3.1 Macro- and micro-scale changes
Denudation of the granite samples by Na2SO4 solution was observed and analysed using macroscopic and microscopic images. Figures 6 and 7 show the macroscopic changes to granite samples after wet-dry cycling in the four Na2SO4 solutions of different concentrations. Figure 6 shows circular granite samples with signs of spalling at the edges after 50 cycles in 10% and 20% Na2SO4 solutions. Compared with the other three concentrations, the edge of the sample at 20% concentration showed uneven peeling. In Fig. 7, the semicircular granite sample shows slight peeling at the edges and corners after 50 cycles with 5% and 10% solutions. At the 20% concentration, there were obvious signs of spalling at 40 cycles that increased with further cycling. Because the semicircular granite sample was eroded by Na2SO4 solution, there were three eroded contact surfaces at the edges and corners. Therefore, peeling started from the edges and corners and then gradually spread to the centre, thereby expanding the peeling area. As a result, the overall shape of the sample became smaller and irregular, even losing the semicircular shape. When the granite sample was eroded by Na2SO4 solution, the rate of mineral exfoliation inside the rock was higher than that on the surface.
Figure 8 shows the microscopic changes to the granite samples. At both 5% and 10% concentrations, the edges of mineral particles were first eroded and then gradually peeled-off with increasing cycles, and only a small range of mineral particles directly peeled off. At the 20% concentration, whole dark mineral particles were directly peeled off. Under constant erosion by the Na2SO4 solution, the peeling area of the mineral particles gradually expanded and deepened.
3.2 Changes in mass
Table 1 shows the mass-loss rates of the granite samples. Figure 9 shows the relationship between the mass-loss rate of granite samples and number of wet-dry cycles based on the data in Table 1. While the 0% curve does not change significantly, those of the other three Na2SO4 solutions generally decrease first and then increase. At the beginning of a cycle, the mass-loss rate is negative, which indicates that the mass of the granite sample increases. After a certain amount of cycling damage, the mass of the sample decreases, mass-loss rate results in a positive increase. This phenomenon was more significant with the 20% Na2SO4 solution.
At 50 wet-dry cycles, the mass-loss rate of semicircular samples was nearly double that of circular samples. In 20% Na2SO4 solution, the mass-loss rate of the circular sample was 0.9907% while that of the semicircular sample was 1.7045%. The reason is that the semicircular granite samples had a greater surface area in contact with the solution than the circular samples.
Table 1 Mass-loss rates of granite samples after wet-dry cycling (%)
Na2SO4
concentration
|
Cycles NC
|
0
|
5
|
10
|
15
|
20
|
30
|
40
|
50
|
Circular samples
|
0%
|
0.0000
|
0.0413
|
0.0165
|
0.0247
|
0.0661
|
0.0413
|
0.0661
|
0.0909
|
5%
|
0.0000
|
-0.0642
|
-0.0964
|
-0.0964
|
-0.0562
|
-0.0482
|
0.0241
|
0.0482
|
10%
|
0.0000
|
-0.1490
|
-0.1255
|
-0.1412
|
-0.0784
|
-0.0156
|
0.0863
|
0.1255
|
20%
|
0.0000
|
-0.2964
|
-0.1716
|
-0.0936
|
0.0780
|
0.2028
|
0.6630
|
0.9907
|
Semicircular samples
|
0%
|
0.0000
|
0.0653
|
0.0490
|
0.0653
|
0.0980
|
0.0817
|
0.1144
|
0.1307
|
5%
|
0.0000
|
-0.0494
|
-0.0824
|
-0.0824
|
-0.0494
|
-0.0329
|
0.0329
|
0.0824
|
10%
|
0.0000
|
-0.1136
|
-0.1136
|
-0.1299
|
-0.0649
|
-0.0324
|
0.0812
|
0.2273
|
20%
|
0.0000
|
-0.2110
|
-0.1298
|
-0.0974
|
0.0974
|
0.2597
|
0.7792
|
1.7045
|
3.3 Changes in surface roughness
The surface roughness of granite samples is expressed by the arithmetic mean deviation (Ra). As shown in Fig. 10, in the case of the 0% Na2SO4 solution, the roughness did not change significantly with cycle number. In the 5%, 10% and 20% Na2SO4 solutions, with increasing cycles, roughness shows an increasing trend overall. This process of increase can be divided into two stages: 1) a slow stage (0–40 cycles) where roughness increases slowly with cycle number. Because the Na2SO4 solution left crystals on the surface of the granite or filled areas where crystals fell off, the data for each cycle fluctuated up and down slightly, with a slight increase overall. With 5%, 10% and 20% Na2SO4 solutions, the initial roughnesses of the circular granite samples were 0.031, 0.035 and 0.041 mm, respectively. After 40 cycles, the roughnesses of circular samples were 0.045, 0.044 and 0.050 mm, respectively, which are 0.014, 0.009 and 0.009 mm higher than their initial values. The initial roughnesses of semicircular granite samples were 0.042, 0.040 and 0.039 mm and, after 40 cycles, these increased by 0.002, 0.005 and 0.029 mm to 0.044, 0.045 and 0.068 mm, respectively. 2) The second stage is a rapid growth stage (40-50 cycles). Here, roughness increases rapidly with cycle number, most obviously with the 20% Na2SO4 solution. After 50 cycles with 5%, 10% and 20% Na2SO4 solutions, the roughnesses of circular granite samples were 0.052, 0.064 and 0.072 mm and those of semicircular granite samples were 0.054, 0.074 and 0.171 mm, respectively. Compared with that at 40 cycles, the roughnesses of circular samples were 0.007, 0.020 and 0.022 mm greater, while those of semicircular samples were 0.010, 0.029 and 0.103 mm greater.
The above results show that the concentration of Na2SO4 solution has a significant effect on the surface roughness of granite. The higher the concentration of salt solution, the greater the change in surface roughness.
3.4 Changes in indentation hardness
Figure 11 shows the load-displacement deformation curves derived from the point-load indentation hardness tests, which shows the deformation characteristics of the granite samples during the indentation process. The peak load of the initial sample was the highest, with the peak values of the curves decreasing obviously with increases in the number of cycles. At a given number of cycles, the peak values decrease with increases in concentration. At 10 cycles, the peak load decreased from 4604.31 to 3382.65 as the Na2SO4 solution concentration increased from 0% to 20%. At 20 cycles, the peak load decreases from 4848.33 to 3658.95. At 30 cycles, the peak load decreases from 4623.56 to 3002.87. At 40 cycles, the peak load decreases from 4813.93 to 3098.40. At 50 cycles, the peak load decreases from 5122.38 to 3032.51. At a given number of cycles, the curve peak is lowest and the vertical displacement is the largest with the 20% Na2SO4 solution. With 0% and 5% concentrations, the deformation curves show obvious brittleness characteristics, and the relationship between load and vertical displacement is almost linear. At concentrations of 10% and 20%, the samples exhibit plastic and brittle characteristics and the curve shows a convex shape in a certain range before reaching the peak load.
Figure 12 shows the variation in indentation hardness Py in the point-load test obtained by Eq. (2). With increasing cycle number, the fluctuation of the granite with the sample 0% solution shows almost no change, while those of samples with 5%, 10% and 20% solutions all show downward trends. The average hardness of the granite samples stays almost the same (2322.62–2333.18) between 10-50 cycles with the 0% Na2SO4 solution. With the 5% solution, the average hardness decreases from 2116.62 to 1839.66. At 10%, the average hardness decreases from 2041.67 to 1582.04. At 20%, the average hardness decreases from 1850.99 to 1507.38. At a given number of cycles, the greater the Na2SO4 concentration, the lower the indentation hardness.
Figure 13 is a graph from which the plasticity coefficient can be derived according to the initial load-displacement curves of the granite samples and Eq. (3). The plasticity coefficients were obtained by calculation, as shown in Fig. 14. It shows that at a given Na2SO4 concentration, the plasticity coefficient gradually increases with cycle number. The plasticity coefficient shows a linear increase with the 10% and 20% Na2SO4 solutions (Fig. 15). The plasticity coefficient increases from 1.41 to 1.53 over cycles 10-50 with the 5% Na2SO4 solution. The plasticity coefficient increases from 1.46 to 1.53 with the 10% solution and from 1.59 to 2.27 with the 20% solution. It can be seen from Fig. 16 that at a given cycle number, the higher the concentration, the lower the indentation hardness and the larger the plasticity coefficient of a granite sample. After 50 cycles, the indentation hardness and plasticity coefficient of the sample in 20% Na2SO4 solution were significantly different from the initial values.