Indentation load response

The indentation size effect (ISE) is a trend wherein hardness decreases with increasing indentation size or indentation load as shown schematically in Figure 2 i,2,3,4,5,6,7,8,9,10,11,12,13 plateau Knoop hardness is reached at loads from 5 N to 20 N in glasses. The ISE exists for both conventional Knoop and Vickers hardness, but usually with different trends due to different amounts of deformation, densifica-tion, displacement, displacement rate, and fracture induced by the two indenters. The ISE has been variously attributed to test procedure artifacts, frictional forces, environmental effects, or various material responses including elastic recovery, densifica-... 

After median-crack generation, when loading to larger indentation loads, lateral cracks may appear eventually. These propagate parallel to the indented surface and are circular in form. They are observed easily with optical microscopy since they produce optical interferences because of imperfect closure at the crack interftice. Under elevated loads, the craclcs emerge at the surftice and then form a chip which is later removed from the glass surface. These craclcs are responsible for glass erosion and abrasion (see Section 8.4). 

More recently, Schuh, Lund and Nieh (2004) using instrumented indentation showed that metallic glass loading response was dependent on the strain rate, with emphasized discontinuities in the curves observed under low strain rate. In fact, serrated flow has been reported extensively for many metallic glasses. Such serrations are observed in Figure J.6 as small jumps on the curve. The correlation between serration and slip lines is however still debated. 

Perhaps the most significant complication in the interpretation of nanoscale adhesion and mechanical properties measurements is the fact that the contact sizes are below the optical limit ( 1 t,im). Macroscopic adhesion studies and mechanical property measurements often rely on optical observations of the contact, and many of the contact mechanics models are formulated around direct measurement of the contact area or radius as a function of experimentally controlled parameters, such as load or displacement. In studies of colloids, scanning electron microscopy (SEM) has been used to view particle/surface contact sizes from the side to measure contact radius [3]. However, such a configuration is not easily employed in AFM and nanoindentation studies, and undesirable surface interactions from charging or contamination may arise. For adhesion studies (e.g. Johnson-Kendall-Roberts (JKR) [4] and probe-tack tests [5,6]), the probe/sample contact area is monitored as a function of load or displacement. This allows evaluation of load/area or even stress/strain response [7] as well as comparison to and development of contact mechanics theories. Area measurements are also important in traditional indentation experiments, where hardness is determined by measuring the residual contact area of the deformation optically [8J. For micro- and nanoscale studies, the dimensions of both the contact and residual deformation (if any) are below the optical limit. 

From a more fundamental point of view, the selection of different inden-ter geometries and loading conditions offer the possibility of exploring the viscoelastic/viscoplastic response and brittle failure mechanisms over a wide range of strain and strain rates. The relationship between imposed contact strain and indenter geometry has been quite well established for normal indentation. In the case of a conical or pyramidal indenter, the mean contact strain is usually considered to depend on the contact slope, 0 (Fig. 2a). For metals, Tabor [32] has established that the mean strain is about 0.2 tanG, i.e. independent of the indentation depth. A similar relationship seems to hold for polymers although there is some indication that the proportionality could be lower than 0.2 for viscoelastic materials [33,34], In the case of a sphere, an... 

Some difficulties also arise for the interpretation of scratch tests carried out at progressively increasing normal load or indentation depth. Figure 3 indicates, for example, that a transition from ductile deformation to brittle cracking can occur when increasing the normal load whilst the contact strain is nominally fixed by the conical indenter angle. This is indeed observed in many polymer systems and the notion of a critical load at the ductile-brittle transition is largely used to characterize the scratch response. This depth... 

Extensive TEM studies by Page et al. [65] delivered all previous low-temperature electron microscopy results to the consistent view that (i) silicon becomes amorphous in response to the high contact stresses under a hardness indenter and (ii) limited dislocation arrays are generated around the deformed volume at contact loads exceeding some threshold value. The authors also argued that the dislocation arrays might occur as a means of accommodating the displacements from the densification transformation, rather than as a primary response to the indenter intrusion. 

Therefore, in order to obtain reliable values of the hardness, one should always use at least >4 pm thick films, make sure that the maximum indentation depth, Amax. does not exceed 10% of the film thickness, check for possible dynamic response of the material by sustaining the indentor at the maximum load Ln x... 

In particular, the Vickers indentation method seems favourable to compare the response of inorganic matrix composites containing CNTs to contact loads due to its simplicity and easy sample preparation and test operation. It should be noted that the fracture toughness values shown in Table 1 were all determined by the Vickers indentation method as this has been the accepted practice in the field in the last years and in most cases the fracture toughness of CNT/ceramic composites has been obtained by this method. 

---