Previous   |   Next

doi:10.2204/iodp.proc.319.103.2010

Geochemistry

Gas geochemistry results

Scientific online mud gas monitoring was performed during several phases of drilling at Site C0009: during drilling of a 12¼ inch hole from 703 to 1510 m MSF (drilling Phase 2 operations, see C0009_T1.XLS in GEOCHEM in "Supplementary material"), core drilling of a 10⅝ inch hole from 1510 to 1594 m MSF (drilling Phase 3), and enlargement of the hole to 17 inches from 703 to 1569 m MSF (drilling Phase 8, see C0009_T2.XLS in GEOCHEM in "Supplementary material"). Core drilling (Phase 3) was conducted with low ROP in combination with the same mud pumping rate used during Phase 2, which resulted in low gas concentrations in the drilling mud. Low mud gas concentrations are a common feature during core drilling. Because of similar compositions and distributions of gas with depth, the discussion of results from both the initial drilling (Phase 2) and hole opening (Phase 8) phases are integrated here.

Gas composition

The principal formation gases extracted from the returning drilling mud were hydrocarbons, mainly methane (CH₄). As much as 14 vol% CH₄ was detected during Phase 2 and as much as 3 vol% was detected during Phase 8. During Phase 2, the gas chromatograph (GC) was not in operation; however, diagnostic ion currents measured by the mass spectrometer did not indicate the presence of higher hydrocarbons (C₂₊). During Phase 8, the GC detected traces of ethane (C₂H₆; up to 16 parts per million by volume [ppmv]) and propane (C₃H₈; up to 3 ppmv). The CH₄/C₂H₆ ratio was consistently >500 and generally ~1000.

Because of the constant high pH of the drilling mud (>10), CO₂ concentration was always low (i.e., significantly lower than air concentrations [~370 ppmv]). Any CO₂ entering the hole would have immediately reacted with drilling mud to form hydrogen carbonate according to the following reaction:

The consistently low H₂ content (<100 ppmv) positively correlates with the occurrence of CH₄. However, H₂ could also be produced by CH₄ fragmentation on the ion source of the mass spectrometer. The CH₄/H₂ ratio from postexpedition measurements on pure methane samples will confirm whether the H₂ was produced artificially. Helium concentration never exceeded 6 ppmv and was always derived from air, as shown by constant He/Ar ratios.

Gas distribution with depth

The depth distribution of gas from online measurements from 716 to 1454 m MSF during Phase 2 is discussed and compared with results from Phase 8 drilling from 716 to 1270 m MSF (Fig. F48). Mud gas measurements from below 1270 m MSF in Phase 8 were affected by very high methane background concentrations because the gas-loaded drilling mud was not circulated out after passing a gas-rich interval at 1270 m MSF; therefore, these data are not discussed here.

Down to ~800 m MSF, methane concentration was relatively low during Phase 2. From 800 to 930 m MSF, methane concentration increased, reaching a local maximum at ~880 m MSF, and decreased sharply below this depth. Methane concentration was also high between 930 and 1000 m MSF, peaking at ~970 m MSF. The highest methane concentration in the entire well (maximum value = 14 vol% at 1130 m MSF) was observed between 1050 and 1220 m MSF. A sharp peak was observed between 1250 and 1280 m MSF, with maximum values of ~12 vol% at ~1270 m MSF. Below 1280 m MSF, methane concentration decreased abruptly to ~4–5 vol%.

During Phase 8, better depth resolution was achieved by modifying the gas extraction setup (see "Geochemistry" in the "Methods" chapter) and in general lower gas concentrations were observed. Reduced gas concentration is due to higher mud weight that slowed gas liberation from cuttings into the drilling mud. The interval from ~800 to 930 m MSF, characterized by enhanced methane concentration during drilling Phase 2, did not exhibit high methane. This might be due to heterogeneous methane distribution at depth (i.e., locally enriched methane was liberated during Phase 2 only). Apart from this, both data sets exhibit similar downhole trends; however, most features in the gas profiles identified during Phase 2 are shifted ~10–20 m deeper in Phase 8. A possible explanation could be an error in conversion from lag time to lag depth during Phase 8.

The distribution of methane correlates with the presence of organic fragments (wood and lignite) in drill cuttings (Fig. F48). This observation implies that wood and coal are the primary source of hydrocarbons in the mud gas. Two degradation processes of organic material can account for hydrocarbon genesis: thermal degradation of organic matter (thermogenic gas) and microbial degradation of organics (biogenic gas). The CH₄/(C₂H₆ + C₃H₈) ratio provides an indication of hydrocarbon origin. Thermal degradation of organic matter generates hydrocarbons with a CH₄/(C₂H₆ + C₃H₈) ratio <50, whereas microbial gas has a ratio >100. During Phase 8, the CH₄/(C₂H₆ + C₃H₈) ratio was uniformly >500, and generally ~1000 (Fig. F49). The molecular composition of hydrocarbons measured during hole opening clearly suggests a microbial source of natural gas, although this ratio might be somewhat biased by the different solubility of hydrocarbons in drilling mud (ethane and propane have higher solubility in water-based liquids than methane). Moreover, estimated in situ temperatures of <50°C at the bottom of the hole are too low for significant hydrocarbon generation via thermal degradation of organics (see "Physical properties").

Postexpedition studies on the thermal maturity of organic matter in cuttings will illuminate whether or not thermal degradation can account for the hydrocarbon generation. Isotopic studies of methane (δ¹³C and H/D) and, if possible, ethane, from drilling mud gas and cuttings gas samples will provide conclusive evidence to distinguish between a biogenic or thermogenic origin. At this stage, the correlation between wood and coal fragments and peak methane concentration is interpreted to result from in situ microbial decomposition of organic matter.

Under the assumption of an isotropic linear elastic media, measured V_P/V_S ratio is related to Poisson's ratio. Figure F50 shows Poisson's ratio calculated from V_P/V_S logs and the measured methane concentration during Phase 2. The two data sets are clearly anticorrelated, implying that hydrocarbons are located in the pore space of the formation and these cause the lower Poisson's ratio. The V_P and V_S logs have a depth of investigation of a few meters. On the basis of the correlation between stratigraphic observations and mud gas data (Fig. F48), gas migration through permeable strata or fractures seems unlikely to play a significant role in the observed gas distribution. This interpretation is also consistent with the absence of gases migrating from greater depths (i.e., helium, heavy hydrocarbons) and with indications that hydrocarbons were produced from in situ microbial degradation of organic material.

Organic geochemistry

Inorganic carbon, total carbon, and total nitrogen

TOC, total nitrogen (TN), and CaCO₃ were determined from inorganic carbon, total carbon, and TN measurements in 111 cuttings samples from 1038 to 1588 m MSF (see "Geochemistry" in the "Methods" chapter for analytical procedures) and in 34 core samples from 1509.8 to 1591.5 m CSF. TOC and TN concentrations and the TOC to TN (TOC/TN) ratio are shown in Tables T10 and T11 and plotted together in Figure F51.

The calcium carbonate content in cuttings samples calculated from the inorganic carbon content ranges from 7.56 to 16.31 wt%, with an average of 13.28 wt%. CaCO₃ concentration and its distribution with depth are in good agreement with calcite data from XRD measurements (see "Lithology"). The meaning of the CaCO₃ data, however, is unclear, as they are apparently affected by interaction of the cuttings with the drilling mud. From 1038 to 1460 m MSF, the CaCO₃ profile exhibits values scattered between 11 and 16 wt%, with minimum values at 1103, 1292, 1333, and 1388 m MSF. Below 1460 m MSF, carbonate decreases from ~16 wt% at 1460 m MSF to 7.6 wt% at 1588 m MSF. In core samples, CaCO₃ concentration is noticeably lower than for cuttings in the same depth interval, with a range between ~1 and 9 wt%, and shows considerable scatter.

TOC content of cuttings samples ranges from 0.93 to 8.7 wt%, with an average of 2.51 wt%. The relatively high TOC values, particularly in the interval between 1080 and 1240 m MSF, with maximum values >8 wt% at 1088 and 1183 m MSF, probably derive from wood and coal, which are common in Subunit IIIB (see "Lithology") and also correlate with the occurrence of methane gas (see "Mud gas monitoring" in the "Methods" chapter). TOC content decreases to ~1 wt% at 1250 m MSF. At 1510 m MSF, TOC increases sharply and remains at ~2 wt% until the deepest sample (1592 m MSF). This increase, however, is probably artificial and may be caused by a different drilling mud used during core drilling. A similar feature is observed in CaCO₃ data at this depth. TOC concentration in core samples between ~1510 and 1600 m MSF is significantly reduced relative to cuttings samples, with a range of 0.25 to 0.9 wt%, and exhibits minimal scatter.

TOC and TN show a very similar distribution with depth, suggesting that both derive from the same source. TN content ranges from 0.051 to 0.128 wt%, with an average of 0.077 wt%. Core samples have similar but slightly higher average values relative to cuttings samples, but exhibit much greater scatter. The TOC/TN ratio ranges from 13.8 to 74.5, with an average of 29.8. Core sample ratios are reduced relative to cuttings samples, driven by the lower TOC values, and range from ~3 to 18. TOC/TN of marine organic matter typically ranges from ~4 to ~10, whereas terrestrial organic matter shows higher values. This distinction arises from the absence of cellulose in algae, its abundance in terrestrial plants, and the protein richness of algae (Meyers, 1997). Therefore, these results suggest that the source of organic matter in much of the borehole is terrestrial, which is in agreement with the wood and coal fragments observed in drill cuttings, particularly from Subunit IIIB. Deviations between core and cuttings samples in the 1510–1595 m MSF depth interval, specifically in CaCO₃ and TOC concentrations, suggest that cuttings are subject to contamination by drilling mud; however, trends in the cuttings data may reflect changes in the formation.

Interstitial water geochemistry

A total of seven whole-round sections from Cores 319-C0009A-3R to 9R were collected for interstitial water analyses. The sediments from the cored interval were highly lithified; it therefore took a special effort to extract a minimum volume of pore water for analysis. No pore water could be obtained from Core 319-C0009A-3R, which was squeezed using the conventional IODP procedure. From Cores 319-C0009A-4R to 9R (1537–1584 m CSF), pore water could be liberated only by squeezing for up to 54 h (see "Geochemistry" in the "Methods" chapter for the modified squeezing technique and Table T12 for squeezing times and composition of each sample). Water extraction from drilling mud for reference was unsuccessful, although several different filtering and extraction techniques (including centrifuging) were attempted. Coring during riser drilling and with continuous circulation of drilling mud during Expedition 319 was different than in most previous IODP coring operations; therefore, comparison between data sets from this and other expeditions should be treated with caution. The following interpretation is based on the assumption that any core material contaminated by drilling mud was removed prior to interstitial water extraction.

Squeezing was completed after obtaining 4 cm³ of pore water. The long extraction period may have caused degradation of species involved in microbial activities. NH₄⁺ and PO₄^3– concentrations were below detection limits for all samples, and sulfate concentration was very low (<0.4 mM). Assuming that no sulfate reduction took place during the sample treatment, the low sulfate concentration suggests that pore fluids were not diluted with seawater during sample recovery. The high chloride concentration (Fig. F52) probably indicates evaporation during the many hours of sample extraction. This conclusion is based on the large difference in these data (up to 1053 mM chloride, almost twice the seawater value) and those collected during IODP Expeditions 315 and 316 (Ashi et al., 2008; Kimura et al., 2008). During Expeditions 315 and 316, the general trend is toward chloride concentrations lower than seawater with depth (with the exception of IODP Site C0004). Given the probable evaporation of water before and/or during processing, we can only provide an estimate of the in situ chloride concentration (500 mM). If we correct the other data for this assumed evaporation, then much of the data are consistent with the deepest data from Site C0002, where pore waters were recovered from as deep as 1043 m CSF (Expedition 315 Scientists, 2009).

The six pore water samples cover a depth range of 47 m; formation pore waters from this narrow depth range should exhibit highly similar compositions (Figs. F52, F53, F54). Standard deviation of the six samples for the major ions in seawater is greatly reduced after taking into account the likelihood of evaporation during processing. With the evaporation correction (assuming 500 mM chloride), Na⁺ and sulfate values are consistent with deep pore water from Site C0002; Mg²⁺ values are lower and Ca²⁺ values are higher, probably reflecting continued in situ alteration of ash at the base of the drilled section; and Mn, Fe, V, Zn, Mo, and U values are generally similar. In contrast, these six pore water samples, corrected for evaporation, have much higher K⁺, Li⁺, Sr²⁺, Ba²⁺, Rb⁺, and Cs⁺ than at Site C0002 and slightly higher Br^– concentrations (Ashi et al., 2008). The higher Br^– values are likely linked to the presence of wood and microbial processes that degrade organic matter. The alkali metals are reactive during clay diagenesis and are affected by the temperature of reaction, generally being removed from solution at low temperatures and added to solution at high temperatures. Given greater concentrations for all of these elements relative to results from Expeditions 315 and 316, specifically K⁺, the deep samples from Site C0009 are generally consistent with warmer in situ temperatures. The high alkaline earth elements (Sr and Ba) are likely linked to carbonate remobilization (Sr) and low sulfate concentration (Ba). Overall, the deep samples from Site C0009 are consistent with warmer in situ temperatures or a more reactive sediment matrix than at Site C0002. Given a maximum temperature of 50°–60°C, these high alkaline and alkali values are likely diagenetic. Two elements, B and Si, have lower concentrations than those from Expeditions 315 and 316. The Si data are likely an artifact of sampling and precipitation of quartz during extraction. The solubility of quartz is ~120 mM, the observed concentration of several of the samples, at 30°C.

Top of page   |   Previous   |   Next