Optimization of Transport Performance and Strength of the Filling Slurry in Tailings Reservoir Waste by Adding Air Entraining Agent

: At present, many mines adopt the ﬁlling method. It is particularly important to solve the problem of the long-distance transportation of slurry during the ﬁlling process. Based on the high-density ﬁlling material of Sanning mine, the experiments were designed to add sodium abietate (SA), triterpene saponin (SJ) and sodium dodecyl sulfonate (K 12 ) with concentrations of 0.0%, 0.2%, 0.4% and 0.6%, respectively, which were used as air entraining agents (AEA). The ﬁlling body with the curing age of 7 and 28 days was prepared for various tests, including nuclear magnetic resonance (NMR), and alternating current (AC) impedance tests. The e ﬀ ects of the air entraining agent and curing time on the physical properties, pore structure and AC impedance properties of the ﬁlling were obtained. The results show that: (1) within the frequency range of 10 − 1 –10 5 Hz, the variation trend of AC impedance of the ﬁlling body cannot be changed by adding the air entraining agent, and the ﬁlling body with the same ratio had a similar topological structure. (2) The ﬁlling body with di ﬀ erent AEA and curing times can be represented by the same equivalent circuit model, while the maximum chi-square coe ﬃ cient was 0.46%. (3) Under the condition of a high frequency of 10 5 Hz, the porosity and uniaxial compressive strength of the ﬁlling body with 7 day curing age were linearly correlated with the AC impedance. However, the porosity and uniaxial compressive strengths of the ﬁlling body with 28 days curing time were a ﬀ ected by the type of AEA at a high frequency of 10 5 Hz. of the ﬁlling body is not regular at a low frequency, and it has the similar trend for high frequency. The AC impedance for the test blocks cured for 28 days with the same AEA concentration and the same frequency were all greater than those of the test blocks cured for 7 days. The results show that the longer the curing time, the weaker the connectivity of the pores inside the backﬁll. This is caused by the hydration expansion reaction, which is the property of the backﬁll itself. However, the AC impedance value of the sample preserved for 28 days was lower than that of the sample without AEA. When the curing time was less than 28 days, the e ﬀ ect of AEA still existed, making the porosity inside the backﬁll larger than that of the sample without AEA. The properties of this kind of AEA can help pump the ﬁlling slurry and increase the ﬂuidity of ﬁlling slurry in actual ﬁlling engineering.


Introduction
Similar to other rock masses, backfill has a complex internal structure. Cement tailings or macadam cement tailings consolidation fill not only has solid particles, but also has water and air. The strength of the filling body is related to the nature of solid particles, the liquid phase of the filling body, and the distribution of the liquid phase in the filling body. Since the first application of electrochemical AC impedance to study the microstructure characterization of filling [1], some scholars have paid more attention to researching this field of engineering [2,3]. Due to the limitations of early testing techniques, AC impedance testing is performed only within the kilohertz (kHz) range [4,5]. With the development of AC impedance testing equipment and electrochemical measurement technology, the AC impedance can be measured in the high frequency range [6,7]. Previous studies had shown that the AC impedance spectrum can characterize the changes in pore structure during the hydration process of the filling body. Shi et al. (1998) [8] studied the influence of various admixtures on the ionic concentration of concrete's pore solution and pore structure, and showed that the conductivity of concrete can be reduced by as the filling aggregate, while fly ash was used as the cementing materials. The particle size distribution is shown in Figures 1 and 2. According to the GB/T 14685-2011 [19], the apparent density, the packing density, and the contents of surface water of aggregates were analyzed. The corresponding results are presented in Table 1.
Minerals 2020, 10, x FOR PEER REVIEW 3 of 17 as the filling aggregate, while fly ash was used as the cementing materials .The particle size distribution is shown in Figures 1 and 2. According to the GB/T 14685-2011 [19], the apparent density, the packing density, and the contents of surface water of aggregates were analyzed. The corresponding results are presented in Table 1.   The mineral compositions of materials were obtained using XRD, and the results are provided in Table 2. The dolomite content is up to 69.65% in gangue. Therefore, it has little effect on the cementation activity of the filling body. The hydroxyllapatite in tailings is as high as 60.94%. This will have an effect on the cement hydration reaction and reduce the amount of tri-calcium silicate (C3S) [10,[15][16][17]. It will affect the strength of the cemented paste backfill. The Al element is mainly present in muscovite and albite, which will affect the activity of fly ash. as the filling aggregate, while fly ash was used as the cementing materials .The particle size distribution is shown in Figures 1 and 2. According to the GB/T 14685-2011 [19], the apparent density, the packing density, and the contents of surface water of aggregates were analyzed. The corresponding results are presented in Table 1.   The mineral compositions of materials were obtained using XRD, and the results are provided in Table 2. The dolomite content is up to 69.65% in gangue. Therefore, it has little effect on the cementation activity of the filling body. The hydroxyllapatite in tailings is as high as 60.94%. This will have an effect on the cement hydration reaction and reduce the amount of tri-calcium silicate (C3S) [10,[15][16][17]. It will affect the strength of the cemented paste backfill. The Al element is mainly present in muscovite and albite, which will affect the activity of fly ash.  The mineral compositions of materials were obtained using XRD, and the results are provided in Table 2. The dolomite content is up to 69.65% in gangue. Therefore, it has little effect on the cementation activity of the filling body. The hydroxyllapatite in tailings is as high as 60.94%. This will have an effect on the cement hydration reaction and reduce the amount of tri-calcium silicate (C 3 S) [10,[15][16][17]. It will affect the strength of the cemented paste backfill. The Al element is mainly present in muscovite and albite, which will affect the activity of fly ash. In this experiment, three AEAs, namely sodium abietate (SA), sodium dodecyl sulfonate (K 12 ) and triterpene saponins (SJ) were selected. The three AEAs are SA, the main chemical component of which was sodium abietate, SJ, the main chemical component of which was triterpenoid saponin, and K 12 , whose main chemical component was sodium dodecyl sulfate, which are widely used in China. Table 3 shows the physical properties of three AEAs. The main instruments used were UCS (WDW-2000, Shanghai new sansi Measuring Instrument Manufacturing Co., Ltd., Shanghai, China), NMR (MiniMR-60, Shanghai Newmai Co. Ltd., Shanghai, China), an electrochemical workstation and an electron microscope scanner (SEM, MIRA3 (LMH/LMU), TESCAN(China) Co. Ltd., Shanghai, China). The main experimental equipment and processes are shown in Figure 3.  In this experiment, three AEAs, namely sodium abietate (SA), sodium dodecyl sulfonate (K12) and triterpene saponins (SJ) were selected. The three AEAs are SA, the main chemical component of which was sodium abietate, SJ, the main chemical component of which was triterpenoid saponin, and K12, whose main chemical component was sodium dodecyl sulfate, which are widely used in China. Table 3 shows the physical properties of three AEAs. The main instruments used were UCS (WDW-2000, Shanghai new sansi Measuring Instrument Manufacturing Co., Ltd., Shanghai, China), NMR (MiniMR-60, Shanghai Newmai Co. Ltd., Shanghai, China), an electrochemical workstation and an electron microscope scanner (SEM, MIRA3 (LMH/LMU), TESCAN(China) Co. Ltd., Shanghai, China). The main experimental equipment and processes are shown in Figure 3.

Methods
The experiment mainly studied the cementing fill body with the same proportion for different curing ages (7 and 28 days). The changes in porosity, strength, and internal microstructure of the filling body were studied by changing the contents of different AEAs. The grouping and proportioning of this experiment are presented in Table 4. Each group had three filling samples. The main experimental steps are as follows.
(1) According to the proportioning scheme presented in Table 4, the filling material was poured into the mixer for full and even mixing. According to the Sanning mine, the slump height of the filling body needs to reach 22 cm; when the mixing amount of AEA was 0.2%, 0.4% and 0.6%, the slump of the filling body increased approximately by 3%, 6% and 8.5%, respectively. It was then poured into the cylindrical standard test mold (50 mm × 100 mm) after stirring. The samples were cured at ambient temperature (20 °C) and a relative humidity of 99% for 7 and 28 days, respectively.

Methods
The experiment mainly studied the cementing fill body with the same proportion for different curing ages (7 and 28 days). The changes in porosity, strength, and internal microstructure of the filling body were studied by changing the contents of different AEAs. The grouping and proportioning of this experiment are presented in Table 4. Each group had three filling samples. The main experimental steps are as follows.
(1) According to the proportioning scheme presented in Table 4, the filling material was poured into the mixer for full and even mixing. According to the Sanning mine, the slump height of the filling body needs to reach 22 cm; when the mixing amount of AEA was 0.2%, 0.4% and 0.6%, the slump of the filling body increased approximately by 3%, 6% and 8.5%, respectively. It was then poured into the cylindrical standard test mold (50 mm × 100 mm) after stirring. The samples were cured at ambient temperature (20 • C) and a relative humidity of 99% for 7 and 28 days, respectively.
(2) After 7 and 28 days of curing, the filling body was vacuumed and saturated. The dry pumping mode was selected for 4 h, whereas the wet pumping mode was selected for 2 h to ensure that the sample was fully saturated with water.
(3) The porosity and T 2 spectrum of the filling body under the condition of saturated water were measured using the NMR instrument. The NMR tests used the MiniMR-60 magnetic resonance imaging (MRI) analysis system, which was manufactured by Shanghai Newmai Co. Ltd., Shanghai, China. The experimental parameters were as follows: receiver bandwidth SW = 100 KHz; sampling points TD = 4502; waiting time Tw = 4800 ms; analog gain RG1 = 20 db; cumulative sampling NS = 64; 90 • pulse application time P1 = 11.52 µs; 180 • pulse application time P2 = 22.48 µs; number of echo cycles NECH = 3000; digital gain DRG1 = 3; coil waiting time RFD = 0.25 ms; number of times NS = 64.
(4) The electrochemical workstation (Integrated electrochemical test system of AMETEK, the experimental parameters were as follows: AC properties = value; start frequency = 10 5 Hz; end frequency = 10 −1 Hz; amplitude = 10 mV; current range = 2 mA; Voltage range = ±6 V; points per decade = 10; data quality = 3), the UCS (the tests were conducted by a computer-controlled fully automatic pressure testing machine (WDW-2000) with a maximum capacity of 200 kN. The specimens were loaded at a constant displacement rate of 1mm/min), and SEM (a specimen of 1 mm 2 was cut from the center part of each cemented paste backfill specimen, then take the prepared samples to the TESCAN MIRA3 field-emission scanning electron microscope to obtain the results) was used to measure the water-filled filling body with different AEAs and curing ages of 7 and 28 days.

Results and Discussion
3.1. AC Impedance of the Filling Body after 7 Days of Curing Figure 4 shows the electrochemical impedance results of the filling body with different mass concentrations of SA (a), K 12 (b), SJ (c) and without AEA (d) for a curing period of 7 days. As can be seen from Figure 4, with the increase in frequency, the AC impedance of the filling body with different mass concentrations of SA, K 12 and SJ showed a similar trend to that of without AEA, both of which decreased gradually with the increase in frequency. This indicates that whether AEA is added or not within the frequency range of 10 −1 -10 5 Hz, the variation trend of AC impedance of backfill remains unchanged, and the filling body with the same proportion of AEA has a similar topological structure. In the low-frequency range of 10 −1 -10 4 Hz, the impedance value of the filling body without AEA is the maximum for the same frequency. Furthermore, the higher the mass concentration of SA and K 12 , the lower the impedance value. However, with the increase in mass concentration, the AC impedance value of SJ first increased, and then decreased. Within the range of 10 4 -10 5 Hz at high frequency, the impedance value of the filling body without AEA was the highest. The variation trends of impedance of the filling body for the three AEAs were basically consistent with each other, indicating that the AC impedance of the filling body decreased with the increase in the mass concentration of AEA. with each other, indicating that the AC impedance of the filling body decreased with the increase in the mass concentration of AEA.  Figure 5 shows the electrochemical impedance results of the filling body with different mass concentrations of SA (a), K12 (b), SJ (c) and without AEA (d) for the curing age of 28 days. As can be seen from Figure 5, with the increase in frequency, the AC impedance of the filling body with different mass concentrations of SA, K12 and SJ showed the same variation trend as that of without AEA, both of which decreased gradually with the increase in frequency. This indicates that whether AEA is added or not within the frequency range of 10 −1 -10 5 Hz, the variation trend of AC impedance of backfill remains unchanged, and the filling body with the same proportion of AEA shows a similar topological structure. In the low-frequency range of 10 −1 -10 4 Hz, the impedance value of the filling body without AEA is the maximum for the same frequency. Adding SA and K12 did not show regularity in the whole low-frequency region; however, regularity was observed only in a certain frequency range. In the high frequency range of 10 4 -10 5 Hz, the AC impedance variation patterns of the filling body for the curing periods of 28 and 7 days were consistent with each other.   Figure 5 shows the electrochemical impedance results of the filling body with different mass concentrations of SA (a), K 12 (b), SJ (c) and without AEA (d) for the curing age of 28 days. As can be seen from Figure 5, with the increase in frequency, the AC impedance of the filling body with different mass concentrations of SA, K 12 and SJ showed the same variation trend as that of without AEA, both of which decreased gradually with the increase in frequency. This indicates that whether AEA is added or not within the frequency range of 10 −1 -10 5 Hz, the variation trend of AC impedance of backfill remains unchanged, and the filling body with the same proportion of AEA shows a similar topological structure. In the low-frequency range of 10 −1 -10 4 Hz, the impedance value of the filling body without AEA is the maximum for the same frequency. Adding SA and K 12 did not show regularity in the whole low-frequency region; however, regularity was observed only in a certain frequency range. In the high frequency range of 10 4 -10 5 Hz, the AC impedance variation patterns of the filling body for the curing periods of 28 and 7 days were consistent with each other.  Figure 6 shows the Nyquist plots of AC impedance of the filling body with AEAs present in the same mass concentration for a curing period of 7 days. The impedance spectrum of the filling body with the curing age of 7 days is the same. The AC impedance of the filling body with AEA is lower than that of the test blocks without AEA for the same curing time. When the mass concentration of AEA was 0.2%, the degree of influence on AC impedance was found in the following descending order: SA > K12 > SJ. When the mass concentration of AEA was 0.4%, the degree of influence on AC impedance was found in the following descending order: SA > SJ > K12. When the mass concentration of AEA was 0.6%, the degree of influence on AC impedance was found in the following descending order: SA > SJ > K12. This phenomenon indicates that the addition of AEA reduces the degree of adhesion of the backfill, increases the connectivity of internal pores, and makes the migration of ions in pore water smoother.   Figure 6 shows the Nyquist plots of AC impedance of the filling body with AEAs present in the same mass concentration for a curing period of 7 days. The impedance spectrum of the filling body with the curing age of 7 days is the same. The AC impedance of the filling body with AEA is lower than that of the test blocks without AEA for the same curing time. When the mass concentration of AEA was 0.2%, the degree of influence on AC impedance was found in the following descending order: SA > K 12 > SJ. When the mass concentration of AEA was 0.4%, the degree of influence on AC impedance was found in the following descending order: SA > SJ > K 12 . When the mass concentration of AEA was 0.6%, the degree of influence on AC impedance was found in the following descending order: SA > SJ > K 12 . This phenomenon indicates that the addition of AEA reduces the degree of adhesion of the backfill, increases the connectivity of internal pores, and makes the migration of ions in pore water smoother. As shown in Figure 7, the Nyquist plots of AC impedance of the filling body with AEA with the same mass concentration for the same curing age. For the curing age of 28 days, the variation of AC impedance of the filling body is not regular at a low frequency, and it has the similar trend for high frequency. The AC impedance for the test blocks cured for 28 days with the same AEA concentration and the same frequency were all greater than those of the test blocks cured for 7 days. The results show that the longer the curing time, the weaker the connectivity of the pores inside the backfill. This is caused by the hydration expansion reaction, which is the property of the backfill itself. However, the AC impedance value of the sample preserved for 28 days was lower than that of the sample without AEA. When the curing time was less than 28 days, the effect of AEA still existed, making the porosity inside the backfill larger than that of the sample without AEA. The properties of this kind of AEA can help pump the filling slurry and increase the fluidity of filling slurry in actual filling engineering. As shown in Figure 7, the Nyquist plots of AC impedance of the filling body with AEA with the same mass concentration for the same curing age. For the curing age of 28 days, the variation of AC impedance of the filling body is not regular at a low frequency, and it has the similar trend for high frequency. The AC impedance for the test blocks cured for 28 days with the same AEA concentration and the same frequency were all greater than those of the test blocks cured for 7 days. The results show that the longer the curing time, the weaker the connectivity of the pores inside the backfill. This is caused by the hydration expansion reaction, which is the property of the backfill itself. However, the AC impedance value of the sample preserved for 28 days was lower than that of the sample without AEA. When the curing time was less than 28 days, the effect of AEA still existed, making the porosity inside the backfill larger than that of the sample without AEA. The properties of this kind of AEA can help pump the filling slurry and increase the fluidity of filling slurry in actual filling engineering.

Equivalent Circuit Analysis of the AC Impedance
As shown in Figure 8, the conductive schematic of the microstructure of the filling body is mainly composed of three conductive modes, namely the continuous conductive channel (CCP), the discontinuous conductive channel (DCP), and the insulator channel (ICP), in which water in isolated pores forms a solid-liquid contact interface with the backfill. According to the measured impedance value of the filling body based on the electrochemical AC impedance data, Zview software was used to simulate the equivalent circuit, and the corresponding

Equivalent Circuit Analysis of the AC Impedance
As shown in Figure 8, the conductive schematic of the microstructure of the filling body is mainly composed of three conductive modes, namely the continuous conductive channel (CCP), the discontinuous conductive channel (DCP), and the insulator channel (ICP), in which water in isolated pores forms a solid-liquid contact interface with the backfill.

Equivalent Circuit Analysis of the AC Impedance
As shown in Figure 8, the conductive schematic of the microstructure of the filling body is mainly composed of three conductive modes, namely the continuous conductive channel (CCP), the discontinuous conductive channel (DCP), and the insulator channel (ICP), in which water in isolated pores forms a solid-liquid contact interface with the backfill. According to the measured impedance value of the filling body based on the electrochemical AC impedance data, Zview software was used to simulate the equivalent circuit, and the corresponding  According to the measured impedance value of the filling body based on the electrochemical AC impedance data, Zview software was used to simulate the equivalent circuit, and the corresponding results are shown in Figure 9. The equivalent circuit diagram of the filling body shows samples with SA, K 12 and SJ with the mass concentration of 0.6% and for different curing ages. In the equivalent circuit model shown in Figure 9, Ws is the warbug impedance of the diffusion process occurring in the filling body, CPE1 is the original normal-phase angle reflecting the properties of the solid-liquid interface in the backfill, and R s is the pore water resistance. Furthermore, the resistance of charge transfer in the backfill is represented by R p .
By simulating the equivalent circuit model of the filling body, it can be seen that when the concentration of SA, K12 and SJ was 0.6% for the same curing age, the equivalent circuit diagram remained unchanged. When the curing ages for the filling body with the same AEA were 7 days and 28 days, the equivalent circuit diagrams remained the same, and the chi-square coefficient had the maximum value of 0.46%. The results show that the main structure of backfill did not change due to different AEAs and curing ages. This means that the components in the equivalent circuit were consistent. However, the internal microstructure of the filling body changed due to changes in AEA and curing ages. Table 5 presents the values of each property parameter of the circuit components. For example, for 0.6% mass concentration of SA for the curing age of 28 days, Rs was 2365 Ω, whereas that for the curing age of 7 days was 560.2 Ω, which is almost 4.2 times less than that for 28 days. These results indicate that longer the curing age, the more adequate the hydration reaction of the backfill, the smaller the internal porosity and the weaker the connectivity.   By simulating the equivalent circuit model of the filling body, it can be seen that when the concentration of SA, K 12 and SJ was 0.6% for the same curing age, the equivalent circuit diagram remained unchanged. When the curing ages for the filling body with the same AEA were 7 days and 28 days, the equivalent circuit diagrams remained the same, and the chi-square coefficient had the maximum value of 0.46%. The results show that the main structure of backfill did not change due to different AEAs and curing ages. This means that the components in the equivalent circuit were consistent. However, the internal microstructure of the filling body changed due to changes in AEA and curing ages. Table 5 presents the values of each property parameter of the circuit components. For example, for 0.6% mass concentration of SA for the curing age of 28 days, R s was 2365 Ω, whereas that for the curing age of 7 days was 560.2 Ω, which is almost 4.2 times less than that for 28 days. These results indicate that longer the curing age, the more adequate the hydration reaction of the backfill, the smaller the internal porosity and the weaker the connectivity.

Analysis of the Porosity of the Filling Body
In this paper, the porosities of water-filled backfill with different concentrations of SA, K 12 and SJ cured for 7 and 28 days were measured using the NMR instrument, and the corresponding results are shown in Figure 10. For the same curing age, the porosity inside the backfill gradually increased with the increase in AEA. For the curing age of 7 days, the minimum porosities of SA and SJ were greater than that of 8.686% without AEA, whereas the porosity inside the backfill was greater than that without AEA when the concentration of K 12 was only 0.6%. This shows that SA and SJ can increase the porosity of the filling body more significantly than K 12 . When the curing age was 28 days, the porosities inside the backfill with SA, SJ and K 12 were all greater than that without AEA. This shows that all three AEAs can increase the porosity of backfill, and their effect is found in the following descending order: SA > SJ > K 12 . Obviously, the porosity of backfill with the curing age of 7 days is greater than that of the filling body with the curing age of 28 days. This indicates that the longer the curing age, the more sufficient the hydration reaction and the smaller the porosity.
SEM was used to analyze the pores in the filling body with the curing ages of 7 and 28 days. The changes of pore structure in the filling body with the addition of AEA and curing age were obtained. It can be seen from Figure 11 that the filling body without AEA for the curing period of 7 days had a smaller porosity with a large number of pores and an uneven distribution. However, the filling body with AEA was more porous with a higher number of pores and even distribution. Due to a certain volume of backfill, the connectivity between the pores inside the filling body will be enhanced. From the microscopic point of view, it is reasonable that the AC impedance value of the filling body with AEA at the same curing age was less than that without AEA. It can be clearly seen from the comparison between the curing period of 7 days and 28 days for 0.6% SA that the longer the curing period, the more sufficient the hydration reaction inside the backfill, which will lead to a smaller porosity. Meanwhile, the larger pore size will gradually decrease, and the smaller pore size will gradually increase. Additionally, the connectivity between the pores will gradually weaken. This indicates that the longer the curing age under the same conditions, the smaller the porosity of backfill and the larger the AC impedance. Liu et al. (1998) [10] showed that the hydration reaction can reduce the pore diameter of the filling body. The experiment in this paper proves this point from the microscopic view, and shows that the filling body with AEA has stronger fluidity, which is conducive to improving the current situation of the long-distance transportation of the filling body. SEM was used to analyze the pores in the filling body with the curing ages of 7 and 28 days. The changes of pore structure in the filling body with the addition of AEA and curing age were obtained. It can be seen from Figure 11 that the filling body without AEA for the curing period of 7 days had a smaller porosity with a large number of pores and an uneven distribution. However, the filling body with AEA was more porous with a higher number of pores and even distribution. Due to a certain volume of backfill, the connectivity between the pores inside the filling body will be enhanced. From the microscopic point of view, it is reasonable that the AC impedance value of the filling body with AEA at the same curing age was less than that without AEA. It can be clearly seen from the comparison between the curing period of 7 days and 28 days for 0.6% SA that the longer the curing period, the more sufficient the hydration reaction inside the backfill, which will lead to a smaller porosity. Meanwhile, the larger pore size will gradually decrease, and the smaller pore size will gradually increase. Additionally, the connectivity between the pores will gradually weaken. This indicates that the longer the curing age under the same conditions, the smaller the porosity of backfill and the larger the AC impedance. Liu et al. (1998) [10] showed that the hydration reaction can reduce the pore diameter of the filling body. The experiment in this paper proves this point from the microscopic view, and shows that the filling body with AEA has stronger fluidity, which is conducive to improving the current situation of the long-distance transportation of the filling body.
(a)  SEM was used to analyze the pores in the filling body with the curing ages of 7 and 28 days. The changes of pore structure in the filling body with the addition of AEA and curing age were obtained. It can be seen from Figure 11 that the filling body without AEA for the curing period of 7 days had a smaller porosity with a large number of pores and an uneven distribution. However, the filling body with AEA was more porous with a higher number of pores and even distribution. Due to a certain volume of backfill, the connectivity between the pores inside the filling body will be enhanced. From the microscopic point of view, it is reasonable that the AC impedance value of the filling body with AEA at the same curing age was less than that without AEA. It can be clearly seen from the comparison between the curing period of 7 days and 28 days for 0.6% SA that the longer the curing period, the more sufficient the hydration reaction inside the backfill, which will lead to a smaller porosity. Meanwhile, the larger pore size will gradually decrease, and the smaller pore size will gradually increase. Additionally, the connectivity between the pores will gradually weaken. This indicates that the longer the curing age under the same conditions, the smaller the porosity of backfill and the larger the AC impedance. Liu et al. (1998) [10] showed that the hydration reaction can reduce the pore diameter of the filling body. The experiment in this paper proves this point from the microscopic view, and shows that the filling body with AEA has stronger fluidity, which is conducive to improving the current situation of the long-distance transportation of the filling body. For the frequency of 10 5 Hz, Figure 12a,b show that the relationship between the porosity of the filling body with different AEA concentrations and the AC impedance for the curing age of 7 days, and the relationship between the porosity of the filling body with different AEA concentrations and the AC impedance for the curing age of 28 days, respectively. The AC impedance and porosity of the filling body adding SA are given after the curing ages of 7 days, as shown in Table 6. Because some of the pores in the filling body are small or closed, the porosity measured by NMR is smaller than the true value. Therefore, the change in filling porosity studied in this paper is relative, mainly based on Figure 11. SEM structure diagram of the filling body. (a) represents the filling body with different AEAs at the 7 day curing age; (b) represents the filling body with different AEAs at the 28 day curing age, the same below.
For the frequency of 10 5 Hz, Figure 12a,b show that the relationship between the porosity of the filling body with different AEA concentrations and the AC impedance for the curing age of 7 days, and the relationship between the porosity of the filling body with different AEA concentrations and the AC impedance for the curing age of 28 days, respectively. The AC impedance and porosity of the filling body adding SA are given after the curing ages of 7 days, as shown in Table 6. Because some of the pores in the filling body are small or closed, the porosity measured by NMR is smaller than the true value. Therefore, the change in filling porosity studied in this paper is relative, mainly based on the nuclear magnetic resonance technology to study the change law of filling porosity. According to Figure 12a, when the curing age was 7 days, the experimental data were fitted linearly, and the values of correlation coefficient (R 2 ) were 0.7116, 0.9129 and 0.9487, respectively. This indicates that porosity is linearly correlated with AC impedance, and the AC impedance decreases with the increase in porosity. As can be seen from Figure 12b, when the curing age was 28 days, the experimental data were fitted linearly. However, such a feat could not be achieved for SJ. The values of correlation coefficient for SA and K 12 were 0.9571 and 0.9655, respectively. The larger the porosity of the backfill with SA, the higher the impedance, while the larger porosity of the backfill with K 12 , the smaller the impedance.

Analysis of the Strength of the Filling Body
In this paper, the electro-hydraulic servo material testing machine of 200 kN was used to measure the uniaxial compressive strength of water saturated backfill with different concentrations of SA, K12 and SJ and the curing ages of 7 and 28 days, respectively. The results are shown in Figure  13. For the same curing age, the uniaxial compressive strength of the filling body decreases gradually with the increase in AEA concentration. When the curing age was 7 days, the uniaxial compressive strengths of the backfill with SA, SJ and K12 with the concentration of 0.4% were close to that of the backfill without AEA at 1.23 MPa. This shows that the three AEAs in this paper have similar effects on the uniaxial compressive strength of the filling body. When the curing age was 28 days, the relationship between the uniaxial compressive strength of the filling body and the AEA was similar to that of the curing age of 7 days. When the concentration of AEA was 0.4%, the strength of the filling body was close to that of the sample without AEA. Obviously, with the same type of AEA, the uniaxial compressive strength of the backfill with the curing age of 28 days was greater than that for the curing age of 7 days. The results show that the longer the curing time, the more sufficient the   Figure 12. Relationship between porosity and AC impedance at 10 5 Hz; (a) represents the filling body with different AEAs at the 7 days curing age; (b) represents the filling body with different AEAs at the 28 days curing age. The experimental results show that the porosity and AC impedance at the curing age of 7 days show a negative linear correlation, while at the curing age of 28 days, the porosity and AC impedance do not show a good regularity, but are related to the type of AEA added. The porosity and AC impedance of backfill with the curing age of 7 days with different concentrations of AEA show a linear negative correlation. However, the relationship between the porosity and AC impedance of the filling body with different concentrations of AEA for a curing age of 28 days at high frequency is affected by the type of AEA added to the system.

Analysis of the Strength of the Filling Body
In this paper, the electro-hydraulic servo material testing machine of 200 kN was used to measure the uniaxial compressive strength of water saturated backfill with different concentrations of SA, K 12 and SJ and the curing ages of 7 and 28 days, respectively. The results are shown in Figure 13. For the same curing age, the uniaxial compressive strength of the filling body decreases gradually with the increase in AEA concentration. When the curing age was 7 days, the uniaxial compressive strengths of the backfill with SA, SJ and K 12 with the concentration of 0.4% were close to that of the backfill without AEA at 1.23 MPa. This shows that the three AEAs in this paper have similar effects on the uniaxial compressive strength of the filling body. When the curing age was 28 days, the relationship between the uniaxial compressive strength of the filling body and the AEA was similar to that of the curing age of 7 days. When the concentration of AEA was 0.4%, the strength of the filling body was close to that of the sample without AEA. Obviously, with the same type of AEA, the uniaxial compressive strength of the backfill with the curing age of 28 days was greater than that for the curing age of 7 days. The results show that the longer the curing time, the more sufficient the hydration reaction inside the backfill, the smaller and more uniform the porosity inside the filling body, and the stronger the backfill strength. WhileŞahin et al., (2011) [16] showed that the moderate concentration of AEA results in the optimal UCS of concrete, in accordance with the experimental results, this shows that the filling body with 0.4% concentration of AEA can improve the production of the mine due to the insufficient strength of the filling body compared with the filling body without AEA, thus relieving the pressure of mining on the ecological environment. For the frequency of 10 5 Hz, Figure 14a,b show the relationship between the UCS and the AC impedance of the filling body with different concentrations of AEA for the curing age of 7 days, and the relationship between the UCS and the AC impedance of the filling body with different concentrations of AEA for the curing age of 28 days, respectively. As can be seen from Figure 14a, when the curing age was 7 days, the experimental data were fitted linearly, and the values of correlation coefficient (R 2 ) were 0.8542, 0.9984 and 0.981, respectively. These results show that the UCS was linearly correlated with the AC impedance, and the AC impedance increased with the increase in strength. As can be seen from Figure 14b, the experimental data for the curing age of 28 days were linearly fitted (except for the data for SJ). The values of correlation coefficient of SA and K12 were 0.9979 and 0.8668, respectively. The higher the strength of the backfill with SA, the lower the impedance value was, while the higher the strength of the filling body with K12, the higher the impedance value.
The experimental results show that the UCS was positively linearly correlated with the AC impedance value for the curing age of 7 days, while the UCS and impedance value at the curing age of 28 days do not show a good regularity, but are related to the type of AEA added to the system. The porosity and impedance of backfill with the curing age of 7 days with different concentrations of AEA were positively linearly correlated. The relationship between the UCS and AC impedance of the filling body with different concentrations of AEA for a curing age of 28 days at high frequency was For the frequency of 10 5 Hz, Figure 14a,b show the relationship between the UCS and the AC impedance of the filling body with different concentrations of AEA for the curing age of 7 days, and the relationship between the UCS and the AC impedance of the filling body with different concentrations of AEA for the curing age of 28 days, respectively. As can be seen from Figure 14a, when the curing age was 7 days, the experimental data were fitted linearly, and the values of correlation coefficient (R 2 ) were 0.8542, 0.9984 and 0.981, respectively. These results show that the UCS was linearly correlated with the AC impedance, and the AC impedance increased with the increase in strength. As can be seen from Figure 14b, the experimental data for the curing age of 28 days were linearly fitted (except for the data for SJ). The values of correlation coefficient of SA and K 12 were 0.9979 and 0.8668, respectively. The higher the strength of the backfill with SA, the lower the impedance value was, while the higher the strength of the filling body with K 12 , the higher the impedance value.
The experimental results show that the UCS was positively linearly correlated with the AC impedance value for the curing age of 7 days, while the UCS and impedance value at the curing age of 28 days do not show a good regularity, but are related to the type of AEA added to the system. The porosity and impedance of backfill with the curing age of 7 days with different concentrations of AEA were positively linearly correlated. The relationship between the UCS and AC impedance of the filling body with different concentrations of AEA for a curing age of 28 days at high frequency was affected by the type of AEA added to the system. days were linearly fitted (except for the data for SJ). The values of correlation coefficient of SA and K12 were 0.9979 and 0.8668, respectively. The higher the strength of the backfill with SA, the lower the impedance value was, while the higher the strength of the filling body with K12, the higher the impedance value.
The experimental results show that the UCS was positively linearly correlated with the AC impedance value for the curing age of 7 days, while the UCS and impedance value at the curing age of 28 days do not show a good regularity, but are related to the type of AEA added to the system. The porosity and impedance of backfill with the curing age of 7 days with different concentrations of AEA were positively linearly correlated. The relationship between the UCS and AC impedance of the filling body with different concentrations of AEA for a curing age of 28 days at high frequency was affected by the type of AEA added to the system.

Conclusions
(1) Within the frequency range of 10 −1 -10 5 Hz, the addition of SA, SJ and K 12 AEAs cannot change the variation pattern of the AC impedance of the filling body, which showed a gradual decrease. The AC impedance of the backfill for the curing age of 7 days was stronger than that of the filling body with the curing age of 28 days.
(2) Equivalent circuit fitting shows that the backfill with different kinds of AEAs and curing ages can be represented by the same equivalent circuit model, indicating that the main structure of the backfill samples had not changed, and that only the internal microstructure of the backfill changed due to the AEAs and curing ages.
(3) For the same curing age, the porosity inside the backfill increased with the increase in AEA. With the same type of AEA, the porosity of backfill with the curing age of 7 days was greater than that of the backfill with the curing age of 28 days.
(4) Under the condition of high frequency of 10 5 Hz, the porosity of the backfill with 7 days of curing age was negatively correlated with the AC impedance, whereas the UCS was positively correlated with the AC impedance. For the backfill with the curing age of 28 days, the relationship between the porosity, the UCS and the AC impedance was affected by the type of AEA at the high frequency of 10 5 Hz.