Investigation of intermolecular interactions of mixed extractants of quaternary phosphonium or ammonium chlorides and bis(2,4,4-ethylhexyl)phosphoric acid for metal separation

Abstract

Mixed extractants composed of di(2-ethylhexyl)phosphoric acid (P204) and ionic liquids (trihexyl(tetradecyl)phosphonium chloride (IL101) or methyltrioctylammonium chloride (A336)) for Zn(II)/Cu(II) separation were investigated. An antagonistic effect can be found for the extraction of both zinc and copper when mixing ionic liquids with P204, especially for Cu(II), thus the Zn(II)/Cu(II) separation can be greatly improved by adjusting the kinds and mole fractions of ionic liquids in the mixed extractants. Attenuated total reflection infrared (ATR-IR) and nuclear magnetic resonance (NMR) spectroscopies elucidate that the diverse intermolecular interactions between P204 and ionic liquids can be tailored by changing their compositions. Two-dimensional correlation analysis on IR spectra further extracts the overlapped structural information of the mixed extractants. And the dependence of metal extraction on intermolecular interactions has been elucidated.

---

1. Introduction

Solvent extraction is a common technique used to separate and recover valuable metals from various complex liquors and effluents.¹ However, the development and commercialization of any new extractant are extremely time-consuming and expensive. An alternative way to solve this problem is using specific mixtures of the commercially available extractants.² The mixed extractants exhibit potential advantages to achieve highly efficient separation of metal ions due to their special synergistic or antagonistic effects and independence of synthesis of new ligands, thus opening a new choice for extractive hydrometallurgy.^3–5

Various mixed extractants, consisting of two or more commercially available acidic, neutral and basic reagents, have been widely used for the separation and recovery of transition metals,^6,7 rare earth metals^4,8 and so on. Organophosphorus acids (Cyanex 272, D2EHPA, HEHEHP) exhibit greatly synergistic effects for vanadium extraction after mixing with primary amine N1923.⁹ For the Cyanex 272/Alamine 336 combination, both synergistic and antagonistic effects were found for the extraction of uranium(VI) and iron(III) in mixed sulfate/chloride media, respectively, thereby presenting an excellent selectivity of uranium over iron.¹⁰ For the recovery of spent lithium-ion battery, although an antagonistic effect was observed for both Co and Mn extraction with the mixed extractant of Versatic 10/D2EHPA, the separation factor of Mn over Co can be enhanced from 14.3 to 33.97.⁶

Recently, ionic liquids (ILs) have been applied into metal extraction as solvent or extractant because of their specific properties of high thermal stability, very low flammability and negligible vapor pressure.^11,12 Some novel extractants, mixed with ILs and acidic^13,14 or neutral¹⁵ ligands, seem to be quite exciting. Especially, the quaternary ammonium (e.g. Aliquat 336)^16,17 and quaternary phosphonium compounds (e.g. Cyphos IL101)^18–20 have attracted much attention for the extraction of anions and metal cations. For instance, the combination of Cyanex 272 with Aliquat 336, Cyphos IL101 or Cyphos IL167 could improve removal of boron from chloride solution.²¹ Also, the highly efficient separation of Zn/Cu can be achieved by the mixtures of Cyanex 272 and Aliquat 336 due to their synergistic effect.¹³ Moreover, Zhu et al. found that the synergistic extraction coefficients for Pr(III) can be adjusted by changing the kinds and ratio of commercial extractants (DEHEHP, TBP and TRPO) in the mixed extractants based on Aliquat 336,²² indicating that the synergistic or antagonistic effect for the mixed extractants based on ILs can be easily tuned for metal separation. In our previous work,²³ an evident improvement for Cu(II)/Ni(II) separation in ammoniacal media can be found by mixing Aliquat 336 with LIX84I, and it is noteworthy that the separation factor of Cu/Ni can be adjusted by the mole fraction of Aliquat 336. Therefore, as sharply increasing the demands of the metal recovery, utilizing the ILs-based mixed extractants has a broad prospect in the extraction and separation of polymetallic complex solutions.

Generally, the single extractants are well-known and understood, but the extractant mixtures can provide unique properties, new extraction mechanisms and organic-phase species. In the mixtures, the new added components could react with metal extracts to form more stable adducts or compete with metal ions to react with active components due to some special intermolecular interactions, thus presenting the desired extraction and separation behaviors.^24,25 Petrova found that the extraction of trivalent lanthanides from chloride media can be enhanced in the presence of TOPO, TPPO and TBP due to the coordination of the added phosphine oxide compounds with metal extracts.²⁶ Tkac et al. studied on the extraction of lanthanides by mixtures of D2EHPA and CMPO, and found that the antagonistic effect can be attributed to the formation of the D2EHPA–CMPO adduct through hydrogen bonding, which decreases the concentration of available CMPO for complexation with Eu(III).²⁴ However, there is few efforts to elucidate the intermolecular interactions of the mixed extractants based on ILs and their effect on extraction and separation of metal ions from a molecular level understanding.²⁷ Clearly, the insights into the intermolecular interactions will enable to tailor the performance of the mixed extractants to adapt to various hydrometallurgical and waste management processes.

Meanwhile, separation of zinc and copper in sulfuric media receives much attention during the recovery of zinc from polymetallic sulfide ores, spent zinc–copper catalyst and metallurgical wastewaters.²⁸ Solvent extraction seems to have great potential to selectively recover zinc. Although the commercially available extractants such as P204, Ionquest 801, Cyanex 272,^29,30 LIX 622,³¹ etc. have been used for the extraction of zinc from acidic sulfate solution, the separation for Zn/Cu by these acidic extractants is still difficult due to the coextraction of copper, especially at pH > 2.³² The present research is aimed to investigate the mixed extractants composed of di(2-ethylhexyl)phosphoric acid (P204) and trihexyl(tetradecyl)-phosphonium chloride (IL101) or methyltrioctylammonium chloride (A336) and their application for Zn/Cu separation. The intermolecular interactions of the mixed extractants have been in detail studied under the conditions of different structures or mole fractions of ILs by IR and NMR spectroscopies. Two-dimensional (2D) correlation technology is used to resolve the overlapped structural information from IR spectra. Moreover, the dependence of Zn/Cu separation on the intermolecular interactions of mixed extractants has been elucidated.

2. Experimental

2.1. Reagents and samples

Di(2-ethylhexyl)phosphoric acid (P204, purity of 97%) is purchased from Sinopharm. The ionic liquid extractants of trihexyl(tetradecyl)phosphonium chloride (Cyphos IL101, denoted as IL101, purity of 95%) and methyltrioctylammonium chloride (Aliquat 336, denoted as A336, purity of 99%) are supplied by the J&K (China). Toluene (purity of 99%) is used as a diluent. The extractants and diluents are used without further purification.

2.2. ATR-IR measurement and 2D correlation analysis

For the measurements of ATR-IR spectra, a series of mixed extractants were prepared with P204 and IL101 or A336, where the total extractant concentration is kept at 1 mol L⁻¹ and the mole fractions of ILs components (x_ILs, which is the concentration ratio of IL to total extractant) increases from 0 to 1 with an interval of 0.1. ATR-IR spectra were recorded at room temperature with a Nicolet 6700 FT-IR spectrometer. The attenuated total reflection through cell equipped with a ZeSe crystal of 45° incident angle and 12 reflections was used. The IR spectra with a resolution of 4 cm⁻¹ were collected by 128 scans in the range of 4000–650 cm⁻¹. Each sample was measured three times to check the spectral repeatability, and the average spectra were used for further analysis. The final spectra were obtained by removing the solvent background with the peak 1605 cm⁻¹ of toluene as the reference. In order to clearly clarify the overlapped structural information, the two-dimensional (2D) correlation technology is used to analyze the final IR spectra. The serial spectra of the mixed extractants after baseline correction were used to construct 2D correlation IR spectra using Matlab 7.0 software and the algorithm developed by Noda.³³

2.3. NMR measurements

The ¹H and ³¹P NMR spectra of the mixed extractants were recorded on a 400 MHz NMR spectrometer (AVANCE III). The ¹H and ³¹P NMR chemical shifts were measured with a sealed 2 mm capillary tube filled with the desired reference. Toluene-d₈ is used as the lock solvent and the internal standard for ¹H NMR (δ = 2.13 ppm). Triphenylphosphine (P(C₆H₅)₃, 99%) is for the outer standard for ¹P NMR (δ = −6.05 ppm).

2.4. Extraction experiments

The stock aqueous solutions containing 0.05 mol L⁻¹ Zn(II) or 0.02 mol L⁻¹ Cu(II) were prepared in sulfuric acid solution with pH of 4, respectively. Sodium sulfate was added into stock solutions to keep ionic strength at 0.6 mol L⁻¹. The fresh organic solutions were prepared with the different mole fractions of ILs (x_ILs = 0.1, 0.2, 0.3, 0.5, 0.7, 0.9 and 1) but keeping the total extractant concentration at 0.5 mol L⁻¹. Equal volumes (5 mL) of organic and aqueous phases were mechanically shaken for 30 minutes at room temperature for metal extraction equilibrium. After settling and phase separation, the loaded organic phases were taken for structural characterization. The metal concentration in the aqueous phases was determined with the inductively coupled plasma-atomic emission spectrometry (ICP-AES, Perkin Elmer 5300DV America) before and after extraction. The average value of the analyses for three times was taken in the present work. The metal concentration in the loaded organic phases was calculated by mass balance. The distribution coefficient (D) was calculated as the ratio of metal concentration in the organic phase to that in the aqueous phase at equilibrium. The separation factor of Zn/Cu (β) was calculated from the ratio of distribution coefficient of zinc and copper.

3. Results and discussion

3.1. Extraction and separation of metal ions

The extraction and separation behaviors of Zn(II) and Cu(II) with P204 in the absence and presence of ILs (A336 or IL101) with various mole fractions were shown in Fig. 1. The extraction behaviors of three individual extractants were also investigated for comparison. As shown in Fig. 1a and b, the extraction efficiency of Zn(II) rapidly increases for individual P204, A336 or IL101 with increasing their concentrations and reaches up to 68.14%, 98% and 85.77% at the concentration of 0.5 mol L⁻¹, respectively. Although the extraction ability of three individual extractants for Cu(II) is relatively weaker than Zn(II), the extraction percentage of Cu(II) can still reach to 19.7% for P204 alone, indicating that a certain amount of Cu(II) can be co-extracted with Zn(II). This behavior is unfavorable for their separation and recovery from various effluents. Meanwhile, it's interesting that both Zinc and copper can be extracted by quaternary phosphonium or ammonium chlorides from sulfate media. The Zn(II) and Cu(II) could be directly extracted as the form of MSO₄(R₄PCl)_n by ILs. In fact, the similar phenomena can be found for the extraction of both Co(II) and Ni(II) in sulfuric acid solution with Aliquat 336.³⁴ When IL components were added, the extraction efficiencies of both Zn(II) and Cu(II) have been depressed. Meanwhile, the degree of the antagonistic effect is changed with varying mole fraction of ILs (x_ILs). When concentration of ILs in the mixed systems is 0.1 mol L⁻¹ (x_ILs = 0.2), both Zn(II) and Cu(II) cannot be extracted with the mixed extractant and the extremely strong antagonistic effect has been showed. When increasing the mole fraction of ILs, the extraction percentage has been increased for Zn(II), while it is still kept at the low level for Cu(II). Moreover, the antagonistic effect for Zn(II) seems to be weakened but still strong for Cu(II). At x_ILs = 0.9, the antagonistic effect for Zn(II) almost can be ignored and 99.8% Zn(II) can be extracted, but the extraction percentage of Cu(II) is still kept lower than 10%.

Fig. 1 Extraction and separation behaviors of metal ions. (a) Extraction of Zn(II) and Cu(II) with P204; (b) extraction of Zn(II) with ILs alone or mixed with P204; (c) extraction of Cu(II) with ILs alone or mixed with P204; (d) separation behavior of zinc/copper with P204 in the absence and presence of IL at various mole fractions. Organic phase: total extractant concentration of 0.5 mol L⁻¹ for mixed systems; aqueous phase: 0.05 mol L⁻¹ Zn(II) or 0.02 mol L⁻¹ Cu(II) in acid sulfide solution with pH of 4, 1.0 mol L⁻¹ (NH₄)₂SO₄.

Although the metal extraction is depressed when P204 mixing with ILs, as shown in Fig. 1d, the separation factors of Zn/Cu can be greatly improved by the mixed extractants. For the individual P204 system, the separation factors of zinc over copper are kept at lower than 10. Interestingly, while adding the IL components, the Zn/Cu separation can be dramatically promoted, and the separation factor could be adjustable via changing the kind and mole fraction of ILs. It can be found that P204-IL101 system has a better separation performance compared to P204-A336 system. The separation factor of Zn/Cu increases from 9 to 2985 with increasing the mole fraction of IL101 from 0.2 to 0.8; whereas the separation factor increases from 7.6 to 546 when adding A336. Hence, the antagonistic effect of the mixed extractants shows a prominent advantage over the individual extractant for the separation of metal ions from the complex solutions. Therefore, the microscopic interaction information of the mixed extractants will be of great importance to understand the phenomena discussed above.

3.2. ATR-IR analysis of mixed extractants

As is known, the proton acidic extractants are prone to form the dimers or multimers by self-association.³⁵ As shown in Fig. 2a, the ATR-IR spectra of individual P204 shows the characteristic bands of the overlapped P–O and P–O–C groups around 1024 cm⁻¹ and the vibration peak of P=O group at 1228 cm⁻¹. The adsorption bands around 1689 cm⁻¹ for –OH group and 2300 cm⁻¹ for –POOH bending vibration can be ascribed to the formation of P204 dimers.³⁶ With decreasing P204 concentration from 1.0 mol L⁻¹ to 0.2 mol L⁻¹, there is no peak shift for both P–O and P=O bonds except for the decrease of peak intensity. Moreover, the peaks of –OH and –POOH groups remain no change, thus indicating that the strong self-association role still exists during the dilution process. This phenomenon can be also verified by the corresponding ¹H NMR in Fig. 2b. The chemical shift at δ = 12.47 ppm can be attributed to –OH group of the associated P204. If the P204 dimers are disassociated into free-state, the chemical shifts of hydrogen atoms will shift to up-field due to the increase of their electronic density. However, when continuously decreasing P204 concentration from 1.0 to 0.2 mol L⁻¹, there is no chemical shift for hydrogen atoms, thereby the association state of P204 is still stable even at a low concentration.

Fig. 2 ATR-IR (a) and ¹H NMR spectra (b) of different concentration P204. The inset shows the enlarged IR spectra in range of 1500–2500 cm⁻¹ for clarify.

The interaction study of P204 with ILs are shown in Fig. 3. Keeping total extractant concentration at 1 mol L⁻¹, after adding IL101 or A336, the significant spectral changes of P204 can be observed. With increasing x_ILs to 0.2, a negligible change has been taken for P=O and P–O groups. However, it can be found from Fig. 3b and d that the peaks at 2300–2500 cm⁻¹ and 1689 cm⁻¹ for P–OH group of P204 dimers have disappeared, indicating that the association states of P204 have been destroyed.

Fig. 3 ATR-IR spectra of P204-IL101 (a and b) and P204-A336 systems (c and d) at various mole fractions of ILs. Total extractant concentration of 1 mol L⁻¹; mole fraction of ILs from top to bottom (a and c) increases from 0 to 1 with an interval of 0.1.

Continuing increasing x_ILs to 0.7, the P–O peak at 1024 cm⁻¹ sharply decreases, accompanying with the presence of a new shoulder peak at 985 cm⁻¹. Meanwhile, the P=O peak at 1228 cm⁻¹ broadens and shifts to the higher wavenumbers, which suggesting that a hydrogen bonding interaction may occur between P204 and ILs. The similar phenomena are also observed for the other mixed ligand systems.^24,25 When increasing x_ILs toward 1, the shoulder peak at 985 cm⁻¹ becomes more obvious and the P=O peak shifts to 1265 cm⁻¹, indicating the dominant ion pairs/clusters of ILs could induce the interaction with P204 molecules.³⁷ Thus, it can be suggested that the intermolecular interactions of the mixed extractants can be adjusted via varying the kind and ratio of IL components.

Two-dimensional (2D) correlation analysis is widely used to improve the spectral resolution and can effectively resolve the overlapped spectra.^33,38 In order to further identify the structural information of the mixed extractants, the 2D correlation IR analysis for P–O group in the spectral range of 900–1100 cm⁻¹ is shown in Fig. 4. From the synchronous spectra, a strong positive autocorrelation peak can be observed due to the P204 concentration change. The overlapped information in the spectral range of 900–1100 cm⁻¹ for P–O can be resolved by the corresponding asynchronous spectra. As shown in Fig. 4b and d, two cross-peaks at (1056, 1024) and (1024, 985) can be found, showing that the P–O stretching vibration at 1024 cm⁻¹ has been split into three separated bands around 1024, 1056 and 985 cm⁻¹, respectively. The 2D asynchronous spectra indicate that there are three different states of P–O upon the perturbation of mole fraction of ILs, which have been overlapped in the one dimensional IR spectra. The peak at 1024 cm⁻¹ can be attributed to free P–O group, whereas the bands at 985 cm⁻¹ and 1056 cm⁻¹ could be assigned to the vibration of P–O groups associated with ion clusters/pairs of ILs.³⁷ According to the rule proposed by Noda,³⁸ the positive cross-peak (1056, 1024) and negative cross peak (1024, 985) showed that the peak intensity at 1056 cm⁻¹ and 985 cm⁻¹ varies prior to the peak at 1024 cm⁻¹, thus suggesting that the P204 molecules prefer to interact with IL rather than to keep dimers or monomers. Moreover, by comparison of the asynchronous spectra of P204-IL101 system, the cross peaks of P204-A336 system are elongated, indicating a more pronounced band shift induced by A336. Therefore, the intermolecular interaction of P204-A336 is stronger than that of P204-IL101. Thus, the separation factor of Zn/Cu in the presence of A336 is lower than that of IL101 due to a stronger interaction for P204-A336 system. Although the addition of IL components benefits for the dissociation of P204 dimers, a prominent antagonistic effect can be found around x_ILs of 0.2, indicating the formation of a special associated structure between P204 and ILs. Rey et al. revealed that the mixtures of HDEHP and TOPO are likely to form the core–shell aggregates.³⁹ When mixing the basic ILs with acidic P204 extractant, the chloride anion can be a polar center, thus a similar core–shell aggregate structure could be formed for the P204-ILs system through ion–dipole interaction.

Fig. 4 Contour maps of synchronous (a and c) and asynchronous (b and d) 2D correlation IR spectra for P204-IL101 system (a and b) and P204-A336 system (c and d).

3.3. NMR analysis of mixed extractants

The local environments of the mixed extractants during the component change can be further determined by ¹H and ³¹P NMR spectra. The chemical shift (δ) of ¹H NMR signals for the individual and mixed extractants are presented in Fig. 5. The chemical shift at 12.47 ppm is assigned to the proton hydrogen of –OH group of P204. The chemical shift at 2.83 ppm (Fig. 5a) and at 3.48 ppm (Fig. 5b) belongs to the ethyl group for IL101 (P–CH₂) and A336 (N–CH₂), respectively. With increasing the mole fraction of ILs, the ¹H chemical shift moves to the upfield for the hydrogen bonding on –OH group. It's known that the chemical shift can be always caused by several factors such as the hydrogen bonding or anisotropic effect.⁴⁰ In the case of ILs, the observation on ¹H upfield shifts of ethyl groups can verify the formation of hydrogen bonds in the mixed extractants. The observed ¹H up-field shifts for ethyl group should mainly result from the location of hydrogen atoms in the shielding cone in the aromatic solvent.⁴¹ Besides, as shown in Fig. 5c, the ¹H chemical shift on P–OH group of P204 shows two tuning point upon addition of ILs, which implying the different interactions between P204 and ILs with changing their mole fraction. When x_ILs < 0.3, the inner hydrogen bonds of P204 could be destroyed by the added IL components, and accompanying with the formation of core–shell aggregates between the dissociated P204 and chloride anion. Continuing increasing the mole fraction of ILs, the δ_OH for P204 shifts to the upfield with a steeper slope for 0.3 < x_ILs < 0.7, and both δ_P-CH2 for IL101 and δ_N-CH2 for A336 can shift to upfield, indicating the hydrogen bond interaction between P=O group and ethyl groups of the ILs. At x_ILs > 0.7, the ILs in the mixed extractants is prone to form ion clusters due to the predominant electrostatic interaction among IL components, hence resulting in the formation of hydrogen bond between P204 and IL clusters. Moreover, under the same mole fraction, the δ_OH of P204 shifts to a stronger component interaction for the former because the N atom has a stronger ability to draw electron density than P.

Fig. 5 ¹H NMR spectra of the mixed extractants at various IL mole fractions. (a) P204-IL101 system; (b) P204-A336 system; ¹H chemical shifts for P–OH groups of P204 (c) and for ethyl groups of ILs (d) plotted against the mole faction of ILs. The total extractant concentration is 1 mol L⁻¹.

In addition, the intermolecular interaction of P204-ILs systems can be further justified by ³¹P NMR analysis. As shown in Fig. 6, the ³¹P chemical shift for individual P204 and IL101 can be found at δ = 0.68 ppm and at δ = 32.5 ppm, respectively. For the P204-A336 system, it only exhibits the P chemical shift of P204 because of A336 without P atom. In the P204-IL101 system, it can be found that the P chemical shift of IL101 moves to the downfield as changing the mole fraction of IL101, but presents a turning point at x_ILs = 0.3, and then moves to the upfield. Thus, the diverse intermolecular interactions should exist in the mixed extractants with varying IL components. As seen in Fig. 6c, the δ values showed a slight down shifts towards downfield at x_ILs < 0.3, indicating that the electronic density on P of IL101 has been slightly increased due to the ion–dipole interaction between IL101 and P204. However, when x_ILs > 0.3, the IL101 can interact with P204 via the formation of hydrogen bond, thus causing a pronounced downfield shifts for P atom due to the ethyl of ILs as electron donor. But in the IL101-rich region (x_ILs > 0.7), the slow downfield shifts indicate that the interaction between P204 and IL101 is weaken, because the ion clusters are the dominant form for IL101 in the mixed extractants. Meanwhile, the chemical shifts for P on P204 move toward upfield due to the formation of hydrogen bond in the mixed extractants. Therefore, combining with the 2D IR analysis, it can be verified that the species and their structures in the mixed extractants can be varied with changing the IL ratio, thus achieving the adjustable separation of metal ions.

Fig. 6 ³¹P NMR spectra (a and b) and P chemical shifts (c and d) for P204-IL101 (a and c) and P204-A336 (b and d) systems at various IL mole fractions.

3.4. Dependence of zinc extraction on intermolecular interaction

Generally, per zinc ion can coordinate with two P204 molecules through bonding with the P=O and P–OH groups.^42,43 The added IL components could compete with Zn(II) to react with P204 through molecular interaction, thus changing the extraction performance of P204. The ATR-IR spectra of the mixed extractants before and after Zn(II) extraction were shown in Fig. 7. At x_ILs = 0.2, there is negligible shift for P=O peak at 1248 cm⁻¹ and P–OH peak at 1024 cm⁻¹ before and after extraction, indicating that zinc(II) is hardly extracted into organic phase at this specific mole fraction. This behavior is attributed to the formation of core–shell structure between P204 and ILs through ion–dipole interaction, in which the available functional groups of P204 are confined in the core center. Meanwhile, the core–shell structure could decrease the polarity of organic phases, thus impede the extraction reaction at liquid–liquid interface.⁴⁴ Therefore, both zinc and copper have the minimum extraction efficiency. At x_ILs = 0.5, the intensity of the shoulder peak at 985 cm⁻¹ decreases and a slight red shift for P=O peak is taken at 1265 cm⁻¹. Moreover, a new peak appears around 1150–1200 cm⁻¹, which is attributed to the coordination between P=O groups with Zn(II).⁴² Hence, the hydrogen bond between P204 and ion pairs has been destroyed during the extraction process since the hydrogen bonding is weaker than covalent bond, and the antagonistic effect of the mixed extractants has been weakened. At x_ILs = 0.7, the intensity of the shoulder peak at 985 cm⁻¹ decreases after extraction, accompanying with the increase of peak at 1024 cm⁻¹. Furthermore, a further red shift for P=O peak around 1265 cm⁻¹ was observed and the absorption intensity become more obvious. It can be deduced that the molecular bonding between P204 and ILs has been weaken by the electrostatic interaction of the predominant ion clusters. Hence, there are more released active groups of P204 to react with Zn(II), and the antagonism effect of the mixed extractants becomes weaker in the region of rich ILs. However, because the bonding between copper and ligands could be weaker than that of zinc, the antagonism effect is always stronger for copper, thereby improving the separation of Zn/Cu.

Fig. 7 ATR-IR spectra of mixed extractants before and after Zn(II) extraction.

3.5. Interaction models of mixed extractants

The structural studies clearly reveal that the intermolecular interactions between P204 and ILs can be adjusted by changing the composition of mixed extractants, thus causing the different degree of antagonistic effect on extraction process to achieve an adjustable separation of metal ions. Based on the discussion above, the proposed intermolecular interaction models of mixed extractants at different mole fractions of ILs are given in Scheme 1. At x_ILs < 0.3, the P204 dimers can be destroyed by the added ILs, thus a core–shell structure can be formed between the dissociated P204 molecules and the separated ion pairs through ion–dipole interaction, which is unfavorable for metal extraction. When 0.3 < x_ILs < 0.7, the hydrogen bond interaction between P204 and ILs is predominant because both IL101 and A336 are prone to form ion pairs. Therefore, the intermolecular hydrogen bond can be gradually weakened as increasing the mole fraction of ILs. Although the formed hydrogen bond can depress metal extraction, the separation results indicate that this antagonism effect will benefit for the Zn/Cu separation. In the ILs-rich region (x_ILs > 0.7), the ionic liquids have the stronger electrostatic interaction between anion and cation to form ion clusters. The hydrogen bonds of mixed extractants could be more easily broken by this ion clusters. Thus, the separation of zinc/copper can be achieved through the adjustable intermolecular interactions of mixed extractants based on ionic liquids.

Scheme 1 The proposed intermolecular interaction models of mixed extractants at different mole fractions of ILs.

4. Conclusions

The mixed extractants containing P204 and ionic liquids have been used for the extraction and separation of Zn and Cu. Although an antagonistic effect can be found for the extraction of both zinc and copper when adding A336 or IL101 into P204, the Zn(II)/Cu(II) separation can be greatly improved by adjusting the mole fractions of ionic liquids in the mixed extractants. The separation factor of Zn/Cu increases from 9 to 2985 with increasing mole fraction of IL101 from 0.2 to 0.8; whereas the separation factor increases from 7.6 to 546 for the P204-A336 system. 2D correlation IR analysis resolve three overlapped structural characters of P204 in the mixed extractants. The IR and NMR spectroscopies clearly reveal the diverse intermolecular interactions between P204 and ILs, and which can be adjusted by changing the composition of mixed extractants. Moreover, the dependence of metal extraction on various intermolecular interactions has been elucidated.

Acknowledgements

This work was financially supported by the National Basic Research Program of China (No. 2014CB643401), National Natural Science Foundation of China (Nos. 51134007 and 51304244), China Postdoctoral Science Foundation (2014M552152) and Hunan Provincial Natural Science Foundation of China (2015JJ3154).

References

1. A. M. Wilson, P. J. Bailey, P. A. Tasker, J. R. Turkington, R. A. Grant and J. B. Love, Chem. Soc. Rev., 2014, 43, 123–134.
2. D. Flett, J. Chem. Technol. Biotechnol., 1999, 74, 99–105.
3. W. He, W. Liao, C. Niu and D. Li, Sep. Purif. Technol., 2008, 62, 674–680.
4. A. Chagnes, C. Fosse, B. Courtaud, J. Thiry and G. Cote, Hydrometallurgy, 2011, 105, 328–333.
5. D. Beltrami, A. Chagnes, M. Haddad, A. Varnek, H. Mokhtari, B. Courtaud and G. Cote, Hydrometallurgy, 2013, 140, 28–33.
6. S.-H. Joo, D. Shin, C. Oh, J.-P. Wang and S. M. Shin, J. Power Sources, 2016, 305, 175–181.
7. N. A. Grigorieva and I. Y. Fleitlikh, Hydrometallurgy, 2013, 138, 71–78.
8. Y. Liu, H. S. Jeon and M. S. Lee, Hydrometallurgy, 2014, 150, 61–67.
9. D. Jiang, N. Song, S. Liao, Y. Lian, J. Ma and Q. Jia, Sep. Purif. Technol., 2015, 156, 835–840.
10. J. E. Quinn and K. H. Soldenhoff, Hydrometallurgy, 2015, 152, 7–12.
11. V. T. Nguyen, J.-c. Lee, J. Jeong, B.-S. Kim, G. r. Cote and A. Chagnes, Ind. Eng. Chem. Res., 2015, 54, 1350–1358.
12. J. Castillo, M. T. Coll, A. Fortuny, P. Navarro Donoso, R. Sepúlveda and A. M. Sastre, Hydrometallurgy, 2014, 141, 89–96.
13. B. Pospiech and A. Chagnes, Sep. Sci. Technol., 2015, 50, 1302–1309.
14. B. K. Tait, Solvent Extr. Ion Exch., 1992, 10, 799–809.
15. B. Pospiech, Sep. Sci. Technol., 2014, 49, 1706–1712.
16. G. Helle, C. Mariet and G. Cote, Talanta, 2015, 139, 123–131.
17. B. Wionczyk and W. Apostoluk, Hydrometallurgy, 2005, 78, 116–128.
18. T. Vander Hoogerstraete, S. Wellens, K. Verachtert and K. Binnemans, Green Chem., 2013, 15, 919–927.
19. S. Wellens, B. Thijs and K. Binnemans, Green Chem., 2012, 14, 1657–1665.
20. P. Rybka and M. Regel-Rosocka, Sep. Sci. Technol., 2012, 47, 1296–1302.
21. A. Fortuny, M. Coll and A. Sastre, Sep. Sci. Technol., 2012, 97, 137–141.
22. M. Zhu, J. Zhao, Y. Li, N. Mehio, Y. Qi, H. Liu and S. Dai, Green Chem., 2015, 17, 2981–2993.
23. J. Hu, S. Wang, H. Hu, H. Li, K. Peng, D. Liu, Q. Liu and Q. Chen, Sep. Purif. Technol., 2013, 118, 828–834.
24. P. Tkac, G. F. Vandegrift, G. J. Lumetta and A. V. Gelis, Ind. Eng. Chem. Res., 2012, 51, 10433–10444.
25. R. J. Ellis, Y. Meridiano, J. Muller, L. Berthon, P. Guilbaud, N. Zorz, M. R. Antonio, T. Demars and T. Zemb, Chem.– Eur. J., 2014, 20, 12796–12807.
26. M. A. Petrova, V. B. Kurteva and L. A. Lubenov, Ind. Eng. Chem. Res., 2011, 50, 12170–12176.
27. D. Cholico-Gonzalez, M. Avila-Rodriguez, G. Cote and A. Chagnes, J. Mol. Liq., 2013, 187, 165–170.
28. M. K. Jha, V. Kumar and R. J. Singh, Resour., Conserv. Recycl., 2001, 33, 1–22.
29. D. Pereira, S. Rocha and M. Mansur, Sep. Purif. Technol., 2007, 53, 89–96.
30. L. R. Gouvea and C. A. Morais, Miner. Eng., 2010, 23, 492–497.
31. G. Owusu, Hydrometallurgy, 1999, 51, 1–8.
32. A. D. d. Souza, P. d. S. Pina, V. A. Leão, C. A. d. Silva and P. d. F. Siqueira, Hydrometallurgy, 2007, 89, 72–81.
33. I. Noda, J. Am. Chem. Soc., 1989, 111, 8116–8118.
34. A. A. Nayl, J. Hazard. Mater., 2010, 173, 223–230.
35. R. K. Biswas, R. A. Banu and M. N. Islam, Hydrometallurgy, 2003, 69, 157–168.
36. D. F. Peppard and J. R. Ferraro, J. Inorg. Nucl. Chem., 1959, 10, 275–288.
37. Y. Zheng, N. Wang, J. Luo, Y. Zhou and Z. Yu, Phys. Chem. Chem. Phys., 2013, 15, 18055–18064.
38. I. Noda, J. Mol. Struct., 2014, 1069, 3–22.
39. J. Rey, S. Dourdain, L. Berthon, J. Jestin, S. Pellet-Rostaing and T. Zemb, Langmuir, 2015, 31, 7006–7015.
40. B.-M. Su, S. Zhang and Z. C. Zhang, J. Phys. Chem. B, 2004, 108, 19510–19517.
41. B. Dalai, S. K. Dash, S. K. Singh and B. B. Swain, J. Mol. Liq., 2015, 208, 151–159.
42. B. S. Morais and M. B. Mansur, Hydrometallurgy, 2004, 74, 11–18.
43. D. D. Pereira, S. D. F. Rocha and M. B. Mansur, Sep. Purif. Technol., 2007, 53, 89–96.
44. H. Freiser, Chem. Rev., 1988, 88, 611–616.

---