Introduction
The inability of the coating materials used in the constructions to reflect the solar radiation and trapping heat in their structure’s may cause higher urban temperatures and the formation of heat islands depending on the local climatic conditions (IPCC 2021). Increasing carbon emissions with industrialization, population growth in cities and unsustainable construction have great potential to accelerate the formation of urban heat islands (Haines and Ebi 2019; Tong and Ebi 2019). The energy efficiency of buildings is very important issue that needs to be analyzed and various projects are supported by several research and innovation programs in the world. Leadership in Energy and Environmental Design (LEED), developed by the United States of Green Building Council (USGBC) is a certification program consisting of criteria supporting sustainable and high-performance green building practices. LEED certified green buildings are defined as buildings that contribute positively to human health and the environment throughout the life of the building, depending on criteria such as “Energy and Atmosphere” and “Materials and Resources” (USGBC 2005).
Besides meeting the desired physical and aesthetic properties, dimension stones are generally considered for cool ceiling and exterior cladding applications to reduce energy demand for the buildings and to increase thermal comfort of the interior. In urban areas, the ability of such materials for reducing both surface and surrounding temperatures are generally expressed by their SRI values. SRI quantifies the heat, which would accumulate on a material related to a white and a black pattern surface under standard environmental conditions (Alchapar et al. 2014). When materials having high SRI value are exposed to sunlight, their temperatures will not rise much, so the use of such materials in buildings, parks and squares is important in terms of cooling efficiency especially in hot climates, and significantly prevents formation of urban heat islands. Total terrestrial flux that reaches to the earth’s surface after it has been filtered by the atmosphere, is so decisive for the formation of urban heat islands due to about 3% of the total terrestrial flux is ultraviolet, 47% is visible light, and the remaining 50% is infrared (IR) (ASHRAE 2005).
Anzeige
In recent years, some studies are being carried out to reduce UHI formations and provide satisfactory thermal comfort in outdoor areas (Alchapar et al. 2014; Radhi et al. 2014; Rosso et al. 2014). TE is an important parameter affecting the surface and ambient temperature, which indicates how effectively a material releases absorbed heat once the heat source is removed. Surfaces having low SR absorb most of the radiation from the sun energy (Sailor and Fan 2002; Simpson and McPherson 1997; Doulos et al. 2004; Santamouris et al. 1998; Araújo 2005). While some of the absorbed energy is transmitted to the ground and to the buildings, another part is transmitted to the air by convection causing rise in air temperature and eventually to the atmosphere. Under the same conditions, the lower SRI value means less emittance on the surface, it cools down longer and therefore releases a higher temperature (Sailor and Fan 2002). Shi and Zhang (2011) investigated the effects of SR and TE on the energy saving of the exterior facades of buildings in different climatic conditions and a computer simulation program was implemented on 35 different cities. As a result of the analyses, in the tropical and subtropical climates, the facades of the buildings having high SR and TE would play an important role in energy saving. Studies in recent years revealed that, increasing solar reflectivity of the building facades from an existing 10–20% to about 60% can reduce cooling-energy use in buildings by up to 20% in very hot climate zones. On the contrary, in the mountainous and temperate climates, they shared that lower SR and TE values would be more suitable in terms of preventing energy losses. Kültür and Türkeri (2011, 2012) have worked on the effect of newly coated and aged roofing material widely used in Türkiye on the potential formation of urban heat islands and which surface features of the material are efficient to reduce the ability of them. Ma et al. (2001) made SR measurements on self-color changing coatings in order to benefit more from sunlight. In these coatings, also known as chameleon-type coatings, when the temperature is below 18 °C, the coating is red, while the color is white above 18 °C. Some assessments were performed according to the colors of the coatings and it was concluded that between the red- and purple-colored coatings absorb more sunlight than the white-colored ones. Sleiman et al. (2011) concluded that the demand for air conditioning in residential areas will be reduced by covering the roofs of buildings with high SR materials in hot areas of the USA. Berdahl and Bretz (1997) examined the changes in the SR values of the roofs depending on the material used, surface roughness and the amount of extraneous matter, and stated a strong correlation between the SR and the surface temperatures. Jo et al. (2010) determined the improvement in energy saving around 2,6 − 3,8% monthly with the decrease in surface temperature by changing the surface materials of the roofs of existing buildings. Berdahl et al. (2012) stated in their work that, some changes were occurred in the SR values of asphalt concrete by exposure to chemical and physical changes. The effects of environmental pollution and precipitation on SR values were investigated for some regions and they found that the increase in bacteria in the asphalt concrete surfaces in the hot climates affected the change in SR values. Cheng et al. (2012) investigated the effects of the pigmented roofs on the SR for seven regions in the state of California. Alchapar et al. (2013) measured SRI values and thermal behaviours of pedestrian pavements due to their commonly used compositions, shapes and colors to reduce urban heat islands and investigated the effects of time dependent contamination on SRI and thermal behaviour. Doulos et al. (2004) concluded that the most appropriate material is marble to reduce the urban heat islands in the cities by considering its surface texture, color and thermal properties.
Carbonate-based dimension stones have always been considered the most appropriate material as external cladding material for low-cost buildings and preferably used in many stages of construction projects, especially as blocks, shaped decorative facing or veneer material, pavement, sidewalks, borders, and landscape architecture (Ugur 2014). However, these types of materials used on the exterior facades of the buildings may cyclically be exposed to physical degradation factors, such as rain, wind, thermal shock and freeze-thaw effects. With the presence of time, these types of recurrent cycles may have significant effects on color changes, integrity and strength loss in the rock, of which water absorption rates increase and initial strength values cannot be preserved. (Guler et al. 2021). As studied by many researchers, the long-term rock degradation induced by T-S aging processes is a function of their physico-mechanical characteristics (Royer-Carfagni 1999; Hale and Shakoor 2003; Yavuz et al. 2006; Cantisani et al. 2009; Ozguven and Ozcelik 2014; Vagnon et al. 2019). They stated that, in the case of thermal effects are combined with other environmental factors, such as water ingress to the body and the presence of some chemicals on the stone surface will accelerate deterioration due to increased porosity and loosening of rock cohesion. Sarıcı (2016) and Ozcelik et al. (2012) stated that rocks may have reduced gloss and increased surface roughness values due to freeze-thaw and thermal shock effects.
As can be seen from the literature, several studies were performed dealing mostly with to improve SRI values of roof and ceiling applications, asphalt pavements, paints, pigmented roofs and different types of coating materials expediently. However, there is a lack of studies regarding the use of fresh stone elements through the buildings and aging response of them in terms of SRI values depending on the atmospheric interactions after a certain period of time. This study contributes to the literature to understand the effects of the changes in SRI values of some carbonate-based stone types as the exterior cladding material depend on some surface properties such as color, roughness, and gloss values. Multivariate regression analysis was established for the estimation of SRI as dependent parameter from these surface properties as independent explanatory variables with a high degree of accuracy. A multivariate regression equation has been developed for the estimation of SRI value because of the direct methods measuring the SRI in laboratory is somewhat time-consuming and expensive. In addition, the effect of physical aging process caused by cyclic T-S on the SRI values were also analyzed. Therefore, there is a need for better understanding of the changes in SRI values in such using of selected dimension stone types depending on their modified surfaces with different optical and some surface properties under the effect of T-S cycles.
Experimental materials and testing procedures
Sample description
The main materials for the present study were polished stone slabs, prepared from a total of 30 selected carbonate-based (travertine, limestone, dolomitic limestone and re-crystallized limestone) dimension stone types were brought to the laboratory in order to carry out a series of laboratory tests. For this reason, dimension stone samples with high commercial acceptance were chosen and collected from a marble processing plant located around Antalya, Türkiye. Although there is a wide range of natural building stone types, the study focused on these natural building stones based on their perceptible color and texture characteristics preferably used as building and facing material both for historical and modern structures in Türkiye. General descriptions with appearances and commercial names of these stone varieties used in the analyses are given in Fig. 1; Table 1.
Anzeige
Fig. 1
General appearances of the stone varieties
×
![]()
Table 1
General descriptions of the stone varieties
Sample Code | Origin | Rock Type | Region | Color | Texture | Main Mineral Content |
|---|---|---|---|---|---|---|
1 | Sedimentary | Travertine | Aksaray | Medium brown | Microcrystalline | Calcite and aragonite |
2 | Sedimentary | Travertine | Denizli | Grayish yellow | Microcrystalline | Calcite and aragonite |
3 | Sedimentary | Travertine | Denizli | Grayish yellow | Microcrystalline | Calcite |
4 | Sedimentary | Travertine | Denizli | Grayish yellow | Microcrystalline | Calcite and aragonite |
5 | Sedimentary | Travertine | Denizli | Grayish yellow | Microcrystalline | Calcite |
6 | Sedimentary | Travertine | Denizli | Grayish yellow | Microcrystalline | Calcite |
7 | Sedimentary | Limestone | Antalya | Yellowish gray | Microcrystalline | Calcite |
8 | Sedimentary | Limestone | Antalya | Yellowish gray | Microcrystalline | Calcite |
9 | Sedimentary | Limestone | Antalya | Yellowish gray | Microcrystalline | Calcite |
10 | Sedimentary | Limestone | Antalya | Yellowish gray | Microcrystalline | Calcite |
11 | Sedimentary | Limestone | Antalya | Yellowish gray | Microcrystalline | Calcite |
12 | Sedimentary | Limestone | Isparta | Very light gray | Microcrystalline | Calcite |
13 | Sedimentary | Limestone | Isparta | Medium light gray | Microcrystalline | Calcite and aragonite |
14 | Sedimentary | Limestone | Bilecik | Yellowish gray | Microcrystalline | Calcite |
15 | Sedimentary | Limestone | Afyon | Pale yellowish brown | Microcrystalline | Calcite |
16 | Sedimentary | Limestone | Sivas | Olive gray | Microcrystalline | Calcite |
17 | Sedimentary | Limestone | Bursa | Dark orange | Microcrystalline | Calcite |
18 | Sedimentary | Limestone | Kastamonu | Olive gray | Microcrystalline | Calcite |
19 | Sedimentary | Limestone | Burdur | Yellowish gray | Microcrystalline | Calcite |
20 | Sedimentary | Limestone | Burdur | Yellowish gray | Microcrystalline | Calcite |
21 | Sedimentary | Limestone | Bursa | Yellowish gray | Microcrystalline | Calcite |
22 | Sedimentary | Dolomitic limestone | Manisa | Yellowish gray | Microcrystalline | Calcite and dolomite |
23 | Sedimentary | Dolomitic limestone | Isparta | Very light gray | Microcrystalline | Calcite, dolomite and aragonite |
24 | Sedimentary | Dolomitic limestone | Isparta | Very light gray | Microcrystalline | Calcite and dolomite |
25 | Metamorphic | Re-crystallized limestone | Muğla | White | Granoblastic | Calcite |
26 | Metamorphic | Re-crystallized limestone | Uşak | White | Crystalline | Calcite |
27 | Metamorphic | Re-crystallized limestone | Afyon | White | Crystalline | Calcite |
28 | Metamorphic | Re-crystallized limestone | Balıkesir | Light gray | Granoblastic | Calcite |
29 | Metamorphic | Re-crystallized limestone | Balıkesir | Light gray | Crystalline | Calcite |
30 | Metamorphic | Re-crystallized limestone | Balıkesir | Dark gray | Crystalline | Calcite |
Techniques used for characterization of samples
The optical properties of the building elements used as walls, roofs, ceilings and floors, taking into account all scientific literature, are characterized by their SR and TE features (Akbari et al. 1996; Doulos et al. 2004; Noelia et al. 2014; Schabbach et al. 2018; Di Giuseppe et al. 2019). SR and TE values are used in the LEED certification system to evaluate the performance of these building elements.
Instrumental techniques, applied to determine some properties of the stone samples, including color, gloss, roughness, and optical features such as SR and TE values were emphasized and defined below. Measurements were performed in Süleyman Demirel University, Excavation Mechanics and Natural Stone Technology Laboratory based on relevant standards. Through the measurement procedure, three slab samples with 150 × 150 mm surfaces were divided into 9 grids of 30 × 30 mm dimensions (Fig. 2a). All measurements were performed on the middle point of these 9 grids of each sample and the average of these 27 measurements was taken into account as the relevant value for each stone type. Measurements were repeated after 20 T-S cycles of exposure, selected as the most common type of accelerated physical aging to understand the relationships among these parameters and SRI. The ambient temperature and relative humidity values were recorded for each measurement. The ambient temperature was between 20 and 24 °C and the relative humidity was 40–60% during the experiments.
Fig. 2
Formation of grids and surface measurement procedure (a) Prepared grids (b) Roughness measurement apparatus (c) Glossmeter (d) Spectrophotometer (e) CIE L*a*b* color space
×
![]()
Roughness measurements
Surface roughness was measured using a contact profilometer Mitutoyo SJ-210 (calibrated by standard ISO1997) surface roughness measurement device, equipped with a diamond tip that moves on the surface with constant speed (Fig. 2b). The measurements have been performed by selecting an exploration length ranging from 0 to 6.5 mm and a sensor speed of 0.5 mm/s. Data collected on a straight line on the surface of a sample enables calculating following three parameters to define roughness: Ra (average roughness), Rq (root mean square roughness), and Rz (average maximum profile height).
Ra is a somewhat statistical value as arithmetic mean of the absolute values of the y-coordinates corresponding to the surface roughness profile. Rq parameter is the quadratic average value of the y-coordinates corresponding to the surface roughness profile which provides statistically significant information about the surface. Rz is the maximum roughness depth, or the vertical distance between the highest and lowest points of the surface roughness profile which forms the basis for measuring the y-coordinate ranges corresponding to the surface roughness. Although there are many different roughness parameters in use, Ra is by far the most common roughness parameter as it gives the arithmetic mean of the roughness profile and therefore this parameter was taken into account in the scope of the study.
Gloss measurements
The method of determining the gloss of dimension stones involves the use of a portable measuring device (glossmeter), located at various points of the stone surface. To evaluate gloss changes, the gloss of the stones was measured by a Tasco brand digital glossmeter equipped with a LED light source in the wavelength of 890 nm (Fig. 2c). Gloss measurements were performed by choosing an angle of the reflected light-beam of 60° in a measurement range of 0–100 gloss units.
Spectrophotometric measurements
Initial colors of stone surfaces were visualized and quantified with a ColorFlex EZ spectrophotometer device (Hunter Associates Laboratory, Inc., Virginia, USA) (Fig. 2d) having the ability to measure the mean L*a*b* values in the CIE L*a*b* color space (Rossel et al. 2006). CIE L*a*b* color system represents 3-dimensional color space which is built-up from three axes that are perpendicular to one another in color space diagram, where the center is neutral or achromatic (Fig. 2e). As three values of uniform color system, L* gives the lightness (represented on a vertical axis with values from 0 (black) to 100 (white), while a* and b* gives chromacity coordinate represented on the a* (negative for green and positive for red) and b* (negative for blue and positive for yellow) axis, respectively.
SR measurements
SR (β), often referred to as albedo, is an optical surface feature measured on a scale of 0 to 1, that higher values imply better reflection of solar rays, contributing to reduced heat absorption of the surface. SR covers the reflectivity across the entire spectrum of solar radiation, including IR and high IR reflectivity is desirable for energy efficiency (Cozza et al. 2015). Generally, light-colored materials that in the visible spectrum have high SR on the contrary dark-colored have low SR in that scale. SR values of the stone samples were measured with a solar spectrum reflectometer (SSR) Model SSR-ER version 6 from Devices and Services Company using the procedure in ASTM C 1549 (2009). The SSR has an ability to make accurate measurements on both diffuse and specular materials up to 6.4 mm thick, which is displayed to 0.001 accuracy. In addition, this measurement procedure is acceptable for meeting the requirements of LEED Green Building Rating System for New Construction and Major Renovations (USGBC 2005). The SSR requires zero offset adjustment by a blackbody cavity with a SR of zero and calibration with a white standard reference material, having SR of 0.801 before the measurements (Fig. 3a).
Fig. 3
Measurement procedure of SR (a) and TE (b) values of the stone samples
×
![]()
TE measurements
TE (ε) is also an optical surface property, the ratio of the radiant flux emitted by the sample to that emitted by a blackbody radiator, which completely absorbs all incident radiant energy at the same temperature. It is a measure of how well a surface emits heat on a scale of 0 to 1, e.g., polished aluminium has an emittance less than 0.1 and a black non-metallic surface has an emittance greater than 0.9 (ASHRAE 2005). The measurement procedure of the TE is regulated by the Standard ASTM C1371-04a (2015) with an apparatus having a circular head with a diameter of 50 mm, heated by an electric power supplier up to a temperature of 82 °C. The measuring head of the apparatus includes two pairs of high emittance and low emittance detector elements (Fig. 3b). The emissivity of the sample can be determined by the knowledge of the radiation heat flow and the surfaces temperatures.
Calculation of SRI
In the present study, the reflective ability of the stone slabs, regarding to return solar energy to the atmosphere has been quantified by means of their SRI values. The resulting values of SR and TE have been used as input data to calculate initial and thermally aged SRI values of the stone slabs in accordance with ASTM E1980-11 (2010) standard. Previously described methodology was used as measurement procedure to obtain SR and TE values from the stone surface. For the measurements, 3 samples of each stone variety were prepared in dry state and 810 initial measurements were performed on their surfaces (Fig. 3a, b). A total of 1620 SR and TE readings were taken for 30 different carbonate based natural stone types. After these measurements were completed, the measurements were repeated for all stone samples for also after T-S cycles.
SRI value of each sample was calculated by also considering the convection heat transfer coefficient (h) as external wind conditions using the Eqs. 1–3 given below. Measured values of the SR and TE of a test surface are used to calculate the SRI for three convective coefficients of 5, 12, and 30 W/(m2K), corresponding to low, medium, and high wind conditions, respectively.
$$\:\text{S}\text{R}\text{I}=\text{123,97}-\text{141,35}x+\text{9,655}{x}^{2}$$
(1)
$$\:x=\frac{\left(\alpha\:-\text{0,029}\epsilon\:\right)\left(\text{8,797}+h\right)}{\text{9,5205}\epsilon\:+h}$$
(2)
$$\:\alpha\:\:=\:1-\beta\:$$
(3)
Where \(\:\alpha\:\) is the solar absorbance, \(\:\beta\:\) is the solar reflectance (SR), \(\:\epsilon\:\) is the thermal emissivity (TE), \(\:h\) is the convective coefficient (W/(m2K)).
Since SRI is an index value in the range of 0 to 100%, the materials having closer SRI values to 100% show a tendency to higher reflective ability and so lower body temperatures, makes them more appropriate building materials for diminishing urban temperatures (Fig. 4). The lower and upper limit values of the SRI will vary depend on the minimum and maximum values of SR and TE values that were given in Table 2. Previous studies show that SRI values in the literature are over 100 (104, 107 and 114) for some materials (Asdrubali et al. 2015; Costanzini et al. 2021).
Fig. 4
SRI change with different surfaces (Cool Roof Rating Council (CRRC))
×
![]()
Table 2
Variation of lower and upper limit values of the SRI with SR and TE values
SR, β | TE, ε | SRI | ||
|---|---|---|---|---|
h = 5 W/(m2K) | h = 12 W/(m2K) | h = 30 W/(m2K) | ||
0 | 0 | -192.6 | -7.7 | -7.7 |
0 | 1 | 1.8 | -4.3 | -4.2 |
1 | 0 | 124.0 | 124.0 | 124.0 |
1 | 1 | 127.9 | 127.9 | 128.0 |
Accelerated aging by the process of cyclic T-S
Various experimental standards have been developed as reliable assessment methods to reflect the observed behaviour of natural stones under the influence of unfavourable both environmental and climatic conditions. Of these, T-S cycles refer to stress formations that can cause all the minerals that make up the stone to expand at different rates in different parts at high temperatures. A rapid drop in temperature and related thermal expansion lead to contraction of the outer minerals, causing a stress between the inner expanded hot body and the outer cold part. This phenomenon plays an important role on the crack propagation through the rock structure and results in intergranular decohesion of the stone with increased pore volume (Plevova et al. 2010).
In this study, T-S test was considered as an artificial physical degradation process based on abrupt temperature changes through the stone structure and performed in accordance with the recommendations of TS EN 14,066 (2015) standard. Firstly, the samples were dried to be conditioned in an air-ventilated oven until to reach constant weight at 40 ± 5 °C for 18 h. Following this process, all studied slab samples were subjected to 20 cycles of thermal shock aging process in a testing cabinet that cycle times, cycle numbers, temperature level and other data entry can be made by LCD screen, expediently (Fig. 5). The testing procedure consists of two main stages at a temperature of about 70 °C for 18 h, followed by rapid cooling with submerging them in water and saturated at a temperature of about 20 °C for 6 h. These two stages are assumed to be as one cycle and each cycle requires 24 h to be completed (Fig. 6).
Fig. 5
Automated T-S testing cabinet
×
![]()
Fig. 6
Diagram showing two main stages of the T-S aging process for one cycle
×
![]()
Results and discussion
Chemical composition and colorimetric values
Chemical composition of the studied stone types was determined with XRF (ICP-ES), Rigaku ZSX Primus II device located in Afyon Kocatepe University (AKU) laboratory based on the recommendations of TS EN 15,309 (2008) standard. XRF analysis results reveal that natural stone samples are mainly consist of CaO and MgO with similar chemical features (Table 3). Ca indicates the most abundant chemical element followed by Mg, Si, Al and Fe, whereas lower values were measured for Na, K and Mn. In particular, almost all samples show the highest CaO values, whereas the 22, 23 and 24 series have the lowest with particularly high MgO content.
Table 3
Chemical composition of the samples obtained by means of XRF and colorimetric values
Sample Code | Chemical Composition (%) | CIE L*a*b* Coordinates | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
CaO | MgO | Fe2O3 | SiO2 | Al2O3 | Na2O | K2O | MnO | L* | a* | b* | |
1 | 55.46 | 0.18 | 0.02 | 0.25 | 0.09 | 0.05 | 0.01 | 0.124 | 50.48 | 11.23 | 28.28 |
2 | 56.00 | 0.38 | 0.03 | 0.19 | 0.08 | 0.01 | 0.001 | - | 77.34 | 4.09 | 18.42 |
3 | 55.43 | 0.36 | 0.05 | 0.23 | 0.08 | 0.03 | 0.01 | - | 69.65 | 5.32 | 19.00 |
4 | 55.30 | 0.34 | 0.10 | 0.59 | 0.21 | 0.02 | 0.03 | - | 70.05 | 5.21 | 18.74 |
5 | 55.43 | 0.35 | 0.05 | 0.22 | 0.08 | 0.02 | 0.01 | - | 66.82 | 5.68 | 20.28 |
6 | 56.04 | 0.20 | 0.02 | 0.12 | 0.04 | 0.02 | 0.001 | - | 71.24 | 5.51 | 20.25 |
7 | 56.86 | 0.25 | 0.01 | 0.03 | 0.01 | 0.02 | - | - | 85.91 | 1.73 | 10.41 |
8 | 56.73 | 0.24 | 0.02 | 0.16 | 0.09 | 0.02 | 0.001 | - | 88.07 | 1.67 | 9.33 |
9 | 57.04 | 0.27 | 0.01 | 0.02 | 0.01 | 0.02 | - | - | 89.37 | 1.31 | 8.36 |
10 | 56.04 | 0.20 | 0.02 | 0.28 | 0.09 | 0.02 | 0.001 | - | 88.91 | 1.30 | 8.47 |
11 | 56.60 | 0.16 | 0.01 | 0.03 | 0.02 | 0.01 | - | - | 88.00 | 1.55 | 9.31 |
12 | 55.88 | 1.05 | 0.02 | 0.08 | 0.03 | 0.01 | 0.001 | - | 69.49 | 1.03 | 3.30 |
13 | 55.02 | 0.57 | 0.14 | 0.59 | 0.31 | 0.03 | 0.06 | 0.0293 | 38.23 | 1.59 | 5.04 |
14 | 55.95 | 0.40 | 0.05 | 0.25 | 0.15 | 0.02 | 0.01 | - | 77.88 | 4.07 | 14.29 |
15 | 54.20 | 0.25 | 0.21 | 0.83 | 0.56 | 0.03 | 0.09 | 0.0209 | 37.26 | 3.06 | 7.74 |
16 | 54.80 | 0.38 | 0.09 | 0.55 | 0.21 | 0.02 | 0.05 | 0.0085 | 26.51 | 1.74 | 5.81 |
17 | 54.60 | 0.05 | 0.08 | 0.25 | 0.11 | 0.02 | 0.01 | - | 59.85 | 5.84 | 20.02 |
18 | 55.90 | 0.56 | 0.04 | 0.17 | 0.07 | 0.02 | 0.01 | - | 49.65 | 0.90 | 4.63 |
19 | 56.67 | 0.37 | 0.02 | 0.14 | 0.07 | 0.02 | 0.001 | - | 81.73 | 2.09 | 10.56 |
20 | 56.40 | 0.38 | 0.03 | 0.13 | 0.09 | 0.01 | - | - | 79.75 | 3.63 | 14.09 |
21 | 56.90 | 0.19 | 0.04 | 0.15 | 0.09 | - | - | - | 80.57 | 2.62 | 12.73 |
22 | 38.40 | 15.80 | 0.04 | 0.13 | 0.08 | 0.02 | 0.01 | - | 80.40 | 2.13 | 8.85 |
23 | 36.50 | 17.70 | 0.02 | 0.06 | 0.04 | 0.02 | 0.001 | 0.008 | 78.86 | 0.65 | 4.85 |
24 | 36.48 | 16.73 | 0.03 | 0.06 | 0.04 | 0.02 | 0.001 | 0.0086 | 67.86 | 1.00 | 4.18 |
25 | 56.49 | 0.16 | 0.02 | 0.14 | 0.09 | 0.02 | 0.001 | - | 74.80 | -1.03 | 1.25 |
26 | 56.40 | 0.38 | 0.03 | 0.13 | 0.09 | 0.01 | - | - | 88.30 | -1.43 | 1.97 |
27 | 56.10 | 0.39 | 0.04 | 0.09 | 0.05 | 0.01 | - | - | 85.01 | -0.22 | 1.96 |
28 | 56.40 | 0.53 | 0.01 | 0.07 | 0.01 | - | - | - | 84.52 | -0.73 | -0.05 |
29 | 57.00 | 0.43 | 0.02 | 0.08 | 0.05 | 0.01 | - | - | 84.21 | -0.54 | 0.75 |
30 | 53.60 | 3.00 | 0.04 | 0.18 | 0.14 | 0.02 | 0.02 | - | 73.44 | 0.05 | 0.65 |
In addition, surfaces of 30 different natural stones were analyzed in terms of their colorimetric values. Recently, the CIE L*a*b* color space is used in many applications as standard color space for various areas. Color spaces are mathematical models used to describe colors which are designed in three dimensions to represent all colors. The components of the CIE L*a*b* color space are L*, a* and b*, of which L* is lightness of a color while a and b constitute the color (Yılmaz et al. 2002). The results regarding the colorimetric analyses clearly indicate that the slab samples have considerably different initial color characteristics and show high chromatic variability in the L*, a* and b* coordinates (Table 3).
It is well known that the presence of metal oxide (MO), particularly Fe, Al and Mn oxides, also in low quantities will contribute significantly the formation of darker colors, which means lower L* values in natural stones. As seen from the Fig. 7 that, the increase in the L* values in connection with lower MO percentage indicates that the SRI value of the sample will also increase.
Fig. 7
Relationship between MO percentage and L* values of the stone types
×
![]()
Initial relationships between SRI and other stone properties
As a result of the experiments carried out within the framework of the principles described above, initial optical properties (SR and TE values) of the samples were measured and these values were used for the calculation of initial SRI values of samples as given in Table 4. Basic relationships were analyzed to represent the interactions between surface parameters and the SRI values which were found to have statistically significant. Results varying in a very low testing error through the SR and TE measurements have tend to be low standard deviation for each sample due to homogeneity of the surface and nature of the measurement procedures. Different wind speeds mentioned above were also considered through SRI calculations with using Eq. 1 and the results were given in Table 4. Air temperature and wind conditions are one of the most important parameters affecting SRI values with respect to reflective characteristics of the surface under specified ambient conditions. Values of 5, 12, and 30 W/(m2K) are assigned to the convection heat transfer coefficient (h) corresponding to low (< 2 m/s), intermediate (between 2 and 6 m/s), and high (between 6 and 10 m/s) wind speeds, respectively (Costanzini et al. 2021). As can be seen from Table 4, in the case of increased wind speeds, SRI values also increase slightly. Considering the advantage brought by the wind condition, the following evaluations are based on the SRI values calculated according to h = 5 W/(m2K) where the wind condition is the lowest or stagnant. For this reason, SRI values of 30 different natural stone samples subjected to T-S cycles were calculated by using only the lowest (h = 5 W/(m2K) convection heat transfer coefficient. This is also consistent in terms of energy saving calculations, which also represent the wind condition during the peak period of solar radiation.
Table 4
Initial values of SRI and other properties of the samples
Sample Code | Optical Properties | SRI | Roughness (µm) | Gloss | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
b | a | ε | h = 5 W/(m2K) | h = 12 W/(m2K) | h = 30 W/(m2K) | Ra | Rq | Rz | ||
1 | 0.521 | 0.479 | 0.859 | 58.94 | 59.92 | 60.79 | 1.236 | 2.247 | 16.642 | 61.7 |
2 | 0.664 | 0.336 | 0.861 | 79.05 | 79.71 | 80.29 | 1.613 | 6.489 | 40.667 | 49.0 |
3 | 0.743 | 0.257 | 0.858 | 90.16 | 90.68 | 91.14 | 2.83 | 5.435 | 35.944 | 61.0 |
4 | 0.806 | 0.194 | 0.850 | 99.08 | 99.52 | 99.90 | 2.85 | 5.297 | 33.587 | 57.6 |
5 | 0.726 | 0.274 | 0.848 | 87.43 | 88.09 | 88.67 | 1.799 | 3.757 | 26.784 | 69.7 |
6 | 0.712 | 0.288 | 0.846 | 85.34 | 86.05 | 86.68 | 1.035 | 1.643 | 11.916 | 68.0 |
7 | 0.903 | 0.097 | 0.837 | 113.14 | 113.37 | 113.56 | 1.236 | 2.247 | 16.642 | 31.1 |
8 | 0.897 | 0.103 | 0.833 | 112.16 | 112.41 | 112.64 | 2.57 | 4.412 | 25.700 | 43.8 |
9 | 0.931 | 0.069 | 0.830 | 117.20 | 117.35 | 117.48 | 2.062 | 3.235 | 20.751 | 29.9 |
10 | 0.933 | 0.067 | 0.827 | 117.50 | 117.65 | 117.79 | 1.951 | 3.197 | 20.574 | 29.7 |
11 | 0.905 | 0.095 | 0.841 | 113.42 | 113.63 | 113.82 | 1.786 | 2.858 | 16.317 | 30.4 |
12 | 0.654 | 0.346 | 0.838 | 76.71 | 77.67 | 78.51 | 0.216 | 0.345 | 2.738 | 59.3 |
13 | 0.231 | 0.769 | 0.861 | 19.83 | 21.32 | 22.64 | 0.116 | 0.189 | 1.807 | 55.8 |
14 | 0.753 | 0.247 | 0.861 | 91.60 | 92.09 | 92.52 | 0.166 | 0.339 | 3.443 | 45.2 |
15 | 0.366 | 0.634 | 0.861 | 38.05 | 39.27 | 40.36 | 0.918 | 8.141 | 46.580 | 31.8 |
16 | 0.197 | 0.803 | 0.830 | 12.92 | 15.31 | 17.40 | 0.127 | 0.215 | 2.345 | 90.3 |
17 | 0.676 | 0.324 | 0.852 | 80.48 | 81.21 | 81.86 | 0.276 | 0.389 | 2.868 | 62.1 |
18 | 0.332 | 0.668 | 0.845 | 32.25 | 33.92 | 35.39 | 0.081 | 0.149 | 1.760 | 71.9 |
19 | 0.809 | 0.191 | 0.846 | 99.44 | 99.89 | 100.30 | 0.199 | 5.790 | 34.597 | 47.2 |
20 | 0.738 | 0.262 | 0.852 | 89.35 | 89.94 | 90.46 | 0.186 | 0.408 | 3.581 | 61.4 |
21 | 0.798 | 0.202 | 0.851 | 97.84 | 98.29 | 98.69 | 0.06 | 0.107 | 1.062 | 93.1 |
22 | 0.724 | 0.276 | 0.850 | 87.23 | 87.87 | 88.44 | 0.334 | 0.791 | 7.879 | 73.4 |
23 | 0.693 | 0.307 | 0.849 | 82.80 | 83.52 | 84.16 | 1.004 | 1.913 | 15.091 | 37.9 |
24 | 0.573 | 0.427 | 0.849 | 65.70 | 66.72 | 67.62 | 1.266 | 1.994 | 13.229 | 21.4 |
25 | 0.579 | 0.421 | 0.860 | 67.13 | 67.97 | 68.72 | 0.172 | 0.329 | 3.028 | 73.3 |
26 | 0.504 | 0.496 | 0.844 | 55.88 | 57.15 | 58.26 | 0.199 | 0.368 | 3.961 | 81.5 |
27 | 0.559 | 0.441 | 0.848 | 63.76 | 64.83 | 65.77 | 0.33 | 0.623 | 5.857 | 70.0 |
28 | 0.587 | 0.413 | 0.846 | 67.56 | 68.59 | 69.49 | 0.148 | 0.297 | 2.291 | 105.2 |
29 | 0.450 | 0.550 | 0.850 | 48.70 | 49.99 | 51.13 | 0.261 | 0.478 | 5.114 | 85.6 |
30 | 0.409 | 0.591 | 0.850 | 43.00 | 44.38 | 45.60 | 0.172 | 0.288 | 3.023 | 81.8 |
In that study, the lightness (L*) of the stone samples was chosen as the key parameter strongly correlated with SRI values with respect to other surface and colorimetric properties. There was found to be an exponentially increasing correlation between the mean SRI values for h = 5 W/(m2K) and the L* values of surfaces with the correlation coefficient as 0.74. (Fig. 8).
Fig. 8
Relationship between SRI and L* values of the stone types
×
![]()
Additionally, varying of SRI depend on the roughness and gloss values can be seen in Figs. 9 and 10. These figures demonstrate that, the change in these values have remarkable effects on SRI values, which increase as the roughness increase, whereas higher gloss values lead to decrease in SRI values of the tested stone types in a certain degree.
Fig. 9
Relationship between the SRI and the Ra values
×
![]()
Fig. 10
Relationship between SRI and the gloss values of the stone types
×
![]()
Results show that, in the case of the rougher surfaces resulted in significant increase in SRI value even up to 100 for 7–11 coded limestone samples, while for lower values of the Ra result in a decrease in target value down to 70 especially for 25–30 coded metamorphic originated stone types. This can be interpreted as the increasing of surface area is important for these stone types depending on their undulated rougher surface structure. As Levinson et al. (2010) concluded that, this phenomenon is closely related to global solar irradiance defined as the solar power per unit surface area incident on a surface from all directions, while global solar reflectance is defined as the fraction of this irradiance that is reflected. These results are also compatible with our results for after aging tests, that surfaces become rougher. Further studies on this issue would be necessary to verify and assess this observation.
Multivariate regression analysis
The univariate linear regression models were not exactly satisfactory to identify relationships among independent surface parameters and predicted target SRI value in the present study. Because of only one independent variable is used to predict the dependent variable, it was thought that the correlation is somewhat weak because of the low level of association between two variables depending on the probable error in using the correlation equation is high. In engineering research field, where there may be a greater contribution from complicating factors, high correlation coefficient of at least above 0.7 is preferred. The presence of outliers among the obtained test data, results in obtaining moderate correlation coefficient for the univariate linear regression analyses.
In this study, due to generalized SRI model allows to consider changes in such surface properties at the same time, a multivariate regression equation has been developed for the estimation of SRI value of the natural stones. In addition, direct methods measuring the SRI in laboratory is somewhat time-consuming, limited possession of the testing device and expensive. This model, given in Eq. 4 gives opportunity to possibility of using multiple surface features having importance to taken into account such as L*, Ra, and gloss values.
$$\:SRI=1.111{L}^{*}+9.889Ra-0.29Gloss+4.525\:(r\:=\:0.87)$$
(4)
The analysis of variance (ANOVA) was performed to test the significance of Eq. 4 (Table 5). This test follows an F-distribution with numerator degrees of freedom (df1 of 3) and the denominator degrees of freedom (df2 of 26) so that the critical region consists of the values exceeding 2.98 for a 95% level of confidence. Since the calculated F value for Eq. 4 was greater than the tabulated F value, the null hypothesis that there is no significant relationship between the SRI values of natural stones and other surface properties were rejected. The coefficient of determination (R2) values was also high for the empirical Eq. 4 for the 5% significance level. So, it can be concluded that these properties (L*, Ra, and Gloss) are significant properties for the stone material to predict the SRI values. Also, the relationships between experimental and calculated results were assessed based on the obtained Eq. 4 (Fig. 11). For assessing the validity degree of the results, the 45° line (\(\:y=x\)) has been plotted in each coordinate system. According to this statistical evaluation, developed multi-regression model determined in this study could adequately estimate the SRI value of natural stones whereas rough estimates could be made with simple regression models.
Table 5
Some statistical parameters for the multivariate regression analyses
Model | Coefficient of Determination (R2) | Standard Error of Estimate | Fcalculated | Ftabulated | Sig. |
|---|---|---|---|---|---|
Equation (4) | 0.753 | 14.94 | 26.453 | 2.98 | 0.000 |
Fig. 11
A comparison between measured and predicted SRI values from Eq. (4)
×
![]()
Changes in SRI and other stone properties after T-S cycles
Variations in roughness (Ra, Rq, Rz), gloss, optical properties and SRI values caused by T-S cycles are summarized in Table 6. In addition, Ra and gloss changes of the samples before and after T-S cycles were given in Figs. 12 and 13.
Table 6
SRI and other properties of the samples after T-S cycles
Sample Code | Optical Properties | SRI | Roughness (µm) | Gloss | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
b | a | ε | h = 5 W/(m2K) | h = 12 W/(m2K) | h = 30 W/(m2K) | Ra | Rq | Rz | ||
1 | 0.538 | 0.462 | 0.857 | 61.31 | 62.27 | 63.13 | 1.415 | 2.443 | 17.685 | 43.8 |
2 | 0.749 | 0.251 | 0.860 | 91.13 | 91.62 | 92.06 | 1.898 | 8.144 | 48.859 | 33.2 |
3 | 0.750 | 0.250 | 0.859 | 91.19 | 91.69 | 92.14 | 3.168 | 5.866 | 39.059 | 15.6 |
4 | 0.814 | 0.186 | 0.849 | 100.21 | 100.63 | 101.00 | 2.904 | 5.218 | 35.743 | 34.1 |
5 | 0.773 | 0.227 | 0.858 | 94.52 | 94.98 | 95.39 | 2.121 | 4.225 | 31.167 | 55.2 |
6 | 0.738 | 0.262 | 0.850 | 89.24 | 89.84 | 90.38 | 0.868 | 1.238 | 9.493 | 46.4 |
7 | 0.905 | 0.095 | 0.850 | 113.50 | 113.68 | 113.85 | 1.849 | 6.150 | 37.378 | 18.5 |
8 | 0.897 | 0.103 | 0.841 | 112.23 | 112.46 | 112.67 | 3.09 | 4.867 | 30.596 | 30.4 |
9 | 0.925 | 0.075 | 0.838 | 116.45 | 116.60 | 116.74 | 2.523 | 4.147 | 25.280 | 19.6 |
10 | 0.916 | 0.084 | 0.838 | 115.04 | 115.22 | 115.39 | 1.924 | 3.082 | 20.176 | 20.4 |
11 | 0.906 | 0.094 | 0.848 | 113.69 | 113.88 | 114.04 | 2.157 | 3.224 | 19.682 | 23.4 |
12 | 0.656 | 0.344 | 0.850 | 77.53 | 78.34 | 79.06 | 0.224 | 0.364 | 3.373 | 51.6 |
13 | 0.236 | 0.764 | 0.860 | 20.41 | 21.91 | 23.24 | 0.182 | 0.304 | 2.730 | 40.9 |
14 | 0.754 | 0.246 | 0.855 | 91.67 | 92.19 | 92.65 | 0.705 | 0.534 | 5.198 | 36.1 |
15 | 0.372 | 0.628 | 0.860 | 38.65 | 39.89 | 40.99 | 1.357 | 8.624 | 45.499 | 25.2 |
16 | 0.198 | 0.802 | 0.854 | 14.91 | 16.65 | 18.19 | 0.221 | 0.384 | 3.773 | 57.9 |
17 | 0.687 | 0.313 | 0.859 | 82.25 | 82.88 | 83.44 | 0.166 | 0.273 | 1.888 | 40.7 |
18 | 0.329 | 0.671 | 0.851 | 32.20 | 33.74 | 35.10 | 0.16 | 0.292 | 2.975 | 40.3 |
19 | 0.809 | 0.191 | 0.854 | 99.57 | 99.98 | 100.34 | 0.204 | 0.775 | 6.146 | 43.5 |
20 | 0.752 | 0.248 | 0.856 | 91.40 | 91.92 | 92.38 | 0.332 | 0.639 | 5.572 | 33.1 |
21 | 0.805 | 0.195 | 0.850 | 98.86 | 99.30 | 99.69 | 0.52 | 0.729 | 5.645 | 36.3 |
22 | 0.776 | 0.224 | 0.842 | 94.58 | 95.15 | 95.65 | 0.434 | 0.781 | 7.175 | 65.9 |
23 | 0.691 | 0.309 | 0.848 | 82.37 | 83.12 | 83.78 | 1.31 | 2.219 | 15.510 | 30.9 |
24 | 0.576 | 0.424 | 0.851 | 66.23 | 67.21 | 68.07 | 0.957 | 1.522 | 10.573 | 30.1 |
25 | 0.733 | 0.267 | 0.855 | 88.65 | 89.23 | 89.74 | 0.334 | 0.642 | 5.925 | 46.6 |
26 | 0.623 | 0.377 | 0.855 | 72.97 | 73.79 | 74.52 | 0.214 | 0.683 | 5.305 | 59.0 |
27 | 0.623 | 0.377 | 0.851 | 72.86 | 73.73 | 74.50 | 0.497 | 0.868 | 6.950 | 30.9 |
28 | 0.695 | 0.305 | 0.845 | 82.87 | 83.64 | 84.32 | 0.337 | 0.582 | 4.800 | 80.1 |
29 | 0.590 | 0.410 | 0.841 | 67.86 | 68.94 | 69.90 | 0.255 | 0.326 | 3.266 | 68.6 |
30 | 0.508 | 0.492 | 0.859 | 57.19 | 58.19 | 59.07 | 0.357 | 0.568 | 4.630 | 41.4 |
Fig. 12
Ra change of samples before and after T-S cycles
×
![]()
Fig. 13
Gloss change of samples before and after T-S cycles
×
![]()
As can be seen from the figures that, the decrease in gloss and so increase in roughness values for almost all stone types is noticeable in samples subjected to T-S cycles which lead to an increase in roughness values and in turn to a certain decrease of the gloss due to the formation of irregular surface structure. These results demonstrate that, the interaction of stone surface both with water cyclically and high temperatures regarding to T-S cycles has remarkable effects on the micro-degradation of sample surface results with an increase in roughness together with the probable loss of gloss.
Figure 13 indicated that, the gloss unlike the roughness values of the stone samples appears to be significantly decreased and more matt surfaces were occurred after the T-S cycles. At the end of T-S cycles, while the highest gloss loss values were obtained for 3, 21 and 30 coded samples, lowest variation in gloss loss values were observed for 12, 22 and 23 coded samples. The reason for this differentiation may be the 3, 21 and 30 coded stone samples had the most fractured and degraded surface structure due to sudden temperature changes which increases the number of fractures and pores among all samples. The reason for 12, 22 and 23 coded stone samples having a lower gloss loss value after T-S cycles is that these samples had stronger homogeneous structure with lower porosity values.
The thermal response of a surface due to the combined effects of SR and TE is often given through the SRI values. SRI was calculated by using values of measured SR and TE of the stone slab samples before and after the T-S cycles. Through these measurements, SRI values were calculated for h = 5 W/(m2K) and the column graphs were plotted in Fig. 14.
Fig. 14
Variation of SRI values of samples before and after T-S cycles
×
![]()
In general, the results demonstrate several change levels in SRI values for almost all stone types in parallel with the increase in roughness and so decrease in gloss values due to their thermally-aged surfaces. Although an increasing trend can be observed in SRI values of travertine group as 1 to 6 coded samples, SRI values of limestone group as 7 to 21 coded samples and dolomitic limestone group as 22 to 24 coded samples were slightly higher than that of initial states. This can be explained by their unique mineralogical composition and compact structure with less porosity causing them less susceptible to T-S cycles unlike travertine group. The results demonstrated that most significant change was observed especially for the 25–30 coded re-crystallised samples that can be interpreted as the more degradation effect of T-S cycles on these stone types. As Berdahl et al. (2008) stated that the first reason for this differentiation could be the probability of multiple scattering phenomena due to increasing of roughness altering the SR of these stone types more evidently. The results of our study confirm and further support this concept that the roughness increases the irregularities on the surface. This undulated surface structure will increase the possibility of light encountering the sample surface repeatedly which causes the sample to absorb a portion of the incident light multiple times. As a result, the SRI values of the sample would be a function of the roughness causing micro-reflectance by multiple scattering as described above.
The second cause could be the color differentiation after T-S cycles that partly influence both SR and TE of the surfaces of these stone types. It is well known that the color of a sample effects the SR and many studies focus on altering the surface to be white will cause considerable changes in resultant heat flux. Parker et al. (1996) found that by resurfacing the roof and coloring it white, the heat flux entering the building was decreased by approximately 20%. Another study conducted by Berdahl and Bretz (1997) showed a direct correlation between the L* and the SR values of materials.
SRI variations in terms of cyclic T-S
Alchapar et al. (2013) defined the aging response of the materials in terms of the changes in SRI values quantitatively as stated in Eq. (5). In order to examine the effect of T-S cycles on the changes in SRI values, this equation was used for the comparison of the SRI values before and after T-S test of the slab samples.
$$\:\varDelta\:SRI\:={SRI}_{2}-{SRI}_{1}$$
(5)
Where \(\:\varDelta\:SRI\) is the difference between \(\:{SRI}_{1}\) and \(\:{SRI}_{2}\) values, \(\:{SRI}_{1}\) is the initial SRI value, \(\:{SRI}_{2}\) is the SRI value of aged materials.
According to this approach, it is preferable that the initial SRI value (\(\:{SRI}_{1}\)) will be lower than that of the aged material (\(\:{SRI}_{2}\)), when considering the interaction of a certain material with the surrounding medium. As would be expected, if \(\:{SRI}_{1}\) values are higher than \(\:{SRI}_{2}\), the urban temperatures will increase and thus contributes to the formation of heat islands after a certain period of time. The classification of the \(\:\varDelta\:SRI\) values for aged materials is as follows (Alchapar et al. 2013). When;
\(\:\varDelta\:SRI<-5\) is Degraded,
\(\:-5\le\:\varDelta\:SRI\le\:+5\) is Stable,
\(\:\varDelta\:SRI>+5\) is Improved.
In this study, \(\:\varDelta\:SRI\) values of 30 different natural stone types are given graphically in Fig. 15 by taking into account the changes in SRI values before and after the T-S cycles. It can be seen from the Fig. 15 that 2, 5, 22 and 25–30 coded natural stone samples (majority of metamorphic origin) exhibit certain improvement with \(\:\varDelta\:SRI>+5\) after T-S cycles, while all remaining samples (majority of sedimentary origin) exhibit stable behaviour with \(\:\varDelta\:SRI\le\:\pm\:\:5\) according to classification of the \(\:\varDelta\:SRI\) given above. There were no samples in a degraded scale with \(\:\varDelta\:SRI<-5\).
Fig. 15
Variation of SRI values (ΔSRI) of the samples before and after T-S cycles
×
![]()
Conclusions
This study is primarily focused on some optical and surface features such as SR, TE, gloss, Ra, and L* values of the 30 different carbonate-based natural stone types, due to the fact that the stone properties have significant importance on SRI values that measure the ability of them to reflect incident solar radiation from the surface. It can be concluded that the selection of such light-colored dimension stone types as facade material for outdoor urban applications would contribute both to optimize the energy demands in buildings and so to decrease the formation of urban heat islands in hot climatic conditions.
-
L* value of the stone slab samples was determined as the key parameter strongly correlated with SRI values and there was found to be a clear correlation between SRI and L* values. On the other hand, L* values are negatively affected by the increasing values of Fe, Al and Mn oxides tend to diminish SRI values depending on their lower L* values due to the existence of darker colors, as expected.
-
Multivariate regression analysis was used to predict SRI values from other stone properties because of the mentioned relationships above have meaningful correlations with each other. Multivariate regression analysis reveals that, SRI values can be predicted from these significant surface properties with a high degree of accuracy (\(\:r=0.87\)).
-
Degradation of materials initiated by cyclic T-S was also investigated for the selected stone types. Changes in SRI values under the effect of T-S cycles reveal that, the cyclic exposure to process of cyclic T-S aging mostly caused further improving in SRI in terms of their modified surfaces. Results clearly demonstrate that, SRI values have been increased with the increasing Ra values unlike gloss values of the natural stones before and after T-S cycles. This may be due to the fact that both the natural stone surfaces become more undulated with higher surface area, causing the probability of multiple scattering phenomena by absorbing a portion of the incident light multiple times and color differentiation tends to be lighter after the physical degradation process.
-
While there were no samples in a degraded scale with \(\:\varDelta\:SRI<-5\), it was determined that, two travertine samples, one dolomitic limestone and all metamorphic originated re-crystallised limestone groups exhibit certain improvement with \(\:\varDelta\:SRI>+5\), and all remaining samples exhibit stable behaviour with \(\:\varDelta\:SRI\le\:\pm\:5\) after T-S cycles according to classification of the \(\:\varDelta\:SRI\) given above.
Acknowledgements
This research was funded by TUBITAK-1001 “The Scientific & Technological Research Council of Turkey” (no. 114M569). We are grateful to the institution for their valuable support and cooperation during field work.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
