Experimental evaluation of the shear strength of aerial lime mortar brickwork by standard tests on triplets and non-standard tests on core samples

1 Experimental evaluation of the shear strength of aerial lime mortar brickwork by standard tests on triplets and non-standard tests on core samples Luca Pelà*, Konstantina Kasioumi, Pere Roca Department of Civil and Environmental Engineering, Universitat Politècnica de Catalunya (UPCBarcelonaTech), Jordi Girona 1-3, 08034 Barcelona, Spain * Corresponding author. Phone: +34 934011036 Fax: +34 934054135 [email protected] Abstract – This paper presents the results of an experimental program carried out in the laboratory to evaluate the shear strength of aerial lime mortar brickwork. Masonry triplets and walls were tested after one year from their construction by adopting two different testing methods. The first approach consisted in the shear tests of masonry triplets, whereas the second technique was based on core drilling from walls of 90 mm diameter cylindrical specimens to be subjected to Brazilian tests with varying inclination of the diametric mortar joint. The first method is more adequate to characterize new masonry, whilst the second one is a suitable MDT procedure for the analysis of existing structures. The experimental results from standard and non-standard tests were properly investigated in order to obtain the shear failure envelope of the bond interface and mortar joint. The comparisons between the different tests and their interpretative theories show the possibilities of the novel non-standard testing method for the evaluation of the shear strength of structures of the built cultural heritage. Keywords: Masonry; Triplets; Coring; Double Punch Test; Brazilian Test; Minor Destructive Testing (MDT); Mohr-Coulomb Criterion; Friction; Cohesion; Bond Interface. Abbreviations - MDT: Minor Destructive Testing; LVDT: Linear Variable Differential Transformer; DPT: Double Punch Test. 1. Introduction The mechanical characterisation of existing masonry structures faces significant difficulties due to the intrinsic complexity of masonry as a composite material, the large heterogeneity normally associated to masonry construction and the dependence of the quality of masonry with the construction practice. An adequate characterisation is also hindered by the insufficient standardization available for the mechanical testing of existing masonry materials and structural members. 2 Nevertheless, a sufficient characterisation of the main resisting parameters is essential for a reliable structural verification of masonry structures. The determination of the compressive strength of existing brickwork is necessary to assess the current safety level of structures of the built heritage subjected to gravitational loads and, more specifically, to future loads caused by changes of use. In turn, the investigation of the shear strength is of paramount importance to evaluate the potential capacity of existing structures against horizontal loads induced by wind or earthquake. The recent earthquakes experienced in the Mediterranean basin, and especially in Italy, have motivated an increasing research effort on the validation of experimental procedures for the investigation of the strength capacity of existing masonry. Solid clay bricks and lime mortar joints constitutes a typical material combination in historical masonry structures. However, only a limited number of comprehensive experimental programs is available in the literature about the mechanical characterisation of pure lime mortars in the laboratory [1,2]. One of the main reasons explaining such limited available experience may be found in the difficulties related to the long curing and hardening time required by these materials and especially by aerial lime mortar. Recent experimental researches were carried out in the laboratory using pure lime mortar brickwork in order to evaluate its compressive behaviour [3,4]. Masonry built with mixed cement-lime mortar has deserved more significant attention [5–7] than that prepared with lime mortar. More specifically, the contributions available in the literature on the laboratory characterisation of the shear behaviour of pure lime mortar brickwork [8,9] are definitely more scarce than those focusing on masonry with lime-cement mortar or cement mortar [10,11]. In these works, shear tests on triplets or couplets were mainly investigated in order to obtain a shear failure envelope for the joints and the bond interface. The triplet test is established by the EN 1052-3:2002 [12] and is suitable for the shear characterisation of new masonry. The application of the triplet test to historical structures is not possible due to the impossibility found in extracting this type of sample from existing brickwork. The shear behaviour of existing masonry can be directly evaluated by the in-situ diagonal compression test according to ASTM 509-2010 [13] and RILEM LUM B6 [14]. Several contributions are available in the literature about this experimental technique [15,16] that was also compared with the triplet test in the laboratory [17]. However, this type of experiment is destructive and can be hardly applied to historical buildings. Recent construction standards like the Italian ones [18,19] offer a practical alternative by providing some pre-calculated reference values that may be used by professionals depending on the type of constituent materials and the level of knowledge acquired about the specific case of study through inspection activities. The extraction of sufficiently large samples from existing structures, allowing the application of conventional laboratory mechanical tests, is hampered by the intrinsic geometrical characteristics of 3 the masonry texture, the fragility of the investigated material and the difficulty in bringing to the laboratory intact specimens after the sampling and transportation operations. In the case of historical structures belonging to the built cultural heritage, additional difficulties emerge due the need to minimize the damage caused by the inspection. In this case, the investigation of the mechanical properties must be done in a respectful way so that the damage is limited to an indispensable minimum. This allows the inspected structure to be easily repaired whereas the cost of the minor signs due the inspection is largely surpassed by the benefit provided by the allowed gain of knowledge about the material properties. Following this approach, Minor Destructive Testing (MDT) seems an interesting possibility nowadays, since it is based on in-situ sampling of only small specimens that can be tested in the laboratory afterwards. Among the MDT techniques aimed at the assessment of the shear strength of historical masonry, an approach experiencing rising interest is the extraction of small core samples (typically with 70÷110 mm diameter) to be subjected to Brazilian tests with variable inclinations of the diametric mortar joint interposed between two cylindrical segments of brick. This method was proposed firstly by considering a fixed 45º inclination of the diametric mortar joint during the Brazilian test [20,21]. This preliminary approach was aimed at obtaining an empirical correlation between the results from Brazilian tests on core samples and diagonal compression tests on walls. Recent works proposed a novel comprehensive methodology combining different Brazilian tests with varying inclinations of the diametric mortar joint [22–24]. This method was applied to single leaf walls with half brick thickness, by obtaining core samples with length equal to the thickness of the wall, and also to single leaf walls with one brick thickness. In this second case, as well as in the case of thicker structural members (e.g. two or three bricks thick), the long cylinders extracted from the whole thickness of the wall may present internal vertical joints and thus their transversal cutting is necessary in order to obtain the suitable specimens. In this case, the cut can provide adequate specimens composed of two solid brick cylindrical segments and one diametric mortar joint. The referred researches investigated the behaviour of masonry made of clay bricks and hydraulic lime mortar with low strength, i.e. a frequent material combination in historical materials. The Brazilian tests on core samples showed the shear failure of mortar for specimens with 45º to 60º inclination of the diametric mortar joint. Other researches [20,25] focused on masonry with lime-cement or cement mortar characterised by rather high strengths and thus the results could be considered more representative of masonry of more recent construction. In particular, Brazilian tests on core samples with very strong cement mortar exhibited splitting of the brick and thus the approach was not suitable to provide further evaluations of the shear strength of the mortar. 4 This paper constitutes a novel contribution focusing on the shear behaviour of aerial lime mortar brickwork. Despite the aforementioned difficulties in building and testing this type of material in the laboratory, the experimental program presented herein considered masonry specimens tested one year after their construction. The research presents a comparative study considering both standard shear tests on triplets and non-standard tests on core samples. The triplets were built in the laboratory, as well as masonry walls replicating historical aerial lime mortar brickwork. In the second case, the cylindrical samples were extracted by simulating in the laboratory the in-situ core drilling of an existing structural member. The methodology proposed in recent works by the authors [22–24] is followed since it can efficiently combine the results from Brazilian tests on cylindrical specimens with variable inclination of the diametric mortar joint and the results from double punch test (DPT) of mortar joints extracted from the brickwork. This approach combines the results from different experimental tests into the same interpretative theory. The experimental results on the shear behaviour obtained using the non-standard procedure is compared with the outcomes from standard triplet tests with the same age and kind of aerial lime mortar brickwork. The study provides direct relationships between standard and non-standard tests that may result helpful in works dealing with the material characterisation of structures of the historical heritage built with low-strength masonry. 2. Experimental program The experimental campaign was carried out at the Laboratory of Technology of Structures and Construction Materials of the Technical University of Catalonia (UPC-BarcelonaTech). This section presents information about the material components, the construction of the specimens and the testing setups. 2.1 Material components This experimental program considered material components intended to reproduce one of the most frequent material combinations in historical masonry, i.e. solid clay bricks and lime mortar joints. Handmade terracotta bricks and pure aerial lime mortar CL90 (without cement) [26] were used for this purpose. The mechanical properties of the material components were assessed on prismatic specimens in compliance with the available technical standards. According to EN 1015-11:2007 [2], prismatic samples with dimensions 160 × 40 × 40 mm3 were prepared to evaluate the strength of CL90 mortar. Flexural strength ffm was evaluated on four prismatic specimens (Figure 1a), whereas the compressive strength fcm was assessed on the eight halves produced 5 by the splitting of the prisms from the four flexure tests. Those halves were tested in compression under 40 × 40 mm2 steel platens (Figure 1c). Both flexure and compression tests were carried out using a load cell of 10 kN in load control, but with different loading rates of 10 N/s and 50 N/s, respectively. Table 1 summarizes the experimental results that were obtained after one year from mortar pouring. The average values are ffm = 0.55 MPa and fcm = 1.63 MPa and their coefficients of variation (CV) are 9.5% and 5.3%. The average values of flexural and compressive strengths fall within the typical ranges of these properties for historical lime mortar. The nominal dimensions of the terracotta bricks were 305 × 145 × 45 mm3, even though the effective dimensions of each unit could slightly vary due to the handmade type of construction. Six units were tested under compression according to the EN 772-1:2011 [27], using a load cell of 3000 kN. The average compressive strength was fcb,u = 30.7 MPa with CV = 5.6%. Besides the standard compression tests on whole units, compression tests on prismatic samples cut from the bricks were carried out by following the EN 1015-11:2007 for mortar [2]. Sixteen brick prisms, with dimensions 160 × 40 × 40 mm3, provided an average flexural strength ffb,x = 3.63 MPa (CV = 3.7%) along the stretcher direction and ffb,y = 3.79 MPa (CV = 13.1%) along the header direction (Figure 1b). Six brick cubes of about 35 × 35 × 35 mm3 (Figure 1d) provided an average compressive strength fcb,c = 18.4 MPa with CV = 5.9% in the vertical direction. The cubic samples provided a mean compressive strength lower than that obtained from the whole units. This result was due to the different shapes of the two tested specimens, since the brick units had aspect ratios much higher than those of the brick cubes. As a consequence, the units presented a larger surface in contact with the steel loading platens of the compression machine. This fact generated a higher horizontal confinement effect due to the contact friction. The more pronounced triaxial compressive stress state inside the units produced higher values of the maximum vertical compression force than in cubic samples. Table 2 summarizes the experimental results that were obtained from brick samples. 6 Figure 1 – Testing of prismatic samples of material components: flexure and compression tests on mortar (a,c) and brick (b,d). 2.2 Construction of masonry specimens The materials described in the previous section were employed to build two walls in stretcher bond with dimensions 1605 × 870 × 145 mm3 (Figure 2a). The two walls reproduced portions of structural members that may be encountered in typical historical buildings made of low-strength masonry. They were used in this research to simulate, under laboratory controlled conditions, a generic in-situ sampling activity consisting in the core-drilling of 90 mm diameter cylindrical specimens. Each wall was built in-between two steel C beams that were placed below the base and over the top of the specimen. The beams were interconnected by four low-tensioned tie rods. In this way, a very small pre-compression was applied to the walls with the sole purpose of guaranteeing their easy transportation inside the laboratory, as well as their sufficient stability during the coring operations. 7 At the same time of the construction of the two walls, nine triplets were also built using the same materials, having dimensions 305 × 145 × 165 mm3 and including three units and two mortar joints (Figure 2b). Figure 2 – Prismatic specimens of masonry built in the laboratory: a) walls and b) triplets. After one year from the construction, the two walls were taken out of their storage to carry out the extraction of the cylindrical samples (Figure 3a). The core drilling followed a dry procedure that was proposed in previous works by the authors [4,24]. The absence of water is a remarkable advantage of this improved sampling technique, since it minimizes the risk of disjointing the cylindrical specimens. The wet core drilling, commonly adopted for concrete structures, is not fully adequate for existing masonry buildings since the water may wash the mortar joints away and spoil the samples. On the other hand, the dry core drilling uses an aspirator to cool the bit and to remove the dust. This important feature allows the optimization of the sampling procedure by increasing the number of successful extractions of cores. This is very important in the study of historical constructions, where the minimization of the impact of the inspection on the existing structure is mandatory. As soon as they were extracted, the masonry cylinders were properly confined by strong adhesive tape stuck on the lateral curved surface to avoid any damage during their subsequent storing and transportation. Figure 3b shows the wall after the core-drilling. As it can be noticed, the cut of the bit is precise and does not damage the surrounding material. The procedure proves to be remarkably clean thanks to the absence of dust that is completely drawn by the vacuum cleaner conveniently connected to the drilling 8 equipment. This feature of the technique, together with the total absence of water, makes the dry core drilling a suitable sampling method, even for the extraction from the interior of existing buildings. Masonry core samples with 90 mm diameter were extracted for this research (Figure 3c). They included a diametric mortar joint interposed between two cylindrical segments of brick. Additional 150 mm diameter cores were extracted to be used in another experimental study. Thirty-one core samples with one diametric mortar joint were extracted from the two walls. Although some of them disjoined during the core drilling, fifteen undamaged samples could be successfully assigned to the Brazilian tests. 9 Figure 3 – Extraction of the cylindrical specimens from masonry walls: a) core drilling, b) wall at the end of the extraction and c) 90 mm diameter cylinders for Brazilian tests. 2.3 Testing setups Both core samples and triplets were tested one year after their construction [28]. Nine triplets were considered first to determine the shear strength of horizontal bed joints in a conventional manner according to the standard EN 1052-3:2002 [12]. All the specimens were tested in a machine with a load cell of 200 kN to apply the shear force to the centre brick. The experimental setup is shown in Figure 4a. The force normal to the bed joints was applied by a 10 kN loading jack. Three triplets were tested for each level of normal stress, i.e. 0.3 MPa, 0.6 MPa and 1.0 MPa. Each normal stress level was kept constant during the test, whilst the increasing shear force was applied to the mortar joints of the triplet. Four Linear Variable Differential Transformers (LVDTs) with ±5 mm range and 5 μm precision were used to control the tangential displacements in the mortar joints during the shear tests. Fifteen core samples were subjected to Brazilian tests with variable inclinations of the diametric mortar joint. Due to the novelty of this kind of test, there are not specific standards. The only standard related, to some extent, with the tests carried out is the EN 12390-6:2000 [29] about the splitting test of concrete specimens. Hence, the experimental setup was the same adopted in previous works by the authors [22–24]. The core samples were laid horizontally between the loading platens. Two wooden stripes were inserted between the platens and the specimen, in order to distribute correctly the diametric longitudinal load. The specimen was compressed along two opposite longitudinal generators until failure. The test was carried out under displacement control at a rate of 0.005 mm/s. The core samples were tested for different inclinations α of the diametric mortar joint with respect to the horizontal direction, as shown in Figure 4b. This peculiar experimental setup allows the application of a combined compression-shear load over the mortar joint of the core samples. The magnitudes of the shear and compression stresses acting on the mortar joint during the test is determined by the angle α of inclination of the diametric mortar joint with respect to the horizontal direction, see Section 3.2. Increasing values of α correspond to increasing shear and decreasing normal compression on the mortar during the test. A crucial point before the execution of the experiments was the decision of the inclinations of the diametric mortar joints for the different Brazilian tests. The range of inclinations must be carefully chosen in order to avoid the vertical splitting of the core sample and to trigger the frictional failure in the mortar [23]. Since the mortar had rather low strength, the Brazilian tests were cautiously carried out with lower inclinations than in previous studies [22–24], i.e. α = 40º, α = 45º and α = 50º. Five core samples were assigned to each one of these inclinations. 16 tanc τ σφ = − (4) where c is the cohesion of mortar and φ is its internal friction angle. According to the Mohr-Coulomb’s theory, the material reaches the failure when the Mohr’s circle representing its stress state becomes tangent to the lines given by Equation 4. Given a set of several experimental results, together with their corresponding Mohr’s circles built as shown in Figures 7a-b, the values c and φ of the overall failure envelope are those that minimize the sum of the squares of the distances between the failure envelope and the Mohr’s circles. Figure 7c shows the failure envelope obtained thanks to the Continuum Model applied to the experimental results from double punch tests and Brazilian tests. The Mohr-Columb parameters result φ = 26.35º and c = 0.32 MPa. According to Mohr-Coulomb’s theory, it is possible to evaluate the corresponding values of the compressive and tensile strength of mortar fcm and ftm using the two following expressions: 2 cos 2 cos 1 sin 1 sin cm tm cc ff φφ φφ = = −+ (5a,b) These equations provide fcm = 1.02 MPa and ftm = 0.39 MPa. The first value provides a good estimation of the average compressive strength of mortar obtained from the double punch tests (fcm,DPT = 0.91 MPa, see Table 6). The tensile strength derived from Mohr-Columb theory results correctly lower than the flexural strength obtained from tests on prismatic specimens (ffm = 0.55 MPa, see Table 1). 17 Figure 7 – Interpretation of Brazilian tests on core samples and double punch tests on mortar joints: construction of Mohr’s circle representing the stress state of mortar at failure a) in Brazilian tests and b) in double punch tests; c) failure envelope obtained by the continuum model ( φ = 26.35º and c = 0.32 MPa). 4. Discussion Both the standard shear tests on triplets and the non-standard Brazilian tests on core samples with inclined diametric mortar joint (complemented with DPT results) provided the shear failure envelopes of the structure composed of the mortar joint and the bond interface. These envelopes were derived from the application of either the interface model or the continuum model. From a theoretical point of view, the interface model seems more compatible with the bond failure at the unit-mortar interface, whereas the continuum model seems adequate to describe the shear failure in the mortar material filling the joint. 18 In the shear tests on triplets the failure mechanism was the sliding at the bond interface, proving the appropriateness of processing the experimental results through the interface model. The MohrCoulomb parameters resulted Φ = 35.65° and C = 0.04 MPa. On the other hand, the non-standard Brazilian tests exhibited two different types of failures, i.e. either the sliding along the bond interface or the parasymmetric mechanism with shear failure of the mortar material inside the joint. For the interpretation of the non-standard tests that resulted heterogeneous as for the experimental failure mechanisms, the authors have chosen to apply both the interface and continuum models to the whole sample of experimental results. Then, the results from the two different approaches have been critically compared. From a theoretical point of view, it might seem more rigorous to assign the results with sliding failure to the interface model and those with parasymmetric failure to the continuum model. However, this approach could result cumbersome since it would reduce the experimental redundancy of tests assigned to a specific interpretative theory. Redundancy of experimental results is indeed extremely important in experimental campaigns on historical masonry, where the number of available samples is usually limited and the scatterings of the experimental results may result even higher than those observed in this experimental program performed under controlled laboratory conditions. For all these reasons, a better option seems the consideration of both the interface and the continuum models to be applied separately to the whole set of experimental results and then critically compared. Actually, and due to the fact that some of the failures obtained involve both mortar cracking and interface sliding, the real physical phenomenon is a mixed condition between those represented by the two aforementioned insights. In this experimental program, the interface model applied to the Brazilian tests provided Φ = 35.86° and C = 0.09 MPa, i.e. Mohr-Coulomb parameters quite close to those obtained from triplet tests. The dimensions of the core samples are smaller than those of triplets and this fact produces inevitably a higher experimental scattering since the Brazilian test results depend more on the local conditions of the material components in the small specimen. The coefficients of determination results R2 = 0.97 for standard shear tests triplets and R2 = 0.92 for Brazilian tests on core samples. The discrepancy between the test on triplets and the Brazilian tests was expected due to the different testing setups and natures of the samples and in any case it shows that the interface model can be acceptably applied to both the standard and non-standard shear tests. On the other hand, the continuum model applied to the Brazilian tests and the DPTs provided φ = 26.35º and c = 0.32 MPa, that are parameters obviously completely different than those of the interface model since they were derived from a different failure theory based on continuum mechanics. 19 However, when these parameters are inserted into Equations 5a-b they provide compressive and tensile strengths in remarkable agreement with those obtained from direct tests on mortar. On the basis of these observations, it is highly advisable in practical applications to compare different testing techniques and interpretative models so as to characterize the mechanical properties of existing materials. 5. Conclusions This research has focused on the evaluation of the shear strength of aerial lime mortar brickwork. This kind of material has so far deserved very little attention in the existing literature. The limited research on the mechanical properties of aerial lime mortar is due to the practical difficulties caused by its long curing and hardening time. The experimental program presented in this research has considered masonry samples built in the laboratory and then tested one year after their construction. The research has analysed two different testing techniques to evaluate the shear strength of aerial lime mortar brickwork. The first one is the triplet test established by the EN 1052-3:2002 [12]. Being a standard procedure, this technique presents a straightforward interpretation of the experimental results and can be easily carried out for the mechanical characterisation of new masonry. However, this approach cannot be applied to the investigation of the shear strength of existing masonry structures, since it is practically impossible to extract this type of sample from the existing structural member. For this reason, the present study has investigated a second testing technique based on the in-situ core drilling and subsequent testing of masonry cylindrical specimens in the laboratory. Masonry walls were built to reproduce historical masonry members and then they were core drilled to extract cylindrical specimens with 90 mm diameter. These core samples were subjected to different Brazilian tests with varying inclinations of the diametric mortar joint. This non-standard MDT procedure can be easily applied to existing structures by combining several advantages like low invasivity, easy sampling and limited costs, as demonstrated by recent works [22–24]. Both standard and non-standard testing methods have been investigated on the same materials and under controlled laboratory conditions in order to assess and validate their applicability. The following conclusions can be drawn at the end of the experimental work: - The comparative study of standard and non-standard shear tests has allowed the evaluation of the strength of existing masonry through a novel MDT technique based on the extraction of small size core samples. This approach can be considered a suitable alternative to more 20 destructive approaches, such as the in situ diagonal compression test, aimed at the mechanical characterisation of the shear strength of existing masonry. - The Brazilian tests on core samples have shown their utility for the estimation of the material shear strength. Two different interpretation theories have been presented, called interface and continuum models. They correspond respectively to a purely frictional sliding model over the mortar-unit bond interface and a Mohr-Coulomb failure criterion for the mortar. The first interpretation considers only the results from the Brazilian tests, whereas the second approach considers both the Brazilian tests and DPT results of mortar joints extracted from the walls. This second option shows to be appropriate when it is required a combined interpretation of experimental results from different testing methods, as usual in the analysis of existing structures. - Due to the fact that some of the failures involve mainly mortar cracking and others mainly interface sliding, the ultimate mechanism in Brazilian tests is not unique and both the interpretative theories have been applied and critically compared. The interface model has provided a shear failure envelope in good agreement with that derived from standard tests on triplets. The parameters obtained from the continuum model based on Mohr-Coulomb theory, on the other hand, have provided strength values in very good agreement with those previously obtained from direct tests on mortar samples. - The comparison between the standard and non-standard testing techniques has revealed to be of high importance in order to deepen the understanding of the MDT procedures for the characterisation of the shear strength of existing brickwork. The investigation of masonry composed of solid clay bricks and aerial lime mortar joint has provided insight on the performance of the materials typically found in historical heritage structures. Acknowledgements The authors gratefully acknowledge the financial support from the MINECO (Ministerio de Economia y Competitividad of the Spanish Government) and the ERDF (European Regional Development Fund) through the MULTIMAS project (Multiscale techniques for the experimental and numerical analysis of the reliability of masonry structures, ref. num. BIA2015-63882-P). The authors would like to thank Elisa Canella for her helpful contribution. References [1] Baronio G, Binda L, Saisi A. Analisi di Malte Antiche e Comportamento di Malte Riprodotte in Laboratorio (in Italian). 7° Convegno Naz. - L’ingegneria Sismica Ital., Siena, Italy: 1995, p. 267–76. 21 [2] Baronio G, Binda L, Saisi A. Mechanical and Physical Behaviour of Lime Mortars Reproduced After the Characterisation of Historic Mortar. Int. RILEM Work. Hist. Mortars Charact. Tests, 1999, p. 307–25. [3] Drougkas A, Roca P, Molins C. Compressive strength and elasticity of pure lime mortar masonry. Mater Struct 2016;49:983–99. doi:10.1617/s11527-015-0553-2. [4] Pelà L, Canella E, Aprile A, Roca P. Compression test of masonry core samples extracted from existing brickwork. Constr Build Mater 2016;119:230–40. doi:10.1016/j.conbuildmat.2016.05.057. [5] Page AW. The Biaxial Compressive Strength of Brick Masonry. Proc Inst Civ Eng 1981;71:893–906. doi:10.1680/iicep.1981.1825. [6] Vermeltfoort AT. Compression properties of masonry and its components. 10th Int. Brick Block Mason. Conf., 1994, p. 1433–42. [7] Aprile A, Benedetti A, Grassucci F. Assessment of Cracking and Collapse for Old Brick Masonry Columns. J Struct Eng 2001;127:1427–35. doi:10.1061/(ASCE)0733-9445(2001)127:12(1427). [8] Binda L, Mirabella Roberti G, Tiraboschi C, Abbaneo S. Measuring Masonry Material Properties. In: Abrams D., Calvi GM, editors. U.S.-Italy Work. Guidel. Seism. Eval. Rehabil. Unreinforced Mason. Build., Pavia, Italy: 1994, p. 326–47. [9] Binda L, Tiraboschi C, Abbaneo S. Experimental research to characterise masonry materials. Mason Int 1997;10:92–101. [10] Van Der Pluijm R. Shear Behaviour of Bed Joints. Sixth North Am. Mason. Conf., Philadelphia, Pennsylvania (US): 1993, p. 125–36. [11] Lourenço PB, Barros JO, Oliveira JT. Shear testing of stack bonded masonry. Constr Build Mater 2004;18:125– 32. doi:10.1016/j.conbuildmat.2003.08.018. [12] CEN. EN 1052-3:2002 Methods of test for masonry - Part 3: Determination of initial shear strength 2002. [13] ASTM International. ASTM E519 / E519M-10, Standard Test Method for Diagonal Tension (Shear) in Masonry Assemblages 2010. doi:10.1520/E0519_E0519M-10. [14] RILEM. Technical recommendations for testing and use of constructions materials: LUM B6 Diagonal tensile strength tests of small wall specimens 1991:488–9. doi:10.1617/2351580117.152. [15] Corradi M, Borri A, Vignoli A. Experimental study on the determination of strength of masonry walls. Constr Build Mater 2003;17:325–37. doi:10.1016/S0950-0618(03)00007-2. [16] Brignola A, Frumento S, Lagomarsino S, Podestà S. Identification of Shear Parameters of Masonry Panels Through the In-Situ Diagonal Compression Test. Int J Archit Herit 2008;3:52–73. doi:10.1080/15583050802138634. [17] Alecci V, Fagone M, Rotunno T, De Stefano M. Shear strength of brick masonry walls assembled with different types of mortar. Constr Build Mater 2013;40:1038–45. doi:10.1016/j.conbuildmat.2012.11.107. [18] Decreto Ministeriale 14/01/2008. Nuove norme tecniche per le costruzioni (in Italian) 2008. [19] Circolare 2 febbraio 2009 n. 617. Istruzioni per l’applicazione delle Nuove norme tecniche per le costruzioni di cui al decreto ministeriale 14 gennaio 2008 (in Italian) 2009. [20] Braga F, Dolce M, Filardi B, Masi A, Nigro D. A test method to assess the shear strength of existing masonry structures - Theoretical basis and first experimental results. Int. Work. CNR-GNDT “Effectiveness Inject. Tech. Retrofit. Stone Brick Mason. Walls Seism. Areas,” Milan, Italy: 1992, p. 204–27. [21] Filardi B, Liberatore D, Masi A, Nigro D. Valutazione della Resistenza a Taglio di una Tipologia Muraria 22 Tramite Prove su Pannelli, Carote e Triplette (in Italian). La Mecc. delle Murature tra Teor. e Progett., Messina, Italy: 1996, p. 75–84. [22] Pelà L, Benedetti A, Marastoni D. Interpretation of Experimental Tests on Small Specimens of Historical Mortars. In: Jasieńko J, editor. Struct. Anal. Hist. Constr., Wrocław, Poland: DWE; 2012, p. 716–23. [23] Pelà L, Roca P, Benedetti A. Mechanical Characterization of Historical Masonry by Core Drilling and Testing of Cylindrical Samples. Int J Archit Herit 2016;10:360–74. doi:10.1080/15583058.2015.1077906. [24] Marastoni D, Pelà L, Benedetti A, Roca P. Combining Brazilian Tests on masonry cores and Double Punch Tests for the mechanical characterization of historical mortars. Constr Build Mater 2016;112:112–27. doi:10.1016/j.conbuildmat.2016.02.168. [25] Mazzotti C, Sassoni E, Pagliai G. Determination of shear strength of historic masonries by moderately destructive testing of masonry cores. Constr Build Mater 2014;54:421–31. doi:10.1016/j.conbuildmat.2013.12.039. [26] CEN. EN 459-1:2010 Building lime - Part 1: Definitions, specifications and conformity criteria 2010. [27] CEN. EN 772-1:2011 Methods of test for masonry units. Determination of compressive strength 2011. [28] Kasioumi K. Experimental Testing on Cylindrical Specimens of Aerial Lime Mortar Masonry. SAHC Master dissertation. Technical University of Catalonia, 2015. [29] CEN. EN 12390-6:2000 Testing hardened concrete - Part 6: Tensile splitting strength of test specimens. 2000:1– 8. [30] DIN 18555-9:1999. Testing of mortars containing mineral binders - Part 9: Hardened mortars. Determination of the mortar compressive strength in the bed joint 1999. [31] International Union of Railways. Leaflet 778-3R, Recommendations for the inspection, assessment and maintenance of masonry arch bridges 2011. [32] Henzel J, Karl S. Determination of Strength of Mortar in the Joints of Masonry by Compression Tests on Small Specimens. Darmstadt Concr 1987;2:123–36. [33] Marastoni D. Advanced Minor Destructive Testing for the Assessment of Existing Masonry 2016:PhD Thesis, University of Bologna. 23 Tables Table 1 – Experimental strengths of aerial lime mortar prisms Mortar CL90 Sample f fm (MPa) f cm (MPa) Prism 1 0.48 1.70 1.54 Prism 2 0.52 1.47 1.72 Prism 3 0.60 1.63 1.72 Prism 4 0.60 1.68 1.58 Average 0.55 1.63 CV % 9.5 5.3 Table 2 – Experimental strengths of bricks Brick Sample f cb,u (MPa) f cb,c (MPa) f fb,x (MPa) f fb,y (MPa) 1 31.2 19.6 3.81 3.81 2 29.4 19.5 3.67 3.53 3 30.4 19.1 3.60 3.57 4 30.3 18.3 3.52 3.74 5 34.2 16.6 3.51 3.19 6 28.9 17.5 3.59 3.32 7 3.88 4.40 8 3.48 4.73 Average 30.7 18.4 3.63 3.79 CV % 5.6 5.9 3.7 13.1 24 Table 3 – Experimental results of shear tests on triplets with σ p=0.3 MPa ID Before Test After Test Failure Mode F max (kN) τ max (MPa) TR1 27.73 0.32 TR2 19.50 0.22 TR4 20.75 0.24 Average 22.66 0.26 CV 16% 16% 25 Table 4 – Experimental results of shear tests on triplets with σ p=0.6 MPa ID Before Test After Test Failure Mode F max (kN) τ max (MPa) TR5 39.26 0.45 TR6 38.64 0.44 TR7 39.99 0.46 Average 39.30 0.45 CV 1% 1%