X-ray Computed Tomography Observation of Methane Hydrate Dissociation

Deposits of naturally occurring methane hydrate have been identified in permafrost and deep oceanic environments with global abundance estimated to be twice the total amount of energy stored in fossil fuels. The fundamental behavior of methane hydrate in natural formations, while poorly understood, is of critical importance if the economic recovery of methane from hydrates is to be accomplished. In this study, computed X-ray tomography (CT) scanning is used to image an advancing dissociation front in a laboratory-prepared, heterogeneous gas hydrate/sand sample at 0.1 MPa. The cylindrical methane hydrate and sand aggregate, 2.54 cm in diameter and 6.3 cm long, was contained in a PVC sample holder insulated on all but one end. At the uninsulated end, the dissociated gas was captured and the volume of gas monitored. The sample was initially imaged axially using Xray CT scanning within the methane hydrate stability zone at 77°K. As the sample was subsequently warmed through the methane hydrate dissociation point at 194°K and room pressure, gas was produced and the temperature at the bottom of the sample plug was monitored while CT images were acquired. The experiment showed that CT imaging can resolve the reduction in density (as seen by a reduction in beam attenuation) of the hydrate/sand aggregate due to the dissociation of methane hydrate. In addition, a comparison of CT images with gas flow and temperature measurements reveals the CT scanner is able to resolve accurately and spatially the advancing dissociation front. Future experiments designed to better understand the thermodynamics of hydrate dissociation are planned to take advantage of the temporal and spatial resolution that the CT scanner provides. Introduction In the oil and gas industry, gas hydrate research has traditionally focused on designing improved methods to prevent gas hydrate formation in conduits where it impedes fluid transport in subsea or permafrost operations (1,2,3). More recently, interest in hydrates has been directed toward understanding the vast potential of gas hydrates as a natural gas resource (4). Field, laboratory, and theoretical work have begun focusing on the many aspects of gas hydrates including reserve evaluation, hydrate recovery, gas production, hydrate formation, hydrate properties, dissociation, depressurization modeling, and thermal gas production from hydrates (5). To help refine current models of gas production from natural gas hydrate accumulations (6), a better understanding of the physical processes occurring during methane hydrate dissociation is critically important. Many physical and hydrologic properties of hydrate/sediment aggregates require evaluation to gain a further understanding of the potential for gas recovery from hydrates. These properties include relative permeability, thermal conductivity, heat capacity, and compaction. While various aspects of gas hydrate formation and dissociation have been investigated on both natural (7) and synthetic (8,9) gas hydrate test materials, synthetic hydrates offer the advantage of better sample control and uniformity than natural materials. Furthermore, representative natural hydrate samples have proved to be very difficult to collect. Imaging methods, such as X-ray Computed Tomography (CT), have been used to characterize naturally occurring hydrates (10) and to directly observe their dissociation while reducing the ambient pressure below the stability pressure (11). In this study, we describe CT scanning experiments on an aggregated sample of methane hydrate with quartz sand, and show how this technique can be used successfully to image an advancing dissociation front due to heat influx applied to one end of the sample, at a constant pressure of 0.1 MPa. Methods Sample fabrication and preparation Test material of polycrystalline methane hydrate was synthesized by combining cold, pressurized CH4 gas (250 K, 27 MPa) with granular H2O ice “seeds” (typically 26 to 32 g, at 180-250 μm grain size) in stainless steel reaction vessels SPE 75533 X-ray Computed Tomography Observation of Methane Hydrate Dissociation Liviu Tomutsa, SPE, Barry Freifeld, Timothy J. Kneafsey, Lawrence Berkeley National Laboratory, Laura A. Stern, United States Geological Survey 2 L. TOMUTSA, B. M. FREIFELD, T. J. KNEAFSEY, L. A. STERN SPE 75533 [6,7]. Heating the ice + gas reactants above the H2O melting point promotes in situ conversion of ice grains to hydrate grains by the general reaction CH4 (g) + 5.9H2O (s → l) → CH4·5.9H2O (8,9). Complete reaction was attained by continued warming to 290 K and ~30 MPa for approximately 12 to 15 hours. During hydrate synthesis, sample temperature (T) was monitored by axially positioned thermocouples in side-by-side companion samples, and methane gas pressure (PCH4) was monitored by a Heise bourdon gauge and pressure transducers. Complete reaction was determined from the P-T synthesis record, as hydrate formation consumes a known mass of the vapor phase and thus causes a predictable pressure offset in the P-T record. Samples were then cooled to 250 K, and the lack of discontinuities in their thermal profiles upon crossing the ice point provided further substantiation that no measurable unreacted H2O ice remained. The resulting hydrate product is highly reproducible in composition as well as in grain and pore characteristics; test specimens are nearly pure (> 99 vol. %), porous (~ 29%) cohesive cylinders of 2.54 cm in diameter by 9 to 12 cm in length, depending on the initial amount and packing of the seed ice (Figure 1). The hydrate has controlled grain size, random crystallographic orientation, and no detectable secondary phases. The methane hydrate/quartz sand samples used in this experiment were produced by mixing measured and prespecified amounts of granular “seed” ice with quartz sand (100-150 μm grain size, Oklahoma #1), then packing the ice + sediment mixtures into the reaction vessels prior to admission of CH4 gas (8). Using the same static growth method described above, no detectable migration of either the H2O or sediment was observed during subsequent heating and conversion of the ice grains to hydrate grains. For the hydrate + quartz experiments described in the current study, all test material came from one initial sample, shown at right in Figure 1. For this sample, ice + sand grains were pre-mixed and packed as four discrete layers to yield final compositions of (from top to bottom): 75 vol. % hydrate + 25 vol. % sand; 60 vol. % hydrate + 40 vol. % sand; 40/60; 25/75. The top and bottom segments were removed and we only consider the solid material (methane hydrate and sand) in our classification of the samples as 40% sand/60% hydrate etc. All layers were relatively porous (~49%). Test procedures The sample used in our experiment was a 2.54 cm diameter x 6.3 cm long composite with about half of the sample composed of 60%/40% sand/hydrate (by volume), and the other half 40%/60% sand/hydrate (by volume). Note that we do not include porosity in these descriptions. A gas-tight PVC sample holder with plastic fittings was constructed to contain the sample (Figure 2). Materials were selected to minimize X-ray attenuation. A sheathed type-J thermocouple was inserted through the sample holder bottom such that the sample would rest in contact with the thermocouple tip. The sample holder top was connected to the gas collection system. The sample holder was set into an insulating Styrofoam container and stabilized with Styrofoam