The Effect of an Anionic Polyacrylamide on the Flotation of Chalcopyrite, Enargite, and Bornite

Abstract

Water recycling in mining is essential to decrease water usage, which results in the accumulation of high concentrations of inorganic and organic substances in the process water. Consequently, adverse impacts on the flotation process of copper sulfides may arise. High-molecular-weight polymers based on anionic polyacrylamides (PAMs) are used as tailing flocculants in mineral processing plants. The recirculation of water recovered from the tailing thickeners to the flotation process introduces residual PAMs, which can impact the flotation of important copper sulfides like chalcopyrite, bornite, and enargite. This issue has been rarely studied. In this work, results on the effect of an anionic polyacrylamide (PAM) of medium–low anionicity on the flotation of chalcopyrite, enargite, and bornite are reported and analyzed. The results show that PAM molecules depress the flotation of chalcopyrite, enargite, and bornite under a wide range of pH values. The experimental data indicate that the depressing effect of PAMs on copper sulfides increases with pH. The zeta potential results reveal that this parameter becomes less negative with the addition of PAMs, indicating interactions between PAM molecules and the surfaces of the copper sulfides. PAM adsorption on copper sulfides increases with pH, which correlates with the flotation and zeta potential data. It is proposed that the interactions between PAM molecules and copper sulfides are explained by the presence of surface iron and copper hydroxides that create chemically active adsorption sites.

1. Introduction

The importance of water quality in flotation highlights the necessity for the mining sector to adopt more effective and sustainable practices. This involves improving the design and implementation of water systems to reduce the negative effects of dissolved species on concentration processes [1,2,3,4,5]. The mining industry already incorporates water recycling and the use of seawater (either raw or desalinated), but this practice presents challenges such as increased concentrations of inorganic and organic components in the process water, which directly affect the flotation process [1,2,3,4,5,6,7].

High-molecular-weight polymers based on anionic polyacrylamides (PAMs) are frequently used as tailing flocculants in mineral processing plants [8,9]. The amide is the primary functional group of PAMs and forms molecules via polyaddition processes [10]; the resulting molecules need to be hydrolyzed and modified to be used for industrial purposes, producing acrylic units in the PAM chain containing usually carboxylic, sulfonic, and amine groups that render PAMs anionic or cationic; these ionic units are responsible for molecule stretching and the bridging action on the tailing particles [8]. PAMs’ molecular conformation is significantly affected by the physicochemical properties of the solution they are dissolved in. Additionally, the adsorption and effectiveness of PAMs as flocculants rely on the interactions between the polymer and minerals, as well as on factors such as molecular weight, the degree of anionicity (DA), and molecular conformation [11,12]. It is also known that applying excessive mechanical shearing to these high-molecular-weight flocculants causes chain breakage, significantly diminishing their bridging and flocculating capabilities [5,13].

Although the use of PAMs in mineral processing plants is aimed at improving the settling rates in the thickening stages, it is known that residual flocculants in the recycled water may affect the flotation of different species. Water recovered from the tailing thickeners is recirculated to the milling and flotation stages, bringing with it residual PAMs that may influence the surface properties of copper and molybdenum sulfides. It is important to note that PAMs can be degraded in various mechanical, chemical, thermal, and biological ways [14], which influences flotation, and that in processing plants, these molecules may be strongly degraded by the mechanical action of the pumps. Previous studies showed that tailing flocculants have a significant depressing effect on the floatability of molybdenite, which is an important by-product of the copper mining industry because of the high price of molybdenum [3,4,5,15,16]. It was demonstrated that anionic PAMs have a depressing effect on molybdenite flotation, which is influenced by factors such as pH, mechanical degradation, the degree of anionicity, and molecular weight [3,4,5]. Previous studies indicated that non-mechanically degraded PAMs have a strong depressing effect on molybdenite flotation, and, in contrast, strongly degraded PAMs showed no effect. It was also demonstrated that the depressing effect of PAMs of a low DA decreases at high pH levels, which suggested that the PAM–molybdenite repulsive forces outweigh the attractive forces arising between the active groups of the polymer and the hydroxylated metal sites on the faces of molybdenite; in contrast, more anionic PAMs showed stronger effects at high pH levels.

PAMs can also significantly inhibit coal flotation by altering the interfacial characteristics of the solid/liquid and liquid/gas interfaces [17]. It was also reported that diaspore flotation was negatively affected by a cationic PAM in the selective separation of this mineral from kaolinite using dodecyl amine as a collector [18]. Some modified PAMs showed different effects on metal sulfides [19]; it was previously reported that modified PAMs (hydroxamic, carboxylated, sulfonated, hydroxylated, and modified by xanthation) can interact with pyrite through chemical bonding and depress its flotation [18,19,20]; it was also informed that cationic PAMs can be used as depressants to separate kaolinite and diaspore using flotation with dodecyl amine as a collector [21].

Studies on the effects of anionic PAMs on the flotation of relevant copper sulfides such as bornite, enargite, and chalcopyrite are scarce. In particular, the depressing effect of PAMs on bornite and enargite has not been reported, and this is the main novelty of this work. In this paper, results of the depressing effect of a medium-DA PAM on the flotation of relevant copper sulfides such as bornite, enargite, and chalcopyrite are presented. Copper sulfide flotation is assessed using microflotation, as a function of pH, and PAM concentrations. Zeta potential and adsorption measurements are used to study the PAM–mineral interactions.

2. Methodology

2.1. Materials and Reagents

The enargite (Cu₃AsS₄) mineral sample was obtained from the Quiruvilca mine in Peru while chalcopyrite (CuFeS₂) and bornite (Cu₅FeS₄) were supplied by Ward’s Natural Science Establishment. The three mineral samples were initially crushed manually to achieve particle sizes ranging from −2 to +1.4 mm, and subsequently manually concentrated to eliminate visually noticeable impurities. Subsequently, a laboratory high-intensity induced roll magnetic separator was employed to eliminate magnetic minerals, followed by desliming with a 600# screen to remove ultra-fines. X-ray diffraction (XRD) using Bruker^® D4 Endeavor equipment (Bruker, Billerica, MA, USA)with Cu radiation and Ni Kβ radiation filter was used to analyze the minerals. XRD analyses indicate that enargite, chalcopyrite, and bornite samples were of 99.8%, 99.5%, and 99.7% purity, respectively. Enargite contained minor amounts of tennantite while the chalcopyrite and bornite samples reported low concentrations of pyrite and quartz.

The carboxylated PAM tested in the study was obtained from SNF-Chile (99.99% purity as indicated by the company, Santiago, Chile) and was identified as SNF2350. The PAM sample was presented in dry white granules and was tested in the study with no treatment and/or purification. According to the provider, the PAM flocculant was described as anionic and had a molecular weight of approximately 16 × 10⁶ Da. Previous characterization of the same PAM sample reported a DA of 11.9%, which indicates that the polymer sample was a medium-DA PAM. It must be noted that the number of hydrolyzed units of PAMs defines the DA, i.e., the more the moles of acrylate ions in the polymer chain there are, the higher DA is. In industrial practice, PAMs with DA values above 13–14% are typically classified as anionic whereas those below 10–11% are categorized as having low anionicity. Therefore, the PAM used in this study was classified as having medium anionicity.

Potassium amyl xanthate (PAX) obtained from Solvay (Brussels, Belgium) was purified using ether and acetone and used as collector. Methyl isobutyl carbinol (MIBC) from Merck (Darmstadt, Germany) was used as frother. Milli-Q water of a resistivity of 18.4 MΩ∙cm at 25 °C was used in all the experiments. pH was adjusted with sodium hydroxide (NaOH). The aqueous medium for the experiments was 0.01M NaCl.

2.2. Procedure

2.2.1. Microflotation

Mineral flotation was evaluated through microflotation experiments conducted in a 150 mL Partridge and Smith glass cell following the procedure described in Figure 1. Initially, the copper sulfides in the −2 + 1.4 mm size fraction were wet-ground using a lab carbon steel ball mill (7 cm diameter and 12 cm length) charged with 270 g of carbon steel balls (12 balls of 13 mm diameter; 8 balls of 17 mm diameter). Using wet grinding is important as it mimics the redox conditions obtained in the industrial application of copper sulfide flotation. The microflotation experiments considered 3 g of the mineral samples previously ground to a P80 of 74 μm at a solid content of 25% by weight. The grinding times to obtain the required P80 were 5.2 min for chalcopyrite, 5.35 min for bornite, and 5.05 min for enargite. The grinding product was split into 3 samples later conditioned over 3 min in 150 mL of solution (0.01M NaCl) to adjust pH to the required condition (pH values ranging 7–11, which are typical in industrial practice) using NaOH. Later, PAX and MIBC were added at concentrations of 25 and 60 mg/L and conditioned for 3 min. Finally, PAM was added at concentrations between 0 and 10 mg/L and conditioned for additional 3 min, after which the resulting pulp was transferred to the cell to be floated using 80 mL/min of nitrogen. In this study, nitrogen was employed to prevent the additional oxidation that might occur if air were used as the flotation gas. Following the opening of the gas valve, the froth was removed every 10 s for a duration of 2 min. Tailings and concentrates were then oven-dried at 105 °C for 8 h, and mineral recoveries were calculated by dividing the mass of the concentrate by the total mass of the concentrate plus tailings. All the tests were performed in triplicates and average values are reported with the error bars indicating one standard deviation.

2.2.2. Zeta Potential Measurements

The interactions between the mineral particles and PAM molecules were assessed via zeta potential measurements using a Zetacompact Z9000 (CAD instrument, Paris, France). This device utilizes video analysis to monitor the motion of fine particles induced by an applied electric field. The number of particles observed varied, ranging from a few hundred in closely packed samples to several thousand in evenly dispersed suspensions. A quantity of 10 mL of suspension, prepared with a solid content of 0.02% and containing particles smaller than 20 µm, was introduced into the instrument’s quartz cell. All the tests were conducted using a 0.01M NaCl solution. The experiments covered pH values from 2 to 11 to observe a wide range of behaviors.

2.2.3. Adsorption Tests

PAM adsorption on copper sulfide minerals was assessed following the total organic carbon (TOC) in solution as measured using a TOC-L Shimadzu equipment. In the experiments, 1 g of copper mineral was dispersed over 15 min in 150 mL of PAM solutions prepared at known varying concentrations (0–26 mg/L PAM) of the polymer, different pH values, and at 20 °C, which was controlled with a thermostatic bath. The experiments were conducted using a 0.01M NaCl solution. The resulting suspensions were centrifuged at 2000 rpm for 5 min and filtered, and the liquid streams were analyzed for TOC, from which the concentrations of PAM in solution were obtained from a calibration curve. The quantity of PAM adsorbed on copper sulfides in mg/g was calculated from the PAM-concentration differences in the initial (C_i) and final (C_f) solutions multiplied by the volume of solution (150 mL). All these experiments were performed in triplicate with an average standard error of 3%. The Langmuir [22] isotherm model was fitted to the experimental data to describe the experimental results.

3. Results

3.1. Microflotation

Figure 2 shows the recovery of chalcopyrite, enargite, and bornite as a function of pH and PAM concentration. The results show that the PAM depresses the flotation of the three copper minerals tested in this work. Moreover, the results indicate that the depressing effect of the PAM on the copper sulfides increases with pH.

3.2. Zeta Potentials

Figure 3 shows the zeta potentials of copper sulfides as functions of pH and PAM concentration. The results show that the zeta potentials become less negative with the addition of the PAM, which indicates interactions between the PAM molecules and the surfaces of the copper sulfides.

3.3. Adsorption Isotherms

Figure 4 shows the equilibrium adsorption isotherms of the PAM on chalcopyrite, enargite, and bornite at pH values of 8, 9, and 10. The results show that as the PAM concentration increases, so does the extent of adsorption, reaching a saturation point that suggests predominantly monolayer adsorption [23]. The isotherms also indicate that adsorption increases with pH, a trend that was documented in previous studies on different systems and which correlates with the data presented in Figure 2 [23,24].

The Langmuir [22] isotherm model was used to describe the experimental results. The linearized Langmuir isotherm model is presented in Equation (1).

$$\frac{1}{{\mathrm{q}}_{\mathrm{e}}}=\frac{1}{{\mathrm{q}}_{\mathrm{m}}}+\frac{1}{{\mathrm{K}}_{\mathrm{L}}.{\mathrm{q}}_{\mathrm{m}}}\frac{1}{{\mathrm{C}}_{\mathrm{e}}}$$

(1)

Here, q_e is the quantity of the PAM (mg) adsorbed per g of mineral at equilibrium, q_m is the value of q_e at the saturation point, C_e is the equilibrium concentration (mg/L), and K_L is a Langmuir adsorption isotherm constant (L/mg) that relates to adsorption energy. From a plot of 1/q_e vs. 1/C_e, the constants q_m and K_L can be determined from the intersection and slope of the straight line, respectively.

Table 1. Langmuir model constants.

Chalcopyrite	pH = 8	pH = 9	pH = 10
qm, mg/g	6.67	2.08	2.33
KL, L/mg	0.009	0.048	0.050
Enargite	pH = 8	pH = 9	pH = 10
qm, mg/g	1.89	2.44	2.63
KL, L/mg	0.099	0.075	0.074
Bornite	pH = 8	pH = 9	pH = 10
qm, mg/g	1.64	1.25	1.47
KL, L/mg	0.127	0.372	0.335

shows the Langmuir model constants, and Figure 5 shows the linearization of the model at the three pH values tested.

4. Discussion

PAMs are hydrophilic molecules of high molecular weight that are used to flocculate fine gangue particles in mineral processing plants. PAMs also affect the surface properties of valuable minerals and strongly modify the physicochemical properties of mineral suspensions [3,4,5,25,26]. PAM adsorption on mineral particles can occur because of the physical and chemical interactions occurring between the polymer molecules and the solid surfaces, i.e., electrostatic, dipole–dipole, London–Van der Waals, and hydrophobic forces; chemical interactions through chemical and coordination bonds, and hydrogen bonds also, lead to the attachment of PAMs onto mineral particles, in which case the polymers must surpass the electrostatic repulsion between the polymer and mineral surfaces [27,28,29]. Previous studies also showed that ions dissolved in the aqueous medium neutralize the functional groups of the PAM molecules, leading to polymer coiling and less adsorption on the mineral particles [4,30].

The surfaces of ground copper sulfides are believed to display hydrophobic patches formed by sulfur-based entities and hydrophilic patches of hydroxides compounds [31]. The extent to which hydrophobic patches outweigh the hydrophilic ones determines their floatability, and therefore, less hydrophobic copper sulfide surfaces at highly alkaline conditions create adsorption sites for PAMs. The balance of repulsive and attractive interactions that develops between PAM molecules and copper sulfides defines the extent of polymer adsorption. Repulsive forces originate from the electrostatic interactions between the anionic molecules and the negative surface charge of copper sulfides. Attractive interactions arise due to the presence of surface iron and copper hydroxides that create chemically active adsorption sites on copper sulfides. As can be seen from the zeta potential results presented in Figure 3, repulsive forces become stronger at high pH levels. However, the increased presence of metallic adsorption sites, particularly iron and copper hydroxides on the copper sulfide surfaces, is more significant. This accounts for the intensified adsorption and depressing impact of PAMs observed at higher pH values, as depicted in Figure 2 and Figure 4. The experimental results indicate that at high pH values, the electrostatic repulsion between the polymer and mineral surfaces is surpassed, which induces more adsorption [26,27,29]. PAM adsorption on mineral surfaces also depends on molecule conformation, which strongly depends on the ionization of the carboxylic groups of PAMs. At elevated pH levels, carboxylic groups become highly ionized [3,4], enhancing the interaction of PAM molecules with iron and copper hydroxides that form on metal sulfide surfaces, thereby inducing flotation depression. The experimental data show that the depressing effect of the PAM on copper sulfide flotation (Figure 2) and the specific adsorption (Figure 4) increase as pH increases, which supports the proposed mechanism. The zeta potential results presented in Figure 3 also indicate that the interactions between PAM molecules and the surfaces of copper sulfides become stronger at higher pH values.

The effects of PAM molecules on chalcopyrite, enargite, and bornite flotation, zeta potentials, and adsorption trends presented in Figure 2, Figure 3 and Figure 4 are similarly significant in scale. Chalcopyrite is a copper sulfide that exhibits natural hydrophobicity at low pH and under a range of redox potential of 100–500 mV/NHE [32,33]. The natural floatability of chalcopyrite decreases with pH, causing the potential range to narrow due to the formation of iron and copper hydroxides on the surface sites on which anionic PAM molecules tend to adsorb causing low recoveries. The enargite surface properties in flotation were previously investigated [34,35,36,37,38]. It was reported that enargite flotation is affected by the pH and pulp redox potential, i.e., collectorless flotation is good at low pH and high E_h levels, which is explained by the generation of surface elemental sulfur, and flotation is reduced at high pH levels because of the presence of surface-oxidized copper–iron–arsenic compounds. The experimental results presented in this work suggest that PAM-molecule adsorption on enargite results from the presence of these oxidized compounds. Bornite is a mineral that readily oxidizes during grinding, which affects its surface properties, causing low recoveries in flotation [39,40]. The presence of these oxidized surface compounds explains PAM adsorption.

5. Conclusions

• PAMs depress the flotation of chalcopyrite, enargite, and bornite under a wide range of pH values. The experimental data indicate that the depressing effect of PAMs on the copper sulfides increases with pH levels.
• The results of zeta potentials show that this parameter becomes less negative with the addition of PAMs, which indicates interactions between the PAM molecules and the surfaces of the copper sulfides.
• PAM adsorption on copper sulfides increases with pH levels, which correlates with the flotation and zeta potential data.
• It is proposed that the interactions between PAM molecules and copper sulfides are explained by the presence of surface iron and copper hydroxides that create chemically active adsorption sites.
• The results obtained in this study were generated with pure samples and in a low-ionic-strength solution. It is recommended to address these limitations in future studies.