UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Solvent extraction of copper and cyanide from waste cyanide solution Xie, Feng 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_spring_2010_xie_feng.PDF [ 3.11MB ]
Metadata
JSON: 24-1.0071001.json
JSON-LD: 24-1.0071001-ld.json
RDF/XML (Pretty): 24-1.0071001-rdf.xml
RDF/JSON: 24-1.0071001-rdf.json
Turtle: 24-1.0071001-turtle.txt
N-Triples: 24-1.0071001-rdf-ntriples.txt
Original Record: 24-1.0071001-source.json
Full Text
24-1.0071001-fulltext.txt
Citation
24-1.0071001.ris

Full Text

SOLVENT EXTRACTION OF COPPER AND CYANIDE FROM WASTE CYANIDE SOLUTION by FENG XIE B. Sc., The Northeastern University (PRC), 1992 M. Sc., The Northeastern University (PRC), 1995 M. A. Sc., The University of British Columbia, 2005 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Materials Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2010 © Feng Xie, 2010 ABSTRACT The potential use of two commercial extractants, LIX 7950, a guanidine derivative, and LIX 7820, a solvent mixture of quaternary amine and nonylphenol, for recovery of copper and cyanide from waste cyanide solution has been investigated. Low equilibrium pH favors copper extraction while a high molar ratio of cyanide to copper depresses the copper loading. It is confirmed that Cu(CN)32 is preferentially extracted over Cu(CN)43 and CN by the extractants. Solvent extraction of the mixture of metal cyano complexes shows a selectivity order as follows: Zn > Ni > Cu > Fe. The presence of S042 or S203 shows an insignificant effect on copper extraction while SCN ions may potentially compete for the available extractant with copper cyanide species and thus depress copper extraction significantly. Both extractants exhibit an affinity sequence as SCN > CNO > CN>S203.The selectivity order of different anions with the extractants can be explained by the interrelated factors including anion hydration, charge density, compatibility of the formed complex with the organic phase and the geometry effect. The extraction of Cu(CN)32with LIX 7950 is exothermic with an enthalpy change (AH°) of -191 kJ/mol. The copper extraction with LIX 7820 has little change when the temperature is varied from 25 °C to 45 °C. For both extractants, the loaded copper and cyanide can be stripped efficiently by a moderately strong NaOH solution. Further increase in NaOH concentration results in the formation of a third phase. The presence of NaCN can facilitate stripping of the loaded copper and cyanide by favoring the formation of Cu(CN)43 in the stripping solution. The important findings suggest a possible solution to the separation of metal cyanide species and free cyanide in the cyanide effluent. Both extractants can be used in a SX circuit for pre concentrating copper into a small volume of strip solution which can be further treated by electrowinning, AVR, SART or similar processes to recover copper products and cyanide. The free cyanide will remain in the raffinate solution from solvent extraction circuit which allows for the potential recycling of the barren solution to the gold cyanidation process. 11 TABLE OF CONTENTS ABSTRACT.ii TABLE OF CONTENTS iii LIST OFTABLES vi LISTOFFIGURES ix ACKNOWLEDGEMENTS xiv 1 Introduction 1 1.1 Cyanide classification 2 1.2 Free cyanide 3 1.3 Cyanate and thiocyanate 5 1.4 Metal cyanide complexes 7 1.4.1 Cyanide complex equilibrium 7 1.4.2 Copper cyanides 9 1.4.3 Zinc cyanides 12 1.4.4 Nickel cyanides 14 1.4.5 Iron cyanides 16 1.4.6 Gold and silver cyanide complexes 17 1.4.7 Mixtures of metal cyanide species 18 2 Cyanide Destruction and Recovery 19 2.1 Chemical destruction process 19 2.1.1 Inco S02/Air process 19 2.1.2 Hydrogen peroxide 20 2.1.3 Caro’s acid 21 2.2 Cyanide and metal recovery process 21 2.2.1 AVR/MNR/SART process 22 2.2.2 Activated carbon 24 2.2.3 Ion exchange resin 25 2.2.4 Solvent extraction 29 V 111 2.2.5 Miscellaneous 33 2.3 Research objective 34 3 Experimental 36 3.1 Organic reagents and chemicals 36 3.2 Preparation of the extractant solvents 37 3.3 Preparation of aqueous solutions 37 3.4 Test procedure 39 3.5 Analysis 39 4 Extraction with LIX 7950 42 4.1 Terms and definitions 42 4.2 Equilibrium time 43 4.3 Organic formula 44 4.3.1 Effect of diluents 44 4.3.2 Effect of modifiers 46 4.3.3 Effect of the extractant concentration 49 4.4 Effect of temperature 51 4.5 Effect of CN/Cu ratio 55 4.5.1 Extraction results 55 4.5.2 FTIR analysis 57 4.5.3 Preferential extraction 58 4.6 Effect of phase ratio 63 4.7 Effect of other anions 67 4.7.1 Extraction of the mixture solution 67 4.7.2 Effect of S042 71 4.7.3 Effect of SCN, CNO and S203 73 5 Extraction by LIX 7820 78 5.1 Equilibrium time 78 5.2 Organic formula 79 5.2.1 Effect of the molar ratio of nonyiphenol to Aliquat 336 79 5.2.2 Effect of diluents 84 5.2.3 Effect of the extractant concentration 85 iv 5.3 Effect of CN/Cu ratio.87 5.4 Effect of phase ratio 90 5.5 Effect of temperature 92 5.6 Co-extraction with other anions 94 5.6.1 Effect of non-metal anions 94 5.6.2 Extraction from a mixed solution of metal cyanides 96 6 Discussion 99 7 Stripping of the Loaded Copper and Cyanide 105 7.1 Effect of stripping reagents 105 7.1.1 Stripping with NaOH solution 105 7.1.2 Stripping with NaCN-NaOH solution 107 7.2 Effect of temperature 108 7.3 Effect of phase ratio 110 8 Conclusions and Recommendations 112 8.1 Conclusions 112 8.2 Recommendations 114 Bibliography 115 Appendix I Analysis Methods 126 Free Cyanide 126 Thiocyanate 127 V LIST OF TABLES Table 1-1 Simplified classification of cyanide compounds (modified from Flynn and McGill, 1995) 3 Table 1-2 Solubility of common copper minerals in 0.1 % NaCN solutions (after Hedley and Tabachnik, 1968) 10 Table 1-3 Some properties of copper(I) cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) 12 Table 1-4 Some properties of zinc cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) 14 Table 1-5 Some properties of nickel(II) cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) 15 Table 1-6 Some properties of iron cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) 16 Table 1-7 Some properties of gold and silver cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) 18 Table 2-1 Some commercial base extractants for gold solvent extraction (Rydberg, et a!., 2004) 30 Table 3-1 Some information of LIX® 79 and Aliquat® 336 36 Table 3-2 Some information of diluents and modifiers used in the research 37 Table 3-3 Some information of inorganic chemicals used in the research 38 Table 3-4 The major components in the mixture solution of metal cyanides 39 Table 4-1 Comparison of the effect of diluent types on extraction of metal cyano complexes . .46 Table 4-2 The effect of modifier types on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25°C) 47 Table 4-3 The effect of temperature on copper extraction with LIX 7950 (Org: 10 % v/v 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: [Cu] = i0 mol/L, CN/Cu=3,A/O=1) 52 Table 4-4 Comparison of 1G° and uS0 for different solvent extraction systems 54 vi Table 4-5 Copper extraction with LIX 7950 under pH uncontrolled conditions (Org: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane, aq: [Cu] = 500 mg/L, CN/Cu = 5, initial pH 10.5; 20 °C) 65 Table 4-6 The separation factors for Zn, Ni, and Fe over Cu under different equilibrium pH (Org: 30% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane, initial aqueous solution as in Table 3-4; A/0=1; 25 °C) 70 Table 4-7 The effect of S042 on copper extraction with LIX 7950 (Org: 10%v/v 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 3; AJO = 1; 20°C) 72 Table 4-8 The loaded anion content under different initial concentrations (Org: 10%v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: [Cu] = 3.93 x 10 mol/L, CN/Cu = 3; PHeq = 10.50 ± 0.05; A/0 1; 25°C) 77 Table 5-1 The effect of 1-octanol concentration on copper extraction with Aliquat 336 (Org: 9.4 x i0 mol/L Aliquat 336 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu=5;A/0=1;20°C) 83 Table 5-2 Copper extraction with LIX 7820 under pH uncontrolled conditions (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 500 mg/L, CN/Cu = 5, pH 10.5; 20°C) 90 Table 5-3 The loaded anions under different initial anion concentrations (Org: 2%v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 3; PHeq = 10.50 ± 0.05; A!0 = 1; 20°C) 95 Table 6-1 Summary of the selectivity orders with various extractants 100 Table 6-2 The hydration properties of some anions 102 Table 7-1 Stripping of copper and cyanide from the loaded LIX 7950 by NaOH solutions (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively; 0/A = 1; 20 °C) 106 Table 7-2 Stripping of copper and cyanide from the loaded LIX 7820 by NaOH solutions (Org: 2% v/v 7820 in n-octane, loaded Cu and CN are 2.17 x i0 mol/L and 6.50 x i0 mol/L, respectively; 0/A = 1; 20 °C) 106 Table 7-3 Stripping of copper and cyanide from LIX 7950 by NaOH-NaCN solutions (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively; [NaOH] = 1 mol/L; 0/A = 1; 20°C) ....108 vii Table 7-4 Stripping of copper and cyanide from LIX 7820 by NaOH-NaCN solutions (Org: 2% v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x 10-3 mol/L and 6.50 x 10-3 mol/L, respectively, [NaOHJ = lmol/L; 0/A = 1; 20 °C) 108 viii LIST OF FIGURES Figure 1-1 Plot of equilibrium distribution diagram for free cyanide vs pH at 25 °C 4 Figure 1-2 Eh-pH diagram of CN-H20system ([CN] =102 mol/L, 25 °C) 6 Figure 1-3 Eh-pH diagram for S-CN-H20system ([CN] = i0 mol/L, [S] = i0 mol/L, 25 °C) 7 Figure 1-4 Eli-pH diagram for Cu-CN-H20system ([Cu] = 1 mol/L, [CN] = 1 mol/L, 25 °C) 11 Figure 1-5 Plot of mole fraction of copper cyanide species vs log [CN] ([Cu]= 0.1 mol/L, 25 °C) 11 Figure 1-6 Eh-pH diagram for Zn-CN-H20system ([CN] = mol/L, [Zn] = i0 mol/L; 25 °C) 13 Figure 1-7 Eh-pH diagram for Ni-CN-H20system ([CN] = 103mo1/L, [Ni] = lO4mol/L; 25 °C) 15 Figure 1-8 Eh-pH diagram for Fe-CN-H20system ([CN] = 103mo1/L, [Fe] 104mo1/L; 25 °C) 17 Figure 2-1 Use of solvent extraction in the recovery of copper and cyanide from solution (after Dreisinger, et a!., 2001) 35 Figure 3-1 The schematic experimental set-up for pH and temperature-controlled tests 40 Figure 3-2 Picture of the shaking machine for pH-uncontrolled tests 41 Figure 3-3 Picture of the separation and filtration apparatus 41 Figure 4-1 Plot of variations of copper extraction and pH vs contact time (Org.: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu 5; A/O = 1; 25°C) 44 Figure 4-2 The effect of diluents types on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25 °C) 45 Figure 4-3 The effect of the concentration of 1-dodecanol on Cu extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25°C) 47 ix Figure 4-4 Plot of log Dcu[OW]2vs log [RG]org at different metal concentrations (Org.: LIX 7950 and 50 gIL 1-dodécanol in n-dodecane; aq: CN/Cu = 3, initial pH 10.00 ± 0.05; NO =1; 20°C) 51 Figure 4-5 Plot of the calculated Log Kex vs 1000/T (Org: 10 % v/v 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: [Cu] = i0 mol/L, CN/Cu = 3; NO = 1) 54 Figure 4-6 The effect of CN/Cu ratio on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; NO = 1; 25°C) 56 Figure 4-7 The effect of CN/Cu ratio on cyanide extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu = 5; NO = 1, 25 °C) 56 Figure 4-8 The calculated CN/Cu ratios in the organic phase under different pH (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu = 5; NO = 1; 25 °C) 57 Figure 4-9 Plot of the calculated fraction of copper cyanide complexes vs pH ([Cu] =3.93 x i03 mol/L, CN/Cu = 5; 25 °C) 60 Figure 4-10 Plot of the calculated fraction of copper cyanide complexes vs CN/Cu ratio ([Cu] =3.93 x i0 mol/L, pH =11; 25 °C) 60 Figure 4-11 The schematic extraction of copper cyanide solution with LIX 7950 63 Figure 4-12 The distribution isotherms of copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: CN/Cu = 5; pHeq = 10.50 ± 0.05; 20°C) 65 Figure 4-13 The distribution isotherms of cyanide extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: CN/Cu = 5; pHeq 10.50 ± 0.05; 20°C) 66 Figure 4-14 The schematic McCabe-Thiele diagram for copper and cyanide extraction (Org: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane; aq: CN/Cu = 5; PHeq = 10.50 ± 0.05; 20°C) 66 Figure 4-15 Plot of variations of metal extraction and solution pH vs contacting time (Org: 30% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane; initial aqueous solution as in Table 3-4; NO = 1; 25 °C) 68 x Figure 4-16 The extraction of metals and cyanide with LIX 7950 under different pH (Org: 30 % v/v LIX 7950 and 100 g/L 1-dodecano in n-dodecane; initial aqueous solution as in Table 3-4; A/O = 1; 25 °C) 70 Figure 4-17 The extraction isotherms for CN, SCN, and CNO with LIX 7950 (Org: 10 % v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane; pH = 10.50 ± 0.05; 25 °C) ..74 Figure 4-18 The effect of SCN, CNO and S203 ions on copper extraction by LIX 7950 (Org: 10 % v/v LIX 7950 and 50 g/ L 1-dodecanol in n-dodecane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu =3; pHeq = 10.50 ± 0.05, AJO = 1; 25°C) 77 Figure 5-1 Plot of variations of copper extraction and solution pH vs contact time (Org: 2% v/v LIX 7820 in n-octane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O=1;20°C) 79 Figure 5-2 The effect of the molar ratio of nonylphenol to Aliquat 336 on copper extraction (Org: 9.4 x i0 mol/L Aliquat 336 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O=1;20°C) 82 Figure 5-3 The effect of the 1-octanol concentration on copper extraction with LIX 7820 (Org: 2 % v/v LIX 7820 in n-octane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O=1,20°C) 83 Figure 5-4 The effect of diluent types on copper extraction with LIX 7820 (Org: 20% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 4; A/O=1;20°C) 85 Figure 5-5 The effect of the extractant concentration on copper extraction with LIX 7820 (Org: LIX 7820 in n-octane; aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 20°C) 86 Figure 5-6 The effect of the initial copper concentration on Cu extraction with LIX 7820 (Org: 10% v/v LIX 7820 in n-octane; aq: CN/Cu = 5; A/O = 1; 20°C) 87 Figure 5-7 The effect of CN/Cu ratio on copper extraction with LIX 7820 (Org: 2% LIX 7820 in n-octane; aq: [Cu] = 3.93 x i0 mol/L, A/U = 1; 20°C) 88 Figure 5-8 The effect of CN/Cu ratio on cyanide extraction with LIX 7820 (Org: 2% LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, A/U = 1; 20°C) 89 Figure 5-9 Plot of calculated CN/Cu ratios in organic phase vs equilibrium pH (Org: 2% v/v LIX 7820 in n-octane; aq: [Cu] = 3.93 x i0 mol/L, NO = 1; 20°C) 89 xi Figure 5-10 The distribution isotherms of copper extraction with LIX 7820 (Org.: 2% v/v LIX 7820 in n-octane, aq: CN/Cu = 5, pH = 10.50 ± 0.05; 20°C) 91 Figure 5-11 The distribution isotherms of cyanide extraction with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: CN/Cu = 5; pH = 10.50 ± 0.05; 20°C) 91 Figure 5-12 The schematic McCabe-Thiele diagram for copper and cyanide extraction (Org: 2% v/v LIX 7820 in n-octane, aq: CN/Cu = 5; pH = 10.50 ± 0.05; 20°C) 92 Figure 5-13 The effect of temperature on extraction of copper and cyanide with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x10 mol/L, CN/Cu = 5; P1eq = 11.00 ± 0.05; AJO = 1) 93 Figure 5-14 The effect of SCN, CNO and S2O3 on copper extraction with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x10 mol/L, CN/Cu = 5; pHeq= 10.50±0.05;AJO= 1;20°C) 95 Figure 5-15 The extraction isotherms of SCN, CN, CNO with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: sodium salt; PHeq = 10.50 ± 0.05; 20°C) 96 Figure 5-16 The effect of contacting time on metal extraction and pH with LIX 7820 (Org: 5% v/v LIX 7820 in n-octane; aqueous solution as in Table 3-4; A/O =1; 20°C) 97 Figure 5-17 The extraction of metals and cyanide with LIX 7820 under different pH (Org: 5% v/v LIX 7820 in n-octane, aqueous solution as in Table 3-4; AJO =1; 20 °C) 98 Figure 7-1 Stripping of loaded Cu and CN by NaOH solution under different temperatures (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively; aq: [NaOH] = lmol/L; 0/A = 1) 109 Figure 7-2 Stripping of loaded Cu and CN by NaOH solution under different temperatures (Org: 2% v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x i0 mol/L and 6.50 x i0 mol/L, respectively; aq: [NaOH] = 1 mol/L; 0/A = 1) 109 Figure 7-3 The striping isotherms of copper from the extractant solvent of LIX 7950 (10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively; aq: [NaOHJ = 1 mol/L; 20 °C) 111 xii Figure 7-4 The striping isotherms of copper from the extractant solvent of LIX 7820 (2 % v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x i0 mol/L and 6.50 x i0 mol/L, respectively; aq: [NaOH] = lmol/L; 20 °C) 111 Figure 8-1 The schematic flowsheet for the potential application of SX circuit 114 xlii ACKNOWLEDGEMENTS The author would like to express his utmost gratitude to his supervisor Dr. David Dreisinger for his kind guidance, support and encouragement during the whole thesis work. The contributions and suggestions put forth by Dr. Berend Wassink are gratefully acknowledged. The author would like to express sincere thanks to the people in UBC Hydro lab for their kind help and suggestions. The author also wishes to express gratitude to the Natural Science and Engineering Research Council of Canada (NSERC), the Canadian Institute of Mining, Metallurgy and Petroleum (CIM) - Hydrometallurgy Section of Metallurgy Society, CIM - Vancouver Branch, and the Cy and Emerald Foundation for financial support. Cognis is thanked for supplying the solvent samples. The author is particularly indebted to his family (especially his wife and daughter) for their tolerance, forbearance and encouragement throughout the period of the thesis. xiv 1 Introduction The cyanidation process has been practiced for treating gold ores by most of the gold processing plants for more than 100 years. In 1846, Eisner (Eisner, 1846) first established the important reaction for gold dissolution in oxygenated cyanide solution: 4Au + 8KCN + 02 + 2H0 = 4Au(CN)2+ 4KOH (1-1) The modern cyanidation process was patented between 1887 and 1888 by McArthur and the Forrest brothers, and was rapidly developed into a commercial process in the latter years of the 19th century (Marsden and House, 1992). Though some alternative lixiviants have been developed in recent years due to environmental pressure and process opportunity, none of them has yet found its way to practical application. The cyanidation process is still the mainstream technology for gold extraction from gold ore and it seems that this will continue. However, a big challenge for the process is the treatment of the large amount of cyanide-contaminated effluents since most of the cyanide consumed in the cyanidation process is actually wasted in the effluents, some occurring as free cyanide, with the balance forming metal cyanide complexes (Botz. et al., 2005; Fleming, 2005). For example, the initial reaction of pyrrhotite in the cyanide solution can be expressed as follows: Fe7S8 + NaCN = 7FeS + NaSCN (1-2) FeS and SCN can be further oxidized to form iron cyanide and various aqueous sulfur species such as sulphate: FeS + 6NaCN + 202 = Na4Fe(CN)6+ Na2SO4 (1-3) It is known that less than 2% of the cyanide consumed accounts for the dissolution of gold and silver in many gold operations and the majority of the cyanide is consumed by the cyanide soluble minerals found commonly in gold bearing ores (Marsden and House, 1992). The solution chemistry of some cyanide species commonly encountered in the cyanidation process is summarized as below. 1 1.1 Cyanide classification The term cyanide refers to a large family of compounds each member of which contains a cyano group (CN) as part of its molecular structure (Kunz, et al.. 1978). The cyano group consists of a carbon atom and nitrogen atom joined by three electron pairs. The definition for cyanide covers both organic and inorganic compounds. From the point of view of environmental effect, only those compounds which are capable of releasing the cyanide ion, CN, are of concern due to the high toxicity of CN. Although all cyanide compounds are capable, at least theoretically, of dissociation in an aquatic environment to release cyanide ions, the extent of ionization of different cyanide compounds varies in the range from infinitesimal to complete dissociation (Flynn and McGill, 1995). For example, nitriles and cyanohydrins are both organic cyanide compounds containing the cyano group. Since nitriles do not release cyanide ion in water, they are not classified as cyanide species according to this criterion. On the other hand, cyanohydrins are capable of releasing cyanide ions in water and therefore are considered as the cyanide species. In practice, cyanide pollution is always associated with inorganic cyanides since wastewaters from the principal sources of cyanide-contaminated effluent, e. g., metal electroplating and finishing facilities and the mining industry (particularly gold mill operations) always contain cyanides in the inorganic form. Since the behavior of cyanide in an aqueous medium is affected by a number of factors, including pH, temperature and the presence of other constituents in solution, the clear definition and classification of cyanide species are difficult though traditionally, cyanides may be classified into two general types of structures: free cyanide and complex cyanide, or more specifically, WAD (weak acid dissociable) and non-WAD (or strong acid dissociable) cyanide (Table 1-1, Flymi and McGill, 1995). In this chapter, special interest, particularly from the wastewater treatment viewpoint, is focused on the chemistry and recovery process of free cyanide including HCN and uncomplexed cyanide ion, CN, and some of the transitional metal cyanide complexes including copper, zinc, nickel, and iron cyanide species. Some species derived from cyanide such as CNO and SCN were also discussed. 2 Table 1-1 Simplified classification of cyanide compounds (modified from Flynn and McGill, 1995) Classification Compounds Free cyanide CN, HCN Weak-acid dissociable cyanide Zn(CN)42,Cd(CN)3, Cd(CN)42,Hg(CN)42, Mn(CN)62,Mn(CN)63,Cr(CN)64 Strong-acid dissociable cyanide Moderate strong complexes Cu(CN)j, Cu(CN)32Ni(CN)42,Ag(CN)2 Strong complexes Fe(CN)64,Fe(CN)63,Au(CN)2,Co(CN)64 1.2 Free cyanide Free cyanides are those cyanide compounds that are present in an aqueous solution in the form of CN ion and hydrogen cyanide (HCN). The concentration of free cyanide is in fact the sum of CN and HCN in the aqueous solution (Marsden and House, 1992, Flynn and McGill, 1995). They are related by the acid dissociation of HCN, HCN = CN ÷ H (1-4) Hydrogen cyanide acid is a weak acid in aqueous solution. The pKa for HCN as a function of temperature can be expressed as below: 2347.2pK =1.3440+ (1-5)a T+273.16 where T is temperature given in degrees Celsius (lzatt, et al., 1962). The diagram illustrating equilibrium distribution for free cyanide as a function of pH at 25 °C is shown in Figure 1-1. At pH above 9.21, CN predominates and at pH below 9.21, HCN predominates. As a result, a pH of 10.5 to 11.5 is usually maintained by most gold cyanidation operations to keep the predominant species of cyanide as CN since HCN has a relative high vapor pressure and will volatilize from the leachate and cause a significant loss of cyanide and a detrimental effect to the working environment. 3 0 4-I C.) ( 0 E The electronic configuration of cyanide ion CN is (Gis)2 (a* is)2(cy2s)*2sa2p7t py1t (Sharpe. 1976). Thus the two atoms (C and N) are triply bonded by one bond and two ‘t bonds as [:CN:] Free cyanide (CN) in aqueous solution shows infrared maxima at 2080 cm1. Values of the enthalpy of hydration of CN and the electron affinity of CN are close to those for the corresponding quantities for Br (-306 KJ/mol and -33iKJImol respectively, Flynn and McGill, 1995). When forming complexes, the cyanide ion has the ability to stabilize transition metal ions in a low oxidation state and acts as a monodentate ligand with carbon as the donor atom (Sharpe, 1976). 1 0.8 0.6 0.4 0.2 0 0 2 4 6 8 10 12 14 pH Figure 1-1 Plot of equilibrium distribution diagram for free cyanide vs pH at 25 °C 4 1.3 Cyanate and thiocyanate The most important cyanide compound of oxygen is cyanate (CNO), which is the major oxidation product of cyanide (Flynn and McGill, 1995). The cyanate ion consists of one oxygen atom, one carbon atom, and one nitrogen atom and possesses 1 unit of negative charge, borne mainly by the nitrogen atom. The structure of cyanate can be considered to resonate between two canonical forms: eO_CN O=C=Ne The resonance hybrid resulting from these two contributory structures can be represented as IOCEENI The cyanate ion is isoelectronic with carbon dioxide, and so shares its linear shape (Greenwood and Earnshaw, 1997). In aqueous solution, cyanide is fairly easily oxidized to cyanate as shown by the following potential calculated from the standard free energies of the species involved: CNO + 2W + 2e = CN + H20 E° = -0.14 V (1-6) Oxidation of CN to CNO occurs in the presence of oxidants such as S02/Air, hydrogen peroxide (11202) and hypochiorite (dO). The oxidation of CN by 02 is observable and it was reported that the process is catalyzed in the presence of copper and activated carbon (Adams, 1990), 2 CN +02= 2CN0 (1-7) The equilibrium distribution diagram for cyanate as a function of pH at 25 °C is shown in Figure 1-2. Cyanate is much less toxic than cyanide. It may degrade in aqueous solution to form CO2 and NH4 and decomposes rapidly in acidic solutions. Strong oxidants such as ozone and hypochiorite may further oxidize CNO to C032 and N2. 5 Eh (Volts) 2.0 — 1.5 1.0 HCNO(a) 0.5 CNO(-a) -1.0 -1 HCN(a) CN(-a) 2 4 6 8 10 12 14 pH Figure 1-2 Eh-pH diagram of CN-H20system ([CN] 102 mol/L, 25 °C) The most important sulfur cyanide species is the well-known thiocyanate ion (SCN). Thiocyanate is analogous to the cyanate ion wherein oxygen is replaced by sulfur. The structure and bonding of thiocyanate is as follows (Greenwood and Earnshaw. 1997): °S—CN The formation of SCN in the cyanidation process is attributed to the reaction of CN with solid sulfur or dissolved polysulfide (S2), or thiosulfate (S203)formed by the oxidation of sulfide minerals. The simplified reactions are presented as below, CN + HS + V2 02= CNS + OW (1-8) 8 CN + S8 =8 SCN (1-9) CN + S2032 = SCN+ S032 (1-10) 6 The Eh-pH diagram of S-CN-H20system at 25 °C is shown in Figure 1-3. Those oxidants that oxidize CN to CN0 can also oxidize SCN and oxidation of SCN by 02 is extremely slow at ambient temperature (Wilson and Harris, 1960). Eh (Volts) 2.0 1.5 1 0 HSO4(- 0.5 S04(-2a) -1.0 HCN(a) CN(-a) H2S(a) HS(-a) (-2a -1.5 -2.0 — 0 2 4 6 8 10 12 14 pH Figure 1-3 Eh-pH diagram for S-CN-H20system ([CN] = i0 mol/L, [SI = i0 mol/L, 25 °C) 1.4 Metal cyanide complexes 1.4.1 Cyanide complex equilibrium Cyanide bound to a metal ion is usually referred to as complexed cyanide which may occur as solids or dissolved species.. When a metal ion forms more than one complex with cyanide the corresponding series of equilibrium can be represented as follows (Flynn and McGill, 1995): 7 [MCN] M + CN = MCN K1 = [M][CN] (1-11) [MCN21MCN1 + CN = M(CN)2h1 K2 = [MCN][CNj (1-12) [MCIV’] M(cN)il1 + CN = M(CN)x K = [M(CN)11’j[C ] (1-13) where M11 is a metal ion of charge n+, i is the number of cyanide that can be taken by the metal and K is the stability constant of the complex. The equilibrium between the metal, the cyanide ligands and the complex can also be represented by the following equation, [M (CN) nj + i CN = M(CN) = [M][cN]1 (1-14) where 3, is the formation constant of the i complex. The formation constant can be expressed in terms of the stability constants by, 13=KoKiK2 K1 (1-15) by definition K0 = fib = 1 (free ion) and K1 = fir. The fraction of the individual metal cyanide complex (M(CN)1)can be given as follows, a = fi[CN]1 1 fi[CNjk (1-16) The equilibrium constants for the formation of metal cyanides provide a measure of the strength of the complex species. Many metal complexes with cyanide are stronger than those with other ligands, such as chloride (Cr), ammonium (NH3), and even ethylenediaminetetraacetate (EDTA). Metal cyanide species differ widely in their reactivity with acids (Flynn and McGill, 1995). As a result, metal cyanide complexes have been traditionally classified into two groups in regard to their reactivity with acids, namely, WAD, (weak-acid-dissociable), e.g., Zn, Cd and Ni cyanide complexes and non-WAD, e.g., cobalt and iron cyanide complexes. 8 1.4.2 Copper cyan ides Copper minerals dissolve to varying degrees in alkaline cyanide solutions depending on their reactivity with cyanide and the leaching conditions. Many copper oxide and sulfide minerals are very reactive in cyanide solutions (Hedley and Tabachnik, 1968). Except for chalcopyrite (CuFeS2) and chrysocolla (CuSiO3), the reactivity of most common copper minerals such as chalcocite (Cu2S), covellite (CuS), cuprite (Cu20), and malachite (CuCO3Cu(OH)2)with cyanide is substantial (Sceresini, 2005). Cu2S + 7NaCN + ½ 02 + H20 = Na2Cu(CN)3+ 2NaOH + NaCNS (1-17) 2CuS + 8NaCN + ‘/202 + H20 = 2NaCu(CN)3+ 2NaOH + 2NaCNS (1-18) Cu20 + 6NaCN + H20 = Na2Cu(CN)3+ 2NaOH (1-19) CuCO3 + 8NaCN + 2NaOH= 2NaCu(CN) + 2NaCO3+ NaCNO + H20 (1-20) The solubility of some typical copper minerals in cyanide solution is summarized in Table 1-2 (Hedley and Tabachnik, 1968). The fast dissolution kinetics of copper minerals in cyanide solution requires the maintenance of a high level of free cyanide during gold leaching. This results in a significant economical penalty in excess cyanide consumption and loss of valuable copper in cyanide effluent. The presence of copper in the ore also causes serious problems during cyanide effluent treatment as copper in the cyanide solution may occur as various copper cyanide complexes. In the cyanide tailings, when the concentration of free cyanide decreases with time, more cyanide can be liberated due to the equilibrium shift from Cu(CN)43 and Cu(CN)32 to Cu(CN)2 and finally to CuCN precipitates. This results in a substantial level of free cyanide in the tailing ponds. In contrast, when excess cyanide exists in the tailings, copper cyanide may react with the cyanide ion to form stable complexes with a higher number of CN ligands. Subsequently, the equilibrium among copper cyanide complexes buffer the free cyanide content within the tailing waters and acts as a sink to store the free cyanide (Hedley and Tabachnik, 1968). A typical Eh-pH diagram for Cu-CN-H20system is shown in Figure 1-4 (supposed CuO is a stable species) and the plot of mole fraction of copper cyanide species as a function of free cyanide concentration is shown in Figure 1-5 (based on data on Table 1-3). As shown in the 9 Figures, the formation of Cu(CN)2 only occurs at low pH. In alkaline cyanide solution, copper mainly occur as Cu(CN)32 and Cu(CN)43.Higher concentration of free cyanide favors the formation of Cu(CN)32 and Cu(CN)43.The infrared and Raman spectra for the copper(I) complexes have been studied and the information on their geometry has been confirmed. The schematic molecular structure of Cu(CN)32 and Cu(CN)43 complexes has been reported and the analysis results indicated that Cu(CN)32 has planar triangular shape and Cu(CN)43 is tetrahedral (Torre, et a!., 2006). Some properties of three copper cyanide complexes are summarized in Table 1-3 (according to Flynn and McGill, 1995 and Sharpe, 1976). Table 1-2 Solubility of common copper minerals in 0.1 % NaCN solutions (after 1-Tedlev and Tahachnik 1 96X——--- / Minerals Formula Copper dissolved Copper dissolved at23°C(%) at45°C(%) Azurite Malachite Chalcocite Metallic Copper Cuprite Bornite Enargite Tetrahedrite Chrysocolla Chalcopyrite 2CuCO3.Cu(OH)2 CuCO.C (OH) Cu2S Cu Cu20 FeS2Cu2.CuS Cu3AsS4 4Cu2SSb3 CuSiO2H0 CuFeS2 94.5 90.2 90.2 90.0 85.5 70.0 65.8 21.9 11.8 5.6 100 100 100 100 100 100 75.1 43.7 15.7 8.2 10 10.8 0 .1-i C.) 0.4 0.2 pH Figure 1-4 Eli-pH diagram for Cu-CN-H20system ([Cu] = 1 mol/L, [CN] = 1 mol/L, 25 °C) 0 -8 -6 -4 -2 0 2 4 log [CN]/moIIL. Figure 1-5 Plot of mole fraction of copper cyanide species vs log [CN] ([Cu] = 0.1 mol/L, 25 °C) > . w 1.5 1 0.5 0 -0.5 —1 -1.5 CuO CuCN Cu(CN)32 Cu(CN)43 CN 0 2 4 6 8 10 12 14 11 Table 1-3 Some properties of copper(I) cyanide species (Flynn and McGillj995 and Sharpe, 1976) Cu-C bond, nm Frequency, cm1 Molar absorptivity (in aqueous) cm2 M4 x 103(in Copper(II) cyanide complexes are unstable in the typical cyanide solutions with respect to the reduction of Cu(II) by cyanide, In acid solution, 2Cu + 4HCN 2CuCN(s) +C2N(aq) + 4H In basic solution, 2Cu + 7CN + 20W = 2Cu(CN)3 + CNO + H20 The log K values of these two reactions are so large that there is no known complexing agent for Cu(II) that will prevent the reduction of Cu(II) by CN (Flynn and McGill, 1995). 1.4.3 Zinc cyanides Zinc minerals occur infrequently and usually in small quantities in gold ores. Many of the zinc minerals are moderately soluble in cyanide solution although sphalerite (ZnS) is inert under normal cyanidation conditions (Hedley and Tabachnik, 1968). Metallic zinc has been commonly used for the recovery of gold from cyanide leach solutions (Merrill-Crowe process) before the invention of activated carbon adsorption process. The method is still adopted by some gold plants. The displacement reaction is indicated as below, Zn + 2Au(CN) = 2Au + Zn(CN)42 (1-23) Species Geometry f3 aq) CuCN(aq) Linear N N N 3.16x10’° Cu(CN)2 linear 0.192 2125 0.16 5.01x102’ Cu(CN)32 triangular 0.190-0.193 2094 1.09 7.94x1027 planar Cu(CN)43 tetrahedral 0.199 2076 1.66 3.16x1028 N: No data available (1-21) (1-22) 12 Reactions involving the cyanide complexes of metallic Zn are relatively fast. In the typical cyanide leaching solutions (e.g. 0.01 M free CN), most zinc occurs as the soluble tetra cyanozincate (Zn(CN)42).A typical Eh-pH diagram of Zn-CN-H20system at 25 °C is shown in Figure 1-6. Zinc cyanide complexes are restricted to the oxidation state of +2 and five cyano complexes have been identified. Zinc cyanide (Zn(CN)2)is a white powder which is sparingly soluble in water (K = 3.16 x 1 0). Table 1-4 gives some information on the properties of these zinc cyano complexes (Flynn and McGill, 1995 and Sharpe, 1976). Eh (Volts) 2.0 1.5 1.0 . . . Zn(CN)4(-2a) 0.5 ....Zn(+2a) 0.0 -0 5 ______________ HCN(a) CN(-a) : Zn 2 4 6 8 10 12 14 pH Figure 1-6 Eh-pH diagram for Zn-CN-H20system ([CNJ = 1 0 mol/L, [Zn] = 1 0 mol/L; 25 °C) 13 Table 1-4 Some properties of zinc cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) Species Geometry Zn-C Frequency, Molar absorptivity f3 bond, nm cm1 (in aq) cm2 M1 x 103(in aq) ZnCN linear N N N 2.OOx Zn(CN)2(aq) linear N N N 1.17x101 Zn(CN)3 trigonal N N N 1.12x106 Zn(CN)42 tetrahedral 0.202 2149 0.11 4.17x1019 Zn(CN)53 bipyramidal N N N 1.47x102° N: No data available 1.4.4 Nickel cyanides Nickel minerals are sometimes occurring in gold ores. A typical Eh-pH diagram for the Ni-CN H20 system (at 25 °C) is shown in Figure 1-7. Nickel cyanide (Ni(CN)2)is a green-blue solid powder with a low solubility in water (K = 3.16 x 1020). In dilute cyanide solution, the only important nickel species is the orange-yellow tetra-cyanonickelate ion (Ni(CN)4j.Intermediate species such as Ni(CN)3 are evidently not stable and disproportionate to Ni2 and Ni(CN)42 (Person. 1976). According to the molar ratio of nickel to CN, Ni(CN)53 can also be formed. The infrared and X-ray studies shows that the Ni(CN)42 ion is square planar or quite near so. This is quite different compared with Zn(CN)42 and Cu(CN)43 which have a tetrahedral shape. Some information on the properties of Ni(CN)53 and Ni(CN)42 is shown in Table 1-5 (Flynn and McGill, 1995 and Sharpe, 1976). It should be pointed out that though overwhelming majority of nickel cyano complexes has an oxidation state of +2, there are some well-defined complexes of nickel (0) and nickel (I) (such asK4(Ni(CN) andK4Ni2(CN)6)(Sharpe, 1976). 14 Table 1-5 Some properties of nickel(II) cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) Species Geometry Ni-C Frequency, Molar absorptivity bond, nm cm1 (in aq) cm2 M’ x 103(in aq) Ni(CN)42 square 0.186 2124 1.1 1.58x10° planar Ni(CN)53 bipyramidal N N N 1.58x 1029 N: No data available 4 6 pH Figure 1-7 Eh-pH diagram for Ni-CN-H20system ([CN] 103mo1/L, [Ni] = 10mo1/L; 25 °C) Eh (Volts) 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 Ni(+2a) Ni Ni(C1 HCN(a) [)4(-2a) CN(-a) 2 8 10 12 14 15 1.4.5 Iron cyanides Iron minerals are the most common minerals in gold ores. The formation of ferro- and ferric cyanides is rarely a problem for gold cyanidation since the dominant iron sulfide mineral, pyrite (FeS2), is inert in cyanide solution under normal gold leaching conditions. However, some iron sulfides such as pyrrhotite (Fe1S) are quite reactive in alkaline cyanide solutions and may decompose to form iron cyanide complexes and various sulfur species. Hexacyanoferrate(II), Fe(CN)64,and hexacyanoferrate(III), Fe(CN)63are among the best known of all iron cyanide complexes. The Fe(CN)64 ion can be oxidized to Fe(CN)63 by oxygen in acid. However in alkaline solution the reaction takes places at a very low rate. Some properties of these two complexes are shown in Table 1-6 (Flym and McGill, 1995 and Shame, 1976). A typical Eh-pH diagram for Fe-CN-H20system is shown in Figure 1-8. Though iron cyanides are extremely stable to pH and chemical changes, they will be a source of concern if they should pass to the tailings pond, since although the complexes themselves are of low toxicity, they are slowly decomposed by UV radiation, with liberation of free hydrogen cyanide (Kunz, et al., 1978). The primary irradiation reaction of Fe(CN)64 under ultraviolet or visible light is, [Fe(CN)641 hv/H20 [Fe(CN)5H20]3+ HCN + OH (1-24) The Fe(CN)63 ion is rather more reactive than Fe(CN)64 even though thermodynamically more stable with respect to dissociation into the constituent ions. The photochemistry of aqueous solution of Fe(CN)63 showed that a variety of products including Fe(OH)3, [Fe(CN)520], [Fe(CN)5H20]3,Fe(CN)64,and Prussian blue (a mixed oxidation state cyanide complex of iron with an unknown structure) could be formed. Table 1-6 Some properties of iron cyanide species (Flynn and McGill, 1995 and Shame, 1976) Species Geometry Fe-C Frequency, Molar absorptivity f3 bond, nm cm1 (in aq) cm2 M1 x 103(in aq) Fe(CN)64 octahedral 0.191 2044 N 2.51)4027 Fe(CN)63 octahedral 0.193 2118 N 2.51x1035 N: No data available 16 1.4.6 Gold and silver cyanide complexes Since most of the gold and silver will be recovered in the cyanidation process, their content in the cyanide effluent are usually insignificant. In this research work, no tests have been done on studying their extraction behaviors by the extractants. Au(CN)2 is the only gold species found under gold leaching conditions (pH above 10.5). Silver may form different complexes with cyanide depending on the leaching conditions (pH, cyanide concentration and ionic strength, etc.). Some of the properties of gold and silver cyanide complexes have been summarized in Table 1-7 (Flynn and McGill, 1995, Sharpe, 1976). Eh (Volts) 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 Figure 1-8 Eh-pH diagram for Fe-CN-H20system ([CN] = 1 0 mol/L, [Fe] = 1 0 mol/L; 25 °C) 2 4 6 8 10 12 14 pH 17 Table 1-7 Some properties of gold and silver cyanide species (Flynn and McGill, 1995 and Sharpe, 1976) Species Geometry Fe-C Frequency, Molar absorptivity f3 bond, nm cm1 (in aq) cni2M1 x103(in aq) Au(CN)j linear 0.212 2145 N 3.98x 1036 Ag(CN)2 linear 0.213 2135 0.26 3.02x102° Ag(CN)32 trigonal N 2105 0.38 2.51x10 Ag(CN)43 tetrahedral N 2092 0.56 6.31x102° N: No data available 1.4.7 Mixtures of metal cyanide species Usually the gold leaching solution contains a mixture of metal cyanide complexes in which complex equilibria will be established. In the case of insufficient CN present in the solution to convert all cyanide reactive metal species to their anionic cyanide complexes, metals may precipitate as cyanides, oxide species, or mixtures of compounds. Of these, the heavy metal hexacyanoferrates (II, III) are best known. These compounds nearly always form as colloids or gelatinous precipitates that are not stoichiometric. They contain variable quantities of the alkali ion from the hexacyanometallate salts and variable quantities of water of hydration. The detailed information on the formation of these cyano complexes has been summarized by Flynn and McGill (1995). 18 2 Cyanide Destruction and Recovery 2.1 Chemical destruction process The current available methods to detoxify cyanide-containing effluents include destruction by natural degradation, by biological processes or by chemical oxidants (Palmer, et al, 1988, Goode, et al., 2001). Due to increasing environmental concerns, the natural degradation process is seldom used as the sole detoxification step in cyanide effluent treatment. Biological processes have been successfully practiced but are not extensively used in the gold mining industry, partially because of its high cost and potential instability (Clark, et al., 2001). As the oldest and most widely recognized process for cyanide destruction, the alkaline-chlorination process is sometimes employed primarily in the plating industry though it is occasionally still used at a few mining sites. As the most commonly adopted methods, INCO S02/Air and hydrogen peroxide have been extensively utilized for treating cyanide effluents arising from the gold mining industry. Caro’s acid has been quite popular especially in Europe in recent years. Some other reagents and methods including ozonation (03), permanganate (MnO4), bromine compounds (Br2 and Br02), and photo-catalytic processes were also investigated, but none of them has yet been commercially practiced (Lanouette, 1977, Cooley, 1976, Domenech and Peral, 1988). 2.1.1 moo S02/Air process Developed in the early 1980s, the sulfur dioxide (S02)-air oxidation process from INCO Ltd. offers a reliable means of treating industrial cyanide effluents. By using a mixture of SO2 and air at controlled pH (about 8-10) in the presence of dissolved copper as catalyst, the process can oxidize free cyanide and most of the cyanide complexes with the exception of strong complexes, such as iron cyanides (Borberly, et al., 1984, Devuyst, et al., 1989). NaCN +02+ SO2 + H20 = NaCNO + H2S04 (2-1) The WAD (weak acid dissociable) metal cyanides complexes (such as copper and zinc) can also be removed from the stream, 19 Me(CN) + x SO2 (g) + x 02 (g) + x H20 = x CNO + x H2S04+ Me (2-2) where Me represents a metal element. Iron cyanide in the wastewaters is removed by precipitation as iron cyanide double salts with copper, zinc or nickel. Thiocyanate (SCN) can be also oxidized to cyanate by SO2 and air, but the kinetics are much slower compared with cyanide oxidation (Whittle, et a!., 1989). Hence SCN removal by S02/Air is incomplete. SCN + 4S02 + 402 + 5H20 = CNO + 5 H2S0 (2-3) The process can be applied to clear cyanide effluent and pulps. One of the benefits of the process is the availability of the relative cheap reagents such as sulfur dioxide can be generated on site from roaster gas or by burning elemental sulfur. During the past decades, the INCO SO2/Air process has been successfully used in the treatment of cyanide effluent arising from both metal finishing plants and gold cyanidation operations (Devuyst, et al., 1991). 2.1.2 Hydrogen peroxide Hydrogen peroxide (H20)is a strong oxidant which is capable of oxidizing free cyanide as well as the WAD cyanide species. CN + H20 = CNO + H20 (2-4) The cyanate then hydrolyzes to form carbon dioxide or carbonate depending on the pH of the solution, In acidic conditions, CNO + 2H + H20 = CO2 + NH4 (2-5) In basic conditions, CNO + OH + HO = C032 + NH3 (2-6) Thiocyanate can also be oxidized by hydrogen peroxide, but iron cyanides are not destroyed (Wilson and Harris, 1960, Vickell, et al., 1989, Kunz, et al., 1978). SCN + H2O = S + CNO + H20 (2-7) The oxidation of cyanide with hydrogen peroxide proceeds fairly fast when treating wastewaters with high cyanide concentration, but the reaction is quite slow in dilute waste cyanide solution. A number of metals such as copper can act as the catalyst. The specific advantage of hydrogen peroxide is that it is an ecologically desirable pollution control agent and yields only water and/or oxygen upon decomposition. The process has been successfully practiced by more and more gold mining operations though high consumption of the reagent is sometimes observed, 20 probably due to the vigorous catalytic decomposition by heavy metals and various organic compounds (Goode, et al., 2001). 2.1.3 Caro’s acid Caro’s acid (H2S05)has been quite popular for wastewater treatment in recent years. As a strong acid and oxidant, it can react with cyanide to form cyanate which may further be decomposed by hydrolysis. CN + H2S05= CNO + H2S04 (2-8) In the presence of excess Caro’s acid, cyanide is completely and rapidly destroyed, 2CN + 5H2S0+ 20ff = 2C0 + N2 + 5H2S04+ H20 (2-9) As the addition of Caro’s acid causes a drop in pH, caustic soda is usually added simultaneously to avoid the formation of volatile hydrogen cyanide (Clancy, et al., 1978, Goode, et a!., 2001). Persulfuric acid (H2S08)or persulfate salts such as ammonium persulfate ((NH4)2S08and sodium persulfate (Na2SO8)are also powerful oxidants and react with cyanide in a manner similar to Caro’s acid. These agents are especially suitable for destroying cyanide in concentrated solutions such as the spent cyanide wastewaters from electroplating industry and can readily oxidize ferrocyanide to ferricyanide. Actually regeneration of the spent ferricyanide bleach with persulfate is a widely used method in the photoprocessing industry (Cooley, 1976). The regeneration reaction is expressed as below, 2Fe(CN)64 + S208 = 2 Fe(CN)63 + 2S04 (2-10) 2.2 Cyanide and metal recovery process Though the destruction processes (INCO S02/Air, Caro’s acid or hydrogen peroxide) can be very efficient in destroying free cyanide in the cyanide effluents, the destruction of effluents containing high cyanide concentration or valuable metals (such as copper) could severely decrease the profitability of the gold plant operations. Sometimes this may even render the cyanidation process ineffective if the copper and complexed cyanide are not recovered after gold 21 recovery (Goode, et al., 2001, Jay, 2001). At the same time, a number of accidents involving cyanide contaminated effluents have resulted in the environmental constraints controlling the discharge of cyanide from gold mining industry being tightened by local governments worldwide. In the case of a cyanide spill at the Aural gold mine in Baja Mare, Romania in early 2000, the cyanide tailing dam broke which resulted in the release of cyanide effluent into the Tisza River, one of the major waterways in Europe (DeVries, 2001). Stricter regulations to reduce the effect of future accidents have been implemented as a result and the use of cyanide in new mining operations has even been banned in some countries or districts (Greece, Turkey and Montana of US). The concentration of free cyanide and the WAD cyanide species discharged into the tailing is usually required to be controlled below a strict limit (in many sites it is 50 ppm to tails and 0.1 ppm or less if there is any discharge to a receiving waterway). Subsequently, there has been growing interest in technologies for the recovery of valuable metals and cyanide from cyanide effluents arising from gold mining industry. 2.2.1 AVR/MNR/SART process The AVR (Acidification-Volatilization-Regeneration) process was first developed in the early part of the 20th century (Fleming, 2001). The process concept is relatively simple: the waste cyanide solution is first acidified to weakly acidic pH (usually below 4-5 by addition of sulfuric acid) and then contacted with high-pressure air. Most of the cyanide is converted to HCN which is volatilized by air and then adsorbed in alkaline solutions to produce aqueous NaCN or Ca(CN)2.The main reactions involved in this process are as following, 2CN ÷ H2S04= 2 HCN (aq) + 2 5Q42 (2-11) HCN (aq) = HCN (g) (2-12) HCN (g) + NaOH = NaCN + H20 (2-13) The cyanide-free solution then passes through a neutralization step to precipitate the heavy metals (Riveros, et al., 1997). The Cyanosorb process is a variation of the AVR process which uses the same principle to treat waste cyanide puips instead of the clear solutions (Stevenson, et al., 1996). However, during the acidification stage, copper is precipitated as copper cyanide (CuCN) which is difficult to sell due to the presence of cyanide. 22 Cu(CN)32 + H2S04= Cu(CN)(s) + 2HCN + S042 (2-14) In order to recover valuable copper, some modified AVR processes have been developed. The MNR process was developed by Metallgesellschaft Natural Resources and involves a solid/liquid separation process to obtain a clarified cyanide solution, to which water-soluble sulfide compounds (NaSH or Na2S) are added to precipitate base metals (mainly subject to copper). The solution is then acidified to pH < 5 by the addition of sulphuric acid. The copper sulfide precipitate is recovered by filtration. The acidification and sulfidization reaction of copper cyanides is presented below, 2Na3Cu(CN)4+ 3.5 HSO + NaSH = Cu2S + 3.5 Na2SO4+ 8HCN (aq) (2-15) Other base metals present in the solution will also co-precipitate (Dreisinger. et al., 1995). The neutralization step may be performed directly on the acidified solutions (after filtration of copper sulfide) or may be linked to a volatilization step. The hydrogen cyanide gas generated in the acidification process is volatilized and reabsorbed in an alkaline solution. The SART process (Sulfidization/Acidification — Recycling — Thickening) is based on the same chemistry as that of MNR, except that the copper sulfide precipitates are recovered first by precipitation and thickening rather than direct filtration. The process has been successfully practiced at Telfer Gold Mine in Australia (Dreisinger, et al., 2001, Barter, et al., 2001). The main drawback associated with the AVR/MNR/SART process appears to be the high operational cost - the reagents (acid and base) and the energy required by air sparging. The process can be applied economically to effluent solutions containing > 150 mg/L total cyanide, but in case of low cyanide concentration in the tailings, it is generally considered to be unsuitable for producing a final solution for discharge because of the high cost of reducing the cyanide concentration down to required control levels. The AVR process has been used for cyanide recycle at Flin Flon (Canada) starting in the 1930s and was abandoned in 1975, partially due to the reasons above (Marsden arid House, 1992). 23 2.2.2 Activated carbon Activated carbon is a generic term for a broad range of amorphous, carbon-based materials, prepared so as to exhibit a high degree of porosity and a large associated surface area. Due to the particular affinity of gold and silver cyanide for adsorption, activated carbon has been extensively used in the gold recovery process for the past decades. Since activated carbon can act both as an adsorbent and as a catalyst for the oxidation of cyanide, it has been also suggested for recovery of cyanide and metal-cyanide species from waste cyanide solution. An early technology is the Calgon process which employed columns packed with granular activated carbon to recover cyanide and metals from waste cyanide solution (Bernardin, 1976; Hoffman. 1973). In order to increase the kinetics of cyanide oxidation on the carbon, cupric ions and oxygen were added to the cyanide wastewaters before feeding them into the treatment system. The use of activated carbon as a modification for the AVR process to remove metal and cyanide has also been proposed (Batzias and Sidiras, 2001). Since plain carbon adsorption is believed to be not very efficient at removing free cyanide from the effluents, modification and impregnation technologies, such as Al, Cu, Ag and Ni — impregnated activated carbons, have been developed (Manktelow, et al., 1984, Adams, 1994, Williams, 1997, Adhoum and Monser, 2002). They suggested that the following reactions may occur during adsorption on surface of the impregnated activated carbon (such as when Ag or Ni was added), Ag + -COOH = -COOAg + H (2-16) Ni + -COOH = -COONi + H (2-17) where —COOH represent the acidic carboxyl functional group present on the carbon surface. Therefore, the adsorption capacity and the feasible removal rates of cyanides were substantially boosted since they were not only removed by adsorption on the surface of the plain carbon, but could be removed by those added chemicals (by forming Ag(CN)j and Ni(CN)42 on the surface). For those gold cyanidiation plants where activated carbon is already used for extraction process (CIP/CIL, carbon-in-pulp/carbon-in-leaching), the use of activated carbon for treating cyanide effluent would be simple since it is convenient for installation on-site. However, due to the low adsorption capability of activated carbon (even for those pre-impregnated carbons), it is 24 more suitable for use as a polishing process to remove cyanide to low levels when initial cyanide concentration is already low (for example, 1-5 mg/L, Fleming, 2005). The process has not yet been reported to be used in practice. 2.2.3 Ion exchange resin Though the commercial scale application of ion exchange resins in the gold mining industry was well established in the former Soviet Union in the 1970s, it was not until the late 1980s that such processes had attracted the attention of the Western World with the commissioning of Golden Jubilee resin-in-pulp plant in South Africa (Fleming and Cromberge, 1984 (A) and (). More research and investigations on the fundamental and practical aspects of ion exchange resin technologies on gold cyanidation have been produced since then (Bolinski and Shirley, 1996, Seymore and Fleming, 1986). Ion exchange resins also present a possible alternative for the treatment of waste cyanide effluents. As early as the 1950s, Walker and Zabban (1953) developed a bench scale ion-exchange resin process to concentrate cyanide from the aqueous waste streams produced in electroplating operations. Bessent, et al. (1980) conducted a pilot test on cyanide removal from coke plant wastewaters by selective ion exchange resins. FeSO4 was added to the stream prior to introduction to the ion exchange column where Amberlite IRA-958 (Rohm & Haas) was loaded. The excess of iron is precipitated as Fe(OH)3 under alkaline conditions. Though the cyanide concentration could be reduced below 2 mg/L, the disposal of the sludge generated in the process caused another problem since the sludge contains complicated hazardous materials (multi-metal cyanide precipitates, such as Fe2[Fe(CN)6]). Goldblatt (1956, 1959) developed an ion exchange resin process to recover cyanide and gold from the waste cyanide effluents arising from Stilfontein Gold Mine’ cyanidation operations. The strong base ion exchange resin, Amberlite IRA-400 (Rohm & Haas), was applied to remove cyanide and metals from recycled water containing CN, SCN, Zn, Ni, Co, and Cu. The system was comprised of two adsorption columns. The metal cyano complexes were removed in the first column. The effluent was then forwarded to the second column containing “CuCN-conditioned” resin where the remaining free cyanide was removed as copper cyanide. The treated effluent was returned to the leaching tanks. Both columns were then eluted with 1% 112S04 solution. CN was 25 converted to HCN and recovered as NaCN. The acid solution was then contacted with a strong acid resin IR-120 (Rohm & Haas) to remove dissolved metals. After several adsorption/elution cycles, copper cyanide (CuCN) was found to accumulate in the resins, causing the reduction of its exchange capacity. This solid was further removed as a soluble complex by elution with a ferric sulfate solution. Two typical resin technologies developed in recent years are AuGMENT process and Vitrokele process in both of which strong base resin were used to recover metals and cyanide from cyanide solutions. The use of guanidine-based resin to extract metals from gold leachate is also suggested (Kordosky, et al., 1993, Jermakowicz and Kolarz, 2002). 2.2.3.1 AuGMENT process The AuGMENT process was developed by SGS Lakefield research and the DuPont Corporation in which strong base resin (quaternary amine functionality) was used to recover and pre concentrate copper cyanide from gold-plant tailings (Fleming, 1998, Fleming, et al.. 1998). The chemistry involved in the various unit operations is based on the formation of different copper cyanide complexes as the cyanide to copper molar ratio is varied. CuCN precipitated resin was used as the adsorbent for the adsorption of free cyanide and soluble copper cyanide. The adsorption step is carried out with a barren solution containing a cyanide to copper molar ratio of at least four. It was believed that during adsorption, Cu(CN)32 or a higher complex reacts with CuCN producing Cu(CN)j which allowing the maximum copper loading (60-80 gIL resin) to be achieved. The loading mechanism can be described by the following equation: 2R-S04 (CuCN()) + Cu(CN)32 + 2CN —* 3R-Cu(CN)2+ R-CN + S042 (2-18) where R represents the resin matrix and functional group. Once loaded, the copper cyanide species are eluted from the resin using a concentrated copper cyanide solution having a cyanide to copper molar ratio of approximately four which can convert, the Cu(CN)i to Cu(CN)32.The elution process can be described as, 2(R-Cu(CN)) + Cu(CN)32 + 2CN —* 2R-Cu(CN)3 + Cu(CN)32 (2-19) Finally the resin is regenerated via conversion to the CuCN form with sulfuric acid. The eluate is submitted to electrowinning to produce copper cathodes. Gold has to be recovered prior to copper electrowinning and cyanide recovery. Cyanide can also be recovered via AVR circuit where the copper cyanide is precipitated and re-dissolved in the loaded catholyte ahead of the 26 electrowinning circuit. One potential disadvantage of the processes is that the precipitated CuCN may block the resin pores decreasing the opportunity for additional metal cyanide complexes to be adsorbed into the resin. If cobalt is present in the effluent, the possible polymerization of adsorbed cobalt cyanide complexes under strongly acidic conditions will poison the resins (Goldblatt, 1959, Leao, et al., 1998, Jay, 2001). 2.2.3.2 Vitrokele process Ion exchange resins have also been considered as a modification to the AVR process for indirect recovery of free cyanide and metals cyanide species (Silva, et al., 2003). A typical example is the Vitrokele process which uses strong base resins for recovery of cyanide and metals from either a clear solution or slurry. The resin is based on a highly cross-linked polystyrene structure (VitrokeleTM 911 and 912, which probably have the quaternary amine functionality, Jay, 2001). The loaded resins were eluted with the strong cyanide eluant to recover copper cyanide species. Precious metals and other strongly bound metal cyanide complexes were “crowded” from the resins with tetracyanozincate (Zn(CN)42).Sulphuric acid was used in the last elution cycle to destroy most of the cyanide complexes to regenerate the resins (Whittle, 1992). The chemistry involved in the process using strong base resins can be described as following, Loading: _____________________ {RN(CH3)}2[S04]2 + 2 [Au(CN)f= 2 RN(CH3) [Au(CN)2f+ [S04]2 (2-20) Stripping: 2 RN(CH3)[Au(CN)2f +[Zn(CN)4J2= {RN(CH3)}2[Zn(CN)4]2+ 2 [Au(CN)] (2-21) Regeneration: {RN(CH3)}2[Zn(CN)4]2 +H2S04= {RN(CH3)}2[SO4]2 + ZnSO4 + 4HCN(g) (2-22) One disadvantage of the process is that significant amounts of non-WAD cyanide species such as iron or cobalt cyanide complexes will poison the resins since only WAD cyanide complexes can 27 be removed by acid decomposition. This process has been successfully applied for treating the heap leachate at the Connemara Mine in Zimbabwe (Satalic, et a!., 1996). It was also tested at May Day Mines (Cobar, Australia) from July 1997 to June 1998 (DeVries, 2001). However, some technical problems developed during the operation of the process resulted in the abandonment of the Vitrokele process at the Mines. One of the major problems was that the elution of copper from the resin was not effective. The reaction between the residual copper cyanide in the resin with the eluant (H2S04)led to the formation of CuCN which blocked the active sites on the resin surface. 2.2.3.3 Cognis AURIX resin The AuRIX resin is a weak base ion exchange resin developed by Cognis for the recovery of precious metals (gold and silver) from gold cyanide leachate (Kordosky, et a!. 1993). It is a typical styrene-divinylbenzene resin bead functionalized with a guanidine functional group. Guanidines are strong organic bases having an intermediate basicity between that of simple amines and quaternary amines. Guanidines exhibit a pKa of approximately 12 and are capable of being protonated to form a guanidinium cation at the operating pH of the gold leachate (usually 10 to 11). This guanidinium cation can form an ion-pair with aurocyanide resulting in gold extraction from the cyanide solution. By increasing the basicity of the aqueous phase, the guanidinium cation is converted to the neutral guanidine functionality. The neutral guanidine functionality no longer forms an ion-pair with aurocyanide resulting in gold stripping from the resin. Ideally, the extraction of Au(I) from cyanide solution by the resin can be described by the following equations. RG + H20 = RGW OH (2-23) RGH OH + Au(CN)j RGH Au(CN)2 + OH (2-24) The overall reaction is, RG + H20 + Au(CN)2 = RGW Au(CN)j + OH (2-25) where RG represents the function group of the resin and RGW represents its protonated form (Virnig and Wolfe, et a!., 1996). 28 Ion exchange resins offer certain advantages over activated carbon. They are less easily poisoned by organic matter and can be eluted at room temperature, and selectivity for particular metals can be achieved by the choice of the functional group incorporated into the bead or by the selective elution process. However, the high operational cost of the resins has severely hampered their wide application in practice (Jay, 2001). 2.2.4 Solvent extraction 2.2.4.1 Recovery of gold cyanide complex Solvent extraction has been successfully practiced in many metal recovery processes, such as uranium, copper, and nickel. The use of solvents for purification and concentration of gold from dilute cyanide solution has long been of interest, but has never been practiced commercially. Some commercial base extractants which have been suggested for the recovery of gold and/or silver from cyanide solutions are summarized in Table 2-1. A variety of amines were found to be capable of extracting gold and silver from alkaline cyanide solutions. Both weak and strong base amines can behave in a similar manner with that of weak and strong-base resins. For example, a general equilibrium reaction for Au(I) extraction by the primary amine can be expressed as: RNH2org + H + Au(CN)2org = RNH3Au(CN)2org (2-26) where R represents the alkyl group associated with the amine (Caravaca, et al., 1996 (A); Caravaca, et al., 1996 (B)). More technical papers emerged after the original findings that showed the selective extraction of cyano-anions could be accomplished by addition of modifiers (e.g. organophosphorous compounds) to weak base extractants (such as primary, secondary and tertiary amines), which make it possible to effectively extract gold cyanide in alkaline conditions (Miller and Mooiman, 1984, Mooiman and Miller, 1986). Quaternary amines can strongly extract common anions with less selectivity but are more difficult to strip. Riveros, et al. (1990) have even run a small pilot plant to demonstrate the potential application of quaternary amines (Aliquat 336 dissolved in Solvesso 150) for recovery of gold from cyanide leachate. This process seems difficult for further applicability in industry since acidic thiourea was recommended as the stripping agent. This resulted in a more complicated process. 29 Table 2-1 Some commercial base extractants for gold solvent extraction (Rydberg, et al., 2004) Reagent Class Structure Commercial Extractants Primary amines RNH2 Primene JMT R=(CH3)C(CH4 Secondary R’R2NH . amines 1=C99CHCHCH Amberlite LA-i =CH3C(C)( R1=CH3(C2)11 Amberlite LA-2 =CHC(C( R1,2=CH3()12 Adogen 283 R1,R2=CHH(CH)C(6 HOE F2562 Tertiary amines R123N R1,R2,3=CH()7 Alamine 300 R1,R2, R3 =C8-10 mixture Alamine 336, Adogen 364, Hostarex A 327 R1,R2, R3 = (CH)2CH(CH5 Adogen 381, Alamine 308, Hostarex A 324 R’, R2, R3 = (CH)2CH(CH7 Alamine 310, R1, R2, R3 = (CH3)(C11 Adogen 363, Alamine 304, R1, R2, R3 = CH3(C2)12 Adogen 383 R1,R2, R3 = C28H57 Amberlite XE 204 =CH3()7R2 = CH3(C2)9,R3 = Adogen 368 CH3(C2)11 Quaternary RN(CH3) Cl . Aliquat 336, Adogen 464, amines R1, R2, R3 = C8-C10 mixture HOES 2706 In recent years, solvent extraction systems such as guanidine derivatives were also developed in order to find a suitable extractant for gold (and/or silver) recovery from dilute cyanide solutions. After the introduction of this functionality as AURIXTM 100 ion-exchange resin by Cognis, the solvent product with the similar functionality, LIX® 79, a tri-alkylguanidine extractant was also 30 developed (Kordosky, et al., 1992, Virriig and Wolfe, 1996). The extractant allows the extraction of gold or silver up to an aqueous pH of 11 or more and the loaded gold can be stripped off with strong basic solutions. Following to the same principle as AURIX resins the guanidine functional group undergoes protonation to form guanidinium cation when contacting with an aqueous solution. The guanidinium cation can form an organic soluble ion pair with the anions in the aqueous phase which resulting in their extraction. RGorg + H20 = RGH OH org (2-27) RGHOHorg + Au(CN)2 = RGH Au(CN)2org + OH (228) The overall reaction is, RG org + H20 + Au(CN)i = RGH Au(CN)i org + OH (2-29) where RGorg represents the extractant molecule and RGHorg its prononated form (Vimig and Wolfe, et al, 1996). 2.2.4.2 Recovery of metals and cyanide Due to the potential high selectivity and loading capacity, and the relatively fast extraction rate, solvent extraction technology also offers an alternative method for recovery of metals and cyanide from waste cyanide solution. In the early work of Moore (1975), and Moore and Groenier (1975), solvent extraction of zinc and cadmium from alkaline cyanide solutions was examined. They found that quaternary amines (Aliquat 336 and Adogen 464) have a good ability to extract zinc and cadmium cyanide from highly alkaline solutions. Regeneration of the amine solvent can be achieved by stripping off the loaded metals (Cd and Zn) with sodium hydroxide (NaOH), sodium hypochiorite (NaC1O), or alkaline or acidic formaldehyde (HCHO). Villaverde and Martin (1995) examined the feasibility of solvent extraction of gold and silver from the Gossan barren dam waters originating from the gold cyanidation process. Different extraction solvents including Primene (ATP, a primary commercial amine), TBP (tributylphosphate), and Cyanex (general formulaC24H510P, not specified by the author), and their combinations were investigated. They suggested that it was possible to separate and concentrate gold, and to a lesser extent silver and copper, by means of solvent extraction with the synergistic extractant, Cyanex + ATP. The waste dam water could be treated directly at pH 9 and overall 90 % gold recovery can be achieved. The process using organophosphorous extractants Cyanex 923, di-butyl-butyl 31 phosphonate and tri-butyl-phosphate for the recovery of hydrogen cyanide (HCN) was also developed. The process has been suggested to be used to replace the air sparging step in the AVR process (after acidification of the cyanide solution) since the cost of air sparging is usually substantially high (Larmour-Ship. K, et al., 2005). Dreisinger, et al. (1995) investigated solvent extraction of copper from dilute cyanide solution by LIX 79 and proposed the development of a copper SX/EW process for the treatment of waste cyanide solutions or alternatively as a front end for any of the other final copper cyanide recovery process. The process is carried out in four stages, solvent extraction of copper cyanide complexes from clarified solution by LIX 79; stripping copper cyanide from loaded organic phase using a high pH and copper cyanide rich spent electrolyte; electrolysis of the strip solution in a membrane cell (NafionTM 417 membraneTM,Du Pont) to produce copper metal and liberate free cyanide; and cyanide recovery from a bleed stream from the electrolysis cell (Dreisinger, et a!., 2001). Davis, et al. (1998) proposed the use of LIX 7800 series extractants (the mixture of the quaternary amine Aliquat 336 and nonylphenol at different molar ratios) as the pre-concentration step to recover copper from copper cyanide solutions. The extraction and stripping of copper cyanide complexes are believed to occur via ion exchange mechanism. (Q X)org + (HP)org + OH = (QP)org ÷ X + H20 (1-37) where Q is the quaternary ammonium cation, HP is the protonated form of the nonylphenol, and X is the extracted anion (Mattison and Vimig, 2001). Under low pH conditions, nonylphenol is protonated and the quaternary ammonium compounds extract an anion from the aqueous phase. Under highly alkaline conditions, nonylphenol starts to be significantly converted to phenoxide anion (F) and forms an ion pair with the quaternary ammonium cation (QP). Consequently the extracted anion will be gradually expelled to the aqueous phase with increasing equilibrium pH. The economic aspects of the potential application of the process to recover copper from cyanide solution have been developed based on the extraction results. 32 2.2.5 Miscellaneous Lower and Spottiswood (1983) reported a process of cyanide removal from coke making and blast furnace wastewaters by ion floatation of iron cyanides. The process was found reasonably effective on ferricyanide but not on CN and ferrocyanide. Soto, et al. (1997) developed a method by adjusting solution pH to recover cyanide and copper from cyanide solutions containing copper and thiocyanate. Copper is first precipitated as copper thiocyanate (CuSCN) or as copper cyanide (CuCN) depending on pH and the concentration of thiocyanate and cyanide in the effluent. The precipitates then can be separated from the effluent by filtration after settling. The decant solution and the filtrate containing the bulk of the cyanide were then oxidized with ozone to transform the remaining thiocyanate into cyanide. Cyanide is not oxidized by ozone under the adopted conditions. The regenerated solution rich in free cyanide can be recycled to the cyanidation process. Recovery of over 96 % copper and cyanide were reported in their laboratory tests. Lu, et al. (2002) developed a membrane-electrolysis cell with graphite felt to recover copper and to recycle the cyanides. Copper recoveries up to 60% were achieved with an energy consumption of 1-2 kWh/kg. The process has been suggested as a subsequent process for treating the eluent or stripping solution from copper pre-concentration step in treating cyanide effluent by either ion exchange resin or solvent extraction. The study on the bio-sorption of heavy metals ions from cyanide solutions using a waste fungal biomass containing killed cells of Aspergillus niger was conducted by Nataralan, et al. (1999). The uptake of base metals from the industrial cyanide effluent was examined. It was found that the biomaterials tested in the study have shown high value of metal uptake from cyanide effluent particularly for gold and zinc. The bio-sorption process of metal-cyanide complexes, tetracyanocuprate(II), Cu(CN)42 and tetracyanonickelate(II), Ni(CN)42 from waste cyanide solutions by using different fungal cultures were studied by Patil and Paknikar (1997). They observed that the fungi (C. Cladorporioides) showed maximum loading capacity (40 imo1/g Cu(CN)42,and 34 jimol/g, Ni(CN)42)comparing to the other fungal absorbents studied and the activated charcoal. It was found that 1 mol/L sodium hydroxide was effective to remove the bound metal-cyanide species which could be concentrated to allow recycling in the plating circuit in the user industry. 33 2.3 Research objective To date the detoxification of cyanide effluent arising from the gold cyanidation process is still a challenge to the gold mining industry, especially those gold plants dealing with gold ores containing high concentrations of copper minerals. The current cyanide recovery technologies including the AVR process and adsorption by ion-exchange resins or activated carbons are either too expensive to construct and operate, or have relative low loading capability and hence are impractical. Though still underdeveloped, solvent extraction technology offers an alternative recovery system for treating those high tonnage wastewaters containing cyanide and valuable metals like copper. Since the extraction kinetics of the solvent extraction systems are usually fast and the process can operate through continuous stages, relative small organic inventory will be required. An attractive application of this process is to incorporate a solvent extraction circuit as a pre-concentration step to treat the high copper cyanide-containing waste solutions (shown in Figure 2-1, Dreisinger, et al., 2001). Ideally, the solvent extraction of metal species from waste cyanide solution should be accomplished at around pH of 11 with stripping achieved at a moderate higher pH (for example, pH of 12 -13). This requires that the extractant exhibits an appropriate basicity and has the desired specificity for the targeted anions. The feasibility of the use of LIX 7820 (a mixture of quaternary amine Aliquat 336 and nonyiphenol) and LIX 7950 to recover copper from cyanide solution has been proven. However, the fundamental aspects on these extraction systems for treating cyanide effluents, especially the knowledge of the extraction behavior of cyanide and the potential effect of other anions on the recovery of copper and cyanide are still lacking. Much research work needs to be conducted to elucidate the characteristics involved in the extraction of the common anions occurring in the cyanide solution by the extractants to predict their potential application. In this work, the extraction chemistry of common anions occurring in cyanide effluents arising from gold cyanidation process by two commercial extractants (LIX 7820 and LIX 7950) has been examined. The possible solution to recovery of valuable metal (subjected to copper) and cyanide was examined. More specifically, the research was designed to determine: 34 > The extraction chemistry of metal cyanide complexes (principally copper cyanides) with two extractants under different experimental conditions, including the effect of the organic formula, the molar ratio of cyanide to copper and the temperature; > The co-extraction behavior of mixed metal cyanide solution and the selectivity of the extractants for the specific metal (copper); The potential effect of non-metal cyanides anions including SCN, CNO, S203 on the extraction of copper and cyanide with the extractants; > The appropriate reagents for stripping of loaded copper and cyanide; > The extraction and stripping isotherms of copper and cyanide with the two extractants > The potential solution to recovery of copper and cyanide from waste cyanide solution Reagents Copper — Cyanide _________ Cu-CN Barren Solution Solution Solvent Ext. To Recycle or Disposal Concentrated Cu — CN Solution ‘1 _ Jr Reagents and Power ‘+ Cu EW SART AVR ‘1r 1, Copper Metal Cu2S and CuCN and + Recycle CN Recycle CN Recycle CN Figure 2-1 Use of solvent extraction in the recovery of copper and cyanide from solution (After Dreisinger, et a!., 2001) 35 3 Experimental 3.1 Organic reagents and chemicals Two extractants, LIX 7950 and LIX 7820 which are commercial extractants produced by Cognis were used in the research. Similar to the guanidine extractant LIX 79, LIX 7950 is also based on formulation of an alkylguanidine but has a higher concentration of guanidine and shows a higher basicity than LIX 79 (which has a pKa of 12 or higher, Kordosky, et a!., 1992). LIX 7820 is a solvent mixture of Aliquat 336 (a commercial quaternary amine produced by Cognis) and 4- nonylphenol at a molar ratio of 1: 2. Some information of the extractants, LIX 79 and Aliquat 336 is summarized in Table 3-1 (Kordosky, et a!., 1992, Mattison and Virnig, 2001). Information regarding the diluents and modifiers used in this research is summarized in Table 3-2. Table 3-1 Some information on LIX 79 and Aliquat 336 LIX 79 Aliquat 336 Molecular Formula N C25H541N Molecular weight, g/mol N 404.16 Specific Gravity (25° C) 0.80-0.85 0.884 Schematic structure R1 (CH )7CH3 N )NcN( CH3/ 4 R3 (CH2)7C3 CH3(C2)7 R = H or alkyl N: No information available 36 3.2 Preparation of the extractant solvents The extractant solvent containing LIX 7950 was prepared by dissolving a desired volume of the extractant and a designated mass of the modifier (1-dodecanol) into the diluent. The extractant solvent was then used directly for solvent extraction tests without any pre-treatment. The extractant solvent for LIX 7820 was prepared by dissolving a desired volume of the extractant into the diluent. The solvent mixture was first washed with 1 mol/L sodium hydroxide (NaOH) solution three times and then washed with de-ionized water twice. Through this procedure, the extractant LIX 7820 was converted from chloride form to hydroxide form. The extractant solvents with different molar ratios of Aliquat 336 and 4-nonylphenol were also prepared following the same procedure before use in the extraction tests. Table 3-2 Some information on diluents and modifiers used in the research Chemicals Formula Molecular Specific Purity Supplier weight, g/mol Gravity(25° C) 4-nonyiphenol C15H240 220.35 0.940 95% Cognis 1-octanol C8H180 130.23 0.824 98% Acros 1-dodecanol C12H260 186.34 0.83 1 98% Acros n-octane C8H18 114.23 0.707 97% Acros Decane C10H22 142.29 0.730 99% Acros n-dodecane C12H26 170.34 0.750 97% Acros Toluene C7H8 92.14 0.867 99.9% Fisher 3.3 Preparation of aqueous solutions Information regarding the inorganic chemicals used in this research is summarized in Table 3-3. De-ionized water was used in the preparation of aqueous solutions. Synthetic metal cyanide solution was prepared by dissolving the corresponding metal cyanide salt in an aqueous NaCN solution. A mixture solution of metal cyanides was prepared and the content of the major components is shown in Table 3-4. Copper content in the waste cyanide solution arising from gold mining industry may vary from several mg/L to 100 gIL depending on copper minerals and 37 their content in gold ores. The copper content of 250 mg/L in the mixture solution was chosen in this research. For comparison, the concentration of other metals was established at the equivalent molarity to that of copper. Cyanide, cyanate (CNO), thiocyanate (SCN), and thiosulfate (S203) solutions were made up from their sodium salts, respectively. Table 3-3 Some information of inorganic chemicals used in the research Chemicals Formula Molecular Purity Supplier weight, g/mol Copper(I) cyanide CuCN 89.56 99 % Aldrich Zinc cyanide Zn(CN)2 117.42 99 % Fisher Nickel(II) cyanide tetrahydrate Ni(CN)2 4H20 182.73 99 % Alfa Aesar Potassium hexacyanoferrate(II) K4Fe(CN)63H0 422.41 99.5 % Anachemia Potassium hexacyanoferrate(III) K3Fe(CN)6 329.26 99.2 % Fisher Sodium cyanide NaCN 49.01 95 % Anachemia Sodium thiocyanate NaSCN 82.07 99.6 % Fisher Sodium thiosulfate NaS2O3 158.11 99.9 % Fisher Sodium hydroxide NaOH 40.00 99.2 % Fisher Sodium sulfate Na2SO4 142.04 99.8 % Fisher Sodium chloride NaCl 58.44 99 % Fisher Sodium nitrate NaNO3 84.99 99.1 % Fisher Potassium chloride KC1 74.56 100 % Fisher Lithium chloride LiCl 42.39 99.7 % Fisher Sodium cyanate NaCNO 65.01 97 % Fluka 38 Table 3-4 The major components in the mixture solution of metal cyanides Component Concentration, mg/L Cu(I) 250.0 Zn(II) 255.7 Ni(II) 230.1 Fe(II) 220.3 CNT 2045.6 3.4 Test procedure For the experiments where pH/temperature was controlled, the tests were conducted in a sealed cylinder cell shown in Figure 3-1. Mixing of the two phases was provided by a mechanical agitator with glass impellers. A glass pH probe was used to measure the solution pH. The temperature was controlled through a thermostatic water-bath. For the pH/temperature- uncontrolled tests, a shaking machine (Figure 3-2) was used for mixing the organic and the aqueous solutions in the sealed glass bottles. The extraction and stripping tests were carried out at a relatively small scale (25 to 50 ml volumes of phases) with various phase ratios. Equilibrium pH was adjusted by direct addition ofH2S04 solution (10% v/v) and NaOH solution (2 mol/L). When equilibrium was established, phase separation was conducted in a 125 mL separatory funnel (Figure 3-3). The aqueous solution was first separated and filtered to remove any entrained organic before analysis (with No. 3 filter paper, Fisher). The organic sample was then filtered with 1PS phase separation paper (Fisher) to remove the entrained aqueous solution before any analysis or stripping tests. A duplicate test was conducted to ensure the consistency of the results (less than 5% variation). 3.5 Analysis The metal content in aqueous samples was analyzed by atomic absorption spectroscopy (AAS). Total cyanide content in the aqueous solution was determined by the distillation method (Csikai 39 and Barnard. 1983). The metal content and total cyanide in the organic phase were calculated by the mass balance. Some selected loaded organic samples were stripped with concentrated NaOH solution (1 mol/L) three times and the stripping solutions were collected for analysis of metal and cyanide. The mass balance calculation confirmed that for copper, an accuracy of 97% between analytical and calculated results could be usually obtained. Analysis of the total cyanide shows that an accuracy of 95% could be achieved. The analysis procedure for determination of free cyanide and thiocyanate is attached in Appendix I. The analysis of cyanate, sulfate, thiosulfate, chloride, and nitrate ions in aqueous samples was conducted by IPL (International Plasma Labs Limited), a certified assayer located in Richmond, BC, Canada. The species in the organic phase were calculated based on mass balance. FTIR (Fourier Transform Infrared) spectra were taken to record CN stretching vibrations in the blank solvent and the solvent with loaded copper cyanides. The JR spectra were recorded with 40 scans at a resolution of 10 cni1. Figure 3-1 The schematic experimental set-up for pH and temperature-controlled tests (1 stirring controller; 2 — thermostatic water bath; 3 — extraction/stripping cell; 4 — 1 mol/L NaOH solution; 5 — condenser) 40 Figure 3-2 Picture of the shaking machine for pH-uncontrolled tests Figure 3-3 Picture of the separation and filtration apparatus 41 4 Extraction with LIX 7950 4.1 Terms and definitions Some terms and definitions related to the extraction process are highlighted below: The measure of efficiency for the extraction process is given by the distribution coefficient, D. D is defined as: — [A1org A — [A1aq (4-1) where [A”]org is the molar concentration of the extracted anion in the organic phase (mol/L) and [A”iaq is the molar concentration of the anion in the aqueous phase (mol/L). It should be specified here that for metal species, the value of D is for all the metal species rather than one species. For two extracted metals (A and B), the separation factor is defined as followed: SAIB =DL (4-2) where S is the separation factor of A over B; DA and DB are the distribution coefficients for species A and B, respectively. The extraction of a species, E, can be expressed as a function of the distribution coefficient and the volume ratio of organic and aqueous phase by: E = 0* [A I * 100% * ( D ) * 100% (4-3)A*[A]+0*[A] A A where 0 and A represent the volume of the organic phase and the aqueous phase, respectively and D is the distribution coefficient. For comparative purposes, a term of pH50 is defined as the equilibrium pH at which 50% extraction (E = 50%) is achieved. The value of pH5o can be obtained by plotting E (%) vs equilibrium pH. 42 4.2 Equilibrium time A preliminary test was conducted to evaluate the effect of contact time on copper extraction from copper cyanide solution with LIX 7950. The organic solvent was composed of 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane. The copper content in the aqueous solution was 3.93 x 10 mol/L (equivalent to 250 mg/L copper) and the molar ratio of cyanide to copper (CN/Cu) was 5:1. The initial solution pH was 10. The results are presented in Figure 4-1. Under the established experimental conditions, copper extraction increased significantly in the first minute of mixing. After 3 minutes of contact, further prolonging agitation could not increase copper extraction any more. The solution pH was also constant after 2 minutes of mixing, indicating the copper extraction kinetics were relatively fast and the equilibrium between the two phases could be established rapidly. The influence of the contact time on the extraction of other metal cyanide solutions (Zn, Ni, Fe cyanide solutions) with the extractant also exhibited similar behavior. The results are in accordance with those reports that the extraction kinetics of gold from cyanide solution with guanidine extractants and modified amines were all relatively fast (Caravaca, 1994, Kordosky, et al., 1992). In the subsequent extraction/stripping tests, a contact time of 10 minute has been arbitrarily chosen when equilibrium pH was uncontrolled. For pH controlled tests, an additional 5 minutes of mixing was applied when the solution pH reached a constant value in order to establish equilibrium. 43 100 12 pH A A A 80 0 11 0 I. Cu extraction 40 20 10 0 I I 9 0 2 4 6 8 10 12 Contact time, mm Figure 4-1 Plot of variations of copper extraction and pH vs contact time (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu =5; A/O = 1; 25°C) 4.3 Organic formula 4.3.1 Effect of diluents The term diluent refers to the organic liquid in which the extractant and modifier are dissolved to form the solvent. The effect of two different types of diluents (toluene which is aromatic and n dodecane which is aliphatic) on copper extraction with LIX 7950 was examined. Experiments were carried out with the organic solution of 10% v/v LIX 7950 and 50 g/L 1-dodecanol in each diluent and the aqueous solution of 3.93 x103 mol/L copper (CN/Cu = 5). Plots of copper extraction as a function of equilibrium pH for two diluents are shown in Figure 4-2. The change of the diluents has shown a small but measurably influence on copper extraction and the pH5o value for n-dodecane is slightly higher than that for toluene (11.3 and 11.0, respectively). A 44 similar phenomenon has been observed by Sastre, et al. (1999) who investigated the influence of diluent types on solvent extraction of gold cyanide with the guanidine extractant LIX 79. They found that the pH5o value in n-heptane was slightly higher than that in toluene or in cumene. The extraction of silver from cyanide solution by LIX 79 has also shown a slightly higher pH5o in n heptane than in cumene (Sastre, et a!., 2004). The relative research results are summarized in Table 4- 1. It seems that the guanidine extractants (LIX 79 and LIX 7950) extract metal cyanide complexes more efficiently in aliphatic diluents than in aromatic diluents. However, the effect of diluent types on the pH50 values for the extraction of gold by modified amines showed that the change of the aromatic diluent to aliphatic diluent caused a slight reduction on pH5o values (Miller and Mooiman, 1984). Though there is still no strong evidence, the interaction between the extractant and the diluent is believed to play an important role. It should be pointed that the solubility of toluene in water (0.59 gIL, 25 °C) is higher than that of n-dodecane (8.42 x106 gIL) (Poulson. et al., 1999). This loss of extractant in the aqueous phase due to the high solubility of toluene may also play a role in the reduced copper extraction. The high solvent loss is definitely undesirable in practice. 100 —-—n-dodecane 80 —.-—Toluene C o 60 0 I I 9 9.5 10 10.5 11 11.5 12 12.5 13 Equilibrium pH Figure 4-2 The effect of diluents types on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25°C) 45 Table 4-1 Comparison of the effect of diluent types on extraction of metal cyano complexes LIX 79 Cumene ____________ LIX 79 n-heptane LIX 79 Toluene LIX 79 Cumene LIX 79 n-heptane LIX 79 Kerosene _____ LIX 7950 Toluene LIX 7950 n-dodecane N/A — no information available; 4.3.2 Effect of modifiers It was found that a third phase (a sticky yellowish compound) was produced when the extactant solvent (10% v/v LIX 7950 in toluene or n-dodecane) was contacted with the copper cyanide solution (3.93 x i0 mol/L Cu, CN/Cu = 5). This phenomenon does not happen when a very dilute copper cyanide solution (10 mol/L Cu, CN/Cu = 5) was used. The formation of the third phase can be overcome when a certain amount of the modifier (1-dodecanol or octanol) was added to the organic phase before contacting with the copper cyanide solution. For the same aqueous solution (3.93 x 10 mol/L Cu and CN/Cu = 5), a small amount of the third phase was produced when 10 gIL of 1-dodecanol (10 % LIX 7950 in n-dodecane) was used. When the concentration of 1-dodecanol was increased to 25 gIL, no third phase was produced. Obviously a certain minimum amount of the modifier is required to avoid the formation of third phase. There was little change on copper extraction between the cases of 25 g/L and 50 gIL 1-dodecanol were used and the pH5o values in the two cases are very close (Figure 4-3). Table 4-2 shows that copper extraction exhibited negligible change when the modifier changed from 50 g/L 1- Extractant diluent modifier Target pH5o Reference metal N/A Ag 10.1 Sastre. et al., 2004 N/A Ag 10.3 Sastre. et al., 2004 N/A Au 10.4 Sastre, et al., 1999 N/A Au 10.6 Sastre, et al., 1999 N/A Au 10.7 Sastre, et al., 1999 N/A Au 10.8 Sastre, et al., 1999 1-dodecanol Cu 11 This research 1 -dodecanol Cu 11.3 This research 46 0 ‘I-’ 0 ‘I-. x 0 Equilibrium pH Figure 4-3 The effect of the concentration of 1-dodecanol on Cu extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25 °C) Table 4-2 The effect of modifier types on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu = 5; A/O = 1; 25 °C) pH Cu extraction, % 10% v/v LIX 7950 and 50 g/L 10% v/v LIX 7950 balanced 1-dodecanol in n-dodecane with 1-octanol 10.0 99.1 98.8 11.0 65.1 65.0 11.5 31.2 31.5 12.0 6.5 6.2 - dodecanol (10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane) to 90% v/v 1-octanol (10% v/v LIX 7950 balanced with 1-octanol). 100 75 50 25 0 9 9.5 10 10.5 11 11.5 12 12.5 13 47 The use of alcohols as the modifier in the organic phase to facilitate the extraction of metal cyanide by guanidine extractants has been reported by many researchers. The investigation conducted by Kordosky, et al. (1992) reported that viscous and slimy materials formed at the organic-aqueous interface and on the settler walls in extraction of gold from cyanide solution by a guanidine extractant (N, N’-bis(tridecyl)guanidine). The third phase materials were more soluble in the mixture of Exxon Aromatic 150 with 10% tridecanol than in Exxon Aromatic 150 only. The authors believed that these viscous materials were a mixture of the metal complexes of the guanidine extractant. The research results of Gonzalez (1994) indicated that during the extraction of gold cyanide by a dodecyl guanidine, the gold extraction was not changed when the volume ratio of octanol and toluene changed from 1/3 to 3/1, but a better extraction of gold was obtained when full strength of octanol was used. The author believed that the modifier can facilitate the displacement of the acid-base equilibrium between guanidine and guanidinium ion to the acid form due to the stabilization of the ions by solvation. Since the conjugate acids of the guanidines have PKa values as high as 12 - 13, the guanidine basicity is sufficiently great to react with a protic modifier and form the guanidinium ion through the following reaction: ROll ÷ RG = RGW + R0 (4-4) where ROH is an alcohol modifier and RG is the guanidine extractant (all the species are in the organic phase). However, the investigation on gold solvent extraction with LIX 79 shows that there was not an evident effect on global extraction efficiency when the concentration of the modifier (tridecanol) was varied (Aguayo, et al., 2007), but an emulsion was produced when a low concentration of tridecanol (5% v/v) was used. The emulsion was reduced in size considerably at a high level of tridecanol concentration (i. e., 10% v/v). No relative information on the modifier have been given in the research on the extraction of gold and silver from cyanide solution by LIX 79 conducted by Sastre, et al.(2004) and Sastre, et al. (1999). A much higher concentration of tridodecanol (100 g/L) was used as the modifier by Dresinger, et al. (1996) when the guanidine extractant XI 7950 was used for copper extraction from the solutions with high concentration of copper cyanides (> 2 g/L Cu) with the purpose to avoid emulsion. Ritcey and Ashbrook (1979) suggested that the main function of the modifiers was to increase the solubility of the compound of the extractant and extracted metal complexes in the aprotic diluent. The necessary amount of a modifier in the organic phase will depend on the properties of the extactant, the diluent, the modifier itself and sometimes the properties of the aqueous phase. The 48 results in this research seems to support this viewpoint that a certain minimum amount of the modifier (alcohols) is required to avoid the formation of emulsion when using LIX 7950 to extract copper cyanide complexes. More excess of the modifier exhibits little effect on copper extraction by the extractant. 4.3.3 Effect of the extractant concentration The tests for extractant concentration effect on the extraction of copper and cyanide with LIX 7950 were carried out with relatively dilute copper cyanide solutions ([Cu] = 9.45 x10 mol/L and 9.45 x103 mol/L which are equivalent to 60 mg/L and 600 mg/L of copper, respectively, CN/Cu = 3, initial pH 10.00 ± 0.05). The organic phase was composed of various amounts of LIX 7950 and 50 gIL 1-dodecanol in n-dodecane. The extraction tests were conducted at a phase ratio of unity and 20 °C. The solution pH was uncontrolled during extraction and the equilibrium pH was measured when the equilibrium was established. It was found that the equilibrium pH for the tests varied in the range of 10 to 11. According to the copper speciation, Cu(CN)32 is the dominant anion in the initial aqueous solution which accounts for more than 98% of total copper and cyanide (it may vary in a slight extent depending on the formation constants chosen for the copper cyanide complex species). The analysis of the extraction of copper and cyanide indicated that the molar ratios of the loaded copper to cyanide were all close to 3 and so were the CN/Cu ratios in the raffinate. It is thus assumed that the overall extraction of copper and cyanide under the experimental conditions is equivalent to the following equilibrium: 2RGorg + 2H0 + Cu(CN)32 = 2(RG H)Cu(CN)3org + 20Ff (45) Define the stoichiometric extraction constant, Kex, as — [2(RGH )CU(CN)3]org[OH]2 ex — [Cu (CN )32_ ][RG ]org 2 (4-6) where [2(RGH)Cu(CN)3]orgis the molar concentration of the compound of the extractant and the copper cyanide complexes (which is equal to the molar concentration of loaded copper). [RG]org is the molar concentration of the free extractant in the organic phase. If ideal behaviors for all the related species in both organic and aqueous phase are hypothesized and supposed there is no aggregation of the extractant, [RG]org can be calculated from the mass balance equation: 49 [RGjorg = [RG]Torg — 2 [2(RG H)Cu(CN)3org (47) where [RG]Torg is the concentration of the total extractant (the initial concentration of the extractant. 10% v/v of LIX 7950 is equivalent to 0.0133mo1/L). The distribution coefficient of copper, Dc, can be expressed as follows: D Cu = [Cu ] org = [2 (RGH ) Cu (CN ) ] org (4-8)[Cu [Cu (CN ) ] where [Cu]org and [Cu]aq are the molar concentration of Cu(CN)32 in the organic phase and in aqueous phase, respectively. Combining equations (4-6) and (4-8) leads to the following expression: log (Dc[OFf]2)= 2 log [RG]org + log Kex (4-9) Hence, in the ideal conditions, if log (Dc[OH]2)is plotted against log [RG]org, a straight line should be obtained with a slope of 2 and an intercept equal to log The experimental data are plotted in Figure 4-4. A linear relationship with a slope close to 2 is obtained between log (Dc[OW]2)and log [RG]org when the initial concentration of copper varied from 60 mg/L to 600 mg/L, indicating the effectiveness of equation (4-9). According to the intercept, the logarithm stoichimetric extraction constant of copper (log K) under the experimental conditions is -1.17 which is close to the corresponding average value of log K calculated numerically which is -1.34 ± 0.08 (based on equation (4-6)). 50 -3 -3.5 -J -4 0 I .5 -2.4 -2.2 -2 -1.8 -1.6 -1.4 -1.2 -1 log [RG]org, mol/L Figure 4-4 Plot of log Dcu[OW]2vs log [RG]org at different metal concentrations (Org.: LIX 7950 and 50 gIL 1-dodecanol in n-dodecane; aq: CN/Cu = 3, initial pH 10.00 ± 0.05; A/O=1;20°C) 4.4 Effect of temperature The tests for temperature effect on the extraction of copper and cyanide with LIX 7950 were carried out with the dilute copper cyanide solution ([Cu] 9.45 x104 mol/L, CN/Cu 3, initial pH 10.00 ± 0.05) and the organic solvent of 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n dodecane. The temperature was controlled (varied from 25 °C to 45 °C) while pH was uncontrolled during extraction. The final pH was measured when the equilibrium was established. The loaded organic solvent was stripped with 1 mol/L NaOH solution three times. The stripped solvent was re-contacted with the original copper cyanide solution. It was found that the solvents from different temperature tests exhibited the similar extraction capability for copper and cyanide after stripping, indicating the functionality of the extractant exhibit insignificant change before and after test. y=2.09x-1.17 R2 = 0.99 o[Cu]=600mgJL A [Cu]=60 mg/L 51 The extraction results are summarized in Table 4-3. It shows that the equilibrium pH decreases with an increase of temperature. The distribution coefficients for copper and cyanide also decrease when temperature is increased. Similar results was noticed by Caravaca, et a!. (1996) and Caravaca. et a!.. (1996) (B) who noticed that the distribution coefficients for the extraction of Au(CN)2 by primary amines decreased with an increase of temperature. However, no information on the temperature effect on extraction of metal cyanide complexes by guanidine extractants has ever been reported. The temperature effect can be explained qualitatively in terms of the degree of anion hydration. According to the ion-exchange mechanism (reaction (4-5)), hydroxyl ions will be released into the aqueous phase when copper cyanide complexes are extracted into the organic phase (for example, for the extraction of one molecule of Cu(CN)32, two molecules of OH will be liberated to the aqueous phase). Since OH ion has a relative large hydration number and a large hydration free energy, it is highly hydrated in water. A typical example is the dissolution of NaOH which is a strongly exothermic process due to the large entropy involved in the hydration of OH ions. There is a possibility that the protonation of the extractant is depressed at an elevated temperature. The solvation energy involving the hydration of copper cyanide complexes are much less compared to that of OH since they are large anions and less hydrated in water (Riveros, 1993). As a result, the overall extraction of copper and cyanide with the extractant coupled with the release of OH to the aqueous phase is an exothermic process and a high temperature depresses the copper extraction with LIX 7950. Table 4-3 The effect of temperature on copper extraction with LIX 7950 (Org: 10 % v/v 7950 and 50 g/L 1-dodecanol in n-dodecane; aq: [Cu] = i0 mol/L, CN/Cu = 3; A/O=1) T, °C pHeq* log Dc 25 10.94 0.95 30 10.75 0.77 35 10.62 0.54 40 10.40 0.41 45 10.31 0.21 * pH was measured at the indoor temperature (20 °C) after phase separation. 52 The van’t Hoff equation in term of the extraction constant can be written as, logK =— +C (4-10) 2.303RT where AH° is the enthalpy change for the extraction reaction (4-16) and C is integration constant which is equal to -AS°/2.303R in ideal conditions. The values of log Kex at different temperatures were calculated based on equation (4-9). The plot of log Kex vs 1000!T is presented in Figure 4- 14 and a linear relationship is obtained. It shows that the copper extraction constant (Kex) decreases when temperature is increased, indicating the overall extraction process is exothermic. The enthalpy change of the reaction was evaluated from the slope which gave a value of -191.1 kJ/mol (tsH° = 2.303R*l000*SIope). On this basis, AS° was calculated to be about -0.66 kJ/K mol. The negative value of AS° suggests the formation of the compound of copper cyanide complexes with the extractant (2(RGH)Cu(CN)3)is probably a more ordered structure than that of the protonized extractant (RGHOH). The extraction of copper cyanides decreases appreciably with an increase of temperature is in great extent because of the negative entropy term. Similar results have been reported for the extraction of gold from cyanide solution by Tridecylamine (primary amine, Caravaca. et al., 1996 (B)) and by the solvation extraction system (Miller, et al, 1987). However, positive values for AS° were obtained for some primary amine extraction systems (Table 4-4), indicating the effect of temperature on solvent extraction of metal cyanides is highly depending on the extraction system adopted, and also possibly the properties of the aqueous solution. 53 —1 C, 0 -2 -3 -4 3.1 I 000ITIW1 Figure 4-5 Plot of the calculated Log Kex VS 1000/T (Org: 10% v/v 7950 and 50 gIL 1-dodecanol in n-dodecane, aq: [Cu] = i0 mol/L, CN/Cu = 3; A/O = 1) Table 4-4 Comparison of hG0 and AS° for different solvent extraction systems Solvent AG°, kJ/mol AS°, kJ/K mol Reference Primene JMT -119.9 0.31 Caravaca, et al., 1996 a (Primary amine, RNH2)in xylene Primene 81R -52.6 0.03 Caravaca, et al., 1996 a (Primary amine, RNH2)in xylene Tridecylamine -58.9 -0.04 Caravaca:et al., 1996 b (Primary amine, RNH2)in xylene Dibutyl Butyl Phosphonate -28.1 120.5* Miller, Ct al, 1987 (DBBP) LIX 7950 -- -0.66 This research 50g/L 1-dodecanol in n-dodecane * Probably a typo error; a value of -120.5 J/mol seems more reasonable. y = 1O.OOx- 34.84 R2 = 0.99 3.2 3.3 3.4 54 4.5 Effect of CN/Cu ratio 4.5.1 Extraction results Three copper cyanide solutions that have the same copper concentration ([Cu] 3.93 x i0 mol/L) but with different molar ratios of cyanide to copper (CN/Cu = 3, 5 and 10, respectively) have been used to examine the potential effect of CN/Cu ratio on extraction of copper and cyanide with LIX 7950 (10% LIX 7950 and 50 gIL 1-dodecanol in n-dodecane). The variations of copper and cyanide extraction against equilibrium pH under different CNICu ratios are shown in Figure 4-6. For all the cases, copper extraction decreases with an increase of equilibrium pH (varied from 9 to 13). This is as expected since a low pH favors the protonation of guanidine (from guanidine to guanidinium ion in acid form) and thus facilitate the extraction of copper cyanide complexes. Copper extraction was also influenced by the CNICu ratio — a higher cyanide level tends to depress copper extraction. The results for the extraction of total cyanide by the extractant are shown in Figure 4-7. Similar to copper extraction, the extraction of total cyanide also decreases with an increase of equilibrium pH and CNICu ratio. Comparing the extraction of total cyanide with that of copper, it was found that the molar ratios of loaded cyanide to loaded copper in organic phase were all close to three even when the initial CN/Cu ratio in the aqueous solution was as high as 10 (Figure 4-8). The analysis of stripping solutions of the loaded organic samples further confirmed the results. When the initial CN/Cu is three, the CN/Cu ratios in the aqueous phase show little change after extraction. When the initial CNICu ratios was higher than three, the CN/Cu ratio in the aqueous phase showed an increase when equilibrium was achieved. The change of CN/Cu ratio in the aqueous solution was depending on the equilibrium pH — the lower the equilibrium pH, the more significant change the CN/Cu ratio in the aqueous phase exhibited. However, the variation of equilibrium pH shows little effect on the CNICu ratios in the organic phase which were always close to 3. These results indicate that the extracted copper and cyanide may potentially occur as Cu(CN)32. 55 0 0 1-’ x z C) 100 80 C .9 60 0 0 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 4-6 The effect of CN/Cu ratios on copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25 °C) 100 80 60 40 20 0 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 4-7 The effect of CN/Cu ratio on cyanide extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i03 mol/L, CN/Cu 5; A/O = 1; 25 °C) 56 43 X 2’ 0 2 oCN/Cu=3 z C.) tCN/Cu=5 1 x CN/Cu=1O 0 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 4-8 The calculated CN/Cu ratios in the organic phase under different pH (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; A/O = 1; 25°C) 4.5.2 FTIR analysis The infrared spectroscopic analysis on both the blank extractant solvent and the solvent with loaded copper and cyanide was conducted. Copper and cyanide were extracted from a copper cyanide solution containing 6.0 g/L copper (CN/Cu = 5) with the extractant solvent comprising of 50% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane at equilibrium pH 10. The high concentration was chosen in order to increase the intensity of the CN stretching absorption. On equilibration, the organic phase contained about 3.2 g/L copper. Comparing the adsorption peaks of two FTIR spectra, the only difference is that the spectrum for the solvent with loaded copper and cyanide exhibits a small but marked peak at about 2093 cm1, which agrees well with the reported CN stretching frequency for Cu(CN)32 anions (Flynn and McGill, 1995). The result is in accordance with those IR investigations on the modified amine extractants with loaded gold cyanides which indicated marked adsorption peak around 2140 cm1 (the CN stretching frequency of Au(CN)2)(Mooiman and Miller, 1986). Caravaca (1994) examined the infrared spectrum of the gold loaded solvent mixture of Primene 81R (primary amine) and iso-decanol 57 and found that there was not any appreciable change in the CN stretching frequency for Au(CN)2. The infrared spectrum of the solvent mixture of Amberlite LA2 (secondary amine, Fluka) with loaded gold cyanide complex also indicated that there was not any appreciable change in the CN stretching frequency for Au(CN)j (Alguacil and Alonso, 2005). The examination of the gold coordination environment of Au(CN)2 in the loaded quaternary amine extractant by means of EXAFS (Extended X-ray Absorption Fine Structure) techniques indicated gold coordination did not change before and after extraction (Ma, et a!., 2000). All these evidences support the belief that there is no specific interaction of the amine extractant solvent with the loaded metal cyanide complex. Solvent extraction of copper cyanides by LIX 7950 solvent thus complies with the ion-exchange mechanism (shown in Figure 4-10). Moreover, the FTIR analysis results confirmed that the loaded copper and cyanide in the organic phase mainly occur as Cu(CN)32. 4.5.3 Preferential extraction According to the results obtained above, the preferential extraction of specific anions (i. e., Cu(CN)32)by the extractant can be concluded which means the extractant exhibits different affinity for the anions in the aqueous phase. The speciation of Cu(I) in cyanide solution has been discussed in Section 1.2. In the aqueous phase, the equilibrium speciation of Cu(I) in cyanide solution can be represented by the following reactions (4-11) to (4- 16). [CuCN ]Cu + CN = CuCN K1 = [Cu + ][CN ] (4-11) [Cu (CN)2] CuCN + CN = Cu(CN)2 K2 = [CuCN ][ CN -j (4-12) [Cu (CN)32 ICu(CN)2 + CN = Cu(CN)32 K3 = [Cu(CN)2J[CN] (4-13) [Cu (CN )43] Cu(CN)32 + CN = Cu(CN)43 K4 = [Cu (CN)32 ][CN -] (4-14) CuCN = Cu + CN K5 = [Cu] [C1’T] (4-15) Cu20 + H20 = 2Cu + 20W K6 = [CuJ [Off] (4-16) 58 where K1 to K4 are the equilibrium constants and and K6 are solubility product constants (here the value of 1020 and 1029.5 have been chosen for them, respectively, Lu, et al., 2002). Cyanide occurs as free cyanide (HCN and CN) and complexed cyanide in the solution and their concentrations depend on temperature, pH and total copper and cyanide content, etc. The speciation of copper cyanide complexes was calculated through a spread sheet program based on the following mass and charge balances: [Cu] = [Cu] + [CuCN] + [Cu(CN)2]+ [Cu(CN)321+ [Cu(CN)431 (4-17) [CN] = [CN] + [HCN] + 2[Cu(CN)j] + 3[Cu(CN) + 4[Cu(CN)] (4-18) [Na] + [Cu’] + [Hf] = [CN] + [Cu(CN)j] + 2[Cu(CN)3]+ 3[Cu(CN)4]+ [OH] (4-19) The calculated copper speciation diagrams under different pH and CN/Cu ratios are shown in Figures 4-9 and 4-10. The formation of Cu20, CuCN and Cu(CN)2 are negligible and they are not shown in the Figures. The speciation diagrams show that low pH favors the formation of the complexes with low coordination numbers (such as Cu(CN)i). The concentration of Cu(CN)43 is negligible below pH 7 and increases with an increase of pH when CN/Cu is constant. A high CN/Cu ratio favors the formation of Cu(CN)32 and Cu(CN)43.The formation of Cu(CN)j is negligible when the CN/Cu molar ratio is higher than three (Lu, et al., 2002). At pH 11, about 98% of copper occurs as Cu(CN)32 when CN/Cu ratio is three. When CN/Cu ratio increases up to 10, about 47% of copper occurs as Cu(CN)32with 53 % as Cu(CN)43. 59 00 E 0 U I 0 E Figure 4-10 Plot of the calculated fraction of copper cyanide complexes vs CN/Cu ratio ([Cu] = 3.93 x i0 mol/L; pH =11; 25° C) 60 1.0 0.8 0.6 0.4 0.2 0.0 14 Figure 4-9 Plot of the calculated fraction of copper cyanide complexes vs pH ([Cu] = 3.93 x i0 mol/L, CN/Cu = 5; 25° C) 4 6 8 10 12 pH 1.0 0.8 0.6 0.4 0.2 0.0 Cu(CN) 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 CN/Cu molar ratio According to the composition of the copper cyanide solutions used for the extraction tests described in Section 4.5.1, the anions in the aqueous solution include OW, CN, Cu(CN)32and Cu(CN)43.Generally, the equilibrium reactions for the extraction of the copper cyanide solutions with the LIX 7950 solvent can be expressed as follows: RGorg + H20 + CN = RG HCN org + OW (420) 2RGorg + 2H0 + Cu(CN)32=2(RGH)Cu(CN)3org + 20W (45) 3RGorg + 3H20 + Cu(CN)43 = 3(RGH)Cu(CN)4org + 30W (4-21) where 2(RGH)Cu(CN) org and 3(RGH)Cu(CN) org represent of the compounds of the extractant and the extracted Cu(CN)32 and Cu(CN)43,respectively. Theoretically, the loaded copper is the summation of the loaded Cu(CN)32 and Cu(CN)43.The total loaded cyanide is the summation of the loaded free cyanide (CN) and the complexed cyanide in Cu(CN)32 and Cu(CN)43.However, the fact that the CN/Cu ratios in the organic phase were always about 3 and the loaded copper and cyanide mainly occurred as Cu(CN)32 indicates the extraction of CN and Cu(CN)43 by the extractant are negligible under the experimental conditions. The preferential extraction of Cu(CN)32 over CN and Cu(CN)43 by the extractant was thus confirmed. Due to the preferential extraction, only Cu(CN)32 was extracted during extraction, resulting in an increase of the CN/Cu ratio in the aqueous phase. Some of Cu(CN)43 ions would dissociate to Cu(CN)32 and CN ions till new equilibriums among the reactions (4-12) — (4-16) and (4-20), (4-5) and (4-21) were established. A schematic extraction process is shown in Figure 4-11. Since a high CN/Cu ratio favors the formation of Cu(CN)43 and decreases the content of Cu(CN)32 in the aqueous phase, copper extraction will decrease when the CNICu ratio increases under the same equilibrium pH. The possible reasons for the preferential extraction are believed to be attributed to the charge density effect, the hydration properties of the ions, and the geometric factors which will be discussed in detail in Chapter 6. The potential effect of ionic concentration should also be considered since though the total copper concentration in the solutions is the same, the total anion concentration increases with an increase of CN/Cu ratio. However, the investigations on the effect of S042 and S203 ions on copper extraction by the extractant have shown that the influence of the concentration factor under the experimental conditions is insignificant (See Section 4.7.2). 61 Some investigations on the effect of CN/Cu ratio on copper extraction from alkaline cyanide solution with amine extractants have been conducted. There is a lack of the information on the extraction of cyanide by amine solvents in the literature. The research results of Dreisinger, et al. (1996) indicated that a higher CN/Cu ratio depressed copper extraction with the guanidine extractant XI 7950. The investigation results of Davis, et al (1998) indicated that lower extraction of copper and iron were obtained when a high cyanide concentration was used during extraction of metals from cyanide solution with the solvent mixture of the quaternary amine and nonylphenol (LIX 7820 and LIX 7825), however, the extraction of zinc and nickel showed little change when cyanide concentration was increased. Virnig and Wolfe (1996) examined the effect of various anions on metal extraction by the guanidine extractant LIX 79 and found that the concentration of sodium cyanide has little effect on the extraction of gold. The authors also reported that varying levels of sodium cyanide did not have an appreciable effect on the extraction of the other metals tested (Ag, Zn, Hg, and Ni). However, it seems that the extraction of silver shows a small but marked decrease with an increase of NaCN level according to their experimental results. In the examination of selective solvent extraction of gold from cyanide solution by TBP (Tri-butyl phosphate) and DBBP (dibutyl butyl phosphonate), Miller, et al.(1 987) noticed different behavior between silver and the other metals (gold, zinc and nickel). They found that instead of increasing with an increase of NaCN concentration due to the effect of ionic strength, the amount of silver extraction decreased with increasing cyanide concentration. The authors believed that these results are indicative of the fact that at high cyanide concentrations, silver may add another CN ligand to form Ag(CN)32, Ag(CN)2 + CN = Ag(CN)32 (4-22) The higher charge of the Ag(CN)32 complex reduces the extent of silver extraction due to the charge density effect. Actually, according to the concentrations of cyanide and silver used in their research, there was possibility of the formation of more higher charge complex Ag(CN)43 (Flynn and McGill, 1995). Similar to the effect of CN/Cu ratios on copper cyanide speciation discussed above, the formation of the higher charged silver cyanide complexes depressed the silver extraction as expected. That cyanide concentration exhibited negligible effect on gold extraction is probably because gold coordination in cyanide solution is relatively simple: only gold and cyanide complex, Au(CN)2 occurs and its affinity with the extractant molecules are 62 much higher than that of free cyanide. As a result, the extraction of gold by the extractant (LIX 79) is not sensitive to the variation of sodium cyanide level. Organic phase (RGW)3Cu(CN4 (RGW)2Cu(CN3-RGWCN RGHOH Cu(CN)43 = Cu(CN)32- + CN + W + OH- =H20 Aqueous phase HCN Figure 4-11 The schematic extraction of copper cyanide solution with LIX 7950 (= : main reactions; = possible reactions) 4.6 Effect of phase ratio The extraction of copper and cyanide with the extractant solvent (10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane) under different phase ratios was examined. The aqueous solutions used for the extraction tests contained 500 mg/L copper and 1022 mg/L cyanide (CN/Cu = 5, initial pH was around 10). It was found that when pH was uncontrolled, the equilibrium pH were all higher than 12 when the A/O ratios varied from 4 1 to 1: 2 and there was little extraction of copper and cyanide by the extractant after equilibrium was established (Table 4-5). The extraction of copper and cyanide with the extractant under the pH-controlled conditions (pHeq = 10.50 ± 0.05, 20 °C, A/O ratios varied from 2:1 to 1:8) were thus examined and the results are shown in Figures 4-12 and 4-13. It was found that the extraction isotherms of cyanide are much different from those of copper. For the copper extraction isotherms, the copper content in 63 aqueous tends to zero with decreasing A/O ratio. However, for the cyanide extraction isotherms, the value of cyanide content in aqueous is still significant when AJO ratio is as low as 1:8. The extraction efficiency of cyanide is relatively low under the experimental conditions since the extractant has a weak affinity for CN. Even when a high volume of the extractant solvent was used (NO = 1 : 8), the analysis showed that little free cyanide has been extracted by the extractant. and most remained in the aqueous phase. A concern arising here is when using McCabe-Thiele diagram method to model the operating of the extraction process, both the extraction of copper and cyanide have to be considered. This can be realized by constructing the McCabe-Thiele diagram for copper and cyanide, respectively and then co-relating each step between two extraction isotherms. A semi-graphical method to correlate two isotherms is illustrated in Figure 4-14. The extraction isotherms of cyanide are established by plotting the loaded cyanide as a function of the copper content in the aqueous phase instead of the cyanide content in the aqueous phase. In this schematic diagram, the operating line is established at an A/O ratio of 1.2. The copper content in the aqueous and organic phases from each extraction stage can be obtained graphically. The cyanide content in the organic can be obtained in the graph once the aqueous copper content is known. The aqueous cyanide can be calculated based on mass balance (equal to the amount of total input cyanide minus the loaded cyanide in the organic phase). For example, according to Figure 4-14, two stages of extraction with the designated copper cyanide solution (CN/Cu = 5, initial copper concentration of 500 mg/L) with the extractant solvent (10% LIX 7950 and 1-dodecanol in n dodecane) are sufficient for above 95% of copper extraction efficiency (under a pH of 10.50 ± 0.05, NO =1.2, 20 °C). Accordingly, the extraction efficiency of total cyanide is only about 60% since most of the free cyanide remains in the raffinate. 64 Table 4-5 Copper extraction with LIX 7950 under pH uncontrolled conditions (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 500 mg/L, CN/Cu = 5, initial pH 10.5; 20 °C) A/O Final pH Cu extraction, % 2: 1 12.23 Nil 1: 1 11.98 Nil 1:2 11.66 <1 1:4 11.56 <1 500 400 300 ;;iiiEZZ1 c 200 - C.) 100 1/8 0• I I 0 100 200 300 400 Cu in aq, mg/L Figure 4-12 The distribution isotherms of copper extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: CN/Cu = 5, [Cu] = 500 mg/L; pHeq = 10.50 ± 0.05; 20°C) 65 500 500 400 200 100 0 E 300 2’ 0 . 200 z C.) 0 0 200 400 600 CN in aq, mg/L 400 100 800 1000 Figure 4-13 The distribution isotherms of cyanide extraction with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: CN/Cu = 5, [Cu] = 500 mg/L; pHeq = 10.50 ± 0.05; 20°C) -J 0) E 300 0 z 1) C-) 0 50 100 150 200 250 300 350 Cu in aq, mgIL Figure 4-14 The schematic McCabe-Thiele diagram for copper and cyanide extraction (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: CN/Cu = 5, [Cu] = 500 mg/L; pHeq = 10.50 ± 0.05; 20°C) 66 4.7 Effect of other anions 4.7.1 Extraction of the mixture solution The extraction of the mixture solution of metal cyanides by the LIX 7950 was examined and the results are shown in Table 3-4. Initially all metals in the aqueous solution were at the same concentration of 3.93 x i0 mol/L which is equal to the molarity of a 250 mg/L copper solution. The speciation calculation indicates that metal species in the mixture mainly occur as Zn(CN)42, Ni(CN)42,Fe(CN)64 and Cu(CN)32 (accounting for about 87% of total copper and about 13% as Cu(CN)43).The organic solution was formed with 30% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane. Plot of variations of metal extraction and pH as a function of contacting time are shown in Figure 4-15. The metal extractions and the solution pH tend to be stable after contacting two minutes, indicating the extraction kinetics for the four metals is relatively fast under the experimental conditions. This is as expected since an ion-exchange extraction usually exhibits extremely fast kinetics. The results are in accordance with those reported by the other researchers (Kordosky, et al., 1992, Sastre, et a!., 1999). Virnig and Wolfe (1996) found that the extraction equilibrium could be achieved within 60 seconds when using LIX 79 to extract metals (Au, Cu, Ag and Zn) from cyanide solutions. 67 100 13 nfl - 80 - . A - 12 0 4-’ C.) .1-’ 0. x . 40 20 ____ 11 9 0 0 0 10 0 100 200 300 400 500 600 Contact time, sec —o—Cu —ci--Zn —a—Ni —---Fe —*—pH Figure 4-15 Plot of variation of metal extraction and solution pH vs contacting time (Org: 30 % v/v LIX 7950 and 100 gIL 1-dodecanol in n-dodecane, initial aqueous solution as in Table 3-4; A/O 1; 25°C) Solvent extraction of the mixture solution with LIX 7950 under different equilibrium pH was examined and the results are shown in Figure 4-16. It was found that the extractant exhibited a selectivity order for the four metals as below: Zn > Ni > Cu > Fe. The separation factors for the individual metals over copper have been calculated according to the equation (4-2): SCUIM (4-2) where DM and Dc are the distribution coefficients of metal M and copper, respectively. The separation factors for Zn, Ni, and Fe over Cu under different equilibrium pH are summarized in Table 4-6. The apparent selectivity order follows the series as: SCu/Fe > Scu/Ni> 68 The separation factors are dependent on the individual metals. SCu/Fe is always bigger than one and5Cu/Ni and 5Cu/Zn are always smaller than one, indicating the extractant has a stronger affinity for zinc and nickel cyanide complexes than for copper cyanide complexes, but has a weaker affinity for iron(II) cyanide complex. The separation factors for the individual metals are also dependent of the equilibrium pH. At a lower pH (such as a pH of 10), the extractant will exhibit a higher extraction capability, part of the metal complexes which have weaker affinity with the extractant (such as Fe(CN)64)can also be extracted to the organic phase. At an elevated equilibrium pH (such as pH 11), the extraction capability of the extractant decreases, the weak- associated complexes will be repelled to the aqueous phase while the complexes with a higher affinity for the extractant solution still remain in the organic phase (such as Zn(CN)42 and Ni(CN)42). When the equilibrium pH is further increased, most of the extracted cyano complexes will be repelled to the aqueous phase. These results indicate that when a solvent extraction circuit with LIX 7950 as the extractant is used to treat copper cyanide solution containing impurity metals such as zinc and iron, zinc will be first extracted in the extraction step and the last to be stripped off in the stripping step. On the contrary, iron will be the least extracted and the easiest to be stripped. The analysis of total cyanide extraction by the extractant shows that the extracted cyanide mainly occurred as complexes (assumed that extracted species occur as Zn(CN)42,Ni(CN)42,Cu(CN)32,and Fe(CN)64).Most of the free cyanide remains in the aqueous phase. The potential reasons for the selectivity order for the metal cyano complexes will be discussed in detail in Chapter 6. 69 100 80 60 0 4-I jx 4° 20 0 I I 9.5 10 10.5 11 11.5 12 12.5 13 ____ Equilibrium pH Hi—Cu —a--Zn —h—-Ni —Q—Fe _*—CN1 Figure 4-16 The extraction of metals and cyanide with LIX 7950 under different pH (Org: 30% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane, initial aqueous solution as in Table 3-4; AJO = 1; 25 °C) Table 4-6 The separation factors for Zn, Ni, and Fe over Cu under different equilibrium pH (Org: 30% v/v LIX 7950 and 100 g/L 1-dodecanol in n-dodecane, initial aqueous solution as in Table 3-4; A/O = 1; 25 °C) PHeq ± 0.05 SCu/Zn SCu/Ni SCu/Fe 10.00 0.13 0.17 144.63 11.00 0.02 0.08 224.60 12.00 0.02 0.12 32.29 70 4.7.2 Effect of S042 The tests for the effect of S042 on extraction of copper and cyanide were carried out with the mixture solution of copper cyanide and Na2SO4 ([Cu] = 3.93 x 10 mol/L, CN/Cu =3, varied content of Na2SO4). The extractant solvent was composed of 10% v/v LIX 7950 and 50 g/L 1- dodecanol in n-dodecane. The extraction tests were conducted at 25 °C and phase ratio of unity. Solution pH was uncontrolled during extraction and the equilibrium pH was measured. The extraction results are shown in Table 4-7. Both the copper distribution coefficient and the equilibrium pH exhibit insignificant change when the concentration of S042 increases from zero to 0.01 mol/L. The analysis shows that the extraction of S042 by the extractant is negligible when the initial concentration of S042 is below 0.01 mol/L, indicating the extractant has a weak affinity for S042. The extraction of S042 increases slightly when the initial concentration of S042 is further increased. Accordingly, the copper extraction shows a slight decrease. It is believed that the co-extraction of S042 by the extractant may potentially decrease the availability of the extractant for the extraction of copper cyanide complex. According to the general ion- exchange mechanism, two molecules of the extractant will be used in order to extract one molecule of S042. 2 RG org + 2H0 + S042 2 (RGH) SO4 org + 20ff (4-23) Based on this stoichiometric reaction, when the initial concentration of 5Q42 is 0.1 mol/L, the loaded S042 (0.57 x103mol/L) may consume about 1% of the extractant. The test results also indicate that the ionic strength has negligible effect on extraction of copper cyanides by the extractant under the experimental conditions (Supposed ideal conditions and neglecting the effect of other species, the ionic strength of the aqueous solution is 0.03 when the concentration of Na2SO4is 0.01 mol/L according to the equation, I = c,z?). Thus, it can be deduced that the effect of ionic strength is insignificant during extraction of copper and cyanide by the extractant solvent under varied CN/Cu ratios (Section 4.5.1). Thus, the decrease of copper extraction with an increase of CN/Cu ratio is mainly due to the formation of Cu(CN)43 at a high CN/Cu ratio rather than the effect of the increasing ionic strength caused by the increased cyanide concentration. Caravaca et al. (1996) (A) and Caravaca, et al., (1996) (B) used different concentration of NaCl and Na2SO4solutions (0.01 mol/L to 0.1 mol/L) to examine the effect of 71 ionic strength on the extraction of aurocyanide by three primary amine extractants and found that there was not any appreciable effect on the extraction of gold due to the presence of these salts in the solution. Sastre, et a!. (1999) reported that the pH50 value remained almost constant when the concentration of NaC1 increased from 0.05 to 1 mol/L during extraction of aurocyanide with the guanidine extractant LIX 79 and thus concluded that the effect of the aqueous ionic strength on gold extraction with LIX 79 was insignificant. Table 4-7 The effect of S042 on copper extraction with LIX 7950 (Org: 10% v/v 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu =3,A/O=1;20°C) Initial S042’, mol/L pHeq Log Loaded S042, x103 mol/L 0 10.99 1.13 ND 0.001 10.99 1.13 ND 0.005 10.98 1.13 ND 0.01 10.98 1.13 ND 0.05 10.95 1.12 0.16 0.1 10.93 1.11 0.57 0.2 10.93 1.11 0.69 ND : below the detection limit (< 0.1 mg/L) The effect of ionic strength can be explained qualitatively in terms of salting-out and salting-in effects (Rydberg, et al., 2004). The process of extraction of metal cyanide complexes by the LIX 7950 can be expressed by the following sequential reactions: RGorg + H20 = RGHOH org (4-24) RGHOH org RGH + OH (4-25) RGH + Cu(CN)32 = 2(RGH) Cu(CN)32 (4-26) 2(RGH) Cu(CN)32= 2(RGW) Cu(CN)32org (427) To extract the complex anion from the aqueous phase, the extractant is first protonized and then extracts the target anion by complexation (Figure 4-11). The presence of Na2SO4in the aqueous phase may exhibit two effects: (1) it ties up H20 molecules by forming hydrated ions so that less 72 free water is available for protonation of the extractant molecules and (2) it breaks down the hydrogen bond structure of the water for the extractant molecule to dissolve in the aqueous phase. If the former effect dominates during extraction, it is described as a salting-out effect and the reverse as salting-in effect (Rydberg. et al., 2004). The presence of Na2SO4may also exhibit similar effects on the hydration of Cu(CN)32 and especially OH ions since they are well hydrated in water. It seems that the overall effect of S042 on the extraction process is negligible under the experimental conditions - the variation of the aqueous ionic strength in the low range (<0.01 mol/L ofNa2SO4)has an insignificant effect on copper extraction with LIX 7950. 4.7.3 Effect of SCN, CNJ and S203 4.7.3.1 Extraction of CI’T, CNO, SCI’T andS203 The extraction isotherms for CN, CNO, SCN and S203 with LIX 7950 are presented in Figure 4-17. The extraction tests were carried out with their sodium salts. The organic solvent was formed with 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane. The equilibrium pH was controlled at 10.50 ± 0.05. The extraction tests were conducted at a phase ratio of unity and 25 °C. The speciation calculation indicated the copper cyanide complexes are stable under the experimental conditions. It was found that the extraction of S203 ions by the extractant was negligible under the experimental conditions (the results are not presented in Figure 4-18), indicating the extractant has the weakest affinity for S203 ions among the four anions. Of the other three anions, the extractant exhibits relative weak affinity for CN and CNO, but has a stronger affinity for SCN ions. The sequence of affinity with the extractant for the four anions is as followed: SCN > CNO> CN>>S203. It should be pointed out that in the simple alkaline cyanide solution, free cyanides are present in the form of HCN and CN. The equilibrium reaction of the two species can be described as: HCN = H + CN pKa 9.27 (1-3) 73 HCN will not be extracted by the extractant when the cyanide solution is contacted with the extractant. The general equilibrium reaction of cyanide extraction with LIX 7950 can be described by reaction (4-14): RGorg + H20 + CN = RG HCNorg + OH (4-20) As a result, the extraction of free cyanide is controlled by both equilibrium reactions (1-3) and (4-14). Since there are no dissociation reactions of CNO, SCN, 52032 under the experimental conditions, the general equilibrium extraction for the three anions can be expressed as: RGorg + H20 + CNO = RG HCNOorg + Off (4-28) RGorg + H20 + SCN = RG HSCN0rg ÷ OH (4-29) 2RGorg + 2H0 + S203 = 2(RG H)S203org + 20Ff (4-30) where RGHCNO0rg, RGHCNSorg and 2RG H S203org represent the compound of the extractant and the extracted CNO, SCN, andO3respectively. 0.008 0.006 o 0.004 0 C 0.002 0 0 0.01 0.02 0.03 0.04 0.05 0.06 Figure 4-17 The extraction isotherms for CN, SCN, and CNO with LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, PHeq = 10.50 ± 0.05; 25 °C) The information on the extraction of simple anions by guanidine extractants are lacking. However, quaternary amine extractants are reported to exhibit a much stronger affinity for SCN than CN (Gonzalez, 1994). The equilibrium constants for extraction of a series of anions by the Anion in aq, mol!L 74 quaternary amines such as Aliquat 336 show the order of C104 > SCN> F >NO >Br > HS04 >S042 OH (du Preez, 2000). As a result, the stripping efficiency of gold cyanide with thiocyanate solution was much higher than with cyanide solution (Jiang, et al., 2003). The affinity sequence can be explained in terms of the degree of hydration of the anions and charge and size effect which will be discussed in detail in Chapter 6. 4.7.3.2 Effect of CNO SCK andS203on copper extraction A mixed solution of copper cyanide and sodium thiocyanate (NaSCN) was used to examine the effect of SCN on copper extraction with LIX 7950. The solutions contained copper and cyanide ([Cu] = 3.93 x i0 mo/L, CN/Cu = 3) and different concentration of SCN and were contacted with the organic solvent (10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane). The equilibrium pH was controlled at 10.50 ± 0.05. The tests were conducted at phase ratio of unity and 25 °C. Similar method was used to examine the effect of cyanate (CNO) and thiosulfate (S2032) on copper extraction by the extractant. The extraction results are shown in Figure 4-18. It was found that the presence of a small amount ofS203 (<0.05 mol/L) in the solution exhibits negligible effect on copper extraction. When its concentration is further increased, the copper extraction exhibits a slight decrease. However, copper extraction decreased significantly with increasing SCN concentration. When the concentration of SCN was increased from zero to 0.01 mol/L, copper extraction decreased from 96% down to 78%. The presence of CNO also shows a marked depression effect on copper extraction. The significance of the depression effect is as follows: SCN > CNO>> S2032. The analysis shows that at the same initial concentration in copper cyanide solutions, the loaded SCN in the organic phase is much higher than that ofS203 and CNO (Table 4-8). It is believed that the extraction of the non-copper anions may substantially compete the available extractant with copper cyanide complexes during extraction. The more the non-copper anions are extracted, the lower the copper loading. Since the extractant has the strongest affinity for SCN ions and the least affinity for S2O ions, the present of SCN exhibited significant negative effect on copper extraction and the effect S203 is insignificant. The calculation shows that when initial concentration of SCN was 0.05 mol/L, the loaded SCN in the organic would consume more 75 than 90% of the extractant according to reaction (4-21), resulting in a low extraction of copper (Figure 4-18). That the presence of other anions in cyanide solution may potentially influence the extraction of metal cyanide complexes has been noticed by some researchers. Virnig and Wolfe (1996) examined the effects of varying levels of NaC1, NaSCN, NaCN, and Na2CO3on the extraction of metals from cyanide solution by guanidine extractant LIX 79. They found that varying levels of sodium carbonate did not have an appreciable effect on extraction of Au, Ag, Zn, Hg, and Ni. Higher levels of sodium chloride resulted in marked reduction in extraction of silver, zinc, and nickel cyanide complexes, but had negligible effect on the extraction of gold. However, increasing the level of sodium thiocyanate resulted in a significant decrease in extraction of all the metals mentioned above. The investigation of Davis, et al. (1998) indicated that the extraction efficiency of Cu and Fe decreased significantly with an increase of the concentration of SCN in the aqueous phase when the solvent mixture of a quaternary amine and nonylphenol was used to extract metals from cyanide solutions. However, high levels of SCN exhibited little effect on extraction of zinc and nickel cyanide species. These results can be attributed to the fact that the presence of these anions may compete the available extractant with the metal cyanide complexes. Depending on their affinity with the extractant, these anions may exhibit different degree of depression on extraction of the metal cyanide species. For example, of the non-metal anions tested, the effect of SCN on metal extraction is substantial due to its strong affinity with amine extractants. Comparatively, the presence of S042 or S203 ions has an insignificant effect due to their weak affinity with the extractant. 76 100 0 4.’ 0 4-. x a, C) Table 4-8 The loaded anion content under different initial concentrations (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 3; PHeq = 10.50 ± 0.05; A/O = 1; 25 °C) Initial anion Loaded anion, x103 mol/L concentration, SCN CNO S203 mol/L 0.05 11.37 1.03 0.19 0.1 11.79 2.96 0.19 0.2 12.03 3.87 0.28 0.25 12.21 4.22 0.42 80 60 40 20 0 0 0.1 0.2 0.3 0.4 0.5 Salt concentration, mol/L 0.6 Figure 4-18 The effect of SCN, CNO and S203 ions on copper extraction by LIX 7950 (Org: 10% v/v LIX 7950 and 50 g/ L 1-dodecanol in n-dodecane, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu =3; pHeq = 10.50 ± 0.05; NO = 1; 25 °C) 77 5 Extraction by LIX 7820 5.1 Equilibrium time A preliminary test was conducted to evaluate the effect of contact time on copper extraction from the copper cyanide solution by LIX 7820. The organic solvent was composed of 2% v/v LIX 7820 in n-octane. The copper content in the aqueous solution was 3.93 x i0 mol/L and the CN/Cu ratio was five. The extraction tests were conducted at 20 °C and phase ratio of unity. The initial solution pH is about 10. The extraction results are shown in Figure 5-1. Similar to the case of LIX 7950, under the established experimental conditions, copper extraction increases significantly in the first minute of mixing. After 2 minutes of contact, further prolonging agitation could not further increase copper extraction. The solution pH was also constant after 2 minutes of mixing, indicating the copper extraction kinetics with LIX 7820 was relatively fast and the equilibrium between the two phases could be established rapidly. The influence of the contacting time on extraction of other metal cyanide solutions (Zn, Ni, Fe cyanide solutions) with the extraction system also exhibited similar behaviors. Subsequently, a contact time of 10 minutes was chosen for extraction tests when the solution pH was uncontrolled. During pH controlled experiments, an additional 5 minutes of mixing was applied when the solution pH reached the setpoint in order to achieve equilibrium. 78 100 12 00 Contact time, mm Cu extraction —A-— pH Figure 5.-i Plot of variations of copper extraction and solution pH vs contact time (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; AJO = 1; 20°C) 5.2 Organic formula 5.2.1 Effect of the molar ratio of nonyiphenol to Aliquat 336 The effect of the molar ratio of nonyiphenol to the quaternary amine Aliquat 336 (NP/A) on copper extraction was examined and the results are shown in Figure 5-2. The tests were carried out with aqueous solutions of 3.93 x i0 mol/L copper (CN/Cu = 5) and the organic solutions formed by 9.4 x i03 mol/L of Aliquat 336 and different amount of 4-nonylphenol in n-octane. The molar ratio of 4-nonyiphenol to Aliquat 336 varied from zero to six. The phase ratio of unity was used and the tests are conducted at 20 °C. When there is no addition of 4-nonylphenol, the extractant Aliquat 336 exhibits a strong extraction ability on copper. Nearly all copper could be extracted into the organic phase even when the equilibrium pH was as high as i3. The quaternary amine’s strong affinity for the large metal cyanide complexes has been noticed by the 79 researchers. Moore (1975) and Moore and Groenier (1976) found that zinc and cadmium cyanide complexes can be extracted efficiently by quaternary amines (Aliquat 336 and Adogen 464) from waste cyanide solutions arising from the electroplating industry. The strong base ion-exchange resins with the functional group of quaternary amine were also developed to extract metals from cyanide solutions (Riveros, 1993). However, the big challenge for using quaternary amines to recover metal cyanides is the stripping of the loaded metal cyanide complexes. Since the extraction of copper cyanide complexes is still very high at a high pH, it will be difficult to remove the loaded copper by simply varying solution pH. Moore (1975) and Moore and Groenier (1976) reported that sodium hypochlorite (NaC1O) and formaldehyde (HCNO) can strip the loaded zinc and cadmium cyanide complexes efficiently. Alternatively, the strong NaOH solution (as high as 12 mol/L) has to been used in order to strip the loaded cadmium from the solvent loaded with 0.025 mol/L Aliquat 336. In the case of using strong base resins to recover gold cyanide, zinc cyanide solution was suggested to be used to “crowd off’ the loaded gold cyanide complex. Sulfuric acid is then applied to destroy the zinc cyanides to regenerate the resins back to their original form (Fleming, 1998). However, the presence of nonylphenol in the solvent mixture can render the extraction capability of Aliquat 336 pH-dependent. According to Figure 5-2, when a NP/A ratio of 2 is used (with a concentration of 9.4x10 mol/L Aliquat 336), the copper extraction with the solvent mixture gradually decreases with an increase of equilibrium pH. When the equilibrium pH varies from 9.5 to 12.5, copper extraction decreases from 80% down to less than 10%. Mattison and Virnig (2001) attributed this “binding effect” to the interaction between 4-nonylphenol and the quaternary ammonium cation. At a low equilibrium pH (i. e., pH 11 or below), 4-nonylphenol is protonated and the quaternary ammonium compound extracts an anion from the aqueous phase. But under highly alkaline conditions, nonylphenol become anionic and will form an ion-pair with the quaternary ammonium cation. Consequently, the extracted anion will be expelled to the aqueous phase. The possible reactions in the extraction of an anion in the aqueous phase by the solvent mixture of quaternary amine and nonylphenol are listed as follows: HPorg = HPaq (5-1) FIPorg = H + Porg (5-2) HPaq = W + Faq (53) 80 R4NOHorg = R4Norg + OH (54) R4NOHorg + HPorg = R4NPorg + H20 (5-5) fl RiNOHorg + X = (R4N) Xorg + fl OH (5-6) fl R4NPorg + X’ + fl H20 (R4N) Xorg + fl HPorg + fl OH (5-7) where HP represents 4-nonyiphenol andR4NOH represents the hydroxide form of the quaternary amine (Aliquat 336); X11 represents the target anion with n negative charge. Since the solubility of nonyiphenol (HP), nonylphenol oxide (F) and Aliquat 336 in the water is small, the reactions (5-1) to (5-4) are negligible under the experimental conditions. The extraction of the target anion from the aqueous phase is actually a process of competing for the available quaternary amine with nonylphenol. The general extraction equilibrium can then be described as reaction (5-7) (Mattision and Virnig, 2001). Thus, the extraction capability of the solvent mixture decreases with an increase of NP/A ratio. At a NP/A of 6, the solvent mixture exhibits such a weak affinity for copper cyanides that less than 10% copper could be extracted even at the equilibrium pH of 9.5. Similar effect has been reported by Davis, et al. (1998) who found that the extraction capacity of copper with LIX 7825 (NP/A = 2.5) was much lower than that with LIX 7820 (NP/A = 2). It was also observed that the extraction of gold, zinc and nickel from their cyanide solution with the solvent mixture also decreased with an increase of NP/A ratio (Mattision and Vimig, 2001; Davis, et al. 1998). Comparatively, it was found that copper extraction with Aliquat 336 exhibited negligible change when 1-octanol was used instead of nonyiphenol. At the same equilibrium pH, there is little variation of copper extraction when the content of octanol in the solvent was increased from 30% v/v to full strength (Table 5-1). The addition of 1-octanol to the LIX 7820 solvent mixture also did not have an appreciable effect on copper extraction (Figure 5-3). This evidence indicates that simple hydrogen bonding is not sufficient to produce the “binding effect” on the quaternary ammonium cation. Mattision and Vinig (2001) believed that there is strong role for solvation of the ion-pair of the quaternary ammonium cation and phenoxide, with phenolic concentrations above stoichiometric serving to stabilize the ionized phenoxide and that the solvating species must be acidic enough to share a proton with the phenoxide anion. It was found that the extractant LIX 7950 exhibited almost no extraction of copper complexes when a small amount of 4-nonyiphenol (1% v/v) was added even at a equilibrium pH of 9.5, indicating that the 81 interaction between LIX 7950 and nonyiphenol is so strong that the copper cyanide complexes do not have any chance to form an ion pair with the extractant molecule. Since the extraction and stripping behaviors of the metal cyanide complexes with the solvent mixture of Aliquat 336 and nonyiphenol can be fine-tuned by varying the ratio of nonyiphenol to quaternary amine, the LIX 7800 series solvent products have thus been developed and suggested for the recovery of metals from cyanide solutions by Cognis. In this research, the reagent LIX 7820 (containing 0.47 mol/L Aliquat 336, NP/A = 2) has been used in the following tests. 1OOU 0 -o 80 C o 60 0 Cu 4-’ x C.) 20 9.5 10 10.5 11 11.5 12 12.5 13 Equilrium pH Figure 5-2 The effect of the molar ratio of nonylphenol to Aliquat 336 on copper extraction (Org: 9.4 x i0 mol/L Aliquat 336 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; AJO = 1;20°C) 82 Table 5-1 The effect of the 1-octanol concentration on copper extraction with Aliquat 336 (Org: 9.4 x i0 mol/LAliquat 336 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; AJO = 1;20°C) V Copper extraction, % pH 30% v/v octanol full strength of 1octanol* 9.5 100 100 10.5 99.9 99.9 11 99.5 99.3 11.5 99.2 99.5 12 98.8 98.7 12.5 98.7 98.7 *the designated volume of extractant is dissolved in 1-octanol without any addition of n-octane. 100 . no octanol 30 % v/v octanol 80 . C 0 ‘I-I U x 0 C.) 20 A. 0 I I 9.5 10 10.5 11 11.5 12 12.5 13 Equilrium pH Figure 5-3 The effect of the 1-octanol concentration on copper extraction with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x 10 mol/L, CN/Cu = 5; AJO = 1; 20°C) 83 5.2.2 Effect of diluents The effect of diluents on copper extraction with LIX 7820 under different equilibrium pH was examined. Toluene, n-dodecane, decane, n-octane and n-heptane have been used as diluent, respectively, and the extraction results are shown in Figure 5-4. No remarkable difference on copper extraction was observed while changing the diluents. This is as expected since the copper extraction with LIX 7820 solvent mixture has little change even when 30% v/v octanol was mixed with n-octane (Figure 5-3). This is different from the case for guanidine extractant LIX 7950 where the diluent types have a marked effect on copper extraction. Miller and Mooiman (1984) reported that the effect of diluent type on the pH5o values for amine extraction of gold was insignificant in the case of primary amines, however, the aliphatic diluent (hexane) may significantly reduced the pH50 values for secondary and tertiary amines compared to the aromatic diluent (xylene). However, the investigation on the extraction of gold cyanide by primary amine (tridecylamine) conducted by Caravaca, et al.(1 996A) indicates that there was marked shift on pHso values when changing diluent type (i. e., from toluene to decane). It seems to be difficult to draw a general conclusion on such behaviors. The insignificant effect of diluent types for copper extraction with LIX 7820 is probably because the interaction between the quaternary amine and nonyiphenol plays such a strong role in extraction that the effect of diluents is negligible. 84 100 a - - 80 - o —--decane 4 heptane 40 n-dodecane z C.) + octane \\ toluene 20 +4 C 0 I 9.5 10 10.5 11 11.5 12 12.5 13 Equilibrium pH Figure 5-4 The effect of diluent types on copper extraction with LIX 7820 (Org: 20% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 4; A/O = 1; 20°C) 5.2.3 Effect of the extractant concentration The effect of the concentration of LIX 7820 on copper extraction under different equilibrium pH was examined and the results are shown in Figure 5-5. The concentration of LIX 7820 in n octane varied from 2% v/v to 10% v/v. As expected, when the extractant concentration was increased, an enhanced copper extraction was observed. For the same aqueous solution, there is a significant shift on pH5o values when the concentration of LIX 7820 in the organic phase is increased (from 11 for 2% v/v LIX 7820 to about 12.5 for 10 % v/v LIX 7820). At the same equilibrium pH, higher copper extraction was obtained when a higher concentration of the extractant was used. However, the difference tends to decrease at high alkaline conditions due to the formation of phenoxide at high pH. The effect of initial copper concentration on copper extraction with LIX 7820 was also examined and the results are shown in Figure 5-6. Under the 85 0 C.) x a) C.) 0 9.5 10 10.5 11 11.5 12 12.5 13 Equilibrium pH Figure 5-5 The effect of the extractant concentration on copper extraction with LIX 7820 (Org: LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 5; AJO = 1; 20°C) same equilibrium pH, the copper extraction from the solutions with a lower copper concentration is higher than that from those with higher copper content. When equilibrium pH is higher than 11.5, copper extraction in all cases tends to decrease significantly. This indicates that the quaternary ammonium cations tend to form ion pairs with phenoxide ions at elevated pH and consequently the extraction capability of the solvent mixture decreases significantly. It also indicates that stripping with NaOH solution is feasible even in case of the high concentration of extractant or concentrated copper cyanide solutions are used. 100 80 60 40 20 86 100 80 0 4-I C.) ‘I-I x C) 20 0 9.5 10 10.5 11 11.5 12 12.5 13 Equilibrium pH Figure 5-6 The effect of the initial copper concentration on Cu extraction with LIX 7820 (Org: 10% v/v LIX 7820 in n-octane, aq: CN/Cu = 5; A/O = 1; 20°C) 5.3 Effect of CN/Cu ratio The tests for the effect of CN/Cu ratio on extraction of copper and cyanide were carried out with three copper cyanide solutions that have the same copper concentration ([Cu] = 3.93 x i0 mol/L) with different molar ratios of cyanide to copper. The organic solvent mixture contained 2% v/v 7820 in n-octane. The extraction tests were conducted at phase ratio of unity and 20 °C. The extraction of copper and cyanide under different equilibrium pH were determined and the results are presented in Figure 5-7 and 5-8. The extraction of copper and cyanide with LIX 7820 exhibits the same trends as those with LIX 7950. Both extractions of copper and cyanide decrease with an increase of the equilibrium pH. High cyanide levels tend to depress the extraction of copper and cyanide. It was found that for the solutions with CN/Cu ratios of five and ten, the CN/Cu ratios in the raffinate exhibited a marked increase after extraction, especially at a low equilibrium pH. The plot of calculated CN/Cu ratios in the organic phase under different 87 equilibrium pH is shown in Figure 5-9. The mass balance indicated that the CN/Cu ratios in the organic phase are all close to three. The analysis of the stripping solution of the organic samples further confirmed the results. The infrared spectroscopic analysis on both the blank solvent and the organic solvent with loaded copper and cyanide was conducted (Copper and cyanide were extracted from a copper cyanide solution containing 6000 mg/L copper and CN/Cu = 5 with 10% v/v LIX 7820 in n-octane at equilibrium pH 10. On equilibration, the organic phase contained 4100 mg/L copper). Comparing the absorption peaks of the two JR spectra, the only difference is that the spectrum for the solvent with loaded copper and cyanide exhibits a small but marked peak at about 2095 cni’, which agrees well with the reported CN stretching frequency for Cu(CN)32 anions (2093 cm1 according to Flynn and McGill, 1995). The results are in accordance with those for LIX 7950 solvent, indicating the extracted copper and cyanide mainly occurred as Cu(CN)32 and most of free cyanide remains in the aqueous phase. The preferential extraction of Cu(CN)32by the solvent mixture over Cu(CN)43 and CN is confirmed. 0 C., x 0) C.) 100 80 60 40 20 0 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 5-7 The effect of CN/Cu ratio on copper extraction with LIX 7820 (Org: 2% LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L; A/O = 1; 20°C) 88 2’ 0 z C.) 100 80 C .2 60 0 x G) 40 I-. z C-) 20 0 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 5-8 The effect of CN/Cu ratio on cyanide extraction with LIX 7820 (Org: 2% LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L; A/O = 1; 20°C) 4 3 2 I 0 DAat • CN/Cu =3 i CN/Cu5 o CN/Cw1O 9 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH Figure 5-9 Plot of calculated CN/Cu ratios in organic phase vs equilibrium pH (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L; A/O = 1; 20°C) 89 5.4 Effect of phase ratio The extraction of copper and cyanide with the solvent mixture (2% v/v LIX 7820 in n-octane) under different phase ratios was examined. The initial concentration of copper in the aqueous phase is 500 mg/L (CN/Cu = 5). The extraction isotherm of copper and cyanide with the solvent mixture was established by varying the volume ratio of the aqueous phase to the organic phase (A/O) from 4 : 1 to 1 : 2 (Table 5-2). The preliminary tests show that copper extraction with the solvent mixture was significantly low (<5%) under pH-uncontrolled conditions. Subsequently, the equilibrium pH was controlled at 10.50 ± 0.05. The results are shown in Figures 5-10 and 5-11. Similar to the results with LIX 7950, due to the preferential extraction of Cu(CN)32 over Cu(CN)43 and CN by the extractant, the extraction isotherms for cyanide is different from that of copper. The extracted cyanide mainly occurred as complexed cyanide (Cu(CN)3j.The schematic McCabe-Thiele diagram for copper and cyanide extraction based on the semi-graphical method to co-relate the two isotherms has been given in Figure 5-12. It shows that after two stages of extraction with the designated extractant solution (under a pH of 10.50 ± 0.05 and A/O = 1:2), about 90% of copper extraction can be obtained. The extraction of total cyanide is only about 50% and most of the free cyanide remains in the raffinate. Table 5-2 Copper extraction with LIX 7820 under pH uncontrolled conditions (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 500 mg/L, CN/Cu = 5, initial pH 10.5; 20°C) A/O Final pH Cu extraction, % 2: 1 12.78 Nil 1:1 12.56 <1 1:2 12.44 <1 1 : 4 12.30 1.52 90 250 2/1 ::: . 100 C.) 1/8 50 0 I I 0 100 200 300 400 500 Cu in aq, mg/L Figure 5-10 The distribution isotherms of copper extraction with LIX 7820 (Org.: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 500 mg/L, CN/Cu = 5; pH = 10.50 ± 0.05; 20°C) 300 250 I ___ 0 200 400 600 800 1000 1200 CN inaq,mg/L Figure 5-11 The distribution isotherms of cyanide extraction with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 500 mg/L, CN/Cu = 5; pH = 10.50 ± 0.05; 20°C) 91 300 250 -J 200 150 0 500 Figure 5-12 The schematic McCabe-Thiele diagram for copper and cyanide extraction (Org: 2% v/v LIX 7820 in n-octane, aq: CN/Cu = 5; pH = 10.50 ± 0.05; 20°C) 5.5 Effect of temperature The tests for temperature effect on extraction of copper and cyanide were carried out with the copper cyanide solutions ([Cu] = 3.93 x i0 mol/L, CN/Cu = 5) and the organic solutions of LIX 7820 (2% v/v in n-octane). The extraction test was conducted at a phase ratio of unity and the temperature varied from 25 °C to 45 °C. The equilibrium pH was controlled at 11.00±0.05. The extraction results are shown in Figure 5-13. It was found that the variation of temperature has a negligible effect on the extraction of copper and cyanide by the solvent mixture. Under the experimental conditions, Cu(CN)32 is the only loaded complex according to the mass balance. The loaded solvents were stripped with 1 mol/L NaOH solution three times and were re contacted with the original copper cyanide solutions. The stripped solvent exhibits similar extraction capability for copper and cyanide, indicating the main functionality of the solvent mixture has not changed. The temperature effect on copper extraction with LIX 7820 is different from that with LIX 7950. As discussed in Section 4-6, this is probably attributed to the different 0 100 200 300 400 Cu in aq, mg!L 92 properties between the two extraction systems. The guanidine extractant LIX 7950 has to be first protonated before extracting the metal cyano complexes from the solution. An elevated temperature may significantly influence this process. Comparatively, the quaternary amine Aiiquat 336 has a much higher basicity due to its “permanent protonized structure” and it seems that a small change in temperature (from 25 °C to 45 °C) has little effect on extraction capability for the copper cyanide complex. 80 60 Cu 0 0 40 CN ‘C w 20 0 I 25 30 35 40 45 Temperature, C Figure 5-13 The effect of temperature on extraction of copper and cyanide with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 xi0 mol/L, CN/Cu = 5; pHeq = 11.00 ± 0.05; A/O=1) 93 5.6 Co-extraction with other anions 5.6.1 Effect of non-metal anions Mixed solutions of copper cyanide and NaSCN, NaCNO, or Na2SO3were used to examine the effect of SCN, CNO and S203 on the extraction of copper and cyanide with LIX 7820. The aqueous solutions contained the same amount of copper and cyanide ([Cu] = 3.93 x i0 mo/L, CN/Cu = 3) and different concentrations of SCN, CNO, orS203. The organic solvent of 2% v/v LIX 7820 in n-octane was used. The equilibrium pH was controlled at 10.50 ± 0.05. The tests were conducted at 20 °C and phase ratio of unity. The extraction results are shown in Figure 5-14. Copper extraction decreases slightly with an increase in the concentration of S203, but was significantly suppressed in the presence of SCN ions. CNO ions exhibit a moderate depress on copper extraction. The loaded SCN, CNO or S203 in the organic phase under different initial concentration conditions was calculated based on mass balance and the results are shown in Table 5-3. At same initial concentration, the loaded SCN is much higher than CNO and S203. The extraction isotherms of SCN, CNO and S203 were determined with the equilibrium pH controlled at 10.50 ± 0.05 and the results are shown in Figure 5-15. Similar to LIX 7950, the solvent mixture exhibits an affinity sequence for the four anions as followed, SCN> CNO > CN>>S203. It is believed that SCN ions strongly compete for the available extractant molecules with copper cyano complexes during extraction and consequently exhibits the most significant depression on copper extraction by the solvent mixture. 94 80 0 4-. 0 x C) Figure 5-14 The effect of SCN, CNO and S203 on copper extraction with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x103 mol/L, CN/Cu = 5; pHeq = 10.50 ± 0.05; A/O = 1; 20°C) Table 5-3 The loaded anions under different initial anion concentrations (Org: 2% v/v LIX 7820 in n-octane, aq: [Cu] = 3.93 x i0 mol/L, CN/Cu = 3; pHeq = 10.50 ± 0.05; A/O = 1; 20°C) Loaded anion, x103 mol/L Initial anion, mol/L SCN CNO S203 0.05 6.23 1.45 -- 0.1 6.59 1.85 0.22 0.2 7.02 1.93 0.39 0.25 7.31 -- 0.51 60 40 20 0 0 0.1 0.2 0.3 0.4 Salt concentration, mol/L 0.5 0.6 95 0.008 -j 0 E 2’ o 0.004 0 < 0.002 0 0 0.02 0.04 0.06 0.1 Anion in aq, mol/L Figure 5-15 The extraction isotherms of SCN, CN, CNO with LIX 7820 (Org: 2% v/v LIX 7820 in n-octane, aq: sodium salt; pHeq = 10.50 ± 0.05; 20°C) 5.6.2 Extraction from a mixed solution of metal cyanides The extraction of the metals from the mixed solution (see Table 3-4) with LIX 7820 was examined. As described in Section 4.7.1, all metals in the initial aqueous solution were at the same concentration of 3.93 x i0 mol/L and the speciation calculation indicated that metal species in the mixture mainly occurred as Zn(CN)42,Ni(CN)42,Fe(CN)64 and Cu(CN)32 (accounting for about 87% of total copper and about 13% Cu(CN)43).The organic phase was formed by dissolving 5% LIX 7820 in n-octane. The tests were conducted at phase ratio of unity and 20 ° C. Plot of variations of the metal extraction and the solution pH versus contacting time are shown in Figure 5-16. Similar to LIX 7950, the extraction equilibrium can be established quickly. The metal extraction and solution pH tend to be constant after contacting 2-3 minutes, indicating the extraction kinetics for the four metals is relatively fast under the experimental conditions. The extraction of metal with the solvent mixture under different equilibrium pH was examined and the results are shown in Figure 5-17. The solvent mixture exhibited a selectivity order for the four metals as follows; 0.08 96 Zn>Ni >Cu>Fe. The mass balance calculation indicates that the extracted cyanide mainly occurred as complex (Zn(CN)42,Ni(CN)42,Cu(CN)32,and Fe(CN)64)and most of the free cyanide remained in the aqueous phase. C 0 U x a) ___________________________________ a) r Ni -a-Fe -- Figure 5-16 The effect of contact time on metal extraction and pH with LIX 7820 (Org: 5% v/v LIX 7820 in n-octane, aqueous solution as in Table 3-4; AIO =1; 20°C) 100 80 60 40 20 0 -D-O-O-——D ---— A A *-e.. • . . . . -.:x , )( ;4 r_aaa t ,% , 11 10.5 0. 9.5 0 100 200 300 400 Contact time, sec 9 500 600 97 0 V., .1-i x a, a) -.-- Cu -s--Zn Figure 5-17 The extraction of metals and cyanide with LIX 7820 at different pH (Org: 5% v/v LIX 7820 in n-octane, aqueous solution as in Table 3-4; AJO =1; 20°C) 100 80 60 40 20 0 9 9.5 10 10.5 11 11.5 12 12.5 Equilibrium pH 98 6 Discussion According to the experimental results in the former Sections, the following facts have been ascertained: a) For both extractants, the affinity sequence for four non-metal anions is as follows, SCN > CNO > CN >> 52032. Accordingly, when co-extracted within copper cyanide solution by the extractants, the significance of reduction on extraction of copper and cyanide follows the sequence, SCN> CNO >> S2032; b) Both extractants have a much higher affinity for Cu(CN)32 than for Cu(CN)43.Though both extractants exhibit a relative weak affinity for CN ions, a high level of CN in copper cyanide solution depresses copper extraction by changing the speciation of unextracted copper to Cu(CN)43; c) For both extractants, the following selectivity order for the metals is obtained during the extraction of the mixture of metal cyanides, Zn> Ni> Cu > Fe; d) Both extractants preferentially extract metal cyano complexes over free cyanide (CN). The selectivity orders for the metal cyano complexes in various solvent extraction systems have been summarized in Table 6-1. The selectivity order obtained in this research is in good agreement with other studies in the literature. It seems that an erroneous selectivity order for Cu and Zn was reported by Aguayo, et al (2007) and Valenzuela, et al (2003) since obviously zinc shows much higher extraction than copper based on their experimental results and their calculations on the distribution coefficients and the selectivity for Au over Cu and Zn. The selectivity sequence has been explained qualitatively in terms of the degree of hydration of the anions (Irving and Damodaran, 1971). Mooiaman and Miller (1984) believed that that the selectivity order of the relative metal cyano complexes with modified amines was attributed to charge and size ratio, compatibility of the hydrated anion with the organic phase, and extent of solvation of the amine/anion adduct in the organic phase. 99 Table 6-1 Summary of the selectivity orders with various extractants Author Extractant Selectivity order Mooiman and Modified amine — Primary, Au(CN)2>Ag(C Zn(CN) Miller, 1986 secondary and tertiary amine with >Ni(CN)4Cu35.Fe6 . >Fe(CN)64Tributyphosphate Miller, et al., 1987 Tributylphosphate, Au(CN)2>Ag(C>Cu(CN)42 Dibutylbutyiphosphate >Zn(CN)45NiFe63 Riveros, 1990 Quternary amine Aliquat Au>Zn>Ni>Cu, Fe 336 Figuerola. et al., Trioctylmethylammonium Zn> Ni>Co> Cu> Fe 1992 cloride Virnig and Wolfe, Guanidine LIX 79 Au>Zn>Ag>Ni>Cu >Fe 1996 Sastre, et al, 1999 Guanidine LIX 79 Au>Zn>Cu>Fe Valenzuelea, et al., Guanidine LIX 79 Au>Ag>Cu>Zn>Fe 2003 Aguayo, et al., Guanidine LIX 79 Au>Ag>Cu>Zn>Fe 2007 This research Guanidine LIX 7950 Zn>Ni>Cu>Fe This research LIX 7820 (quaternary amine Zn>Ni>Cu>Fe and nonylphenol) As suggested in Chapter 4, the extraction of Cu(CN)32by LIX 7950 can be expressed as below, 2RGH0Horg + Cu(CN)32 2RGHCu(CN)3org+ 2011 (61) where RGHOH is the protonated form of LIX 7950. Since ions in water are hydrated in varying degrees, the extraction process actually involves ligand displacement. To extract an anion from the aqueous phase to the organic phase, the extraction process may constitute the following steps, I. Breaking the anion hydration shell; 100 II. Forming the complex of the extractant and the anion (ion-pairing); III. Solvation of the formed complex in the organic solvent; IV. Re-arrangement of the aqueous phase due to the release of hydroxyl ions (OH hydration). So the anion hydration, the interation between the extractant and the target anion, and the solvation of the formed compound in the organic phase may play important roles in the extraction process. For the extraction of Cu(CN)32 by LIX 7950, a relative small amount of energy will be needed to break the hyration shell of Cu(CN)32,however, a large amount of energy will be released due to the hydration of OH (2 mol OH will be released; AH° for OH hydration is — 529 U/mo!, Table 6-2). This is probably the main reason that the extraction of Cu(CN)32by LIX 7950 is exothermic with a relative large enthalpy change (AH° = -191 KJ/mol). More specifically, the different affinity of anions with the extractant and the selectivity order can be explained in the following interrelated factors: o charge density (anion charge and size ratio); o anion hydration properties; o compatibility of the formed complex with the organic phase; o geometry effect. 6.1 Non-metal anion extracting behaviors The hydration properties of some anions are summarized in Table 6-2. The hydration number provides a measure of the degree of hydration of an anion in water. The OH and S042 ions have much higher hydration numbers than the others in Table 6-2, indicating they are the most in need of hydration in water, and are the least extracted by the extractant. As a result, the presence of sulfate ions in copper cyanide solution exhibits little effect on copper extraction with the extractant. Though the data on the hydration ofS203 ions in the water are lacking, it is believed that the hydration properties will be similar to those of the S042 ion. An investigation has shown that the solvation of S203 ions in the water is much more complete than N03 ions due to its higher charge density (Afana&ev and Tyunina, 2004), indicating S203 ions are well hydrated in water. SCN ion has a lower hydration number compared with CN, CNO and S203 ions and thus is the least in need of hydration. It is more lipophilic (or hydrophobic) and is the easiest to 101 be extracted. Subsequently, the presence of SCN in copper cyanide solution significantly depresses copper extraction with the extractant due to their high competition for the available extractant with copper cyano complexes. That CNO ions exhibit a slight better affinity with the extractant than CN ion is probably due to their relative larger size which results in their overall lower charge density. The sequence of the ion radius of the three singly charged anions is as below, CN <CNO <SCN. Xie and Dreisinger (2008) found that the presence of nitrate and chloride ions in copper cyanide solution exhibits much higher reduction in copper extraction with LIX 7820 than sulfate ions. According to these principles, the phenomenon that N03 ions exhibit more significant effect than cr ion is as expected since both are carrying a single charge, the former has a relatively large ionic radius (and larger hydration number) and thus is more hydrophobic. Though the size of S042 ions is relative large, they are highly hydrated and tend to form a tight solvation shell in the water since they have two charges and are with the largest hydration numbers. As a result, S042 ions exhibit the least affinity for the extractant solvent compared with C1 and N03. Anion Table 6-2 The hydration properties of some anions (abstracted from Marcus, 1997 and Marcus, 1985) ion radius, L\hydG°, AhdH°, 1\hydS°, nm kJ mof1 kJ moF1 J K’mo11 Ionic hydration numbers (from AhydS°) OH 0.133 -439 -529 11.4 5.9 S042 0.230 -1090 -1138 64.5 6.9 Ci 0.181 -347 -376 78.7 2.3 N03 0.179 -306 -329 168.6 2.0 CN 0.191 -305 -326 116.3 2.1 CNO 0.203 -- -404 128.9 2.5 SCN 0.213 -287 -307 166.5 1.6 Fe(CN)64 0.450 10.3-- -- 183.8 102 6.2 Metal cyano complexes extraction As discussed above, the small CN ions are highly hydrated. Metal cyanide complexes are usually much larger than CN ions, i.e., the ionic radius of Cu(CN)32 is about 0.302 nm compared to 0.19 nm for CN (Torre. et al., 2006, Marcus, 1985). Therefore, the hydration of CN ion is much higher than Cu(CN)32 due to the higher charge density of CN. The compatibility of the anion with the extractant molecule may also play an important role during extraction. Since the extractant molecule is relatively large, the large size of the metal cyano complex may render the distribution of the charge over the compound of the extractant and the extracted complex more evenly. This facilitates compatibility within the organic phase. When the small CN ions are extracted, it will be difficult for the big extractant molecules to ‘trap’ them. The preferential extraction of metal cyano complexes over free cyanide (CN) by the extractant is thus favored. As discussed in Chapter 1, when forming complexes with transition metals, cyanide ligands can stabilize the metal ion by delocalizing charge density from the central metal ion by means of accepting electrons into its t anti-bonding orbital (Sharpe, 1976). This results in the charge delocalization over the metal cyanide complex. For the metal cyano complexes, the negatively charged CN ligands will exhibit dual function on the charge density for the complex. On one hand, the ligands can render the charge distribution of the complex more even. On the other hand, its negative charge will increase the charge density for the complex (assuming the coordinated number of CN ligands exceed the charge of the central transition metal). It can be deduced that for the same central bonding metal, the higher the coordinated CN ligands number, the higher the charge density will be with the complex. Therefore, Cu(CN)43 will have a relatively higher charge density than Cu(CN)32 and the charge density for Fe(CN)64 will be much higher than that for Zn(CN)42,Ni(CN)42 and Cu(CN)32.For the same charged complexes, Cu(CN)32will exhibit a higher charge density than Ni(CN)42,Zn(CN)42 due to its lesser number of coordinated CN ligands. Due to charge density effect, the lower charged metal cyano complexes would be extracted preferentially over those higher charged complexes and those complexes with lower coordination numbers were extracted preferentially over those with higher 103 coordination numbers (Sastre, et al, 1999,). The preferential extraction of Zn and Ni over Cu and Fe by the extractants is in accordance with these observations. Geometrical factors may play important roles on the compatibility of the complexes with the extractant molecules (Riveros. 1993). Cu(CN)43 has a tetrahedral shape and Cu(CN)32 is trigonal planar (Sharpe, 1976). A more symmetrical complex may form between the extractant molecules and Cu(CN)32 than Cu(CN)43.Moreover, the extraction of Cu(CN)43 would require three molecules of the extractant per molecule of copper while Cu(CN)32 would only require two. These factors indicate a better compatibility of Cu(CN)32 with the extractants than Cu(CN)43,therefore, the later one is poorly extracted. The preferential extraction of Zn(CN)42 over Ni(CN)42 is probably due to the fact that Zn(CN)42 has a tetrahedral shape while Ni(CN)42 is square planar. The distribution of charge over Zn(CN)42 may be more uniform than Ni(CN)42 and is thus more compatible with the organic phase when forming compound with the extractant molecules. The complicated octahedral shape of Fe(CN)64 ion results in its poor affinity with the extractant. The relatively low charge density for Au(CN)2 and Ag(CN)2 complexes results in a much stronger affinity with the amine extraction systems compared to those base metal cyanide complexes mentioned above (Sastre, et al, 1999, Riveros, 1990, Mooiman and Miller, 1986). The linear shape of these complexes also makes them more compatible in the organic phase when forming compounds with the extractant molecules. As a result, amine extractants always exhibit better selectivity for Au and Ag than for other metals. However, silver extraction may be also dependent of cyanide concentration since a higher level of NaCN favors the formation of the higher charged complexes such as Ag(CN)32 and Ag(CN)43 which exhibits much weaker affinity with the extractant. This is probably the reason that the selectivity order for Ag over the other metals varies in studies reported in the literature. 104 7 Stripping of the Loaded Copper and Cyanide 7.1 Effect of stripping reagents 7.1 .1 Stripping with NaOH solution It is desirable to recover copper and cyanide from the loaded solvent using a simple technique so that the organic solvent can be reused. As discussed in the former chapters, for both extractant solvents, the extraction of copper and cyanide decreases when equilibrium pH is increased, indicating that the extracted copper and cyanide can be gradually repelled to the aqueous phase with an increase of equilibrium pH. As a result, NaOH solution was tested for stripping the loaded copper and cyanide from the solvents. The organic samples collected from the extraction tests were used for stripping tests. For LIX 7950, the loaded copper and cyanide in the organic are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively. The stripping test was conducted at unity volume ratio of organic phase to aqueous phase (0/A) and 20 °C. The stripping results for LIX 7950 are shown in Table 7-1. It shows that the stripping efficiency of copper and cyanide increases when NaOH concentration is increased from 0.05 mol/L to 1 mol/L. About 93% of copper and cyanide can be stripped off by 1 mol/L NaOH solution, indicating sodium hydroxide is an efficient stripping reagent. However, when the NaOH concentration is further increased (above 5 mol/L), an obvious third phase formed between two clear phases (organic and aqueous). This phenomenon also occurred when the blank extractant solution (10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane) was contacted with the concentrated NaOH solution (5 mol/L or higher). In both cases, the volume of the aqueous phase decreased significantly (from initially 25 mL down to 22 mL), but the change of the volume of the organic phase was negligible. The NaOH concentration in the aqueous phase decreased slightly. The mass balance calculation on copper and cyanide shows that part of copper and cyanide occurred in the third phase. It was found that the clear organic phase exhibited a much poorer extraction capability on copper and cyanide when re-contacted with copper cyanide solution. These facts indicated that the third phase was probably formed by sodium hydroxide, water, the organics (including the extractant), 105 and part of copper and cyanide if they are present. As a result, the stripping efficiency of copper and cyanide decreased significantly. Since the formation of third phase is intolerable in practice, the moderate strong NaOH solutions (< 5 mol/L) are proposed for copper cyanide stripping. The stripping of loaded copper and cyanide from LIX 7820 solvent was examined with the same method. The loaded copper and cyanide in the organic sample are 2.17 x 10 mol/L and 6.50 x i0 mol/L, respectively. Table 7-2 shows the results of stripping copper and cyanide from the loaded LIX 7820 solvent mixture under different Na0H concentrations. Above 90% of copper and cyanide can be stripped by 1 mol/L Na0H solution. However, similar to LIX 7950, when a higher concentration of NaOH (5 mol/L) was used, 0/A ratio changed significantly and a third phase was produced, resulting in reduced stripping of copper and cyanide. Table 7-1 Stripping of Cu and CN from the loaded LIX 7950 by NaOH solutions (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively; 0/A = 1; 20 °C) NaOH, M Cu stripping, % CN stripping, % 0.05 89.9 90.5 0.1 90.8 91.2 1 93.9 93.3 5** 79.4 81.2 10 2.8 3.1 * *Third phase formed. Table 7-2 Stripping of copper and cyanide from the loaded LIX 7820 by NaOH solutions (Org: 2% v/v 7820 in n-octane, loaded Cu and total CN are 2.17 x i0 mol/L and 6.50 x i0 mol/L, respectively; 0/A = 1; 20 °C) NaOH, M Cu stripped, % CN stripped, % 0.1 88.9 89.5 1 92.8 92.2 5** 37.4 38.5 **Third phase formed. 106 7.1.2 Stripping with NaCN-NaOH solution A mixed solution of NaCN and NaOH was tested for stripping loaded copper and cyanide from the organic solvents. The stripping solutions contained 1 mol/L NaOH with varied NaCN concentration. The stripping results for LIX 7950 and LIX 7820 are shown in Table 7-3 and 7-4, respectively. For both cases, the stripping efficiency of copper and cyanide by the NaCN-NaOH solutions is slightly higher than by NaOH solutions only. At the same concentration of NaOH, the stripping efficiency of copper and cyanide increases with an increase of NaCN concentration. It is believed that the addition of NaCN to the stripping system increases the molar ratio of copper to cyanide which favors the formation of Cu(CN)43 and subsequently facilitated the stripping of the loaded copper and cyanide since both the extractants exhibit much weaker affinity for Cu(CN)43 than for Cu(CN)32.Valenzuela, et al. (2003) examined the stripping of loaded gold from the guanidine extractant LIX 79 with the mixture solution of NaCN and NaOH and found that the addition of a small amount of NaCN in NaOH solution can increase the stripping efficiency of gold. The research results of Sastre, et al. (1999) also indicated that both NaCN and NaOH solutions can strip the loaded gold efficiently. The authors claimed that the stripping efficiency with NaOH is higher than with NaCN. However, by examining their results, it shows that the gold stripping with 0.01 mol/L NaCN solution is higher that with 0.1 mol/L NaOH solution, indicating the stripping with NaCN is more efficient than with NaOH solution. Since Au(CN)j is the only gold and cyanide complex formed in gold cyanide solution, the effect of NaCN on gold stripping is not due to the formation of a higher coordinated complex. According to Table 6-2, OH ion has a much higher hydration number than CN ion, indicating the former is more hydrated and is more lipophobic. As a result, stripping with the NaCN solution is more efficient than with the NaOH solution at the same concentration. 107 Table 7-3 Stripping of copper and cyanide from LIX 7950 by NaOH-NaCN solutions (Org: 10% v/v LIX 7950 and 50 gIL 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively, [NaOHJ = 1 mol/L; 0/A = 1; 20 °C) NaCN, mol/L Cu stripping, % CN stripping, % 0 93.9 93.3 0.00 1 94.5 94.9 0.005 95.4 96.5 0.01 96.8 96.2 Table 7-4 Stripping of copper and cyanide from LIX 7820 by NaOH-NaCN solutions (Org: 2% v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x i0 mol/L and 6.50 x i0 mol/L, respectively,[NaOHJ = lmol/L; 0/A = 1; 20 °C) NaCN, mol/L Cu stripping, % CN stripping, % 0 92.8 92.2 0.001 93.9 94.2 0.005 95.4 95.5 0.01 95.8 95.7 7.2 Effect of temperature The stripping of loaded copper and cyanide from LIX 7950 and LIX 7820 solvents with NaOH solutions at different temperatures was examined. The temperature varied from 25 °C to 45 °C and 1 mol/L NaOH solutions was used as stripping solution. The results are shown in Figure 7-1 and 7-2. For LIX 7820, temperature exhibits negligible effect on the stripping efficiency of copper and cyanide at the range of 25 °C to 45 °C. For LIX 7950, the stripping efficiency of copper and cyanide increases slightly when temperature is increased from 25 °C to 45 °C. The results are in accordance with the temperature effect on extraction of copper and cyanide by the two extractants. 108 100 75 .50 Cu 25 a CN 0 I 25 30 35 40 Temperature, C Figure 7-1 The stripping of loaded Cu and CN by NaOH solution under different temperatures (Org: 10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively, aq: [NaOH] = lmol/L; 0/A = 1) 100 75 .Cu a CN 0 25 30 35 40 Temperature, C Figure 7-2 The stripping of loaded Cu and CN by NaOH solution under different temperatures (Org: 2% v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x i0 mol/L and 6.50 x 10 mol/L, respectively, aq: [NaOH] = 1 mol/L; 0/A = 1) 109 7.3 Effect of phase ratio The stripping of the loaded copper and cyanide from LIX 7950 solvent with 1 mol/ L NaOH solution under different phase ratios was examined. The loaded Cu and CN in the organic solvent (10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane) are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively. The stripping isotherm of copper from the extractant was established by varying the volume ratio of the organic phase to the aqueous phase (0/A) from 2:1 to 1:8. The schematic McCabe-Thiele diagram is shown in Figure 7-3, in which the operating line is supposed to be at an 0/A ratio of 10. It shows that a single stage of stripping with 1 mol/L NaOH is sufficient for a 93% stripping efficiency of copper. Since cyanide occurs as Cu(CN)32,the stripping isotherm of CN can be established accordingly. The schematic McCabe-Thiele diagram for LIX 7820 is shown in Figure 7-4. The loaded Cu and CN in the organic solvent (2% v/v LIX 7820 in n-octane) are 2.16 x i0 mol/L and 6.50 x 10 mol/L, respectively). The stripping isotherm of copper from the loaded LIX 7820 solvent was established by varying the volume ratio of the organic phase to the aqueous phase (0/A) from 2:1 to 1:8. A single stage of stripping with 1 mol/L NaOH solution is sufficient for a 92% stripping efficiency of copper. 110 2000 1600 -J 1200 0• Ca 800 C.) 400 0 0 40 80 120 160 Cuinorg,mgIL Figure 7-3 The striping isotherms of copper from the extractant solvent of LIX 7950 (10% v/v LIX 7950 and 50 g/L 1-dodecanol in n-dodecane, loaded Cu and CN are 3.68 x i0 mol/L and 1.11 x 102 mol/L, respectively, aq: [NaOH] = 1 mol/L; 20°C) 1200 stripping isotherms operating line O/A=1O 800 E 0(U c 400 0 0 20 40 60 80 100 120 Cu in org, mg/L Figure 7-4 The striping isotherms of copper from the extractant solvent of LIX 7820 (2% v/v LIX 7820 in n-octane, loaded Cu and CN are 2.16 x i0 mol/L and 6.50 x i0 mol/L, respectively, aq: [NaOH] = lmol/L; 20 °C) 111 8 Conclusions and Recommendations 8.1 Conclusions Solvent extraction of copper and cyanide from alkaline cyanide solution with two amine extractants, LIX 7950 (a guanidine derivative) and LIX 7820 (the solvent mixture of a quaternary amine and 4-nonylphenol) has been examined in the research. (1) For LIX 7950, copper extraction with n-dodecane as diluent is slightly higher than that with toluene. When above the certain minimum concentration to avoid emulsion, the effect of the modifier concentration (1-dodecanol) on copper extraction is insignificant. For LIX 7820, the diluent types exhibit insignificant effect on copper extraction. (2) For both extractants, the extraction kinetics of copper are relatively fast and the equilibrium can be established within two to three minutes under the experimental conditions. Solvent extraction of the mixture solution of metal cyanides indicates that the extraction kinetics of Zn, Ni, and Fe(II) by the two extractants is also relatively fast. (3) For both extractants, low equilibrium pH favors the extraction of copper and cyanide. A high CN/Cu ratio depresses the extraction of copper and cyanide. The preferential extraction Cu(CN)32 over Cu(CN)43 and CN by two extractants is confirmed. Both extractants exhibit a selectivity order for metal cyano complexes as follows: Zn(CN)42 > Ni(CN) > Cu(CN)32 > Fe(CN)64.Cyanide is mainly extracted as complexes and most of the free cyanide remains in the aqueous phase. (4) Both extractants show an affinity sequence for four non-metal anions as follows: SCN > CNO > CN >> S203. The presence of S2032 results in an insignificant effect on copper extraction while copper extraction decreases significantly with an increase in SCN concentration. The presence of SCN ions strongly competes for the available extractant with copper cyanides species during extraction. 112 (5) The selectivity order of the metal cyanide complexes and the affinity sequence of anions with the extractants can be explained by a consideration of anion hydration properties, the charge and size effect, compatibility of the formed complex with the organic phase and the geometric factors. Due to charge density effect, the lower charged metal cyano complexes would be extracted preferentially over those higher charged complexes and those complexes with lower coordination numbers were extracted preferentially over those with higher coordination numbers. For the anions with same charge, the larger the size of the anion, the stronger affinity with the extractants. (6) Temperature shows significant effect on copper extraction with LIX 7950. Both the distribution coefficient and the stoichiometric extraction constant of Cu(CN)32 decrease with an increase of temperature. The extraction of Cu(CN)32 with LIX 7950 is exothermic and the enthalpy change of the reaction (AH° ) was -190 kJ/mol. For LIX 7820, the effect of temperature on copper extraction is insignificant under the experimental conditions. (7) For both extractants, the loaded copper and cyanide can be stripped off efficiently by a moderately strong NaOH solution. Further increase in NaOH concentration may result in the formation of the third phase. The addition of NaCN to NaOH stripping solutions can facilitate the stripping of copper and cyanide. For LIX 7820, temperature exhibits insignificant effect on stripping of copper and cyanide in the range of 25 °C to 45 °C. For LIX 7950, the stripping efficiency of copper and cyanide increases slightly when the temperature is increased from 25 °C to 45 °C. (8) Due to the preferential extraction of Cu(CN)32 over CN by the extractants, the extraction isotherms of copper and of cyanide are different. A semi-graphic method using McCabe-Thiele diagram to model the operation of solvent extraction of copper and cyanide has been constructed. For both extractants, above 90% copper can be extracted after two-stage extraction under the experimental conditions. 113 8.2 Recommendations The important findings in this research suggest a possible solution to the separation of metal cyanide species and free cyanide in the cyanide effluent. Both extractants can be used in the SX circuit for pre-concentrating copper cyanide into a small volume of strip solution which can be further treated by electrowinning, AVR, or similar processes to recover copper products and cyanide (Figure 8-1). Due to the preferential extraction of metal cyanide complexes, most of the free cyanide will remain in the aqueous phase which allows for the potential recycling of the barren solution to the cyanidation process. If zinc and nickel cyanide complexes are present in the waste solution, they will be preferentially extracted into the organic phase and their potential effect on the subsequent copper recovery process should be considered. Iron(II) cyanide complexes will probably accumulate in the recycling water if a closed circuit system is applied in practice and their potential effects on the cyanidation process should be considered. The need for clarified feed solution for solvent extraction will not be a limitation while targeting heap leaching solutions, overflow stream or dam return water from tailings. For operations using carbon-in- pulp (CIP) for the recovery of gold, it will be necessary to thicken and wash the solids in order to produce a clarified feed solution for SX circuit. Cyanide Effluent Loading Stripping Stripping Solution Raffinate Recovery of Cu and CN by AVR, SART, or Recycling to cyanidation Electrowinning Figure 8-1 The schematic flowsheet for the potential application of SX circuit SX Circuit r 114 Bibliography Adams, M. D., 1990. The chemical behavior of cyanide in the extraction of gold, J. S. Mr. Inst. Mm. Metal!., 90, 37-44 Adams, M. D., 1994. Removal of cyanide from solution using activated carbon, Minerals Engineering, 7(9), 1165-1177 Adhoum, N., Monster, L., 2002. Removal of cyanide from aqueous solutions using impregnated activated carbon, Chemical Engineering and Processing, 41, 17-21 Afanas’ev, V. N., Tyuninal E. Yu., 2004. Structural features of ion hydration in sodium nitrate and Thiosulfate, Russia Journal of Generals Chemistry, 74 (5), 673-678 Aguayo, S., Valenzuela, J., Parga, J. R., Lewis, R. G., Cruz, M., 2007. Continuous laboratory gold solvent extraction from cyanide solutions using LIX 79 reagent, Chemical Engineering and Technology, 30(11), 1532 -1536 Alguacil, F. J., Alonso, M., 2005. Liquid-liquid extraction of Au(I) by Amberlite LA2 and its application to polymer-immobilized liquid membrane system, Gold Bulletin, 38 (2) Almanda, E., Lizama, H., Orellana, F., Suarez, C., Huidobro, A., 2002. Catalytic oxidation of Fe(II) by activated carbon in the presence of oxygen, Carbon, 40(15), 2827-2834 Bachiller, D., Torre, M., Rendueles, M., Diaz, M., 2004, Cyanide recovery by ion exchange from gold ore waste effluents containing copper, Minerals Engineering, 17, 767-774 Barter, J., Lane, G., Mitchell, D., Kelson, R., Dunne, R., Trang, C., Dreisinger, D., 2001. Cyanide management by SART, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, Pa 549-562. Batzias, F. A., Sidiras, D. K., 2001. Wastewaters treatment with gold recovery through adsorption by activated carbon, Water Pollution VI, International series on progress in water resources, 3, 143-152 Bernardin, F. E., 1976. Selecting and specifying activated-carbon-adsorption system, Chemical Engineering, 83(22), 77-82 Bessent, R. A., Luther, P. A., Eklund, C. W., 1980. Removal of cyanides from coke wastewaters by selective ion exchange- results of pilot testing program, In: Proceedings of the 34th industrial waste conference, Ann Arbor Science, 47-62 115 Bolinski, L., Shirley, J., 1996. Russian resin-in-pulp technology, current status and recent developments, In: Proceedings of the Randol Gold Forum ‘96, Randol International, 419-423 Borberly G., Devuyst, E., Ettel, V., Mosoiu, M., Schitka, K., 1985. Cyanide removal from aqueous streams, Canadian Patent No. 1,165,474, April 1984 and US Patent No. 4, 537,686, August 1985 Botz, M. M., Mudder, T. I., Akcil, A.U., 2005. Cyanide treatment: physical, chemical and biological process, In: Advances in gold ore processing, Adams, M. D. (ed.), Elsevier, 672-702 Breguncci, V. A., Riveiro, C. P., Martins, A. H., 2001. Gold recovery from alkaline cyanide solutions by electrochemical elution of strong-base anionic polymeric resins, Minerals & Metallurgical Processing, 18(1), 49-53 Caravaca, C., Alguacil, F. J., 1994. Extraction of gold(I) from cyanide aqueous media by Primene JMT, Hydrometallurgy, 35, 67-78 Caravaca, C., 1994. Gold(I) extraction from cyanide media by the primary amine Primene 81R/iso-decanol system, Hydrometallurgy, 35, 53-66 Caravaca, C., Alguacil, F. J., Sastre, A., Martinez, M., 1996. Extraction of gold(I) cyanide by the primary amine tridecylamine, Hydrometallurgy, 40, 89-97 Caravaca, C., Alguacil, F. J., Sastre, A., 1996. The use of primary amine in gold(I) extraction from cyanide solution, Hydrometallurgy, 40, 263-275 Carlson, L. G., 1982. U.S. patent, US 4321145 A Clancy, J., Ellis, J., Rowley, P., Wheway, R., 1978. The detoxification of industrial effluent, Report No. 53, Water Research Foundation of Australia, Kingsford Clark, P., Jordan, D. M., Malloy, T. M., 2001. Cyanide heap biological detoxification, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 237-251 Cooley, A. C., 1976. Regeneration and disposal of photographic processing solutions containing hexacyanoferrate, J. Appl. Photogr. Eng., 2(2), 61-64 Csikai, N. J., Barnard, A. J., 1983. Determination of total cyanide in thiocyanate-containing waste water, Journal of Analytical Chemistry, 55, 1677-1682 Davis, M. R., MacKenzie, M. W., Sole, K. C., Virnig, M. J., 1998. A proposed solvent extraction route for the treatment of copper cyanide solutions produced in leaching of gold ores, Alta Copper Hydrometallurgy Forum, Brisbane, Australia, 14-17 116 Deveci, H., Yazici, E. Y., Alp, I., Ulsu, T., 2006. Removal of cyanide from aqueous solutions by plain and metal-impregnated granular activated carbon, mt., J. Miner. Process., 79, 198-208 DeVries, F., 2001. Brief overiew of the Baia Mare Dam breach, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 11-14 Devuyst, E. A., Conard, B. R., Robbins, G., Vergunst, R, 1989. A cyanide removal process using sulfur dioxide and air, JOM, 41(12), 43-45 Devuyst, E. A., Robbins, G., Vergunst, R. Tandi, B., lamarino, P. F., 1991. Inco’s cyanide removal technology working well, Mining Engineering, 43(2), 205-207 Diaz, X., Caizaguano, R., 1999. Bioremediation of cyanide leaching residue, In: Biohydrometallurgy and the environment toward the mining of the 21st century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A.(eds.), Elsevier, 5 95-606 Dobrevski, I., Dimova-Todorova, M., Panyotova, T., 1994. Ion exchange treatment of gold and silver containing wastewaters from electroplating plants, Environment Protection Engineering, 18(1-2), 81-87 Dreisinger, D., Ji, J., Wassink, B., 1995. The solvent extraction and electrowinning recovery of copper and cyanide using X17950 extractant and membrane cell electrolysis, Randol Gold Forum, Perth, Austalia, 239-244 Dreisinger, D. B., Wassink, B., De Kock, F. P., West-Sells, P., 1996. Solvent extraction and electrowinning recovery of copper and cyanide — Recent Developments, Randol Gold Forum ‘96, Randol International, Golden, CO., 315-319 Dreisinger, D., Wassink, B., Ship, K., King, J., Hames, M., Hackl, R, 2002. Cyanide recovery by solvent extraction, International Solvent Extraction Conference, Cape Town, South Africa, Publisher: S. Mr. Inst. Mm. and Metal., Marshalltown, S. Mr., Mar. 17-21, 2002 (2002), 798- 803. Dreisinger, D., Wassink, B., Lu, J., De Kock, F., Ji, J., West-Sells, P., Dunne, R., 2001. Solvent extraction recovery of copper cyanide from spent gold mill effluents. In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 347-360 du Preez, J. G. H., 2000. Recent advances in amines as separating agent for metal ions, Solvent Extraction and Ion Exchange, 18(4), 679-701 Elsner, 1846. German Journal Für Praktische Chemie, 37, 441-446 117 Figuerola, E., Sastre, A., Miralles, N., 1992. Solvent extraction of some metal cyanide complexes by trioctylmethylammonium chloride, Journal of Chemical Research (S), 24-25 Fleming, C., Cromberge, G., 1984(A). The elution of auro-cyanide from strong and weak base resins, Journal of the South African Institute of Mining and metallurgy, 84(9), 269-280 Fleming, C., Cromberge, G., 1984(B). Small scale pilot-plant tests on the resin-in-pulp extraction of gold from cyanide media, Journal of the South African Institute of Mining and metallurgy, 84(11), 369-378 Fleming, C. A., 1998. The potential role of anion exchange resins in the gold industry. In: EPD Congress, Misha, B. (ed.), 95-118 Fleming, C. A., 2001. The case for cyanide recovery from gold plant tailings — positive economics plus environmental stewardship, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 271-288 Fleming, C. A., Grot, W. G., Thorpe, J. A., 1998. Hydrometallurgy extraction process, US patent, No. 5.807.421 Fleming, C. A., 2005. Cyanide recovery, In, Advances in gold ore processing, Adams, M. D. (ed.), Elsevier, 703-727 Flynn, C. M., McGill, S. L., 1995. Cyanide chemistry-precious metals processing and waste treatment, United States Bureau of Mines Information Circular/1995, 9429 Goldblatt, E., 1956. Recovery of cyanide from waste cyanide solution by ion exchange, Industrial and Engineering Chemistry, 48, 2107-2114 Goldblatt, E., 1959. Recovery of cyanide from waste cyanide solution by ion exchange, Industrial and Engineering Chemistry, 51(3), 241-246 Goldstein, M., 1976. Economics of treating cyanide wastes, Pollution Engineering, 8(3), 36-38 Gomes, C. P., Almeida, M. F., Loureiro, J. M., 2001. Gold recovery with ion exchange used resins, Separation and Purification Technology, 24(1-2), 35-37 Gonzalez, C. A. G., 1994. Solvent extraction chemistry of quaternary organic cations for gold recovery from alkaline cyanide solutions, Ph. D thesis, University of Utah Goode, J. R., McMullen, J., Wells, J. A., Thomas, Ken G., 2001. Cyanide and the environment; Barrick Gold Corporation’s perspective, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, 118 Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 12-15 Greenwood, N. N., Earnshaw, A. 1997. Chemistry of the Elements, 2nd Edition, Pergamon press, New York Gregor, K. H., 1992. Cyanide detoxification with peroxygens, In: Proceedings of 1st International Symposium Chemical Oxidation, Eckenfelder, W. W., Browers, A. R., Roth, J. A.(eds.), Technomic, Lancaster, Pa. Griffiths, A., Knorre, H., Gos, S., Higgins, R., 1987. The detoxification of gold-mine tailings with hydrogen peroxide, J. S. Mr. Inst. Mm. Metal!., 87(9), 279-283 Haristy, D. M., Rosen, H. M., 1977. Industrial wastewater ozonation, Proceedings of 321I Industrial Waste Conference, Purdue University, Lafayette, IN, 294-302 Hedley, N., Tabachnik, H., 1968. Chemistry of cyanidation, Mineral Dressing Notes, No. 23, American Cyanamid Company, New York Hoffman, D. C., 1973. Oxidation of cyanides adsorbed on granular activated carbon, Plating, v. 60(2), 157-161 Izatt, R. M., Christensen, J. J., Pack, R. T., Benchu, R., 1962. Thermodynamics of metal-cyanide coordination, I. pK, H°, and S° values as a function of temperature for hydrocyanide acid dissociation in aqueous solution, Inorganic chemistry, 4, 828-831 Irving, H. M. N. H., Damodaran, A. D., 1971. The extraction of complex cyanides by liquid ion exchangers, Anal. Chim. Acta, 53, 267-275 Jay, W. H., 2001. Copper cyanide recovery system, In: Cyanide: Social, Industrial and Economic Aspects, Proceedings of a Symposium held at Annual Meeting of TMS, New Orleans, LA, United States, Feb. 12-15, 2001, Young, C. A., Twidwell, L. G., Anderson, C. G. (eds.), TMS, Warrendale, PA, 3 17-340 Jermakowicz-Bartkowiak, D., Kolarz, B. N., 2002. Gold recovery from guanidine resins, Separation Science and Technology, 37(15), 3513-3524 Jiang, J., Zhou, W., Gao, H., Wu, J., Xu, G., 2003. Solvent extraction and stripping of gold(I) cyanide in the tetradecyldimethylbenzylammonium choride system, 70, 73-81 John, M. W., 2001. The evaluation of adsorbents for the recovery of gold, In: Symposium Series, S16 (Hidden Wealth), Glen, H.W.(ed.), S. Mr. Inst. Mm. and Metall. (SAIMM), 157-161 Kordosky, G. A., Sierakoski, J. M., Virnig, M. J., Mattison, P. L., 1992. Gold solvent extraction from typical cyanide solutions, Hydrometallurgy, 20, 291 119 Kordosky, G.A., Kotze, M. H., Mackenzie, J. M. W., Virnig, M. J., 1993. New solid and liquid ion exchange extractants for gold, In: Proceedings XVIII International Minerals Processing Congress, v. 5, Sydney, Australia, 1195-1203. Kunz, R. G., Casey, J. P., Huff, J. E., 1978. Refinery cyanides: A Regulatory Dilemma, Hydrocarbon Processing, 57(10), 98-106 Lanoutte, K. H., 1977. Treatment of phenolic wastes, Chem. Eng., 84(22), 99-106 Larmour-Ship, K, Buchalter, E., Dreisinger, D., Wassink, B., Hackl, R., Hames, M., Grinbaum, B., Cyanide recovery by solvent extraction in a continuous plant, In Proceedings of the international Symposium on the Treatment of Gold Ore, Deschenes, G., Hodouin, D., Lorenzen, L. (eds.), CIM, Calgary, Canada, 2005, 469-483 Leão, V. A., Ciminelli, V. S. T., Costa, R. S., 1998. Cyanide recycling using strong-base ion exchange resins, JOM, 50(10), 66-69 Leao, V. A., Ciminelli, V. S. T., 2000. Application of ion exchange resins in gold hydrometallurgy, a tool for cyanide recycling, Solvent extraction and ion exchange, 18(3), 567- 582 Leao, V. A., Lukey, G. C., Ciminelli, V.S.T., 2001. The effect of resin structure on the loading of copper and iron cyanocomplexes, Solvent extraction and ion exchange, 19(3), 507-530 Lower, G. W., Spottiswood, D. J., 1983. Cyanide removal from coke making and blast wastewaters, In: Report (1983), US Government Report Announcement Lu, J., Dreisinger, D. B., Cooper, W. C., 2002. Thermodynamics of the aqueous copper-cyanide system, Hydrometallurgy, 66(1-3), 23-36 Lukey, G. C., Deventer, J. S. J., Chowdhury, R. L., Shallcross, D. C., 1999. The effect of salinity on the capability and selectivity of ion exchange resins for gold cyanide, Minerals Engineering, 12(7), 769-785 Lukey, G. C., Deventer, J. S. J., Chowdhury, R. L., Shallcross, D.C., 2000. Selective elution of copper and iron cyanide complexes from ion exchange resins using saline solutions, Hydrometallurgy, 56, 217-236 Ma, G., Yan, W., Chen, J., Yan, C., Gao., Zhou, W., Shi, N., Wu, J., Xu, G., Huang, K., Yu, J., Cui, N., 2000. Mechanism of gold solvent extraction from aurocyanide solution by quaternary amines: modes of extracting species based on hydrogen bonding, Science in China (series B), 43 (2), 169-177 Manktelow, S. A., Paterson, J. G., Meech, J. A., 1984. Removal of copper and cyanide from solution using activated carbon, Minerals and the Environment, 6(1), 5-9 120 Marcus, Y., 1997. Ion properties, M. Deckker, New York Marcus, Y., 1985. Ion salvation, John Wiley & Sons Ltd., Chichester, UK Marsden, J. and House, I., 1992. The chemistry of gold Extraction, Ellis Horwood Ltd. Matsuda, Y., Takasu, Y., Fujisawa, T., Fujikawa, S., 1981. Ozonation of cyanate ion in alkaline solution, Plating and Surface Finishing, 68(6), 86-88 Mattison, P., Virnig, M., 2001. LIX-7800 series: A new reagent family for reversibly extracting anions from alkaline solutions, Solvent extraction for the 21st century, Proceedings of ISEC’99, v. 1, Barcelona, Spain, July 11-16, 1999, Cox, M., Hidalgo, M, Valiente, M. (eds.), 375-379 McDougall, G. J., Fleming, C. A., 1987. Extraction of precious metals on activated carbon, In: Ion exchange and sorption process in hydrometallurgy, Critical reports on applied chemistry,. Streat, M, Naden, D. (eds.), The society of chemical industry, 19, 56-126 Mickler, W, Reich, A., Uhlemann, E., 2001. Reactive liquid-liquid extraction of heavy metals from landfill leachate, In: Solvent extraction for the 21st century, Proceedings of IESC’99, Barcelona, Spain, July 11-16, 1999, Cox, M., Hidalgo, M, Valiente, M. (eds.), 225-229 Miller, J. D., Mooiman, M. B., 1984. A review of new development in amine solvent extraction systems for hydrometallurgy, Separation Science and Technology, 19, 895-909 Miller, J. D., Wan, R. Y., Sibrell, P. L., 1987. Selective solvation of gold from alkaline cyanide solution by alkyl phosphorus esters, Separation Science and Technology, 22(2&3), 487-502 Mooiman, M. B., Miller, J. D., 1984. Selectivity considerations in the amine extraction of gold from alkaline cyanide solutions, Minerals and Metallurgy Processing, 1(2), 153-157 Mooiman, M. B., Miller, J. D., 1986. The chemistry of gold solvent extraction from cyanide solution using modified amines, Hydrometallurgy, 16, 245-260 Moore, F. L., 1975. Solvent extraction of cadmium from alkaline cyanide solutions with quaternary amines, Environmental Letters, 10(1), 37-46 Moore, F. L., Groenier, W. 5., 1976. Removal and recovery of cyanide and zinc from electroplating wastes by solvent extraction, Plating and Surface Finishing, 63(8), 26-29 Natarajan, K. A., Subramanian, S., Modak, J. M., 1999. Biosorption of heavy metal ions from aqueous and cyanide solutions using fungal biomass, In: Biohydrometallurgy and the environment toward the mining of the 21st century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A (eds.), Elsevier, 351-361 Palmer, S. A. K., Breton, M. A., Nunno, T. J., Sullivan, D. M., 1988. Metal Cyanide Containing Wastes Treatment Technologies; Noyes Data Corp.: Park Ridge, NJ, US 121 Patil, Y. B., Paknikar, K. M., 1999. Removal and recovery of metal-cyanides from industrial effluents, In: Biohydrometallurgy and the environment toward the mining of the 21st century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A. (eds.), Elsevier, 707-715 Person, II., 1976. Kinetics of formation of the tetracyanonickelate(II) ion, Acta Chem. Scand., 30A, 31-38 Prober, R., Kidon, W. E., 1976. Differential-bed study of cyanide with activated carbon, Industrial & Engineering Chemistry Process Design and Development, 15(1), 223-225 Poulson, S. R., Harrington, R. R., Drever, J. I., 1999. The solubility of toluene in aqueous salt solutions, Talanta, 48, 633-641 Ritcey, G. M., Ashbrook, A. W., 1979. Solvent extraction: principles and applications for process metallurgy, Part I, Elsevier, Amsterdam, Netherlands Riveros, P., Koren, D., McNamara, V. M., Binivignat, 1997. Mini pilot plant evaluation of the AVR process for the Las Cristinas gold project, Randol Gold Forum’97, Monterey, CA, 231-235 Riveros, P. A., 1990. Studies on the solvent extraction of gold from cyanide media, Hydrometallurgy, 24, 135-156 Riveros, P. A., Molmar, R., McNamara, V. M., 1993. Alternative technology to decrease the environmental impact of gold milling—a progress report on CANMET research activities in this field, CIM Bulletin, 86(968), 167—171 Rorke, G. V., Muhlbauer, R. M. 1999. Biological processes for thiocyanate and cyanide degradation, In: Biohydrometallurgy and the environment toward the mining of the 21st century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A. (eds.), Elsevier, 731-740 Rosa, L. H., Souza-Fagundes, E. M., Sentos, M. H., Dias, J. C. T., Paimental, P. F., Gomes, N.C.M., 1999. Degradation of thiocyanate by immobilized cells of mixed and pure cultures, In: Biohydrometallurgy and the environment toward the mining of the 21st century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A. (eds.), Elsevier, 819-824 Rowley, W. J., Otto, F. D., 1980. Ozonation of cyanide with emphasis on gold mill wastewaters, Can. J. Chem. Eng., 58(5), 646-653 Rydberg, J., Cox, M., Musikas, C., Choppin, G. R., 2004. Solvent Extraction Principles and Practice, Marcel Dekker Inc., New York 122 Satalic, D., Spenser, P., Paterson, M., 1996. Vitrokele: commercial application in the gold industry, Proceedings from AusIMM Conference, Diversity: The Key to Prosperity, The Australasian Institute of Mining and Metallurgy, Melbourne, Australia, 167-171 Satio, S., 1984. The removal of hexacyanoferrate(II) and (III) ions in dilute aqueous solutions by activated carbon, Water Research, 18(3), 3 19-323 Sastre, A. M., Madi, A., Cortina, J. L., Alguacil, F. J., 1999. Solvent extraction of gold by LIX 79: Experimental and equilibrium studies, Journal of Chemical Technology and Biotechnology, 74, 310 Satre, A. M., Kumar, A., Alguacil, F. J., 2004. Solvent extraction of silver by LIX 79: Experimental and equilibrium studies, Journal of Chemical Technology and Biotechnology, 79, 306-310 Sceresini, B., 2005. Gold-copper ores, In: Advances in gold ore processing, Adams, M. D. (ed.), Elsevier, 789-824 Seymore, D., Fleming, C. A., 1986. Golden Jubilee resin-in-pulp plant for gold recovery, In: Proceedings of the Randol Gold Forum 1989, Randol International, 243-252 Scott, J. 5., 1989. An overview of gold mill effluent treatment, In: Processing of gold mining effluent treatment seminars, Environment Canada, Ottawa, ON, 1-22 Scott, M. B., Zappia, L. R., Franzman, P. D., Miller, P. C., Walting, H. R., Houchin, M. R., 1999. In: Biohydrometallurgy and the environment toward the mining of the 21St century, International Biohydrometallurgy Symposium, IBS’99, Process Metallurgy 9B, Amils, R., Ballester, A. (eds.), Elsevier, 809-8 18 Sharpe, A. G., 1976. The chemistry of cyano complexes of the transition metals, Academic Press, London Silva, A. L., Costa, R. A., Martins, A. H., 2003. Cyanide regeneration by AVR process using ion exchange polymeric resins, Minerals Engineering, 16, 555-557 Soto, H., Nava, F., Jara, J., 1997. Cyanide regeneration and copper recovery from cyanide solutions, In: Global exploitation of heap leachable gold deposits, Hausen, D. M. (ed.), The Mineral, Metal & materials Society, 151-160 Stevenson, J. A., Wilder, A. L., Richinss, R. T., 1996. Recovery of cyanide from mill tailings, In: Proceedings of Randol Gold Forum ?96, Olympic Valley, Calif., Apr. 21-24, 1996, 179-181 Talimadge, J. A., 1967. Ion-exchange treatment of mixed electroplating wastes, Industry and Engineering Chemistry: Process Design and Development, 6(4), 419-423 123 Torre, M., Bachiller, D., Rendueles, M., Menendez, C. 0., Diaz, M., 2006. Cyanide recovery from gold extraction process waste effluents by ion exchange I. Equilibrium and kinetics, Solvent Extraction and Ion Exchange, 24, 99-117 Vachon, D. T., 1985. Removal of iron cyanide from gold effluent by ion exchange, Water Science and Technology, 17(2-3), 3 13-324 Valenzuela, J., Aguayo, S., Parga, J. R., Lewis, R. G., 2003. Gold solvent extraction from alkaline cyanide solution using LIX 79 extractant, In: Proceedings of Hydrometallurgy 2003- Fifth international conference in honor of professor Ian Ritchie, Young, C. A., Alfantazi, A. M., Anderson, C. G., Dreisinger, D. B., Harris, B., James, A.(eds.), TMS, Vancouver, 2003 Van Weert, G., De Jong, I, 1992. Trace cyanide removal by means of silver activated carbon, Emerging Process Technol. Cleaner Environ., In: Proc. Symp., Pub. Soc. Mm. Metall. Expi., Littleton, Colorado, 161-165 Vickell, G. A., Davis, D. T., Gec, R., 1989. Hydrogen peroxide treatment of gold mill waste, In: Processing of gold mining effluent treatment seminars, Environment Canada, Ottawa, ON, 145- 154 Villaverde, J. A. T., Martin, P. R., 1995. Metals recovery from liquid effluent of Gossan waste, In: Proceedings of the international symposium on waste processing and recycling in mineral and metallurgical industries II, Rao, S. R. (ed.), Vancouver, BC, CA, 35-46 Virnig, M. J., Wolfe, G. A., 1996. LIX 79: A new liquid ion exchange reagent for gold(I) and silver. In: ISEC’96 Proceedings, Value added through solvent extraction, v. 1, 311-316 Walker, C. A., Zabban, W., 1953. Disposal of plating-room wastes, V, Treatment of cyanide waste solutions by ion exchange, Plating, 40, 269-278 Warshawsky, A., Kahana, N., Kampel, V., Rogachev, I., Kautzmann, R. M., Cortina, J. L., Sampaio, C. H., 2001. Ion exchange resins for gold cyanide extraction containing a piperrazine functionality, 2. Study of the gold extraction reaction, Macromolecular Materials and Engineering, 286(5), 285-295 Weber, W. J., Corapcioglu, 0., 1981. Catalytic oxidation of cyanides, In: Proceedings of 36th Industrial Waste Conference, Purdue University, Lafayette, 500-508 Whittle, L., Aidi, S. A., Schmidt, J. W., Conard, B. R., Devuyst, E. A., 1989, Full application of the SO2 Air process at a Canadian gold mine, In: Processing of gold mining effluent treatment seminars, Environment Canada, Ottawa, ON, 347-361 Whittle, L., 1992. The piloting of VitrokeleTM for cyanide recovery and waste management at two Canadian gold mines, Randol Gold Forum, Vancouver, 379-384 124 Williams, P., 1997. The optimization of an impregnated carbon system to selectively recover cyanide from dilute solutions, Minerals Engineering, 10(5), 483-490 Wilson, I. R., Harris, G. M., 1960. The oxidation of thiocyanate ion by hydrogen peroxide, J. Am. Chem. Soc., 82, 45 15-4517 Xie, F., Dreisinger, D. B., 2008. Copper solvent extraction from waste cyanide solution by a solvent mixture of a quaternary amine and nonylphenol, In: Proceedings of International Solvent Extraction Conference 2008, Sept.15-19, 2008, Tuscon Arizona, US, 107-113 125 Appendix I Analysis Methods Free Cyanide Free cyanide is defined as the uncomplexed cyanide ion, CN, and molecular hydrogen cyanide, HCN. The concentration of free cyanide is the sum of the CN and HCN concentration. The most common method for determining free cyanide, titration with silver nitrate (AgNO3)was adopted in this research. The end-point is determined by use of rhodamine as the indicator. A blank correction is required before performing the titration. Pour 50 ml of 0.25 M NaOH into a 250 ml Erlenmeyer conical flask and add enough water to make the volume up to 150 ml. Add 10-15 drops rhodamine solution and titrate with standard AgNO3 solution and the blank value can be obtained. The silver nitrate solution can be standardized by any convenient method, such as with standardized NaC1 solution with chromate as the indicator. The titration reactions and calculation are: Ag + 2CN= Ag(CN)j [CN] = {[AgNO3]x (Vt-Vb) x 2 x DF}/V where, [CN] = Cyanide content, mol/l (multiply by 26.018 for CN in gil) [AgNO3]= silver in titrant concentration, mol/l Vt = titration volume, ml Vb = blank correction volume, ml Vs = sample volume, ml DF = dilution factor. Apparatus and Reagents -Radiometer ABU 80 Autoburrette, equipped with a 10-ml burette -250 ml Erlenmeyer conical flask -magnetic stirring plate and Teflon coated magnetic stir bar -deionized water -0.25 M NaOH (lOgll) 126 -1M NaOH (40 Wi) -0.018 M AgNO3 solution (3.06 gIl), prepare weekly and store in a glass vessel in the dark -p-dimethylaminobenzal rhodamine indicator (200m1/l in acetone) -Standard NaC1 solution, 0.04 M -Chromate indicator (50 Wi) Procedure • Add 30-40 ml 0.25 M NaOH into a 250m1 Erlenmeyer flask; • Take 1 ml solution sample (as in most cases in the experiments, this volume of solutions contain at least 0.5 mg of cyanide) and put it into the flask; • Add 6 — 10 drops of rhodamine indicator and put the flask on the magnetic stirring plate; • Titrate with 0.018 M AgNO3 solution using the autotitrator (Radiometer ABU 80). Thiocyanate Thiocyanate, SCN, reacts with Ag to form AgSCN solid, a white precipitate. Standard Ag is added to the acidic sample of SCN containing Fe(III). Ferric forms an intense red complex with SCN. As Ag consumes SCN the color fades and eventually disappears. The silver nitrate solution is standardized with the same method mentioned above. The reactions and calculations are as following Ag + SCN= AgSCN Fe3+ xSCN- <-> [Fe(SCNJ3(red-brown) Ag + 1/x [Fe(SCN]3= AgSCN + Fe3 [SCN] = {[AgNO3Jx Vt x DF}IV where, [SCN] = thiocyanate concentration, mol/l [AgNO3]= silver in titrant concentration, molIl V = titration volume, ml Vs = sample volume, ml DF = dilution factor. 127 The titration must be done in an acidic solution to avoid hydrolysis of Fe(3+) which also gives an orange-brown color. While titrating, the color fades gradually and the end point is when the red- brown color just disappears. Apparatus and Reagents -Radiometer ABU 80 Autoburrette, equipped with a 10 ml burrette -100 ml and 250 ml Erlenmeyer conical flask -magnetic stirring plate and Teflon coated magnetic stir bar -deionized water -0.02 M standard NaSCN solution -36 % wt Fe(N03)solution as indicator -0.02 M AgNO3 solution, prepare weekly and store in a glass vessel in the dark -Standard NaCl solution, 0.04 M -Chromate indicator (50 g/l) -6 M HNO3 solution Procedure • Take out designated volume solutions into a 100 ml beaker (usually 50-100 ml leaching solution in this research as thiocyanate concentration is relatively low); • Add 30- 50 ml deionized water and add enough 6 M HNO3 to make the solution up to 0.8 M HNO3 (about 6 ml 6 M HNO3); • Heat the solution and N2 gas was sparged into the solution for about 30-40 minutes to volatilize HCN; If there is any precipitate, it should be filtered and the filtrate for next step titration; • Add lml Fe(N03)indicator and titrate with the stand AgNO3 solution; • Read the end point is when red brown color just disappears. 128

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0071001/manifest

Comment

Related Items