where Pm is the mean
pressure and Y is the yield stress. If the mean pressure is less than this,
the contact remains elastic.
An indentation process may produce:
(i) a reversible, elastic deformation; (ii) a permanent, plastic
deformation; or (iii) both plastic and elastic deformation. The detailed
behaviour depends on the ratio of the actual strain to the yield strain of
the material. Low ratios (< 2) produce elastic behaviour; high ratios (> 50)
produce plastic behaviour. The actual strain is given by tanb , where b is
the angle between the indenter and sample surfaces, so it is clear that a
spherical indenter will behave in a fundamentally different way from a
pyramidal indenter. In the case of a Vickers pyramid, for example, b is
constant and the strain is therefore constant (about 8%), regardless of
depth.
For a spherical indenter, b , and
therefore the strain, increases as the indentation depth increases. A series
of spherical indentations with progressively higher maximum load can
therefore produce results ranging from purely elastic to elastic + plastic,
and, furthermore, can yield curves of Stress vs. Strain.
In the NanoTest, multiple load - unload cycles with
increasing maximum loads are performed at one point. From this data and the
known diamond radius, the contact area is determined and hardness, Young's
modulus, stress and strain are calculated. The contact areas are corrected
for possible "piling up" or "sinking in" around the contact perimeter.
Experimental
Spherical indentations were produced
at two points on a highly polished E52100 steel surface. Five sets of
load-unload data were obtained at each point, with the maximum load being
increased from P1=2 mN to P5=10 mN. For each
indentation, unloading was continued to 10% of the maximum load. The diamond
indenter was conical with a spherical endform of 5 µm radius.
A transition from
elastic to fully plastic deformation was observed in the load range
investigated. A first order approximate value of a', the radius of the
contact circle at each maximum load, was calculated, with the plastic depth
determined by differentiating the first part of the unloading curve and
taking epsilon as 0.75. A plot of log(Pi) vs. log(a') was
produced for all experiments (i.e., for i = 1 to 5). This gave Meyer's index
2+1/n as the slope. The "pile up" correction parameter c (actual contact
radius = ca') was found to be 0.923, thus indicating "sinking-in" type
deformation.