Previous   |    Next

doi:10.2204/iodp.proc.333.201.2014

Methods

Whole-round samples were sealed and stored at 4°C to minimize water loss prior to testing. Shipboard selection of whole-round specimens was conducted by D/V Chikyu staff based on visual inspection and analysis of X-ray computed tomography (CT) images to identify areas free of cracks and voids and with minimal (no visual) coring disturbance. Material for grain size and MICP measurements was taken immediately adjacent to samples used for geotechnical tests to ensure data consistency.

Constant-rate-of-strain consolidation tests

CRS consolidation tests were performed at room temperature (20°C) with consolidation in the vertical direction (parallel to the z-axis of the core), except for Sample 333-C0012E-3X-4, 31.5–37.5 cm, with which we performed an additional consolidation test normal to the z-axis of the core (i.e., in the horizontal direction). These tests were performed following American Society for Testing and Materials (ASTM) International standards (ASTM International, 2006). Samples were extruded from the core liner and trimmed into a fixed metal ring using a trimming jig, wire saw, sharp-edged spatula, and recess tool. This ensured that all samples had a height of 2.41 cm and diameter of 5.09 cm prior to testing. After trimming, the sample and fixed ring were loaded into the consolidation chamber with porous stones and filter paper on the top and base of the sample. The use of the fixed ring ensured zero lateral strain during consolidation. The consolidation chamber was sealed, filled with distilled water, and pressurized with a constant pressure of 386 kPa overnight (>8 h) to ensure complete saturation of the sample. Following this saturation stage, the drain valve at the base of the sample was locked and the sample was consolidated vertically at a constant strain rate (). During consolidation, the pore pressure ratio (ratio of difference between the pressure at the base of the sample and chamber pressure to the chamber pressure) was monitored, and was adjusted between 0.2% and 0.5%/h to maintain a pore pressure ratio <0.15. During the test, the total axial stress (σ_a), instantaneous sample height (H), and pore pressure at the base of the sample (P_p) were recorded. Each test proceeded until 20% axial strain to ensure a suitable amount of virgin consolidation data (consolidation at vertical effective stress greater than the past maximum effective stress of the sample; we assume the past maximum effective stress equals the preconsolidation stress).

Data recorded during each consolidation test were used to determine compression index (C_c), permeability (k), preconsolidation stress (σ_pc′), and OCR. C_c defines the stress-strain relationship during virgin consolidation (Craig, 1992) (Fig. F2) and was computed as

where

• C_c = compression index,
• e = void ratio, and
• σ_a′ = axial effective stress (Pa).

Void ratio during the consolidation test was determined from strain data and the initial void ratio at laboratory conditions, which was computed from mass and density measurements following the method of Blum (1997). Permeability was computed as

k = (HH0µ)/(2Δu),

(2)

where

• k = permeability (m²),
• = strain rate (1/s),
• H = specimen height (m),
• H₀ = instantaneous specimen height (m),
• µ = dynamic viscosity of pore fluid (Pa·s), and
• Δu = base excess pressure (Pa).

We assumed a pore fluid dynamic viscosity of 0.001 Pa·s. Base excess pressure is defined as the difference between the pore pressure at the base of the specimen (P_p) and the consolidation cell pressure (P_c) (Δu = P_p – P_c). Base excess pressure data were smoothed using a three-point moving average, and strain rate data were smoothed using a six-point moving average. Permeability was extrapolated to initial permeability (k₀) at the initial porosity for each specimen by assuming a log-linear relationship between porosity and permeability during virgin consolidation (e.g., Neuzil, 1994) (Fig. F3).

Permeability and axial effective stress data were used to determine the coefficient of consolidation (c_v) during each test (ASTM International, 2006; Craig, 1992):

where

• c_v = coefficient of consolidation (m²/s),
• k = permeability (m²),
• µ = dynamic viscosity of pore fluid (Pa·s), and
• m_v = coefficient of volume compressibility (1/Pa).

The coefficient of volume compressibility (m_v) is defined as the change in axial strain per unit increase in axial effective stress (Craig, 1992) and is computed as

where ε = axial strain.

Preconsolidation stress (σ_pc′) was determined for all samples except the horizontally oriented measurement on Sample 333-C0012E-3X-4, 31.5–37.5 cm. σ_pc′ represents an estimate of the maximum vertical effective stress that a sample has experienced. We determined σ_pc′ using the work-stress method of Becker et al. (1987). This method analyzes the slope of the preyield and postyield behavior of the sample in axial stress-work per unit volume space (Fig. F4). The work per unit volume (ΔW) resulting from an incremental increase in axial stress was computed by

where W = work per unit volume (J/m³).

This defines σ_pc′ for each specimen. Sample disturbance can result in a poorly defined σ_pc′ (Santagata and Germaine, 2002). See Saffer (2003) and Dugan and Germaine (2008) for further discussion of errors associated with interpretation of σ_pc′. OCR is the ratio of σ_pc′ to the hydrostatic vertical effective stress (σ_vh′). σ_vh′ is determined as the difference between the total vertical stress and the hydrostatic fluid pressure and was determined for each vertically oriented specimen from shipboard moisture and density (MAD) data:

(6)

where

• ρ_b = bulk density from MAD data (kg/m³),
• ρ_w = pore fluid (water) density (kg/m³),
• g = acceleration due to gravity (m/s²),
• ζ = constant of integration (m), and
• z = depth below seafloor (m).

We assumed ρ_w = 1024 kg/m³. The first term on the right-hand side of Equation 6 is the total vertical stress, and the second term is the hydrostatic fluid pressure. The estimated in situ vertical effective stress from Equation 6 and the preconsolidation stress from the work-stress analysis were used to determine OCR. An OCR >1 suggests that the sample has been unloaded from a previous greater hydrostatic vertical effective stress, and an OCR <1 suggests in situ overpressure conditions. Cements or other diagenetic mineralization may result in an OCR >1 because of enhanced sediment strength (e.g., Morgan et al., 2007). An OCR = 1 is interpreted to represent that the specimen was at its greatest hydrostatic vertical effective stress at in situ conditions.

Grain size measurements

Grain size measurements were conducted following the ASTM standard for particle size analysis (ASTM International, 2007). All measurements were conducted by hydrometer analysis in a settling column, as no particles were retained by passing through a 2 mm sieve. Samples were oven-dried at 105°C for at least 24 h and powdered with a ceramic mortar and pestle. The powdered samples were then mixed with distilled water and 5 g of sodium hexametaphosphate dispersant and left to soak overnight (>16 h). After soaking, the samples were further dispersed for 1 min using a Hamilton Beach milkshake mixer, poured into a settling column, and diluted with distilled water to make 1 L of solution. The column was then agitated for 1 min and left to settle. During settling, measurements of the bulk density of the solution were made periodically using ASTM hydrometer 151H. The mass fraction of particles remaining in suspension (m_p) at the time of measurement is given by

where

• m_p = mass fraction of particles remaining in suspension,
• ρ_s = specimen grain density (taken as 2700 kg/m³),
• V = volume of solution (m³),
• m_s = dry mass of specimen (kg),
• ρ = hydrometer reading (kg/m³), and
• ρ_f = density of solution fluid without sediment (kg/m³).

The particle diameter (D) corresponding to a mass fraction of particles obtained from Equation 7 is given by

(8)

where

• D = equivalent particle diameter (m),
• µ = solution fluid viscosity (Pa·s),
• L = effective depth from solution surface to center of hydrometer bulb (m),
• t = time of hydrometer measurement (s), and
• g = acceleration due to gravity (m/s²).

L was determined for hydrometer 151H from Table 2 of the ASTM standard (ASTM International, 2003).

Specific surface measurements

Specific surface measurements were conducted by methylene blue adsorption using the spot-test method (Santamarina et al., 2002). Samples were oven-dried at 60°C for at least 72 h to allow evaporation of pore water but prevent any clay alteration. Following drying, samples were powdered with a ceramic mortar and pestle. During measurement, 10 g of powdered sediment was mixed with a solution of 1 g methylene blue powder (C₁₆H₁₈ClN₃S) and 200 mL distilled water in 0.5 mL increments dispensed by pipette. The mixture was stirred continuously with a magnetic stirrer throughout the measurement process. After stirring for 1 min following addition of a methylene blue solution increment, a sample of the mixture was taken by eyedropper and a drop placed on Fisher brand filter paper P5. When all mineral surfaces are coated with methylene blue, the excess methylene blue in solution will bleed out around the drop on the filter paper and form a halo around the drop. When a halo was observed, the measurement process was deemed complete. The specific surface (S_a) is given by

where

• S_a = specific surface (m²/g),
• M_m = molar mass of methylene blue (319.87 g/mol),
• N = number of methylene blue increments,
• A_v = Avogadro’s number (1/mol), and
• A_MB = mineral surface area covered by one molecule of methylene blue (1.3 × 10^–18 m²).

Because methylene blue adsorption is performed on wet samples, the measured surface area includes the surfaces within clay interlayers (Santamarina et al., 2002).

Mercury injection capillary pressure measurements

MICP measurements were performed at room temperature (20°C) using a Micrometrics AutoPore device. Prior to measurement, samples were oven-dried at 115°C for at least 24 h. During the measurement, each sample was immersed in mercury within a pressure-sealed chamber, which was attached to a capillary stem with a cylindrical coaxial capacitor. The mercury pressure was increased incrementally to a maximum of 380 MPa, and each pressure step was held until volume equilibrium was reached as determined from the change in the capacitance of the system. The volume of mercury injected at each increment was determined by capacitance measurements. Pore volume was computed from the bulk sample volume determined by immersion in mercury, and porosity was determined following the method of Blum (1997).

The volume of mercury injected at each pressure increment was used along with pore volume and pressure data to construct a pore size distribution for each sample. Mercury injection pressure (P_Hg) was converted to pore radius (r_p) using the Young-Laplace equation:

where

• r_p = pore radius (m),
• σ_Hg = air-mercury interfacial tension (0.485 N/m),
• θ_Hg = mercury-sediment contact angle (140°), and
• P_Hg = mercury injection pressure (Pa).

The median pore radius (r₅₀) was determined as the median of the pore size distribution.

Mercury injection pressure (P_Hg) was converted to air-water capillary pressure (P_c) by

where

• P_c = air-water capillary pressure (Pa),
• σ_aw = air-water interfacial tension (0.072 N/m),
• θ_aw = air-water contact angle (180°), and
• P_Hg = mercury injection pressure (Pa).

Air-water entry pressure was determined from the minimum mercury injection pressure at which the volume of injected mercury was nonzero. This corresponds to the mercury percolation threshold (Bear, 1972).

Top of page   |    Previous   |    Next