Geology Reference
In-Depth Information
Strength at actual contact points between grains of soil or rock is
largely derived from electrochemical bonds over the true area of con-
tact, which is only a very small proportion of the apparent cross-
sectional area of a sample. At each contact between grains, elastic
deformation, plastic
flow and dissolution may take place, spreading
the contact point so that the actual contact area is directly proportional
to normal load. The attractive force over the true area of contact gives
rise to frictional behaviour (Hardy & Hardy, 1919; Terzhagi, 1925;
Bowden & Tabor, 1950, 1964). Bowden & Tabor, in particular,
established that the area of asperity contact changed linearly with
normal load for metals by measuring electrical resistance across the
junctions. Power (1998) carried out similar tests using a graphite-
based, rock-like model material (Power & Hencher, 1996).
The lower-bound friction angles for dry samples of quartz and calcite is
reportedly about 6 degrees but higher when wet (Horn & Deere, 1962).
The opposite behaviour was reported for mica and other sheet minerals.
Perhaps linked to Horn & Deere
s observations, mineral species that
reportedly give higher friction values whenwet are the same minerals that
commonly form strong bonds during burial diagenesis through dissolu-
tion and authigenic cementation (Trurnit, 1968). It is possible that the
presence of water allows asperity contacts to grow in these minerals, even
in laboratory tests. Conversely mica, chlorite and clay minerals are rarely
associated with pressure solution bonding and inhibit pressure solution
and cementation of quartz (Heald & Larese, 1974). Some authors have
questioned whether Horn & Deere
'
s data are valid because of possible
contamination and natural soil does not exhibit the same phenomena
(Lambe &Whitman, 1979), but there is other evidence that basic friction
of rock-forming minerals can be so low. Hencher (1976, 1977) used
repeated tilt tests on steel-weighted, saw-cut samples of sandstone and
slate to reduce the sliding angle from about 32 degrees to almost
12 degrees, which is approaching the low values of Horn and Deere.
The reduction in strength was attributed to polishing (
Figure 5.2).
'
Whilst basic friction the lower bound of minerals, originating from
adhesion at asperities, might be of the order of 10 degrees or even
lower, friction angles even for planar rock joints and non-dilational soil
are often greater than 30 degrees yet the additional resistance (above
basic) is still directly proportional to normal load. This additional fric-
tional component varies with surface
finish of planar rock joints and can
be reduced by polishing (Coulson, 1971) or by reducing the angularity of
sand (e.g. Santamarina &Cho, 2004).
Figure 5.3
shows results from two
series of direct shear tests on saw-cut and ground surfaces of granite. As
