MEASUREMENT OF WATER PERMEABILITY UNDER THE PRESENCE OF METHANE HYDRATE

Hideki Minagawa¹, Yukiko Hirakawa¹, Mitsutaka Sato¹, Ryo Ohmura¹, Yasushi Kamata¹, Satoshi Takeya¹, Jiro Nagao¹, Takao Ebinuma¹, Hideo Narita¹, Yoshihiro Masuda^2
1 National Institute of Advanced Industrial Science and Technology (AIST),
2-17-2-1 Tsukisamu-Higashi, Toyohira-ku, Sapporo, 062-8517, Japan;
² University of Tokyo, School of Engineering, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan;

Introduction

Methane hydrates existing in the sea floor sediment and permafrost is expected to be an unconventional methane resource. The methods of recovering methane gas from the hydrate, such as (1) depressurization method, (2) thermal stimulation method and (3) inhibitor injection method, have been proposed. In any methods, permeability to gas/water is an important property of the hydrate sediment to determine recovery efficiency of methane gas. In this work, we studied the relations between water permeability and hydrate saturation by using different types of hydrate sediments. We points out water permeability under the presence of hydrate would be seriously dependent on hydrate distributions in pores related to hydrate growth process.

Experimental Apparatus

The experimental apparatus consists of a temperature-controlled core holder, water and gas inlet-outlet lines with pressure gauges, and data-recording system. The core holder is combined with hydrostatic tri-axial pressurized system. The sample of sediment is set in a cylindrical rubber sleeve and pressurized from both axial and radial direction with a pressure of 20 MPa at maximum.

The temperature of the core holder is controlled from 0 to 50°C by circulation of the coolant. Temperatures of the sediment sample are monitored at three different positions by t-type thermocouples with the precision of 0.4°C. Methane gas can be injected into the sample through the gas inlet line at a flow rate controlled by a mass flow controller. The purity of methane gas used in the experiment is above 99.99%.

In case of water permeability measurement, distilled and de-ionized water with electric resistance of approximately 1.67x10⁵ m was injected from the bottom end of the core by a precisely-control injection pump. The flow rate of water can be changed from 0.02 to 20 cm³/min. The pressures of the inlet and the outlet of the core and its pressure difference were measured by pressure gauges. The confining pressure supplied to the rubber sleeve was controlled by another pump. Pore pressure was controlled by a back pressure regulator. All data were monitored and simultaneously recorded by a data-logging system.

Two Types of Methane Hydrate Sediment

Toyoura sand was used for making the sediment sample. Two methods of making methane hydrate sediment (MH sediment) were applied in order to elucidate the relations between permeability and hydrate saturation. One is the conventional method for permeability measurement. After the sand-packed core in the rubber sleeve was saturated with water under supersonic vibration, methane gas was flowed through the core. Gas displaced water from the core until connate water saturation. The connate water saturation was controlled by changing flow rate and pressure of methane gas. Next, by increase of methane gas pressure and decrease of core temperature the connate water was converted into hydrate and MH sediment was made in the rubber sleeve (Type-1 sediment). The feature of Type-1 sediment is that hydrate starts to grow in pores at relatively-high gas saturations.

Another was the method that had been developed by Chuvilin and Ebinuma et al.(1) It is called gas diffusion method. At first the mixture of water and sand grains was put into the rubber sleeve and stamped. After setting the rubber sleeve into the core holder, remaining air was displaced by methane gas. Later methane gas was injected into the core at a high pressure. Methane hydrate was formed in the pores by decreasing temperature and MH sediment was made in the rubber sleeve (Type-2 sediment). The feature of Type-2 sediment is that hydrate starts to grow in pores at relatively-low gas saturations.

Results and Discussions

After making the methane hydrate sediment, water was injected into the sediment. The gas remaining in the pores was gradually displaced by water, and finally water passed through the sediment. Though the pore pressure was kept above the hydrate equilibrium pressure during successive water injection, the amount of methane gas produced increased with time and the measured pressure difference decreased with time. As a result, the calculated water permeability increased gradually with time. These results would be caused by two phenomena. One is the displacement of residual gas in pores by water. Another is the decomposition of hydrate in pores above the equilibrium pressure. From material-balance calculation considering the gas volume produced and phase saturations in the sediment, the residual gas saturation during permeability measurement was estimated to be less than 5%.

Now we consider the effective water permeability calculated at water-breakthrough time as a representative permeability and name it apparent water permeability. The data set of apparent water permeability vs. hydrate saturation could be obtained from the experiments using Type-1 and Type-2 sediments. When the apparent water permeability was plotted against hydrate saturation (Sh), we found that the permeability of Type-1 sediment was 20 times bigger than that of Type-2 sediment at maximum. Surprisingly, the water permeability at the same hydrate saturation was different between Type-1 and Type-2 sediments. This result indicates that water permeability would be dependent on hydrate distribution in pores related to hydrate growth process.

From the logarithmic plot of the data of apparent water permeability vs. (1-Sh), we could find the tendency of decreasing permeability with increasing hydrate saturation. When we use Equation (1) for expressing the tendency of this decrease and fit the experimental data to this curve, N equals to 2.6 for Type-1 sediment, whilst N equals to 9.8 for Type-2 sediment.

K = K₀(1-Sh)^N                                         (1)

K₀: apparent permeability at Sh= 0
N: constant.

According to Kleinberg⁽²⁾, the tendency of gentle decreasing permeability appeared in Type-1 sediment can be explained by the Kozeny grain model where hydrate assumes to coat grain surface. But, the tendency of steep decreasing permeability appeared in Type-2 sediment is difficult to be explained by conventional physical models of porous media and then we need an empirical approach similar to that of Masuda et al.⁽³⁾ Or we need a more sophisticated model to express partial plugging of pore channels. It is clear that the different tendency on permeability vs. hydrate saturation was caused by the different methods of making hydrate sediments. Hence, we think that in-situ water permeability of sediments under the presence of hydrate is dependent on distribution of hydrate in pores related to hydrate growth.

Conclusion

The model samples of MH sediment were made by different methods of forming hydrate and the data set of apparent water permeability vs. hydrate saturation was obtained. As expected water permeability decreased with increasing hydrate saturation, but we found that the tendency of decreasing permeability differed with types of MH sediment. This result indicates that water permeability would be seriously dependent on hydrate distribution in pores related to hydrate growth process.

Acknowledgements

This work was financially supported by the Research Consortium for Methane Hydrate Resources in Japan (MH21 Research Consortium) on the National Methane Hydrate exploitation program planned by Ministry of Economy Trade and Industry (METI).

References

(1) Chuvilin, E.M., T. Ebinuma, Y. Kamata, T. Uchida, S. Takeya, J. Nagao, and H. Narita, Effects of temperature cycling on the phase transition of water in gas-saturated sediments, Can. J. Phys., 81 (1-2), 343-350, 2003.

(2) Kleinberg, R.L., C. Flaum, D.D. Griffin, Methane hydrate growth habit in porous media and its relationship to hydraulic permeability, deposit accumulation, and submarine slope stability, Journal of Geophysical Research B, in printing.

(3) Masuda, Y., S. Naganawa, S. Ando and K. Sato, Numerical calculation of gas production performance from reservoirs containing natural gas hydrates, SPE 382901, 1997.