Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 Apr 3;9(15):17344–17353. doi: 10.1021/acsomega.3c10513

Evolution of a Filling Paste Cementitious System and Impact Patterns of Sulfate Mine Water Erosion on Filling Paste Performance

Yao Lu †,‡, Changxiang Wang †,*, Yongqiang Zhou †, Wenqiang Mou §, Ning Jiang ‡,*
PMCID: PMC11024973  PMID: 38645362

Abstract

graphic file with name ao3c10513_0015.jpg

To assess the impact of sulfate mine water on filling material performance, an accelerated sulfate erosion process was used to analyze the effects of various erosion concentrations, aging periods, and cation types on the macroscopic properties of the filling paste. These properties encompassed apparent phenomena, mass changes, and alterations in the uniaxial compressive strength. Observations revealed sulfate erosion, causing the formation of white substances and salt crystals on specimen surfaces. Initially, all solution-eroded specimens exhibited increased mass and strength. Over time, specimens in 5 and 10% MgSO4 solutions displayed the first signs of decline, while variations in other solutions were relatively small. Increasing the erosion concentration led to greater variations in mass and strength during the initial erosion phase. Specimens in 5 and 10% MgSO4 solutions initially peaked in mass and compressive strength, followed by a decline, while other filling paste specimens continued slow increases. Under equivalent conditions, the MgSO4 solution exhibited stronger erosion than the Na2SO4 solution. Composite erosion by Na2SO4 and MgSO4 involved initial strengthening and gel pore filling, intermediate expansion and crystallization, and late-stage substantial degradation, with MgSO4 exhibiting a more pronounced and complex impact. Gray relational analysis ranked factors affecting mass and uniaxial compressive strength variations as erosion concentration > erosion ion type > erosion aging period. Correlation degrees for factors influencing mass variations were 0.8822, 0.8714, and 0.4754, while for factors influencing uniaxial compressive strength variations, the correlation degrees were 0.8336, 0.7943, and 0.6125, respectively.

1. Introduction

In response to the environmental degradation caused by coal mining, green mining has gradually become a consensus in the mining industry. It is widely believed that paste backfilling is an ideal solution for addressing environmental issues associated with coal mining.1−6 The stability of the goaf is closely related to the structural properties of the paste backfill after its implementation. Factors such as temperature, pressure, and chemical composition affect the paste backfill, leading to physicochemical changes and subsequently influencing the filling effect.7−13

In China, the distribution of sulfate-rich soils is extensive, particularly in coal mining areas, where the presence of sulfate-rich water is prominent.14 These waters typically contain high concentrations of sulfates, resulting in low pH values and an acidic environment.15,16 This characteristic significantly intensifies their erosive effect on filling materials, potentially leading to a noticeable decline in the physical properties of the filling paste or even causing severe damage. Additionally, sulfate-rich mine water may dissolve ore components, forming corrosive solutions that can cause considerable damage to mine facilities and equipment. Therefore, studying the impact of sulfate-rich mine water on the stability of the filling paste is crucial.

Chen et al.17 investigated the resistance of the drying (upper) portion and immersed (lower) portion of cement mortars partially immersed in 10% Na2SO4 solution. Their findings indicated that the deterioration of cement mortars partially immersed in sulfate solution is not only solely attributed to chemical sulfate attack but also to physical sulfate attack. Li and Fall18 conducted an experimental study on the sulfate effect on the early age strength and self-desiccation of cemented paste backfill (CPB). They found that sulfate has a significant effect on the early age strength and self-desiccation of CPB. In the early ages, sulfate can have negative effects, leading to a decrease in the CPB strength and a reduction in the intensity and rate of self-desiccation within the CPB. The magnitude of these effects is dependent on the initial sulfate concentration. In a sulfate environment, Liu et al.15 analyzed the macroscopic and microstructural changes of filling paste through uniaxial compression tests, X-ray diffraction (XRD), and scanning electron microscopy (SEM), and discussed the mechanism of deterioration and cracking.

Studies by Pokharel and Fall19 examined the permeability of paste backfill under the combined action of temperature and sulfates. Under constant temperature conditions, the permeability of paste backfill initially decreases and then increases with increasing initial sulfate concentration. This is attributed to the formation of gypsum and ettringite, which fill internal pores of the paste backfill, reducing its permeability. Once the pores are filled, expansion minerals formed due to the increased sulfate content lead to the generation of cracks within the paste backfill, thereby enhancing its permeability. Research by Ercikdi et al.20 emphasized the impact of desliming on the short-term and long-term strength, stability, and rheological properties of CPB produced from different mill tailings. Desliming proves beneficial for CPB made from sulfide-rich mill tailings as it enhances strength and stability, particularly in the long term, while also reducing binder consumption. Stroh et al.21 conducted simulations of combined sulfate–chloride attack under laboratory conditions using solutions containing NaCl and Na2SO4 in varying concentrations. The exact mechanisms of the phase transitions induced by this combined attack are subjected to debate. The distribution of phases in the samples was determined using synchrotron XRD, and a mechanism of phase developments under combined sulfate–chloride attack was derived.

Furthermore, Fall and Benzaazoua22 employed artificial neural networks to simulate the sulfate corrosion resistance of paste backfill. Their research indicated that at lower sulfate concentrations, it is beneficial for increasing the uniaxial compressive strength of paste backfill. However, when the concentration exceeds 15,000 ppm, the compressive strength of paste backfill begins to decrease. Orejarena and Fall23 conducted an assessment of sulfate erosion on paste backfill. They employed variables including binder proportion, water-to-cement ratio, and sulfate content while maintaining tailings and temperature as constants. By establishing an artificial neural network model, they were able to predict variations in the strength of the backfill. Giordano et al.24 applied the theory of fuzzy concepts to assess the safety level of concrete structures, considering the stochastic behavior of the involved key parameters. They conducted probabilistic and fuzzy probabilistic analyses on prestressed bridges to evaluate their behavior under maintainable conditions. Kawamura et al.,25 in their existing performance evaluation system for concrete bridge decks, introduced a novel approach. They assessed the deterioration of concrete decks through expert knowledge and established a performance evaluation index system for bridges by using artificial neural networks.

When it comes to studies on the erosion caused by sulfates on materials, the focus has generally been on qualitative investigations of sulfate erosion mechanisms and influencing factors in concrete or cement. There has been relatively limited quantitative research specifically targeting sulfate erosion of coal mine backfill materials. Given the widespread distribution of sulfate-rich soils in China and the frequent occurrence of sulfate-rich water in coal mines, it is imperative to accurately assess the corrosive impact of sulfate-containing mine water on filling materials. The stability of backfill materials under sulfate-containing mine water erosion is influenced by various factors such as sulfate erosion in mine water and erosion duration. The degree of relationship between these factors and the stability of the backfill material is uncertain. In response to these concerns and practical engineering needs, this study conducts high-concentration-accelerated erosion experiments, simulating the erosion of backfill materials by two primary types of sulfate-containing mine water. Through theoretical analysis and gray relational degree assessment, an evaluation of the durability of coal mine backfill materials under sulfate erosion is carried out. This research not only supports the wider application of filling methods in sulfate-rich environments but also provides scientific guidance and technical support for the assessment of utilizable materials containing sulfur in backfilling, ensuring the safe and efficient operation of mines and the long-term stability of goaf areas.

2. Experimental Methods

2.1. Experimental Materials

2.1.1. Coal Gangue

The majority of the coal gangue had a particle size smaller than 100–250 mm. In accordance with the requirements for preparing filling materials, the maximum particle size of coal gangue should generally be less than 25 mm, with around 40% of it being smaller than 5 mm. Most of the coal gangue met the filling requirements, with only a small portion required to undergo crushing processing. Additionally, the proportion of particles smaller than 5 mm was relatively low. Therefore, during the experiments, the coal gangue was crushed to ensure that its maximum particle size was controlled within 25 mm. It was also subjected to sieving, with the sieving results shown in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Composition of coal gangue particles.

In the paste filling material, the primary purpose of the coal gangue was to serve as an aggregate. The inorganic components of the coal gangue mainly consisted of minerals and water, typically dominated by silicon dioxide and aluminum oxide. Additionally, there were varying amounts of Fe2O3, CaO, MgO, SO3, Na2O, and so forth. The main chemical composition of the coal gangue used in this experiment is detailed in Table 1.

Table 1. Chemical Composition of Coal Gangue %.
component SiO2 Al2O3 Fe2O3 CaO MgO K2O Na2O
proportion 59.10 18.90 4.30 2.36 1.41 1.89 0.43
Open in a new tab

2.1.2. Cement

The cement adopts no. 32.5 ordinary Portland cement produced by Shandong Shanshui Cement Group Co., Ltd. The main mineral components are shown in Table 2.

Table 2. Average Mineral Content of Portland Cement Clinker %.
chemical name chemical formula shorthand notation percentage
tricalcium silicate 3CaO·SiO2 C3S 54
dicalcium silicate 2CaO·SiO2 C2S 20
tricalcium aluminate 3CaO·Al2O3 C3A 7
iron aluminum acid tetracalcium 4CaO·Al2O3·Fe2O3 C4AF 14
gypsum CaSO4·2H2O CSH 3.5
Open in a new tab

2.1.3. Fly Ash

In this paper, the fly ash of the Qingdao Huangdao Power Plant was used. The main chemical composition and physical properties are shown in Table 3. The coal combustion was relatively complete with low carbon content, density 2.24 t/m3, and bulk weight 0.65 t/m3.

Table 3. Chemical Composition and Physical Properties of Fly Ash.
source of fly ash SiO2/% Al2O3/% CaO/% Fe2O3/% MgO/% loss/%
Huangdao Power Plant 53.94 30.91 6.53 2.38 0.92 6.34
apparent density/kg/m3 stacking density/kg/m3 45 μm square hole sieve residue/% standard consistency water requirement ratio/%
1950 780 19.0 91
Open in a new tab

By using SEM, the microscopic morphology of Class II fly ash was observed, as shown in Figure 2. The particles of Class II fly ash were very small and mostly exhibited a spherical glassy shape, demonstrating a good rolling ball effect. The particle size distribution of Class II fly ash was determined with a laser particle-size analyzer. The two curves in the figure represent the cumulative distribution curve and the differential distribution curve. The particle sizes were mainly concentrated around 50 μm, and particles larger than 40 μm accounted for 40% of the overall particle size distribution.

Figure 2.

Figure 2

Open in a new tab

Size distribution and SEM of fly ash.

2.2. Experimental Scheme

2.2.1. Mixing and Casting

Portland cement, fly ash, and appropriate amounts of sand and aggregate were mixed in a mass ratio of 1:4. The water-to-binder ratio was 1:2. The slurry concentration was 76%, and the fine aggregate proportion was 45%. The specific preparation process is shown in Figure 3. Cubic specimens with side lengths of 100 mm were prepared according to the paste formulation requirements. The coal gangue, cement, fly ash, and water were accurately weighed according to the proportions of the filling material. The mixed filling material was placed in a mixer for thorough mixing. The well-mixed filling material was then placed into molds, the surface was smoothed with a scraper, and vibration was applied to eliminate air bubbles within the specimen. After completion of the vibration process, the paste specimens were left to stand for 24 h before demolding. The demolded specimens were then subjected to specialized curing for 28 days.

Figure 3.

Figure 3

Open in a new tab

Experimental scheme.

2.2.2. Erosion Solution Preparation and the Immersion Test

For the erosion test, two types of solutions, Na2SO4 and MgSO4, were selected, with water used as the control group. The mass fraction concentrations of Na2SO4 and MgSO4 solutions were 3, 5, and 10%. The specimens for immersion were the 28 day cured coal gangue paste filling specimens, and poly(vinyl chloride) (PVC) barrels were used as the immersion containers. The 28 day cured specimens were numbered and immersed in the erosion solution. Before the immersion test was conducted, one set of specimens was dried, and their mass and compressive strength were tested. Sodium sulfate and magnesium sulfate solutions were prepared at three concentrations, 3, 5, and 10%, each concentration immersing 9 sets of specimens, with 3 specimens in each set, as shown in Figure 3. The immersion periods were 6, 30, 60, 90, 120, 150, and 180 days. In addition, a comparative test was set up with specimens immersed in water. The changes in mass and compressive strength of the filling specimens in solutions of different concentrations and durations were tested. Furthermore, the external appearance of the specimens in each age and erosion solution was observed and recorded.

2.2.3. Experimental Process

2.2.3.1. Solution Configuration

When preparing salt solutions, considering their hygroscopicity, anhydrous sodium sulfate (see Figure 3) produced by Tianjin Dingshengxin Chemical Co., Ltd. was selected as the reagent. It was weighed using an electronic scale with an accuracy of 0.01 g (see Figure 3) and made into a salt solution. Since anhydrous sodium sulfate easily combines with water to form a hydrate, resulting in white precipitation, it can be prepared into a sodium sulfate solution using warm water. Glass rods were used for stirring to prevent precipitation.

2.2.3.2. Specimen Immersion

The prepared solution was poured into the prepared PVC barrels and each barrel was labeled. Then, the 28 day cured specimens were placed into each barrel. The specimens selected should be relatively intact, without obvious holes, cracks, significant unevenness, or sharp corners on the surface.

2.2.3.3. Immersion Time

In this experiment, immersion periods of 6, 30, 60, 90, 120, 150, and 180 days were chosen. Whenever the designated immersion period was reached, one set of specimens was taken out from each barrel. Each set consisted of 3 specimens, which were used for subsequent testing.

3. Experimental Result

3.1. Apparent Phenomenon

With varying soaking durations and sulfate solution concentrations, the stress–strain behaviors of the tested filling paste blocks exhibit noticeable disparities, leading to irregularities in the elastic modulus. Subsequently, fractures emerge in alternating, staggered patterns, accompanied by splintering around the core of the test block (see Figure 3). This distinctive behavior ensures that the core of the block remains a key load-bearing component, resulting in a less precipitous decline in stress levels compared to the response of rock under analogous conditions of damage.

Taking the erosion by 5% sodium sulfate as an example, before the erosion period, the coal gangue paste specimens were taken out, dried, and appeared white, as shown in Figure 4. At the 90 day erosion period, a large amount of white substance appeared at the edges and corners of the coal gangue paste specimen. After XRD analysis, the white substance generated on the surface is mainly the Na2SO4 crystal. This is because there are sodium ions in the solution, which easily cause crystal erosion, so that the appearance of the filling paste has obvious salt crystallization.

Figure 4.

Figure 4

Open in a new tab

Filled paste sample after soaking in sulfuric acid solution.

3.2. Mass Variation

The mass variation under different immersion periods and various immersion solutions is presented in Table 4.

Table 4. Mass Variation of Each Immersion Solution during Different Immersion Periods.

solution period/d water/g 3%Na2SO4/g 5%Na2SO4/g 10%Na2SO4/g 3%MgSO4/g 5%MgSO4/g 10%MgSO4/g
6 0.35 8.12 12.06 16.23 14.12 18.09 20.04
30 0.12 3.04 4.16 6.13 7.24 9.36 10.23
60 0.02 2.01 3.05 4.17 5.05 6.12 7.15
90 0 0.05 0.02 0.03 0.01 –11.02 –18.06
120 0.14 –0.59 –1.15 –2.10 –3.50 –13.88 –20.11
150 0.5 –1.12 –4.23 –6.32 –9.05 –17.20 –23.55
180 0 –4.51 –9.50 –13.40 –15.77 –21.33 –32.10
Open in a new tab

3.2.1. Influence of Different Erosion Ages on the Mass of the Filling Paste

The variation in the mass of the filling paste eroded by sulfate solutions of different concentrations at different ages is illustrated in Figure 5.

Figure 5.

Figure 5

Open in a new tab

Changes in the mass of different concentrations of sulfate solution at different ages.

Figure 5 reveals that before 60 days, the mass of test specimens subjected to erosion by all solutions increased with the age, albeit at a progressively slower rate with longer aging periods. After 60 days, the specimens eroded by 5% MgSO4 and 10% MgSO4 solutions showed a declining trend in the mass, while the mass of specimens eroded by other solutions gradually increased and approached a plateau. The primary reason for this phenomenon lies in the initial stage of erosion, where the porosity of the filling paste is relatively large. Erosion ions from various solutions enter the interior of the filling paste through surface pores, react with the chemical substances inside the filling paste, and generate insoluble substances that adhere to the pores. This leads to a gradual reduction in the porosity of the filling paste, resulting in an increased compactness and a slow increase in mass.

When the concentration reaches a certain level and the erosion reaction occurs for a certain period, the pores inside the filling paste will gradually be filled with insoluble erosion materials. At this point, as the reaction continues, the accumulation of insoluble substances reached a certain level. This may lead to the emergence of tiny cracks within the interior of the filling paste, increasing the porosity. Erosion ions from the solution can more easily enter the interior of the filling paste, for further erosion. Consequently, the compactness of the filling paste decreases rapidly, leading to a significant reduction in the mass. After 90 days, for all solution-eroded test specimens, the mass shows a negative growth trend with the increase of the aging period. Moreover, the higher the solution concentration, the faster the negative growth.

3.2.2. Influence of Different Erosion Concentrations on the Mass of the Filling Paste

Figures 6 and 7, respectively, show the variation in the mass of specimens immersed in Na2SO4 and MgSO4 solutions of different concentrations.

Figure 6.

Figure 6

Open in a new tab

Changes in the mass of different concentrations of the Na2SO4 solution.

Figure 7.

Figure 7

Open in a new tab

Changes in the mass of different concentrations of the MgSO4 solution.

From Figures 6 and 7, it can be observed that the trend of mass variation for specimens eroded by Na2SO4 and MgSO4 solutions is consistent. With an increase in concentration, the growth in the mass becomes larger. However, when the concentration exceeds a certain threshold and the erosion aging period reaches a certain level, the mass experiences negative growth. This indicates that at this point, sulfate erosion has caused deterioration in the filling paste. The difference lies in the fact that after 60 days, the mass of specimens eroded by 5% MgSO4 and 10% MgSO4 solutions begins to decline, while the mass of specimens eroded by other solutions gradually increases and approaches a plateau. After 90 days, for all solutions (including Na2SO4 solution), the mass of eroded specimens shows a negative growth trend with the increase of the aging period. Furthermore, the higher the concentration of the solution, the faster the negative growth.

The main reason for this phenomenon lies in the initial stage of erosion, where the porosity of the filling paste is relatively large. Erosion ions from various solutions enter the interior of the filling paste through surface pores, react with the chemical substances inside the filling paste, and generate insoluble substances that adhere to the pores. This leads to a gradual reduction in the porosity of the filling paste, resulting in increased compactness and a slow increase in the mass.

When the concentration reaches a certain level and the erosion reaction occurs for a certain period, the pores inside the filling paste will gradually be filled with insoluble erosion materials. At this point, as the reaction continues, the accumulation of insoluble substances reaches a certain level. This may lead to the emergence of tiny cracks within the interior of the filling paste, increasing the porosity. Erosion ions from the solution can more easily enter the interior of the filling paste for further erosion. Consequently, the compactness of the filling paste decreases rapidly, leading to a significant reduction in the mass.

3.2.3. Influence of Different Erosion Ions on the Mass of the Filling Paste

From Figure 8, it can be observed that under equivalent aging periods and concentrations, the absolute change in the mass of the filling paste immersed in MgSO4 solution is greater than that of the filling paste immersed in Na2SO4 solution. Under equivalent aging conditions, except for the case of a 6 day immersion, the absolute change in the mass of the filling paste immersed in 3% MgSO4 solution is greater than that of the filling paste immersed in 10% Na2SO4 solution. This indicates that under similar conditions, magnesium sulfate solution has a stronger erosive effect compared to sodium sulfate solution.

Figure 8.

Figure 8

Open in a new tab

Mass change of specimens immersed in different cation solutions.

3.3. Uniaxial Compressive Strength

The changes in the uniaxial compressive strength of different immersion solutions at different immersion periods are shown in Table 5.

Table 5. Change of Uniaxial Compressive Strength of Each Immersion Solution during Different Immersion Periods.

solution period/d water/MPa 3%Na2SO4/MPa 5%Na2SO4/MPa 10%Na2SO4/MPa 3%MgSO4/MPa 5%MgSO4/MPa 10%MgSO4/MPa
6 4.1 5.4 6.02 6.45 7 7.4 7.9
30 3.7 4 3.21 4.62 3.39 4.24 4.32
60 2.12 1.19 2.41 1.63 2.59 3.26 3.93
90 0.68 1.43 1.76 1.05 0.39 –4.8 –7.15
120 –0.1 1.28 –1.4 –2.44 –0.56 –0.79 –0.32
150 0.13 –1.87 –0.73 –1.02 –3 –0.5 –0.79
180 –0.05 –1.11 –1.46 –1.28 –0.98 –0.5 –0.55
Open in a new tab

3.3.1. Influence of Different Erosion Ages on the Strength of the Filling Paste

Figure 9 shows the variation of the uniaxial compressive strength with erosion ages in different sulfate solutions.

Figure 9.

Figure 9

Open in a new tab

Uniaxial compressive strength changes with age in each sulfate solution.

As observed in Figure 9, from 6 to 60 days, there was a certain degree of increase in compressive strength for the specimens. The increase in compressive strength was inversely proportional to the age; the larger the age, the slower the increase in compressive strength. Simultaneously, the strength of the specimens soaked in sulfate solutions was significantly higher than those soaked in water. This was because the sulfate solutions underwent chemical reactions with the specimens, generating insoluble substances that filled the internal pores of the specimens, thereby enhancing their compactness and resulting in an increase in compressive strength.

At 90 days, the change in the compressive strength of the specimens soaked in 5 and 10% MgSO4 solutions became negative, while the compressive strength of the remaining specimens increased slowly. Among them, the specimens soaked in 10% MgSO4 solution exhibited a greater decrease in compressive strength compared to those soaked in 5% MgSO4 solution. The reduction in strength was attributed to the expansion of previously formed condensates after filling the pores, leading to internal stress within the specimens and the formation of cracks. These cracks made it easier for sulfate ions to penetrate the interior of the specimens, creating a vicious cycle. For the other solution-soaked specimens, the increase in the strength gradually leveled off. This was because, after a series of reactions, the condensates generated by the reactions in the solution did not completely fill the pores of the specimens, causing a slow and gradual increase in strength.

At 120 days, the compressive strength of the specimens soaked in water was almost unchanged compared with 90 days. The specimens soaked in 3% Na2SO4 solution showed a positive increase in compressive strength, while the strength of the remaining specimens exhibited varying degrees of decline. Among them, specimens soaked in 5% Na2SO4, 10% Na2SO4, and 3% MgSO4 solutions showed negative changes in compressive strength for the first time. The compressive strength decrease was most significant after soaking in a 10% Na2SO4 solution, with a decrease of 2.44 MPa. The specimens soaked in 5% MgSO4 and 10% MgSO4 solutions exhibited successive decreases of 0.79 and 0.32 MPa, respectively, compared to soaking for 90 days.

From 150 to 180 days, the change in compressive strength for specimens soaked in water was approximately zero, while for the other specimens, the change in strength was negative. Among them, the 3% Na2SO4 solution exhibited the largest decrease, with a reduction of 3.98 MPa.

3.3.2. Influence of Different Erosion Concentrations on the Strength of the Filling Paste

Figures 10 and 11 depict the trend of compressive strength variation of specimens immersed in solutions of different concentrations of Na2SO4 and MgSO4, respectively.

Figure 10.

Figure 10

Open in a new tab

Compressive strength change of specimens immersed in different concentrations of Na2SO4 solution.

Figure 11.

Figure 11

Open in a new tab

Compressive strength change of specimens soaked in different concentrations of MgSO4 solution.

As observed in Figure 10, the change in the compressive strength of specimens immersed in different concentrations of Na2SO4 solution was positive within the 90 day corrosion period. This indicates that all specimens experienced an increase in compressive strength to varying degrees within 90 days. Specifically, at 6 days of corrosion, the change in compressive strength was directly proportional to the concentration, i.e., 10% Na2SO4 > 5% Na2SO4 > 3% Na2SO4 > water. With the increase in the age, the change in compressive strength for all specimens gradually decreased, suggesting that during this stage, the compressive strength of the specimens increased slowly with the age. When the age reached 120 days, the change in compressive strength for specimens soaked in water was near zero. The change in compressive strength for specimens immersed in 3% Na2SO4 solution remained positive, while for the remaining specimens, the change in compressive strength was negative. This indicates that the compressive strength of the specimens corroded by water had a tendency to stabilize at this point. The specimens corroded by 3% Na2SO4 solution continued to increase in compressive strength, while those corroded by 5% Na2SO4 and 10% Na2SO4 solutions experienced a decrease in compressive strength. From 150 to 180 days, except for the specimens corroded by water, the change in compressive strength for specimens immersed in Na2SO4 solutions was negative, indicating a decrease in compressive strength.

Figure 11 reveals that the change in the compressive strength of specimens immersed in different concentrations of MgSO4 solution is positive within the 60 day period of erosion. This indicates that within this time frame, all specimens experience varying degrees of strength improvement. Specifically, at 6 days of erosion, the change in strength is directly proportional to the concentration, i.e., 10% MgSO4 > 5% MgSO4 > 3% MgSO4 > water. As the duration of erosion progresses, the change in compressive strength for all specimens gradually decreases, suggesting a slow increase in strength during this phase.

When the duration extends to 90 days, the change in compressive strength for specimens immersed in water is close to zero, while the change for specimens in a 3% MgSO4 solution remains positive. On the other hand, the change in compressive strength for all other specimens turns negative. This indicates that at this point, the compressive strength of the specimens undergoing water erosion stabilizes. The compressive strength of the specimens in 3% MgSO4 solution continues to increase. However, for specimens in 5 and 10% MgSO4 solutions, the compressive strength starts to decline with a larger change in strength observed for higher concentrations.

With the increase in concentration of the aforementioned solutions, there is a higher presence of ions, leading to the generation of more insoluble substances, such as gypsum, calcium aluminate, salt crystals, etc., through reactions within the specimen. Consequently, the internal voids within the specimen become smaller, resulting in a higher compressive strength. However, as the concentration of the corroding solution increases, the formation of crystals also intensifies. Eventually, after the pores have been filled and undergoing expansion, this leads to the formation of cracks, causing a gradual reduction in specimen pressure strength.

3.3.3. Influence of Different Erosion Ions on the Strength of the Filling Paste

From Figure 12, it is evident that at aging stages of 6, 30, and 60 days, the change in the compressive strength of specimens immersed in MgSO4 solution is greater than that of specimens immersed in Na2SO4 solution at the same concentration, which in turn is greater than that of specimens immersed in water. For instance, after 60 days of immersion, the change in the compressive strength of specimens immersed in 5% MgSO4 solution surpasses that of the 5% Na2SO4 solution. Moreover, the compressive strength variations during these three aging stages generally exhibit an ascending trend, with smaller variations observed as the aging period prolongs. At the 90 day mark, the compressive strength of specimens soaked in MgSO4 solution begins to decline, while those immersed in Na2SO4 solution slowly rise to a stable state. Beyond 120 days of aging, all specimens immersed in the solutions experience negative variations in compressive strength, indicating a varying degree of decrease in compressive strength.

Figure 12.

Figure 12

Open in a new tab

Compressive strength of specimens immersed in different cation solutions.

During the initial stage of immersion corrosion, both solutions exhibited a positive promotion effect on the compressive strength of the coal mine backfill. However, the hardening effect of MgSO4 was significantly stronger than that of Na2SO4. With the extension of immersion time, the growth rate of the compressive strength of the coal mine backfill gradually slowed down, and even a decline in compressive strength occurred. This indicates a trend of the gradual weakening of the hardening effect and strengthening of the deterioration effect. Under specific concentration and age conditions, the effect of MgSO4 solution on the specimen strength was significantly greater than that of Na2SO4.

4. Experimental Analysis

4.1. Mechanism Analysis

Based on existing relevant research26−30 and experimental results presented in this study, it is evident that the composite erosion of filling paste specimens through Na2SO4 and MgSO4 undergoes three distinct stages of transformation. These stages are intricately linked to the erosion concentration, eroding ions, and the erosion duration.

Initial stage (approximately ≤ 3 months): during the initial erosion phase, the surface of the filling paste remained essentially unchanged and intact. As depicted in Figure 13a,b, with the infiltration of eroding ions, the generated erosion products fill the gel pores of the filling paste, enhancing its compactness. Due to the reinforcement of the filling paste structure, there is a growing trend in the paste strength and mass during this stage. The primary reactions occurring at this stage are as follows

4.1. 1
4.1. 2

Figure 13.

Figure 13

Open in a new tab

Evolution of the sulfate-eroded filling paste during different stages.

Intermediate stage (3 to 5 months): as illustrated in Figure 13c, over time, erosion products such as CaSO4·2H2O and AFt crystallize within the filling paste. CaSO4·2H2O and AFt cause volumetric expansions exceeding 2.0 and 2.5 times, respectively, resulting in volume expansion of the filling paste. This expansion leads to the extension of pre-existing cracks and even delamination at the corners. Under the influence of water, soluble NaOH and sparingly soluble Mg(OH)2 increase the porosity of the filling paste, gradually diminishing its mechanical properties.

Late stage (after 5 months): as shown in Figure 13d, MgSO4 infiltrates the filling paste, reacting to generate MgO·SiO2·H2O. This product lacks cohesive strength, causing the internal structure of the filling paste to become loose. The mechanical properties of the filling paste experience a significant decline with severe surface mortar detachment, coarse and fine aggregate separation, and overall structural integrity being severely compromised. This suggests that MgSO4 erosion, in comparison to Na2SO4, exhibits a composite effect, exerting a more pronounced influence on the structure and performance of the filling paste. The primary reaction at this stage is

4.1. 3

Specimens aged six months were selected for erosion immersion and the products were analyzed from the visibly corroded areas through energy spectrum analysis. The results of the analysis are shown in Figure 14. According to the feedback from the energy spectrum, it is evident that in the corrosion products of magnesium sulfate, the levels of magnesium and oxygen elements are significantly higher than those in sodium sulfate solution corrosion. This indirectly suggests that compared to sodium sulfate corrosion, additional reactions leading to the formation of MgO·SiO2·H2O and Mg(OH)2 occur in magnesium sulfate corrosion. These additional reaction products have expansive characteristics, exacerbating the degree of erosion.

Figure 14.

Figure 14

Open in a new tab

Energy spectrum analysis for corrosion products in different solutions soaked for 60 d.

4.2. Gray Relational Analysis

The gray relational analysis employs mathematical methods to assess the degree of proximity between a reference sequence and several comparative sequences. The closer the geometric shapes of the curves, the greater the gray relational degree between them. Conversely, the farther apart they are, the smaller the degree. In this study, we utilized the modified correlation analysis method proposed by Su.31

In the analysis, compressive strength and mass variation are taken as the reference sequences, denoted as X0 = {X0(k), k = 1, 2,···,m}. The erosion aging period, erosion concentration, and erosion ion category are considered as comparative sequences. Based on the test results and the data in Figures 5–12 above, there are a total of 64 data sets, with Xi = {Xi(k), k = 1, 2, ... , m; i = 1, 2, ... , n}, n = 3, m = 64. The correlation results between the erosion aging period, solution concentration, cation erosion, and mass variation as well as compressive strength variation are presented in Table 6.

Table 6. Correlation between Influencing Factors vs Mass and Strength Variation.

reference sequence erosion ages erosion concentrations erosion ions
mass variation 0.4754 0.8822 0.8714
strength variation 0.6125 0.8336 0.7943
Open in a new tab

From Table 6, it can be observed that the correlation between each influencing factor and the mass variation of the filling paste follows the order from highest to lowest: erosion concentration > cation erosion > erosion aging period. The concentration of sulfate solution exhibits the highest correlation with the mass variation of the filling paste, reaching 0.8822. The presence of cation erosion also significantly affects the mass variation of the filling paste, with a correlation coefficient of 0.8714. On the other hand, the correlation between the erosion aging period and the mass variation of the filling paste is relatively lower, at 0.4754.

Furthermore, the order of the gray correlation degree between each influencing factor and the uniaxial compressive strength variation of the filling paste aligns with the trend observed for mass variation. Specifically, the correlation between solution concentration, cation erosion, and uniaxial compressive strength is relatively higher, with correlation coefficients of 0.8336 and 0.7943, respectively. The correlation between the erosion aging period and uniaxial compressive strength is relatively lower, with a correlation coefficient of 0.6125.

While the correlation of the erosion aging period is relatively smaller compared to the other two factors in relation to the mass and strength variation of the filling paste, it is important to note that in the case of long-term immersion in a sulfate mine water environment, sulfate erosion is a prolonged process that can still lead to both mass loss and strength weakening of the filling paste.

5. Conclusions

  • (1)

    In the initial phase of immersion erosion, the mass of specimens subjected to all solutions exhibited an initial increase with aging, albeit at a gradually decelerating rate. Beyond 60 days, specimens immersed in 5 and 10% MgSO4 solutions experienced a decline in the mass, indicative of paste filling material degradation due to sulfate erosion. Notably, MgSO4 solution demonstrated a more pronounced erosion impact on the paste filling material compared to Na2SO4 solution under identical aging and concentration conditions.

  • (2)

    Throughout the 6 to 60 day immersion erosion period, compressive strength displayed an ascending trend, albeit with a diminishing rate of increase. Specimens immersed in sulfate solutions manifested higher compressive strength than those in water. Post 90 days, compressive strength decreased in 5 and 10% MgSO4 solutions, contrasting with a gradual rise in other specimens. Elevated concentrations resulted in reduced strength, particularly between 120 to 180 days. MgSO4 solution exerted a more substantial influence on compressive strength than Na2SO4 under equivalent conditions.

  • (3)

    Gray correlation analysis delineated the impact of factors on mass and strength variations in paste filling material, ranking them as follows: erosion concentration > cation erosion > erosion aging period. Erosion concentration wielded the most significant influence (correlation coefficient: 0.8822), succeeded by cation erosion (0.8714), and followed by the erosion aging period (0.4754). Factors influencing uniaxial compressive strength were ranked with correlation coefficients of 0.8336, 0.7943, and 0.6125. In a sulfate mine water environment, sulfate erosion unfolds as a protracted process causing both mass loss and a decline in paste filling material strength.

Data Availability Statement

All data in this paper are available.

Author Contributions

All authors contributed to publishing this paper. C.W. and N.J. conceived the main idea of the paper; Y.L. contributed to theoretical analysis; Y.Z. analyzed the data; C.W. and Y.L. wrote the paper; and Y.L. and W.M. modified figures and proofread the revised version.

The study was supported by the National Key Research and Development Program (2023YFC3009100 and 2023YFC3009102), National Natural Science Foundation of China (no. 52304198), , and Anhui Provincial Key Research and Development Project (2022m07020006).

The authors declare no competing financial interest.

References

  1. Wang Y.; Na Q.; Yang J.; Zhang L.; Zhang J.; Li J.; Jin F. Monitoring of Barricade Pressure During The Entire Backfilling Process For a High Iron Mine Stope. Case Stud. Constr. Mater. 2023, 19, e02456. [Google Scholar]
  2. Bulusu S.; Aydilek A. H.; Petzrick P.; Guynn R. Remediation of Abandoned Mines Using Coal Combustion By-Products. J. Geotech. Geoenviron. Eng. 2005, 131 (8), 958–969. 10.1061/(asce)1090-0241(2005)131:8(958). [DOI] [Google Scholar]
  3. Karfakis M. G.; Bowman C. H.; Topuz E. Characterization of Coal-Mine Refuse As Backfilling Material. Geotech. Geol. Eng. 1996, 14 (2), 129–150. 10.1007/bf00430273. [DOI] [Google Scholar]
  4. Li D.; Wang C.; Xiao J.; Lu W.; Zhang B.; Li Z.; Tong X. Raw Gangue Filling Mining Under Construction—A Case Study In China. Engineering 2023, 15, 176–195. 10.4236/eng.2023.153014. [DOI] [Google Scholar]
  5. Changxiang W.; Hao H.; Meng Z.; Huaibin L.; Gengshuo Z.; Mengtang X. Experimental Study on Chemical–Physical Hardening Mechanism of Early Strength of Filling Paste. Geotech. Geol. Eng. 2022, 41, 1715–1724. 10.1007/s10706-022-02363-8. [DOI] [Google Scholar]
  6. Wang C.; Lu Y.; Shen B.; Li Y.; Liang Y. Design And Monitoring of CPB Replacement Mining RSCP: a Case Study in China. Energy Sources Part A 2021, 43 (1), 80–95. 10.1080/15567036.2019.1623944. [DOI] [Google Scholar]
  7. Mashifana T.; Sithole T. Clean Production of Sustainable Backfill Material From Waste Gold Tailings And Slag. J. Cleaner Prod. 2021, 308, 127357. 10.1016/j.jclepro.2021.127357. [DOI] [Google Scholar]
  8. Nasharuddin R.; Luo G.; Robinson N.; Fourie A.; Johns M. L.; Fridjonsson E. O. Understanding the Microstructural Evolution of Hypersaline Cemented Paste Backfill With Low-Field NMR Relaxation. Cem. Concr. Res. 2021, 147, 106516. 10.1016/j.cemconres.2021.106516. [DOI] [Google Scholar]
  9. Martins A. C. P.; Franco De Carvalho J. M.; Costa L. C. B.; Andrade H. D.; de Melo T. V.; Ribeiro J. C. L.; Pedroti L. G.; Peixoto R. A. F. Steel Slags In Cement-Based Composites: An Ultimate Review on Characterization, Applications and Performance. Constr. Build. Mater. 2021, 291, 123265. 10.1016/j.conbuildmat.2021.123265. [DOI] [Google Scholar]
  10. Niroshan N.; Sivakugan N.; Veenstra R. L. Flow Characteristics of Cemented Paste Backfill. Geotech. Geol. Eng. 2018, 36 (4), 2261–2272. 10.1007/s10706-018-0460-8. [DOI] [Google Scholar]
  11. Cui B.; Feng G.; Bai J.; Xue G.; Wang K.; Shi X.; Wang S.; Wang Z.; Guo J. Failure Characteristics and the Damage Evolution of a Composite Bearing Structure in Pillar-Side Cemented Paste Backfilling. Int. J. Miner. Metall. Mater. 2023, 30, 1524–1537. 10.1007/s12613-022-2545-x. [DOI] [Google Scholar]
  12. Cui B.; Feng G.; Bai J.; Wang K.; Shi X.; Wu H. Acoustic Emission Characteristics and Damage Evolution Process of Backfilling Body–Coal Pillar–Backfilling Body Composite Structure. Bull. Eng. Geol. Environ. 2022, 81 (8), 300. 10.1007/s10064-022-02779-9. [DOI] [Google Scholar]
  13. Yao D.; Jiang N.; Wang X.; Jia X.; Lv K. Mechanical Behaviour and Failure Characteristics of Rocks With Composite Defects of Different Angle Fissures Around Hole. Bull. Eng. Geol. Environ. 2022, 81 (7), 290. 10.1007/s10064-022-02783-z. [DOI] [Google Scholar]
  14. Tariq A.; Nehdi M. Developing Durable Paste Backfill From Sulphidic Tailings. Proc. Inst. Civ. Eng. Waste Resour. Manag. 2007, 160 (4), 155–166. 10.1680/warm.2007.160.4.155. [DOI] [Google Scholar]
  15. Liu Y.; Lu Y.; Wang C.; Cui B.; Guo H.; Li H.; Guo Y. Effect Of Sulfate Mine Water On The Durability Of Filling Paste. Int. J. Green Energy 2018, 15 (13), 864–873. 10.1080/15435075.2018.1529582. [DOI] [Google Scholar]
  16. Cui B.; Liu Y.; Guo H.; Liu Z.; Lu Y. Experimental Study On The Durability Of Fly Ash-Based Filling Paste In Environments With Different Concentrations Of Sulfates. Adv. Mater. Sci. Eng. 2018, 2018, 1–12. 10.1155/2018/4315345. [DOI] [Google Scholar]
  17. Chen F.; Gao J.; Qi B.; Shen D. Deterioration Mechanism Of Plain And Blended Cement Mortars Partially Exposed To Sulfate Attack. Constr. Build. Mater. 2017, 154, 849–856. 10.1016/j.conbuildmat.2017.08.017. [DOI] [Google Scholar]
  18. Li W.; Fall M. Sulphate Effect On The Early Age Strength And Self-Desiccation Of Cemented Paste Backfill. Constr. Build. Mater. 2016, 106, 296–304. 10.1016/j.conbuildmat.2015.12.124. [DOI] [Google Scholar]
  19. Pokharel M.; Fall M. Combined Influence Of Sulphate And Temperature On The Saturated Hydraulic Conductivity Of Hardened Cemented Paste Backfill. Cem. Concr. Compos. 2013, 38, 21–28. 10.1016/j.cemconcomp.2013.03.015. [DOI] [Google Scholar]
  20. Ercikdi B.; Baki H.; İzki M. Effect Of Desliming Of Sulphide-Rich Mill Tailings On The Long-Term Strength Of Cemented Paste Backfill. J. Environ. Manage. 2013, 115, 5–13. 10.1016/j.jenvman.2012.11.014. [DOI] [PubMed] [Google Scholar]
  21. Stroh J.; Meng B.; Emmerling F. Deterioration Of Hardened Cement Paste Under Combined Sulphate-Chloride Attack Investigated By Synchrotron XRD. Solid State Sci. 2016, 56, 29–44. 10.1016/j.solidstatesciences.2016.04.002. [DOI] [Google Scholar]
  22. Fall M.; Benzaazoua M. Modeling The Effect Of Sulphate On Strength Development Of Paste Backfill And Binder Mixture Optimization. Cem. Concr. Res. 2005, 35 (2), 301–314. 10.1016/j.cemconres.2004.05.020. [DOI] [Google Scholar]
  23. Orejarena L.; Fall M. The Use Of Artificial Neural Networks To Predict The Effect Of Sulphate Attack On The Strength Of Cemented Paste Backfill. Bull. Eng. Geol. Environ. 2010, 69 (4), 659–670. 10.1007/s10064-010-0326-7. [DOI] [Google Scholar]
  24. Giordano L.; Recupero A.; Tondolo F. Serviceability Behaviour Of Pc Structures By Probabilistic And Fuzzy Probabilistic Approaches. Struct. Infrastruct. Eng. 2008, 4 (2), 153–162. 10.1080/15732470601155565. [DOI] [Google Scholar]
  25. Kawamura K.; Miyamoto A.; Frangopol D. M.; Kimura R. Performance Evaluation Of Concrete Slabs Of Existing Bridges Using Neural Networks. Eng. Struct. 2003, 25 (12), 1455–1477. 10.1016/s0141-0296(03)00112-3. [DOI] [Google Scholar]
  26. Zhang X.; Luo Y.; Yao W. Research On The Sulfate Resistance Of Cement-Based Materials Incorporating Dolomite Powder. Fullerenes Nanotubes Carbon Nanostruct. 2022, 30 (12), 1180–1192. 10.1080/1536383X.2022.2081684. [DOI] [Google Scholar]
  27. Cheng S.; Shui Z.; Sun T.; Gao X.; Guo C. Effects Of Sulfate And Magnesium Ion On The Chloride Transportation Behavior And Binding Capacity Of Portland Cement Mortar. Constr. Build. Mater. 2019, 204, 265–275. 10.1016/j.conbuildmat.2019.01.132. [DOI] [Google Scholar]
  28. Maes M.; De Belie N. Influence of Chlorides On Magnesium Sulphate Attack For Mortars with Portland Cement And Slag Based Binders. Constr. Build. Mater. 2017, 155, 630–642. 10.1016/j.conbuildmat.2017.07.201. [DOI] [Google Scholar]
  29. Matos A. M.; Sousa-Coutinho J. Asr And Sulphate Performance Of Mortar Containing Industrial Waste. Struct. Concr. 2016, 17 (1), 84–95. 10.1002/suco.201400095. [DOI] [Google Scholar]
  30. Li J.; Gao Y.; Yang T.; Zhang P.; Deng W.; Liu F. Effect of water on the rock strength and creep behavior of green mudstone. Geomech. Geophys. Geo-Energy Geo-Resour. 2023, 9 (1), 101. 10.1007/s40948-023-00638-9. [DOI] [Google Scholar]
  31. Su X.Research on the Durability of Concrete Under Saline Soil Environment in the Western Region of Jilin Province [D]; Jilin University, 2013. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All data in this paper are available.

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES