Introduction

Natural stones used in building construction are expected to meet certain physical and mechanical requirements. Their physical properties, such as colour and texture, and mechanical properties, i.e., compressive strength, bending strength, and rupture modulus, are taken into account in the selection of natural stone as building and cover materials. The susceptibility of the materials to different atmospheric factors, such as air pollutants and salts, and their capillary water absorption capacity are other parameters considered in the use of natural stones.

Limestone has been widely used as a construction material in buildings and monument structures for hundreds of years due to its workability. Limestones vary depending on their geological structure, mineralogical and chemical composition, grain size, and textural properties. These properties lead to differences in weathering characteristics and behaviour under different environmental conditions and applications (Souza et al. 2021). Limestones are particularly vulnerable to moist weather conditions and air pollutants (SOx and NOx). Therefore, complex weathering processes involving physical, chemical, and biological effects are observed in limestones over time (Pápay et al. 2021; Steiger and Charola 2011; Korkanç and Savran 2015).

Water in combination with other environmental factors intensifies the degree of weathering in natural stone. When the pores of stone are filled with water, soluble salts, or air pollutants (SOx/NOx), natural stones weather and their mechanical are also adversely affected (Rothert et al. 2007; Cueto et al. 2009; Tomašić et al. 2011). Therefore, the pore size distribution and capillary water absorption parameters of natural stones are very important in understanding and predicting weathering.

The most common type of weathering in limestones is crystallization and dissolution of salts in the pore systems (Doehne 2002; Ruffolo et al. 2015; Al Omari et al. 2016). Soluble salts delay drying by increasing the critical water content in capillary cracks and/or pores. Once the water evaporates, the salts crystallize and form unsightly stains, blooms, flakes, or bands on the stone surface (Siegesmund and Dürrast 2011; Charola 2000; Benavente et al. 2007). This process results in the surface degradation of limestone over many crystallization/dissolution cycles of various soluble salts in dry and wet weather (Grossi and Brimblecombe 2007; Monna et al. 2008; Přikryl et al. 2017). Pore size distribution is a key parameter affecting the damage caused by salts to stone, with particularly detrimental effects on stone strength (Nicholson 2001; Yu and Oguchi 2010; Lόpez-Doncel et al. 2016).

Weathering is a complex process that accelerates the degradation of rocks in the presence of soluble salts and air pollutants, such as sulfur dioxide (SO2), nitrogen oxide (NOx), carbon monoxide (CO), carbon dioxide CO2, and atmospheric particulate matter (PM) (Molina et al. 2017; De la Fuente et al. 2011; Steiger and Charola 2011; Ozga et al. 2011; Spezzano 2021). Sulphuric acid (H2SO4) acts on the limestone in the presence of moisture and causes the formation of calcium sulfide hemihydrate (CaSO3·0.5 H2O) and gypsum (CaSO4·2H2O) on the surface of stones (Rodriguez-Navarro and Sebastian 1996). Gypsum and calcium sulfide affect the presence and distribution of hemihydrate and cause deformations on the surface of natural stone (Böke et al. 2002; Malaga-Starzec et al. 2003). Weathering occurs by the action of gaseous pollutants on stone surfaces, it varies in dry and wet climates, and it depends on temperature and humidity conditions (Grossi and Murray 1999; Elfving et al. 1994; Böke et al. 1999; Luque et al. 2013; Lan et al. 2005; Olaru et al. 2010). Major air pollutants, such as NO and NO2, produce strong acids such as HNO2 and HNO3 in the presence of water, and these acids cause deterioration of the surface of natural stone. Studies assessing the effects and weathering mechanisms of nitrogen oxide on natural stone are quite limited (Kirkitsos and Sikiotis 1995, 1996). Nitrogen oxides act mainly as catalysts for SO2 oxidation during dry and wet deposition processes (Massey 1999; Haneef et al. 1992, 1993; Bai et al. 2006). The secondary crystallization of calcium sulfate and calcium nitrate has been reported to be much higher in wet conditions (Allen et al. 2000). Ageing tests carried out in mixed acid atmospheres under laboratory conditions showed that the effects caused by HNO3 and H2SO3 acid mixtures on stone surfaces occurred faster than with pure NO2 under wet atmospheric conditions (Vazquez et al. 2016). Recent studies have focused on the capillary water absorption–pore relationship in natural stone and the effects of weathering due to salt and air pollutants on mineralogical, physical, and mechanical properties. The findings of these studies show that salt-induced weathering in limestones is generally a function of various mineralogical–petrographical and physico-mechanical parameters, such as mineral type and distribution, strength, freeze‒thaw resistance, ultrasonic velocity, roughness, and colour (Espinosa Marzal and Scherer 2008; Barbera et al. 2012; Derluyn et al. 2014). Many studies have discussed the importance of capillary water absorption and pore distribution and their effects on weathering. According to these studies, pore size distribution has an effect on permeability and is also influenced by rock texture (Lucia 1983). Small pore sizes in rocks produce low water absorption coefficient values (Benavente et al. 2002), and this coefficient may be inversely proportional to rock strength. The pore structure effect strongly affects the water absorption coefficient when used as a strength indicator (Benavente et al. 2004, 2007). Various studies on weathering processes due to air pollutants have noted that open porosity and surface weathering lead to a loss of material in different rock types, resulting in changes in colour and surface roughness (Tecer 1999; Vazquez et al. 2016; Çetintaş and Akboğa 2020).

Although the effects of the capillary water absorption coefficient, pore size distribution, salt crystallization, and air pollutants (SOx and NOx) on the weathering of limestones have been addressed individually in many studies, a few studies have assessed the collective action of such effects on weathering processes.

The aim of our study is to elucidate the decomposition mechanisms and physical changes that occur due to salt and air pollutants by observing the changes in capillary water absorption and porosity in limestones from the Antalya region. The variations in the capillary water absorption coefficient and pore size distribution of the studied limestones were determined based on weathering tests (salt crystallization and air pollutants), and then, the results were interpreted and discussed according to physical properties (weight and microstructure).

Materials and methods

For the laboratory experiments, three limestones were obtained from marble plants in the Antalya region. Limestone characterization is necessary to understand the behaviour of the stone in the presence of different factors that play an important role in the weathering process. For this reason, simulated weathering tests were conducted on the Antalya limestones to determine the changes in the properties of the limestones due to weathering. In the petrographical characterization of the limestones, thin sections were first prepared for polarizing light microscopy (PLM) investigations. The petrographical and micromorphological features of the studied samples were examined using a Nikon Eclipse 2V100POL polarizing microscope, and the resulting mineralogical and petrographic descriptions are given in Table 1. A geochemical analysis of the limestones was conducted using XRF (X-ray fluorescence spectrometry), and the results of the geochemical analysis are presented in Table 2.

Table 1 The mineralogical and petrographical descriptions of the limestones from the Antalya region
Full size table
Table 2 Geochemical analysis of the limestone samples
Full size table

The resistance to salt crystallization (TS EN 12370) and ageing through the action of H2SO3 and HNO3 (TS EN 13919) were determined based on the methods described in standards to identify changes in limestones under different atmospheric conditions (TS EN 13919 2004; TS EN 12370 2020). For the determination of resistance to salt crystallization, seven samples 40 × 40 × 40 mm in size were prepared, and one of these was used as the reference sample. The samples were left in a 14% sodium sulphate decahydrate solution (14 g Na2SO4·10H2O per 86 g of deionized water) for 2 h at 20 ± 0.5 °C, and then, the limestones were removed from the solution and dried in an oven for at least 16 h. The samples were cooled at room temperature for 2.0 ± 0.5 h before being reimmersed in a cold sodium sulphate solution. The process was repeated 15 times, and the samples were washed with deionized water (Fig. 1a). After 15 test cycles, the samples were dried at 70 ± 5 °C. The final dry weight of the samples was measured, and the relative mass difference (decrease or increase in mass) was calculated.

Fig. 1
figure 1

Images of the tests: a Salt crystallization and b Chamber with acid solutions

Full size image

For the determination of resistance to ageing through the action of H2SO3 and HNO3, experiments were carried out in accordance with the procedure specified in TS EN 13919. For each experiment, seven samples 120 × 60 × 10 mm in size were selected from different limestones, and the mass of the dry samples was measured (m0). Prior to the ageing tests, the samples were immersed in water (20 ± 5 °C) at atmospheric pressure for 24 h. SA solution (150 ± 10 mL deionized water and 500 ± 10 mL H2SO3 acid) was prepared for the determination of resistance to ageing through the action of H2SO3 (sulphurous acid, solution in water containing 5–6% by mass of sulfur dioxide), and NA solution (150 ± 10 mL deionized water and 500 ± 10 mL HNO3 acid) was prepared for the determination of resistance to ageing through the action of HNO3 (nitric acid, solution in water containing 5–6% by mass of nitrogen dioxide). Each stone was placed in a test container without direct contact with the solution (Fig. 1b). The samples were kept at different concentrations of the SA and NA solutions in the test container and were removed from the container after 21 days, washed with deionized water, and dried (70 ± 5 °C) again. The final mass (m1) of the dry sample was measured, and the change in mass was calculated with a precision of 0.01%.

Before and after the various atmospheric tests (weathering tests), the capillary water absorption coefficient of the limestone samples was measured in accordance with the TS EN 1925 standard, and a mercury porosimetry (MIP) analysis was conducted to determine the pore throat diameter distribution, which has bearing on the weathering mechanism in the study. Parameters such as pore throat diameter distribution, pore throat diameter, and pore volume were determined using an Autopore IV 9500 Micromeritics mercury porosimeter before and after the weathering tests of the limestone samples. Then, a ZEISS scanning electron microscope (SEM) was used to determine the microtextural properties of the limestone samples.

Results and discussion

Changes in weight

Weight change is one of the primary physical parameters in the determination of material loss and/or crystal formation during the weathering of limestones. In the present study, the weight loss in the limestones was obtained after weathering (salt, SA, and NA) tests, and the weight change data are presented in Fig. 2.

Fig. 2
figure 2

Weight change in the studied samples after weathering tests

Full size image

Figure 2 shows that the weights of the limestones aged by salt action increased from 0.71 to 1.25%. The greatest weight increase was obtained in the Crema Bella, and the lowest weight increase was observed in the Myra Beige. The penetration of salt solution and precipitation of salt crystals in pores and capillary fissures during the ageing cycles induced an increase in weight in the limestones. Salt crystallization due to salt ageing was generally found at the microscale and in trace amounts in SEM studies (Fig. 3).

Fig. 3
figure 3

SEM images of the studied limestones after salt weathering (P: pore, S: salt)

Full size image

The weight increase in Myra Beige (0.11%) was greater than that in Crema Bella (0.06%). The crystallization of calcium sulfide hemihydrate (CaSO3·0.5 H2O) and gypsum (CaSO4·2H2O) on the limestone surfaces after the SO2 treatment resulted in weight gain without direct contact with the solution. During macroscopic examination, thin, white-coloured crust-shaped blooms were observed on the surface of the limestones (Fig. 4). The distribution and quantity of these particles varied and correlated with the mineralogical and petrographical properties of the limestones. The existence of fine-grained calcite crystals, capillary cracks, and secondary calcite fillings may lead to more rapid weathering through the action of acid. The conversion of CaO to CaSO4 occurs rapidly in small calcite crystals (Wang et al. 2015). Secondary gypsum crystals were identified in surrounding smaller calcite crystals in SEM investigations (Fig. 5).

Fig. 4
figure 4

Macroscopic observations of limestones after SA treatment: a Hera Beige, b Myra Beige, and c Crema Bella

Full size image
Fig. 5
figure 5

SEM images after SA treatment: a Hera Beige, b Myra Beige, and c Crema Bella (Gp: gypsum, P: pore, and Cal: calcite)

Full size image

Ageing in an NO2-laden atmosphere induced a reduction in the weight of the studied limestones. The greatest weight loss was found in Hera Beige (0.80%), and the weight losses of Crema Bella and Myra Beige were 0.49% and 0.20%, respectively. The SEM studies revealed that the surface of the limestones had deteriorated as a result of the acidity of the NO2-containing atmosphere (Fig. 6). Secondary minerals were not detected after acid treatment. However, a porous structure was formed along the laminas and fissures. Signs of physical weathering, such as fracturing and disconnections, were observed in the capillary cracks along the secondary calcite crystals. The limestones also included veins with secondary calcite (sparite) fillings in different directions. The rock matrix had a homogeneous micrite texture with low porosity. According to the previous studies, the main causes of disintegration and weathering in limestones are the presence of secondary calcite veins and micritic textures (Barnoos et al. 2020; Freire-Lista et al. 2021).

Fig. 6
figure 6

SEM images of limestones after NA treatment: a Hera Beige, b Myra Beige, and c Crema Bella (P: pore, Cal: calcite)

Full size image

Although the NA cycle in the laboratory was more effective than natural environmental conditions, weight loss in the NA treatment limestones was less than 1.0%, which is considered minimal.

Impact on capillary absorption

The coefficient of capillary water absorption (CCWA) expresses how quickly a limestone absorbs water when it is placed in contact with water. The results of the coefficient of capillary water absorption based on weathering tests are shown in Table 3, and illustrations of capillary water absorption are presented in Fig. 7. The CCWA values increase from 0.46, 4.35, and 0.27 g/m2√s (in the reference samples) to 0.93, 4.74, and 0.49 g/m2√s (in the salt-treated samples) after the salt crystallization tests. The Hera Beige and Crema Bella samples reached maximum saturation after the first 30 min, as shown in Fig. 7a, but water absorption in the Myra Beige continued until the end of the test. The increase in the CCWA occurs, because the surface tension results in salt crystallization. Because the salt crystals in the pores weight down the pore walls, internal stress is created in the stone, resulting in capillary cracks and increased porosity (Angeli et al. 2008; Dinçer and Bostancı 2019). The CCWA increases due to capillary fissures or pores after physical weathering. However, the CCWA varies depending on important factors such as environmental conditions, the amount of salt in the stone, the properties of the pore network, and salt concentration.

Table 3 The results of the coefficient of capillary water absorption (CCWA) based on weathering tests
Full size table
Fig. 7
figure 7

Illustrations of capillary water absorption: a salt crystallization, b SA solution, and c NA solution. Abbreviations: Ref HB reference Hera Beige, Ref MB reference Myra Beige, Ref CB reference Crema Bella, Salt HB after salt crystallization test of Hera Beige, Salt MB after salt crystallization test of Myra Beige, Salt CB after salt crystallization test of Crema Bella, SA HB after SA solution test of Hera Beige, SA MB after SA solution test of Myra Beige, SA CB after SA solution test of Crema Bella, NA HB-NA MB after NA solution test of Myra Beige, NA CB after NA solution test of Crema Bella

Full size image

In the SA tests, the CCWA values of the Hera Beige and Crema Bella also increased from 1.57 to 1.72, whereas those of the Myra Beige decreased from 4.35 to 3.15 g/m2√s during weathering. The surface properties of limestones deteriorate under the acidic effect of SA. The type and degree of weathering and secondary mineral types change depending on the difference in the mineralogical and petrographic properties of the rocks (Scrivano et al. 2018; Israeli and Emmanuel 2018) and often lead to the formation of pores and the decomposition of minerals (Janvier-Badosa et al. 2014). The secondary mineral fillings in the pores cause a decrease in the CCWA values following mineral formation on the surface. The variation in the CCWA values of the studied limestones confirms the results of previous studies.

Figure 7c shows the variation in the capillary water absorption of the reference and NA acid-treated samples. The CCWA values of Hera Beige, Myra Beige, and Crema Bella were 1.21, 5.27, and 1.02 g/m2√s, respectively. Capillary water absorption in Hera Beige and Crema Bella reached saturation after 30 min and remained stable until the end of the test period in the reference and acid-treated samples (Fig. 7). However, water absorption (g/m2) increased in the reference and acid-treated samples of Myra Beige. These findings indicate prove that limestones of the same origin may show different water absorption tendencies due to different physical and mineralogical properties (pore network systems, mineral sizes).

The capillary water absorption illustrations of the weathering tests for the studied limestones are presented in Fig. 8. The values of the capillary water absorption obtained during the weathering tests of the Hera Beige sample remained stable between 60 and 294√s after capillary water absorption increased rapidly from 8 to 42√s. The capillary water absorption coefficient expresses how quickly a stone can absorb water when it comes into contact with water. In this case, water absorption was fast in Hera Beige, in the range of 8–42 √s.

Fig. 8
figure 8

Capillary water absorption illustrations of the weathering tests for the studied limestones: a Hera Beige, b Myra Beige, and c Crema Bella. Abbreviations: Salt HB after salt crystallization test of Hera Beige, SA HB after SA solution test of Hera Beige, NA HB after NA solution test of Hera Beige, Salt MB after salt crystallization test of Myra Beige, SA MB after SA solution test of Myra Beige, NA MB after NA solution test of Myra Beige, Salt CB after salt crystallization test of Crema Bella, SA CB after SA solution test of Crema Bella and NA CB after NA solution test of Crema Bella

Full size image

According to the different weathering tests, the Hera Beige sample had the highest capillary water absorption after SA treatment, but the Crema Bella sample showed similar capillary water absorption after three weathering tests (Fig. 8a, c). The values for capillary water absorption in Myra Beige increased until the end of the test, and saturation was not reached during the weathering tests (Fig. 8c). Microporous structures can easily absorb solutions through capillary action, and the presence of salt crystals in limestone enhances the crystallization pressure in the pores at the microscale (Rijniers et al. 2005; Van et al. 2007). While the porosity dramatically decreased under the action of salt at the beginning, an increase in porosity was observed after 15 cycles. Pores that were disconnected from the outer surface could not be detected, but the internal pressure created by salts or acids can lead to the formation of new fissures connecting new pores and the open pore structure (La Russa et al. 2013). The porosity increased in the Myra Beige sample after the weathering tests, and this increase was confirmed by the MIP measurements.

Change in pore size

Porosity and pore throat distribution may have an impact on the strength of rocks. These properties were detected using the MIP technique. The MIP data of the reference and acid-treated (salt, SA, and NA tests) samples were compared to determine the change in porosity and pore throat distribution of the limestones with the weathering tests. The results of the average pore diameter, open porosity, apparent and bulk density, and pore size distribution (%) (calculated according to test results) as determined by mercury porosimetry for the studied limestones are presented in Table 4.

Table 4 The pore size distribution values of the studied limestones
Full size table

The MIP analysis revealed that the porosity of Hera Beige was 0.64% in the reference sample and 0.50, 0.61, and 0.66% after the salt, SA, and NA tests, respectively. The average pore diameter increased from 2.66 µm in the reference sample to 5.60 µm after the NA test in Hera Beige.

In Myra Beige, the average pore diameter (volume) of the reference sample was 78.55 µm. However, an increase (95.95 µm) occurred after the salt test and a reduction (37.51 and 55.91 µm) was observed after the SA and NA tests, respectively, in the average pore diameter (volume). The porosity was 5.23% in the reference sample and was 6.66%, 5.33%, and 3.86% after the weathering tests in the MIP analysis of Myra Beige. The average pore diameter was 0.23 µm in the reference sample and increased from 0.43 to 0.47 µm after the weathering tests. While the average pore diameter (volume) was 1.17 µm in the reference sample, it decreased by 0.76–1.02 µm after the weathering tests. The spherical volume decreased after weathering of Sabucina rocks due to salt crystallization in the microtextural and microstructural examinations conducted by Barone et al. (2015). The decrease in the pore diameter volume of Myra Beige after the weathering tests confirmed that the spherical volume available for mercury decreased due to new crystal formation or pore deterioration. According to the MIP results of Crema Bella, the porosity decreased from 1.31% in the reference sample to 0.45% and 0.88% after the weathering tests. The average pore diameter increased from 0.02 µm (in the reference) to 19.48, 1.08, and 11.24 µm after the salt, SA, and NA tests, respectively. A similar change was observed in the average pore diameter (volume) of Crema Bella.

In a study by Angeli et al. (2008), a comparison of unweathered and weathered pore size distribution curves provided information on the weathering mechanism induced by salt. Based on their results, if salt crystals fill a pore entrance, the modal pore radius remains low, and spherical porosity remains unchanged in the weathered samples. The degree of filling of salt crystals in pores has an effect on the pore size mode and spherical volume. Partial salt filling ensured a fixed pore mode and a lower spherical volume. However, complete filling of the pore entrance and partial filling of the pore interior by salt crystals induced a decrease in the modal pore size and spherical volume.

The pore size distribution curves of the studied limestones are presented in Fig. 9. After the tests, the studied limestones had a multimodal pore size distribution. Differences in pore sizes were observed between the reference and weathered samples. The pore size modes between 0.2 and 0.5 µm and 50 and 346 µm reached 2–3 µm after salt crystallization. This new pore mode shows that salt crystals formed new pores in Hera Beige (Fig. 9a). The post-test pore size mode was in the range of 0.03–8 µm, and a reduction in spherical volume was observed in Myra Beige (Fig. 9b). An increase in spherical volume was observed in the 20–30 µm pore size mode. However, the pore size did not change, and the spherical volume decreased in the 50–346 µm pore size range. The pore size mode was not changed, but the spherical volume decreased in the range of 9–345 µm after the weathering test in Crema Bella. The pore ranges of 0.004–0.005, 0.01–0.02, 0.03–0.05, 0.07–0.14, and 0.6–1 µm in Crema Bella could not be identified after the weathering test (Fig. 9c). The variation in pore range was interpreted as the disappearance or filling of microsized pores with salt crystals during the weathering process.

Fig. 9
figure 9

The pore size distribution curves of studied limestones with salt tests: a Hera Beige, b Myra Beige, and c Crema Bella

Full size image

The graphics of the pore size distribution curves show that the pore sizes changed to multimodal based on the SA test (Fig. 10). New pores ranging from 0.9 to 1.5 µm were found after the SA test in Hera Beige (Fig. 10a). The new pores similarly formed in the 0.14–0.20 µm and 1–2 µm ranges in Crema Bella (Fig. 10c). The modal pore size was shifted towards lower values along with a decrease in incremental pore volume in the range of 0.15–0.20 µm in Hera Beige. The pore size mode between 6 and 346 µm was not changed, but the incremental pore volume shifted towards lower values. The incremental pore volume decreased in the pore size range of 9–346 µm, and the pore size did not change in Crema Bella. The multiple modes confirmed the existence of gypsum minerals that formed as a result of the interaction of calcite minerals with SA. Secondary gypsum crystals filled the pores, causing a decrease in incremental pore volume. The destructive and abrasive effect of SA gave rise to the formation of new pores (Fig. 5). The pore size mode of the weathered test sample was in the range of 0.03–8 µm, the incremental pore volume decreased, and the pore size range increased in Myra Beige (Fig. 10b). Pore size was not changed, but the incremental pore volume decreased in the pore size range of 9–346 µm. The increase in pore size resulted in the formation of more pores through the action of SA. Incremental pore volume was reduced due to the presence of the newly formed crystals.

Fig. 10
figure 10

Pore size distribution curves with SA tests: a Hera Beige, b Myra Beige, and c Crema Bella

Full size image

The pore size distribution curves in the NA weathering test show marked changes and differences in pore sizes (Fig. 11). New pores or fissures in the range of 0.6–7 µm were observed after the NA test, the pore size mode in the range of 30–345 µm was not changed, and the incremental pore volume also decreased in Hera Beige (Fig. 11a). From the absence of pore volumes in the range of 0.2–0.5 µm after the NA test, it can be concluded that the pore sizes of the rocks decreased, and newly formed porous structures were also observed in SEM studies (Fig. 6). Under the action of concentrated solutions, small pores and microcracks begin to grow during the cycle and decompose the limestones. The incremental pore volume increased both in the range of 0.05–8 µm and in the range of 9–346 µm, and the incremental pore volume decreased in the range of 4–8 µm in Myra Beige (Fig. 11b). New pores or fissure were formed in the range of 2–4 µm, the pore size was not changed in the range of 8–346 µm, and the incremental pore volume decreased in Crema Bella (Fig. 11c).

Fig. 11
figure 11

Pore size distribution curves during the NA weathering tests: a Hera Beige, b Myra Beige, and c Crema Bella

Full size image

The gypsum was a sequential reaction product and formed as a result of the dissolution of calcite by acids. This conversion is known as the acidity effect and occurs by the substitution of carbonate with sulfates (Gibeaux et al. 2018). The MIP analysis revealed new pore formation in limestones that were exposed to air pollutants (SA and NA), the pore size was not changed, and the spherical volume decreased due to the dissolution of calcite. The porous network of rocks affects dissolution and crystallization patterns. The good pore connection leads to an increase in porosity due to rock dissolution by acid rain. Dissolution occurs mainly on the surface in the case of low connectivity, even if the pore distribution changes (Eyssautier-Chuine et al. 2016). This phenomenon accounts for the difference in the grain size distribution values of the reference samples used in the study and the values after the weathering tests, namely, the increase and decrease in porosity (%).

Conclusion

Limestone, as a building material, is subject to weathering by salt and air pollutants (SA and NA). In the present study, laboratory tests were carried out to evaluate the weathering behaviours of three limestones from the Antalya region under the effects of salt and air pollutants SA and NA through an evaluation of their capillary water absorption and pore distribution characteristics. Capillary water absorption, weight change, and mercury porosimetry analyses were carried out to determine physical and microstructural changes in the limestones during the tests using scanning electron microscope (SEM) examinations. The experimental results obtained from the study are presented below.

Weight increased in the salt and SA weathering tests. No change was observed at a macroscopic scale due to the effects of salt. However, salt crystals developed in the pores at the microscopic scale. The reaction of the SO2 atmosphere in the test chamber with the limestone influenced the presence and distribution of calcium sulfite hemihydrate and gypsum on the surface of the limestones and resulted in weight gain without direct contact with the solution. At a macroscopic scale, thin, white crust-shaped blooms formed on the surface of the studied limestones. The reduced weights of the limestones under the action of NA and the formation of the porous structure observed under microscale examination confirmed the loss of material. An evaluation of the strength of the limestones as a building material in our study revealed the Antalya limestones to be “very strong” according to a classification system based on resistance to salt attack. Classifications according to the effect of air pollutants do not exist. However, the studied limestones are considered to be very strong due to their low weight loss (< 1%) after exposure to H2SO3 and HNO3 solutions.

The values for capillary water absorption increased in the limestones after the salt weathering tests, since the salt increased surface tension. The formation of new pores in the structure of the studied limestones under the action of SA and NA increased the capillary water absorption coefficient. Secondary mineral filling of the pores in the limestones resulted in decreased water absorption and the formation of secondary minerals on the rock surface.

After the limestone weathering tests, the pore size distribution curves became multimodal, and the pore sizes varied. Gypsum minerals were formed as a result of the interaction of calcite minerals with SA and filled the pores, resulting in decreased incremental pore volumes. The destructive and abrasive effect of the weathering tests led to the formation of new pores.

According to MIP analysis of the new pores observed in the studied limestones exposed to air pollutants (SA and NA), pore size did not change due to the dissolution of calcite under the action of acid, whereas the spherical volume decreased due to the substitution of carbonate with sulfates.

To conclude, a better understanding of weathering mechanisms can be gained by determining the physical and microstructural (capillary water absorption coefficient and pore distribution) properties of limestones during weathering tests. The approach used in our study provides significant benefits, particularly during the use of limestone as a building material in regions with harsh weather conditions.