GAS FRACTIONATION EXPERIMENTS
IN SIMPLE ANALOG SYSTEMS

Experimental

The range in CH₄ ¹³C values for PCS gas samples from individual cores (as much as 9) is significantly greater than the analytical uncertainty (2 = ±0.4). A possible explanation for these large variations is that kinetic fractionation of CH₄ ¹³C is occurring during degassing of the PCS (as well as in uncontrolled degassing of APC and XCB cores). The observation that CH₄ ¹³C values vary depending on the volume of CH₄ released indeed suggests that carbon isotope fractionations result from kinetic (nonequilibrium) effects during degassing.

To isolate causes of the observed fractionation of CH₄ carbon isotopes, we performed shore-based degassing experiments in simple systems analogous to the shipboard degassing experiments with the PCS. The first set of experiments involved releasing methane from a pressure vessel into a bubbling chamber identical to that used during Leg 164 PCS experiments (Fig. 8; see also Paull, Matsumoto, Wallace, et al., 1996). The second set of experiments involved degassing of CH₄ gas-saturated water from the pressure vessel (Fig. 9).

Fractionation During Degassing of Pressurized Methane

The first set of experiments was designed to test whether significant fractionation of carbon isotopes could occur during high pressure release of CH₄. Four series of degassing experiments were performed, one each at 500, 1000, 1500, and 2000 psi. For each experiment, the pressure vessel did not contain water. Plastic spacers were used inside the pressure vessel to reduce the total gas volume. This made the void space inside the vessel comparable to that in the second set of experiments in which water was present in the vessel. The pressure vessel was pressurized using CH₄ with 2% CO₂. After the final pressure was reached, the intake valve was closed, and the pressure vessel was allowed to equilibrate for ~5 min. Both the pressure and temperature in the vessel stabilized in less than 30 s after closing the intake valve. After recording the pressure at the pressure transducer (located in the sampling tube), a valve was closed to isolate gas at pressure in a short section of tubing to which the pressure transducer was attached (Fig. 8). The outlet valve was then opened, allowing the gas to escape into a bubbling chamber consisting of an overturned graduated cylinder in NaCl-saturated water. The volume of gas released was then measured to the nearest 10 mL, and a sample of gas was taken for isotopic analysis using a plastic syringe. The outlet valve was then closed, and the valve connecting the pressure vessel to the sampling tube and pressure transducer was again opened. After waiting 2 min, this valve was again closed and the pressure at the pressure transducer was recorded. The outlet valve then was opened releasing gas into the bubbling chamber. This procedure was repeated a third time. For the experiments conducted at 1500 and 2000 psi, in any single degassing step, all of the gas in the sampling tube could not be released at once, as the volume would have exceeded the 1 L capacity of the overturned graduated cylinder in the bubbling chamber.

Fractionation During Degassing of Pressurized Methane and Gas-Saturated Water

The second set of experiments was designed to assess carbon isotope fractionation between CH₄ gas and CH₄ gas-saturated water during rapid degassing. All experiments were performed at ~1000 psi. For each experiment, the pressure vessel was pressurized using the same tank of CH₄ (with 2% CO₂) used in the first series of experiments. The inlet tube extended to near the bottom of the pressure vessel, which was about 90% filled with water, so that the methane bubbled through the water to facilitate equilibrium. After a pressure of ~1050 psi was reached, the vessel was allowed to equilibrate for 10-12 hr. The apparatus was designed so that a small aliquot of the headspace gas could be isolated in the sampling tube and, almost simultaneously, CH₄-saturated water could be released back out of the inlet port directly into syringes (Fig. 9). When this occurred, the water effervesced vigorously, and the final amounts of water and evolved gas in the syringes could be measured. After the water samples were taken, the outlet valve was opened and the sample of headspace gas was introduced into the bubbling chamber, from which the gas was sampled with a syringe (Fig. 9). A second sample of headspace gas was isolated in the sampling tube following the large decrease in pressure in the vessel caused by release of CH₄-saturated water. This gas was then released to the bubble chamber for sampling.

Three separate experiments were performed at 1000 psi. Between each, the pressure vessel was entirely vented to atmospheric pressure, and nitrogen bubbled through the water before the vessel was resealed and pressurized again. In the second and third experiments, a sample of CH₄-saturated water at lower pressure was also taken after higher pressure samples were taken according to the procedures described above.

Experimental Results

Measurements of CH₄ ¹³C for gas samples derived from the degassing experiments were made using the same instrument and procedures as already described for analysis of the PCS gas samples. Results of the experiments are presented in Table 2 and Table 3. Methane ¹³C values of gases from the first set of experiments on pressurized methane are shown in Figure 10. The results show that during a single degassing experiment, ¹³C values may vary by as much as 1, but no systematic patterns are observed. Because no water was present in the pressure vessel during the experiments, the only processes that could be leading to these fractionations are (1) rapid flow of gas through tubing to the bubble chamber, (2) bubbling of methane through NaCl-saturated water in the bubbling chamber, and (3) sample handling as gas is removed from the bubbling chamber using a syringe and then transferred to an inverted gas bottle submerged in water.

Results of the CH₄-saturated water degassing experiments are shown in Figure 11. Three separate experiments were run at ~1000 psi. Variability of methane ¹³C values in each of the experiments exceeds the 2 analytical uncertainty of ± 0.4. Although there is considerable scatter in the data, there is no evidence of an equilibrium carbon isotope fractionation between CH₄ gas and CH₄ dissolved in water, because the high-pressure headspace gas and dissolved CH₄ (water + gas in Fig. 11) are indistinguishable within analytical uncertainty in the second and third experiment (Figs. 11B, 11C). In the first and third experiments (Figs. 11A, 11C), isotope fractionation as great as 1.5 is observed for the headspace gas samples collected after passing through the bubbling chamber. In contrast, aliquots of CH₄ exsolved from CH₄-saturated water (water + gas in Fig. 11) show a more restricted range of ¹³C variation.

Release of CH₄-saturated water caused a large pressure drop inside the pressure vessel. Several minutes after this pressure drop (to ~200 psi) occurred, a new sample of headspace gas, which undoubtedly contained some CH₄ just released from solution, was isolated in the sampling tube and then released to the bubbling chamber. In the second and third experimental runs (Figs. 11B, 11C), an additional sample of CH₄-saturated water at ~200 psi pressure was released out the top of the inlet port. These samples of headspace gas and dissolved CH₄ at ~200 psi were extracted before the water and gas had a chance to fully reequilibrate after the rapid pressure drop from ~1000 psi. The data for this lower pressure, unequilibrated system show a consistent ¹³C fractionation of 1.5-2 between the headspace gas and dissolved CH₄, with headspace gas having lighter ¹³C (Figs. 11B, 11C).

We conclude that the main factors contributing to carbon isotope fractionation during the degassing experiments are the release of gas from high pressure and the lack of equilibrium between water and gas that can occur after rapid pressure drops within the pressure vessel.