Flocculation and Settlement Characteristics of Ultraﬁne Tailings and Microscopic Characteristics of Flocs

: Aiming to solve the problems related to the slow settling speed and the long-term consumption of ultra-ﬁne tailings in mine ﬁlling, the effect of ﬂocculant type on the ﬂocculation and settling performance of ultra-ﬁne tailings was studied through static sedimentation experiments on tailings. The microstructure of the ﬂocculation was observed and analyzed using an electron microscope. On this basis, the selection of the optimum ﬂocculant type and dosage parameters was carried out. The results show that the best addition amount of the AZ9020 anionic ﬂocculant was 30 g/t, a solution concentration of 0.3%, and a stirring time of more than 45 min. The ﬂoc structure of the full-tailings ﬂocculation solution was formed by the AZ9020 anionic ﬂocculant. Moreover, the size of less than 0.1 µ m was still relatively large; thus, the overall size of the structure was small and uniformly dispersed. The ﬂoc solution had the smallest porosity, the fractal dimension was the largest, the molecular weight of the ﬂoc was the largest, and the ﬂoc was the most compact, making it appropriate for the rapid removal of ﬂoc structures from water. Sedimentation is also the best ﬂocculant for ﬂocculation and sedimentation. The size of the ﬂocs decreased as the height of the ﬂocculation sediment bed increased during ﬂocculation and sedimentation. The research results provide a microscopic view for the selection of the best ﬂocculant type. settling velocity experienced two peaks: the ﬁrst peak appeared within the ﬁrst 3 min, and the second peak appeared between 5 and 15 min. The calculation shows that the maximum settlement velocity of L4 is K4 = 7.80 mm/min, which is higher than that of the other groups. The limit concentrations of experimental groups L1, L2, L3, L4, and L5 after 24 h of sedimentation were 60%, 58.48%, 58.82%, 58.03%, and 58.59%, respectively, which met the mine-ﬁlling requirements for underﬂow concentrations. Experimental group L4 (AZ9020) generally showed the best reduction in the maximum settlement velocity and in the 1 h height of the clariﬁcation layer. heights show different characteristics. The upper tailings are dispersed, the size of the middle tailings is obviously larger, and the lower tailings become denser as a whole. small-sized ﬂocs during the ﬂocculation and sedimentation processes, forming a uniform and dense ﬂoc pore structure.


Introduction
As a cement filling, tailings can prevent surface subsidence and can reduce the occupation of ground space by tailing stacking. Thus, the use of tailings as a cement filling is an inevitable choice for the green development of current mines [1][2][3]. However, ultrafine tailings encounter problems, such as a long natural settlement time and slow speed, due to their high content of fine particles, thus failing to meet the large-scale continuous filling requirements of mines [4][5][6]. Increasing the sedimentation efficiency by adding flocculants is a common practice in current mines to meet the needs of large-scale continuous filling underground. The sedimentation speed and concentration of tailings particles in the flocculation sedimentation process are affected by many factors, such as the type of flocculant [7,8]. Therefore, selecting a reasonable flocculation type according to the characteristics and properties of tailings is crucial to ensure the flocculation and settlement effect of mine-filling systems.
In recent years, scholars at home and abroad have conducted various studies on the sedimentation laws of ultrafine tailings. Jiao created a tailings sedimentation velocity model through experiments and divided the tailings sedimentation process into the following six stages: turbulent flow affected, accelerated sedimentation, final sedimentation velocity, Ultrafine tailings from the Daye Iron Mine were selected for the experiment. The physical parameters are shown in Table 1, and the particle size composition is presented in Figure 1. The figure shows that the gradation of ultrafine tailings is not particularly uniform. Several coarse and fine particles are observed, and only a few intermediate particles are found. The average median and surface area volume and average particle sizes of the tailings were relatively small, belonging to the category of ultrafine particle tailings. The tailings sorting coefficient was relatively large, making it suitable as a raw material for downhole filling.   Table 2 shows that the main mineral components of the tailing TFe and SFe and contain active oxides, such as CaO, Al2O3, and Mg tailings are inert materials. However, these tailings have correspon ity and can be stimulated for comprehensive utilization.

Flocculant Parameters
Numerous studies have shown that anionic flocculants have effects on tailings [20,21]. Therefore, the flocculants used in this article, AZ9020, and AZ505, are all anionic. The relevant parameters are show White granules or  Table 2 shows that the main mineral components of the tailings are inert oxides SiO 2 , TFe and SFe and contain active oxides, such as CaO, Al 2 O 3 , and MgO. Therefore, ultrafine tailings are inert materials. However, these tailings have corresponding cementing activity and can be stimulated for comprehensive utilization.

Flocculant Parameters
Numerous studies have shown that anionic flocculants have superior flocculation effects on tailings [20,21]. Therefore, the flocculants used in this article, namely AZ358, AZ625, AZ9020, and AZ505, are all anionic. The relevant parameters are shown in Table 3.

Experimental Program and Process
Regardless of the cross-effect between the ultrafine tailings concentration and the additional amount of flocculant, the type of flocculant, unit consumption, concentration, and mixing time using a fixed ultrafine tailings concentration of 30% were determined through indoor sedimentation experiments via the single-factor analysis method. The research, experimental program, and process are shown in Table 4. The height and time of the solid-liquid separation surface drop during the experiment were observed and recorded, as shown in Figure 2. mixing time using a fixed ultrafine tailings concentration of 30% were determined through indoor sedimentation experiments via the single-factor analysis method. The research, experimental program, and process are shown in Table 4. The height and time of the solidliquid separation surface drop during the experiment were observed and recorded, as shown in Figure 2.

Nuclear Magnetic Resonance (NMR) Analysis Experiment
The experimental system used for the NMR analysis in this experiment (MESOMR23-060H-I, Suzhou, China), which has a hydrogen spectrum, is a kind of nuclear magnetic resonance effect of H −1 that can be used for nuclear magnetic resonance spectroscopy and has the following parameters: resonance frequency: 23 MHz, magnet temperature: 25-35 °C, temperature control accuracy: ±0.05 °C. The NMR spectrum is shown in Figure 3a.
The flocculants AZ358, AZ625, AZ9020, and AZ505 were used to perform the flocculation and sedimentation experiments with a solution concentration of 0.3%, a unit consumption of 30 g/t, and a stirring time of 45 min. The stirrer was taken out, and the solution was allowed to stand, and it was then timed to observe the height of the clarified layer in the cylinder. The flocculation solution was extracted at 200 mL of the graduated cylinder after 30 min and was placed in a sealed glass bottle (Figure 3b). The experimental system used for NMR analysis was also used for observation.

Nuclear Magnetic Resonance (NMR) Analysis Experiment
The experimental system used for the NMR analysis in this experiment (MESOMR23-060H-I, Suzhou, China), which has a hydrogen spectrum, is a kind of nuclear magnetic resonance effect of H −1 that can be used for nuclear magnetic resonance spectroscopy and has the following parameters: resonance frequency: 23 MHz, magnet temperature: 25-35 • C, temperature control accuracy: ±0.05 • C. The NMR spectrum is shown in Figure 3a.

Micro-Electron Microscope Scanning Experiment (SEM)
For the SEM observation experiment, an SEM EVO 18 tungsten filament from Ca The flocculants AZ358, AZ625, AZ9020, and AZ505 were used to perform the flocculation and sedimentation experiments with a solution concentration of 0.3%, a unit consumption of 30 g/t, and a stirring time of 45 min. The stirrer was taken out, and the solution was allowed to stand, and it was then timed to observe the height of the clarified layer in the cylinder. The flocculation solution was extracted at 200 mL of the graduated cylinder after 30 min and was placed in a sealed glass bottle (Figure 3b). The experimental system used for NMR analysis was also used for observation.

Micro-Electron Microscope Scanning Experiment (SEM)
For the SEM observation experiment, an SEM EVO 18 tungsten filament from Carl Zeiss (ZEISS) was adopted, along with an image size of 1000 × 750 pixels and an acceleration voltage of 20 kV. SEM experiment 1 : Figure 3 shows the extraction of the four flocculant floc solutions at a volume of 200 mL in a cylinder. This extraction was performed using a long pipette according to the exact scale of the extraction floccules and by gently dropping the cut-out on good filter paper. Liquid nitrogen freezing and fixed, conductive adhesive spraying carbon treatment procedures, the prepared samples were into the micro-electron microscope for scanning observation of the samples, and the electron microscope was operated at a magnification of 2000 times.

Micro-Electron Microscope Scanning Experiment (SEM)
For the SEM observation experiment, an SEM EVO 18 tung Zeiss (ZEISS) was adopted, along with an image size of 1000 × 75 tion voltage of 20 kV. SEM experiment ①: Figure 3 shows the extraction of the four at a volume of 200 mL in a cylinder. This extraction was perform according to the exact scale of the extraction floccules and by gen on good filter paper. Liquid nitrogen freezing and fixed, condu carbon treatment procedures, the prepared samples were into t scope for scanning observation of the samples, and the electron m at a magnification of 2000 times.
SEM experiment ②: The flocculant with the smallest porosi was selected to conduct the flocculation settlement experiment o ings a second time. The flocculant solutions at different bed posi 3, located at 300, 200, and 100 mL, respectively, in the measuring those shown in Figure 4, were extracted and prepared for use as S according to the aforementioned method. The sample was observ tron microscope with magnification of 2000 times.   settling velocity experienced two peaks: the first peak appeared within the first 3 min, and the second peak appeared between 5 and 15 min. The calculation shows that the maximum settlement velocity of L4 is K4 = 7.80 mm/min, which is higher than that of the other groups. The limit concentrations of experimental groups L1, L2, L3, L4, and L5 after 24 h of sedimentation were 60%, 58.48%, 58.82%, 58.03%, and 58.59%, respectively, which met the mine-filling requirements for underflow concentrations. Experimental group L4 (AZ9020) generally showed the best reduction in the maximum settlement velocity and in the 1 h height of the clarification layer.
of the ultrafine tailings solution in the early stages (within 60 min), and the declining height of the clear layer of L4 is always lower than that of the four other groups (L1, L2, L3, and L5). The maximum difference is 80 mm. Figure 5b reveals that the curve of the average settling velocity experienced two peaks: the first peak appeared within the first 3 min, and the second peak appeared between 5 and 15 min. The calculation shows that the maximum settlement velocity of L4 is K4 = 7.80 mm/min, which is higher than that of the other groups. The limit concentrations of experimental groups L1, L2, L3, L4, and L5 after 24 h of sedimentation were 60%, 58.48%, 58.82%, 58.03%, and 58.59%, respectively, which met the mine-filling requirements for underflow concentrations. Experimental group L4 (AZ9020) generally showed the best reduction in the maximum settlement velocity and in the 1 h height of the clarification layer.  Figure 6a shows that the height of the clarified layer of the tailings solution at first increased and then simultaneously decreased when the unit consumption of the flocculant increased. The unit consumption of experiment D3 was 30 g/t, representing the best consumption that was reached. Figure 6b indicates that the maximum settlement velocity first increased and then decreased as the single consumption of the flocculant increased. The single consumption in experiment D4 was 30 g/t, and the maximum settlement velocity was 8.20 mm/min. shows that the height of the clarified layer of the tailings solution at first increased and then simultaneously decreased when the unit consumption of the flocculant increased. The unit consumption of experiment D3 was 30 g/t, representing the best consumption that was reached. Figure 6b indicates that the maximum settlement velocity first increased and then decreased as the single consumption of the flocculant increased. The single consumption in experiment D4 was 30 g/t, and the maximum settlement velocity was 8.20 mm/min.

Flocculation Sedimentation Characteristics of Ultrafine Tailings
3.1.1. Analysis of the Influence of the Flocculant Type on the Settlement Effect Figure 5a shows that the addition of a flocculant accelerates the sedimentation rate of the ultrafine tailings solution in the early stages (within 60 min), and the declining height of the clear layer of L4 is always lower than that of the four other groups (L1, L2, L3, and L5). The maximum difference is 80 mm. Figure 5b reveals that the curve of the average settling velocity experienced two peaks: the first peak appeared within the first 3 min, and the second peak appeared between 5 and 15 min. The calculation shows that the maximum settlement velocity of L4 is K4 = 7.80 mm/min, which is higher than that of the other groups. The limit concentrations of experimental groups L1, L2, L3, L4, and L5 after 24 h of sedimentation were 60%, 58.48%, 58.82%, 58.03%, and 58.59%, respectively, which met the mine-filling requirements for underflow concentrations. Experimental group L4 (AZ9020) generally showed the best reduction in the maximum settlement velocity and in the 1 h height of the clarification layer. shows that the height of the clarified layer of the tailings solution at first increased and then simultaneously decreased when the unit consumption of the flocculant increased. The unit consumption of experiment D3 was 30 g/t, representing the best consumption that was reached. Figure 6b indicates that the maximum settlement velocity first increased and then decreased as the single consumption of the flocculant increased. The single consumption in experiment D4 was 30 g/t, and the maximum settlement velocity was 8.20 mm/min.  Figure 7a shows that the clarification layer height increases at the initial sedimentation stage (0−10 min) as the concentration of the flocculant solution rises. The height of the clarification layer is the largest at the initial sedimentation stage, when the experimental N5 concentration is 0.3%, resulting in the best flocculation sedimentation effect. Figure 7b shows that the average sedimentation velocity at each concentration reached its maximum between 5 and 10 min, and the maximum appeared when the experimental N5 concentration was 0.3%, showing a height of 71.00 mm/min.  Figure 7a shows that the clarification layer height increases at the initial sedimentation stage (0−10 min) as the concentration of the flocculant solution rises. The height of the clarification layer is the largest at the initial sedimentation stage, when the experimental N5 concentration is 0.3%, resulting in the best flocculation sedimentation effect. Figure 7b shows that the average sedimentation velocity at each concentration reached its maximum between 5 and 10 min, and the maximum appeared when the experimental N5 concentration was 0.3%, showing a height of 71.00 mm/min.  Figure 8a shows that the height of the clarified layer increased under the same settling time conditions even though the stirring time increased. Experiments T1, T2, and T3 revealed the presence of transparent floccules of undissolved flocculant that were visible to the naked eye, indicating that the flocculant was only partially dissolved in the water at this time and that the concentration did not reach 0.3%. Figure 8b shows that the clarified liquid height curve is close to the curve when the flocculant concentration is 0.05% when the stirring time is 15 min and that the curve is close to the curve when the concentration is 0.20% and at the stirring times of 25 and 35 min. To achieve an improved settling effect, the stirring time should be more than 45 min.  Figure 8a shows that the height of the clarified layer increased under the same settling time conditions even though the stirring time increased. Experiments T1, T2, and T3 revealed the presence of transparent floccules of undissolved flocculant that were visible to the naked eye, indicating that the flocculant was only partially dissolved in the water at this time and that the concentration did not reach 0.3%. Figure 8b shows that the clarified liquid height curve is close to the curve when the flocculant concentration is 0.05% when the stirring time is 15 min and that the curve is close to the curve when the concentration is 0.20% and at the stirring times of 25 and 35 min. To achieve an improved settling effect, the stirring time should be more than 45 min.

Microscopic Characteristics of Floc Structure of Ultrafine Tailings
In order to study the flocculation effect of different flocculants from the microscopic point of view further, nuclear magnetic resonance analysis was used to observe all of the tailings flocculant solutions formed by the different flocculants and to analyze the porosity and other microscopic parameters of the flocculant solutions.

Analysis of Pore Distribution Characteristics of Ultrafine Tailing Flocculent Solution
The pore size distribution curve of the flocculation solution under the four flocculants is shown in Figure 9a based on the NMR detection and analysis. The spectrum is shown in Figure 9b. tially dissolved in the water at this time and that the concentration did not reach 0.3%. Figure 8b shows that the clarified liquid height curve is close to the curve when the flocculant concentration is 0.05% when the stirring time is 15 min and that the curve is close to the curve when the concentration is 0.20% and at the stirring times of 25 and 35 min. To achieve an improved settling effect, the stirring time should be more than 45 min.

Microscopic Characteristics of Floc Structure of Ultrafine Tailings
In order to study the flocculation effect of different flocculants from the microsco point of view further, nuclear magnetic resonance analysis was used to observe all of t tailings flocculant solutions formed by the different flocculants and to analyze the por ity and other microscopic parameters of the flocculant solutions.

Analysis of Pore Distribution Characteristics of Ultrafine Tailing Flocculent Solu tion
The pore size distribution curve of the flocculation solution under the four floccula is shown in Figure 9a based on the NMR detection and analysis. The spectrum is show in Figure 9b.  Figure 9 reveals that the pore size distribution of the floc solution under the AZ3 AZ625, and AZ505 flocculants is only slightly different. The main peaks are at the sam position, and the width of the pore size distribution is insignificantly different. The tw main peaks appear at 0.01−0.1 μm. The main peak at 0.1−1 μm and the width of the po size distribution are both large, indicating that the flocs with a pore size larger than μm account for a substantial proportion and that the size of the floc structure is also lar The pore size distribution of the AZ9020 flocculant was significantly different fro that of the other flocculants. The two main peaks were located at 0.01 and 0.1 μm. T largest main peak appeared at 0.01 μm, and the pore size distribution between 0.01 a 0.1 μm was also large, indicating that the flocs with a pore size smaller than 0.1 μm counted for a substantial proportion. The results show that the flocs were small in s and that the whole solution was homodispersed. Solutions that were formed by addi the flocculants had a certain proportion of particles with a small pore size.
Research shows that the tightness of the material structure is negatively related the porosity [22]; that is, a smaller porosity leads to a tighter material structure. The p  Figure 9 reveals that the pore size distribution of the floc solution under the AZ359, AZ625, and AZ505 flocculants is only slightly different. The main peaks are at the same position, and the width of the pore size distribution is insignificantly different. The two main peaks appear at 0.01−0.1 µm. The main peak at 0.1−1 µm and the width of the pore size distribution are both large, indicating that the flocs with a pore size larger than 0.1 µm account for a substantial proportion and that the size of the floc structure is also large.
The pore size distribution of the AZ9020 flocculant was significantly different from that of the other flocculants. The two main peaks were located at 0.01 and 0.1 µm. The largest main peak appeared at 0.01 µm, and the pore size distribution between 0.01 and 0.1 µm was also large, indicating that the flocs with a pore size smaller than 0.1 µm accounted for a substantial proportion. The results show that the flocs were small in size and that the whole solution was homodispersed. Solutions that were formed by adding the flocculants had a certain proportion of particles with a small pore size.
Research shows that the tightness of the material structure is negatively related to the porosity [22]; that is, a smaller porosity leads to a tighter material structure. The porosities of the flocculation solution corresponding to flocculants AZ358, AZ625, AZ9020, and AZ505 are 35.96%, 38.84%, 32.65%, and 37.37%, respectively. Among these solutions, the porosity of the flocculation solution corresponding to flocculant AZ9020 is the smallest. Flocculant AZ9020 has the best flocculation and sedimentation effects under the tailings concentration, the amount of additional flocculant, and the concentration.

Fractal Characteristics of Floc Structure of Ultrafine Tailings
The SEM images are shown in Figure 10 and were obtained through the scanning experiment 1 using the electron microscope. The dark parts of the figure represent the pores, and the other parts represent the floc structure. The SEM images are shown in Figure 10 and were obtained through the s experiment ① using the electron microscope. The dark parts of the figure repre pores, and the other parts represent the floc structure. Ultrafine tailings flocculation is under the action of Brownian motion and tur in which the tailings particles collide with the flocculant and combine to form i clusters with fractal characteristics. The fractal dimension is important to charact fractal characteristics of the flocculation index, which can quantitatively describe structure [23,24]. Take N (N = 1, 2, 3, ...) squares with side length r to divide the im divided areas do not overlap, and the area containing the flocs is denoted as N(r).
(1) is then established as follows: (2) is obtained after taking the logarithm: where D represents the fractal dimension. The quantitative analysis based on the SEM scanning images reveals that im narization work is the basis for obtaining fractal analysis. Binarization is als threshold segmentation. Binarization is the process of setting the pixels of an im or 255 and then presenting the entire image with a clear black and white effec Ultrafine tailings flocculation is under the action of Brownian motion and turbulence, in which the tailings particles collide with the flocculant and combine to form irregular clusters with fractal characteristics. The fractal dimension is important to characterize the fractal characteristics of the flocculation index, which can quantitatively describe the floc structure [23,24]. Take N (N = 1, 2, 3, ...) squares with side length r to divide the image. The divided areas do not overlap, and the area containing the flocs is denoted as N(r). Formula (1) is then established as follows: Formula (2) is obtained after taking the logarithm: where D represents the fractal dimension. The quantitative analysis based on the SEM scanning images reveals that image binarization work is the basis for obtaining fractal analysis. Binarization is also called threshold segmentation. Binarization is the process of setting the pixels of an image to 0 or 255 and then presenting the entire image with a clear black and white effect. Thus, binarization is essentially the process of classifying each pixel. Assuming that the size of the SEM picture is M × N, f (x, y) represents the gray value of the pixel in the (x−1)th row and (y−1)th column of the image, where 0 ≤ x ≤ M, 0 ≤ y ≤ N, and x, y are integers. The principle of the gray-scale image binarization process is then presented as follows: where T refers to the threshold. All pixels in the overall image are either black or white after the binarization process. The binarization of the image markedly reduces the amount of data in the image, thus making the contour of the target prominent. Figure 11 shows the binarized image obtained after processing the SEM image, and Figure 12 reveals the obtained fractal characteristic curve.  The calculated fractal dimensions of the samples under the four 1.894, 1.903, and 1.914. The largest fractal dimension indicates the c large molecular weights of the flocculants, which are conducive to from the water. A small particle spacing inside the flocs results in su between the densities of the flocs and the liquid, and large sedimenta improved flocculation effects [25]. The fractal theory indicates that culation effects of the four flocculants is as follows: AZ9020 > AZ62 Flocculant AZ9020 has the best flocculation and sedimentation effe ings. The calculated fractal dimensions of the samples under the four flocculants are 1.887, 1.894, 1.903, and 1.914. The largest fractal dimension indicates the compact flocs and the large molecular weights of the flocculants, which are conducive to rapid sedimentation from the water. A small particle spacing inside the flocs results in substantial differences between the densities of the flocs and the liquid, and large sedimentation speeds facilitate improved flocculation effects [25]. The fractal theory indicates that the order of the flocculation effects of the four flocculants is as follows: AZ9020 > AZ625 > AZ505 > AZ358. Flocculant AZ9020 has the best flocculation and sedimentation effects for ultrafine tailings.

Microscopic Characteristics of Floc Structure of Ultrafine Tailings
In this experiment, scanning electron microscopy was used to observe the different positions of the whole tailings floc solution formed by the same flocculant (AZ9020) to analyze the spatial morphological characteristics and microscopic parameters of the floc structure.
between the densities of the flocs and the liquid, and large sedimentation improved flocculation effects [25]. The fractal theory indicates that the culation effects of the four flocculants is as follows: AZ9020 > AZ625 > Flocculant AZ9020 has the best flocculation and sedimentation effects ings.

Analysis of the Particle and Morphological Characteristics of Tailings Floccules
SEM images of different sedimentation bed heights of flocculant AZ9020 were obtained through the scanning electron microscope experiment 2 , and the tailings flocs were acquired through binarization processing, image contour perfection, edge detection, and boundary discrete-point sealing treatment [26][27][28][29]. The block area is shown in Figure 13. The white part in the picture is the tailings flocs, and their outlines are clearly demonstrated. The particles of tailings flocs of various bed settlement heights show different characteristics. The upper tailings are dispersed, the size of the middle tailings is obviously larger, and the lower tailings become denser as a whole.

Microscopic Characteristics of Floc Structure of Ultrafine Tailings
In this experiment, scanning electron microscopy was used to observe the differe positions of the whole tailings floc solution formed by the same flocculant (AZ9020) analyze the spatial morphological characteristics and microscopic parameters of the fl structure.

Analysis of the Particle and Morphological Characteristics of Tailings Floccules
SEM images of different sedimentation bed heights of flocculant AZ9020 were o tained through the scanning electron microscope experiment ②, and the tailings flo were acquired through binarization processing, image contour perfection, edge detection, an boundary discrete-point sealing treatment [26][27][28][29]. The block area is shown in Figure 13. T white part in the picture is the tailings flocs, and their outlines are clearly demonstrate The particles of tailings flocs of various bed settlement heights show different characte istics. The upper tailings are dispersed, the size of the middle tailings is obviously large and the lower tailings become denser as a whole.  area delineated by the selected floc profile, and the norm ings floc size was acquired as shown in Figure 14. The f upper part of the small-sized flocs accounted for a large were concentrated between 5 and 18 μm; the middle part were distributed between 10 and 30 μm; the lower part o for substantially concentrated distributions between 20 a lent diameters of the upper, middle, and lower tailings floc μm, respectively. The diameter of a tailings floc is positive height.

Analysis of Gray-Scale Characteristics of Ultrafine Tailing Floc Structure
The gray value refers to the dark range of the image, where the white value is 255, and the black value is 0, which can intuitively represent the difference between pores and entities [30]. The original SEM images obtained from the scanning electron microscope 2 experiment are analyzed on the basis of the following gray-scale characteristics: Figure 15a shows that the gray value of flocs at the top of the settlement bed is approximately 180, and the gray value of only one pore is below 50, with an average gray value of 112. Figure 15b reveals that the gray value of the flocs in the middle is approximately 110, with an average gray value of 78. Figure 15c shows that the gray value of the flocs at the bottom is lower than 130, and the gray value of most pores is lower than 50, with an average gray value of 52. Thus, the average gray value of flocs decreases with the bed height. This finding indicates that the water content of the upper flocs is high, and the flocs are bright white. The internal water is constantly drained during the floc sedimentation process, the flocs gradually become dark, and the gray value slowly decreases, thus forming a high concentration of ultrafine total tailings floc solution.
flocs at the bottom is lower than 130, and the gray value of most pores is lower than 5 with an average gray value of 52. Thus, the average gray value of flocs decreases with t bed height. This finding indicates that the water content of the upper flocs is high, and t flocs are bright white. The internal water is constantly drained during the floc sediment tion process, the flocs gradually become dark, and the gray value slowly decreases, th forming a high concentration of ultrafine total tailings floc solution. The different gray values of the original SEM images obtained by SEM experime (2) were defined as yellow, green, cyan, and blue, and the gray values of some regio were extracted and converted into 3D graphics [31][32][33], such as those shown in Figure 1 The upper part of the bed settlement ( Figure 16a) reveals that the size of flocs is larg most of which are over 20 μm. Moreover, the distribution is concentrated, and macropor that are 10 μm in size are found. The middle of the bed settlement (Figure 16b) shows th the size of flocs ranges from 15 μm to 20 μm, and the pore size is evenly distributed. T lower part of the bed settlement (Figure 16c) demonstrates that the size of most flo ranges from 3 μm to 10 μm and that there are many small flocs and a relatively increase number of pores distributed around the flocs. The flocs were also evenly distributed. Th size of tailings flocs decreased as the settlement height decreased. This finding indicat that large-sized flocs gradually settle, dehydrate, and disperse into small-sized flocs du ing the flocculation and sedimentation processes, forming a uniform and dense floc po structure. The different gray values of the original SEM images obtained by SEM experiment (2) were defined as yellow, green, cyan, and blue, and the gray values of some regions were extracted and converted into 3D graphics [31][32][33], such as those shown in Figure 16. The upper part of the bed settlement ( Figure 16a) reveals that the size of flocs is large, most of which are over 20 µm. Moreover, the distribution is concentrated, and macropores that are 10 µm in size are found. The middle of the bed settlement ( Figure 16b) shows that the size of flocs ranges from 15 µm to 20 µm, and the pore size is evenly distributed. The lower part of the bed settlement (Figure 16c) demonstrates that the size of most flocs ranges from 3 µm to 10 µm and that there are many small flocs and a relatively increased number of pores distributed around the flocs. The flocs were also evenly distributed. The size of tailings flocs decreased as the settlement height decreased. This finding indicates that large-sized flocs gradually settle, dehydrate, and disperse into small-sized flocs during the flocculation and sedimentation processes, forming a uniform and dense floc pore structure.

Conclusions
The flocculation and sedimentation characteristics of ultrafine tailings were stud in this paper through indoor sedimentation experiments, NMR analysis, and micro-el tron microscope scanning observations. A reasonable type of flocculant was selected, a the microscopic characteristics of the floc structure in the flocculation and sedimentat processes were analyzed. The main conclusions are as follows:

Conclusions
The flocculation and sedimentation characteristics of ultrafine tailings were studied in this paper through indoor sedimentation experiments, NMR analysis, and micro-electron microscope scanning observations. A reasonable type of flocculant was selected, and the microscopic characteristics of the floc structure in the flocculation and sedimentation processes were analyzed. The main conclusions are as follows: (1) The sedimentation characteristics of ultrafine tailings under different types of flocculants were studied by indoor flocculation sedimentation experiments, NMR monitoring and analysis, and scanning electron microscope observations. The results indicate that the optimal additional flocculant amount of ultrafine tailings from the Daye Iron Mine was 30 g/t, and the optimal concentration for the flocculant solution was 0.3%. The flocculant was completely dissolved when the flocculant was stirred more than 45 min, and the flocculation sedimentation effect was optimal under these conditions.
(2) The analysis of the pore distribution and fractal characteristics of the ultrafine tailings flocculation solution revealed that the size of flocs in the ultrafine tailings flocculation solution formed by the AZ9020 flocculant was less than 0.1 µm and that the overall structure size was small and evenly dispersed. The flocculant solution had the smallest porosity, the largest fractal dimensions, and the most compact flocs, indicating that it had the best flocculant sedimentation effect.
(3) The analysis of the spatial morphology and gray-scale characteristics of the ultrafine total tailings flocs showed that the average gray-scale value and size of the flocs decreased as the height of the flocculation sedimentation bed decreased. In the flocculation and sedimentation processes, the large-sized flocs gradually dehydrated and dispersed into small-sized flocs, forming a more uniform and compact pore structure for the flocs.