PSI - Issue 42

Andrea Zanichelli et al. / Procedia Structural Integrity 42 (2022) 118–124 Author name / Structural Integrity Procedia 00 (2022) 000 – 000

119

2

1. Introduction Nowadays, environmental-friendly materials are increasingly attracting the interest of many researchers. Among them, earthen construction materials are spreading out in the field of civil engineering, due to the need of finding more sustainable building techniques, which could represent eco-friendly alternatives for the construction industry. As a matter of fact, the construction industry is responsible for about 30% of world carbon dioxide emissions and consumes more raw materials than any other sector (Pacheco-Torgal and Jalali (2012)). In this context, earth building techniques guarantee low carbon dioxide emissions and very low pollution impacts. As a matter of fact, earth is cheap, locally available in large quantities and can be obtained directly from the excavation of foundations, avoiding the transportation costs. Low energy is also required for these building techniques and, after demolition, raw earth is completely recyclable and reusable (Losini et al. (2022)). Moreover, earthen construction materials are characterised by both low thermal conductivity, maintaining a comfort temperature inside the buildings, and high hygroscopic properties, regulating the building relative humidity (Vyncke et al. (2018), Anglade et al. (2022)). Despite the above advantages, earth constructions are characterised by lower mechanical strengths and durability with respect to those made with traditional construction materials (Kebao and Kagi (2012)). Therefore, in order to endorse the use of such materials in advanced engineering applications, a deep characterisation is needed and, in such a context, numerical models can represent a worthwhile tool. Indeed, the strength and durability design must consider the typical damage phenomena occurring in such materials under in-service loading. The degrading effects, mainly related to matrix deformation or cracking, must be properly taken into account by a suitable numerical model. The present research work deals with the numerical simulation of an experimental campaign carried out on the shot-earth 772 by Vantadori et al. (2022a) at the University of Parma. The numerical analyses are performed by employing a numerical model (Scorza (2015), Scorza et al. (2022)) proposed by some of the present authors which allows to simulate both flexural and fracture tests on the shot-earth considered. Then, the Modified Two-Parameter Model (Vandatori et al. (2018), Vantadori et al. (2021), Zanichelli et al. (2018)) is used in order to compute the fracture toughness, from both the experimental fracture tests and the corresponding numerical simulations. The paper is structured as follows. The main features of the numerical model employed are summarized in Section 2. The experimental campaign analysed is briefly described in Section 3. In Section 4 the results determined by employing the above numerical model are discussed and compared with the experimental data. Finally, some conclusions are drawn in Section 5. 2. FE numerical model The numerical model here employed has been implemented in a non-linear 2D FE homemade code developed in standard Fortran language (Scorza (2015), Scorza et al. (2022)). After the definition of the input data, the code framework may be summarised in the following steps: • For each load step, the stiffness matrix is assembled; • After the definition of the nodal load increment vector, the current unknown displacement increment vector is determined; • The stress and strain fields are computed according to the material constitutive law; • Both stress and strain values are used in each finite element to check if the condition of matrix cracking is reached; • if the matrix cracking condition is reached, both stress and strain fields are updated, and the vectors of the internal and unbalanced forces are written; • the convergence at the current load step is checked in terms of incremental displacement norm and unbalanced forces and, if the convergence conditions are satisfied, the further load increment follows; • All the above steps are then repeated until the final load increment is reached.