GEOCHEMISTRY

Geochemistry during Leg 193 concentrated on the analysis of igneous rocks, altered rocks, and in situ downhole water samples. The analytical methods used were elemental analyses of total nitrogen, carbon, hydrogen, and sulfur in rock samples and ICP-AES analyses of all rock types and waters as well as ion chromatography (IC) of waters. A sulfide methodology for the ICP-AES analyses was developed for digestion and analysis of sulfide samples and is described in "Geochemistry" in the "Site 1188" chapter. Water samples were analyzed for major cation and anion concentrations as well as salinity, alkalinity, temperature, and pH. In addition, the concentration of H₂O-soluble sulfate of altered rocks was determined gravimetrically.

Before undergoing elemental analysis for carbon, hydrogen, nitrogen, and sulfur and ICP-AES analysis, samples were sawed into small blocks and the saw marks and exterior surfaces (with associated contamination) were removed by grinding on a diamond lap wheel. The sample blocks were subsequently rinsed for 15 min in deionized water, then rinsed again in a methyl alcohol ultrasonic bath for 10 min and subsequently dried at 110°C. The clean blocks were crushed in a Spex 8510 shatter box using tungsten carbide barrels. Whenever recovery permitted, at least 5 cm³ of material was ground to ensure a representative sample was analyzed. Weighing measurements were made on a Scientech balance with weighing error of ±0.0005 g.

The samples selected for ICP-AES analysis and splits of the samples selected and prepared for XRD were analyzed on an NA1500 Carlo Erba NCS (nitrogen, carbon, and sulfur) elemental analyzer for total sulfur, nitrogen, hydrogen, and carbon. The sample powders (5 mg) were combusted at 1000°C in an oxygen atmosphere with the addition of vanadium pentoxide to convert sulfur to SO₂, carbon to CO₂, and nitrogen to NO₂. The NO₂ was then reduced to N₂ using copper metal. The gases were separated by gas chromatography and measured with a thermal conductivity detector. The precision of these analyses is better than 1%-2% of the analyzed values of the standards. The total sulfur values were initially used to screen the samples with high sulfur values from those with low sulfur values for method development on the ICP-AES. Total sulfur values were also used in the calculation of the H₂O-soluble sulfate phase content using the equation:

TS = S_{soluble sulfate} + S_{sulfide + insoluble sulfate} ,

where TS is total sulfur, S_{soluble sulfate} is sulfur in H₂O-soluble sulfate phases, and S_{sulfide + insoluble sulfate} is sulfur in sulfide phases plus H₂O-insoluble sulfates. The total sulfur values are reported in the tables together with the major element oxide, trace element, and loss on ignition (LOI) data for samples from each site. In general, the values range from trace amounts in the unaltered rocks to 9% TS in the highly altered units. The precision for this method is 1%-2% of the measured quantities. Subsequently, all samples from Hole 1188A were analyzed following the methods for geochemistry described below.

In addition to the rock samples, the residue from the LOI analysis (ignited in quartz crucibles) was also measured on the Carlo Erba NCS elemental analyzer for sulfur contents to check for the extent of oxidation and the removal of sulfur from the samples during ignition. The pre- and post-ignition values obtained for the elemental analysis of total sulfur showed that not all sulfur-bearing components were removed in the LOI roasting process. Because small amounts of barite have been found in thin section and by XRD, it is possible that the residual sulfur in the samples is, in part, attributed to barite, which would not be affected by the LOI temperature of 1050°C. The carbon values are quite low and are ignored because the samples were milled in a tungsten carbide shatter box. Also, no carbonate minerals have been identified in thin section or by XRD. The carbon dioxide component of the LOI is therefore modeled as CO₂ = 0. The nitrogen blank was unusually high, potentially because of a small leak in the sample inlet assembly, therefore these data, also generally low values, are also ignored. Hydrogen values from the NCS analysis were used to calculate the H₂O⁺ content of the samples.

An ICP-AES instrument (Jobin Yvon JY2000) was used to determine the major oxide and selected trace element abundances of the ignited whole-rock powders. Major oxides and the concentrations for the trace elements Zr, Y, Zn, Cu, Ba, and Sr were determined by ICP-AES analyses following the general procedure outlined by Murray et al. (2000) and Shipboard Scientific Party (2001). Analyses were carried out on lithium metaborate fusions prepared from 100 mg of ignited powder that was mixed with 400 mg of preweighed (on shore) lithium metaborate. This mixture was melted in air at 1150°C in a Pt-Au crucible for 10 min. The resulting fused bead was cooled, removed from the crucible, and dissolved in 50 ml of nitric acid (2.3 M) during agitation for 1 hr. The 4:1 flux-to-sample ratio was used for samples, standards, and blanks. This reduces the matrix effects for a range of igneous rock compositions. Therefore, the relationship between plasma emission intensity and concentration becomes linear and can be described by

C_i = (I_i · m_i) - b_i ,

where C_i is the concentration (in either oxide weight percent or parts per million) of element i, I_i is the net peak intensity (in voltage) of element i, m_i is the slope of the calibration curve (in concentration per voltage) for analyte i, and b_i is the measured blank of element i (in either oxide weight percent or parts per million). The slope m_i was calculated by regressing a line through intensity data obtained by measurement of well-characterized standard reference materials (U.S. Geological Survey STM-1 and RGM-1 and Geological Survey of Japan JA-2, JA-3, JB-2, and JR-2).

Methods for Igneous and Sulfide Rock Geochemistry

Major oxide concentrations, along with a set of six trace elements, were acquired by ICP-AES. Cu and Zn were added to an existing igneous-rock analytical methodology because of their importance in mineralized systems, whereas Zr and Y were included to help assess parentage of the altered rocks and Sr and Ba to help characterize alteration style.

An ICP-AES sulfide method was developed for samples containing >10% total sulfur as sulfides. This involved roasting the samples at 600°C for 6 hr to oxidize all the expected sulfide sulfur. Subsequently, the oven temperature was raised to 1050°C and the standard LOI methodology was followed. This two-step heating procedure was employed so as not to damage the quartz crucibles. The fluxing of the samples, and reference rocks, was then completed in ultrapure graphite crucibles. Standard dissolution techniques were used to digest the fused bead into a solution.

Sulfate Analysis

To obtain the ratio of H₂O-soluble sulfate to sulfide in the rocks from Holes 1188A and 1189A, a portion of the powdered rock prepared for the ICP-AES analysis was used for a gravimetric sulfate determination. First, 0.5 g of sample was weighed in a 50-mL polyethylene centrifuge tube and then 50 mL of nanopure water (18.2 M) was added. The samples were agitated and then placed in a rack in a 4°C refrigerator for 12 hr. The samples were then centrifuged, and the supernatant liquid was filtered. The filtered solution was transferred to an acid-washed beaker, to which barium chloride was added promoting the precipitation of BaSO₄. Samples were covered with a watch glass and allowed to react for another 12 hr, after which the barium sulfate was filtered through preweighed 0.45-µm filters. The concentration of sulfate in the former solution was determined gravimetrically. The difference between the total sulfur from the elemental analyzer and the gravimetric measurement of soluble sulfate was considered to represent the insoluble sulfide fraction of the sample along with any barite present in the rock.

Water Samples

Fe, Mn, Ba, Li, and Sr concentrations were determined by ICP-AES analysis following the general procedure outlined by Murray et al. (2000). During Leg 193, given the close proximity of the sites, three water samples were collected over the side of the ship and are suggested to be generally representative of the bulk seawater contribution of the hydrothermal system. Water samples were collected into acid-cleaned plastic buckets from the side of the ship and filtered through sterile 0.45-µm Gelman polysulfone disposable filters. Samples for shipboard analyses were stored in plastic vials and those for shore-based analyses were stored in heat-sealed acid-washed plastic tubes and/or glass vials.

Analyses of water samples followed the procedures outlined by Gieskes et al. (1991). Salinity was measured with a Goldberg optical handheld refractometer. The pH and alkalinity were measured by Gran titration with a Brinkmann pH electrode and a Metrohm autotitrator. The Cl^- concentration was measured by titration against silver nitrate. Concentrations of NH₄⁺ were measured by spectrophotometric methods with a Milton Roy Spectronic 301 spectrophotometer. The concentrations of K⁺, Mg²⁺, Na⁺, Ca²⁺, and SO₄^2- were measured by ion chromatography using a Dionex DX-120 instrument. Analytical precision was determined by replicate analyses of natural samples and by reanalyzing standards as unknowns. Values of precision (expressed as percent of the measured value) are as follows for the respective constituents: alkalinity is <1.5%; C^- is 0.4%; Ca²⁺ is ~<1%; Mg²⁺ is 0.5%; NH₄⁺ is ~5%; K⁺ is <3%; SO₄^2- is <4%; and Na⁺ is <5%. Chemical data for borehole waters are reported in molar units.

Some rock samples used for porosity measurements (see "Physical Properties") yielded such high porosity values that it was hypothesized that soluble sulfate minerals were dissolving during the measurement. To test this, selected rock samples were soaked for 24 hr in seawater, and the water was subsequently analyzed by IC for SO₄^2- concentration.