OS22A-01
The Methane Hydrate Reservoir System
We use multi phase flow modeling and field examples (Hydrate Ridge, offshore Oregon and Blake Ridge, offshore North Carolina) to demonstrate that the methane hydrate reservoir system links traditional and non- traditional hydrocarbon system components: free gas flow is a fundamental control on this system. As in a traditional hydrocarbon reservoir, gas migrates into the hydrate reservoir as a separate phase (secondary migration) where it is trapped in a gas column beneath the base of the hydrate layer. With sufficient gas supply, buoyancy forces exceed either the capillary entry pressure of the cap rock or the fracture strength of the cap rock, and gas leaks into the hydrate stability zone, or cap rock. When gas enters the hydrate stability zone and forms hydrate, it becomes a very non traditional reservoir. Free gas forms hydrate, depletes water, and elevates salinity until pore water is too saline for further hydrate formation: salinity and hydrate concentration increase upwards from the base of the regional hydrate stability zone (RHSZ) to the seafloor and the base of the hydrate stability zone has significant topography. Gas chimneys couple the free gas zone to the seafloor through high salinity conduits that are maintained at the three-phase boundary by gas flow. As a result, significant amounts of gaseous methane can bypass the RHSZ, which implies a significantly smaller hydrate reservoir than previously envisioned. Hydrate within gas chimneys lie at the three-phase boundary and thus small increases in temperature or decreases in pressure can immediately transport methane into the ocean. This type of hydrate deposit may be the most economical for producing energy because it has very high methane concentrations (Sh > 70%) located near the seafloor, which lie on the three-phase boundary.
OS22A-02
Formation of Regional Bottom-Simulating Seismic Reflectors (BSRs) and Their Use in Quantifying Upward Fluid Flow
It is becoming apparent that the character of regional hydrate and free-gas distributions in passive-margin environments is different to the equivalent distributions in accretionary wedges at convergent margins. Accretionary wedges typically have widespread BSRs with underlying gas zones of a few tens of metres in thickness; passive margins have comparatively rare BSRs with gas zones typically a few hundred metres thick. In both cases, the sub-BSR free gas typically occupies a few percent, or less, of pore space. Hydrate recycling caused by seabed uplift, sedimentation, bottom-water warming etc. (where free gas comes from dissociating hydrate) is one mechanism for producing sub-BSR free gas that works well in accretionary wedges. However, hydrate recycling does not work well in passive margins where recycling mechanisms are slower than in accretionary wedges (there is no seabed uplift), but the sub-BSR free-gas zones are thicker: we expect the opposite if hydrate recycling is the primary mechanism producing sub-BSR free gas. Here we present a mechanism for the production of gas beneath hydrate-bearing sediments that is complementary to hydrate recycling in all environments, but which becomes the dominant mechanism in passive-margin settings. This mechanism produces a downward decreasing distribution of gas in low concentrations (few percent of pore space) across thick zones (hundreds of metres) when the upward fluid flow is low (less than a few tenths of mm/yr) and when the gas--water solubility curve is downward decreasing beneath the gas hydrate stability zone. The latter requires moderate to high geothermal gradients and pressures. We find that the sub-BSR free-gas zone produced by the solubility-curvature mechanism achieves a steady-state thickness that is controlled by the rate of upward fluid flow and the concentration of methane dissolved within it. Using the seismically derived free- gas zone thickness for a test case offshore Svalbard (Norwegian Arctic) as the steady-state thickness, we show that the solubility curvature mechanism can be used to forward model the observed gas distribution and, consequently, to constrain the rate of upward fluid flow.
OS22A-03
Grain Scale Study of Hydrate Formation in Sediments From Methane Gas: A Coupled Fluid- Solid Interaction Model
Ocean sediments bearing methane hydrates exhibit a range of behavior, from cold seeps where solid and gas phases co-exist in the hydrate stability zone (HSZ), to essentially static accumulations where solid and liquid co- exist. This paper and its companion (Behseresht, Prodanovic and Bryant) describe the development and application of models for grain-scale phenomena governing in situ gas-to-hydrate conversion. The motivation is the following hypothesis: as gas phase pore pressure varies, the competition between brine displacement and sediment fracturing determines the extent of conversion of methane gas entering the HSZ to hydrate. Here we present a discrete element method (DEM) to model the strong coupling that takes place between the pore fluids (brine and methane gas) and the mechanical behavior of the sediment. In a discrete element method, each element or grain is an individual entity, identified by its size, mass and moments of inertia. Newton's second law dictates the motion of the assembly of grains. For dry systems, the grain-scale forces are limited to interactions at grain contacts. In contrast, when one or more fluids are present, additional pore-scale forces play a significant role: a set of forces due to pore fluid pressure, and another set due to surface tension between fluids. We develop a self-consistent fluid-solid interaction (FSI) model at the grain scale, in which these additional sets of forces are introduced rigorously. Our computational model captures the two-way coupling between multiphase fluid flow and sediment mechanics, which we validate by means of triaxial laboratory experiments. In particular, this allows us to determine the conditions under which gas invasion fractures the sediment. This determines the distribution of methane gas and hydrate which, in turn, has direct implications on the likelihood that gas and hydrate will co-exist, and on the overall size of the energy resource. Work is under way to couple this grain mechanics model with a capillarity-controlled displacement model, described in the companion paper by Behseresht, Prodanovic and Bryant. http://www.netl.doe.gov/technologies/oil- gas/FutureSupply/MethaneHydrates/projects/DOEProjects/MH_43
OS22A-04
Pore Scale Mechanistic Study of the Preferential Mode of Hydrate Formation in Sediments: Fluid Flow Aspects
A spectrum of behavior is encountered in ocean sediments bearing methane hydrates, ranging from essentially static accumulations where hydrate and brine co-exist, to active cold seeps where hydrate and a methane gas phase co-exist in the hydrate stability zone (HSZ). In this and a companion paper (Jain and Juanes) we describe methods to test the following hypothesis: the coupling between drainage and fracturing, both induced by pore pressure, determines whether methane gas entering the HSZ is converted completely to hydrate. Here we describe a novel implementation of the level set method (LSM) to determine the capillarity-controlled displacement of brine by gas from sediment and from fractures within the sediment. Predictions of fluid configurations in infinite-acting model sediments indicate that the brine in drained sediment (after invasion by methane gas) is better connected than previously believed. This increases the availability of water and the rate of counter-diffusion of salinity ions, thus relaxing the limit on hydrate build-up within gas- invaded grain matrix. Simulated drainage of a fracture in sediment shows that points of contact between fracture faces are crucial. They allow residual water saturation to remain within an otherwise gas-filled fracture. Simulations of imbibition, which can occur for example after drainage into surrounding sediment reduces gas phase pressure in the fracture, indicate that the gas/water interfaces at contact points significantly shifts the threshold pressures for withdrawal of gas. During both drainage and imbibition, the contact points greatly increase water availability for hydrate formation within the fracture. We discuss coupling this capillarity-controlled displacement model with a discrete element model for grain-scale mechanics. The coupled model provides a basis for evaluating the macroscopic conditions (thickness of gas accumulation below the hydrate stability zone; average sediment grain size; principal earth stresses) favoring co- existence of methane gas and hydrate in the HSZ. Explaining the range of behavior is useful in assessing resource volumes and evaluating pore-to-core scale flow paths in production strategies.
OS22A-05
Growth kinetics and microstructure of methane hydrates formed in porous media
The occurrence of natural gas hydrates within sediments is known from a large number of locations. They commonly occupy pore spaces cementing sedimentary deposits. Yet, detailed information about the influence of mineral composition on the formation process in porous media is still very limited. Laboratory investigations of the microstructure of gas hydrate in porous media, as a function of p-T conditions, mineral composition and water/gas supersaturation are therefore of considerable interest. Such studies may allow a better understanding of the formation process and even the prediction of accumulation /decomposition rates of some natural gas hydrates in a given geological setting. As a model study, we carried out various reactions with methane gas and water in three types of media: 1) quartz, 2) quartz + kaolinite, 3) quartz + montmorillonite. The progress of the reactions was recorded by gas consumption (pressure drop) at 3°C. Samples recovered at various stages of the formation or decomposition reactions were investigated using field-emission scanning electron microscopes (FE-SEM) equipped with a cryo-stage [1]. In the SEM investigations, methane hydrates appeared between the quartz grains acting as cement. Kaolinite particles were observed as a filigree network on the surface of hydrate cement, while montmorillonite form flakes or crust like features. Each of the minerals may play individual/coupled interaction with water and gas hydrate, and thereby display a characteristic configuration in the SEM images. Dissimilar kinetic features, using different porous media at the investigated conditions, confirm that mineral composition directly influences the progress of gas hydrate formation. Medium 3 shows the fastest hydrate saturation. With increasing water content of the porous media the formation tends to proceed in a multi-stage process with a slower diffusion-limited later stage. Reference: [1] A. Klapproth, K. Techmer, S.A. Klapp, M.M. Murshed and W.F. Kuhs., In Physics and Chemistry of ice (ed. by W. F. Kuhs), RSC Publishing, Cambridge 2007, pp.321-328.
OS22A-06
Molecular and dissociation studies of natural gas hydrates collected from different oceanic environments
Natural gas hydrates occur globally in marine sediments or in permafrost regions when specific conditions of high pressure, low temperature and sufficiently methane concentration are combined to initiate their formation and stabilize their structure. As well as they appear attractive for gas industry, natural gas hydrates can have an important impact in continental slope stability or climate change. Therefore, it is important to focus our attention on structural evolution and thermodynamical stability of these natural minerals. For this, high-resolution powder X-ray synchrotron diffraction and Raman spectroscopy techniques are efficient and powerful tools to determine the hydrate structures. We performed a first physical characterization of two intact natural gas hydrates from the Congo-Angola and the Nigerian margin by X-ray synchrotron diffraction. The collected samples exhibit a preponderance of structure I (sI) (cubic lattice with space group Pm n). The Rietveld refinement of lattice parameters for the type I structure gives values intermediate between lattice constant of less pure methane specimens and pure artificial methane hydrates. This indicates that lattice constant can be affected by the presence of encaged CO2, H2S and other gas molecules, even in small amount. Thermal expansion is also presented for Congo-Angola hydrate in the temperature range 90-200 K and coefficients are comparable with values reported for synthetic hydrates at low temperature, whereas they tend to approach ice thermal expansion coefficient at higher temperature. In a second step, we performed a physical characterization by Raman spectroscopy of natural gas hydrates recovered from Haakon Mosby Mud Volcano (Norwegian Margin) during the Vicking cruise (HERMES project, 2006). These samples exhibit as well a preponderance of structure I (sI) embedded in ice originating from frozen pore water and hydrate dissociation during recovery. The dissociation temperature (Td) of these hydrates is investigated by Raman spectroscopy at atmospheric pressure from 77 K to 260 K. Td shows to depend on the size of the hydrate particles. It is found to increase as the particle size increases. These results are consistent with previous data reported by Takeya et al. (2005). A "multi-layer" dissociation mechanism can be established for large hydrate particles thanks to the high spatial resolution of the micro-Raman technique. This effect will be discussed in the context of the stability of hydrates in natural environments. Takeya et al. (2005), Particle size effect of CH4 hydrate for self-preservation, Chem. Eng. Sci., 60, 1383-1387.
OS22A-07
Micromechanics of Hydrate-Bearing Sediments by Grain-Scale Simulations
Dissociation of gas-hydrates in marine sediments converts the solid hydrate structure into liquid water and gas. Weakening of the solid skeleton causes a reduction of the elastic moduli. The increased pore pressure reduces the effective stress. As a consequence, a point of fracturing or fluidizing of the sediment can be reached. If such events occur, seafloor subsidence and landslides can severely damage off-shore infrastructure. We seek to quantify the impact of hydrate dissociation on the strength of hydrate-bearing sediments. The sediment weakening can be attributed to the reduction of the elastic moduli as hydrates become liquid and gas. We calculate these moduli using numerical simulations of deformations of a random disordered pack of spherical grains. Our model is discrete, accounting for the interactions between individual grains by calculating the loads which develop at each contact. We use a quasi-static approach by presenting deformation as a sequence of equilibrium configurations of the grain pack. Each configuration is characterized by the minimum of the total mechanical work in the pack. We find this minimum numerically, using a modified conjugate-gradient algorithm. In natural sediments, the distribution of hydrates in the pore space is a result of geologic history of hydrate formation. It can be affected, among other factors, by the saturations of gas and water, by the pressure and temperature, and by the pore geometry. There is a big uncertainty regarding the actual hydrate distribution. Therefore, we consider three different models: (a) pore-filling hydrate grains, (b) small amounts of hydrates in the pore bodies, and (c) small amounts of hydrate forming cement bonds at the contacts. To model dissociation, we first reduce the volume of solid hydrate. Then, we change the effective stress by imposing a macroscopic strain at the boundary, using poroelastic constitutive relations. Our simulations reveal the microscopic mechanisms that lead to the nonlinear, path-dependent stress-strain relations which are inherent to granular media. The consequences of dissociation are different for different models of hydrate saturation. The weakening of hydrate- bearing sediments due to the dissociation is captured in our simulations as a reduction in macroscopic moduli.
OS22A-08
Formation of Massive Hydrate From the Accumulation of Free CH4 gas in Sediments
In many cases, hydrate samples recovered from seafloor sediments occur as veins or nodules, suggesting that there are strong geologic controls on where hydrate is likely to accumulate. Large-volume experiments have been conducted in Oak Ridge National Laboratory's high-pressure Seafloor Process Simulator examining massive hydrate accumulation from free methane gas bubbles within natural and synthetic sediments. Pressure and temperature data were supplemented by visual observation. Presence of methane hydrate was first recorded as a thin film at the gas/water interface of methane bubbles which had accumulated within sediment void spaces. Massive hydrate growth occurred when hydrate-covered bubbles aggregated. Observations of hydrate growth indicate that accumulation of gas bubbles within void spaces and at facies boundaries likely results in the formation of massive hydrate deposits in systems containing free gas. These experiments suggest that geologic controls, such as bubble pathways and accumulation points likely control the location and habit of massive hydrate deposits in systems containing free methane gas.