PSI - Issue 10

I. Karatasios et al. / Procedia Structural Integrity 10 (2018) 211–218

212

I. Karatasios et al. / Structural Integrity Procedia 00 (2018) 000 – 000

2

a weakness zone for the composite mechanical behavior of stone units (Bosiljkov et al. (2003); Vasconcelos and Lourenço (2009)). The above process becomes more complex in the case of archaeological monuments since different types of stone blocks used in the structures may present distinct mechanical properties even if there is an internal homogeneity (e.g. marble). It is apparent that the situation becomes even more complicated when stone is heterogenous (e.g. sandstone, marls etc.), depending on their lithology and the subsequent microstructure. Theoretically, the properties of stone and mortar should be fully matched. However, in practice this is not desirable, since the join mortar must exhibit the highest possible adhesion to the stone surface and at the same time lower (tensile) strength than the stone, in order to secure that any potential failure will take place at the mortar mass or at the stone-mortar interface, rather than on the stone body. These requirements lead to several studies on the complex behavior of stone-mortar interface under mechanical load (Lindqvist (2009); Isebaert et al. (2014); Szemerey-Kiss and Torok (2017)), especially in the cases of masonry mortars. The development and selection of compatible mortars for joining archaeological stone fragments requires therefore an in-depth study of mechanical and microstructural properties of both stone substrate and mortar mixtures. For mortars, the main selection criteria (Szemerey-Kiss and Torok (2017)) encompass performance characteristics such as: enhanced adhesion and lower strength than the stone substrate for avoiding failure of original stone, elevated surface area values for enhancing adhesion potential, good workability and rheology for ensuring easiness in application as well as, sort setting and hardening time. Moreover, other critical parameters for successful joining interventions include the pre-treatment of the surfaces before mortar application (Courard et al. (2014)) as well as prevention of crack development during setting through proper maintenance of the humidity and temperature. Although there are several norms and standards for testing individual mortar properties, the methodology for evalu ating the adhesion strength on stone-mortar interface is not clear. For instance, the use of pull-off tests for evaluating adhesion strength has been proposed (Ramos et al. (2012); Szemerey-Kiss and Torok (2017)), while compressive and shear tests has been suggested by Vasconcelos and Lourenço (2009). In practice the basic failure modes that occur at the level of stone-mortar interface are two (Vasconcelos and Lourenco (2009)): tensile failure (mode I) associated to stresses acting to joints, and shear failure (mode II) corresponding to a sliding mechanism of the mortar joint. In prac tice, the predominant or the combination of different failure modes is related to the orientation of the principal stresses. This work is considering the effect of different binder combinations in the adhesion strength of restoration mortars for joining fragments of archaeological stones of varying mechanical properties. The development of compatible repair mortars was based on the interpretation of mechanical properties of both individual stone and mortar specimens subjected to unconfined compression and three-point flexural bending test as well as, on joined stone specimens with different types of mortar mixtures under four-point flexural bending test. In the later case, the joins were parallel to the direction of the load applied. Within this broader project context this study had three specific aims:  To investigate the range of compressive, flexural and adhesion strengths resulted by different hydraulic binders.  To identify the compatible combinations between hydraulic mixtures and different types of natural stones, according to the compressive and bending strengths achieved at 28-days.  To investigate the failure mode of mortar-stone joins and conclude on the adhesive strength and compatibility of restoration mortars with various types of soft marls. This paper presents the methodological approach followed for designing repair mortars for joining stone fragments without reinforcement. The evaluation of their compatibility and adhesion performance were studied in stone lithotypes of the same geological formation - soft marly limestones, similar to those found in the archaeological sites of Piraeus, Greece. Overall, the work focuses on the experimental evaluation of the adhesion strength between stone and repair mortar interface, when different types of mortar mixtures are used.

2. Materials and methods

2.1. Stone specimens

Eight (8) different representative stone samples were collected from the wider area of Piraeus, being representative of the different forms and layers of the so- called “Marl of Piraeus” . Petrographic examination of thin-sections was carried out under petrographic microscope, in crossed and parallel Nichols (EN 12407 (2007); Ingham (2013)), in order to describe their fabric, microstructure and mineralogy. Their physical properties, such as water absorption co-