Enhancement of ultrafine coal slime flotation using high-shear pretreatment and polyvinylpyrrolidone

ABSTRACT Flotation has been used to treat coal over the last century. However, the conventional flotation condition can hardly be applied to ultrafine coal slime (UCS). An alternative combined method, which has the potential to improve UCS separation efficiency, is using high-shear pretreatment (HSP) and polymer regulator. Presented in this paper are the findings from an investigation into the flotation performance and aggregate characteristics of UCS with the assistance of HSP and polyvinylpyrrolidone (PVP). Electrokinetic, laser particle size, and rheology experiments were employed to analyze the influence of HSP and PVP on ultrafine particles. The UCS studied has a high ash content of 53.62%. The flotation results were analyzed through flotation efficiency and selectivity indexes. The results show that the flotation concentrate has less gangue entrainment and high combustible recovery, low ash content (12.74%) and higher flotation efficiency (66.26%) were obtained. Moreover, the results of flotation kinetics illustrate the advantage of using PVP, which improves the cumulative combustible recovery and flotation rate. The introduction of HSP was conducive to the improvement of separation selectivity. These findings provide fundamental insights into the effect of HSP and PVP on ultrafine particles and offer an effective technique for the enhancement of UCS flotation.


Introduction
The significance of coal for energy consumption in the world is beyond parallel, recycling coal tailings to produce clean coal is a valuable topic.As an effective method for the separation of fine minerals, flotation has been used over the last century (Norori-McCormac et al. 2017).However, the challenges of coal separation have been escalating, with the application of large machinery and the increasing mechanization of the modern mining industry, resulting in large amounts of coal slime (Wang et al. 2018).Although there has been no consensus on the definition of ultrafine coal particle, the coal particles with diameter less than 100 μm certainly can be classified as ultrafine coal (Wang and Wang 2018).The ultrafine size caused many problems for the separation.On the one hand, the gangue entrained in the ultrafine coal slime (UCS) were usually difficult to remove, which constrained the ulterior separation and utilization (Zhao et al. 2021).On the other hand, the UCS was often oxidized due to the exposed storage conditions.Therefore, UCS has surface cover of hydrophilic matters and low floatability (Barry, Klima, and Cannon 2017).All these difficulties in recycling UCS are troublesome so as to the inappropriate classification of them into the solid waste.Thus, recycling of the UCS not only saves energy resources but also protects the environment.
Over the years, attempts have been made to adapt to the benefits of UCS.Raw coal ores are often associated with clay or gangue minerals.Thus, efficient liberation techniques were developed to reduce the association of minerals (Ma, Tao, and Xian 2021).In recent years, improvements in separation equipment have supported the recycling of coal slimes.For example, the flotation cell was enhanced and well-matched with particles smaller than 20 μm, wherein the ultrafine particles obtained sufficient collision probability with bubbles (Jameson 2010).For UCS, the separation difficulty grew with more entrainment of other minerals.Flotation of UCS was highly related to the slurry condition.Increasing the agitation energy input in flotation could be a good method for improving flotation performance (Engel, Middlebrook, and Jameson 1997;Wang et al. 2020).This is interpreted as the requirement of high-intensity conditions and water turbulence for the sufficient contact between particles and bubbles (Pascoe and Wills 1994).However, the desliming performance and recovery are often unable to be satisfied simultaneously.Hence, the shear treatment was separated from flotation, which alleviated the influence of shear treatment on flotation recovery (Raghavan et al. 2004).
To further mitigate the weakness of high shear conditions to flotation recovery, efficient flotation reagents were developed to improve flotation recovery.Hydrophobic combustible components of UCS were adapted to be gathered in the concentrate, whereas hydrophilic incombustible components tend to be dispersed in water.Traditional nonpolar reagents, such as kerosene, obtained poor efficiency in the UCS flotation, which can be attributed to the poor dispersity of kerosene droplets (Chen et al. 2022).The surface properties of UCS, such as oxidized surface, high specific surface area, and higher surface roughness, also deteriorated flotation performance (Xu et al. 2017).Another method for recovering hydrophobic components from tailings is oil agglomeration, which also received extensive attention (Mehrotra, Sastry, and Morey 1983;Tian et al. 2017).The aggregation method was also expressed as floc-flotation (Coleman et al. 1995).The kinetics of flocculation coalescence and breakage can be affected by many parameters, such as shear conditions, reagents, and pH (Ozer, Basha, and Morsi 2016;Patil, Andrews, and Uhlherr 2001).The floatability and wettability of coal can be regulated by reagents, such as surfactants, which modified the surface properties of coal (Dey 2012).For instance, the dodecyl trimethyl ammonium bromide was used as the surfactant in coal flotation (Xia et al. 2019).The adsorption of reagents on the surface of particles is significant, wherein collision and aggregation of particles are susceptible (Ozmak and Aktas 2006).Compound reagents, such as emulsion binders, were used for fine particle flotation through selective agglomeration (van Netten, Moreno-Atanasio, and Galvin 2014).In dealing with the separation of fine particles, polymers possess the advantages of better aggregation performance.Recently, the application of polymers to mineral flotation has received attention.The flotation of fine copper sulfide was improved by the tuneable polymer (Ng et al. 2018).The combustible recovery of coal flotation was increased with the introduction of polyacrylamide (Zou et al. 2019).The polymeric aluminum was also used for separation of coal, to enhance the flotation recovery (Li, Xu, and Zhang 2020).Studies on the effect of polymers on UCS separation are relatively rare, and the application of polymers in UCS flotation is of exploratory significance.
In this study, to recycle combustible matter from UCS, we tested a combined method using HSP and polymer regulators.A rarely used polymer in coal beneficiation, polyvinylpyrrolidone, was adopted in this study.PVP has been adopted in the material and chemical engineering before this study (Alabresm et al. 2018;Palchoudhury and Lead 2014).PVP can play a regulatory role in ultrafine particle agglomeration in our previous study (Lin et al. 2022).To date, no study has explored the synergistic effect of HSP and PVP on UCS flotation.This study intended to investigate how the HSP and PVP facilitate the separation of UCS.
To understand the collaboration of HSP and PVP, this study investigated their effects through various methods.First, zeta potential analyses were conducted to explain the impact of HSP and PVP on the surface electrical properties of ultrafine particles, wherein the purified minerals were used to simulate different components of UCS.The structure of aggregates was then investigated in terms of size and strength.The size and strength of aggregates were measured by a laser particle size analyzer and a rheology analyzer, respectively.In order to illustrate the influence of HSP and PVP on flotation, laboratory flotation tests with HSP and PVP were conducted to observe how they facilitated the separation of UCS.Flotation efficiency and selectivity index (SI) were introduced to evaluate the results.This study provides a novel technique for the separation of UCS.

Materials
The UCS used in this study was sourced from a plant located in Shandong, China.To figure out the basic properties of the UCS, the proximate and elemental analyses were carried out according to GB/T 212-2008, and results are shown in Table 1.The ash content of UCS is 53.62%, higher than the defined value of high ash coal in GB/T 15,224.1-2018.A laser particle size instrument (Beckman, LS 13,320, Brea, CA, USA) was used to test the particle size distribution of UCS.Results in Fig. 1 showed that the mean size (D 50 ) of the UCS was 24.15 μm, with 90% of particles (D 90 ) smaller than 69.13 μm.
To identify the primary mineral compositions of UCS, the UCS sample was scanned by XRD (Rigaku Smart Lab, Tokyo, Japan).The results in Fig. 2 showed that the main gangue components (coal deducted) in UCS were quartz, kaolinite, and calcite.
For theoretical analyses, a purified low ash coal sample (A ad < 5%) was obtained from the same plant in Shandong, China.The pure kaolinite, quartz, and calcite were obtained from Mairuida Technology Co., Ltd., Beijing, China.All the pure minerals were ground to

Zeta Potential Measurement
A zeta potential analyzer (Zeta plus, Brookhaven, NY, USA) was used to detect the electrical properties of pure minerals, in terms of surface charge and zeta potential value.Deionized water (Milli-Q Integral 5, Chesnes, France) with a resistivity of less than 18.2 MΩ⋅cm was used.To enhance the accuracy of the test, a standard solution was used.The pH of the solution was maintained at 6.8 during the measurement.The PVP solution was prepared with several concentrations (1, 3, 5, 7, 10, 15, 20 mg/L).For the tests with PVP, the pure minerals were added to PVP solution to prepare the suspension (Mei et al. 2022).The effects of HSP on zeta potentials were measured at a rotation speed of 9000 rpm.Subsequently, 10 mL of centrifuged suspension (Avanti J-26×PI, Beckman, Brea, CA, USA) was filtered to prepare the sample for zeta potential tests.

Aggregates Size Measurement
Aggregates size was measured by the laser particle size instrument (Beckman, LS 13,320, Brea, CA, US).The value of refractive index used in this study was set to 1.85, 1.46, 1.53, and 1.56 for coal, quartz, kaolinite, and calcite, respectively.The deionized water was circulated to transfer the solid samples.For each test, 4 g minerals were added.The tests with 5 mg/L PVP and 9000 rpm HSP were performed to investigate the effect of HSP and PVP on aggregates size.All these tests were conducted with multiple replications to ensure the accuracy (Lin et al. 2022).

Aggregates Strength Measurement
In this study, the rheological behaviors of particle suspension were tested to characterize the strength of aggregates.A rotational rheometer (Kinexus, Netzsch, Bavaria, Germany) was employed with a round plate (Hu et al. 2021).The samples were prepared as suspensions (7 wt%) before tests.The 5 mg/L PVP and 9000 rpm HSP were selected to represent the effect of HSP and PVP on aggregates strength.As a valid method to characterize the aggregates strength, rheological measurements provided the data on apparent viscosity and shear stress (Turian et al. 2002).The apparent viscosity and yield stress help characterize the ability of aggregates to resist shearing.The apparent viscosity of shear rate between 0 and 100 s −1 was recorded, and the corresponding shear stress was also recorded.

HSP and Flotation Experiments
The HSP was conducted in a tank with a stirrer whose speed was controlled by a digital modifier (Eurostar 20 digital, IKA Co., Ltd., Staufen, Germany).The volume and diameter of the tank for the HSP are 5 L and 100 mm (Figure S3).The distance of the blender from the bottom of the tank is 50 mm.Laboratory flotation tests were carried out in an XFD flotation cell (Figure S4) purchased from Wuhan Exploration Machinery Co., Ltd., Wuhan, China.The cell volume was 3 L, and the stirring speed was controlled at 1800 rpm.
For the flotation with HSP, UCS (210 g), and water (3000 mL) were mixed in the tank for 2 min at 9000rpm HSP to prepare the flotation feed (7 wt%).The flotation tests with the PVP were performed at 5 mg/L of the PVP solution.The optimization steps are shown in Figure S3 and Figure S4.DDA and 2-Octanol were added as collector and frother.All the tests were conducted at an environmental temperature (25°C) and natural pH (6.8).
The results of flotation were evaluated by combustible recovery (ε) and flotation efficiency (η) as Equation (1) and Equation (2): where γ j is the yield of froth concentrate (mainly combustible matters), A j is the ash content of froth concentrate, and A y is the ash content of the flotation feed (UCS).
In addition, we calculated the kinetics parameters to measure the effect of HSP and PVP on UCS flotation.The first-order kinetics model was used (Barbian, Ventura-Medina, and Cilliers 2003).The first-order dynamics equation is shown as Equation (3): Where ε was the concentrate recovery, ε 1 was the theoretical maximum value of concentrate recovery, k was the fitted value of rate constant, and t was the duration for the collection of concentrate.
To calculate the selectivity index (SI), a modified rate constant (K m , min −1 ) is first introduced as Equation ( 4): Subsequently, the SI is calculated by Equation ( 5): where K m;com , K m;ash were the modified rate constants of combustible matter and ash, respectively (Mei et al. 2022).

Zeta Potential Analyses
As a widely used method to analyze the electrical property, the variation of zeta potential value explains the influence of promoters on electrostatic attraction or repulsion between particles (Mei et al. 2022).The zeta potential values of pure minerals were shown in Figs. 3  and 4. On the one hand, the zeta potential values of the pure minerals increased with the presence of PVP (Fig. 3), which indicated that the introduction of PVP in solution decreased the repulsion between ultrafine particles.The zeta potential values of initial minerals (without PVP) were −31.83, −33.37, −34.67, and −36.48 mV for coal, calcite, kaolinite, and quartz (Table S1), respectively.However, for different minerals, the increase in the value of the zeta potential with the presence of PVP was different.The maximum  increase occurred in coal, which was 8.29 mV.For the calcite, kaolinite, and quartz, the increases were 6.15, 5.48, and 6.70 mV, respectively.The differences demonstrate that the attraction between coal particles is more adapted to be influenced by PVP.On the other hand, the zeta potential values of pure minerals were decreased by HSP.The zeta potential decrease in coal was 2.13 mV (from −31.83 mV to −33.95 mV), whereas that for calcite, kaolinite, and quartz were decreased to 4.15, 4.31, and 3.99 mV (Table S2).The introduction of HSP caused an electrostatic repulsion between particles, which inhibited their aggregation.Wherein the effect of HSP on coal was much weaker than other minerals (Fig. 4).In conclusion, the HSP and PVP played the opposite role in the interaction of pure minerals.It can be deduced that the HSP dispersed mineral particles before flotation, released the heterogeneous aggregation, wherein the coal would be more difficult to be dispersed and be easier to aggregate with the assistance of PVP.These effects offered the opportunity to enhance the separation of UCS.The effectiveness of our method still needs to be evaluated with the characterization of aggregates.

Aggregates Size Analyses
The size distribution of pure minerals is shown in Figs. 5 and 6.It can be seen in Fig. 5 that the size of aggregates increased with the presence of PVP.The amplification in the size of coal aggregates was the most remarkable.However, the increase in other minerals was not so indistinctive.These effects can be attributed to the selective regulatory role of PVP on aggregates.In Fig. 6, the D 10 , D 50 , and D 90 indicated the cumulated volume of particles, which occupied 10%, 50%, and 90% of the total volume.These indicators were commonly used to characterize the size distribution of aggregates.The initial D 50 of coal, quartz, kaolinite, and calcite were 20.23, 6.67, 5.94, and 6.13 μm, respectively (Table S3).The D 50 of coal aggregates was increased from 20.23 μm to 45.96 μm, and the D 90 was increased from 57.72 μm to 105.06 μm (Table S5) with presence of PVP.The increase in aggregates size was beneficial to the adhesion of coal on bubbles, leading to a faster flotation rate.Meanwhile, the aggregates size of other minerals was much less than that of coal (attained an amplification of 127% in the D 50 ).For comparison, the recorded details of size can be found in Table S3-S5.The increase in the D 50 of quartz, kaolinite, and calcite both were around 30%, reached at 8.66 μm, 7.81 μm, and 8.14 μm (Table S5), respectively.These variations proved the role of PVP in the enhancement of coal aggregates.In contrast, the opposite effect in the size distribution was observed with the addition of HSP (Fig. 6a and Table S4).The decrease in the size of coal aggregates was less than that of other minerals.It can be deduced that the coal aggregates were stronger than other mineral aggregates.The use of HSP can be detrimental to the entrainment of other minerals in the flotation concentrate.And the deduction would have to be proved by other measurements.According to the comprehensive analysis, part of the other minerals adhered to the coal was removed with the help of HSP.The subsequent introduction of PVP in flotation enhanced the flotation of coal aggregates.This is the purpose of the separation of coal from UCS.To further investigate the aggregates structure, the strength of aggregates was studied.

Aggregates Strength Analyses
The results in Figs.7 and 8 approved the effects of HSP and PVP on aggregates strength.The strength of aggregates was analyzed through rheology behavior analyses.In principle, more hydrophobic coal aggregates were adapted to possess the larger values of apparent viscosity and shear stress.However, the hydrophilic gangue aggregates were not resistant to shear force.The more intense aggregates would be beneficial to the collision and adhesion with bubbles.As for the influence of PVP on rheology behaviors of aggregates, the values of apparent viscosity and shear stress of coal were always greater than those of other minerals.However, the effects of HSP were opposite.As the parameters for evaluating the interaction between particles, the greater value of apparent viscosity and shear stress illustrated means the stronger aggregates.Therefore, the results shown in Figs.7 and 8 implied that the introduction of PVP would benefit the aggregates strength, whereas the aggregates were weakened by HSP.It is worth noting that the aggregates strength of different minerals were distinguishing.The strength of coal aggregates is more likely to be strengthened by PVP, rather than other minerals.And the dispersion effect of HSP on coal aggregates was lower than that of other minerals.Thus, it can be deduced that the collaboration of PVP and HSP would selectively enhance the coal aggregates and disperse the other minerals in water.The deduction would be proved by the analyses of flotation results.

Laboratory Flotation
Laboratory flotation tests were used to evaluate the influence of HSP and PVP on flotation.Flotation tests with different promoters were performed and recorded.To calculate the recovery and flotation rate parameters, the flotation concentrates and tailings were analyzed.The results in Table 2 and Fig. 9 showed the effect of HSP and PVP on UCS flotation.Theoretical maximum recovery and kinetic parameters (R max , k f , and K m ) are calculated using Equation (3)and Equation (4).It can be found that an increase occurred in both combustible matter recovery and rate constant with PVP as the only promoter.The influence can be attributed to the enhancement of aggregates structure.In contrast, the combustible matter recovery and rate constant both decreased with HSP.As for the entrainment of other minerals, the ash content of the concentrate illustrated that the collaboration of HSP and PVP was beneficial for the inhabitation of heterogeneous aggregation of coal and ash.The variation in flotation rate of ash content also supported the conclusion.As modified rate constant, K m took both the flotation recovery and rate constant into consideration.The K m of concentrate reflected the role of HSP and PVP on different components of UCS.The values of K m;com were increased with the PVP as promoter, and performed opposite with HSP as promoter.It is worth noting that K m;ash were decreased by HSP, which proved the improvement of HSP on flotation selectivity.In conclusion, the ash entrainment was inhibited without affecting concentrate recovery through the combined promoters of HSP and PVP.These effects were attributed to the collaboration of HSP and PVP.The purpose of enhancing UCS flotation is to recycle the coal resources and reduce the residual of coal in tailing, realizing the resource utilization of UCS.With the collaboration of HSP and PVP, the aim of reducing ash entrainment in the concentrate and reducing coal residual in tailings were realized simultaneously.To illustrate the effects of PVP and HSP on separation efficiency, the flotation indexes were drawn in Fig. 10.As a blank control group, the test without HSP and PVP was performed.The concentrate of the blank control group obtained a high ash content (15.78%), low flotation efficiency (62.91%), and a SI value of 13.95.With the HSP as the only promoter, a lower ash content (11.81%) of the concentrate was achieved.The SI (16.44) was excellent and should be attributed to the role of HSP in inhibition of ash entrainment.However, the flotation recovery decreased with HSP, leading to the restriction of flotation efficiency.Tests with PVP as the only promoter did not yield well in the value of SI.The combined method of HSP and PVP obtained the highest flotation efficiency (66.26%) and a low ash content (12.74%) of the concentrate.The selectivity index and flotation efficiency both satisfied with the collaboration of PVP and HSP, which proved the advantage of the novel method.In previous research of our study, it was found that the impacts of low dosage of PVP can play a regulatory role on coal agglomeration (Palchoudhury and Lead 2014).The modified coal surface would reduce the energy barrier between particles and bubbles, leading to a faster flotation rate.The HSP provided a better condition for dispersion of different minerals, which offered an opportunity for the adsorption of PVP on coal as shown in Fig. 11.Thus, the agglomeration impacts of PVP on coal would be stronger than other minerals.In this study, the zeta potential analyses and aggregate analyses proved the effects of PVP and HSP on the attraction and repulsion between particles and illustrated the collaboration of HSP and PVP on UCS flotation.

Conclusion
In this work, the flotation performance of UCS was improved by the collaboration of HSP and PVP.This novel combined method was an attempt to find an economical solution for the enhancement of UCS separation and utilization.The tests with HSP indicate that the HSP plays a role in dispersing particles.The HSP inhibited the entrainment of ash content in concentrate.Thus, it can be used as a substitute for dispersant or inhibitor.The flotation  performance of UCS is illustrated through the analyses of aggregates.The influence of HSP and PVP on aggregates is measured by laser particle size and rheology measurements.The size and strength of aggregates are selectively influenced by the collaboration of PVP and HSP, wherein the different minerals were separated.The HSP and PVP were introduced into the laboratory flotation, and test results proved the benefits of the HSP and PVP on the UCS flotation.The main conclusions are summarized below.
(a) The UCS was a mixture of coal and other minerals.The promoter of HSP enhanced the repulsion between particles.The size and strength of aggregates were also affected by HSP.The particles can be dispersed by HSP, thus fewer ash entrainment would be obtained.The purified coal aggregates lead to the lower ash content of the concentrate.However, the flotation recovery would be decreased due to the weak structure of aggregates.Thus, the single promoter of HSP was difficult to play an effective role in UCS flotation.(b) The adsorption of PVP on particles affected the surface properties of minerals.These effects are directly reflected in the zeta potential value of particles.Meanwhile, the aggregates were regulated by PVP before flotation.The flotation rate and combustible recovery were significantly improved with PVP.However, it was found that the entrainment of ash in the concentrate was still high.Thus, the PVP as a single promoter cannot yield well in UCS flotation.(c) The separation of UCS needs to address many issues.The entrainment and aggregates were the main issues solved by this study.After the dispersion of HSP, the PVP enhanced the aggregates, which would be beneficial for the inhabitation of heterogeneous aggregation and flotation recovery.The collaboration of PVP and HSP showed advantage in enhancement of UCS flotation and improved the separation efficiency.

Figure 3 .
Figure 3.Effect of PVP on zeta potentials of pure minerals.

Figure 4 .
Figure 4. Effect of HSP on zeta potentials of pure minerals.

Figure 5 .
Figure 5. Size distribution of pure minerals and their aggregates (with PVP adsorbed).

Figure 6 .
Figure 6.The variation in D 10 , D 50 , and D 90 of pure minerals with the introduction of HSP (Fig.6a) and PVP (Fig.6b).

Figure 7 .
Figure 7. Apparent viscosity of aggregates with the increase of shear rate.

Figure 8 .
Figure8.Shear stress of aggregates with the increase of shear rate (The legend is the same as Fig.7).

Figure 10 .
Figure 10.Indexes of UCS flotation (a.without HSP and PVP; b. with HSP; c. with PVP; d. with HSP and PVP).

Figure 11 .
Figure 11.Schematic illustration of HSP and PVP on UCS flotation.

Table 1 .
Proximate and elemental analysis of UCS.