Abstract
Chemical corrosion plays a significant role in affecting the properties of rock materials. To understand the effects of chemical corrosion on the pore structure and mechanical properties of sandstones, porosity, spectrum distribution, and NMR images of sandstone specimens were measured after every 10 days of immersion in chemical solutions using the nuclear magnetic resonance (NMR) technique. Static uniaxial compressive tests and dynamic compressive tests were conducted using a conventional servo-controlled testing machine and a split Hopkinson pressure bar (SHPB) system for specimens treated with chemical corrosion. The test results showed that after being treated with chemical corrosion, the porosity of a specimen increased, the spectrum distribution would successively shift towards the right, and the distribution of pores tended to become more irregular. Additionally, all of the compressive strength and elastic modulus of sandstone treated with chemical corrosion under static and dynamic loads decreased, and the peak strain increased. The effect order of a chemical solution on the pore structure and mechanical properties of sandstone was H2SO4>NaOH>distilled water, which would be related to the different mechanisms of a water-rock reaction. According to the experimental results, the correlations between the mechanical properties and porosity were established. The results can serve as a reference for research in related fields.
1. Introduction
Rock, as a natural geological body, has a pore structure that is closely related to the surrounding environment [1]. Water is one of the most active factors in the environment that affects the porosity of rocks and the safety of rock mass engineering [2, 3]. The change of the microscopic structure and mineral composition of rocks is the essence of a water-rock reaction, which is able to increase the porosity and then affect the physical and mechanical properties of the rock. In recent years, the influence of chemical corrosion on the physical-mechanical properties of rocks has received increasing attention from researchers, especially in the study of large-scale geotechnical engineering problems, such as oil extraction, subgrade, slope, chamber, nuclear waste geological repository, energy underground reservoir, and carbon dioxide geological storage reservoir. A large number of research results have been obtained [3–10].
The effect of chemical corrosion on the pore structure of rocks is time dependent and manifests as the variety of rock porosities with the difference of corrosion time. It is helpful to analyze the corrosion mechanism of chemical solutions and the degradation characteristics of rock mechanical properties to develop the study of the evolution of rock pore structure under the condition of chemical corrosion. Many advanced techniques, such as CT scanning [11–14], SEM [15–18], and -wave velocity [1, 19–21], have been applied to the measurement of pore structures of rocks. These techniques have unique advantages in some aspects, but they cannot be used to obtain the comprehensive effect of chemical corrosion on the pore structure of rocks, such as the distribution and size of pores. Nuclear magnetic resonance (NMR) technology [22, 23], which is a nondestructive spectroscopic testing technique, has been widely used to analyze the internal pore structure of rocks [14, 24–30]. The NMR technique has powerful abilities to obtain the evolution of the pore structure inside rocks by testing porosity, spectrum distribution [31], and NMR images [23, 29, 32]. Therefore, the NMR technique is used to investigate the pore structure of rocks in this work.
Rocks exist in a multifield-coupled environment, and many geological disasters of rock mass engineering related to chemical corrosion are caused by the coupling of mechanical action and chemical effect [9, 10]. The deterioration of the static mechanical properties of various rocks treated with different types of chemical corrosion has been widely investigated [4, 11, 33–37]. The mechanical properties which were investigated include UCS, elastic modulus, and Poisson’s ratio. These researches were mainly done in the context of static loading conditions. The previous researches mainly focused on the static mechanical properties, and limited studies have been conducted on the dynamic mechanical behaviors of rocks treated with chemical corrosion.
Dynamic stress or impact loads, such as rockbursts, earthquakes, and blasting, have significant effects on the safety and stability of rock engineering, and many rock engineering disasters are caused by dynamic stresses or shocks. Under the action of dynamic loading, the dynamic mechanical behaviors of rocks, such as compressive strength, deformation, and failure process, are quite different from those of rocks under the static state [38–43]. Therefore, it is a very important issue to determine the rock dynamic properties accurately [44]. However, the results of the current limited researches are still far from a complete understanding of the dynamic mechanical behaviors of rocks treated with chemical corrosion.
In this study, to investigate the effect of chemical corrosion on the pore structure and the static and dynamic mechanical properties of sandstone, the NMR technique was applied to detect porosity, distribution, and NMR images of sandstone specimens. Then, static and dynamic uniaxial compression tests were carried out to determine the static and dynamic UCS and elastic modulus. In addition, correlations of porosity and mechanical properties of sandstone were established based on the experimental data.
2. Materials and Experimental Methods
2.1. Material Preparation
The rock material tested is fine- to medium-grained red sandstone from the subgrade of a road engineering project in Liuyang City, Hunan Province. All samples were cored from an intact sandstone block with a high degree of integrity and uniformity to eliminate interference factors and ensure the reliability of the test results. The X-ray diffraction (XRD) technique was used to detect the mineral components of the sandstone, and the detected results are shown in Table 1. From Table 1, it can be seen that the sandstone is mainly composed of 71% quartz, 15% feldspar, 8% mica, 4% calcite, and 2% chlorite. Note that the results listed in Table 1 are the average of the samples, and the change of mineral components is little due to the high uniformity of the sandstone block. Thus, the effect of the variety of mineral components can be negligible.
As required by the standards of ISRM suggested in [30, 44, 45], cylindrical specimens with a diameter of were selected. To ensure the roughness of end surfaces less than 0.2 mm and the nonparallelism within ±0.1%, the ends of specimens were polished. To enhance the accuracy of the tests, the specimens with larger discreteness were excluded by comparing the density, longitudinal wave velocity (LWV), and porosity. The mean porosity, density, and LWV of specimens retained were 5.06%, 2.32 g/cm3, and 2320 m/s, respectively. In general, two sets of specimens were prepared. The specimens of the first set, with a length/diameter ratio of 2.0, were used in static tests. Specimens of the other set which have a length/diameter ratio of 1.0 were used in dynamic SHPB tests.
Rock mass engineering is always in a complex chemical environment where the ionic composition is complex and the pH value varies. In order to investigate the effect of chemical solutions on rock mass in a laboratory environment and simplify the experimental study, three chemical solutions with varying pH values were prepared in this work, i.e., an acidic solution of H2SO4 (pH 2.0), distilled water (pH 7.0), and an alkaline solution of NaOH (pH 12.0). There are four test groups, i.e., A, B, C, and D. Each group concluded six specimens (labeled as 4, 5, 6, 7, 8, and 9). The specimens labeled as A, B, and C were immersed in H2SO4 solution, distilled water, and NaOH solution, respectively. Specimens from group D represent the natural group.
2.2. Experimental Procedure
In this work, the specific experimental procedure is as follows: (1) The sandstone specimens are prepared. Specimens were cut into cylinders and excluded by comparing the quality, LWV, and porosity, then they were divided into four groups. (2) The chemical solutions are prepared. H2SO4 solution (pH 2.0), distilled water (pH 7.0), and NaOH solution (pH 12.0) were prepared, and the specimens from groups A, B, and C were soaked in H2SO4 solution, distilled water, and NaOH solution, respectively. (3) NMR tests were conducted. The porosity, spectrum distribution, and NMR images of the specimens every 10 days of immersion in chemical solutions were tested until the porosity stays unchanged. (4) Mechanical tests were conducted. For all of the groups, the specimens (labeled as 4, 5, and 6) were tested under the static uniaxial compressive loading, and the specimens (labeled as 7, 8, and 9) were tested by the SHPB system. The complete experiment schedule is shown in Figure 1.
2.3. NMR Tests
The principle of nuclear magnetic resonance detection technology is that the atomic nucleus is magnetized by a magnetic field producing a response to the radio frequency; the response can then be reflected by signals that can be measured by the NMR system [23]. Because the rock skeleton has no signal, NMR technology detects the signal of the fluid in the pores of the rocks. When the pores are filled with fluid, the fluid volume is equivalent to the pore volume, so the porosity measured by NMR technology can truly reflect the actual porosity of the rock. The distribution of pores inside the rock is not uniform, and their sizes vary. These result in the different attenuation characteristics obtained in NMR detection. This difference can be expressed by a curve which is called spectrum distribution. Magnetic resonance imaging is a technique for obtaining the distribution and sizes of pores inside the rock by analyzing the signal attenuation of the atomic nucleus.
The NMR technique has been widely used to measure the porosity, spectrum distribution, and NMR images of rocks [23, 25–30, 33]. In this work, the AniMR-150 NMR imaging system was used to measure the porosity, distribution, and NMR images of sandstone specimens every ten days of immersion in chemical solutions (H2SO4 solution, distilled water, and NaOH solution).
2.4. UCS and SHPB Tests
To investigate the mechanical behavior of sandstone immersion in chemical solutions, the static and dynamic compressive loading tests were carried out.
The static compression experiments were performed using an Instron-1346 servo-controlled material testing system with a 2000 kN-capacity loading frame. During a test, the axial load and displacement are recorded by a linear variable differential transducer (LVDT). The experiment used the axial displacement control, and specimens were compressed with a loading rate of 0.15 mm/min until failure.
The Splite Hopkinson pressure bar (SHPB) experimental technique is a reliable experimental method for studying the dynamic mechanical properties of materials. The dynamic mechanical tests were carried out on the SHPB test system of Central South University. The schematic of the system is shown in Figure 2. The system consists of a specially shaped striker, an incident bar, a transmission bar, an absorption bar, and a damper. The incident bar, transmission bar, and absorption bar are all 40Cr steel with a longitudinal wave velocity of 5400 m/s, a density of 7810 kg/m3, and an elastic modulus of 240 GPa [30, 38, 46]. Two strain gauges are attached to the incident and transmission bars to record the deformation of the two bars.
During testing, the gas gun is opened and the specially shaped striker is shot out and impacts the incident bar. Then, an elastic compressive wave is generated and propagates along the incident bar. At the incident bar/specimen interface, the wave is partially reflected and partially transmitted. A portion of the incident wave is reflected back along the incident bar, and the transmission wave propagates along the transmission bar. The strains can be measured with the strain gauges on the bars, stored in digital waveform memory via a dynamic strain gauge, and finally processed by the data processing system.
According to the assumption of 1- wave propagation [47], the relationship of the stress and strain of the specimen can be calculated as and the forces and acting on the specimen shown in Figure 3 can be calculated as where , , and represent the wave velocity, the cross-sectional area, and the elastic modulus of the bars, respectively. and are the original area and length of the specimen, respectively., , and are the incident strain, reflected strain, and transmission strain, respectively.
(a) H2SO4 solution
(b) Distilled water
(c) NaOH solution
3. Evolution of Pore Structure of Sandstones
3.1. Porosity Increment of Specimens under Different Chemical Conditions
The effect of a water-rock reaction on the pore structure of sandstone is time dependent, and the reactions are different with different chemical solutions. In this study, the porosity of specimens was measured by the NMR technique. Since the porosity of specimens from all groups is stable after 50 days of immersion in chemical solutions, the porosity of sandstone specimens with immersion times of 0, 10, 20, 30, 40, and 50 days is shown in Table 2. In order to analyze the variation of porosity more intuitively, the curves of the average porosity of specimens in different chemical solutions are shown in Figure 4.
As Figure 4 displays, the variation of the porosity of sandstone is different with different chemical solutions. When soaked in distilled water, the porosity of sandstone increases significantly in the first 20 days (with an increase rate of 20.77%) and remains unchanged in the last 30 days (with an increase rate of 3.42%). When soaked in H2SO4 and NaOH solutions, the characteristics of the two curves are similar. In the first 40 days, the porosity increases obviously and the increase rate decreases with the increase of immersion time. After 40 days of immersion, the curve is almost parallel to the axis and the porosity tends to be stable. Compared with the curve of specimen immersion in NaOH solution, it can be seen that the increase rate of samples soaked in acid solution is larger, which results in the larger porosity of specimens in the acid solution. Moreover, it also can be concluded that the effect order of chemical solutions on the development of pores inside sandstones is H2SO4>NaOH>distilled water.
3.2. Spectrum Distribution
The characteristics of pore size distribution within the rock can be represented by the curve of the spectrum. The transverse relaxation time is positively correlated with pore size, and the signal component corresponding to the peak is correlated with the ratio of the pore [24, 32, 48–50]. The shorter the transverse relaxation time is, the smaller the pore size is, and the larger the signal component corresponding to the peak is, the larger the ratio of the pores with respect to the time (pore sizes) is.
The curves of spectrum distribution of sandstone specimens with different times soaked in chemical solutions are shown in Figure 5. From Figure 5, it can be seen that the pores and their size distribution are dynamically changing. The characteristics of pore development inside the rocks are as follows: (1)There are three peaks (labeled 1, 2, and 3) in distribution of all the samples at the natural state (without any solutions), indicating that the pore structure of the rock specimens is distributed approximately in three sizes, and the concentrations of the pore distribution in the three sizes are different. It can also be seen that the first two spectrum peaks of each specimen have a larger porosity component, which means that the number of tiny and medium pores is greater than that of the large pores inside the rock.(2)When soaked in acid or alkaline solutions, the variation characteristics of the pore structure are different with the increase of soaking time, and the pore structures of specimens change significantly. A new peak appears on the left side of the curve (labeled as peak 4) and the time is the smallest, indicating that smaller pores (secondary pores) are generated in the specimen caused by chemical corrosion, and the number of secondary pores increases first and then decreases. After 30 days of immersion, the number of secondary pores reaches a maximum. The curves shifted to the right with different degrees, and the signal amplitude values are larger than those in the initial state, indicating that the sizes and number of original pores also increase.(3)In distilled water, the curve of the spectrum distribution shifted toward the right gradually and the signal amplitude values corresponding to the peaks increase. However, no new peaks are generated. This indicates that the size and number of pores in the sandstone specimen increase under the effect of water dissolution, and no secondary pores are generated inside the specimens(4)With the increase of soaking time, the signal amplitudes of peaks 1 and 2 on the curves increased significantly, indicating that there is a large increase of tiny and medium pores during the process of the water-rock reaction. These also illustrate the relevance of the evolution of pore size within the specimen. Moreover, it can be drawn that the chemical damage of H2SO4 and NaOH solutions is larger than that of distilled water
(a)
(b)
(c)
3.3. Evaluation of NMR Images
The NMR technique can vividly reflect the distribution of pores inside rocks. The NMR image can directly reflect the NMR signal of the fluid that fills the pores, and the distribution and migration of the pore fluid can exactly reflect the development of the pore damage inside the rocks. The bright spots in the images indicate the distribution of pores inside the samples. The larger the bright spot area, the stronger the signal, and the greater the pore volume. Then, the distribution and evolution of pores inside the sandstone specimens can be obtained. The NMR images of sandstone specimens with different times of immersion in different chemical solutions are shown in Figure 3.
There are two types of pores inside specimens with different days of immersion: one is the secondary pore (labeled as ①) and the other one is the pore results from the initial pore size becoming larger (labeled as ②), as shown in Figure 3. From Figure 3, the number and area of light spots increase significantly with the increase of the soaking time, which means that the number and volume of pores inside specimens increase. When soaked in H2SO4 solution or NaOH solution, the secondary pores are generated after ten days of immersion, and the number of secondary pores increases with the increase of immersion time in the first 30 days and then decreases in the last 20 days. It can also be seen that the light spots in some areas show a trend of connectivity, but there is no continuity. This indicates that the chemical damage of the area in the rock specimen is serious and the pores are more developed, but there is no macroscopic crack. In addition, the borders of NMR images are obviously brighter. This is mainly because the chemical reaction is related to the contact area, the surface of the specimen is larger in the contact area, and the chemical reaction is more intense than in the internal area. Moreover, the number of NMR images of specimens from groups A and C is larger than that of group B, indicating that the chemical damage of specimens soaked in H2SO4 solution and NAOH solution is greater compared with the damage of specimens soaked in distilled water.
In summary, the dissolution and water-rock reaction between the sandstone and H2SO4 solution, NAOH solution, and distilled water will cause chemical damage to the sandstone with different degrees, and the order of influence on the pore development within the sandstone is H2SO4>NaOH>distilled water.
4. Mechanical Property Deterioration of Sandstone
4.1. Static Mechanical Property of Sandstone
Figure 6 shows the typical uniaxial compressive stress–strain curves of sandstone specimens at different chemical solutions. As shown in Figure 6, four different phases are contained in the typical stress–strain curves: the phase of compaction, the phase of linear elasticity, the phase of plastic yielding, and the phase of failure. The first phase is significant because of the larger porosity of specimens. Moreover, the axil strain corresponding to the peak strength is larger under the condition of chemical immersion, which means that the ductility of sandstone treated with chemical immersion increases. Compared with the failure strain of a specimen under the natural state, the increment of failure strain of sandstone specimens soaked in H2SO4 solution is the largest, followed by the increment in NaOH solution, while the increment in distilled water is the smallest.
The results of the static UCS and elastic modulus of sandstone specimens treated with different chemical solutions are shown in Table 3. Figure 7 displays the reductions of the average value of UCS and elastic modulus. Compared with the specimens without any solutions, the chemical damage results in a decrease in the UCS and the elastic modulus. After 50 days of immersion in distilled water, NaOH solution, and H2SO4 solution, the static UCS was reduced, respectively, to 3.66%, 17.09%, and 18.93%, and the elastic modulus decreased, respectively, by 6.45%, 12.44%, and 15.31%. The reductions have a decisive influence on the stability of practical engineering. The results suggest that the deterioration of static mechanical properties occurs during the reaction between the rock and chemical solutions, and the order of chemical solutions affecting the deterioration of static mechanical properties of sandstones is H2SO4>NaOH>distilled water.
(a)
(b)
Previous studies have indicated that some relations existed between the static mechanical properties and the porosity of sandstones. Based on the experimental data in this study, the relationships between porosity and the static UCS and static elastic modulus are shown in Figure 8. It can be seen that the relationships between porosity and the static mechanical properties of the sandstones are nonlinear. Moreover, exponential function has been experimentally investigated by many scholars [1, 51, 52] to describe the relationship. Therefore, exponential function was also used in this study. The results based on the test data in this study show that equations (3) and (4) fit well on the static UCS and elastic modulus with the fitting coefficient of determination () of 0.957 and 0.974, respectively. where and are the dynamic UCS and elastic modulus, respectively, and represents the porosity of sandstone specimens.
(a)
(b)
4.2. Dynamic Mechanical Property of Sandstone
The original and extracted signals of sandstone specimens in a typical SHPB test are shown in Figures 9 and 10. In Figure 9, the signal is continuous and stable and the reflected wave has a good platform segment, indicating that the system is stable and the specimen is in a state of constant strain rate during loading, and the obtained result is reliable. In Figure 10, the transmitted wave () basically overlaps with the curve of the sum () of the incident and reflected waves, indicating that the specimen is basically in the state of stress equilibrium before reaching the peak strength during dynamic loading and the test result is reliable. In addition, the inertial effects are ignored because there is no global force difference in the sandstone specimen to induce inertial forces [38, 46].
The dynamic mechanical parameters are listed in Table 4. By combining Tables 3 and 4, it can be seen that the dynamic strength and elastic modulus are larger than the static strength and elastic modulus of specimens under the same conditions. This may be caused by the specimens that are destroyed instantaneously at the high loading rate.
According to equation (1), the dynamic stress–strain curves of the sandstone specimens treated with different chemical solutions are obtained in Figure 11. The dynamic stress–strain curves can be divided into three phases: the elastic phase, the plastic yielding phase, and the post peak phase. In the first phase, the curve approximately reveals a trend of straight line, which indicates that a compaction still happens but there is only a slow response to the high dynamic load. In the plastic yielding phase, the slope of the curve decreases gradually, which is caused by the continuous mechanical damage of internal materials. At the phase of the post peak, the strength of specimens without any solutions drops rapidly, which means that the effect of the water-rock reaction could weaken the dynamic mechanical capability of the sandstone. Moreover, the slope of stress–strain curves in the phase of elastic and peak stress varies with the different chemical solutions, which results from gradual accumulation of inner chemical damage. The slope and peak stress of specimens without any solutions are the largest, followed by specimens immersed in distilled water, NaOH solution, and H2SO4 solution, respectively. Combined with the results of NMR tests, it can be referred that the initial pore size becomes larger and large numbers of secondary pores are generated when the specimens are soaked in chemical solutions, which results in a significant increase of the failure strain of specimens.
Figure 12 displays the reduction of the dynamic compressive strength and elastic modulus of sandstone specimens subjected to different chemical solutions. It can be seen that the original average dynamic compressive strength is 85.79 MPa, and then the dynamic strength reduces to 77.42 MPa, 65.03 MPa, and 60.08 MPa after treatment with corrosion of distilled water, NaOH solution, and H2SO4 solution, which are 9.76%, 24.2%, and 29.97% reductions, respectively. Similarly, compared with the original dynamic elastic modulus (30.77 GPa), the reductions of the dynamic elastic modulus are 24.73% (23.16 GPa), 43.81% (17.29 GPa), and 50.76% (15.15 GPa) after 50 days immersion in distilled water, NaOH solution, and H2SO4 solution, respectively.
(a)
(b)
Figure 13 displays the dynamic compressive strength and elastic modulus of samples with different porosities. Based on the experimental data, the relations among porosity, the dynamic compressive strength, and elastic modulus of sandstone are regressed. The results reveal that the exponential functions in equations (5) and (6) fit well with the dynamic compressive strength and elastic modulus with the fitting coefficient of determination () of 0.989 and 0.991, respectively. where and are the dynamic compressive strength and elastic modulus, respectively.
(a)
(b)
4.3. Comparison of Experimental Results and the Proposed Relationships
According to the relationships between porosity and the mechanical properties of sandstones under the static and dynamic states, the static UCS and elastic modulus and dynamic compressive strength and elastic modulus of sandstones treated with different times of immersion in a chemical solution can be predicted. To verity the reliability of the relationships, the experimental data of sandstone specimens (labeled as A10, A11, B10, B11, C10, and C11) treated with 20 days of immersion in a chemical solution was applied. The results are shown in Table 5.
As shown in Table 5, the predicted values are close to the experimental results, and the relative errors are all less than 5%, which indicate that the relationships obtained in this study have good basic prediction capability and are applicable for evaluating the strength of sandstones treated with chemical solutions. However, relative errors still exist. This is because only the porosity could be considered in the relationships, but in the test results, all of the factors such as porosity and the distribution and size of pores affect the mechanical properties of the rock. It also can be found that the error rates of specimens treated with immersion in distilled water are the smallest. It may be due to the weak reaction between distilled water and sandstone, which results in weaker pore development inside the rock. These are for future studies.
5. Corrosion Mechanism Analysis
Natural rocks are formed by cemented mineral compounds and internal defects, such as pores and microcracks. At the presence of a chemical solution, the fluid will penetrate into the pores and cracks and interacts with the minerals. Water-rock interactions, including hydrophysical effects and hydrochemical effects, result in the deterioration of rock physical and mechanical properties. In terms of hydrophysical effects, the dissolution of water on rocks reduces the physical and mechanical properties of rocks by reducing the interconnection between the mineral particles and the effectiveness of confining pressure. In terms of hydrochemical effects, the effects deteriorate the physical and mechanical properties of rocks by changing the mineral components and microstructure of rocks, such as the size of particles, pore structures, and crack morphology. The hydrochemical effects are more significant [36], thus the corrosion mechanism of hydrochemical effects are mainly analyzed.
As described in Section 2.1, the sandstone was composed primarily of quartz, feldspar, mica, calcite, and chlorite. When immersed in acid solutions, minerals such as feldspar, calcite, and chlorite can easily react with the hydrogen ion (H+). Quartz is unstable in alkaline solution. Minerals such as calcite are soluble in water. All of these will result in the change of the physical and mechanical properties of the sandstones. The key corrosion mechanisms for the sandstone studied in this work are summarized below.
When soaked in distilled water, these minerals are not likely to dissolve in water over the experiment timescales. Thus, the corrosion mechanisms of the sandstone treated in distilled water are not analyzed in this study and the detailed reaction equations can be found in the literature [10, 32].
In acidic solution, the main reactions between the major mineral compositions and acidic solution are as follows:
Additionally, the mineral composition of sandstone contains chlorite which is active in acid solutions [53, 54], thus the following reactions may also happen when the studied sandstone is soaked in acid solution:
In alkaline solution, the following reactions mainly occur between the sandstone and the hydroxide ion:
Moreover, the concentrations of Ca2+, Na+, K+, and SiO2 were detected in the test and the results are shown in Figure 14. From Figure 14, it can be seen that the ion concentration is dynamically changed under the effect of chemical corrosion. In acid solution, the concentration of Ca2+, Na+, and K+ is higher than that in neutral and alkaline solutions. In alkaline solution, the concentration of SiO2 is higher than that in neutral and acid solutions. These illustrate that different minerals have varying chemical solution sensitivities. Chlorite, feldspar, and calcite (the main component of cement) are unstable in acid solution, and quartz (composition of coarse aggregate) is more active in alkaline solution. The deterioration of the mechanical properties of rocks caused by the dissolution of cement is more significant than that caused by the dissolution of the aggregate [10, 36]. Therefore, the chemical corrosion effect of acid solution is more severe compared with that of alkaline and neutral for the sandstone studied.
6. Conclusions
In this study, the alterations on the pore structure and mechanical properties of sandstone caused by chemical corrosion have been experimentally investigated. Based on the experiments, the following conclusions can be drawn: (1)The pore structure of sandstone treated with chemical corrosion is more developed. When sandstone is soaked in chemical solutions, the porosity increases, the spectrum distribution shifts towards the right, secondary pores are generated, the pore size becomes larger, and the distribution of pores tends to be more irregular. The effect of H2SO4 solution is the greatest, followed by NaOH solution, and distilled water.(2)The mechanical properties of sandstone treated with different types of chemical corrosion will deteriorate. Under static and dynamic states, the compressive strength and elastic modulus decrease, and the failure strain increases after treatment with chemical corrosion. The corrosion of chemical solutions results in a weakening of the strength and an increase in the ductility of sandstone. The effect order of a chemical solution on the mechanical properties of sandstone is H2SO4>NaOH>distilled water, which is due to the larger loss of feldspar and calcite of the sandstone in acid solution and the significant loss of quartz of the sandstone in alkaline solution.(3)Nonlinear relations existed between the static mechanical properties and porosity of sandstone. The relationships among porosity, the static and dynamic compressive strength, and the static and dynamic elastic modulus were established based on the experimental data. The porosity is of exponential relation with the static and dynamic mechanical properties. The larger the porosity is, the more reductions of mechanical properties are there.
The results in this work would provide some points to help understand the effects of chemical corrosion on the pore structure and mechanical properties of the sandstones and have a guiding effect on the stability analysis on the subgrade engineering in Liuyang City, Hunan Province. It is also hoped that the results can serve as a reference for research in related fields. However, it should be noted that the mechanical properties of rocks are affected by many factors, such as mineral contents, temperature, and concentration of chemical solutions. Since we did not include these factors in this work, it should be included in future work.
Data Availability
Most of the data generated or analysed during this study are included in this manuscript and all of the data are available from the corresponding author on reasonable request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The research presented in this paper was jointly supported by the National Natural Science Foundation of China (Grant Nos. 51474252, 51774323, and 41502327), the Project of the State Key Laboratory of Safety and Health for Metal Mines of China (Grant No. 2016-JSKSSYS-02), and the Fundamental Research Funds Project for the Central South University (Grant No. 2016zzts095). The first author would like to thank the Chinese Scholarship Council for financial support to the joint PhD studies at the University of Adelaide.