Mechanics of Materials
Impact of Clay Mineral Type on Rock Properties under Drying and Wetting Cycles
2024 Volume 65 Issue 11 Pages 1390-1396
Details
2024 Volume 65 Issue 11 Pages 1390-1396
Rocks containing clay minerals exhibit diminished strength and undergo physical and chemical deterioration of their texture due to the presence of these clay minerals. Specifically, rocks containing swelling clay minerals experience a significant decrease in strength when subjected to repeated drying and wetting compared to rocks with non-swelling clay minerals. This study conducted a comprehensive series of experiments (physical, uniaxial compressive, and swelling pressure tests) using artificial soft rock mixed with clay minerals. The primary objective was to elucidate the impact of the types of clay minerals present on the physical properties of rock materials under repeated drying and wetting. The clay minerals investigated in this study included Na-type and Ca-type smectites (swelling clay minerals) and mica (non-swelling clay minerals). The smectite-mixed specimen demonstrated higher swelling pressure than the mica-clay-mineral-mixed specimen. Moreover, the uniaxial compressive strength and P-wave velocity of the specimens were remarkably reduced due to repeated drying and wetting process. Hence, the difference in clay mineral content and type in the clay mineral-bearing rock material specimens directly influences the physical and mechanical properties of rock materials when subjected to dry and wet cycles. Furthermore, the outcomes suggest a close association between the reduction in rock material strength, inclusive of clay minerals, and the swelling pressure observed in the specimen.
This Paper was Originally Published in Japanese in J. Soc. Mater. Sci., Japan 73 (2024) 198–204.
The rocks that make up the surface layer of the Earth’s crust often contain a wide variety of clay minerals, which are the products of various alteration processes, including weathering. Rocks containing clay minerals usually exhibit much physical and chemical deterioration in texture due to the influence of clay minerals. Because of this deterioration, clay minerals are often cited as a factor in disasters such as slope failures, deformation, and collapse during construction work. Therefore, when carrying out bedrock construction, the clay minerals present in the bedrock are often treated as a nuisance due to their unique properties. Clay minerals, such as smectite, also exhibit swelling properties upon water absorption and shrinkage upon drying. Therefore, rocks containing swelling clay minerals are more severely degraded by repeated drying and wetting than those containing non-swelling clay minerals.
To date, many studies have been conducted to evaluate the physical properties of geomaterials and rock materials containing clay minerals such as smectite, including water absorption, swelling pressure characteristics [1], deterioration characteristics [2], specific resistance [3], point load strength [4, 5], and uniaxial compressive strength [4–7]. Kohno et al. [7] investigated the effects of the type and content of clay minerals on the physical properties of geomaterials and rock materials containing clay minerals using a gypsum-clay mineral-mixed specimen (artificial soft rock). This study revealed that the uniaxial compressive strength of a specimen under dry conditions varies depending on the type of chemical bond that acts between the layers of clay minerals. However, it has also been reported that the uniaxial compressive strengths of specimens containing smectite (a swelling clay mineral) and mica clay mineral (a non-swelling clay mineral) are comparable. This is because the test results were obtained by performing a uniaxial compressive test on the specimen in dry conditions without wetting after the specimen was prepared. Smectite is a swelling clay mineral that absorbs water between its layers when wet and swells; when dry, it shrinks due to dehydration, causing the material to deteriorate. Therefore, when a specimen containing swelling clay minerals is subjected to a uniaxial compressive test after repeated drying and wetting, the uniaxial compressive strength will be lower than that of a specimen containing non-swelling clay minerals, even under the same dry conditions. On the other hand, although some studies [8–11] have reported on changes in the physical properties of rocks subjected to repeated drying and wetting, very few studies have focused on the presence or absence and types of clay minerals in rocks.
This study conducted drying-wetting cycle tests using artificial soft rock [7] mixed with various clay minerals (or unmixed). We then experimentally investigated how the physical properties of rock materials change due to repeated drying and wetting, focusing on the presence or absence of clay minerals and swelling properties. We attempted to consider the influence of clay mineral content on the strength reduction of rock materials due to repeated drying and wetting, focusing on the swelling pressure generated in the specimen when immersed in water.
In this study, we investigated the effect of clay minerals on the strength reduction of rock materials due to repeated drying and wetting, focusing on the swelling pressure generated in the specimen. Therefore, as mixed clay minerals were considered in this study, we used powder samples with Na-type and Ca-type smectite as the main components (swelling clay minerals) and mica clay minerals (non-swelling clay minerals). As shown in Fig. 1, clay minerals are layered silicate minerals with a 2:1 structure and a layered crystal structure consisting of a combination of tetrahedral and octahedral sheets [12]. Another characteristic is that the unit layers are mainly connected to each other by ionic bonds, and the distances between the unit layers under dry conditions are approximately the same.
Schematic crystal structure of smectite and mica (based on Shirozu [12]).
Figure 2 shows the X-ray powder diffraction (XRD) patterns (Rigaku Ultima IV diffractometer, CuKα, 40 kV, 20 mA, receiving slit: 0.15 mm, divergence slit: 0.5°, scattering slit: 2°, scanning range: 5–40°) of mixed clay minerals in dry and wet conditions. From the XRD pattern, accompanying minerals such as quartz and feldspar were confirmed in the powder sample, which had smectite as the main component. Smectite was quantified using an internal standard method. The smectite content was 64 mass% in the sample mainly composed of Na-type smectite and 82 mass% in the sample mainly composed of Ca-type smectite. These values are comparable to those of Na-type bentonite (Kunigel-V1) and Ca-type bentonite (Kunibond), which are representative samples whose main components are montmorillonite (smectite) (Na-type montmorillonite content: 57%; Ca-type montmorillonite: 84%) [13].
Unoriented X-ray powder diffraction (XRD) patterns of mixed clay minerals under dry and wet conditions. A: Na-smectite, B: Ca-smectite, and C: mica powders.
The artificial soft rock specimen was created using the method described by Kohno et al. [7]. The specimen was prepared by mixing bassanite powder as a solidifying material, quartz powder, and clay mineral powder. Half of the distilled water was added to these powders to form a slurry, which was poured into a mold and solidified. Cylindrical specimens (diameter: 30 mm, height: 60 mm) with different clay mineral mixing ratios were created by maintaining a constant ratio of bassanite and varying the ratio of quartz powder to clay mineral powder. As shown in Table 1, the clay mineral mixing ratio Cc (= percentage of (mass of clay minerals/total mass of specimen) under dry conditions) was set to Cc = 0% and 10%. For the Ca-type smectite, specimens with Cc = 20% were prepared, and the influence of the clay mineral mixing ratio was investigated. For example, in the case of a specimen with Cc = 20%, the ratio was 7:1:2 (bassanite:quartz:clay mineral; mass ratio), and in the case of a specimen with Cc = 0%, the ratio was 7:3:0. The dry density and effective porosity of all prepared specimens were approximately 1.36 Mg/m3 and 50%, respectively, regardless of the type of clay mineral and whether it was mixed. In addition, no major changes in these physical properties were observed because of repeated drying and wetting.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/MT-Z2024015tab01.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
Figure 3 shows a flow diagram of the drying and wetting cycles. In the drying and wetting cycles, the wetting process (the specimen was immersed in distilled water at a temperature of 22 ± 1°C), air-drying process (the specimen was naturally dried in a constant temperature room at 22 ± 1°C), and drying process (the specimen was oven dried at 60°C) were set to 14 days, 1 day, and 3 days, respectively, based on the results of previous studies [8–11] and preliminary experiments. One cycle lasted 18 days for each process, and 14 cycles were conducted in this study. The number of days for each process was set based on the number of days until the mass of the specimen became constant after wetting or oven-drying. In this study, the mass of the specimen was measured using an electronic balance (resolution: 0.001 g). It was determined to be constant on the day that the value displayed on the electronic balance was the same as that displayed on the previous day. When the specimen transitioned from wet to dry conditions, an air-drying process was provided to maintain the drying temperature environment for 3 days, considering the capacity of the drying oven. In addition, a previous study [7] confirmed that the hydration water of gypsum (CaSO4·2H2O) is not dehydrated if the drying temperature is below 70°C. In this study, the drying temperature was set at 60°C to keep the drying temperature as low as possible within the range of 70°C or less and to shorten the drying time as much as possible.
Flow chart of drying and wetting cycles test.
During the drying and wetting cycle tests, the mass (wet, submerged, and dry) of the specimen was measured to calculate the density and effective porosity, as well as the P-wave velocity, uniaxial compressive, and swelling pressure tests. P-wave velocity measurements and uniaxial compressive tests (axial strain rate of 0.1%/min) were performed at 0 and 1 cycles and at the end of an even number of cycles. Therefore, P-wave velocity measurements and uniaxial compressive tests in this study were conducted on specimens in dry conditions. P-wave velocity measurements were performed using an oscillator (resonant frequency of 75 kHz), a function generator, and an oscilloscope. The trigger signal generated by the function generator was set to a square wave with a frequency of 60 Hz and an amplitude of 10 Vpp. The propagation time Tp was read up to 0.1 µs three times for each specimen, and the elastic wave velocity Vp (l/Tp, l: specimen length) was calculated using these average values. As shown in Fig. 4, the swelling pressure test was conducted by immersing the specimen in distilled water in an environment in which no confining pressure other than water pressure was applied to the side surface of the specimen. The load F generated in the vertical direction was measured at 1-second intervals, and the swelling pressure (F/A, A: cross-sectional area of the specimen) was calculated. Swelling pressure tests were conducted during a 14-day wetting process. A series of tests, including drying and wetting cycles, were conducted in a constant temperature room (22 ± 1°C).
Schematic diagram of the swelling pressure test for artificial soft rock mixed with clay minerals.
Figure 5 shows the change in P-wave velocity in the specimen’s dry condition due to repeated drying and wetting. The average values of the test results for four or more specimens are plotted in Fig. 5. Measurements were also attempted for the Na- and Ca-type smectite mixed specimens after the 12th cycle. However, data were not obtained under the measurement conditions used in this study; therefore, the results were considered missing.
P-wave velocity of the artificial soft rock mixed with clay minerals under dry condition by the drying and wetting cycles test.
The P-wave velocity at the 0th cycle is Vp0 = 2.46 to 2.55 km/s, and there are no major differences due to the type and content of mixed clay minerals. However, as shown below, after the 1st cycle of repeated drying and wetting, differences in the P-wave velocity change trends were observed due to the differences in the type and content of mixed clay minerals. The P-wave velocity of the clay mineral-unmixed specimen (Cc = 0%) in 0 to 14th cycles was similar at Vp0-14 = 2.49 to 2.63 km/s, and the influence of repeated drying and wetting cycles could hardly be confirmed. Similarly, almost no changes were observed in the mica-clay-mineral-mixed specimens until the 6th cycle. After the 8th cycle, the P-wave velocity showed a slight decreasing tendency, and in the 14th cycle, the P-wave velocity was Vp14 = 2.26 km/s, which was approximately 10% lower than the initial value (0th cycle). In contrast, the P-wave velocity of the specimen mixed with smectite (swelling clay mineral) showed a decreasing trend from the 1st cycle onwards. The P-wave velocity at the 10th cycle was the lowest in the Na-type smectite mixed specimen (Vp10 = 1.13 km/s). Because of repeated drying and wetting, the P-wave velocity decreased by more than 50% compared to the P-wave velocity at the 0th cycle. Furthermore, when comparing the decreasing trends of the P-wave velocity of the Na- and Ca-type smectite mixed specimens at Cc = 10%, the degree of decrease was greater for the former. The degree of decrease in the P-wave velocity of the Ca-type smectite-mixed specimen was greater when Cc = 20% than when Cc = 10%. From the above, the type (swelling or not) and content of clay minerals influence the change in the P-wave velocity of rock materials due to repeated drying and wetting.
3.2 Changes in uniaxial compressive strength due to repeated drying and wettingTable 2 shows the uniaxial compressive strength before drying and wetting (0th cycle) and at the end of the 14th cycle. As shown in Table 2, the uniaxial compressive strength before the drying and wetting cycles (0th cycle) differed depending on the specimen. Therefore, in this study, as shown in Fig. 6, the changes in the uniaxial compressive strength of the specimen under dry conditions due to repeated drying and wetting are summarized as Sci/Sc0 (i = 0 to 14 cycles). The average values of the test results for four or more specimens are plotted in Fig. 6. The coefficients of variation of the unconfined compressive strength obtained in this study were all less than 15%, and more than half were less than 10%.
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/MT-Z2024015tab02.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
Uniaxial compressive strength ratio Sci/Sc0 of the artificial soft rock mixed with clay minerals under dry condition by the drying and wetting cycles test.
From Table 2 and Fig. 6, the uniaxial compressive strength of all the specimens decreased due to repeated drying and wetting, regardless of the type and content of the mixed clay minerals. The strength tended to decrease slowly from the 6th to 8th cycles, except for the mica-clay-mineral-mixed specimen. This may be related to the material properties of the gypsum used for the artificial soft rock rather than the influence of clay minerals and should be verified in the future. Uniaxial compressive strengths Sc0 and Sc14 at the end of the 0th and 14th cycles. The uniaxial compressive strength of the clay mineral-unmixed specimen (Cc = 0%) was Sc14 = 5.2 MPa, approximately 53% lower than that of Sc0 = 11.1 MPa. In other words, the unconfined compressive strength decreased because of repeated drying and wetting, even if clay minerals were not included. In addition, the uniaxial compressive strength of the mica-clay-mineral-mixed specimen was Sc14 = 4.9 MPa, which is approximately 58% lower than Sc0 = 11.7 MPa. A slight decrease in strength was observed compared to the specimen without the clay mineral mixture (Cc = 0%). In the smectite-mixed specimen, Sc14 decreased relative to Sc0 as follows: it decreased by approximately 78% in the Na-type smectite-mixed specimen, 70% in the Ca-type smectite-mixed specimen (Cc = 10%), and 75% in the Ca-type smectite-mixed specimen (Cc = 20%). When comparing the three types of specimens containing 10% clay minerals, we found that the uniaxial compressive strength of specimens mixed with swelling clay minerals (Na- and Ca-type smectites) was significantly lower than that of specimens mixed with non-swellable clay minerals (mica clay minerals). In other words, the degree of decrease in the uniaxial compressive strength due to repeated drying and wetting differed depending on the presence or absence of swelling in the clay mineral. In addition, the degree of decrease in the uniaxial compressive strength of the Ca-type smectite-mixed specimen was greater when Cc = 20% than when Cc = 10%. Therefore, the degree of uniaxial compressive strength reduction depends on the clay mineral content. Furthermore, the degree of the uniaxial compressive strength reduction in the specimen containing 10% Na-type smectite was lower than that in the specimen containing 10% Ca-type smectite. Therefore, the uniaxial compressive strength due to repeated drying and wetting varied depending on the type of smectite used (degree of swelling). From the above, the type (swelling or not) and content of clay minerals affect the change in the uniaxial compressive strength of rock materials due to repeated drying and wetting.
Figure 7 shows the relationship between the P-wave velocity and the uniaxial compressive strength for each cycle in the drying and wetting cycle tests. Regardless of the type and content of the clay minerals mixed in the specimen, the uniaxial compressive strength Sc tended to increase as the P-wave velocity Vp increased. This trend is consistent with previous reports [14]. The slope of the Vp-Sc relationship differs depending on the presence or absence of clay minerals and the type and content of the mixed clay minerals. The decrease in the P-wave velocity and uniaxial compressive strength due to repeated drying and wetting was remarkable for the specimen mixed with smectite. In contrast, if we focus on the clay mineral-unmixed specimen (Cc = 0%) and the mica-clay-mineral-mixed specimen (non-swelling clay mineral), although the uniaxial compressive strength decreased due to repeated drying and wetting, the P-wave velocity hardly decreased compared to the specimen mixed with smectite. From the above, although the P-wave velocity and uniaxial compressive strength of the specimen decreased due to repeated drying and wetting, the degree of decrease became even greater due to the presence of swelling clay minerals. In the next section, we discuss the changes in the P-wave velocity and uniaxial compressive strength of rock materials due to repeated drying and wetting based on the difference in swelling pressure in the specimen when immersed in water.
Figure 8 shows the swelling pressure generated in the specimens during the 1st cycle of the drying and wetting test. The swelling pressure varies depending on the type and content of the mixed clay minerals. The swelling pressures of the clay mineral-unmixed specimen (Cc = 0%) and mica-clay-mineral-mixed specimen were approximately the same. The swelling pressures of both specimens were extremely low (less than 0.01 MPa), indicating that almost no swelling pressure was generated. All the specimens mixed with smectite generated a swelling pressure of 0.01 MPa or more, and the Ca-type smectite mixed specimen (Cc = 20%) had the highest swelling pressure, approximately 0.05 MPa. The swelling pressure of the Na-type smectite-mixed specimen exceeded 0.04 MPa immediately after the measurement started and then gradually decreased. However, the swelling pressures of the Na- and Ca-type smectite mixed specimens (Cc = 10%) were higher for the former. These features are also visible in the XRD pattern (Fig. 2). The diffraction lines of Na-type and Ca-type smectites under the dry condition (2θ CuKα = 7.6° [d = 11.6 Å] and 2θ CuKα = 5.9° [d = 15.0 Å]) were shifted to lower angles of 2θ CuKα = 5.9° [d = 15.0 Å] and 2θ CuKα = 5.7° [d = 15.5 Å], respectively, due to wetting (note that d is the lattice spacing.). This indicates that the interlayer distance of smectite increased due to water absorption; that is, it swelled. In addition, the magnitude of the peak migration distance was larger in the Na-type smectite than in the Ca-type smectite, which could have generated a larger swelling pressure. Conversely, no difference was observed in the peaks in dry and wet conditions for mica clay minerals, suggesting that the interlayers did not spread and almost no swelling pressure was generated. Therefore, the type (swelling or not) and content of clay minerals influence the magnitude of swelling pressure generated in rock materials.
Swelling pressure of the artificial soft rock mixed with clay minerals during water immersion (1 cycle) by the drying and wetting cycles test.
Figure 9 shows the swelling pressure generated in the Ca-type smectite-mixed specimen during the 1st to 4th cycles. Considering the swelling pressure of Ca-type smectite mixed specimens with different mixing ratios, the swelling pressure of the Cc = 20% specimen was higher than that of the Cc = 10% specimen. This is because the swelling pressure increases with the clay mineral content, contributing to swelling. Additionally, the swelling pressure of all specimens decreased as the number of cycles increased. In the 4th cycle, the swelling pressure was less than 0.01 MPa for both Cc = 20% and Cc = 10% specimens. As explained later, the 1st cycle of water immersion caused micro-cracks to occur inside the specimen due to the swelling pressure. Consequently, when measuring the swelling pressure from the 2nd cycle onwards, the micro-cracks generated by water immersion in the 1st cycle become a place for the swelling pressure to escape. Alternatively, the transmission of swelling pressure may be inhibited by micro-cracks, resulting in low swelling pressure. No major changes in the macroscopic characteristics of the specimens were observed because of the repeated drying and wetting. Therefore, the micro-cracks used here do not refer to the macroscopic cracks that divide the specimen; however, the microscopic cracks that occur in the specimen cannot be observed with the naked eye.
Swelling pressure of the artificial soft rock mixed with Ca-smectite (Cc = 10% and 20%) during water immersion (e.g., 1 to 4 cycles) by the drying and wetting cycles test. The numbers in the figure show the number of cycles.
We focused on the degree of decrease in the P-wave velocity of the specimens during the 1st cycle. The degree of decrease in the clay mineral-unmixed specimen (Cc = 0%) and the mica-clay-mineral-mixed specimen was approximately ±2%, and the degree of decrease in the swelling clay mineral-mixed specimen was approximately 6 to 15%. However, the degree of uniaxial compressive strength reduction tendency of the specimen in the 1st cycle was similar to that of the P-wave velocity. There was a difference of approximately 10% between the clay mineral-unmixed specimen (Cc = 0%) and the mica-clay-mineral-mixed specimen and approximately 25% between the swelling clay mineral-mixed specimen. This trend was confirmed from the 2nd cycle onward. In other words, the swelling pressure generated in the specimen due to swelling clay minerals reduced the P-wave velocity and uniaxial compressive strength. Ichinose and Goto [15] investigated the changes in the pore volume of rocks due to repeated drying and wetting using the mercury intrusion method. Consequently, it has been reported that repeated drying and wetting cause micro-cracks and the expansion of latent cracks within the rock, and the overall pore volume increases, leading to a progressive decline in strength. In this study, we assumed that the uniaxial compressive strength reduction of the specimen was due to the weakening of the specimen due to the following two factors: The first is deterioration due to minute swelling and contraction of the specimen due to repeated drying and wetting. Second, especially in the case of specimens mixed with swelling clay minerals, the swelling pressure generated during wetting promotes the generation of micro-cracks and the expansion of latent cracks inside the specimen. We believe this decreased the strength of the specimen and changed the P-wave velocity.
In this study, we conducted drying and wetting cycle tests using artificial soft rock mixed with smectite (swelling clay mineral) or mica clay minerals (non-swelling clay mineral) and experimentally investigated how the physical properties (mainly P-wave velocity and uniaxial compressive strength) of rock materials change due to drying and wetting cycles. In addition, we attempted to consider the changes in the physical properties of rock materials due to repeated drying and wetting, focusing on the swelling pressure that occurs in the specimen when immersed in water. The results are summarized as follows:
Thus, rock materials containing swelling clay minerals, such as smectite, may experience a significant decrease in strength due to repeated drying and wetting. Therefore, understanding the swelling properties of clay minerals when evaluating the mechanical properties of ground and rock materials under such conditions is important. In the future, it will be necessary to clarify the rate and scale of crack occurrence due to the generation of swelling pressure and to quantitatively evaluate the relationship between changes in physical properties due to repeated drying, wetting, and swelling pressure. Furthermore, it is necessary to verify that natural rocks contain clay minerals.
This work was partly supported by Grants-in-Aid for Scientific Research ‘KAKENHI’ (grant number 19K15489) from the Japanese Society for the Promotion of Sciences (JSPS). We gratefully acknowledge this support.
