GASES FROM THE PRESSURE CORE SAMPLER

Sample Collection

Forty-two deployments (runs) of the PCS on Leg 164 successfully recovered a sediment core at high pressure (>3.45 MPa) at Sites 994, 995, 996, and 997 (Paull, Matsumoto, Wallace, et al., 1996; Dickens et al., Chap. 43 and Chap. 11,  this volume). Data collection for most of these cores generally proceeded as follows (Paull, Matsumoto, Wallace, et al., 1996; Dickens et al., Chap. 43, this volume). After core recovery, the PCS was placed in an ice bath, and a gas manifold system and sampling chamber were attached to an outlet port. Incremental volumes of gas were released from the PCS over time until the inside of the PCS was at atmospheric pressure. The PCS was removed from the ice bath and warmed to ambient temperature (~15ºC). Additional volumes of gas were then collected. Aliquots of gas were taken from gas volume increments for compositional analyses.

Due to a variety of technical and operational reasons, there was a lack of experimental consistency with PCS operations during Leg 164 (see Paull, Matsumoto, Wallace, et al. [1996] and Dickens et al., [Chap. 43 and Chap. 11, this volume], for details). Of particular importance for interpreting carbon isotope compositions of gas samples released from the PCS are (1) most cores were not given sufficient time to equilibrate after changes in pressure (and gas concentration), and (2) individual cores had different initial pressures, temperatures, gas concentrations, and sediment volumes.

The predominant hydrocarbon gas in all gas samples released from the PCS was CH₄ (C₁/C₂ > 1100; Paull, Matsumoto, Wallace, et al., 1996).

Carbon Isotope Analyses

Ninety gas samples from stepwise degassing of 18 PCS cores recovered at Sites 994, 995, 996, and 997 were analyzed for CH₄ ¹³C using a Finnigan MAT 252 mass spectrometer with a combustion interface at the University of North Carolina at Chapel Hill. The analytical procedure for the PCS samples was the same as that for CH₄ collected from gas voids by the vacutainer method and is described by Paull et al. (Chap. 7, this volume). Measured carbon isotope ratios for PCS gas samples are expressed in delta notation relative to the Peedee belemnite (PDB) standard (Table 1). Analytical precision for the CH₄ ¹³C values based on replicate gas analyses was typically ±0.2 or less (1).

Observations

Figure 1 shows downhole profiles of CH₄ ¹³C for gas samples that were taken from gas expansion voids in APC and XCB cores from Sites 994, 995, and 997 (Paull et al., Chap. 7, this volume). Methane ¹³C for gas voids increases from minimum values of -85 to -80 at ~40 mbsf to relatively constant values of -65 to -62 between 300 mbsf and the bottom of the holes (700-750 mbsf) at Sites 995 and 997. At depths greater than ~200 mbsf, CH₄ recovered from gas voids at Site 994 shows considerably more scatter in carbon isotopic composition than CH₄ recovered from gas voids at Sites 995 and 997. The cause of this scatter is unknown. It could be the result of greater natural variability of CH₄ ¹³C in Site 994 sediments compared with 995 and 997. Alternatively, it may be an artifact of sample handling and storage of the Site 994 gases.

Also shown in Figure 1 are CH₄ ¹³C values for PCS gas samples from Sites 994, 995, and 997. Although only a small number of PCS gas samples from Site 994 were analyzed for CH₄ ¹³C, values fall within the range defined by the data for gas voids. At Sites 995 and 997, multiple gas samples were taken from numerous PCS cores during depressurization of the PCS. For PCS gas samples from Site 995, CH₄ ¹³C values from an individual core vary by as much as 9 (Fig. 1). The range in CH₄ ¹³C values for successive gas samples from individual PCS cores at Site 995 typically vary from a value that is significantly lighter than the trend defined by gas void data to maximum values that lie on the trend. In contrast to Site 995, the range of CH₄ ¹³C values for PCS gas samples from Site 997 straddles the trend defined by the gas void data (Fig. 1).

Gas samples collected from the PCS have been numbered sequentially according to when the gas was released from the PCS during pressure loss (Dickens et al., Chap. 43, this volume). Shown in Figure 2 and Figure 3 are time series plots of CH₄ ¹³C values for gas samples taken from individual PCS cores at Sites 995 and 997, respectively. All cores from Site 995 are characterized by an increase in CH₄ ¹³C over time and pressure loss. The large (up to 9) spread in CH₄ ¹³C for gas from Cores 164-995A-27P, 36P, and 45P results from an initial degassing step in which ¹³C is much lighter than subsequent steps. Patterns for PCS cores from Site 997 are less systematic, although Cores 164-997A-25P and 21P also show gradual increases in the CH₄ ¹³C throughout the degassing sequence (Fig. 3). In contrast to all other PCS cores, the initial gas volume released from Core 164-997A-49P has a significantly heavier ¹³C (by ~2) than samples released at lower pressure. Data for CH₄ ¹³C in gas samples from two PCS cores taken at Site 996, a shallow hole located above the Blake Ridge Diapir (Paull, Matsumoto, Wallace, et al., 1996), show patterns generally similar to those observed in the Site 995 and 997 PCS data, with variations as much as 3 (Fig. 4).

During controlled degassing of a PCS core, the first (and sometimes second) degassing step usually releases a small volume of CH₄ poor gas at high pressure (Paull, Matsumoto, Wallace, et al., 1996; Dickens et al., Chap. 43 and Chap. 11, this volume). The composition of this gas is dominantly air that is trapped inside the PCS chamber during deployment, or helium that is used to purge the manifold prior to gas release. Subsequent degassing steps at lower pressure contain mostly CH₄ (Table 1). In Figure 5, Figure 6, and Figure 7, the CH₄ ¹³C value of each gas sample is plotted against the volume of CH₄ released in the degassing step from which the gas sample was taken for isotopic analysis. Also shown is the average value for CH₄ ¹³C weighted according to the volume of CH₄ released in each degassing step. The results show that anomalous CH₄ ¹³C values in PCS gas samples commonly correspond to degassing steps involving small volumes of CH₄ The results also show that the volumetrically weighted average of CH₄ ¹³C values for all gas samples collected from a PCS core are within 1 of CH₄ ¹³C values collected from gas voids of APC and XCB cores at similar depth (Fig. 1; Table 1).

In contrast, CO₂ ¹³C values of PCS gas samples are highly variable and mostly much lighter than CO₂ ¹³C of gas voids from APC and XCB cores recovered at comparable depths (Paull et al., Chap. 7, this volume). Given that CO₂ accounts for only a small fraction of the total gas (2% by volume), it is likely that CO₂ ¹³C values of PCS gas samples are biased to some extent by methane oxidation during sample handling. However, no correlation is observed between CH₄ ¹³C and CO₂ ¹³C in PCS gas samples, suggesting that oxidation effects do not change CH₄ ¹³C appreciably.