ESD05-048/ESD04-002

Goal
The project is bringing new laboratory measurements and evaluation techniques to bear on the difficult problems of characterization and gas recovery from methane hydrate deposits.

Performer
Lawrence Berkeley National Laboratory, Berkeley, CA 94720

Background
LBNL is performing laboratory tests to provide data to support the characterization and development of methane hydrate deposits. Two major work efforts are underway: hydrological measurements, and combined geomechanical/geophysical measurements.

Hydrologic Measurements
Relatively little research has been done to experimentally determine permeability (a measure of the resistance to fluid flow through the medium) in hydrate-bearing sediments (HBS) or relative permeability (the effect of another phase that interferes with flow) in a three phase—hydrate/water/gas—system where the hydrate habit (how the hydrate occupies space) is not well understood. The presence of hydrate in the porespace will hinder flow, and alter the gas/water capillary pressure, affecting the relative amounts of each phase present in the porespace. Numerical models account for relative permeability in a hydrate/water/gas system based on estimations and idealizations, however these estimations are not currently based on measurements.

LBNL is performing experimental tests to provide data that will allow for numerical inversion of relative permeability relationships. In performing these inversions, LBNL has found that unique relative permeability functions have not been obtainable without knowledge of how hydrate affects capillary pressure functions. In hydrate-bearing porous media, pore geometry, surface wettability, interfacial tension, and fluid phase saturations affect capillary pressure. LBNL has developed a technique to continuously measure gas/water capillary pressure in a hydrate-bearing porous media, and is making measurements to understand this important effect.

Geomechanical/Geophysical Measurements
It is important to understand the effect of hydrate on the geomechanical strength of hydrate-bearing sediments, because offshore oil and gas development requires placement of equipment on the seafloor. The strength of the sediments will change with the amount of hydrate present in the sediment, and the relationship between the sediment strength and the hydrate abundance is poorly understood. Numerical simulators have been constructed that can compute the effects of hydrate formation or dissociation on sediment strength, but these depend on measurements of these strength parameters.

LBNL is performing triaxial stress tests on samples containing methane hydrate in order to better understand and measure the geomechanical behavior of oceanic HBS as they undergo thermomechanical changes. Concurrent x-ray CT scanning to quantify sample uniformity and observe failure modes, and geophysical (acoustic) property measurements are being made in an attempt to relate the stress in the samples to field measureable properties.

Potential Impact
The primary benefits of this research are improved empirical relationships between experimental and theoretical relative permeability which will result in improvements to hydrologic and reservoir engineering applications, and to determine the envelope of HBS stability under conditions typical of those related to the construction and operation of offshore platforms.

Accomplishments
Work under this project is being conducted under three separate tasks that include:

• laboratory studies of relative permeability and capillary pressure relationships,
• development of a more rapid technique for estimating relative permeability and capillary pressure relationships based on laboratory data, and
• enhanced measurements of geophysical and geomechanical properties of hydrate-bearing sediments.

Geomechanical Cell

Hydrologic Measurements
LBNL has performed a number of tests to provide data to allow numerical inversion of relative permeability relationships. In these tests, sand at specified moisture contents is packed into a confining sleeve. A confining pressure is then applied and the sample is subject to a series of changes including freezing, thawing, hydrate formation, where gas permeability and water, ice, or hydrate saturation distribution is measured using CT at each step, and then water is flowed through the sample while monitoring with CT, and finally the sample is dissociated.

Tests for three initial water saturations for two sands and a sand/silt mixture have been completed. Numerical inversion is also underway. Initial inversions showed that it was not possible to provide unique relative permeability functions without appropriate capillary pressure functions, thus measurements of capillary pressure versus water and hydrate saturations are ongoing. In the tests performed so far, the hydrate was likely to be present as a grain-cementing agent. This is because the water in the moist sand samples is initially held in pendular structures between sand grains and in films surrounding the mineral grains. When the water converts to solid hydrate, it is thought to remain in these pendular regions, cementing the grains together instead of moving into the pore bodies. Thus, the hydrate formation technique used may produce hydrate that is representative of some but not all natural formations.

The technique developed by LBNL to measure gas/water capillary pressure in hydrate-bearing porous medium relies on pressurizing packed moist sand with methane and maintaining the system at the appropriate temperature allowing hydrate formation. Then the system is depressurized to the hydrate equilibrium point to preclude hydrate formation upon the addition of water, and the gas volume, pressure, and temperature are maintained constant. The sample is then carefully water saturated and then desaturated at a very slow rate with intervals of no flow to allow water saturation equilibration throughout the sample. The pressure difference between the gas phase and the water phase (the capillary pressure) is continuously measured using a high-pressure differential pressure transmitter calibrated over the expected range of differential pressure.

Geomechanical/Geophysical Measurements
LBNL has designed and constructed a device to enable performing triaxial stress tests on HBS. In addition to applying triaxial stress, the device is x-ray transparent allowing CT scanning to be performed during the tests to observe the distribution of hydrate, and failure modes. The device also has the ability to apply and detect ultrasonic compressional and shear waves.

Current Status

Hydrologic Measurements
LBNL is completing their current set of relative permeability measurements and related numerical inversions, as well as performing repeated tests to validate previous measurements. LBNL will perform waterflood tests on hydrate-bearing samples and use inverse modeling to estimate relative permeability from pressure, temperature, differential pressure, and CT data. Capillary pressure will be directly measured as a function of water and hydrate saturation.

To date, three capillary pressure tests have been performed using the LBNL technique. The measurement of capillary pressure and relative permeability of hydrate-bearing sediments has proved to be experimentally and computationally challenging, with each experiment and numerical inversion requiring significant effort. Ultimately, to simulate gas production from specific hydrate-bearing reservoirs, reservoir-specific measurements using native samples will be needed and quicker measurement techniques with less specialized equipment will be required for this purpose. LBNL is now investigating a method similar to that of Jennings et al. (1988) in which a more rapid transient test combined with numerical modeling may not only speed up the capillary pressure measurements, but also simultaneously provide relative permeability data.

Geomechanical Measurements
For the geomechanical behavior studies, LBNL has designed and manufactured a test vessel to enable performing the needed tests. Current efforts include vessel shake-down, preparing an engineering note documenting the vessel design and manufacture, and pressure testing, prior to testing HBS. Triaxial tests with concurrent compressional and shear-wave velocities will be conducted for a selected number of material parameters, including initial sediment porosity, hydrate saturation, pore pressure, and temperature. Initially fine-grained sand will be used but ultimately several sediment types, including a mixture of fine sand and silica flour or clay (kaolinite is considered), and silica flour or clay without fine sand will be evaluated. The impact of long-term loading on the sample strength by means of a creep test will also be investigated.

In addition to the triaxial tests with CT scanning and geophysical property measurements, LBNL will also measure acoustic properties of HBS undergoing thermomechanical changes using a new resonant bar apparatus. Conventionally, acoustic properties of most laboratory samples are measured using ultrasonic waves at frequencies of 100 kHz to 1 MHz. However, for testing sediments containing relatively low concentrations of hydrate that was formed in situ, measured acoustic properties may not be accurate, owing to wave scattering caused by wavelength-scale heterogeneities. Such heterogeneities can also develop as a thin layer along the core surface during dissociation of hydrate as a result of temperature and pressure changes in a sample.

To avoid the negative effects that could occur with ultrasonic measurements, LBNL will use a new, low-frequency resonant bar device that is capable of determining the acoustic properties from a small core sample. For typical unconsolidated and weakly cemented sediments, this device can determine the Young’s modulus and shear modulus of an HBS core at frequencies near 1 kHz – a frequency more akin to field frequencies. From the measured elastic moduli, more accurate acoustic (seismic) wave velocities and their attenuation can be determined. Such measurements can be conducted continuously as the hydrate in an HBS core forms and dissociates under controlled conditions.

Project Start: July 1, 2003
Project End: September 30, 2009

Project Cost Information:
All DOE Contribution

Under ESD04-002
FY04 - $141,000
FY05 - $20,000

Under ESD05-048
FY06- $330,000
FY07 - $320,000
FY08 - $330,000
FY09 - $240,000

Total Funding - $1,380,000

Contact Information:
NETL – Richard Baker (Richard.Baker@netl.doe.gov or 304-285-4714)
LBNL –Timothy J. Kneafsey (tjkneafsev@lbl.gov or 510-486-4414)

Additional Information
In addition to the information provided here, a full listing of project related publications and presentations as well as a listing of funded students can be found in the Methane Hydrate Program Bibliography [PDF].

Additional LBNL hydrate-related publications can also be found on the LBNL Gas Hydrate Publications webpage.

Quarterly Progress Report – October - December 2007 [PDF-398KB]

2008 ICGH Paper - Fluid Flow Through Heterogeneous Methane Hydrate-Bearing Sand: Observations Using X-Ray CT Scanning [PDF] - July, 2008

Quarterly Progress Report – January - March 2008 [PDF-498KB]

2008 Hydrate Peer Review [PDF-4.49MB]

Publications
Kneafsey, T.J., Seol, Y., Gupta, A., and Tomutsa, L., 2008, Permeability of Laboratory-Formed Methane-Hydrate-Bearing Sand, paper presented at the 2008 Offshore Technology Conference held in Houston, Texas, U.S.A., 5-8 May 2008, (OTC 19536-PP).

Kneafsey, T.J., Tomutsa, L., Moridis, G.J., Seol, Y., Freifeld, B.M., Taylor, C.E., and Gupta, A., 2007, Methane hydrate formation and dissociation in a partially saturated core-scale sand sample, Journal of Petroleum Science and Engineering 56 (1-3), 108-126.

Waite, W.F., Kneafsey, T.J., Winters, W.J. and Mason, D.H., 2008, Physical property changes in hydrate-bearing sediment due to depressurization and subsequent repressurization, Journal of Geophysical Research, 113, 12 pages, B07102, doi:10.1029/2007JB005351,