- Email: [email protected]
Thermal effects on the physical properties of limestones from the Yucatan Peninsula
International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
Contents lists available at ScienceDirect
International Journal of Rock Mechanics & Mining Sciences journal homepage: www.elsevier.com/locate/ijrmms
Technical Note
Thermal effects on the physical properties of limestones from the Yucatan Peninsula W.S. González-Gómez a,n, P. Quintana a, A. May-Pat b, F. Avilés b, J. May-Crespo a, J.J. Alvarado-Gil a a b
Applied Physics Department, Cinvestav-Unidad Mérida, Carr. Ant. a Progreso km 6, C.P. 97310 Mérida, Yucatán, México Centro de Investigación Científica de Yucatán, Unidad de Materiales, Calle 43, No.130, Col. Chuburná de Hidalgo, CP 97200 Mérida, Yucatán, México
art ic l e i nf o
a b s t r a c t
Article history: Received 22 July 2014 Received in revised form 10 October 2014 Accepted 29 December 2014 Available online 21 February 2015
The effect of thermal degradation on the compression strength, ultimate compression strain, color and mass loss of four limestones extracted from the Yucatan Peninsula was studied. The samples sets were independently subjected to 25, 100, 200, 300, 400, 500 and 600 °C during one hour for each temperature. The thermal degradation of the investigated rocks and their chemical composition, microstructure and porosity is also investigated. The results show a correlation between the mechanical properties, thermal behavior, the presence of minerals and porosity in each type of limestone rock. The variation of the limestones color with temperature was described in terms of standard space color parameters and their diffuse reflectance. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Limestones Thermal heating Compressive properties Mineralogy
1. Introduction The historical monuments and archeological sites in the Maya area comprising southern Mexico, the central part of Honduras and Guatemala, are constructed with different types of limestone, with calcium carbonates (CaCO3) as the basic constituent. Currently in the Yucatan Peninsula there are a variety of historical constructions and archeological sites built with stones. The Peninsula of Yucatan is a smooth carbonate platform developed from Cretaceous, Tertiary and Quaternary limestones, with predominantly shallow calcareous soil [1], containing different grain sizes, porosities, and texture. Also, many modern buildings use limestone as a primary construction material. Limestones can be obtained from a variety of sources and may differ considerably in their chemical compositions and physical microstructure. The chemical reactivity show a large variation due to their differences in crystalline structure and the nature of impurities (silica, iron, magnesium, manganese, sodium, and potassium, to name a few) [2]. The study of changes in the physical properties of rocks as a function of thermal excursions is relevant to various engineering and industrial applications. Rocks are composed of minerals, bounding matrix, cracks and pores, and heat may produces important physical and chemical changes in their microstructure [3]. Temperature
n
Corresponding author. E-mail address: [email protected] (W.S. González-Gómez).
http://dx.doi.org/10.1016/j.ijrmms.2014.12.010 1365-1609/& 2015 Elsevier Ltd. All rights reserved.
changes can induce micro-cracks in the rocks due to the mismatch in the thermal properties of the different mineral compositions (intergranular) or within grains (intragranular) [4]. Temperature gradients can affect the rock dilation, changing the coefficients of thermal expansion [5]. Qualitative changes produced by heat can also occur when the rocks undergo a phase transition such as the α/β phase transition in quartz [6]. One of the traditional forms of tropical agriculture in the Yucatan peninsula is the “slash-and-burn” system that is applied every year, during the dry periods [7,8]. Temperature measurements on these traditional practices can reach non-uniform high temperatures from 300 1C up to 700 1C [9]. These repeated burnings and extreme heating changes can lead to long term cumulative thermal effects on the rock properties and ecosystem. This is because the modifications of the mechanical properties of limestones have a strong influence on the soil conformation and induce erosion, promoting water infiltration, increased permeability and porosity, which are one of the major problems in these regions. Temperature variations are one of the major factors influencing the intrinsic properties of rocks [10,11]. It has been shown that the variation of expansion increases with temperature [3], which is primarily responsible for the changes in the microstructure of rocks, inducing new fractures and microfractures. A common form of chemical action caused by temperature is the oxidation of iron, which provokes significant color changes even in the presence of minor amounts of iron (ferrous minerals) in the calcareous matrix. For example, red coloration is due to the presence of hematite, whereas less common yellows and browns generally result from
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
2. Material and methods 2.1. Description of rock samples The four types of rocks were selected from quarries of the Yucatan State, Mexico, which are used in the construction industry. These materials were selected considering their regional significance, abundance and frequency of usage as building and restoration materials [18]. The rocks were named “XCA”, “CH”, “LB”, and “LR” after their Spanish acronyms. The sample XCA is a limestone with shades ranging from white to beige, inlaid with shells and fossils of snails with high porosity. CH is a limestone with different pale beige shades and high porosity. LB is a white compact limestone with low porosity, whereas the sample LR is a red compact limestone with low porosity. The physical properties of the four limestone samples such as bulk density, water absorption, water content and effective porosity, were determined by water immersion based on mass difference and according to the ASTM standard D2216-98 [19] with three repetitions for each rock. For this purpose, a set of 12 cubic limestone samples were cut into coupons of 1 cm3 obtained employing a low-speed diamond saw (Buehler Isomet). 2.2. Thermal treatment Seven sets of rocks containing ten replicated limestone coupons were cut with a low-speed diamond saw (Buehler Isomet) for each temperature. The limestone coupons for thermal treatment comply with the dimensions set for (subsequent) compression testing with average dimensions of 21 mm 10.5 mm 10.5 mm, having a total of 280 prismatic coupons. The samples were polished with sandpaper in order to have a flat surface finish, and then cleaned using a sonicating bath. The samples were then washed with a 2:1 (v/v) mixture of water:acetone (J.T. Baker, 99.8%); afterwards they were left to dry at room temperature and set inside of an oven at 80 1C during 24 h. Each set of samples were subjected to one of the six thermal treatments investigated with the targeted temperature fixed at 100, 200, 300, 400, 500 and 600 1C, using a heating rate of 3 1C/min to ensure thermal equilibrium of the limestone coupons. The targeted temperature was reached sequentially (step-wise) in steps of 100 1C maintained during 1 h, as depicted in Fig. 1. Additionally, a set of samples was kept at 25 1C to act as a blank. Once the targeted
ramp 6
600 ramp 5
500
Temperature (°C)
limonite and goethite, respectively. The green-gray to black color in rocks is related to total organic carbon where darker colors corresponding to higher carbon content; this relationship has been confirmed on sediments [12]. The presence of kaolinite and smectite provide white to light neutral colors. Compounds of other transition metals (Ti, Mn, Co, Cu and Zn) may also show an influence of the pigment [13]. Also, the color in the rocks changes when heated at temperatures above 250–300 1C, which correspond to the dehydration of iron compounds. Pale brown colored rock change to reddish brown upon heating, which may not be apparent until the stones haves reached temperatures above 400 1C when goethite transforms to hematite [4,14,15]. Several investigations on the weathering mechanisms and degradation kinetics of Yucatan limestones have been reported [16,17], but little is known about the effect of thermal degradation on their mechanical properties. In this contribution, four types of limestone rocks, collected from the state of Yucatan (Southern Mexico), were investigated to elucidate correlations between rock compressive properties, mineral transformations and color change with thermal treatments at different temperatures as an attempt to simulate temperature changes similar to those observed in the “slash-and-burn” traditional systems in Yucatan.
183
ramp 4
400 ramp 3
300 ramp 2
200 100 0
ramp 1
0
200
400
600
800
1000 1200 1400 1600 1800
Time (min) Fig. 1. Temperature treatment performed on the rocks before the compression tests. Heating was performed at a rate of 3 1C/min. The temperature steps (ramps) correspond to 100 1C, 200 1C, 300 1C, 400 1C, 500 1C and 600 1C.
temperature was reached, the samples were maintained during one hour at the corresponding temperature and the furnace was then cooled down slowly to room temperature in order to avoid sample cracking if suddenly are cooled. For example, for the test at 400 1C, the heat treatment followed the four initial ramps depicted in Fig. 1 requiring a total time of 16 h. 2.3. Sample characterization The composition of specimens was analyzed using the powder X-ray diffraction technique (XRD Siemens D-5000) with a BraggBrentano geometry, a Cu-Kα radiation (λ ¼1.5418 Å), scans with a step size of 0.021, an integration time of 10 s and with a 2θ angular range of 51r 2θr 501. Thermogravimetric analysis (TG) and the derivative of the thermogram (DTG) were recorded in a Discovery TGA (TA Instruments). Powder samples (2 mg) were heated from 50 to 1000 1C applying a heating rate of 10 1C/min under an inert atmosphere (N2). Compression tests were conducted in a Shimadzu AGI-100 universal testing machine equipped with load cells of 5 kN and 100 kN and using a cross-head speed of 0.05 mm/min according to the ASTM standard D3148 [20]. The prismatic coupons for compression testing had a length of 21 mm and 10.5 mm 10.5 mm cross-sectional area and ten replicates were tested for each thermal treatment. To measure the reflectance spectra of the limestone materials as a function of the thermal treatment, the samples were illuminated with a deuterium-halogen light source (AvaLight DH-S-BAL) using an optical fiber; the light reflected was then collected using an integrating sphere (Labsphere USRS-99-010) and sent to a spectrometer (AvaSpec-2048) equipped with an optical fiber. The reflectance spectra were measured from 400 nm to 1000 nm. The color change was determined by applying the model of the Commission Internationale de l'Eclaraige (CIE), using as a standard illuminant D65 under an angle of 101 [21,22]. The color was described in terms of Lnanbn space color by CIE-LAB and CIELCH systems (CIE, 1976) [23], where the value of Ln represents lightness (Ln ¼0: black y Ln ¼ 100: white), and an and bn chromaticity (þ an ¼red and -an ¼green; þbn ¼yellow and bn ¼blue) [24,25].
3. Results and discussion 3.1. Physical properties of the limestone rocks The physical properties of the investigated limestone rocks such as bulk density and effective porosity were determined using
184
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
the saturation and buoyancy techniques, as recommended by ASTM [19]. The value of the bulk density of the four rocks is between 1.98 g/cm3 and 2.36 g/cm3 (Table 1), which are lower than other values reported for limestone rocks [16,26,27]. A comprehensive study of porosity can provide valuable information to determine whether a given type of rock is susceptible to thermal stress or not. The effective porosity percentages for the rocks LB and LR are similar to each other with values of 6.02% and 4.05%, respectively contrasting with the rocks XCA and CH, which have higher effective porosity values (20.1% and 18.6%, respectively) (Table 1). Water retained by the rocks at 25 1C, is 12.2% and 11.6% for limestone types XCA and CH, respectively, which is about three times more than that retained by rocks LB and LR with 3.96% and 2.38%, respectively (Table 1). The highest water content was for the CH rock with 1.06% as compared to the other limestone samples ranging from 0.14% to 0.22%.
Table 1 Physical properties of four limestone samples. Rock name
Bulk density (g/ Water content Water cm3) (%) absorption (%)
Effective porosity (%)
XCA CH LB LR
2.00 70.12 1.98 70.04 2.36 70.10 2.30 70.15
20.17 1.50 18.6 7 1.12 6.02 7 1.35 4.05 7 0.58
0.22 7 0.08 1.067 0.07 0.147 0.02 0.167 0.02
12.2 70.95 11.6 70.80 3.96 70.70 2.38 70.48
3.2. Mineral composition of the limestone rocks XRD analysis of the limestones samples was performed to determine the mineral composition of the different lithotypes. The X-ray diffraction patterns for the powdered samples showed mainly the presence of calcite (CaCO3), except for limestone type CH which presented another polymorph of calcium carbonate, aragonite. To detect the presence of other minerals, the powders were treated with 5% HCl (v/v) in order to dissolve all carbonates. The XRD powder patterns of the investigated rocks after treatment with HCl (Fig. 2), show the presence of quartz (SiO2) in all samples, as an essential component, which is directly related to some physical properties such as hardness and strength [28]. Some clays were also identified in all the limestones rocks such as tosudite (Na0.5(Al,Mg)6(Si, Al)8O18(OH)12∙5H2O); kaolinite (Al2Si2O5(OH)4) and montmorillonite (Ca0.2(Al, Mg)2Si4O10(OH)2∙4H2O) was additionally detected in XCA and LB. Anatase (TiO2) was identified in all samples except for LR. Orthoclase (KAlSi3O8) was found in CH and LB. Iron oxyhydroxide as goethite (FeO(OH)) and hematite (Fe2O3) were present in LR. 3.3. Effect of temperature on the compression properties Representative stress–strain (σ ε) diagrams chosen from the ten replicates of the compressive tests for each thermal treatment and rock type, are shown in Fig. 3. Even though the thermic influence on the behavior of the limestone is difficult to quantify, it was possible to observed, despite the relatively high standard deviation (inherent
Q
Relative Intensity (a.u.)
Relative Intensity (a.u.)
T Q Q A Q T
M K
5
10
15
20
25
30
35
40
45
50
TQ
10
O T
K
5
Q A
T
15
20
25
30
35
40
45
50
45
50
T
Relative Intensity (a.u.)
Relative Intensity (a.u.)
G
Q M
M
K
5
10
15
A
20
25
O T
30
35
40
45
50
T Q
G
K
5
10
15
G H
T
20
25
30
T
35
G
40
Fig. 2. XRD patterns of limestone coupons after the dissolution of carbonates: (a) XCA, (b) CH, (c) LB and (d) LR. The observed phases were: quartz (Q), kaolinite (K), tosudite (T), anatase (A), goethite (G), montmorillonite (M), hematite (H) and orthoclase (O).
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
strength with a decreasing trend as temperature was increased (Fig. 4a). On the other hand, the ultimate strain (εmax) is lower for LB and LR (Fig. 4b). The variations of such ultimate strain with the thermal treatments are within the experimental scattering for XCA, LB and LR, but tend to increase for CH when the temperature increases (Fig. 4b). This behavior indicates that LB and LR rocks have a significantly higher strength and are more brittle than XCA and CH specimens according to ISRM [29].
to the natural heterogeneity of rocks), important trends defining the mechanical behavior of the investigated rocks. The analysis of the stress–strain diagrams shows significant differences between the four rocks (Fig. 4). The group of rocks (LB and LR) showed the highest compression strength (σmax), and such strength was maintained for all thermal treatments (Fig. 4a). The group of rocks with more inlaid shells and other fossils, XCA and CH, showed lower compression
15
35
25°C 100°C 200°C 300°C 400°C 500°C 600°C
10
185
25°C 100°C 200°C 300°C 400°C 500°C 600°C
30 25 20 15
5
10 5
0
0 0.0
0.5
1.0
1.5
2.0
80
60 50 40
0.5
1.0
1.5
2.0
100
25°C 100°C 200°C 300°C 400°C 500°C 600°C
70
0.0
2.5
25°C 100°C 200°C 300°C 400°C 500°C 600°C
d) LR 80
60
40
30 20
20
10 0
0 0.0
0.5
1.0
1.5
2.0
0.0
0.5
1.0
1.5
2.0
Fig. 3. Representative compressive stress–strain (σ-ε) diagrams of the limestone rocks at different heat levels: (a) XCA, (b) CH, (c) LB and (d) LR.
8
2.5
100
XCA
90
CH
LB
LR
XCA
CH
LB
XCA
LR
CH
LB
LR
7
80 6
2.0 70
5
60
4
1.5
50 40
3
30 2
1.0 20
1
10
0
0.5
0 25
100
200
300
400
Temperature (°C)
500
600
25
100
200
300
400
Temperature (°C)
500
600
25
100
200
300
400
Temperature (°C)
Fig. 4. Values of compressive stress–strain (σ-ε) diagrams of the limestone rocks at different heat levels: (a) σmax, (b) εmax and (c) E.
500
600
186
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
The elastic modulus (E) at 25 1C (Fig. 4c), is also significantly higher for the more brittle rocks (LR and LB) than for XCA or CH cases. There is a trend of decreasing E as temperature increases for XCA and CH, and in less proportion for LB; however, this behavior was not observed for LR which displayed an increase in the
modulus of elasticity as for the case-hardening argument presented in [30] in a study of heating effect on sandstone, limestone and granite rocks. These results show a good correlation with the relative porosity values summarized in Table 1. Because temperature influences the porosity and mechanical properties of rocks
Fig. 5. Photographs of fractured rocks after the compression testing.
TG (%)
-0.05 70 Crytal (OH)
60 50
- 26.4%
-0.10
Organic matter + Water (OH)
200
300
400
500
600
700
800
Water (OH)
-0.2 50 Organic matter
40
900
100
200
300
Temperature (°C) 0.04
90
0.02
40
- 8.5%
400
500
-0.3 900
0.04 - 5.2%
0.02
600
Temperature (°C)
700
800
90
- 4.3%
-0.10 900
0.00
85
80
75
-0.02
- 5.8%
-0.06 -0.08
300
800
0.06
95
TG (%)
-0.02 -0.04
Organic matter + Water (OH)
200
700
100
DTG (%/min)
TG (%)
- 9.6%
100
600
Goethite
-0.04
Organic matter + Water (OH)
-0.06
Hematite 100
200
300
400
500
600
Temperature (°C)
Fig. 6. TG-DTG curves of limestone: (a) XCA, (b) CH, (c) LB and (d) LR.
700
800
-0.08 900
DTG (%/min)
Crystal (OH)
70
50
500
0.00
- 20.4%
60
400
Temperature (°C)
100
80
-0.1
- 9.9%
70 60
-0.20 100
0.0
80
-0.15
40
- 22.2%
DTG (%/min)
80
0.1
- 2.6%
90
0.00
- 18.4%
TG (%)
90
100
0.05
- 3.8%
DTG (%/min)
100
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
[31,32], the rocks LB and LR with significantly lower porosity have a higher elastic modulus and compressive strength and are more brittle. The stress–strain an elastic modulus behavior for LR and LB rocks are similar to the mechanical properties obtained on limestones from China [33,34]; however, in this work the obtained values for the σmax and E, at each temperature are lower, than the reported. In order to visualize better these mechanical behaviors, Fig. 5 displays pictures of the rocks upon fracture for representative limestone coupons treated at different temperatures. Overall, rocks LR and LB show a single crack running along the loading direction (vertical) or slightly inclined, which indicates failure dominated by normal stresses and typically found in brittle materials. Rocks CH and XCA, on the other hand present cracks inclined at 25 1C, with the loading direction (vertical), indicating influence of shear stresses in the failure mode. As the temperature treatment increases, the influence of shear stresses in the failure mode of CH and XCA becomes more prominent, as indicated by the inclination of the main crack within the coupon. For temperatures above 400 1C, the rocks CH and XCA pulverize upon failure (which is not the case for LR and LB,) indicating the lack of molecular cohesion of these rocks at such high temperatures. 3.4. Effect of temperature treatment on thermogravimetric analysis The curves representing the thermogravimetric analysis (TG, left vertical axis) and the derivative of the thermogram (DTG, right vertical axis) for the four natural limestone types in Fig. 6, initially show a small mass loss due to the adsorbed water ( 200 1C), followed by a loss of CO2 from calcium carbonate (730 1C) with a
total mass loss of 40–45%. These values are lower than those reported in the literature on calcination kinetics and surface area of dispersed limestone particles, which are around 45–55% [35,36]. The same analysis (TG/DTG) was applied to the limestones after the dissolution of carbonates (Fig. 6). The DTG curve of XCA initially shows the loss of adsorbed water on clays and the elimination of organic matter at 330 1C, with a mass loss of 22.2%. Thereafter, the loss of structural water is observed in the temperature range from 380 1C up to 650 1C [37], with a mass loss of 26.4%. In the case of the CH rock the behavior was similar to XCA, and the mass loss was 24.8%. However, a rapid mass loss in the shorter temperature range of 240–400 1C, with a maximum at 330 1C, is observed as this rock has higher organic matter produced by the marine microorganisms which are inherent to the rock matrix [38,39]. The second loss of mass is 9.9% and ends at 500 1C, ascribed to dehydroxylation of clays, due to the presence of kaolinite. The thermal behavior of LB shows two mass losses within the temperature range of 230–530 1C which correspond to the elimination of organic matter and the removal of the hydroxyl groups between the interlayers of the mixed clays (kaolinite, tosudite and montmorillonite) with a total mass loss of 30%. Subsequently, a mass loss of 8.5% in the temperature range of 550–850 1C (with an inflection at 705 1C) corresponds to the elimination of the structural water from the aluminosilicates. The DTG curve of LR during the heating process, the temperature range for the removal of the water is 25–150 1C, following by a mass loss at 260 1C that corresponds to the dehydroxylation of goethite which is transformed into hematite during the heating process; the transformation temperature usually depends on the
100
100
100°C
200°C 80
80
100°C
70
Reflectance (%)
Reflectance (%)
25°C
90
25°C
90
300°C 400°C
60 50
500°C
40
600°C
30
187
500
600
700
800
70
400°C
60 50 40
500°C
30
600°C
20
900
200°C 300°C
500
600
Wavelength (nm)
700
800
900
800
900
Wavelength (nm)
90
80
25°C 80
70
100°C
25°C 100°C
70
Reflectance (%)
Reflectance (%)
300°C 400°C 200°C 60
500°C 50
60
300°C 200°C
50
400°C 500°C
40
600°C
600°C 40 30
30
500
600
700
Wavelength (nm)
800
900
20
500
600
700
Wavelength (nm)
Fig. 7. Diffuse reflectance spectra variation with thermal treatment for different limestone: (a) XCA, (b) CH, (c) LB and (d) LR.
188
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
crystallinity and the substitution of iron by aluminum [40,41]. The second mass loss at 430 1C is related to the removal of organic matter and the OH from the clays, with a total mass loss of 15%.
3.5. Effect of temperature on reflectance spectra The diffuse reflectance spectra of the samples after each thermal treatment were obtained in the visible region (Fig. 7). Initially, the reflectance spectra of the samples were similar to each other the temperature range of 25 to 200 1C. However, regardless of the limestone type, a strong decrease in the reflectance (from 10 up to 30%) is observed at higher temperatures. The decrease in reflectance is due to the color change of the limestone as the color becomes darker with the increase in temperature. Furthermore, the spectra of LR show changes at 580 nm at low temperatures (25–200 1C) due to the presence of the two iron oxides (goethite and hematite) as shown in Fig. 7d. At higher temperatures only the hematite signal is observed because the transformation of goethite to hematite occurs at 260 1C, which is also observed in the TG/DTG analysis. The parameterization of the spectra applying the function of Kubelka-Munk (K-M) allows identifying and quantifying the iron oxide minerals as shown by Scheinost's studies of soils [42]. The reflectance values are used to calculate the absorption and scattering coefficients, which corresponds to the two phenomena that occur in the interaction between the light-sample processes. Because the transformation of the derivative is particularly sensitive to high frequency instrumental noise, an appropriate filtering procedure was applied using the Savitzky-Golay function [43].
Hematite Goethite
-4
F(R) 2nd Derivative (x10 )
25°C 100°C 200°C 300°C 400°C 500°C 600°C
LR
2.0
1.5
1.0
0.5
0.0 300
350
400
450
500
550
600
650
700
Wavelength (nm) Fig. 8. Second derivative of the LR sample, showing the bands of hematite and goethite.
The second-derivative from the spectra reflectance of the iron oxides show four bands in the range from 350 to 1030 nm according to Sherman and Waite [44], due to three single electron transitions: 4t1’6A1 at 870–1030 nm, 4T2’6A1 at 650–710 nm, and 4E; 4A1’6A1 at 405–410 nm as well as the electron pair transition (EPT) at 488–493 nm. The EPTs are the most intense absorption edges which, in turn, are closely related to the hue. The EPTs for the hematite occur around 521–565 nm, clearly separated from the EPTs related to yellow iron oxides at 479–499 nm. The second derivative of the reflectance spectra was applied only to LR limestone in the range from 400 to 1000 nm. The iron oxides bands can be identified in Fig. 8, which changes according to the temperature. Hematite displays two signals at 538 nm and 423 nm that gives the red characteristic color of the rock and the signals of goethite at 488 and 413 nm are associated to the yellows hues, according to literature reports [41,42]. Analyzing in more detail the spectra, the signals of goethite and hematite can be identified at room temperature (25 1C). In this regard, as the temperature increases, a displacement of the signals is observed at higher wavelength due to the transformation from goethite to hematite. After 300 1C only the hematite signal is detected. For 600 1C, no signal was detected since the sample becomes completely dark. 3.6. Temperature effect on the color of the rock The color changes due to the thermal treatment, in the case of lightness (Ln) are similar in all four rocks and at 25 1C have the highest value and diminishes successively as the temperature increases (Fig. 9a). The first color change from yellowish-beige to gray took place at 400 °C for the XCA, CH and LB rocks. An increase in the red color could be observed from 100 up to 300 0 1C for LR rocks and this color changes again from red to gray at 400 1C. The initial values of an for LR are higher and constant due to the presence of iron oxides (goethite and hematite), which render a red color to the rock. After 300 1C values of an diminishes because the LR rock becomes darker (Fig. 9b). The difference of the values of an for XCA, CH and LB does not change with increasing temperature and with smaller values comparing them with the values of LR. The values of bn show a variation with temperature, initially decreasing and subsequently increases in the temperature range from 300 1C to 400 1C, where the samples becomes gray (Fig. 9c.). In this study, we prove experimentally the process of reddening of limestone rocks by oxidative conditions during heating. Changes in the physical properties and color (namely reddening) of rocks, which typically occur in stones after temperatures changes, it has recently been described in many monuments [45–47]. The observed changes can, however, be associated to other alteration processes such as chemical changes (e.g. oxidation of Fe2 þ to Fe3 þ ). 5
25
120
XCA
CH
LB
XCA
LR
CH
LB
LR
XCA
LR
0
80
b*
a*
15
L*
LB
20
100
10 60
40
CH
-5
5
0
100
200
300
400
500
Temperature (°C)
600
0
0
100
200
300
400
500
Temperature (°C)
600
-10
0
100
200
300
400
500
Temperature (°C)
Fig. 9. Color variation of the limestone rocks after the corresponding thermal treatments: (a) L*, (b) a* and (c) b*.
600
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
4. Conclusions The compressive properties of four limestone rocks with different physical characteristics and chemical composition were analyzed as a function of thermal treatments. LR is a red rock due to the presence of iron oxides and has a lower porosity than XCA and CH which are white rocks, with higher amount of carbonates. LB is a highly compacted and heterogeneous white rock with a low porosity. Compression testing showed similar mechanical behavior for rocks LB and LR, but different to those of XCA and CH, which also showed similar mechanical behavior between them. LB and LR rocks showed a brittle behavior with the highest compressive strength and elastic modulus among the four groups of rocks, and such mechanical properties were maintained at temperatures up to 600 1C. On the contrary, XCA and CH rocks showed lower strength and elastic modulus, and a decreasing trend on their mechanical properties with increased treatment temperature. This mechanical behavior is strongly correlated to the mineral composition or the rocks, as observed in LR limestone due to the transformation of goethite to hematite. Also, the porosity of the rocks changes upon temperature exposure, and less porous rocks have higher mechanical properties.
Acknowledgments The authors acknowledge the financial support at LANNBIO Cinvestav, Mérida from projects FOMIX-Yucatán 2008-108160 and CONACYT LAB-2009-01 No. 123913. Technical assistance of Daniel Aguilar is acknowledged. References [1] Cabadas HV, Solleiro E, Sedov S, Pi T, Alcalá JR. The complex genesis of red soils in Peninsula de Yucatan, Mexico: mineralogical, micromorphological and geochemical proxies. Eurasian Soil Sci 2010;13:1439–57. [2] Kılıc Ӧ. The influence of high temperatures on limestone P-wave velocity and Schmidt hammer strength. Int J Rock Mech Min Sci 2006;43:980–6. [3] Yavuz H, Demirdag S, Caran S. Thermal effect on the physical properties of carbonate rocks. Int J Rock Mech Min Sci 2010;47:94–103. [4] Brotóns V, Tomás R, Ivorra S, Alarcón JC. Temperature influence on the physical and mechanical properties of a porous rock: San Julian's calcarenite. Eng Geol 2013;167:117–27. [5] Jansen DP, Carlson SR, Young RP, Hutchins DA. Ultrasonic-imaging and acoustic-emission monitoring of thermally-induced microcracks in lac-dubonnet granite. J Geophys Res-Sol Ea 1993;98:22231–43. [6] Glover PWJ, Baud P, Darot M, Meredith PG, Boon SA, Leravalec M, Zoussi S, Reuschle T. Alpha/beta phase-transition in quartz monitored using acoustic emissions. Geophys J Int 1995;120:775–82. [7] Faust BB. Maya environmental successes and failures in the Yucatan Peninsula. Environ Sci Policy 2001;4(4–5):153–69. [8] Eastmond A, Faust B. Farmers, fires, and forests: a green alternative to shifting cultivation for conservation of the Maya forest? Landscape Urban Plan 2006;74:267–84. [9] McCabe S, Smith BJ, Warke PA. Exploitation of inherited weakness in firedamaged building sandstone: the ‘fatiguing’ of ‘shocked’ stone. Eng Geol 2010;115:217–25. [10] Aurangzeb LAK, Maqsood A. Prediction of effective thermal conductivity of porous consolidated media as a function of temperature: a test example of limestone. J Phys D Appl Phys 2007;40:4953–8. [11] Chaki S, Tarkali M, Agbodjan WP. Influence of thermal damage on physical properties of a granite rock: porosity, permeability and ultrasonic wave evolutions. Constr Build Mater 2008;22:1456–61. [12] Myrow PM. A new graph for understanding colors of mudrocks and shales. J Geol Educ 1990;38:16–20. [13] Dionísio A, Sequeira-Braga MA, Waerenborgh JC. Clay minerals and iron oxides-oxyhydroxides as fingerprints of firing effects in a limestone monument. Appl Clay Sci 2009;42:629–38. [14] Chakrabarti B, Yates T, Lewry A. Effect of fire damage on natural stonework in buildings. Constr Build Mater 1996;10:539–44. [15] Sheu DD, Presley BJ. Variations of calcium carbonate, organic carbon and iron sulfides in anoxic sediment from the Orca Basin, Gulf of Mexico. Mar Geol 1986;70:103–18. [16] Alonzo-Salomón L. Características geotécnicas de suelos y rocas de los estados de Yucatán y Campeche. México: UADY; 2010.
189
[17] Ozsoy EA, Yilmaz G, Arman H. Physical, mechanical and mineralogical properties of ophiolitic rocks at the Yakakayi dam site, Eskisehir, Turkey. Sci Res Essays 2010;5:2579–87. [18] Maldonado L, Veleva L, Díaz-Ballote L. Characterization of limestone for building in the Yucatan Peninsula. Appl Phys A 2011;103:1105–10. [19] ASTM D2216-98. Standard test method for laboratory determination of water (Moisture) content of soil and rock by mass. West Conshohocken (PA, USA): ASTM International; 2007. [20] ASTM D3148. Standard test for elastic moduli of intact rock core specimens in uniaxial compresion. West Conshohocken (PA, USA): ASTM International; 2007. [21] Yokoyama T, Nakashima S. Color developed of iron oxides during rhyolite weathering over 52,000 years. Chem Geol 2005;219:309–20. [22] Yamanio Y, Takeuchi S, Okumura S, Nakashima S, Yokoyama T. Color measurements of volcanic ash deposits from three different styles of summit activity at Sakurajima volcano, Japan: conduit processes recorded in color of volcanic ash. J Volcanol Geoth Res 2008;178:81–3. [23] CIE (Commission Internationale de l' Eclairage). Recommendations on uniform color spaces, color difference equations, psychometric color terms. Supplement No 2 to CIE Publication No 15, Colorimetry, Bureau Central de la CIE, Paris; 1976. [24] Pospíšil J, Hrdў J, Hrdў Jr. J. Basic methods for measuring the reflectance color of iron oxides. Optik 2007;118:278–88. [25] Grossi CM, Brimblecombe P, Esbert RM, Alonso FJ. Color changes in architectural limestones from pollution and cleaning. Color Res Appl 2007;32 (4):320–31. [26] May-Crespo J, Martínez-Torres P, Alvarado-Gil JJ, Quintana P, Vilca-Quispe L. Photothermal radiometry characterization of limestone rocks from the Península of Yucatán. Int J Thermophys 2012;33:1908–15. [27] Ozguven A, Ozcelik Y. Investigation of some property changes of natural building stones exposed to fire and high heat. Constr Build Mater 2013;38:813–21. [28] Özkahraman HT, Selver R, Işık EC. Determination of the thermal conductivity of rocks from P-wave velocity. Int J Rock Mech Min Sci 2004;41:703–8. [29] ISRM. Rock characterization testing and monitoring. In: Brown ET, editor. ISRM suggested methods, 113–116. Oxford: Pergamon Press; 1981. p. 135–40. [30] Allison RJ, Goudie AS. The effect of fire on rock weathering: an experimental study. In: Robinson DA, Williams RBG, editors. Rock weathering and landform evolution. Chichester: John Wiley & Sons; 1994. p. 41–56. [31] Allison RJ. A non-destructive method of determining rock strength. Earth Surf Process Landf 1988;13:729–36. [32] Moses C, Robinson D, Barlow J. Methods for measuring rock surface weathering and erosion: a critical review. Earth-Sci Rev 2014;135:141–61. [33] Mao XB, Zhang LY, TZ LI, Liu HS. Properties of failure mode and thermal damage for limestone at high temperature. Min Sci Technol 2009;19:0290–4. [34] Tang F, Mao X, Zhang L, Yin H, Li Y. Effects of strain rates on mechanical properties of limestone under high temperature. Min Sci Technol (China) 2011;21:857–61. [35] Ar I, Doğu G. Calcination kinetics of high purity limestones. Chem Eng J 2001;83:131–7. [36] Mustakimah M, Yusup S, Maitra S. Decomposition of calcium carbonate in cockle shell. J Eng Sci Technol 2012;7:1–10. [37] Rizov B. Phase transformations from goethite to hematite and thermal decomposition in various nickeliferous laterite ores. J Univ Chem Technol Metall 2012;47:207–10. [38] Lee JT, Keener TC, Knoderera M, Khang SJ. Thermal decomposition of limestone in a large scale thermogravimetric analyzer. Thermochim Acta 1993;213:223–40. [39] Salehi MH, Beni OH, Harchegani HB, Borujeni IE, Motaghian HR. Refining soil organic matter determination by loss-on-ignition. Pedosphere 2011;21 (4):473–82. [40] Gualtieri AF, Venturelli P. In situ study of the goethite-hematite phase transformation by real time synchrotron powder diffraction. Am Mineral 1999;84:895–904. [41] Cornell RM, Schwertmann U. The iron oxides: structure, properties, reactions, occurences and uses. 2nd ed.Wiley-Vch Verlag GmbH & Co. Weinheim; 2003. [42] Scheinost AC, Chavernas A, Barron V, Torrent J. Use and limitations of secondderivative diffuse reflectance spectroscopy in the visible to near-infrared range to identify and quantity fe oxide minerals in soils. Clay Clay Miner 1998;46(5):528–36. [43] Savitzky A, Golay MJE. Smoothing and differentiation of data by simplified least squares procedures. Anal Chem 1964;36:1627–39. [44] Sherman DM, Waite TD. Electronic spectra of Fe3 þ oxides and oxide hydroxides in the near IR to near UV. Am Mineral 1985;70:1262–9. [45] Hajpál M. The behavior of natural building stones by heat effect. In: Kourkoulis SK, editor. Fracture and failure of natural building stones. Dordrecht: Springer; 2006. p. 439–45. [46] Obojes U, Tropper P, Mirwald PW, Saxer A. The effects of fire and heat on natural building stones: first results from the Gröden Sandstone. In: Fort R, Alvarez De Buergo M, Gomez-Heraz M, Vazques-Calvo C, editors. Heritage, weathering and conservation. Leiden: Taylor and Francis/Balkema; 2006. p. 521–4. [47] Sippel J, Siegesmund S, Weiss T, Nitsch KH, Korzen M. Decay of natural stones caused by fire damage. In: Prikril R, Smith BJ, editors. Building Stone Decay: From Diagnosis to Conservation, 271. Geological Society, London, Special Publications; 2007. p. 139–51.
Contents lists available at ScienceDirect
International Journal of Rock Mechanics & Mining Sciences journal homepage: www.elsevier.com/locate/ijrmms
Technical Note
Thermal effects on the physical properties of limestones from the Yucatan Peninsula W.S. González-Gómez a,n, P. Quintana a, A. May-Pat b, F. Avilés b, J. May-Crespo a, J.J. Alvarado-Gil a a b
Applied Physics Department, Cinvestav-Unidad Mérida, Carr. Ant. a Progreso km 6, C.P. 97310 Mérida, Yucatán, México Centro de Investigación Científica de Yucatán, Unidad de Materiales, Calle 43, No.130, Col. Chuburná de Hidalgo, CP 97200 Mérida, Yucatán, México
art ic l e i nf o
a b s t r a c t
Article history: Received 22 July 2014 Received in revised form 10 October 2014 Accepted 29 December 2014 Available online 21 February 2015
The effect of thermal degradation on the compression strength, ultimate compression strain, color and mass loss of four limestones extracted from the Yucatan Peninsula was studied. The samples sets were independently subjected to 25, 100, 200, 300, 400, 500 and 600 °C during one hour for each temperature. The thermal degradation of the investigated rocks and their chemical composition, microstructure and porosity is also investigated. The results show a correlation between the mechanical properties, thermal behavior, the presence of minerals and porosity in each type of limestone rock. The variation of the limestones color with temperature was described in terms of standard space color parameters and their diffuse reflectance. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Limestones Thermal heating Compressive properties Mineralogy
1. Introduction The historical monuments and archeological sites in the Maya area comprising southern Mexico, the central part of Honduras and Guatemala, are constructed with different types of limestone, with calcium carbonates (CaCO3) as the basic constituent. Currently in the Yucatan Peninsula there are a variety of historical constructions and archeological sites built with stones. The Peninsula of Yucatan is a smooth carbonate platform developed from Cretaceous, Tertiary and Quaternary limestones, with predominantly shallow calcareous soil [1], containing different grain sizes, porosities, and texture. Also, many modern buildings use limestone as a primary construction material. Limestones can be obtained from a variety of sources and may differ considerably in their chemical compositions and physical microstructure. The chemical reactivity show a large variation due to their differences in crystalline structure and the nature of impurities (silica, iron, magnesium, manganese, sodium, and potassium, to name a few) [2]. The study of changes in the physical properties of rocks as a function of thermal excursions is relevant to various engineering and industrial applications. Rocks are composed of minerals, bounding matrix, cracks and pores, and heat may produces important physical and chemical changes in their microstructure [3]. Temperature
n
Corresponding author. E-mail address: [email protected] (W.S. González-Gómez).
http://dx.doi.org/10.1016/j.ijrmms.2014.12.010 1365-1609/& 2015 Elsevier Ltd. All rights reserved.
changes can induce micro-cracks in the rocks due to the mismatch in the thermal properties of the different mineral compositions (intergranular) or within grains (intragranular) [4]. Temperature gradients can affect the rock dilation, changing the coefficients of thermal expansion [5]. Qualitative changes produced by heat can also occur when the rocks undergo a phase transition such as the α/β phase transition in quartz [6]. One of the traditional forms of tropical agriculture in the Yucatan peninsula is the “slash-and-burn” system that is applied every year, during the dry periods [7,8]. Temperature measurements on these traditional practices can reach non-uniform high temperatures from 300 1C up to 700 1C [9]. These repeated burnings and extreme heating changes can lead to long term cumulative thermal effects on the rock properties and ecosystem. This is because the modifications of the mechanical properties of limestones have a strong influence on the soil conformation and induce erosion, promoting water infiltration, increased permeability and porosity, which are one of the major problems in these regions. Temperature variations are one of the major factors influencing the intrinsic properties of rocks [10,11]. It has been shown that the variation of expansion increases with temperature [3], which is primarily responsible for the changes in the microstructure of rocks, inducing new fractures and microfractures. A common form of chemical action caused by temperature is the oxidation of iron, which provokes significant color changes even in the presence of minor amounts of iron (ferrous minerals) in the calcareous matrix. For example, red coloration is due to the presence of hematite, whereas less common yellows and browns generally result from
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
2. Material and methods 2.1. Description of rock samples The four types of rocks were selected from quarries of the Yucatan State, Mexico, which are used in the construction industry. These materials were selected considering their regional significance, abundance and frequency of usage as building and restoration materials [18]. The rocks were named “XCA”, “CH”, “LB”, and “LR” after their Spanish acronyms. The sample XCA is a limestone with shades ranging from white to beige, inlaid with shells and fossils of snails with high porosity. CH is a limestone with different pale beige shades and high porosity. LB is a white compact limestone with low porosity, whereas the sample LR is a red compact limestone with low porosity. The physical properties of the four limestone samples such as bulk density, water absorption, water content and effective porosity, were determined by water immersion based on mass difference and according to the ASTM standard D2216-98 [19] with three repetitions for each rock. For this purpose, a set of 12 cubic limestone samples were cut into coupons of 1 cm3 obtained employing a low-speed diamond saw (Buehler Isomet). 2.2. Thermal treatment Seven sets of rocks containing ten replicated limestone coupons were cut with a low-speed diamond saw (Buehler Isomet) for each temperature. The limestone coupons for thermal treatment comply with the dimensions set for (subsequent) compression testing with average dimensions of 21 mm 10.5 mm 10.5 mm, having a total of 280 prismatic coupons. The samples were polished with sandpaper in order to have a flat surface finish, and then cleaned using a sonicating bath. The samples were then washed with a 2:1 (v/v) mixture of water:acetone (J.T. Baker, 99.8%); afterwards they were left to dry at room temperature and set inside of an oven at 80 1C during 24 h. Each set of samples were subjected to one of the six thermal treatments investigated with the targeted temperature fixed at 100, 200, 300, 400, 500 and 600 1C, using a heating rate of 3 1C/min to ensure thermal equilibrium of the limestone coupons. The targeted temperature was reached sequentially (step-wise) in steps of 100 1C maintained during 1 h, as depicted in Fig. 1. Additionally, a set of samples was kept at 25 1C to act as a blank. Once the targeted
ramp 6
600 ramp 5
500
Temperature (°C)
limonite and goethite, respectively. The green-gray to black color in rocks is related to total organic carbon where darker colors corresponding to higher carbon content; this relationship has been confirmed on sediments [12]. The presence of kaolinite and smectite provide white to light neutral colors. Compounds of other transition metals (Ti, Mn, Co, Cu and Zn) may also show an influence of the pigment [13]. Also, the color in the rocks changes when heated at temperatures above 250–300 1C, which correspond to the dehydration of iron compounds. Pale brown colored rock change to reddish brown upon heating, which may not be apparent until the stones haves reached temperatures above 400 1C when goethite transforms to hematite [4,14,15]. Several investigations on the weathering mechanisms and degradation kinetics of Yucatan limestones have been reported [16,17], but little is known about the effect of thermal degradation on their mechanical properties. In this contribution, four types of limestone rocks, collected from the state of Yucatan (Southern Mexico), were investigated to elucidate correlations between rock compressive properties, mineral transformations and color change with thermal treatments at different temperatures as an attempt to simulate temperature changes similar to those observed in the “slash-and-burn” traditional systems in Yucatan.
183
ramp 4
400 ramp 3
300 ramp 2
200 100 0
ramp 1
0
200
400
600
800
1000 1200 1400 1600 1800
Time (min) Fig. 1. Temperature treatment performed on the rocks before the compression tests. Heating was performed at a rate of 3 1C/min. The temperature steps (ramps) correspond to 100 1C, 200 1C, 300 1C, 400 1C, 500 1C and 600 1C.
temperature was reached, the samples were maintained during one hour at the corresponding temperature and the furnace was then cooled down slowly to room temperature in order to avoid sample cracking if suddenly are cooled. For example, for the test at 400 1C, the heat treatment followed the four initial ramps depicted in Fig. 1 requiring a total time of 16 h. 2.3. Sample characterization The composition of specimens was analyzed using the powder X-ray diffraction technique (XRD Siemens D-5000) with a BraggBrentano geometry, a Cu-Kα radiation (λ ¼1.5418 Å), scans with a step size of 0.021, an integration time of 10 s and with a 2θ angular range of 51r 2θr 501. Thermogravimetric analysis (TG) and the derivative of the thermogram (DTG) were recorded in a Discovery TGA (TA Instruments). Powder samples (2 mg) were heated from 50 to 1000 1C applying a heating rate of 10 1C/min under an inert atmosphere (N2). Compression tests were conducted in a Shimadzu AGI-100 universal testing machine equipped with load cells of 5 kN and 100 kN and using a cross-head speed of 0.05 mm/min according to the ASTM standard D3148 [20]. The prismatic coupons for compression testing had a length of 21 mm and 10.5 mm 10.5 mm cross-sectional area and ten replicates were tested for each thermal treatment. To measure the reflectance spectra of the limestone materials as a function of the thermal treatment, the samples were illuminated with a deuterium-halogen light source (AvaLight DH-S-BAL) using an optical fiber; the light reflected was then collected using an integrating sphere (Labsphere USRS-99-010) and sent to a spectrometer (AvaSpec-2048) equipped with an optical fiber. The reflectance spectra were measured from 400 nm to 1000 nm. The color change was determined by applying the model of the Commission Internationale de l'Eclaraige (CIE), using as a standard illuminant D65 under an angle of 101 [21,22]. The color was described in terms of Lnanbn space color by CIE-LAB and CIELCH systems (CIE, 1976) [23], where the value of Ln represents lightness (Ln ¼0: black y Ln ¼ 100: white), and an and bn chromaticity (þ an ¼red and -an ¼green; þbn ¼yellow and bn ¼blue) [24,25].
3. Results and discussion 3.1. Physical properties of the limestone rocks The physical properties of the investigated limestone rocks such as bulk density and effective porosity were determined using
184
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
the saturation and buoyancy techniques, as recommended by ASTM [19]. The value of the bulk density of the four rocks is between 1.98 g/cm3 and 2.36 g/cm3 (Table 1), which are lower than other values reported for limestone rocks [16,26,27]. A comprehensive study of porosity can provide valuable information to determine whether a given type of rock is susceptible to thermal stress or not. The effective porosity percentages for the rocks LB and LR are similar to each other with values of 6.02% and 4.05%, respectively contrasting with the rocks XCA and CH, which have higher effective porosity values (20.1% and 18.6%, respectively) (Table 1). Water retained by the rocks at 25 1C, is 12.2% and 11.6% for limestone types XCA and CH, respectively, which is about three times more than that retained by rocks LB and LR with 3.96% and 2.38%, respectively (Table 1). The highest water content was for the CH rock with 1.06% as compared to the other limestone samples ranging from 0.14% to 0.22%.
Table 1 Physical properties of four limestone samples. Rock name
Bulk density (g/ Water content Water cm3) (%) absorption (%)
Effective porosity (%)
XCA CH LB LR
2.00 70.12 1.98 70.04 2.36 70.10 2.30 70.15
20.17 1.50 18.6 7 1.12 6.02 7 1.35 4.05 7 0.58
0.22 7 0.08 1.067 0.07 0.147 0.02 0.167 0.02
12.2 70.95 11.6 70.80 3.96 70.70 2.38 70.48
3.2. Mineral composition of the limestone rocks XRD analysis of the limestones samples was performed to determine the mineral composition of the different lithotypes. The X-ray diffraction patterns for the powdered samples showed mainly the presence of calcite (CaCO3), except for limestone type CH which presented another polymorph of calcium carbonate, aragonite. To detect the presence of other minerals, the powders were treated with 5% HCl (v/v) in order to dissolve all carbonates. The XRD powder patterns of the investigated rocks after treatment with HCl (Fig. 2), show the presence of quartz (SiO2) in all samples, as an essential component, which is directly related to some physical properties such as hardness and strength [28]. Some clays were also identified in all the limestones rocks such as tosudite (Na0.5(Al,Mg)6(Si, Al)8O18(OH)12∙5H2O); kaolinite (Al2Si2O5(OH)4) and montmorillonite (Ca0.2(Al, Mg)2Si4O10(OH)2∙4H2O) was additionally detected in XCA and LB. Anatase (TiO2) was identified in all samples except for LR. Orthoclase (KAlSi3O8) was found in CH and LB. Iron oxyhydroxide as goethite (FeO(OH)) and hematite (Fe2O3) were present in LR. 3.3. Effect of temperature on the compression properties Representative stress–strain (σ ε) diagrams chosen from the ten replicates of the compressive tests for each thermal treatment and rock type, are shown in Fig. 3. Even though the thermic influence on the behavior of the limestone is difficult to quantify, it was possible to observed, despite the relatively high standard deviation (inherent
Q
Relative Intensity (a.u.)
Relative Intensity (a.u.)
T Q Q A Q T
M K
5
10
15
20
25
30
35
40
45
50
TQ
10
O T
K
5
Q A
T
15
20
25
30
35
40
45
50
45
50
T
Relative Intensity (a.u.)
Relative Intensity (a.u.)
G
Q M
M
K
5
10
15
A
20
25
O T
30
35
40
45
50
T Q
G
K
5
10
15
G H
T
20
25
30
T
35
G
40
Fig. 2. XRD patterns of limestone coupons after the dissolution of carbonates: (a) XCA, (b) CH, (c) LB and (d) LR. The observed phases were: quartz (Q), kaolinite (K), tosudite (T), anatase (A), goethite (G), montmorillonite (M), hematite (H) and orthoclase (O).
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
strength with a decreasing trend as temperature was increased (Fig. 4a). On the other hand, the ultimate strain (εmax) is lower for LB and LR (Fig. 4b). The variations of such ultimate strain with the thermal treatments are within the experimental scattering for XCA, LB and LR, but tend to increase for CH when the temperature increases (Fig. 4b). This behavior indicates that LB and LR rocks have a significantly higher strength and are more brittle than XCA and CH specimens according to ISRM [29].
to the natural heterogeneity of rocks), important trends defining the mechanical behavior of the investigated rocks. The analysis of the stress–strain diagrams shows significant differences between the four rocks (Fig. 4). The group of rocks (LB and LR) showed the highest compression strength (σmax), and such strength was maintained for all thermal treatments (Fig. 4a). The group of rocks with more inlaid shells and other fossils, XCA and CH, showed lower compression
15
35
25°C 100°C 200°C 300°C 400°C 500°C 600°C
10
185
25°C 100°C 200°C 300°C 400°C 500°C 600°C
30 25 20 15
5
10 5
0
0 0.0
0.5
1.0
1.5
2.0
80
60 50 40
0.5
1.0
1.5
2.0
100
25°C 100°C 200°C 300°C 400°C 500°C 600°C
70
0.0
2.5
25°C 100°C 200°C 300°C 400°C 500°C 600°C
d) LR 80
60
40
30 20
20
10 0
0 0.0
0.5
1.0
1.5
2.0
0.0
0.5
1.0
1.5
2.0
Fig. 3. Representative compressive stress–strain (σ-ε) diagrams of the limestone rocks at different heat levels: (a) XCA, (b) CH, (c) LB and (d) LR.
8
2.5
100
XCA
90
CH
LB
LR
XCA
CH
LB
XCA
LR
CH
LB
LR
7
80 6
2.0 70
5
60
4
1.5
50 40
3
30 2
1.0 20
1
10
0
0.5
0 25
100
200
300
400
Temperature (°C)
500
600
25
100
200
300
400
Temperature (°C)
500
600
25
100
200
300
400
Temperature (°C)
Fig. 4. Values of compressive stress–strain (σ-ε) diagrams of the limestone rocks at different heat levels: (a) σmax, (b) εmax and (c) E.
500
600
186
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
The elastic modulus (E) at 25 1C (Fig. 4c), is also significantly higher for the more brittle rocks (LR and LB) than for XCA or CH cases. There is a trend of decreasing E as temperature increases for XCA and CH, and in less proportion for LB; however, this behavior was not observed for LR which displayed an increase in the
modulus of elasticity as for the case-hardening argument presented in [30] in a study of heating effect on sandstone, limestone and granite rocks. These results show a good correlation with the relative porosity values summarized in Table 1. Because temperature influences the porosity and mechanical properties of rocks
Fig. 5. Photographs of fractured rocks after the compression testing.
TG (%)
-0.05 70 Crytal (OH)
60 50
- 26.4%
-0.10
Organic matter + Water (OH)
200
300
400
500
600
700
800
Water (OH)
-0.2 50 Organic matter
40
900
100
200
300
Temperature (°C) 0.04
90
0.02
40
- 8.5%
400
500
-0.3 900
0.04 - 5.2%
0.02
600
Temperature (°C)
700
800
90
- 4.3%
-0.10 900
0.00
85
80
75
-0.02
- 5.8%
-0.06 -0.08
300
800
0.06
95
TG (%)
-0.02 -0.04
Organic matter + Water (OH)
200
700
100
DTG (%/min)
TG (%)
- 9.6%
100
600
Goethite
-0.04
Organic matter + Water (OH)
-0.06
Hematite 100
200
300
400
500
600
Temperature (°C)
Fig. 6. TG-DTG curves of limestone: (a) XCA, (b) CH, (c) LB and (d) LR.
700
800
-0.08 900
DTG (%/min)
Crystal (OH)
70
50
500
0.00
- 20.4%
60
400
Temperature (°C)
100
80
-0.1
- 9.9%
70 60
-0.20 100
0.0
80
-0.15
40
- 22.2%
DTG (%/min)
80
0.1
- 2.6%
90
0.00
- 18.4%
TG (%)
90
100
0.05
- 3.8%
DTG (%/min)
100
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
[31,32], the rocks LB and LR with significantly lower porosity have a higher elastic modulus and compressive strength and are more brittle. The stress–strain an elastic modulus behavior for LR and LB rocks are similar to the mechanical properties obtained on limestones from China [33,34]; however, in this work the obtained values for the σmax and E, at each temperature are lower, than the reported. In order to visualize better these mechanical behaviors, Fig. 5 displays pictures of the rocks upon fracture for representative limestone coupons treated at different temperatures. Overall, rocks LR and LB show a single crack running along the loading direction (vertical) or slightly inclined, which indicates failure dominated by normal stresses and typically found in brittle materials. Rocks CH and XCA, on the other hand present cracks inclined at 25 1C, with the loading direction (vertical), indicating influence of shear stresses in the failure mode. As the temperature treatment increases, the influence of shear stresses in the failure mode of CH and XCA becomes more prominent, as indicated by the inclination of the main crack within the coupon. For temperatures above 400 1C, the rocks CH and XCA pulverize upon failure (which is not the case for LR and LB,) indicating the lack of molecular cohesion of these rocks at such high temperatures. 3.4. Effect of temperature treatment on thermogravimetric analysis The curves representing the thermogravimetric analysis (TG, left vertical axis) and the derivative of the thermogram (DTG, right vertical axis) for the four natural limestone types in Fig. 6, initially show a small mass loss due to the adsorbed water ( 200 1C), followed by a loss of CO2 from calcium carbonate (730 1C) with a
total mass loss of 40–45%. These values are lower than those reported in the literature on calcination kinetics and surface area of dispersed limestone particles, which are around 45–55% [35,36]. The same analysis (TG/DTG) was applied to the limestones after the dissolution of carbonates (Fig. 6). The DTG curve of XCA initially shows the loss of adsorbed water on clays and the elimination of organic matter at 330 1C, with a mass loss of 22.2%. Thereafter, the loss of structural water is observed in the temperature range from 380 1C up to 650 1C [37], with a mass loss of 26.4%. In the case of the CH rock the behavior was similar to XCA, and the mass loss was 24.8%. However, a rapid mass loss in the shorter temperature range of 240–400 1C, with a maximum at 330 1C, is observed as this rock has higher organic matter produced by the marine microorganisms which are inherent to the rock matrix [38,39]. The second loss of mass is 9.9% and ends at 500 1C, ascribed to dehydroxylation of clays, due to the presence of kaolinite. The thermal behavior of LB shows two mass losses within the temperature range of 230–530 1C which correspond to the elimination of organic matter and the removal of the hydroxyl groups between the interlayers of the mixed clays (kaolinite, tosudite and montmorillonite) with a total mass loss of 30%. Subsequently, a mass loss of 8.5% in the temperature range of 550–850 1C (with an inflection at 705 1C) corresponds to the elimination of the structural water from the aluminosilicates. The DTG curve of LR during the heating process, the temperature range for the removal of the water is 25–150 1C, following by a mass loss at 260 1C that corresponds to the dehydroxylation of goethite which is transformed into hematite during the heating process; the transformation temperature usually depends on the
100
100
100°C
200°C 80
80
100°C
70
Reflectance (%)
Reflectance (%)
25°C
90
25°C
90
300°C 400°C
60 50
500°C
40
600°C
30
187
500
600
700
800
70
400°C
60 50 40
500°C
30
600°C
20
900
200°C 300°C
500
600
Wavelength (nm)
700
800
900
800
900
Wavelength (nm)
90
80
25°C 80
70
100°C
25°C 100°C
70
Reflectance (%)
Reflectance (%)
300°C 400°C 200°C 60
500°C 50
60
300°C 200°C
50
400°C 500°C
40
600°C
600°C 40 30
30
500
600
700
Wavelength (nm)
800
900
20
500
600
700
Wavelength (nm)
Fig. 7. Diffuse reflectance spectra variation with thermal treatment for different limestone: (a) XCA, (b) CH, (c) LB and (d) LR.
188
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
crystallinity and the substitution of iron by aluminum [40,41]. The second mass loss at 430 1C is related to the removal of organic matter and the OH from the clays, with a total mass loss of 15%.
3.5. Effect of temperature on reflectance spectra The diffuse reflectance spectra of the samples after each thermal treatment were obtained in the visible region (Fig. 7). Initially, the reflectance spectra of the samples were similar to each other the temperature range of 25 to 200 1C. However, regardless of the limestone type, a strong decrease in the reflectance (from 10 up to 30%) is observed at higher temperatures. The decrease in reflectance is due to the color change of the limestone as the color becomes darker with the increase in temperature. Furthermore, the spectra of LR show changes at 580 nm at low temperatures (25–200 1C) due to the presence of the two iron oxides (goethite and hematite) as shown in Fig. 7d. At higher temperatures only the hematite signal is observed because the transformation of goethite to hematite occurs at 260 1C, which is also observed in the TG/DTG analysis. The parameterization of the spectra applying the function of Kubelka-Munk (K-M) allows identifying and quantifying the iron oxide minerals as shown by Scheinost's studies of soils [42]. The reflectance values are used to calculate the absorption and scattering coefficients, which corresponds to the two phenomena that occur in the interaction between the light-sample processes. Because the transformation of the derivative is particularly sensitive to high frequency instrumental noise, an appropriate filtering procedure was applied using the Savitzky-Golay function [43].
Hematite Goethite
-4
F(R) 2nd Derivative (x10 )
25°C 100°C 200°C 300°C 400°C 500°C 600°C
LR
2.0
1.5
1.0
0.5
0.0 300
350
400
450
500
550
600
650
700
Wavelength (nm) Fig. 8. Second derivative of the LR sample, showing the bands of hematite and goethite.
The second-derivative from the spectra reflectance of the iron oxides show four bands in the range from 350 to 1030 nm according to Sherman and Waite [44], due to three single electron transitions: 4t1’6A1 at 870–1030 nm, 4T2’6A1 at 650–710 nm, and 4E; 4A1’6A1 at 405–410 nm as well as the electron pair transition (EPT) at 488–493 nm. The EPTs are the most intense absorption edges which, in turn, are closely related to the hue. The EPTs for the hematite occur around 521–565 nm, clearly separated from the EPTs related to yellow iron oxides at 479–499 nm. The second derivative of the reflectance spectra was applied only to LR limestone in the range from 400 to 1000 nm. The iron oxides bands can be identified in Fig. 8, which changes according to the temperature. Hematite displays two signals at 538 nm and 423 nm that gives the red characteristic color of the rock and the signals of goethite at 488 and 413 nm are associated to the yellows hues, according to literature reports [41,42]. Analyzing in more detail the spectra, the signals of goethite and hematite can be identified at room temperature (25 1C). In this regard, as the temperature increases, a displacement of the signals is observed at higher wavelength due to the transformation from goethite to hematite. After 300 1C only the hematite signal is detected. For 600 1C, no signal was detected since the sample becomes completely dark. 3.6. Temperature effect on the color of the rock The color changes due to the thermal treatment, in the case of lightness (Ln) are similar in all four rocks and at 25 1C have the highest value and diminishes successively as the temperature increases (Fig. 9a). The first color change from yellowish-beige to gray took place at 400 °C for the XCA, CH and LB rocks. An increase in the red color could be observed from 100 up to 300 0 1C for LR rocks and this color changes again from red to gray at 400 1C. The initial values of an for LR are higher and constant due to the presence of iron oxides (goethite and hematite), which render a red color to the rock. After 300 1C values of an diminishes because the LR rock becomes darker (Fig. 9b). The difference of the values of an for XCA, CH and LB does not change with increasing temperature and with smaller values comparing them with the values of LR. The values of bn show a variation with temperature, initially decreasing and subsequently increases in the temperature range from 300 1C to 400 1C, where the samples becomes gray (Fig. 9c.). In this study, we prove experimentally the process of reddening of limestone rocks by oxidative conditions during heating. Changes in the physical properties and color (namely reddening) of rocks, which typically occur in stones after temperatures changes, it has recently been described in many monuments [45–47]. The observed changes can, however, be associated to other alteration processes such as chemical changes (e.g. oxidation of Fe2 þ to Fe3 þ ). 5
25
120
XCA
CH
LB
XCA
LR
CH
LB
LR
XCA
LR
0
80
b*
a*
15
L*
LB
20
100
10 60
40
CH
-5
5
0
100
200
300
400
500
Temperature (°C)
600
0
0
100
200
300
400
500
Temperature (°C)
600
-10
0
100
200
300
400
500
Temperature (°C)
Fig. 9. Color variation of the limestone rocks after the corresponding thermal treatments: (a) L*, (b) a* and (c) b*.
600
W.S. González-Gómez et al. / International Journal of Rock Mechanics & Mining Sciences 75 (2015) 182–189
4. Conclusions The compressive properties of four limestone rocks with different physical characteristics and chemical composition were analyzed as a function of thermal treatments. LR is a red rock due to the presence of iron oxides and has a lower porosity than XCA and CH which are white rocks, with higher amount of carbonates. LB is a highly compacted and heterogeneous white rock with a low porosity. Compression testing showed similar mechanical behavior for rocks LB and LR, but different to those of XCA and CH, which also showed similar mechanical behavior between them. LB and LR rocks showed a brittle behavior with the highest compressive strength and elastic modulus among the four groups of rocks, and such mechanical properties were maintained at temperatures up to 600 1C. On the contrary, XCA and CH rocks showed lower strength and elastic modulus, and a decreasing trend on their mechanical properties with increased treatment temperature. This mechanical behavior is strongly correlated to the mineral composition or the rocks, as observed in LR limestone due to the transformation of goethite to hematite. Also, the porosity of the rocks changes upon temperature exposure, and less porous rocks have higher mechanical properties.
Acknowledgments The authors acknowledge the financial support at LANNBIO Cinvestav, Mérida from projects FOMIX-Yucatán 2008-108160 and CONACYT LAB-2009-01 No. 123913. Technical assistance of Daniel Aguilar is acknowledged. References [1] Cabadas HV, Solleiro E, Sedov S, Pi T, Alcalá JR. The complex genesis of red soils in Peninsula de Yucatan, Mexico: mineralogical, micromorphological and geochemical proxies. Eurasian Soil Sci 2010;13:1439–57. [2] Kılıc Ӧ. The influence of high temperatures on limestone P-wave velocity and Schmidt hammer strength. Int J Rock Mech Min Sci 2006;43:980–6. [3] Yavuz H, Demirdag S, Caran S. Thermal effect on the physical properties of carbonate rocks. Int J Rock Mech Min Sci 2010;47:94–103. [4] Brotóns V, Tomás R, Ivorra S, Alarcón JC. Temperature influence on the physical and mechanical properties of a porous rock: San Julian's calcarenite. Eng Geol 2013;167:117–27. [5] Jansen DP, Carlson SR, Young RP, Hutchins DA. Ultrasonic-imaging and acoustic-emission monitoring of thermally-induced microcracks in lac-dubonnet granite. J Geophys Res-Sol Ea 1993;98:22231–43. [6] Glover PWJ, Baud P, Darot M, Meredith PG, Boon SA, Leravalec M, Zoussi S, Reuschle T. Alpha/beta phase-transition in quartz monitored using acoustic emissions. Geophys J Int 1995;120:775–82. [7] Faust BB. Maya environmental successes and failures in the Yucatan Peninsula. Environ Sci Policy 2001;4(4–5):153–69. [8] Eastmond A, Faust B. Farmers, fires, and forests: a green alternative to shifting cultivation for conservation of the Maya forest? Landscape Urban Plan 2006;74:267–84. [9] McCabe S, Smith BJ, Warke PA. Exploitation of inherited weakness in firedamaged building sandstone: the ‘fatiguing’ of ‘shocked’ stone. Eng Geol 2010;115:217–25. [10] Aurangzeb LAK, Maqsood A. Prediction of effective thermal conductivity of porous consolidated media as a function of temperature: a test example of limestone. J Phys D Appl Phys 2007;40:4953–8. [11] Chaki S, Tarkali M, Agbodjan WP. Influence of thermal damage on physical properties of a granite rock: porosity, permeability and ultrasonic wave evolutions. Constr Build Mater 2008;22:1456–61. [12] Myrow PM. A new graph for understanding colors of mudrocks and shales. J Geol Educ 1990;38:16–20. [13] Dionísio A, Sequeira-Braga MA, Waerenborgh JC. Clay minerals and iron oxides-oxyhydroxides as fingerprints of firing effects in a limestone monument. Appl Clay Sci 2009;42:629–38. [14] Chakrabarti B, Yates T, Lewry A. Effect of fire damage on natural stonework in buildings. Constr Build Mater 1996;10:539–44. [15] Sheu DD, Presley BJ. Variations of calcium carbonate, organic carbon and iron sulfides in anoxic sediment from the Orca Basin, Gulf of Mexico. Mar Geol 1986;70:103–18. [16] Alonzo-Salomón L. Características geotécnicas de suelos y rocas de los estados de Yucatán y Campeche. México: UADY; 2010.
189
[17] Ozsoy EA, Yilmaz G, Arman H. Physical, mechanical and mineralogical properties of ophiolitic rocks at the Yakakayi dam site, Eskisehir, Turkey. Sci Res Essays 2010;5:2579–87. [18] Maldonado L, Veleva L, Díaz-Ballote L. Characterization of limestone for building in the Yucatan Peninsula. Appl Phys A 2011;103:1105–10. [19] ASTM D2216-98. Standard test method for laboratory determination of water (Moisture) content of soil and rock by mass. West Conshohocken (PA, USA): ASTM International; 2007. [20] ASTM D3148. Standard test for elastic moduli of intact rock core specimens in uniaxial compresion. West Conshohocken (PA, USA): ASTM International; 2007. [21] Yokoyama T, Nakashima S. Color developed of iron oxides during rhyolite weathering over 52,000 years. Chem Geol 2005;219:309–20. [22] Yamanio Y, Takeuchi S, Okumura S, Nakashima S, Yokoyama T. Color measurements of volcanic ash deposits from three different styles of summit activity at Sakurajima volcano, Japan: conduit processes recorded in color of volcanic ash. J Volcanol Geoth Res 2008;178:81–3. [23] CIE (Commission Internationale de l' Eclairage). Recommendations on uniform color spaces, color difference equations, psychometric color terms. Supplement No 2 to CIE Publication No 15, Colorimetry, Bureau Central de la CIE, Paris; 1976. [24] Pospíšil J, Hrdў J, Hrdў Jr. J. Basic methods for measuring the reflectance color of iron oxides. Optik 2007;118:278–88. [25] Grossi CM, Brimblecombe P, Esbert RM, Alonso FJ. Color changes in architectural limestones from pollution and cleaning. Color Res Appl 2007;32 (4):320–31. [26] May-Crespo J, Martínez-Torres P, Alvarado-Gil JJ, Quintana P, Vilca-Quispe L. Photothermal radiometry characterization of limestone rocks from the Península of Yucatán. Int J Thermophys 2012;33:1908–15. [27] Ozguven A, Ozcelik Y. Investigation of some property changes of natural building stones exposed to fire and high heat. Constr Build Mater 2013;38:813–21. [28] Özkahraman HT, Selver R, Işık EC. Determination of the thermal conductivity of rocks from P-wave velocity. Int J Rock Mech Min Sci 2004;41:703–8. [29] ISRM. Rock characterization testing and monitoring. In: Brown ET, editor. ISRM suggested methods, 113–116. Oxford: Pergamon Press; 1981. p. 135–40. [30] Allison RJ, Goudie AS. The effect of fire on rock weathering: an experimental study. In: Robinson DA, Williams RBG, editors. Rock weathering and landform evolution. Chichester: John Wiley & Sons; 1994. p. 41–56. [31] Allison RJ. A non-destructive method of determining rock strength. Earth Surf Process Landf 1988;13:729–36. [32] Moses C, Robinson D, Barlow J. Methods for measuring rock surface weathering and erosion: a critical review. Earth-Sci Rev 2014;135:141–61. [33] Mao XB, Zhang LY, TZ LI, Liu HS. Properties of failure mode and thermal damage for limestone at high temperature. Min Sci Technol 2009;19:0290–4. [34] Tang F, Mao X, Zhang L, Yin H, Li Y. Effects of strain rates on mechanical properties of limestone under high temperature. Min Sci Technol (China) 2011;21:857–61. [35] Ar I, Doğu G. Calcination kinetics of high purity limestones. Chem Eng J 2001;83:131–7. [36] Mustakimah M, Yusup S, Maitra S. Decomposition of calcium carbonate in cockle shell. J Eng Sci Technol 2012;7:1–10. [37] Rizov B. Phase transformations from goethite to hematite and thermal decomposition in various nickeliferous laterite ores. J Univ Chem Technol Metall 2012;47:207–10. [38] Lee JT, Keener TC, Knoderera M, Khang SJ. Thermal decomposition of limestone in a large scale thermogravimetric analyzer. Thermochim Acta 1993;213:223–40. [39] Salehi MH, Beni OH, Harchegani HB, Borujeni IE, Motaghian HR. Refining soil organic matter determination by loss-on-ignition. Pedosphere 2011;21 (4):473–82. [40] Gualtieri AF, Venturelli P. In situ study of the goethite-hematite phase transformation by real time synchrotron powder diffraction. Am Mineral 1999;84:895–904. [41] Cornell RM, Schwertmann U. The iron oxides: structure, properties, reactions, occurences and uses. 2nd ed.Wiley-Vch Verlag GmbH & Co. Weinheim; 2003. [42] Scheinost AC, Chavernas A, Barron V, Torrent J. Use and limitations of secondderivative diffuse reflectance spectroscopy in the visible to near-infrared range to identify and quantity fe oxide minerals in soils. Clay Clay Miner 1998;46(5):528–36. [43] Savitzky A, Golay MJE. Smoothing and differentiation of data by simplified least squares procedures. Anal Chem 1964;36:1627–39. [44] Sherman DM, Waite TD. Electronic spectra of Fe3 þ oxides and oxide hydroxides in the near IR to near UV. Am Mineral 1985;70:1262–9. [45] Hajpál M. The behavior of natural building stones by heat effect. In: Kourkoulis SK, editor. Fracture and failure of natural building stones. Dordrecht: Springer; 2006. p. 439–45. [46] Obojes U, Tropper P, Mirwald PW, Saxer A. The effects of fire and heat on natural building stones: first results from the Gröden Sandstone. In: Fort R, Alvarez De Buergo M, Gomez-Heraz M, Vazques-Calvo C, editors. Heritage, weathering and conservation. Leiden: Taylor and Francis/Balkema; 2006. p. 521–4. [47] Sippel J, Siegesmund S, Weiss T, Nitsch KH, Korzen M. Decay of natural stones caused by fire damage. In: Prikril R, Smith BJ, editors. Building Stone Decay: From Diagnosis to Conservation, 271. Geological Society, London, Special Publications; 2007. p. 139–51.