1245
2009, 113, 1245–1248 Published on Web 01/09/2009
Effect of Interlayer Ions on Methane Hydrate Formation in Clay Sediments Sun-Hwa Yeon,† Jiwoong Seol,† Young-ju Seo,† Youngjune Park,† Dong-Yeun Koh,† Keun-Pil Park,‡ Dae-Gee Huh,‡ Jaehyoung Lee,‡ and Huen Lee*,† Chemical and Biomolecular Engineering Department, KAIST, 373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, Republic of Korea, and Korea Institute of Geoscience and Mineral Resources, 30 Gajeong-dong, Yuseong-gu, 305-350 Daejeon, Republic of Korea ReceiVed: NoVember 17, 2008; ReVised Manuscript ReceiVed: December 24, 2008
Natural methane hydrates occurring in marine clay sediments exhibit heterogeneous phase behavior with high complexity, particularly in the negatively charged interlayer region. To date, the real clay interlayer effect on natural methane hydrate formation and stability remains still much unanswered, even though a few computer simulation and model studies are reported. We first examined the chemical shift difference of 27Al, 29 Si, and 23Na between dry clay and clay containing intercalated methane hydrates (MH) in the interlayer. We also measured the solid-state 13C MAS NMR spectra of MH in Na-montmorillonite (MMT) and Camontmorillonite (MMT) to reveal abnormal methane popularity established in the course of intercalation and further performed cryo-TEM and XRD analyses to identify the morphology and layered structure of the intercalated methane hydrate. The present findings strongly suggest that the real methane amount contained in natural MH deposits should be reevaluated under consideration of the compositional, structural, and physical characteristics of clay-rich sediments. Furthermore, the intercalated methane hydrate structure should be seriously considered for developing the in situ production technologies of the deep-ocean methane hydrate. Methane hydrate has received sustained attention due to its tremendous value as a future energy resource.1,2 Deep-sea methane hydrate occurs naturally in large landfills but exists in these locations mostly in dispersed particles of sediments. Most research concerning methane hydrate has focused on the homogeneous ice-like pure crystalline phase combined with various guest molecules,3,4 through microscopic or macroscopic approaches, despite the fact that methane hydrates occur in heterogeneous sediments just below the sea floor under certain concentration, pressure, and temperature conditions.5-9 Clay minerals have a layered structure with unit layers that are approximately 1 nm thick containing micrometer-sized crystalline particles, and are involved in important surface activity phenomena such as industrial catalysis and the oceanic buffering of atmospheric CO2 on a global scale.10 However, acceptable and clear evidence to attain a solid understanding of methane hydrate formation in confined clay layers has not yet been reported. The unique surroundings of clay interlayer, where there is a high interaction complexity among clay, water, light hydrocarbons, and cations, are expected to strongly influence hydrate nucleation characteristics such as host water-lattice formation and, more importantly, competitive occupancy patterns of guest species, methane, and cations in cages. A comprehensive understanding of the dynamic guest inclusion phenomenon in clay interlayer will make it possible to more accurately evaluate the genuine amount of methane stored in deep-sea floor gas hydrate sediments. * Corresponding author. E-mail:
[email protected]. † KAIST. ‡ Korea Institute of Geoscience and Mineral Resources.
10.1021/jp810079c CCC: $40.75
A microscopic analysis to gain fundamental insight into the growth of intercalated methane hydrate has revealed the important role of the interlayer cations in elucidating hydrate stability by providing an ion nucleation site. We first confirm abnormal methane popularity established in the course of intercalation through solid-state MAS NMR. Furthermore, cryoTEM and XRD analyses were carried out to support the morphology and layered structure of the intercalated methane hydrate. CH4 gas with a purity of 99.9 mol % was supplied by World Gas Co., Korea. Deionized water was purchased from Merck Chemical Co., Germany. The clay was Na-montmorillonite (purchased from Crook County, Wyoming type, U.S.A.) with a scientific designation of SWy-2 (Na-MMT). The average pore size of the clay minerals is 90.3315 Å, the average surface area of the pores is 35.1711 m2/g, and the average volume of the pores is 0.079426 cm3/g. These clays include members of the dioctahedral and trioctahedral series with various tetrahedral Si/ Al ratios and with different interlayer compositions. The 2:1 layer type has two tetrahedral sheets fused to an octahedral sheet. Their generic chemical formula is Na0.32[Al3.01Fe(III)0.41Mn0.01Mg0.54Ti0.02][Si7.98Al0.02]O20(OH)4. In addition, Ca-montmorillonite was obtained from the Cheto Mine (Cheto type), Apache County, AZ; its scientific designation is SAz-1 (CaMMT), and its generic chemical formula is Ca0.39[Al2.71Mg1.11Fe(III)0.12Mn0.01Ti0.03][Si8]O20(OH)4. The average surface area of the pores is 97.42 m2/g. Two clay mineral materials (Namontmorillonite Swy-2 and Ca-montmorillonite Saz-1) were purchased from Society’s Source Clay Repository. Samples with various clay contents (7.5-80 wt %) were prepared by dispersing the clay in water. In order to obtain highly varied and 2009 American Chemical Society
1246 J. Phys. Chem. B, Vol. 113, No. 5, 2009
Letters
Figure 2. Solid-state 13C MAS NMR spectra of (a) Na-MMT MH and (b) Ca-MMT (Cheto type) MH.
Figure 1. Solid-state MAS NMR spectra of (a) 27Al MAS NMR, (b) 29 Si MAS NMR, and (c) 23Na MAS NMR. Na-MMT (blue, down) and 25 wt % Na-MMT MH (red, up). The asterisks (*) show the sidebands.
complex rheological behavior, the swollen clay was ground in a jewel bowl until a gel phase appeared. For a specifically designed spectroscopic analysis, gel-phase clay samples prepared at 213 K were finely (100-150 µm mesh) pulverized using a mortar and pestle at a temperature equal to that of liquid nitrogen temperature. They were then exposed to pure methane gas at 120 bar and a temperature of 267 K for 3 days for favorable hydrate synthesis in the clay interlayer. We also used a parallel plate rheometer (model: Ares 4, Rheometric Scientific). For the solid-state 13C, 27Al, 29Si, and 23Na MAS NMR spectra experiments, respectively, the hydrate sample was powdered under liquid nitrogen temperature conditions, and was then placed under atmospheric conditions in a 4 mm o.d. zirconia rotor loaded into a variable temperature probe. The NMR rotor was sealed using a rotor cap after the hydrate sample was packed. The working temperature for the NMR experiment was 223 K for 13C, 27Al, 29Si, and 23Na MAS NMR. The structure of the sample was determined using a D/max-RB (Rigaku) diffractometer with Cu KR1 radiation (λ ) 1.5406 Å), and the diffraction peaks were at 1 bar and 123.15 K. To obtain a TEM image, the clay MH sample was sectioned at 153 K with a thickness of 70 nm using CryoUltra microtorming and was laid on a 300-mesh copper carbon grid. Transmission electron microscopy (TEM, Tecnai G2 Spirit, FEI Company, U.S.A.)
Figure 3. (a) Cage occupancy ratio of Na-MMT MH (black) and CaMMT (Cheto type) MH (red) and (b) viscosity (shear stress vs shear rate) comparison data of three clay types (Otay Ca-MMT, Cheto CaMMT, and Wyoming Na-MMT) hydrated by 10 wt %.
Letters
J. Phys. Chem. B, Vol. 113, No. 5, 2009 1247
Figure 4. Cryo-TEM images (up) measured at 77 K and PXRD (down) measured at 123.15 K of pure Na-MMT powder (a) and 15 wt % Na-MMT MH (b). The PXRD data of 15 wt % Na-MMT MH were indexed with clay, methane hydrate(sI), and hexagonal ice phases.
with an integrated cryo system at 120 kV with a resolution of 3.4 Å (lattice image 2 Å) at 77 K was used. Clay minerals have a layered structure with unit layers that are approximately 1 nm thick with micrometer-sized crystalline particles, implying that the properties of methane hydrates confined in the clay interlayer are vastly different from those of methane hydrates in the bulk. A noticeable change in the specific interactions between clay elements and methane was initially found by comparing the NMR spectra of pure and methane-hydrated clays. The solid-state 27Al, 29Si, and 23Na MAS NMR spectra of pure Na-MMT and 25 wt % Na-MMT methane hydrate (MH) samples are shown in Figure 1. The 27Al MAS NMR spectra of two samples show structural interpretations of Al atoms for octahedral (6Al) and tetrahedral (4Al) sites in the clay layer. Both yield identical 4Al and 6Al peaks with chemical shifts of 60.34 and 1.44 ppm, respectively. The 29Si MAS NMR spectra of these two samples were found to be identical to those shown in Figure 1b. It is known that, in Namontmorillonite, approximately one-fourth of the Si in the tetrahedral sheet is replaced by Al. These substitutions induce a negative charge in a layer compensated by interlayer Na+ ions that is located between two adjacent layers.11,12 Accordingly, the environmental differences in the vicinity of Si depend solely on the Al and Si distributions, as all atoms of the tetrahedral sheet have the same second neighbors in both the interlayer space and the octahedral sheet. The lines in the figure are assigned to Si surrounded by three Si, at -133.44 ppm, and two Si + one Al, at -119.15 ppm. The consistency in the chemical shift confirms that a noticeable structural transformation caused by Al and Si interactions with neighboring elements
does not occur during hydration. However, it was observed that Na chemical shifts appear at -49.68 ppm for pure clay and at -24.87 ppm for hydrated clay, implying that the Na+ ions behave differently in the hydrated clay interlayer, where Na-OH2O bonds are much weaker and longer than Si-Oclay or Al-Oclay bonds. In particular, the 23Na NMR shielding with methane hydrate confirms that Na+ ions in hydrated clay are more strongly bound with a hydrated, well-structured clay-water network than in dehydrated pure clay. To identify the actual nature of sodium ions, the solid-state 13C MAS NMR spectra of MH in Na-MMT and Ca-MMT (Cheto type) were evaluated, revealing chemical shifts of -6.7 ppm for CH4 in sI-L and -4.3 ppm for CH4 in sI-S13(Figure 2). All methane hydrates formed under various clay concentrations have identical chemical shifts without any structure changes. However, it was noted that the CH4 distribution ratio between the sI-S and sI-L significantly changes with the clay concentration (Figure 3). The CH4 area ratio of large to small cages in pure bulk methane hydrate is known to be approximately 3.7:1, but the corresponding ratio in Na-MMT MH increases with the clay concentration, approaching 7.1:1 for 45 wt % Na-MMT MH. The significant 23Na chemical shift change strongly supports the idea that the interlayer Na+ cations somehow affect CH4 occupancy in sI-S. With respect to molecule size, the Na+ ion (0.97 Å, ionic radius) has much smaller structural dimensions than the CH4 molecule (2.28 Å, radius). A similar intercalated structure, as shown in the 13C NMR of Figure 2b, was obtained for Ca-MMT (Cheto type) with a different interlayer cation of Ca2+ in which a gelphase formation was possible up to a high clay concentration of 80%. The ACH4L/ACH4S value of Ca-MMT MH appears to be 23
1248 J. Phys. Chem. B, Vol. 113, No. 5, 2009 nearly identical to that of Na-MMT MH over clay concentrations of 0-40%; however, at nearly 60%, it reaches a maximum while at higher clay concentrations it remains unchanged. In Na-MMT MH, the experimentally feasible maximum concentration was 45-50 wt %, because the Wyoming-type montmorillonite (NaMMT) suspensions with a high percentage of exchangeable Na+ show a higher viscosity value, yielding higher stress and thixotropy than the Cheto-type montmorillonite (Ca-MMT) suspensions (Figure 3b). Figure 4 shows cryo-TEM images (up) and XRD patterns (down) of pure Na-MMT powder and 15 wt % Na-MMT MH. The clay interlayers of pure Na-MMT powder and 15 wt % Na-MMT MH are discernible as solid dark lines with interspacing distances of 1.2 and 1.6 nm, respectively. The XRD powder data at 123 K of the pure Na-MMT powder exhibit a (001) diffraction pattern with an interlayer d-spacing of 12.01 Å. In the XRD result of the 15 wt % Na-NNT MH, three phases, such as clay with an increased interlayer d-spacing of 16.01 Å, methane hydrate of cubic sI with a ) 11.897 ( 0.002 Å, and ice of hexagonal structure, were observed, indicating that MH is intercalated in the clay layer. The clay interlayer d-spacings of pure Na-MMT powder and 15 wt % Na-MMT MH agree well with those observed in TEM. In conclusion, the present findings strongly suggest that the real methane amount contained in natural MH deposits should be reevaluated under consideration of the compositional, structural, and physical characteristics of clay-rich sediments. Furthermore, the intercalated methane hydrate structure should be seriously considered for developing the in situ production technologies of the deep-ocean methane hydrate. Acknowledgment. This work was supported by the Korea Science and Engineering Foundation (KOSEF) through the National Research Laboratory Program funded by the Ministry
Letters of Education, Science and Technology (No. R0A-2005-00010074-0(2007)), “Identifying the Nature Phenomenon of Methane Hydrate Formation in Heterogeneous Clay Layer and Its Replacement Mechanism” funded by the Ministry of Knowledge Economy, Nuclear R&D Program of the Ministry of Education, Science and Technology (M20702000164-07B0200-16410), and partially funded by the Brain Korea 21 Project. We are grateful to Korea Basic Science Institute for experimental help of TEM images. References and Notes (1) Lu, H.; Seo, Y.; Lee, J.; Moudrakovski, I.; Ripmeester, J. A.; Chapman, N. R.; Coffin, R. B.; Gardner, G.; Pohlman, J. Nature 2007, 445, 303–306. (2) Udachin, K. A.; Lu, H.; Enright, G. D.; Ratcliffe, C. I.; Ripmeester, J. A.; Chapman, N. R.; Riedel, M.; Spence, G. Angew. Chem., Int. Ed. 2007, 46, 8220–8222. (3) 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. (4) Yeon, S. H.; Seol, J.; Lee, H. J. Am. Chem. Soc. 2006, 128, 12388– 12389. (5) Kim, D.-Y.; Lee, J.-w.; Seo, Y.-T.; Ripmeester, J. A.; Lee, H. Angew. Chem., Int. Ed. 2005, 44, 7749–7752. (6) Kim, D.-Y.; Park, J.; Lee, J.-w.; Ripmeester, J. A.; Lee, H. J. Am. Chem. Soc. 2005, 128, 15360–15361. (7) Dillon, W. P.; Lee, M. W.; Doleman, D. F. Ann. N.Y. Acad. Sci. 1994, 715, 364–380. (8) Kevenvolden, K. A. ReV. Geophys. 1993, 31, 173–187. (9) Kevenvolden, K. A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3420– 3426. (10) Michalopoulos, P.; Aller, R. C. Science 1995, 270, 614–617. (11) Sanz, J.; Serratosa, J. M. J. Am. Chem. Soc. 1984, 106, 4790– 4793. (12) Fette, G.; Tichit, D.; Menorval, L. C.; Figueras, F. Appl. Catal., A 1995, 126, 165–176. (13) Seo, Y. T.; Lee, H.; Moudrakovski, I. L.; Ripmeester, J. A. ChemPhysChem 2003, 4, 379–382.
JP810079C
