Issue 40

Z.S. Metaxa et al, Frattura ed Integrità Strutturale, 40 (2017) 61-73; DOI: 10.3221/IGF-ESIS.40.06 62 I NTRODUCTION he ad-hoc accurate and thorough knowledge of the mechanical behaviour of building materials used in the restoration of Cultural Heritage Monuments is fundamental for the protection of their structural integrity. However, most of the traditional techniques used in the laboratory and in the field, such as strain-gauges, dial- gauges, extensometers, and Linear Variable Differential Transformers (LVDTs), provide data drawn from the materials’ external surface. Therefore, gathering information about the internal events, for example local failures and micro-fracture, which precede those detected on the materials’ surface, is challenging. The technological need to mine data from the interior of the restored joints and specimens were reported by Kourkoulis et al [1]. In a series of articles, e.g. [2-4] the important role of the interphases of the connecting materials used in the interior of the restored joints are noted. Additionally, several researchers underlined the important aspects of straining of the connecting members of the joint that are well in the interior of the restored Monument of Cultural Heritage, e.g. [5-7]. To cope with this problem, innovative measuring techniques should be used in parallel with the traditional ones for calibrating purposes. An already mature technique for damage monitoring of cement-based materials is the embedded glass fiber optical sensors. In the areas of local strain change, loss of the transmitted light signal [8] occurs that corresponds to damage within the monitoring region [9]. The detection can be performed with small embedded Bragg grating sensors with nominal diameter of approximate 120 μm, e.g. [10-12]. Nevertheless, this technique has several limitations, e.g. monitoring of the matrix transverse cracking. In addition, a dense network of optical fibers would be needed to fully monitor the complete structure that is economically not sustainable. Acoustic emission is another technique to monitor damage development in cement-based materials. Initiation of damage mechanisms (tension or shear) induce different crack tip motions, resulting in quite different AE characteristics. These distinct acoustic events can therefore be directly linked to a specific type of failure or can be evaluated in a cumulative manner to characterize the state of damage of the matrix, e.g. [13, 14]. Despite the numerous articles in this scientific field, the provided AE data are usually not sufficient to solely characterize the structural health of the matrix and another technique is required to cross-plot the necessary information [15, 16]. In this study, the exploitation of different types of advanced fibers made from polyvinyl alcohol (PVA) reinforced with carbon nanotubes (CNTs) for sensing the mechanical performance of the cement mortar used for the restoration of Acropolis’ Parthenon will be investigated. These new innovative PVA-CNT fibers have small dimensions ( d = 40 to 60 μm) and exhibit excellent piezoresistive characteristics and ductility that exceeds 100 % elongation [17]. Manufacturing of the specific fibers can be done by a potentially scalable process already reported [18]. Previous research on epoxy resin composites reinforced with glass fibers showed that the aforementioned PVA-CNT microfibers can be successfully embedded in the composite [19] and can be used for strain/damage monitoring purposes of non-conductive composites under tensile and bending loading tests [20, 21]. E XPERIMENTAL PROCEDURE Manufacturing of prismatic cement mortar specimens hite cement, Portland type, with the code name AALBORG WHITE and class CEM I 52,5R was used in the present study. The specific cement is manufactured from exceptional pure limestone and fine-grain quartz sand. The mineralogical phases of the white cement used as well as several characteristics of the cement can be seen in Tabs. 1 and 2, respectively. Two different sand types were used: (a) coarse grained quartz sand from 1 to 2 mm as well as (b) fine-grain quartz sand Μ32 with average grain size of 260 mm. The selection of the above materials has been performed according to several criteria, extensively discussed in [4]. C3S C2S C3A C4AF 77 (wt%) 16 (wt%) 5 (wt%) 1 (wt%) Table 1 : Mineralogical phases of white cement. In the present work, the following kinds of sensors were used: (a) 2 different types of CNT reinforced PVA fibers (PVA- CNT fibers). The first type will be called in the following as “coated PVA-CNT” fiber, while the second one as “coated and annealed PVA-CNT” fiber, that had different process parameters as discussed in the following. This kind of sensors T W