Dalton Transactions View Article Online
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
PAPER
Cite this: Dalton Trans., 2014, 43, 1630
View Journal | View Issue
Structural properties of the inner coordination sphere of indium chloride complexes in organic and aqueous solutions† Hirokazu Narita,*a Mikiya Tanaka,a Hideaki Shiwaku,b Yoshihiro Okamoto,b Shinichi Suzuki,b Atsushi Ikeda-Ohnob and Tsuyoshi Yaitab The nature of the inner coordination sphere of In3+ present in both the organic and aqueous solutions during the solvent extraction of In3+ from an aqueous HCl solution with tri-n-octyl amine (TOA) was investigated by In K-edge XAFS. This information was then used to clarify the details of the extraction properties of indium chloride anion complexes with TOA. In aqueous HCl solution (0.1–10 M), In3+ exists as octahedral anion complexes, [InCln(H2O)6−n]3−n (n ≥ 4); the [InCl6]3− complex is dominant at 10 M HCl. The extraction of In3+ from HCl solution with TOA was performed using two kinds of diluents: nitrobenzene (NB) or n-dodecane (DD), which contained 20 vol% of 2-ethylhexanol as an additive. The stoichiometric composition of the extracted complexes, which is estimated from the distribution ratios of In3+, is affected by the diluents and the HCl concentration of the aqueous phase; the apparent values of TOA/In3+ in the extracted complex are 3 for DD–1 M HCl (diluent–aqueous phase) and DD–5 M HCl, 2
Received 10th September 2013, Accepted 30th October 2013 DOI: 10.1039/c3dt52474d www.rsc.org/dalton
for NB–1 M HCl and NB–5 M HCl, and 1 for NB–10 M HCl. The EXAFS analysis of these extracted complexes shows that the In3+ has ∼4 Cl− at ∼2.336 Å and no H2O in the inner coordination sphere; additionally, the shape of the XANES suggests that their coordination geometry is tetrahedral. Therefore, the same tetrahedral [InCl4]− complex is formed during the extraction in spite of the variation in the stoichiometric composition (TOA/In3+ = 1–3) of the extracted complexes.
Introduction Because of the increased importance of indium in industrial use (e.g., indium tin oxide (ITO) for electronic products), the recycling of indium from indium-containing scrap has received much attention in recent years.1 For indium recovery, solvent extraction is one of the most versatile methods. Thus, many extraction studies have been carried out using various indium extraction reagents: acidic organic reagents, solvating organic reagents and basic organic reagents.2 These studies mainly concerned the extraction equilibrium, while structural information on the extracted complexes has been limited.
a Research Institute for Environmental Management Technology, National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan. E-mail: [email protected]; Fax: +81-29-861-8481; Tel: +81-29-861-8486 b Quantum Beam Science Directorate, Japan Atomic Energy Agency (JAEA), 1-1-1 Koto, Sayo, Hyogo 679-5148, Japan † Electronic supplementary information (ESI) available: Figures of In EXAFS spectra and corresponding FTs for In3+ in additional HCl solutions and the extracted complexes, and their structural parameters. See DOI: 10.1039/c3dt52474d
1630 | Dalton Trans., 2014, 43, 1630–1635
In hydrometallurgical processings, chloride media are often employed due to the high solubility of many metal chloride complexes in water.3 Therefore, the knowledge of In3+ chloride complexing in solution under different conditions is required in terms of selecting appropriate extractants, since the extractability of In3+ depends strongly upon the structure of In3+ complexes in the aqueous solution. However, systematic information about the coordination of In3+ complexes in concentrated HCl solutions is still scarce. Most of the early studies used solvent extraction, ion exchange methods, and polarographic and potentiometric measurements to obtain the stability constants of In3+ chloride complexes. They found that In3+ exists as [InCln(H2O)6−n]3−n complexes (n ≤ 4), while the stability constants are very scattered;4,5 exceptionally, Ferri et al. suggested the presence of penta- and hexachloroindate in addition to [InCln(H2O)6−n]3−n complexes (n ≤ 4) at ≤5 mol kg−1 Cl− from a potentiometric investigation.6 The geometry of In3+ in aqueous solutions has been determined by spectroscopic studies. Jarv et al. found evidence for octahedral [InCln(H2O)6−n]3−n: 1≤ n ≤ 4 in the range of 0.48 ≤ [Cl−]/[In3+] ≤ 15.9 by an investigation of the Raman band profiles.7 Seward et al. revealed that the dominant species of In3+ in 0.10 M HCl solution changes from the
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
Dalton Transactions
octahedrally coordinated [InCln(H2O)6−n]3−n complexes at 25 °C to the tetrahedrally coordinated [InCl4]− complex at 300 °C using EXAFS.8 The extraction of metal anion complexes from acidic aqueous solution is extensively performed with an ion-pair type extractant; especially, amine compounds are very popular,9 since they are readily protonated. According to a review,2 In3+ extraction in hydrochloric acid solution with amine compounds mostly involves the ion-pair extraction of [InCl4]− with the protonated amine compounds; some authors suggested the predominance of [InCl4]− or [InCl5]2− in the organic phase from IR or Raman spectra. Other than for the amine extractants, only a few structural investigations have been conducted for the extracted complexes. The Raman spectra of the In3+ species extracted with ether from an In3+– HCl solution is attributed to the tetrahedral [InCl4]− ion.10 Haraguchi et al. investigated the In3+ complexes extracted with methyl isobutyl ketone, ethyl ether, isopropyl ether, n-butyl acetate, cyclohexanone and ethyl acetoacetate using 115In NMR spectroscopy, and found the formation of the tetrahedral [InCl4]− ion in all the systems.11 However, these studies did not attempt to correlate the coordination structures and the extraction properties. In this work, we used a tertiary amine compound: tri-n-octylamine (TOA), which is one of the most popular extractants for industrial use that can readily extract metal anion complexes by ion-pair formation. We report herein the extraction behavior of In3+ from HCl solution with TOA diluted in nitrobenzene or n-dodecane, which contains 20 vol% of 2-ethylhexanol as an additive to prevent a third phase formation during the extraction of In3+, and the structural properties of In3+ chloride solution complexes in concentrated HCl solutions and the extracted species using the EXAFS method. EXAFS data display the average structure of all the species existing in a solution; therefore, we investigated systematic changes in the coordination structure of In3+. In addition, we discuss the relationship between the inner coordination sphere structure and the extraction behavior of In3+.
Experimental Sample preparation TOA was purchased from Wako Chemical and used without further purification. The other chemicals were commercially available (reagent grade). Deionized water was used for all solution preparations. A stock solution of In3+ in HCl (0.1–10 M) was prepared from InCl3·4H2O. For XAFS measurements, the solutions were transferred to polyethylene or glass cells, which were then tightly sealed; samples of the extracted complex in the organic solution were prepared by the solvent extraction method detailed below. Solvent extraction Two organic solutions with different dielectric constants, nitrobenzene or n-dodecane, were used as diluents in which 20 vol%
This journal is © The Royal Society of Chemistry 2014
Paper
of 2-ethylhexanol was added. The organic solution (TOA dissolved in the diluents) was pre-equilibrated with the same volume of HCl solution in the absence of In3+. An aliquot of the pre-equilibrated organic solution and the same volume of the HCl solution containing In3+ ([In]: 0.001 M for the extractability measurements; 0.1 M for the XAFS samples) were shaken for 60 minutes in a glass tube at room temperature (23 ± 2 °C), and then centrifuged. The metal contents in the aqueous phase were measured by ICP-AES (Horiba ULTIMA2). The concentrations in the organic phase were calculated by the mass balance of metals before and after the extraction. When using n-dodecane (with 2-ethylhexanol), a third phase was formed on contact with 10 M HCl solution; therefore, in this case, no sample was available for XAFS measurements. XAFS measurements The XAFS measurements were carried out at the BL11XU of the SPring-8. The operating energy and the ring current were 8 GeV and ∼99 mA. The SPring-8 synchrotron radiation was monochromatized using Si(111) crystal monochromators. The In K-edge (27 940 eV) X-ray absorption spectra were collected in transmission mode using two ion chambers (both filled with N2) at room temperature (∼25 °C). Energy calibration was performed using an In metal foil. XAFS data analysis The experimental data were analysed using the EXAFS data analysis software package WinXAS, Ver. 3.1.12 The EXAFS oscillation χ(k) was weighted by k3 and windowed using a Gaussian window function. The EXAFS data were fitted using theoretical phases and amplitudes were calculated using the program FEFF8.13 The crystal structures of tetrahedral tetrachloroindate [InCl4]−,14 and octahedral diaquatetrachloroindate [InCl4(H2O)2]− 15 and hexachloroindate [InCl6]3− 16 were used for the FEFF calculations as model compounds. All the interactions were modeled using single scattering (SS) and multiple scattering (MS) paths based on the model compounds; however, the MS paths were ignored since their contribution was not significant in the systems studied. For In3+ aqueous solutions, the coordination number, CN, of In3+ is known to be six;17 therefore, the CN was held constant at 6 in the aqueous solutions, and the amplitude reduction factor, S02, was determined by the use of In3+ in 10 M HCl solutions. The obtained S02 value (0.95) was fixed during all the fits. The Debye–Waller factor squared, σ2, of the In–Cl and –O(H2O) shells for the aqueous samples was determined using typical samples and all subsequent fits were performed by the use of the σ2 values so obtained: In3+ in 10 M HCl (highest [HCl]) for In–Cl, and In3+ in 0.1 M HCl (lowest [HCl]) for In–O(H2O). For the other shells, the CN, bond distance, r, and the shift in threshold energy, ΔE0 (the same ΔE0 was used for each shell), were allowed to vary in the fit. Error values were determined by the F-test18 (95% confidence). The fitting quality was checked using the R-factor: {∑|k3χ(k)obs − k3χ(k)calc|/∑|k3χ(k)obs|} × 100.
Dalton Trans., 2014, 43, 1630–1635 | 1631
View Article Online
Paper
Results and discussion
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
In3+ in HCl solution The dominant species of In3+ ions in concentrated HCl solutions were first investigated. Fig. 1 and S1, ESI† show the raw In K-edge EXAFS and corresponding Fourier transforms (FTs) for the In3+ ion in 0.1–10 M HCl solutions. In the FTs, the shape of the intense peak (R = 1–2.5 Å) gradually changes with the HCl concentration; the peak for 0.1 M HCl is broader than that for 10 M; the highest point shifts to higher R with an increase in the HCl concentration. This intense peak corresponds to the In–O/–Cl correlation in these solutions. In order to determine the presence or absence of O atoms, 1-shell fits (Cl− only) and 2-shell fits (Cl− and O) were performed on this peak and then the reasonableness of the structural parameters and the R-factor were checked. For the spectra of 0.1–8 M HCl, the 2-shell fits give reasonable parameters (see Tables 1 and S1, ESI†) and lower R-factors (e.g., 9.3→4.7 for 6 M HCl, 6.5→4.4 for 8 M HCl). This is consistent with the broader peak shape, especially on the lower side, than that for the spectrum of 10 M HCl. In contrast, the CN value of the In–O shell for 10 M HCl is
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
PAPER
Cite this: Dalton Trans., 2014, 43, 1630
View Journal | View Issue
Structural properties of the inner coordination sphere of indium chloride complexes in organic and aqueous solutions† Hirokazu Narita,*a Mikiya Tanaka,a Hideaki Shiwaku,b Yoshihiro Okamoto,b Shinichi Suzuki,b Atsushi Ikeda-Ohnob and Tsuyoshi Yaitab The nature of the inner coordination sphere of In3+ present in both the organic and aqueous solutions during the solvent extraction of In3+ from an aqueous HCl solution with tri-n-octyl amine (TOA) was investigated by In K-edge XAFS. This information was then used to clarify the details of the extraction properties of indium chloride anion complexes with TOA. In aqueous HCl solution (0.1–10 M), In3+ exists as octahedral anion complexes, [InCln(H2O)6−n]3−n (n ≥ 4); the [InCl6]3− complex is dominant at 10 M HCl. The extraction of In3+ from HCl solution with TOA was performed using two kinds of diluents: nitrobenzene (NB) or n-dodecane (DD), which contained 20 vol% of 2-ethylhexanol as an additive. The stoichiometric composition of the extracted complexes, which is estimated from the distribution ratios of In3+, is affected by the diluents and the HCl concentration of the aqueous phase; the apparent values of TOA/In3+ in the extracted complex are 3 for DD–1 M HCl (diluent–aqueous phase) and DD–5 M HCl, 2
Received 10th September 2013, Accepted 30th October 2013 DOI: 10.1039/c3dt52474d www.rsc.org/dalton
for NB–1 M HCl and NB–5 M HCl, and 1 for NB–10 M HCl. The EXAFS analysis of these extracted complexes shows that the In3+ has ∼4 Cl− at ∼2.336 Å and no H2O in the inner coordination sphere; additionally, the shape of the XANES suggests that their coordination geometry is tetrahedral. Therefore, the same tetrahedral [InCl4]− complex is formed during the extraction in spite of the variation in the stoichiometric composition (TOA/In3+ = 1–3) of the extracted complexes.
Introduction Because of the increased importance of indium in industrial use (e.g., indium tin oxide (ITO) for electronic products), the recycling of indium from indium-containing scrap has received much attention in recent years.1 For indium recovery, solvent extraction is one of the most versatile methods. Thus, many extraction studies have been carried out using various indium extraction reagents: acidic organic reagents, solvating organic reagents and basic organic reagents.2 These studies mainly concerned the extraction equilibrium, while structural information on the extracted complexes has been limited.
a Research Institute for Environmental Management Technology, National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan. E-mail: [email protected]; Fax: +81-29-861-8481; Tel: +81-29-861-8486 b Quantum Beam Science Directorate, Japan Atomic Energy Agency (JAEA), 1-1-1 Koto, Sayo, Hyogo 679-5148, Japan † Electronic supplementary information (ESI) available: Figures of In EXAFS spectra and corresponding FTs for In3+ in additional HCl solutions and the extracted complexes, and their structural parameters. See DOI: 10.1039/c3dt52474d
1630 | Dalton Trans., 2014, 43, 1630–1635
In hydrometallurgical processings, chloride media are often employed due to the high solubility of many metal chloride complexes in water.3 Therefore, the knowledge of In3+ chloride complexing in solution under different conditions is required in terms of selecting appropriate extractants, since the extractability of In3+ depends strongly upon the structure of In3+ complexes in the aqueous solution. However, systematic information about the coordination of In3+ complexes in concentrated HCl solutions is still scarce. Most of the early studies used solvent extraction, ion exchange methods, and polarographic and potentiometric measurements to obtain the stability constants of In3+ chloride complexes. They found that In3+ exists as [InCln(H2O)6−n]3−n complexes (n ≤ 4), while the stability constants are very scattered;4,5 exceptionally, Ferri et al. suggested the presence of penta- and hexachloroindate in addition to [InCln(H2O)6−n]3−n complexes (n ≤ 4) at ≤5 mol kg−1 Cl− from a potentiometric investigation.6 The geometry of In3+ in aqueous solutions has been determined by spectroscopic studies. Jarv et al. found evidence for octahedral [InCln(H2O)6−n]3−n: 1≤ n ≤ 4 in the range of 0.48 ≤ [Cl−]/[In3+] ≤ 15.9 by an investigation of the Raman band profiles.7 Seward et al. revealed that the dominant species of In3+ in 0.10 M HCl solution changes from the
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
Dalton Transactions
octahedrally coordinated [InCln(H2O)6−n]3−n complexes at 25 °C to the tetrahedrally coordinated [InCl4]− complex at 300 °C using EXAFS.8 The extraction of metal anion complexes from acidic aqueous solution is extensively performed with an ion-pair type extractant; especially, amine compounds are very popular,9 since they are readily protonated. According to a review,2 In3+ extraction in hydrochloric acid solution with amine compounds mostly involves the ion-pair extraction of [InCl4]− with the protonated amine compounds; some authors suggested the predominance of [InCl4]− or [InCl5]2− in the organic phase from IR or Raman spectra. Other than for the amine extractants, only a few structural investigations have been conducted for the extracted complexes. The Raman spectra of the In3+ species extracted with ether from an In3+– HCl solution is attributed to the tetrahedral [InCl4]− ion.10 Haraguchi et al. investigated the In3+ complexes extracted with methyl isobutyl ketone, ethyl ether, isopropyl ether, n-butyl acetate, cyclohexanone and ethyl acetoacetate using 115In NMR spectroscopy, and found the formation of the tetrahedral [InCl4]− ion in all the systems.11 However, these studies did not attempt to correlate the coordination structures and the extraction properties. In this work, we used a tertiary amine compound: tri-n-octylamine (TOA), which is one of the most popular extractants for industrial use that can readily extract metal anion complexes by ion-pair formation. We report herein the extraction behavior of In3+ from HCl solution with TOA diluted in nitrobenzene or n-dodecane, which contains 20 vol% of 2-ethylhexanol as an additive to prevent a third phase formation during the extraction of In3+, and the structural properties of In3+ chloride solution complexes in concentrated HCl solutions and the extracted species using the EXAFS method. EXAFS data display the average structure of all the species existing in a solution; therefore, we investigated systematic changes in the coordination structure of In3+. In addition, we discuss the relationship between the inner coordination sphere structure and the extraction behavior of In3+.
Experimental Sample preparation TOA was purchased from Wako Chemical and used without further purification. The other chemicals were commercially available (reagent grade). Deionized water was used for all solution preparations. A stock solution of In3+ in HCl (0.1–10 M) was prepared from InCl3·4H2O. For XAFS measurements, the solutions were transferred to polyethylene or glass cells, which were then tightly sealed; samples of the extracted complex in the organic solution were prepared by the solvent extraction method detailed below. Solvent extraction Two organic solutions with different dielectric constants, nitrobenzene or n-dodecane, were used as diluents in which 20 vol%
This journal is © The Royal Society of Chemistry 2014
Paper
of 2-ethylhexanol was added. The organic solution (TOA dissolved in the diluents) was pre-equilibrated with the same volume of HCl solution in the absence of In3+. An aliquot of the pre-equilibrated organic solution and the same volume of the HCl solution containing In3+ ([In]: 0.001 M for the extractability measurements; 0.1 M for the XAFS samples) were shaken for 60 minutes in a glass tube at room temperature (23 ± 2 °C), and then centrifuged. The metal contents in the aqueous phase were measured by ICP-AES (Horiba ULTIMA2). The concentrations in the organic phase were calculated by the mass balance of metals before and after the extraction. When using n-dodecane (with 2-ethylhexanol), a third phase was formed on contact with 10 M HCl solution; therefore, in this case, no sample was available for XAFS measurements. XAFS measurements The XAFS measurements were carried out at the BL11XU of the SPring-8. The operating energy and the ring current were 8 GeV and ∼99 mA. The SPring-8 synchrotron radiation was monochromatized using Si(111) crystal monochromators. The In K-edge (27 940 eV) X-ray absorption spectra were collected in transmission mode using two ion chambers (both filled with N2) at room temperature (∼25 °C). Energy calibration was performed using an In metal foil. XAFS data analysis The experimental data were analysed using the EXAFS data analysis software package WinXAS, Ver. 3.1.12 The EXAFS oscillation χ(k) was weighted by k3 and windowed using a Gaussian window function. The EXAFS data were fitted using theoretical phases and amplitudes were calculated using the program FEFF8.13 The crystal structures of tetrahedral tetrachloroindate [InCl4]−,14 and octahedral diaquatetrachloroindate [InCl4(H2O)2]− 15 and hexachloroindate [InCl6]3− 16 were used for the FEFF calculations as model compounds. All the interactions were modeled using single scattering (SS) and multiple scattering (MS) paths based on the model compounds; however, the MS paths were ignored since their contribution was not significant in the systems studied. For In3+ aqueous solutions, the coordination number, CN, of In3+ is known to be six;17 therefore, the CN was held constant at 6 in the aqueous solutions, and the amplitude reduction factor, S02, was determined by the use of In3+ in 10 M HCl solutions. The obtained S02 value (0.95) was fixed during all the fits. The Debye–Waller factor squared, σ2, of the In–Cl and –O(H2O) shells for the aqueous samples was determined using typical samples and all subsequent fits were performed by the use of the σ2 values so obtained: In3+ in 10 M HCl (highest [HCl]) for In–Cl, and In3+ in 0.1 M HCl (lowest [HCl]) for In–O(H2O). For the other shells, the CN, bond distance, r, and the shift in threshold energy, ΔE0 (the same ΔE0 was used for each shell), were allowed to vary in the fit. Error values were determined by the F-test18 (95% confidence). The fitting quality was checked using the R-factor: {∑|k3χ(k)obs − k3χ(k)calc|/∑|k3χ(k)obs|} × 100.
Dalton Trans., 2014, 43, 1630–1635 | 1631
View Article Online
Paper
Results and discussion
Published on 31 October 2013. Downloaded by Lomonosov Moscow State University on 23/12/2013 05:40:49.
In3+ in HCl solution The dominant species of In3+ ions in concentrated HCl solutions were first investigated. Fig. 1 and S1, ESI† show the raw In K-edge EXAFS and corresponding Fourier transforms (FTs) for the In3+ ion in 0.1–10 M HCl solutions. In the FTs, the shape of the intense peak (R = 1–2.5 Å) gradually changes with the HCl concentration; the peak for 0.1 M HCl is broader than that for 10 M; the highest point shifts to higher R with an increase in the HCl concentration. This intense peak corresponds to the In–O/–Cl correlation in these solutions. In order to determine the presence or absence of O atoms, 1-shell fits (Cl− only) and 2-shell fits (Cl− and O) were performed on this peak and then the reasonableness of the structural parameters and the R-factor were checked. For the spectra of 0.1–8 M HCl, the 2-shell fits give reasonable parameters (see Tables 1 and S1, ESI†) and lower R-factors (e.g., 9.3→4.7 for 6 M HCl, 6.5→4.4 for 8 M HCl). This is consistent with the broader peak shape, especially on the lower side, than that for the spectrum of 10 M HCl. In contrast, the CN value of the In–O shell for 10 M HCl is
