MATERIALS TRANSACTIONS
Online ISSN : 1347-5320
Print ISSN : 1345-9678
ISSN-L : 1345-9678
Separation of Silver (I) and Zinc(II) from Nitrate Solutions by Solvent Extraction with LIX63
Pan-Pan SunByoung-Jun MinSung-Tae KimSung-Yong Cho
Author information
JOURNALS FREE ACCESS FULL-TEXT HTML

2017 Volume 58 Issue 2 Pages 287-290

Details
Abstract

The separation of Ag(I) and Zn(II) from nitrate solutions using 5,8-diethyl-7-hydroxy-dodecan-6-oxime (LIX63) as the extractant was investigated as a function of nitric acid (HNO3) and extractant concentrations. Selective extraction of Ag(I) over Zn(II) was achieved with LIX63 when HNO3 concentration was in the range 0.001–1 mol/dm3. Quantitative stripping of Ag(I) from loaded LIX63 was accomplished using 5 mol/dm3 of HNO3. The McCabe-Thiele diagrams for the extraction and stripping of Ag(I) with LIX63 were constructed, and the results were verified by simulated cross-current extraction and stripping experiments. Finally, Ag(I) and Zn(II) solutions of high purity (> 99.95%) were obtained.

1. Introduction

Silver has been widely used in aesthetic, industrial, and technical applications. Recently, its utilization in technical applications, such as production of brazing alloys, photographic paper and film, catalysts, and electrical and electronic devices, has increased significantly. The growing demand for silver due to the depletion of natural resources and the increasing strictness of environmental policy have made its recovery from secondary resources an attractive proposition.1) The recovery of silver from spent silver oxide batteries is one such approach and is a widespread trend in global recycling markets.2,3)

Among the various types of media used in the recovery process, silver oxide-zinc button batteries have been investigated for recovering silver via leaching with nitric acid (HNO3) solution and fermentation liquor.4,5) In the past, mercury was extracted from spent silver oxide batteries by distillation or adsorption.5) However, due to stricter environmental regulations, silver oxide batteries without mercury were developed and came to be widely used instead.

Separation of silver and zinc has been investigated via precipitation,5) ion exchange followed by selective elution,6) solvent extraction,7,8) etc. The solvent extraction process offers significant advantages as a potential technique for recovering metal ions from solutions.9,10) In particular, when the concentration of the metal ion to be separated is higher than 1 g/dm3, solvent extraction is considered a better technique than ion exchange in terms of process efficiency.11)

In this study, 5,8-diethyl-7-hydroxy-dodecan-6-oxime (LIX63), a chelating agent, was employed to separate Ag(I) and Zn(II) from nitrate solutions. The effect of HNO3 and extractant concentrations on the extraction efficiency of LIX63 was examined. McCabe-Thiele diagrams were constructed for the extraction and stripping of Ag(I). Simulated cross-current extraction and stripping experiments were also performed to validate the results.

2. Experimental

2.1 Materials

The feed solution was prepared by dissolving the necessary amounts of AgNO3 (99.8%, Daejung Chemicals and Metals Co., Ltd.) and Zn(NO3)2･6H2O (98.0%, Daejung Chemical and Metals Co., Ltd.) in doubly distilled water. In all the experiments, the concentrations of Ag(I) and Zn(II) were maintained at 30 and 6 g/dm3, respectively, to mimic the leach solution.5) HNO3 was used to adjust the acidity of the solution.

The extractant, LIX63 was purchased from Cognis and used without further purification. Pure kerosene (Daejung Chemicals and Metals Co., Ltd.) was used as the diluent. All other reagents used were of analytical grade.

2.2 Solvent extraction and stripping procedure

The extraction and stripping experiments were carried out by contacting equal aliquots (20 dm3) of the aqueous and organic phases, in a unit (i.e., 1/1) aqueous-to-organic (A/O) phase ratio, for 30 min using a wrist shaker (Burrell Wrist Action Shaker Model 75, USA). All the experiments were performed at ambient temperature (25 ± 1℃). After shaking, the two phases were separated using separation funnels, and the the metal ion concentration in the aqueous phase was measured via inductively coupled plasma mass spectrometry (ICPS-7500, Shimadzu). The metal ion concentration in the organic phase was determined by mass balance. The experimental chart for batch extraction and stripping experiments is shown in Fig. 1.

Fig. 1

The experimental flow chart for the batch extraction and stripping experiment.

The number of extraction and stripping stages in a continuous operation was derived by constructing the McCabe-Thiele diagrams. The equilibrium isotherms were obtained from shakeout tests with aqueous and organic solutions in different phase volume ratios and mass balance after equilibration.

Cross-current simulations of the extraction and stripping processes were performed to verify the results obtained from the McCabe-Thiele diagrams. In the cross-current simulation of the extraction of Ag(I) with LIX63, a feed solution containing 30 g/dm3 of Ag(I), 6 g/dm3 of Zn(II), 0.01 mol/dm3 of HNO3, and 1 mol/dm3 of the extractant LIX63 (diluted in kerosene) were employed in a unit A/O phase ratio. Fresh LIX63 was fed at each extraction stage and the corresponding extract was obtained. All extract streams were combined together to obtain the overall extraction, while the raffinate was acquired at the last stage.

3. Results and Discussions

3.1 Effect of the concentration of HNO3

Extraction of Ag(I) and Zn(II) from the feed nitrate solution using LIX63 was investigated. In these experiments, the concentration of Ag(I) and Zn(II) was kept constant (30 and 6 g/dm3, respectively). Figure 2 exhibits the variation of the extraction percentages of Ag(I) and Zn(II) as a function of the concentration of HNO3, with the concentration of LIX63 fixed at 0.5 mol/dm3.

Fig. 2

Variation of the extraction of the metal ions as a function of the concentration of HNO3 at concentrations of Ag(I), Zn(II), and LIX63 fixed at 30, 6, g/dm3 and 0.5 mol/dm3, respectively.

As can be seen from Fig. 2, extraction of Ag(I) with LIX63 initially decreased with increasing HNO3 concentration in the range of 0.001–0.1 mol/dm3, and then increased with further increase in the HNO3 concentration up to 1 mol/dm3. In contrast, no Zn(II) was extracted at any of the HNO3 concentrations tested. The obtained results agree well with a previous study12) that reported that Ag(I) extraction from HNO3 solutions using LIX63 proceeds by different mechanisms, depending on the HNO3 concentration. Accordingly, a typical cation exchange reaction occurs in the acid concentration range of 0.001–0.1 mol/dm3, while Ag(I) extraction with LIX63 follows a solvating extraction mechanism when the HNO3 concentration increased further to 1 mol/dm3. Extraction of Ag with LIX63 at 0.001–0.1 mol/dm3 and 0.1–1 mol/dm3 can be presented through eqs. (1) and (2), respectively.

 $\begin{split} & {\rm [Ag^+]_{aq}} + {\rm [NO_3^- ]_{aq}} + {\rm n [ H_2 R ]_{org}} = {\rm [Ag (H_2 R )NO_3 ]_{org}} \\ & \quad {\rm (n = 1,2 )} \end{split}$ (1)

 ${\rm [Ag^+]_{aq}} + {\rm [H_2 R]_{org}} = {\rm [Ag(HR)]_{org}} + {\rm H^+ }$ (2)
where H2R represents LIX63.

The plot in Fig. 2 indicates that LIX63 exhibits selectivity for the extraction of Ag(I) over Zn(II); therefore, their separation from HNO3 solutions is possible with LIX63.

3.2 Effect of the concentration of LIX63

The separation efficiency of Ag(I) and Zn(II) by extraction with LIX63 was investigated by varying the LIX63 concentration at initial nitric acid concentrations ranging from 0.01 to 1 mol/dm3. The LIX63 concentration was varied from 0.3 to 1 mol/dm3.

As shown in Fig. 3, the extraction percentage of Ag(I) increased with increasing LIX63 concentrations at a fixed acid concentration, while that of Zn(II) did not exhibit any variation and was ～0%. HNO3 concentration, in the range of 0.01–0.1 mol/dm3, had minimal effect on Ag(I) extraction in these experiments. At each LIX63 concentration, the extraction percentage of Ag(I) increased with increasing concentration of HNO3 from 0.1 to 1 mol/dm3, in agreement with Fig. 1. Similar to the interpretations in Fig. 2, neither the concentration of LIX63 nor that of HNO3 had any effect on the extraction of Zn(II).

Fig. 3

Variation of the extraction of the metal ions as a function of the concentration of LIX63 at HNO3 concentrations from 0.01 to 1 mol/dm3.

Based on these extraction behaviors (shown in Figs. 2 and 3), it is inferred that Ag(I) and Zn(II) can be separated completely by extraction with LIX63.

3.3 McCabe-Thiele diagram for extraction of Ag(I) with LIX63

In order to determine the theoretical number of stages required for complete extraction of Ag(I) from the synthetic solution containing Ag(I) and Zn(II), a McCabe-Thiele diagram was constructed by varying the A/O ratio from 5/1 to 1/3 at 0.01 mol/dm3 of HNO3. The aqueous solution was contacted with 1 mol/dm3 of LIX63. The results thus obtained show that Ag(I) extraction increased from 24.4 to 98.5% as the A/O ratio decreased from 5/1 to 1/3. The results in Fig. 4 indicate that quantitative extraction of Ag(I) using LIX63 took place in two stages at a unit A/O ratio.

Fig. 4

McCabe-Thiele diagram for the extraction of Ag with 1 mol/dm3 of LIX63 at an aqueous-to-organic (A/O) phase ratio of 1/1.

3.4 Stripping of Ag(I) from loaded LIX63

3.4.1 Selection of the stripping agent

Stripping of Ag(I) from loaded LIX63 was investigated using various stripping agents like HNO3, ammonia water, thiourea, and a mixture of thiourea and HNO3. The loaded LIX63 was obtained by contacting 1 mol/dm3 of LIX63 in kerosene with the aqueous phase, in which Ag(I) and HNO3 concentrations were 30 g/dm3 and 0.1 mol/dm3, respectively. The results of Ag(I) stripping are summarized in Table 1.

Table 1 Stripping of Ag from the loaded LIX63.
Agent Stripping percentage (%)
3 mol/dm3 HNO3 18.3
5 mol/dm3 HNO3 37.2
0.1 mol/dm3 thiourea 71.8
0.1 mol/dm3 thiourea + 0.1 mol/dm3 HNO3 72.3
25% Ammonia water 79.0

Although Ag(I) loaded in LIX63 could be efficiently stripped using thiourea and a mixture of thiourea and HNO3 (see Table 1), the stripped silver compound containing sulfur was relatively difficult to deal with. Furthermore, as rules for using concentrated ammonia water are stringent, HNO3 was selected as the stripping agent in order to obtain convenient recovery streams.

3.4.2 McCabe-Thiele plot for the stripping of Ag

The stripping distribution isotherm of Ag(I) stripped from 1 mol/dm3 of loaded-LIX63 using 5 mol/dm3 of HNO3 was generated by varying the A/O volume ratio from 1/2 to 5/1. The loaded LIX63 was obtained by mixing 1 mol/dm3 of LIX63 and the mixture of Zn(II) and Ag(I) was prepared in 0.01 mol/dm3 of HNO3. While no Zn(II) was found in the loaded LIX63, the concentration of Ag(I) present in it was determined to be 24.89 g/dm3. The stripping percentage of Ag(I) increased from 20.49 to 76.94% by increasing the A/O ratio from 1/2 to 5/1. The McCabe-Thiele diagram for the stripping of Ag(I) from the loaded LIX63 is shown in Fig. 5. The results indicate that four stages are needed to strip most of the Ag(I) from the loaded LIX63 at an A/O ratio of 3/1 in cross-current stripping with 5 mol/dm3 of HNO3.

Fig. 5

McCabe-Thiele diagram for the stripping of Ag with 5 mol/dm3 of HNO3 at an aqueous-to-organic (A/O) phase ratio of 3/1.

3.5 Proposal of a separation process flowsheet and its verification

A process flowsheet for the separation of Zn(II) and Ag(I) developed in this study is presented in Fig. 6. Ag(I) in HNO3 solution of concentration 0.01 mol/dm3 was selectively extracted using 1 mol/dm3 of LIX63. Zn(II) was left in the raffinate with a purity higher than 99.95%. Quantitative stripping of Ag(I) from the loaded LIX63 was achieved with 5 mol/dm3 of HNO3. From the Ag(I) stripping solution, pure HNO3 can be recovered by extraction with tributyl phosphate (TBP) and stripping with distilled water.13)

Fig. 6

Flowsheet of the proposed process for the separation of Ag(I) and Zn(II) from nitrate solution.

In this study, results of batch simulations of experiments were compared with those derived from the McCabe-Thiele plots for extraction and stripping. The starting feed solution contained 29.4 g/dm3 Ag(I), 6 g/dm3 Zn(II), and 0.1 mol/dm3 of HNO3, while 1 mol/dm3 LIX63 in kerosene was used as the extractant. Batch simulations on the cross-current operation were employed in this study for convenience and to check the validity of the prediction.

Figure 7 shows the cross-current simulation of the extraction of Ag(I) with LIX63 at an A/O ratio of 1/1. The three stages of cross-current extraction removed ～99.8% of Ag(I), leaving Zn(II) in the raffinate (the concentration of Zn was 6 g/dm3). After four stages of cross-current extraction, the concentration of Ag(I) in the raffinate was lower than 3 mg/dm3, indicating that the purity of Zn(II) in the raffinate solution was 99.95%.

Fig. 7

Cross-current simulation of the extraction of Ag(I) with 1 mol/dm3 of LIX63.

The loaded LIX63 employed for the cross-current stripping simulation was obtained by extraction from 0.01 mol/dm3 HNO3 solution with 1 mol/dm3 of LIX63 at unit A/O ratio. The concentration of Ag(I) in the loaded LIX63 was 23.6 g/dm3. No concentration change was detected for Zn(II) after extraction. The cross-current stripping experiments of Ag (I) were carried out using 5 mol/dm3 of HNO3 at an A/O ratio of 3/1 with five stages.

The results presented in Fig. 8 show that the stripping percentage of Ag(I) increased rapidly with increasing number of stripping stages. After four stages, Ag(I) was nearly stripped completely from the loaded LIX63. No Zn(II) was detected in the stripping solution.

Fig. 8

Cross-current simulation of the stripping of Ag(I) from loaded LIX63 with 5 mol/dm3 of HNO3.

The results are very close to the values predicted by the McCabe-Thiele equilibrium isotherms in Figs. 4 and 5. The proposed separation process was thus verified by the results obtained in the batch simulation of the cross-current experiments. Continuous experiments, using a mixer-settler type extractor with real leaching solution, are necessary in future studies.

4. Conclusions

Ag(I) was selectively extracted with LIX63 from a solution containing Ag(I) and Zn(II) in HNO3 with concentration ranging from 0.001 to 1 mol/dm3. McCabe-Thiele diagrams were constructed to propose the loading and stripping strategy and the number of stages.

A process flow sheet was proposed for the separation of Ag(I) and Zn(II) from HNO3 solutions. In this process, Ag(I) was selectively extracted with LIX63, leaving Zn(II) in the raffinate solution. The loaded Ag(I) was recovered by stripping with 5 mol/dm3 of HNO3. In addition, this process was verified by batch simulation experiments of the cross-current extraction and stripping to obtain Ag(I) and Zn(II) solutions with high purity.

Acknowledgement

This research was supported by the Basic Science Research Program (No: 2014R1A1A2007063) and BK21 plus program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education. The authors are grateful for the financial support. The authors gratefully thank the Gwangju branch of the Korea Basic Science (KBSI) for ICP data.

© 2017 The Japan Institute of Metals and Materials
Top