1184
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
mineralogy of Pierre Shale is primary clay minerals, specifically
smectite and mixed-layer clays.
Unweathered samples were obtained from a boring at the
Oahe Dam site, South Dakota, at depths below 60 meters. Firm
shale, without any weathering, has an average dry density of
1.71 g/cm
3
, a moisture content of 25%, and specific density of
2.7 at the site. Unweathered Pierre Shale at the site exhibits
unconfined strengths of 0.5 to 17.4 MPa and a Young’s
modulus between 137 and 965 MPa (Johns
et al.
1963). The
wide variation in strength shows the engineering complications
in classifying Pierre Shale as either a rock or soil. Extensive
problems were encountered during the construction of the Oahe
Dam due to the rapid strength loss in the material (Knight
1963). After many failures, slopes had to be redesigned with
´
=8.5 degrees and 14.4 kPa for cohesion (Johns
et al.
1963).
Weathering was observed to cause the durable, brittle, rock-like
shale to turn into a weathered, soil-like material at the site. The
peak strength of the weathered material was found to be
´
=11.9
degrees and 24.9 kPa for cohesion, but due to the possibility of
slip surfaces in the shale, the residual strength controlled the
design parameters. In a laboratory slaking test, Botts (1986)
observed nearly a 75% drop in the shear strength of Pierre Shale
samples, or a 6-degree drop in the internal friction angles and a
reduction of the cohesion from 848 kPa to 0 kPa, after one
wetting and drying cycle.
1.2
Sample Preparation
1.2.1
Residual Shear Strength
The testing procedure for the Bromhead Ring Shear device used
a modified version of ASTM D6467. The material was crushed
and passed through a US standard #50 sieve, remolded with
distilled water to their plastic limit and allowed to hydrate for
48 to 72 hours. Higher water contents are suggested by ASTM
(2003) but Bromhead (1979) advises drier samples in the
Bromhead Ring Shear device to prevent excess settlement of the
top platen.
1.2.2
Micromorphology
Central to the quality of any fabric analysis in soils is a
preparation method preventing disturbance. Dry samples are
required for analysis in the vacuum environment of the scanning
electron microscope (SEM). The freeze-drying technique was
used to obtain relatively undisturbed samples. A sample from
each weathering cycle was placed in an intermediate freezer for
twelve hours at a temperature of -75 degrees Celsius and then
placed in a VirTis Ultra 35 8-shelf model freeze dryer at a
temperature of -25 degrees Celsius until a vacuum of <100
millitorr was achieved. The temperature was raised to 26
degrees Celsius and held constant for 48 to 72 hours. Tension
fractured and placement on carbon stubs preceded the analysis
in the SEM.
1.2.3
Mineralogical and Chemical Analysis
X-ray fluorescence (XRF) and x-ray diffraction (XRD) analyses
used both bulk samples and size fractions to determine the
composition of the material. Bulk, silt, and sand mineralogy
was obtained from random-oriented mount samples mounted in
the XRD unit and analyzed at a speed of 2
o
two theta per minute
with copper K-alpha radiation. Minerals were identified by a
computerized catalog of the Joint Committee on Powder
Diffraction Standards (JCPDS) Powder Diffraction File system
(The International Centre for Diffraction Data® 2004).
Particle-size separation was performed on the material on the
basis of Stoke’s law, where particles <2
m equivalent spherical
diameter were obtained. This fraction was further separated by
the same method into the coarse clay fraction, particles less than
2
m, and the fine clay fraction, particles less than 0.5
m. The
oriented-mounted samples intensify the (001) reflections and
reduce (hk0) reflections by removing non-platy minerals,
dispersing clay minerals into individual colloidal particles, and
laying the clay particles flat (Moore and Reynolds 1997). The
filter transfer method, as recommended by Moore and Reynolds
(1997), was used to transfer the material to a filter. The type of
clay mineral is identified by the characteristic expansion,
contraction, or collapse of the clay mineral’s d-spacing through
five subsequent treatments: air drying, glycolation with ethylene
glycol, heating to 400°C, and heating to 550°C. The methods
described in Shultz (1978) were used to estimate the
concentrations of the clay minerals in the mixed-layer.
2 TESTING AND ANALYITICAL METHODS.
2.1
Weathering Cycle
ASTM standard C593 (ASTM 2003) was used as an outline for
the saturation of the Pierre Shale. The vacuum-saturation
strength testing procedure described in the standard was
modified to allow for saturation of the soil samples. The test
consisted of obtaining multiple two cubic-inch samples of
unweathered Pierre Shale and placing them on a filter in a
vacuum chamber. Samples were vacuumed at 24 inch Hg (11.8
psi) for one hour in the chamber; the chamber was then flooded;
and the samples were soaked for one hour in the distilled water
bath. The samples were removed from the bath and allowed to
air-dry for 48 to 72 hours, completing one weathering cycle.
2.2
Classification and Ring Shear
Classification testing followed applicable American Society for
Testing and Materials (ASTM 2003) standards. Atterberg limit
and hydrometer analyses were performed on the samples. The
Bromhead Ring Shear Device was used for measurement of the
residual strength. The device was first proposed by Bromhead
(1979) and has provided results that are in good agreement with
back-calculations for slope stability analysis (Bromhead and
Dixon 1986, Skempton 1985). The mold dimensions are 100-
mm outer diameter, 70-mm inner diameter, and a thickness of 5-
mm. Replicate testing was not conducted for the cycling studies
due to sample limitations.
The prepared sample was molded into the ring at a water
content near its plastic limit and placed in a distilled water bath.
Multistage testing was used to obtain the residual strength. This
technique uses progressive loading after the formation of a
shear surface. The samples were initially consolidated at the
normal stress of 100 kPa and then sheared at a rate of 0.16
degrees per minute (0.119 mm per minute) for a displacement
of one revolution to form the shear surface. The rate was
slowed down to 0.048 degrees per minute (0.036 mm per
minute) and allowed to shear until the residual strength was
reached, typically an additional 10 to 15 mm. The residual
strength was obtained for 100, 200, and 400 kPa normal stress
increments.
2.3
Scanning Electron Microscopy
Freeze-dried samples were analyzed with a Hitachi S-2460N VP
scanning electron microscope (SEM) with energy dispersive
spectrographic (EDS) and digital imaging capability. Two
sources were used to obtain surface topography: backscattered
electron (BSE) and secondary electron (SE). BSE mode was
used for lower-resolution imaging and elemental analyses while
SE mode was used for higher magnifications. Samples were
coated with gold to prevent charging for samples analyzed in
the SE mode. Digital elemental maps were produced using the
EDS function on the SEM. Mineral contents were inferred by
coinciding elemental maps.
