Search for memory effects in methane hydrate: structure of water before hydrate formation and after hydrate decomposition

Neutron diffraction with HD isotope substitution has been used to study the formation and decomposition of the methane clathrate hydrate. Using this atomistic technique coupled with simultaneous gas consumption measurements, we have successfully tracked the formation of the sI methane hydrate from a water/gas mixture and then the subsequent decomposition of the hydrate from initiation to completion. These studies demonstrate that the application of neutron diffraction with simultaneous gas consumption measurements provides a powerful method for studying the clathrate hydrate crystal growth and decomposition. We have also used neutron diffraction to examine the water structure before the hydrate growth and after the hydrate decomposition. From the neutron-scattering curves and the empirical potential structure refinement analysis of the data, we find that there is no significant difference between the structure of water before the hydrate formation and the structure of water after the hydrate decomposition. Nor is there any significant change to the methane hydration shell. These results are discussed in the context of widely held views on the existence of memory effects after the hydrate decomposition.     

Gas hydrates of argon and methane synthesized at high pressures: composition, thermal expansion, and self-preservation

For the first time, the compositions of argon and methane high-pressure gas hydrates have been directly determined. The studied samples of the gas hydrates were prepared under high-pressure conditions and quenched at 77 K. The composition of the argon hydrate (structure H, stable at 460-770 MPa) was found to be Ar.(3.27 +/- 0.17)H(2)O. This result shows a good agreement with the refinement of the argon hydrate structure using neutron powder diffraction data and helps to rationalize the evolution of hydrate structures in the Ar-H(2)O system at high pressures. The quenched argon hydrate was found to dissociate in two steps. The first step (170-190 K) corresponds to a partial dissociation of the hydrate and the self-preservation of a residual part of the hydrate with an ice cover. Presumably, significant amounts of ice Ic form at this stage. The second step (210-230 K) corresponds to the dissociation of the residual part of the hydrate. The composition of the methane hydrate (cubic structure I, stable up to 620 MPa) was found to be CH(4).5.76H(2)O. Temperature dependence of the unit cell parameters for both hydrates has been also studied. Calculated from these results, the thermal expansivities for the structure H argon hydrate are alpha(a) = 76.6 K(-1) and alpha(c) = 77.4 K(-1) (in the 100-250 K temperature range) and for the cubic structure I methane hydrate are alpha(a) = 32.2 K(-1), alpha(a) = 53.0 K(-1), and alpha(a) = 73.5 K(-1) at 100, 150, and 200 K, respectively.     

High-pressure transitions in methane hydrate

Three recent studies of the high-pressure transformations of methane hydrate [Chem. Phys. Lett. 325 (2000) 490; Proc. Natl. Acad. Sci. 97 (2000) 13484; Nature 410 (2001) 661] have reported apparently different behaviours. Detailed comparison of our X-ray and neutron diffraction data with those obtained in earlier work shows that there is in fact consistent behaviour on isothermal compression at room temperature with a transition from the clathrate I structure hydrate to a hexagonal hydrate with an unknown structure at 0.9 GPa.     

NMR study of methane + ethane structure I hydrate decomposition

The thermally activated decomposition of methane + ethane structure I hydrate was studied with use of 13C magic-angle spinning (MAS) NMR as a function of composition and temperature. The observed higher decomposition rate of large sI cages initially filled with ethane gas can be described in terms of a model where a distribution of sI unit cells exists such that a particular unit cell contains zero, one, or two methane molecules in the unit cell; this distribution of unit cells is combined to form the observed equilibrium composition. In this model, unit cells with zero methane molecules are the least stable and decompose more rapidly than those populated with one or two methane molecules leading to the observed overall faster decomposition rate of the large cages containing ethane molecules.     

Phase transformation of methane hydrate under high pressure

In situ Raman spectroscopy is employed to study the phase behavior of methane hydrate at high pressure. The structure 1 of methane hydrate can be maintained up to 950 MPa at 299 K. The transformation of structure I<-->structure H+water+CH4 occurs at 880 MPa and 323 K. The structure H of methane hydrate, however, decomposes to methane and water at 960 MPa and 348 K. The initiation mechanism of methane hydrate sI is also discussed.     

Molecular simulation of non-equilibrium methane hydrate decomposition process

We recently performed constant energy molecular dynamics simulations of the endothermic decomposition of methane hydrate in contact with water to study phenomenologically the role of mass and heat transfer in the decomposition rate [S. Alavi, J.A. Ripmeester, J. Chem. Phys. 132 (2010) 144703]. We observed that with the progress of the decomposition front temperature gradients are established between the remaining solid hydrate and the solution phases. In this work, we provide further quantitative macroscopic and molecular level analysis of the methane hydrate decomposition process with an emphasis on elucidating microscopic details and how they affect the predicted rate of methane hydrate decomposition in natural methane hydrate reservoirs. A quantitative criterion is used to characterize the decomposition of the hydrate phase at different times. Hydrate dissociation occurs in a stepwise fashion with rows of sI cages parallel to the interface decomposing simultaneously. The correlations between decomposition times of subsequent layers of the hydrate phase are discussed.     

Phase diagram and high-pressure boundary of hydrate formation in the ethane-water system

Dissociation temperatures of gas hydrate formed in the ethane-water system were studied at pressures up to 1500 MPa. In situ neutron diffraction analysis and X-ray diffraction analysis in a diamond anvil cell showed that the gas hydrate formed in the ethane-water system at 340, 700, and 1840 MPa and room temperature belongs to the cubic structure I (CS-I). Raman spectra of C-C vibrations of ethane molecules in the hydrate phase, as well as the spectra of solid and liquid ethane under high-pressure conditions were studied at pressures up to 6900 MPa. Within 170-3600 MPa Raman shift of the C-C vibration mode of ethane in the hydrate phase did not show any discontinuities, which could be evidence of possible phase transformations. The upper pressure boundary of high-pressure hydrate existence was discovered at the pressure of 3600 MPa. This boundary corresponds to decomposition of the hydrate to solid ethane and ice VII. The type of phase diagram of ethane-water system was proposed in the pressure range of hydrate formation (0-3600 MPa).     

Experimental investigation of methane hydrate decomposition by depressurizing in porous media with 3-Dimension device

In order to simulate the behavior of gas hydrate formation and decomposition, a 3-Dimension experimental device was built, consisting of a high-pressure reactor with an inner diameter of 300 mm, effective height of 100 mm, and operation pressure of 16 MPa. Eight thermal resistances were mounted in the porous media at different depthes and radiuses to detect the temperature distribution during the hydrate formation/decomposition. To collect the pressure, temperature, and flux of gas production data, the Monitor and Control Generated System (MCGS) was used. Using this device, the formation and decomposition behavior of methane hydrate in the 20～40 mesh natural sand with salinity of 3.35 wt% was examined. It was found that the front of formation or decomposition of hydrate can be judged by the temperature distribution. The amount of hydrate formation can also be evaluated by the temperature change. During the hydrate decomposition process, the temperature curves indicated that the hydrate in the top and bottom of reactor dissociated earlier than in the inner. The hydrate decomposition front gradually moved from porous media surface to inner and kept a shape of column form, with different moving speed at different surface position. The proper decomposition pressure was also determined.     

Direct measurement of methane hydrate composition along the hydrate equilibrium boundary

The composition of methane hydrate, namely n(w) for CH4.n(w)H2O, was directly measured along the hydrate equilibrium boundary under conditions of excess methane gas. Pressure and temperature conditions ranged from 1.9 to 9.7 MPa and 263 to 285 K. Within experimental error, there is no change in hydrate composition with increasing pressure along the equilibrium boundary, but n(w) may show a slight systematic decrease away from this boundary. A hydrate stoichiometry of n(w) = 5.81-6.10 H2O describes the entire range of measured values, with an average composition of CH4.5.99(+/-0.07)H2O along the equilibrium boundary. These results, consistent with previously measured values, are discussed with respect to the widely ranging values obtained by thermodynamic analysis. The relatively constant composition of methane hydrate over the geologically relevant pressure and temperature range investigated suggests that in situ methane hydrate compositions may be estimated with some confidence.     

Investigation of hydrate formation in the system H2-CH4-H2O at a pressure up to 250 MPa

Phase equilibria in the system H2-CH4-H2O are investigated by means of differential thermal analysis within hydrogen concentration range 0-70 mol % and at a pressure up to 250 MPa. All the experiments were carried out under the conditions of gas excess. With an increase in hydrogen concentration in the initial gas mixture, decomposition temperature of the formed hydrates decreased. X-ray diffraction patterns and Raman spectra of the quenched hydrate samples obtained at a pressure of 20 MPA from a gas mixture containing 40 mol % hydrogen were recorded. It turned out that the hydrate has cubic structure I under these conditions. The Raman spectra showed that hydrogen molecules are not detected in the hydrate within the sensitivity of the method, that is, almost pure methane hydrate is formed. The general view of the phase diagram of the investigated system is proposed. A thermodynamic model was proposed to explain a decrease in hydrate decomposition temperature in the system with an increase in the concentration of hydrogen in the initial mixture.     

On the pressure-induced phase transformation in the structure II clathrate hydrate of tetrahydrofuran

A high-pressure phase of the clathrate hydrate of tetrahydrofuran was prepared by freezing a liquid phase of overall composition THF · 7 H₂O under a pressure of 3.0 kbar, or by pressurizing the solid structure II THF hydrate of 255K to 3.4 kbar. Unfortunately, the products recovered at 77K were always mixed phase materials as shown by X-ray powder diffraction. A number of diffraction lines could be indexed in terms of the cubic structure I hydrate with a slightly expanded lattice parameter, 12.08 Å, giving some support to Dyadin's idea that the high pressure phase transition involves a conversion of Structure II to Structure I. Other phases observed in the recovered product include Ice IX and amorphous materials. The reversion of the high pressure sample to the structure II hydrate was followed by differential scanning calorimetry. At ambient pressure, the high pressure sample converts slowly back to Structure II hydrate event at 77K.NRCC No. 35786.     

Methane hydrate formation and decomposition: structural studies via neutron diffraction and empirical potential structure refinement

Neutron diffraction studies with hydrogen/deuterium isotope substitution measurements are performed to investigate the water structure at the early, medium, and late periods of methane clathrate hydrate formation and decomposition. These measurements are coupled with simultaneous gas consumption measurements to track the formation of methane hydrate from a gas/water mixture, and then the complete decomposition of hydrate. Empirical potential structure refinement computer simulations are used to analyze the neutron diffraction data and extract from the data the water structure in the bulk methane hydrate solution. The results highlight the significant changes in the water structure of the remaining liquid at various stages of hydrate formation and decomposition, and give further insight into the way in which hydrates form. The results also have important implications on the memory effect, suggesting that the water structure in the presence of hydrate crystallites is significantly different at equivalent stages of forming compared to decomposing. These results are in sharp contrast to the previously reported cases when all remaining hydrate crystallites are absent from the solution. For these systems there is no detectable change in the water structure or the methane hydration shell before hydrate formation and after decomposition. Based on the new results presented in this paper, it is clear that the local water structure is affected by the presence of hydrate crystallites, which may in turn be responsible for the "history" or "memory" effect where the production of hydrate from a solution of formed and then subsequently melted hydrate is reportedly much quicker than producing hydrate from a fresh water/gas mixture.     

Molecular dynamics study of structure H clathrate hydrates of methane and large guest molecules

Methane storage in structure H (sH) clathrate hydrates is attractive due to the relatively higher stability of sH as compared to structure I methane hydrate. The additional stability is gained without losing a significant amount of gas storage density as happens in the case of structure II (sII) methane clathrate. Our previous work has showed that the selection of a specific large molecule guest substance (LMGS) as the sH hydrate former is critical in obtaining the optimum conditions for crystallization kinetics, hydrate stability, and methane content. In this work, molecular dynamics simulations are employed to provide further insight regarding the dependence of methane occupancy on the type of the LMGS and pressure. Moreover, the preference of methane molecules to occupy the small (5(12)) or medium (4(3)5(6)6(3)) cages and the minimum cage occupancy required to maintain sH clathrate mechanical stability are examined. We found that thermodynamically, methane occupancy depends on pressure but not on the nature of the LMGS. The experimentally observed differences in methane occupancy for different LMGS may be attributed to the differences in crystallization kinetics and/or the nonequilibrium conditions during the formation. It is also predicted that full methane occupancies in both small and medium clathrate cages are preferred at higher pressures but these cages are not fully occupied at lower pressures. It was found that both small and medium cages are equally favored for occupancy by methane guests and at the same methane content, the system suffers a free energy penalty if only one type of cage is occupied. The simulations confirm the instability of the hydrate when the small and medium cages are empty. Hydrate decomposition was observed when less than 40% of the small and medium cages are occupied.     

The structure of methane hydrate under geological conditions a combined Rietveld and maximum entropy analysis

We present a study of the structure of a fully deuterated methane hydrate under the geological conditions found in the world's oceans. In situ high-resolution neutron diffraction experiments have been performed at temperatures of 220, 275, and 280 K and a pressure of 100 bar, corresponding to the conditions at 1000 m water depth. The data were analyzed with a combination of Rietveld refinement and maximum entropy methods. From the Rietveld refinement, precise atomic parameters of the host lattice could be determined, indicating increasing distortions of the structure of the cages at elevated temperatures and pressures. Debye-Waller factors of the encaged CD(4) molecules have been found to exceed the values of the Debye-Waller factors of the D(2)O molecules considerably. In the large cage of structure type I the thermal center-of-mass displacements of the guests are 5-10 times larger than those of the water molecules. From the maximum entropy analysis maps of the scattering length density have been obtained, showing details of the vibrational amplitudes of the atoms in methane hydrate. The Debye-Waller factors of all molecules have been found to deviate considerably from a simple spherical geometry.     

Experimental studies of the formation and dissociation of methane hydrate in loess

In order to study the nature of gas hydrate in porous media, the formation and dissociation processes of methane hydrate in loess were investigated. Five cooling rates were applied to form methane hydrate. The nucleation times of methane hydrate formation at each cooling rate were measured for comparison. The experimental results show that cooling rate is a significant factor affecting the nucleation of methane hydrate and gas conversion. Under the same initial conditions, the faster the cooling rate, the shorter the nucleation time, and the lower the methane gas conversion. Five dissociating temperatures were applied to conduct the dissociation experiment of methane hydrate formed in loess. The experimental results indicated that the temperature evidently controlled the dissociation of methane hydrate in loess and the higher the dissociating temperature, the faster the dissociating rates of methane hydrate.     

Molecular vibrations of methane molecules in the structure I clathrate hydrate from ab initio molecular dynamics simulation

Vibrational frequencies of guest molecules in clathrate hydrates reflect the molecular environment and dynamical behavior of molecules. A detailed understanding of the mechanism for the vibrational frequency changes of the guest molecules in the clathrate hydrate cages is still incomplete. In this study, molecular vibrations of methane molecules in a structure I clathrate hydrate are calculated from ab initio molecular dynamics simulation. The vibrational spectra of methane are computed by Fourier transform of autocorrelation functions, which reveal distinct separation of each vibrational mode. Calculated symmetric and asymmetric stretching vibrational frequencies of methane molecules are lower in the large cages than in the small cages (8 and 16 cm(-1) for symmetric and asymmetric stretching, respectively). These changes are closely linked with the C-H bond length. The vibrational frequencies for the bending and rocking vibrational modes nearly overlap in each of the cages.     

Thermodynamic calculations in the system CH4-H2O and methane hydrate phase equilibria

Using the Gibbs function of reaction, equilibrium pressure, temperature conditions for the formation of methane clathrate hydrate have been calculated from the thermodynamic properties of phases in the system CH4-H2O. The thermodynamic model accurately reproduces the published phase-equilibria data to within +/-2 K of the observed equilibrium boundaries in the range 0.08-117 MPa and 190-307 K. The model also provides an estimate of the third-law entropy of methane hydrate at 273.15 K, 0.1 MPa of 56.2 J mol(-1) K(-1) for 1/nCH4.H2O, where n is the hydrate number. Agreement between the calculated and published phase-equilibria data is optimized when the hydrate composition is fixed and independent of the pressure and temperature for the conditions modeled.     

Cage occupancies in the high pressure structure H methane hydrate: a neutron diffraction study

A neutron diffraction study was performed on the CD(4) : D(2)O structure H clathrate hydrate to refine its CD(4) fractional cage occupancies. Samples of ice VII and hexagonal (sH) methane hydrate were produced in a Paris-Edinburgh press and in situ neutron diffraction data collected. The data were analyzed with the Rietveld method and yielded average cage occupancies of 3.1 CD(4) molecules in the large 20-hedron (5(12)6(8)) cages of the hydrate unit cell. Each of the pentagonal dodecahedron (5(12)) and 12-hedron (4(3)5(6)6(3)) cages in the sH unit cell are occupied with on average 0.89 and 0.90 CD(4) molecules, respectively. This experiment avoided the co-formation of Ice VI and sH hydrate, this mixture is more difficult to analyze due to the proclivity of ice VI to form highly textured crystals, and overlapping Bragg peaks of the two phases. These results provide essential information for the refinement of intermolecular potential parameters for the water-methane hydrophobic interaction in clathrate hydrates and related dense structures.     

Molecular insights into the heterogeneous crystal growth of si methane hydrate

In this paper we report a successful molecular simulation study exploring the heterogeneous crystal growth of sI methane hydrate along its [001] crystallographic face. The molecular modeling of the crystal growth of methane hydrate has proven in the past to be very challenging, and a reasonable framework to overcome the difficulties related to the simulation of such systems is presented. Both the microscopic mechanisms of heterogeneous crystal growth as well as interfacial properties of methane hydrate are probed. In the presence of the appropriate crystal template, a strong tendency for water molecules to organize into cages around methane at the growing interface is observed; the interface also demonstrates a strong affinity for methane molecules. The maximum growth rate measured for a hydrate crystal is about 4 times higher than the value previously determined for ice I in a similar framework (Gulam Razul, M. S.; Hendry, J. G.; Kusalik, P. G. J. Chem. Phys. 2005, 123, 204722).     

Water transfer characteristics in the vertical direction during methane hydrate formation and dissociation processes inside non-saturated media

In order to study water transfer characteristics inside non-saturated media during methane hydrate formation and dissociation processes, water changes on the top, middle and bottom locations of experimental media during the reaction processes were continuously followed with a novel apparatus with three pF-meter sensors. Coarse sand, fine sand and loess were chosen as experimental media. It was experimentally observed that methane hydrate was easier formed inside coarse sand and fine sand than inside loess. Methane hydrate formation configuration and water transfer characteristics during methane hydrate formation processes were very different among the different non-saturated media, which were important for understanding methane hydrate formation and dissociation mechanism inside sediments in nature.