Age effect on the mechanical properties of natural hydraulic and aerial lime mortars

Construction and Building Materials 236 (2020) 117573

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Age effect on the mechanical properties of natural hydraulic and aerial lime mortars Lucía Garijo a, Xiaoxin Zhang a,b,⇑, Gonzalo Ruiz a, José J. Ortega a a b

ETS de Ingenieros de Caminos, Canales y Puertos, Universidad de Castilla-La Mancha, Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain EI Minera e Industrial de Almadén, Universidad de Castilla-La Mancha, Plaza Manuel Meca 1, Almadén, 13400 Ciudad Real, Spain

h i g h l i g h t s
Advanced mechanical characterization of two lime mortars in the long-term.
Measurement of the fracture energy and splitting tensile strength with age.
Fast increase of the mechanical properties in both lime mortars up to 56 days.
Empirical equations of age effect are proposed.

a r t i c l e

i n f o

Article history: Received 29 April 2019 Received in revised form 22 October 2019 Accepted 9 November 2019

Keywords: Natural hydraulic lime mortar Aerial lime mortar Long-term behavior Mechanical characterization Fracture energy Empirical equations

a b s t r a c t This paper focuses on the advanced mechanical characterization of natural hydraulic and aerial lime mortars presenting a lime/aggregate ratio of 1:3 and a water/lime ratio of 0.9. Seven properties, such as density, flexural, compressive and splitting tensile strengths, fracture energy, elastic modulus from prisms and carbonation depth by means of the phenolphthalein method, were measured at different moments. The results show that the greatest increase in the mechanical properties of both mortars occurs up to the first 56 days. This increase ranges between 60% and 90% of their corresponding value at an age of 448 days depending on the mechanical property and type of mortar. Between day 56 and day 224, there is a more moderate, but progressive evolution. However, after day 224, the evolution of the mechanical properties in light of their mean value, is very slow for the aerial lime mortar and shows a slight increase for the natural hydraulic mortar. Furthermore, various empirical equations of such behavior over time are proposed for both mortars. Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction In accordance to the Standard EN 459-1 [1], there are several types of hydraulic and aerial limes. The sub-classes of hydraulic limes are natural hydraulic lime (NHL), formulated lime (FL) and hydraulic lime (HL). The first is formed by burning argillaceous or siliceous limestones without additions. The second is mainly comprised of air lime and/or natural hydraulic lime with hydraulic and/or pozzolanic material. The last is formed by lime and other materials, such as cement, blast furnace slag, fly ash, limestone filler and other components in very small proportions [1]. There also exist various sub-families of aerial limes, and differentiations may be made between calcium lime (CL) and dolomitic ⇑ Corresponding author at: ETS de Ingenieros de Caminos, Canales y Puertos, Universidad de Castilla-La Mancha, Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain. E-mail address: [email protected] (X. Zhang). https://doi.org/10.1016/j.conbuildmat.2019.117573 0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.

lime (DL). The former is mainly composed of calcium oxide and/or calcium hydroxide while the latter mainly of calcium magnesium oxide and/or calcium magnesium hydroxide; both of which do not have any hydraulic or pozzolanic additions. Likewise, these two types of aerial limes may be in the form of quicklime (Q) or hydrated lime. Quicklime mainly exists in an oxide form and reacts exothermically with water; while hydrated lime exists mainly in a hydroxide form and is produced by slaking quicklime. Moreover, hydrated lime can be a powder (S), putty (S PL) or slurry or milk of lime (S ML). Moreover, mortars produced with aerial lime differ from mortars produced with natural hydraulic lime. Both are used extensively for restoration purposes [2–4]. Both can harden by the reaction of portlandite with carbon dioxide from the atmosphere, referred to as ‘‘carbonation”. Similarly, natural hydraulic limes also harden by hydration of their mineral compounds, that is, mainly by combining dicalcium silicates, C2S with water to produce C-S-H and portlandite [3,5].

2

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

Both lime mortars are mainly used to join blocks, bricks and stones in masonry walls [2,5–12]. Natural hydraulic lime mortars are adopted when early strength gain and a faster setting time are necessary [13] or when soluble salts or high humid conditions [14] require better behavior. Aerial lime mortars rather are also used as renders, or to bond material for decorative purposes, mosaics or ceramic tiles [7]. Mortars with pozzolanic properties made of lime have mostly been present in historic masonry structures over long periods of time. Nowadays, both aerial and natural hydraulic lime mortars are used to simulate ancient masonry mortars [2,12]. It is wellknown that the carbonation process of both mortars can be longlasting [4,15,16], and therefore, their strengths at early ages are frequently less than those corresponding to the long term. Thus, there is a need to quantify the corresponding increase of mechanical properties. In this regard, Drougkas et al. [2] measured the flexural and compressive strengths of an aerial mortar and a natural hydraulic lime mortar at different ages, from day 14 to day 49 for the former and from day 7 to day 49 for the latter. They found that the flexural strength increased approximately 75% for the aerial lime mortar, while it was approximately 87% in the case of the natural hydraulic mortar within the corresponding testing periods. Regarding the compressive strength of prisms, the increase was 50% for the aerial lime mortar and 22% for the natural hydraulic mortar. Furthermore, logarithmic adjustments for them were provided and also an observed increase between day 28 and day 49 for lime mortars was significant in comparison to Portland cement-based materials. Kalagri et al. [13] obtained flexural and compressive strengths, dynamic modulus of elasticity (through the ultrasonic pulse velocity method), mean pore radius and open porosity on four types of mortars fabricated with NHL 5 and NHL 3.5-Z (with pozzolanic or cementitious additives up to 20%). The mortars presented a lime/ aggregate ratio of 1:2.3 by weight. They conducted tests at days 28, 90 and 270 and established relationships between the aforementioned parameters. Karaveziroglou-Weber and Papayianni [17] studied the evolution of the dynamic modulus of elasticity determined by the ultrasonic pulse velocity method, flexural and compressive strengths of eighteen mortars and ten grouts until day 1540. The results show that the compressive strength of mortars without ceramic material or cement exhibited an increase in their compressive strength over time, while it decreased slightly after one year. The flexural strength increased up to 60% from day 28 to one year, the modulus of elasticity followed almost the same tendency as compressive strength. Stefanidou and Papayianni [18] analyzed the role of aggregates in the structure and properties of aerial lime mortars at different ages until day 730. They observed that the aggregates with a particle size distribution of 0/2 mm had a positive impact on strength in the long-term and coarse aggregates improved volume stability. Lanas et al. [5] measured the compressive strengths of different mortars fabricated with aerial and natural hydraulic limes until day 365. They discovered different stages of hardening in the mortars with the two types of limes. Aerial lime mortars presented two main stages: an initial one characterized by excess water loss, and another consisting of the carbonation process. These phenomena were also observed in reference [4]. However, in the case of the natural hydraulic lime mortars, they presented three phases of hardening in function to their chemical composition, according to the reference [5]. The first phase until day 28 included the hydration of several hydraulic compounds that formed hydrated calcium silicates (C-S-H) and an increase in strength due to the presence of C3S; the second lasted approximately 182 days with a slight increase in the strengths as C3S almost finished and finally, the third phase of 365 days exhibited an increase of the long-term

strengths due to the carbonation process and the contribution of C2S. In the reference [3], Lanas et al. measured flexural and compressive strengths of several mortars fabricated with NHL 5 at different lime/aggregate ratios. They observed that NHL mortars with high lime/aggregate ratios, such as 1:1 and 1:2, followed the three stages of hardening as mentioned previously, while in mortars with lower lime/aggregate ratios, such as 1:3, 1:4 and 1:5, the last stage presented less of an increase in strength as the carbonation process and the presence of C2S were lower. Manita and Triantafillou [19] studied the compressive strength of mortars with lime, pozzolana, cement and brick fragments over 1095 days. They verified that some fluctuations could appear in the long-term in the compressive strength of mortars with lime, resulting in an increase, stabilization or decrease in strength. They explained that this behavior depended mainly on the amount of portlandite, Ca(OH)2 that remained uncarbonated. Lanas et al. [3,4] also observed a similar behavior for pure lime mortars. Up to the present time, most researchers have focused on the evolution of compressive and flexural strengths or a dynamic modulus of elasticity [2–5,13,17,18]. However, regarding the evolution of fracture energy, which is an important parameter to characterize the ductility and fracture behavior of mortars, it is still unknown. Thus, in this paper, an advanced mechanical characterization of a natural hydraulic mortar and an aerial lime mortar is performed in the long-term, that is, up to day 448. For this purpose, density, carbonation depth, flexural, compressive and splitting tensile strengths, fracture energy and elastic modulus were measured at different points and the evolution of these properties were also studied. Furthermore, various non-dimensional laws are provided for all these mechanical properties with respect to their corresponding values at day 28. This could be useful for numerical simulations of masonry structures in the long term with these materials. The remainder of the paper is organized as follows: the subsequent section describes the experimental procedure. Section 3 presents the results and analysis. Finally, conclusions are reached in Section 4.

2. Experimental procedure 2.1. Raw materials In this work, a natural hydraulic (NHL 3.5) and an aerial (CL 90S) lime were used to fabricate the corresponding lime mortars and study the age effect on the physical and mechanical properties. Both limes are widely used for restoration purposes [2–4]. The name of the former means that at day 28 it is expected to provide a compressive strength of between 3.5 MPa and 10 MPa. The name of the other indicates that it is a calcium lime in the form of powder (S) with CaO + MgO in a proportion higher than 90%. [1]. From a mineralogical point of view, both lime mortars are composed of portlandite, Ca(OH)2, according to their technical sheets, in a proportion of approximately 30% for the natural hydraulic lime and higher than 80% for the aerial one. They can also contain some amount of calcite (CaCO3) produced by the transformation of portlandite with the CO2 from the atmosphere [20]. Natural hydraulic lime, instead, also contains calcium silicates (C2S is the major hydraulic phase). Gehlenite (C2AS) may still be present, and C3S, C3A and C4AF may be detected in small amounts due to local overheating in the limekiln [3,8,20–22]. The natural hydraulic lime was supplied by ‘‘Socli, Italcementi Group” (France) and had an apparent density of 850 kg/m3. The aerial lime was provided by ‘‘Calcasa Calcinor” (Spain) and had an apparent density of 490 kg/m3. Both chemical and mineralogical analyses of the limes were performed by corresponding X-ray flu-

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

orescence (XRF) and X-ray diffraction (XRD) analyses. For each test, a Philips (PANALYTICAL) Magis Pro X-ray fluorescence spectrometer and a Philips (PANALYTICAL) X’Pert MPD diffractometer were used, respectively. The results show that indeed the natural hydraulic lime presents a content of portlandite higher than 30%, and the aerial lime contains a corresponding one that is higher than 80% (see Table 1 and Fig. 1). In addition, some portlandite has been transformed into calcite due to carbonation during storage. Moreover, the laser particle size distribution curves of both limes (Fig. 2) were obtained by means of a PSD Mastersizer 3000 from Malvern Instruments. A commercial crushed limestone was used as an aggregate. It had a particle size distribution curve, presented in Fig. 3 and determined according to EN 1015-1 [23]. Its apparent particle density was 2680 kg/m3 in accordance with EN 1097-6 [24] and its apparent density was 1820 kg/m3 according to EN 1097-3 [25]. 2.2. Fabrication of the natural hydraulic and aerial lime mortars Both mortars were fabricated with a lime/aggregate ratio of 1:3 by volume according to the traditional historic mortars [12,26,27]. The water/lime ratio was 0.9 by volume, which provided plastic consistencies according to EN 1015-3 [28] and EN 1015-6 [29]. Thus, the consistency of the natural hydraulic lime mortar was 150–155 mm, and the consistency of aerial lime mortar was 140–150 mm of the flow diameter. To obtain a convenient measurement during the fabrication process, volume proportions of compounds were converted to weight by using their corresponding apparent density. The specimens were prepared according to EN 1015-11 [30]. They were cast in prismatic molds measuring 40 mm  40 mm  160 mm. Natural hydraulic lime mortar specimens were just cured since fabrication in the climatic chamber (RH 97% ± 0.5% and 20 °C ± 0.5 °C) with the first two days in the molds. However, aerial lime mortars were cured for seven initial days inside the climatic chamber (RH 97% ± 0.5% and 20 °C ± 0.5 °C) and the remaining days under laboratory conditions (RH 51% ± 10% and 17 °C ± 3 °C). During the seven initial days in the climatic chamber, Table 1 Chemical and mineralogical compositions of the limes. Natural hydraulic lime, NHL 3.5 Chemical composition (%) Na2O 0.125 MgO 3.078 Al2O3 2.051 SiO2 13.606 P2O5 0.044 SO3 0.990 Cl – K2O 0.456 CaO 59.949 TiO2 0.132 MnO 0.014 V2O5 – Fe2O3 0.994 NiO 0.014 CuO 0.007 ZnO 0.004 SeO2 0.002 Rb2O 0.003 Br – SrO 0.169 CO2 18.365 Mineral phases (%) Portlandite, Ca(OH)2 40–45 Calcite, CaCO3 40–45 Calcium silicates 10–15

3

the aerial lime mortars were initially cured for five days inside the molds and then for two additional days outside the molds. Subsequently, during the curing in the laboratory, it is noted that the initial 14 days were especially favorable for the carbonation process (RH 60% ± 6% and 16.6 °C ± 0.5 °C) [31–35]. 2.3. Mechanical and physical tests on lime mortars Seven physical and mechanical properties were measured on both lime mortars. They included density, carbonation depth, flexural and compressive strengths on days 7, 14, 28, 56, 112, 224 and 448. Splitting tensile strength (the specimens were obtained from the remaining prisms of the three-point bending test to obtain fracture energy), fracture energy and elastic modulus from prisms were obtained on those same days, but starting at day 28 due to the fact that the prismatic notch specimens could be damaged during the installation process for the test prior to that time. The aerial lime mortar was not tested on day 7 as the specimens were not sufficiently hard. Fig. 4a–d shows a picture of each mechanical test configuration and Fig. 4e presents the resulting specimen surface after the carbonation test using the phenolphthalein method. 2.3.1. Apparent density The apparent density of the fresh mortar was obtained according to EN 1015-6 [29,36] by measuring the weight of one liter of mortar at the moment of fabrication. Subsequently, the apparent density of the material was measured on the prismatic specimens prior to performing the flexural test by simply dividing the mass of the specimen into its volume. 2.3.2. Flexural and compressive strengths from prisms The flexural and compressive strengths were measured in accordance with EN 1015–11 [30] by using an Instron 1011 testing machine. The flexural strength was obtained from three-point bending tests on three specimens at a loading rate of 10 N/s and a span of 100 mm (Fig. 4a) [36,37]. The compressive strength was measured at a loading rate of 50 N/s on the remaining six half-prisms from the bending tests. Two steel plates (40 mm  40 mm  10 mm) were placed on opposite faces of the specimen to centralize the load. Furthermore, an individualized ball-and-socket joint over the steel plate was used to reduce the eccentricity during the loading process (Fig. 4b) [36,37].

Aerial lime, CL 90-S 0.070 0.545 0.336 0.944 0.020 1.022 0.006 0.039 70.609 0.029 – 0.017 0.164 0.017 0.009 – – – 0.002 0.034 26.137 90–95 5–10 –

2.3.3. Fracture energy The fracture energy is defined as the energy required to open a unit area of crack surface. It was measured by three-point bending tests, using an Instron 8805 testing machine in accordance with the procedure described in RILEM [38] and the improvements recommended by Planas, Guinea and Elices [39–41]. The specimens were the same size as those used for flexural tests, but with a pre-cast notch (Fig. 4c). This notch was made by introducing a cardboard piece (2 mm in thickness and 20 mm in depth) into the middle of the beam during fabrication. The tests were performed in a displacement control at a rate of 5.0  104 mm/s up to a displacement of 0.3 mm, and subsequently, at a rate of 2.5  103 mm/s until the conclusion of the tests (at approximately 3 mm of displacement). Therefore, the fracture energy was calculated by measuring the area under the load-displacement curve obtained from the three-point bending test on the notch specimen. As mentioned previously, the improvements of Planas, Guinea and Elices [39–41] were followed, insofar as they involve the corrections in the tail of the load-displacement curve in order to obtain the unmeasured work. In the references [36,37], more detailed information may be found on how to determine the fracture energy.

4

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

Fig. 1. X-ray diffraction patterns of (a) natural hydraulic lime, NHL 3.5, and (b) aerial lime, CL 90-S. CH = portlandite, CC = calcite, C2S = dicalcium silicate (belite), C3S = tricalcium silicate (alite), Qtz = quartz.

tb ðaÞ ¼ tb ða=DÞ ¼ 0:8  1:7a þ 2:4a2 þ þ

0:66 ð1  aÞ2


4 0:04  0:58a þ 1:47a2  2:04a3 b

ð2Þ

where B is the specimen width, Ci is the initial compliance determined from Load-CMOD curve and tb ðaÞ is a dimensionless shape function dependent on b and the relative notch/depth (a/D) ratio a. 2.3.5. Splitting tensile strength (indirect tensile strength) Splitting tensile tests (Brazilian tests) were performed on the four prismatic halves from the preceding bending test to obtain the splitting tensile strength by using an Instron 1011 testing machine at a loading rate of 50 N/s in accordance with the procedure recommended by the Standard EN 12390-6 [43]. The proportion between the load-bearing width and the height of the specimens was kept as low as 1/10 (Fig. 4d) according to the recommendations in [44–46]. The bearing strips were made of plywood and were placed in the middle of the longest side of the halves [36,37]. The splitting tensile strength was obtained from:

Fig. 2. Laser particle size distribution curves of the limes.

ft ¼

2F

pBD

ð3Þ

where f t is the splitting tensile strength, F is the maximum load, B and D are, respectively, the specimen width and depth, as previously mentioned. 2.3.6. Characteristic length The characteristic length of both lime mortars was also obtained at different moments according to Eq. (4):

lch ¼ Fig. 3. Aggregate grading curve.

2.3.4. Elastic modulus from prisms From the three-point bending tests for obtaining fracture energy, the elastic modulus may also be measured by attaching an extensometer (strain gauge extensometer Instron 2620) in order to obtain the crack-mouth opening displacement (CMOD) (Fig. 4c). Thus, the elastic modulus (Epr) was obtained from prisms by applying general Eqs. (1) and (2) for span/depth (S/D) ratios (b) between 2.5 and 16 [36,42].

Epr ¼ 6

Sa C i BD2

tb ðaÞ

ð1Þ

EGF 2

ft

ð4Þ

where E, GF and ft are, respectively, the elastic modulus, the fracture energy and the splitting tensile strength, as previously mentioned. The characteristic length is a parameter proposed by Hillerborg et al. [47] and it is an indicator of the brittleness of a material. As a material becomes more brittle, it decreases. It should be mentioned that not only were the flexural and compressive strength from prisms measured, as is common practice, but also the fracture energy and the splitting tensile strength. From these two properties and the elastic modulus, the characteristic length can be obtained, which, as mentioned, is an indicator of the ductility of a material. Moreover, the fracture energy and splitting tensile strength are intrinsic properties. This means that they do not change with the specimen size, while the compressive strength does in fact vary. For example, in our previous work

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

5

Fig. 4. Tests for: (a) flexural strength, (b) compressive strength, (c) fracture energy, (d) splitting tensile strength. (e) Specimen surface after carbonation test with phenolphthalein method.

[37], it was observed that the compressive strength measured from the standard prism was approximately 30–40% higher than that measured from the cylinder (75 mm in diameter, 150 mm in height). Thus, using the value from a standard prism would result in unsafe design of structures.

2.3.7. Carbonation test method As for the carbonation test method, there is no standard for lime mortars. The purpose of performing such a test is to analyze the evolution of the carbonation state on specimen surfaces. Carbonation in concrete is not a desired chemical reaction because it reduces the pH value of the material thereby causing cracking and carbonation-induced steel corrosion [48]. However, in the case of lime mortars, carbonation is one of the hardening mechanisms [4,16]. In fact, the increase of the carbonation depth is linked to the evolution of the mechanical properties [3,5]. For this reason, in order to better understand the hardening mechanisms of both lime mortars and their evolution, the carbonation depth was determined by the phenolphthalein method. This consisted of spraying this substance (in this case with an alcohol solution containing a 1% concentration of phenolphthalein, C20H14O4) on the broken surface of the mortar sample [49], which in our case, was the prismatic surfaces after the compressive tests. Phenolphthalein causes some of the central part of the broken surface to turn purple, which means that this part remains uncarbonated. The part with no change of color (usually on the borders) reveals that it has been carbonated. Carbonation depth is the length between the external surface of the mortar and the end of the colored region [49]. The tests were performed on the natural hydraulic and aerial lime mortars from day 14 to day 448, on the remaining parts of the prismatic specimens tested under compression.

3. Results and discussion In this section, firstly, the evolution of the carbonation depth with the phenolphthalein method is introduced. Subsequently, the results of the mechanical tests at various moments, described in Section 2 for the natural hydraulic and the aerial lime mortars, are presented. Finally, non-dimensional evolution curves of such properties are shown for both mortars. 3.1. Evolution of the carbonation depth The evolution of the carbonation depth on the samples of natural hydraulic and aerial lime mortars by means of the phenolphthalein method can be observed in Figs. 5 and 6, respectively. The results of this method must be treated with care because the carbonation front is not sharply defined as the reaction tends to proceed along cracks and voids beyond the front. However, it is still a suitable method for studying the progress of carbonation [27,33,50]. For the natural hydraulic lime mortar, the carbonation depth was also measured on day 7, but it is not included in the paper as almost the entire surface turned purple, thereby indicating that the specimens were not carbonated at such an early stage. From Figs. 5 and 6, it is observed that the carbonation process did not start until day 14 or day 28 (shown by a narrow unstained area on the sample borders). Due to the fact that the carbonation process only begins once excess pore blocking water is evaporated [15,16,33,48,51], little carbonation can occur within the first 14 days. For the natural hydraulic lime mortar, the carbonation process continued progressively from the outside to the inside [52] up to day 448, as shown by the increase of the unstained depth in Fig. 5. It shows that the shape of the carbonation front can be irregular, especially as time passes and it usually gets wider

6

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

Fig. 5. Carbonation degree of natural hydraulic lime mortar on different days.

Fig. 6. Carbonation degree of the aerial lime mortar at different days.

at the surface where the specimen is laid as the atmospheric CO2 is prevented from getting inside [27]. For the aerial lime mortar, rather, a configuration similar to the Liesegang pattern from day 112 to day 448 (Fig. 6) can be observed. This phenomenon is a ‘‘quasi-periodic self-organized precipitation of a sparingly soluble product in the wake of a moving reaction front” [15,53,54]. This pattern consists of concentric rings of stained and unstained material. The pale rings represent areas with a higher level of carbonation than the purple rings [15]. This phenomenon is more typical of calcium lime in a putty form, especially the aged one that is stored

underwater for a long time, but it can also happen, although more seldom, in calcium lime in the form of powder on long curing days [27,55] as in the case of this research. More information about the Liesegang pattern can be found in [15,27,53,54]. At day 448, it was observed that there were still purple areas in both mortars, at the core of the sample of the natural hydraulic lime mortar and in the form of a Liesengang pattern, as explained, for the aerial lime mortar. This demonstrates that portlandite still remains in the mortars and therefore, the carbonation process could continue beyond day 448 [15,16].

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

Analyzing the carbonation depth of this research with the ones obtained by other researchers, a comparison can be established. For example, for quicklime mortars the carbonation depth is larger as shown by Oliveira et al. in [55]. Lawrence et al. [15] also observed larger carbonation depths in non-hydraulic mortars with bioclastic and oolitic binders ranging between 1.0 mm and 2.5 mm at day 14. In this research, the carbonation depth was between 0.5 mm and 1.0 mm at the same time for the natural hydraulic lime mortar and between 0.5 mm and 1.5 mm for the aerial one. In the thesis of Cazalla [27], a similar carbonation depth of the specimens of this research at day 56 can be found in a calcitic lime mortar with siliceous sand at day 49. That research obtained a carbonation depth of approximately 1.0 mm and 3.5 mm, while our specimens presented a carbonation depth of between 2.1 mm and 2.3 mm in the case of the natural hydraulic lime mortar and between 1.5 mm and 3.5 mm for the aerial mortar. Nevertheless, it is difficult to establish a comparison because small variations in the binder type, the dosage, the water content or the curing conditions can affect the porosity of the material and therefore, the CO2 access and carbonation process. 3.2. Evolution of density and mechanical properties The evolution of density and some mechanical properties for both natural hydraulic and aerial lime mortars is observed in Tables 2 and 3 and Fig. 7. In those tables, q, fflex, fcpr, ft, GF and Epr are, respectively, the apparent density, the flexural strength, the compressive strength from prisms, the splitting tensile strength, the fracture energy and the elastic modulus measured through prisms. Furthermore, Std. Dev. and CV stand for the standard deviation and the coefficient of variation, respectively. In Fig. 7, the error bars are the standard deviation. It has been observed that the density and mechanical properties of the natural hydraulic lime mortar are higher than that of the corresponding aerial lime mortar. This is mainly due to the different mineralogical compositions of both binders and the different hardening processes of both mortars. As explained in Section 1, aerial lime is mainly composed of portlandite, while natural hydraulic lime has calcium silicates (mainly dicalcium silicates, C2S) in addition to portlandite (Table 1 and Fig. 1) [8,20–22]. The former hardens by carbonation (Eq. (5)), while the latter also hardens due to the hydration of its mineralogical compounds (Eq. (6)) [5,33].

CaðOHÞ2 þ CO2 ! CaCO3 þ H2 O

ð5Þ

2ð2CaO  SiO2 Þ þ 4H2 O ! 3CaO  2SiO2  3H2 O þ CaðOHÞ2

ð6Þ

Table 2 and Fig. 7a show that for both mortars the density decreases considerably from the moment of the fabrication (apparent density of fresh mortar was obtained according to EN 1015-6 [29,36]) until day 56. This decrease in density is due to the water evaporation of the mortars during the air drying process of the samples [16,51]. At this age, the hydraulic lime mortar loses 8.88% of its density compared to the value at the fabrication moment, while the loss is 9.76% for the aerial one. This water loss is logically more pronounced for the aerial lime mortar than for the natural hydraulic mortar as the former is cured under laboratory conditions after seven days while the latter remains in the climatic chamber at higher RH conditions for the entire time. After day 56, the density of the hydraulic lime mortars remained almost constant. In reference [3], a similar tendency was also found for mortars fabricated with NHL 5 and low lime/aggregate ratios. For the aerial lime mortar, however, the density increased slightly after day 56. This is due to the transformation of portlandite into calcite that causes a weight increase as the larger pores are filled with the expanding calcium carbonate without any noticeable change in the volume [16,51,52].

7

Regarding the evolution of mechanical properties (see Tables 2 and 3 and Fig. 7b–f), the results show that there is a sharp increase up to day 56 for both lime mortars. That is to say, in the case of the natural hydraulic lime mortar, taking the mechanical properties at day 448 as a reference, the flexural strength (Fig. 7b), the compressive strength (Fig. 7c), the splitting tensile strength (Fig. 7d), the fracture energy (Fig. 7e) and the elastic modulus (Fig. 7f), gain 88%, 73%, 61%, 88%, 69%, respectively. Similar behavior was found by Lanas et al. in the reference [3] for mortars with NHL type 5 instead of 3.5 of this research. They observed that mortars with low lime/aggregate ratios (that is 1:3, 1:4 and 1:5) experienced an increase in their flexural and compressive strengths close to 85–90% of their maximum strength up to day 28. Mortars with higher lime/aggregate ratios (1:1 and 1:2) experienced lower increase percentage, which is approximately 50% of their maximum strength. This increase in the mechanical properties was attributed to Lanas et al. [3] to the hydration of several hydraulic compounds that form hydrated calcium silicates (C-S-H phases) [3,56]. They are mainly tricalcium silicates, C3S, that can be present in small amounts in NHL mortars and have an influence from the beginning to day 28 [3,5,57]. Dicalcium silicates, C2S, that start to contribute to the hardening process after day 28 as explained in [3,5] (Eq. (5)). Furthermore, within 14 or 28 days, the beginning of the carbonation process can play a role (Fig. 5a and b) [3,5,33,58]. In the case of the aerial lime mortars and taking the mechanical properties at day 448 as a reference, the flexural strength (Fig. 7b), the compressive strength (Fig. 7c), the splitting tensile strength (Fig. 7d), the fracture energy (Fig. 7e) and the elastic modulus (Fig. 7f) gain 83%, 88%, 75%, 75%, 89%, respectively, at day 56. The aerial lime mortars start to set and then harden by a loss of water, once their pores have reached an optimum moisture content (for Portland cement with hydrated lime it is approximately 50% of its pore volume [52]), and the carbonation process is maximized [16,48,51]. If the pores are saturated with water, carbonation is not developed as the diffusion of CO2 is not allowed. Only when the specimens start to dry, does carbonation begin to play a role [15,16,33,48,51]. As mentioned previously, this process usually starts within 14 or 28 days, see Fig. 6a and b [4,5,33] (Eq. (4)). Therefore, carbonation increases considerably at approximately 2 months of curing, after which time it increases more moderately [58]. This increase in the carbonation process of aerial lime mortars has an effect on its mechanical properties. Similar tendencies were observed for aerial lime mortars by Lanas and Alvarez in the reference [4]. They found an initial hardening phase with the drying of the water excess and a second phase with the carbonation process. After day 56, the mechanical properties of both lime mortars present a more moderate increase up to day 224. A similar tendency was observed by Lanas et al. [3] for natural hydraulic lime mortars up to day 182 approximately. They attributed this phenomenon to the fact that the hydration of the C3S had already been finished as it contributed to the strength at early ages. From this moment and until day 448, considering the mean values, the evolution of the mechanical properties is very slow for the aerial lime mortars and with a slight increase for the natural hydraulic mortars. For example, at day 448 for the natural hydraulic lime mortar the flexural strength reached 1.45 MPa, the compressive strength 4.45 MPa, the splitting tensile strength 0.64 MPa, the fracture energy 13.6 N/m and the elastic modulus from prisms 8.3 GPa. Similar results were obtained by Lanas et al. [3] for mortars with NHL type 5 and a lime/aggregate ratio of 1:3. They reached corresponding flexural and compression strengths of approximately 1.5 MPa and 4.0 MPa, respectively at day 365. In this mortar, this moderate increase is mainly due to the fact that the remaining C2S continues reacting in the long-term [3,5,57,59] and also due to the carbonation process itself, as evidenced by Fig. 5d–f and

8

Table 2 Physical and mechanical properties of both natural hydraulic and aerial lime mortars at various ages. Type of mortar

Age (days)

Natural hydraulic lime 7

14

56

112

224

448 Aerial lime 14

28

56

112

224

448 *

Flexural strength, fflex (MPa)

Compressive strength from prisms, fcpr (MPa)

Splitting tensile strength, ft (MPa)

Fracture energy, GF (N/m)

Elastic modulus from prisms, Epr (GPa)

Characteristic length, lch (mm)

Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%)

2160 10 0.5 2140 20 0.9 2060 20 1 2050 20 1 2060 10 0.5 2040 10 0.5 2040 10 0.5

0.37 0.03 8 0.51 0.01 2 0.89 0.20 22 1.28 0.10 8 1.52 0.05 3 1.75 0.03 2 1.45 0.10 7

1.10 0.07 6 1.53 0.02 1 2.41 0.15 6 3.24 0.10 3 3.36 0.16 5 4.35 0.50 12 4.45 0.13 3

–

–

–

–

–

–

–

–

0.30 0.01 3 0.39 0.02 5 0.49 0.04 8 0.53 0.04 8 0.64 0.04 6

7.6 0.8 11 11.9 2.0 17 12.5 2.0 16 12.5 1.0 8 13.6 2.0 15

4.7 0.8 17 5.7 1 18 6.7 0.5 8 7.3 1.0 14 8.3 0.9 11

400 140* 34* 450 200 45 350 140 40 380 140 38 280 110 38

Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%) Mean Std.Dev. CV (%)

2050 10 0.5 2010 20 1 2010 10 0.5 2020 20 1 2030 20 1 2040 10 0.5

0.27 0.01 4 0.57 0.07 12 0.62 0.04 7 0.73 0.03 4 0.76 0.05 7 0.75 0.05 7

0.69 0.05 7 1.27 0.08 6 1.79 0.15 8 1.88 0.10 5 2.02 0.09 5 2.03 0.12 6

–

–

–

0.13 0.02 15 0.15 0.02 13 0.18 0.01 6 0.18 0.01 6 0.20 0.01 5

3.7 0.4 11 3.9 0.5 13 5.3 0.8 15 5.6 1.5 27 5.2 2.4 46

2.5 0.3 12 3.1 0.2 7 3.4 0.2 6 3.5 0.1 3 3.5 0.1 3

Calculated from its error as a derived magnitude.

550 290* 53* 540 250 46 560 190 33 605 250 42 455 270 59

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

28

Apparent density,

q (kg/m3)

9

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573 Table 3 Percentage of increase in mechanical properties of both natural hydraulic and aerial lime mortars at various ages with respect to day 448. Type of mortar

Age (days)

Flexural strength, (fflex/fflex448)100 (%)

Compressive strength from prisms, (fcpr/fcpr448)100 (%)

Splitting tensile strength, (ft/ft448)100 (%)

Fracture energy, (GF/GF448)100 (%)

Elastic modulus from prisms, (Epr/Epr448)100 (%)

Natural hydraulic lime

7 14 28 56 112 224 448

26 35 61 88 105 121 100

25 34 54 73 76 98 100

– – 47 61 77 83 100

– – 56 88 92 92 100

– – 57 69 81 89 100

Aerial lime

14 28 56 112 224 448

36 76 83 97 101 100

34 63 88 93 99 100

– 65 75 90 90 100

– 71 75 102 108 100

– 71 89 97 100 100

Fig. 7. Evolution of the physical and mechanical properties of both natural hydraulic and aerial lime mortars: (a) density, (b) flexural strength, (c) compressive strength from prisms, (d) splitting tensile strength, (e) fracture energy, (f) elastic modulus from prisms.

10

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

observed in [3,27]. However, according to Lanas et al. [3], the amount of NHL is lower than in mortars with higher lime/aggregate ratios (1:1 and 1:2) and therefore, the effect of the C2S and the carbonation process is not so significant. This results in a lower strength increase than in mortars with high lime/aggregate ratios. Furthermore, they observed some fluctuations in the evolution of mechanical properties in the mortars with high lime/aggregate ratios (1:1 and 1:2), resulting in an increase, stabilization or decrease of the strength. Subsequently, Manita and Triantafillou [19] observed the same phenomenon on the compressive strength of mortars with lime, pozzolana, cement and brick fragments up to day 1095. They explained that this behavior depended mainly on the amount of portlandite that remained uncarbonated. Such fluctuations were not observed in the case of mortars with low lime/ aggregate ratios as the mortars in this research. The evolution of flexural strength of the natural hydraulic lime mortar at day 448

is irregular. At this age, the flexural strength tests performed on three specimens according to EN 1015-11 [30] provided rather inclined crack paths in the three cases, which could cause a reduction on the flexural strength at this age as observed in Fig. 5b. For the aerial lime mortar, at day 448 the flexural strength is 0.75 MPa, the compressive strength 2.03 MPa, the splitting tensile strength 0.20 MPa, the fracture energy 5.2 N/m and the elastic modulus from prisms 3.5 GPa. Lanas and Alvarez [4] obtained similar results for aerial lime mortars with a lime/aggregate ratio of 1:3. Thus, they reached corresponding flexural and compressive strengths of approximately 0.80 MPa and 2.0 MPa, respectively at day 365. In the case of this mortar, the increase in the mechanical properties is much less pronounced as it is only due to the carbonation process itself [5,33]. As for the evolution of the characteristic length, it should be mentioned that it remains constant in both lime mortars from

Fig. 8. (a) Density loss of both natural hydraulic and aerial lime mortars with time. (b–f) Non-dimensional graphs of the evolution of the mechanical properties of the mortars with respect to their corresponding values at day 28: (b) flexural strength, (c) compressive strength from prisms, (d) splitting tensile strength, (e) fracture energy, (f) elastic modulus from prisms.

11

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573 Table 4 Coefficients for Eq. (7). Yt

fflex fpr ft GF Epr

Natural hydraulic lime mortar

Aerial lime mortar 2

m

n

R

m

n

R2

0.99 0.94 1.08 1.18 1.05

0.27 0.28 0.25 0.18 0.19

0.91 0.92 0.97 0.69 0.99

0.88 1.00 1.05 1.05 1.10

0.19 0.21 0.15 0.11 0.10

0.71 0.74 0.93 0.94 0.98

day 28 to day 448 (see Table 2) as the small variations at different moments are within the margin of error. It is noted that on average, the characteristic length of aerial lime mortars is higher than that of natural hydraulic lime mortars, and accordingly, the former are more ductile. 3.3. Non-dimensional evolution of density and mechanical properties Non-dimensional tendencies of the evolution of the mechanical properties of both lime mortars with respect to their corresponding values at day 28 were obtained by the least square method (see Fig. 8) where the error bars are the standard deviations of the experimental results. Moreover, empirical equations are also proposed, as shown in Eq. (6) and Table 4. The mechanical properties at day 28 were selected as references, due to the fact that they are measured in general according to the standard [30]. It is worth noting that the natural hydraulic and aerial lime mortars of this research have only reached at this age 54% and 62% respectively, of their corresponding compressive strengths at day 448. From Fig. 8, it is observed that the relative increase of all mechanical properties tested is higher for the natural hydraulic lime mortar than for the aerial one, especially, after day 28. For example, at day 448, the non-dimensional compressive strength is approximately 14% higher for the natural hydraulic lime mortar than the aerial one, while the splitting tensile strength is 26% higher for the former. This is probably due mainly to the hydration of dicalcium silicates, C2S, that contributes to the strength of natural hydraulic lime mortars after 28 days [3,5]. The tendency of the flexural strength for the natural hydraulic lime mortar is obtained up to day 224. As explained in Section 3.1, the corresponding results at day 448 were irregular, so they were not included in the fitting to obtain the empirical formula. For that reason, the tendency is estimated according to the expected regular behavior after 224 days (red dashed line in Fig. 8b). The correlation coefficient, R2, is more than 70% in most of the cases. Only for the evolution of fracture energy, GF, in the case of the natural hydraulic lime mortar, R2 is a little bit lower. The underlying reason is not clear. It could be due to the fact that the measurement of such property usually involves higher scatter. Notwithstanding this fact, the obtained fitted curve lies within the range of the standard deviations of the measurements. It is obvious that these empirical equations would be useful for the numerical simulations of the long-term behavior of masonry structures built with aerial or natural hydraulic lime mortars thereby presenting a lime/aggregate ratio of 1:3, a water/lime ratio of 0.9 and limestone aggregates. Further studies would be needed to obtain similar equations to apply to other types of lime mortars with varying water/lime, lime/aggregate ratios and types of aggregates.

 n Yt t ¼m t0 Y 28

4. Conclusions This paper studies the evolution of a few of the physical and mechanical properties of a natural hydraulic mortar and an aerial lime mortar with a lime/aggregate ratio of 1:3 until day 448. Seven properties, such as, density, flexural, compressive and splitting tensile strengths, fracture energy, elastic modulus from prisms and carbonation depth by means of the phenolphthalein method were assessed at different moments. The results show that for both mortars there is a faster increase of their mechanical properties until day 56. Subsequently, the increase of their mechanical properties is more moderate, but progressive up to day 224. From this point until day 448, the evolution of the mechanical properties, by considering their mean values, is very slow for the aerial lime mortar while it shows a slight increase for the natural hydraulic mortar. At this point, some portlandite still remains as shown by the phenolphthalein method. Furthermore, some empirical equations were obtained to describe the evolution of the mechanical properties, which could be useful for numerical simulations with natural hydraulic and aerial lime mortars thereby presenting a lime/aggregate ratio of 1:3, a water/lime ratio of 0.9 and limestone aggregates. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors acknowledge the funding from INCRECYT, the Ministerio de Ciencia, Innovación y Universidades, Spain, under grant BIA2015-68678-C2-1-R and RTC-2017-6736-3 the Junta de Comunidades de Castilla-La Mancha (JCCM) & Fondo Europeo de Desarrollo Regional, Spain, under grant PEII-2014-016-P. L. Garijo gratefully acknowledges the financial support from the scholarship FPU014/05186 awarded by the Ministerio de Educación, Cultura y Deporte, Spain, and J. J. Ortega from the scholarship 2016/12998 awarded by JCCM, Spain. We appreciate the helpful advice on the dosage of the lime mortars of Prof. Pere Roca, from Universidad Politécnica de Cataluña, and assistance with the chemical and mineralogical characterization of the raw limes from the research group of Prof. Anselmo Acosta, from Universidad de Castilla-La Mancha. We also thank the contributions from Iesmat, Instrumentación Específica de Materiales in regard to the laser particle size distribution of the limes. References

ð7Þ

Yt = the magnitude of a mechanical property at time t, Y28 = the magnitude of a mechanical property at day 28, t = time in days, t0 = 28 days.

[1] BS EN 459-1. Building lime – Part 1: Definitions, specifications and conformity criteria. Brussels, Belgium: BSI, 2015, p. 52. [2] A. Drougkas, P. Roca, C. Molins, Compressive strength and elasticity of pure lime mortar masonry, Mater. Struct. 49 (3) (2016) 983–999. [3] J. Lanas, J.L.P. Bernal, M.A. Bello, J.I.A. Galindo, Mechanical properties of natural hydraulic lime-based mortars, Cem. Concr. Res. 34 (12) (2004) 2191–2201.

12

L. Garijo et al. / Construction and Building Materials 236 (2020) 117573

[4] J. Lanas, J.I. Alvarez, Masonry repair lime-based mortars: Factors affecting the mechanical behavior, Cem. Concr. Res. 33 (11) (2003) 1867–1876. [5] J. Lanas, M. Arandigoyen, J.L. Alvarez, J.L. Pérez Bernal, M.A. Bello, Mechanical behavior of masonry repair mortars: aerial and hydraulic lime-based mixtures. 10th International Congress on Deterioration and Conservation of Stones Stockholm, 2004. [6] L. Pelà, K. Kasioumi, P. Roca, Experimental evaluation of the shear strength of aerial lime mortar brickwork by standard tests on triplets and non-standard tests on core samples, Eng. Struct. 136 (2017) 441–453. [7] R. Veiga, Air lime mortars: what else do we need to know to apply them in conservation and rehabilitation interventions? A review, Constr. Build. Mater. 157 (2017) 132–140. [8] B.A. Silva, A.P.F. Pinto, A. Gomes, Influence of natural hydraulic lime content on the properties of aerial lime-based mortars, Constr. Build. Mater. 72 (2014) 208–218. [9] D. Marastoni, L. Pelà, A. Benedetti, P. Roca, Combining Brazilian tests on masonry cores and double punch tests for the mechanical characterization of historical mortars, Constr. Build. Mater. 112 (2016) 112–127. [10] L. Pelà, E. Canella, A. Aprile, P. Roca, Compression test of masonry core samples extracted from existing brickwork, Constr. Build. Mater. 119 (2016) 230–240. [11] J. Grilo, A.S. Silva, P. Faria, A. Gameiro, R. Veiga, A. Velosa, Mechanical and mineralogical properties of natural hydraulic lime-metakaolin mortars in different curing conditions, Constr. Build. Mater. 51 (2014) 287–294. [12] D. Marastoni, A. Benedetti, L. Pelà, G. Pignagnoli, Torque Penetrometric Test for the in-situ characterisation of historical mortars: fracture mechanics interpretation and experimental validation, Constr. Build. Mater. 157 (2017) 509–520. [13] A. Kalagri, I. Karatasios, V. Kilikoglou, The effect of aggregate size and type of binder on microstructure and mechanical properties of NHL mortars, Constr. Build. Mater. 53 (2014) 467–474. [14] M. Apostolopoulou, E. Aggelakopoulou, L. Siouta, A. Bakolas, M. Douvika, P.G. Asteris, A. Moropoulu, A methodological approach for the selection of compatible and performable restoration mortars in seismic hazard areas, Constr. Build. Mater. 155 (2017) 1–14. [15] R.M.H. Lawrence, T.J. Mays, P. Walker, D. D’Ayala, Determination of carbonation profiles in non-hydraulic lime mortars using thermogravimetric analysis, Thermochim. Acta 444 (2) (2006) 179–189. [16] G. Cultrone, E. Sebastián, M.O. Huertas, Forced and natural carbonation of lime-based mortars with and without additives: mineralogical and textural changes, Cem. Concr. Res. 35 (12) (2005) 2278–2289. [17] M. Karaveziroglou-Weber, I. Papayianni, Long-term strength of mortars and grouts used in interventions. Proceedings of IABSE symposium on structural preservation of the architectural heritage. Rome, 1993, p. 527–532. [18] M. Stefanidou, I. Papayianni, The role of aggregates on the structure and properties of lime mortars, Cem. Concr. Compos. 27 (9–10) (2005) 914–919. [19] P. Manita, T.C. Triantafillou, Influence of the design materials on the mechanical and physical properties of repair mortars of historic buildings, Mater. Struct. 44 (9) (2011) 1671–1685. [20] A. Arizzi, G. Cultrone, M. Brummer, H. Viles, A chemical, morphological and mineralogical study on the interaction between hemp hurds and aerial and natural hydraulic lime particles: implications for mortar manufacturing, Constr. Build. Mater. 75 (2015) 375–384. [21] J. Lanas, R. Sirera, J.I. Alvarez, Compositional changes in lime-based mortars exposed to different environments, Thermochim. Acta 429 (2) (2005) 219– 226. [22] A. Arizzi, G. Martinez-Huerga, E. Sebastián-Pardo, G. Cultrone, Mineralogical, textural and physical-mechanical study of hydraulic lime mortars cured under different moisture conditions, Materiales De Construccion. (2015;65(318).). [23] BS EN 1015-1. Methods of test for mortar for masonry – Part 1: Determination of particle size distribution (by sieve analysis): BSI; 1998/A1, 2006. p. 8. [24] BS EN 1097-6. Test for mechanical and physical properties of aggregates – Part 6: Determination of particle density and water absorption: BSI, 2013. p. 54. [25] BS EN 1097-3. Test for mechanical and physical properties of aggregates – Part 3: Determination of loose bulk density and voids: BSI, 1998, p. 10. [26] A. Moropoulou, A.S. Cakmak, G. Biscontin, A. Bakolas, E. Zendri, Advanced Byzantine cement based composites resisting earthquake stresses: the crushed brick/lime mortars of Justinian’s Hagia Sophia, Constr. Build. Mater. 16 (8) (2002) 543–552. [27] O. Cazalla, Morteros de cal. Aplicación en el Patrimonio Histórico, Universidad de Granada, Granada, Spain, 2002. [28] BS EN 1015-3. Methods of test for mortar for masonry – Part 3: Determination of consistence of fresh mortar (by flow table): BSI; 1999/A2, 2006, p. 10. [29] BS EN 1015-6. Methods of test for mortar for masonry – Part 6: Determination of bulk density of fresh mortar: BSI; 1998/A1, 2006, p. 8. [30] BS EN 1015-11. Methods of test for mortar for masonry – Part 11: Determination of flexural and compressive strength of hardened mortar: BSI; 1999/A1, 2006, p. 12. [31] D.R. Moorehead, Cementation by the carbonation of hydrated lime, Cem. Concr. Res. 16 (5) (1986) 700–708.

[32] R.M. Dheilly, J. Tudo, Y. Sebai, M. Queneudec, Influence of storage conditions on the carbonation of powdered Ca(OH)2, Constr. Build. Mater. 16 (3) (2002) 155–161. [33] O. Cizer, Competition between Carbonation and Hydration on the Hardening of Calcium Hydroxide and Calcium Silicate Binders, Katholieke Universiteit Leuven, Haverlee (Belgium), 2009. [34] P. Faria, A. Martins, Influence of curing conditions on lime and limemetakaolin mortars. International Conference on Durability of Building Materials and Components Porto (Portugal), 2011, p. 1–8. [35] P. Faria, A. Martins, Influence of type of air lime and curing conditions on lime and lime-metakaolin mortars. in: Vasco Peixoto de Freitas JMPQD, editor. Durability of Building Materials and Components: Building Pathology and Rehabilitation, 2013, p. 105–126. [36] L. Garijo, X.X. Zhang, G. Ruiz, J.J. Ortega, Z.M. Wu, The effects of dosage and production process on the mechanical and physical properties of natural hydraulic lime mortars, Constr. Build. Mater. 169 (2018) 325–334. [37] L. Garijo, X.X. Zhang, G. Ruiz, J.J. Ortega, R.C. Yu, Advanced mechanical characterization of NHL mortars and cohesive simulation of their failure behavior, Constr. Build. Mater. 153 (2017) 569–577. [38] RILEM TC 50-FMC, Determination of the fracture energy of mortar and concrete by means of the three-point bend tests on notched beams, Mater. Struct. 18 (1985) 285–290. [39] M. Elices, G.V. Guinea, J. Planas, Measurement of the fracture energy using three point bend tests. 3. Influence of cutting the P-d tail, Mater. Struct. 25 (1992) 327–334. [40] M. Elices, G.V. Guinea, J. Planas, Measurement of the fracture energy using three point bend tests. 1. Influence of experimental procedures, Mater. Struct. 25 (1992) 121–218. [41] J. Planas, M. Elices, G.V. Guinea, Measurement of the fracture energy using three point bend tests. 2. Influence of bulk energy dissipation, Mater. Struct. 25 (1992) 305–312. [42] G.V. Guinea, J.Y. Pastor, J. Planas, M. Elices, Stress intensity factor, compliance and CMOD for a general three-point-bend beam, Int. J. Fract. 89 (2) (1998) 103–116. [43] BS EN 12390-6. Testing hardened concrete – Part 6: Tensile splitting strength of test specimens: BSI, 2009, p. 14. [44] C. Rocco, G.V. Guinea, J. Planas, M. Elices, Size effect and boundary conditions in the Brazilian test: theoretical analysis, Mater. Struct. 32 (220) (1999) 437– 444. [45] C. Rocco, G.V. Guinea, J. Planas, M. Elices, Size effect and boundary conditions in the Brazilian test: experimental verification, Mater. Struct. 32 (217) (1999) 210–217. [46] I. Iglesias, B. Acosta, R. Yu, G. Ruiz, M. Aineto, A. Acosta, Study of mechanical characterization of ceramic specimens from a Brazilian test adaptation, Materiales De Construcción. 61 (303) (2011) 417–429. [47] A. Hillerborg, M. Modéer, P.E. Peterson, Analysis of crack formation and crack growth in concrete by means of fracture mechanics and finite elements, Cem. Concr. Res. 6 (1976) 73–782. [48] V. Rostami, Y.X. Shao, A.J. Boyd, Z. He, Microstructure of cement paste subject to early carbonation curing, Cem. Concr. Res. 42 (1) (2012) 186–193. [49] S. Boualleg, M. Bencheikh, L. Belagraa, A. Daoudi, M.A. Chikouche, The combined effect of the initial cure and the type of cement on the natural carbonation, the portlandite content, and nonevaporable water in blended cement, Adv. Mater. Sci. Eng. 1–17 (2017). [50] A. El-Turki, M.A. Carter, M.A. Wilson, R.J. Ball, G.C. Allen, A microbalance study of the effects of hydraulicity and sand grain size on carbonation of lime and cement, Constr. Build. Mater. 23 (3) (2009) 1423–1428. [51] K. Van Balen, D. Van Gemert, Modeling lime mortar carbonation, Mater. Struct. 27 (171) (1994) 393–398. [52] D.R. Moorehead, Cementation by the carbonation of hydrated lime, Cem. Concr. Res. 16 (5) (1986) 700–708. [53] C. Rodríguez-Navarro, O. Cazalla, K. Elert, E. Sebastián, Liesegang pattern development in carbonating traditional lime mortars, Proc. R. Soc. Math. Phys. Eng. Sci. 458 (2025) (2002) 2261–2273. [54] O. Cazalla, C. Rodríguez-Navarro, E. Sebastián, G. Cultrone, M.J. De la Torre, Aging of lime putty: effects on traditional lime mortar carbonation, J. Am. Ceram. Soc. 83 (5) (2000) 1070–1076. [55] M.A. Oliveira, M. Azenha, P.B. Lourenço, A. Meneghini, E.T. Guimaraes, F. Castro, D. Soares, Experimental analysis of the carbonation and humidity diffusion processes in aerial time mortar, Constr. Build. Mater. 148 (2017) 38–48. [56] P.J.P. Gleize, A. Mullier, H.R. Roman, Microstructural investigation of a silica fume-cement-lime mortar, Cem. Concr. Compos. 25 (2) (2003) 171–175. [57] F.M. Lea, The Chemistry of Cement and Concrete. in: E. Arnold (Ed.), Glasgow, 1970. [58] A. Arizzi, G. Cultrone, The influence of aggregate texture, morphology and grading on the carbonation of non-hydraulic (aerial) lime-based mortars, Q. J. Eng. Geol. Hydrogeol. 46 (4) (2013) 507–520. [59] K.S. Wang, K.L. Lin, T.Y. Lee, B.Y. Tzeng, The hydration characteristics when C2S is present in MSWI fly ash slag, Cem. Concr. Compos. 26 (4) (2004) 323–330.