Previous   |   Next

doi:10.2204/iodp.proc.301.205.2008

Methods

A total of 83 samples were collected from Hole U1301B for shipboard determination of physical properties. Analyses included bulk density, grain density, porosity, thermal conductivity, and seismic velocity, but not all measurements were made on each sample. Physical properties were measured on cubes (usually 2 cm on a side) cut from larger pieces of the core. After being measured on the ship, the cubes were transferred to our laboratory for SSA measurement of 13 selected samples. The selected samples spanned the range of depth and the range of shipboard physical properties.

Physical property methods

Methods for physical property analyses are presented in detail in the "Methods" chapter and are briefly summarized here. Bulk density, grain density, and porosity properties are measured by resaturating samples in natural seawater under a partial vacuum for 24 h and then weighing them to obtain a wet mass. Samples are dried in an oven for 24 h at 105°C and then a dry mass is obtained. Sample volume is measured using a Quantachrome gas pycnometer (using He). Density and porosity properties can be calculated from these three measurements plus an assumed salinity of the resaturation water. Seismic velocity was measured using a P-wave logger system (Hamilton frame PWS-3). Samples were resaturated under a partial vacuum in seawater for at least 2 h, and then they were sanded, polished, and ultrasonically cleaned. The compressional wave velocity calculation is based on the delay time of a 500 kHz square wave signal traveling between a pair of piezoelectric transducers in contact with opposing surfaces of the sample cube.

Surface area measurement

The standard method for measuring SSA is by gas adsorption and the application of the Brunauer, Emmett, and Teller (BET) equation (Brunauer et al., 1938), which is described in detail in a number of reference books (e.g., Gregg and Sing, 1982; Lowell and Shields, 1991; and Condon, 2006). The principle of BET analysis is that at low relative pressure, gas adsorbs to a solid in a monolayer (multilayers form at higher pressures). By knowing the number of gas molecules in a monolayer and the dimensions of an individual molecule, the surface area covered by the monolayer can be calculated (see Table T3 for an example). The BET equation models a measured number of moles of adsorbate (n) adsorbed on 1 g of sample with the following relationship (Rouquerol et al., 1994):

,

(1)

where

n_m = calculated number of moles adsorbed as a monolayer on 1 g of adsorbent,

P = gas pressure,

P₀ = saturation vapor pressure of the gas, and

c = a constant that is dependent upon the shape of the isotherm.

We used an instrument with a volumetric approach (ASAP 2000, Micromeritics, USA) so n_m is obtained by plotting P/V_a(P₀ – P) against P/P₀, where V_a is the volume of gas adsorbed per gram of sample normalized to standard temperature and pressure (STP) (cm³/g). Generally the plot is linear at low relative pressures (P/P₀ < 0.3). The slope of the linear part of the graph has a slope (s) of c – 1/V_mc and an intercept (i) of 1/V_mc, where V_m is the volume of gas required to form a monolayer on a unit gram of the sample (cm³/g). Both s and i have units of cm³/g at STP. By algebraic substitution, V_m = (s + i)^–1. Finally, the BET SSA can be calculated with the equation

,

(2)

where

a_m = area occupied by a molecule of the adsorbate (1.49 × 10^–19 m² for Ar),

L = Avogadro's number, and

V_l = molar volume of the analysis gas (22 L for Ar) at STP.

The 8 cm³ cubes (2 cm × 2 cm × 2 cm) used for shipboard physical property measurements were too large to fit in standard BET analysis tubes, which have an 11 mm opening. Therefore, we sectioned the samples into a few rectangular blocks (minimum dimension ≤ 7.5 mm) with a rock saw to allow them to pass through the analysis tube openings. In order to remove cutting debris and precipitated salt, sectioned samples were ultrasonically cleaned, resaturated with deionized water under a partial vacuum for 24 h, and then dried at 105°C for 72 h. Samples were placed in acetone-rinsed tubes and then degassed under a partial vacuum at 150°C for 24 h (until the system pressure dropped below 1.3 kPa). Sample size was limited by the size of the degassing station heating element. In each case, we used the maximum amount of sample that could be uniformly heated by the element, which was generally ~6 g. After degassing, samples were immediately analyzed using N₂ or Ar as the adsorbate. Both gases have similar physical properties and are commonly used to make surface area measurements (Gregg and Sing, 1982). Ar is a slightly smaller molecule and has a lower vapor pressure. Analyses performed with Ar, with lower vapor pressure, result in a smaller dead space correction; Ar is therefore used for materials with small surface areas because it is more sensitive than N₂ (Anbeek, 1992; Sing et al., 1985). Although Ar measurements are generally more precise, they are not necessarily more accurate because the molecules may interact with the sample differently. Therefore, we present values derived from both gases. Multipoint surface area (five points) was calculated for relative pressures (P/P₀) ranging from 0.058 to 0.225. All samples were measured at least twice to confirm repeatability of adsorption isotherms; some were measured as many as four times. A commercially prepared kaolinite sample was periodically run as a standard.

Top of page   |   Previous   |   Next