The pressure-temperature (P-T) stability conditions for the formation of methane hydrate have been determined on synthetic methane hydrate (e.g., Deaton and Frost, 1946; Yokoi et al., 1993); however, those of natural gas hydrate formed in marine sediments have not been measured, because natural gas hydrate samples have not been available for such an experiment. Considering that the environment of formation of natural gas hydrate is quite different from that of laboratory experiments, the P-T conditions for the three-phase equilibrium of natural gas hydrate may not be the same as that determined on synthetic gas hydrate. Leg 164 scientists observed that the depth of the BSR was significantly shallower than that of the theoretically determined base of methane hydrate stability (Paull, Matsumoto, Wallace, et al., 1996). Therefore, it is crucial to determine equilibrium dissociation conditions of the Blake Ridge gas hydrate to investigate the discrepancy between the BSR and the base of gas hydrate stability (BGHS).
One gas hydrate sample (Sample 164-997A-42X-3, 25-35 cm) was observed to contain a gray part and black part. The former was expected to contain gas and hydrate, and the latter one seemed to be mainly the host sediment. It is crucial for this experiment to develop the method to detect a slight increase in pressure caused by the dissociation of natural gas hydrate. Because the equipment used for this experiment was originally designed for synthesis of a large gas hydrate, the volume of the vessel was too big (630 mL) to detect the dissociation of a small amount of gas hydrate. Therefore, the volume of the vessel was reduced to 160 mL by putting stainless steel blocks inside (Fig. 15). Consequently, the accuracy of temperature measurement was approximately ±0.5 K. The equilibrium conditions were obtained only by wrapping the gas hydrate sample in a vinyl bag.
A 6.5 g of the frozen sample that consists of mixture of both a black part and a grey part was placed in a thin vinyl bag, and the bag was tightly sealed after the air in the bag was evacuated. The sample was placed in a pressure vessel and the remaining volume of the chamber was filled with nitrogen and water.
The temperature of the vessel was increased from 268 to 275 K by circulating coolant and the pressure was increased up to 11 MPa. This condition is well within the methane hydrate stability. Equilibrium dissociation was initiated by careful pressure release by 0.05 MPa steps at a temperature of 274.5- 275.0 K (Fig. 16). Gas hydrate did not dissociate upon step-wise pressure release while the pressure was maintained within the stability condition. However, when the pressure decreased down to 3.25 MPa at a temperature of 274.5-274.8 K, a slight pressure recovery was recognized in a few minutes. This pressure increase is explained to reflect partial dissociation of gas hydrate. The pressure finally attained constant value of 3.27 MPa in 15 min. These P-T values (3.27 MPa and 274.7K) are considered to be a close approximation of the three-phase equilibrium conditions of the Blake Ridge gas hydrate.
As shown in Figure 17, the three-phase equilibrium condition of the Blake Ridge hydrate is almost the same as the equilibrium condition of pure methane hydrate in pure water, with an error of about 0.5 K. As documented by the discrepancy between BSR and BGHS, the apparent phase boundary condition of methane hydrate in the Blake Ridge sediment is a few degrees lower than that of the pure methane hydrate in pure water or in seawater; however, the three-phase equilibrium condition as determined on recovered solid methane hydrate samples does not greatly differ from the experimental data (e.g., Deaton and Frost, 1946).