Separation of manganese from aqueous solution using an emulsion liquid membrane

Abstract

The present study deals with the carrier-facilitated transport of manganese from this acidic leach solution in an emulsion liquid membrane system. The ELM consists of MDEHPA as the extractant, industrial solvent as the inert diluent, non-ionic polyethylene glycol as a surfactant and sulfuric acid as the stripping solution. The main parameters influencing the ELM stability and the extraction of manganese are the pH of the acidic leach solution, mixing speed, concentrations of the stripping solution, extractant and surfactant, phase ratio and treatment ratio. With respect to the parameters mentioned, the optimum conditions for the process were determined. The results demonstrated the possibility of 93% extraction of manganese by ELM from the acidic leach solutions at optimum operational conditions.

---

Introduction

Manganese is associated with nickel in laterite waste effluents¹ and has industrial applications in the manufacture of various chemicals, dry cell batteries, catalysts, adsorbents, plant fertilizers, animal feed, glass, ceramics, paints as well as numerous medicinal purposes. In particular, within the steel industry, manganese has distinctive properties compared to other metals with respect to hardness, corrosive resistance, etc.² The industrial waste effluents containing manganese are good potential sources of this valuable metal. Therefore, due to economic and environmental considerations, the recovery of manganese from these streams is of interest.³

Solvent extraction is a common method of metal extraction that has been extensively used in hydrometallurgical operations in the treatment of wastewaters to remove soluble metals. However, metal extraction with an emulsion liquid membrane is an attractive process due to its high selectivity and it could be used for a dilute metals system instead of solvent extraction.^4,5 This process is technically favorable because of its low energy consumption and negligible losses of organic liquid. For the quantitative recovery of manganese and the co-recovery of other associated metal ions, the emulsion liquid membrane process may offer the cheapest and the most suitable technique.^6–13 Emulsion liquid membranes are typically prepared by emulsifiying the internal phase of an immiscible liquid and then dispersing this emulsion in a third liquid phase (called the external phase); the internal and the external phases are miscible, but both are immiscible with the membrane phase.¹⁴

In facilitated transport, the carrier promotes solute transfer through the membrane. The carrier used in ELM can be regenerated after its reaction with the inner reagent at the interface between the membrane and the internal phases, so it diffuses to the outer interface of the ELM and the cycle is restarted.¹⁵ The carrier for metal ion recovery may be acidic, basic or chelating.¹⁶ Some reported organic extractants for the extraction of metal ions by ELM are D2EHPA, Cyanex® 272, Cyanex® 302, LIX® 63, LIX® 860 and Aliquat® 366.¹⁷ The extraction of silver ions from neutral and weakly acidic solutions by D2EHPA, Cyanex® 272, Cyanex® 302, LIX® 63 has been reported in the literature.^18–22 Extraction and separation of Ag(I), Mn(II), Zn(II) and Co(II) using various extractants such as D2EHPA, tri-n-butyl phosphate and Cyanex® 272, Cyanex® 302 from different solutions have been reported (Ritcey and Lucas, 1971; Devi et al., 1997). MDEHPA, a mixture of mono and di(2-ethylhexyl) phosphoric acid, showed superior performance due to its strong chelating capability and less sterically hindered sites.^14,23 However, so far there is no work reported on extraction and separation of manganese ions with MDEHPA as an extractant, even though economically MDEHPA is cheaper than D2EHPA. For quantitative recovery of manganese and co-recovery of other associated metal ions, the emulsion liquid membrane process may offer the cheapest and the most suitable technique.^12,13

As far as we know, no work has been conducted on the extraction of manganese from the aqueous phase using a mixture of mono and di(2-ethylhexyl) (MDEHPA) as the extractant in an ELM system. In this research, extraction of manganese from an aqueous solution by an emulsion liquid membrane with MDEHPA as a carrier diluted with an industrial paraffinic solvent and polyethylene glycol (PEG) as a non-ionic surfactant was performed. The external phase was the source of manganese.

Hence, the transport of manganese ions occurred from the external to the internal phase. The parameters affecting the transport of manganese ions through the ELM, such as surfactant concentration, phase ratio, stripping phase concentration, carrier concentration and treatment ratio were studied to determine the most suitable process conditions among those investigated. Moreover, a real industrial wastewater sample – the effluent of a steel plant (Esfahan’s Mobarakeh Steel Co., Iran) – was employed for the extraction of manganese ions.

Experimental

Apparatus

The ELM experiments were carried out in a cylindrical glass container of 105 mm diameter and 145 mm height equipped with four round glass baffles with 8 mm diameter and equally spaced in the tank. The impeller for the extraction experiments was made of Teflon with a 95 mm diameter and had three pitched blades with 45° angles (Fig. 1).

Fig. 1 Dimensions of the preparation and processing tanks.

For experiments with the emulsion system, a water in oil (W/O) emulsion was prepared using organic and strip solutions using a high shear homogenizer (Omni Programmable Digital Homogenizer (PDH) model, USA) with variable speed (1000–28 000 rpm). An electric driven agitator, model Heidolph RZR 2020, Germany, with variable speed from 50–6000 rpm was used in the experiments (Fig. 2).

Fig. 2 A scheme of the experimental set up.

In order to maintain a stable milky white W/O emulsion, a sonicator (Elmasonic Germany, model: P60H) was used. The samples were taken at pre-determined time intervals by disposable sanitary syringes and centrifuged by a laboratory centrifuge (ALC PK130 Bench Top Centrifuge) thus separating the primary emulsion phase from the external phase. For the determination of the concentration of manganese ions in the aqueous samples, an atomic absorption spectrometer (Perkin Elmer model PinAAcle 900T) with deuterium background corrector equipped with a 10 cm long slot-burner head and an air–acetylene flame was used. Viscosity determinations were made on a programmable rheometer, Brookfield Model DV-III, at room temperature. The pH values of the aqueous solutions were set by a digital pH meter (model Metrohm 780).

Reagents

Span 80 and polyethylene glycol (PEG) were supplied by Fluka and Marun Petrochemical Co. (Iran) respectively. PEG is a by-product of the EO/EG plant of Marun Petrochemical Company. The chemical composition of PEG is shown in Table 1.

Table 1 The compositions and viscosities of PEG

Molecular weight of PEG (g mol^−1)	PEG	Viscosity, cP
20 000	%74	—
12 000	%22	—
6000–10 000	%6	1960
PEG (mixture)	—	2850

---

The latter was used as a non-ionic surfactant for stabilizing the emulsion. MDEHPA is a mixture of 55% di-2-ethyl hexylphosphoric acid ester (D2EHPA) and 45% mono-2-ethyl hexylphosphoric acid ester (M2EHPA). This was purchased from Sigma-Aldrich and used as the mobile carrier without further purification. MIPS, an industrial solvent supplied from Marun Petrochemical Co. (Iran), was used as an inert diluent for the extraction, with a consistent formulation, mainly of paraffinic and naphthenic hydrocarbons in the range C₁₀–C₁₄. The industrial solvent (MIPS) was mainly paraffin. A stock solution of manganese (1000 mg L⁻¹) was prepared by dissolving 3076 mg of manganese sulphate in 1000 mL of demineralized water. The source phases for various experimental runs were prepared from this stock solution by dilution with water up to the desired concentration. Manganese sulphate, sulfuric acid, sodium hydroxide and other chemicals were of analytical grade and bought from Chem-Lab NV, Belgium. Demineralized water (conductivity < 0.5 μS cm⁻¹ and pH = 5.5–6.0) was used for preparation of the solutions.

Diluent

The diluent plays an important role in making an emulsion liquid membrane. The importance of the correct diluent choice for an extraction processes has been long recognized²³ and since the membrane organic phase consists of an extractant and a diluent, the role of the diluent on the transport and in the membrane stability becomes the most important factor. Cheapness, negligible loss, good solubility and reduced toxicity are characteristics of a good diluent. In addition, some of the other properties that should be considered when selecting a suitable diluent are specific gravity, viscosity, flash point and polarity. The most common diluents are n-hexane, kerosene, cyclohexane, benzene, toluene, carbon tetrachloride, and chloroform.^24,25

In general, aromatic materials have higher specific gravities than aliphatic materials, which could inhibit dispersion and coalescence of the solvent. At high metal loading of the solvent, the difference between the specific gravity of the loaded solvent and the aqueous phase is small, which might present problems in separating the two phases. The density of the solvent usually increases as the concentration of extractant is increased so the density of the diluent is a factor that could be taken into consideration.²³

Polarity is one of the most important properties of a diluent. The extraction of metal ions decreases with an increase in the polarity of the diluent. Furthermore, the interaction between diluent and extractant can result in lower extraction efficiency. Therefore diluents with lower dielectric constants are suggested for the highest extraction. Likewise, aliphatic hydrocarbons are less hydrophilic than aromatics. The solubility of these solvents in water is typically on the order of one to five parts per million. Aliphatic diluents are generally preferred compared to aromatic ones due to the polarity. The extraction of metals declines with an increase in the polarity of the diluent. Interaction of the diluent with the extractant can result in a lower extraction coefficient for metal ions. MIPS, a locally produced diluent, was selected owing to its aliphatic character, which provides low volatility. The chemical composition of MIPS is shown in Table 2.

Table 2 Specifications of the industrial paraffinic solvent

Parameter	Value	Unit
<n-C10	Max 0.5%	% mol
n-C10+ to n-C11	Min 40%	% mol
	Max 50%
n-C12+ to n-C13	Min 43%	% mol
	Max 56%
>n-C13	Max 1.5%	% mol
Aromatic and water	Max 100	mg kg^−1
Total sulfur content	Max 1	mg kg^−1
Kinematic viscosity	1.96 @ 20 °C	cSt
Density	0.83	g mL^−1
Dielectric constant	1.93	—

---

Preparation of the ELM

The internal aqueous standard solution was prepared by dissolving the required amount of H₂SO₄ in demineralized water. The organic phase was prepared by dissolving the required quantity of carrier (MDEHPA) and non-ionic surfactants (PEG, Span 80) in the MIPS solvent. To prepare a stable W/O emulsion under stirring (7000 rpm) conditions, the internal or stripping phase (aqueous H₂SO₄ solution) was added to the organic phase to obtain the required volume ratio of organic phase to internal phase. The blended solution with a volume of 50 mL was sonicated for 1–2 h until a stable milky white W/O emulsion was formed. During the sonication process, the temperature was controlled below 30 °C to avoid de-emulsification. The prepared emulsion was dispersed into the continuous (source) phase with a proper stirring arrangement. The organic membrane phase was prepared by dissolving the appropriate amount of PEG or Span 80 and MDEHPA in MIPS gently mixed by a magnetic stirrer. A pre-determined volume ratio of 1 : 1 was maintained for the organic phase to the internal stripping phase.

Feed solution preparation

The steel plant effluent (Esfahan’s Mobarakeh Steel Co., Iran) consisted of 1045 ppm Fe, 970 ppm Mn, 134 ppm Ni, 113 ppm Cd, 98 ppm Zn, 47 ppm Co and 40 ppm Cu. In order to obtain a relatively rich solution in this process, with the exception of manganese, cadmium, and nickel, the other metal ions in the acidic leach solution were gradually precipitated by adding various reagents and adjusting the solution pH. After adding sulphuric acid to adjust the pH of the feed phase, re-analysis of the feed phase concentration showed that apart from manganese, cadmium, iron and nickel, the concentrations of other metal ions in the acidic leach solution were changed. Moreover, cadmium was selectively separated out by the emulsion liquid membrane (ELM) process. The decrease of the concentrations of Fe, Cu, Co, and Zn ions in the feed solution was not studied. It was found that the corresponding concentration of Mn in the feed solution scarcely changed. The Mn, Fe, Ni and Zn concentrations in the prepared acidic leach solution were 939 ppm, 645 ppm, 104 ppm and 78 ppm, respectively. The prepared acidic leach solution was used as the feed solution in the ELM experiments.

Membrane stability

A certain volume of the prepared W/O emulsion was added to 500 mL of external aqueous solution (demineralized water) in a cylindrical glass vessel attached to an overhead motor driver (Fig. 2). The impeller agitator used had four 45° pitch blades for down pumping (diameter 95 mm). The content of the vessel was stirred in order to disperse the emulsion in the external phase for different contact times to make a W/O/W double emulsion. The pH of the external phase was continuously measured to follow its evolution against time. The increase of H ions in the external phase decreased with the pH of the aqueous solution (demineralized water) and indicated rupture of the W/O emulsion. Each experiment was performed at least twice, and the average value is reported.

Experimental procedure

The ELM experiments were carried out in a 1000 mL glass beaker as shown in Fig. 1. A measured quantity of freshly prepared W/O emulsion was poured slowly into 500 mL of the continuous phase (aqueous manganese solution) in a glass beaker. The solution in the beaker was stirred by a motor driven mechanical stirrer regulated by a voltage regulator at 200–600 rpm to disperse the emulsion phase. To determine the effect of the important governing variables on the permeation and the separation of the manganese, surfactant and carrier concentrations, internal phase concentration, stirring speed, volume ratio of emulsion to external phase, manganese ions concentration and pH of the feed solutions were investigated under different conditions.

Analytical method

First, samples were collected periodically from the beaker and set aside for a few seconds (around 10–20 seconds). As the continuous phase was heavier than the emulsion phase, it settled at the bottom. After the separation of the phases, the aqueous phase was carefully separated from the organic phase and the equilibrium pH was measured. The concentration of manganese ions in the emulsion phase was calculated from the difference between the external phase concentration before and after extraction. In all experiments the results were taken in terms of the concentration of metal ions extracted to the membrane at different time intervals. The extraction percentage (%E) was calculated by the relation:where C₀ is the initial concentration of metal ions in the feed solution and C_t is the metal ion concentration in the feed solution after time interval, t. All experiments were performed at the constant temperature of 28 ± 0.1 °C using a Shaking Water Bath (Julabo, TW20).

Results and discussion

Transport mechanism

A scheme of the transport of manganese ions through the liquid membrane facilitated by the carrier is depicted in Fig. 3. In order to determine the formation mechanism of the metal–extractant complex during manganese ions extraction, an ideal extraction system was assumed (liquid phase activity = 1)¹³ leading us to propose eqn (1) for metal–extractant complex formation:

M_(aq)ⁿ⁺ + m(HL)_x(org) ⇄ ML_n(HL)_mx−n(org) + nH_(aq)⁺ (2)

Fig. 3 Extraction and stripping of manganese ions from the acidic leach solutions by ELM process using MDEHPA as extractant or carrier.

In this equation, M_(aq)ⁿ⁺ is the metal ion in the aqueous phase, (HL)_org is the extractant used, ML_n(HL)_mx−n(org) is the metal–extractant complex and H_(aq)⁺ is the hydrogen ion released during metal complex formation. Additionally, n is the oxidation number of the metal, m is the number of molecules of extractant engaged in the reaction (coordination number) and x is the polymerization degree of the extractant, i.e., it indicates how the molecules of the extractant are associated with themselves (monomer, dimer, trimer, etc.)

The extraction reaction for Mn²⁺ can be written as:

Mn_(aq)²⁺ + m(HL)_x(org) ⇄ MnL₂(HL)_mx−2(org) + 2H_(aq)⁺ (3)

from eqn (3), the equilibrium constant (K_ex) may be expressed as:

and eqn (3) can be also written as:

For the extraction of manganese, the distribution coefficient (D) can be expressed as a function of metallic species concentration ratio in the organic and aqueous phases at the equilibrium condition as:

By introduction of eqn (6) into eqn (5), the linear form of the distribution coefficient is obtained as:

log D = log K_ex + m log[(HL)_x]_org + 2pH (7)

The mechanism of separation by extraction and re-extraction (stripping) in the ELM system considered here corresponds to a liquid–liquid cation exchange reaction. The log–log plot of the distribution coefficient (D) versus [HL]_org gives an estimate of the number of the extractant molecules associated with the extracted manganese complex (Fig. 4).

Fig. 4 Plot of log(D) vs. log[MDEHPA], pH = 6.0; [MDEHPA]₀ = 10% (v/v); [Mn(II)]₀ = 1000 ppm; T = 28 ± 2 °C.

The slope in Fig. 4 (2.098) indicates that two molecules of MDEHPA participated in the formation of the extractable complex and this supports the proposed equilibrium expression in eqn (3). The extraction reaction for manganese ions can be written as:

M_(aq)²⁺ + 2(HL)_x(org) ⇄ ML₂(HL)_{2(x−1)(org)} + 2H_(aq)⁺ (8)

The 1 : 2 stoichiometry of the reaction was confirmed by the result of the equilibrium experiment. The mechanism of the counter transport may be represented by the scheme shown in Fig. 3. This shows transport of manganese ions from the feed solution to the receiving solution facilitated by the carrier in a solvent as the membrane phase. After the formation at the outer interface, the complex diffuses through the membrane phase to the inner interface between the membrane and the stripping phases. If reaction (8) occurs in reverse then the manganese ions are released into the stripping phase.

Membrane stability

The stability of the ELM is one of the most important parameters that affects the permeation process. Membrane break-up causes a decrease in the separation efficiency due to the leakage of separated ions from the internal to the external phase.¹⁶ First, stability tests were performed in order to understand whether emulsion components such as surfactants, diluent, extractant and stripping agent were compatible with each other or not. Before each test, using trial-and-error, stability testing was performed. The purpose of this experiment was to choose appropriate diluents to stabilize the emulsion during the experiments. Stability tests were conducted on some diluents (MIPS, kerosene, toluene and chloroform) where the amount of each compound differed in the organic phase. Equal volumes of organic and stripping phases were mixed using a high speed homogenizer (1000–8000 rpm) for 10 minutes.

After that, the tracer method was used to determine the stability of the liquid membrane. In order to determine the stability of the ELM experiments were carried out with different concentrations of PEG and emulsification time. Furthermore, during the stability experiments we used demineralised water with marked acidity as the external phase instead of an aqueous solution containing the manganese ion. The concentration of H ions was used as a tracer in the internal aqueous phase in all stability experiments for controlling the effect of mass transfer. Breakage results in transfer of the tracer from the internal to the external phase. The stability of the ELM was quantified as the percentage of H ions transferred from the internal to the external phase. The breakage ratio is defined as follows:¹⁷

where B is the breakage ratio of the emulsion which represents the stability of the ELM system, [H⁺]₀ is the initial concentration of total H ions in the internal aqueous phase, [H⁺]_e is the concentration of H ions in the external aqueous phase at contact time, [H⁺]_i is the initial concentration of total H ions in the external aqueous phase and V_e and V_i are the volume of the external and the internal aqueous phases, respectively. By measuring the concentration of H ions in the external phase at various time intervals, the stability of the ELM with time can be determined by eqn (9).

Effect of emulsification time on stability

In order to determine the influence of emulsification time on the stability of the ELM system, studies were conducted with various emulsification times in the range of 5.0–20.0 min.

The effect of emulsification time on the emulsion stability indicates that less breakage occurred for an emulsification time of 10 to 15 min and further increases in emulsifying time decreased the stability (Fig. 5). For insufficient emulsification times (<5 min), the breakage was high due to the large size leading to the coalescence of the droplets. In contrast, for a long emulsification time, breakage was important because of high internal shearing conducive to a very high number of small droplets by volume unit. This increases the contact frequency between small droplets contributing to emulsion breakage. Accordingly, to ensure the stability of the ELM, a 10 min emulsification was selected as the optimum emulsification time. Each stability test of the ELM system was conducted more than twice.

Fig. 5 Effect of emulsification time on the stability of the ELM system (PEG: 4% (w/v), MDEHPA: 10% (v/v), stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; feed phase (demineralized water; pH = 5.5–6.0); phase ratio: 5/5; treatment ratio: 1/10).

Effect of surfactant concentration on stability

Surfactant concentration is an important factor affecting the ELM stability, the extraction rate and the emulsion swelling. The emulsion stability, in particular, is strongly dependent on surfactant concentration. The stability of the emulsion is determined by the molecular layer formed by the surfactant between the oil and the aqueous phases. By increasing the surfactant concentration, more surfactant molecules are arrayed between the surface of the oil and the aqueous phase. On the other hand, excessive surfactant increases the resistance of the interface. This diminishes the rate of mass transfer (Tang et al., 2010): Fig. 6 shows the variation of emulsion stability for various surfactant concentrations. This indicates that the breakup of the emulsion or emulsion stability was strongly dependent on the surfactant concentration. From Fig. 6, it is observed that the emulsion stability is improved by increasing the surfactant concentration up to 6% (w/v). At low surfactant concentrations (less than 2%), emulsions are easily broken. This thin layer does not have a strong influence on the interaction energy and produces no barrier on oil film rupture. This is due to the fact that as the concentration of surfactant decreases the surfactant is not adsorbed to a greater extent. Therefore a more compact and more strongly adsorbed interfacial film of surfactant molecules would not be formed. So this amount is insufficient for surrounding all of the internal aqueous phase. Consequently, the quantity of surfactant in the membrane must be adequate for emulsion stabilization.

Fig. 6 Effect of surfactant (PEG) concentration on stability of the ELM system (PEG: 3–6% (w/v), MDEHPA: 10% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; feed phase (demineralized water, pH = 5.5–6.0); emulsification time: 12 min; phase ratio: 5/5; treatment ratio: 1/10).

When the concentration of PEG in the membrane phase was 3 or 4% (w/v), the breakage was remarkably enhanced after 15 min. By increasing the concentration of PEG to 5 or 6% (w/v) a linear relationship versus contact time during a 30 minute period was obtained. There was no substantial difference between the breakage ratios for 5 and 6% surfactant concentration (see Fig. 6). On the basis of the results shown in Fig. 6, the concentration of surfactant was set at 5% (w/v) in all subsequent experiments since the breakage ratio at 6% (w/v) was rather close to that at 5.0% (w/v).

In summary, the optimal experimental conditions with the aim of obtaining a stable W/O emulsion can be summarized as: emulsion volume: 50 mL; external feed phase volume: 500 mL; volume ratio of internal phase to organic phase: 1/1; emulsification time: 12 min; stirring speed 500 rpm; concentration of PEG: 5% (w/v); feed phase pH 6; volume ratio of emulsion to external phase: 50/500; internal phase concentration (H₂SO₄): 0.5 M; diluent: MIPS.

Extraction

Effects of surfactant type and surfactant concentration. The nature of the surfactant and its concentration play an important role in the ELM separation processes. The stability and emulsion type depend on the nature and concentration of the surfactant.¹⁸ Surfactants are usually organic compounds that are amphipathic, in that they contain both hydrophilic and hydrophobic groups. Therefore, they are soluble in both organic and water phases. The surfactant must be properly chosen in order to minimize the co-transport of water during the extraction process. Increasing the concentration of the surfactant increases the stability of the liquid membrane and decreases the breakup rate. However, an increase in the surfactant concentration decreases the removal efficiency due to a greater mass transfer resistance. Accordingly, there is an optimum concentration for the surfactant.¹⁹ Span 80 (HLB = 4.7) is a widely used surfactant for the preparation of W/O type emulsions²⁰ in various ELM processes. In this study, we considered two non-ionic surfactants, one was the widely used Span 80 and the other was a nonconventional surfactant i.e. PEG. PEG acts as a bi-functional surfactant in the extraction of some metal ions such as Zn, Mo, Co, Li, etc.²¹ In such separations, PEG functions as a carrier as well as a surfactant. However, in the separation of manganese ions, PEG acts only as a surfactant, as discussed in the section on the effect of the carrier concentration.

First, the effect of Span 80 concentration was studied in the range 4–7% (w/v). The experimental conditions were applied on the basis of solvent extraction. Fig. 7 shows the extraction of manganese ions at various Span 80 concentrations. At the lowest surfactant concentration (4%), the extraction was enhanced for up to 2 minutes. This was followed by a sharp decline. Increasing the surfactant concentration from 4 to 6%, resulted in a sharp increase in the amount of extracted manganese ions. However, after 10 min due to the instability of the emulsion, the amount of extracted manganese ions decreased gradually. Furthermore, around 95% of manganese ions were separated within 10 min; but at longer time, the manganese ion concentration in the feed phase was increased. This is more prominent at lower concentrations of Span 80. It is known that Span 80 affects solute transport due to its influence on water transport and swelling via osmotic effects.¹⁴ Swelling results in breaking of the emulsion, and hence the manganese ion concentration in the feed is increased over time. Moreover, Span 80 is chemically unstable and undergoes the reaction of hydrolysis in moderate acidic or basic media.²⁰ Because of these drawbacks of Span 80, we considered another surfactant; PEG.

Fig. 7 Effect of surfactant (Span 80) concentration on the extraction rate of Mn (Span 80: 4–7% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

Fig. 8 shows the extraction of manganese ions at various PEG concentrations. The extraction of manganese ions was improved over time reaching an almost constant rate after 20 min of operation. However, in the case of Span 80 (at a lower concentration, Fig. 7) the extraction of manganese ions declined over time. On the other hand, at low concentration, the emulsions are not stable for a long enough time to allow separation of emulsion phase from the continuous phase. Accordingly, PEG was selected as the surfactant of choice for this separation.

Fig. 8 Effect of surfactant (PEG) concentration on the extraction rate of Mn (PEG: 3–6% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

Fig. 7 shows manganese ion extraction with various concentrations of Span 80. The experiments were carried out with a feed phase concentration of 1000 mg L⁻¹ of manganese ions and a strip phase concentration of 0.5 M H₂SO₄. On the basis of solvent extraction,²³ the separation of manganese ions was maximized at a pH of 6.0. Therefore, the feed phase pH was maintained at 6.0 by adding the required quantity of H₂SO₄. Surfactant concentration is very important in order for the ELM to stabilize the emulsion during the process. The main disadvantage is the viscosity increment with an improvement in mass transfer resistance.¹⁸

For PEG, Fig. 8 shows that at the lowest surfactant concentration (3%), about 96% of manganese ions were separated within 5 min but after that, due to the instability of the emulsion, the amount of extracted manganese ions decreased sharply. Increasing the surfactant concentration from 4 to 5%, resulted in a sharp increase in the amount of extracted manganese ions. When the surfactant concentration of the organic phase was increased to 6%, the extraction efficiency for manganese ions was decreased. Excessive surfactant increases the resistance at the interface and this can be attributed to a number of possible factors caused by high interfacial occupancy of the surfactant, including a decrease in metal ion extraction at the membrane phase–feed phase interface, an increase in interfacial viscosity and a decrease in the movement of inner droplets within the emulsion globule.^6–10 The extraction of manganese ions using 5% surfactant was similar to that at 4%. However, subsequent experiments were carried out at the higher PEG concentration of 5% (w/v) to ensure the ELM’s stability.

Effect of feed solution pH. The extraction reaction on the membrane–feed solution side plays a vital role in the transfer of metal ions from the feed side to membrane side. So the extraction kinetics was studied with different feed phase pH values varied from 5.0–7.0. Fig. 9 shows the effect of pH (H ion concentration) on the kinetics of manganese ion extraction and on the final manganese ion concentration. It is clear from eqn (7) that transport of ions through the ELM is dependent on the H ion concentration in the feed solution. As seen from Fig. 9, manganese extraction was increased by increasing the pH between 5.0 and 6.0. However, for pH between 6.0 and 6.5, the extraction efficiency of manganese ions hardly changed when increasing the pH. When the pH of the feed phase was increased to 7.0, the extraction and enrichment of manganese ions decreased due to swelling of the emulsion. The osmotic pressure difference due to the pH increment was responsible for water transport into the internal phase leading to swelling. Accordingly, an acidic feed phase facilitates manganese enrichment.

Fig. 9 Effect of feed solution pH on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 5.0–7.0; phase ratio: 5/5; treatment ratio: 1/10).

The poor performance at low pH could be explained by the competition of H ions with the solute due to the release of H ions from the extractant to the acidic leach solution. As a result, maximum extraction was achieved at pH 6.5. At this pH, swelling of the emulsion was not observed. Thus, pH 6.5 was found to be the optimal pH.

Effect of stripping solution type. As a general rule, the most important factor for application of an emulsion liquid membrane is the emulsion stability. In addition to mixing speed, extractant, and surfactant concentration, another parameter is the type of stripping agent.²⁶ The selection of a suitable stripping solution is considered to be one of the key factors for an effective ELM system. Initially we examined HCl, H₂SO₄, and HNO₃ as stripping solutions for the current ELM. The results are presented in Fig. 10. The stripping solution with 0.5 M H₂SO₄ solution provided higher manganese extraction and more stable emulsion. Therefore, 0.5 M sulfuric acid was selected as the stripping solution of choice.

Fig. 10 Effect of stripping solution type on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

Effect of internal phase concentration. The solute extraction rate increases with an increase in the concentration of the internal reagent in the emulsion.¹⁹ The influence of H₂SO₄ concentration in the strip phase on the transport of manganese was studied in the range of 0.3–0.6 M H₂SO₄. By increasing the sulfuric acid concentration from 0.3 to 0.5 M, the extraction efficiency was improved. However, this was diminished by increasing the concentration of sulfuric acid from 0.5 to 0.6 M. Again, the extraction efficiency of manganese was increased (Fig. 11) by increasing the acidity in the stripping phase. The differences of the hydrogen ion chemical potentials between the two aqueous phases are the main driving force in the emulsion liquid membrane for the counter-transport mechanism. Thus, the extraction efficiency was increased by enhancing the concentration of H₂SO₄ in the stripping solution from 0.3 to 0.5 M. However, for a concentration of 0.6 M sulfuric acid, the emulsion swells up due to osmosis, which may lead to the dilution of the internal phase and a decrease of its acidity causing a lower stripping and a consequent lower extraction. Additionally, for longer times the osmotic effect causes an increase in external phase concentration and in the ratio C/C₀.²⁷

Fig. 11 Effect of stripping solution acid concentration on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.3–0.6 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

Effect of carrier concentration. Manganese ions are insoluble in organic solvents without the assistance of a mobile carrier. The carrier in the membrane phase has two important roles: one is to transport manganese ions between the external and the internal interface of the membrane phase; and the other is to improve the selectivity of the ELM. Theoretically, MDEHPA enhances the extraction by forming complexes with manganese ions, which are permeable through the solvent (membrane phase). The influence of the carrier (MDEHPA) concentration on the extraction of manganese was studied in the range of 5.0–12.5% (v/v). The effect of carrier concentration on the extraction of manganese ions is displayed in Fig. 12. It is evident from Fig. 12 that extractant concentration has a significant effect on extraction of manganese ions from acidic solution.

Fig. 12 Effect of carrier concentration on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 0.0–12.5% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

The transport of manganese ions was negligible in the absence of a carrier. This indicates that PEG does not show any capability as an extractant and acts as a surfactant in the present study. With the increase of carrier concentration in the membrane phase the extraction of manganese was improved. This is in good agreement with eqn (3). An increase in extractant concentration from 5% to 10% leads to an increase in extraction efficiency from 90 to 96%. However, increasing the extractant concentration from 10% to 12.5% did not result in any significant change in the extraction efficiency. The unchanged extraction efficiency can be also attributable to the increased membrane phase viscosity as the carrier concentration increases in this phase. The higher concentration of carrier at the interface between the emulsion and external phase promotes the transport of the solute. However, it is expected that a very high content of carrier in the membrane does not result in a benefit due to the respective increase in viscosity, which leads to larger globules. Consequently, the kinetics is slower as shown in Fig. 12. Similar results have been obtained by other researchers.^8,9 Accordingly, an extractant concentration of 10% was considered the optimum value for manganese ion extraction.

Effect of mixing speed. A key parameter affecting the process is the mixing speed. The efficiency of ELM extraction was improved by increasing the mixing speed. This is due to a decline in the size of emulsion globules providing a higher surface area for mass transfer. However, a large enhancement of this speed affects the stability of the emulsion globules leading to longer times and to their breakage because of the excessive shearing energy input in the system.^19,23

Therefore, the extraction efficiency of manganese decreased with long term operation. Fig. 13 indicates that the maximum extraction (97%) occurred at 25 min with a mixing speed of 600 rpm. Beyond 10 min, extraction efficiency was decreased for 700 rpm.

Fig. 13 Effect of mixing speed on the extraction rate of Mn (PEG5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 400–700 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 1/10).

Effect of organic/internal phase volume ratio. The volume ratio of the aqueous internal phase to the organic phase has a significant effect on extraction using ELMs. The phase ratio is defined as:

An increase in the stripping solution volume fraction results in an enhancement of the capacity of the emulsion to extract the solute. When the volume of the stripping solution increases (i.e. the phase ratio is increased), the thickness of the film in droplets becomes thinner due to dispersion of the stripping solution in the membrane phase by mixing.^28,29 This is favorable to the transport of manganese ions into the stripping solution and results in an increase in the extraction of manganese ions. The effect of the volume ratios of the internal stripping solution to the membrane phase in the range of 4/5–7/5 on extraction of manganese ions, by maintaining a constant membrane volume, is depicted in Fig. 14. The extraction efficiency was improved by increasing the phase ratio from 4/5 to 6/5 because of the greater availability of extractant for the manganese extraction. Beyond the ratio of 6/5, a further increase in the volume of the internal aqueous solution decreased both the rate and the efficiency of the extraction. This may be due to an increase in the emulsion viscosity and an improvement in the diameter of the internal droplets.^8,28 The increase of the droplet diameter decreases the interfacial contact area between the emulsion and feed solution. This decreases the extraction efficiency. Additionally, the emulsion stability may decrease rapidly with an increase of the phase ratio from 6/5 to 7/5. Increasing the stripping phase volume makes the emulsion unstable with a leakage of the internal phase into the continuous phase.

Fig. 14 Effect of phase ratio on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 4/5–7/5; treatment ratio: 1/10).

In order to obtain a better extraction efficiency in the membrane solution, the volume ratio of the internal aqueous phase to the membrane phase of 6/5 was selected as the optimum volume ratio.

Effect of treatment ratio. The volume ratio of the primary emulsion to the feed solution (V_E/V_F) affects the interfacial mass transfer across an ELM. This is a measure of the emulsion holdup in the system. Improving the treatment ratio (V_E/V_F) results in an increase in the emulsion phase holdup and simultaneously an increase in the extraction capacity of the emulsion.⁸ As a general expectation, a higher treatment ratio leads to higher extraction efficiency.

Fig. 15 shows that, as the treatment ratio is varied from 0.1 to 0.2, while the emulsion phase holdup was low (treatment ratio around 0.1), the rate of extraction was at its minimum. When maintaining all other parameters as constant, there is an increase in the rate of extraction. This increase occurs when the phase ratio increases from 0.1 to 0.15, but increasing the emulsion holdup to 0.16 slightly increased the extraction efficiency. Beyond the ratio of 0.16 (treatment ratio around 0.2) the extraction efficiency was diminished. This behavior can be explained by two opposite phenomena occurring simultaneously. Increasing the treatment ratio increases the quantity of membrane and internal phase, the amount of carrier in the membrane phase and the membrane phase surface area for transport. This enhances the permeation and stripping of manganese.^21,30 An increase in the emulsion volume leads to a reduction of the mass transfer area and hence the rate of extraction. In this case, a longer extraction time is required and consequently more emulsion breakage is expected.²¹ This reduces the permeability and stripping of manganese. Therefore, a treatment ratio of 0.15 was selected as the treatment ratio for optimum conditions.

Fig. 15 Effect of treatment ratio on the extraction rate of Mn (PEG: 5% (w/v); MDEHPA: 10.0% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; Mn concentration of feed solution: 1500 mg L⁻¹; feed solution pH: 6.0; phase ratio: 5/5; treatment ratio: 0.1–0.2).

Extraction of manganese from industrial wastewater at optimum conditions

The optimum conditions for extraction of manganese ions, obtained from the previous experiments using the specified ELM, are presented in Table 3. The same process at optimum conditions was employed for the extraction of manganese from real industrial wastewater to elucidate the feasibility of the ELM process in treating real effluent.

Table 3 Optimum conditions for the extraction of manganese with the ELM

Parameter	Value
Surfactant (PEG)	5% (w/v)
Extractant (MDEHPA)	10.0% (v/v)
Diluent (MIPS)	Table 1
Stripping solution type	25 mL H2SO4 solution
Stripping solution H2SO4 concentration	0.5 M
pH of the feed solution	6.5
Mixing speed	600 rpm
Phase ratio	6/5
Treatment ratio	0.15

---

The experiments were carried out by adjusting the pH of industrial wastewater from 6.8 to 6.5 by adding H₂SO₄. The effect of the optimum conditions on the extraction efficiency of manganese from the acidic leach solution-containing Mn, Fe, Ni and Zn ions is presented in Fig. 16 indicating an extraction of around 92–93% of the manganese as well as low extraction of iron, nickel and zinc within a period of 30 min of permeation. This is expected due to the interference of other ions present in the feed phase. The present work provides a basis for extraction and removal of manganese ions from real industrial wastewater using an emulsion liquid membrane indicating an acceptable performance at the conditions established in this work.

Fig. 16 Effect of the optimum conditions on the extraction rate of Mn from real industrial wastewater (PEG: 5% (w/v); MDEHPA: 12.5% (v/v); stripping solution: 25 mL 0.5 M H₂SO₄; mixing speed: 500 rpm; feed solution: real industrial wastewater; feed solution pH: 6.5; phase ratio: 6/5; treatment ratio: 1/10).

Conclusions

An emulsion liquid membrane process using MDEHPA as an extractant to selectively extract manganese ions from an acidic leach solution containing various metal ions has been investigated. From this study, the following conclusions can be drawn.

1. The extraction of manganese ions is influenced by a number of variables including the pH of the feed aqueous solution, mixing speed, concentrations of surfactant, stripping phase and carrier, and treatment ratio.

2. The proposed ELM process was efficient in the separation of manganese ions from industrial wastewater with removal of 92–93% of manganese ions.

3. The experimental results obtained also demonstrated the validity of the ELM method for the treatment of industrial wastewaters contaminated with manganese ions.

References

1. W. Zhang and C. Y. Cheng, Manganese metallurgy review, Part II: manganese separation and recovery from solution, Hydrometallurgy, 2007, 89(3–4), 160–177.
2. N. B. Devi and S. Mishra, Solvent extraction equilibrium study of manganese(II) with Cyanex 302 in kerosene, Hydrometallurgy, 2010, 103, 118–123.
3. D. H. Fatmeshari, D. Darvishi, S. Etenadi, A. R. E. Hollagh, E. K. Almadri and A. A. Salardini, Interaction between TBP and D2EHPA during Zn, Cd, Mn, Cu, Co and Ni solvent extraction: a thermodynamic and empirical approach, Hydrometallurgy, 2009, 98, 143–147.
4. J. Gega, W. Walkowiak and B. Gajda, Separation of Co(II) and Ni(II) ions by supported and hybrid liquid membranes, Sep. Purif. Technol., 2001, 22–23, 551–558.
5. R. A. Wood and R. V. Roscoe, The isolation of manganese in a system between bis(2-ethyl-hexyl) hydrogen phosphate and aqueous nitric acid, J. Inorg. Nucl. Chem., 1970, 32, 1351–1355.
6. P. S. Kulkarni and V. V. Mahajani, Application of liquid emulsion membrane (LEM) process for enrichment of molybdenum from aqueous solutions, J. Membr. Sci., 2002, 201, 123–135.
7. L. Yang, Z. Zhang, Y. Guo, X. Gao and H. Takeuchi, Uranium(VI) extraction by liquid surfactant membrane with N-alkylcaprolactams as a carrier, Sep. Purif. Technol., 2005, 47, 88–94.
8. R. A. Kumbasar, Extraction of chromium(VI) from multicomponent acidic solutions by emulsion liquid membranes using TOPO as extractant, J. Hazard. Mater., 2009, 167, 1141–1147.
9. N. Othman, H. Mat and M. Goto, Separation of silver from photographic wastes by emulsion liquid membrane system, J. Membr. Sci., 2006, 282, 171–177.
10. N. B. Devi, K. C. Nathsarma and V. Chakravortty, Separation of divalent manganese and cobalt ions from sulfate solutions using sodium salt of D2EHPA, PC 88A and Cyanex 272, Hydrometallurgy, 2000, 54(2–3), 117–131.
11. S. C. Lee, J. H. Chang, B. S. Ahn and W. K. Lee, Mathematical modeling of silver extraction by an emulsion liquid membrane process, J. Membr. Sci., 1996, 114, 171–185.
12. A. P. Argekar and A. K. Shetty, Extraction of manganese(II) with bis(2,4,4-trimethylpentyl) mono-thio-phosphinic acid and its spectrophotometric determination with form aldoxime, Anal. Sci., 1997, 13, 131–135.
13. G. M. Ritcey and B. H. Lucas, Purification of manganese solutions containing copper and zinc by liquid–liquid extraction, using di-(2-ethyl-hexyl) phosphoric acid, Can. Metall. Q., 1971, 10, 223–228.
14. G. C. Sahoo and N. N. Dutta, Studies on emulsion liquid membrane extraction of cephalexin, J. Membr. Sci., 1998, 145, 15–26.
15. A. Kargari, T. Kaghazchi, M. Sohrabi and M. Soleimani, Experimental and modeling of selective separation of gold(III) ions from aqueous solutions by emulsion liquid membrane system, J. Membr. Sci., 2006, 279, 389–393.
16. A. Kargari, T. Kaghazchi, M. Sohrabi and M. Soleimani, Batch extraction of gold(III) ions from aqueous solutions using emulsion liquid membrane via facilitated carrier transport, J. Membr. Sci., 2004, 233, 1–10.
17. B. Tang, G. Yu, J. Fang and T. Shi, Recovery of high-purity silver directly from dilute effluents by an emulsion liquid membrane-crystallization process, J. Hazard. Mater., 2010, 177, 377–383.
18. K. Chakrabarty, P. Saha and A. K. Ghoshal, Separation of lignosulfonate from its aqueous solution using emulsion liquid membrane, J. Membr. Sci., 2010, 360, 34–39.
19. H. J. Park and T. S. Chung, Removal of phenol from aqueous solution by liquid emulsion membrane, Korean J. Chem. Eng., 2003, 20, 731–735.
20. I. A. Nemeh and A. P. V. Peteghem, Sorbitan monooleate (Span 80) decomposition during membrane ageing, a kinetic study, J. Membr. Sci., 1992, 74, 9–17.
21. C. F. Jee, T. B. Long, X. M. Xia, Q. Q. Jin and Z. L. Ying, A study on a two components liquid membrane system, J. Membr. Sci., 1985, 23, 137–154.
22. R. Mohapatra and S. B. Kanungo, Kinetics of Mn(II) transport from aqueous sulfate solution through a supported liquid membrane containing di(2-ethylhexyl) phosphoric acid in kerosene, Sep. Sci. Technol., 1992, 27, 1759–1773.
23. N. B. Devi, K. C. Nathsarma and V. Chakravortty, Extraction and separation of Mn(II) and Zn (II) from sulfate solutions by sodium salt of Cyanex 272, Hydrometallurgy, 1997, 45(1–2), 169–179.
24. L. Soko, L. Chimuka, E. Cukrowska and S. Pole, Extraction and preconcentration of manganese (II) from biological fluids (water, milk and blood serum) using supported liquid membrane and membrane probe methods, Anal. Chim. Acta, 2003, 485, 25–35.
25. S. S. Kolekar and M. A. Anuse, Solvent extraction separation of rhodium(III) with N-noctylaniline as an extractant, Talanta, 2002, 58, 761–771.
26. C. P. Mane and M. A. Anus, Extraction behaviour of 2-octylaminopyridine towards lead(II) from succinate media and its separation from other toxic metals, J. Hazard. Mater., 2008, 152, 1146–1154.
27. A. Reis, M. Teresa and J. M. R. Carvalho, Recovery of zinc from an industrial effluent by emulsion liquid membranes, J. Membr. Sci., 1993, 84(3), 201–211.
28. C. Das, M. Rungta, G. Arya, S. DasGupta and S. De, Removal of dyes and their mixtures from aqueous solution using liquid emulsion membrane, J. Hazard. Mater., 2008, 159(2–3), 365–371.
29. A. Dâas and O. Hamdaoui, Extraction of anionic dye from aqueous solutions by emulsion liquid membrane, J. Hazard. Mater., 2010, 178, 973–981.
30. L. Boyadzhiev and E. Bezenshek, Carrier mediated extraction: application of double emulsion technique for mercury removal from wastewater, J. Membr. Sci., 1983, 14, 13–18.

---