Kinetic Study of Manganese Oxidative Precipitation Reaction by Using SO₂/Air Gas Mixture

Abstract

Removing manganese from zinc electrolytes is necessary to pave the way for replacing lead-based anodes with mixed metal oxide (MMO) anodes. MMO anodes offer significantly lower overpotential towards oxygen evolution reactions, thus are attractive from an energy consumption viewpoint. Previous studies had shown that, thanks to the catalytic effect of cobalt, manganese can be removed successfully from the zinc purification solution through the oxidative precipitation method using a simulated roasting off-gas plant. This study focuses on understanding the primary mechanism behind manganese oxidation precipitation and investigating the influence of various operating parameters such as temperature, dissolved oxygen (DO), and solution potential on the reaction kinetics. The results revealed that the kinetics of the reaction was highly dependent on the temperature and catalyst activity rather than on the reactant concentration. Additives, with radical scavenging effects, were added to identify the radicals responsible for the oxidation of Mn. The manganese oxidation reaction was dramatically suppressed when methanol was added. However, in the presence of tert-butyl alcohol (TBA), a sensible reduction in manganese removal was not observed, suggesting sulfate radical as the predominant species for oxidizing manganese. The physical and chemical characteristics of the sediments were also presented.

1. Introduction

Various electrochemical methods are used to recover metals in hydrometallurgy including electrowinning and electrorefining. Electrowinning involves the reduction of metal ions from a solution onto a cathode, resulting in the formation of a solid metal deposit. Electrorefining is used to purify an impure cast anode while electrowinning is used when the metal is in an already dissolved form [1,2]. Nowadays, the majority of metallic zinc is obtained by the electrowinning process since the raw material is zinc sulfide, which is converted into zinc oxide by roasting, then purified through a series of hydrometallurgical processes (Figure S1). Electrowinning is therefore a suitable process to recover the dissolved zinc from the purified liquor. In this process, lead–silver alloys are used as anode materials, posing several environmental challenges, including lead pollution and high power consumption [3]. The corrosion of lead anodes in highly acidic environments is unavoidable, leading to considerable anode slime generation [4,5]. The anode slime is mainly composed of manganese dioxide, with lead contamination as a hazardous waste [6,7]. Lead-based anodes also impose a nearly 30% extra power requirement due to the high overpotential of the oxygen evolution reaction (OER), which is heavily dependent on the structure and composition of the anode material [8,9]. Such high energy costs, along with environmental pressure for carbon neutrality, have stimulated significant interest in searching for a suitable substitution for lead anodes [10].

Mixed metal oxide (MMO) anodes could be an alternative to lead-based anodes in zinc cell houses to address the high OER overpotential. MMO anodes are made of a titanium substrate, coated by mixed metal oxides (TiO₂, RuO_2, and IrO₂), having good electrocatalytic activity towards OER [11,12]. However, manganese ions in zinc electrolytes seriously limit the practical applications of MMO anodes in zinc plants. During the electrowinning process, manganese dioxide is created on the surface of MMO anodes, resulting in the poisoning of its catalytic sites [13]. The efficient and economical removal of manganese ions from the zinc purification solution is therefore the first step toward using MMO anodes in zinc plants.

Various technologies such as ion exchange, adsorption, solvent extraction, and precipitation have been introduced to remove manganese from solutions with high manganese concentrations. However, these methods are limited by high cost, low removal efficiency, potential risks, etc. The ion exchange method suffers from a lack of selectivity for removing manganese ions. It increases the cost of the operation due to the consumption of extra reagents for eliminating other ions present in the zinc purification solution [14,15]. The cost of the chemical agents is significant, making it economically less viable [15]. Carbonate and hydroxide precipitation methods are not sufficient for removing manganese from zinc purification solutions due to the significant co-precipitation of zinc along with manganese, particularly when the pH exceeds 8 [16].

Advanced Oxidation Processes (AOPs), involving highly reactive oxidants such as peroxymonosulfate (${\mathrm{S}\mathrm{O}}_5^{•-}$), hydroxyl (${\mathrm{O}\mathrm{H}}^{•-}$), and sulfate radicals (${\mathrm{S}\mathrm{O}}_4^{•-}$), have gained significant attention for their effectiveness in degrading and mineralizing stubborn organic and inorganic compounds [17,18]. Zhang et al. [19,20] studied the reaction mechanism of iron oxidation by SO₂/O₂ in acidic media. They showed that the radical chain mechanism plays a crucial role in iron oxidation by the formation of peroxymonosulfate radicals (${\mathrm{S}\mathrm{O}}_5^{•-}$) followed by peroxymonosulfate anions (${\mathrm{H}\mathrm{S}\mathrm{O}}_5^{-}$), which are responsible for the oxidation of iron and sulfite species. This mechanism was verified by quenching tests and the addition of hydroquinone (${\mathrm{C}}_6{\mathrm{H}}_6{\mathrm{O}}_2$) as a radical scavenger to the system, resulting in a considerable reduction in the oxidation rate [19,20].

Typically, ${\mathrm{S}\mathrm{O}}_4^{•-}$ can be generated by activating peroxydisulfate (PDS) or peroxymonosulfate (PMS) by heating, ultraviolent radiations, ultrasound, and the addition of transition metals [21,22,23]. High-energy inputs, required for most of these treatments, pose challenges for practical applications. In contrast, the utilization of transition metals as catalysts (Co²⁺, Mn²⁺, Fe²⁺, Ni²⁺, Cu²⁺) was shown to facilitate the homogeneous activation of persulfate (PS) or peroxymonosulfate (PMS) [24,25].

The activation process of peroxymonosulfate anion is described by the following equations (Equations (1)–(3)). Sulfate and hydroxyl radicals are formed by the homolytic cleavage of peroxide bond (-O-O-) in peroxymonosulfate structures [26,27].

$${\mathrm{H}\mathrm{S}\mathrm{O}}_5^{-}\to {\mathrm{S}\mathrm{O}}_4^{•-}+{\mathrm{O}\mathrm{H}}^{•}$$

(1)

$${\mathrm{H}\mathrm{S}\mathrm{O}}_5^{-}+{\mathrm{M}}^{\mathrm{n}+}\to {\mathrm{S}\mathrm{O}}_4^{•-}+{\mathrm{M}}^{\left(\mathrm{n}+1\right)+}+{\mathrm{O}\mathrm{H}}^{-}$$

(2)

$${\mathrm{H}\mathrm{S}\mathrm{O}}_5^{-}+{\mathrm{M}}^{\mathrm{n}+}\to {\mathrm{S}\mathrm{O}}_4^{2-}+{\mathrm{M}}^{\left(\mathrm{n}+1\right)+}+{\mathrm{O}\mathrm{H}}^{•}$$

(3)

Some scavenging tests, known as chemical probes, can determine whether hydroxyl radicals and/or sulfate radicals dominate the reaction rates. In this method, conventional alcohols such as ethanol (EtOH), methanol (MeOH), and tert-butyl alcohol (TBA) are introduced to the solution [28,29,30]. The oxidation rate of metallic ions is expected to be suppressed in the presence of these alcohols. Ethanol is capable of scavenging both sulfate (1.6 $\times {10}^7-7.7\times {10}^7 {\mathrm{M}}^{-1}{\mathrm{S}}^{-1}$) and hydroxyl radicals (1.2 $\times {10}^9-2.8\times {10}^9 {\mathrm{M}}^{-1}{\mathrm{S}}^{-1}$) with high rates, whereas TBA reacts with hydroxyl radicals (3.8 $\times {10}^8-7.6\times {10}^8 {\mathrm{M}}^{-1}{\mathrm{S}}^{-1}$) with a higher rate than with sulfate radicals (4.0 $\times {10}^5-9.1\times {10}^5 {\mathrm{M}}^{-1}{\mathrm{S}}^{-1}$) [31]. In fact, TBA reacts with hydroxyl radicals approximately 1000 times faster than that with the sulfate radical. Therefore, TBA is usually recognized as a hydroxyl radical scavenger [32]. On the other hand, EtOH is known as a sulfate and hydroxyl radical scavenger with similar rates.

The oxidative precipitation method using a SO₂/air gas mixture, due to its high level of accessibility, and cost–benefit considerations, has received significant attention for removing manganese from aqueous solutions. Zhang et al. studied the kinetics of manganese oxidation with SO₂/O₂ from nickel/cobalt leach liquors within the pH range of 1–6 and a temperature range of 25–80 °C [33]. Menard et al. [34] investigated the effect of agitation and impeller configuration on the gas transfer kinetics. However, the literature on manganese elimination from zinc electrolytes is quite rare. Although the SO₂/O₂ mixture seems to be efficient for removing manganese, the high cost of this synthetic gas is prohibitive for industrial deployment. Askarian et al. [35] recently showed that the off-gas of the roasting plant, containing (7–9)% SO₂ and around 10% of O₂, could be utilized as an oxidant to reduce manganese from 1000 mg/L to 1 mg/L in zinc purification solution. They showed that, by taking advantage of 10 mg/L cobalt as a catalyst, present in the zinc purification solution as an impurity, the maximum manganese removal was obtained at a SO₂/air ratio of 6% at 80 °C (resembling diluted off-gas of the roasting plant) [35]. The main goals of this study are to examine the kinetics of the manganese oxidation reaction and to investigate the effect of some operating parameters, i.e., temperature, dissolved oxygen, and solution potential, on the manganese removal process and to identify the dominant radicals involved in the reactions for better understanding of the reaction mechanisms.

2. Materials and Methods

2.1. Materials and Reagents

Synthetic zinc-rich solutions were prepared with demineralized water. All chemicals were of analytical research grade and used as purchased without further purification. Sulfuric acid with a purity of (95–98)% was purchased from Fisher brand, and nitric acid with a purity of 70% was obtained from Anachemia.

Manganese sulfate pentahydrate (MnSO₄·5H₂O) and cobalt sulfate heptahydrate (CoSO₄·7H₂O) were used as sources of manganese and cobalt and obtained from Laboratoire MAT INC and Sigma-Aldrich Company Ltd., Burlington, MA, USA, respectively. Potassium hydroxide (KOH), used as a neutralizing agent to maintain the solution pH constant, was purchased from Sigma-Aldrich Company Ltd., USA. Activated zinc oxide was used as the source of zinc to prepare feeding solutions. Methanol (CH₄O) and tert-butyl alcohol (C₄H₁₀O) were used as radical scavengers and provided by Fisher-Scientific and Sigma-Aldrich Company Ltd. USA, respectively. Air (99.9%) and sulfur dioxide (99.98%) were provided by PRAXAIR (Mississauga, ON, Canada).

2.2. Apparatus

A schematic representation of the Mn removal set-up is shown elsewhere [35]. Two different mass flow controllers were used to control the flow rate of air and SO₂ mass flow controllers were provided by POLYCONTROLS (models SLA5850S1BBC6C2A1 and 58505, respectively) (Brossard, QC, Canada). A BROOKS instrument model 0254 (POLYCONTROLS, Brossard, QC, Canada) was used as a power supplier and set point controller for the air and SO₂ mass flow controllers. SO₂ and air were mixed through a T-conjunction and introduced to the reactor by a single 316-stainless steel sparger.

The reactor was made of a 3000 mL water-jacketed glass beaker with four 316 stainless steel baffles symmetrically mounted near the cell wall. Agitation was provided by a Caframo mixer motor model BDC 3030 (Caframo Lab Solutions, Georgian Bluffs, ON, Canada) with two different 316 stainless steel impellers, six blades, and a diameter of 50.8 mm. A Rushton impeller (model #6BR-3) and a 45° upward pitched blade impeller (model #ATI-3) were provided by INDCO Inc., New Albany, OH, USA. A water bath Isotemp^TM model 6200 H7 by Fisherbrand (Waltham, MA, USA) was connected to the jacketed glass beaker to circulate hot water and maintain a constant temperature during the experiments. A Cole-Parmer combination pH electrode and Ag/AgCl refillable reference electrode were accommodated into the reactor to monitor the solution pH continuously. The pH signals were processed by an OAKTON pH meter model 700 (Quebec, QC, Canada). A Sensorex double-junction oxidation–reduction potential (ORP) prob model 3000 was applied to measure the potential of the solution (Garden Grove, CA, USA). It features a platinum electrode with a built-in Ag-AgCl/3.5KCl-sealed reference. A VisiFerm Arc optical dissolved oxygen sensor coupled with the Arc Air software analyzer version 3.9.2 was used to monitor continuously the concentration of dissolved oxygen (DO) provided by Hamilton Company, Reno, NV, USA.

2.3. Methods

The jacketed glass beaker was filled with 1300 mL of the synthetic zinc-rich solution containing 150 g/L zinc, 1 g/L manganese, and 10 mg/L cobalt. The pH was adjusted with some drops of sulfuric acid to reach a pH of around 4. According to our previous results [35], the SO₂/air gas mixture with the volumetric gas ratio of 6% SO₂ and 94% air was injected through the sparger. The solution was agitated at a rotation speed of 2000 rpm under atmospheric pressure. A potassium hydroxide solution (20 wt.%) was added to the reactor throughout the experiments to neutralize the sulfuric acid produced during the test and to keep the pH constant at around 4 in all experiments. Aqueous samples were taken periodically during the test. Temperature, OPR, DO, and pH were recorded during the whole experiment. These conditions were applied for all experiments unless otherwise stated.

2.4. Analysis and Characterization

The manganese concentration was measured using Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) (Perkinelmer, Waltham, MA, USA, Avio 220 Max) coupled with the Syngistix analyzer software version 5.1. Aqueous samples were taken during the test and filtered through 0.2 μm pore-size wool glass. Then, 1 mL of the transparent filtered solution was diluted 100 times with ${\mathrm{H}\mathrm{N}\mathrm{O}}_3$ 5 vol.% and was kept in the sealed tube to prevent contaminations. The morphology of precipitates was examined by a field-emission gun scanning electron microscope (FEG-SEM) (FEI, Inspect F50, Hillsboro, OR, USA). The elemental analysis of sediments was performed by energy-dispersive X-ray spectrometry (EDS) (Edax Ametek, Octane Super-A, Warrendale, PA, USA) coupled with SEM. The working voltage for this analysis was 15 kV. The SEM analyses were performed in secondary electron (SE) mode.

To prepare a solid phase sample, the precipitates were allowed to settle spontaneously at the bottom of the cell for 24 h without a clarifying agent. Next, the sediments were washed multiple times with hot distilled water (80 °C) to remove any zinc ions from the sediments and were filtered by filter paper (WhatmanTM was purchased from Sigma-Aldrich Company Ltd. Burlington, MA, USA) with 125 mm diameter and 0.45-micron size). Finally, the obtained particles were placed in an oven at 70 °C for 24 h. The oven operated without any protective atmosphere. X-ray photoelectron spectroscopy (XPS) analyses were carried out using a PHI 5600-ci spectrometer (Physical Electronics, Eden Prairie, MN, USA). A monochromatic aluminum X-ray source (1486.6 eV, 300 W) was used to record the survey spectra (1400–0 eV) with a pass energy of 187.85 eV at a high vacuumed atmosphere (${10}^{-10}\mathrm{T}\mathrm{o}\mathrm{r}$). High-resolution (HR) spectra were obtained using an achromatic magnesium X-ray source (1253.6 eV, 300 W) with a pass energy of 5.85 eV. No charge neutralization was applied for the survey and HR spectra. Moreover, HR curve fittings were determined using the least-squares method, Gauss–Lorentz functions with a Shirley background subtraction. The powder X-ray diffraction (XRD) analyses were conducted using the X-ray diffractometer model Aeris, manufactured by Pan Analytical.

The radiation source was Cu-Kα ($\mathsf{\lambda }=1.541 \mathrm{A})$ at 40 kV and 40 mA operating from 2$\mathsf{\theta }$ angle of 10°–80°. The diffraction patterns were analyzed by the software X’Pert HighScore (Pan Analytical version 4.8).

3. Results and Discussion

A series of experiments were conducted to investigate the effect of various operating factors, such as temperature, solution potential, and dissolved oxygen concentration, on manganese removal kinetics. Some radical scavengers, namely methanol and tert-butyl alcohol, were added to the solution as chemical probes to identify the predominant radicals responsible for manganese oxidation. Table S1 (Supplementary Information) presents a chronological description of the experiments conducted in this study.

3.1. Effect of Temperature on the Kinetics of the Reaction

It has already been demonstrated that cobalt is an effective catalyst to accelerate the manganese oxidation reaction [35]. It has also been reported that cobalt could be co-precipitated within the manganese dioxide structure and be removed from the system during the manganese oxidation reaction. In this context, we also aim to reveal the effect of temperature on Mn and cobalt removal as well as to clarify the role of cobalt as a catalyst of the Mn removal reaction.

Manganese concentration as a function of time at different reaction temperatures is shown in Supplementary Materials (Figure S2). These experiments were conducted within the temperature range of 30 °C to 80 °C, maintaining a pH of 4 and a SO₂/air ratio of 6 vol.%. The initial concentration of Mn was targeted at 18 mmol/L (1000 mg/L) during solution preparation, and each electrolyte was analysed before processing to obtain the actual initial concentration. It can be seen that the rate of Mn removal, the slope of the curves, increases with temperature. The reaction rates are 0.4, 0.5, 0.6 and 0.7 mmol/s for reaction temperatures of 30, 50, 70 and 80 °C. In addition, one can see that the rate of Mn removal is constant at a given temperature. As the concentration of Mn varies with time but the rate of Mn removal remains constant, it suggests that the Mn removal rate is independent of the Mn concentration. In other words, the Mn removal reaction is a zeroth-order reaction.

The linear parts of the Mn concentration vs. time are plotted in Figure 1a and allow for calculating the reaction rates (k) at each temperature. Figure 1b presents the Arrhenius plot for four temperatures. A good linear correlation calculates the activation energy of the reaction as 9.6 kJ/mol.

Figure 2a,b show, respectively, the concentrations of dissolved oxygen and cobalt as a function of time. The concentration of oxygen, as a reactant, at a given temperature is constant. This was expected since oxygen is continuously supplied to the system so that its concentration depends only on the temperature. The concentration of the dissolved oxygen decreases from 0.50 mmol/L (8 mg/L) at ambient temperature to approximately 0.11 mmol/L (1.86 mg/L) at 80 °C. Although the concentration of dissolved oxygen, as a reactant, decreases by increasing temperatures, the Mn removal rate is higher at elevated temperatures. This observation suggests that the effect temperature on the Mn removal rate outweighs the impact of the dissolved oxygen concentration. The concentration of cobalt also decreases monotonically with time and its removal rate increases with temperature. The removal of cobalt was also expected since it has been demonstrated that it coprecipitates with Mn, being trapped within the Mn structure [36,37].

Previous results [35] showed that the Mn removal rate is too small when cobalt is not present in the system, while adding a small amount of cobalt accelerates the Mn removal (Supplementary Materials Figure S3). Our results also suggest that cobalt acts as a catalyst for the Mn removal reaction, resulting in zeroth-order reactions. However, the rate of Mn removal does not corelate with the cobalt concentration. Once again, at a given temperature, the cobalt concentration decreases with time while the rate of Mn removal is constant. This may suggest that cobalt does not directly act as catalyst, since otherwise the reaction rate would have increased with its concentration, but possibly through the generation of intermediate catalytic species whose concentration does not vary with cobalt concentration.

Cobalt ions have the highest catalytic activity for peroxymonosulfate (PMS) activation, resulting in the creation of highly reactive oxygen species such as hydroxyl (${\mathrm{O}\mathrm{H}}^{•-}$) and sulfate radicals (${\mathrm{S}\mathrm{O}}_4^{•-}$) [25,26,31], which can affect the reaction rate of manganese removal reactions. Therefore, we designed a series of experiments to identify the possible reactive oxygen species responsible for reactions.

3.2. Solution Potential in Mn-Co-SO₂-Air System

The presence of reactive oxygen species can affect the solution potential. To investigate the effect of manganese and cobalt ions on the solution potential and determine at which point manganese oxidation precipitation was completed, we continuously monitored the solution potential during experiments. We designed two distinct experiments for this purpose. One test was conducted in the absence of manganese and cobalt ions. The other one was carried out with the presence of both, with the pH adjusted to 4 and the temperature maintained at 80 °C. Figure 3 depicts the solution potential with respect to Ag/AgCl and the percentage of precipitated manganese over a 60 min experiment. Without the presence of manganese and cobalt ions, it was observed that the solution potential remained relatively constant, averaging around 340 mV throughout the entire test duration. However, the solution potential notably increased when we introduced manganese and cobalt ions under the same conditions. The solution potential is significantly higher in the presence of cobalt and manganese. This higher potential could be attributed to the possible formation of oxidizing species such as ${\mathrm{S}\mathrm{O}}_3^{–},{\mathrm{S}\mathrm{O}}_5^{–},{\mathrm{H}\mathrm{S}\mathrm{O}}_5^{–}, {\mathrm{S}\mathrm{O}}_4^{2–}, {\mathrm{S}\mathrm{O}}_4^{–},{{\mathrm{S}}_2\mathrm{O}}_6^{2–}$ due to the catalytic effect of cobalt ions in the system [38]. Mulaudzi and Mahlangu [39] reported also a correlation between the redox potential and SO₂ concentration at pH 3 and found that the redox potential is highly dependent on the SO₂ concentration in the solution, determining the redox potential and oxidizing power for manganese removal. At the beginning of the experiment and upon the introduction of SO₂/air, the solution potential increases rapidly from 540 mV to 650 mV, then stays at this level for 25 min. This period corresponds to the manganese removal from the solution at a constant rate (zeroth-order reaction). At this stage, oxidizing species are generated and consumed by manganese at the constant rate, thus the equilibrium potential of the solution reaches approximately 650 mV. When the manganese removal is completed, and all manganese is precipitated (after 25 min), the solution potential increases again to 735 mV and stays constant. This observation suggests that there is still enough cobalt in the solution to generate oxidizing species and their equilibrium concentration results in a constant solution potential of 735 mV.

3.3. Determination of Principal Reactive Radicals

As previously discussed, Co ions could generate strong oxidizing agents such as hydroxyl $\left({\mathrm{O}\mathrm{H}}^{•–}\right),\mathrm{s}\mathrm{u}\mathrm{l}\mathrm{f}\mathrm{a}\mathrm{t}\mathrm{e}{(\mathrm{S}\mathrm{O}}_4^{•–}),$ and peroxymonosulfate radical (${(\mathrm{S}\mathrm{O}}_5^{•–}$) [26]. To determine whether sulfate and/or the hydroxyl radical play a dominant role in manganese removal, we introduced some radical scavengers into the reaction system. Methanol (MeOH) and tert-butyl alcohol (TBA) were selected due to their distinct reactivity rates with ${\mathrm{O}\mathrm{H}}^{•–} \mathrm{a}\mathrm{n}\mathrm{d}{\mathrm{S}\mathrm{O}}_4^{•–}$ radicals. Methanol quenches both sulfate and hydroxyl radicals at almost the same rate. However, TBA is recognized as a hydroxyl radical scavenger due to its high-rate reaction with hydroxyl compared to sulfate radicals [31]. In other words, these alcohols compete with manganese and cobalt ions to react with the radical species.

First, we added different concentrations of scavengers in order to find the concentrations at which the effect of scavenger on the reaction rate is observed (Figure S4). Based on these preliminary tests, 100 mmol/L of each scavenger was chosen for further analyses since their effect was measurable. Figure 4a illustrates the influence of 100 mmol of MeOH and TBA on the reaction kinetics at 6% SO₂/air ratio and a constant temperature of 30 °C. According to Figure 4a, the addition of TBA did not change the rate of the Mn removal reaction (rate constant = 0.33 mmol/L.s) compared to the sample without scavenger (rate constant = 0.35 mmol/L.s). However, the addition of 100 mmol of MeOH significantly suppressed the Mn removal reaction (rate constant = 0.21 mmol/L.s); even after 50 min of reaction, 6.3 mmol/L of manganese still remained in the final solution. In other words, TBA, which is recognized as a strong ${\mathrm{O}\mathrm{H}}^{•–}$ scavenger, has no effect on the reaction compared to MeOH, which scavenges both ${\mathrm{O}\mathrm{H}}^{•–} \mathrm{a}\mathrm{n}\mathrm{d} {\mathrm{S}\mathrm{O}}_4^{•–}$. This observation confirms that the main radical responsible for Mn oxidation is ${\mathrm{S}\mathrm{O}}_4^{•–}$.

Figure 4b indicates that the cobalt removal rate was not affected by scavengers, suggesting that none of the scavengers significantly affected the rate of cobalt removal. At the end of the experiments, there was still around 0.10 mmol/L of cobalt present in the solutions. These results suggest that while cobalt plays a critical role as a catalyst in the radical chain mechanism, its removal is not affected by the generation of the oxidizing species. This finding supports our previous hypotheses that cobalt ions become trapped within the manganese dioxide structure and precipitate from the solution, as also reported elsewhere [35].

3.4. Characterization of the Precipitates

3.4.1. Scanning Electron Microscopy Analysis

The effect of temperature on the morphology of precipitates was investigated at a SO₂/air ratio of 6% without a radical scavenger. Figure 5a–c¹ and

Table 1. EDS elemental analysis of precipitates (wt.%) of SEM images (a–c).

Precipitate	Mn	O	Zn	S	K	Co
30 °C	53.16	24.28	17.22	4.08	0.94	0.32
50 °C	44.84	29.45	18.94	5.07	1.49	0.21
80 °C	44.30	21.40	29.54	3.88	0.58	0.29

depict SEM images at two different magnifications and a corresponding EDS elemental analysis of generated precipitates at various operating temperatures after 40 min. According to the EDS results (Table 1), these particles are mainly composed of Mn, O, and Zn, verifying the formation of manganese dioxide particles with the coprecipitation of other cations such as zinc, cobalt, and potassium. It is worth mentioning that potassium came from potassium hydroxide, which was added to the system as a neutralizing agent to control the pH. The EDS spectra of precipitates, collected at different temperatures, are shown in Figures S5–S7, Supplementary Materials.

Based on the SEM micrographs (Figure 5a–c), the morphology of the precipitates is cauliflower-like shape with no remarkable discrepancy at different temperatures in low magnification (10 kx). In contrast, at higher magnification (Figure 5a¹–c¹), one can see that the fraction and size of porosities were enhanced by raising the temperature from 30 °C to 80 °C. For example, there was no visible porosity in the precipitates collected at 30 °C (Figure 5a¹), while more porosity was generated (red circles) at 80 °C, which was turned into a sponge-like shape (Figure 5c¹). Additionally, the entrapment of cobalt ions within the manganese dioxide precipitates could be attributed to their sponge-like structure at elevated temperatures. These data correspond with data presented in Figure 2b and the ICP results of the solutions, showing that by increasing the temperature from 30 °C to 80 °C, the cobalt removal is improved, and that the final concentration was reduced from 0.09 mmol/L (5 mg/L) to 0.02 mmol/L (1 mg/L), respectively.

3.4.2. X-ray Diffraction Analysis

Figure 6 shows the XRD patterns of the precipitates at two different temperatures. The diffractogram of the precipitates obtained at a lower temperature (30 °C) was weak, indicating poor crystallinity or the formation of nanoscale crystals. Four broad peaks at 2$\mathsf{\theta }$ positions of around 12.32°, 25.21°, 37.83°, and 65.89° were observed, which can be well assigned to the (001), (002), (11$\overline{1}$), (021) planes of a layered birnessite-type δ-MnO₂ (JCPDS #13-0105) [40]. At a higher temperature (80 °C), the intensity of the peaks is stronger, indicating a higher degree of crystallinity; one additional peak emerged at a 2θ position of 34.68°, corresponding to ($\overline{2}$01). This peak, together with the peak at 25.23°, correlate to (002) and the crystallographic planes of potassium birnessite hydrate (K_0.23MnO₂(H₂O)_0.7, JCPDS# 80–1098) [41]. Ma et al. [42] pointed out that the diffraction pattern showing the (001) plane of as-δ-MnO₂ reveals the layered structure characteristic of birnessite MnO₂. Additionally, the (002) reflection is attributed to the presence of hydrated cation layers (in this case, potassium) inserted between the (001) layers. It is worth noting that pure δ-MnO₂ (JCPDS: 18-0802) does not exhibit a reflection at 2θ∼25°. This observation suggests that the incorporation of cations, i.e., ${\mathrm{K}}^{1+},{\mathrm{C}\mathrm{o}}^{2+}{,\mathrm{Z}\mathrm{n}}^{2+}$, in the structure of MnO₂ is most likely easier at 80 °C than that at 30 °C.

3.4.3. X-ray Photoelectron Spectroscopy

XPS analysis was performed on the precipitates collected at 30 °C and 80 °C to examine the effect of temperature on the surface elemental composition and chemical states of the precipitates. Figures S8 and S9 (Supplementary Materials) present the survey spectra, showing the peaks related to Mn 2p, Co 2p, S 2p, Zn 2p, and O 1S, supporting the EDS analysis results.

Figure 7a,b present high-resolution spectra of ${\mathrm{M}\mathrm{n} 2\mathrm{p}}_{3/2}$ at 30 °C and 80 °C, respectively. ${\mathrm{M}\mathrm{n} 2\mathrm{p}}_{3/2}$ was deconvoluted at (642.5–642.6) eV and 645.1 eV, and both were attributed to ${\mathrm{M}\mathrm{n}}^{4+},$ indicating the formation of the MnO₂ and MnSO₄ compound, respectively [35,43]. Figure 7c,d depict the high-resolution XPS spectra of O 1s at 30 °C and 80 °C, respectively. At both temperatures, the O 1s high-resolution spectra were further deconvoluted into two peaks at (530.2–530.3) eV and (532.5–532.6) eV, which correspond to the Mn-O-Mn bond for the tetravalent oxide and O=S−O bond energy for sulfate, respectively [35].

The high-resolution spectra of ${\mathrm{Z}\mathrm{n} 2\mathrm{p}}_{3/2}$ at 30 °C (Figure 7e) exhibit only a single chemical bond at 1021.5 eV, which could be ascribed to zinc in combination with O₂- in Zn=O lattice [44,45]. At 80 °C, the high-resolution spectrum of ${\mathrm{Z}\mathrm{n} 2\mathrm{p}}_{3/2}$ (Figure 7f) was deconvoluted at 1021.6 eV and 1023.9 eV; the former is similar to what was observed at 30 °C (Zn=O), and the latter belongs to the Zn-OH compound, which may be arisen from the addition of KOH as a neutralizing agent [46]. These results show that zinc also co-precipitates along with the Mn structure in the form of ZnO. This zinc should be within the MnO₂ crystal, since no XRD peak related to a separate ZnO phase was detected. The presence of Zn-OH peaks in the sample treated at 80 °C suggests that zinc is also present in the form of ZnOH in the precipitates at 80 °C. This zinc hydroxide is most likely related to the electrolyte, trapped in the porous structure of these precipitates, which upon drying, is converted to ZnOH.

Overall, it can be concluded that the chemical compositions of the precipitates at both temperatures are more or less similar. Notably, the zinc content (in atomic percent) of precipitates increased from 5.9% to 13.8% (Figures S8 and S9, Supplementary Information) by increasing the temperature from 30 °C to 80 °C, which is in good agreement with EDS measurements. The morphology of the precipitates at 80 °C was more porous and enhanced the accommodation of cations inside the manganese dioxide structure.

4. Conclusions

Our goal in this work was to investigate the kinetics parameters of reactions for Mn removal from the zinc electrowinning solution and reveal the effect of a few process parameters on the efficiency of the process. An SO₂/air gas mixture of 6% (volume ratio of SO₂) was used as an oxidating agent. This gas mixture, which mimics the off-gas of the roasting plant, had already been optimized for this reaction.

Our results showed that the temperature is an important parameter for the manganese oxidation precipitation reaction. It favors reaction kinetics and accelerates the manganese elimination. By evaluating the reaction rates at four different temperatures, we calculated the activation energy of the reaction to be 9.6 kJ/mol.

We demonstrated that the manganese oxidation reaction rate, in the presence of cobalt, is independent of the concentration of manganese in the solution. This kinetic behavior indicates a zeroth-order reaction. The results also showed that Co acts as catalyst, indirectly controlling the Mn removal reaction rate through generating radicals. In the presence of cobalt, radicals are formed, being the main responsible species for manganese oxidation. Based on the radical quenching test results, we revealed that sulfate radicals are mainly responsible for manganese removal when an SO₂/air gas mixture is used as an oxidating agent.

The XRD patterns confirmed the precipitation of birnessite MnO₂ at both low and high temperatures. SEM images demonstrate that the morphology of manganese dioxide particles is cauliflower-like, with the porous structure at an elevated temperature, and available cations (Co²⁺, Zn²⁺, K⁺¹) can be trapped in their layered structure. The high resolution XPS analysis of Mn 2p and O 1s evidenced the presence of manganese dioxide in the precipitates, which agrees with the other characteristic analysis (EDS and XRD).