AU2011342379A1 - Extraction of gold - Google Patents

Abstract

Disclosed is a process for extracting gold ions from an aqueous phase containing gold ions. The process comprises contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the extractant phase; and separating the extractant phase from the aqueous phase.

Description

WO 2012/079129 PCT/AU2011/001632 EXTRACTION OF GOLD This patent application claims priority from Australian Provisional Patent Application No. 2010905537 titled "Extraction of Gold" and filed 17 December 2010, the entire contents of which are 5 hereby incorporated by reference. FIELD The present invention relates to processes for extracting gold ions from an aqueous phase containing gold ions. BACKGROUND 10 Recovery of the valuable metal gold (Au) is typically a tedious process requiring repeated application of pyrometallurgical, hydrometallurgical or electrowinning processes to achieve acceptable metal recovery and metal purity. Most source materials such as ores, ore concentrates and smelter mattes are chemically complex because of the diversity of elements contained in them and the presence of non-precious metals in the source materials. Current industry practice for the recovery of gold largely 15 relies on numerous chemical processes. These are, at times, used in combination with solvent extractions, often with high cost or high toxicity, to recover, separate and purify gold. Hydrometallurgy is the recovery of metals from ores using an aqueous medium. One aspect of hydrometallurgy is solvent extraction which, as mentioned, is an important step in the recovery of gold. One major drawback of existing industrial methods (using Mixer-Settlers) is the potential for 20 fine particles (precipitated from the leach solution) or surfactants to inhibit rapid phase separation, leading to an unwanted "crud" phase, i.e. a particle-stabilized emulsion. Other disadvantages include a significant 'footprint size', large volumes of hazardous organic liquids, and relatively long separation times. There is a need for improved solvent extraction processes suitable for the recovery of gold. 25 SUMMARY. The present invention arises from research into solvent extraction systems based on an ionic liquid extractant phase.. In particular, we have found that gold ions can be extracted from an aqueous phase using an extractant phase comprising an ionic liquid or an ionic liquid with a miscible volatile organic solvent. Our research has shown that the ionic liquid itself functions as an extractant and, as such, 30 extraneous organic extractants do not need to be added to the extractant phase. We have surprisingly found that gold extractions using ionic liquids are highly efficient and rapid with, in some cases, greater than 90% extraction efficiency after less than two seconds residence time.

WO 2012/079129 PCT/AU2011/001632 2 In a first aspect, the present invention provides a process for extracting gold ions from an aqueous phase containing gold ions, the process comprising: contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the 5 extractant phase; and separating the extractant phase from the aqueous phase. In a second aspect, the present invention provides a process for recovering gold from an aqueous phase containing gold ions, the process comprising: contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the 10 extractant phase; and separating the extractant phase from the aqueous phase. In some embodiments, the process of the first and second aspects further comprises recovering the gold or gold ions from the extractant phase. In some embodiments, the ionic liquid is selected from the group consisting of: ethyl-3 methylimidazolium bis(trifluoromethanesulfonyl)imide, I -hexyl-3-methylimidazolium 15 bis(trifluoromethanesulfonyl)imide, 1-hexyl-3-methylimidazolium hexafluorophosphate, 1-dodecyl-3 methylimidazoliumbis(trifluoromethylsulfonyl)imide, 1-methyl-l-propylpiperidinium bis(trifluoromethylsulfonyl)imide, 1-methyl-l-propylpyrrolidinium bis(trifluoromethylsulfonyl)imide, tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, methyltrioctylammonium 20 bis(trifluoromethylsulfonyl)imide, and butyltrimethylammonium bis(trifluoromethylsulfonyl)imide. In some specific embodiments, the ionic liquid is selected from the group consisting of: I -hexyl-3 methylimidazolium hexafluorophosphate, tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, and methyltrioctylammonium bis(trifluoromethylsulfonyl)imide. In some more specific embodiments, the 25 ionic liquid is 1-hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)iinide. In some embodiments, the extractant phase comprises the ionic liquid and a volatile organic solvent that is miscible with the ionic liquid. The volatile organic solvent may be selected from the group consisting of: chloroform, methanol, ethanol, acetone, toluene, and acetonitrile. In some embodiments, the process further comprises treating the aqueous phase to increase the 30 concentration of inorganic anions in the phase prior to or during contact with the extractant phase. The inorganic anion may be a halide ion, a thiocyanate ion, a thiosulfate ion, a cyanide ion or a perchlorate ion. The halide ion may be selected from iodide, bromide, chloride, and fluoride.

WO 2012/079129 PCT/AU2011/001632 3 BRIEF DESCRIPTION OF THE ACCOMPANYING FIGURES Figure 1 shows a schematic of the microfluidic chip for solvent extraction (a: top-view; b: cross-section). Figure 2 shows UV absorption of gold as a function of gold concentration. 5 Figure 3 shows an image of micro-streams of aqueous phase and ionic liquid. The flow rate ratio of aqueous phase to ionic liquid phase was 25. Figure 4 shows the extraction percentage of gold versus residence time for (a) 100 ppm (b) 500 ppm (c) 1000 ppm and (d) all three concentrations of gold solutions. Figure 5 shows the extraction percentage of gold versus gold concentration at 0.2 seconds of 10 residence time. Figure 6 shows the extraction percentage of gold versus residence time in different ILs. Figure 7 shows the extraction percentage of gold as a function of residence time in IL/CHCl 3 mixtures at 20 v/v % IL and 2 v/v % IL concentrations. Figure 8 shows the extraction percentage of Au versus ILs at different HCl concentrations. 15 Figure 9 shows the logarithm of distribution ratio of Au versus ILs at 0.02 M HCl solution. Figure 10 shows the distribution ratio of Au versus number of CH 2 group in cations of ILs at 0.02 M HCl solution (TFSI as anions). Figure 11 shows the extraction percentage of gold as a function of IL concentration in CHCl 3 . Figure 12 (a) shows an FTIR spectrum of the pure IL showing the regions of interest to this 20 study; (b) cation and (c) anion. (b) shows FTIR spectra for pure IL and IL after extraction of gold at two different feed mole ratios (RAWIL = 0.25 and 0.50) showing the -C=C- peak at 1573 cm', associated with the IL cation, hmim* , which is normalised to unity for all spectra. (c) shows FTIR spectra as per (b) but showing depletion of the C-S peak at 790 cm 1 , characteristic of the loss of IL anion from the 25 IL phase. (d) shows the measured anion to cation peak ratio plotted against the predicted ratio, based on a 1:1 anion exchange mechanism and the depletion of AuCl 4 front the aqueous phase (from UV-vis analysis).

WO 2012/079129 PCT/AU2011/001632 4 Figure 13 shows a plot showing the capacity of AuCl 4 anion in the IL phase (mass of anion per litre of IL) versus the feed mole ratio of Au to IL. Closed and open symbols represent the results after 3 min mixing and after 3 min mixing followed by > 2 weeks in contact (unmixed). The maximum concentration of AuC1 in the pure IL was -1050 5 g/L, corresponding to a mole ratio in the IL phase of -0.87. The stoichiometric saturation limit is shown as the broken line. The solid line represents 100% extraction of the AuC1 4 anion. Figure 14 shows a series of representative samples, showing the appearance of the crystallisation of hmim.AuCl 4 after > 2 weeks for feed mole ratios of (a) 1.34, (b) 10 1.13, (c) 0.47, and (d) 0.37. Samples (a-c) exhibit IL phases that have solidified, despite their different appearance. Crystal growth is clearly seen in (b), expanded for clarity. The IL phase in sample (d) remains as a liquid, although some solid particles appeared within the liquid, which may indicate precipitated hmim.AuC4. Figure 15 shows a plot showing the ion exchange isotherm for extraction of AuCl 4 into IL for 3 15 min mixing (black symbols) and standing for 2 weeks (white symbols). Figure 16 shows a plot showing the rate of extraction determined by contacting the aqueous phase with either pure IL (triangles) or 20% IL diluted with chloroform (squares) in a microfluidic chip (schematic shown inset). The curves represent the best fit of the second order rate equation, A 1 = A 0 1 + kt where A and Ao are the concentration 20 of AuCI4- in the aqueous phase at time t and initially, respectively, and k is the rate constant. DETAILED DESCRIPTION In a first aspect, the present invention provides a process for extracting gold ions from an aqueous phase containing gold ions. The process comprises. contacting the aqueous phase with an extractant 25 phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the extractant phase. In a second aspect, the present invention provides a process for recovering gold from an aqueous phase containing gold ions, the process comprising: contacting the aqueous phase with an extractant 30 phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the extractant phase; and separating the extractant phase from the aqueous phase.

WO 2012/079129 PCT/AU2011/001632 5 The aqueous phase may be any water containing solution, suspension, emulsion, etc containing gold ions. An example of an aqueous phase is an aqueous leach solution obtained from processing gold containing ore. Processes for obtaining aqueous phases containing gold ions from gold containing ore are known in the art. The aqueous phase may not be a solution and it may, for example, contain 5 particulate materials, etc. The extractant phase consists of or comprises an ionic liquid. By "consists of' an ionic liquid, we mean the extractant phase contains ionic liquid without the addition of any other solvent. By "comprises" an ionic liquid, we mean the extractant phase contains the ionic liquid and may also contain another solvent, such as a volatile organic solvent, as discussed in more detail later. In both 10 cases, the person skilled in the art will appreciate that the ionic liquid may contain other materials and/or compounds that are either in the ionic liquid obtained from commercial sources or by synthesis, or that are dissolved in the ionic liquid during use. - The terms "ionic liquid" and "IL" are used interchangeably and mean a salt that is in the liquid state at room temperature. Ionic liquids consist of a bulky, asymmetric organic cation and a smaller anion and 15 they are liquids at relatively low. temperatures (eg below about 100 0 C). A range of ionic liquids are available commercially or can be synthesised using known methods. Specific ionic liquids that are suitable for use in processes of the present invention have imidazolium, piperidinium, pyrrolidiunium, ammonium or phosphonium cations. The anion of the ionic liquid may be bis(trifluoromethanesulfonyl) imide, chloride or hexafluorophosphate. 20 Specific ionic liquids that we have found to be suitable for use in the processes of the present invention include ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (emim.NTf 2 ), I hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (hmim.NTf 2 ), 1-hexyl-3 methylimidazolium hexafluorophosphate ( hmim.PF 6 ), I -dodecyl-3 methylimidazoliumbis(trifluoromethylsulfonyl)imide (dmim.NTf 2 ), 1-methyl-I-propylpiperidinium 25 bis(trifluoromethylsulfonyl)imide (mppip.NTf 2 ), 1-methyl- 1 -propylpyrrolidinium bis(trifluoromethylsulfonyl)imide (mpPyr.NTf 2 ); tradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide PI4.

6

.

6

,

6 .NTf 2 ), tetradecyl(trihexyl)phosphonium chloride

(P

1 4.

6

.

66 .Cl), methyltrioctylammonium bis(trifluoromethylsulfonyl)imide (N 8 s 8

.

8 .. NTf 2 ), butyltrimethylammonium bis(trifluoromethylsulfonyl)imide (N 4

,.

1 ,.NTf 2 ). 30 We have found that ionic liquids that are particularly suitable for the extraction of gold ions are 1 hexyl-3-methylimidazolium hexafluorophosphate, tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, and methyltrioctylammonium bis(trifluoromethylsulfonyl)imide. 1-hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide is particularly suitable for the liquid-liquid extraction of gold.

WO 2012/079129 PCT/AU2011/001632 6 Surprisingly, we have found that the extraction of gold ions from the aqueous phase using the ionic liquids described herein is highly efficient in the absence of an extraneous organic extractant in the extractant phase. In other words, we have surprisingly found that the ionic liquid itself is an efficient extractant whilst also acting as the solvent. 5 We have also found that ionic liquids such as hmim.NTf 2 , emim.NTf 2 , N 4

.

11 .NTf 2 , mppip.NTf 2 , mpPyr.NTf 2 , and P 14

.

6

.

6

.

6 .NTf 2 are able to selectively extract gold ions from an aqueous phase that contains other metal ions, such as Pt and Pd ions. We have further found that a volatile organic solvent can be included in the extractant phase (in addition to the ionic liquid) and the extraction of gold ions is still highly efficient. Thus, the 10 extractant phase may also comprise a volatile organic solvent and the ionic liquid. Some of the ionic liquids used, such as 1 -hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (hmim.NTf 2 ), may be immiscible with relatively non-polar solvents such as hydrocarbon solvents (eg. hexane and cyclohexane) and aryl solvents (eg. toluene) and, therefore, the volatile organic solvent used in these cases is preferably more polar than a hydrocarbon or aryl solvent. Chloroform, methanol, ethanol, 15 acetone, and acetonitrile may be particularly suitable volatile organic solvents for use in these cases. Alternatively, some ILs, such as tetradecyl(trihexyl) phosphonium chloride ( P 14

.

6

.

6

.

6 .Cl), are miscible with relatively non-polar solvents such as toluene and, therefore, relatively non-polar solvents such as hydrocarbons and aryl solvent can be used in these cases. The person skilled in the art is able to readily determine which solvents specific ILs are soluble in by mixing the two and assessing whether 20 they are miscible. The concentration of ionic liquid in the volatile organic solvent may be less than or equal to about 20 v/v/o. In some embodiments, the concentration of ionic liquid in the volatile organic solvent is less than or equal to about 2 v/v/o. In some embodiments, the concentration of ionic liquid in the volatile organic solvent is about 20 25 v/v/o. In some other embodiments, the concentration of ionic liquid in the volatile organic solvent is about 2 v/v%. After contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at least some of the gold ions from the aqueous phase to the extractant phase, the extractant phase is separated 30 from the aqueous phase. The gold or gold ions may then be recovered from the extractant phase using any of the techniques known for that purpose in the art, including for example electrodeposition, reduction, etc.

WO 2012/079129 PCT/AU2011/001632 7 The step of contacting the aqueous phase with the extractant phase may be carried out under bulk liquid-liquid extraction conditions or under microfluidic extraction conditions. Bulk phase liquid-liquid extraction processes are well known in the art and typically comprise adding the two phases to a suitable vessel and mixing for a time sufficient to allow transfer of at least some 5 of the solute from one phase to the other. In the processes described herein, the ratio of the volume of the aqueous phase to the volume of the extractant phase in bulk phase extractions is about 1:1 to about 10:1. In some embodiments, the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 2:1 to about 6:1. In some other embodiments, the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 4:1. 10 The aqueous phase and the extractant phase may be mixed for a period of about 1 minute to about 30 minutes. In some embodiments, the aqueous phase and the extractant phase are mixed for a period of about 5 minutes to about 15 minutes. In some specific embodiments, the aqueous phase and the extractant phase are mixed for a period of about 10 minutes. The step of contacting the aqueous phase with the extractant phase may alternatively be carried out 15 under microfluidic extraction conditions. Microfluidics involves the manipulation of fluids at the microscopic scale. Fluids behave very differently at small scales, due to the increasing importance of interfacial phenomena as the interface-to-bulk ratio increases. These phenomena include interfacial tension, surface charge, and adsorption (of surfactant or particles), amongst others. In conventional microfluidics applications, small volumes of liquid are manoeuvred within a microchannel network to 20 achieve a variety of chemical, physical, and/or biological processes. The precise control permitted by the microscopic environment over physical parameters, e.g. flow behaviour, heat/mass transfer, and mixing, facilitates significant improvements in reaction yields, product purity, sampling rates, and efficiencies. Typically, microfluidics is employed for high value, low volume samples (biological material, pharmaceuticals, or specialty chemicals); however in recent years, chip stacking 25 ("numbering iip") has permitted higher throughputs and new applications.' The step of contacting the aqueous phase with the extractant phase may be carried out under microfluidic extraction conditions using a microfluidic liquid-liquid extraction device as described herein and/or as described in our co-pending published application WO 2010/022441 titled "Extraction Processes" (the disclosure of which is incorporated herein in its entirety) and/or our co 30 pending unpublished Australian provisional patent application 2010905349 titled "High Throughput Microfluidic Device" and/or using any of the microfluidic separation techniques known in the art. Briefly, the microfluidic device may comprise a microchip containing an aqueous phase microchannel and an extractant phase microchannel. A pressure driven, co-current laminar flow technique may be applied during the process of solvent extraction in the microchannel, where the aqueous phase and the WO 2012/079129 PCT/AU2011/001632 8 extractant phase converge at a Y-j unction and flow together without mixing along the channel length, before separating at another Y-junction downstream. The residence time for extraction can be manipulated by altering the flow rate of the two phases using syringe pumps. The ratio of flow rate of the aqueous phase to the extractant phase may be greater than or equal to 5 about 25:1 when the extractant phase consists essentially of the ionic liquid. In this case, the residence time between the aqueous phase and the ionic liquid may be less than about 2 seconds. We have found that the extraction percentage of the gold ions using this process is greater than 90% in most cases, greater than 95% in many cases, and greater than 99% in some cases. In any of the processes described herein it may be advantageous to treat the aqueous phase to increase 10 the concentration of inorganic anions in the phase prior to or during contact with the extractant phase. The inorganic anion may be a halide, a thiocyanate, a thiosulfate, a cyanide or a perchlorate ion. The halide ion may be selected from iodide, bromide, chloride, and fluoride. Chloride may be particularly suitable. The concentration of inorganic anions in the aqueous phase may be increased by adding a salt 15 containing the inorganic anion, such as a halide salt, to the aqueous phase. In some embodiments, the process comprises treating the aqueous phase with HCI to increase the chloride concentration in the aqueous phase prior to contact with the extractant phase. The aqueous phase may be about 0.02M HCl, about 3M HC or about 7M HCI. In some other embodiments, the process comprises treating the aqueous phase with KCl to'increase 20 the chloride concentration in the aqueous phase prior to contact with the extractant phase. The aqueous phase may be about 3M KCl. The invention is hereinafter described by way of the following non-limiting examples. EXAMPLES Materials 25 . Solvent extraction (SX) was carried out via dispersion of an aqueous phase containing chloroauric acid with the IL phase for 3 min. In our experiments, the desired feed mole ratio (the mole ratio of total gold to total IL, RAuJIL) was achieved by adjusting the volume or concentration of the respective phases to reduce the use of reagents (particularly the IL). The SX was carried out in 20 mL glass vials with polypropylene ring seal caps 21 . The feed mole ratios ranged from 1.7 x 10- (2 g of HAuCl 4 : 1 L 30 of IL) to 2.66 (3200 g of HAuCl 4 : 1 L of IL) and were generally prepared using 10 g/L aqueous WO 2012/079129 PCT/AU2011/001632 9 solution of HAuCl 4 . The necessary amounts of pure IL were measured gravimetrically. A 20% IL in chloroform stock solution was prepared and used in the microfluidic solvent extractions. Gold (III) chloride trihydrate HAuCl 4 93H 2 0 (99.9+ %) was obtained from Sigma Aldrich. AR grade Chloroform, CHC1 3 , (Spectroscopic Grade) was obtained from Ajax Chemicals. Hydrochloric acid 5 (36%) was obtained from Ajax Chemicals. Ethanol (AR grade) was obtained from Chem-Supply). Acetone (AR grade) was obtained from Merck. These materials were used without further purification. Ten ILs used in this investigation (listed below) were purchased from IoLiTec, Germany. Five types of cations were studied: imidazolium (Im), piperidinium (Pip), pyrrolidinium (Pyr), ammonium (N) 10 and phosphonium (P). In all cases the anion was bis(trifluoromethanesulfonyl) imide (NTf 2 ). The anions hexafluorophosphate (PF 6 ) and chloride (Cl) were further chosen for hmim and phosphonium cations, respectively. Imidazolium and ammonium cations with different chain lengths were also used. Ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, 98 %, (emim.NTf 2 ) 15 1-Hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, 98 %, (hmim.NTf 2 ) I -Hexyl-3-methylimidazolium hexafluorophosphate, 99 %, (hmim.PF 6 ) 1 -Dodecyl-3-methylimidazoliumbis(trifluoromethylsulfonyl)imide, 98 %, (dmim.NTf 2 ) 1-Methyl-i -propylpiperidinium bis(trifluoromethylsulfonyl)imide, 99 %, (mppip.NTf 2 ) 1-Methyl-1-propylpyrrolidinium bis(trifluoromethylsulfonyl)imide, 99 %: mpPyr.NTf 2 20 Tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, 97 %, (P1 4

,

6

.

6 .NTf 2 ) Tetradecyl(trihexyl)phosphonium chloride, 96 %, ( P 1 46

.

6

.

6 .Cl) Methyltrioctylammonium bis(trifluoromethylsulfonyl)imide, 99 %, (N 8 8

.

8

.

1 .NTf 2 ) Butyltrimethylammonium bis(trifluoromethylsulfonyl)inide, 99 %, (N 4 m, 11 . NTf 2 ) Preparation of gold standard solutions 25 A stock gold solution of 10000 ppm was prepared in a volumetric flask by dissolving gold salt in neutral water, resulting in a solution pH value of 1.7. The gold standard solutions with different concentrations ranging from 20 to 1000 ppm were prepared by diluting the gold stock solution with aqueous HC solution at pH 1.7. To study the influence of HCl concentration on extraction WO 2012/079129 PCT/AU2011/001632 10 percentage, 500 ppm of gold salt was dissolved in pure water with different HCl concentrations (typically 0.02, 3 and 7 M HQl). Preparation of standard solutions of ionic liquid/CHC 3 mixture Standard solutions of ionic liquid, hmim.NTf2, in CHC1 3 with different volume by volume 5 concentrations ranging from 0 to 100 % were prepared by dissolving hmimTFSI in CHC 3 . The selected concentrations of ionic liquid in CHC1 3 are given below: 0 (pure CHC 3 ), 0.1 , 0.3 , 0.5 , 0.8 , 1.0 , 2.0 , 5.0, 7.0, 10.0 , 20.0, and 100 (pure ionic liquid) v/v % respectively. Example 1 - Bulk solvent extraction of gold by ionic liquid or ionic liquid/CHC 3 mixture Bulk solvent extraction of gold was carried out in a glass vial of 10 ml capacity by the following 10 procedure: 2 ml of 500 ppm gold solution were added to 0.5 ml of ionic liquid or different ionic liquid/CHCl13 mixtures. They were mixed vigorously with a magnetic stirrer for 10 min to achieve equilibrium distribution. When extraction was completed, the solution was left for enough time to phase separate. After extraction, the maximum UV absorption A* of the aqueous phase at a wavelength of 311 mu was measured. The extraction percentage was calculated as: 15 E %= [(Ao - A*) /Ao] x 100 AO: absorbance before extraction, A*: absorbance after extraction. An ICP technique was also used to measure the concentration of metal ions in aqueous solution after extraction. The extraction percentage was calculated as: E %= [(Co - C*) / Co] x 100 20 CO: concentration before extraction, C*: concentration after extraction. Example 2 -Microfluidic solvent extractionof gold using ILs PyrexTM microchips (Model DY- lOG) were kindly provided by IMT (Institute of Microchemical Technology, Japan,). An Olympus Optical microscope (Model BH2-UMA) was used to monitor the flow behaviour. The rate of extraction was determined by UV-vis absorption spectroscopy (Model 25 QE65000, Ocean Optics). The UV flow cell (volume s;8 piL) had a pass length of 2.5 mm. For bulk extraction, a quartz UV cell with a pass length of 2 mm was used to measure the UV absorption. Syringe pumps (Model KDS2 lOP, KD Scientific) were used to control the flow rate. The residence time t between two phases was calculated using the following equation: t = A*L / Q, where A is the area of the cross-section of the channel, L is the length of the channel and Q is the faster flow rate for WO 2012/079129 PCT/AU2011/001632 11 one of the two liquid phases. The residence times for typical flow rates used in the extraction are listed in Table 1. Table 1 - Residence time versus flow rate Flow rate 0.5 0.75 1.00 2.00 4.00 6.00 8.00 10.00 12.00 (ml/h) Residence time (s) 1.647 1.098 0.824 0.412 0.206 0.137 0.103 0.082 0.069 5 The structure and dimensions of the microchannel are shown in Figure 1. The dimensions of the microchannel are 40 sm in depth and 160 pm in width with a 'guide' structure in the middle. The length of the microchannel used in this experiment was 80 mm. A pressure driven, co-current laminar flow technique was applied during the process of solvent extraction in the microchannel, where the aqueous gold solution and organic phase containing IL converge at a Y-junction and flow together 10 without mixing along the channel length, before separating at another Y-junction downstream. The flow behavior inside the microchannel was monitored via a microscope equipped with a digital camera.The extraction progress was monitored using UV-vis absorption measurements in a microvolume flow cell connected to the organic or aqueous phase outlet. The residence time for extraction was manipulated by altering the flow rate of the two phases using syringe pumps. The 15 influence of extraction time on transfer, and thus the transfer kinetics, was investigated in the microfluidic chip. Results and discussion The UV absorption of gold solutions with different concentrations was measured using the flow cell method. The maximum absorption of gold solution near a wavelength of 311 nm as a function of 20 concentration is shown in Figure 2. A linear relationship was found for gold concentrations up to 500 ppm and can be fitted to the equation A = 0.0028 C, where A is the UV absorption of gold solution and C is the gold concentration. Therefore, at concentrations below 500 ppm, the transfer percentage of gold was calculated by the following equation: E % = {(AO - At) / Ad} * 100%, where AO is the gold absorption before extraction and At is gold absorption after the extraction at residence time t. 25 Above 500 ppm, however, the transfer percentage of gold was calculated using the following equation: E % = {(Ao - At / 0.0028) / AO} * 100%, where At needs to be less than the maximum absorption of gold at 500 ppm.

WO 2012/079129 PCT/AU2011/001632 12 Flow stability between ionic liquid and aqueous phase in the microchannel Liquid flow was driven by a precision syringe pump fitted with Hamilton gas tight syringes. As shown in Figure 1, a guide structure was fabricated in the middle of the microchannel, which helps to pin the three-phase contact line and, therefore, maintains the position of the liquid-liquid interface. 5 The flow rate of the aqueous phase was in the range from 0.5 ml/h to 12 ml/h, where the Reynolds number Re is less than 50 and thus the flow was laminar. For the most stable microfluidic flow of the two phases (as shown in Figure 3), it was found that the optimal flow rate ratio of aqueous to ILs including hmim.NTf2, emim.NTf2, mppyr.NTf2 and N4,1,1,1.NTf2 need to be fixed at 25, 8, 25 and 30, respectively. It is well-known that the viscosity of ionic liquid is generally much higher than that 10 of aqueous phase, thus the higher flow rate ratio of aqueous to ionic liquid phase was expected. As the cross-sections of the aqueous and ionic liquid streams were similar, the flow rate ratio was generally proportional inversely to that of viscosity ratio. However, the system of ionic liquid/aqueous liquid does not follow the trend mentioned above. The underlying causes still remain unclear. Extraction percentage versus residence time at different gold concentrations 15 The extraction percentage of gold as a function of residence time at different concentration of gold solutions was investigated. The results are shown in Figure 4. For all three concentrations, the extraction percentage increased with an increase in residence time. Gold extraction was extremely fast and efficient, for the extraction equilibrium was achieved in less than two seconds while the extraction percentage reached above 90 % for 1000 ppm gold and nearly 20 100 % for both the 100 ppm and 500 ppm gold solutions. As the ionic liquid is very viscous, its flow rate was 25 times less than that of aqueous phase. Therefore, the gold was not only extracted fast and efficiently, but was also concentrated effectively in the ionic liquid. With an increase in gold concentration, however, the extraction percentage decreased slowly. For 1000 ppm gold solution, only 90 % of gold was extracted. 25 The influence of concentration of gold on extraction percentage At a fixed flow rate of 4 ml/h, where the residence time is 0.2 second, the influence of gold concentration on extraction percentage was investigated. The results are shown in Figure 5. At a fixed residence time, the extraction percentage of gold decreased with increasing gold concentration. For a 20 ppm gold solution, 100 % of gold was extracted into the ionic liquid in 0.2 30 second, while 67 % of gold was extracted from a 500 ppm gold solution at the same time. Therefore, the ionic liquid, hmim.NTf2, could be an ideal medium to extract low concentrations of gold ions from an aqueous phase.

WO 2012/079129 PCT/AU2011/001632 13 Gold extraction using different ILs The extraction of gold with hmim.NTf2, emim.NTf2, mppyr.NTf2 and N4, 1,1,1 .NTf2 were carried out in microchannel. In order to stabilize the multiphase flow, the flow rate ratio of IL phase to aqueous phase for hmim.NTf2, emim.NTf2, mppyr.NTf2 and N4, 1, 1, .NTf2 was controlled at 25, 8, 5 25 and 30, respectively. The extraction percentage as a function of residence time is shown in Figure 6. The extraction efficiency decreased in the following order: hmim.NTf2 > emim.NTf2 mppyr.NTf2 > N4,1,1,1.NTf2. ILs of hmim.NTf2 and mppyr.NTf2 have same flow rate ratio, but different extraction efficiency, while emim.NTf2 and N4, 1,1,1 .NTf2 have different flow rate ratio, but the extraction efficiency is similar. Flow rate ratio thus did not show much influence on gold 10 extraction. As all ILs have the same anion NTf2, the variations in extraction efficiency is attributed to the different cations in the ILs. Since the diffusion of gold ions in the aqueous phase is the same. The exchange of gold ions at the IL/water interface might be the rate determining step for gold extraction. The extraction efficiency of gold in hmim.NT2/CHC 3 mixture The extraction of gold in hmim.NTf2/CHC 3 mixtures was further examined in the microchannel at 20 15 v/v % and 2 v/v % ionic liquid concentrations for 500 ppm gold. After diluting with CHC 3 , the viscosity of hmim TFSI decreased sharply. As a result, the flow rate ratio of aqueous phase to IL/CHCl 3 mixture decreased from 25 (pure ionic liquid) to 1 (20 v/v % IL) and 0.56 (2 v/v % IL) respectively. The extraction percentage as a function of residence time is shown in Figure 7 a and b. As expected, the extraction percentage increased with increasing residence time. At 0.8 second 20 residence time, the extraction percentage of gold was 95 % (pure ionic liquid), 64 % (20 v/v % IL) and 91 % (2 v/v % IL) from a 500 ppm gold solution, respectively. The extraction percentage increased with decreasing Vaq:VIL ratio (here Qaq/QIL is flow rate ratio), but increasing concentration of extractant. Although the concentration of 20 v/v % IL is higher than 2 v/v % IL, the QaqIQnt flow rate ratio is 1 for the former, which is much higher than that for the latter (0.56) and a lower 25 extraction percentage of gold was thus observed for the 20 v/v % IL. However, a higher extraction percentage of gold was found for the pure ionic liquid at a high Qaq/QIL flow rate ratio (25). The underlying transfer kinetics and mechanism need to be explored. Example 3 - Extraction of gold using different ionic liquids The conventional batch solvent extraction of gold from HCl solutions using pure ILs was carried out. 30 The performance of different ionic liquids on the extraction of gold from 0.02 and 3 M HCl solutions is shown in Figure 8. All ILs selected extract more than 90% of the Au from a 0.02 M HCl solution. The extraction percentage decreases slightly in 3 M HCl solutions. In order to understand how the type and structure of ionic liquids affect the extraction of gold, the distribution ratio for each of the WO 2012/079129 PCT/AU2011/001632 14 ionic liquid was calculated (Fig. 9). The distribution ratios for both hmim.PF 6 and P 1 4

.

6

.

6 .Cl is around 10000 and much higher than those for other ionic liquids with a NTf 2 anion. The distribution ratio decreases in the following order: Cl > PF 6 > NTf 2 . For ILs with the same NTf 2 anion, the influence of the centre atom and chain structure of the cation is given in Fig. 10. As the number of CH 2 groups 5 increases the distribution ratio increases as well. Similar distribution ratios were found with ILs having a similar number of CH 2 groups in the chain. The central atom - N or P - did not have an impact, indicating no strong affinity to the Au chloride complex Example 4 - Extraction of gold using ionic liquid/CHC 3 mixtures The ionic liquid, hmim.NTf2, was found to be immiscible with non-polar or less polar organic 10 solvents such as hexane, cyclohexane or toluene, but miscible with medium polar or polar organic solvents including CHCl 3 , methanol, ethanol, acetone, toluene, acetonitrile, etc. For extraction purposes, CHC1 3 was found to be the most suitable solvent to dilute hmim.NTf2 ionic liquid. Solutions of hmim.NTf2 in CHCl 3 with different concentrations (v/v) were prepared and subjected to bulk extraction. 0.5 ml of the IL/CHCl 3 solution with different IL concentrations was mixed 15 thoroughly with 2 ml of 500 ppm of gold solution for 90 seconds. The extraction percentage of gold as a function of IL concentration is shown in Figure 11. The extraction percentage of gold increased sharply with increasing concentration of ionic liquid in the concentration range less than 2 v/v %. Extraction equilibrium was reached within 90 seconds. It was observed that 96 % and 100 % of gold were extracted into IL/CHCl 3 mixtures at 2 v/v % and 10 20 v/v % of ionic liquid concentration respectively. Therefore, the Ionic liquid, hmim TFSI, is a highly efficient extractant for gold in HC1 solutions in both pure ionic liquid form, as well as in ionic liquid/CHCl 3 mixtures. Solvent extraction of gold in HCl solution by ionic liquids or hmim.NTf2/CHCl 3 mixture was carried out as both bulk and microchannel processes. Extraction was extremely fast and efficient in 25 microfluidic extraction. Extraction equilibrium reached within several seconds in the microchannel while almost 100 % gold was extracted into pure ionic liquid or IL/CHCl 3 mixture for 500 ppm gold. ILs affect the transfer rate of gold in the microchannel. Conventional batch experiments showed gold can be well extracted by all ILs selected, while many other metals including Pt (IV), Pd (II), Zn (II) and Cu (II) in HCl solutions were not extracted by most of ILs. Therefore, ionic liquid is a promising 30 medium and extractant for gold extraction, combining with high selectivity and efficiency. Example 5 - Measuring extraction performance Bulk solvent extractions were carried out via dispersion of an aqueous phase containing chloroauric acid with the IL phase for 3 min. The desired feed mole ratio (the mole ratio of total gold to total IL, WO 2012/079129 PCT/AU2011/001632 15 RA.L) was achieved by adjusting the volume or concentration of the respective phases to reduce the use of reagents (particularly the IL). The SX was carried out in 20 mL glass vials with polypropylene ring seal caps. The feed mole ratios ranged from 1.7 x 10-3 (2 g of HAuCL4: 1 L of IL) to 2.66 (3200 g of HAuCl 4 : 1 L of IL) and were generally prepared using 10 g/L aqueous solution of HAuCl 4 . The 5 necessary amounts of pure IL were measured gravimetrically. A 20% IL in chloroform stock solution was prepared and used in the microfluidic solvent extractions. The extraction performance was measured using the characteristic UV-vis or Infrared Spectroscopy (FTIR) peaks for AuCl4- and the IL, respectively. UV-vis spectra of the aqueous phase were obtained using an Ocean Optics QE65000 miniature spectrometer. Samples from bulk and microfluidic 10 extractions were analysed in a cuvette (2 mm path length) and flow cell (2.5 mm path length), respectively. FTIR spectra were obtained using a Nicolet FT-IR spectrometer. The IL phase was separated from the aqueous phase and immobilized as a thin film between dehydrated NaCl plates for analysis (note that these samples were not diluted prior to measurements). The plates were rinsed with acetone before each measurement and background spectra were collected between measurements to 15 exclude cross-contamination. Microfluidic extractions were carried out using the microfluidic device described in Example 2 and according to the protocol used in our previous study on copper extraction 4 , with a variable flow rate to adjust the contact time in each extraction experiment. Liquids were introduced into the microchip using precision syringe pumps (KD Scientific) with glass syringes (Hamilton, I mL and 2.5 mL) that 20 were fitted with PEEK adaptors and tubing (Upchurch Scientific, 150 ym inner diameter). Microchip experiments were monitored optically (Olympus microscope, BH2-UMA, and Moticam 2000 digital camera). Flow rates ranged from 0.125 to 10 ml/h and three different liquid-liquid contact lengths (channel lengths) were used (20, 80 and 240 mm) to access a wide range of extraction times. The UV vis flow cell was connected to the outlet of the microchip and positioned at an angle to avoid bubble 25 entrapment when first introducing liquid. A T-piece connection fitted between the microfluidic chip and flow cell, allowed rinsing of the flow cell with pure water between measurements, which also served as a reference solution. The flow of the reference solution was then stopped and the aqueous solution leaving the microfluidic chip was allowed to pass through the UV-vis flow cell. The relevant absorbance peak was monitored continuously until a stable plateau was reached (indicating that the 30 reference solution had been completely flushed from the flow cell), typically 5 min. Results and Discussion Aqueous/Ionic Liquid Solvent Extraction WO 2012/079129 PCT/AU2011/001632 16 Solvent extraction of metal ions is usually carried out via complexation with a ligand present in the organic phase. Where ionic liquids are employed as the organic phase, different mechanisms for transfer and different forms of the metal ion may be involved due to the ionic nature of the organic phase. In the present work Au is in the form of chloroauric acid (HAuC4), which dissociates in the 5 aqueous phase to yield the AuCL4 anion. The concentration of chloroauric acid in the aqueous phase can be determined using UV-vis spectroscopy (peak position 311.6 run) according to Beer's Law for concentrations up to 0.5 g/L. To establish that gold can be extracted by IL as the chloride complex, we observed the depletion of the UV-vis peak associated with chloroauric acid in the aqueous phase after mutually dispersing equal volumes of the two phases for 3 min. In every case, the post 10 extraction absorbance was less than (2.2 ± 0.8)% of the original value, indicating near complete extraction (98%) of the dissolved gold. UV-vis spectroscopy of the IL phase (post-extraction) revealed a peak at 321 nm, which was slightly shifted from the aqueous phase peak at 311.6 un. These experiments demonstrate quite clearly that very high extraction efficiencies of chloroauric acid can be achieved using the water-immiscible IL, hmim.NTf 2 . Although the results presented thus far 15 are very promising for rapid and efficient gold extraction in the form of HAuC1 4 , the nature of the extraction (mechanism) and the limitations of the process (capacity of the IL phase and extraction rate) are also important considerations and are addressed in the following sections. Extraction Mechanism The dissociation of HAuCl 4 in water is complete at pH 1.6, with the gold present in the form of 20 AuC1 4 in our experiments. For charge neutrality and based on previous studies of analogous systems", we expected that transfer occurs via anion exchange of NTf 2 by AuCl 4 . This mechanism, which should exhibit 1:1 exchange stoichiometry, would require the affinity of AuCl 4 for the IL cation to be higher than that of the NTf 2 to drive the extraction. An appreciable solubility of the NTfj anion in the aqueous phase (note that the IL itself is not miscible with water) is also required to 25 maintain charge neutrality in the two phases. The former condition can be inferred by comparing literature values of the melting temperatures of Rmim.AuCl 4 salts (emim.AuC4, 58 *C, and bmim.AuCl 4 , 50 'C)' 0 and Rmim.NTf 2 liquids (emim.NTf 2 , -15 to -21"C"'; bmim.NTf 2 , -6 to -2. 'C"'").Note that the range of reported IL melting temperatures may be due to trace contamination, e.g. with water.' 4 In this study, we sought evidence for anion exchange by observing the depletion of 30 the NTf 2 anion from the IL after extraction, using Fourier Transform Infrared Spectroscopy (FTIR). We chose two absorption bands which are characteristic of the IL anion (C-S; 790 cm-')''" and the IL cation (-C=C-; 1573 cm 1

)

7 and, importantly, rather insensitive to the molecular environment, e.g. pH, hydrogen bonding Figure 12(a) shows the full FTIR spectrum for the pure IL prior to extraction. The two regions of interest, labelled (b) and (c), are expanded in Figure 12(b) and (c) for the pure IL 35 and IL samples after gold extraction at two different feed mole ratios (0.25 and 0.50). All spectra WO 2012/079129 PCT/AU2011/001632 17 were normalised against an absorbance of unity for the cation-specific peak; -C=C-; 1573 cm- , to allow peak comparisons. The anion to cation peak ratio for the pure IL was found to be 0.825, corresponding to a 1:1 mole ratio of the IL anion and cation. The measured peak ratios determined for the IL phases after solvent extraction were then correlated with the depletion of AuCl4 from the 5 aqueous phase (as measured using UV-vis), to estimate the stoichiometry of the exchange. The predicted anion/cation peak ratio (assuming a 1:1 ion exchange of the anions) is compared with the actual peak ratio (measured after extraction) in Figure 12(d). The results indicate a 1:1 anion exchange mechanism within the experimental uncertainty. For the highest concentrations of gold studied here, the fraction of NTf2 in the IL exchanged with 10 AuCL 4 is quite large. In fact, the AuCL 4 to hmim* mole ratio in the organic phase, nAuc,4Jnhmim+, approaches 0.5 in Figure 12. At such high ratios, the physical properties of the IL phase must be different from those of hmim.NTf 2 (due to the large amount of hmim.AuC4 present). At the limit of 100% exchange, solidification of the post extraction IL phase is anticipated, due to the relatively high melting temperatures for similar Au compounds (emim.AuCl 4 , 58 'C, and bmim.AuC4, 50 'C)'. 15 Although, as will be shown here, less than 100% exchange may also lead to the formation of precipitated hmim.AuC 4 . For feed mole ratios greater than ~ 0.5, we observed the formation of a yellow precipitate over periods (hours to days) much longer compared with the initial extraction time of 3 min. The appearance of the precipitate formed in our study was consistent with that of Rmim.NTf 2 compounds prepared by Hasan et al.' 0 As the nature of the IL phase is changing, one may 20 expect implications for the distribution coefficient at high loadings of gold. Loading Capacity and Selectivity Although the stoichiometric saturation limit is 1:1 for equally charged ions in ion exchange processes, some limitations may be expected based on the physical properties of the exchange medium (e.g. resin or solvent). Therefore, the capacity of the IL to extract the AuCl4 anion was investigated for 25 different feed ratios of Au to IL ( RWIIL). Figure 13 shows the loading capacity (mass of AuCl 4 per litre of IL) of the IL against the feed mole ratio. Two data sets are shown to reflect different time-. scales of the extractions. Closed symbols represent the results collected immediately after 3 min of vigorous mixing. Further mixing (up to 20 min) showed no further extraction. However, this was in fact a non-equilibrium result, as after leaving the same solutions for several weeks (in contact but with 30 no further mixing) .improved further the extraction; see open symbols in Figure 13. For feed mole ratios less than unity and long extraction times (open symbols), almost all of the gold added to the system is extracted into the IL phase - a condition indicated by the solid line in Figure 13. Departure from this condition is only evident at, or very near, the feed mole ratio of unity (the onset of the plateau observed in Figure 13). In both cases, the plateau is found at - 1050 g/L, corresponding to 35 87% of maximum loading capacity estimated from the stoichiometry of the anion exchange (1204 WO 2012/079129 PCT/AU2011/001632 18 g/L, see dashed line in Figure 13). The reason for the incomplete exchange of anions is not'clear, however, it is known that the water immiscible Rmim.NTf 2 ILs are readily contaminated by water up to molar ratios as high as 0.25 (emim.NTf 2 ) and 0.18 (bmim.NTf 2 ) at 81% relative humidity, with the uptake being rather slow (hours).' 8 For hmim.NTf 2 , the solubility of water is 11 g/L or a molar ratio of 5 - 0.20,19.20 and the water molecules in the pure IL phase hydrogen bond with two IL anions.

2 ' In our FTIR data, the O-H stretch peak at 3635 cm-' is hardly detectable from the baseline in Figure 12 (absorbance < 0.04), indicating a "dry" IL phase prior to contacting the aqueous phase. After extraction, the intensity of the O-H stretch peak becomes significant (absorbance > 0.1, not shown). Just as the extraction of AuCl 4 ~ alters the nature of the organic phase, an appreciable extraction of 10 water may impact the ion exchange of the NTf 2 by AuCI 4 . The results presented in Figure 13 suggest two time-scales are relevant in these extractions. Within several minutes the extraction appeared to stall, as further extraction could not be detected (up to 20 min mixing). However, allowing the two phases to remain in contact for several weeks allowed the stoichiometric limit of the extraction to be approached. In many cases, the continued extraction was 15 accompanied by solidification of the IL phase to form a yellow solid. Figure 14 shows representative images of the solid salt formed after 2 weeks contacting the two phases (i.e. without further mixing). The partitioning of the AuC1 4 and NTfI anions between the two phases can be summarised using an ion exchange isotherm, as shown in Figure 15. The two axes are the ionic fractions of AuC14 in the organic and aqueous phases. The ionic fraction of AuC14 in the organic phase is equal to the mole 20 ratio of AuC1 4 to hmim* in the organic phase at equilibrium, i.e. nAuCI4JnL = nAuC4-(nNTf2 + nAuCI4-) In the aqueous phase the ionic fraction of AuC1 4 is simply the mole ratio of AuCl 1 at equilibrium relative to AuCl 4 initially, i.e. nAuc,4JnAuCI4- - nAuCl4J(nNTf2 + nAuCI4-). The straight line represents equal distribution of the two anions between the IL and aqueous phases, i.e. no selectivity. Above and below this line, the IL phase is more selective for the AuC1 4 and NTfj anions, respectively. Thus, 25 Figure 15 reveals a distinct selectivity of the IL phase for the AuCl 4 anion. Where RAu/IL is small, such that almost all of the AuCl' is extracted (see Figure 14), the ionic fraction of gold in the organic phase (IL) and the aqueous phase are both very low (less than ~ 0.1) and the selectivity of the IL for AuCl 4 is negligible. However, beyond this region, we observed an abrupt step in the data which sharpens and shifts with contact time (see arrow). This behaviour is usually attributed to a physical 30 change in the ion exchange medium which alters the selectivity of the medium itself." In the present case, the phase change is from a liquid organic phase, hmim.NTf 2 , to a solid organic phase (hmim.NTf 2 with large amounts of hmim.AuCl 4 ). Prior to this step in the isotherm, the IL remains liquid irrespective of the extraction time (within the limits studied here). At and beyond this step, the organic phase solidified over days to weeks. In some samples (usually intermediate feed mole ratios), 35 yellow, needle-like crystals formed at the interface between the two phases and extended into the WO 2012/079129 PCT/AU2011/001632 19 aqueous phase, Figure 14(b). Nonetheless, when isolated from the two-phase system, the solid phase was identical in all solidified samples. The solidification of the IL phase was accompanied by further depletion of AuC 1 4 from the aqueous phase until the ionic fraction of AuCl 4 in the aqueous phase dropped below - 0.1 (for the maximum contact times studied here). 5 Extraction Rate When microfluidic chips are used to perform solvent extraction of metal species the phase separation step can be essentially eliminated. In the microchannels, the aqueous and organic phases are contacted as microscopic laminar streams to allow extraction to occur without dispersing the two immiscible liquids (Figure 16, inset). Thus, the two liquids can be separated conveniently and instantaneously 10 using a branched microchannel, and analysed online with UV-vis spectroscopy. While our bulk experiments exhibited rather slow phase separation, phase separation in the microchip is instantaneous and therefore allows us to accurately determine the rate of the ion exchange process (the extraction ceases at the second y-junction). The concentration of chloroauric acid that was used in the microfluidic experiments was fixed at 0.5 g/L. Figure 16 shows the progress of extractions 15 using both pure IL and IL diluted with chloroform (20%vol IL). The diluted IL was chosen to reflect solvent extractions in which an extractant (in this case the IL) is present in a non-ionic organic solvent (chloroform). The flow rate ratios (organic:aqueous) for the pure IL and diluted IL experiments were 1:25 and 1:1, respectively, to stabilise the two streams of liquid along the length of the microchannel. For both extractions, the percentage of AuCl 4 extracted increases towards equilibrium within only a 20 few seconds. The rapid transfer of AuCl 4 is characteristic of the small stream dimensions involved (high surface to bulk ratio), which allows the diffusion-limited ion transport to take place. We estimate the contact time required for the microfluidic extraction to occur to be ~ 1.6 s, based on a diffusion length equal to the stream widths, 80 im, and the order or magnitude of the diffusion coefficient for the AuC14 anion, i.e. 10- cm 2 /s in water. The estimated contact time is in good 25 agreement with our experimental results (Figure 16) where extraction is complete within 2 s and 4 s for the pure and diluted IL extractions, respectively. The experimental data can be fitted (R 2 > 0.96) using a second-order rate equation which is known to describe ion exchange kinetics rather well.

37 ; AL 4- Ai" + kt where A and AO are the concentration of AuCl 4 in the aqueous phase at time t and t = 0, respectively, and k is the rate constant. The different rate constants (k,,,, IL = 6.0 and kdiluted IL 30 2.0) observed for the pure and diluted IL experiments reflect the different flow rate ratios and IL concentrations in the organic phase. While the diluted IL has 5 times fewer moles of IL per unit volume than the pure IL, the flow rate is 25 times higher. Thus, while the lower concentration of IL in the organic phase will slow the extraction kinetics (due to lesser availability of the exchangeable "sites" at the liquid-liquid interface), the higher flow rate maintains a steeper concentration gradient 35 across the interface, i.e. a higher driving force for diffusion of the AuCl4 and NTf2 anions. These WO 2012/079129 PCT/AU2011/001632 20 combined effects lead to the small difference in the observed extraction kinetics. We can, however, discount any influence of the slower kinetics associated with the solidification of the IL because no solidification of hmim.AuCl 4 forms when a large excess of IL is present. Furthermore, the longer extraction time used for the bulk extractions in this paper (3 min) is clearly much longer than is 5 actually required to achieve equilibrium where the feed mole ratio of Au/IL is low (an excess of IL). The latter point may prove very convenient for application of ionic liquids in solvent extraction processes. In particular, coupling ionic liquids with the microfluidic approach not only offers the advantages of microfluidics, i.e. ease of handling small sample volumes, but also the potential for higher microfluidic extraction throughput due to the very short residence times necessary. 10 Conclusion We can now build a physical picture for the extraction of AuC 4 by the IL [hmim][Tf 2 N]. Chloroauric acid dissociates in the aqueous phase to form AuC1-, which diffuses to the liquid-liquid interface. At the interface, NTf2~, which is the counteranion for anion exchange, is replaced by AuCl 4 due to the stronger interaction between AuC1 4 and the IL cation, hmim*. As the concentration of AuC1 4 in the 15 IL phase increases, a phase transition occurs from liquid, hmim.NTf 2 , to solid, hmim.AuCl 4 , which shifts the distribution coefficient (cf. ion exchange isotherm, Figure 15) due to an effective change in the solvent properties. The latter process is apparently slow and, in our experiments, occurs over several weeks (without further agitation). Saturation of IL phase with AuCl 4 appears to occur prior to the stoichiometric limit at -1050 g/L (mole ratio of 0.87). The extraction (for low concentrations of 20 AuCl 4 , 0.5 g/L) proceeds rapidly limited only by diffusion, with more than 90% of the gold complex extracted in under 2 s using pure IL. For diluted IL systems this time is extended to 4 s. Thus, we have demonstrated that hmim.NTf 2 is capable of rapidly extracting AuCl 4 from acidic aqueous solutions (pH 1.6) at high efficiencies. Furthermore, high loading capacities (1050 g/L), similar to those achieved using ion exchange resins, were possible; however, at high loading of the IL the 25 process is significantly slowed due to a phase change (liquid to solid) in the organic phase. It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive. 30 Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

WO 2012/079129 PCT/AU2011/001632 21 All publications mentioned in this specification are herein incorporated by reference. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common 5 general knowledge in the field relevant to the present invention as it existed in Australia or elsewhere before the priority date of each claim of this application.

WO 2012/079129 PCT/AU2011/001632 22 REFERENCES 1. Microreaction technologies: microtechnologies for chemical process intensification. Yole Developpement: Lyon, 2007. 2. (a) Klink, S.; Priest, C.; Ralston, J.; Sedev, R.; Mawatari, K.; Kitamori, T., Selective two 5 phase mineral separation on a microfluidic chip. In The 12th International Conference on Miniaturized Systems for Chemistry and Life Sciences, San Diego, USA, 2008; p 1796; (b) Priest, C.; Zhou, J.; Sedev, R.; Ralston, J.; Aota, A.; Mawatari, K.; Kitamori, T., Microfluidic Extraction from Particle Laden Systems. Submitted 2010. 3. MCT Redbook: Solvent Extraction Reagents and Applications. Cognis Group: 2007. 10 4. Priest, C.; Zhou, J.; Sedev, R.; Ralston, J.; Aota, A.; Mawatari, K.; Kitamori, T., International Journal of Mineral Processing 2011, 98, 168-173. 5. Paclawski, K.; Fitzner, K., Metallurgical and Materials Transactions B 2004, 35, 1071-1085. 6. Dietz, M. L.; Dzielawa, J. A., Chemical Communications 2001, 2124-2125. 7. Jensen, M. P., Dzielawa, J. A.; Rickert, P.; Dietz, M. L., Journal of The American Chemical 15 Society 2002, 124, 10664-10665. 8. Jensen, M. P.; Neuefeind, J.; Beitz, J. V.; Skanthakumar, S.; Soderholm, L., Journal of The American Chemical Society 2003, 125, 15466-15473. 9. Visser, A. E.; Jensen, M. P.; Laszak, I.; Nash, K. L.; Choppin, G. R.; Rogers, R. D., Inorganic Chemistry 2003, 42, 2197-2199. 20 10. Hasan, M.; Kozhevnikov, I. V.; Siddiqui, M. R. H.; Steiner, A.; Winterton, N., Inorganic Chemistry 1999, 38, 5637-5641. 11. Dzyuba, S. V.; Bartsch, R. A., ChemPhysChem 2002, 3, 161-166. 12. Ngo, H. L.; LeCompte, K.; Hargens, L.; McEwen, A. B., Thermochimica Acta 2000, 357 358, 97-102. 25 13. Fredlake, C. P.; Crosthwaite, J. M.; Hert, D. G.; Aki, S. N. V. K.; Brennecke, J. F., Journal of Chemical & Engineering Data 2004, 49, 954-964. 14. Seddon, K. R.; Stark, A.; Torres, M.-J., Pure Appl. Chem. 2000, 72, 2275-2287.

WO 2012/079129 PCT/AU2011/001632 23 15. Rey, I.; Johansson, P.; Lindgren, J.; Lassegues, J. C.; Grondin, J.; Servant, L., The Journal of Physical Chemistry A 1998, 102, 3249-3258. 16. Herstedt, M.; Smirnov, M.; Johansson, P.; Chami, M.; Grondin, J.; Servant, L.; Lassegues, J. 5 C., Journal of Raman Spectroscopy 2005, 36, 762-770. 17. H6fft, 0.; Bahr, S.; Kempter, V., Langmuir 2008, 24, 11562-11566. 18. Francesco, F. D.; Calisi, N.; Creatini, M.; Melai, B.; Salvo, P.; Chiappe, C., Green Chemistry 2011, 13, 1712-1717. 19. Endres, F.; Abbott, A. P.; MacFarlane, D. R., Electrodeposition from Ionic Liquids. Wiley 10 VCH: Weinheim, 2008. 20. Minami, I., Molecules 2009, 14, 2286-2305. 21. Cammarata, L.; Kazarian, S. G.; Salter, P. A.; Welton, T., Physical Chemistry Chemical Physics 2001, 3, 5192-5200. 22. Helfferich, F., Ion Exchange. McGraw-Hill Book Company: New York, 1962.

Claims (37)

1. A process for extracting gold ions from an aqueous phase containing gold ions, the process comprising: contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at 5 least some of the gold ions from the aqueous phase to the extractant phase; and separating the extractant phase from the aqueous phase.

2. The process according to claim 1, the process further comprising recovering gold or gold ions from the extractant phase.

3. The process according to any one of the preceding claims, wherein the ionic liquid is selected 10 from the group consisting of: ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, I hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, 1 -hexyl-3-methylimidazolium hexafluorophosphate, 1-dodecyl-3-methylimidazoliumbis(trifluoromethylsulfonyl)imide, 1-methyl-i propylpiperidinium bis(trifluoromethylsulfonyl)imide, 1-methyl-I -propylpyrrolidinium bis(trifluoromethylsulfonyl)imide, tetradecyl(trihexyl)phosphonium 15 bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, methyltrioctylammonium bis(trifluoromethylsulfonyl)imide, and butyltrimethylammonium bis(trifluoromethylsulfonyl)imide.

4. The process according to claim 3, wherein the ionic liquid is selected from the group consisting of: 1-hexyl-3-methylimidazolium hexafluorophosphate, tetradecyl(trihexyl)phosphonium 20 bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, and methyltrioctylammonium bis(trifluoromethylsulfonyl)imide.

5. The process according to claim 3, wherein the ionic liquid is 1 -hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide.

6. The process according to any one of the preceding claims, wherein the extractant phase 25 comprises a volatile organic solvent and the ionic liquid.

7. The process according to claim 6, wherein the volatile organic solvent is more polar than a hydrocarbon or aryl solvent.

8. The process according to claim 7, wherein the volatile organic solvent is selected from the group consisting of: chloroform, methanol, ethanol, acetone, toluene, and acetonitrile. 30

9. The process according to any one of claims 6 to 8, wherein the concentration of ionic liquid in the volatile organic solvent is less than or equal to about 20 v/v%. WO 2012/079129 PCT/AU2011/001632 25

10. -The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is less than or equal to about 2 v/v%.

11. The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is about 20 v/v/o. 5

12. The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is about 2 v/v%.

13. The process according to any one of the preceding claims, wherein contacting the aqueous phase with the extractant phase is carried out under bulk liquid-liquid extraction conditions.

14. The process according to claim 13, wherein the ratio of the volume of the aqueous phase to 10 the volume of the extractant phase is about 1:1 to about 10:1.

15. The process according to claim 14, wherein the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 2:1 to about 6:1.

16. The process according to claim 15, wherein the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 4:1. 15

17. The process according to any one of claims 13 to 16, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 1 minute to about 30 minutes.

18. The process according to any one of claims 13 to 17, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 5 minutes 20 to about 15 minutes.

19. The process according to any one of claims 13 to 18, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 10 minutes.

20. The process according to any one of claims 1 to 12, wherein contacting the aqueous phase 25 with the extractant phase is carried out under microfluidic solvent extraction conditions.

21. The process according to claim 20, wherein the extractant phase consists essentially of the ionic liquid.

22. The process according to claim 21, wherein the ratio of flow rate of the aqueous phase to the extractant phase is greater than or equal to about 25:1. WO 2012/079129 PCT/AU2011/001632 26

23. The process according to claim 22, wherein the residence time between the aqueous phase and the ionic liquid is less than about 2 seconds.

24. The process according to claim. 23, wherein the extraction percentage of the gold ions is greater than 90%. 5

25. The process according to claim 24, wherein the extraction percentage of the gold ions is greater than 95%.

26. The process according to claim 25, wherein the extraction percentage of the gold ions is greater than 99%.

27. The process according to any one of the preceding claims, wherein the process further 10 comprises treating the aqueous phase to increase the concentration of inorganic anions in the phase prior to or during contact with the extractant phase.

28. The process according to claim 27, wherein the inorganic anion is a halide ion selected from iodide, bromide, chloride, and fluoride.

29. The process according to claim 28, wherein the halide ion is chloride. 15

30. The process according to any one of claims 27 to 29, wherein the concentration of inorganic anions in the aqueous phase is increased by adding a salt containing the inorganic anion to the aqueous phase.

31. The process according to claim 30, wherein the concentration of halide ion in the aqueous aqueous phase is increased by adding a halide salt to the aqueous phase. 20

32. The process according to claim 31, wherein the process comprises treating the aqueous phase with HCI to increase the chloride concentration in the aqueous phase prior to contact with the extractant phase.

33. The process according to claim 32, wherein the aqueous phase is about 0.02M HCl.

34. The process according to claim 32, wherein the aqueous phase is about 3M HCl. 25

35. The process according to claim 32, wherein the aqueous phase is about 7M HCl.

36. The process according to claim 31, wherein the process comprises treating the aqueous phase with KC to increase the chloride concentration in the aqueous phase prior to contact with the extractant phase. WO 2012/079129 PCT/AU2011/001632 27

37. The process according to claim 36, wherein the aqueous phase is about 3M KCl. WO 2012/079129 PCT/AU2011/001632 28 AMENDED CLAIMS received by the International Bureau on 13 April 2012 (13.04.2012) THE CLAIMS 1. A process for extracting gold ions from an aqueous phase containing gold ions, the process comprising: contacting the aqueous phase with an extractant phase consisting of or comprising an ionic liquid (IL) under liquid-liquid extraction conditions for a time sufficient to allow transfer of at 5 least some of the gold ions from the aqueous phase to the extractant phase; and separating the extractant phase from the aqueous phase, wherein the extractant phase does not contain an extraneous organic extractant. 2. The process according to claim 1, the process further comprising recovering gold or gold ions from the extractant phase. 10 3. The process according to any one of the preceding claims, wherein the ionic liquid is selected from the group consisting of: ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, 1 hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide, 1-hexyl-3-methylimidazolium hexafluorophosphate, 1-dodecyl-3-methylimidazoliumbis(trifluoromethylsulfonyl)imide, 1-methyl-1 propylpiperidinium bis(trifluoromethylsulfonyl)imide, 1-methyl-1-propylpyrrolidinium 15 bis(trifluoromethylsulfonyl)imide, tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, methyltrioctylammonium bis(trifluoromethylsulfonyl)imide, and butyltrimethylammonium bis(trifluoromethylsulfonyl)imide. 4. The process according to claim 3, wherein the ionic liquid is selected from the group 20 consisting of: 1 -hexyl-3 -methylimidazolium hexafluorophosphate, tetradecyl(trihexyl)phosphonium bis(trifluoromethanesulfonyl)imide, tetradecyl(trihexyl)phosphonium chloride, and methyltrioctylammonium bis(trifluoromethylsulfonyl)imide. 5. The process according to claim 3, wherein the ionic liquid is 1-hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide. 25 6. The process according to any one of the preceding claims, wherein the extractant phase comprises a volatile organic solvent and the ionic liquid. 7. The process according to claim 6, wherein the volatile organic solvent is more polar than a hydrocarbon or aryl solvent. 8. The process according to claim 7, wherein the volatile organic solvent is selected from the 30 group consisting of: chloroform, methanol, ethanol, acetone, toluene, and acetonitrile. WO 2012/079129 PCT/AU2011/001632 29 9. The process according to any one of claims 6 to 8, wherein the concentration of ionic liquid in the volatile organic solvent is less than or equal to about 20 v/v%. 10. The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is less than or equal to about 2 v/v%. 5 11. The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is about 20 v/v%. 12. The process according to claim 9, wherein the concentration of ionic liquid in the volatile organic solvent is about 2 v/v%. 13. The process according to any one of the preceding claims, wherein contacting the aqueous 10 phase with the extractant phase is carried out under bulk liquid-liquid extraction conditions. 14. The process according to claim 13, wherein the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 1:1 to about 10:1. 15. The process according to claim 14, wherein the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 2:1 to about 6:1. 15 16. The process according to claim 15, wherein the ratio of the volume of the aqueous phase to the volume of the extractant phase is about 4:1. 17. The process according to any one of claims 13 to 16, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 1 minute to about 30 minutes. 20 18. The process according to any one of claims 13 to 17, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 5 minutes to about 15 minutes. 19. The process according to any one of claims 13 to 18, wherein the bulk liquid-liquid extraction conditions include mixing the aqueous phase and the extractant phase for a period of about 10 25 minutes. 20. The process according to any one of claims I to 12, wherein contacting the aqueous phase with the extractant phase is carried out under microfluidic solvent extraction conditions. 21. The process according to claim 20, wherein the extractant phase consists essentially of the ionic liquid. WO 2012/079129 PCT/AU2011/001632 30 22. The process according to claim 21, wherein the ratio of flow rate of the aqueous phase to the extractant phase is greater than or equal to about 25:1. 23. The process according to claim 22, wherein the residence time between the aqueous phase and the ionic liquid is less than about 2 seconds. 5 24. The process according to claim 23, wherein the extraction percentage of the gold ions is greater than 90%. 25. The process according to claim 24, wherein the extraction percentage of the gold ions is greater than 95%. 26. The process according to claim 25, wherein the extraction percentage of the gold ions is 10 greater than 99%. 27. The process according to any one of the preceding claims, wherein the process further comprises treating the aqueous phase to increase the concentration of inorganic anions in the phase prior to or during contact with the extractant phase. 28. The process according to claim 27, wherein the inorganic anion is a halide ion selected from 15 iodide, bromide, chloride, and fluoride. 29. The process according to claim 28, wherein the halide ion is chloride. 30. The process according to any one of claims 27 to 29, wherein the concentration of inorganic anions in the aqueous phase is increased by adding a salt containing the inorganic anion to the aqueous phase. 20 31. The process according to claim 30, wherein the concentration of halide ion in the aqueous aqueous phase is increased by adding a halide salt to the aqueous phase. 32. The process according to claim 31, wherein the process comprises treating the aqueous phase with HCl to increase the chloride concentration in the aqueous phase prior to contact with the extractant phase. 25 33. The process according to claim 32, wherein the aqueous phase is about 0.02M HCl. 34. The process according to claim 32, wherein the aqueous phase is about 3M HCl. 35. The process according to claim 32, wherein the aqueous phase is about 7M HCl. WO 2012/079129 PCT/AU2011/001632 31 36. The process according to claim 31, wherein the process comprises treating the aqueous phase with KCI to increase the chloride concentration in the aqueous phase prior to contact with the extractant phase. 37. The process according to claim 36, wherein the aqueous phase is about 3M KCl. 5 WO 2012/079129 PCT/AU2011/001632 STATEMENT UNDER ARTICLE 19(1) International Appln No. International Filing Date PCT/AU2011/001632 16 December 2011 Applicant University of South Australia et al The applicant forwards herewith replacement pages 24 to 27 for originally filed pages 24 to 27. Claim 1 has been amended to further distinguish the claimed invention from the prior art listed in the International Search Report. Independent Claim 1 now recites in the relevant part that the extractant phase does not contain an extraneous organic extractant.