Extraction of Ce(IV) from sulphuric acid solution by emulsion liquid membrane using D2EHPA as carrier

Abstract

In order to provide a potential method for extracting Ce(IV), the extraction of Ce(IV) from sulphuric acid solution by an emulsion liquid membrane using D2EHPA as a carrier was investigated. The ELM system consisted of sulfonated kerosene as diluent, Span 80 as surfactant, liquid paraffin as intensifier and hydrochloric acid containing hydrogen peroxide as the inner aqueous solution. The influences of various parameters on the extraction of Ce(IV) were investigated. The optimum conditions for Ce(IV) extraction can be summarized as follows: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2–3% (v/v); liquid paraffin concentration, 2–4% (v/v); hydrochloric acid concentration in the internal phase, 4–5 mol l⁻¹; hydrogen peroxide concentration, 0.02 mol l⁻¹; volume ratio of membrane phase to internal phase (R_oi), 1.5; external phase acidity, 0.4–0.5 mol l⁻¹; volume ratio of external phase to membrane phase (R_we), 2; extraction time, 15 min; and stirring speed, 250 rpm. Experiments in which Ce(IV) was separated from RE(III) were then carried out under the optimum conditions, and the results indicated that the system is extremely selective for Ce(IV). The mechanism of Ce(IV) extraction has also been discussed. The loss for the experimental process was within 3%. The results reveal that the ELM method is a clean and cost-effective process for the extraction of Ce(IV) from sulphuric acid solution.

---

1 Introduction

Cerium, one of most abundant of the light rare earth elements, is widely used in industry, mainly in the form of ceria (CeO₂), as a material for ceramics, catalysts, polishing powders and so on.^1,2 Solvent extraction is an important method for the separation and purification of cerium and is widely used in industrial production because of its simplicity, speed, and applicability in the extraction and separation of rare earth elements.^3,4 The most commonly used extractants are organophosphorus acids that belong to a compound formation class, such as di-(2-ethylhexyl) phosphoric acid (D2EHPA) and 2-ethylhexyl phosphonic acid-2-ethylhexyl ester (HEHEHP).⁵ The conventional solvent extraction method possesses many deficiencies, such as complex multi-stage extraction operation, larger volumes of solvent and associated equipment, high energy consumption, and higher investment costs.⁶ Therefore, the development of more efficient extraction techniques has attracted much attention and has important significance in the development of the rare earth industry. Liquid membrane technology is gaining importance and is emerging as an attractive and promising technology for the concentration and separation of metal ions from aqueous solutions.⁷

There are several different types of liquid membranes, such as bulk (BLM), supported (SLM), and emulsion (ELM) liquid membranes, polymer inclusion membranes (PIMs) and hollow fiber liquid membranes (HFSLMs).^8,9 BLMs consist of an aqueous feed and stripping phase separated by an immiscible membrane phase. BLMs are often used to study the transport properties of novel carriers; however, their small membrane surface areas make them unattractive.¹⁰ SLMs have essentially the same configuration as BLMs, except that the organic phase is contained in the pores of a macroporous polymer sheet.⁹ SLMs are very useful for laboratory research and development purposes, but the surface area to volume ratio of flat sheets is too low for industrial applications.¹¹ Among SLMs, HFSLM is the most promising. It possesses several advantages such as simultaneous one-step extraction and stripping, high selectivity, large surface area, high mass transfer rate, and low extractant and energy consumption.¹² Despite its advantages, the large-scale application of HFSLM is still very limited due to its insufficient membrane stability.¹³ In addition, its surface active effects can lead to fouling, and the variable geometry makes it difficult to model.⁷ The PIMs, which are composed of polymers, plasticizers, and ion carriers, have gained considerable attention due to their higher selectivities for target metal ions, lower carrier consumption, and the integrated extraction and stripping process; however, the selection of a suitable polymer for PIMs is one of the important factors in membrane separation.¹⁴

Emulsion liquid membrane (ELM), a separation technique proposed by Li,¹⁵ has been regarded as an emerging energy-efficient technology that combines the characteristics of solid membrane separation and solvent extraction. The main advantages of ELM are as follows: high interfacial area for mass transfer because of the small sizes of the aqueous droplets; high diffusion rate of the target ion through the membrane; simultaneous, one-step extraction and stripping; the capability of treating a variety of elements and compounds in the industrial setting at a greater speed and with a high degree of effectiveness, with varying contaminant concentrations and volume requirements.¹⁶ Compared with HFSLM, ELM requires a much smaller inventory of solvent and carrier since the volume ratios of the two aqueous phases to the membrane are large; however, the large volume ratios are a disadvantage in that very hydrophobic membrane solvents and carriers are required to maintain membrane integrity and operate the system.⁷

The ELM contains three phases, the membrane phase, internal phase and external phase. The liquid membrane consists of a diluent, a surfactant and a carrier reagent (extractant) to separate the solute by chemical reaction. The external continuous phase and internal phase are separated by the liquid membrane. The selective component is transported through the liquid membrane from the external phase and concentrated in the internal phase.¹⁷ After extraction, the loaded emulsion is separated from the feed solution, and demulsification yields a membrane phase that can be recycled.⁶ The extraction mechanism involved in liquid membrane transport is essentially the same as that in solvent extraction. It has been reported that various ions can be successfully extracted by ELM.^18–23 Few papers have been reported on the extraction of lanthanides by ELM.^24–26 Teramoto et al.²⁷ and Tang and Wai²⁸ explored the extraction and transport of trivalent lanthanides by liquid membrane systems.

The transport of Ce(IV) with a membrane system has been reported before;^29–31 however, studies on Ce(IV) extraction from sulphuric acid solution by ELM are lacking.

In this paper, the extraction of Ce(IV) from sulphuric acid solution by ELM was investigated. D2EHPA was chosen as the extractant. Various parameters influencing the extraction of Ce(IV) were investigated with the aim of providing potential applications for extracting Ce(IV) by a clean and cost-effective separation method, replacing the existing system.

2 Experimental

2.1 Reagents and apparatus

Liquid paraffin with a specific gravity of 0.835–0.890 was supplied by Tianjin Bodi Chemical Industry Co., Ltd. Span80 (C₂₄H₄₄O₆), hydrogen peroxide (H₂O₂) and cerium(IV) sulphate (Ce(SO₄)₂·4H₂O) were of analytical grade and supplied by Shenyang Guoyao Group Chemical Reagent Co., Ltd. Sulphuric acid (A.R. 98%) and hydrochloric acid (A.R. 37.5%) were supplied by Shenyang Lab Science and Trade Co., Ltd. n-Dodecane, D2EHPA and sulfonated kerosene were supplied by Shanghai Laiyashi Chemical Industry Co., Ltd. All solutions were prepared using deionised water with resistivities >18.23 MΩ cm⁻¹.

An FJ200-SH digital high-speed dispersion homogenizer (Shanghai Specimen Model Factory) was used to prepare the liquid membrane solution. A DW-3 speed control magnetic stirrer (Gongyi Yuhua Instruments Co., Ltd) was used to carry out the extraction experiments.

2.2 Procedure

2.2.1 Membrane preparation. Using sulfonated kerosene as a membrane diluent, Span80 as surfactant, D2EHPA as carrier and liquid paraffin as the liquid membrane intensifier, an organic solution was prepared by mixing certain volume fractions of sulfonated kerosene, Span80, D2EHPA and liquid paraffin in a 250 ml beaker at stirring speed of 1000 rpm using a dispersion homogenizer. The hydrochloric acid aqueous solution with hydrogen peroxide as reductant was then added dropwise to the stirred membrane solution until the desired volume ratio of membrane solution to stripping solution was obtained. The solution was stirred continuously for 20 min at a stirring speed of 5000 rpm to obtain a stable W₁/O-type ELM.

2.2.2 Feed solution preparation. The stock solution of Ce(IV) was prepared by adding the sulphuric acid and cerium(IV) sulphate in stoichiometric amounts to a volumetric flask and then adding water to obtain the desired concentrations of cerium and acid.

2.2.3 Extraction. All experiments were performed at a constant temperature of 25 ± 0.2 °C. The prepared emulsion was added into a specific volume of external aqueous solution. The contents were stirred by a speed-controlled magnetic stirrer at speed of 250 rpm for 15 min to produce the W₁/O/W₂ double emulsions. The stirred solution was allowed to separate to an emulsion (W₁/O) and an external aqueous solution (W₂) by gravity in a 125 ml separatory funnel. After two-phase separation, the concentration of Ce(IV) in the external aqueous phase was analysed. The concentration in the internal aqueous phase was obtained by mass balance. The emulsion was recovered and subsequently broken into its constituent organic and internal aqueous cerium concentrated solutions.

To determine the important variables governing the extraction of Ce(IV), the extractant concentration, surfactant concentration, intensifier concentration, hydrochloric acid and hydrogen peroxide concentrations of the stripping solution, volume ratio of the membrane phase to the internal phase (R_oi), volume ratio of the external phase to the membrane phase (R_we), acidity in the feed solution, extraction time and stirring speed were varied to observe their effects on Ce(IV) extraction. The separation of Ce(IV) from RE(III) was also investigated.

2.2.4 Demulsification of emulsion. After the extraction experiment, the loaded emulsion was broken using a 45 kHz KQ-700VDE dual-frequency NC ultrasonic processor (Kunshan Ultrasonic Instrument Co., Ltd) into the internal Ce(IV)-concentrated phase and the organic phase. The internal phase was separated and centrifuged, and the Ce(IV) concentration was determined.

2.3 Analyses methods

The chemical structure of the D2EHPA complex with Ce(IV) was detected by the Nicolet-380 Fourier transform infrared spectrometer (Thermo Electron Corporation). The viscosities of the emulsions were measured by DV2TLVTJ0 viscometer (Brookfield).

The average sizes of emulsions were measured using a BT-9300H laser particle size analyser (Dandong Bettersize Instruments Co., Ltd) within a few minutes after preparation. The globule size of the emulsion was measured by diluting the sample with deionised water, and the internal droplet size of the emulsion was measured by diluting the sample with n-dodecane. The average diameters of the emulsion liquid globules and droplets were represented by Sauter mean diameter (d₃₂ = ∑n_id_i³/∑n_id_i², representing a surface average value). Additionally, a BT-1600 particle image analysis system with an optical microscope (Dandong Bettersize Instruments Co., Ltd) was used to view emulsion morphology and assess the average thickness of the membrane.

The concentration of D2EHPA was determined by titration with standard NaOH solution using phenolphthalein and bromophenol blue as indicators. The acidity was determined by neutralisation titration with standard NaOH solution using mixed indicators of methyl red and bromocresol green after masking the metal ions with Ca-EDTA solution. The concentration of RE(III) was determined by titration with standard EDTA solution with xylenol orange as the indicator. The concentration of Ce(IV) was determined by titration with standard [(NH₄)₂Fe(SO₄)₂] solution using sodium diphenylamine sulfonate as indicator.

The extraction efficiency η was obtained by eqn (1):

η = (C₀ − C_t)/C₀ × 100% (1)

where C₀ is the Ce(IV) concentration in the external aqueous phase before extraction (mol l⁻¹), and C_t is the Ce(IV) concentration in the external aqueous phase after extraction (mol l⁻¹).

3 Results and discussion

3.1 Extraction mechanism in ELM system

D2EHPA is supposed to be present as a dimeric species such as H₂A₂ in the non-polar diluent sulfonated kerosene due to the presence of intermolecular hydrogen bonds;³² the chemical structure of D2EHPA is shown in Fig. 1. Solvent extraction studies on rare earth elements using D2EHPA suggested that Ce(IV) can be extracted from an aqueous acid medium with D2EHPA via a cation-exchange mechanism.³³ The FTIR spectra of D2EHPA complexes with Ce(IV) are presented in Fig. 2. The adsorption peaks at 2316 cm⁻¹, 1685 cm⁻¹, 1233 cm⁻¹ and 1035 cm⁻¹ are attributed to the stretching modes of P–OH bonds, the dimeric peak of hydrogen bonds, the P=O bonds and the stretching modes of P–O–C bonds, respectively.³⁴ The band at 2316 cm⁻¹ nearly disappears after extraction, demonstrating the extraction mechanism involving cation-exchange in which Ce(IV) replaces the hydrogen atom in P–OH bonds. The band at 1685 cm⁻¹ also disappears, indicating that the intermolecular hydrogen bonds in D2EHPA are eliminated. The shift of the P=O stretching vibration from 1233 cm⁻¹ to about 1283 cm⁻¹ in the Ce(IV)-loaded D2EHPA reveals the strong interaction between Ce(IV) and P=O. The peak at 1035 cm⁻¹ has no obvious variation, indicating that there is no chemical bond between Ce(IV) and P–O–C. The extraction reaction between Ce(IV) and D2EHPA in the ELM system is the same as the solvent extraction. According to the results, the saturated extraction mechanism of Ce(IV) by ELM using D2EHPA as carrier in sulphuric acid medium (pH < 1) can be expected to follow eqn (2):³

nCe(IV)_(a) + 2nH₂A_2(o) ⇆ (CeA₄)_n(o) + 4nH(I)_(a) (2)

where the subscripts ‘(a)’ and ‘(o)’ represent species present in the aqueous and the organic phases, respectively. Because of the strong stability of the extracted complex, it is difficult to directly reverse the extraction of Ce(IV).

Fig. 1 The chemical structure of D2EHPA in sulfonated kerosene.

Fig. 2 FTIR spectra of (a) sulfonated kerosene, (b) sulfonated kerosene + 12%D2EHPA (v/v) and (c) sulfonated kerosene + 12%D2EHPA (v/v) + Ce(IV).

Hydrogen peroxide can be used as an oxidant as well as a reductant. The standard electrode potentials in hydrochloric acid conditions are as follows:^35,36

From eqn (3), the oxidizing ability of Ce(IV) in acid solution is stronger than that of hydrogen peroxide, so Ce(IV) can oxidize hydrogen peroxide. It can be concluded that using hydrogen peroxide as a reductant to reduce Ce(IV) is thermodynamically feasible and will not bring any impurity ions into the solution. Therefore, hydrochloric acid containing a certain amount of hydrogen peroxide was employed as the internal aqueous phase. The hydrogen peroxide first reduces Ce(IV) to Ce(III), and Ce(III) is then stripped with hydrochloric acid following the equilibrium equation shown as eqn (4):

2(CeA₄)_n(o) + 6nHCl_(a) + 3nH₂O_2(a) ⇆ 2nCeCl_3(a) + 4nH₂A_2(o) + 2nH₂O + 2nO₂↑ (4)

It is concluded that both the equilibrium reactions take place in the ELM system in the following sequence (Fig. 3):

Fig. 3 Possible scheme for Ce(IV) extraction by ELM using D2EHPA as extractant.

(a) Transport of Ce(IV) ions from the bulk of the external phase to interface1 via the D2EHPA-based membrane.

(b) Formation of the extracted complex at the external phase-liquid membrane interface.

(c) Diffusion of the extracted complex from the membrane phase to interface2.

(d) Stripping of the metal ion from the extracted complex by the internal phase stripping solution (HCl–H₂O₂) and enrichment of the Ce(III) ions in the internal phase droplets.

As discussed above, the extraction of Ce(IV) by ELM is a reverse-transfer process in which the Ce(IV) and H(I) are transferred from the high-concentration region to the low-concentration region. The Ce(IV) is transferred from the external phase to the internal phase, while the H(I) is transferred from the internal phase to the external phase.

3.2 Emulsion characterisation

Liquid membranes were prepared by emulsifying an aqueous solution of stripping phase (HCl–H₂O₂) with an organic phase (D2EHPA, Span80 and liquid paraffin in sulfonated kerosene), as discussed in Section 2. Studies carried out with varying emulsifier speed (1000–7000 rpm) indicated that a stable emulsion was obtained only when the emulsifier speed was higher than 2000 rpm. Therefore, an emulsifier speed of 5000 rpm was selected for the preparation of emulsion in all subsequent studies. The ELM was allowed to stand and kept under observation for any physical changes to determine the stability of ELM with respect to time. Photographs of the emulsion at different times are presented in Fig. 4. The ELM was found to be a milky emulsion that is stable for up to 2 h. The emulsion becomes unstable 6 h later and separates into two phases (organic and strip phases) after 12 h.

Fig. 4 Photographs of the emulsion at different times (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; R_oi, 1.5).

For further observation of the emulsion, a typical photograph of the emulsion is presented in Fig. 5. As shown, the clear and stable W₁/O type globules distribute uniformly in the emulsion, and the Sauter mean diameter of the globules is 47.26 μm. There are many droplets in the globules with an average diameter of 1.20 μm. The average thickness is estimated to be approximately 5 μm. Table 1 shows the effects of different components of D2EHPA, Span 80 and liquid paraffin on the viscosity. The table shows that increasing various components concentrations results in increased emulsion viscosity. Based on these results, the prepared emulsion can be applied for all subsequent experiments in the present study, including the extraction of Ce(IV) under various experimental conditions.

Fig. 5 Photograph of emulsion.

Table 1 Viscosities at 10 s⁻¹

D2EHPA (%v/v)	Viscosity (cp)	Span 80 (%v/v)	Viscosity (cp)	Liquid paraffin (%v/v)	Viscosity (cp)
8	20.40	1	12.00	0	24.00
10	23.70	2	24.60	2	28.50
12	24.60	3	33.10	3	29.70
14	27.90	4	45.32	4	31.32
16	31.20	8	86.70	6	33.90

---

3.3 Effect of emulsion composition variables

3.3.1 Effect of D2EHPA concentration. As shown in eqn (2), the extraction of Ce(IV) is positively influenced by the concentration of D2EHPA. ELM extraction studies were carried out by preparing emulsions containing various D2EHPA concentrations in the range of 8–16% (v/v), and the results are presented in Fig. 6. The extraction efficiency was found to increase with increasing extractant concentration from 8% to 16% (v/v). The viscosities of emulsions with various D2EHPA concentrations are shown in Table 1. The globule size of the emulsion containing 8% D2EHPA was measured to be d₃₂ = 39.01 μm, while that of the emulsion containing 16% D2EHPA was d₃₂ = 54.57 μm. Sengupta et al.³⁷ found a similar tendency. The inner droplet sizes d₃₂ were found to be 1.20 μm for 8% D2EHPA and 1.09 μm for 16% D2EHPA.

Fig. 6 Effect of D2EHPA concentration on the extraction of Ce(IV) (experimental conditions: Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; R_oi, 1.5; Ce(IV) concentration in external phase, 0.01mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

The higher carrier concentration at the interface between the external phase and the emulsion promotes the transport of solute and the complex of Ce(IV) and D2EHPA. Increasing the amount of carrier to 12% (v/v) results in a Ce(IV) extraction efficiency of over 95%. Upon further increases in carrier concentration, the extraction efficiency shows no obvious change, which may be attributed to an increase in the viscosity, leading to larger globules.³⁸ The literature shows that increases in emulsion viscosity with increasing carrier concentration will limit the extraction rate by affecting the dispersion behaviour of the emulsion.²⁰ Chiha et al.³⁹ reported that increasing the carrier concentration decreases the stability of the emulsion because the larger globules can aggregate quickly. In addition, the increase in the carrier concentration will promote permeation swelling, diluting the stripping solution. The swelling effect is more serious when the D2EHPA concentration is increased.⁴⁰ A higher concentration of carrier will also increase the cost. Therefore, the optimal carrier concentration is found to be 12% (v/v).

3.3.2 Effect of surfactant concentration. The effects of various Span 80 concentrations in the range of 1–8% (v/v) were investigated, as shown in Fig. 7. It was observed that the extraction efficiency increases with increasing surfactant concentration up to 3% (v/v) and decreases thereafter. The surfactant is expected to play a role as an emulsifier for the liquid membrane and act as a protective barrier between the external phase and the internal phase, preventing the leakage of emulsion.³⁸ At lower surfactant concentrations (less than 2%), the emulsion breaks easily, and the extraction efficiency is poor because the coverage of the membrane interface is incomplete at low surfactant concentration. The increase in surfactant concentration lowers the membrane's surface tension and yields smaller globules, leading to a higher contact area.^41,42 At higher surfactant concentrations (beyond 3%), the membrane stability increases, while the extraction efficiency decreases. This can be attributed to the fact that the increase in emulsion viscosity resulting from the increasing surfactant concentration (Table 1) leads to the augmentation of mass transfer resistance due to presence of excessive surfactant at the aqueous–organic phase interface,^16,43 resulting in less transfer of Ce(IV) molecules to the internal phase. A similar phenomenon was observed previously.³⁹ Therefore, Span 80 concentrations ranging from 2–3% (v/v) were found to be optimal, producing extraction efficiencies greater than 96%.

Fig. 7 Effect of Span80 concentration on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; R_oi, 1.5; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

3.3.3 Effect of liquid paraffin concentration. Liquid paraffin was added as a membrane intensifier for the liquid membrane; it promoted the mechanical strength and stability of the membrane. Photographs of emulsions with different liquid paraffin contents are shown in Fig. 8. As shown, a higher liquid paraffin concentration corresponds to a thicker globule membrane. When the membrane intensifier content is too low, the liquid membrane is very thin, resulting in poor stability; when the content is excessively high, the thickness of the membrane increases, leading to increases in diffusion and mass transfer resistance and resulting in lower mass transfer velocity and extraction efficiency.

Fig. 8 Photographs of emulsions with different liquid paraffin contents: (a) 2% (v/v) and (b) 6% (v/v).

The effect of the liquid paraffin concentration on the extraction of Ce(IV) by ELM is shown in Fig. 9. An extraction efficiency higher than 96% is obtained in the range of 2–4% (v/v), which is considered to be the optimum concentration.

Fig. 9 Effect of liquid paraffin concentration on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; R_oi, 1.5; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

3.3.4 Effect of hydrochloric acid concentration in the internal phase. Acids such as sulphuric acid, nitric acid and hydrochloric acid are the most commonly used stripping agents for the reverse extraction of rare earth elements. Sulphuric acid and nitric acids have certain causticity and oxidizability at high concentration, which will damage the stability of the liquid membrane and make the reduction of Ce(IV) more difficult. Therefore, hydrochloric acid was used as the stripping acid solution in this study. The capacity to trap and concentrate the solute in the stripping solution is determined by the concentration of acid solution in the internal phase. The effect of the hydrochloric acid concentration in the internal phase on Ce(IV) extraction by ELM is shown in Fig. 10. The yield of Ce(IV) extraction increases with increasing hydrochloric acid concentration, reaching 97% at 4–5 mol l⁻¹. This result can be attributed to the fact that a higher hydrochloric acid concentration leads to greater dissociation of the Ce–D2EHPA complex at the interface between the membrane and the stripping phase. Consequently, the driving force for the transport of the Ce–D2EHPA complex is larger due to the greater difference in Ce–D2EHPA complex concentration across the membrane. Upon further increases in acid concentration, the extraction efficiency decreases, possibly due to membrane damage that leads to the leaching of the internal phase. Consequently, the optimal concentration of hydrochloric acid in the stripping solution for the effective extraction of Ce(IV) was 4–5 mol l⁻¹.

Fig. 10 Effect of HCl concentration in the internal phase on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; R_oi, 1.5; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

3.3.5 Effect of hydrogen peroxide concentration. Because it is difficult to strip Ce(IV) directly with hydrochloric acid, hydrogen peroxide is important in the stripping process. The effect of hydrogen peroxide concentration on Ce(IV) extraction was studied. The obtained results are presented in Fig. 11.

Fig. 11 Effect of H₂O₂ concentration in the internal phase on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; R_oi, 1.5; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

It can be seen that the extraction efficiency is extremely low when hydrogen peroxide is absent in the stripping solution, and the extraction efficiency increases with increasing hydrogen peroxide concentration. When the hydrogen peroxide concentration exceeds 0.02 mol l⁻¹, the extraction efficiency increases slowly. The addition of hydrogen peroxide should be carefully controlled; when the amount added is too little, Ce(IV) cannot be completely reduced and stripped and accumulates in the organic phase, decreasing the extraction ability. However, excess hydrogen peroxide leads to the entrainment of redundant hydrogen peroxide in the organic phase, affecting the extraction efficiency. Thus, a hydrogen peroxide concentration of 0.02 mol l⁻¹ was selected as the most appropriate value.

3.3.6 Effect of volume ratio of membrane phase to internal phase (R_oi). The volume ratio of the membrane phase to the internal phase (R_oi) plays an important role in determining the effectiveness of the ELM system. The effect of R_oi on the extraction of Ce(IV) by ELM was investigated, and the obtained results are shown in Fig. 12. At low R_oi, the volume of membrane solution is not sufficient to enclose all of the stripping solution,⁴⁴ leading to low extraction efficiency. When R_oi is increased from 0.5 to 1.5, the transport rate and extraction efficiency of Ce(IV) increase. This can be attributed to the fact that an increase in the membrane phase volume fraction increases the thickness of the membrane phase, resulting in enhanced mass transfer.³⁸ The extraction of Ce(IV) reaches 96% at an R_oi of 1.5. Beyond 1.5, further increases in R_oi decrease both the rate and efficiency of extraction. This may be due to the fact that at high R_oi, membrane thickness and the viscosity of the emulsion phase are high due to the relatively high organic content. In addition, a high R_oi also means that less stripping agent is available for solute stripping. Thus, an R_oi of 1.5 is selected as the optimal volume ratio.

Fig. 12 Effect of volume ratio of the membrane phase to internal phase (R_oi) on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; acidity in external phase, 0.5 mol l⁻¹; R_we, 2; extraction time, 15 min; stirring speed, 250 rpm).

3.4 Effect of extraction process variables

3.4.1 Effect of acidity in external phase. The acidity in the external phase determines the speciation of cerium and affects the migration and extraction efficiency of Ce(IV). In order to explore the role of acidity in the external phase during the transport of Ce(IV) in the ELM system, experiments were performed with various sulphuric acid concentrations ranging from 0.2–1 mol l⁻¹. The obtained results are shown in Fig. 13. This figure shows that the extraction efficiency is strongly dependent on the acidity in the external phase solution. The transport of Ce(IV) from the external phase to the internal phase is mainly driven by the difference in H⁺ concentration between the external and internal phases. Theoretically, the greater the difference in H⁺ concentration between the external and internal phases, the easier the extraction.³ However, the extraction efficiency is extremely low at lower acidity, which can be attributed to the fact that Ce(IV) is easy to reduce and hydrolyse in low-acidity conditions, leading to the loss of Ce(IV). Thus, the Ce(IV) extraction efficiency decreases with increasing acidity from 0.5 to 1 mol l⁻¹. The maximum extraction efficiency is achieved at an acidity of approximately 0.4 mol l⁻¹. At higher acidity in the external phase, the properties of the surfactant are deteriorated,⁴⁵ leading to the destabilization of the emulsion and a diminution in extraction efficiency. Thus, 0.4–0.5 mol l⁻¹ was chosen as the optimal range of acidity in the external phase.

Fig. 13 Effect of acidity in the external phase on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; R_oi, 1.5; R_we, 3; extraction time, 15 min; stirring speed, 250 rpm).

3.4.2 Effect of volume ratio of external phase to membrane phase (R_we). The volume ratio of the external phase to the membrane phase (R_we) controls interfacial mass transfer and plays an important role in determining the efficiency of ELM. The effect of R_we on Ce(IV) extraction efficiency was studied, and the results are shown in Fig. 14. The figure shows that the extraction efficiency decreases gradually in the range of R_we from 1.5 to 6 and reaches 96% when R_we decreases below 2. The change in Ce(IV) concentration (C_Ce) in the external phase with time can be expressed as follows:^46,47where V_F is the volume of the feed phase, V_E is the total volume of emulsion droplets, A is the total surface area of emulsion droplets, and d is their average diameter. The area is given by A = V_E (6/d), and J_Ce is the transfer rate of Ce(IV) per unit surface area of emulsion droplets.

Fig. 14 Effect of volume ratio of external phase to membrane phase (R_we) on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; R_oi, 1.5; acidity in external phase, 0.5 mol l⁻¹; extraction time, 15 min; stirring speed, 250 rpm).

It is evident from eqn (6) that the Ce(IV) extraction efficiency is affected by the volume ratio of the external phase to the membrane phase and by the emulsion droplet diameter.⁴⁶ An increase in R_we corresponds to a decrease of the amount of emulsion in the external phase, leading to decreases in the amount of available globules and interfacial surface area per unit volume of the external phase. In addition, the transfer efficiency of Ce(IV) is weakened. Considering the cost, the volume ratio of external phase to membrane phase (R_we) of 2 was selected as the optimal value.

3.4.3 Effect of extraction time. The effect of extraction time (5–25 min) on the extraction efficiency was studied, as shown in Fig. 15. The results show that by increasing the extraction time from 5 to 15 min, the Ce(IV) extraction efficiency increases from 86 to 97%. When the extraction time further increases from 15 to 25 min, the extraction efficiency of Ce(IV) decreases slightly, possibly because some membrane droplets begin to break during the long period of stirring. Therefore, the extraction time of 15 min was selected.

Fig. 15 Effect of extraction time on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; R_oi, 1.5; R_we, 2; acidity in external phase, 0.5 mol l⁻¹; stirring speed, 250 rpm).

3.4.4 Effect of stirring speed. The stirring speed is a key factor in the rate of mass transfer through ELM.³⁸ In order to investigate the effect of stirring speed on extraction, the external solution and emulsion were mixed at stirring speeds ranging from 100 to 350 rpm. The obtained results are shown in Fig. 16. It is observed that increasing the stirring speed from 100 to 250 rpm increases the efficiency of extraction. This may be due to the fact that bigger globules disintegrate to form smaller globules at higher stirring speeds, increasing the interfacial area along with the mass transfer coefficient.⁴⁸ However, further increasing the stirring speed from 250 to 350 rpm decreases the extent of extraction because the higher stirring speed decreases the stability of the emulsion. In addition, an increase in shear on the emulsion phase induces the breakage of fragile emulsion droplets near the tip of the impeller.⁴⁹ Thus, there is a trade-off between the decrease in external mass transfer resistance and emulsion stability. Based on the results, 250 rpm was selected as the most appropriate stirring speed.

Fig. 16 Effect of stirring speed on the extraction of Ce(IV) (experimental conditions: D2EHPA concentration, 12% (v/v); Span 80 concentration, 2% (v/v); liquid paraffin concentration, 2% (v/v); hydrochloric acid concentration in internal phase, 5 mol l⁻¹; hydrogen peroxide concentration in internal phase, 0.02 mol l⁻¹; Ce(IV) concentration in external phase, 0.01 mol l⁻¹; R_oi, 1.5; R_we, 4; acidity in external phase, 0.5 mol l⁻¹; extraction time, 15 min).

3.5 The separation of Ce(IV)/RE(III)

The parameters that affect the extraction of the ELM process were experimentally investigated. The optimum conditions for Ce(IV) extraction can be summarized as follows: D2EHPA concentration, 12% (v/v); Span80 concentration, 2–3% (v/v); liquid paraffin concentration, 2–4% (v/v); hydrochloric acid concentration in the internal phase, 4–5 mol l⁻¹; hydrogen peroxide concentration, 0.02 mol l⁻¹; volume ratio of membrane phase to internal phase (R_oi), 1.5; acidity in external phase, 0.4–0.5 mol l⁻¹; volume ratio of external phase to membrane phase (R_we), 2; extraction time, 15 min; and stirring speed, 250 rpm. The extraction of Ce(IV) was conducted under the experimentally determined optimal conditions, and the obtained extraction efficiency of Ce(IV) was over 98%.

RE(III) elements such as La(III), Pr(III), Nd(III) and Sm(III) often accompany Ce(IV) in solution. Under the determined optimum conditions, the separation of Ce(IV) from other RE(III) was investigated.

The separation factor β_Ce/RE is calculated as follows:

where C_Ce and C_RE are the concentrations of Ce(IV) and RE(III) in the stripping and initial feed phases, respectively, in mol l⁻¹. The results of the extraction of Ce(IV) and RE(III) are given in Table 2. Clearly, the system is extremely selective for Ce(IV), and the separation factors are relatively high.

Table 2 Separation factors of Ce(IV) over RE(III) for various feed mixtures under the optimised conditions

Initial concentration in the feed solution (mol l^−1)	Extraction efficiency (%)		βCe/RE
	Ce(IV)	RE(III)
Ce 0.01 + La 0.01	98.25	0.56	175.45
Ce 0.01 + Pr 0.01	97.78	0.89	109.87
Ce 0.01 + Nd 0.01	98.54	0.75	131.39
Ce 0.01 + Sm 0.01	98.80	0.45	219.56

---

At present, many membrane separations for metals have been reported. Table 3 shows some previous studies on metal separation using different emulsion compositions. As can be seen, Span 80 and kerosene are the most widely used surfactant and diluent, respectively. The choices of extractant and internal phase depend on the types of metals being extracted. For rare earth metals, sulphuric acid is most commonly used as the internal phase.

Table 3 Compositions and conditions of ELM on metal extraction and separation

Solute	Surfactant	Extractant	Internal phase	Diluent	Reference
Nickel and cobalt	ECA 4360J	8-HQ	EDTA	Kerosene	6
Chromium, cobalt, nickel, copper and zinc	Span 80	TBP	(NH4)2CO3	Kerosene	48
Cadmium, zinc, cobalt and nickel	Span 80	Triotylamine	NH3	Kerosene	16
			Na2CO3
			(NH4)2CO3
			CHsCOONH4
Rare earth metals	2C18Δ^9GE	^tOct[4]CH2COOH	H2SO4	Toluene	50
Rare earth metals	Span 80, 2C18Δ^9GEC2QA	PC88A	H2SO4	n-Heptane	51
Rare earth metals	2C18Δ^9GE	PC88A	H2SO4	n-Heptane	52
	2C18Δ^9GEC2QA
Rare earth metals	2C18Δ^9GE	^tOct[1]CH2COOH	H2SO4	Toluene	53
		^tOct[4]CH2COOH
		^tOct[6]CH2COOH
Lead and cadmium	ECA5025	D2EHPA	HCl, H2SO4	Tetradecane	54
Zinc and copper	Span80	D2EHPA	H2SO4	Iso-dodecane	55
Cobalt and nickel	Span80	TOPO	NH3	Kerosene	56
Cobalt and nickel	ECA4360J	Alamine300	NH4OH	Kerosene	57
Platinum and palladium	Span80	TLA,TOMAC,TBP,TOPO,TIBPS	HCl	n-Heptane	58
	PX100
Trace elements: Cd, Co, Cu, Fe, Mn, Ni, Pb, Zn	Span80	D2EHPA,PC88A	HCl + H2SO4	Kerosene	59

---

3.6 Demulsification

The Ce(IV)-loaded emulsion obtained after extraction and separation must be demulsified to recover the metal ions and quantify the extraction efficiency.

The stripping efficiency is calculated by eqn (8):

where C_in is the final concentration of Ce(IV) in the internal phase (mol l⁻¹), C_exo is the initial concentration of Ce(IV) in the external phase (mol l⁻¹), C_ex is the final concentration of Ce(IV) in the external phase (mol l⁻¹), V_in is the final volume of the internal phase (ml), V_exo is the initial volume of the external phase (ml), and V_ex is the final volume of the external phase (ml).

Five experimental tests were carried out, and the results are illustrated in Fig. 17. A satisfactory mass balance was obtained, and the loss between the external phase and the recovery from the internal phase was within 3%.

Fig. 17 The stripping efficiency after demulsification (experimental conditions: optimum conditions).

4 Conclusions

In this study, the extraction of Ce(IV) from sulphuric acid solution by a emulsion liquid membrane comprised of D2EHPA dissolved in sulfonated kerosene as carrier containing Span80 as the emulsifier was investigated. Hydrochloric acid containing hydrogen peroxide was used as the stripping solution. The obtained results can be summarised as follows:

1. Stable W₁/O-type liquid membranes were prepared at an emulsifier speed of 5000 rpm for 20 min and could be applied for all subsequent experiments.

2. The optimum conditions for Ce(IV) extraction can be summarised as follows: D2EHPA concentration, 12% (v/v); Span80 concentration, 2–3% (v/v); liquid paraffin concentration, 2–4% (v/v); hydrochloric acid concentration in the internal phase, 4–5 mol l⁻¹; hydrogen peroxide concentration, 0.02 mol l⁻¹; volume ratio of membrane phase to internal phase (R_oi), 1.5; acidity in external phase, 0.4–0.5 mol l⁻¹; volume ratio of external phase to membrane phase (R_we), 2; extraction time, 15 min; and stirring speed, 250 rpm. The results demonstrate that among the studied parameters, the D2EHPA concentration, hydrogen peroxide concentration, acidity in the external phase, R_we and stirring speed play vital roles in Ce(IV) extraction.

3. Under the optimum operating parameters, the extraction efficiency of Ce(IV) is over 98%. The separation of Ce(IV) from RE(III) is realised, showing that the system is extremely selective for Ce(IV).

4. The results obtained demonstrate the validity of the ELM method, which represents an interesting advanced process for the extraction of Ce(IV) from sulphuric acid solution.

Acknowledgements

The financial aid from the following programs are gratefully acknowledged: the key program of National Natural Science Foundation of China (NSFC: 50934004), National Natural Science Foundation of China (51274061), Major State Basic Research Development Program of China (973 Program: 2012CBA01205), and Fundamental Research Supporting Project of Northeastern University (N110602006).

References

1. X. L. Ren, Q. F. Wei, S. R. Hu and S. J. Wei, Hydrometallurgy, 2010, 103, 205–210.
2. L. Jelinek, Y. Z. Wei and K. Mikio, J. Rare Earths, 2006, 24, 257–263.
3. W. Y. Wu, Rare Earth Metallurgy, Chemical Industry Press, Beijing, 2005, pp. 71, 112 and 163, in chinese.
4. D. B. Wu, Q. Zhang and B. Borong, Hydrometallurgy, 2007, 88, 210–215.
5. S. H. Yin, W. Y. Wu, X. Bian and F. Y. Zhang, Hydrometallurgy, 2013, 131–132, 133–137.
6. R. A. Kumbasar and S. Kasap, Hydrometallurgy, 2009, 95, 121–126.
7. R. M. Izatt, J. Inclusion Phenom. Macrocyclic Chem., 1997, 29, 197–220.
8. L. Brinchi, R. Germani, M. V. Mancini, G. Savelli and N. Spreti, Eur. J. Org. Chem., 2004, 1330–1335.
9. H. C. Visser, D. N. Reinhoudt and F. de Jong, Chem. Soc. Rev., 1994, 75–81.
10. N. M. Kocherginsky, Q. Yang and L. Seelam, Sep. Purif. Technol., 2007, 53, 171–177.
11. R. A. Bartsch and J. D. Way, ACS Symp. Ser., 2009, 642, 1–10.
12. P. Ramakul, U. Mooncluen, Y. Yanachawakul and N. Leepipatpiboon, J. Ind. Eng. Chem., 2012, 18, 1606–1611.
13. A. M. Neplenbroek, D. Bargeman and C. A. Smolders, J. Membr. Sci., 1992, 67, 121–132.
14. L. Guo, Y. H. Liu, C. Zhang and J. Chen, J. Membr. Sci., 2011, 372, 314–321.
15. N. N. Li, Separating hydrocarbons with liquid membranes, US Pat., 3, 410, 794, 1968.
16. R. A. Kumbasar, Hydrometallurgy, 2009, 95, 290–296.
17. M. T. A. Reis and J. M. R. Carvalho, J. Membr. Sci., 2004, 237, 97–107.
18. A. B. Lende, M. K. Dinker, V. K. Bhosale, S. P. Kamble, P. D. Meshramc and P. S. Kulkarni, RSC Adv., 2014, 4, 52316–52323.
19. Z. F. Guo, H. D. Su, R. Cai and X. L. Ma, RSC Adv., 2015, 7, 1860–1865.
20. B. Sengupta, R. Sengupta and N. Subrahmanyam, Hydrometallurgy, 2006, 84(1–2), 43–53.
21. J. Fang, M. Li and Z. Xu, Sep. Sci. Technol., 2003, 38, 3553–3574.
22. N. Othman, H. Mat and M. Goto, J. Membr. Sci., 2006, 282, 171–177.
23. D. He and M. Ma, Hydrometallurgy, 2000, 56, 157–170.
24. T. Oshima, T. Kakoi, F. Kubota, K. Ohto and M. Goto, Sep. Sci. Technol., 1998, 33, 1905–1917.
25. Z. N. Xie, Q. L. Chen and L. J. Zhao, Chin. J. Process Eng., 2013, 13, 197–201 , in chinese.
26. R. H. Zhang and L. Xiao, J. Chin. Rare Earth Soc., 1984, 2, 30–34.
27. M. Teramoto, T. Sakuramoto, T. Koyama, H. Matsuyama and Y. Miyake, Sep. Sci. Technol., 1986, 21, 229–250.
28. J. Tang and C. Wai, J. Membr. Sci., 1988, 35, 339–345.
29. A. M. Chaudry, A. Shahid and M. M. Tayyib, Sep. Sci. Technol., 1996, 31, 1309–1326.
30. R. Prakorn, P. Weerawat and P. Ura, Korean J. Chem. Eng., 2006, 23, 85–92.
31. L. Pei, L. M. Wang and Z. Y. Ma, Front. Environ. Sci. Eng., 2014, 8, 503–509.
32. G. X. Xu and C. Y. Yuan, Solvent extraction of rare earths, Science Press, Beijing, 1987, p. 127 and 128, in chinese.
33. D. Q. Li, Z. H. Wang, G. F. Zeng and Z. Y. Xue, J. Chin. Rare Earth Soc., 1984, 2, 9–19.
34. X. Q. Sun, Y. Ji, F. C. Hu, B. He, J. Chen and D. Q. Li, Talanta, 2010, 81, 1877–1883.
35. X. L. Lv, Inorganic peroxide chemical, Science and Technology Press, Beijing, 1987, in chinese.
36. E. wadsorth, F. K. Dukeand and C. A. Goetz, Anal. Chem., 1957, 29, 1824.
37. B. Sengupta, M. S. Bhakhar and R. Sengupta, Hydrometallurgy, 2007, 89, 311–318.
38. S. Chaouchi and O. Hamdaoui, Sep. Purif. Technol., 2014, 129, 32–40.
39. M. Chiha, M. H. Samar and O. Hamdaoui, Desalination, 2006, 194, 69–80.
40. R. S. Juang and K. H. Lin, Colloids Surf., A, 2004, 238, 43–49.
41. S. H. Lin, C. L. Pan and H. G. Leu, Chem. Eng. J., 2002, 87, 163–169.
42. H. R. Mortaheb, M. H. Amini, F. Sadeghian, B. Mokhtarani and H. Daneshyar, J. Hazard. Mater., 2008, 160, 582–588.
43. L. Yang, Z. Zhang, Y. Guo, X. Gao and H. Takeuchi, Sep. Purif. Technol., 2005, 47, 88–94.
44. R. S. Juang and K. H. Lin, Colloids Surf., A, 2004, 238, 43–49.
45. R. Sabry, A. Hafez, M. Khedr and A. El-Hassanin, Desalination, 2007, 212, 165–175.
46. R. A. Kumbasar and O. Tutkun, Hydrometallurgy, 2004, 75, 111–121.
47. K. Kondo, K. Kita, I. Koida, J. Irie and F. Nakashio, J. Chem. Eng. Jpn., 1979, 12, 203–209.
48. R. A. Kumbasar, Hydrometallurgy, 2010, 178, 875–882.
49. P. S. Kulkarni and V. V. Mahajan, J. Membr. Sci., 2012, 201, 123–135.
50. T. Kakoi, T. Oshima, T. Nishiyori, F. Kubota, M. Goto, S. Shinkai and F. Nakashio, J. Membr. Sci., 1998, 143, 125–135.
51. K. Uezu, S. Irie, O. Yoshimura, M. Goto and F. Nakashio, Trans. IchemE, 1997, 75, 513–518.
52. M. Goto, T. Kakoi, N. Yoshii, K. Kondo and F. Nakashio, Ind. Eng. Chem. Res., 1993, 32, 1681–1685.
53. T. Kakoi, T. Nishiyori, T. Oshima, F. Kubota, M. Goto, S. Shinkai and F. Nakashio, J. Membr. Sci., 1997, 136, 261–271.
54. B. J. Raghuraman, N. P. Tirmizi, B. S. Kim and J. M. Wiencek, Environ. Sci. Technol., 1995, 29, 979–984.
55. E. A. Fouad, Chem. Eng. Technol., 2008, 31, 370–376.
56. R. A. Kumbasar, Sep. Purif. Technol., 2009, 68, 208–215.
57. R. A. Kumbasar and O. Tutkun, Desalination, 2008, 224, 201–208.
58. T. Kakoi, M. Goto and F. Nakashio, J. Membr. Sci., 1996, 120, 77–88.
59. Y. Li, J. C. van Loon and R. R. Barefoot, Fresenius' J. Anal. Chem., 1993, 345, 467–470.

---