers of cavities that capture guest molecules can vary according to the formed structure.3,4 In addition, the sH hydrate having the largest cavity size among the hydrate structures requires a second guest (a so-called “help gas”) in order to stabilize the entire crystal structure. Since a huge amount of natural gas is known to be buried in the form of the gas hydrate, a lot of attention has been paid to how to use it as a new energy source in the future. Naturally occurring gas hydrate is known to exist under the permafrost and deep ocean sediments, where specific temperature and pressure conditions for hydrate formation are naturally satisfied. The amount of the natural gas hydrate is estimated to be about twice as much as the total energy in fossil fuel resources.5,6 Such a huge amount is mainly attributed to characteristics of * Corresponding author. E-mail:
[email protected]. Tel: +82-42-8603475, Fax: +82-42-860-3134. † Korea Institute of Energy Research. ‡ Kongju National University.
the gas hydrate, which can hold a large amount of gas contents in a unit volume of solid gas hydrates. Such characteristics are used for many applications such as storage of energy gases such as hydrogen7 and separation/sequestration of greenhouse gases such as carbon dioxide.8 In addition, the gas hydrate has also been studied to store or transport natural gas in the solid form by synthesizing the natural gas hydrate.9–11 When designing a hydrate-based process, formation kinetics in addition to hydrate-phase equilibria also play an important role. Since Handa and Stupin12 investigated the effect of porous media on the equilibrium pressures using CH4 and C3H8 hydrates for the first time, other researchers have expanded such investigation to other guest species.13,14 In addition, Anderson et al.15 experimentally studied the formation behaviors of CH4, CO2, and CH4 + CO2 hydrates using meso-porous silica gels and suggested a modeling equation. In the phase-equilibrium point of view, the capillary effect of the porous media results in inhibition of the hydrate-phase equilibrium curve (that is, requiring higher pressure at a given temperature or lower temperature at a given pressure than bulk phase to form solid hydrates), which leads to increased energy cost. However, Kang et al.16 reported that higher conversion yield can be achieved in a short period of time without mechanical agitation when porous water contents in the silica gel are used to form the gas hydrate. Therefore, from the kinetic point of view, usage of porous media can have some advantages. Seo et al.17 supported such faster kinetics of CO2 hydrate in the silica gel by means of a microimaging technique. In addition, Kang et al.16 and Seo et al.18 recently reported a variety of phase-equilibrium and formation kinetics using single- and multicomponent systems in the silica gel.
10.1021/jp100812p 2010 American Chemical Society Published on Web 04/29/2010
6974
J. Phys. Chem. B, Vol. 114, No. 20, 2010
Kang and Lee
TABLE 1: Dry-Based Natural Gas Composition
thermal equilibrium. During the experiments, the temperature and pressure of the high-pressure reactor was recorded using a data acquisition system with an interval of 10 min. For measuring formation kinetics, about 250 cm3 of the silica gel was placed in the same high-pressure cell. After charging the silica gels containing pore water in the high-pressure cell, the reactor was cooled to the desired temperature and allowed to stabilize. Then, sufficiently pressurized natural gas in the reservoir was introduced to the reactor up to the desired experimental conditions (5.0 MPa). During the hydrate formation, an ISCO syringe pump was used to compensate for pressure drop in the reactor. The data acquisition system recorded the temperature, pressure, and gas flow rate (with digital thermal mass flow controller 5850E, Brooks Instrument, LLC), ranging from 0 to 200 L/min with an accuracy of 1% of the full scale, throughout the experiments at an interval of 1 min. All the experimental measurements were repeated three times in order to check repeatability and experimental reliability. In addition, the gas composition of the vapor phase and the gas retrieved from dissociation of the hydrate phase were analyzed by gas chromatography.
gas
composition (%)
gas
composition (%)
CH4 C 2H 6 C 3H 8 iso-C4H10
89.86 6.40 2.71 0.48
n-C4H10 n-C5H12 N2
0.49 0.02 0.04
TABLE 2: Physical Properties of Porous Silica Gels sample name
6 nm SG
15 nm SG
100 nm SG
mean particle diameter (µm) mean pore diameter (nm) specific pore volume (m3/kg) specific surface area (m2/kg)
34 to 75 5.51 8.4 × 10-4 586 × 103
33 to 74 14.6 11.3 × 10-4 308 × 103
40 to 75 94.5 8.3 × 10-4 42.4 × 103
However, experimental approaches so far have been limited to single or binary guest systems due to convenient handling and analysis. Moreover, formation kinetics of various gas hydrates in the porous media has not been sufficiently reported. Therefore, in this study, thermodynamic phase equilibria of natural gas hydrate in the porous silica gel were investigated in order to examine hydrate formation behaviors of multicomponent guests and calculate the hydrate-water interfacial tension in macro- and meso-sized pores. In addition, formation kinetics of the natural gas hydrate was also measured for the purpose of applying the gas hydrate as a storage/transportation media of natural gas from the engineering point of view. Natural gas is a mixture of hydrocarbons and a few nonhydrocarbons and has various compositions depending on production area. In this report, the natural gas composition distributed in the Korean domestic natural gas grid was used. The experimental results obtained in this study can provide valuable information for identifying multicomponent behaviors during hydrate formation in porous media as well as for designing or operating any natural gas hydrate-based process in the future. Experimental Methods The gas mixture used in this study was supplied by Rigas (Korea) and was ultrahigh purity (UHP) grade. Dry-based gas composition is summarized in Table 1, which simulates the natural gas composition distributed in the Korean domestic natural gas grid. HPLC grade water was supplied by SigmaAldrich Chemical Co. with a minimum purity of 99.99 mol %. The spherical silica gels with nominal pore diameters of 6, 15, and 100 nm were selected and purchased from Sigma-Aldrich Chemical Co. (6 and 15 nm) and Silicycle (100 nm). The properties of the silica gels with 6 and 15 nm pore diameters were measured by nitrogen adsorption/desorption experiments with ASAP 2400 (Micrometrics), and that with a 100 nm pore diameter was measured by mercury intrusion with Autopore VI 9500 (Micrometrics), as listed in Table 2. In addition, normalized pore size distributions are presented in Figure 1. Details regarding preparation hydrate samples in porous media can be found in our previous report.16 Phase equilibria and formation kinetics were measured with a specifically constructed high-pressure vessel. For phaseequilibrium experiments, approximately 250 cm3 of silica gels containing pore water was charged in the high-pressure cell made from 316 stainless steel (maximum working pressure is 15 MPa and internal volume is about 350 cm3). Then, after the cell was purged by natural gas to remove the remaining air in the system, the cell was pressurized to the desired pressure and cooled to 263 K at a cooling rate of 1.5 K/h. When pressure drop due to hydrate formation reached a steady-state condition, the cell was heated stepwise by 0.1 K and kept for 1 h to reach
Results and Discussion As noted by the previous literature,13,15,16,19 hydrate-phase equilibria in the porous media are shifted to the inhibition region compared with the bulk phase. Such inhibition can be explained by an increased pressure in the solid phase, which is induced by the high curvature of the solid-liquid interface.15 The hydrate-phase equilibria, measured by the dissociation curve of the hydrate phase, are also affected because the solid-liquid transition is depressed to lower temperatures at any given pressure and to higher pressures for any given temperature. Figure 2 illustrates phase equilibria of the natural gas hydrate in the porous media. As it can be seen in this figure, the hydratephase equilibrium curve in the bulk condition is shifted to the inhibition region when the natural gas hydrate is formed in the silica gels. In addition, the hydrate equilibrium curve in the 100 nm silica gel is not significantly different from that of the bulk phase, while the 6 nm silica gel shows the largest inhibitions effect. In particular, the inhibition effect change from 100 to 15 nm was found to be larger when compared with pore size decrease from 15 to 6 nm. Observed results lead us to conclude that the inhibition effect of the porous media is gradually increased as the pore size decreases. Therefore, special care should be taken when forming the natural gas hydrate in mesoporous silica gels. In addition, the Gibbs-Thomson equation was used to relate hydrate formation temperature depression with pore size of the porous media.20 According to the relationship, the temperature depression of hydrate dissociation in a cylindrical pore, ∆Tm,pore, relative to the bulk dissociation temperature, Tm,bulk, can be defined in terms of pore diameter as15,21,22
(
2RHW cos θ ∆Tm,pore )Tm,bulk Fs∆Hm,sd
)
where Fs is the solid density, ∆Hm,s is the latent heat of melting, θ is the contact angle between the solid phase and the pore wall, and d is the pore diameter. The plot of the reciprocal pore diameter (1/d) versus the temperature depression (∆Tm,pore/Tm,bulk) is presented in Figure 3. For the natural gas hydrate, a hydrate density of 936 kg/m3 was used, which is obtained on the basis of experimental cage occupancies of the components in our
Natural Gas Hydrates in Porous Silica Gels
J. Phys. Chem. B, Vol. 114, No. 20, 2010 6975
Figure 1. Pore-size distribution of the silica gels used in the experiments.
Figure 2. Hydrate-phase equilibria of natural gas hydrates in porous silica gels.
previous report.19 In addition, 72.498 kJ/mol is used for ∆Hm,s of the natural gas hydrate. This value is obtained from the Clasius-Clapeyron equation using the measured phase-equilibrium data in porous silica gels. Using these values and the slope of the experimental data in Figure 3, the hydrate-water interfacial tension was estimated to be 59.74 ( 2 mJ/m2 for the natural gas hydrate. This value is larger than the reported values of 45 ( 1 mJ/m2 and 50 ( 2 mJ/m2 for pure C3H8 and mixed CH4 + C2H6 + C3H8 hydrates, respectively.14,18 The interfacial tension, expressed as the energy acting perpendicular to a unit area (J/m2), can be interpreted as energy needed to separate two surfaces of the material. In the case of the gas hydrate, this energy is strongly dependent on the attractive force or a pair potential between the hydrate particles, that is, hydrate guest molecules.23 Therefore, higher density can yield larger interfacial tension assuming that all the other conditions (such as hydrate structure) are identical. In this study, natural gas components other than methane are identified to occupy only the large cages (θL,C2H6 ) 0.3247, θL,C3H8 ) 0.3600, θL,i-C4H10 and n-C4H10 )
0.0424), while only methane can fill both small and large cages (θS,CH4 ) 0.8695, θS,CH4 ) 0.1905). Comparing this result with the literature, small cage occupation by CH4 molecules or large cage occupation by heavier molecules than C3H8 will make the density of the natural gas hydrate larger (that is, attractive force among guest species) than that of pure C3H8 hydrate. In addition, large cage occupation by heavier hydrocarbons will also increase the density value compared with mixed CH4 + C2H6 + C3H8 hydrate. As a result, such guest occupation is mainly contributed to the increased hydrate density and larger interfacial tension than the reported values for pure C3H8 and mixed CH4 + C2H6 + C3H8 hydrates. Formation kinetics as well as the equilibrium curve is also critical for applications of the gas hydrate. In this study, the formation kinetics of the natural gas hydrate in the 100 nm silica gel was investigated using various formation temperatures at 3.92 MPa, which is summarized in Figure 4. As it can be seen in this figure, the final gas consumption showed little difference for all the formation conditions, while the initial formation rate
6976
J. Phys. Chem. B, Vol. 114, No. 20, 2010
Kang and Lee
Figure 3. Reciprocal pore diameter (1/d) versus ∆T/Tm.
Figure 4. Formation kinetics of natural gas hydrates in porous silica gels at 3.92 MPa.
at 275.2 K is found to be the fastest. When initial abrupt gas consumption due to hydrate formation is defined as the hydrate formation rate (Rf), the formation rates for 273.2, 275.2, and 278.2 K are obtained as 0.4261, 0.7396, and 0.4221 mmol gas/ mol H2O · min, respectively. In the experimental results, the fastest formation rate at 275.2 K is about 1.75 times larger than those at other conditions. This result indicates that hydrate formation can be processed much faster at specific conditions. In addition, it should be noted that the formation rate at 275.2 K is found to be larger than that at 273.2 K even though driving force increases in accordance with lowering temperature. Such a faster formation rate at 275.2 K than 273.2 K can be explained by combined two factors. First, from the mass transfer point of view, both viscosity and temperature increase as temperature is increased, leading to increased mass transfer at higher temperature. Therefore, hydrate formation is limited due to limited mass transfer of guest species at low temperature. Second, considering larger driving force, hydrate formation should be faster at lower temperature. In other words, hydrate formation occurs faster in spite of slower mass transfer at 273.2 K than 275.2 K. After such hydrate formation, formed solid-
phase may act as additional barrier for mass transfer of guest species. Initial formation rate determined by these two factors make us to conclude that there exists a possible optimum temperature condition, even though further investigations are necessary to identify such descriptions at other sizes of porous media and other natural gas compositions. Although this experiment was performed using the porous silica gels, the same consideration can be expanded to the bulk-phase formation. In this regard, a Japanese pilot plant, developed by Chugoku Electric Power Co., Inc. and Mitsui Engineering & Shipbuilding Co. Ltd., is operating continuously at 277 K for hydrate formation, dewatering, and pelletizing.24 During the hydrate formation, composition changes of the gas phase were monitored by means of the gas chromatograph, which is provided in Figure 5a. Moreover, the formed hydrate samples were collected and allowed to dissociate in order to analyze the retrieved gas compositions, which is presented in Figure 5b. As shown in the figures, CH4 concentration in the natural gas is increased initially during the hydrate formation. More stable hydrate formers of ethane (C2H6) and propane (C3H8) are thought to be consumed first due to their milder
Natural Gas Hydrates in Porous Silica Gels
J. Phys. Chem. B, Vol. 114, No. 20, 2010 6977
Figure 5. (a) Gas phase composition changes during the natural gas hydrate formation, and (b) gas phase composition changes during the retrieved natural gas hydrate dissociation.
formation conditions, and such consumption leads to the increased concentration of CH4 in appearance. However, when CH4 molecules start to be encaptured to the gas hydrate after 40-80 min, the CH4 concentration in the mixed natural gas is slightly decreased and fluctuated until reaching the steady composition. When comparing the final gas composition with the initial one, ethane and propane (in particular, propane) concentrations were found to be decreased. Such a decreased concentration can be explained by fast participation of the guests in the hydrate formation because propane is the most stable hydrate former. Meanwhile, such guest stability resulted in the reversed behavior in the hydrate dissociation. In the dissociation, methane, ethane and propane successively start to escape from the hydrate phase, which is reflected in the composition changes. Therefore, when designing a hydrate-based process using the natural gas and operating the process in a continuous way, additional unit or balancing flow is required on the basis of the different guest stability (that is, composition change during the formation). Moreover, milder hydrate formation by ethane or propane relative to that of methane is attributed to the CH4 concentration drop from the initial 90% to the final 70%. From our previous spectroscopic investigation,14 propane and ethane molecules were identified to be captured only into large cages
of the sII hydrate, while CH4 molecules are found to occupy both small and large cages of the sI and sII hydrates. As reported by Lee et al.,25 such a concentration decrease of CH4 molecules can occur both in the bulk phase and the porous media. Accordingly, the encapturing tendency and accurate occupation behavior of various guest molecules should be identified first before designing and operating a continuous hydrate-based process. Gas storage in an easy and economic manner has become very important in the last few decades, and some methods have been widely investigated as alternatives to conventional compressed natural gas (CNG) and liquefied natural gas (LNG) technologies. Among a variety of such methods, adsorbed natural gas (ANG) in activated carbon is a promising technology because of its relatively low pressure and high temperature condition, which is especially useful for nonstationary storage such as a transportation fuel. To enhance the amount of deliverable gas by the ANG method, activated carbon in wet conditions has been applied where additional gas hydrate formation is taken into account. However, the heavy components in natural gas are more strongly adsorbed than methane so that the composition before and after the gas hydrate formation is slightly changed.11 Therefore, the heating value should be also
6978
J. Phys. Chem. B, Vol. 114, No. 20, 2010
considered when the gas hydrate formation is accompanied14 as examined in this study. In this regard, Najibi et al. reported that it is not proper to use the wet activated carbon for enhancing the amount of gas storage. Even though activated carbon in their study has larger surface area and pore volume than silica gel pores in this work, the possible surface areas to gas molecules decreased when water is applied to activated carbon because water covers the surface and competes with gas molecules. In this work, water contents are intended to exist only in silica gel pores so that a large amount of gas consumption can be achieved even with smaller pore volume in the silica gel: 120 vol/vol for this work, 97 vol/vol for wet activated carbon.11 Meanwhile, Rogers et al.10 suggested a micellar gas hydrate storage process using a surfactant without any mechanical stirring. Because mechanical stirring for rapid gas hydrate formation requires high energy cost, any nonstirring method becomes attractive in terms of the process cost. Moreover, water contents dispersed in pores make it faster to form gas hydrates. Therefore, a nonstirring method for gas hydrate formation should be investigated in depth and considered as a possible option for gas storage and transport. Conclusions In this report, the phase equilibria and formation kinetics of the natural gas hydrate were investigated using the porous silica gels. The hydrate-phase equilibria in the porous media are found to shift to the inhibition area compared to that in the bulk phase. In addition, the inhibition effect is observed to be more significant in the mesosized pores than the macrosized ones. The obtained results are also used to calculate the interfacial tension in combination with the Gibbs-Thomson equation. Such physical properties of the multicomponent hydrate system in porous media can provide useful information in understanding the formation behaviors of multiple guests in silica gels. In the formation kinetics, it was found that the hydrate formation reached the steady state in a short period of time without mechanical stirring. Furthermore, the formation rate was found to be faster at 275.2 K than at 273.2 K even though the driving force at 273.2 K is larger than that of 275.2 K. Such a faster hydrate formation can be explained by relatively faster guest transfer at 275.2 K than near the freezing point. From an engineering point of view, the existence of possible optimum conditions and hydrate formation without stirring can be advantageous when applying porous media for hydrate formation in a hydrate-based process. In addition, a gas chromatograph was used to monitor the gas compositions during the hydrate formation/dissociation. The composition analysis suggests that methane (the most unstable guest in the natural gas components) is the slowest and the fastest guest for the hydrate formation and the dissociation, respectively. Therefore, when the natural
Kang and Lee gas is used for a hydrate-based process, methane concentration is thought to increase more than the initial feed concentration. In order to remove such methane accumulation in the natural gas and operate a continuous hydrate-based process, recycling or addition gas feeding is required for balancing the gas composition. Although this investigation was performed with the natural gas composition in the Korean domestic grid, the results can also be expanded for designing or operating any hydrate-based process using various gas compositions. References and Notes (1) Hammerschmidt, E. G. Ind. Eng. Chem. 1934, 26, 851. (2) Wilcox, W. I.; Carson, D. B.; Katz, D. L. Ind. Eng. Chem. 1941, 33, 662. (3) Sloan Jr, E. D.; Koh, C. A. Clathrate Hydrates of Natural Gases, 3rd ed.; CRC Press: Boca Raton, FL, 2008. (4) Ripmeester, J. A.; Tse, J. S.; Ratcliffe, C. I.; Powell, B. M. Nature 1987, 325, 135. (5) Collett, T. S.; Kuuskraa, V. A. Oil Gas J. 1998, 96, 90. (6) Park, Y.; Kim, D. Y.; Lee, J. W.; Huh, D. G.; Park, K. P.; Lee, J.; Lee, H. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 12690. (7) Lee, H.; Lee, J. W.; Kim, D. Y.; Park, J.; Seo, Y. T.; Zeng, H.; Moudrakovski, I. L.; Ratcliffe, C. I.; Ripmeester, J. A. Nature 2005, 434, 743. (8) Koide, H.; Takahashi, M.; Shindo, Y.; Tazaki, Y.; Iijima, M.; Ito, K.; Kimura, N.; Omata, K. Energy 1997, 22, 279. (9) Gudmundsson, J. S.; Parlaktuna, M.; Khokhar, A. A. SPE Prod. Facil. 1994, 9, 69. (10) Rogers, R. E.; Zhong, Y.; Etheridge, J. A.; Arunkumar, R.; Pearson, L. E.; Hogancamp, T. K. Micellar gas hydrate storage process. The 5th International Conference on Gas Hydrates, 2005, Trondheim, Norway. (11) Najibi, H.; Chapoy, A.; Tohidi, B. Fuel 2008, 87, 7. (12) Handa, Y. P.; Stupin, D. J. Phys. Chem. 1992, 96, 8599. (13) Uchida, T.; Ebinuma, T.; Ishizaki, T. J. Phys. Chem. B 1999, 103, 3659. (14) Uchida, T.; Ebinuma, T.; Takeya, S.; Nagao, J.; Narita, H. J. Phys. Chem. B 2002, 106, 820. (15) Anderson, R.; Llamedo, M.; Tohidi, B.; Burgass, R. W. J. Phys. Chem. B 2003, 107, 3500. (16) Kang, S. P.; Lee, J. W.; Ryu, H. J. Fluid Phase Equilib. 2008, 274, 68. (17) Seo, Y. T.; Moudrakovski, I. L.; Ripmeester, J. A.; Lee, J. W.; Lee, H. EnViron. Sci. Technol. 2005, 39, 2315. (18) Seo, Y.; Lee, S.; Cha, I.; Lee, J. D.; Lee, H. J. Phys. Chem. B 2009, 113, 5487. (19) Seo, Y.; Kang, S. P.; Jang, W. J. Phys. Chem. A 2009, 113, 9641. (20) Clarke, M.; Pooladi-Darvish, M.; Bishnoi, P. Ind. Eng. Chem. Res. 1999, 38, 2485. (21) Clennell, M. B.; Hovland, M.; Booth, J. S.; Henry, P.; Winters, W. J. J. Geophys. Res. 1999, 104, 22985. (22) Henry, P.; Thomas, M.; Clennell, M. B. J. Geophys. Res. 1999, 104, 23005. (23) Kreiter, S.; Feeser, V.; Kreiter, M.; Mo¨rz, T.; Grupe, B. Comput. Geosci. 2007, 11, 117. (24) Nogami, T.; Oya, N.; Ishida, H.; Matsumoto, H. Development of natural gas supply chain by means of natural gas hydrate (NGH). The 6th International Conference on Gas Hydrates, 2008, Vancouver. (25) Lee, Y. C.; Cho, B. H.; Baek, Y. S.; Mo, Y. G. Korean Chem. Eng. Res. 2004, 42, 168.
JP100812P
