PSI - Issue 5

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

1073

2

during time (due e.g. to heavy rainfalls); (c) the shear strength could be reduced when the material is subjected to wetting/drying cycles. Apart from the conventional identification and classification procedures, the unconfined compression, the triaxial and the direct shear tests could be considered. The first one involves a simple laboratory procedure. The second test involves a relatively complex laboratory procedure (for instance, Cividini, 1999), however provides detailed results on the stress-strain behavior of the tested material. The third one is less accurate than the second one, the test has to be considered as a boundary value problem for identifying the elastic properties but it requires an apparatus available almost everywhere in the geotechnical laboratories. In the present context both 'natural' and re-constituted samples of adobe reinforced with plant fibers were tested. The 'natural' samples were trimmed from small adobe blocks and the blocks were, in turn, recovered from the collapsed walls in a building of the historical heritage of Aliano (Basilicata, Italy). The investigation on the re-constituted samples was necessary for verifying the influence of the fiber content, even if experimental results are obtained from uniaxial compression tests. With reference to the materials considered in the experimental investigation, the study intends to provide some additional information for the development of constitutive laws applicable in the numerical analysis (for instance, Desai et al., 1991; Shieh-Beygi and Pietruszczak, 2008; Cividini, 2005). In turn, the numerical analysis can be used in analyzing the time-dependent (viscous) behaviour of masonry prisms (e.g. Cividini et al., 1993) or in obtaining an insight into the possible causes of observed severe damages in ancient structures (e.g. Cividini et al., 1998), or in comparing the effectiveness of various alternative solutions applicable in the field for the restoration of buildings that characterize monument sites. The geotechnical laboratory investigation referred in this section was carried out in the framework of a research project concerning the historical buildings of Aliano (Basilicata, Italy) (Liberatore et al., 2006). Adobe reinforced with plant fibers was recovered from a collapsed wall and used in the laboratory investigation, aimed at identifying the material and at evaluating its shear resistance. The identification was carried out with the support of the grain size distribution curve and of the consistency limits, while the second aspect was investigated by means of direct shear (DS) tests. The geotechnical laboratory testing was carried out in general accordance with ASTM D2487 (2006), ASTM D3080 (2004) and BS1377-7 (1990). The determination of the particle size distribution indicates that adobe is silty sand, since sand is the predominant fraction (62%), with a significant silt fraction (about 30%). The plasticity of the finer part of Aliano adobe was investigated by determining the water content associated with the transition from solid to semi-plastic solid state, then to plastic state and finally to the liquid state (for instance, Whitlow 1988). The values of the three water contents (usually referred to as consistency, or Atterberg, limits) are respectively: shrinkage limit SL=17, plastic limit PL=17 and liquid limit LL=20. On the basis of the liquid limit and of the plasticity index PI=LL−PL, the plasticity chart was used to establish the sub -group of the fine-graded part of the adobe material and, in the present case, it is confirmed that the fine part is silt of low plasticity (with symbol ML). Due to the limited amount of available material and observing that in the field the adobe block walls are lined with an almost impervious facing, the testing was aimed only at evaluating the shear resistance properties of the dry adobe material. For testing dry samples the direct shear apparatus seems preferable to other conventional device (as for instance the triaxial one), since it allows the evaluation of the specimen volume variation. This in turn permits to assess the dilatancy properties and consequently one of the material parameters required in various nonlinear stress analyses. 2.2. Direct shear tests 2. 'Natural' samples of adobe material 2.1. Geotechnical identification and classification