Comparative Study on the Effect of Pyrophosphate and Tripolyphosphate on the Flotation Separation of Arsenopyrite and Muscovite

Abstract

The aim of the study was to compare the effects and mechanism of tetrasodium pyrophosphate (TSPP) and sodium tripolyphosphate (STPP) as dispersants on the selective flotation of arsenopyrite from muscovite. The results of single-mineral flotation showed that the recovery of arsenopyrite was 81.4% when no dispersant was added. The recovery of arsenopyrite slightly decreased with increasing dosage of TSPP. When the dosage of STPP was 6 × 10⁻⁵ mol/L, the recovery of arsenopyrite was only 28.6%. Neither of the dispersants had significant influence on the muscovite flotation (<10%). However, in a mixed-mineral system, the recovery of arsenopyrite dropped significantly, and then under the action of dispersants, its recovery back up. The RPM results showed that the dispersion effect of TSPP was superior to that of STPP. The electrokinetic potential showed that the potential change value of muscovite with TSPP was −17.37 mV, while that of muscovite with STPP was −8.33 mV (pH = 8). The adsorption of TSPP onto muscovite was stronger than that of STPP. FTIR and XPS analysis confirmed that dispersants exhibited chemical adsorption with the Al atoms on muscovite and that dispersant STPP exhibited stonger adsorption than TSPP on arsenopyrite, which was consistent with flotation experiments.

1. Introduction

Due to the year-by-year exploitation of high-grade gold mines, low-grade gold mines with high clay mineral content have gradually become the mainstay of beneficiation plants [1]. The high content of clay minerals in the ores, such as chlorite and mica, has caused many problems for mineral processing workers due to their low hardness and layered structure [2,3]. The gold-bearing minerals in the ores processed by the plant of Dulan Jinhui Mining Co., Ltd. (Golmud, Qinghai province, China) are primarily arsenopyrite that is found in association with mica [4]. Relevant studies had shown that due to the typically lower hardness of clay minerals than that of sulfide minerals, the particle size of clay minerals entering the flotation workshop was usually fine, which could cover the surface of sulfide minerals and hinder the adsorption with collectors [5]. Mechanical entrainment could mix clay minerals with the flotation concentrate, causing deteriorated flotation performance.

In recent years, beneficiation practitioners have conducted extensive research to determine how to eliminate the negative impact of clay minerals on flotation. This can be achieved through physical and chemical methods. The physical methods primarily consist of high-intensity stirring and ultrasonic pretreatment to remove the covering layer of clay minerals on the surface of sulfide minerals [6]. Although such physical methods had been demonstrated to be feasible in laboratory studies, they were rarely used in industrial production due to their high costs [7,8]. Chemical dispersants had been developed to improve flotation by removing clay layers that cover the surface of sulfide minerals or by changing the network structure of clay minerals [9,10]. Dispersants have a wide range of applications in dressing plants and are also commonly used in coatings, inks, and food industries. Andres Ramirez et al. [11] used sodium hexametaphosphate and sodium silicate as dispersants to alleviate the negative effects of kaolinite on chalcopyrite during seawater flotation. Selim et al. [12] investigated the effect of different dispersants on the dispersion behavior of bentonite. It was discovered that dispersants could increase the absolute value of the particle surface potential, which improved electrostatic repulsion, which prevented particle aggregation and gravity sedimentation, improving the dispersion degree of slurry. Liu Shiqi et al. [13] found that high clay content in gold ores during flotation resulted in highly mechanical entrainment in clay minerals, which led to a low gold grade. By adding DP-1777 (a lignosulfonate-based biopolymer) to slurry, it was found that the reagent significantly increased the gold grade and recovery by reducing mechanical entrainment and slurry viscosity.

Tetrasodium pyrophosphate (TSPP) and sodium tripolyphosphate (STPP) are dispersants in flotation. TSPP was used in the flotation separation of calcite from apatite, and STPP was used to selectively depress calcite in magnesite flotation. In the study, TSPP and STPP are used as dispersants in arsenopyrite/muscovite flotation to investigate the effect on clay minerals. By comparing the effects of dispersants on the recovery of minerals, it is hoped that references can be provided to related concentrators for eliminating the adverse effects of clay minerals in sulfide flotation. The mechanism by which dispersants affect the surface of minerals is revealed, from the minerals’ surface to the microscopic level of atoms, through a reflection polarizing microscope (RPM), electrokinetic potential, Fourier transform infrared spectrum (FTIR), and X-ray photoelectron spectroscopy (XPS).

2. Experimental Section

2.1. Minerals and Reagents

Arsenopyrite and muscovite used in the experiment were provided by Dulan Jinhui Mining Co., Ltd., Golmud, Qinghai province, China. First, the mineral samples were manually selected and crushed, and then placed in a ceramic ball mill for dry grinding. Finally, dry screening was performed to obtain −74 + 45 μm arsenopyrite and −38 μm muscovite for flotation experiments and RPM testing. The mineral samples were ground to −5 μm for electrokinetic potential FTIR and XPS detection. The X-ray diffraction patterns of arsenopyrite and muscovite were verified in our previous experiments to confirm the purity for requirements [5] because the XRD patterns of the two minerals are very similar to their PDF cards; they prove to be of higher purity. The arsenopyrite and muscovite samples were mixed at a mass ratio of 1:1 to produce artificially mixed minerals.

Rin En Technology Development Co., Ltd., Shanghai, China, provided the TSPP and STPP (analytically pure, AR) used in experiments. The compounds dissolved easily in water. Sulfuric acid (H₂SO₄) and sodium hydroxide (NaOH) (all AR, and from Sinopharm Group Chemical reagent Co., LTD, Shanghai, China) were used to adjust the pH. Terpineol (AR) was used as frother, and isoamyl xanthate (IAX, technical pure) was provided by Dulan Jinhui Mining Co., Ltd. Deionized water (18.25 MΩ·cm) was used for all tests.

2.2. Flotation Tests

The XFGII-type laboratory flotation machine with a 60 mL cell was used in microflotation. Pure mineral particles (2.0 g) and 38 mL of deionized water were mixed in the flotation cell, and the rotation speed was adjusted to 1900 RPM. After stirring for 2 min, the pH regulators, desired amount of dispersant, collector (8 × 10⁻⁵ mol/L), and frother (1 × 10⁻⁴ mol/L) were added sequentially into the slurry, with 3 min of stirring after each addition. After being filtered and dried, the concentrate and tailings were collected. For pure mineral flotation, the flotation recovery was calculated based on mass balance. For flotation of artificial mixed minerals (mass ratio of arsenopyrite and muscovite 1:1), the recovery was calculated based on the solid weight ratio and As grade. Three flotation tests were conducted under the same experimental conditions, and their average values were reported. The tests were carried out in an air atmosphere.

2.3. Reflection Polarizing Microscope Tests

RPM was used to research the coagulation and dispersion behaviors of arsenopyrite and muscovite. The suspension was prepared by combining 1.0 g of arsenopyrite and 1.0 g of muscovite in 50 mL of distilled water. The mixture was stirred for 30 min and then air-dried. The resulting suspension was examined using a reflection polarizing microscope (BK-POL1810158, Chongqing Aote Optical Instrument Co., LTD, Chongqing, China), and photos were taken.

2.4. Electrokinetic Potential

An amount of 30 mg of pure minerals was added to 50 mL of KCl (1 mM) to obtain suspension, which was conditioned through magnetic agitation. While stirring, the desired pH regulators, dispersants, and collector were added separately to the beaker in sequence and conditioned for 10 min. After stopping the stirring, the mixture was allowed to precipitate for at least 30 min. The electrokinetic potential was measured using the Delsa 440sx Electrokinetic potential Analyzer (Malvern Instruments Ltd., Malvern, UK). The average of at least three independent experiments and the standard deviation of parallel results were calculated and presented.

2.5. FTIR Measurements

The KBr diffuse reflectance method and IRAffinity-1 Fourier transform infrared spectrometer (Shimadzu, Japan) were used to obtain the spectra of mineral samples before and after treatment with flotation reagents. Ultrasound was performed on 30 mg of pure mineral sample (−5 μm) in 100 mL of deionized water. The reagents were then added in the same order that they were added in the flotation tests (excluding collector and frother). The pulp was stirred at 25 °C (pH = 6) with a magnetic agitator for 30 min, then centrifuged at 4500 rpm. The precipitation was washed with deionized water 3 times and dried in a vacuum oven. In the end, the dry mineral samples were collected for infrared detection at room temperature. The wave number of the spectrum ranges from 400 to 4000 cm⁻¹. Each spectrum was recorded with 20 scans at a resolution of 4 cm⁻¹.

2.6. XPS Detection

The XPS spectra for arsenopyrite and muscovite particles with and without treatment by dispersants at the same concentration used in flotation were recorded with a K-Alpha 1063 (Thermo Scientific Co., Waltham, MA, USA) spectrometer, which employs Al Kα as a sputtering source at 12 kV and 6 mA with 1.0 × 10⁻⁹ Pa pressure in the analytical chamber. The C 1s peak served as a reference to binding energy (BE) for uncharged hydrocarbon at 284.8 eV, and binding energies in all other spectra for that sample corrected for this shift [14]. The quantification and curve fitting of the spectra were determined using Thermo Scientific Avantage software (version 5.5, Thermo Scientific Co., Waltham, MA, USA).

3. Results and Discussion

3.1. Single-Mineral Flotation

Figure 1 shows the results of single-mineral flotation, which is used to study the effect of dispersant type and dosage on the flotability of arsenopyrite and muscovite. In the previous exploration experiments, the optimum dosage of the collector was determined as 8 × 10⁻⁵ mol/L and the frother was 1 × 10⁻⁴ mol/L.

As shown in Figure 1, the addition of TSPP had a slight effect on the recovery of arsenopyrite. Even with the dosage of TSPP increased to 1 × 10⁻⁵ mol/L, the recovery of arsenopyrite could still be maintained above 70.0%. However, when adding 2 × 10⁻⁵ mol/L STPP into slurry, the recovery of arsenopyrite decreased to below 40.0%; the recovery was only 28.6% (STPP = 6 × 10⁻⁵ mol/L), which was in sharp contrast to the recovery of 81.4% without dispersants. STPP had a depressing effect on arsenopyrite, which was inconsistent with the original intention of the study. At the same time, the addition of dispersants had a slight effect on the recovery of muscovite. It was worth noting that the recovery of arsenopyrite was 81.4% and muscovite was only 9.5% when no dispersants were added. The recovery difference was 71.9%, which seems to indicate that flotation separation could be achieved between the two minerals.

3.2. Artificial Mixed-Mineral Flotation

Artificial mixed flotation experiments are conducted under the premise of keeping other conditions unchanged to research the influence of STPP and TSPP on flotation. The experiment results are shown in Figure 2. From Figure 2a, we could clearly reach the conclusion that, in artificial mixed flotation, the difference of recovery of minerals becomes smaller, especially since the value is only about 19% (Ph = 6), compared to the value of 71.9% in single-mineral flotation. Although at pH = 2, the difference in recovery is 60%, there is a risk of serious corrosion of the equipment. This reflects that muscovite has a significant adverse effect on the flotation of arsenopyrite. The overall trend from the curve indicates that with the amount of TSPP increase, the recovery of arsenopyrite exhibited an upward trend, while the recovery of muscovite decreased significantly. This indicates that TSPP can enhance the recovery of arsenopyrite in slurry containing fine muscovite. The recovery of arsenopyrite has not shown a significant upward trend with the increase in STPP dosage, while the downward trend in muscovite recovery is similar to TSPP. Therefore, STPP does not improve the flotation of arsenopyrite in the presence of muscovite. The phenomenon shows that TSPP can improve the flotation of arsenopyrite more effectively than STPP.

3.3. RPM Detection

A reflective polarizing microscope is used to study the RPM images of pure mineral and artificial mixed flotation concentrate (before and after the addition of dispersants, pH = 6), as shown in Figure 3.

In Figure 3a, the black granular minerals represent arsenopyrite. As shown in Figure 3b, the transparent flake minerals are muscovite. The RPM images of arsenopyrite/muscovite artificial mixed flotation concentrate are shown in Figure 3c,d. Comparing Figure 3c,d with Figure 3a, it is evident that during artificial mixed flotation, the surface of arsenopyrite is covered by muscovite, which forms an aggregate structure [15,16]. The hydrophilic muscovite adhering to the arsenopyrite’s surface resulted in a decrease in the recovery of arsenopyrite. Additionally, floating arsenopyrite entrains muscovite, which increases the recovery of muscovite. When TSPP (Figure 3e) and STPP (Figure 3f) are added separately to the slurry as dispersants, distinct phenomena occur. When using TSPP as a dispersant, muscovite that was originally adsorbed on the surface of arsenopyrite dispersed. However, when STPP is added to artificial mixed flotation, muscovite is almost not dispersed. The results show that the dispersion effect of TSPP is significantly more effective compared to STPP, which is consistent with the flotation results.

3.4. Electrokinetic Potential

The results of microflotation and artificial mixed flotation show that TSPP and STPP had significantly different effects on the recovery of arsenopyrite and muscovite. To investigate the mechanism, the changes in the electrokinetic potential of minerals are studied in the presence of different reagents. The results are shown in Figure 4. It is evident that both arsenopyrite and muscovite are negatively charged across the entire pH range.

The electrokinetic potential of arsenopyrite and muscovite is consistent with previous research [17,18]. However, studies have shown that coagulation may occur despite both minerals being negatively charged [19,20]. Figure 4a indicates an electrokinetic potential change diagram of muscovite under different conditions. It can be observed that the electrokinetic potential value of muscovite/TSPP is consistently lower than that of muscovite/STPP. At pH = 8, the electrokinetic potential change value of muscovite/TSPP is −17.37 mV, which is significantly lower than muscovite/STPP (−8.33 mV). This suggests that TSPP adsorbed onto muscovite more strongly than STPP. Simultaneously, both TSPP and STPP did not influence the adsorption of IAX, for the electrokinetic potential of muscovite/TSPP/IAX, muscovite/STPP/IAX, and muscovite/IAX are very close.

Figure 4b shows the electrokinetic potential change of arsenopyrite under different conditions. The overall trend shows that the potential value of arsenopyrite/STPP is lower than arsenopyrite/TSPP, indicating that the adsorption of STPP is stronger than TSPP. Combined with the results of microflotation, the adsorption of STPP is believed to have caused the decrease in the recovery of arsenopyrite. TSPP could better adsorb to muscovite, but its adsorption on arsenopyrite is weaker than STPP. The results of the electrokinetic potential further verify that TSPP is more suitable as a dispersant in the arsenopyrite/muscovite system.

3.5. FTIR Detection

To gain a deeper understanding of how dispersants are absorbed on the surface of minerals, FTIR is conducted to investigate the adsorption mechanism. The FTIR of minerals is tested before and after treatment with different reagents, and the results are shown in Figure 5 and Figure 6.

As shown in Figure 5a, the bending vibration peak of -OH appears at 3624 cm⁻¹ in muscovite. For muscovite/TSPP, the peak has shifted to 3619 cm⁻¹, while for muscovite/STPP, the corresponding peak position is 3614 cm⁻¹. This phenomenon indicates that regardless of which dispersant is added, the -OH is involved in the adsorption, demonstrating that hydrogen bond adsorption takes place. Figure 5b shows a partial enlargement of Figure 5a. In Figure 5b, the peaks at 824 and 799 cm⁻¹ correspond to the stretching vibration peak of Si-O-Si [21]. When muscovite is treated with TSPP, the peaks shift to 830 and 801 cm⁻¹. When muscovite is treated with STPP, the peaks shift to 826 and 795 cm⁻¹. The peaks at 746 and 690 cm⁻¹ correspond to the stretching vibration peak of Si-O-Al. After muscovite is treated with TSPP, the peaks shift to 750 and 694 cm⁻¹. When muscovite is treated with STPP, the peaks shift to 748 and 685 cm⁻¹. At the same time, following the dispersant treatment, new peaks appeared in the FTIR of muscovite. New peaks appeared in the FTIR of muscovite/TSPP at 861 and 634 cm⁻¹, corresponding to the stretching vibration absorption peak of P=O and the bending vibration absorption peak of O-P-O [22]. The bending vibration absorption peak of P-O at 631 cm⁻¹ is observed for muscovite/STPP [23]. This phenomena indicate that both TSPP and STPP chemically adsorb onto muscovite.

As shown in Figure 6a, the bending vibration peaks of -OH are observed at 3426 and 1636 cm⁻¹, while 1066 cm⁻¹ corresponded to SO₄²⁻ [24]. The -OH vibration peaks of arsenopyrite/TSPP shift to 3438 and 1636 cm⁻¹, while the SO₄²⁻ peak shifts to 1093 cm⁻¹. Simultaneously, a stretching vibration absorption peak of P=O is observed at 1193 cm⁻¹, and an O-P-O bending vibration absorption peak is observed at 1135 cm⁻¹. This indicates that chemical adsorption occurred with TSPP on arsenopyrite. The peak position of the -OH vibration peak in arsenopyrite/STPP shifts to 3463 and 1637 cm⁻¹, while the peak position of the SO₄²⁻ peak shifts to 1178 cm⁻¹. The bending vibration absorption peak of P-O appeared at 998 cm⁻¹ (as shown in Figure 6b), indicating that STPP also demonstrated chemical adsorption onto arsenopyrite.

3.6. XPS Detection

To investigate the valence bond on the surfaces of muscovite and arsenopyrite before and after treatment with TSPP and STPP, XPS detection is conducted in this study. The results are shown in Figure 7, Figure 8, Figure 9 and Figure 10.

As shown in Figure 7a, the binding energies of Al-OH and Al-O in muscovite are measured to be 73.57 and 74.14 eV [14]. The binding energies of Al-OH and Al-O in muscovite/TSPP are measured to be 73.69 and 73.89 eV, with offsets of 0.12 and 0.25 eV. This indicates that the addition of TSPP results in adsorption with the Al atoms on the surface of muscovite (as shown in Figure 7b). The binding energy positions of the two groups in muscovite/STPP shifted to 73.63 and 73.99 eV, with offsets of 0.06 and 0.15 eV (as shown in Figure 7c). This indicates that TSPP is more readily adsorbed with Al atoms on the surface of muscovite. Figure 8 shows that the deviation of the Si binding energy is extremely small in both TSPP- and STPP-treated muscovite, clearly indicating that Si atoms do not participate in the reaction between minerals and reagents.

Figure 9 shows the spectrum of arsenopyrite Fe2p before and after treatment [25]. As shown in Figure 9a, the peaks at 712.93 eV and 710.89 eV represent Fe(III)-SO and Fe(III)-OH. The peaks at 707.10 eV and 706.35 eV represent Fe(II)-(AsS). The spectrum of Fe2p after treatment with TSPP is shown in Figure 9b. At this stage, Fe(III)-SO shifted to 711.49 eV, Fe(III)-OH shifted to 709.34 eV, and Fe(II)-(AsS) shifted to 707.45 and 707.05 eV, indicating that TSPP adsorbed Fe atoms on the surface of arsenopyrite. The spectrum of Fe2p after treatment with STPP is shown in Figure 9c. Fe(III)-SO shifted to 712.33 eV, Fe(III)-OH shifted to 711.04 eV, and Fe(II)-(AsS) shifted to 707.50 eV and 707.16 eV, indicating that STPP also adsorbed Fe atoms on the surface of arsenopyrite.

Figure 10 shows the spectrum of arsenopyrite As3d before and after treatment. As shown in Figure 10a, 44.53 eV represents As(V)-S, 43.24 eV represents As(III)-O, and 40.68 eV represents As(-I)-S. The spectrum of As3d after treatment with TSPP is shown in Figure 10b. At the time, As(V)-S shifted to 44.73 eV, As(III)-O shifted to 43.27 eV, and As(-I)-S shifted to 41.08 eV. It indicates that STPP was adsorbed along with As atoms on the surface of arsenopyrite. The spectrum of As3d after treatment with STPP is shown in Figure 10c. As(V)-S shifted to 45.17 eV, As(III)-O shifted to 43.70 eV, and As(-I)-S shifted to 41.32 eV, indicating that STPP is adsorbed along with As atoms on the surface of arsenopyrite.

In a word, TSPP and STPP can adsorb muscovite, but TSPP is more readily adsorbed on the surface of muscovite. At the same time, in the microflotation experiment, dispersants have different degrees of inhibitory effect on arsenopyrite because dispersants can react with Fe and As atoms on the surface of arsenopyrite.

3.7. Suggested Adsorption Model

Figure 11 shows the possible interface interaction model between dispersants and the surface of minerals. In artificial mixed flotation, the flotability of arsenopyrite is poor, while the flotability of muscovite is high because muscovite can be adsorbed on the surface of arsenopyrite, hindering the adsorption of collector. When TSPP is added as a dispersant, the muscovite that covered the surface of arsenopyrite can be dispersed, improving the recovery of arsenopyrite. When STPP is used as the dispersing agent, its dispersing effect on muscovite is poor, reducing its effectiveness in improving arsenopyrite flotation. Therefore, TSPP can better eliminate the adverse effects of muscovite on the surface of arsenopyrite compared to STPP. It is reasonable to use TSPP as a dispersant for arsenopyrite/muscovite in actual flotation.

4. Conclusions

In the flotation of pure mineral arsenopyrite and muscovite, it is found that the recovery between them is different, and separation can be achieved by flotation. However, in the flotation of artificial mixed flotation, the adsorption of muscovite on the surface of arsenopyrite prevents the adsorption of collector. When adding dispersants, TSPP can better recover flotation index. By comparing RPM, electrokinetic potential, FTIR, and XPS, it is found that dispersants can all adsorb Al atoms in muscovite through P-O groups, but the adsorption of TSPP is significantly stronger than STPP. At the same time, the flotation experiment shows that STPP has a certain degree of inhibition on arsenopyrite. The phenomenon indicates that TSPP is more suitable for dispersing muscovite adsorbed on the surface of arsenopyrite than STPP.