PSI - Issue 5

Annamaria Cividini et al. / Procedia Structural Integrity 5 (2017) 1072–1077 Cividini / Structural Integrity Procedia 00 (2017) 000 – 000

1074

3

The tests were carried out in general accordance with (ASTM D3080, BS 1377-7) on three cylindrical specimens (60.mm in diameter, 23.mm in height) trimmed from one of the adobe blocks. Four steps characterize the performed direct shear tests, namely: 1) application of normal (vertical) stress  n under oedometric condition (i.e. the lateral deformation is constrained); 2) shearing stage for determining the ‘peak’ resistance (displacement controlled test, at 0.5mm/h); 3) a series of five backward/forward traverses carried out by hand, for particle rearrangement toward the residual condition; 4) re- shearing stage for determining the ‘residual’ properties. The experimental results are plotted in Figure 1 where the variation is shown of the shear stress and of the vertical displacement (change in height) versus the horizontal displacement. It can be observed that during shearing (solid lines) one sample is affected by a loss of strength (or strain softening) after the peak stress has been reached, whilst the other two present a more “ductile” behaviour. The di fferent behaviour could depend on both the inhomogeneous nature of the adobe material and on the increase in the applied normal stress. During re-shearing (dotted curves in Figure 1) the behaviour of the three specimens was comparable and characterized by two main aspects, that are: (a) a marked reduction of the attained shear resistance and (b) a contractive behaviour preceding the dilatant one.

Fig. 1. Evolution of shear stress (a) and of change in height (b) versus horizontal displacement from direct shear (DS) tests on dry Aliano adobe. The positive vertical displacement indicates a reduction in the specimen height.

For evaluating the variation of the shear resistance, normal and shear stresses are plotted in Figure 2 for three different conditions as follow. The stress values measured on the horizontal plane at failure (‘peak’ condition in shearing stage, DS pk ) are represented by square symbols and it can be noted that the shear resistance at low normal stress does not compare with the resistance of the other two samples. On the contrary the stress data at the end of shearing stage ( DS e-o-sh ) and those of the re-shearing stage ( DS re sh e-o-s , i.e. triangular symbols in Figure 2) lie on two straight lines, so that the Mohr-Coulomb failure criterion could be adopted for interpolating the experimental results. It is worthwhile to note that a non-negligible value of cohesion characterizes the resistance at the end of re-shearing, indicating that the residual condition was not fully attained in the adobe reinforced with plant fibers. This in turn confirms the relevant reinforcing effect provided by plant fibers in dry adobe.

2.3. Unconfined compression tests

Liberatore and co-workers carried out also six uniaxial compression (UC) tests on prismatic blocks from Aliano adobe buildings. They provided the measured compressive strength at failure (see Table 3 in Liberatore et al., 2006) observing a significant scatter of the results. The corresponding Mohr circles are shown in Figure 2 and the data seem consistent with the results from the DS tests, even if the specimens are different in size.