Pore characteristics and mechanical behaviors of autoclaved aerated concrete produced from coal bottom ash and tailings

Abstract

The solid wastes such as coal bottom ash and tailings are currently widely used in various building materials, but further research on their application in autoclaved aerated concrete (AAC) is needed. To study the physical and mechanical properties of AAC produced from coal bottom ash and tailings, the pore characteristics of two different density grades of AAC were extracted and analyzed, and wave velocity and uniaxial compressive tests under water-saturated and dry conditions were carried out. The results show that the AAC produced from solid waste has excellent performance. The specimen is anisotropic, and the pore size distribution of its cross-section is mainly concentrated in the ranges of <0.1, 0.1–0.5, and 0.5–1.0 mm. The pore size perpendicular to the direction of rise is smaller and the pores number is greater. The wave velocity and uniaxial compressive strength of the test specimens in the water-saturated state are significantly lower than those in the dry state. The strengths of specimens loaded perpendicular to the direction of rise are higher than those loaded parallel to the direction of rise whether water-saturated or dry. The failure mode of the water-saturated specimens are different from those of dry specimens, which is manifested in the act that there are fewer surface cracks when the specimens fail in the water-saturated state, while there are more surface cracks in the dry specimens, and their morphology is irregular. The physicochemical effect between water and components and the mechanical effect of pore water are the reasons for the deterioration of AAC's mechanical properties. The results help to reveal the performance of AAC produced from solid waste and have good practical significance.

INTRODUCTION

Autoclaved aerated concrete (AAC) is a porous concrete product made of calcareous and siliceous materials as the main raw materials, with air-generating agents added, whose fabrication entails multiple processes. Its porosity can reach 70%–80%. It has the advantages of light weight, strong thermal and sound insulation performance, and good seismic performance. It is a new type of green building material. At present, it has been widely used in various industrial and civil construction for interior walls and to envelop the structure of buildings.^1–3

Concrete structures (e.g., exterior walls and roof panels of houses) are often eroded by water during use. To better understand the influence of water on the mechanical properties of concrete, many researchers have conducted a considerable amount of research.⁴ Huang et al.⁵ performed quasi-static compression tests on saturated concrete and normal concrete and found that free water reduced the quasi-static compressive strength of concrete and accelerated the damage evolution rate. Liu et al.⁶ experimentally studied the effect of moisture content on the elastic modulus of static compression of concrete, and their results showed that the elastic modulus decreased with the increase in moisture content. Wang et al.⁷ have demonstrated that the mechanical properties of concrete in a water-pressure environment are not only related to the water content and loading rate but also closely related to the scale of pore water.

As a porous material, AAC itself has a high porosity, resulting in a very high saturated moisture content, and the influence of water on its physical and mechanical properties cannot be ignored. Gottfredsen et al.⁸ showed that AAC changes in length as a result of changes in moisture. Jerman et al.⁹ study revealed that the thermal conductivity of AAC increases with increasing water content and that its freeze–thaw resistance increases with the increase in AAC compressive strength. Jin et al.¹⁰ established two-phase and three-phase fractal models to predict the thermal conductivity of dry and unsaturated and wet AAC, respectively, finding that the thermal conductivity of AAC increases with the increase in moisture content. Cong et al.¹¹ showed that moisture content has a major influence on AAC performance, especially on strength and frost resistance. Jerman et al.¹² found that drying shrinkage of different types of AAC increases with decreasing moisture content in the AAC pore system and this increase is fastest in the range of very low moisture (much lower than 6%).

In recent years, the utilization of solid waste resources has become a research hotspot, and the production of AAC by solid waste is undoubtedly a good solution for waste disposal. It not only solves the problem of solid waste utilization, but also reduces the adverse impact of sand mining on the environment by replacing sand with coal bottom ash or tailings, which has good economic and social benefits. However, the performance of AAC produced by solid waste materials is also a problem that needs to be studied, and its physical and mechanical properties need to be further understood. Previous studies have shown that water has a significant effect on the physical and mechanical properties of concrete materials. It is also very important to investigate the mechanical properties of AAC under saturated conditions. In this study, the pore characteristics of AAC produced from coal bottom ash and tailings are first extracted and analyzed, and the uniaxial compressive strength under saturated and dry conditions is measured to investigate the influence of water on the physical and mechanical properties of AAC. Through these studies, it is helpful to reveal the physical and mechanical properties of AAC produced from coal bottom ash and tailings, promote its application fields and improve the design theory of AAC structure.

MATERIALS AND METHODS Materials

The AAC specimens in this experiment were divided into two density levels (B05 and B06), which were cut from slabs. The raw materials of AAC are mainly solid wastes, including coal bottom ash and desulfurization gypsum from thermal power plant, tailing mud and tailing sand from quartzite deposit, which can realize the reuse of solid wastes. In addition, it also contains cement, lime, and aluminum paste. The specific ratio information of AAC raw materials is listed in Table 1, in which composition of the solid waste is: silicon tailings: silicon tailings mud: coal bottom ash: desulfurization gypsum = 4: 3: 3: 1.5, and the waste mortar is the leftover material cut after curing before pouring (recycling of leftover materials). The water to solid ratio is 0.61.

TABLE 1 Mixing proportion of raw materials of AAC.

[Table omitted. See PDF]

The size of the test specimens were 100 mm × 100 mm × 100 mm (Figure 1) cutted from the slab. In the production of AAC slabs, different directions of rise may lead to differences in their physical and mechanical properties through aluminum powder aeration. Therefore, the direction of rise was marked with arrows. When cutting the specimen, the surface of the specimen should be flat without cracks or obvious defects. The allowable deviation of the size should be ±1 mm, the flatness should not be greater than 0.5 mm, and the verticality should not be greater than 0.5 mm. The physical properties of the AAC are listed in Table 2, where the determination of the basic parameters according to the Chinese standard GB/T 11969-2020¹³ and comes from three randomly selected specimens.

TABLE 2 Material parameters.

Level	Dry density/$$$ \mathrm{kg}/{\mathrm{m}}^3 $$$	Natural moisture content/%	Saturated water content/%	Porosity/%
B05	524.7	2.1	71.3	79
B06	598.5	2.0	69.4	78

Experiment method AAC pore feature extraction

Images of the AAC were captured using a digital camera model Sony ILCE-7M2 (Figure 2) with a maximum image size of 25 million pixels. And the cross-sectional photographs are processed to analyze AAC porosity, pore number, pore size, and other parameters.^14–16 The primary AAC cross-section photographic processing entailed the following:

1. Sample preparation: The specimen was cut into cubes of 50 mm × 50 mm × 50 mm, and the inner section of the specimen was cleaned with a fine brush and ear-washing ball to remove the dust in the pores. It is worth noting that pores caused by the spalling of fine aggregates should be avoided when cutting and cleaning samples.
2. Taking cross-section photographs: A digital camera was used to record images of the cross sections of the specimen perpendicular and parallel to the direction of rise to obtain cross-sectional photographs. A fixed distance between the camera and the specimen is required, in this experiment the distance is about 40 cm, and a ruler was placed next to the specimen. After the photograph was obtained, the cross-sectional dimensions were cut to 30 mm × 30 mm. After converting the pixel size of the image and the actual size, an average of 1 pixel = 0.03839 mm can be obtained, and an aperture of 38 μm can theoretically be identified.
3. Cross-section reading: MATLAB version R2021a was used as the test platform, the photographs obtained by digital camera are imported into MTALAB, so as to use its own image processing toolbox to carry out subsequent processing of the photographs.
4. Image graying: The photographs obtained by digital cameras are in RGB format; that is, each pixel needs to store three primary colors in three bytes, which makes image processing difficult and requires a large amount of data. After the image is converted into a grayscale image (Figure 3), the image information of each pixel can be stored in one byte, that is, a grayscale value. The cross-sectional structure of aerated concrete is mainly composed of pore walls and holes, and the two are significantly different in the image. After grayscale processing, the hole walls and holes can be effectively separated.
5. Image enhancement: Improving image quality and recognizability requires performing image enhancement on the grayscale-processed photographs. In this study, the images underwent image median filtering, image background color subtraction, and image contrast adjustment.
6. Binarization processing: The enhanced grayscale image needs to undergo binarization processing. The principle of binarization is given by

[Image Omitted. See PDF] where $$$ f\left(i,j\right) $$$ is the pixel value of a point in the image matrix and $$$ T $$$ is the gray value threshold. After binarization (Figure 4), the hole in the section is black, and its pixel value is 0, and the hole wall is white, and its pixel value is 1.

1. Inverse color processing: For pore statistics and calculation, the binarized image needs to be inverted. After inverting the color, the hole in the section is white, and its pixel value is 1, and the hole wall is black, and its pixel value is 0.
2. Image data statistical processing: The processed data were imported into Excel for statistical processing to calculate the porosity, pore area, number of pores, and proportion of different pore sizes of AAC. It is worth noting that, because the pore shape is not a regular circle, for the convenience of analysis, after the pore area was obtained, its equivalent area to that of a circle was used to calculate the pore size.

The macroscopic pores of AAC with a pore size of >60 μm have a significant effect on its performance.¹⁷ After image processing, macroscopic pores of >38 μm could be identified. Therefore, method followed in this study can be used to better characterize the pore structure of AAC.

Preparation of dry and water-saturated specimens

The specimens were dried and saturated according to the provisions of the Chinese standard GB/T 11969-2020.¹³ The specimens were placed in an electric heating blast drying oven, kept at 60 ± 5°C for 24 h, then kept at 80 ± 5°C for 24 h, then baked at 105 ± 5°C to a constant mass, and finally cooled to room temperature to obtain dry specimens.

The dried specimens were put in a constant-temperature water tank with a water temperature of 20 ± 2°C. Water was added to 1/3 of the height of the specimen. After 24 h, more water was added to 2/3 of the height of the specimen. After another 24 h, more water was added to a level of 30 mm above the height of the specimen. the saturated specimen was then considered prepared after another 24 h.

P-wave velocity test

The ultrasonic wave velocity is related to the characteristics and state factors of the concrete. In this test, the wave velocity for the prepared test specimen was measured, and only the P-wave velocity was considered. A nonmetallic ultrasonic detector model ZBL-U5100 was used for the test (Figure 5). During the test, petroleum jelly as a couplant is applied to both sides of the specimen, and then the probe is pressed against both sides of the specimen and measured for 3 times then averaged the results, and both directions of rise were detected.

Uniaxial compression test

A uniaxial compression test was performed on the prepared test specimens. Loading was divided into directions perpendicular and parallel to that of the rise. In this test, the displacement control method was used for loading, and the loading rate was set to 0.12 mm/min. The loading diagram is shown in Figure 6.

RESULTS AND ANALYSES Pore characteristics

Figure 7 and Table 3 show the results of processing the cross-sectional images perpendicular and parallel to the direction of rise. It can be seen from Figure 7 that, in the two levels of test specimens, the test specimens perpendicular and parallel to the direction of rise were mainly concentrated in the ranges of <0.1, 0.1–0.5, and 0.5–1.0 mm. Table 2 presents the statistical results of the pore size distribution. The data in the table reveal that, in the porosity ranges of <0.1 and 0.1–0.5 mm, the perpendicular pore of B05 and B06 density levels of AAC perpendicular to the direction of rise accounted for 84.8% and 84.2%, respectively, while the proportions of pores parallel to the direction of rise were 79.1% and 79.3%, respectively. This shows that the pore size perpendicular to the direction of rise is smaller in the two density levels of AAC. In addition, in terms of the number of pores, in the two density levels of AAC, the number of pores perpendicular to the direction of rise is greater than the number of pores parallel to the direction of rise, so that AAC is anisotropic.

TABLE 3 Statistical table of pore size distribution.

[Table omitted. See PDF]

It has been shown in the literature^17,18 that the pore size and the number of pores have a significant effect on the physical and mechanical properties of porous materials. The difference in microscopic pore characteristics of the two sections parallel and perpendicular to the direction of rise also leads to differences in their uniaxial compressive strength. Therefore, in the following tests, the effects of perpendicular and parallel directions of rise were considered.

P-wave velocity

The AAC P-wave velocity measurement results are shown in Figure 8. When the measurement direction was the same and under the same temperature conditions, the results of the two density levels of specimens show that the P-wave velocity in the saturated state was significantly lower than that in the dry state, which is similar to the research results of Jasiński et al.¹⁹

In addition, under the same state, the P-wave velocity perpendicular to the direction of rise is higher than that parallel to the direction of rise, and this phenomenon is more obvious in the B06 specimen. In Section 3.1 above, the test specimen perpendicular to the direction of rise section has more pores and a smaller pore size distribution range than the section parallel to the direction of rise. The propagation of ultrasonic wave velocity in AAC mainly follows the internal pores, and it is precisely because of the different pore characteristics of the two directions of rise that the difference in P-wave velocity in different measurement directions under other conditions being equal.

Uniaxial compressive strength

Figure 9 shows the compressive strength of the two density levels of AAC specimens. When loaded perpendicular to the direction of rise, the compressive strength of B05 and B06 test specimens in dry state is 3.75 and 5.47 MPa, respectively, showing high compressive performance. In addition, the data in Figure 9 show that, when the loading direction was the same, the compressive strength of the two density levels of AAC test specimens in the saturated state was lower than that in the dry state. In this test, when the load was parallel to the direction of rise, the compressive strength of B05 and B06 test specimens from dry to saturated state decreased by 32.5% and 35.1%, respectively; and when the load was perpendicular to the direction of rise, the compressive strength decreased by 35.2% and 40.4%, respectively. It can be seen that the strength of the AAC test specimen in the saturated state was significantly lower than that in the dry state.

The results indicate that, when the specimen was loaded perpendicular to the direction of rise, its strength was higher than that when it was loaded parallel to the direction of rise. Different loading directions lead to profound changes in the compressive strength of test specimens of the same level. As shown in Figure 9, in the dry state, the compressive strength of B05 and B06 was reduced by 5.6% and 26.0%, respectively, when the test specimens were loaded perpendicularly to parallel; and in the saturated state, the compressive strength of B05 and B06 was reduced by 3.2% and 19.3%, respectively. It can be seen that in the same state, different loading directions also affect the strength.

In addition, in the two density levels of B05 and B06, the compressive strength of the B06 specimen is greatly affected by the water and loading direction. In summary, the AAC in dry state exhibits excellent compressive properties. The water and loading direction both affected the compressive strength of AAC, and water saturation had a greater influence on the strength.

Failure characteristics

The representative failure morphology of the AAC cube under uniaxial compression in the test is shown in Figure 10.

Comparison of the failure characteristics of AAC in saturated and dry states reveals that the main difference between the two is that, with the progress of loading, the AAC in the dry state undergoes considerable minor surface cracking and the debris falls off, whereas for the AAC in the saturated state, the cracks on the surface of the test specimen were not obvious at first, and most of them had only one crack, but the cracks continued to develop as the loading time progressed. In the final failure form, there were numerous cracks on the surface of the AAC test specimen in the dry state, and the expansion direction was irregular. There were many small cracks scattered around the main crack, and individual test specimens were seriously damaged. This indicates the brittleness of the AAC specimens in the dry state. After the saturated water treatment, the water has a certain softening effect on the AAC, reducing its brittleness, so that there were fewer cracks when the AAC was damaged, and the final instability failure mainly resulted from a main crack.

DISCUSSION

The influence of different direction of rise on the physical and mechanical properties of AAC is due to the different pore characteristics, and the influence of pore characteristics on strength mainly depends on the size of pores and the number of pores. When the porosity is the same, the smaller the pore size and the greater the number of pores, the higher the AAC strength.

The influences of water on the mechanical properties of concrete are mainly related to physico-chemical and mechanical interaction.^20–23 Physicochemical effects are caused by the interaction of water and solid components in concrete. The hydration products of AAC mainly contain quartz, tobermorite, anhydrite and C-S-H gel. Althougth, the strength of AAC is mainly related to tobermorite formation and its crystallinity,²⁴ the influence of other components cannot be ignored. Among these components, tobermorite and C-S-H gel will not produce large changes during the water saturation process, while anhydrite will have a softening effect in the process of water saturation, and anhydrite combines with water to produce gypsum dihydrate: [Image Omitted. See PDF] and dihydrate gypsum is softer than anhydrite, which will reduce the strength. In addition, during the saturation process, water molecules (H₂O) react with quartz (SiO₂), causing stress corrosion.²⁵ The interaction can be expressed as²⁶: [Image Omitted. See PDF]

During hydrolysis, the strong SiO bonds are hydrolyzed to weaker hydrogen bond hydroxyl groups linking the silicon atoms, and the concentration of hydroxyl ions increases, which accelerates the development rate of microcracks under stress.

Mechanical effects are mainly caused by pore water pressure. When the concrete is saturated with water, its internal pores and microcracks are filled with free water. Before the load is applied, the free water inside the concrete is in a stable state. When the load is applied, the free water in the saturated concrete cannot be discharged immediately, and pore water pressure is formed inside the concrete (Figure 11). As the loading continues, microcracks in the concrete gradually develop, and the pore water in the cracks pushes to the tip of the crack, acting as a wedge, which accelerates the development of the microcracks, resulting in faster damage of the test block and reducing the strength of the test specimen.

The weakening of AAC properties by water is also closely related to the properties of AAC itself, porosity, density, composite, etc. Generally, AAC with low density, large porosity, poor cementation are more susceptible to water weakening.

CONCLUSION

This paper aims to investigate the physical and mechanical properties of AAC produced from coal bottom ash and tailings. Experimental analysis were conducted on samples of two density levels, including pore analysis, wave velocity measurement and uniaxial compression test under dry and saturated conditions. The main conclusions of this paper are as follows:

1. Image analysis is used to present the AAC pore structure. The results show that the number of pores in the section perpendicular to the direction of rise is much greater than that in the parallel direction of rise, and the pore size is relatively smaller. In the two density level specimens, the B06 specimen has more pores in the same direction of rise.
2. The wave velocity of AAC in saturated state decreases obviously, and the wave velocity perpendicular to the direction of rise is greater than that parallel to the direction of rise.
3. The uniaxial compressive strength of AAC in the saturated state is lower than that in the dry state. No matter in saturated or dry state, the strength loaded perpendicular to the direction of rise was higher than that when it was loaded parallel to the direction of rise. And these phenomena are more pronounced in B06 specimens. The failure modes of the water-saturated and dry states are different. In the dry test specimen, there are numerous cracks on the surface of the test specimen, the expansion direction is irregular, and individual samples are seriously damaged. In the water-saturated specimen, there are few cracks on the surface, and the final instability failure is mainly caused by a main crack.
4. The physical and chemical interaction between water and solid components in AAC in the process of saturation, as well as the mechanical effect of pore water pressure in the during loading, are the main reasons for the deterioration of its mechanical properties.
5. Overall, the test results show that the AAC produced from coal bottom ash and tailings has good performance.

AUTHOR CONTRIBUTIONS

Mingyuan Zhang: Writing – original draft (equal). Huayan Yao: supervision (supporting); writing – review and editing (supporting). Yuting Liu: Visualization (equal). Jiarui Gan: Data curation (equal). Lingxiao Tang: Investigation (equal).

ACKNOWLEDGMENTS

This research is supported by Key Research and Development Projects in Anhui Province (Grant No. 202004a07020027 and 202104h04020024), and National Innovation and Entrepreneurship Training Program for college students (No. 202210359031). These supports are gratefully acknowledged.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

PEER REVIEW

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1002/eng2.12834.

DATA AVAILABILITY STATEMENT

Data available on request from the authors.