IODP Proceedings    Volume contents     Search

doi:10.2204/iodp.proc.337.102.2013

Inorganic geochemistry

Interstitial water collection

During Expedition 337, WRC samples were squeezed for interstitial water analyses. One or two WRC sections measuring 30–75 cm in length were sampled per core. The particular locations for interstitial water WRC sampling were selected after close inspection of the X-ray CT scan images of the cored sections. Core intervals characterized by high fracture density, disturbed lithologies, or cleat structures were disregarded, as such sediment material is more susceptible to contamination from drilling fluids. After examination for internal structures with X-ray CT scan, selected WRC samples were taken to the sample preparation laboratory and placed in nitrogen-flushed glove bags at room temperature (~23°C).

A portion of the sediment was immediately extruded from the core liner. Once extruded, ~1 cm on each end of the WRC was cut off, as these sediments had probably undergone significant degassing during WRC sample processing. Furthermore, ~0.5–1.0 cm of the exterior portion of WRCs was peeled off to remove portions of the section that were potentially contaminated by riser drilling mud or materials generated during liner splitting contamination. Peeling included internal portions of the cores adjacent to fractures. In the case of LDC, the time span between core recovery and arrival of interstitial water WRC samples in the Sample Preparation Laboratory amounted to as long as half a day because of the longer processing time involved with this type of coring method at the rig floor and core cutting area. This processing delay greatly enhanced the risk for sample contamination. Therefore, more sediment was peeled off from the LDC WRC sample surface and ends (i.e., up to 2–5 cm). Additionally, the integrity of the inner part of the sample was visually inspected and, where necessary, portions showing infiltrated drilling mud were removed. The peeled-off residue from WRC samples was saved for shipboard smear slides and core descriptions.

The cleaned sediment was placed into two 9 cm in diameter Manheim-type titanium squeezers (Manheim, 1966) on top of a stack of two pieces of filter paper rinsed with ultrapure deionized water (18.2 MΩ Milli-Q) and two titanium screens (40 mesh). Fluids released by squeezing were then passed through a sterile 0.22 µm disk filter connected to an acid-washed 20 mL plastic syringe. This process was repeated with the remaining sediment in each section. In cases of low yield, squeeze cakes were combined in one squeezer for further interstitial water recovery. Interstitial water was distributed into aliquots for both shipboard and shore-based geochemical analyses. Squeeze cakes were divided, placed in bags, and stored as requested for shore-based research. In the time between sampling and squeezing, the remaining interstitial water WRC samples were stored in a 4°C refrigerator in nitrogen-flushed, doubly sealed oxygen-impermeable ESCAL bags without applying a vacuum.

Because cores were not recovered during the drilling through Unit I (647–1256.5 m MSF), it was decided to sample and squeeze interstitial water from drilling cuttings across this interval. The unrinsed cuttings were placed on a tray and then transferred with a spatula to a Manheim squeezer. No further attempt was made to remove the drilling fluid that had soaked into the sediment cuttings. This procedure was continued at selected depths of Units II and III, in order to compare cuttings water with interstitial water from cores at the same corresponding depth.

Interstitial water analysis

The only other IODP riser drilling expedition to date encountered problems with limited interstitial water yields at elevated depths (Expedition 319 Scientists, 2010b). In anticipation of potentially the same problems during Expedition 337, several standard analytical shipboard procedures were modified. Interstitial fluid splits were not set aside for chlorinity titration; instead, a very small amount of interstitial water was diluted and used for chloride determination by ion chromatography (IC). Second, the IC analytical run for sulfate was also used to analyze for nitrate. Finally, in cases where the total interstitial water yield from the 60 cm WRC sample was <5 mL, the pH measurement was disregarded and alkalinity was determined by diluting a known volume of interstitial water dilution with 0.7 M KCl. In anticipation of potentially long squeezing times (Expedition 319 Scientists, 2010b) and the possibility that some of the chemical and isotopic components of the interstitial water could be compromised by sediment degassing—in particular, the dissolved inorganic carbon (DIC) prepared for postcruise analysis and shipboard pH/alkalinity determinations—small aliquots of interstitial water for these time-sensitive analyses were prepared first, before the rest of the sediment was squeezed.

Interstitial water samples were analyzed using the standard IODP methods of shipboard geochemical measurements described previously (e.g., Expedition 319 Scientists, 2010a; Expedition 333 Scientists, 2012). A refractometer (Atago RX-5000) was used to determine salinity based on the refractive index. Immediately following interstitial water extraction, alkalinity and pmH were determined with a pH electrode and an autotitrator (Metrohm) with a 3 mL aliquot of interstitial water. The concentration of protons is reported as pmH rather than pH because the calibration was carried out with buffers in an artificial seawater solution of KCl, which approximates the ionic strength of the pore water samples themselves (Dickson, 1984). In cases where the total yield of interstitial water was <5 mL, we carried out a 1:10 dilution using 0.3 mL of interstitial water plus 2.7 mL of 0.7 M KCl solution in order to provide the minimum operational volume required by the alkalinity autotitrator. The 0.7 M KCl solution was used as a diluent both with the 0.3 mL sample and with the calibration standards, serving to maintain the ionic strength at values similar to seawater.

Sulfate, bromide, and nitrate concentrations were analyzed from an unacidified sample diluted with Milli-Q water to 1:100, whereas chloride was determined from a 1:1,000 dilution of the same sample using an IC (Dionex ICS-1500 ion chromatograph) with an anion column. A seawater standard (International Association for the Physical Sciences of the Oceans [IAPSO], Kanto Chemical Co., Inc., 07197-96) was analyzed for QC and to determine accuracy. The residue of the unacidified sample was analyzed with an UV-visible spectrophotometer (Shimadzu UV-2550) to determine dissolved ammonium and phosphate (Gieskes et al., 1991). A cation column was used on the Dionex IC to analyze major cations (Na, K, Mg, and Ca) on acidified (4% of 6 M HCl, trace metal grade) interstitial water samples diluted 1:200–1:800, depending on the concentration of the analyzed compound. The IAPSO standard was used for calibration, and dilutions of IAPSO were used for internal QC of the instrument. Inductively coupled plasma–atomic emission spectroscopy (Horiba Jobin Yvon Ultima2) was used to determine interstitial water minor element concentrations (B, Ba, Fe, Li, Mn, Si, and Sr). Samples were acidified and diluted 1:20 with Milli-Q water. Ultrapure primary standards (SPC Science PlasmaCAL) were prepared in a matrix solution of sulfate-free artificial seawater to fit the sample matrix, and yttrium was used as an internal standard (Shipboard Scientific Party, 2001).

Formation fluid analysis

Formation fluid samples were taken using an IFA coupled with a MDT, collected in a SPMC under in situ conditions, and then brought to the surface and transferred to SSBs for analysis (Dong et al., 2007; Mullins, 2008). Downhole Fluid Analysis (DFA) provides real-time data acquisition of resistivity and color data to identify target strata and to make a preliminary assessment of the potential to collect a water sample at depth (see “Downhole logging”). Once hydraulic communication was established with the sediment on the walls of the drill hole, 250 mL of formation water was pumped into the SPMC, where it was then pressurized and retrieved shipboard. This procedure was carried out at six selected depth horizons, providing a unique suite of samples to be divided between microbiologists, organic geochemists, and inorganic geochemists. After recovery, the sample fluids were transferred from the SPMC into a preevacuated glass SSB (see “Organic geochemistry”). After the gas extraction procedures were completed, the samples were placed in a glove box, where they were filtered and distributed into aliquots for both shipboard and shore-based geochemical analyses in much the same way as the interstitial water samples were handled (see above).

Analysis of drilling mud to correct for contamination

Sediment intervals that presented sulfate in the interstitial water bear the potential earmarks of contamination by seawater used in riser drilling mud and borehole fluids. As has been mentioned, an attempt to minimize this problem was carried out by peeling off the outer portion of the WRC sample prior to squeezing. However, sediment cores that have been fractured by the drilling process with the RCB and LDC, as well as those which present natural foliations or cleat structures, are susceptible to drilling fluid contamination that cannot be eliminated by simply removing the outer part of the WRC sample.

In order to assess and correct for possible contamination, mud from the circulation tanks was sampled twice before drilling, five times during the course of drilling, and once after drilling. Prior to analysis, dilution of drilling mud was necessary in order to facilitate separation of the water-soluble components of drilling mud liquid (DML) from the colloidal components that include bentonite and a polymer viscosifier. A volume of 1 mL of drilling mud was diluted to a total volume of 10 mL with Milli-Q water, homogenized in an ultrasonic bath for 1 h, and then centrifuged for 1 h.

A small correction had to be made for the volume of drilling mud that was diluted, centrifuged, and analyzed because the volume of separated DML used to determine mud concentrations is not identical to the volume of drilling mud that was initially pipetted and diluted. For each mud sample, a separate ~20 mL aliquot of drilling mud was weighed and dried to find the proportion of salts and solids in a given volume of mud. Large-volume pipetting of the drilling mud proved inaccurate because of the highly viscous properties of the mud additives. In order to determine the volume of mud (VDM), we relied on the density from the daily shipboard mud logging sheet form (ρDM) and the mass of the drilling mud before drying.

The major water-soluble constituents of drilling mud used in operations are generally 50 g KCl, 170 g NaCl, and 5 g soda ash (Na2CO3) dissolved in 1 L of seawater (assumed to contain 40 g/L salts) adjusted to a pH of 9.0–10.5 with KOH solution. These are combined with a number of commercial products, which act as colloidal lubricants and viscosifiers, including 5 g TelGel (bentonite), 8 g TelPolymerL (cellulose derivative), 10 g TelPolymerDX (starch derivative), 2.5 g Xanvis (xanthan gum derivative), 50 g CleanLubeL (lubricant and gas-hydrate inhibitor), 100 g RevDust (pseudocuttings), and 1 g TelniteGXL (antiseptic agent). Although actual concentrations may vary, we assume that the relative amounts of salts and nonwater-soluble components are roughly constant, such that the mass ratio of soluble salts to the net sum of soluble and insoluble and colloidal components (Rwss) is 0.600. The bulk dry density of water-soluble salts (ρwss) based on their relative weight abundance and respective densities is assumed to remain constant at 2.13 g/cm3.

Using the initial mass of the wet mud and the mass of the residue (soluble and insoluble components) after freeze-drying, we can calculate the ratio of VDM to the volume of DML (VDML) as

VDM/VDML = VDM/[(mH2OH2O) + Rwss(msrwss)], (24)

where

  • VDM = volume of drilling mud dried (calculated from the mass of the drilling mud and the density of the mud),

  • mH2O = mass of water lost through freeze-drying (g),

  • ρH2O = density of pure water (1.000 g/cm3),

  • msr = mass of solid residue (g),

  • Rwss = ratio of water-soluble solids to total solids in the mud (0.600 g/g), and

  • ρwss = bulk density of the water-soluble salts (2.13 g/cm3).

Because VDM > VDML, the correction factor is a value >1. In cases where the mass correction is small (mH2OρH2O >> msrRwssρwss), the volume ratio approaches unity. DML concentrations were calculated from the diluted samples as follows:

[A]DML = [A]dil(VDM/VDML)Rdil, (25)

where

  • [A]DML = concentration of species A in the DML (expressed in mmol/L of DML),

  • [A]dil = diluted concentration measured in the laboratory, and

  • Rdil = dilution factor for the drilling mud that was centrifuged (10×).

Raw data for the freeze-drying procedure and volume correction results of the drilling mud are presented in MUD_VOL_CORR.XLSX in GEOCHEM in “Supplementary material.” In all five mud samples, the volume correction is small, ranging from 1.035 to 1.078. The volume correction has been applied to all of the drilling mud results (see Table T12 in the “Site C0020” chapter [Expedition 337 Scientists, 2013]).