UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Copper electrowinning from cyanide solutions Lu, Jianming 1999

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

Item Metadata

Download

Media
831-ubc_2000-486559.pdf [ 12.88MB ]
Metadata
JSON: 831-1.0078481.json
JSON-LD: 831-1.0078481-ld.json
RDF/XML (Pretty): 831-1.0078481-rdf.xml
RDF/JSON: 831-1.0078481-rdf.json
Turtle: 831-1.0078481-turtle.txt
N-Triples: 831-1.0078481-rdf-ntriples.txt
Original Record: 831-1.0078481-source.json
Full Text
831-1.0078481-fulltext.txt
Citation
831-1.0078481.ris

Full Text

COPPER ELECTROWINNING FROM CYANIDE SOLUTIONS by J i anming L u B . E n g . , Northeastern Un ive r s i t y , P . R . C h i n a , 1983 M . E n g . , Shanghai Un ive r s i t y , P . R . C h i n a , 1990 M . A . Sc . , Un ive r s i t y o f B r i t i s h C o l u m b i a , 1996 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y i n T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department o f M e t a l s and Mate r ia l s Eng inee r ing W e accept this thesis as confo rming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A October 1999 © J i anming L u , 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The objective o f this research was to explore an efficient process to recover cyanide and copper f rom barren go ld cyanide solut ion. The research w o r k described here concerns an invest igat ion into fundamental and pract ical aspects o f two options for e lec t rowinning copper f rom cyanide solut ion. These two options are: (a) the use o f an alternative anode react ion to • l imit the e lectro-oxidat ion o f cyanide i n concentrated cyanide solutions and (b) the use o f a graphite fibre cathode to e lec t rowin copper f rom dilute cyanide solut ion. (1) A cr i t i ca l literature survey was conducted to examine the stabil i ty constants o f copper cyanide species. The distributions and the equ i l i b r ium redox potentials o f copper cyanide species were calculated us ing the most rel iable stabili ty constants. T h e y are dependent on the mole ratio o f cyanide to copper, total cyanide concentrat ion, p H and temperature. Potent ia l measurements have conf i rmed the va l id i ty o f the calcula ted results. The pH-poten t ia l d iagram was d rawn using the G i b b s free energy data der ived by select ing the most rel iable stabili ty constants. (2) Di rec t copper e lec t rowinning f rom dilute cyanide solutions was conducted i n a membrane ce l l . The accumula t ion o f deposited copper on the graphite felt as the p la t ing proceeds s ignif icant ly improves the conduct iv i ty o f the graphite felt, increases the specif ic surface area and benefits copper deposit ion. Copper can be deposited on the graphite felt f rom l o w concentrat ion solutions (1-2 gL" 1 C u and C N : C u mo le ratio = 3-4) w i t h 50-80 % current eff ic iency, the r emova l o f around 4 0 % C u and an energy consumpt ion o f 1 -2 k W h / k g C u i n the superf ic ia l current density range 30 - 100 A m " 2 at 40 ° C . (3) Copper e lec t rowinning f rom concentrated copper cyanide solu t ion (70 g L " 1 C u ) was conducted us ing four sacr i f ic ia l species (sulphite, methanol , thiocyanate and ammonia) at 40 to 60 ° C . O n l y sulphite can decrease the anodic current eff ic iency o f cyanide ox ida t ion f rom ~ 100 to 10-20 % over the current density range o f 250-500 A m " 2 . W i t h increasing C N : C u mo le ratio f rom 3 to 4.5, the anodic current eff ic iency o f cyanide ox ida t ion increased and the copper deposi t ion current eff iciency decreased. A s regards the recovery o f copper f rom barren go ld cyanide solut ion, it has been shown that us ing sulphite ox ida t ion as an alternative anode reaction, copper can be e lect rowon from a cyanide electrolyte conta in ing I l l about 70 gL~' C u ( C N : C u = about 3) and 0.5 M N a 2 S 0 3 at a cathode current ef f ic iency o f about 9 5 % w i t h a energy consumpt ion o f about 0.8 k W h / k g C u at 250 A m " 2 . (4) In a lkal ine solutions, sulphite is o x i d i z e d to sulphate on the graphite anode i n a two-electron reaction. The reaction order w i t h respect to sulphite ions is b e l o w 1 at l o w potentials(< 0.4 V vs. S C E ) and 1 at h igh potentials. The react ion order for hydrox ide ions is close to zero. T w o Tafe l slopes were observed, 0.060 - 0.64 V decade"' at l o w potentials and 0.19-0.20 V decade" 1 at h igh potentials i n the temperature range 40 - 60 ° C . Sulphi te ox ida t ion i n a lkal ine solu t ion appears to undergo an electron-radical mechan ism. (5) The anodic ox ida t ion o f copper cyanide has been studied us ing a graphite rotat ing disk w i t h reference to cyanide concentration (0.05-4 M ) , C N : C u mo le ratio (3-12), temperature (25-60 ° C ) and hydroxide concentrat ion (0.01-0.25 M ) . Copper had a s ignif icant catalytic effect on cyanide oxidat ion . In the l o w polar iza t ion region (< about 0.4 V vs. S C E ) , cuprous cyanide is o x i d i z e d to cupr ic cyanide complexes w h i c h further react to f o r m cyanate. A t a C N : C u ratio o f 3 and [OH"] = 0.25 M , the Tafe l slope was about 0.12 V decade"'. C u ( C N ) 3 2 " was discharged o n the electrode surface. W i t h increasing C N : C u mo le ratio and decreasing p H , the dominant discharged species shifted to C u ( C N ) 4 3 " . In the h igh po la r iza t ion reg ion (about 0.4 -0.6 V vs. S C E ) , cuprous cyanide complexes were o x i d i z e d to copper ox ide and cyanate. W h e n the concentration o f cyanide was h igh and the p H l o w , cyanogen was formed, but no copper oxide . (6) Sulphite ox ida t ion is enhanced by the presence o f copper cyanide. The effect o f sulphite on l i m i t i n g the ox ida t ion o f copper cyanide decreases w i t h increasing mole ratio o f cyanide to copper. Th i s is related to the shift i n the discharged species f rom C u ( C N ) 3 2 " to C u ( C N ) 4 3 " w i t h increasing mole ratio o f cyanide to copper. Sulphi te is o x i d i z e d to sulphate. A t [Cu] = around 1 M , C N : C u = 3 -3.2, [OH"] = 0.05-0.25 M , [S0 3 2 "] = 0.4-0.6 M and the temperature = 50 - 60 ° C , the anode current eff ic iency o f sulphite ox ida t ion reached 80-90%) as the anodic current eff ic iency o f cyanide fe l l to 20 to 10 % . iv T A B L E O F C O N T E N T S Abstract . i i Tab le o f Contents i v L i s t o f Tables i x L i s t o f F igures x i A c k n o w l e d g m e n t s X X J S Nomenc la tu re xxv : 1. Introduct ion 1 2. Literature R e v i e w 8 2.1 A q u e o u s Chemis t ry o f the Copper -Cyan ide Sys tem 8 2.2 Depos i t i on o f Copper f rom Copper -Cyan ide So lu t ion 12 2.2.1 Pract ice o f Copper Depos i t ion f rom Cyan ide So lu t ion 12 2.2.2 Effect o f Parameters on Copper Depos i t ion 13 2.2.3 K i n e t i c s and M e c h a n i s m o f Copper Depos i t i on 15 2.3 E lec t rochemica l O x i d a t i o n o f Cyan ide 16 2.3.1 Cyan ide O x i d a t i o n i n A l k a l i n e So lu t ion 17 2.3.2 Cyan ide O x i d a t i o n i n W e a k l y A c i d i c , or A l k a l i n e or Neu t ra l Solut ions 17 2.3.3 A n o d i c O x i d a t i o n o f Copper Cyan ide 18 2.4 E lec t rochemica l O x i d a t i o n o f Thiocyanate 21 2.5 E lec t rochemica l O x i d a t i o n o f Sulphite 23 2.6 E lec t rochemica l O x i d a t i o n o f M e t h a n o l 26 2.7 E lec t rochemica l Ox ida t i on o f A m m o n i a 27 2.8 S u m m a r y 27 V 3. The rmodynamics o f Copper Cyan ide 30 3.1 Di s t r ibu t ion o f Copper Cyan ide Species 30 3.2 E q u i l i b r i u m Potent ial Measurement o f Copper Cyan ide 37 3.2.1 Exper imen ta l 37 3.2.2 Resul ts and Discuss ions 38 3.3 Po ten t i a l -pH Diagrams for Copper Cyan ide Sys tem 41 3.4 S u m m a r y 45 4. E lec t rodepos i t ion o f Copper on Graphite Fel t f rom Di lu t e Cyan ide Solut ions 46 4.1 Some Fundamenta l Aspects o f Graphite F ibre Electrodes 46 4.2 Exper imen ta l 50 4.2.1 E lec t ro ly t i c C e l l and Exper imenta l Set-up 50 4.2.2 Mate r ia l s 52 4.3 Resul ts and Discuss ions 52 4.4 S u m m a r y 61 5. E lec t ro w i n n i n g f rom Copper Cyan ide Solut ions U s i n g Al te rna t ive A n o d e 62 React ions 5.1 Exper imen ta l Apparatus and Set-up for E l ec t rowinn ing 62 5.2 Selec t ion o f Sacr i f i c ia l Mater ia l s 63 5.2.1 Thiocyanate 64 5.2.2 M e t h a n o l 65 5.2.3 A m m o n i a 65 5.2.3 Sulphite 66 5.3 Effect o f Some Parameters on the A n o d i c and Cathodic Processes i n the 67 Presence o f Sulphite vi 5.3.1 Effect o f Current Dens i ty 67 5.3.2 Effect o f Sulphite Concentra t ion 68 5.3.3 Effect o f Thiocyanate and M o l e Ra t io o f Cyan ide to Copper 68 5.3.4 Effect o f Temperature 73 5.4 Summary 74 6. A n o d i c O x i d a t i o n o f Sulphite on the Graphi te A n o d e i n A l k a l i n e So lu t ion 75 6.1 Some Fundamenta l Aspec ts o f Rota t ing D i s k Electrodes 75 6.2 Thermodynamics o f Sulphite O x i d a t i o n 79 6.3 Exper imen ta l Apparatus and Set-up 80 6.4 Po la r i za t ion Measurements 82 6.5 Cou lome t r i c Measurements 88 6.6 Reac t ion Order 89 6.7 Effect o f p H 95 6.8 Ca l cu l a t i on o f A c t i v a t i o n Energy for the K i n e t i c Current 97 6.9 D i f f u s i o n Coeff ic ient Es t ima t ion 97 6.10 Potent ia l Sweep Study 99 6.11 Poss ib le Reac t ion M e c h a n i s m 101 6.12 S u m m a r y 103 7. A n o d i c O x i d a t i o n o f Copper Cyan ide on a Graphi te A n o d e i n A l k a l i n e So lu t i on 104 7.1 Exper imen ta l Apparatus and Set-up 104 7.2 Po la r i za t ion Measurements and Identif ication o f the Precipitate 105 7.2.1 A n o d i c Behav iou r for Di lu te Copper Cyan ide So lu t ion 105 7.2.2 A n o d i c Behav io r o f Concentrated Copper Cyan ide So lu t ion 111 v i i 7.3 Cou lome t r i c Measurement 120 7.4 Effect o f C N : C u M o l e Rat io 122 7.5 Effect o f p H 129 7.6 Reac t ion Order 140 7.7 Reac t ion between Cyan ide and Copper(II) 144 7.8 C y c l i c V o l t a m m e t r y 147 7.9 Poss ib le Reac t ion M e c h a n i s m 148 7.10 D i f f u s i o n Coeff ic ient Es t ima t ion 154 7.11 A c t i v a t i o n Energy Ca lcu l a t i on for the K i n e t i c Current 156 7.12 S u m m a r y 157 8. A n o d i c O x i d a t i o n o f M i x e d Copper Cyan ide and Sulphite i n A l k a l i n e So lu t ion 158 8.1 Exper imenta l Apparatus and Set-up 158 8.2 A n o d i c Behav iou r o f M i x e d Sulphite and Copper Cyan ide So lu t ion 159 8.2.1 A n o d i c Behav iou r o f Di lu te Copper Cyan ide So lu t ion w i t h Sulphi te 159 8.2.2 A n o d i c Behav iou r o f Concentrated Copper Cyan ide So lu t ion w i t h 167 Sulphite 8.3 Cou lome t r i c Measurements 178 8.4 Poss ib le A n o d i c React ions 183 8.5 S u m m a r y 184 9. Conc lus ions 186 10. Recommendat ions 190 11. References 191 A p p e n d i x 1 Ini t ia l E c o n o m i c Assessment 211 A p p e n d i x 2 To ta l Cyan ide A n a l y s i s 219 v i i i A p p e n d i x 3 Copper Ti t ra t ion U s i n g E D T A 227 A p p e n d i x 4 Determina t ion o f Sulphite Ions by the Iodimetr ic M e t h o d 229 A p p e n d i x 5 Ca l cu l a t i on o f A c t i v i t y Coeff ic ient U s i n g P i tze r ' s M e t h o d 232 A p p e n d i x 6 Measurement o f the K i n e m a t i c V i s c o s i t y 235 A p p e n d i x 7 Ca l cu l a t i on o f L i q u i d Junct ion Potent ial 237 A p p e n d i x 8 Figures 239 LIST of TABLES ix Table 2-1 A s s o c i a t i o n constants for copper cyanide complexes 8 Tab le 2-2 Copper cyanide bath composi t ions and condit ions 13 Table 3-1 E q u i l i b r i u m constants for copper cyanide system 30 Tab le 3-2 G i b b s free energy data for copper and cyanide species 41 Tab le 4-1 Conduct iv i t i es o f copper cyanide solutions w i t h different cyanide 53 concentrations at f ixed copper concentrat ion Table 4-2 Copper cathodic current eff ic iency and power consumpt ion at 40 ° C and 54 in i t i a l [Cu] = 1 g L " 1 for experiments w i t h oxygen evo lu t ion at an anode Table 4-3 Copper cathodic current eff iciency and power consumpt ion at 40 ° C and 54 in i t i a l [Cu] = 2 g L " 1 for experiments w i t h oxygen evolu t ion at an anode Table 4-4 Di s t r ibu t ion and potentials o f copper cyanide solut ion at [OH"] = 0.01 M 55 at 40 ° C Table 4-5 Resul ts o f cyc le run at 40 ° C (an in i t ia l C N : C u ratio o f 3) 60 Table 5-1 Resul ts for the selection o f sacr i f ic ia l species at 60 ° C 64 Table 5-2 Effect o f current density on the anodic current eff ic iency o f cyanide and 68 the cathodic current eff iciency o f copper at 60 °C. Elec t ro ly te : 70 g L " 1 C u , C N : C u mole ratio = 3, 10 g L " 1 N a O H and 113 g L " 1 N a 2 S 0 3 Table 5-3 Effect o f sulphite concentration on the anodic current eff ic iency o f 68 cyanide and the cathodic current eff ic iency o f copper at 60 C and 250 A m" 2 . E lec t ro ly te : 70 g L 1 C u , C N : C u mole ratio = 3, 10 g L " 1 N a O H Table 5-4 Resul ts o f copper e lec t rowinning at 250 A m" 2 and 60 ° C . Elec t ro ly te : 70 69 g L " 1 C u , C N : C u mole ratio = 3-4.5, 63 g L 1 N a 2 S 0 3 and 10 g L " 1 N a O H i n the presence and absence o f S C N " Tab le 5-5 Resul ts o f copper e lec t rowinning at 250 A m" 2 and different 74 temperatures. Electrolytes : 70 g L " 1 C u , C N : C u mo le ratio = 3, 63 g L " 1 N a 2 S 0 3 and 10 g L " 1 N a O H i n the presence and absence o f S C N " Table 6-1 The activit ies and act ivi ty coefficients for 0.1 M N a 2 S 0 3 , 0.25 M N a O H , 80 1 M N a 2 S 0 4 at 25, 40, 50 and 60 °C X Table 6-2 N u m b e r o f the electrons transferred for the anodic ox ida t ion o f sulphite 88 Table 6-3 Reac t ion order and the kinet ic current calculated us ing different 90 methods for 0.1 M N a , S 0 3 Table 6-4 Reac t ion order and the kinet ic current calculated us ing different 90 methods for 0.4 M N a 2 S 0 3 Table 6-5 Tafe l slopes ( V decade" 1) for the different potential ranges at 25, 40 , 50 94 and 60 ° C Table 7-1 A m o u n t o f cyanide and copper (I) o x i d i z e d per Faraday at 100 r p m and 121 different C N : C u mole ratios and hydroxide concentrations Table 7-2 A m o u n t o f cyanide and copper (I) o x i d i z e d per Faraday at 400 A m" 2 , 122 100 rpm, different C N : C u mole ratios and hydroxide concentrations Table 8-1 Current efficiencies f rom copper cyanide us ing contro l led potential 179 coulometr ic measurements Table 8-2 Current efficiencies f rom copper cyanide us ing control led current 180 coulometr ic measurements Tab le 8-3 Current efficiencies f rom copper cyanide us ing control led current 181 coulometr ic measurements at 100 r p m Table 8-4 Current eff ic iency for copper cyanide and sulphite us ing cont ro l led 182 current coulometr ic measurements at 100 r p m Table 8-5 Current eff ic iency for copper cyanide and sulphite us ing cont ro l led 183 potential coulometr ic measurements at 100 r p m xi LIST OF FIGURES Figure 1-1 F lowshee t for solvent extraction - e lec t rowinning process for the 6 recovery o f copper cyanide F igure 1-2 F lowshee t for direct e lec t rowinning o f copper f rom cyanide solutions 7 F igure 3-1 Copper cyanide species dis tr ibut ion and E ( C u ( I ) / C u ) vs. mole ratio o f 32 cyanide to copper for various solutions at 25 ° C and p H 9 Figure 3-2 Copper cyanide species dis tr ibut ion and E ( C u ( I ) / C u ) vs . mo le ratio o f 33 cyanide to copper for various solutions at 25 ° C and p H 12 Figure 3-3 Copper cyanide species dis t r ibut ion and E ( C u ( I ) / C u ) vs . mo le ratio o f 34 cyanide to copper for various solutions at 60 ° C and p H 12 Figure 3-4 (a) E ( C u ( I ) / C u ) vs. mole ratio o f cyanide to copper at 25 ° C , p H 12 35 and different C u concentrations and (b) E ( C u ( I ) / C u ) vs. p H at 25 ° C , 0.1 M C u and different mole ratios o f cyanide to copper F igure 3-5 Copper concentrations i n the fo rm o f copper complexes and the 36 equ i l i b r i um potential vs. total copper concentrat ion at [CN"] = 2.455 g L " ] and [OH"] = 0.01 M Figure 3-6 Copper concentrations i n the form o f copper complexes and the 36 equ i l i b r ium potential vs . total copper concentrat ion at [ C N ] = 1.227 g L " 1 and [ O H ] = 0.01 M Figure 3-7 Exper imen ta l set-up for the equ i l i b r ium potential measurement 38 F igure 3-8 Elec t rode potential vs. t ime at 25 °C, C N : C u = 3 and [ C u ] t o t a l = 0.1 M 39 F igure 3-9 Elec t rode potential vs. the mole ratio o f cyanide to copper at 25 , 40, 40 50 and 60 ° C , [ C u ] t o t a l = 0.1 M and [OH"] = 0.01 M F igure 3-10 Elec t rode potential vs . the mo le ratio o f cyanide to copper at 25, 40 , 40 50 and 60 ° C , [ C u ] t o t a l = 0.01 M and [OH"] = 0.01 M Figure 3-11 C N - H 2 0 poten t ia l -pH diagram at a l l solute species act ivi t ies o f 1 and 42 P ( C N ) 2 = 1 atm and 25 ° C . (a) assuming H C N O and C N O are stable and (b) assuming ( C N ) 2 is stable F igure 3-12 Po ten t i a l -pH diagrams for C u - C N - H 2 0 system at 25 ° C and the 43 act ivi t ies o f a l l solute species = 1, 10"2, 10"4 and 10"6 cons ider ing C u O as a stable species. H C N O , C N O " and ( C N ) 2 are not considered Figure 3-13 Po ten t i a l -pH diagrams for C u - C N - H 2 0 system at 25 ° C and the 44 act ivi t ies o f a l l solute species = 1, 10"2, 10"4 and 10"6 cons ider ing C u ( O H ) 2 as a stable species. H C N O , C N O " and ( C N ) 2 are not considered F igure 3-14 Po ten t i a l -pH diagram for C u - C N - H 2 0 system at 25 ° C and solute 45 copper species activit ies o f 0.01 and cyanide species" act ivi t ies o f 0.1 consider ing C u ( O H ) 2 as a stable species. H C N O , C N O " and ( C N ) 2 are not considered F igure 4-1 Schematic diagram o f porous electrode 47 F igure 4-2 Schematic d iagram o f electrolyt ic c e l l 51 F igure 4-3 Schematic d iagram o f experimental set-up 51 F igure 4-4 Current eff ic iency and the power consumpt ion o f copper depos i t ion 56 vs. the mole ratio o f cyanide to copper at different cathodic current densities and 40 ° C . The electrolyte: (a) 1 g L " ' C u , 0.01 M N a O H and 0.00862 M N a S C N , and (b): 2 g L " 1 C u , 0.01 M N a O H and 0.01724 M N a S C N . The f l o w ve loc i ty : 2.97, 5.93 and 9.83 c m min." 1 respect ively for 30, 60 and 100 A m " 2 . F igure 4-5 C o n v e r s i o n o f Cu(I) to C u vs. the mole ratio o f cyanide to copper at 56 different cathodic current densities and 40 ° C . The electrolyte: (a) 1 g L " 1 C u , 0.01 M N a O H and 0.00862 M N a S C N , and (b): 2 g L " 1 C u , 0.01 M N a O H and 0.01724 M N a S C N . The f l o w ve loc i ty : 2.97, 5.93 and 9.83 c m m i n . ' 1 respectively for 30, 60 and 100 A m" 2 . F igure 4-6 C e l l voltage vs. t ime at the cathodic current density = 30 A m" 2 and 40 57 ° C . The electrolyte: l g L " 1 C u , C N : C u = 3, 0.01 M N a O H and 0.00862 M N a S C N and the f l o w ve loc i ty : 2.97 c m min." 1 . F igure 4-7 C e l l voltage vs. the mole ratio o f cyanide to copper at different 58 cathodic current densities and 40 ° C . The electrolyte: 2 g L " 1 C u , 0.01 M N a O H and 0.00862 M N a S C N , the f l o w ve loc i ty : 2.97, 5.93 and 9.83 c m min." 1 respect ively for 30, 60 and 100 A m" 2 . F igure 4-8 Graphi te fibre felt on w h i c h copper has been deposited. 59 F igure 4-9 Cross-sec t ion o f the graphite fibre felt o n w h i c h copper has been 59 deposited. F igure 4-10 Concentra t ion o f copper vs. the number o f the solu t ion passes through 61 the graphite felt at [ C u ] i n i t a l = 1 and 2 g L " 1 and 40 ° C . The electrolyte: xiii (1) l g L " 1 C u , C N : C u = 3, 0.01 M N a O H and 0.00862 M N a S C N and (2) 2 g L " 1 C u , C N : C u =3, 0.01 M N a O H and 0.01724 M N a S C N , and the f l o w ve loc i ty : 2.97 c m min." ' . F igure 5-1 Schematic diagram o f the experimental set-up 63 F igure 5-2 Concentra t ion o f cyanide vs. the electrolysis t ime for obta ining the 63 current eff ic iency o f cyanide ox ida t ion at 60 ° C . Elec t ro ly te : 70 g L " ' C u , C N : C u = 3, 113 g L " ' N a 2 S 0 3 , 10 g L " ' N a O H Figure 5-3 C e l l voltage vs. the t ime o f electrolysis i n the presence o f a m m o n i a 67 and sulphite as a sacr i f ic ia l species at 500 A m" 2 and 60 ° C . Elec t ro ly te : 70 g L " ' C u , C N : C u = 3, and 10 g L " ' F igure 5-4 Ca thodic current eff ic iency o f copper deposi t ion and power 71 consumpt ion vs. the mole ratio o f cyanide to copper at 60 ° C and 250 A m" 2 . E lec t ro ly te :70 g L " ' C u , 63 g L " 1 N a 2 S 0 3 , 10 g L " ' N a O H , and different cyanide concentrations i n the presence and absence o f 40 g L " 1 S C N " 1 Figure 5-5 A n o d i c current eff ic iency for cyanide ox ida t ion vs. the mo le ratio o f 71 cyanide to copper at 250 A m" 2 and 60 °C. Elec t ro ly te :70 g L " ' C u , 63 g L " 1 N a 2 S 0 3 , 10 g L " 1 N a O H , and different cyanide concentrations i n the presence and absence o f 40 g L " 1 S C N " 1 Figure 5-6 C e l l voltage vs. t ime o f electrolysis at 250 A m" 2 and 60 ° C . 73 Elec t ro ly te :70 g L " 1 C u , 63 g L - l N a 2 S 0 3 , 10 g L " 1 N a O H , and different cyanide concentrations i n the absence o f S C N " 1 Figure 6-1 Rota t ing disk coordinate system used i n calculat ions o f l i q u i d f l o w 76 near the rotating disk F igure 6-2 Schemat ic d iagram o f rotating disk 81 F igure 6-3 Schematic diagram o f the experimental set-up 82 F igure 6-4 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25, 40 , 84 50 and 60 ° C . Elec t ro ly te : 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 6-5 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating disk at 25 , 40 , 87 50 and 60 ° C . Elec t ro ly te : 0.1 M N a 2 S 0 3 , 0.05 M N a O H and 1 M N a 2 S 0 4 Figure 6-6 C o m p a r i s o n o f the polar iza t ion curves w i t h different sulphite and 88 hydrox ide concentrations at 25 °C and 400 r p m xiv Figure 6-7 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25 ° C . 90 Elec t ro ly te : 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 6-8 L o g i vs . L o g ( l - i / i , ) at constant potentials and 25 ° C . Elec t ro ly te : 0.1 91 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H Figure 6-9 1/i vs . 1/i, at constant potentials ( V vs. S C E ) 25 ° C . Elec t ro ly te : 0.1 M 91 N a 2 S 0 3 , I M N a 2 S 0 4 and 0.25 M N a O H Figure 6-10 L o g i vs . L o g ( l - i / i , ) at constant potentials and 25 ° C . Elec t ro ly te : 0.4 92 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H Figure 6-11 1/i vs . 1/i, at constant potentials ( V vs. S C E ) and 25 ° C . Elec t ro ly te : 92 0.4 M N a 2 S 0 3 , I M N a 2 S 0 4 and 0.25 M N a O H Figure 6-12 L o g i vs . l og [SO, 2"] at 25 ° C and 4900 rpm. Elec t ro ly te : I M N a 2 S 0 4 93 and 0.25 M N a O H Figure 6-13 Potent ia l vs . l og ( ( i / ( l - i / i , ) ) at different temperatures. E lec t ro ly te : 0.1 95 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H Figure 6-14 Po la r i za t ion curves at different hydrox ide concentrations and 25 ° C . 96 Elec t ro ly te : 0.1 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure 6-15 Effect o f p H on sulphite ox ida t ion at different potentials and 25 ° C . 96 Elec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 at variable p H Figure 6-16 L o g i k v s . l / T at different potentials ( V vs. S C E ) . E lec t ro ly te : 0.1 M 97 N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H Figure 6-17 D i f f u s i o n current density vs. the square root o f rotational speed at 98 different temperatures. Elec t ro ly te : 0.05 M N a 2 S 0 3 , l M N a 2 S 0 4 , 0.25 M N a O H Figure 6-18 L o g plot o f diffusion coefficient vs. 1/T 99 F igure 6-19 V o l t a m m o g r a m s at different scan rates at 25 ° C . Elec t ro ly te : 0.1 M 100 N a 2 S 0 3 , 1 M N a 2 S 0 4 , 0.25 M N a O H Figure 6-20 Peak current vs. potential scan rate at 25 ° C . Elec t ro ly te : 0.1 M 101 N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H Figure 7-1 Schematic d iagram for detection o f cupric cyanide species 105 Figure 7-2 Po la r i za t ion curves at different rotational speeds and temperatures. 109 XV Elec t ro ly te : 0.05 M C N " , C N : C u mo le ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-3 C y c l i c vol tammetry at 25 and 40 ° C . Elec t ro ly te : 0.05 M C N " , C N : C u 110 mole ratio = 3, 0.25 M N a 2 S 0 4 and 1 M N a 2 S 0 4 Figure 7-4 Po la r i za t ion curves at different rotational speeds and temperatures. 114 Elec t ro ly te : 3 M C N " , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-5 Po la r i za t ion curves at different rotational speeds and temperatures. 115 Elec t ro ly te : 3.5 M C N " , C N : C u mole ratio = 3.5, 0.25 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-6 Po la r i za t ion curves at different rotational speeds and temperatures. 116 Elec t ro ly te : 4 M C N " , C N : C u mole ratio = 4, 0.25 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-7 Po la r iza t ion curves at different rotational speeds and temperatures. 117 Elec t ro ly te : 3 M C N " , C N : C u mole ratio = 3, 0.05 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-8 Po la r i za t ion curves at different rotational speeds and temperatures. 118 Elec t ro ly te : 4 M C N " , C N : C u mole ratio = 1, 0.05 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-9 Po la r iza t ion curves at different rotational speeds and temperatures. 119 Elec t ro ly te : 4 M C N " , C N : C u mole ratio = 1, 0.50 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-10 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - 125 current vs. potential o n a graphite rotating d i sk at 4900 r p m and different temperatures. Electrolytes : 0.05 M C N " , C N : C u mo le ratio = 3, 4, 6, 12 and no copper, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-11 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - 126 potential vs . l og (current density) o n a graphite rotating d i sk at 4900 r p m (25 and 6 0 ° C ) . Electrolytes : 0.05 M C N " , C N : C u mo le ratio = 3, 4, 6, 12 and no copper, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-12 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - 126 potential vs . l og current density o n a py ro ly t i c graphite rotating electrode at 4900 rpm and 25 ° C . Electrolytes : 0.05 M C N " , C N : C u mole ratio = 3, 4, 6, 12 and 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-13 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - 127 x v i potential vs. l og (current density) on a graphite rotating disk at 4900 r p m (25 and 60 ° C ) . Electrolytes : [Cu + ] = 0.00833, [CN"] = 0.025, 0.05, 0.1, 0.2 and 0.4 M , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-14 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - 127 potential vs . l og (current density) on a py ro ly t i c graphite rotating d i sk at 4900 r p m and 25 ° C . Electrolytes : [Cu + ] = 0.00833 M , [CN"] = 0.025, 0.05, 0.1, 0.2 and 0.4 M , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-15 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - 128 current vs. potential o n a graphite rotating disk at 4900 r p m and 60 ° C . Elect rolytes : 1 M C u + , [CN"] = 3, 3.5 and 4, 0.25 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-16 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - 128 potential vs . l og (current density) on a graphite rotating d i sk at 4900 r p m and 60 ° C . Electrolytes : 1 M C u + , [CN"] = 3, 3.5 and 4, 0.25 M N a O H and 0.5 M N a 2 S 0 4 Figure 7-17 Effect o f p H on cyanide ox ida t ion - current vs. potential o n a graphite 131 rotating disk at 4900 r p m and different temperatures. Elec t ro ly tes : 0.05 M C N " , C N : C u mo le ratio = 3, [OH"] = 0.25, 0.05 and 0 . 0 1 M and 1 M N a 2 S 0 4 Figure 7-18 Effect o f p H on cyanide oxida t ion - potential vs . l og (current density) 132 o n a graphite rotating disk at 4900 r p m (25 and 6 0 ° C ) . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 3, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 I M and 1 M N a 2 S 0 4 Figure 7-19 Effect o f o f p H o n cyanide oxida t ion - potential vs . l og (current 132 density) on a pyro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 3, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 1 M and 1 M N a 2 S 0 4 Figure 7-20 Effect o f p H on cyanide ox ida t ion - current vs. potential o n a graphite 133 rotating disk at 4900 r p m and different temperatures. Elec t ro ly tes : 0.05 M C N " , C N : C u mo le ratio = 4, [OH"] = 0.25, 0.05, and 0 . 0 I M and 1 M N a 2 S 0 4 Figure 7-21 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) 134 on a graphite rotating disk 4900 r p m (25 and 60 ° C ) . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 4, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a 2 S 0 4 Figure 7-22 Effect o f p H o n cyanide ox ida t ion - potential vs . l og (current density) 134 o n a py ro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . XVII Electrolytes : 0.05 M C N " , C N : C u mole ratio = 4, [OH"] = 0.25, 0.05, and 0 . 0 I M and 1 M N a 2 S 0 4 Figure 7-23 Effect o f p H on cyanide ox ida t ion - current vs. potential o n a graphite 135 rotating disk at 4900 r p m and different temperatures. Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 12, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a 2 S 0 4 Figure 7-24 Effect o f p H on cyanide ox ida t ion - potential vs . l o g (current density) 136 o n a graphite rotating disk at 4900 r p m (25 and 6 0 ° C ) . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 12, [OH"] = 0.25, 0.05 and 0 . 0 1 M and 1 M N a ^ O , F igure 7-25 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) 136 o n a pyro ly t i c graphite rotating disk at 4900 rpm and 25 ° C . Elect rolytes : 0.05 M C N " , C N : C u mole ratio = 12, [OH"] = 0.25, 0.125, 0.05, 0.025 and O . O l M a n d 1 M N a 2 S 0 4 Figure 7-26 Effect o f p H o n cyanide oxida t ion - the plot o f the current vs. the 137 potential on a graphite rotating disk at 4900 r p m and different temperatures. Electrolytes : 3 M C N " , C N : C u mole ratio = 3, [OH"] = 0.5, 0.25 and 0.05 M and 0.5 M N a 2 S 0 4 Figure 7-27 Effect o f p H on cyanide ox ida t ion - potential vs . l o g (current density) 138 o n a graphite rotating disk at 4900 r p m and 60 ° C . Elec t ro ly tes : 3 M C N , C N : C u mole ratio = 3, [ O H ] = 0.5, 0.25, and 0.05 M and 0.6 M N a 2 S 0 4 Figure 7-28 Effect o f p H o n cyanide ox ida t ion - current vs. potential on a graphite 139 rotating disk at 4900 r p m and different temperatures. Elec t ro ly tes : 4 M C N " , C N : C u mole ratio = 4, [OH"] = 0.5 and 0.25 and 0.05 M and 0.5 M N a 2 S 0 4 Figure 7-29 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) 139 o n a graphite rotating disk at 60 ° C . Elect rolytes : 4 M C N , C N : C u mo le ratio = 4, [OH"] = 0.50, 0.25 and 0.05 M and 0.5 M N a 2 S 0 4 Figure 7-30 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) o n a graphite 141 rotating disk at 4900 r p m (25 and 6 0 ° C ) . Elect rolytes : [CN"] = 0.025, 0.05, 0.1 and 0.20 M , C N : C u mole ratio = 3, [OH"] = 0.25 M and 1 M N a 2 S 0 4 Figure 7-31 Plots o f l o g (current density) vs. l og ( [Cu(CN) 4 2 " ] ) on a graphite 141 rotating disk at 4900 r p m and 25 ° C . Electrolytes : [CN"] = 0.05, 0.1, 0.20 and 0.40 M , [Cu + ] = 0.00833 M , [OH"] = 0.25 M and 1 M N a 2 S 0 4 XV l l l Figure 7-32 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) o n a py ro ly t i c 142 graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : [CN"] = 0.05, 0.10, 0.20 and 0.40 M , [Cu + ] = 0.0833 M , [OH"] = 0.25 M and 1 M N a 2 S 0 4 Figure 7-33 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) on a graphite 143 rotating disk at 4900 r p m and 25 ° C . Elect rolytes : [CN"] = 0.40 M , [Cu + ] = 0.0167, 0.00833, 0.00417, 0.00208, 0.00104 M , [OH"] = 0.25 M and 1 M N a 2 S 0 4 Figure 7-34 Absorbance vs. t ime when 2.5 c m 3 o f 0.05 M cyanide so lu t ion w i t h 144 0.25 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solu t ion at 25 ° C F igure 7-35 Absorbance vs. t ime when 2.5 c m 3 o f 0.05 M cyanide solu t ion w i t h 145 0.05 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solu t ion at 25 ° C F igure 7-36 Absorbance vs. t ime when 2.5 c m 3 o f 1 M cyanide solut ion w i t h 0.25 145 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate so lu t ion at 25 ° C F igure 7-37 The plot o f (Absorbance)" ' vs. t ime when 2.5 c m 3 o f 1 M cyanide 146 solut ion w i t h 0.25 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solut ion at 25 °C F igure 7-38 C y c l i c vol tammetry at 25 ° C . Elec t ro ly te : 0.025 M C N " , C N : C u mo le 148 ratio =3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-39 Plots o f potential vs. l og (current density) us ing data measured and 154 predicted us ing Equa t ion 7-12 at 25 °C. Elec t ro ly te : 0.1 M C N " , C N : C u mole ratio = 12, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-40 L i m i t i n g current vs. rotational speed at 40 , 50 and 60 ° C . E lec t ro ly te : 155 0.05 M C N " , C N : C u mole ratio= 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 7-41 A c t i v a t i o n energy ca lcu la t ion -p lo t o f l og (current density) vs . 1 /Tat 156 constant potentials. Elec t ro ly te : 0.05 M C N " , C N : C u mo le ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 8-1 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 0.05 M C N " , 162 0.0167 M C u + ( C N : C u mole ratio = 3), 0.25 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure 8-2 Effect o f potential scanning rate on the anodic behaviour o f m i x e d 163 sulphite and copper cyanide at 4900 r p m and 60 °C. Elec t ro ly te : 0.05 XIX M C N " , 0.0167 M C u + ( C N : C u mole ratio = 3), 0.25 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure 8-3 Po la r i za t ion curves for for (1) 0.05 M C N " , 0.0167 M C u + and 0.4 M 163 N a 2 S 0 3 , (2) the same compos i t ion as (1), o n the electrode coated w i t h copper oxide at 0.5 V vs. S C E for 10 minutes i n the same solut ion. (3) 0.4 M N a 2 S 0 3 , (4) 0.4 M N a 2 S 0 3 o n the electrode coated w i t h copper ox ide i n the same solut ion as (1), and (5) 0.4 M N a 2 S 0 3 o n the electrode coated w i t h copper oxide f rom 0.05 M C N " and 0.0167 M C u + . Support ing electrolyte: 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 8-4 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0167 M C u + ( C N : C u mo le 164 ratio = 3) and 0.4 M N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 and (3) 0.05 M C N " and 0.0167 M C u + at 400 r p m and 60 °C. Suppor t ing electrolyte: 0.25 M N a O H and 1 M N a 2 S 0 4 Figure 8-5 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 165 0.0167 M C u + ( C N : C u mole ratio = 3), 0.05 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure 8-6 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0167 M C u + ( C N : C u mole 166 ratio = 3) and 0.4 M N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 and (3) 0.05 M C N " and 0.0167 M C u + at 400 r p m and 60 °C. Support ing electrolyte: 0.05 M N a O H and 1 M N a 2 S 0 4 Figure 8-7 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0125 M C u + and 0.4 M 166 N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 , (3) 0.05 M C N " , 0.0125 M C u + and 0.2 M N a 2 S 0 3 , (4) 0.2 M N a 2 S 0 3 and (5) 0.05 M C N " and 0.0125 M C u + at 400 r p m and 60 ° C . Support ing electrolyte: 0.05 M N a O H and 1 M N a 2 S 0 4 . F igure 8-8 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 3 M C N " , 1 170 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 Figure 8-9 Current density vs. t ime at constant potentials, 400 r p m and different 171 temperatures. Elec t ro ly te : 3 M C N " , 1 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 Figure 8-10 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 3 M C N " , 1 172 M C u \ 0.1 M N a O H , 0.5 M N a 2 S O Figure 8-11 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 3 M C N " , 1 173 M C u + , 0.05 M N a O H , 0.5 M N a 2 S 0 3 Figure 8-12 Po la r i za t ion curves for (1) 3 M C N " , 1 M C u + ( C N : C u mole ratio = 3), 174 0.25 M N a O H and 0.5 M N a 2 S 0 3 , (2) 0.5 M N a 2 S 0 3 , 0.25 M N a O H XX and 1 M N a 2 S 0 4 and (3) 3 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 4 at 400 r p m and 60 ° C Figure 8-13 Po la r i za t ion curves at 400 r p m and 25 ° C for (1) 3 M C N " + 1 M C u + + 174 0.4 M N a 2 S 0 3 + 0.1 M N a 2 S 0 4 (2) 3 M C N " + 1 M C u + + 0.2 M N a 2 S 0 3 + 0.3 M N a , S 0 4 , (3) 0.4 M N a 2 S 0 3 + 1 M N a ^ O , , (4) 0.2 M N a 2 S 0 3 + 1 M N a j S C ^ (5) 3 M C N " + 1 M C u + + 0.5 M N a 2 S 0 4 at [ N a O H ] = 0.05 M N a O H Figure 8-14 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 3.5 M C N " , 175 1 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 Figure 8-15 Po la r i za t ion curves for (1) 3.5 M C N " , 1 M C u + ( C N : C u mo le ratio = 176 3), 0.25 M N a O H and 0.5 M N a 2 S 0 3 , (2) 3 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 4 and (3) 0.5 M N a 2 S 0 3 , 0.25 M N a O H at 400 r p m and 60 ° C F igure 8-16 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 4 M C N " , 1 177 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 Figure 8-17 Po la r i za t ion curves for (1)4 M C N " , 1 M C u + ( C N : C u mo le ratio = 3), 178 0.25 M N a O H and 0.5 M N a 2 S 0 3 , (2) 0.5 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 and (3) 4 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 4 at 400 r p m and 60 ° C F igure A - l Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25, 40 , 239 50 and 60 ° C . Elec t ro ly te : 0.05 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 2 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25, 40 , 240 50 and 60 ° C . Elec t ro ly te : 0.2 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 3 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25, 40 , 241 50 and 60 ° C . Elec t ro ly te : 0.4 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 4 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating disk at 25 , 40 , 242 50 and 60 ° C . Elec t ro ly te : 0.5 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 5 B a c k g r o u n d current density vs. potential o n graphite rotating d i sk at 243 25, 40, 50 and 60 ° C . Elec t ro ly te : 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 6 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating disk at 25 , 40 , 244 50 and 60 ° C . Elec t ro ly te : 0.2 M N a 2 S 0 3 , 0.05 M N a O H and 1 M XXI N a 2 S 0 4 Figure A - 7 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating disk at 25 , 40, 245 50 and 60 ° C . Elec t ro ly te : 0.4 M N a 2 S 0 3 , 0.05 M N a O H and 1 M N a 2 S 0 4 Figure A - 8 L o g (i) vs . L o g (1-1/i,) (a) and 1/i vs . 1/i, (b) at 40 (1), 50 (2) and 60 246 (3) ° C and the corresponding fitted function (y vs. x) are i n the d iagram. Elect royte : 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 . F igure A - 9 C o m p a r i s o n o f the effects o f CuO-coa t ed graphite and copper ions i n 247 the solu t ion at 100 r p m and different temperatures. Elect rolyte :0 .25 M N a O H and 1 M N a 2 S 0 4 Figure A - 1 0 X P S spectrum o f the precipitate prepared at 25 ° C and 0.5 V vs . S C E . 248 Elec t ro ly te : 0.05 M N a C N , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - l 1 X P S spectrum o f the precipitate prepared at 60 ° C and 0.5 V vs. S C E . 248 Elec t ro ly te : 0.05 M N a C N , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 1 2 X P S spectrum o f the precipitate prepared at 25 ° C and 0.5 V vs. S C E . 249 Elec t ro ly te : 0.05 M N a C N , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 1 3 X P S spectrum o f the precipitate prepared at 60 ° C and 0.5 V vs . S C E . 249 Elec t ro ly te : 0.05 M N a C N , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 1 4 Po la r i za t ion curves on the graphite coated w i t h C u O and no C u O i n 250 the absence o f cyanide and copper at different temperatures. Elec t ro ly te : 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 1 5 Po la r i za t ion curves at different rotational speeds and temperatures. 251 Elec t ro ly te : 0.05 M C N ' , C N : C u mole ratio = 3.5, 0.25 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper ox ide and 3 - evolu t ion o f oxygen F igure A - 1 6 Po la r i za t ion curves at different rotational speeds and temperatures. 252 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 3.5, 0.25 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper oxide , 2 - prec ipi ta t ion o f copper oxide and 3 - evo lu t ion o f oxygen F igure A - 1 7 Po la r i za t ion curves at different rotational speeds and temperatures. 253 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 6, 0.25 M N a O H and 1 xxn M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper oxide , 3 - evo lu t ion o f oxygen and 2+3 - copper ox ide and o x y g e n appeared almost at the same potential F igure A - 1 8 Po la r i za t ion curves at different rotational speeds and temperatures. 254 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 12, 0.25 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper ox ide , 3 - evolu t ion o f oxygen and 2+3 - copper ox ide and o x y g e n appeared almost at the same potential F igure A - 1 9 Po la r i za t ion curves at different rotational speeds and temperatures. 255 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 3, 0.05 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper oxide and 3 - evolu t ion o f oxygen F igure A - 2 0 Po la r i za t ion curves at different rotational speeds and temperatures. 256 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 4, 0.05 M - N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper oxide and 3 - evolu t ion o f oxygen F igure A - 2 1 Po la r i za t ion curves at different rotational speeds and temperatures. 257 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 12, 0.05 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - precipi ta t ion o f copper oxide and 3 - evolu t ion o f oxygen F igure A - 2 2 Po la r i za t ion curves at different rotational speeds and temperatures. 258 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 3, 0.01 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper ox ide and 3 - evolu t ion o f oxygen Figure A - 2 3 Po la r i za t ion curves at different rotational speeds and temperatures. 259 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 4, 0.01 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper oxide , 2 - prec ipi ta t ion o f copper oxide and 3 - evo lu t ion o f oxygen F igure A - 2 4 Po la r i za t ion curves at different rotational speeds and temperatures. 260 Elec t ro ly te : 0.05 M C N " , C N : C u mole ratio = 12, 0.01 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi tat ion o f copper ox ide , 2 - precipi ta t ion o f copper ox ide and 3 - evolu t ion o f oxygen F igure A - 2 5 Po la r i za t ion curves at different rotational speeds and temperatures. 261 Elec t ro ly te : 0.5 M C N " , C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no precipi ta t ion o f copper ox ide , 2 - prec ipi ta t ion o f copper ox ide and 3 - evo lu t ion o f o x y g e n F igure A - 2 6 X - r a y diffract ion pattern o f the anodic precipitate prepared under the 262 condi t ions: 3 M C N " , 1 M C u (I), 0.25 M N a O H , 0.5 M N a 2 S 0 4 , 25 XX111 ° C , 0.5 V vs. S C E , and 100 r p m Figure A - 2 7 X - r a y diffract ion pattern o f the anodic precipitate prepared under the 262 condi t ions: 3 M C N " , 1 M C u (I), 0.25 M N a O H , 0.5 M N a 2 S 0 4 , 60 ° C , 0.5 V vs. S C E , and 100 r p m Figure A - 2 8 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - 263 potential vs . log (current density) on a graphite rotating d i sk at 4900 r p m (40 and 50 ° C ) . Electrolytes : 0.05 M C N " , C N : C u mo le ratio - 3, 4, 6, 12 and no copper, 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 2 9 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - 263 potential vs. l og (current density) on a graphite rotating disk at 4900 r p m (40 and 50 ° C ) . Electrolytes : [Cu + ] = 0.00833 M , [ C N ] = 0.025, 0.05, 0.1, 0.2 and 0.4 M , 0.25 M N a O H and 1 M N a 2 S 0 4 Figure A - 3 0 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) 264 o n a graphite rotating disk at 4900 r p m (40 and 5 0 ° C ) . Elec t ro ly tes : 0.05 M " C N " , a C N : C u mole ratio o f 3, [ O H ] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 I M and 1 M N a 2 S 0 4 Figure A - 3 1 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) 264 o n a Pt graphite rotating disk at 4900 r p m and 25 ° C . Elec t ro ly tes : 0.05 M " C N " , a C N : C u mole ratio o f 3, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 I M and 1 M N a 2 S 0 4 Figure A - 3 2 Effect o f p H on cyanide oxida t ion - potential vs . l og (current density) 265 o n a graphite rotating disk 4900 r p m (40 and 50 ° C ) . Elec t ro ly tes : 0.05 M " C N " , a C N : C u mole ratio o f 4, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a 2 S 0 4 Figure A - 3 3 Effect o f p H o n cyanide ox ida t ion - potential vs . l og (current density) 265 on a graphite rotating disk at 4900 r p m (40 and 50 ° C ) . Elec t ro ly tes : 0.05 M " C N , a C N : C u mole ratio o f 12, [OH"] = 0.25, 0.05 and 0 . 0 I M and 1 M N a 2 S 0 4 Figure A - 3 4 Plots o f potential vs . l og (current density) on a graphite rotating d i sk at 266 4900 r p m and different temperatures. Elect rolytes : [ C N ] = 0.025, 0.05, 0.1 and 0.20 M , a C N : C u mo le ratio - 3, [OH"] = 0.25 M and 1 M N a 2 S 0 4 Figure A - 3 5 Plots o f the potential vs . l og (current density) o n a py ro ly t i c graphite 267 rotating disk at 4900 r p m and 25 ° C . Elect rolytes : [CN"] = 0.025, 0.05, 0.1 and 0.20 M , a C N : C u mole ratio = 3, [OH"] = 0.25 M and 1 M N a 2 S 0 4 xxiv Figure A - 3 6 Plots o f l og (current density) vs. l o g ( [Cu(CN) 3 2 " ] ) o n a py ro ly t i c 267 graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : [CN"] = 0.025, 0.05, 0.1 and 0.20 M , a C N : C u mo le ratio = 3, [OH"] = 0.25 M and 1 M N a ^ O , F igure A - 3 7 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 268 0.0167 M C u + ( C N : C u mole ratio = 3), 0.25 M N a O H , 0.2 M N a 2 S 0 3 and 1 M N a 2 S 0 4 . F igure A - 3 8 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 268 0.0167 M C u + ( C N : C u mole ratio = 3), 0.25 M N a O H , 0.1 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 3 9 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 0.05 M C N " , 270 0.0125 M C u + ( C N : C u mole ratio = 4), 0.25 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 4 0 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 271 0.0125 M C u + ( C N : C u mole ratio = 4), 0.25 M N a O H , 0.2 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 4 1 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 272 0.0125 M C u + ( C N : C u mole ratio = 4), 0.25 M N a O H , 0.1 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 4 2 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 0.05 M C N " , 273 0.0167 M C u + ( C N : C u mole ratio = 3), 0.05 M N a O H , 0.2 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 4 3 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 274 0.0125 M C u + ( C N : C u mole ratio = 4), 0.05 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 Figure A - 4 4 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 275 0.0125 M C u + ( C N : C u mole ratio = 4), 0.05 M N a O H , 0.2 M N a 2 S 0 3 and 1 M N a 2 S 0 4 . F igure A - 4 5 Po la r i za t ion curves at different temperatures. E lec t ro ly te : 0.05 M C N " , 276 0.0125 M C u + ( C N : C u mole ratio = 4), 0.05 M N a O H , 0.1 M N a ^ and 1 M N a j S O , . Nomenclature Lis t s o f symbols a specif ic area (m" 1) aj ac t iv i ty o f species i C concentrat ion (moi dm" 3) C b bu lk concentrat ion (moi dm" 3) C s surface concentration (mole dm" 3) C . E . current eff ic iency D di f fus ion coefficient ( m 2 s"1) d diameter o f graphite fibre E potential o f the electrode ( V ) e porosi ty o f matr ix e 0 i n i t i a l porosi ty o f matr ix E ° standard potential ( V ) E P peak potential i n l inear potential sweep ( V ) E P / 2 potential where i = i p / 2 i n l inear potential sweep ( V ) E . C . energy consumpt ion ( k W h r kg" 1) E D T A ethylenediaminetetra-acetic ac id en ethylenediamine F Faraday constant = 96487 A s moi" 1 G e x excess G i b b s free energy I current ( A ) or ion ic strength I, current i n the l i q u i d phase ( A ) xxvi I s current i n the so l id phase ( A ) i current density, ( A m"2) i C u current density o f copper deposi t ion ( A m" 2) i d d i f fus ion current density (Am" 2 ) i H current density o f hydrogen evolu t ion ( A m" 2) i k k ine t i ca l ly control led current density ( A m"2) i , l i m i t i n g current density (Am" 2 ) i p peak current density ( A m"2) k heterogeneous rate constant ( m s"1) K a constant o f H C N dissocia t ion K s p so lubi l i ty product o f C u C N k m mass transfer coefficient ( m 2 s"1) K , equ i l i b r i um constant between S0 2 ( aq . ) and H S 0 3 " K 2 e q u i l i b r i u m constant between H S 0 3 " and S 0 3 2 " K 2 3 equ i l i b r i um constant for C u ( C N ) 2 " + C N " -> C u ( C N ) 3 2 " K 3 4 e q u i l i b r i u m constant for C u ( C N ) 3 2 " + C N " - » C u ( C N ) 4 3 " M molar i ty ( m o i dm" 3) m f m o l a l i t y o f species i , (moi kg"') n number o f electrons transferred or moles o f solutes i , j or k n r react ion order n w k i log rams o f solvent p H negative logar i thm to base 10 o f the act ivi ty o f hydrogen i o n R gas constant (8.314 J K " ' moi"') xxvii R e Reyno lds number S Siemens (Q" 1) S C E saturated ca lomel electrode S h She rwood number S H E standard hydrogen electrode T absolute temperature, ( ° K ) u ve loc i ty o f the l i q u i d ( m s"1) U * act ivat ion energy at the potential = 0 (J moi" 1 ) U* (E) ac t ivat ion energy at potential (E) (J moi" 1 ) v scan rate o f potential sweep ( V s"1) Z , ion ic charge L i s t s o f Greek S y m b o l s a charge transfer coefficent a a anodic charge transfer coefficient a , P i t ze r ' s parameters (= 2.0 for 1-1, 2-1 , 1-2,3-1, 4-1 electrolyte) a2 P i t ze r ' s parameters (= 0.0 for 1-1, 2-1 , 1-2,3-1, 4-1 electrolyte) P0 P i t ze r ' s parameter (3, P i t ze r ' s parameter P2 P i t ze r ' s parameter for 2-2 electrolyte or the associat ion constant for C u ( C N ) 2 " Yi ac t iv i ty coefficient o f species i 0 P i tze r ' s interaction parameter for l ike charged ions Xn te rm for descr ibing the short-range inter ionic effects as a funct ion o f ion ic xxi: strength to d isplay the type o f behaviour caused by the hard core effect p f chemica l potential o f the solute (mola l i ty) u°, chemica l potential i n the solute (mola l i ty) standard state Liijk term for tr iple i o n interactions w h i c h ignores any ion ic strength dependence 8 d thickness o f the diffusion layer (m) r) overpotential ( V ) v k inemat ic v iscos i ty ( m 2 s"1) Q o h m or number o f moles o f solvent i n a k i l o g r a m (55.51 for water) co angular ve loc i ty (s"1) K effective conduct iv i ty o f the l i q u i d (solution) phase (S m" 1) a effective conduct iv i ty o f the so l id (graphite fibre) phase (S m"') c3> potential difference bewteen the so l id phase and the l i q u i d ( V ) ®, potential o f the l i q u i d phase ( V ) O s potential o f the so l id phase ( V ) (j) osmot ic coefficient \\i P i t ze r ' s ternary parameter XXI X ACKNOWLEDGEMENTS I w o u l d l i ke to express m y sincere appreciat ion to D r . D . B . Dre i s inger for his thoughtful supervis ion and constructive discussions and r e v i e w i n g and edi t ing this thesis. I a m very grateful to D r . W . C . Cooper for r ev i ewing and edi t ing this thesis. I w o u l d also l ike to acknowledge D r . D . Tromans , D r . G . H . K e l s a l l and Prof. C . O l o m a n for p r o v i d i n g constructive ideas. D r . B . W a s s i n k ' s k i n d help especial ly i n chemica l analysis is ve ry m u c h appreciated.. Thanks are extended to m y fe l low graduate students and the staff o f the hydrometa l lurgy group and w i t h w h o m I have enjoyed w o r k i n g . The f inancia l support f rom the Facu l ty o f Graduate Studies i n the f o r m o f a U B C Graduate F e l l o w s h i p is greatly appreciated. F i n a l l y , I w o u l d l ike to thank m y wi fe , m y parents, brothers and sisters for g i v i n g me mora l support. 1 1. INTRODUCTION Cyan ide leaching has been w i d e l y accepted as an excellent industr ial method to recover g o l d and s i lver [1, 2] . Howeve r , the cyanida t ion o f copper -gold ores conta in ing the c o m m o n oxide and secondary sulfide copper minerals e.g. chalcoci te ( C u 2 S ) , bornite ( C u 5 F e S 4 ) , malachi te ( C u C 0 3 , C u ( O H ) 2 ) , covel l i te ( C u S ) and cuprite ( C u 2 0 ) results i n cyanide degradation and copper so lub i l i za t ion as cuprous cyanide complexes . In convent ional go ld processing, the copper and complexed cyanide are not recovered after the g o l d is r emoved f rom solut ion. T h i s leads to a significant economic penalty i n excess cyanide consumpt ion , loss o f a valuable copper by-product and significant cost i n cyanide destruction dur ing effluent treatment. Several ways have been proposed to solve the above problems. F o r example , (1) pretreating ores to remove copper etc. before cyanida t ion such as pressure ox ida t ion leaching [3], roast ing-leaching [4-6], and b io leaching [7, 8]; (2) the appl ica t ion o f alternative l ix iv ian t s to recover go ld such as thiourea [9], a m m o n i u m thiosufate [10], chlor ide [2], and b romide [2, 11]; (3) the addi t ion o f other reagents such as a m m o n i a to decrease the consumpt ion o f cyanide [12]. H o w e v e r , the above methods have their o w n drawbacks or appl ica t ion l imi t s and so i n most cases cyanide is s t i l l used to leach go ld ores conta ining copper. Therefore the recovery o f copper and associated complexed cyanide f rom leach solutions has been approached i n a variety o f ways such as ac id i f i ca t ion-vo la t i l i za t ion-regeneration ( A V R ) , i o n exchange and electrolysis [13-25]. The basic A V R process consists o f the f o l l o w i n g steps: (1) ac id i fy ing the barren solut ion to p H 2-3 w i t h sulphur ic ac id to dissociate copper cyanide complexes to fo rm H C N and precipitate copper as C u C N or a mixture o f C u C N and C u S C N , i f there is S C N " i n the solut ion; (2) v o l a t i l i z i n g H C N f rom the so lu t ion by intense air sparging, and (3) recover ing the H C N by absorpt ion i n an a lka l ine solu t ion ( N a O H or C a ( O H ) 2 ) [13 - 18]. In order to recover cyanide f rom the precipitates, oxidants such as H 2 0 2 and 0 3 have been tested to convert C u C N and C u S C N to C u 2 + and H C N i n the A V R process [15, 17]. N a H S was tested to precipitate copper as C u 2 S and recover a l l o f the cyanide i n the A V R process [18]. Several ion-exchange process has been proposed to improve the recovery o f cyanide i n a combina t ion w i t h the A V R process [15, 19, 20] . 2 E l e c t r o w i n n i n g was used to recover copper as metal and cyanide [21-25]. D u r i n g e lec t rowinning , cyanide is o x i d i z e d to cyanate, decreasing the recovery o f cyanide , and copper deposi t ion current eff ic iency was l o w due to the l o w copper concentrat ion. Severa l methods have been proposed to solve the above problems. T o increase the copper depos i t ion current eff ic iency, porous electrodes were used to deposit copper [22-25]. O r o c o n Inc. [22] reported that the thiocyanate i n the solut ion cou ld be o x i d i z e d to C N " and S 0 4 2 " to decrease the consumpt ion o f cyanide. Howeve r , the anodic current eff ic iency o f thiocyanate was not g iven . T o prevent the cyanide oxida t ion at the anode, an ion-exchange membrane was used to separate the anode and the cathode [25]. Recent ly a process was proposed w h i c h combines i o n exchange, A V R , membrane ce l l electrolysis and improves the eff ic iency for recover ing copper and cyanide [26, 27] . H o w e v e r , generally these processes suffer f rom the f o l l o w i n g drawbacks: incomple te recovery o f cyanide, incomplete recovery o f copper, low-va lue copper products (e.g. C u C N , C u S C N and C u 2 S ) and compl ica ted flowsheets. In order to overcome the above drawbacks , a solvent extract ion-elect rowinning process has been developed to recover copper and cyanide f rom g o l d m i n i n g effluents [28]. In summary, copper cyanide is extracted us ing a guanidine-based extractant (XI7950) or a m i x e d strong base extractant w i t h nony lpheno l ( X I 7 8 ) , str ipped w i t h strong a lkal ine electrolyte and f ina l ly e lec t ro lyzed i n a membrane ce l l to produce copper metal and a bleed stream for A V R to recover cyanide. The chemis t ry o f the process is shown be low: Copper extract ion: N a 2 C u ( C N ) 3 + 2 R + 2 H 2 0 -> R 2 H 2 C u ( C N ) 3 + 2 N a O H (1-1) where R species refers to the guanidine solvent extractant. Copper s t r ipping: R 2 H 2 C u ( C N ) 3 + 2 N a O H - > 2 R + N a 2 C u ( C N ) 3 + H 2 0 (1-2) Coppe r e lec t rowinning ( in a membrane ce l l ) : N a 2 C u ( C N ) 3 + N a O H - > C u + 3 N a C N + l / 4 0 2 + 1 / 2 H 2 0 (1-3) The use o f a membrane (Nat ion) c e l l i n the copper e lec t rowinning c e l l is necessary to prevent cyanide ox ida t ion at the anode. H o w e v e r , the N a f i o n membrane is expensive and m a y be subject to mechanica l damage by the g rowing metal deposit. I n order to e l iminate the use o f a membrane ce l l , an alternative anode react ion is used to prevent the ox ida t ion o f 3 cyanide. T h i s w i l l result i n a s impler c e l l design (no membrane) w i t h reduced capi ta l cost and l o w c e l l voltage ( l o w energy consumption) . Al te rna t ive anode reactions w h i c h have been suggested and tested are: (1) the ox ida t ion o f thiocyanate to cyanide and sulphate, (2) the ox ida t ion o f methanol to C O , and H 2 0 , (3) the ox ida t ion o f sulphite to sulphate and (4) the ox ida t ion o f a m m o n i a to N 2 and H 2 0 . The inc lus ion o f the above sacr i f ic ia l species was tested i n some proof-of-concept e lec t rowinning experiments i n our lab and was s h o w n to be p r o m i s i n g on ly for sulphite. W i t h sulphite addi t ion, the c e l l chemistry becomes: N a 2 C u ( C N ) 3 + l / 2 N a 2 S 0 3 + N a O H = C u + 3 N a C N + 112Na2S04 + 1 / 2 H 2 0 (1-4) Therefore a process has been proposed by the Hydrometa l lu rgy G r o u p at U B C us ing the flowsheet shown i n F igure 1-1 to recover copper and cyanide. In the first step ( loading) , barren cyanide solut ion (0.5 -2 g L " ' C u , C N : C u mo le ratio = 3-4) is m i x e d w i t h organic phase (extractant and solvent) and copper cyanide is extracted to the organic phase. In the second step (stripping), the organic phase loaded w i t h copper cyanide is m i x e d w i t h strong a lkal ine electrolyte (60 g L " ' C u , C N : C u = about 3, 4-10 g L " 1 N a O H and 50-60 g L " 1 N a 2 S 0 3 ) and copper cyanide is transferred to the electrolyte and the copper concentrat ion o f the electrolyte increases to about 70 g L " ' . In the th i rd step (e lectrowinning) , the electrolyte is returned to the e lec t rowinning ce l l and copper is deposited on the cathode. In the fourth step (acidif icat ion) , a bleed stream o f electrolyte is taken out and m i x e d w i t h H 2 S 0 4 and copper cyanide is dissociated to fo rm H C N and C u C N at p H 2-3. C u C N was returned to the e lec t rowinning ce l l and H C N is removed by sparging air and f ina l ly absorbed i n a lka l ine solu t ion ( N a O H or C a ( O H ) 2 ) . The direct e lec t rowinning o f copper f rom a barren cyanide leach solu t ion m a y be preferred i n some cases for the recovery o f copper and recycle o f cyanide. H o w e v e r , careful study o f this process has not been reported. Therefore, the efficient deposi t ion o f copper f rom a barren cyanide solut ion is a p romis ing alternative approach to the recovery o f cyanide and copper. The process for the direct e lec t rowinning process has been developed and consists o f the f o l l o w i n g steps: (1) barren cyanide solut ion ( 1-2 g L " 1 C u and C N : C u mo le ratio = 3-4) enters the membrane c e l l and f lows through the graphite felt cathode o n w h i c h copper is deposited and the copper depleted cyanide solut ion returns to go ld leaching and (2) copper is deposited o n a metal sheet and then refined i n a second electrorefining c e l l conta in ing copper sulphate solut ion. The flowsheet is shown i n F igure 1-2. A n in i t i a l economic assessment has been performed o n direct e lec t rowinning and o n the S X - E W system(see A p p e n d i x 1). The assessment has been made us ing an assumed ore grade. The analysis indicates that a significant benefit m a y be avai lable by app ly ing one o f these processes. In order to improve the above processes, the two e lec t rowinning processes should be studied as regards both the pract ical and fundamental aspects. Therefore the present research was undertaken w i t h the f o l l o w i n g objectives: (1) T o study the aqueous chemistry o f copper cyanide solutions i n the temperature range 25 -60 ° C w i t h reference to copper concentration, C N : C u mole ratio and p H . The results c o u l d be generated by ca lcu la t ion us ing the equ i l i b r ium copper cyanide constants and then conf i rmed by potential measurement. It was expected that this study w o u l d lead to an i m p r o v e d understanding o f the dis t r ibut ion o f copper cyanide complexes under prac t ica l condi t ions and their role i n the electrodeposi t ion and the anodic ox ida t ion o f copper cyanide. (2) T o study the e lec t rowinning o f copper f rom concentrated cyanide solutions us ing an alternative anode react ion so as to l i m i t the ox ida t ion o f cyanide. The study w o u l d be conducted w i t h reference to C N : C u mole ratio (3-4.5), temperature (40-60 ° C ) and the concentrat ion o f sacr i f ic ia l species (for sulphite 50 -120 g L " ' ) . These parameters w i l l s ignif icant ly affect the cathode and anode processes. Copper concentrat ion should be contro l led at 60-70 g L " 1 to get a reasonable copper deposi t ion current eff ic iency and to simulate the copper content o f the strong electrolyte i n the e lec t rowinning process. (3) T o study the e lec t rowinning o f copper on a graphite felt cathode w i t h reference to copper concentrat ion (1-2 g L " 1 ) , mole ratio o f cyanide to copper (3-4.5) and f l o w rate and current density (30-100 A m" 2) at an ambient temperature (25-40 ° C ) f rom v iewpo in t o f industr ia l practice. (4) T o study the ox ida t ion o f sulphite on graphite w i t h reference to temperature (25 -60 ° C ) , N a 2 S 0 3 concentrat ion (0.05-0.5 M ) and hydroxide concentrat ion (0.05-0.25 M ) us ing rotat ing disk technique and linear potential sweep. The anodic behaviour o f sulphite on the graphite (Tafel slope and rate constant) and the mass transfer (diffusion coefficient) can be obtained and compared to those o f copper cyanide to decrease the anodic ox ida t ion o f cyanide . (5) T o study the ox ida t ion o f copper cyanide on graphite w i t h reference to temperature (25-60 ° C ) , mo le ratio o f cyanide to copper (3-12), cyanide concentrat ion (0.05-4 M ) and 5 hydrox ide concentrat ion (0.01-0.25 M ) us ing the rotating disk technique. T h i s research c o u l d lead to k n o w i n g h o w these parameters affect the anodic behaviour o f copper cyanide. (6) T o study the anodic ox ida t ion o f copper cyanide and sulphite solutions w i t h reference to their concentrations, C N : C u mole ratio, temperature (25-60 ° C ) , hydrox ide concentrat ion (0.05-0.25 M ) and the current density us ing the rotating disk technique. The anodic behaviour o f m i x e d sulphite and copper cyanide m a y not be the same as w h e n they are present separately i n the solut ion. Therefore it is necessary to k n o w the anodic behaviour o f the mixture . The results o f this study should help to increase the eff ic iency o f recover ing copper and cyanide f rom a barren go ld solut ion and to decrease the cost. T h i s thesis consists o f seven major chapters: Chapter 2 deals w i t h a r e v i e w o f the literature, p r o v i d i n g a summary o f current ideas about the deposi t ion o f copper f rom cyanide solut ion, the anodic ox ida t ion o f copper cyanide and the anodic ox ida t ion o f sulphite, thiocyanate, methanol and ammonia . Chapter 3 considers the thermodyanics o f copper cyanide. Chapters 4 - 8 present the experimental aspects, results and d i scuss ion o f the direct copper e lec t rowinning f rom a dilute cyanide solut ion, copper e lec t rowinning us ing an alternative anodic reaction, the anodic ox ida t ion o f sulphite, anodic ox ida t ion o f copper cyanide and the anodic ox ida t ion o f m i x e d sulphite and copper cyanide solutions respect ively. Chapter 9 summarizes the research w o r k and Chapter 10 gives some suggestions for future studies. 6 Barren cyanide solution H 2 S 0 4 a Loading ->Raffinate Organic phase Organic phase NaOH- a Stripping Electrolyte NaOH-Na2S03-Electrolyte Electrowinning Bleed H 2 S O 4 -CaO-CuCN Acidification H C N Neutralization Cu -> Barren solution Ca(CN)2 Figure 1-1 Flowsheet for solvent extraction - electrowinning process for the recovery of copper cyanide 7 Water Gold Heap leaching U Gold recovery Solution storage < Reagent Barren solution Cu Electrowinning Cu depleted solution Impure Cu Cu electrorefining Pure Cu Figure 1 -2 Flowsheet for direct electrowinning of copper from cyanide solutions 8 2. LITERATURE REVIEW 2.1 Aqueous Chemistry of the Copper-Cyanide System Copper cyanide can be d isso lved i n the presence o f excess cyanide to fo rm cyanocuprate ions, C u ( C N ) 2 " , C u ( C N ) 3 2 " and C u ( C N ) 4 3 " i n aqueous solut ion. T h i s d i sso lu t ion has been studied by var ious methods [29-59]. These species undergo the f o l l o w i n g successive equ i l i b r i um steps i n react ion w i t h free cyanide and undissociated hydrocyan ic ac id . CuCN = Cu+ + C A T K s p (2-1) CuCN + CN- =Cu(CN)~ K i , 2 (2-2) Cu+ +2CN' = Cu(CNy2 (2-3) Cu(CN)2~ + CN~ = Cu(CN)32~ K 2 , 3 (2-4) Cu(CN)32- +CN~ = Cu(CN)43~ K 3 , 4 (2-5) HCN = H+ + C A T K a (2-6) Table 2-1 The associat ion constants for copper cyanide complexes Method Temperature Concentration log P2 log K 2 3 log K 3 4 Potentiometry [32] 25 °C 10"1 - 1 0 " 7 M C N - 23.72 - -Potentiometry [42] 20 °C 0.5 -5 M C N 21.7+ 1.0 4.6 ±0 .30 2.3 ±0 .15 Potentiometry [46] 22 °C 0. 01 M C u 21.7 + 0.2 5.1 ± 0 . 2 1.1 ± 0 . 2 Potentiometry [47] 25 °C 0.15 M C u 24 ± 0.23 4.8 2.25 Potentiometric titration [59] 25 °C 1 M NaCl 23.97 ±0.01 5.43 + 0.04 2.38 + Infrared spectroscopy [33] 25 °C 0.1 -0.2 M C u - 4.89 1.72 Ultraviolet spectroscopy [58] 25 °C ionic strength: 0.01M - 5.34 1.74 Ultraviolet spectroscopy [39] 25 °C 0.001 M Cu - 4.1 -Calorimetry 25 °C 1.0 M C u - 5.0 2.6 Calorimetry [38] 25 °C ionic strength—» 0 - 5.3 +0.01 1.5 ± 0 . 2 The so lub i l i ty product (IC^) o f cuprous cyanide differs s l ight ly between authors, a value o f 10" 2 0 at 2 5 ° C being generally accepted [44]. There is good agreement for the H C N dissocia t ion constant ( K J amongst the publ i shed data. The recommended value for K a at 25 ° C is 10" 9 2 1 w h i c h was obtained by extrapolat ion to the ion ic strength = 0 or by ca lcu la t ion us ing an extended fo rm o f the D e b y e - H i i c k e l equation [44, 52]. The d issoc ia t ion constant i n aqueous solu t ion conta ining different ion ic m e d i a has also been reported [52, 54-57] . 9 The equ i l i b r i um constants for copper cyanide complexes (Table 2-1) differ between authors due to the different methods o f measurement and the process ing o f the data. V l a d i m i r o v a and K a k o v s k y [32] obtained a value o f 1 0 2 3 7 2 for (32 us ing potential measurements w i t h pure copper and copper amalgam at p H 4.2. T h i s was consistent w i t h the value estimated f rom the equ i l i b r ium constant between C u C N S and C u ( C N ) 2 \ T h i s value was corrected to 1 0 2 4 by some authors [33, 38, 39, 44] us ing the D e b y e - H i i c k e l equat ion and a more rel iable d issocia t ion constant for H C N . Ro thbaum [42] reported a value o f 1 0 2 1 , 7 for P2 at 20 ° C by measur ing the copper potential i n a solut ion o f h i g h copper cyanide concentrat ion i n the presence o f air and s imp l i fy ing the copper cyanide species for ca lcu la t ion wi thout cons ider ing the act ivi ty coefficient, leading to some error. H a n c o c k et a l . [46] obtained a value o f 1 0 2 1 7 for P2 us ing potential measurement i n solutions conta ining 0.01 M Cu(I ) and 0.025 - 0.1 M C N " at p H 11 and 22 ° C under an A r atmosphere. H o w e v e r , p2 c o u l d be underestimated because some o f the potential data used for the ca lcu la t ion o f p2 were measured at a C N : C u mo le ratio > 4 and were w e l l b e l o w the hydrogen e q u i l i b r i u m potential . The measured potentials were m i x e d potentials and higher than the corresponding e q u i l i b r i u m potentials. B e k and Z h u k o v [47] reported a value for 1 0 2 4 for P2 us ing potential measurements i n solutions w i t h 0.15 M Cu(I) , C N : C u = 4, and 0.1 M N a O H and an extended fo rm o f the D e b y e - H i i c k e l equation. Kappens te in and H u g e l [48] obtained a value o f 1 0 1 6 7 for p2 us ing U V spectroscopy, changing the p H and assuming C u ( C N ) 2 " was the on ly copper complex i n the solut ion. H o w e v e r , this value is m u c h lower than the format ion constant (1 /Ksp) for C u C N and C u ( C N ) 2 " was not dominant under such condi t ions accord ing to its P2 value. Recent ly Hefter et a l . [59] reported a value o f 1 0 2 3 9 7 for P2 w h i c h was obtained by potentiometric t i tration us ing a C u + solut ion produced by reduct ion o f C u 2 + w i t h an excess o f copper and s tabi l ized by chlor ide . So the most rel iable value for p2 appears to be 10 2 4 . The differences among the reported values o f K 2 3 , and K 3 4 are re la t ively sma l l . The most re l iable values o f K 2 3 and K 3 4 are those reported by Izatt et a l . [38]. T h e y were obtained under wel l -def ined condit ions us ing p H measurements and calor imetry and the D e b y e -H i i c k e l equation. A 6 3 C u and 6 5 C u magnetic resonance study showed that C u ( C N ) 4 3 " retains a tetrahedral symmetry and C u ( C N ) 3 2 " has a distorted tetrahedral rather than a plane t r iangular conf igura t ion [43]. Cuprous ions fo rm m i x e d complexes w i t h the cyanide l i gand and other 10 l igands such as thiourea, thiocyanate, iodide , a m m o n i a and chlor ide , for example , C u ( C N ) 3 S C N 3 - and C u ( C N ) 4 S C N 4 " [43]. The complexed cyanide rap id ly exchanges w i t h aqueous free cyanide [29]. W h e n the mole ratio o f cyanide to copper is less than 3, the copper cyanide is readi ly o x i d i z e d by air, suggesting that C u ( C N ) 2 " is less stable [60]. C u p r i c ions react w i t h C N " and form cupr ic complexes , w h i c h are unstable and decompose rap id ly [29, 58]. It was reported that w h e n the mole ratio o f C N : C u is not h igh , cupr ic ions react w i t h the cyanide i n aqueous solut ion to give cupr ic cyanide as a y e l l o w i s h -b r o w n precipitate, w h i c h decomposes into cupr ic cyanide and cyanogen accord ing to the f o l l o w i n g equations [61, 62]: C u 2 + + 2 C N " -> C u 1 1 ( C N ) 2 (2-7) 2 C u n ( C N ) 2 2 C u ' C N + ( C N ) 2 (2-8) The cyanogen thus formed is evo lved as a gas f rom ac id ic solut ion, or it is decomposed i n a lka l ine solu t ion as fo l lows : ( C N ) 2 + 2 0 F T -> C N " + C N O " + H 2 0 (2-9) W h e n the mo le ratio o f cyanide to copper is h igh , not copper(II) d icyanide but a purple intermediate is formed w h i c h rapid ly decomposes into cyanogen and a copper species. E v e n at ordinary temperature a transient v io le t co lour m a y be noted i n neutral or s l igh t ly a lka l ine m e d i a [29, 63, 64]. The kinet ics studies p rov ided the first strong evidence for the format ion o f C u ( C N ) 4 2 " i n reactions between C u 2 + or its E D T A complex and C N " [29, 58, 65, 66] . L o n g o and B u s h [67] conducted the C u 2 + - C N " react ion i n methanol or d imethy l formamide f rom -60 to -30 ° C and conc luded that the unstable purple species is a square planar c o m p l e x C u ( C N ) 4 2 " . M o n s t e d and B j e r r u m [68] studied the react ion between C u 2 + and C N " i n aqueous methanol at - 70 ° C and reported that the absorpt ion m a x i m u m at 535 n m was nearly i n the same pos i t ion as that for C u ( e n ) 2 2 + , suggesting a distorted tetrahedral structure. Nei ther the electron spin resonance nor the opt ica l spectrum is inf luenced by the presence o f excess o f cyanide, showing that no pentacyano c o m p l e x is formed. There are two reports about the format ion constant o f cupr ic tetracyanide [69, 70] . Paterson and B j e r r u m [69] estimated the format ion constant o f C u ( C N ) 4 2 " as 1 0 2 6 7 by potentiometric experiment i n water-methanol so lu t ion (mole fraction o f methanol = 0.45) at -45 ° C , w i t h the ion ic strength va ry ing between 0.05 and 0.1 M ( N a C N ) . K a t a g i r i et a l . [70, 11 71] o x i d i z e d C u ( C N ) 4 3 " on a p la t inum electrode to generate C u ( C N ) 4 2 " and measured the redox potential for the C u ( C N ) 4 2 7 C u ( C N ) 4 3 " c o u p l e . They reported that the standard potential for the C u ( C N ) 4 2 7 C u ( C N ) 4 3 " redox couple was 0.54 V vs. S H E and the overa l l format ion constant for C u ( C N ) 4 2 ' was 10 2 4 . Baxenda le and Westcott [58] studied the react ion between C u 2 + and C N " i n w e a k l y ac id ic so lu t ion to keep the concentration o f free cyanide i o n l o w and decrease the react ion rate. T h e y found that the reaction was second order i n C u 2 + and 6th order i n C N " f rom the change i n the concentrat ion o f the reaction product, C u ( C N ) 2 " us ing a U V spectrophotometer. T h e y proposed the f o l l o w i n g mechanism: C u 2 + + 3 C N " <^ C u ( C N V (2-10) 2 C u ( C N ) 3 ~ -> 2 C u ( C N ) 2 " + ( C N ) 2 (2-11) N o r d and Mat thes [72] used the stopped - f l o w technique to study the react ion between C u 2 + and C N " i n aqueous solutions at 0 to 25 ° C and found that the react ion was second order w i t h respect to C u ( C N ) 4 2 " and inversely proport ional to the concentrat ion o f the free cyanide . O n the basis o f these results, they proposed the f o l l o w i n g react ion mechan ism: C u ( C N ) 4 2 " C u ( C N ) 3 " + C N " (2-12) C u ( C N ) 4 2 " + C u ( C N ) 3 " -> C u ( C N ) 3 2 " + C u ( C N ) 2 " + ( C N ) 2 (2-13) Reac t ion 2-13 is considered to be the rate-control l ing step. K a t a g i r i et a l . [70, 73] studied the kinet ics and mechan i sm o f the decompos i t ion o f C u ( C N ) 4 2 " generated by the anodic ox ida t ion o f C u ( C N ) 4 3 " and found that the rate o f the decompos i t ion was second order w i t h respect to C u ( C N ) 4 2 " and inversely proport ional to the square o f the concentrat ion o f the free cyanide concentrat ion. The f o l l o w i n g decomposi t ion mechan i sm was proposed: 2Cu(CN)24~ <=> Cu2(CN)l + 2CN~ (2-14) Cu2(CN)2- -> 2Cu(CN)2~ +(CN)2 (2-15) Reac t i on 2-15 is proposed as the rate-determining step. The rap id decompos i t ion o f cupric cyanide results i n the ox ida t ion o f cyanide w h i c h has led to the use o f cupric ions as a catalyst to destroy cyanide i n waste water [74 - 76]. 12 2.2 The Electrodeposition of Copper from Copper-Cyanide Solution 2.2 A Practice of Copper Deposition from Cyanide Solution The electrodeposi t ion o f copper f rom cyanide solut ion has been w i d e l y reported [21-29, 31 , 77-128]. H o w e v e r , there are very few reports o n copper e lec t rowinning f rom copper cyanide solu t ion and most reports deal w i t h copper p la t ing. A n early copper e lec t rowinning operat ion was carr ied out at the San Sebastian M i n e i n 1904 [79]. C levenger [84, 85] reported that copper was recovered i n N e v a d a and M e x i c o , but cyanide consumpt ion was h i g h (30% o f cyanide was destroyed) and the current eff ic iency for copper depos i t ion was l o w . L o w e r [21] reported that the direct e lec t rowinning o f copper f rom a leach so lu t ion conta ining 13.7 - 24 g L " 1 C u at ambient temperatures gave about 70 %> current ef f ic iency and a energy consumpt ion o f about 1.3 k W h / k g C u at 47-93 A r n 2 . Shantz and R e i c h [77] ran l o c k e d leaching-e lec t rowinning tests on a copper rougher concentrate and obtained 62 % current eff ic iency and a energy consumpt ion o f 0.7 k W h / k g C u at 70-80 A m" 2 . Coppe r e lec t rowinning f rom dilute barren copper cyanide solutions was carr ied out w i t h a h i g h surface area cathode [22-25], but no details such as copper deposi t ion current ef f ic iency, cyanide consumpt ion and energy consumpt ion are reported. D u Pont [26] has patented a process for the recovery o f cyanide and copper by e lec t rowinning f rom cyanide solut ions i n a c e l l i n w h i c h the anolyte is separated f rom the catholyte by a membrane to a v o i d the anodic ox ida t ion o f cyanide. A c i d i f i c a t i o n , i o n exchange or carbon adsorpt ion was used to concentrate the copper cyanide solutions and adjust the ratio o f cyanide to copper to b e l o w 3.0 - 4.0. Coppe r e lec t rowinning has been conducted at U B C us ing membrane cel ls w i t h the effects o f temperature, compos i t ion , current density being studied [28]. Solvent extract ion was used to concentrate copper cyanide. Solvent extract ion is more effective i n the extract ion o f copper cyanide f rom dilute copper cyanide solutions than the use o f ac id i f ica t ion , i o n exchange or carbon adsorption. The U B C S X - E W process m a y have advantages over the D u Pont process. Copper p la t ing f rom cyanide solutions has been used throughout the meta l f in i sh ing industry since E l k i n g t o n discovered this technology i n 1840 [80]. U n d e r the proper condi t ions , the metal dis t r ibut ion over i r regular ly shaped articles is excel lent because o f the good th rowing power . T y p i c a l copper cyanide bath composi t ions and condi t ions are l is ted i n Table 2-2. Current eff ic iency is a function o f compos i t ion , temperature and current density. 13 Coppe r cyanide solut ion was used to plate copper o n porous materials [121] or carbon fibres i n the presence o f support ing electrolyte [125]. Tab le 2-2 Copper cyanide bath composi t ions and condi t ions [82] B a t h T y p e Strike Roche l l e H i g h E f f i c i e n c y C u (g/1) 11.0 1 5 - 3 0 3 4 - 8 9 Free cyanide (g/1) 6.0 4 - 9 11 - 19 N a 2 C 0 3 or N a O H (g/1) 1 5 ( N a 2 C 0 3 ) 15-60 ( N a 2 C 0 3 ) 2 2 - 2 7 ( N a 0 H ) Temperature (°C) 41-60 55-70 60-80 Cathode Current ( A / m 2 ) 1 0 0 - 320 160 - 650 1 0 0 - 1110 Cathode current eff ic iency (%) 1 0 - 6 0 3 0 - 7 0 >99 2.2.2 The Effect of Parameters on Copper Deposition The copper current eff ic iency decreases w i t h increasing mo le ratio o f cyanide to copper [91, 92, 124]. W i t h increasing ratio o f cyanide to copper, the e q u i l i b r i u m potential decreases. B y L e Chate l ie r ' s p r inc ip le we should expect increasing cyanide to inh ib i t the d issocia t ion o f copper cyanide complexes and to retard the discharge reactions. H o w e v e r , it has a more important effect i n shifting the complex dis t r ibut ion towards the less active c o m p l e x state ( C u ( C N ) 2 " -> C u ( C N ) 3 2 " —» C u ( C N ) 4 3 " ) . Therefore the copper discharge potential decreases result ing i n more hydrogen evolu t ion . The ratio o f cyanide to copper close to 3 is o p t i m u m for the h i g h eff ic iency electrolyte. The equ i l i b r i um potential for H 2 0 / H 2 (expressed as E ( H 7 H ) = -0.0591 p H V vs . S H E ) decreases w i t h increasing p H , but p H has a relat ively smal l effect on the redox potential for C u V C u at a p H above 9. In a lkal ine solut ion H 2 0 is discharged on the electrode and so the current o f hydrogen evolu t ion at a f ixed potential m a y not be dependent o n p H as expected f rom the change i n the equ i l i b r ium potential for H 2 0 / H 2 . The copper current ef f ic iency m a y not s ignif icant ly increase w i t h increasing p H . H y d r o x i d e or carbonate salts have to be added to get a higher p H . H o w e v e r , addi t ion o f carbonate and hydrox ide ions is also associated w i t h a reduct ion i n the current for copper deposi t ion, w i t h the relat ionship be ing approximate ly l inear [92, 124]. These effects are not on ly due to the presence o f C 0 3 2 " and O H " ions , but probably to the concomitant increase i n the a lka l i metal i o n concentrat ion and surface adsorption. 14 The current eff ic iency decreases w i t h increasing current density. O b v i o u s l y , at a higher current density and a higher polar iza t ion potential [87, 88, 91 , 92] , the ratio o f cyanide to copper i n the solu t ion near the cathode surface is higher due to a l im i t ed d i f fus ion rate and hydrogen evo lu t ion increases faster than copper deposi t ion. The cathodic current eff ic iency increases w i t h increasing temperature. A t higher temperatures, the copper-cyanide dissocia t ion constant is larger and the balance shifts to the format ion o f l o w l y coordinated copper complexes ( C u ( C N ) 2 " ) , w h i c h w i l l be discussed i n the next chapter, and cuprous complexes diffuse faster to the cathode surface and are more readi ly reduced. H o w e v e r , w i t h increasing temperature, the hydro ly t i c decompos i t ion o f cyanide increases [91]. A g i t a t i o n increases the cathodic current eff ic iency [92]. D u e to the reduct ion o f cuprous ions at the cathode, the ratio o f copper to cyanide i n the cathode boundary layer decreases resul t ing i n a lower current eff iciency. A g i t a t i o n accelerates the rate o f cuprous i o n movement to the cathode surface and cyanide movement away f rom the cathode. Therefore the concentrat ion o f cuprous ions near the cathode surface increases, resul t ing i n a higher current eff ic iency. I ron and c h r o m i u m i n the copper-cyanide solu t ion decrease the current ef f ic iency [92]. B i s m u t h , z inc , ant imony and other metals w i l l cause a rough deposit at t imes [89]. The incorporat ion o f thiocyanate and specif ic surface-active agents permits the deposi t ion o f bright, smooth deposits [90-98, 101-104]. Thiocyanate also increases the cathodic current eff ic iency [92, 94-96, 98]. It is possible that the adsorpt ion o f S C N " at the copper cathode suppresses the discharge o f H + (or H 2 0 ) and therefore increases the copper cathodic current eff ic iency. S h i v i r i n et a l [99, 100] reported that the addi t ion o f thiocyanate had li t t le effect o n the overpotential o f hydrogen evolu t ion . It was reported that thiocyanate c o u l d be used i n place o f cyanide for copper pla t ing [105]. S o d i u m sulphite and bisulphite have been recommended as addit ions to copper cyanide baths to improve the brightness o f the deposits [90]. 2.2.3 The Kinetics and Mechanism of Copper Deposition 15 The kinet ics and mechan i sm o f copper deposi t ion f rom copper-cyanide so lu t ion have not been w i d e l y studied. B l a n c [108] reported that the species discharged c o u l d not be free C u + accord ing to his w o r k on the effect o f alternating current. Glasstone [32] proposed the d i rec t ion reduct ion o f copper f rom C u ( C N ) 2 " . Cos ta [110] studied the e lectrochemical behaviour o f copper-cyanide solut ions (0 .01-0.08 M C u C l and 0.06-0.93 M K C N ) and proposed the f o l l o w i n g mechan i sm: Cu(CN)32~ -> CuCN + 2C7V" (2-16) CuCN + e ^ C u + CN~ (2-17) The transfer coefficient was 0.38 ± 0.04 and the exchange current density was propor t ional to C u + concentration. The curve o f l og I, as a funct ion o f log[CN"] exhibi ts a changing slope for a free C N " concentrat ion greater than 0.21 M . The v a r y i n g slope is considered to be a result o f the var ia t ion i n the phys i ca l surface o f the electrode rather than a change i n the e lec t rochemical process. L o w e n h e i m [111] thought that the direct discharge o f C u ( C N ) 4 3 " was more poss ible than the two-step discharge mechan i sm 0 < C / V ) 4 3 - - » Cu+ + 4 C A T +" >Cu (2-18) R a u b and M u l l e r [112] thought that the reaction mechan i sm is: Cu(CN)32- -> Cu(CN)2 + C A T - ^ ^ C u (2-19) B e k and Z h u k o v [113-116] studied the deposi t ion o f copper f rom a so lu t ion w i t h 0.1 M C u + and a C N : C u + mole ratio o f 2.8-3.2 at p H 13 and thought that copper depos i t ion results i n a significant var ia t ion i n the dis t r ibut ion o f the copper-cyanide species and a s ignif icant concentrat ion polar iza t ion. T h e y found that C u ( C N ) 2 " was the discharged species and the charge transfer coefficient was 0.1 after correct ing for the concentrat ion change. T h e y proposed the f o l l o w i n g reaction mechanism: Cu(CN)32- -> Cu{CN)2 + CN~ at C N : C u ratio < 3 (2-20) Cu(CN)43- -> Cu(CN)2 + 2C7V" at C N : C u ratio > 4 (2-21) 16 Cu{CN)2~ + e -> Cu + 2CN~ (2-22) S in i t sk i et a l . [118] reported that a distinct l i m i t i n g current c o u l d be obtained i n dilute copper cyanide solutions at p H 4.95. The Tafe l slopes ranged f rom 0.130-0.165 V decade" 1 and the transfer coefficient was 0.40 ± 0.03. C h u and F e d k i w [122] have used the vol tammetr ic and steady-state po la r i za t ion response o f a copper-disk electrode to study the kinet ics o f copper deposi t ion f rom a cyanide bath us ing the solut ion: 0.1 M N a 2 C 0 3 + 0.2 M C u C N + 0.6 M N a C N and p H 12. The major species discharged was considered to be C u ( C N ) 3 2 " , a l though C u ( C N ) 4 3 " is the predominant complex . The cyanide released dur ing deposi t ion shifts the d is t r ibut ion o f the complexes at the surface to the coordinately saturated state and results i n a decreased copper depos i t ion rate since the discharge o f C u ( C N ) 4 3 " is considerably s lower than that o f C u ( C N ) 3 2 " . Hather ley et a l . [124] measured the polar iza t ion curves o f copper depos i t ion f rom cyanide solut ion. It was conc luded that C u ( C N ) 2 " was first d ischarged and subsequently C u ( C N ) 3 2 " . C u ( C N ) 4 3 " does not seem to take part i n the deposi t ion process. A t a certain l i m i t i n g current density these processes break d o w n and there is a loss o f cathode current eff ic iency. Steponavicius et a l . [127] studied the mechan i sm o f copper deposi t ion us ing l inear potential sweep, l inear current scan and single galvanostatic pulse methods and found that the preceding react ion for copper deposi t ion is the d issocia t ion o f C u ( C N ) 3 2 " into C u ( C N ) 2 " and C N " and then C u ( C N ) 2 " is discharged o n the cathode. H s u and T ran [129] studied the reduct ion o f copper cyanide us ing a rotating disc and found that the e lec t rochemical active species is C u ( C N ) 2 " . 2.3 Electrochemical Oxidation of Cyanide Great attention has been pa id to the study o f the e lec t rochemical ox ida t ion o f C N " i n order to m i n i m i z e the destruction o f cyanide i n metal e lec t rowinning f rom cyanide so lu t ion and m a x i m i z e the eff ic iency o f the destruction o f cyanide i n effluent streams to meet envi ronmenta l requirements [21-24, 85, 130 - 168]. The products and mechan i sm o f cyanide 17 ox ida t ion depend m a i n l y o n p H , potential and concentration. F r o m the f o l l o w i n g redox reactions, hydrocyan ic ac id is more diff icul t to ox id ize and is m u c h less electro-active [138]. 2HCN = (CN)2 + 2H+ +2e E° = 0.373 V vs. S H E (2-23) 2CN~ = (CAT) 2 +2e E° = - 0.176 V vs. S H E (2-24) 2.3.1 Cyan ide Oxidation in Alkal ine Solution U n d e r a lkal ine condi t ions, the reaction for the ox ida t ion o f cyanide is [131, 132, 149, 150, 158]: CN~ + 20H~ = CNO' +H20 + 2e E° = - 0.97 V vs. S H E (2-25) Cyanate can be further o x i d i z e d at higher potentials to C 0 3 2 " and N 2 [131, 137, 158], but its current eff ic iency has not been reported. 2CNO~ + %OH~ = 2C02~ + N2+ 4H20 + 6e E° = - 0.95 V vs . S H E (2-26) A r i k a d o et a l . [143] reported that the Tafe l slope for cyanide ox ida t ion o n a graphite electrode was about 0.12 V decade" 1 and the reaction orders were uni ty and zero for C N " and O H " respect ively. Cyan ide is not o x i d i z e d by atomic o x y g e n formed by w a y o f o x y g e n evolu t ion . The f o l l o w i n g mechan i sm was proposed: OH~ + CN~ "2e ) HOCN + Q / r > CNO~ + H20 (2-27) The rate o f cyanide ox ida t ion increases w i t h increasing cyanide concentrat ion and is independent o f O H " concentrat ion (> 0.01 M ) . The discharge o f cyanide i o n determines the overa l l react ion rate. The apparent number o f electrons part ic ipat ing i n the react ion decreases f rom 2 to 1 w i t h decreasing O H " concentration (1 to 10"4 M ) [143]. The current eff ic iency o f cyanide oxida t ion depends o n the anode materials , current density and concentrat ion [156]. 2.3.2 Cyan ide Oxidation in Weak ly Acidic , or Alkal ine or Neutral Solutions In neutral and weak ly alkal ine solutions ( p H 7.0 - 11.7), cyanogen is the m a i n cyanide ox ida t ion product accord ing to Reac t ion 2-24 [138, 143, 149, 150, 163]. T h i s cond i t ion is referred to as hydroxide-s tarved ox ida t ion o f cyanide. Cyanogen can react subsequently w i t h hydrox ide i n solu t ion to give cyanate and cyanide: 18 (CJV)2 + 20H~ = CNO~ + CN~ + H20 (2-28) The cyanide radica l can also po lymer ize to fo rm paracyanogen ( C N ) n . A z u l m i n , ( H C N ) n is formed due to the po lymer i za t ion o f aqueous hydrocyan ic a c i d [136, 149,156, 157]. H i n e et a l . [156] reported that a z u l m i n format ion is c lose ly related to the ratio o f CNT to O H " . In neutral or s l ight ly a lkal ine solutions ( p H 7.0 - 8.6) [157] or i n w e a k l y a lka l ine carbonate-buffer solutions ( p H 9.3) [149], the cyanate i o n m a y cont inuous ly undergo hydro lys i s to produce a m m o n i u m and carbonate ions ( C N O - + 2 H 2 0 —> N H 4 + + C 0 3 2 " ) . In w e a k l y ac id ic solut ion ( p H 5.2-6.8), ( C N ) 2 is hyd r o lyzed to f o r m oxamide , ( C O N H 2 ) 2 and oxalate, C 2 0 4 2 " and N H 4 + : 2.3.3 The Anod ic Oxidation of Copper Cyan ide There are some reports on the anodic ox ida t ion o f copper cyanide , but most o f them are about the products and phenomena o f the electrolyt ic ox ida t ion and are incomple te[135, 139-142, 144, 145, 147, 149-152, 156, 157, 160]. Sperry and C a l d w e l l [135], Dar t et a l . [139], and Eas ton [141] thought that copper deposi t ion releases free cyanide at the cathode and then the free cyanide is o x i d i z e d to cyanate at the anode. D r o g e n and Pasek [140] and Daubaras [151] proposed a direct ox ida t ion route (copper cyanide complexes are d i rec t ly o x i d i z e d to cyanate and cuprous ions. T a n et a l . [160] be l ieved that copper cyanide complexes are first o x i d i z e d to cyanate releasing cuprous ions, w h i c h are o x i d i z e d to copper hydrox ide accord ing to their electrolyt ic products. B y e r l e y et a l . [142] observed that cuprous ions sufficient to complex 10 - 3 0 % o f total cyanide exhib i ted the best catalytic effect o n cyanide ox ida t ion at p H 10 - 11. Hofse th and C h a p m a n [168] reported that the cyanide concentrat ion can be reduced f rom 100 to 1 p p m i n a porous f low-through reticulated vitreous carbon cata lyzed by copper ions. Y o s h i m u r a and K a t a g i r i et a l . [144, 145, 147, 149, 150] measured the steady-state po la r iza t ion curves at a (CN)2 +2H20 = (CONH2)2 (2-29) (CN)2 +4H20 = C202- + 2 M 4 + (2-30) 19 p la t inum anode i n cyanide solutions containing a very sma l l amount o f copper ( C N : C u > 5) and 0.5 M K 2 S 0 4 as supporting electrolyte, and found that the Tafe l slope was about 0.158 V decade - 1 i n a l o w potential region, suggesting that a s imple one-electron react ion was occur r ing at the electrode. The current at a constant potential was propor t ional to the total cuprous i o n concentrat ion but it was almost independent o f the total cyanide concentrat ion. It was assumed that a l l o f the copper exists i n the fo rm o f C u ( C N ) 4 3 " wi thout check ing the dis t r ibut ion o f copper species. In fact, i n the ranges o f cyanide and copper concentrat ion studied by these authors, a significant amount o f copper exists i n the fo rm o f C u ( C N ) 3 2 " and their assumpt ion is not appropriate. The calculated react ion order w i t h respect to C u ( C N ) 4 3 " (actually Cu(I)) was 0.9. C u ( C N ) 4 2 _ was detected by E S R spectroscopy. It was thought that C u ( C N ) 4 3 ~ is o x i d i z e d to C u ( C N ) 4 2 _ , w h i c h is the rate-determining step. The f o l l o w i n g mechan i sm was proposed [150]: Cu(CN)34' -> Cu(CN)42~ + e (2-31) 2Cu(CN)2- <=> Cu2 (CN)2; + 2CN~ (2-32) Cu2 (CN)2- 2Cu(CN)2 + (CN)2 (2-33) Cu(CN)2 + 2CN~ -> Cu(CN)/- (2-34) (CN)2 + 20H~ -> C A T + OCN~ + H20 (2-35) H o w e v e r , no kinet ic data are g iven for a lkal ine copper cyanide solutions except for a po la r iza t ion curve i n 1 M K O H solut ion. The react ion products o f the anodic ox ida t ion o f cyanide at 0.6 and 1.2 V vs. S C E were determined. In a lka l ine solutions ( p H 11.8-14), the react ion can be expressed by Equa t ion 2-25 and cyanate i o n was not o x i d i z e d further. H i n e et a l . [156] studied the anodic ox ida t ion o f copper cyanide o n a P b 0 2 -coated anode and found that on ly copper exhibi ted a catalytic effect o n cyanide ox ida t ion . The Tafe l slope for the ox ida t ion o f the solut ion containing 1 M N a C N and 0.3 M copper was 0.070 -0.110 V decade - 1 i n the current density range o f 50 -1000 A m - 2 . The current ef f ic iency o f cyanide ox ida t ion decreased w i t h decreasing total cyanide concentrat ion at constant copper concentrat ion. It was thought that the f o l l o w i n g react ion was occur r ing : C u ( C N ) 3 2 - + 2 0 H " = C u ( O H ) 2 + 3 C N " + e (2-36) H w a n g et a l . [157] studied the electrolyt ic ox ida t ion o f copper cyanide so lu t ion w i t h C N : C u mo le ratios o f 2.8 to 20 and at different p H ' s us ing a p l a t inum anode. In s trongly 20 alkal ine solu t ion ( p H > 12), the copper-cyanide complex is o x i d i z e d di rect ly to cyanate and copper oxide . The f o l l o w i n g react ion sequence was proposed: Cu(CN)(;-iy + 2nOH~ = Of + nCNO~ + nH20 + 2ne (2-37) 2 C V + 20H~ = Cu20 + H20 (2-38) Cu20 + 20H' = CuO + 2H20 + 2e (2-39) H o w e v e r , the potential was contro l led at 0.71 and 1.2 V vs . S C E and so the o x y g e n evo lu t ion m a y have affected the coulometr ic measurement. A l s o , i n their experiment, the cathode and the anode were not separated. Therefore copper deposi t ion m a y have affected s ignif icant ly the mo le ratio o f cyanide to copper dur ing the course o f the experiment. I n neutral or w e a k l y a lkal ine or ac id ic solutions, the complex does not undergo the direct ox ida t ion , but dissociates to free cyanide due to copper deposi t ion and then free cyanide is o x i d i z e d o n the anode. Apparen t ly the above ox ida t ion procedure is not reasonable. I f the anode and the cathode are separated, the anodic ox ida t ion w i l l not happen. E v e n i f the anode and the cathode are not separated, at l o w C N : C u mole ratios (e.g. 3), the free cyanide released f rom the cathode w i l l immedia te ly bond to the l o w l y coordinated copper cyanide complexes ( C u ( C N ) 3 2 " and C u ( C N ) 2 " ) . Furthermore i f on ly free cyanide is o x i d i z e d , at C N : C u < 3, the concentrat ion o f free cyanide is so l o w (less than 1/1000 o f the total cyanide) that the cyanide ox ida t ion can be neglected. La te r H w a n g et a l . [166] adopted the direct ox ida t ion mechan i sm reported by their group [157] and the catalytic mechan i sm by K a t a g i r i et a l . [147, 150] and reported that the Tafe l slope increased f rom 0.040 V decadent C N : C u = 3 to 0.120 V d e c a d e 1 at C N : C u > 10. The anodic ox ida t ion o f copper cyanide undergoes both direct ox ida t ion and catalyt ic oxida t ion . A t C N : C u < 3, there is on ly the direct ox ida t ion and at C N : C u > 10, there is on ly the catalytic ox ida t ion . The anodic ox ida t ion o f copper cyanide resulted i n the format ion o f copper ox ide , w h i c h ca ta lyzed the ox ida t ion o f free cyanide [159, 163, 165, 167]. F r o m the above discuss ion, there are incomplete and conf l i c t ing results o n the anodic ox ida t ion o f copper cyanide i n alkal ine solut ion and the informat ion f rom the literature is insufficient for the present research, especia l ly w i t h respect to the des ign o f an e lec t rochemical process for C u - C N e lec t rowinning. 21 2.4 The Electrochemical Oxidation of Thiocyanate T h i o c y a n i c ac id exists i n two isomeric forms, H - S -C = N or H - N = C = S i n e q u i l i b r i u m w i t h each other and thiocyanate ions exist i n two tautomeric forms, " S - C ss TV (I) and S = C = N' (II) [169]. The redox react ion is : 2SCN- = (SCN)2 +2e E ° = 0.77 V vs. S H E (2-40) The structure o f th iocyanogen i s N = C - S - S - C = N . ( S C N ) 2 undergoes rap id hydro lys i s i n water to give S 0 4 2 " , H C N , and H + . The e lec t rochemical ox ida t ion o f thiocyanate has been studied for a l ong t ime both i n aqueous and non-aqueous med ia [130, 131, 170-204]. The products o f the ox ida t ion o f thiocyanate i n aqueous solut ion are sulphate and either cyanide or cyanate or, further, a m m o n i u m and carbonate or ni trogen depending on the p H o f the so lu t ion and the anodic potential . The ox ida t ion o f thiocyanate is i rreversible. In ac id solutions, the react ion can be expressed m a i n l y by the f o l l o w i n g equation [174, 177-180, 185, 192-195, 198, 199, 203] : SCN~ + 4H20 = SOl' + HCN + 7H+ + 6e E ° = 0.515 V vs. S H E (2-41) Other products are also formed, e.g. parathicyanogen ( S C N ) X , [177-179], a pass iva t ion f i l m C 6 N 4 S 4 , and ( S C N ) 3 " [192]. The in i t i a l step o f the anodic ox ida t ion m a y be the r emova l o f an electron f rom one thiocyanate i o n [185, 192]. H o w e v e r , L o u c k a et a l . [203] reported that the first step is the decompos i t ion into sulfur and cyanide and then the sulfur is o x i d i z e d to sulphate and the ox ida t ion o f thiocyanate occurs at potentials higher than 0.7 V vs . S H E . In basic solutions, the ox ida t ion reaction can be expressed as: [133, 170]: SCN' + WH' = SO]' + C N ~ +4H20 + 6e E ° = - 0.61 V vs. S H E (2-42) G a u g u i n [174-176] gave the f o l l o w i n g expressions for the potential o f S C N " oxida t ion : E = 0.57 - 0.058 l o g [ S C N ] f rom p H 0 to 7 and E = 1.17 + (0.058/6) l o g ( [ H + ] 8 / [ S C N " ] 6 ) f rom p H 9 to p H 14. The potential for C N " ox ida t ion is E = 0.10 - 0.058 l o g [ C N ] f rom p H 0 to 13 and E = 0.88 + (0.058/2) l og ( [H + ] 2 / [CN"] ) above p H 13. 22 The electrolyt ic convers ion o f thiocyanate to cyanide has been studied as a means o f regenerating cyanide and m i n i m i z i n g the consumpt ion o f cyanide i n hydrometa l lu rgy [23, 87, 130, 170-172, 189-191, 193-195]. The convers ion o f thiocyanate to cyanide is never complete and depends o n cyanide and thiocyanate concentrations, p H , potential , and anode materials. In 1911, C l e n n e l l [78, 170] reported that the product ion o f cyanide rose to a m a x i m u m and then ceased. I f the electrolysis was continued, the cyanide produced f rom thiocyanate ox ida t ion gradually d imin i shed and f ina l ly disappeared. C r o o k et a l . [171] investigated the electrolysis o f thiocyanate on graphite anodes and gave results at different current densities. Wi thou t the addi t ion o f K O H , no C N " was detected. Th i s was probably because H C N was formed and vo la t i l i z ed at l o w p H . The increase i n C N " was propor t ional to the decrease i n S C N " and K O H . K e r n [172] found that thiocyanate i n cyanide solutions reduced the consumpt ion o f cyanide i n the electrolysis and was converted into cyanide to some extent. Varen t sov and B e l y a k o v a [189-191] studied the e lec t rochemical ox ida t ion o f thiocyanate and cyanide at a ruthenium oxide or cobalt oxide coated t i tan ium anode and graphite. T h e y found that the relative rates o f thiocyanate and cyanide ox ida t ion depended on their concentrations and at higher concentrations o f thiocyanate, more thiocyanate was o x i d i z e d and less cyanide. The graphite anode favored the ox ida t ion o f thiocyanate. H o w e v e r , the graphite broke d o w n leading to contaminat ion o f the solut ion. O r o c o n Inc. reported that thiocyanate f rom barren leach solutions can be o x i d i z e d to C N ' and sulphate on graphite fibre [23]. N o current eff ic iency o f the anodic ox ida t ion o f thiocyanate was g iven . B y e r l e y and Enns [193-195] studied the e lect rochemical regeneration o f cyanide f rom thiocyanate at graphite anodes and found that the recovery o f C N " f rom thiocyanate increased w i t h decreasing p H . A t l o w p H thiocyanate is e lec t rooxid ized to produce cyanide i o n w h i c h is immedia te ly protonated by H + . The ac id ic anode boundary layer functions to preserve cyanide f rom rapid electrooxidat ion at the anode by conver t ing the cyanide i o n into H C N , the m u c h more d i f f i cu l t ly o x i d i z e d neutral protonated form. The p H should be kept b e l o w 4 to real ize the better convers ion o f thiocyanate into cyanide. 23 2.5 The Electrochemical Oxidation of Sulphite The anodic ox ida t ion o f sulphite has been studied over a w i d e range o f p H . The anodic behaviour changes w i t h p H due to the change i n the speciat ion o f sulphite. It is important to k n o w the dis t r ibut ion o f the sulphite species w i t h p H i n order to understand the anodic ox ida t ion o f S 0 3 2 \ In solut ion, sulphite exists i n the fo rm o f S 0 2 (aq), H S 0 3 ~ and S 0 3 2 ' w i t h the f o l l o w i n g equ i l ib r i a between these species [205]: S02(aq) + H20 = HSO; + H+ K, = 1.6 x 10"2 (25 ° C ) (2-43) HSO; +H20 = SO2' +H+ K 2 = 1.0 x 10"7 (25 ° C ) (2-44) S 0 2 (aq), H S 0 3 ' and S 0 3 2 ' species are predicted to predominate over the p H ranges < 1.8, 1.8 - 7 and > 7, respect ively. A t p H > 12, the dominant species i n solut ion is S 0 3 2 ' . The redox react ion o f sulphite -sulphate on graphite can be expressed by the f o l l o w i n g equations: SO2'+2e + 4H+= H2S03+H20 E ° = 0.158 V vs. S H E (2-45) SOl~ +2e + H20 = SO2' + 20H~ E ° = -0.936 V vs. S H E (2-46) S20 2' +2e = 2S02' E ° = 0.037 V vs. S H E (2-47) S 0 3 2 ' cannot be reduced cathodical ly , w h i l e H S 0 3 ' ( p H 6-3) m a y be reduced to dithionite S 2 0 4 2 ' [206]. The sulphite-sulphate redox systems are irreversible . The e lec t rochemical ox ida t ion o f sulphur d iox ide or sulphite has been studied and tr ied as an anode depolariser to reduce the overa l l c e l l voltage i n the produc t ion o f hydrogen and i n copper e lec t rowinning i n ac id sulphate m e d i u m [207-252]. H o w e v e r , there are o n l y a few reports o n the e lect rochemical ox ida t ion o f sulphite i n a lkal ine solutions [243, 253-255] . Sulphate and dithionate are formed dur ing the ox ida t ion o f sulphite i n a lka l ine and neutral solutions. The amount o f dithionate produced at the anodic surface was s h o w n to depend on operating condi t ions, namely the anode mater ial , its preparation, current density, so lu t ion p H , and the presence o f additives i n the electrolyte. Friessner et a l . [207,208] s tudied the ox ida t ion o f sulphite earlier and conc luded that the format ion o f dithionate takes place at higher potentials than that o f sulphate. E s s i n [209] reported that the addi t ion o f N H 4 F increases the anodic potential and this benefits the format ion o f dithionate. The anneal ing o f p la t inum leads to the format ion o f p la t inum oxide w h i c h favours the format ion o f dithionate. 24 Glasstone et a l . [210, 211] investigated the effect o f electrolysis condi t ions o n the y i e l d o f dithionate and found that dithionate y ie lds o f up to 30 % can be obtained o n n i c k e l or g o l d electrodes, whereas the y i e l d d i d not exceed 3 % for graphite electrodes. P r e l im ina ry anodic po la r iza t ion increases dithionate y i e l d f rom 22 to 3 3 % . In the current density range f rom 10 to 30 A / m 2 , the dithionate format ion rate does not change, but it decreases not iceably at b e l o w 10 A / m 2 . Di thionate does not fo rm at current densities above 300 A / m 2 . Increasing temperature i n the range o f 18 - 60 ° C had a litt le negative effect o n dithionate format ion. The sulphite concentrat ion has no effect on the dithionate y i e l d . The op t ima l p H value for the format ion o f dithionate is f rom 7 to 9. Rozen ta l et a l . [216] reported that the ox ida t ion o f sulphite i n ac id m e d i a takes place at m u c h smaller posi t ive potentials (about 0.7 V vs . S H E ) than the evo lu t ion o f o x y g e n and conc luded that the ox ida t ion takes place v i a the surface oxides o f p la t inum. L e z h n e v a et a l . [219] investigated the rate o f the ox ida t ion o f sulphur d iox ide o n g o l d and p la t inum-go ld a l loys and found that the presence o f water, cations, and anions near the metal surface sharply changes the properties o f the surface oxygen compounds . Therefore, data o n the properties o f surface oxygen compounds obtained by e lec t rochemical methods cannot a lways be used i n s tudying the mechanism o f sulphite oxida t ion . S h l y g i n et a l . [220, 221] studied the ox ida t ion o f sulphur d iox ide and sulphite at a p l a t inum electrode and conc luded that the anodic ox ida t ion o f sulphur d iox ide i n a c i d and neutral solutions takes place at l o w potentials ( 0.65-1.2 V S H E ) by a reversible electron-radica l mechan ism: The appearance o f adsorbed oxygen can complete ly stop the ox ida t ion by the electron-radical mechan i sm at above 1.2 V vs. S H E . The e lect rochemical ox ida t ion o f S 0 3 2 " and H S 0 3 " begins at 1.2 V vs . S H E and is i rreversible. The ions cannot be o x i d i z e d by the e lectron-radical mechan i sm. The i r ox ida t ion mechan i sm consists i n the addi t ion o f an O H rad ica l at re la t ive ly h i g h anodic potentials and the mechan i sm may be expressed by the f o l l o w i n g reactions: S02 + AH20<^ H2SOA + 2H}0+ + 2e (2-48) 2H20^> 2H20+ + 2e (2-49) 25 2H20+ + 2H20 -> 20H + 2H30 (2-50) S03~ + 20H -> SO,2' + H20 (2-51) Tarasev ich et a l . [239-240, 243] studied the ox ida t ion o f sulphite o n p l a t i num and carbon materials at 22 ° C . The anchoring o f the ac id ic oxides o n the surface o f the carbon materials decreases the react ion rate. The react ion order o f the e lec t rochemical ox ida t ion depends o n the sulphite concentration, being i n a l l cases less than 1. T h i s behaviour m a y be due to adsorpt ion effects. A t l o w concentrations o f sulphite, the coverage is l o w and the react ion rate is propor t ional to the concentration o f sulphite i n the bu lk solut ion. A t h i g h concentrations, the current is propor t ional to the concentrat ion to a fract ional power . In a lka l ine solutions, sulphite seems to be adsorbed to a lesser extent than i n a c i d so lu t ion and the react ion is first-order up to 0.1 M . The dependence o f the react ion rate o n p H p lays an essential role . The dE/dpH value for both pyrographite and activated carbon is c lose to -40 m V i n the range o f p H 0 - 7 and becomes zero i n the reg ion o f higher p H values. The 3E/31ogi value i n the case o f pyrographite amounts to ca. 150 m v decade"' for p H < 7, and increases up to ca. 280 m v decade" 1 for p H > 7. The shape o f the po la r iza t ion curves o n the activated carbon is w e a k l y dependent on the type o f anion. In the reg ion o f intermediate p H values, the curves exhibi t two or even three Tafe l slopes. The first slope i n the ac idic and neutral p H region is 35 to 50 m V decade" 1, whereas i n a lka l ine solutions it is 60 - 70 m v decade" 1. The e lect rochemical ox ida t ion o f sulphite to sulphuric ac id proceeds most l i k e l y v i a the mechan i sm i n v o l v i n g the direct loss o f an electron f rom the o x i d i z e d species. The dependence o f the react ion rate o n p H for carbon materials is due to a var ia t ion i n the compos i t ion o f the species w h i c h are subject to ox ida t ion (at p H < 1.8 H 2 S 0 3 , H S 0 3 \ at p H 1.8-7, H S 0 3 " , S 0 3 2 " and at p H > 7, S 0 3 2 " ) . The adsorbed species that are subject to ox ida t ion undergo deprotonation ( p H < 7): The s l o w step may invo lve the transfer o f the first and the second electron f r o m the adsorbed species: H2S03 -> HS03 ADS + H •+ (2-52) HSO; -»so32-ads + H~ •+ (2-53) 26 HSO;ads -> HSOf + e (2-54) S032~+ H20-+H2SQ4-+e (2-55) Hunger et a l . [253, 254] studied the e lect rochemical ox ida t ion o f sulphite (0.012 M -0.09 M ) o n a graphite electrode at p H 9 and 25 ° C and observed that the current gradual ly increased at about 0.2 V vs. S C E w i t h increasing electrode potential . A p o o r l y def ined current density plateau was observed i n the range 0.5-0.7 V vs. S C E . The onset o f o x y g e n started at 1.5 V vs. S C E . Based o n the K o u t e c k y - L e v i c h equation, they calcula ted the k ine t ic current at different sulphite concentrations and f ina l ly obtained react ion rate constants, react ion orders o f 0.68 and 1.34, and charge transfer coefficients o f 0.058 and 0.048 respect ively for natural graphite and graphite impregnated w i t h phenol . It shou ld be noted that the K o u t e c k y - L e v i c h equation is v a l i d on ly for the first order react ion and therefore their results are not conv inc ing . Brevet t and Johnson [255] studied the anodic ox ida t ion o f sulphite (0.02-0.18 M ) o n pure and doped P b 0 2 f i l m electrodes at 25 and 65 ° C i n a N a H C 0 3 / N a 2 C 0 3 buffer ( p H 10). T h e y obtained a react ion order o f -0.2 us ing the same method as Hunge r et a l . [253, 254] . The reason for their obta ining negative react ion order m a y be that the current was corrected by subtracting the background i n the absence o f sulphite w h i c h was m u c h smaller than that i n the presence o f sulphite and the K o u t e c k y - L e v i c h equation was not v a l i d for their ca lcu la t ion o f the k ine t ic current. S tankovic et a l . [256] reported that the concentrat ion o f sulphite ions and temperature greatly influence the react ion rate. The number o f transferred electrons for the s l o w step was nearly one. 2.6 The Electrochemical Oxidation of Methanol The catalytic e lect rochemical ox ida t ion o f methanol has been w i d e l y studied for about 70 years [257-280]. The react ion i n a lkal ine solutions can be wri t ten as: CH3OH+WH- = CO2' +6H20 + 6e E ° = - 0.895 V vs. S H E (2-56) 27 The best catalytic anode materials are p la t inum metals and their a l loys [258, 259, 270-275] . The ox ida t ion o f methanol is m a i n l y used i n fuel cel ls . M e t h a n o l has been studied for use i n metal e lec t rowinning for depolar izat ion [275-280]. A m o n g soluble fuels, methanol is the most pract ical to use i n an e lec t rowinning ce l l . Ve reecken et a l . [275] used methanol for z i n c e lec t rowinning and observed its ox ida t ion at a l o w potential o n a pla t inum-act ivated graphite anode. The electrode potential , however , started to drift upwards after some t ime, and eventual ly the react ion shifted to oxygen evolut ion . V i n i n g et a l . [276, 277] proposed the use o f a precious metal coated t i tanium anode to extend the catalytic ac t iv i ty o f the electrode. The anode materials for an ac id ic electrolyte are ma in ly plat inum-based and a R u 0 2 - b a s e d catalytic f i l m on t i tanium. There is no report o n the appl ica t ion o f methanol e lec t rochemica l ox ida t ion i n a lkal ine solutions i n electrometallurgy. 2.7 The Electrochemical Oxidation of Ammonia In aqueous solutions the oxida t ion o f a m m o n i a to ni trogen is on ly possible i n a lka l ine solutions and is dependent on the electrode materials and their pretreatment [281]. Therefore the study o f a m m o n i a ox ida t ion was conducted i n concentrated hydrox ide solut ions [282-291] . The react ion can be expressed as: 2NH3 + 60H~ = 6H20 +N2+6e E ° = -0.74 V vs . S H E (2-57) The best catalysts are p la t inum metals and their a l loys and these materials were studied for fuel c e l l appl ica t ion [281-289]. The anodic ox ida t ion o f a m m o n i a was also conducted o n a T i / T i 0 2 / R u 0 2 electrode [291]. D u e to the s l o w kinet ics for a m m o n i a ox ida t ion , ch lor ide i o n was used as a catalyst to ox id i ze a m m o n i a [292, 293] . 2.8 Summary Copper and cyanide can form three stable cuprous complexes (dicyanide , t r icyanide and tetracyanide) and their dis t r ibut ion depends o n the concentrations o f copper and cyanide and the mo le ratio o f cyanide to copper. C u p r i c cyanide complexes are not stable and rap id ly decompose and cyanide is ox id i zed . Copper has a catalytic effect o n the anodic ox ida t ion o f 28 cyanide. In a lkal ine solutions, cyanide is more readi ly o x i d i z e d than thiocyanate and the relative ox ida t ion rates are dependent on the ratio o f cyanide to thiocyanate concentrat ion, current density, temperature and anode materials. H o w e v e r , the si tuation c o u l d be different i n a copper cyanide solut ion. The anodic ox ida t ion o f sulphite and cyanide begins at an approximate ly potential . Howeve r , there are no data w h i c h afford a direct compar i son . F o r methanol , an anode w i t h a plat inum-based f i l m has to be used to decrease the overpotent ial for methanol ox ida t ion . The anode w i l l probably lose its catalyt ic effect w i t h t ime. A m m o n i a can be readi ly o x i d i z e d i n strongly alkal ine solut ion at a p l a t inum electrode. H o w e v e r , such a h i g h hydrox ide concentrat ion is not suitable for the copper-cyanide system. The copper deposi t ion f rom cyanide solut ion has been w i d e l y reported. H o w e v e r , most o f these reports focus on copper plat ing. The copper e lec t rowinning f rom cyanide solu t ion has not been studied extensively and the operating condi t ions should be op t imized . H o w e v e r , some condi t ions used for p la t ing can be appl ied to improve the eff ic iency o f copper e lec t rowinning . In order to obtain a h igh current eff ic iency o f copper deposi t ion, the temperature should be above 40 ° C , the copper concentrat ion should be above 50 g L " 1 and the C N : C u mo le ratio should be around at 3. The addi t ion o f thiocyanate can improve the copper cathodic current eff iciency. There is very litt le informat ion on copper electrodeposi t ion f rom dilute cyanide solutions. In order to get a reasonable current eff ic iency o f copper deposi t ion, porous h i g h surface area electrodes have to be used. The graphite fibre has a large surface area and has been used to remove metal ions f rom waste effluent eff icient ly. It is possible to use graphite fibre felt to deposit copper f rom dilute cyanide eff iciently. T o prevent cyanide ox ida t ion , a membrane c e l l should be used. F r o m the above discussions, the informat ion avai lable i n the literature is insufficient for this project and the further study must be done to develop a successful process. The anodic and cathodic behaviour o f copper cyanide is dependent o n the d is t r ibut ion o f the concentrations o f copper cyanide species. The first step toward understanding the anodic and cathodic behaviour o f copper cyanide is to k n o w the d is t r ibut ion o f copper cyanide species at different concentrations, p H ' s , and temperatures. The d is t r ibut ion o f copper cyanide species can be calculated us ing rel iable complex constants. Coppe r e lec t rowinning us ing an alternative anode react ion i n an und iv ided ce l l should be conducted 29 i n a m i n i - c e l l to select the best sacr i f ic ia l species. F i n a l l y us ing the best sacr i f ic ia l species, copper e lec t rowinning can be improved by changing the temperature and the compos i t ions o f the electrolyte. Therefore the anodic ox ida t ion o f the sacr i f ic ia l species, copper cyanide and their mix ture should be studied to (a) understand h o w the sacr i f ic ia l species l im i t s the anodic ox ida t ion o f cyanide and (b) provide some fundamental informat ion to further improve the copper e lec t rowinning process. A l t h o u g h the graphite fibre felt can be used effectively to deposit copper f rom very dilute solut ion, copper is more diff icul t to deposit f rom cyanide solut ion. A feas ibi l i ty test should first be done and then further research can be conducted to investigate the direct e lec t rowinning on a graphite felt cathode w i t h reference to copper concentrat ion, mo le ratio o f cyanide to copper and f l o w rate. 30 3. THERMODYNAMICS OF COPPER CYANIDE 3.1 Distribution of Copper Cyanide Species Copper cyanide species establish an equ i l i b r i um speciat ion (React ions 2-1 to 2-6). The corresponding equ i l i b r ium constants selected for 25 ° C are l is ted i n Tab le 3-1 [38, 44, 49] . In some cases, we have to k n o w the dis t r ibut ion o f copper cyanide species and the e q u i l i b r i u m potentials for C u ( I ) / C u to understand copper deposi t ion and cyanide ox ida t ion at higher temperatures. H o w e v e r , so far the publ i shed data are inadequate for such a study. Therefore addi t ional data must be generated by calcula t ion. The A H ° values for React ions 2-1 and 2-3 to 2-6 are 128, -121.8, -46.4, -46.9 and 43.6 k J m o i ' 1 respect ively, the absolute values o f w h i c h are larger than 40 kJ /mole . A s s u m i n g that A H ° is approximate ly constant i n the range o f 25 - 60 ° C , we can calculate the equ i l i b r ium constant us ing the equation: d l n K / d T = A H ° / R T 2 [294]. Some calculated constants are l isted i n Table 3-1. Table 3-1 E q u i l i b r i u m constants for copper cyanide system [38, 44, 49 , 57] T e m p e r a t u r e ( ° C ) K a K s p P2 K 2 , 3 K 3 j 4 25 6 . 1 7 x l 0 - 1 0 l . O x l O ' 2 0 l . O x l O 2 4 2 . 0 0 x l 0 5 31.63 40 1.43X10"09 8 .44x10 ' 2 0 9 . 4 7 x 1 0 2 2 8 . 1 4 x l 0 4 12.77 50 2 .40x10" 0 9 5 . 3 3 x l 0 " 1 9 2 . 2 2 x 1 0 2 2 4 . 6 9 x l 0 4 7.317 60 3 . 9 1 x l 0 ' 9 2 . 2 7 x l 0 ' 1 8 5 . 6 1 x l 0 2 1 2 . 7 9 x l 0 4 4.333 The concentrat ion distr ibutions o f these species are dependent o n p H , temperature and the total concentrations o f copper and cyanide. The mass balances o f the copper and cyanide species are described by the f o l l o w i n g equations: [Cu( I ) ] T o t a l = [ C u + ] + [ C u ( C N ) 2 - ] + [ C u ( C N ) 3 2 - ] + [ C u ( C N ) 4 3 - ] (3-1) [ C N ] T o f t l l = [ C N - ] + [ H C N ] + 2 [ C u ( C N ) 2 - ] + 3 [ C u ( C N ) 3 2 - ] + 4 [ C u ( C N ) 4 3 - ] (3-2) 31 B y cons ider ing the equ i l ib r i a (Reactions 2-1 to 2-6) and so lv ing the above equations for the mass balance o f these species, the dis t r ibut ion o f copper cyanide species has been calculated. Since the exact values o f the equ i l i b r ium constants used to calculate the concentrat ion dis t r ibut ion are not suff iciently accurate and the parameters to calculate the ac t iv i ty coefficients o f a l l the species are not avai lable , the act ivi ty coefficients have not been considered i n this study. Therefore the calculated values should be interpreted as ind ica t ing trends rather than absolute values. Howeve r , the va l id i ty o f the pred ic t ion is conf i rmed by the exper imental potential measurements reported i n the next section. F igures 3-1 and 3-2 show the cyanocuprate dis t r ibut ion and the redox potential for C u ( I ) / C u vs. mo le ratio o f total cyanide to copper at p H 9 and 12. There is the f o l l o w i n g relat ion between the potential and the act ivi ty o f cuprous ions: E(Cu( I ) / C u ) = E ° ( C u ( I ) / C u ) + Y " l n ( a C u + ) (3-3) where E ( C u ( I ) / C u ) is the equ i l i b r ium potential for the C u ( I ) / C u couple , E ° ( C u ( I ) / C u ) the standard potential (0 .521, 0.520, 0.5195, and 0.519 V vs. S H E respect ively for 25 , 40 , 50 and 60 ° C , w h i c h were calculated us ing the data f rom the literature [295]). The other symbols have their c o m m o n meanings. Therefore the potential reveals the ac t iv i ty (or concentration) o f cuprous ions. The dis t r ibut ion o f the copper cyanide species depends m a i n l y o n the m o l e ratio o f total cyanide to copper and also o n the concentrat ion o f total copper and the p H . A t C N : C u mo le ratio < 3, the dis t r ibut ion o f the cyanocuprate species depends o n the C N : C u mo le ratio, and less o n the concentration o f copper at p H > 9. The dominant species are copper t r icyanide and dicyanide , and copper tetracyanide can be neglected. A t a mole ratio o f cyanide to copper = 3, copper t r icyanide dominates and most o f copper exists i n the f o r m o f t r icyanide. A t a mo le ratio o f cyanide to copper > 3, the d is t r ibut ion o f the copper-cyanide species depends o n the C N : C u mole ratio, the total copper concentrat ion and p H . F o r example , A t [Cu( I ) ] X o t a l = 0.001 M , copper t r icyanide dominates and s l o w l y decreases w i t h increasing C N : C u mo le ratio and p H . A t [Cu(I ) ] X o t a , = I M and C N : C u mo le ratio = 3-4, t r icyanide dominates and decreases greatly w i t h increasing C N : C u mole ratio and s l o w l y 32 w i t h increasing p H . W i t h further increase i n the mole ratio o f cyanide to copper, tetracyanide is dominant . 0.0010 T 0.0009 - j. 0.0008 -. 2 0.0007 --c o 0.0006 .= re 0.0005 --c a> 0.0004 o c o 0.0003 -o 0.0002 -, 0.0001 -I 0.0000 -• 0.0 -0.1 -0.2 > -0.3 LU X -0.4 V) -0.6 o -0.7 | -0.8 Si LU 0.9 i -1.0 14 Mole ratio of cyanide to copper ( a ) [ C u ( I ) ] T o t a l = 0 . 0 0 1 M , p H 9 12 14 to copper Mole ratio of cyanide (b) [Cu( I ) ] T o t a l = 0 . 0 1 M , p H 9 F igure 3-1 Copper cyanide species dis t r ibut ion and E ( C u ( I ) / C u ) vs. mo le ratio o f cyanide to copper for var ious solutions at 25 ° C and p H 9 33 0.10 -, r 0.0 2 4 6 8 10 12 14 Mole ratio of cyanide to copper (c) [Cu( I ) ] T o t a l = 0 . 1 M , p H 1 2 2 4 6 8 10 12 14 Mole ratio of cyanide to Copper (d) [ C u ( I ) ] T o t a ] = l M , p H 12 F igure 3-2 Copper - cyanide species dis t r ibut ion and E ( C u ( I ) / C u ) vs . mo le ratio o f cyanide to copper for var ious solutions at 25 ° C and p H 12. F igure 3-3 shows the dis t r ibut ion o f copper cyanide species vs. the mo le ratio o f cyanide to copper at 60 ° C . Compared to F igure 3-2b and d (25 ° C ) , at C N : C u mo le ratio < 3, the d is t r ibut ion o f copper cyanide almost does not change. A t a C N : C u mo le ratio > 3, the dis t r ibut ion shifts to l o w l y coordinated complexes to some extent w i t h increas ing temperature due to the decrease i n the stabili ty constants o f copper cyanide complexes . 34 0.010 , ,- 0.0 1.0 -, r 0.0 Mole ratio of cyanide to copper Mole ratio of cyanide to copper (a) [Cu(IVJ = 0.01 M and p H 12 (b) [Cu(I)] = 1 M and p H 12 F igure 3-3 Copper cyanide species dis tr ibut ion and E ( C u ( I ) / C u ) vs. mole ratio o f cyanide to copper for var ious solutions at 60 ° C and p H 12 The redox potential for C u ( I ) / C u decreases w i t h increasing ratio o f total cyanide to copper and to some extent w i t h increasing p H . F r o m Figure 3-4 a, w i t h increasing C N : C u mo le ratio, the redox potential for C u 7 C u decreases greatly at a C N : C u mo le ratio < 4 and decreases re la t ively s l o w l y at a C N : C u mole ratio > 4. A t a C N : C u mo le ratio < about 3, the higher the total copper concentration, the higher the redox potential for C u ( I ) / C u . A t a C N : C u ratio = 3, the redox potential is almost independent o f the total copper concentrat ion. A t a C N : C u mo le ratio > about 3, the higher the total C u + concentrat ion, the l ower the redox potential . F igure 3-4 b shows the redox potential for C u ( I ) / C u vs. p H at [Cu( I ) ] T o t a , = 0.1 M and different C N : C u mole ratios. The effect o f p H on the redox potential depends o n the C N : C u mo le ratio and p H range. Increasing p H is s imi la r to increasing free cyanide concentration, because at a h igher p H , less hydrogen ions compete for C N ' w i t h copper to fo rm H C N . 35 Figure 3-4 (a) E ( C u ( I ) / C u ) vs. mo le ratio o f cyanide to copper at 25 ° C , p H 12 and different copper concentrat ion and (b) E (Cu( I ) /Cu) vs. p H at 25 ° C , 0.1 M C u and different mo le ratios o f cyanide to copper In the direct e lec t rowinning process for copper deplet ion f rom solutions, the cyanide concentrat ion is mainta ined at a constant value and the copper concentrat ion changes due to copper cathodic deposi t ion. F o r example , copper concentrat ion decreases f rom 2 g L " 1 (or 1 g L" 1 ) to 1 g L " 1 (or 0.5 g L" 1 ) due to copper deposi t ion o n the graphite felt electrode and the cyanide concentrat ion is kept at 2.445 g L " 1 (or 1.228 g L " 1 ) . Therefore it is necessary to k n o w the dis t r ibut ion o f copper cyanide species at a constant cyanide concentrat ion and different copper concentrations. Figures 3-5 and 6 show the concentrat ion d is t r ibut ion o f copper cyanide species at [CN"] = 0.09442 M (2.455 g L" 1 ) and 0.04721 M (1.228 g L" 1 ) respect ively . F r o m Figure 3-5, w i t h decreasing total copper concentrat ion f rom 2 to 1.2 g L " 1 , the concentrations o f d icyanide and t r icyanide decrease. H o w e v e r , the tetracyanide concentrat ion increases. The calculated redox potential for C u ( I ) / C u decreases q u i c k l y . W i t h further decrease i n the total copper concentration, a l l copper cyanide species decrease and the redox potential for C u ( I ) / C u decreases. A s imi la r trend is shown i n F igure 3-6. The stabil i ty o f the copper-cyanide solu t ion depends not on ly o n the ratio o f total cyanide to copper, but also on the concentrations o f total copper, p H and temperature. F o r example , the c r i t i ca l cyanide concentrations for stable solutions conta ining I M Cu(I ) are 2.8, 2.7, 2.6 and 2.5 M for 25, 40, 50 and 60 ° C respectively, and i f the cyanide concentrations are 36 l ower than the above values, the product o f the equ i l i b r ium [ C N ] and [Cu + ] w i l l be larger than the K s p o f C u C N and C u C N w o u l d precipitate. The product o f [Cu + ] and [ C N ' ] for 0.0021 M cyanide and 0.001 M copper solut ion is less than the K s p o f C u C N . Therefore the so lu t ion is stable. The lower the total cyanide concentration, the lower the c r i t i ca l mo le ratio o f cyanide to copper. Total copper concentration / g L" (a) 25 ° C Total copper concentration / g l -ib ) 40 ° C F igure 3-5 Copper concentrations i n the fo rm o f copper complexes and the e q u i l i b r i u m potential vs . total copper concentration at [CN"] = 2.455 g L " 1 and [OH' ] = 0.01 M . (a) 25 ° C (b) 40 ° C F igure 3-6 Coppe r concentrations i n the fo rm o f copper complexes and the e q u i l i b r i u m potential vs . total copper concentrat ion at [CN"] = 1.227 g L " 1 and [OH"] = 0.01 M . 37 3.2 The Equilibrium Potential Measurement of Copper Cyanide The cuprous equ i l i b r ium potential can be expressed by Equa t ion 3-3. The ac t iv i ty o f the cuprous i o n depends on the dis t r ibut ion o f the cyanide copper species. Therefore E ( C u ( I ) / C u ) is a funct ion o f the copper cyanide associat ion constants at constant temperature, p H , and the copper and cyanide concentrations. W e can evaluate the va l id i t y o f the ca lcula ted value us ing thermodynamic constants by compar ing the calculated e q u i l i b r i u m potentials to the measured values for different composi t ions . It is very important to conduct the measurement o f the equ i l i b r i um potentials. There are many var ied reports o n the potential measurement for the C u ( I ) / C u couple i n copper cyanide solut ion [31, 32, 42 , 46, 47] w i t h different measur ing methods and condi t ions. In this thesis, a few measurements o f copper cyanide e q u i l i b r i u m potentials were made to conf i rm the calculated values i n Sec t ion 3.1. 3.2.1 Experimental Equipment : The copper cyanide solutions were p laced i n a 1 0 0 - m L airtight water-jacketed electrolyt ic c e l l whose temperature was mainta ined at constant ( ± 0.2 ° C ) us ing a water bath circulator . The solutions were rendered free o f o x y g e n by bubb l ing w i t h h igh ly pure argon gas w h i c h passed a F I S H E R O X I C L E A R gas purif ier to reduce o x y g e n to b e l o w 5 ppb. The copper electrode was a 2 - m m diameter 99 .999% pure copper wi re w h i c h was first po l i shed by s i l i c o n carbide sand, then washed w i t h acetone and f ina l ly put i n 0.01 M pure sod ium cyanide solutions at p H 10 awai t ing for use. A Solar t ron 1286 e lec t rochemical interface was used to measure the potential between the copper w i r e and the saturated c a l o m e l reference electrode and the potential data over t ime were recorded by a computer. The experiment set-up is s h o w n i n F igure 3-7. The l i q u i d j unc t ion potential , estimated by the Hender son equat ion ( A p p e n d i x 7), is less than 2 m V and negl ig ib le . Reagents: 99.99%) sod ium cyanide, 99 .99% copper cyanide, standard 1 M N a O H solu t ion and ultrapure de ionized water were used to prepare the required copper cyanide solutions conta in ing 0.01 M N a O H . 38 Oxyclear gas purifier Solatron voltmeter Computer F igure 3-7 Exper imen ta l set-up for the equ i l i b r ium potential measurement 3.2.2 Results and Discussion A n y o x y g e n i n the solut ion has a significant effect o n the potential measurement. F igure 3-8 shows the electrode potential decreasing w i t h cont inued A r gas bubb l ing and s tab i l iz ing after 3 hours. In general, 3 hours were required to s tabi l ize the potential and so the f ina l va lue was taken after 3 hours. Figures 3-9 and 3-10 show both the ca lcula ted and measured potentials vs. the mole ratio o f copper to cyanide at 25 , 40 , 50 and 60 ° C for the solutions conta ining 0.1 M and 0 . 0 I M copper. A t a C N : C u mo le ratio < 4, the measured potentials are a litt le higher than the calculated values and the differences between the measured and calculated potentials are i n the range o f 5-20 m V for 0.1 M and 1 0 - 2 5 m v for 0.01 M C u . Th i s difference might be caused by a trace amount o f o x y g e n and the so lu t ion ion ic strength or change i n the concentration equ i l i b r ium constant. The exchange current for the lower concentrat ion is lower than that for the higher concentrat ion and m a y be easi ly affected by some factors such as oxygen and hydrogen ions. Therefore the difference for the so lu t ion conta in ing 0.01 M C u is larger than that for the solut ion conta ining 0.1 M C u . A t a C N : C u mo le ratio > 4, the difference between the calculated and measured potentials became larger. The reason c o u l d be that at 0.01 M O H - , the hydrogen potential is about 0.70 V vs. S H E and m u c h higher than the potential for C u ( I ) / C u . Therefore the measured potential 39 might be a m i x e d potential . So us ing the measured potentials to evaluate or calculate the e q u i l i b r i u m constants may be inappropriate. The equ i l i b r i um constants obtained by Ro thbaum [42] and H a n c o c k [46] us ing the potentials measured at C N : C u mole ratio > 4 are less rel iable i n spite o f the h i g h overpotential o f hydrogen o n copper. The potential trend w i t h C N : C u mo le ratio, total copper concentrat ion and temperature is the same as that predicted by ca lcula t ion . F o r example , at a C N : C u mole ratio < 3, the potential decreases w i t h increasing temperature and increases w i t h increasing copper concentration. A t a C N : C u mole ratio = 3-4, the potential is less dependent on the temperature and concentration. A t a C N : C u mole ratio > 4, the potential increases w i t h increasing temperature and decreases w i t h increasing copper concentrat ion. The above dependence o f the equ i l i b r ium potential for C u ( I ) / C u o n the temperature and C N : C u mole ratio is s imi la r to those measured i n 0.5 to 0.4 M C u + solutions w i t h C N : C u mo le ratio = 2.4 - 40 at 20 and 80 ° C [42] and i n 0.15 M C u + solutions w i t h C N : C u mo le ratio = 2.9 - 4.03 i n the temperature range 10 to 50 ° C [47]. F r o m the above statements, it w o u l d appear that the use o f the ci ted equ i l i b r ium constants to calculate the d is t r ibut ion o f copper cyanide species w i l l not result i n a significant error. F igure 3-8 Elec t rode potential vs . t ime at 25 ° C , C N : C u mole ratio = 3 and [ C u ] t o t a ] = 0.1 M 40 Mole ratio of cyanide to copper Figure 3-9 Elec t rode potential vs . the mole ratio o f cyanide to copper at 25 , 40 , 50 and 60 ° C , [ C u ] t o t a l = 0.1 M and [OH"] = 0.01 M . F igure 3-10 Elect rode potential vs . the mole ratio o f cyanide to copper at 25, 40 , 50 and 60 ° C , [ C u ] t o t a l = 0.01 M and [OH"] = 0.01 M . 41 3.3 Potential-pH Diagrams for Copper Cyanide In Sec t ion 3.1, the dis t r ibut ion o f copper cyanide species has been discussed. H o w e v e r , since the stabil i ty o f the copper cyanide species is related to the potential and p H , po ten t ia l -pH diagrams are required to discuss the stabil i ty o f the copper cyanide species. Po ten t i a l -pH diagrams show w h i c h species are stable at a f ixed species concentrat ion, potential and p H . Because the stabil i ty o f copper cyanide changes w i t h concentrat ion, the poten t ia l -pH diagrams for the different species concentrations should be used. The free energy data p rov ided by B a r d et a l . [296] are thought to be the most re l iable and therefore the free energy data for copper and cyanide are ci ted f rom this source. H o w e v e r , the data for copper d icyanide are questionable because its free energy was calcula ted f rom the stabi l i ty constant ((32) reported by Kappens te in and H u g e l [48] w h i c h is on ly 1 0 1 6 7 and m u c h smal ler than the format ion constant ( K ^ ' X I O 2 0 ) o f C u C N . Th i s value was discussed i n Chapter 2 and considered to be unrel iable. A c c o r d i n g to the free energy data o f d icyanide and t r icyanide reported by B a r d et a l . [296], K 2 3 is 10 1 1 ' 7 , m u c h larger than 1 0 5 3 the value w h i c h is considered to be most rel iable . Therefore i n this study, the free energy data for d icyan ide , t r icyanide and tetracyanide have been calculated f rom the free energy data for C u + , C N " and equ i l i b r i um constants ( P 2 =10 2 4 , K 3 = 1 0 5 3 and K 4 = 1 0 1 5 ) o f the copper cyanide complexes . The free energy data for a l l species are l isted i n Table 3-2. Table 3-2 G i b b s free energy data for copper and cyanide species (J moi" 1 ) at 25 ° C [38, 44, 4 9 , 2 8 4 ] Cu Cu+ Cu^+ Cu 2 0 CuO Cu(OH)2 HCu0 2 -0 50,300 65,700 -148,100 -134,000 -359,500 -258,900 Cu022" H 2 0 H+ H 2 0 2 CN" HCN -183900 -237178 0 0 0 166,000 113,423 CNO- HCNO (CN)2 CuCN Cu(CN)2- Cu(CN)3^- Cu(CN)4^" -98700 -12,100 296,300 102,126 245,291 381,035 538,471 O n the basis o f the change i n G i b b s free energy, C u O is more stable than C u ( O H ) 2 . H o w e v e r , C u ( O H ) 2 m a y exist or coexist w i t h C u O . Therefore both C u O and C u ( O H ) 2 are considered i n po ten t ia l -pH diagrams. F igure 3-11 shows the po ten t ia l -pH d iagram for the C N - H 2 0 system assuming that C N " , C N O " , H C N , H C N O and ( C N ) 2 are stable, a l though a l l o f them are not stable. I n the h igh potential range, C N " and H C N are not stable and are o x i d i z e d 42 i n accordance w i t h thermodynamics . Howeve r , H C N and C N " are metastable and the potentials for the ox ida t ion o f H C N and C N " are m u c h higher (1.0-1.2 V ) than those s h o w n i n F igure 3-11. Therefore C N " and H C N are considered to be stable i n the C u - C N - H 2 0 po ten t ia l -pH diagram. Figures 3-12 and 13 show the C u - C N - H 2 0 po ten t i a l -pH diagrams at the act ivi t ies o f a l l o f the solute species = 1, 10"2, 10"4 and 10"6 assuming C u O , C u ( O H ) 2 and C N " are stable. F r o m these two diagrams, at the activit ies o f a l l o f the solute species = 1, C u C N , C u ( C N ) 3 2 " and C u ( C N ) 4 2 " are stable i n the three regions. A t the act ivi t ies o f a l l o f the solute species = 0.01 and 0.0001, C u C N , C u ( C N ) 2 " and C u ( C N ) 3 2 " are stable i n the three p H regions. A t the activit ies o f a l l o f the solute species = 0 .000001, on ly C u C N and C u ( C N ) 2 " are stable. F r o m Figure 3-14, at the activit ies o f the copper solute species = 0.01 and the activit ies o f cyanide species = 0.1, a l l copper cyanide species are stable i n their corresponding p H regions. Copper cyanide species are stable i n certain potential and p H regions. W i t h increasing potential , copper cyanide w i l l be o x i d i z e d to C u 2 + , C u O ( C u ( O H ) 2 ) and C u 0 2 " . Cyan ide can also be o x i d i z e d to cyanate f rom Figure 3-11. Coppe r cyanide complexes can be o x i d i z e d to copper ox ide and cyanate f rom the point o f v i e w o f thermodynamics . F igure 3-11 C N - H 2 0 poten t ia l -pH diagram at a l l solute species act ivi t ies o f 1 and P ( C N ) 2 = 1 a tm and 25 ° C . (a) assuming H C N O and C N O " are stable and (b) assuming ( C N ) 2 is stable. 43 Figure 3-12 Po ten t i a l -pH diagrams for C u - C N - H 2 0 system at 25 ° C and the act ivi t ies o f a l l solute species = 1, 10"2, 10 - 4 and 10"6 cons ider ing C u O as a stable species. H C N O , C N O " and ( C N ) 2 are not considered. 44 Figure 3-13 Po ten t i a l -pH diagrams for C u - C N - H 2 0 system at 25 ° C and the act ivi t ies o f a l l solute species = 1, 10"2, 10"4 and 10"6 consider ing C u ( O H ) 2 as a stable species. H C N O , C N O " and ( C N ) 2 are not considered. 45 PH Figure 3-14 Po ten t i a l -pH diagram for C u - C N - H 2 0 system at 25 ° C and solute copper species act ivi t ies o f 0.01 and cyanide species activit ies o f 0.1 cons ider ing C u ( O H ) 2 as a stable species. H C N O , C N O " and ( C N ) 2 are not considered. 3.4 Summary The distr ibutions and equ i l i b r ium potentials o f copper cyanide species are functions o f the mo le ratio o f cyanide to copper, total cyanide concentration, p H and temperature. W i t h increasing C N : C u mo le ratio, the dis t r ibut ion o f copper cyanide species shifts more comple te ly to the h igh ly coordinated complex ( C u ( C N ) 4 3 " ) at a h i g h cyanide concentrat ion than that at a l o w cyanide concentration. The equ i l i b r i um potential for Cu( I ) / C u decreases w i t h increasing C N : C u mole ratio. Increasing p H is s imi la r to increasing free cyanide concentrat ion. Increasing temperature results i n decreasing stabil i ty constants. Therefore the dis t r ibut ion o f copper cyanide shifts to the l o w l y coordinated complexes . The potential measurements have conf i rmed the va l id i ty o f the calculated results. In the p H -potential diagrams, C u C N , C u ( C N ) 2 " , C u ( C N ) 3 2 " and C u ( C N ) 4 3 " can predominate i n the different p H regions. F r o m the above discuss ion, it is expected that C u deposi t ion current e f f ic iency decreases w i t h increasing C N : C u mole ratio and increases w i t h increasing temperature. The change i n the dis t r ibut ion o f copper cyanide m a y affect its anodic behaviour . 46 4. ELECTRODEPOSITION OF COPPER ON GRAPHITE FELT FROM DILUTE CYANIDE SOLUTIONS Porous 3-dimensional electrodes such as carbon felt and c lo th , ret iculated vi treous carbon and metal mesh are being used increasingly i n e lect rochemical process ing due to their h igh area per unit electrode vo lume and their moderately h i g h mass transport characteristics. One o f their applicat ions is to recover and remove metals f rom dilute waste water because 2-d imens iona l electrodes (e. g. planar) are inefficient for this appl ica t ion [297-315]. N o careful study o n the electrodeposi t ion o f copper f rom dilute cyanide solu t ion has been reported [22-25] . In this chapter, a careful study o f direct e lec t rowinning o f C u o n a graphite fibre electrode is reported. Copper complexed w i t h cyanide is m u c h more di f f icul t to deposit f rom dilute solut ion. E s p e c i a l l y when the C N : C u mole ratio is h igh , the e q u i l i b r i u m potential for the C u ( I ) / C u couple is m u c h lower than the equ i l i b r i um potential for H 7 H 2 and so hydrogen evo lu t ion w i l l s ignif icant ly decrease the current eff ic iency. Graphi te fibre has a h i g h surface area, g i v i n g a m a x i m u m pla t ing area for copper deposi t ion and m i n i m i z i n g the overpotent ial for copper p la t ing and the concentrat ion polar iza t ion . Graphi te also has a re la t ive ly h i g h overpotential for hydrogen evolu t ion w h i c h should m a x i m i z e the current e f f ic iency o f copper deposi t ion i n the in i t i a l deposi t ion stage. Therefore i n this study, graphite fibre felt was used as the porous cathode. 4.1 Some Fundamental Aspects of Graphite Fibre Electrodes Figure 4-1 shows the schematic d iagram o f a one-dimensional porous electrode. The e lec t rochemical react ion takes places i n the porous electrode. A consequence o f electroneutrali ty is that the charge is conserved between the porous electrode mat r ix and pore-solu t ion phases. The f o l l o w i n g equation must be appl ied: Is+I,=0 (4-1) where I s is the matr ix current density and I, the solut ion current density. 47 Metal^backing I Porous electrode Metal backing I x = 0 x = L (a) Figure 4-1 Schemat ic d iagram o f porous electrode x = 0 x = L (b) In the porous electrode, at x = x, the potential difference ( O ) between the so l id phase (<PS) and the so lu t ion phase (O,) is O s - O , . The increase i n O (dO) due to the increase i n the distance (dx) is: dO = d(<$>s -O,) = d<$>s -d<b, = —dx--dx cr K (4-2) where a is the effective conduct iv i ty o f the so l id phase and K the effective conduc t iv i ty o f the solut ion. F r o m Equa t ion 4-2, the f o l l o w i n g equation can be der ived: Is / , J O --- = -r (4-3) a K dx The increase i n the s o l i d phase (d l s ) or the l i q u i d phase (dl,) is due to the e lec t rochemica l react ion o n the interface between the so l id phase and the solut ion. Therefore we have: dls = -di, = Taidx (4-4) where a is the specif ic area ( m 2 / m 3 ) , i the loca l Faradaic current densi ty o f the e lec t rochemical react ion o n the surface ( A m"2) (negative for the cathodic process and posi t ive for the anodic process) and the s ign - for F igure 4 - l a and the s ign + for F igure 4 - l b . F r o m Equa t ion 4-4, the f o l l o w i n g equation can be der ived: dx dx F r o m Equat ions 4-3 and 4-5, we have the fo l l owing equation: (4-5) d2G> 1 1 2 = +a(— + —)i dx2 ~ V KJ' (4"6> In the case o f copper deposi t ion f rom cyanide solutions, the f o l l o w i n g equat ion can be appl ied: 48 H (4-7) dC, Cu ai, Cu dx = + F (4-8) where i the loca l Faradaic current density o n the electrode surface, i C u copper depos i t ion current density ( A m" 2), i H the hydrogen evolu t ion current density ( A m" 2) and C C u the copper concentrat ion ( M ) . The overpotential (n) can replace <3> because O can be expressed as (n + const.). F r o m the above equations, the distributions o f the potential and current are non-un i fo rm due to the resist ivit ies o f the fibre and the electrolyte. In the case o f the copper deposi t ion, the d r i v i n g force (I <t»s - O, I) o f copper deposi t ion decreases l o c a l l y to a value so l o w that copper deposi t ion stops. In order to remove more copper, the potential difference must be increased. H o w e v e r , this may result i n more hydrogen evo lu t ion and lower copper current eff ic iency. Signif icant hydrogen evo lu t ion can b l o c k the electrolyte f rom the fibre, stop copper deposi t ion and dramat ical ly increase the effective resis t ivi ty o f the electrolyte. Z a m y a t i n and B e k [310] studied the effect o f hydrogen evo lu t ion o n g o l d depos i t ion i n graphite fibre felt and found that the current eff ic iency decreased w i t h increas ing total current (potential difference) and the deposi t ion rate o f g o l d first increased to a m a x i m u m value and decreased w i t h increasing current due to hydrogen evolu t ion . The copper deposi t ion also depends o n the electrolyte compos i t ion , temperature and f l o w rate (mass transfer). The m a x i m u m potential difference between the fibre and the electrolyte or- the m a x i m u m current is selected by experiment w i t h reference to the electrolyte compos i t i on and temperature. The thickness o f the fibre electrode is determined by the desired extent o f copper r emova l f rom the electrolyte and the m a x i m u m potential difference between the fibre and the electrolyte [297]. B e y o n d a certain thickness, the electrode s i m p l y adds a barren zone where no copper deposi t ion w i l l take place. In the case o f p la t ing, supporting electrolytes are used to increase the conduc t iv i ty to obtain a un i fo rm copper deposi t ion. F o r example , B e k and Ze reb i l ov [125] deposi ted a th in layer o f copper o n carbon fibres us ing 0.01 M C u + + 0.03 M C N " so lu t ion conta in ing 1 M N a j S C ^ and 0.5 M N a 2 S 0 3 as supporting electrolytes. M a s s transfer i n graphite fibre felt is important to be able to predict the effect o f f l o w rate (ve loci ty) o n copper deposi t ion eff iciency. There are several reports o n the mass transfer 49 i n graphite fibre [316-320]. B e k and Z a m y a t i n [316] reported the f o l l o w i n g relat ions for f low-through fibre w i t h 10 u m diameter: k m = 1 . 9 0 x 1 0 - 2 u ° 3 5 2 ( 0.02 < R e < 0.15) (4-9) where k m is the mass transfer coeff ic ient(cm s"1), u the ve loc i ty o f the l i q u i d ( c m s"1), R e the R e y n o l d s number (ud/v), d the fibre diameter (cm) and v the k inemat ic v i scos i ty ( c m 2 s). Transformed into dimensionless form, Equa t ion 4-7 reads S h = 6 . 1 R e 0 3 5 2 (0.02 < R e < 0.15) (4-10) where S h is the She rwood number ( k m d /D) and D the di f fus ion coefficient. S c h m a l et a l . [318] gave the f o l l o w i n g relat ion for the single fibre w i t h 8 -um diameter: S h = 7 R e 0 4 (0.04 < R e < 0.2) (4-11) The above re la t ion was consistent w i t h the results der ived f rom heat transfer. The S h value for f l o w paral le l to the fibre is 40 % lower than that for f l o w perpendicular to the fibre. K i n o s h i t a and L e a c h [317], Vatis tas et a l . [319] and Car ta et a l . [320] studied the mass transfer for f l ow-by fibre felts and their S h numbers are smaller than that for the f l o w -through fibre reported by B e k and Z a m y a t i n [316]. The compress ion o f fibre felt also s ignif icant ly changes its conduc t iv i ty w h i c h depends p r i m a r i l y on the contact resistance between fibres. The degree o f mat r ix compress ion is accounted for by the change i n porosi ty and the mat r ix conduc t iv i ty can be calcula ted approximate ly by the correlat ion [321]: o-= 10 + 2800(1 -e/e0)155 (S m"1) (0.68 < e/e 0 < 1 at 20 ° C ) (2-12) where e is the porosi ty o f the matr ix and e 0 the in i t i a l porosi ty o f the matr ix . The conduct iv i ty o f a typ ica l aqueous electrolyte falls i n the range 1 - 100 S m" 1 . Therefore the degree o f matr ix compress ion has a significant effect o n the potential dis t r ibut ion. M a t r i x compress ion also changes the specif ic electrode surface, and w h e n the react ion is mass-transfer control led , the compress ion affects the l oca l current density by the re la t ion o f the mass transfer coefficient to the porosi ty [317]. M e t a l deposi t ion also s ignif icant ly increases the conduct iv i ty and the specif ic surface area o f the fibre matr ix and decreases the porosi ty o f the fibre matr ix . The current and potential distr ibutions change w i t h t ime. 50 4.2 Experimental 4.2.1 Electrolytic Cel l and Experimental Set-up Genera l ly the m a i n types o f f l o w for porous electrodes are flow-through and f l o w - b y . F l o w - t h r o u g h was employed for this study, i.e. the f l o w i n the fibre felt is para l le l to the current flow. The graphite fibre felt suppl ied by the N a t i o n a l E lec t r i c C a r b o n C o . has a specif ic surface area o f 0.7 m 2 g"1 and a porosi ty o f 96.5 % . SS316 stainless steel mesh was used to fix the fibre felt o n two sides and conduct electr ici ty to the fibre. E x c e p t for e lec t r ica l contact parts, the stainless steel mesh was painted. The superf icial cathode surface area was 12 c m 2 . The catholyte was separated f rom the anolyte by a D u Pont N a t i o n 450 membrane to prevent the anodic ox ida t ion o f cyanide. The anodes were n i c k e l sheet for o x y g e n evo lu t ion and T I R 2 0 0 0 (Ir and T a coated t i tanium) for chlor ine evolu t ion and their surfaces were 6 c m 2 . The e lectrolyt ic c e l l consisted o f two parts o f polycarbonate w h i c h were connected by screws and sealed by rubber. F igure 4-2 shows the schematic d iagram o f ce l l . T o start an experiment, approximately 18 liters o f electrolyte i n a container were preheated to about 40 ° C using a water bath and then pumped to the e lectrolyt ic c e l l us ing a Co le -Pa rmer pump M o d e l 7 5 1 9 - 2 0 A equipped w i t h a d ig i ta l variable-speed console dr ive for precise un i fo rm f l o w rate control . The electrolyt ic c e l l was put i n a water bath to main ta in the electrolyte temperature at 40 ° C . Af ter the electrolyte had passed through the c e l l , it was pumped to a container i n order to main ta in a un i fo rm f l o w rate. T w o tubes and pumps were used to add N a O H and N a C l and circulate the anolyte. F igure 4-3 shows the schematic d iagram o f electrolyte flow. A coulometer was used to record the amount o f charge consumed. In the case o f chlor ine evolut ion , a B a c h - S i m p s o n L t d . P H M 8 2 standard p H meter was used to moni tor the p H o f the anolyte, keeping it above 4 and a v o i d i n g the s ignif icant migra t ion o f hydrogen ions through the membrane. A Jenway M o d e l 5310 conduct iv i ty meter was used to measure the conduc t iv i ty o f the electrolyte w h i c h was p laced i n a 100 m L tube whose temperature was con t ro l led by a water bath. The copper concentrat ion i n the solut ion was ana lyzed by a tomic absorpt ion and the cyanide concentrat ion was analyzed us ing dist i l lat ion-absorption-t i t rat ion method (see 51 Appendix 2). The copper deposited in the graphite felt was dissolved in nitric acid and analyzed by atomic absorption. Copper conductor Figure 4-2 Schematic diagram of electrolytic cell (size: 18(H)xl3(L)xl2(W) cm) Figure 4-3 Schematic diagram of experimental set-up 52 4.2.2 Materials Reagent grade sodium cyanide, copper cyanide, sodium hydroxide, sodium thiocyanate and sodium chloride were used to prepare the required synthetic solutions. Solid sodium cyanide and copper cyanide were analyzed prior to preparation of the solutions to ensure that the required compositions were achieved. 4.3 Results and Discussion The conductivity of dilute copper cyanide solutions is expected to be low and this low conductivity significantly affects the potential and current distribution. The conductivity was therefore measured. The results are listed in Table 4-1. The conductivity is very low and will affect the potential distribution resulting in nonuniform copper deposition. From Equation 4-12, the approximate conductivities of graphite fibre felt are 10, 37, 89, 158, 241, 336 and 443 S m"' respectively for 0. 5, 10, 15, 20, 25, 30 % compression. In order to increase the conductivities of the graphite felt and decrease the potential difference in the graphite felt, the compression of the graphite felt should be increased. The compression of the graphite felt also increases the specific surface area. However, when the graphite felt is compressed to some degree, the conductivity of the graphite felt is much larger than that of the solution and the further compression will not significantly affect the potential and current distribution according to Equation 4-12. If the compression is too high, the porosity becomes low and the amount of the deposited copper per unit volume becomes low. The compression mainly affects the deposition of copper when the deposition begins. When a certain amount of copper is deposited in the graphite felt, the contact resistivity between the fibres becomes negligible and the conductivity of the graphite felt becomes much higher. Therefore the selection of the degree of the compression is important. In the preliminary test, at a low mole ratio of cyanide to copper (e.g. 3), the compression had less effect on the current efficiency. At a high mole ratio of cyanide to copper, the compression had a significant effect on the current efficiency. The reasons are: at a low mole ratio of cyanide to copper, copper is easily deposited on the graphite and then significantly increases the conductivity of the graphite and improves the surface condition. At a high mole ratio of cyanide to copper, copper is difficult to deposit 53 o n the graphite and hydrogen evolu t ion is dominant and the conduct iv i ty o f the graphite felt does not improves greatly w i t h t ime. F r o m these tests, 2 5 % o f compress ion is required to get an acceptable and reproducible current eff iciency. Therefore, 25%o o f compress ion was used for a l l the experiments. In the dilute solutions discussed, the migra t ion o f copper cyanide complexes is important, result ing i n a decrease i n the mass transfer toward the cathode. A l s o the effect o f the diffuse double layer can decrease the reduct ion o f copper complexes due to their negative charge when the potential is w e l l b e l o w the zero-charge potential . Table 4-1 Conduc t iv i t i e s o f copper cyanide solutions w i t h different C N : C u mo le ratios at f ixed C u concentrations (unit: S m - 1 ) * [ C u ] / g L - ' T e m p . ( ° C ) C N : C u = 3 C N : C u = 3.5 C N : C u = 4 C N : C u = 4 . 5 2 25 1.105 1.241 1.375 1.512 2 40 1.410 1.588 1.769 1.955 1 25 0.703 0.788 0.873 0.952 1 40 0.902 1.002 1.121 1.220 * [ N a O H ] = 0.01 M , [ N a C N S ] = 0.01724 and 0.00862 M respect ively for 2 and 1 g L " 1 C u . Copper deposi t ion on graphite fibre was first conducted i n an und iv ided c e l l i n an attempt to use the anodic ox ida t ion o f thiocyanate to prevent the cyanide ox ida t ion . H o w e v e r , the thiocyanate d i d not protect against the anodic ox ida t ion o f cyanide and the anodic ox ida t ion current eff ic iency was around 100%. Therefore the catholyte was separated f rom the anolyte by a D u Pont N a f i o n 450 membrane. The anolytes were 5 M N a O H and 5 M N a C l respect ively for the oxygen and chlor ine evolu t ion experiments. The current ef f ic iency o f copper deposi t ion and the conduct ivi t ies o f the solut ion are expected to increase w i t h increasing solu t ion temperature. Operat ing copper deposi t ion at elevated temperatures needs heating a large vo lume o f dilute solut ion, result ing i n significant energy consumpt ion . H o w e v e r , operating at a l o w temperature results i n a l o w current eff ic iency w h i c h increases the power consumpt ion . A temperature range o f 25-40 ° C was selected for the invest igat ion. The ve loc i ty o f f l o w used, was i n the range 3-10 c m m i n / 1 and the estimated mass transfer coefficient is i n the range 0.55 to 1.01 x l O " 2 c m s accord ing to Equa t ion 4-10. In a l l the experiments, the total cyanide concentration d i d not change after e lec t rowinning and the amount o f the deposited copper matched c lose ly the change o f the copper concentrat ion i n the solut ion. 54 The results o f copper deposi t ion and energy consumpt ion for o x y g e n evo lu t ion as anode react ion are l is ted i n Table 4-2 ( in i t ia l copper concentrat ion = 1 g L _ 1 ) and Tab le 4-3 ( in i t ia l copper concentrat ion = 2 g L " 1 ) . In the case o f chlor ine evolu t ion , the c e l l voltages are 0.78, 0.57 and 0.41 V higher than those i n the case o f oxygen evo lu t ion respect ively for 30, 60 and 100 A m" 2 and the other results are the same. Table 4-2 Copper cathodic current eff ic iency and energy consumpt ion at 40 ° C and in i t i a l [Cu] = 1 g L " 1 for experiments w i t h oxygen evolu t ion at anode CN:Cu 3 3.5 4 4.5 3 3.5 4 4.5 3 3.5 4 4.5 Current Density / Am"2 30 30 30 30 60 60 60 60 100 100 100 100 Flow velocity / cm min."1 2.97 2.97 2.97 2.97 5.93 5.93 5.93 5.93 9.83 9.83 9.83 9.83 [ C u l / g L ' O ) * 0.713 0.837 0.905 0.925 0.740 0.877 0.950 0.978 0.778 0.902 0.956 0.980 [Cu+] / g L"1 (2)** 0.686 0.767 0.868 0.900 0.725 0.823 0.919 0.970 0.751 0.864 0.948 0.975 C.E. / % (average) 64.2 40.0 22.4 22.0 57.4 38.6 23.2 19.6 47.0 30.8 11.7 7.5 Cell voltage / V 2.64 2.65 2.63 2.56 3.66 3.64 3.52 3.42 5.01 4.85 4.62 4.60 * The samples were taken after the solut ion passed the ce l l i n the midd le course o f the experiments. ** The samples were taken after the solut ion passed the c e l l at the end o f the experiments. Table 4-3 Copper cathodic current eff ic iency and energy consumpt ion at 40 ° C and in i t i a l [Cu] = 2 g L " 1 for experiments w i t h oxygen evolu t ion at anode CN:Cu 3 3.5 4 4.5 3 3.5 4 4.5 3 3.5 4 4.5 Current Density / Am"2 30 30 30 30 60 60 60 60 100 100 100 100 Flow velocity / cm min."1 2.97 2.97 2.97 2.97 5.93 5.93 5.93 5.93 9.83 9.83 9.83 9.83 [Cul /gL/ 'O)* 1.663 1.703 1.845 1.849 1.667 1.708 1.810 1.856 1.672 1.788 1.920 1.950 [Cu+] / g L"1 (2)** 1.612 1.658 1.712 1.741 1.633 1.661 1.767 1.800 1.642 1.708 1.82 1.86 C.E. / % (average) 88.6 69.4 42.9 37.6 84.4 58.4 38.0 31.6 80.6 47.4 23.2 18.0 Cell voltage / V 2.17 2.28 2.30 2.15 2.91 2.94 2.89 2.82 3.81 3.64 3.63 3.56 * The samples were taken after the solut ion passed the c e l l i n the m i d d l e course o f the experiments. ** The samples were taken after the solut ion passed the c e l l at the end o f the experiments. F r o m Figure 4-4, w i t h increasing C N : C u mole ratio, the current e f f ic iency decreases s ignif icant ly and the energy consumpt ion increases s ignif icant ly . T h i s is due to the fact that the l o w l y coordinated copper cyanide complexes (dicyanide or t r icyanide is electroactive species) and the calculated equ i l ib r ium potential for C u ( I ) / C u redox couple decreased w i t h increasing C N : C u mole ratio (see Table 4-4). The exchange current is expected to decrease 55 w i t h increasing mole ratio o f cyanide to copper. The l o w concentrat ion o f electroactive species and the l o w exchange current result i n a h igh po la r iza t ion at a f ixed current. Therefore at a h i g h mole ratio o f cyanide to copper, hydrogen evo lu t ion is dominant and decreases the current eff ic iency s ignif icant ly. Table 4-4 Dis t r ibu t ion and potentials o f copper cyanide at [OH"] = 0.01 M at 40 ° C [ C u ] / g L - l Species & potential C N : C u = 3 C N : C u = 3.5 C N : C u = 4 C N : C u = 4.5 1 C u ( C N ) 2 " C u ( C N ) 3 2 -C u ( C N ) 4 3 " E C u ( n / C u v s . S H E / V 3.00 % 96.51 % 0.49 % -0.632 0 . 1 7 % 92.01 % 7.82 % -0.851 0.08 % 85.30 % 14.62 % -0.907 0.05 % 79.34 % 20.61 % -0.941 2 C u ( C N ) 2 " C u ( C N ) 3 2 -C u ( C N ) 4 3 " Ecudvcu vs . S H E / V 2.30 % 97.06 % 0.64 % -0.656 0.09 % 86.95 % 12.96 % -0.878 0.04 % 76.47 % 23.49 % -0.937 0.02 % 67.86 % 32.11 % -0.974 W h e n the ratio o f the current to the f l o w rate was mainta ined constant, the current eff ic iency decreased w i t h increasing current and f l o w rate, suggesting that the effect o f f l o w rate o n mass transfer and o n current eff ic iency was lower than that o f the current density. T h i s phenomenon becomes more apparent w h e n the ratio o f cyanide to copper is h igh . The reasons are: (1) the mass transfer coefficient i n graphite felt is on ly propor t ional to ( f l ow ra te ) 0 4 f rom Equat ions 4-9 to 4-11 . Therefore the increase i n the mass transfer does not match the increase i n the current density, result ing i n higher concentrat ion po la r iza t ion and hence l o w current eff ic iency; (2) the charge transfer coefficient for hydrogen evo lu t ion (e.g. about 0.45 [113]) is larger than that for copper deposi t ion (0.1 [116] or 0.38 [110]). Therefore the increase i n the current density poss ib ly results i n more increase i n the current densi ty for hydrogen evo lu t ion than that for copper deposi t ion. A t a h i g h mo le ratio o f cyanide to copper, the hydrogen evolu t ion is a dominant react ion and the mass transfer has less effect o n the current eff ic iency o f copper deposit ion. The increase i n the current results i n s ignif icant hydrogen evo lu t ion and hydrogen bubbles cou ld b l o c k the so lu t ion f rom contact ing the graphite, resul t ing i n a significant decrease i n the current ef f ic iency and the effective conduct iv i ty o f the solut ion, g i v i n g a h igh energy consumpt ion . 56 D u e to the above dependence o f current eff ic iency on C N : C u m o l e ratio and current density, the convers ion o f C u (I) to C u decreases w i t h increasing C N : C u m o l e ratio and increasing current density at a fixed ratio o f current density to f l o w ve loc i ty (Figure 4-5). (a) 1 g L " 1 C u (b) 2 g L " ' C u Figure 4-4 Current eff ic iency ( C . E . ) and the energy consumpt ion ( E . C . ) o f copper depos i t ion vs. the mo le ratio o f cyanide to copper at different cathodic current densities and 40 ° C . The electrolyte: (a) 1 g L " 1 C u , 0.01 M N a O H and 0.00862 M N a S C N , and (b): 2 g L " 1 C u , 0.01 M N a O H and 0.01724 M N a S C N . The flow ve loc i ty : 2 .97, 5.93 and 9.83 c m min." 1 respect ively for 30, 60 and 100 A m " 2 . Mole ratio of cyanide to copper Mole ratio of cyanide to copper (a) 1 g L " 1 C u (b) 2 g L " 1 C u F igure 4-5 C o n v e r s i o n o f Cu(I) to C u vs. the mole ratio o f cyanide to copper at different cathodic current densities and 40 ° C . The electrolyte: (a) 1 g L " 1 C u , 0.01 M N a O H and 0.00862 M N a S C N , and (b): 2 g L " 1 C u , 0.01 M N a O H and 0.01724 M N a S C N . The f l o w ve loc i ty : 2.97, 5.93 and 9.83 c m min." 1 respectively for 30, 60 and 100 A m" 2 . 57 From Figure 4-6, the cell voltage decreased with increasing time. This is due to the increasing amount of copper deposited on the graphite fibre electrode giving improved conductivity of the graphite fibre electrode with time. 2.85 -r 2.80 -Time / hours Figure 4-6 Cell voltage vs. time at the cathodic current density = 30 A m"2 and 40 °C. The electrolyte: lg L ' 1 Cu, CN:Cu = 3, 0.01 M NaOH and 0.00862 M NaSCN and the flow velocity: 2.97 cm min."1. From Figures 4-7, the relation between the cell voltage and the mole ratio of cyanide to copper is dependent on the current density and the copper concentration. The cell voltage is the sum of the anode potential drop, the anolyte IR drop, the membrane IR drop, the catholyte IR drop, the cathode potential drop and the hardware IR drop. At a constant potential, only the cathode potential drop and the catholyte IR drop change with CN:Cu mole ratio. According to Table 4-1, with increasing CN:Cu mole ratio, the solution conductivity increases, resulting in a decrease in the cell voltage. From Table 4-4, with increasing CN:Cu mole ratio, the redox potential for Cu(I)/Cu decreases, the concentration of dicyanide or triycyanide decreases, leading to a lower exchange current for copper reduction. Also the potential for hydrogen evolution moves negatively due to the inhibiting effect of cyanide ions on hydrogen evolution [113]. The above factors result in a decrease (more negative) in the cathode potential (i.e. an increase in the cathode potential drop) and an increase in the cell voltage at a fixed current density. Therefore the relation between the cell voltage and CN:Cu 58 mole ratio depends o n w h i c h one (the changes i n the solut ion I R drop and the cathode potential) is predominant . 3 3.5 4 4.5 3 3.5 4 4.5 Mole ratio of cyanide to copper Mole ratio of cyanide to copper (a) 1 g L " 1 C u (b) 2 g L " 1 C u F igure 4-7 C e l l voltage vs. the mole ratio o f cyanide to copper at different cathodic current densities and 40 ° C . The electrolyte: (a) 1 g L " 1 C u , 0.01 M N a O H and 0.00862 M N a S C N , (b) 2 g L " 1 C u , 0.01 M N a O H and 0.01724 M N a S C N , the f l o w ve loc i ty : 2.97, 5.93 and 9.83 c m min." 1 respect ively for 30, 60 and 100 A m" 2 . F igures 4-8 and 4-9 show the graphite fibre felt after the depos i t ion o f copper f rom copper cyanide solut ion. Less copper was deposited where the graphite felt contacted the stainless steel mesh probably due to the shie ld ing effect o f stainless steel and the poor mass transfer because most o f the solut ion d i d not pass this area. The amount o f deposited copper decreased w i t h increasing distance f rom the surface to the inside o f the graphite due to the non-un i fo rm potential d is t r ibut ion caused by the l o w conduct ivi t ies o f the so lu t ion and the graphite fibre felt. Hence there was a decrease i n the d r i v i n g force (polar izat ion) o f copper deposi t ion. A t a mo le ratio o f cyanide to copper > 4, copper was m a i n l y deposited i n a very nar row area near the surface o f the graphite felt. T h i s m a y be caused by the s ignif icant hydrogen evo lu t ion w h i c h greatly decreased the effective conduct iv i ty o f the so lu t ion and even b l o c k e d the solu t ion f rom contacting the graphite fibre. D u e to the fact that no copper was deposited inside the graphite felt, the conduct iv i ty o f the graphite felt was not i m p r o v e d . Figure 4-9 Cross-sec t ion o f graphite fibre felt on w h i c h copper has been deposited 60 F r o m Tables 4-2 and 4-3 , after the solut ion passed through the c e l l , the copper concentrat ion i n the solut ion taken at the end o f the experiment was lower than that i n the midd le o f the experiment. Th i s means the current eff ic iency increased w i t h t ime. T h i s phenomenon is due to the increasing amount o f copper deposited o n the graphite g i v i n g i m p r o v e d conduct iv i ty o f the graphite fibre electrode, the specif ic surface area and the surface condi t ion . Therefore the effect o f deposited copper o n the current ef f ic iency was tested us ing cyanide solutions w i t h a h igh mole ratio o f cyanide to copper. The experiments were conducted by three-cycle runs w i t h l g I / 1 and 2 g L " 1 C u solut ion w i t h an in i t i a l C N : C u mo le ratio o f 3. The results are g iven i n Table 4-5. Table 4-5 Resul ts o f cyc le run at 40 ° C (the in i t i a l C N : C u ratio = 3) N o . o f cyc le 1 [ C u ] / C . E . 2 [ C u ] / C . E . 3 [ C u ] / C . E . Average C . E . Ene rgy consumpt ion 1 g L ' 1 C u 0.76 g L 1 / 61 % 0.53 g L " 1 / 5 8 % 0.34 g L " ' / 5 3 % 57.5 % 1 . 8 k W h / k g C u 2 g L " ' C u 1.67 g L " 1 / 8 6 % 1.38 g L " 1 / 73 % 1.12 g L " 1 / 6 8 % 78.7 % 1 . 1 5 k W h / k g C u Af te r three-cycle runs, copper concentrations decreased f rom 1 g L " 1 to 0.34 g L " ' w i t h an average current eff ic iency o f 57.5 % and a energy consumpt ion o f 1.8 k W h / k g C u and f rom 2 g L _ 1 to 1.1 g L'] w i t h an average current eff ic iency o f 78.7 % and a energy consumpt ion o f 1.15 k W h / k g C u . F r o m Figure 4-10, the copper concentrat ion decreased approximate ly l inear ly after every single solut ion pass through the graphite felt and the current ef f ic iency decreased very l i t t le . The mole ratio o f cyanide to copper increased f rom 3 to 9.4 and 5.5 respect ively for the in i t i a l concentrations o f 1 g L _ 1 and 2 g L 1 . Appa ren t ly for the first s ingle pass, the current eff ic iency was h igh because copper was ready to deposit. F o r the second and th i rd passes, the current efficiencies were s t i l l h igh because a certain amount o f copper was deposited o n the graphite felt, i m p r o v i n g the conduct iv i ty o f the graphite felt and increas ing the specif ic surface. F r o m Figures 3-5 b and 3-6 b, as expected, after the first s ingle passes through the graphite felt, the equ i l i b r ium potential for the C u ( I ) / C u changed s igni f icant ly . Af te r the second and th i rd passes, the equ i l i b r i um potential changed modest ly , w i t h the copper t r icyanide species being a lways dominant. Copper deposi t ion releases free cyanide w h i c h not on ly suppresses the cathodic reduct ion o f copper (I), but also the hydrogen 61 evo lu t ion [113]. Therefore copper can be r emoved eff icient ly f rom cyanide so lu t ion even w i t h a h i g h C N : C u mole ratio. 2.0 r 10 o 0.2 -4- 1 0.0 0 1 2 Number of cycle runs Figure 4-10 Concentra t ion o f copper vs. the number o f the solu t ion passes through the graphite felt at [ C u ] i n i t a l = 1 and 2 g L " 1 and 40 ° C . The electrolyte: (1) l g I / 1 C u , C N : C u = 3, 0.01 M N a O H and 0.00862 M N a S C N and (2) 2 g L " 1 C u , C N : C u =3, 0.01 M N a O H and 0.01724 M N a S C N , and the f l o w ve loc i ty : 2.97 c m min." 1 . 4.4 Summary The current eff ic iency o f copper deposi t ion on a graphite felt electrode decreases w i t h increasing C N : C u mole ratio and current density. D u e to the l o w conduct iv i t ies o f the so lu t ion and the graphite felt, the potential and current d is t r ibut ion o f copper through the 3-d imens iona l electrode are not un i form. The accumula t ion o f deposited copper w i t h the graphite felt as the pla t ing proceeds, s ignif icant ly improves the conduct iv i ty o f the graphite felt increasing the specific surface area and benefi t ing copper deposi t ion. Coppe r can be deposited eff icient ly o n the graphite felt f rom l o w concentrat ion solutions event at a h i g h C N : C u mo le ratio. U p to 60 % o f the C u can be r emoved eff icient ly f rom the solut ion. The energy requirement for copper deposi t ion was as l o w as 1-2 k W h / k g C u (1000-2000 kWh/ tonne C u ) i n the current range 30-100 A m " 2 , w h i c h compares favorably w i t h the va lue obtained i n convent ional copper e lec t rowinning f rom sulphuric ac id-copper sulphate solutions. The obtained results meet the requirement for industr ial practice. 62 5. ELECTROWINNING FROM COPPER CYANIDE SOLUTION USING ALTERNATIVE ANODE REACTIONS A s discussed, copper cyanide can be extracted f rom dilute solutions us ing solvent extract ion to produce a concentrated copper cyanide solut ion f rom w h i c h copper can be recovered us ing the copper e lec t rowinning process. U s i n g an alternative anodic react ion was selected as a w a y to prevent the anodic ox ida t ion o f cyanide and el iminate the use o f a membrane c e l l . Thiocyanate , methanol , sulphite, and a m m o n i a were selected as sacr i f ic ia l species for addi t ion to the electrolyte. 5.1 Experimental Apparatus and Set-up for Electrowinning E l e c t r o w i n n i n g was carr ied out i n a 1.5-L m i n i - c e l l made f rom polycarbonate . The electrolyte was c i rcula ted us ing a C O L E - P A R M E R peristalt ic p u m p at a f l o w rate o f 0.18 L min" 1 . The electrolyte was a l l owed to ove r f low into a 2 5 0 - m l Er l enmeyer flask f rom w h i c h a b leed was taken pe r iod ica l ly to remove free cyanide. C u C N , N a O H and sacr i f ic ia l species were added pe r iod ica l ly to main ta in their respective concentrations due to copper depos i t ion and the anodic consumpt ion o f N a O H and the sacr i f ic ia l species. A magnet ic stirrer was used to accelerate the d issolu t ion and the m i x i n g o f C u C N , N a O H and the sacr i f ic ia l species. In order to keep a constant vo lume o f the electrolyte, de ion ized water was added as required. The electrolyte was heated w i t h quartz-shielded i m m e r s i o n heaters connected to a temperature control ler . A power supply was used to supply the current and a coulometer was used to measure the amount o f electr ici ty passed. The anode materials selected for study were T I R 2000 D S A (t i tanium coated w i t h i r i d i u m and tantalum oxide) for S C N " , C H 3 O H , N H 3 and S 0 3 2 " and graphite on ly for S 0 3 2 " and S C N " . SS316 stainless steel was used as the cathode mater ial . In the case o f the ox ida t ion o f S 0 3 2 " , ni t rogen gas was used to prevent air ox ida t ion . The exper imental set-up is shown i n F igure 5-1. The deposited copper was recovered, washed, dr ied and we ighed to determine the cathodic current eff ic iency. The current efficiencies for the ox ida t ion o f thiocyanate, methanol , a m m o n i a and sulphite were based o n the cyanide analysis ( A p p e n d i x 2). Reagent grade chemica ls were used i n a l l experiments. 63 Power supply I Erlenmeyer flask Coulometer Heater ft Pump Stirrer 1 Stirring plate ft Pump 3-rri Temperature controller Figure 5-1 Schemat ic d iagram o f the experimental set-up 5.2 Selection of Sacrificial Materials The anodic current eff ic iency was obtained us ing least-squares f i t t ing accord ing to the concentrations o f the supposed o x i d i z e d species i n every bleed sample and the mass balance: the amount taken out for the bleed, the amount added, and the amount i n the e lectrolyt ic c e l l for a f ixed vo lume o f the electrolyte. A s s u m i n g a part icular anodic current ef f ic iency o f cyanide , the cyanide concentrat ion i n the electrolyte can be predicted and least-squares can be used to fit the current eff ic iency to the measured concentration. F igure 5-2 shows the d iagram o f the fitted and measured concentrations o f cyanide. 86 85 • Analyzed Fitted C.E. for cyanide = 14.4 % 4 6 8 Time / Hours Figure 5-2 Concentra t ion o f cyanide vs. the electrolysis t ime for obta ining the current ef f ic iency o f cyanide ox ida t ion at 60 ° C . Elec t ro ly te : 70 g L " 1 C u , C N : C u = 3, 113 g L " 1 N a 2 S 0 3 , 10 g L 1 N a O H . Table 5-1 Results for selection o f sacr i f ic ia l species at 60 ° C Additive Anodes current density /Am" 2 Time / hours Average cell voltage / V Anodic C.E. / % Copper C. E. /% Anode surface condition cathode copper condition SCN" 30 g L-1 Graphite 500 13 2.26 9.10 94.6 some black coating dendrite SCN" 30 g L-1 Graphite 1000 13 2.42 6.3 95.2 some black coating sponge dendrite SCN" 40 g L"1 TIR2000 500 12 2.32 12.98 94.96 some black coating sponge dendrite SCN" 40 g L"1 TIR2000 1000 8 2.68 10.54 94.82 some black coating sponge dendrite CH 3OH 22.4 g L"1 TIR2000 500 12 4.00 9.2 91.4 thick black coating sponge-like dendrite C H 3 O H 22.4 g L"1 TIR2000 1000 8 4.16 ? 108 for CN-88.1 thick black coating strong dendrite NH 3 54.2 g L"1 TIR2000 500 12 3.67 12.2 91.5 thick black coating and some foam coral-like strong dendrite NH 3 54.2 g L"1 TIR2000 1000 6 8.74 ? 110 for CN" 82.7 thick black coating and a lot of foam coral-like strong dendrite Na 2S0 3 113 gL"1 TIR2000 500 12 2.18 84.5 91.7 a very little black coating coral-like strong dendrite Na 2S0 3 113 g L"1 TIR2000 1000 6 3.57 40 91.3 a very little black coating coral-like strong dendrite Na 2S0 3 113 gL"1 TIR2000 250 14 1.85 87.5 91.9 a very little black coating small dendrite Na 2S0 3 H3gL-' Graphite 500 12 2.20 84.5 91.9 a very little black coating dendrite Na 2S0 3 113gL-' Graphite 250 14 1.90 86.5 92.1 a very little black coating small dendrite 5.2.1 Thiocyanate In the case o f thiocyanate as a sacr i f ic ia l species, f rom Table 5-1, the anodic current eff ic iency o f thiocyanate ox ida t ion was very l o w on the graphite anode and a li t t le h igher o n the T I R 2000 anode. T h i s means that thiocyanate was more dif f icul t to o x i d i z e o n the above two anodes than copper cyanide. B o t h the graphite anode and the T I R 2000 anode were coated w i t h a b lack so l id substance, w h i c h was readi ly d i s so lved i n cyanide so lu t ion or H C 1 solut ion. Af te r d issolu t ion o f the b lack substance, the H C 1 solu t ion became blue. A n a l y s i s showed that copper was i n both the cyanide and H C 1 solutions. Therefore the b lack substance was presumed to be cupric oxide or a mixture o f cupr ic ox ide and hydrox ide . N o gas evo lu t ion was observed o n the graphite anode, meaning there was no o x y g e n evo lu t ion or no 65 oxida t ion o f cyanate to carbonate and ni trogen gas. A t 500 A m* 2, o n the cathode, very tight sma l l dendrites were observed and at 1000 A m" 2 , large dendrites l ike sponge cora l were formed o n the cathode. The current eff ic iency was h i g h poss ib ly due to the format ion o f cora l - l ike copper w h i c h made the real surface area m u c h larger than the apparent cathode surface area and the specif ic current density m u c h lower than the observed value. Ano the r reason is that thiocyanate suppresses hydrogen evolu t ion and increases the current e f f ic iency as reported i n the literature [92, 94-96, 98]. 5.2.2 Methanol In the case o f methanol as a sacr i f ic ia l species, at 500 A m" 2 the anodic current eff ic iency for methanol was about 9.2 % based on the cyanide analysis , but at 1000 A m" 2 the anodic current eff ic iency was negative and the anodic current eff ic iency for cyanide was about 108%. The anodic ox ida t ion current for cyanide was over 100% probably due to chemica l ox ida t ion by air i n the presence o f methanol . A little gas evo lu t ion was observed and some gas bubbles adhered to the anode surface. The gas was probably o x y g e n or n i t rogen due either to o x y g e n evo lu t ion or the ox ida t ion o f cyanate to ni t rogen gas. The anode was coated w i t h a very th ick layer o f a b lack substance w h i c h d i s so lved readi ly i n H C 1 so lu t ion w h i c h became blue. Therefore the substance was again thought to be copper ox ide . D u e to the th ick b lack coat ing and gas bubble effects, the c e l l voltage became very h igh (4 V at the shutdown o f the experiment. The format ion o f cupric ox ide c o u l d contribute to the l o w consumpt ion o f cyanide and so the anodic current eff ic iency for methanol m a y be l ower than the value based o n analysis . 5.2.3 Ammonia In the case o f a m m o n i a as a sacr i f ic ia l species, at 500 A m" 2 the anodic current eff ic iency for a m m o n i a ox ida t ion was 12.2 % based on the cyanide analysis . H o w e v e r , at 1000 A m" 2 , the anodic current eff ic iency was about zero and the anodic current e f f ic iency for cyanide was about 120 % . W h i t e foam formed around the anode and the higher the current density, the greater the foam. A black and b r o w n substance heav i ly coated the anode surface and formed the passivat ing f i l m . F r o m Figure 5-3, the c e l l voltage increased w i t h increas ing 66 t ime o f e lectrolysis f rom 2.22 to 5.41 V at 500 A m" 2 due to increasing format ion o f copper ox ide . The c e l l voltage increased f rom 2.89 to 19.0 V for 1000 A m" 2 at shut-down. The b l ack substance d i sso lved i n cyanide and H C 1 solutions. H o w e v e r , the whi te b r o w n substance d i d not d issolve . T h i s substance and white foam were probably produced by the react ion o f a m m o n i a and cyanide at the anode result ing i n the h igh consumpt ion o f cyanide ( C . E . for cyanide is over 100%). A t 500 A m" 2 the cupric ox ide format ion c o u l d contribute to the l o w consumpt ion o f cyanide. Therefore the anodic current eff ic iency o f a m m o n i a was l ower than the above value based on the cyanide analysis. 5.2.4 Sulphite In the case o f sulphite as a sacr i f ic ia l species, at 500 A m" 2 the anodic current eff ic iency for sulphite ox ida t ion was about 8 5 % both o n graphite and T I R 2000 based o n the cyanide analysis . H o w e v e r , at 1000 A m" 2 the current eff ic iency decreased to 40 % at T I R 2000. T h i s means that increasing the current results i n the ox ida t ion o f more cyanide . T h i s m a y be due to a change i n the e lectrochemical kinet ics o f the two anode reactions at h i g h current density. W i t h an increasing c i rcula t ing f l o w rate o f electrolyte, the c e l l vol tage decreased due to i m p r o v e d mass transfer o f both copper ions to the cathode and sulphite ions to the anode. O n l y a very smal l amount o f b lack material coated the upper side o f the anode near the surface o f the electrolyte. Therefore sulphite addi t ion can effect ively prevent or decrease the format ion o f copper oxide at the anode. F r o m Figure 5-3, the c e l l vol tage first increased a l i t t le and then decreased s l ight ly w i t h increasing t ime o f e lect rolysis . T h e decrease i n the c e l l voltage may be caused by the g rowing cathode and increasing real surface area due to the format ion o f the dendrit ic copper deposit. N o gas evo lu t ion was observed o n the anodes. F r o m the above discuss ion, thiocyanate, methanol and a m m o n i a d i d not effect ively protect against cyanide ox ida t ion and the anode surface became coated w i t h b lack copper ox ide and lost its catalyt ic act ivi ty . O n l y sulphite ox ida t ion was found to effect ively l i m i t the ox ida t ion o f cyanide. The anodic current eff ic iency o f sulphite was the same o n T I R 2 0 0 0 and graphite anodes. Therefore sulphite ox ida t ion and graphite were selected as the sacr i f ic ia l addit ive and the anode mater ial respectively for further tests. 67 0 2 4 6 8 10 12 Time / hours Figure 5-3 C e l l voltage vs. the t ime o f electrolysis i n the presence o f a m m o n i a and sulphite as a sacr i f ic ia l species at 500 A m" 2 and 60 ° C . Elec t ro ly te : 70 g L " 1 C u , C N : C u = 3, and 10 g L " 1 . 5.3 Effect of Some Parameters on the Anodic and Cathodic Processes in the Presence of Sulphite 5.3.1 Effect of Current Density The current density usual ly affects the anodic and cathodic processes s igni f icant ly . Thus experiments were conducted to determine the effect o f current density. Three current densities were tested and the results are l isted i n Table 5-2. The anodic current ef f ic iency o f cyanide decreases s ignif icant ly w i t h decreasing current density f rom 1000 to 500 A m" 2 , but decreases s l igh t ly f rom 500 to 250 A m" 2 . The cathodic current ef f ic iency was almost independent o f the current density. Th i s phenomenon is probably related to the m o r p h o l o g y o f the copper deposits. A t a h igh current density, more and larger dendrites were p roduced and at a l o w current density, fewer and smaller dendrites were obtained, resul t ing i n approximate ly the same real current density. 68 Table 5-2 Effect o f current density o n the anodic current eff ic iency o f cyanide and the cathodic current eff ic iency o f copper at 60 °C. Elec t ro ly te : 70 g L " 1 C u , C N : C u mo le ratio =3, 10 g L " 1 N a O H and 113 g I / 1 N a 2 S 0 3 . Current density / A m" 2 1000 500 250 C . E . for cyanide ox ida t ion / % 59.9 14.4 12.8 ± 3 C . E . for copper deposi t ion / % 92.2 92.0 92.1 ± 1 5.3.2 Effect of sulphite concentration A t 250 A m ' 2 solutions w i t h 50, 63 and 113 g L " 1 N a 2 S 0 3 were tested w i t h the results be ing l is ted i n Table 5-3. In this range o f sulphite concentration, the anodic current ef f ic iency o f cyanide was not affected very m u c h by sulphite concentrat ion and the cathodic current eff ic iency o f copper deposi t ion was almost independent o f the sulphite concentrat ion. The sulphite concentrat ion d i d not affect the morpho logy o f the cathodic deposit. Therefore the use o f 50-60 g L " 1 N a 2 S 0 3 is sufficient to get a reasonable anodic current e f f ic iency ( m i n i m u m consumpt ion o f sulphite). Table 5-3 Effect o f sulphite concentration on the anodic current eff ic iency o f cyanide and the cathodic current eff ic iency o f copper at 60 °C and 250 A m" 2 . E lec t ro ly te : 70 g L " ' C u , C N : C u mo le ratio =3, 10 g L " 1 N a O H . [ N a 2 S 0 3 ] / g L 1 50 63 113 C . E . for cyanide ox ida t ion / % 13 13.2 ± 3 12.8 ± 3 C . E . for copper deposi t ion / % 91.8 91.9 ± 2 92.1 ± 1 5.3.3 Effects of thiocyanate and mole ratio of cyanide to copper Thiocyanate is expected to be present i n the copper-cyanide e lec t rowinning solut ion. The mo le ratio o f cyanide to copper is a very important parameter affecting the anodic and cathodic processes. Therefore experiments have been conducted on solutions w i t h different mo le ratios o f cyanide to copper i n the presence and absence o f thiocyanate. The results are l is ted i n Table 5-4. 69 Table 5-4 Resul ts o f copper e lec t rowinning at 250 A m" 2 and 60 ° C . Elec t ro ly te : 70 g L " 1 C u , C N : C u mo le ratio = 3-4.5, 63 g L " ' N a 2 S 0 3 and 10 g L " 1 N a O H i n the presence and absence o f S C N " . CN:Cu mole ratio Average cell voltage / V Copper C. E. / % Energy consumption / kWh kg'1 Anodic C.E. for CN' / % Anode surface cathode copper condition 3 (no SCN') 1.92 93.1 0.873 11.3 a very little black coating small dendrite 3 (40 g L'1 SCN') 1.72 95.6 0.759 11.6 a very little black coating coral-like deposits 3.2 (no SCN') 2.05 89.2 0.968 13.8 no black coating small dendrite 3.2 (40 g L'1 SCN) 1.93 93.8 0.867 14.0 no black coating small dendrite 3.5 (no SCN') 2.08 77.85 1.13 17.9 no black coating small dendrite 3.5 (40 g L'1 SCN') 1.97 89.0 0.934 18.0 no black coating small dendrite 4 (no SCN') 2.15 40.9 2.22 37.0 no black coating almost no dendrite 4 (40 g L'1 SCN) 2.08 58.8 1.49 38.8 no black coating very small dendrite 4.5 (no SCN') 2.11 7.85 11.4 54.9 no black coating no dendrite 4.5 (40 g L' 1 SCN') 2.05 8.53 10.1 54.1 no black coating no dendrite F r o m Figure 5-4, the cathodic current eff ic iency o f copper deposi t ion decreases w i t h increasing mole ratio o f cyanide to copper. A t a C N : C u mole ratio < about 3.3, the current eff ic iency decreases s l ight ly w i t h increasing mole ratio o f cyanide to copper and at a C N : C u mo le ratio > 3.3, it decreases s ignif icant ly w i t h increasing C N : C u mo le ratio. F r o m F igure 3-3, w i t h increasing C N : C u mole ratio, the equ i l i b r ium potential for the C u ( I ) / C u couple decreases s ignif icant ly and the species o f copper cyanide shifts f rom the l o w l y coordinated complexes ( C u ( C N ) 2 " and C u ( C N ) 3 2 " to the h igh ly coordinated c o m p l e x ( C u ( C N ) 4 3 " ) . Therefore at a higher C N : C u mole ratio, the discharge o f copper (I) takes place at a more negative potential where more hydrogen was evo lved . In another aspect, free cyanide suppresses the hydrogen evolu t ion [113]. W i t h increasing C N : C u mo le ratio, the hydrogen evo lu t ion should be suppressed. Increasing mole ratio o f cyanide to copper has m u c h more inh ib i t ing effect o n copper deposi t ion than on the hydrogen evolu t ion . Therefore the current eff ic iency decreases w i t h increasing mole ratio o f cyanide to copper. 70 Thiocyanate can increase the current eff ic iency o f copper deposi t ion. The effect o f thiocyanate is dependent o n the C N : C u mole ratio. A t a C N : C u mole ratio = 3-3.3, the effect is sma l l and at a C N : C u mole ratio = 3.3 - 4.4, the effect is significant. A t a C N : C u m o l e ratio = 4.5, the effect is very smal l . A t a l o w C N : C u mole ratio (3-3.3), the current e f f ic iency o f copper deposi t ion is very h igh and w i l l not be improved s ignif icant ly by thiocyanate. A t a C N : C u m o l e ratio - 4.5, the free cyanide concentrat ion is h igh (about 0.5 M ) . Free cyanide also suppresses the hydrogen evolu t ion s ignif icant ly [113]. The effect o f free cyanide o n hydrogen evo lu t ion m a y be m u c h higher than that o f thiocyanate or the co-effect o f free cyanide and thiocyanate on the hydrogen evolu t ion is close to that o f free cyanide . Therefore thiocyanate does not improve the current eff ic iency very m u c h . H o w e v e r , thiocyanate accelerates the format ion o f dendrites o n the cathode and produces poor qual i ty copper. T h i s m a y be another reason for the increase i n copper current eff ic iency i n the presence o f thiocyanate. A t a C N : C u mole ratio = 3, the electrolyte became b r o w n w h e n adding C u C N , N a S C N , N a 2 S 0 3 and N a O H into the Er lenmeyer flask. T h i s m a y be caused by an u n k n o w n react ion between thiocyanate and sulphite. The energy consumpt ion increases s l igh t ly w i t h increasing C N : C u mo le ratio f rom 3 to 3.5 and increases s ignif icant ly at a C N : C u mo le ratio > 4 due to the significant decrease i n the current eff ic iency. Since thiocyanate increases the current eff ic iency and the conduct iv i ty o f the electrolyte, the energy consumpt ion i n the presence o f thiocyanate is lower than that i n its absence. F igure 5-5 shows the anodic current eff ic iency o f cyanide vs . C N : C u mo le ratio at 60 ° C and 250 A m" 2 . The anodic current eff ic iency o f cyanide increases w i t h increasing C N : C u mo le ratio. In the C N : C u mole ratio range 3-3.2, the anodic current ef f ic iency o f cyanide increases s l ight ly w i t h increasing C N : C u mole ratio. A t a C N : C u mole ratio > 3.5, the anodic current eff ic iency o f cyanide increases rapid ly w i t h increasing C N : C u mo le ratio. A t a C N : C u mo le ratio = 3, a very smal l amount o f b lack copper ox ide was observed o n the anode and at a C N : C u mo le ratio > 3.2, no precipitate was observed o n the anode. T h i s is apparently due to the fact that cyanide stabil izes copper i n the fo rm o f copper(I) cyanide complex . The presence o f thiocyanate does not decrease the consumpt ion o f cyanide. 71 3 3.5 4 4.5 Mole ratio of Cyanide to copper Figure 5-4 Ca thodic current eff ic iency o f copper deposi t ion ( C . E . ) and power consumpt ion (P .C . ) vs . the C N : C u mole ratio at 60 ° C and 250 A m" 2 . E lec t ro ly te :70 g L " 1 C u , 63 g L " 1 N a 2 S 0 3 , 10 g L " 1 N a O H , and different cyanide concentrations i n the presence and absence o f 40 g L " 1 S C N " 1 . 60% -r 0% J i 1 1 3 3.5 4 4.5 Mole ratio of cyanide to copper Figure 5-5 A n o d i c current eff ic iency for cyanide ox ida t ion vs. the mo le ratio o f cyanide to copper at 250 A m" 2 and 60 °C. Elect rolyte :70 g L ' 1 C u , 63 g L " 1 N a 2 S 0 3 , 10 g L " 1 N a O H , and different cyanide concentrations i n the presence and absence o f 40 g L " 1 S C N " 1 . The c e l l voltage vs. t ime o f electrolysis at different C N : C u mo le ratios is s h o w n i n F igure 5-6. In the C N : C u mo le ratio range 3-4, the ce l l voltage first increases q u i c k l y to a certain value and then decreases w i t h increasing t ime o f the electrolysis . H o w e v e r , at a 72 C N : C u mole ratio = 4.5, the ce l l voltage increases s l o w l y w i t h increasing t ime o f e lectrolysis and then quite rap id ly w i t h t ime and exceeds the values for lower C N : C u mole ratios, and f ina l ly reaches a m a x i m u m value and decreases s l ight ly w i t h t ime. Genera l ly , the c e l l vol tage increases w i t h increasing C N : C u mole ratio. These phenomena are probably largely related to the cathodic process. A t the beginning o f the electrolysis , the cathode was not covered w i t h copper and the overpotential for hydrogen on S S 316 stainless steel is m u c h lower than o n copper and hydrogen evolu t ion accounted for a significant part o f the cathodic current. Af t e r the cathode was covered w i t h copper, the overpotential for hydrogen evo lu t ion became larger. Therefore the cathodic potential had to move to a more negative potential to main ta in a constant current, resul t ing i n the increase i n the c e l l voltage. T h i s is consistent w i t h the observations o f the cathode: at the beginning, more hydrogen bubbles appeared o n the cathode and after the cathode was covered w i t h copper, the quantity o f bubbles decreased and c e l l voltage increased. A t a C N : C u mole ratio = 3-4, the copper was ready to deposit o n the cathode and comple te ly covered the cathode i n a short t ime. A t a C N : C u mo le ratio = 4.5, it was diff icul t to deposit copper o n the cathode and it took a longer t ime (5 hours) to cover the cathode w i t h copper comple te ly . W h e n the t ime o f the electrolysis was i n the range 0.5 to 4 hours, the coverage o f copper was l o w and so hydrogen overpotential was l o w , the c e l l voltage was lower than the values w i t h l o w e r C N : C u mo le ratios. The deposit and its dendrites were g r o w i n g w i t h t ime and the distance between the cathode and the anode decreased and the real surface area became larger, resul t ing i n a l o w polar iza t ion . Therefore a lower c e l l voltage is needed to keep a constant current. The increase i n the ce l l voltage w i t h increasing mo le ratio o f cyanide to copper can be expla ined by the decrease i n the redox potential for C u ( I ) / C u and the increase i n the overpotential o f hydrogen evolut ion . The increase i n the C N : C u mo le ratio s igni f icant ly shifts the potential for C u ( I ) / C u to more negative values and the dis t r ibut ion o f copper cyanide shifts f rom the electroactive species (dicyanide or probably t r icyanide) to the non-electroactive species (tetracyanide), result ing i n a l o w exchange current. Fur thermore, the increase i n the C N : C u mole ratio also increases the free cyanide concentrat ion, w h i c h i n turn increases the overpotential o f hydrogen evolu t ion [113].Therefore the cathode has to be kept at a lower potential to main ta in a constant current. The increase i n the mo le ratio o f cyanide 73 to copper can increase the conduct iv i ty o f the electrolyte and decrease the c e l l vol tage, but the decrease i n IR i n the electrolyte is smaller than the increase i n the absolute value o f cathode potential . 2.3 0 2 4 6 8 10 12 Time / Hours Figure 5-6 C e l l voltage vs. t ime o f electrolysis at 250 A m" 2 and 60 ° C . Elec t ro ly te :70 g L " 1 C u , 63 g L " 1 N a j S C ^ , 10 g L " 1 N a O H , and different cyanide concentrations i n the absence o f S C N ' 1 . 5.3.4 Effect of Temperature Temperature is expected to be an important factor affecting both the anodic and cathodic processes. Three temperatures (40, 50 and 60 ° C ) were tested and the results are g i v e n i n Tables 5-4 and 5-5. The cathodic current eff ic iency o f copper depos i t ion decreases w i t h decreasing temperature. The anodic current eff ic iency o f cyanide ox ida t ion decreases w i t h increasing temperature and the c e l l voltage decreases w i t h increasing temperature. The increase i n the c e l l voltage was partly caused by the decrease i n the conduc t iv i ty o f the electrolyte. 74 Table 5-5 Resul ts o f copper e lec t rowinning at 250 A m" 2 and different temperatures. Elect rolytes : 70 g L " 1 C u , C N : C u mole ratio = 3, 63 g L ' 1 N a 2 S 0 3 and 10 g L " 1 N a O H i n the presence and absence o f S C N " . [SCN1] / g L"1 Temp. / °C Average cell voltage / V Copper C. E. / % Anodic C.E. for CN" / % Anode surface cathode copper condition 0 50 2.12 85 ± 3 21 ± 5 a very little black coating small dendrite 40 50 2.01 88 ± 3 20 ± 5 a very little black coating coral-like deposits 0 40 2.20 82 + 3 25 ± 5 a very little black coating small dendrite 40 40 2.07 86 ± 3 25 + 5 a very little black coating coral-like deposits 5.4 Summary O f four sacr i f ic ia l species (sulphite, methanol , thiocyanate and ammonia ) , on ly sulphite can effect ively l i m i t the ox ida t ion o f cyanide. W h e n the compos i t i on o f the electrolyte was contro l led at 50-60 g L " 1 N a 2 S 0 3 , 70 g L " 1 C u , C N : C u mo le ratio = 3-3.2, the anodic current eff ic iency o f cyanide decreased f rom about 100 % to 10-20 % i n the current range 250-500 A m" 2 and the temperature range 50-60 ° C . The copper depos i t ion current eff ic iency was 90-96 % and the energy consumpt ion was 0.76-1.0 k W h / k g C u . The anodic current eff ic iency o f cyanide increased f rom about 15 % to 56 % w i t h increas ing C N : C u mole ratio f rom 3 to 4.5 at [Cu] = 70 gL" 1 . W i t h increasing current density, the anodic current ef f ic iency o f cyanide decreases greatly at the current > 500 A m" 2 and s l ight ly at the current < 500 A m" 2 . The anodic current eff ic iency o f cyanide decreases s l ight ly w i t h increas ing temperature. The copper deposi t ion current eff ic iency decreases w i t h increasing C N : C u m o l e ratio and decreasing temperature. The presence o f thiocyanate increases the copper depos i t ion current eff ic iency at C N : C u mole ratio > 4. 5. 75 6. THE ANODIC OXIDATION OF SULPHITE ON A GRAPHITE ANODE IN ALKALINE SOLUTION In Chapter 5, it was noted that o f the addit ives tested on ly sulphite c o u l d effect ively l i m i t the ox ida t ion o f cyanide o n a graphite anode. In order to further the development , it therefore is important to understand the kinet ics o f the anodic ox ida t ion o f sulphite o n graphite. H o w e v e r , the anodic ox ida t ion o f sulphite i n a lkal ine solutions has not been invest igated thoroughly and the publ i shed results are inconsistent. F o r the purpose o f us ing sulphite ox ida t ion as an alternative anode react ion i n copper cyanide e lec t rowinning , the avai lable in format ion is inadequate and further studies on the anodic ox ida t ion o f sulphite i n a lkal ine solu t ion are needed. Therefore a study o f the anodic ox ida t ion o f sulphite was conducted o n a graphite electrode us ing the rotating disc technique and the potential sweep method. 6.1 Some Fundamental Aspects of Rotating Disk Electrodes Rota t ing disk electrodes ( R D E ) have been employed for the study o f a great var ie ty o f e lec t rochemical processes due to certain advantages over other types o f s o l i d electrodes. The major advantage l ies i n the development o f a un i fo rm diffusion layer, the thickness o f w h i c h can be calculated at a g i v e n rotational speed. So , the un i fo rm mass transfer towards and away f rom the electrode surface can be changed by changing the rotational speed i n a pre-determined wa y . R D E theory has been described by L e v i c h [322] w h i l e a comprehensive presentation o n these electrodes is discussed i n two monographs by P l e s k o v and F i l i n o v s k i i [323] and by Opekar and B e r a n [324]. W h e n a rotating disk rotates i n a v i scous and incompress ib le l i q u i d at an angular ve loc i ty co, the l i q u i d layer immedia te ly adjacent to the disc surface takes part i n the rotational mot ion . The layers not immedia te ly adjacent to the disc must also rotate o w i n g to the v iscous forces. U s i n g c y l i n d r i c a l coordinates (r, (p, z) the l i q u i d ve loc i ty can be d i v i d e d into three components: V r - rad ica l d i rec t ion caused by centrifugal force, V^, - az imutha l d i rec t ion due to the l i q u i d v i scos i ty and V z - ax i a l d i rec t ion resul t ing f rom the pressure drop. These ve loc i ty components descr ibed by the Nav ie r -S tokes equation are a funct ion o f rotational speed, l i q u i d v i scos i ty , ver t ica l distance f rom the d i sk (z) 7 6 and rad ia l distance (r). U n d e r these condit ions: (1) the f l o w is non-turbulent ( d V / d t = 0), (2) the f l o w is independent o f the coordinate cp, because o f ax ia l symmetry , (3) the f l u i d is incompress ib le and the boundary is hor izonta l , (4) variat ions i n the pressure i n the boundary layer are dependent on ly o n z and a sufficient angular ve loc i ty generates strong convec t ion , so that contributions f rom extraneous forces are e l iminated [324]. These three components can be represented by the f o l l o w i n g equations [323, 324] : Vr = rcoF(Q Vv = rcoG(%) V__ = Jr~^H(^) (6-1) where £, = (co/v) 1 / 2 z - dimensionless distance f rom the d isk surface, v is the k inemat ic v i scos i ty , co the angular ve loc i ty , r the radial distance and z the ver t ica l distance f rom the disk. F ( ^ ) , G ( ^ ) and H ( ^ ) are dimensionless functions w h i c h have different formulae. z Figure 6-1 Rota t ing disk coordinate system used i n calculat ions o f l i q u i d f l o w near the rotating disk. The thickness o f the dif fus ion layer (8) depends on Schmid t number (Sc = v / D ) [323]. W h e n Sc is larger than 1000, the w e l l - k n o w n L e v i c h equation (Equa t ion 6-2) has sufficient accuracy to express the thickness o f the dif fus ion layer. F o r aqueous solutions, the L e v i c h equation can be appl ied since Sc = v / D « 1 0 3 . 8 = 1.611(D / v ) 1 / 3 ( v / co)1'2 = 1.61 WmvlV2 (6-2) W h e n Sc is b e l o w 1000, the f o l l o w i n g equations should be used: 5= 1 . 6 1 1 ( J D / v ) 1 / 3 ( v / < y ) - 1 / 2 ( l + 0 . 3 5 3 9 ( £ > / v ) 0 3 6 ) 250 < Sc < 1000 (6-3) S- 1 .611(D/ v ) 1 / 3 ( W f f l ) " 1 / 2 ( l + 0 . 3 5 3 9 ( D / v ) " 1 / 3 + 0 . 1 4 5 1 4 ( D / v 2 / 3 ) S c > 100 (6-4) 77 The above equations are based on the laminar f l o w condi t ion . W h e n the R e y n o l d s number (Re = cor2/v) exceeds a c r i t i ca l value, the f l u id f l o w changes qual i ta t ively f rom laminar to turbulent. C o n v e r s i o n is gradual. Firs t , the edge o f the d i sk is affected by turbulence and this gradual ly extends toward the center w i t h increasing ve loc i ty o f rotat ion. The c r i t i ca l R e value is 1.8-3.Ix 10 5 [324]. T o avo id turbulence, the m a x i m u m rotat ional speed for 10 m m rotating disk is (1.8-3.1)xl0 5(60v/27cr 2) = 17200-29600 r p m . W h e n the d i sk vibrates ax i a l l y or rad ia l ly and the surface is uneven, turbulence appears at R e values b e l o w the c r i t i ca l values. Ano the r extreme occurs for R e « 10, when the thickness layer becomes comparable w i t h the d imensions o f the d isk and natural convec t ion i n the so lu t ion begins to p l ay a role. Therefore the rotat ional speed must be m u c h larger than 10(60v/27rr2) « 1 r p m . The thickness o f the boundary layer is sufficiently smaller than the radius o f the d i sk and R e is suff ic ient ly large to make natural convec t ion negl ig ib le . The rotational speed e m p l o y e d is usua l ly f rom 100 to 6000 r p m and so the L e v i c h equation can be accurately appl ied . The ratio o f the diameters o f the outer insulator to the electrode disc should be large enough to m i n i m i z e the edge effects. The l i m i t i n g current density (i,) for the s imple react ion ( O + ne = R ) equals the dif fus ion current density ( i d ) and can be expressed as where n is the number o f electrons transferred, F the Faraday constant (96487 A s moi" 1 ) , C b the bu lk concentrat ion ( m o i dm" 3) and D the dif fus ion coefficient. It should be noted that Equa t ion 6-5 is on ly v a l i d w h e n the transport number o f electroactive species i (t ;) is zero. W h e n the transport number ( t ) is not negl ig ib le but is smaller than 0.1 and the charge number o f the i o n i c species is equal to n , the l i m i t i n g current density can be expressed by the f o l l o w i n g equation [325]: F r o m E q u a t i o n 6-5, the dif fus ion coefficients for e lec t rochemical species can be ca lcula ted f rom the slopes o f the straight l ines for the plot o f i , vs . co,/2. nFDCb S (6-5) h = nFDCb 0.62nFD2/3y-mG)U2Cb 3 ( 1 - 0 ' (6-6) 78 The rotating disk is a powerfu l tool for determining react ion order and the rate constant. There are many methods to determine the react ion order us ing a rotating d i sk and some o f them are discussed i n publ i shed monographs [323, 324] . The der iva t ion o f a fo rmula w h i c h can be appl ied to determine the react ion order and rate constant w h e n the l i m i t i n g currents have been measured is g iven be low. The current density for m i x e d kinet ics at a rotating disk electrode is determined by the heterogeneous react ion w i t h the dif fus ion o f the reactant and the rate o f the heterogeneous react ion be ing equal to the dif fus ion rate under steady-state condi t ions . Therefore w h e n the charge transfer coefficient is independent o f the reactant concentrat ion and the reverse react ion is neg l ig ib le , the current density for a s imple redox react ion (O + ne = R ) can be expressed as: i = nFk(Cs)"' (6-7) i = nFD(—) sinface = nFD—^—^ = - ^ - ) (6-8) where i is the current density, n r the react ion order, k the react ion rate constant, and C s the electrode surface concentration. F r o m Equat ions 6-7 and 6-8, w e have the f o l l o w i n g equations: Q = Q ( 1 - - ) (6-9) h I I, i = nFkCb"r (1 - T)"' = ik 1 -7- (6-10) V i,J i log i = log ik + nr l og ( l - 7 ) (6-11) h where ik = nFkCb"r is the k ine t i ca l ly cont ro l led current. The react ion order can be calcula ted f rom the plot o f l o g i vs . l o g ( l - i / i , ) and the k ine t i ca l ly cont ro l led current can be obtained f r o m the intercept o n the y-axis . The react ion order is obtained at constant i on ic strength and the effects o n the double layer and the act ivi ty coefficient are neg l ig ib le due to the change i n the reactant concentrat ion. Furthermore, i n this method it is not necessary to k n o w the concentrat ion o f the reactant. The exchange current and Tafe l slope can be obtained f rom the plot o f i k vs . overpotential . I f n r = 1 (first order), we get the K o u t e c k y - L e v i c h equat ion f rom Equa t ion 6-10: 79 I__L 1 /' zt i, (6-12) Equat ions 6-7, 6-8, 6-11 and 6-12 are also v a l i d for redox reactions such as O + X +ne = R w h e n the react ion order w i t h respect to X is zero or the concentrat ion o f X is kept at an elevated l eve l so that there is no difference between the surface and the bu lk concentrat ion. In these cases, the kinet ic expression can be reduced to E q u a t i o n 6-7. 6.2 Thermodynamics of Sulphite Oxidation A s was discussed i n Chapter 2, Sect ion 2.5, sulphite exists i n the f o r m o f S 0 2 (aq), H S O y and S 0 3 2 " i n aqueous solut ion. S 0 2 (aq), H S 0 3 " and S 0 3 2 " species are predicted to predominate over the p H ranges < 1.8, 1.8 - 7 and > 7, respect ively. A t p H > 12, the dominant species i n solu t ion is S 0 3 2 " . The anodic ox ida t ion o f sulphite i n a lka l ine so lu t ion o n graphite can be expressed by the f o l l o w i n g equations: S 0 3 2 " + 2 0 F T = S 0 4 2 " + H 2 0 + 2e (6-13) 2 S 0 3 2 " = S 2 0 6 2 ' + 2 e (6-14) The produc t ion o f dithionate o n graphite (Equat ion 6-14) can be neglected accord ing to the literature [211].The standard equ i l i b r ium potentials for Equa t ion 6-13 are -0.936 , - 0.957, -0.971, -0.985 V vs. S H E at 25, 40, 50 and 60 °C respect ively obtained by ca lcu la t ion us ing rel iable thermodynamics data [295, 296] . The Nernst equation for the e q u i l i b r i u m potential for Equa t ion (6-13) is expressed as: n R T , aSO/-aH20 \aSO*-aOH-2J (6-15) There are many methods for ca lcula t ing act ivi ty coefficients i n strong electrolytes such as the Guggenhe im , B r o m l e y , Meissner , C h e n and P i t ze r ' s methods [326]. P i t ze r ' s i o n - i o n interact ion m o d e l is good for calcula t ing the act ivi ty coefficient o f a single species i n m u l t i -component strong electrolytes [326-328] and it has been used i n this study. In P i t ze r ' s 80 method, the concentrat ion is expressed i n mola l i ty and so the act ivi ty o f species i , is a, = m^ y (. The m o l a l i t y o f species i ( m j has the f o l l o w i n g relat ionship w i t h the molar i ty ( Q ) [329] m< = p - 0 . 0 0 1 E C , M , ( 6 ' 1 6 ) where p is the density o f electrolyte. In the presence o f 1 M N a 2 S 0 4 support ing electrolyte, mj « 1.02 Cj (p = 1.12 [330]). F o r convenience, the molar i ty replaces the m o l a l i t y as an approx imat ion for ca lcula t ion . The interaction o f S 0 3 2 " w i t h N a + and O H " is rough ly s imi l a r to that o f S 0 4 2 " [328] and the act ivi ty coefficients o f S 0 3 2 " and S 0 4 2 " are close [331]. Therefore the ac t iv i ty coefficient o f S 0 3 2 " is assumed to equal to that o f S 0 4 2 " . The ac t iv i ty coefficients o f water and hydrox ide ions have been calculated us ing P i t ze r ' s method (see A p p e n d i x 5). The calculated water act ivi ty , act ivi ty coefficient o f hydrox ide and the potent ial for S 0 4 2 " and S 0 3 2 " couple at 25 , 40, 50 and 60 ° C are l is ted i n Table 6-1. The water ac t iv i ty is almost independent o f the temperature and the hydrox ide ac t iv i ty coefficient decreases s l ight ly w i t h increasing temperature. Table 6-1 The act ivi t ies and act ivi ty coefficients for 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 at 25 , 40 , 50 and 60 ° C Temperature / °C 25 40 50 60 a^v 1.03 1.03 1.03 1.03 YOH(I) 0.486 0.470 0.459 0.448 E ( S 0 4 2 7 S 0 3 2 " ) vs S H E / V -0.822 -0.837 -0.8.46 -0.855 6.3 Experimental Apparatus and Set-up A n N E - 1 5 0 graphite rod (impregnated w i t h res in and carbonized at 500 ° C i n vacuum) f rom N a t i o n a l E lec t r i c C a r b o n C o . was used to make a graphite rotating disk. The graphite was mach ined to 4 m m and t ight ly surrounded w i t h a plast ic shie ld . A spr ing was used to conduct the electr ici ty f rom the shaft to the graphite electrode. F igure 6-2 shows the schematic d iagram o f the rotating disk. 81 Steel holder Plastic holder -(Outer insulator) Graphite electrode Shaft Screw Cu for electrical contact Spring Cu for electrical contact Figure 6-2 Schematic d iagram o f rotating disk The electrode surface was first ground using 600-gri t sandpaper, po l i shed w i t h 4000-grit s i l i c o n carbide sandpaper and then soft tissue paper. F i n a l l y the surface was checked under a microscope for surface smoothness. T o ensure reproducible results, the electrode was first treated by c y c l i c vo l tammetry between 0 and 0.75 V vs. S C E at 100 m V s"1 for 30 minutes and po la r i zed at l m V s"1 unt i l the electrode reached a stable condi t ion . The electrode was tested i n ferrous and ferric cyanide solut ion. The l i m i t i n g current density was the same as o n a Pt rotating disk f rom E G & G C o . and was propor t ional to the square root o f the rotat ional speed. Therefore the graphite electrode was considered to be un i fo rm. The graphite hav ing a 12-mm diameter was fashioned as a rotating d i sk for the coulometr ic measurements. The rotating disc electrode system was an E G & G P A R C M o d e l 636 Elec t rode Rotator. The potentiostat was a S O L A R T R O N 1286 E lec t rochemica l Interface. 100 m L o f the solu t ion o f the required compos i t ion were p laced i n an E G & G water-jacketed electrolyt ic c e l l whose temperature was contro l led by a water bath c i rculator . The experiments were carr ied out under an argon atmosphere to protect the sulphite f rom ox ida t ion by air. The reference electrode was a F I S H E R saturated ca lome l electrode ( S C E ) w h i c h was connected to the c e l l electrolyte by an electrolyte br idge. The c a l o m e l electrode was p laced i n a tube containing the same electrolyte as i n the ce l l . The temperature was kept at 25 ° C us ing a water bath. The o h m i c drop between the w o r k i n g electrode and the reference electrode was compensated by the current interruption technique. A schematic d iagram o f the 82 exper imental set-up is shown i n F igure 6-3. Excep t as noted, the po la r iza t ion curves were generated us ing the potential sweeping method at 1 m V s"1. C E O Pt counter electrode Ag-C bush S C E e i e c t r ode Electrolyte bridge Cover Figure 6-3 Schematic d iagram o f the experimental setup A Cannon-Fenske routine viscometer (size 25) was used to measure the k inemat ic v i scos i ty o f the solutions studied. The experimental set-up and the measur ing procedure are shown i n A p p e n d i x 6. The concentrat ion o f sulphite was measured by adding an excess o f standard iodine solu t ion f o l l o w e d by back ti tration w i t h standard thiosulphate so lu t ion (see A p p e n d i x 4). The l i q u i d j unc t i on potential , estimated by the Henderson equat ion [332] (see A p p e n d i x 7), was b e l o w 2 m V and so can be neglected. The thermal l i q u i d j u n c t i o n potential was measured us ing two ca lome l reference electrodes w h i c h were p laced o n the two sides o f an electrolyte br idge. Reagent grade chemicals were used throughout the invest igat ion. 6.4 Polarization Measurements The po la r iza t ion measurements were carr ied out at 25, 40, 50 and 60 ° C i n 1 M N a 2 S 0 4 solutions containing 0.025 to 0.5 M N a 2 S 0 3 and 0.025 - 0.25 M N a O H . I f the app l ied potential was larger than about 1.0 V vs. S C E , the surface o f the electrode was cor roded and 83 became rough, affecting the current measurements (e.g. the l i m i t i n g current became m u c h lower and the current vs. potential was non-reproducible) . Therefore the electrode surface was repol ished for every polar iza t ion measurement to ensure reproducible results. T y p i c a l po la r iza t ion curves for 0.1 M N a 2 S 0 3 solutions containing 0.25 M N a O H are s h o w n i n F igu re 6-4 and those for 0.05, 0.2, 0.4 and 0.5 M N a 2 S 0 3 i n Figures A - l to A - 4 i n A p p e n d i x 8. The anodic ox ida t ion o f sulphite began at 0.16, 0.12, 0.08 and 0. 04 V vs. S C E for 25, 40 , 50 and 60 ° C respect ively. D u e to the presence o f sulphite ions, o x y g e n evo lu t ion was suppressed and the cor ros ion o f the electrode was d imin ished . The higher the concentrat ion o f sulphite, the greater were these effects. The oxygen evolu t ion increases w i t h increasing temperature. A t [ N a 2 S 0 3 ] > 0.4 M and 25 - 60 ° C , almost no oxygen bubbles were fo rmed and the graphite was o n l y s l ight ly corroded. W h e n the current reached a l i m i t i n g value, it became independent o f the potential . A t [ N a 2 S 0 3 ] = 0.05-0.4 M , the l i m i t i n g current was approximate ly propor t ional to the sulphite concentration. H o w e v e r , the increase i n the l i m i t i n g current due to the increase i n sulphite concentrat ion f rom 0.4 to 0.5 M was m u c h smal ler than expected. The l i m i t i n g current was l im i t ed probably by O H " dif fus ion at 0.5 M N a 2 S O s . The background current i n the absence o f sulphite is independent o f the rotat ional speed (Figure A - 5 i n A p p e n d i x 8). H o w e v e r , the current measured i n the presence o f sulphite is sensit ive to the rotational speed and the l i m i t i n g current is propor t ional to the square root o f the rotat ional speed. O x y g e n evolu t ion and the cor ros ion o f graphite are greatly suppressed i n the presence o f sulphite. A t 100 r p m (Figure 6-4), the oxygen evo lu t ion even decreased the current poss ib ly because the oxygen bubbles were not r emoved eff ic ient ly . Therefore the background current i n the presence o f sulphite cou ld be m u c h smaller than that measured i n the absence o f sulphite and cou ld make a negl ig ib le contr ibut ion to the total current. The background current i n the absence o f sulphite was inappropriate for correct ing the current for the sulphite ox ida t ion due to oxygen evolu t ion at h i g h potentials. The cond i t ion o f the surface o f the graphite electrode var ied after the electrode surface was renewed each t ime. Therefore after the same treatment o f the electrode, the values o f current vs. potential scattered to some extent ( ± 1 5 % ) . H o w e v e r , the l i m i t i n g currents scattered less ( ± 2 % ) . 84 0.2 0.4 0.6 Potential vs 0.8 1.0 SCE/V 0.40 0.60 0.80 1.00 Potential / V vs. SCE 1.20 (a) 25 ° C (b) 40 ° C 3000 3500 0.00 0.20 0.40 0.60 0.80 1.00 Potential / V vs. SCE 1.20 c £ 1000 t_ 3 ° 500 4-3000 - _»_4900 rpm CM -m- 3600 rpm E 2500 - _ * _ 2 5 0 0 rpm < _*_ 1600 rpm £< 2000 - _*_ 900 rpm 'in c - • - 4 0 0 rpm ® 1500 - _ , _ 100 rpm 0.2 0.4 0.6 0.8 1.0 Potential vs. SCE /V (c) 50 ° C (d) 60 ° C F igure 6-4 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25 , 40 , 50 and 60 ° C . E lec t ro ly te : 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 . F r o m Figure 6-5, the polar iza t ion curves for 0.1 M N a 2 S 0 3 solutions conta in ing 0.05 M N a O H are quite different f rom those containing 0.25 M N a O H . The current first reached a l i m i t i n g value, and then increased s l ight ly w i t h increasing potential and f ina l ly increased w i t h increasing potential and reached a second l i m i t i n g value. V e r y litt le o x y g e n was e v o l v e d at potentials > 1.4 V vs. S C E . S i m i l a r po lar iza t ion curves for 0.2 and 0.4 M N a 2 S 0 3 are s h o w n i n F igures A - 6 and 7 i n A p p e n d i x 8. The first l i m i t i n g current increased s l igh t ly w i t h increasing sulphite concentrat ion f rom 0.1 to 0.4 M and was m u c h smal ler than the va lue expected for the corresponding sulphite concentration. The second l i m i t i n g current (observed 85 at 100, 400 and 900 rpm) was propor t ional to the square root o f the rotat ional speed and was a l i t t le higher than the l i m i t i n g value obtained i n the solut ion conta in ing 0.25 M N a O H at the same sulphite concentration. A t [ N a O H ] = 0.05 M , the mole ratios o f sulphite to hydrox ide i n the so lu t ion are 2, 4, 8 respect ively for 0.1, 0.2 and 0.4 M N a 2 S 0 3 . The ox ida t ion o f one sulphite i o n needs two hydrox ide ions accord ing to Reac t ion 6-13. So the equivalent ratios o f sulphite to hydrox ide are 4, 8 and 16 respect ively for 0.1, 0.2 and 0.4 M N a 2 S 0 3 . So the mass transfer rates o f hydrox ide ions have to be 4, 8, 16 t imes those o f sulphite ions respect ively for 0.1, 0.2 and 0.4 M to main ta in the a lkal ine condi t ion o n the electrode surface. A t infini te d i l u t i on , the di f fus ion coefficient o f hydroxide (5.26x10" 9 m 2 s"1 at 25 ° C ) is 4.96 t imes that o f sulphite ions (1.06 x l O " 9 m 2 s"1) [318]. A t [ N a 2 S 0 3 ] = 0.2 and 0.4 M , the current becomes so h i g h that the mass transfer rate o f hydrox ide ions is not h igh enough to main ta in the hydrox ide concentrat ion above a certain value (close to zero). Therefore the species o f sulphite shifts f rom S 0 3 2 " to H S 0 3 " and S 0 2 and the properties o f the surface oxygen-carbon function groups can be changed due to the pro ton exchange [334]. The anodic ox ida t ion o f H S 0 3 " and S 0 2 begins at h igher potentials [240] and the change i n the properties o f the surface funct ion group m a y result i n a pass ivat ing effect. Therefore w h e n the first l i m i t i n g current appeared, the o x i d i z e d species o f sulphite changed f rom S 0 3 2 " to H S 0 3 " and S 0 2 and w i t h further increase i n potential , the current increased due to the ox ida t ion o f H S 0 3 " and S 0 2 . F i n a l l y the current reached a second l i m i t i n g value related to the m a x i m u m diffusion rate o f the sulphite species. In the presence o f 1 M N a 2 S 0 4 as support ing electrolyte, the d i f fus ion coeff icient o f hydrox ide ions c o u l d decrease more than that o f sulphite ions and so the ratio o f the d i f fus ion coefficients m a y be lower than 4. Therefore two l i m i t i n g currents for the so lu t ion conta in ing 0.1 M N a 2 S 0 3 and 0.05 M N a O H appeared. Th i s c o u l d be the same as that observed for the so lu t ion conta in ing 0.5 M N a 2 S 0 3 and 0.25 M N a O H because the mo le ratio o f sulphite to N a O H (2) is the same. A t [ N a O H ] = 0.05 and 0.25 M , the polar iza t ion curves measured i n the solut ions conta in ing 0.1 M , 0.2 and 0.4 M N a 2 S 0 4 at 400 r p m are shown i n F igure 6-6. C o m p a r e d to the po la r iza t ion curves w i t h the same sulphite concentrations, we can see: (1) at a current < about 380 A i n 2 , the current for 0.05 M N a O H is almost the same as that for 0.25 M N a O H , 86 (2) at a current density > about 380 A m" 2 , w i t h further increase i n potential , the current densities for 0.05 M N a O H are lower than those for 0.25 M N a O H . T h i s phenomenon m a y be related to the l im i t ed mass transfer o f hydrox ide w h i c h should be the same at a constant concentrat ion o f hydroxide and rotational speed. A t a current density < about 380 A m" 2 , when the concentrations o f hydrox ide at the surface for a l l the solutions are above a certain value (probably p H > 9). The sulphite o n the surface exists on ly i n the fo rm o f S 0 3 2 " w h i c h is discharged on the anode and the react ion order w i t h respect to O H " is zero. Hence , the current is dependent on ly o n the potent ial and the concentrat ion o f sulphite. A t a current density > about 380 A m" 2 , the concentrat ion o f hydrox ide at the surface for solutions w i t h 0.05 M N a O H becomes so l o w that H S 0 3 " and S 0 2 increase o n the electrode surface and S 0 3 2 " decreases, w h i c h decreases the current density. H o w e v e r , the concentrat ion o f hydroxide at the surface for the solutions w i t h 0.25 M N a O H is s t i l l h i g h and the concentrat ion o f S 0 3 2 " does not decrease due to the shift o f the sulphite species f rom S 0 3 2 " to H S 0 3 " and S 0 2 . The second l i m i t i n g current for the solut ion w i t h 0.05 M N a O H is larger than that for the so lu t ion w i t h 0.25 M N a O H and the ratios o f the former to the later are 1.08, 1.12 and 1.18 respect ively for 0.1, 0.2 and 0.4 M N a 2 S 0 4 . The reason c o u l d be: (1) the decrease i n the concentrat ion o f hydroxide f rom 0.25 to 0.05 M decreases the v i scos i ty o f the so lu t ion and weakens the interaction o f ions, result ing i n a higher di f fus ion coefficient and a h igher d i f fus ion l im i t ed current, (2) the anode react ion consumes hydrox ide and even generates hydrogen ions w h i c h diffuse to the bu lk solu t ion and react w i t h S 0 3 2 " i n the d i f fu ison layer to fo rm H S 0 3 " . The dif fus ion coefficient o f H S 0 3 " (1.33 x 10"9 m 2 s"1 at infini te d i l u t i on [318]) is larger than that o f S 0 3 2 " (1.06 x 10"9 m 2 s - 1at infini te d i lu t ion [330]). The concentrat ion gradient o f S 0 3 2 " is increased, result ing i n a larger l i m i t i n g current. T h i s effect increases w i t h increasing sulphite concentrat ion because more hydrogen ions are generated at a constant hydrox ide concentrat ion i n the bu lk solut ion and therefore the ratio o f the l i m i t i n g currents increases w i t h sulphite concentration. 87 Potential vs. S C E / V Potential vs. SCE (a) 25 ° C (b) 40 ° C £ < 2000 --densi 1500 --rrent 1000 --Cui 4900 rpm -.- 3600 rpm -tr- 2500 rpm —M— 1600 rpm —MC— 900 rpm —•— 400 rpm - 4 - 100 rpm 3500 3000 CM E 2500 --< '—2000 --(A C 4) •a 1500 --c a k_ k_ 1000 --3 o 500 --0 i t 0.4 0.6 0.8 1 1.2 Potential vs. S C E / V 1.6 0.4 0.6 0.8 1 1.2 Potential vs. S C E / V (c) 50 ° C (d) 60 ° C F igure 6-5 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25 , 40 , 50 and 60 ° C . E lec t ro ly te : 0.1 M N a j S O j , 0.05 M N a O H and 1 M N a ^ . 88 2500 Potential vs. S C E / V Figure 6-6 C o m p a r i s o n o f the polar iza t ion curves w i t h different sulphite and hydrox ide concentrations at 25 ° C and 400 rpm. 6.5 Coulometric Measurements C o n t r o l l e d potential coulometry was used to determine the number o f the electrons transferred (n) for the anodic ox ida t ion o f the sulphite i on . The electrode potentials were cont ro l led at 0.6 and 0.9 V vs. S C E to a v o i d oxygen evo lu t ion and cor ros ion o f the graphite. The results are g i v e n i n Table 6-2. In a l l cases, the number o f the electrons transferred per one sulphite i o n ranges f rom 1.92 to 1.98. Th i s means that a lmost a l l o f the sulphite was o x i d i z e d to sulphate i n two-electron reaction. Hence the ox ida t ion o f sulphite to dithionate can be neglected. Potent ial and temperature had almost no effect o n the products o f the anodic ox ida t ion o f sulphite. These results are i n agreement w i t h those reported by Glass tone a n d H i c k l i n g [211]. Tab le 6-2 N u m b e r o f the electrons transferred for the anodic ox ida t ion o f sulphite Concent ra t ion o f sulphite / m o i dm" 3 Potent ial / V vs. S C E Temperature / ° C N u m b e r o f electrons transferred (n) per sulphite i o n 0.1 0.6 25 1.94 ± 0 . 0 3 0.1 0.9 25 1.98 ± 0 . 0 2 0.1 0.6 60 . 1.93 ± 0 . 0 3 0.1 0.9 60 1.97 ± 0 . 0 3 0.4 0.6 25 1.92 ± 0 . 0 4 89 6.6 Reaction Order F o r the anodic ox ida t ion o f sulphite, the concentrations o f sulphite and hydrox ide can affect the react ion rate. Therefore the kinet ics were first studied b y changing the concentrat ion o f one species w h i l e the potential and the concentrations o f the other species were mainta ined constant. W h e n the potential and p H were mainta ined constant, the current increased w i t h increasing sulphite concentration, indica t ing that the ra te-control l ing step i n v o l v e d sulphite ions. H o w e v e r , when the potential and sulphite concentrat ion were mainta ined constant, the current was independent o f p H , suggesting that the react ion order w i t h respect to hydrox ide is zero. Therefore on ly the sulphite concentrat ion affects the rate o f the sulphite ox ida t ion and the k inet ic expression for the anodic ox ida t ion o f sulphite ions can be reduced to Equa t ion 6-7 over the p H range studied (11.9-13). In the m i x e d cont ro l reg ion , Equa t ion 6-11 can be appl ied to calculate the react ion order w i t h respect to sulphite. The data (current vs . potential) scattered to some extent due to the inherent surface va r i ab i l i t y after the e lec t rochemical condi t ion ing . The data i n Figures 6-4 were generated w i t h some var ia t ion o f surface cond i t ion and therefore cannot be used di rect ly to calculate the react ion order. F o r the present experiments, the stabil i ty o f the graphite surface was mainta ined by l i m i t i n g the potential range o f the experiments (0 - 0.7 V vs. S C E ) . F igure 6-7 shows the polar iza t ion curves measured o n the same electrode surface i n the potential range o f 0 - 0.7 V vs. S C E and i n a so lu t ion conta ining 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 . U s i n g the data shown i n F igure 6-7, the p lo t o f l o g i vs . l o g ( l - i / i , ) at 25 ° C is a straight l ine (Figure 6-8). A c c o r d i n g to Equa t ion 6-11, the slope o f the l ine (i.e. the react ion order) and the intercepts o n the l o g i axis ( log i k ) were calculated by least squares fi t t ing and are g i v e n i n Table 6-3. The react ion order w i t h respect to the sulphite i o n is 1. F o r the first order reaction, Equa t ion 6-12 can be appl ied and the plot o f 1/i vs . 1/i, is a straight l ine and the intercept o n the 1/i axis is l / i k . F r o m Figure 6-9, the plots o f 1/i vs . 1/i, are l inear and the slopes are 1. The intercepts o f the plot o f l og i vs . l og (1-i / i i) are the same as - l o g o f the intercepts o f the plots o f 1/i vs . at the same potential (see Tab le 6-2). T h i s means that the react ion order is 1 and therefore the two methods match very w e l l . The same results have been obtained i n solutions 0.4 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 . T h e y are s h o w n i n Figures 6-10 and 6-11 and Table 6-4. The react ion order w i t h respect to the sulphite i o n at 90 40, 50 and 60 ° C was measured at potentials b e l o w 0.65 V vs. S C E and was s t i l l one. The results at 40 , 50 and 60 ° C for 0.1 M N a ^ C ^ are shown i n F igure A - 8 . 900 , 0 0.2 0.4 0.6 0.8 Potential vs. S C E / V Figure 6-7 Po la r i za t ion curves o f sulphite ox ida t ion us ing rotating d i sk at 25 ° C . E lec t ro ly te : 0.1 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a ^ . Table 6-3 Reac t ion order and kinet ic current calculated us ing different methods for 0.1 M N a 2 S 0 3 . Potent ia l vs . S C E / V 0.50 0.55 0.60 0.65 0.70 Slope o f the p lo t o f L o g i vs . L o g ( l - i / i , ) 1.05 1.01 1.01 1.01 0.99 Intercepts o f the p lo t o f L o g i vs . L o g ( l - i / i , ) , i.e. i k / A m" 2 2.08 2.35 2.62 2.90 3.14 Slope o f plot o f 1/i vs . 1/i, 1.04 1.01 1.00 1.01 1.00 - L o g ( intercepts o f plot o f 1/i vs . 1/i,), i.e. i k / A m" 2 2.08 2.35 2.62 2.90 3.14 L o g ( i / ( l - i / i , ) ) , i.e. i k / A m" 2 2.06 2.33 2.61 2.89 3.15 Table 6-4 Reac t ion order and k inet ic current calculated us ing different methods for 0.4 M N a 2 S Q 3 . Potent ia l vs . S C E / V 0.50 0.55 0.60 0.65 0.70 Slope o f the plot o f L o g i vs . L o g ( l - i / i , ) 1.04 1.04 1.00 1.00 1.03 Intercepts o f the plot o f L o g i vs . L o g ( l - i / i , ) , i.e. i k / A m" 2 2.64 2.90 3.19 3.39 3.62 Slope o f plot o f 1/i vs . 1/i, 1.03 1.01 1.01 1.02 1.03 - L o g ( intercepts o f plot o f 1/i vs . 1/ij), i.e. i k / A m" 2 2.64 2.90 3.19 3.39 3.62 L o g ( i / ( l - i / i , ) ) , i.e. i k / A m" 2 2.65 2.92 3.21 3.42 3.63 91 Figure 6-8 L o g i vs . L o g ( l - i / i , ) at constant potential and 25 ° C . Elec t ro ly te : 0.1 M N a j S C ^ , 1 M N a j S C ^ and 0.25 M N a O H . 0.012 0.01 0.008 0.006 +L 0.004 0.002 4-0 • 0.50 V • 0.55 V • 0.60 V X 0.65 V X 0.70 V 0.0005 0.001 0.0015 (i, / A rrf 2 ) - 1 0.002 0.0025 Figure 6-9 1/i vs . 1/i, at constant potential ( V vs. S C E ) 25 ° C . Elec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . 92 2.4 -0.5 -0.4 -0.3 -0.2 -0.1 0 LOG(1-l/i|) Figure 6-10 L o g i vs . L o g ( l - i / i , ) at constant potential and 25 ° C . E lec t ro ly te : 0.4 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . F igure 6-11 1/i vs . 1/i, at constant potential ( V vs. S C E ) 25 ° C . Elec t ro ly te : 0.4 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . 93 In the l o w polar iza t ion region, the current is sma l l and therefore the concentrations at the electrode and i n the bu lk are the same. In this case, on ly E q u a t i o n 6-7 is needed to analyze the kinet ics . Equa t ion 6-8 is not required because the mass transfer is not important. Therefore the react ion order was not calculated us ing the slope o f the p lo t o f l og i vs . l o g (1 -i / i , ) , rather it was calculated f rom the plot o f l og i as a funct ion o f the sulphite concentrat ion. The plots o f log( i ) vs . l og [S0 3 2 " ] at 0.2 and 0.4 V vs. S C E at 25 ° C are s h o w n i n F igure 6-12. A t 0.4 V vs . S C E , the react ion order was close to 1. A t 0.2 V vs . S C E , the react ion order was b e l o w 1 and appeared to be nonl inear w i t h increasing reactant concentrat ion. T h i s nonl inear i ty c o u l d be caused by the variable adsorption o f sulphite ions o n the graphite electrode surface. 2.5 -2 -1.5 -1 -0.5 0 L o g ( [S0 3 2 - ] /mo i d m 3 ) Figure 6-12 L o g i vs . l o g [S0 3 2 - ] at 25 ° C and 4900 rpm. Elec t ro ly te : 1 M N a 2 S 0 4 and 0.25 M N a O H . 1 I f the react ion order is 1, the plot o f l og ( i /(1-i/i,)) (corrected for the difference i n concentrat ion o f sulphite between the bu lk electrolyte and that at the electrode surface) vs. potential should be a straight l ine . A t l o w current, (1-i/i,) is close to 1 and the concentrat ion difference can be neglected. The plots o f l og (i /(1-i/i ,)) vs . potential at 25 , 40, 50 and 60 ° C are s h o w n i n F igure 6-13. The corrected current (i/( 1-i/i,)) is the same as the k ine t ic current ( i ,J ca lcula ted us ing the above methods (see Table 6-2). There are two Tafe l slopes. The first Tafe l slope at l o w potentials was 0.059 -0.066 V decade" 1 and the charge transfer coeff icient 94 was about 1. The second Tafe l slope at h igh potentials was 0.19-0.22 V decade" 1 w i t h the charge transfer coefficient be ing i n the range o f 0.29 - 0.31. The Tafe l slopes for the different potentials ranges and temperatures are l isted i n Table 6-5. Table 6-5 Tafe l slopes ( V decade" 1) for the different potential ranges at 25, 40 , 50 and 60 ° C Temperature 25 ° C 40 ° C 50 ° C 60 ° C L o w potential range (vs. S C E / V ) 0 . 1 6 - 0 . 2 5 0.11 - 0 . 2 2 0 . 0 8 - 0 . 1 8 0 . 0 4 - 0 . 1 5 Tafe l slopes for l o w potential range 0.059 0.061 0.064 0.066 H i g h potential range (vs. S C E / V ) 0.4 - 0.7 0 . 3 8 - 0 . 6 6 0.38 - 0.64 0.36 -0.64 Tafe l slopes for h i g h potential range 0.19 0.20 0.21 0.22 The first Tafe l slope (0.060 V decade" 1) corresponds to a nonl inear react ion order (less than 1) at l o w potential (0.16 -0 . 25 V vs. S C E ) and the second Tafe l slope corresponds to a first order react ion at h i g h potentials (0.4 - 0.7 V vs . S C E ) at 25 ° C . T h i s in format ion suggests that there are two reaction mechanisms. The change i n Tafe l slope, hence i n the mechan i sm was not due to the potent ia l-dependent change i n the nature o f electrode surface because after e lec t rochemica l condi t ion ing , the electrode surface was stable over the potential range 0 - 0.7 V vs. S C E . F o r example , at 25 ° C , the background current was almost constant over the potential range 0 -0.6 V vs. S C E , but the change i n the Tafe l slope happened between 0.3 - 0.4 V vs . S C E (see F igure 6-13). The Tafe l slope change c o u l d be due to the f o l l o w i n g : at l o w potential , the ox ida t ion o f the adsorbed sulphite was dominant and at h i g h potential , the ox ida t ion o f unadsorbed sulphite was dominant . Tarasev ich et a l . [240, 243] reported that the first Ta fe l slope was 0.060 - 0.070 V decade" 1 and the react ion order obtained by the change o f sulphite concentrat ion was close to 1, H o w e v e r , these authors d i d not g ive the other Tafe l s lope. 95 0.7 0 -I 1 1 -H \ 1 1 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 L o g / A m"2] Figure 6-13 Potent ia l vs . l o g ( ( i / ( l - i / i , ) ) at different temperatures. E lec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . 6.7 Effect of pH The effect o f p H was studied by changing the sod ium hydrox ide concentrat ion i n the electrolyte conta ining 1 M N a 2 S 0 4 . Howeve r , the electrolyte contained 1 M N a 2 S 0 4 and the p H measurement was not accurate because the electrolyte had a large background concentrat ion o f N a 2 S 0 4 . Therefore the act ivi ty coefficient o f O H " was calcula ted by P i t ze r ' s m o d e l (see A p p e n d i x 5). F igure 6-14 shows the polar iza t ion curves i n 0.1 M N a 2 S 0 3 so lu t ion w i t h different concentrations o f hydroxide . The plots o f the current (corrected for the difference o f concentrat ion between the electrode surface and the bu lk solut ion) vs . p H are shown i n F igure 6-15. The current at a constant potential appears to be almost independent o f p H . Therefore the react ion order w i t h respect to O H " is almost zero. T h i s result is consistent w i t h those reported by Tarasev ich et a l . [240, 243] and means that the ra te-control l ing step does not i nvo lve O H " . 96 800 Potent ia l v s . S C E / V Figure 6-14 Po la r i za t ion curves at different hydrox ide concentrations and 25 ° C . E lec t ro ly te : 0.1 M N a 2 S 0 3 and 1 M N a 2 S 0 4 . E < o 3.5 3.0 2.5 2.0 1.5 1.0 4-0.5 0.0 4--0.5 -1.0 0.7 V vs. S C E 11.9 — • • 0.6 V vs. S C E • * * — * * 0.5 V vs. S C E -X X x x 0.4 V vs. S C E A • A 4 0.3 V vs. S C E • • — • •—• 0.2 V vs. S C E I • • i i i • l I 12.1 i i i 12.3 12.5 12.7 PH I 12.9 13.1 Figure 6-15 Effect o f p H o n sulphite ox ida t ion at different potentials and 25 ° C . E lec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 at var iable p H 6.8 Calculation of Activation Energy for the Kinetic Current 97 A t a constant potential , the f o l l o w i n g equation can be wri t ten: T . Ui(£) £ / : - a a F E L o g i t = constant + — = constant + (6-17) B k 2 . 3 0 3 R T 2 . 3 0 3 R T ^ ; Where U* (E) is the act ivat ion energy at potential E , U++ the act ivat ion energy at potential = 0, a a the anodic charge transfer coefficient and R the gas constant. The act ivat ion energy can be calculated f rom the slope o f the plot o f l og i k vs . 1/T (Figure 6-16). The slopes o f these l inear plots were calculated by least squares fi t t ing. The act ivat ion energy decreases q u i c k l y w i t h increasing potential at l o w potentials and f ina l ly behaves l inear ly w i t h potent ial at potentials > 0.4 V vs . S C E . Th i s is due to a change i n the react ion mechan i sm w h i c h results i n a change i n the charge transfer coefficient. 4 .0 -3.5 J 3.0 -^ — 4 5 . 3 kJ m o r 1 (0.6 V ) 2 .5 - — ~ ^ 4 8 ^ 4 kJ m o r 1 (0.5 V ) 2 .0 . * " — — ^ J 5 1 5 kJ m o r 1 (0.4 V ) 1.5 -~ " ~ ~ s — - . ^ 5 7 . 2 kJ m o r 1 (0.3 V ) 1.0 -0 .5 -^ ~ ^ ~ - - ^ 8 5 . 2 kJ m o r 1 (0.2 V ) 0.0 -0.5 -0 . 0 0 3 0.0031 0 . 0 0 3 2 0 . 0 0 3 3 0 . 0 0 3 4 (T/K)-1 Figure 6-16 L o g i k v s . l / T at different potentials ( V vs. S C E ) . E lec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . 6.9 Diffusion Coefficient Estimation The plots o f di f fus ion current vs. rotational speed at different temperatures are s h o w n i n F igure 6-17. These plots permit the ca lcula t ion o f the di f fus ion coefficients o f S 0 3 2 " us ing 98 the slopes o f the l ines and Equa t ion 6-4. The slopes were calculated us ing least squares f i t t ing. The k inemat ic v i scos i ty values for 0.05 M N a 2 0 3 , 0.25 M N a O H and I M N a 2 S 0 4 were 1.345, 0.982, 0.818 and 0.695 x 10' 6 m 2 s respect ively for 25 , 40 , 50 and 60 ° C . The dif fus ion coefficients at 25, 40, 50 and 60 C were 5.6, 8.6, 9.99 and 12.4 x 10" 1 0 m 2 s"1 respect ively. The dif fus ion coefficient obtained at 25 °C (5.6 x 10" 1 0 m 2 s"1) is m u c h l o w e r than the value at infini te d i lu t ion ( 1 . 0 6 x l 0 " 9 m 2 s"1) [330]. Th i s difference c o u l d be caused by the h i g h ion ic strength (above 3.1 M ) where the i o n - i o n interaction is s ignif icant and the k inemat ic v i scos i ty is 35 % greater than that for water, decreasing the d i f fus ion coefficient . The coefficient at 25 ° C is close to the values (6 - 7 x l 0 " 1 0 m 2 s"' i n 0.5 M N a 2 S 0 4 ) reported by Hunge r et a l . [254]. A t infinite d i lu t ion , the di f fus ion coefficient has the f o l l o w i n g temperature dependence: L o g D = constant - „ ^ (6-18) B 2 . 3 0 3 R T v ' where D is the di f fus ion coefficient, E a the diffusion act ivat ion energy, R the gas constant, T the absolute temperature. The di f fus ion act ivat ion energy calculated f rom the slope o f the l o g plot o f d i f fus ion coefficient vs . 1/T (Figure 6-18) is 18 k J mole" 1 . F igure 6-17 D i f f u s i o n current density vs. the square root o f rotational speed at different temperatures. Elec t ro ly te : 0.05 M N a 2 S 0 3 , l M N a 2 S 0 4 , 0.25 M N a O H . 99 -8.9 0.003 0.0031 0.0032 0.0033 0.0034 (T/K)-1 Figure 6-18 L o g plot o f di f fus ion coefficient vs. 1/T. 6.10 Potential Sweep Study The potential sweep method was used to study the anodic ox ida t ion o f sulphite. F igure 6-19 shows the c y c l i c vo l tammograms after subtraction o f the background current for different scan rates. There is no negative current corresponding to the reduct ion o f the o x i d i z e d products (or intermediates) and the ox ida t ion o f sulphite is therefore i r reversible . The peak current density ( i p ) is g i v e n by the f o l l o w i n g equation for the i r revers ible react ion [333]: i p = (2.99 x 1 0 > ( a ) 1 / 2 C b D , / 2 v 1 / 2 = B v 1 / 2 (6-19) where n is the number o f transferred electrons, a the rate-control l ing step charge transfer coefficient, C b the bu lk reactant concentration, D the dif fusion coefficient, v the potent ial scan rate and B = (2.99 x 1 0 5 ) n ( a ) 1 / 2 C b D 1 / 2 . The peak current is propor t ional to the square root o f the potential scan rate. The plot o f i p vs . v 1 / 2 gave a l inear relat ionship (F igure 6-20). The slope ( B) was calculated by least squares fi t t ing. The f o l l o w i n g relat ionship obtains: 100 1.857i?r Ep-Epn=—^T- (6-20) where E p is the peak potential and E p / 2 the potential w h e n i = i /2 . F r o m the above equation we obtain an apparent charge transfer coefficient o f 0.33 w h i c h is close to that (0.30-0.31) calculated us ing the Tafe l slope at h igh potentials. The total number o f the electrons transferred is 1.98, 1.98, 2.00, 1.98 by combina t ion o f B , C b , oca, D at 25 , 40 , 50 and 60 ° C respect ively. T h i s number corresponds to the s toichiometry indicated by E q u a t i o n 6-13. F igure 6-19 V o l t a m m o g r a m s at different scan rates at 25 ° C . Elec t ro ly te : 0.1 M N a j S C ^ , 1 M N a 2 S 0 4 , 0.25 M N a O H . 101 600 500 ~ 400 tfl c o 73 C 0) l_ k-3 U ro Q . 300 i -200 100 0.1 0.2 0.3 0.4 0.5 (Potential scanning rate / V s"1)1'2 0.6 Figure 6-20 Peak current vs. potential scan rate at 25 ° C . Elec t ro ly te : 0.1 M N a 2 S 0 3 , 1 M N a 2 S 0 4 and 0.25 M N a O H . 6.11 Possible reaction mechanism F r o m Figure 6-13, there are two Tafe l regions. The first one is 0.059 -0.066 V decade" 1 f rom 25 to 60 ° C at l o w potentials and the second is 0.19-0.22 V decade" 1 at h igher potentials. The corresponding charge transfer coefficients are 1 and 0.3 respect ively. These values suggest a change i n the react ion mechan i sm or i n the ra te-control l ing step. The react ion order at l o w potentials is b e l o w 1 and nonlinear . It decreases s l ight ly w i t h increas ing sulphite concentrat ion indica t ing that the adsorbed sulphite c o u l d beg in to be o x i d i z e d at l o w potentials. There are no peaks corresponding to the adsorption i n the vo l t ammograms . T h i s means that on ly a very sma l l amount o f sulphite adsorbs o n the electrode surface. Ta rasev ich et a l . [239] studied the adsorption and elect rooxidat ion o f sulphite o n p l a t i num us ing radioact ive tracers. T h e y found that S 0 3 2 " was w e a k l y adsorbed o n the surface and the amount o f adsorbed S 0 3 2 " d i d not change over the potential range -0.24 - 0. 26 V vs . S C E . It decreased to zero w i t h increasing potential f rom 0.26 to 0.56 V vs . S C E at 22 ° C . In the present study, it was found that S 0 3 2 " begins to be o x i d i z e d o n a graphite anode at 0.16 V vs . S C E at 25 ° C and the Tafe l slope was a constant value o f 0.060 V decade" 1 over the potent ial range 0.16 - 0.25 V vs. S C E . W i t h further increase i n potential , the Tafe l slope increased w i t h 102 increasing potential . W h e n the potential exceeded 0.4 V vs. S C E , the Tafe l slope remained at 0.19 V decade" 1 and was independent o f the potential . The above phenomenon can be expla ined as fo l lows : (1) at 0.16 -0.25 V vs . S C E , the adsorbed S 0 3 2 " is o x i d i z e d and the coverage o f adsorbed S 0 3 2 " is independent o f potential and therefore the Tafe l slope (0.060 V decade "') is independent o f the potential and the react ion order w i t h respect to S 0 3 2 " is b e l o w 1 and nonlinear; (2) at 0.25 - 0.4 V vs. S C E , the coverage o f adsorbed S 0 3 2 " decreases w i t h increasing potential . Therefore the Tafe l slope increases w i t h increasing potential ; (3) at potential > 0.4 V , the amount o f adsorbed S 0 3 2 " is neg l ig ib le and the direct ox ida t ion o f unadsorbed S 0 3 2 " dominates. Thus the Tafe l slope becomes independent o f the potential and the reaction order w i t h respect to S 0 3 2 " become uni ty . The react ion order w i t h respect to O H " ions is almost zero. Th i s means that the rate-control l ing steps for the two Tafe l slope regions do not invo lve O H " . There are numerous carbon ox ide surface groups o n graphite [243, 334] and sulphur c o u l d be bound to these surface groups dur ing the adsorption. In accordance w i t h the these phenomena, the f o l l o w i n g react ion mechan i sm is proposed: A t l o w potentials (< 0.25 V S C E ) , sulphite first adsorbs on the graphite, then loses the first electron, f ina l ly undergoing oxygen transfer and los ing the second electron. F o r example , Step 1 : S032~ <^> 5 0 3 2 - ( a r f . ) Step 2: SO2'(ad.) « S03\ad.) + e Step 3: S03\ad.) -> S03 Step 4: SO,1 + 20H~ -> SO2' + H20 + e Cons ide r ing the theory o f multistep electrode reactions [335, 336] , i f step 1 is rate-cont ro l l ing , the current should be independent o f potential . I f step 2 is ra te-control l ing, the Tafe l slope should be above 0.059 V decade" 1 at 25 ° C (because the charge transfer coefficient < 1). I f step 4 is the rate-control l ing step, the Tafe l slope should be around 0.040 V decade" 1 at 25 ° C and the reaction order w i t h respect to O H " ions should be 1 or more . I f step 3 is rate-control l ing, the Tafe l slope is 0.059 V decade" 1 at 25 ° C . The react ion order w i t h respect to O H " ions c o u l d be zero. L o o k i n g at the experimental results, Step 3 c o u l d be rate-con t ro l l ing . 103 A t h i g h potentials ( > 0.4 V vs. S C E ) , sulphite first loses one electron, subsequently undergoes oxygen transfer and loses the second electron. Step 1: SO2' -> S03'+e Step 2 : S03 + 20H' SO2' +H20 + e The charge transfer coefficient is on ly about 0.3, suggesting that the loss o f the first e lectron is the rate-control l ing step. Th i s is i n agreement w i t h the react ion order w i t h respect to sulphite ions. The react ion order w i t h respect to hydroxide ions is zero, suggesting that the ra te-control l ing step does not i nvo lve hydroxide ions. Therefore step 1 c o u l d be the rate-con t ro l l ing step at h i g h potentials. It should be noted that a smal l amount o f S03~ c o u l d combine to fo rm dithionate and therefore the number o f the electrons transferred is s l igh t ly b e l o w 2. 6.12 Summary A t l o w potentials ( e.g. < 0.25 V vs. S C E at 25 ° C ) , the react ion order for the ox ida t ion o f sulphite is b e l o w 1 and decreases w i t h increasing sulphite concentrat ion. The Tafe l slope is 0.059 -0.065 V decade" 1 i n the temperature range 25-60 ° C . A t h i g h potentials (> 0.4 V vs. S C E ) , the react ion order w i t h respect to sulphite ions is 1 up to 0.4 M sulphite and the Ta fe l slope is 0.19 - 0.21 V decade" 1. The react ion order w i t h respect to hyd rox ide ions is close to zero at surface p H > about 9. The act ivat ion energy for the k inet ic current decreases f rom 85.2 kJmol" 1 at 0.2 V vs. S C E to 45.3 kJmol" 1 at 0.6 V vs. S C E . The dif fus ion coefficients o f sulphite ions were obtained and shown to have an act ivat ion energy o f 18 k J moi" 1 . Sulphi te ox ida t ion i n a lkal ine solut ion appears to undergo a radical -e lec t ron mechan i sm. A t l o w potentials, the adsorbed sulphite ox ida t ion is dominant and at h i g h potentials, the sulphite ions are o x i d i z e d direct ly on the electrode surface. The loss o f the first electron f rom sulphite ions appears to be the rate-control l ing step at h i g h potentials. 104 7. ANODIC OXIDATION OF COPPER CYANIDE ON A GRAPHITE ANODE IN ALKALINE SOLUTION T o decrease the consumpt ion o f cyanide, it is important to understand the anodic ox ida t ion o f copper cyanide. H o w e v e r , the informat ion avai lable is inadequate and further studies are needed. Therefore a study o f the anodic ox ida t ion o f copper cyanide was conducted us ing the rotating disk technique. 7.1 Experimental Apparatus and Set-up The graphite rotating disk was the same as described i n Sec t ion 6.3. T o ensure reproducible results, the electrode was first treated by c y c l i c vo l tammetry between 0 -0.75 V vs. S C E i n 0.25 M N a O H and I M N a 2 S 0 4 solut ion at 100 r p m for 30 minutes and po la r i zed at 1 m V s"1 un t i l the electrode reached a stable condi t ion . Graphi te h a v i n g diameters o f 12 and 24 m m was fashioned as a rotating disk for coulometr ic measurements. A p y r o l y t i c graphite rotating disk hav ing a diameter o f 4 m m and a p la t inum electrode hav ing a diameter o f 5 m m were made by the E G & G C o . The rotating disk electrode system was an E G & G P A R C M o d e l 636 Elec t rode Rotator. The potentiostats were M o d e l S O L A R T R O N 1286 and P A R C 2 7 3 A e lec t rochemica l Interface. A r g o n gas was first bubbled through the solut ion and the experiments were conducted under an argon atmosphere to a v o i d the possible effect o f the air. The exper imental set-up was the same as shown i n Figure 6-3. The po la r iza t ion curves were generated us ing the potential sweep method at 1 m V s"1 as noted. A L E Y B O L D M A X 200 X P S instrument was used to analyze the anode precipitate. A Siemens diffractometer D 5 0 0 0 0 was used to obtain the X - r a y dif f ract ion pattern o f the precipitate. Samples o f the anode precipitate for X P S and X - r a y diffract ion were p laced i n a bottle f i l l ed w i t h A r gas to protect against ox ida t ion by air. The s topped-f low technique and spectrometry us ing a S H I M A D Z U M o d e l U V -2 4 0 I P C U V spectrometer were employed to detect the cupr ic cyanide species. N a C N and C u S 0 4 solutions were injected into a T -tube i n one second and w e l l m i x e d , finally entering the quartz c e l l for U V detection. The experimental set-up is shown i n F igure 7-1. 105 T - tube Syringe Figure 7-1 Schemat ic d iagram for detection o f cupric cyanide species The k inemat ic v i scos i ty o f the solu t ion was measured us ing a Cannon-Fenske routine v iscometer (size 25) (see A p p e n d i x 6). The l i q u i d j unc t i on potential for a dilute copper cyanide solu t ion was ca lcula ted by the Henderson equation (see A p p e n d i x 7). The l i q u i d j u n c t i o n for concentrated copper cyanide was not considered because there are no data for copper cyanide species and the mobi l i t i e s o f copper cyanide species are expected to be close to that o f sod ium ion . The thermal l i q u i d j u n c t i o n potential was measured us ing two ca lome l reference electrodes w h i c h were p laced o n the two sides o f an electrolyte bridge. The cyanide concentrat ion was measured us ing the dis t i l la t ion-absorpt ion-t i t ra t ion procedure (see A p p e n d i x 2). The copper concentrat ion was measured by o x i d i z i n g copper cyanide to cupr ic nitrate us ing concentrated ni t r ic ac id and ti trat ion w i t h E D T A (see A p p e n d i x 3). Reagent grade chemica ls were used i n a l l the experiments. 7.2 Polarization Measurements and Identification of the Precipitate 7.2.1 Anod ic Behaviour for Dilute Copper Cyan ide Solution T o develop an understanding o f the anodic ox ida t ion o f copper cyanide , the study o f the electrode kinet ics was first carr ied out i n dilute copper cyanide so lu t ion i n the presence o f an excess o f inert support ing electrolyte. A s a result, a l l the observed potential difference was 106 concentrated o n the electrode solut ion interface and avai lable for affecting the actual rate o f the electrode reaction. The polar iza t ion measurements were conducted at 25, 40 , 50 and 60 ° C i n a lka l ine solu t ion w i t h different concentrations o f cyanide, copper and sod ium hydrox ide . A s copper ox ide and hydrox ide were precipitated on the electrode surface dur ing the po la r i za t ion measurement, the electrode surface was repol ished after every po la r iza t ion to ensure reproducible results. The polar iza t ion curves for 0.05 M cyanide and a C N : C u mo le ratio o f 3 at 25, 40, 50 and 60 ° C are shown i n F igure 7-2. The anodic ox ida t ion o f copper cyanide can be d i v i d e d into three potential regions. In the first region (approximately 0 - 0.4 V vs. S C E ) , no precipitate was formed on the electrode. In the second reg ion (approximate ly 0.4 - 0.6 V vs. S C E ) , copper ox ide and hydrox ide were formed on the electrode surface and the current increased sharply w i t h increasing potential . In the th i rd reg ion (about > 0.6 V S C E ) , the o x y g e n was evo lved . The behavior o f current vs. potential was dependent o n the temperature and the rotat ional speed. A t 25 ° C (Figure 7-2a), when the rotational speed was 100 r p m , the current reached a l i m i t i n g value and d i d not decrease w i t h increasing potential . H o w e v e r , w h e n the rotat ional speed was above 100 rpm, the current reached a m a x i m u m value and then decreased w i t h increasing potential . A t 40 ° C (Figure 7-2b), w h e n the rotat ional speed was b e l o w 1600 rpm, the current d i d not decrease w i t h potential . W h e n the rotat ional speed was above 2500 rpm, the current reached a m a x i m u m value, then s tabi l ized and f ina l ly decreased w i t h increasing potential . A t 50 and 60 ° C (Figure 7-2c and d), the current d i d not decrease w i t h increasing potential . Th i s anodic ox ida t ion behaviour o f copper cyanide is related to the precipi ta t ion o f copper oxide . F r o m the c y c l i c vol tammetry (Figure 7-3), the effect o f the precipi ta t ion o f the copper ox ide was dependent o n the appl ied potential . A t 25 ° C and 100 r p m (Figure 7-3a), w h e n the potential was swept f rom 0 to 0.55 V vs. S C E and then back to 0 V vs. S C E , the current for the negat ive-going sweep was larger than that for the pos i t ive-going sweep. T h i s means that the precipitate had a catalytic effect on the anodic ox ida t ion o f copper cyanide . W h e n the potential was swept f rom 0 to 0.60 V vs. S C E and then back to 0 V vs. S C E , the current for the negat ive-going sweep was smaller than that 1 for the pos i t ive-going current. T h i s indicates that the precipitate had a passivat ing effect on the anodic ox ida t ion o f copper cyanide . The 107 change i n the catalyt ic properties o f copper ox ide m a y be caused by the adsorpt ion o f o x y g e n produced i n the electrode reaction. The c y c l i c vol tammetry at 40 ° C and 100 and 1600 r p m (Figure 7-3b) shows again the catalytic effect o f the precipitate o f copper ox ide . F r o m Figure A - 9 ( A p p e n d i x 8), the precipi tated copper ox ide has a m u c h more pronounced catalytic effect on the cyanide ox ida t ion than the graphite and copper ions i n the solut ion. F r o m the X P S analysis o f the precipitate (Figures A - 9 and A - 1 0 i n A p p e n d i x 8), the precipitate was found to be copper oxide . The curve fi t t ing o f the X P S spectrum (Figures A -11 and A - 1 2 i n A p p e n d i x 8) conf i rmed that the precipitate was a combina t ion o f copper ox ide and copper hydroxide . The contents o f C u O and C u ( O H ) 2 on the surface were respect ively about 50 % at 25 ° C and 70 % at 60 ° C . So the ratio o f C u O to C u ( O H ) 2 i n the precipitate increased w i t h increasing temperature. The precipi ta t ion o f copper oxide and hydrox ide suggests that copper cyanide can be o x i d i z e d to copper oxide and cyanate. The onset o f the precipi ta t ion o f copper ox ide depends o n the C N : C u mo le ratio and potential . A t l o w rotational speeds, the onset o f the prec ip i ta t ion o f copper ox ide appears at lower potentials than at h i g h rotational speeds leading to h igher currents. The reason c o u l d be that at the same potential , the C N : C u mo le ratio at the electrode surface for a l o w rotational speed is lower than that at the h i g h rotat ional speed, also the l o w l y coordinated copper cyanide complexes are less stable than the h i g h l y coordinated complexes and are easier to ox id i ze to copper oxide and cyanate. The onset o f the format ion o f copper ox ide occurs at higher potentials at a higher C N : C u m o l e ratio f rom the po la r iza t ion measurement. In the th i rd reg ion (potentials > about 0.6 V vs . S C E ) , a gas was evo lved , w h i c h c o u l d be o x y g e n or ni t rogen due to the further ox ida t ion o f cyanate. The current d i d not change un i fo rmly w i t h increasing rotational speed because the f i l m o f copper ox ide o n the graphite was fo rmed i r regular ly . E v e n part o f it dropped f rom the electrode. The coat ing o f C u O s igni f icant ly increases the o x y g e n evolu t ion (see F igure A - 1 4 i n A p p e n d i x 8). The current decreased w i t h increasing potential after it reached a m a x i m u m value because the ox ide f i l m became loose ly adherent o n the graphite. In fact, some o f it dropped f rom the electrode due to the o x y g e n evolu t ion . In the absence o f copper cyanide, the po la r iza t ion curves for the electrode w i t h precipi tated copper ox ide i n the case o f the solu t ion conta in ing 0.05 M cyanate 108 were almost the same as those without cyanate. T h i s suggests that the evo lu t ion o f o x y g e n was dominant . Increasing the C N : C u mole ratio (decreasing [Cu]) results i n a change i n the anodic behaviour o f copper cyanide (Figures A - 1 5 to A - 1 8 i n A p p e n d i x 8). The po la r i za t ion curves for the solutions w i t h C N : C u mole ratios o f 3.5 and 4 are s imi la r to those for the so lu t ion w i t h a C N : C u mole ratio o f 3. The difference is that the onset o f the precipi ta t ion o f copper ox ide begins at a higher potential . Howeve r , at C N : C u mole ratios o f 6 and 12, there were no w e l l defined l i m i t i n g currents because the precipi ta t ion o f copper ox ide began at about 0.6 V vs. S C E and o x y g e n is ready to be evo lved o n the copper ox ide , affecting the ox ida t ion o f copper cyanide. A t a C N : C u mole ratio o f 6 (25 to 50 °C) or 12, the precipi tated ox ide was not t ight ly adherent to the graphite. Therefore the evo lu t ion o f o x y g e n was not ca ta lyzed s ignif icant ly by the copper ox ide as observed at lower mole ratios o f cyanide . H o w e v e r , at a C N : C u mole ratio o f 6 and 60 ° C the current increased cont inual ly w i t h increasing potential because the copper ox ide was re la t ively w e l l deposited o n the electrode and ca ta lyzed s igni f icant ly the evo lu t ion o f oxygen . Decreas ing hydrox ide concentration also leads to the change i n the anodic behaviour o f copper cyanide (see F igure A - 1 9 to A - 2 4 i n A p p e n d i x 8). A t [OH"] = 0.05 M , the polar iza t ion curves for 0.05 C N " solutions w i t h C N : C u m o l e ratios o f 3, 4 and 12 are shown i n Figures A - 1 9 to A - 2 1 ( A p p e n d i x 8) respect ively . The anodic behaviour o f the copper cyanide solut ion can be d i v i d e d into the three potential regions s imi la r to those w i t h 0.25 M N a O H . H o w e v e r , the format ion o f copper ox ide and o x y g e n evo lu t ion was suppressed. A t [OH"] = 0.01 M , the polar iza t ion curves for 0.05 M C N " solutions w i t h C N : C u mole ratios o f 3, 4 and 12 are shown i n Figures A - 2 2 to A - 2 4 respect ively. The format ion o f copper ox ide and o x y g e n evolu t ion was s ignif icant ly decreased. A t C N : C u m o l e ratio = 1 2 , almost no copper ox ide was formed. C o m p a r i n g the anodic behaviour o f copper cyanide w i t h 0.25 M N a O H (Figure 7-2, 7-11 and 7-13), 0.05 M N a O H (Figures A - 1 9 to A - 2 1 ) and 0.01 M N a O H (Figures A - 2 2 to A - 2 4 ) , hydrox ide and copper concentrations affect the anodic ox ida t ion o f copper cyanide s ignif icant ly i n some potential regions and the effect o f hydroxide concentrat ion is dependent 109 o n copper concentrat ion because the anodic behaviour is related to the d is t r ibut ion o f copper cyanide species, as discussed i n Sections 7-4 and 7-5. 0.00 0.20 0.40 0.60 0.80 Potential vs. S C E / V 1.00 1.20 E < 3000 , 2500 2000 (0 c •a 1500 1000 3 o 500 0 J 0.00 4900 rpm 2 3 / 3600 rpm " . . . . 2500 rpm — . -1600 rpm . 900 rpm 400 rpm 100 rpm precipitation -jof copper oxide oxygen evolution _ / Si j \ \ / / , */ A 1 no precipitation of copper oxide 0.20 0.40 0.60 0.80 Potential vs. S C E / V 1.00 1.20 (a) 25 ° C 5000 4500 ^ 4000 ^ 3500 3000 c 2500 0) 2 2000 £ 1500 3 O 1000 500 o 4 0.00 -4900 rpm . 3600 rpm . 2500 rpm -1600 rpm _900 rpm _400 rpm . 100 rpm 0.20 0.40 0.60 0.80 1.00 Potential vs. S C E / V (b) 40 ° C 12000 10000 E < to c a •a c 2 i— 3 o 8000 4J 6000 4000 2000 1.20 0.00 0.20 0.40 0.60 0.80 Potential vs. S C E / V 1.00 1.20 (c) 50 ° C (d) 60 ° C F igure 7-2 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 0.05 M C N " , C N : C u = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 . 110 0 0.1 0.2 0.3 0.4 0.5 0.6 Potential v s . S C E / V (a) 25 ° C a n d 100 r p m 1000 -, Potentia l v s . S C E / V (b) 40 ° C , 100 r p m and 1600 r p m F igure 7-3 C y c l i c vol tammetry at 25 and 40 ° C . Elec t ro ly te : 0.05 M C N ' , C N : C u = 3, 0.25 M N a 2 S 0 4 and 1 M N a ^ C v I l l 7.2.2 Anod ic Behavior of Concentrated Copper Cyan ide Solution In order to obtain a reasonable copper deposi t ion eff ic iency, the copper cyanide concentrat ion should be cont ro l led around 1 M . Therefore the study o f the anodic ox ida t ion o f copper cyanide was also conducted for a h igh concentrat ion copper cyanide solut ion. The anodic behaviour for 0.5 M C N " (Figure A - 2 5 i n A p p e n d i x 8) was quite different f rom that for 0.05 M C N " (Figure 7-2). The current also d i d not increase as expected f rom the increase i n copper cyanide concentration. The precipi ta t ion o f copper ox ide resulted i n a s ignif icant passivat ion. O x y g e n evolu t ion was suppressed s ignif icant ly . F r o m Figure 7-4, at [CN"] = 3 M , a C N : C u mole ratio o f 3, [OH"] = 0.25 and [ N a 2 S 0 4 ] = 0.5 M , the anodic ox ida t ion o f copper cyanide can be described by the three potent ial regions. In the first region, anodic ox ida t ion proceeded wi thout the format ion o f copper ox ide . In the second region, copper oxide was precipitated, resul t ing the pass iva t ion o f the electrode. W h e n the electrode was coated w i t h copper oxide by sweeping the potential o f the electrode to 0.48 V vs. S C E (Figures 7-4b, c and d), there was no dist inct peak and the current was lower at potentials be low about 0.4 V vs. S C E . Hence the decrease i n the current was due to the format ion o f copper oxide . In the th i rd reg ion ( > about 1.0 V vs. S C E ) , several bubbles were observed. The precipi ta t ion o f copper ox ide o n the electrode increased s igni f icant ly the resistance between the graphite and the solut ion. The I R drop at 1.0 V vs . S C E was over 1.0 V , w h i c h was estimated f rom the difference between the potentials measured by a potentiostat (using current interruption technique) and a mult imeter . A c c o r d i n g to the X - r a y diffract ion patterns o f the anode precipitate p roduced at 25 and 60 ° C (Figures A - 2 6 and A - 2 7 i n A p p e n d i x 8), the precipitates were def ined as a combina t ion o f copper hydrox ide and copper ox ide . A t 25 ° C , there was no dist inct peak corresponding to C u O and so most o f the precipitate was C u ( O H ) 2 . A t 60 ° C , there were on ly sma l l peaks corresponding to the strongest peaks o f C u O , and C u ( O H ) 2 was dominant . C u ( O H ) 2 is supposed to be have less catalytic effect o n cyanide ox ida t ion . C o m p a r e d to F igure 7-2 (0.05 M C N " , a C N : C u mole ratio o f 3), the current d i d not increase, and even decreased a l though the concentrat ion o f copper cyanide increased by 59 t imes. T h i s m a y be related to the compos i t i on o f the anode precipitate. M o r e than h a l f o f the anode precipitate produced i n 0.05 M C N " solu t ion w i t h a C N : C u mole ratio o f 3 was C u O , w h i c h had a g o o d 112 catalyt ic effect o n the ox ida t ion o f copper cyanide. H o w e v e r , f rom the X - r a y dif f ract ion patterns, we can predict that the amount o f C u O i n the precipitate produced i n 3 M C N " solut ion w i t h a C N : C u mole ratio o f 3 is very smal l and the precipitate exhib i ted a poor catalytic effect o n the ox ida t ion o f copper cyanide. It is also possible that the concentrat ion o f copper cyanide is so h i g h that it po isoned the catalytic properties o f copper ox ide and suppressed the evo lu t ion o f o x y g e n s ignif icant ly poss ib ly due to the s ignif icant adsorpt ion o f copper cyanide species. W h e n the cyanide concentrat ion was increased f rom 3 to 3.5 M , the anodic behavour o f copper cyanide became different (Figure 7-5). A t 25 ° C , the copper ox ide was precipi ta ted at a l l rotat ional speeds. H o w e v e r , at 50 and 60 ° C , there was a litt le copper ox ide (a spira l b lack l ine) formed o n the electrode on ly at 1600 and 4900 rpm. There was no copper ox ide at 100 and 400 rpm. The format ion o f copper oxide resulted i n an increase or decrease or auto-osc i l l a t ion i n the current due to the change i n the cond i t ion o f the electrode (passivat ion and act ivat ion poss ib ly related to the format ion and d issolu t ion o f copper oxide) w i t h increas ing potential . A t 60 ° C and potentials > 0.45 and 0.54 V vs. S C E respect ively for 100 r p m and 400 rpm, a s ignif icant amount o f bubbles was evo lved . These gas bubbles immedia te ly d i s so lved w h e n the current was turned off. The gas was thought to be cyanogen because the graphite was not corroded and oxygen was not readi ly evo lved . The current became so h i g h that the mass-transfer rate o f hydroxide was lower than the rate o f cyanogen generation and cyanogen gas was formed. W h e n the cyanide concentrat ion was increased to 4 M (Figure 7-6), no anode precipitate was formed on the electrode. A t 40 ° C (100 and 400 rpm), 50 and 60 °C (100 -1600 rpm), w h e n the potential exceeded a certain value ( shown i n F igure 7-6), large bubbles (1-4 m m diameter) were rapid ly evo lved , result ing i n a sharp increase i n the current. La rge bubbles formed and broke d o w n result ing i n the irregular change i n the current w i t h potential . The bubbles d i sso lved rap id ly after the current was turned o f f and the graphite was not ser iously corroded and so the gas was be l ieved to be cyanogen. The rapid evo lu t ion o f large bubbles s ignif icant ly changed the mass transfer o n the rotating disk. Thus the current changed i r regular ly w i t h increasing rotational speed. A t a h i g h rotat ional speed, the bubbles evo lved o n the electrode were removed rapid ly , hav ing less chance to combine and f o r m 113 large bubbles. The h i g h rotational speed also increases the mass transfer o f hydrox ide to the electrode and reduces the format ion o f cyanogen. The format ion o f copper oxide is related to the p H o f the solut ion. Therefore a decrease i n hydrox ide concentrat ion should affect the anodic ox ida t ion o f copper cyanide . F igure 7-7 shows the polar iza t ion curves for the solu t ion w i t h 3 M C N " , a C N : C u m o l e ratio o f 3, 0.05 M N a O H and 0.5 M N a j S Q , . A t 25 ° C , a th in film o f copper ox ide was precipitated, result ing i n changes i n the current w i t h the potential . C o m p a r e d to the anodic behaviour o f copper cyanide i n the solu t ion w i t h 0.25 M N a O H (Figure 7-4), the current was m u c h smaller and m u c h less copper oxide was formed on the electrode. A t 40 ° C , the current vs . potential for 1600 and 4900 r p m was s t i l l s imi la r to that at 25 ° C . H o w e v e r , at 100 and 400 rpm, the current increased cont inuously to a m a x i m u m value and then decreased s l igh t ly . A t 50 ° C and 100 rpm, some gas was evo lved at 0.38 V vs. S C E and there was a lmost no copper ox ide formed o n the electrode. A t 400-4900 rpm, a very sma l l amount o f copper ox ide was precipi tated o n the anode. A t 60 ° C , the gas bubbles were observed at potentials > 0.29 and 0.32 V vs . S C E for 100 and 400 r p m and no copper ox ide was formed. F igure 7-8 shows the polar iza t ion curves for the solu t ion w i t h 4 M C N " , 1 M C u + , 0.05 M O H " and 0.5 M N a 2 S 0 4 . Compared to the anodic behaviour o f copper cyanide at 0.25 M O H " (Figure 7-6), the evolu t ion o f cyanogen began at a re la t ively l ower potential , l ead ing to the difference i n the current. The potential for the rapid evo lu t ion o f large cyanogen bubbles increased w i t h increasing rotational speed because the h i g h rotat ional speed increased the hydrox ide mass transfer and so suppressed the format ion o f large cyanogen bubbles. A t 100 or 400 rpm, the bubbles o f cyanogen were not r emoved eff ic ient ly , resul t ing i n the osc i l l a t ion o f the current. The rapid evolu t ion o f large bubbles s igni f icant ly affected the mass transfer. Therefore the current for a l o w rotational speed was even larger than that for the higher rotational speed i n some potential region. The increase i n the concentration o f hydrox ide should suppress the format ion o f cyanogen and promote the format ion o f copper oxide . W h e n the concentrat ion o f hyd rox ide increased to 0.5 M and the concentrations o f cyanide, copper and sod ium sulphate were kept at 4 M , I M and 0.5 M respectively, the anodic behaviour o f copper cyanide (F igure 7-9) became quite different f rom that for the solutions w i t h 0.25 and 0.05 M N a O H (Figures 7-6 and 8). A th in film o f copper oxide was precipitated o n the anode at the potential > a cer tain 114 value ( shown i n Figures 7-28 a and b). The evolu t ion o f massive gas bubbles considered to be cyanogen was on ly observed at 100 and 400 r p m for 50 ° C , and 100 to 1600 r p m for 60 ° C . 300 250 E < in c O) 73 C a> i— 3 o 150 100 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V 4900 rpm 1600 rpm 400 rpm 100 rpm 4900 rpm-CuO 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V 1.2 (a) 25 ° C (b) 40 ° C 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V 0.4 0.6 0.8 1.0 Potential vs. SCE / V (c) 50 ° C (d) 60 ° C F igure 7-4 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 3 M C N " , C N : C u mo le ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 . K e y s : 1 - no prec ip i ta t ion o f copper ox ide , 2 - precipi ta t ion o f copper oxide and 3 - evo lu t ion o f oxygen . 115 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 4900 rpm 1600 rpm 400 rpm . . 100 rpm Precipitation ^ -of copper oxide / ' \ ^ ' / r"\ 1-\ \ \ / A 1 \ \ \ / / y i A V ^ "Vr-l 1 ^ — '• '• [' —-''/ y r " ! 'An ~~~~ \ '•'\ \ \ . V - ' ' - - ' " D 0.2 0.4 0.6 3.8 1 1 Potential vs . S C E / V (a) 25 ° C 12000 10000 E < c O TJ 4-* C o 8000 4-6000 4000 2000 0.4 0.6 0.8 1.0 Potential v s . S C E / V (b) 40 ° C V) c 0) 73 30000 25000 20000 15000 10000 5000 4900 rpm 1600 rpm 400 rpm 100 rpm Precipitation of copper oxide • "4 Evolution of cyanogen bubbles 0.4 0.6 0.8 1.0 Potential vs . S C E / V 0.0 0.2 0.4 0.6 0.8 1.0 Potential vs . S C E / V 1.2 (c) 50 ° C (d) 60 ° C F igure 7-5 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 3.5 M C N " , C N : C u mole ratio = 3.5, 0.25 M N a O H and 0.5 M N a 2 S 0 4 . 116 14000 12000 --CN E 10000 --< 8000 (A C u •o 6000 --3 4000 o 2000 0 -4900 rpm .1600 rpm . 400 rpm .100 rpm 30000 25000 0.2 0.4 0.6 0.8 1.0 Potential vs . S C E / V 1.2 CN E < 20000 -•— in £ 15000 . •a £ 10000 -3 o 5000 4900 rpm r ' 1600 rpm 400 rpm . . 100 rpm Rapid evolution / A / V \ / /» / \ - of large bubbles / A V ^ ./// ri / ' - Affected by bubbles 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential v s . S C E / V (a) 25 ° C (b) 40 ° C 30000 25000 CN E < 20000 in den 15000 --den £ 0) 10000 --i_ 3 o 5000 .4900 rpm .1600 rpm . 400 rpm . 100 rpm of big bubbles - ' Affected by bubbes 0.2 0.4 0.6 0.8 1.0 Potential v s . S C E / V 1.2 45000 40000 35000 30000 25000 20000 4-15000 10000 5000 0 -4900 rpm . 1600 rpm . 400 rpm -100 rpm Rapid evolution of large bubbles ; / Affected by bubbles 0.0 0.2 0.4 0.6 0.8 Potential vs . S C E / V 1.0 (c) 50 ° C (d) 60 ° C F igure 7-6 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 4 M C N " , C N : C u mo le ratio = 4, 0.25 M N a O H and 0.5 M N a 2 S 0 4 . 117 120 N 1 0 0 E < 80 in § 60 3 o 40 20 -4900 rpm . 1600 rpm . 400 rpm 0.0 1400 0.4 0.6 0.8 1.0 Potential vs . S C E / V 0 0.2 0.4 0.6 0.8 1.0 Potential vs . S C E / V 1.2 (a) 25 ° C (b) 40 ° C •6000 5000 fM E < 4000 in c 3000 0) •a £ 2000 3 o 1000 4900 rpm 1600 rpm 400 rpm 100 rpm Precipitation of copper oxide / V 0.0 0.4 0.6 0.8 Potential vs . S C E / V 9000 8000 <N 7000 E < 6000 ••^  5000 in c •g 4000 O 3000 i-O 2000 1000 4-4900 rpm 1600 rpm 400 rpm _. _ 100 rpm 0.0 0.4 0.6 0.8 1.i Potential vs . S C E / V 1.2 (c) 50 ° C (d) 60 ° C Figure 7-7 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 3 M C N \ C N : C u mo le ratio = 3, 0.05 M N a O H and 0.5 M N a 2 S 0 4 118 (a) 25 °C (b) 40 °C 45000 40000 N 35000 E < 30000 & 25000 .g 20000 S 15000 rj 10000 5000 0 _4900 mm __ 1600 rpm ... 400 rpm ..100 rpm Rapid evolution of large bubbles affected by bubbles 0.2 0.4 0.6 Potential vs . S C E / V 0.8 60000 50000 CM E < 40000 c 30000 § 20000 3 o 10000 0 J 4900 rpm 1600 rpm 400 rpm 100 rpm Rapid evolution of large bubbles V v Affecte'cTby bubbles 0.2 0.4 0.6 0.8 Potential vs . S C E / V (c) 50 °C (d) 60 °C Figure 7-8 Polarization curves at different rotational speeds and temperatures. Electrolyte: 4 M CN", CN:Cu mole ratio = 1, 0.05 M NaOH and 0.5 M Na 2 S0 4 119 (a) 25 ° C (b) 40 ° C 35000 30000 25000 20000 a; 15000 TJ *~> C £ 10000 k-3 o 5000 0 4900 rpm 1600 rpm 400 rpm 100 rpm Precipitatio of, jcopper oxide 0.0 0.2 0.4 0.6 0.8 1.0 Potential vs . S C E / V 30000 25000 CM E < 20000 c 15000 0) TJ S 10000 3 o 5000 4900 rpm 1600 rpm 400 rpm _.-100 rpm Precipitati of copper oxide / ^ •A/ Rapid evolution of xyanogen bubbles 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential vs. SCE / V (c) 50 ° C (d) 60 ° C Figure 7-9 Po la r i za t ion curves at different rotational speeds and temperatures. E lec t ro ly te : 4 M C N " , C N : C u mo le ratio = 1, 0.50 M N a O H and 0.5 M N a 2 S 0 4 120 7.3 Coulometric Measurement F r o m the literature rev iew, it appears that the s toichiometry for the anodic ox ida t ion o f copper cyanide has not been carefully studied and the results are incomple te and conf l i c t ing . In this study, cont ro l led potential coulometry was used to determine the s toichiometry o f the anodic ox ida t ion o f copper cyanide per Faraday. The anode potent ial was cont ro l led at values to m i n i m i z e the rates o f side reactions such as o x y g e n evo lu t ion . The w o r k i n g electrode (anode) was separated f rom the counter electrode (the cathode) to m i n i m i z e the effect o f the change i n the C N : C u mole ratio due to copper depos i t ion at the cathode. The v o l u m e o f the catholyte was on ly about 1-2 c m 3 and its in i t i a l concentrat ion o f N a O H was 10 t imes that o f the anolyte. The evo lu t ion o f hydrogen caused a h i g h concentrat ion o f hydrox ide w h i c h can be transported to the anode compartment to ma in ta in the concentrat ion o f hydroxide i n the anolyte whose p H was moni tored . The results are g i v e n i n Table 7-1. Tests 1-4 show the amount o f o x i d i z e d cyanide and copper (I) per Faraday at 0.05 M C N " , a C N : C u mole ratio = 3, 0.25 M N a O H and 1 M N a ^ O , . I f the react ion o f the anodic ox ida t ion proceeds accord ing to the f o l l o w i n g react ion: the amounts o f cyanide and Cu(I) i on o x i d i z e d per Faraday are 0.429 and 0.143 mo le respect ively. The corresponding values o f these tests are close to the values indica ted by Reac t i on 7-1. Therefore the anodic reaction o f copper cyanide under these condi t ions can be expressed approximate ly by Reac t ion 7-1. The current eff iciencies for this react ion were found to be 99.8, 102, 103, 105 % respect ively at 25, 40 , 50 and 60 ° C . Tests 5 - 9 s h o w the amount o f o x i d i z e d cyanide to copper per Faraday at 0.05 M C N " , C N : C u m o l e ratio = 4, 0.25 M N a O H and 1 M N a 2 S 0 4 . Tests 5 - 8 were conducted at the potential where the current reached the l i m i t i n g current at 100 rpm. I f the react ion for the anodic ox ida t ion proceeds as fo l l ows : the amounts o f o x i d i z e d cyanide and copper (I) ions per Faraday electr ic i ty are 0.444 and 0.111 mo le respect ively. The amount o f cyanide o x i d i z e d per Faraday is a l i t t le h igher than 0.444 and the amount o f copper (I) o x i d i z e d is a lit t le lower than 0.111 mo le . Therefore the anodic react ion o f copper cyanide under these condi t ions can be expressed approximate ly b y C u ( C N ) 3 2 " + 8 0 H " = 3 C N O " + C u ( O H ) 2 + 3 H 2 0 + 7e (7-1) C u ( C N ) 4 3 _ + 1 0 O H - = 4 C N O " + C u ( O H ) 2 + 4 H 2 0 + 9e (7-2) 121 Reac t ion 7-2. The current efficiencies for this react ion were found to be 99.5, 100, 103, 105 % respect ively at 25, 40, 50 and 60 °C. W h e n the potential was cont ro l led at a value where no copper ox ide was formed (Test 9), the amount o f cyanide o x i d i z e d per Faraday is 0.505 m o i F" 1 . The react ion can be expressed as fo l lows : C N " + 2 0 F T = C N O " + H 2 0 + 2e (7-3) Tests 10-13 show the amount o f cyanide to copper o x i d i z e d per Faraday at 0.05 M C N " , C N : C u mo le ratio = 12, 0.25 M N a O H and 1 M N a 2 S 0 4 . The amount o f cyanide o x i d i z e d was very close 0.5 m o i F" 1 , i.e. cyanide is o x i d i z e d to cyanate. W h e n the concentrat ion o f hydrox ide was 0. 01 M N a O H , the amount o f cyanide o x i d i z e d was s t i l l c lose to 0.5 m o i F" 1 (Tests 14 -17 ) . Table 7-1 A m o u n t o f cyanide and copper (I) o x i d i z e d per Faraday at 100 r p m and different C N : C u mo le ratios and hydroxide concentrations N o . C o m p o s i t i o n o f Temperature Potent ial cyanide copper o x i d i z e d solu t ion / ° C vs. S C E / V o x i d i z e d / m o i F" 1 / m o i F" 1 1 0.05 M C N " , 25 0.5 0.435 0.126 2 0.01667 M C u + 40 0.5 0.443 0.136 3 C N : C u = 3 50 0.48 0.445 0.139 4 0.25 M N a O H 60 0.46 0.439 0.159 5 0.05 M C N " , 25 0.5 0.447 0.102 6 0.0125 M C u + 40 0.5 0.449 0.106 7 C N : C u = 4 50 0.48 0.467 0.110 8 0.25 M N a O H 60 0.46 0.470 0.110 9 25 0.3 0.505 0 10 0.05 M C N " , 25 0.45 0.508 0 11 0.0125 M C u + 40 0.45 0.509 0 12 C N : C u = 12 50 0.45 0.510 0 13 0.25 M N a O H 60 0.45 0.512 0 14 0.05 M C N " , 25 0.6 0.510 0 15 0.0125 M O T 40 0.6 0.511 0 16 C N : C u = 12 50 0.6 0.512 0 17 0.01 M N a O H 60 0.6 0.515 0 Table 7-2 lists the coulometr ic results for the solu t ion w i t h h i g h concentrations o f copper cyanide us ing the contro l led current method (400 A m" 2). F o r the so lu t ion w i t h 3 M C N a n d 1 M C u + , at [OH" ] = 0.25 M , the anodic current ef f ic iency for cuprous ions is a lmost that expected f rom Reac t ion 7-1 and the cyanide current ef f ic iency is s l igh t ly l o w e r than that 122 expected f rom Reac t ion 7-1, poss ib ly due to the evo lu t ion o f oxygen . A t [OH"] = 0.1 M , the current eff ic iency for cuprous ions became m u c h lower and the current ef f ic iency for cyanide increased to about 100 % . F o r the solut ion w i t h 4 M C N " and I M C u + , the cyanide current eff iciencies were almost 100 % and the current eff iciencies for cuprous i o n were zero. Therefore the s toichiometry o f the anodic ox ida t ion o f copper cyanide is dependent o n the so lu t ion compos i t ion , temperature and potential . Tab le 7-2 A m o u n t o f cyanide and copper (I) o x i d i z e d per Faraday at 400 A m" 2 , 100 r p m different C N : C u mole ratios and hydroxide concentrations N o . C o m p o s i t i o n o f Temperature / ° C Cyan ide o x i d i z e d Copper o x i d i z e d solut ion / m o i F" 1 / m o i F" 1 1 3 M C N " , 1 M C u + 50 0.412 0.135 C N : C u = 3 2 0.25 M N a O H 60 0.408 0.138 3 3 M C N " , 1 M C u + 50 0.498 0.034 C N : C u = 3 4 0.10 M N a O H 60 0.501 0.037 5 4 M C N " , 1 M C u + 50 0.492 0 C N : C u = 4 6 0.25 M N a O H 60 0.496 0 7 4 M C N " , 1 M C u + 50 0.498 0 C N : C u = 4 8 0.10 M N a O H 60 0.503 0 7.4 Effect of CN.Cu Mole Ratio The po la r iza t ion curves for the anodic ox ida t ion o f copper cyanide w i t h different C N : C u mo le ratios and a constant cyanide concentrat ion (0.05 M ) are g i v e n i n F igu re 7-10. T h e y show that copper has a significant catalytic effect o n cyanide ox ida t ion . A t a C N : C u mole ratio o f 3, the anodic ox ida t ion o f copper cyanide began at 0.090, 0.045, 0.016 and 0.00 V vs. S C E respect ively for 25, 40, 50 and 60 ° C . A t a C N : C u mo le ratio > 4, the anodic ox ida t ion o f copper cyanide began at 0.170, 0.145, 0.115, 0.085 respect ively for 25 , 40 , 50 and 60 ° C . The lower the mole ratio o f cyanide to copper, the l ower the potential for the onset o f the format ion o f copper oxide . W h e n the C N : C u mole ratio exceeded 6, no w e l l -123 defined l i m i t i n g current was obtained because oxygen was evo lved before the current reached a l i m i t i n g value. The plot o f potential vs. l og (current density) for 25 and 60 ° C is shown i n F igure 7-11 and that for 40 and 50 ° C i n F igure A - 2 8 ( A p p e n d i x 8). The cond i t i on o f the surface o f the graphite electrode var ied after the e lec t rochemical cond i t ion ing due to the inherent surface var iab i l i ty . The data for F igure 7-10 were generated w i t h some var ia t ion i n the surface cond i t ion because every measurement was conducted o n a renewed electrode surface. H o w e v e r , the data for F igure 7-11 were generated o n the same electrode surface by l i m i t i n g the potential w e l l b e l o w the value at w h i c h copper ox ide began to precipitate. Therefore the data for F igure 7-11 cannot compared direct ly to those i n F igure 7-10. T h i s explanat ion w i l l also be appl ied i n the next paragraphs. A l t h o u g h there is no correct ion for the concentrat ion difference between the bu lk and the surface solut ion, at l o w potentials, the current was m u c h lower than 10 % o f the l i m i t i n g current. Thus the concentrat ion difference can be neglected. W h e n the concentrat ion difference became significant , the format ion o f copper ox ide began and the current increased sharply. W h e n copper ox ide was precipitated on the anode, even at a constant potential , the current kept increasing. A t a C N : C u mole ratio o f 3, the Tafe l slope was about 0.12 V decade" 1. A t a C N : C u mole ratio > 4, two Tafe l slope ranges appear w i t h the first Ta fe l slope be ing about 0.060 V decade" 1 and the second one about 0.17-0.20 V decade" 1. F r o m F igure 7-12, there is o n l y one wel l -def ined Tafe l slope o n a py ro ly t i c graphite electrode at a C N : C u mo le ratio > 4 and the current at a C N : C u mole ratio o f 3 was larger than those at a C N : C u mo le ratio > 4. Therefore the anodic behaviour o f copper cyanide is dependent o n the anode materials. The increase f rom 3 to over 4 i n the C N : C u mole ratio resulted i n a s ignif icant change i n the potential vs . l o g (current) curves (Figures 7-11 and 7-12). T h i s can be due to a change i n the discharged species. A t a constant potential and cyanide concentrat ion, the current at a C N : C u mo le ratio o f 6 ( lower copper concentration) is larger than that at a C N : C u m o l e ratio o f 4 (higher copper concentration) and the current at a C N : C u mo le ratio o f 4 is larger than that at a C N : C u mo le ratio o f 12. T h i s phenomenon should be due to the change i n the concentrat ion o f the discharged species as conf i rmed i n Sec t ion 7.6. W h e n the concentrat ion o f copper was fixed at 0.00833 M and the cyanide concentrations were fixed at 0.025, 0.05, 0.1, 0.2 and 0.4 M (the corresponding C N : C u m o l e 124 ratios were 3, 6, 12, 24 and 48 respect ively) , the anodic behaviour o f copper cyanide changed w i t h cyanide concentrat ion (Figure 7-13 for 25 and 60 ° C and F igure A - 2 9 ( A p p e n d i x 8 ) for 40 and 50 ° C ) . A t [CN"] = 0.025 M (a C N : C u mo le ratio o f 3), the plots o f potential vs . l o g (current density) are linear. A t [CN"] > 0.05 M (a C N : C u mo le ratio > 6), s imi la r to F igure 7-11, there are two Tafe l slopes i n the plots o f potential vs . l o g (current density). The curves for potential vs . l og (current density) are para l le l to each other, but do not shift un i fo rmly w i t h increasing concentration o f cyanide. Therefore probably the discharged species is not free cyanide ions, but one copper cyanide species. F r o m the plots o f potent ial vs. l og (current density) on a pyro ly t i c graphite electrode (Figure 7-14), the Tafe l slope for 0.025 M C N " (a C N : C u mole ratio o f 3) is a lit t le different f rom those for higher cyanide concentrations. H o w e v e r , at a C N : C u mole ratio > 6, the Tafe l slopes are the same and the curves are para l le l to each other and shift non-un i fo rmly w i t h increasing cyanide concentration. F r o m Figure 7-15, the increase i n cyanide concentrat ion f rom 3 to 4 M results i n a s ignif icant increase i n the current density. A t [ C N ] = 3 (i.e. a C N : C u mole ratio o f 3), the current is m u c h lower than those for 3.5 and 4 M C N " due to the pass iva t ion effect o f the precipi ta t ion o f copper ox ide at a lower potential (0.2 V vs. S C E ) . F r o m the plots o f potential vs . l og (current density) (Figure 7-16), at [CN"] = 3 M , there is a Tafe l slope o f 0.10 V decade" 1. H o w e v e r , at [CN"] = 3.5 and 4 M , there are two Tafe l slopes,- the first one be ing 0.66 V decade ( R T / F ) and the second 0.16 V decade" 1. The second Tafe l slope appears to increase s l igh t ly w i t h increasing potential probably due to the concentrat ion change o f cyanide o n the surface. The results at 25, 40 and 50 ° C are s imi lar . 125 1200 0.2 0.4 0.6 0.8 Potential vs. SCE/ V < 2000 1800 1600 1400 1200 g 1000 2 800 3 o 400 200 0 _*_CN:Cu = 3 - D - C N : C U = 4 -4_CN:Cu = 6 -*_CN:Cu = 12 _«_NoCu 0.2 0.4 0.6 Potential vs. SCE / V (a) 25 °C (b) 40 °C 3000 2500 --E < 2000 --'in c a> 1500 --T3 c o k_ 1_ 1000 3 o 500 --0 --3500 3000 CN E 2500 < 2000 in c 0) TJ 1500 3 o 1000 a. 500 0 _»_CN:Cu = 3 _o-CN:Cu = 4 _*_CN:Cu = 6 _x_CN:Cu = 12 -as— no Cu 0.2 0.4 0.6 Potential vs. SCE/V 0.2 0.4 0.! Potential vs. SCE/V (c) 50 °C (d) 60 °C Figure 7-10 Effect of the mole ratio of cyanide to copper on cyanide oxidation -current vs. potential on a graphite rotating disk at 4900 rpm and different temperatures. Electrolytes: 0.05 M CN", CN:Cu mole ratio = 3, 4, 6, 12 and no copper, 0.25 M NaOH and 1 M Na 2 S0 4 . 126 Figure 7-11 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - potent ial vs . l o g (current density) o n a graphite rotating disk at 4900 r p m (25 and 60 ° C ) . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 3, 4, 6, 12 and no copper, 0.25 M N a O H and 1 M N a 2 S 0 4 . 0.10 1 1 1 1 1 -1.5 -0.5 0.5 1.5 L o g (current dens i ty / A m"2) Figure 7-12 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - potent ial vs . l o g current density o n a pyro ly t i c graphite rotating electrode at 4900 r p m and 25 ° C . Elec t ro ly tes : 0.05 M C N " , C N : C u mole ratio = 3, 4, 6, 12 and 0.25 M N a O H and 1 M N a 2 S 0 4 . 127 Figure 7-13 Effect o f the mole ratio o f cyanide to copper on cyanide ox ida t ion - potent ial vs . l og (current density) on a graphite rotating disk at 4900 r p m (25 and 60 ° C ) . E lec t ro ly tes : [Cu + ] = 0.00833, [CN"] = 0.025, 0.05, 0.1, 0.2 and 0.4 M , 0.25 M N a O H and 1 M N a 2 S 0 4 . 0.6 0.5 0.4 LU O </) oi > § 0.3 c o o °" 0.2 0.025 M C N ~ - Q _ 0.050 M C N ~ -±_ 0.100 M C N ~ - e - 0.200 M C N ~ -•- 0.400 M C N ~ 0.1 -1 -0.! L o g (current dens i ty / A m"z) 5 0 0.5 1 1.5 -2v Figure 7-14 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - potent ial vs . l o g (current density) o n a py ro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . Elec t ro ly tes : [Cu + ] = 0.00833 M , [CN"] = 0.025, 0.05, 0.1, 0.2 and 0.4 M , 0.25 M N a O H and 1 M N a 2 S 0 4 . 128 40000 Potential vs. S C E / V Figure 7-15 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - current vs . potential o n a graphite rotating disk at 4900 r p m and 60 ° C . Elec t ro ly tes : 1 M C u + , [CN"] = 3, 3.5 and 4 M , 0.25 M N a O H and 0.5 M N a ^ . 0.4 -, -0.1 I | | | | I 0 1 2 3 4 5 Log (current density / A m"2) Figure 7-16 Effect o f the mole ratio o f cyanide to copper o n cyanide ox ida t ion - potent ial vs . l og (current density) on a graphite rotating disk at 4900 r p m and 60 ° C . Elec t ro ly tes : 1 M C u + , [CN"] = 3, 3.5 and 4 M , 0.25 M N a O H and 0.5 M N a ^ 129 7.5 Effect of pH A t a C N : C u mo le ratio o f 3, and [ C N ' ] = 0.05 M , the concentrat ion o f hydrox ide s ignif icant ly affects the anodic behaviour o f copper cyanide (Figure 7-17). In the l o w pola r iza t ion region, w i t h decreasing concentration o f hydrox ide , the Tafe l slope decreases f rom 0.130 to 0.060 V decade" 1 and the current decreases at a constant potential (Figure 7-18 for 25 and 60 ° C and F igure A - 3 0 for 40 and 50 ° C ( A p p e n d i x 8)). T h i s suggests that the ra te-control l ing step changes or the mechan i sm changes. In the h i g h po la r iza t ion reg ion , copper cyanide is o x i d i z e d to copper ox ide and cyanate. The current is sensit ive to the hydrox ide concentrat ion and does not reach a wel l -def ined l i m i t i n g value at l o w hydrox ide concentrat ion. T h i s means that hydroxide ions are i n v o l v e d i n the ra te-control l ing step. The results obtained o n pyrographite (Figure 7-19) and Pt rotating disks (F igure A - 3 0 i n A p p e n d i x 8) are s imi la r . A t a C N : C u mole ratio o f 4, [CN"] = 0.05 M , the effect o f p H o n the anodic ox ida t ion depends o n the appl ied potential (Figure 7-20, and F igure 7-21 and F igure A - 3 2 ( A p p e n d i x 8)). In the l o w polar iza t ion region, p H has l i t t le effect o n the anodic ox ida t ion o f cyanide . The Tafe l slope was independent o f p H and the current decreased s l igh t ly w i t h decreasing p H . S i m i l a r results were obtained o n py ro ly t i c graphite (Figure 7-22). T h i s means that hydrox ide is not i n v o l v e d i n the rate-control l ing step. In the h i g h polar iza t ion region (> about 0.5-0.6 V vs. S C E ) , copper cyanide was o x i d i z e d to copper ox ide and cyanate w i t h the current depending greatly o n the hydrox ide concentrat ion. Genera l ly , the current decreases w i t h decreasing hydrox ide concentrat ion. A t 25 ° C and a potential > 0.65 V vs. S C E , the current for 0.25 M N a O H was b e l o w that for 0.05 M N a O H due to pass ivat ion (possibly the adsorption o f the oxygen) . The o x y g e n evo lu t ion and the format ion o f copper oxide decreased s ignif icant ly w i t h decreasing concentrat ion o f hydrox ide . Therefore the current is dependent o n the concentrat ion o f hydrox ide and hydrox ide is i n v o l v e d i n the rate-control l ing step. F r o m Figures 7-23, 7-24 and F igure A - 3 3 , the effect o f p H at a C N : C u m o l e ratio = 12 is s imi la r to that at C N : C u mole ratio = 4. In the l o w polar iza t ion reg ion (< about 0.5 V S C E ) , the current was s l ight ly affected by p H and the Tafe l slope was independent o f p H . The results obtained o n a py ro ly t i c graphite electrode (Figure 7-25) also show that p H has 130 almost no effect o n the anodic ox ida t ion o f copper cyanide at a potential < 0.6 V vs . S C E . A t a potential > about 0.5 V vs. S C E (Figure 7-23), p H affected the current. The difference between the currents for 0.25 and 0.05 M O H " is re la t ively smal l and the difference between the currents for 0.25 (or 0.05 ) and 0.01 M O H " is significant. A t 25 and 40 ° C , the current for 0.25 M O H " was even lower than that for 0.05 M O H " i n one potential r eg ion poss ib ly because the evo lu t ion o f oxygen d imin i shed the ox ida t ion o f copper cyanide. A t a C N : C u mo le ratio o f 3 and [CN"] = 3 M , the effect o f the hyd rox ide concentrat ion was dependent o n the temperature (Figure 7-26). A t 25 ° C , the current decreased w i t h decreasing concentrat ion o f hydrox ide and the anodic ox ida t ion o f copper cyanide was affected by the precipi ta t ion o f copper ox ide o n the electrode. A t the temperature > 40 ° C , i n the in i t i a l potential reg ion the current decreased w i t h increasing concentrat ion o f hydroxide . In the higher potential region, the currents for 0.50 M O H " was larger than that for 0.25 M O H " . H o w e v e r , i n some potential regions, the current for 0.05 M O H " was larger than that for 0.25 M O H " or even 0.50 M O H " . T h i s phenomenon is probably related to the fact that the amount o f the precipitated copper ox ide for 0.05 M O H decreased s ignif icant ly w i t h increasing temperature, result ing i n the less pass ivat ion o f the electrode. A t 60 ° C , there was almost no precipitate o n the electrode at 0.05 M O H " , but at 0.25 or 0.50 M O H " , a th ick copper oxide f i l m was formed, leading to the difference i n the anodic behaviour. F r o m Figure 7-27, at 0.25 and 0.50 M O H " 1 , the Tafe l slope was about 0.10 V decade" 1. A t 0.05 M O H " , there were two Tafe l slopes, the first be ing about 0.66 V decade" 1 and the second one 0.11 V decade' 1 . The change i n p H c o u l d result i n a change i n the discharged species or the rate-determining step. A t a C N : C u mo le ratio o f 4 and [CN"] = 4 (Figure 7-28), i n the l ower po la r i za t ion reg ion (< 0.50 V vs. S C E ) , the current was s l ight ly affected by the change i n the concentrat ion o f hydroxide . A t potentials > about 0.5 V vs. S C E , the concentrat ion o f hydrox ide s igni f icant ly affected the behaviour o f the anodic ox ida t ion o f copper cyanide . A t [OH"] = 0.5 M , w h e n the current increased to a certain value, the mo le ratio o f cyanide to copper o n the surface became l o w , but the hydrox ide concentrat ion o n the surface was s t i l l h i g h and reacted w i t h cupr ic ions to fo rm copper cyanide, resul t ing i n pass iva t ion o f the anodic ox ida t ion o f cyanide. A t [OH"] = 0.05 M , w h e n the current became so h i g h that 131 the pH on the surface was low and (CN) 2 gas was formed. At [OH"] = 0.25 M , the anodic behaviour of copper cyanide is between those at [OH] = 0.5 and 0.05 M . From the plot of potential vs. log (current density) (Figure 7-29) in the low polarization region, the current decreases slightly with decreasing hydroxide concentration. From the above discussion, we can see that the anodic behaviour of copper cyanide is a function of the total cyanide concentration, the mole ratio of cyanide to copper, hydroxide concentration and temperature. 700 600 ^ 500 < & 400 '35 c a> •O 300 c a> t 200 3 o 100 0 1600 1400 CN E 1200 < 1000 800 in £ •a C 600 i— = 400 200 0 - 0.25 M OH" -0.05 MOH" .0.01 MOH" 0.1 0.2 0.3 0.4 0.5 Potential vs . S C E / V 0.1 0.2 0.3 0.4 0.5 Potential vs . S C E / V (a) 25 °C (b) 40 °C 4500 T 4000 --CN E 3500 < 3000 --nsity 2500 O) TJ 2000 --rrent 1500 --Cm 1000 500 .0.25 M OH .0.05 MOH" .0.01 MOH" 3500 3000 CM E 2500 < jfr 2000 '35 c .g 1500 + J c 8j 1000 500 .0.25 M OH .0.05 M OH" .0.01 M OH" 0.1 0.2 0.3 0.4 Potential vs . S C E / V 0.6 0.1 0.2 0.3 0.4 0.5 Potential vs . S C E / V 0.6 (c) 50 °C (d) 60 °C Figure 7-17 Effect of pH on cyanide oxidation - current vs. potential on a graphite rotating disk at 4900 rpm and different temperatures. Electrolytes : 0.05 M CN", a CN:Cu mole ratio of 3, [OH"] = 0.25, 0.05 and 0.01M and 1 M Na 2 S0 4 . 132 Log (current density IA m'2) Log (current density / A m"2) Figure 7-18 Effect o f p H o n cyanide ox ida t ion - potential vs . l og (current density) o n a graphite rotating d i sk at 4900 r p m (25 and 60 ° C ) . Elect rolytes : 0.05 M C N " , a C N : C u mo le ratio o f 3, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 1 M and 1 M N a ^ . 0.5 O- 0.1 i -o -! i i i i i i | -2 -1.5 -1 -0.5 0 0.5 1 1.5 L o g (current dens i ty / A m"2) Figure 7-19 Effect o f o f p H on cyanide ox ida t ion - potential vs . l o g (current density) o n a py ro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : 0.05 M C N " , a C N : C u mo le ratio o f 3, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 1 M and 1 M N a 2 S 0 4 133 1400 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential vs . S C E / V (a) 25 ° C 3500 , 3000 i-Potential v s . S C E / V (b) 50 ° C 2500 Potential v s . S C E / V (b) 40 ° C 0.00 0.20 0.40 0.60 0.80 E vs . S C E / V (b) 60 ° C Figure 7-20 Effect o f p H o n cyanide ox ida t ion - current vs . potential o n a graphite rotat ing disk at 4900 r p m and different temperatures. Elect rolytes : 0.05 M C N " , a C N : C u m o l e ratio o f 4, [ O H ] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a 2 S 0 4 . 134 L o g (current density / A m'2) Log (current density / A m"2) (a) 25 ° C (b) 60 ° C F igure 7-21 Effect o f p H o n cyanide ox ida t ion - potential vs . l o g (current density) o n a graphite rotating disk 4900 r p m (25 and 60 ° C ) . Elec t ro ly tes : 0.05 M C N " , a C N : C u m o l e ratio o f 4, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a ^ . 0.1 -I 1 1 1 1 - 2 - 1 0 1 2 L o g (current dens i ty / A m"2) Figure 7-22 Effect o f p H on cyanide ox ida t ion - potential vs . l o g (current density) o n a py ro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : 0.05 M C N " , a C N : C u mo le ratio o f 4, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M N a 2 S 0 4 . 135 1800 Potential v s . S C E / V (a) 25 ° C 3000 -. Potential vs . S C E / V (c) 50 ° C 2500 Potential v s . S C E / V (b) 40 ° C 4000 -. Potential vs . S C E / V (d) 60 ° C Figure 7-23 Effect o f p H o n cyanide ox ida t ion - current vs . potential o n a graphite rotating d i sk at 4900 r p m and different temperatures. Elect rolytes : 0.05 M C N " , a C N : C u mo le ratio o f 12, [OH"] = 0.25, 0.05, and 0 . 0 1 M and 1 M NajSCv 136 Figure 7-24 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) o n a graphite rotating d i sk at 4900 r p m (25 and 60 ° C ) . Elect rolytes : 0.05 M C N " , a C N : C u m o l e ratio o f 12, [OH"] = 0.25, 0.05 and 0 . 0 I M and 1 M N a 2 S 0 4 . F igure 7-25 Effect o f p H o n cyanide ox ida t ion - potential vs . l o g (current density) o n a py ro ly t i c graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : 0.05 M C N " , a C N : C u mo le ratio o f 12, [OH"] = 0.25, 0.125, 0.05, 0.025 and 0 . 0 1 M and 1 M N a 2 S 0 4 . 137 400 T Potential vs . S C E / V (a) 25 ° C 5000 -i ! 4500 l Potential vs . S C E / V (c) 50 ° C 1600 Potential vs . S C E / V (b) 40 ° C 9000 Potential vs . S C E / V (d) 60 ° C Figure 7-26 Effect o f p H o n cyanide ox ida t ion - the plot o f the current vs . the potential o n a graphite rotating d i sk at 4900 r p m and different temperatures. Elec t ro ly tes : 3 M C N " , a C N : C u mo le ratio o f 3, 0.50, 0.25 and 0.05 M O H " and 0.5 M N a j S O ^ 138 -0.1 L o g (current dens i ty / A m"2) Figure 7-27 Effect o f p H on cyanide ox ida t ion - potential vs . l og (current density) o n a graphite rotating d i sk at 4900 r p m and 60 ° C . Elect rolytes : 3 M C N " , a C N : C u mo le ratio o f 3, 0.50, 0.25, and 0.05 M O H " and 0.6 M N a 2 S 0 4 . 18000 , Potential vs. SCE / V 25000 Potential vs. SCE / V (a) 25 ° C (b) 40 ° C 139 35000 30000 25000 -0.50 M O H - 0.25 M OH" -0.05 M O H " 0.2 0.4 0.6 0.8 1.0 Potential vs. SCE / V 0.2 0.4 0.6 O.i Potential vs. SCE/V (c) 50 ° C (d) 60 ° C F igure 7-28 Effect o f p H o n cyanide ox ida t ion - current vs. potential o n a graphite rotat ing d i sk at 4900 r p m and different temperatures. Elect rolytes : 4 M C N " , a C N : C u mo le ratio o f 4, [OH"] = 0.5 and 0.25 and 0.05 M and 0.5 M N a 2 S 0 4 . 0.4 > 0.3 111 O co to > c CD o a. 0.2 0.1 -0.1 .0 .50 M O H .0 .25 M O H " .0 .05 M O H " Log (current density / A m ) Figure 7-29 Effect o f p H o n cyanide ox ida t ion - potential vs . l o g (current density) o n a graphite rotating disk at 60 ° C . Electrolytes : 4 M C N " , a C N : C u mo le ratio o f 3, [OH"] = 0.50, 0.25 and 0.05 M and 0.5 M N a 2 S 0 4 . 140 7.6 Reaction Order In order to determine w h i c h o f the copper cyanide species is d ischarged at the electrode surface, the react ion order w i t h respect to copper cyanide species was ca lcula ted b y changing the copper cyanide concentrat ion and the mole ratio o f cyanide to copper and measur ing the current vs. concentration o f copper cyanide species at a constant potential . The concentrations o f copper cyanide species ( C u ( C N ) 2 " , C u ( C N ) 3 2 " and C u ( C N ) 4 2 " ) were calcula ted by so lv ing the mass balance equations (Equat ions 3-1 and 3-2) related to React ions 2-3 to 2-6. A t a C N : C u mo le ratio o f 3, the polar iza t ion curves were measured i n the cyanide concentrat ion range 0.025 - 0.2 M and the temperature range 25 to 60 ° C . The current increased un i fo rmly w i t h increasing concentration o f copper cyanide and the Ta fe l slope remained at about 0.120 V decade" 1 (Figure A - 3 4 i n A p p e n d i x 8). Th i s means that the k ine t ic parameters do not change w i t h changing concentration. A b o u t 97 % o f the copper exists i n the f o r m o f C u ( C N ) 3 2 " and its concentration is propor t ional to the concentrat ion o f the total copper cyanide. The concentrations o f C N " , C u ( C N ) 2 " and C u ( C N ) 4 2 " are very l o w and do not increase un i fo rmly w i t h increasing concentration o f the total copper cyanide . The plots o f l o g (current) vs . l o g (concentrations o f t r icyanide) at constant potentials gave straight l ines hav ing slopes 0.97-0.99 (Figure 7-30). T h i s suggests that the react ion order w i t h respect to t r icyanide is one. Therefore C u ( C N ) 3 2 " cou ld be discharged at the electrode fo rming C u ( C N ) 3 " . The same results were obtained o n a py ro ly t i c graphite rotating d i sk (Figures A - 3 5 and A -36 i n A p p e n d i x 8). F r o m Figures 7-13 and 7-14, at [Cu] = 0.00833 M , the increase i n cyanide concentrat ion f rom 0.025 M (a C N : C u mole ratio o f 3) to 0.05 M (a C N : C u mo le ratio o f 6) resulted i n a change i n the Tafe l slope. Th i s means that the discharged species or the rate-con t ro l l ing step changed. H o w e v e r , w h e n the concentrat ion o f cyanide increased f rom 0.05 to 0.4 M , the Tafe l slope d i d not change. The polar iza t ion curves shifted and were a lmost para l le l to each other. T h i s shift c o u l d be due to a change i n the concentrat ion o f some copper cyanide species. The current at a constant potential was almost propor t ional to the concentrat ion o f C u ( C N ) 4 3 " but not the other copper cyanide species. A t 25 ° C , the plots o f log current vs. l og ( [Cu(CN) 4 3 " ] ) at 0.2 and 0.4 V vs. S C E gave straight l ines h a v i n g slopes o f 141 0.96 and 1.0 respect ively (Figure 7-31). The slopes obtained o n a py ro ly t i c graphite rotating d i sk at 0.4 and 0.6 V vs. S C E were 1.01 and 0.98 (Figure 7-32). The results at 40 , 50 and 60 °C are the same. (a) 25 ° C (b) 60 ° C F igure 7-30 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) o n a graphite rotat ing d i sk at 4900 r p m (25 and 60 ° C ) . Elect rolytes : [CN"] = 0.025, 0.05, 0.1 and 0.20 M , a C N : C u mo le ratio = 3, [ O H ] = 0.25 M and 1 M N a 2 S 0 4 . 2.5 (A C CD T3 c £ 3 Oi o 2.0 E < >, 1.5 1.0 0.5 0.0 -0.5 -2.5 o 0.2 V vs. SCE . 0.4 V vs. SCE -2.4 -2.3 -2.2 Log([Cu(CN) 4 3T / moi dm"3) -2.1 Figure 7-31 Plots o f l o g (current density) vs. l og ( [Cu(CN) 4 2 " ] ) o n a graphite rotat ing d i sk at 4900 r p m and 25 ° C . Elect rolytes : [ C N ' ] = 0.05, 0.1, 0.20 and 0.40 M , [Cu + ] = 0.00833 M , [OH"] = 0.25 M and 1 M N a 2 S 0 4 . 142 2.0 T CM i E 5 1 .5 -'55 c a> ~° 1.0 -C £ I 0.5 -TO o - I 0.0 --2.5 -2.4 -2.3 -2.2 -2.1 Log ([Cu(CN) 4 3 l / moi dm"3) Figure 7-32 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) o n a p y r o l y t i c graphite rotating disk at 4900 r p m and 25 ° C . Elect rolytes : [CN"] = 0.05, 0.10, 0.20 and 0.40 M , [Cu + ] = 0.0833, [OH ' ] = 0.25 M and 1 M N a j S O , . F igure 7-33 shows the plots o f l o g (current density) vs. l og ( [Cu(CN) 4 2 " ] ) w h e n the total cyanide concentrat ion was kept at 0.4 M and the copper concentrat ion was changed. The slopes o f the curves were 0.96 and 0.93 respect ively for 0.2 and 0.4 V vs. S C E , w h i c h correspond to the two Tafe l slope ranges. The react ion order w i t h respect to C u ( C N ) 4 3 ~ obtained on a py ro ly t i c graphite electrode was 1.0. Y o s h i m u r a et a l . [144] studied the anodic ox ida t ion o f copper cyanide o n p la t inum and thought that almost a l l o f the copper exists i n the fo rm o f C u ( C N ) 4 3 " . T h e y plotted l og current vs . l og [ C u ] T o t a l and obtained a slope o f 0.9. H o w e v e r , f rom our ca lcula t ion, 32 -24 % o f the copper exists i n the fo rm o f C u ( C N ) 3 2 " i n the concentrat ion range studied and the concentrat ion o f C u ( C N ) 4 3 " is not exact ly propor t iona l to the total copper concentration. The plot o f l og [Cu(CN) 4 3 " ] vs . l og [ C u ] T o t a l gave a slope o f 0.901. Therefore the corrected react ion order w i t h respect to C u ( C N ) 4 3 " should be 0.99 for Ref . 144. F r o m Figure 7-11 and 7-12, at [ C N ] t o t a l = 0.05 M , the po la r iza t ion curves for C N : C u mo le ratios o f 4, 6, and 12 are very close and the current for a C N : C u mo le ratio o f 6 ( [Cu] = 0.00833 M ) at a constant potential is even larger than that for a C N : C u mo le ratio o f 4 ( [Cu] 143 = 0.0125 M ) . T h i s is because the concentrat ion o f C u ( C N ) 4 3 " for a C N : C u mo le ratio 6 is larger than that for a C N : C u mole ratio o f 4. The react ion order w i t h respect to the copper cyanide species was also calcula ted by changing the total copper cyanide concentrat ion and keeping C N : C u mo le ratio at 48. A t this mo le ratio, most o f copper exists i n the fo rm o f C u ( C N ) 4 3 " . The current is a lmost p ropor t iona l to the concentrat ion o f C u ( C N ) 4 3 " , but not to that o f C u ( C N ) 3 2 " . The plots o f log(current) vs . l og ( [Cu(CN) 4 3 " ] ) gave straight l ines hav ing slopes o f 1.1 and 1.0 respect ively at 0.2 V and 0.4 V vs. S C E . The react ion order measured o n a py ro ly t i c graphite electrode was 1.0. The react ion order w i t h respect to hydrox ide was determined by changing the hydrox ide concentrat ion. F r o m Figures 7-18 and 7-19, at a C N : C u mo le ratio o f 3, the Ta fe l slope changes w i t h hydrox ide concentrat ion and the rate-control l ing step or the react ion mechan i sm changes. F r o m Figures 7-21, 7-22, 7-24 and 7-25, i n the l o w po la r iza t ion reg ion , at a C N : C u mo le ratio > 4, the current changes on ly s l ight ly w i t h hydrox ide concentrat ion and the react ion order w i t h respect to hydroxide is close to zero. Thus the ra te-control l ing step does not i nvo lve hydroxide . 2.7 -, -3.5 -3 -2.5 -2 -1.5 Log([Cu(CN) 4 3 l / moi dm"3) Figure 7-33 Plots o f l og (current density) vs. l og ( [Cu(CN) 3 2 " ] ) o n a graphite rotat ing d i sk at 4900 r p m and 25 ° C . Electrolytes : [CN"] = 0.40 M , [Cu + ] = 0.0167, 0 .00833, 0 .00417, 0.00208, 0.00104 M , [OH"] = 0.25 M and 1 M N a 2 S 0 4 . 7.7 Reaction Between Cyanide and Copper(ll) 144 The react ion between cyanide and Cu(II) ions produces C u ( C N ) 4 2 " ions w h i c h have a v io le t co lor and rap id ly decompose into cyanogen and a copper cyanide species [58-73]. The condi t ions i n the literature reports are different f rom those i n this study and the results cannot be compared. Therefore the experiments o n the react ion between cyanide and Cu(II ) were conducted to understand the phenomena observed i n this study. M i x i n g s o d i u m cyanide and C u S 0 4 gave a transient v io le t co lor w h i c h disappeared i n less than one second. U s i n g U V spectroscopy and s top-f low technology a transient species was detected at 535 n m w h i c h was assumed to be long to C u ( C N ) 4 2 " [49]. F igures 7-34 and 7-35 show the absorbance vs . t ime at 535 ± 3 n m w h e n 2.5 c m 3 o f 0.05 M cyanide solutions w i t h 0.25 and 0.05 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solut ion. F r o m Figure 7-36, the decompos i t ion rate o f C u ( C N ) 4 2 " was decreased w h e n the concentrat ion o f cyanide was increased to 1 M . The plot o f (1/absorbance) vs. t ime (Figure 7-37) is a straight l ine g i v i n g a react ion order w i t h respect to C u ( C N ) 4 2 " o f two . The select ion o f the t ime range for F igure 7-37 is based o n the fact that at the t ime < 6.5 s, the concentrat ion o f C u ( C N ) 4 2 " was too h i g h to be propor t iona l to the absorbance and at the t ime > 9 s, the concentrat ion o f C u ( C N ) 4 2 " was too sma l l and was interfered by the environment. 2.5 -, 2.0 J- 1 0) « 1.5 -re n 0.0 0.5 l 0 2 4 8 10 Time Is Figure 7-34 Absorbance vs. t ime w h e n 2.5 c m 3 o f 0.05 M cyanide so lu t ion w i t h 0.25 M O H ' were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solu t ion at 25 ° C . 145 2.5 , Time Is Figure 7-35 Absorbance vs. t ime w h e n 2.5 c m 3 o f 0.05 M cyanide solu t ion w i t h 0.05 M O H were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solut ion at 25 ° C . Time Is Figure 7-36 Absorbance vs. t ime when 2.5 c m 3 o f 1 M cyanide solu t ion w i t h 0.25 M O H " were m i x e d w i t h 0.4 c m 3 o f 0.05 M copper sulphate solut ion at 25 ° C . 146 0.1 M C u S 0 4 so lu t ion was also gradual ly added to 0.05 M sod ium cyanide solut ions w i t h 0.25 M N a O H and 0.01 M N a O H . In the case o f the solut ion conta in ing 0.25 M N a O H at 25 ° C , after C u S 0 4 was added to the reactor, the l oca l so lu t ion became blue and b lack . T h e n the who le so lu t ion became purple, this co lour disappearing i n less than one second. W h e n the amount o f copper added made the mo le ratio o f cyanide to copper exceed about 2.85, the who le so lu t ion became a l ight blue, w h i c h co lour gradual ly disappeared. N o purple co lor was observed. A t a mo le ratio o f cyanide to copper < 2.75, blue C u ( O H ) 2 began to precipitate u p o n the further addi t ion o f cupr ic ions. A t 50 ° C , the precipitate had a m i x e d co lor o f b lack , b r o w n and blue. Apparen t ly cupric ions reacted w i t h cyanide ions and formed cupr ic cyanide w h i c h decomposed into cuprous cyanide and cyanogen. C u p r i c ions also reacted w i t h hydrox ide to fo rm copper hydroxide or oxide . W h e n o n l y a sma l l amount o f cupric ions was added, the mo le ratio o f cyanide to copper and free cyanide was h i g h and the react ion between cyanide and cupr ic ions was favored. W h e n a large amount o f cupric ions was added, the concentrat ion o f free cyanide became so l o w due to the format ion o f very stable cuprous complexes that the react ion between cupr ic i o n and hydroxide is favored. W h e n the concentrat ion o f sod ium hydrox ide was decreased to 0.01 M , no blue precipitate was observed. Af te r the addi t ion o f cupr ic ions, a loca l y e l l o w i s h co lo r appeared 147 w h i c h became purple and disappeared i n one second. W h e n the mo le ratio o f cyanide to copper was b e l o w 1.6, a whi te precipitate appeared and the p H decreased to 4-4.3 due to the consumpt ion o f hydrox ide ions i n the react ion between cyanide and cupr ic i on : 2 C u 2 + + C N " + 2 0 F T = 2Cu(I) + C N O " + H 2 0 (7-4) U p o n further addi t ion o f cupric ions, more whi te precipitate was produced. The whi te precipitate was apparently C u C N . In the l ight o f the above observations, the phenomenon o f the anodic ox ida t ion o f copper becomes easy to understand. A t a l o w potential , the current is l o w and the mo le ratio o f cyanide to copper on the surface is h igh , prevent ing the precipi ta t ion o f copper ox ide due to the ox ida t ion o f copper cyanide. W h e n the potential exceeds a c r i t i ca l va lue and the current is so h i g h that the mo le ratio o f cyanide to copper is l o w , the ox ida t ion o f copper cyanide produces copper ox ide . W h e n the copper concentrat ion is l o w , the potential needs to be higher to make the current reach a c r i t i ca l value where the mole ratio o f cyanide to copper o n the surface becomes l o w enough to favour the precipi ta t ion o f copper ox ide . Decreas ing the hydrox ide concentrat ion suppresses the format ion o f copper ox ide f rom the v i e w p o i n t o f both thermodynamics and kinet ics . Th i s is i n agreement w i t h the results o n the anodic ox ida t ion o f copper cyanide. 7.8 Cyclic Voltammetry U s i n g c y c l i c vol tammetry , we can evaluate the revers ib i l i ty o f the anodic ox ida t ion o f copper and k n o w the stabil i ty o f the intermediate products. D u r i n g the pos i t ive potential scanning, cuprous cyanide species are o x i d i z e d to cupr ic cyanide species w h i c h m a y be reduced dur ing the negative potential scanning. The scanning rate was kept above 10 V s"1 to a v o i d the precipi ta t ion o f copper oxide and the m a x i m u m potential was b e l o w 1 V vs. S C E to a v o i d the significant evo lu t ion o f oxygen . F igure 7-38 shows the c y c l i c vo l tammet ry o f the solu t ion conta in ing 0.05 M C N - , 0.01666 M C u and 0.25 M N a O H at 25 ° C after subtraction o f the background current. There was no reduct ion current dur ing the negative-go ing scanning. The anodic ox ida t ion o f copper cyanide seems to be i r revers ible . The chemica l react ion o f the o x i d i z e d copper cyanide species is too fast to be detected dur ing the 148 negative scanning. B y increasing the mole ratio o f cyanide to copper and decreasing the p H , there was s t i l l no reduct ion current dur ing the negative potential scanning. 2000 CM • E 1500 --< >» 1000 -densi 500 --*ent Curi 0 „ -500 -0.4 .200V/S .100V /S 50V /S . 20 V/s 10V /S 0.1 0.6 Potent ia l v s . S C E / V Figure 7-38 C y c l i c vol tammetry at 25 ° C . Elec t ro ly te : 0.025 M C N " , C N : C u mo le ratio =3, 0.25 M N a O H and 1 M N a 2 S 0 4 . 7.9 Possible Reaction Mechanism A t a C N : C u mole ratio = 3 and [OH"] = 0.25 M , over 97 % o f the copper exists i n the fo rm o f C u ( C N ) 3 2 " and the current is proport ional to the concentrat ion o f C u ( C N ) 3 2 " but not the concentrations o f C u ( C N ) 2 " and C u ( C N ) 4 3 " . T h i s suggests that C u ( C N ) 3 2 " can be discharged and o x i d i z e d to C u ( C N ) 3 " at the electrode. C u ( C N ) 3 " is unstable and decomposes to fo rm cyanogen. Cyanogen reacts w i t h hydroxide to produce cyanate. W h e n the electrode potential exceeded a certain value, copper ox ide was formed o n the electrode (both o n the graphite and the outer insulator) and the current increased sharply to a l i m i t i n g value. Copper ox ide or hydrox ide can be formed by three ways : (1) copper cyanide decomposes into free cyanide and cuprous ions w h i c h are o x i d i z e d to copper ox ide o n the anode, (2) copper cyanide is o x i d i z e d to free cyanide and cupr ic ions w h i c h react w i t h hydrox ide to fo rm copper ox ide and (3) cuprous cyanide is o x i d i z e d to cupr ic cyanide w h i c h reacts w i t h hydrox ide to fo rm copper ox ide . The format ion o f copper ox ide o n the outer 149 insulator means that cupr ic species diffuse to the surface o f the outer insulator and react w i t h O H " to fo rm copper ox ide and hydroxide . The most l i k e l y mechan i sm is: the cuprous c o m p l e x ( C u ( C N ) 3 2 " ) is o x i d i z e d to cupr ic complex (Cu(CN) 3 " ) and some o f the cupr ic c o m p l e x decomposes to fo rm cyanogen. Some o f it reacts w i t h hydrox ide to f o r m hydrox ide o n the anode and a sma l l amount diffuses to the surface o f the outer insulator to f o r m copper ox ide . The mechan i sm o f the anodic ox ida t ion o f cyanide at h i g h potentials is different f rom that at l o w potentials. W i t h decreasing concentrat ion o f hydroxide , the current and the Tafe l slope decrease. T h i s means that hydrox ide affects the rate-control l ing step. The decrease i n the Tafe l slope f rom 0.12 V to 0.060 V decade" 1 means the rate-control l ing step changes or even the mechan i sm changes. A t h i g h hydroxide concentration, the the rmodynamic s tabi l i ty o f the copper (I) species is re la t ively l o w and the species are more easi ly o x i d i z e d . W i t h decreasing hydrox ide concentration, the e lectrochemical stabil i ty o f C u ( C N ) 3 2 " increases and copper cyanide becomes less e lec t rochemical ly activated. A s the C N : C u mole ratio increases, the current and Tafe l slope also decrease at l o w potentials. The Tafe l slope decreases to about 0.060 V decade" 1 w h e n the mo le ratio exceeds 4 at [CN"] = 0.05 M . T h i s means that the rate-control l ing step or the react ion mechan i sm changes. W i t h further increase i n potential , the second Tafe l slope (0.160 to 0.200 V decade" ') appeared. The current at a constant potential is propor t ional to the concentrat ion o f C u ( C N ) 4 3 " and but independent o f the concentrat ion o f hydrox ide . The discharged species are not sensit ive to hydrox ide i o n and therefore it is u n l i k e l y that C u ( C N ) 3 2 " is discharged. C u ( C N ) 4 " 3 is most l i k e l y to be discharged at the electrode. A t [OH"] = 0.01 M and a constant potential , the ratio o f the current measured i n 0.05 M C N " solutions w i t h C N : C u mole ratios o f 3 and 4 is close to the mo le ratio o f C u ( C N ) 4 3 " o f the solutions. Therefore the discharged species c o u l d transfer f rom C u ( C N ) 3 2 " and C u ( C N ) 4 3 " to C u ( C N ) 4 3 " w i t h decreasing hydrox ide concentration. A t h i g h hydrox ide concentrat ion, C u ( C N ) 3 2 " is discharged m u c h faster than C u ( C N ) 4 3 " and the current contr ibuted by C u ( C N ) 4 3 " can be neglected compared to that o f C u ( C N ) 3 2 " . W i t h decreasing hydrox ide concentrat ion, the discharge o f C u ( C N ) 3 2 " is suppressed and the discharge o f C u ( C N ) 4 3 " maintains the constant rate and becomes the dominant discharged species. 150 The amount o f copper oxide formed decreases w i t h increasing C N : C u mo le ratio and decreasing hydrox ide concentration. N o copper oxide formed at the outer insulator at a C N : C u mo le ratio >3.5 or [OH"] < 0.05 M . W h e n copper ox ide was precipi tated o n the electrode, w i t h decreasing hydroxide concentration, the current decreases w i t h decreasing hydrox ide concentration. The higher the C N : C u mole ratio, the less the effect o f hydrox ide , the higher the potential for the precipi ta t ion o f copper ox ide and the more stable the copper (I) species. The f o l l o w i n g possible mechanisms are proposed: (1) In the l o w potential region (< about 0.4 V vs. S C E ) : A t a C N : C u mo le ratio = 3 and a h i g h concentration o f hydrox ide (0.25 M OH") Step 1 C u ( C N ) 3 2 " -> C u ( C N ) 3 - + e Step 2 2 C u ( C N ) 3 " -> 2 C u ( C N ) 2 " + (CN)2 Step 3 ( C N ) 2 + 2 0 H " -> C N O " + C A T + H 2 0 Step 1 c o u l d be the rate-control l ing step f rom a Tafe l slope o f 0.12 V decade" 1 [335, 336] and the discharge o f C u ( C N ) 4 3 " is negl ig ib le compared to C u ( C N ) 3 2 " . Step 1 is ca ta lyzed by hydrox ide ions. H y d r o x i d e ions might be w e a k l y bound to C u ( C N ) 3 2 " to f o r m a surface complex such as C u ( C N ) 3 2 " O H " w h i c h is more readi ly discharged o n the anode. W i t h decreasing p H , the above react ion is suppressed probably due to the decrease i n the surface complex concentration, the current decreases, and the discharge o f C u ( C N ) 4 3 " becomes the dominant anodic react ion. Increasing C N : C u mole ratio has a s imi la r effect because it shifts the d is t r ibut ion o f copper cyanide species f rom l o w l y coordinated complexes to a h igh ly coordinated c o m p l e x ( C u ( C N ) 4 3 " ) and probably also suppresses the format ion o f a surface c o m p l e x (such as C u ( C N ) 3 2 " O H " ) . The c r i t i ca l value for the C N : C u mole ratio depends o n the total copper cyanide concentrat ion because the dis t r ibut ion o f copper cyanide species is dependent o n cyanide concentration. F o r example , at [CN"] = 0.05 M , w h e n a C N : C u m o l e ratio > about 4, the discharge o f C u ( C N ) 4 3 " is dominant. H o w e v e r , at [CN"] =3.5, w h e n C N : C u m o l e ratio > 3.5, the discharge o f C u ( C N ) 4 3 " becomes dominant. W h e n the dominant discharged species is C u ( C N ) 4 3 " , the anodic react ion p robab ly consists o f the f o l l o w i n g steps according to the observed kinet ics : 151 Step 1 C u ( C N ) 4 3 - C u ( C N ) 4 2 - ( a A ) + e 4 Step 2 C u ( C N ) 42 - ( f l A ) » C u ( C N ) k_2 Step 3 2 C u ( C N ) 4 2 ' 2 C u ( C N ) 3 2 - + ( C N ) 2 Step 4 ( C N ) 2 + 2 0 F T -> C N O + C N " + H 2 0 The adsorption rate for the coverage o f C u ( C N ) 4 2 " , (d9/dt), can be expressed b y the f o l l o w i n g equation: dB — = kx(\- 0)[Cu(CN)43-]- k_x6- k20+ k_2(\ - 9)[Cu{CN)2~} (7-5) where 9 is the coverage o f C u ( C N ) 4 2 " on the electrode, k, the rate constant for the e lec t rochemica l adsorption, k , the rate constant for the e lec t rochemical desorption, k 2 the rate constant for the chemica l desorption and k 2 the rate constant for the chemica l adsorption. A t steady state, d9/dt = 0 and i f 9 « 0 and k , » k 2 , the f o l l o w i n g equat ion can be obtained f rom Equa t ion 7-5: e=^mli (7.6) k_[ + k 2 In the in i t i a l l o w potential region, i f k , » k 2 , the f o l l o w i n g equation can be obtained: k , , k0, exp(aFE IRT) k 0 , FE 6 = - - 4 C u ( C N ) 4 3 - ] = T , ^ , g < n = T ^ e x p C — ) (7-7) k. , A : 0 _ 1 e x p ( - ( l - a ) F £ / / ? 7 / ) «:„_, RT where a is the charge transfer coefficient, k 0 , and kg., the rate constants respect ively for ox ida t ion and reduct ion at E = 0 and k, = ko j e x p ( a F E / R T ) and k , = k o . , e x p ( - ( l - a ) F E / R T ) . Reac t ion rate =k 2 9 (7-8) F k k F E i = F k 2 0 = — ^ [ C ^ C N ) , 3 - ] e x p ( — ) (7-9) K 0 - l K 1 F r o m E q u a t i o n 7-9, the react ion order w i t h respect to C u ( C N ) 4 3 " is one and the Ta fe l slope is R T / F ( a b o u t 0.06 V decade" 1). Therefore the above assumption is consistent w i t h the exper imental results. W h e n the potential increases to a value where k , « k 2 , f rom E q u a t i o n 7-6, the coverage o f the adsorbed C u ( C N ) 4 2 " can be expressed as: 152 k , [ C u ( C N ) 4 3 - ] e= 1 4 (7-10) k 2 t^ zFE i = F k 2 # = F k , [ C u ( C N ) 4 3 - ] = F k 0 , [ C u ( C N ) 4 3 - ] e x p ( ^ r ) (7-11) F r o m the above equation, the react ion order w i t h respect to C u ( C N ) 4 3 " is one and the Tafe l slope is R T / a F . T h i s is consistent w i t h the experiment results. It should be poin ted out that Step 3 invo lves some elementary reactions. F r o m the plot o f log(current density) vs. potent ial (or potential vs . l og (current density) according to Equat ions 7-9 and 7-11, w e can calculate F k 2 k 0 ^ [ C ^ C N ) / " ] / ] ^ . , and Fkg , [ C u ( C N ) 4 3 " ] . Therefore k o y k 2 can be calcula ted f rom the above values. A t [CN"] = 0.1 M , C N : C u mo le ratio = 12 and 25 ° C , F k 2 k 0 , [ C u ( C N ) 4 3 \ | / k M and F k o , [ C u ( C N ) 4 3 " ] are about 8.33x10" 4 and 0.546 A m" 2 respect ively. So kQJk2 is 726 and k , / k 2 is 726 exp ( - ( l -oc )FT/RT) . A t potentials < 0.20 V vs. S C E , k , /k 2 is above 10 and so the Tafe l slopes are about 0.060 V decade" 1. A t a potential > 0.35 V vs. S C E , k , /k 2 is b e l o w 1/10 and the Tafe l slope is about 0.171 V decade" 1 ( a = 0.35). Equa t ion 7-6 can be rearranged as: (k, / k , ) [ C u ( C N ) 4 3 - ] _ ( k 0 , / k 0 H ) [ C u ( C N ) 4 3 - ] e x p ( F E / R T ) l + k 2 / k _ , l + t k . / k ^ e x p ^ l - c O F E / R T ) ( " ) i = bk~,t) = (7-13) F ( k 2 k 0 f l / k 0 _ 1 ) [ C u ( C N ) 4 3 - ] e x p ( F E / R T ) 1 + ( k 2 / k 0 _!)exp((l - a ) F E / R T ) F r o m the above equation, the react ion order w i t h respect to C u ( C N ) 4 2 " is one at any potential . F igure 7-39 shows the plots o f potential vs . l o g (current density) us ing data measured and predicted us ing Equa t ion 7-13. The predicted data are consistent w i t h the data measured at a potential < 0.45 V vs. S C E . H o w e v e r , at potentials > 0.45 V , the measured data appear to deviate f rom the predicted value. T h i s is because at potentials > 0.45 V , the assumptions are not v a l i d and the difference i n the copper concentrat ion between the bu lk so lu t ion and the surface is not negl ig ib le . The current is so h igh that the coverage o f C u ( C N ) 4 2 " cannot be neglected and the chemica l desorption determines the who le react ion rate. C u ( C N ) 4 2 " is m u c h less adsorbed o n a py ro ly t i c graphite electrode. Therefore there appears to be on ly one w e l l -defined Tafe l slope. W i t h further increase i n potential , the current reaches a c r i t i ca l va lue and the C N : C u mo le ratio o n the electrode surface decreases to such a l o w value that copper ox ide or hydrox ide is precipi tated on the anode. F r o m the standpoint o f thermodynamics , d icyan ide 153 and t r icyanide are less stable and more readi ly o x i d i z e d to copper ox ide and cyanide . The effect o f the precipi tated copper ox ide on the anodic ox ida t ion o f copper cyanide depends o n the appl ied potential , temperature and total cyanide concentration. A t [CN"] = 0.05 M and a temperature > 40 °C, copper ox ide catalyzes the ox ida t ion o f copper cyanide. A t a temperature < 40 °C, copper ox ide has a l i m i t e d catalyt ic effect o n the cyanide ox ida t ion . It m a y even exhibi t an inh ib i t ing effect at a potential > 0.6 V vs. S C E . A t [Cu + ] = I M and [CN"] = 3 M , the format ion o f copper oxide or hydrox ide s igni f icant ly inhibi ts the anodic ox ida t ion o f copper cyanide. T h i s m a y be related to the properties o f the precipi tated copper oxide and to the adsorption o f copper cyanide. The ox ida t ion o f free cyanide was cata lyzed by cupric ox ide formed o n the electrode because i n the absence o f copper, the anodic current o f free cyanide o n the copper ox ide -coated anode is s ignif icant ly higher than that on the anode wi thout copper ox ide . Cu(III) species such as C u 2 0 3 can be produced i n the potential range studied [341-345]. F o r example , Cu(III) ox ide phase was s tabi l ized at approximately 0.48 V vs . S C E and 0 ° C i n a lka l ine so lu t ion [341]. The ox ida t ion o f the Cu(II) species began at about 0.35 V vs . S C E and 24 ° C i n 1 M N a O H [343] and the intr insic redox potential for Cu(III) /Cu(II) i n the s o l i d ox ide is 0.42 V vs . S C E at p H 14 and 20 ° C [344]. It is possible for Cu(III) to f o r m o n the surface and catalyze the cyanide ox ida t ion as was suggested by W e l l s and Johnson [157]. O x y g e n evo lu t ion was also cata lyzed poss ib ly by the format ion and decompos i t ion o f C u 2 0 3 [345]. The react ion procedure can be expressed by the f o l l o w i n g set o f poss ible reactions: Step 1 C u ( C N ) n - ( n " , ) -> C u ( C N ) n - ( " - 2 ) M , ) +e ( n=2 , 3, 4) Step 2 C u ( C N ) n " ( n " 2 ) ( a * ) + 2 0 H ~ -> nCN~(ads) + C u ( O H ) 2 (or C u O + H 2 0 ) Step 3 C u t C N ) ^ " - 1 ' - » Cu{CN);(n'2) + e Step 4 Cu(CN)„-("-2) = Cu(CN)(n_]}-{"-2) + \/2(CN)2 Step 5 C N - ( f l * ) + 2 0 H - ) C N Q - + H 2 0 W i t h decreasing p H and increasing mole ratio o f cyanide to copper, Step 1 (n = 2 and 3), Step 2, Step 3 (n = 3) and Step 5 are suppressed. T h i s results i n a decrease i n the current and it is i n agreement w i t h the experimental results. A t a h i g h C N : C u ratio and l o w p H , no copper ox ide is formed. The catalysis o f copper ox ide was prevented w i t h increas ing the potential and the copper cyanide concentration. 154 0.5 > 0.4 LU W 0.3 -CO > .5 0.2 --c o t? 0.1 --0 -I . . - 1 0 1 2 3 L o g (current dens i ty / A m"2) Figure 7-39 Plo ts o f potential vs. l og (current density) us ing data measured and predic ted us ing E q u a t i o n 7-12 at 25 °C. Elec t ro ly te : 0.1 M C N " , C N : C u mole ratio = 12 0 25 M N a O H and 1 M N a 2 S 0 4 . 7.10 Diffusion Coefficient Estimation In the presence o f a large amount o f supporting electrolyte, the l i m i t i n g current for a s imple e lec t rochemical react ion on the rotating disk can be expressed by Equa t ion 6-5. The di f fus ion coefficients can be calculated f rom the slopes o f the straight l ines for the plots o f i , vs . co 1 / 2 . In this study, w h e n the current reaches the l i m i t i n g value, cuprous cyanide is o x i d i z e d to cupr ic cyanide w h i c h undergoes two further react ion paths. One is that cupr ic cyanide reacted w i t h hydrox ide to produce copper oxide or hydrox ide and free cyanide w h i c h is further o x i d i z e d to cyanate. Ano the r is that cupr ic cyanide species diffuse f rom the surface and rap id ly decompose to fo rm cyanogen and lower coordinated copper cyanide . The di f fus ion o f cupr ic species to the bu lk solu t ion has the f o l l o w i n g effect o n the l i m i t i n g current: (1) the decompos i t ion i n the dif fus ion layer results i n the shift o f the d is t r ibut ion to the format ion o f the l o w l y coordinated copper (I) complex and affects the concentrat ion gradient o f copper cyanide species and affects the l i m i t i n g current; (2) the undecomposed cupr ic species dur ing the dif fus ion b r ing cyanide to the bu lk resul t ing i n the decrease i n the l i m i t i n g current. 155 F r o m the coulometr ic measurement, at a C N : C u mo le ratio = 3 and [OFT] = 0.25 M , the anodic ox ida t ion o f copper cyanide can be expressed as Reac t ion 7-1, i.e. the ox ida t ion o f one c o m p l e x gave 7 electrons and C u ( C N ) 3 2 " is comple te ly o x i d i z e d to cyanate and copper ox ide . So the amount o f cupr ic cyanide reaching the bu lk solu t ion is very s m a l l . Otherwise more cyanide and less cuprous ions are ox id i zed . The decompos i t ion o f cupr ic cyanide (main ly C u ( C N ) 3 " ) produces C u ( C N ) 2 \ w h i c h does not affect the concentrat ion o f C u ( C N ) 3 2 " i n the d i f fus ion layer accord ing to the ca lcula t ion at C N : C u < 3. The plots o f the l i m i t i n g current vs. co 1 / 2 for 0.05 M C N " and a C N : C u m o l e ratio = 3 (Figure 7-40) are l inear. The slopes were calculated us ing least-squares f i t t ing. A t C N : C u = 3, 97 % o f copper and cyanide exist i n the fo rm o f C u ( C N ) 3 2 " and the calcula ted d i f fus ion coefficients can be assumed to be that o f C u ( C N ) 3 2 " . The dif fus ion coefficients for C u ( C N ) 3 2 " at 40 , 50 and 60 °C were found to be 1.05xl0" 9 , 1 .29xl0" 9 and 1 .52xl0" 9 m 2 s"1 respect ively . The di f fus ion act ivat ion energy is 16.6 kJ /mole . F r o m the act ivat ion energy and E q u a t i o n 6-18, the predicted di f fus ion coefficient at 25 °C is 0.76x10" 9 m 2 s" \ F igure 7-40 L i m i t i n g current vs. rotational speed at 40, 50 and 60 ° C . E lec t ro ly te : 0.05 M C N " , C N : C u mo le ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 . W h e n the mole ratio o f cyanide to copper is very large, C u ( C N ) 4 3 " is dominant and is o x i d i z e d to C u ( C N ) 4 2 " w h i c h diffuses f rom the surface and decomposes to f o r m C u ( C N ) 3 2 " 156 and cyanogen. C u ( C N ) 3 2 " reacts w i t h free cyanide to regenerate C u ( C N ) 4 3 " . Therefore the observed l i m i t i n g current is larger than that expected f rom the L e v i c h equation. 7.11 Activation Energy Calculation for the Kinetic Current A t a constant potential , Equa t ion 6-17 can be appl ied. The act ivat ion energy can be calcula ted f rom the slope o f the plot o f l og i vs . 1/T. The slopes o f these l inear plots were calculated by least-squares fi t t ing. W h e n the mo le ratio o f cyanide to copper is 3 and the concentrat ion o f hydrox ide is 0.25 M , the discharged species is C u ( C N ) 3 2 " . F r o m the ca lcula t ion , the concentrat ion o f C u ( C N ) 3 2 " is almost constant i n the temperature range o f 25 to 60 °C and the change i n the concentration o f C u ( C N ) 3 2 " does not need to be cons idered for the ac t ivat ion energy calcula t ion. A t [CN"] =0.05 M , a C N : C u mole ratio = 3, [OH"] = 0.25 M , the plots o f l o g i vs . 1/T and the act ivat ion energies are shown i n F igure 7-41. 2.0 CM i 1.0 0.003 0.0031 0.0032 0.0033 0.0034 (T / K) ,-1 7-41 A c t i v a t i o n energy ca lcula t ion- plot o f l og (current density) vs . 1/T at constant potentials. E lec t ro ly te : 0.05 M C N " , C N : C u mo le ratio = 3, 0.25 M N a O H and 1 M N a 2 S 0 4 . 157 7.12 Summary The e lec t rochemical k ine t ic behavior o f copper cyanide is dependent on C N : C u mo le ratio, p H and total cyanide concentration. A t l o w potentials ( roughly 0 to 0.4 V vs . S C E ) , cuprous cyanide is o x i d i z e d to cupric cyanide complexes w h i c h produce cyanogen, w h i c h i n turn reacts w i t h hydrox ide to fo rm cyanate. A t a C N : C u mole ratio = 3 and [OH"] = 0.25 M , the Tafe l slope is about 0.12 V decade" 1 and the react ion order w i t h respect to C u ( C N ) 3 2 " is one. C u ( C N ) 3 2 " is discharged at the electrode. The current and Tafe l slope decrease w i t h decreasing hydrox ide concentrat ion and so hydrox ide is i n v o l v e d i n the rate-determining step. Increasing C N : C u mole ratio also results i n the change i n the anodic behaviour o f copper cyanide . W h e n the C N : C u mole ratio is larger than a certain value w h i c h depends o n the total cyanide concentration, e. g. about 4 at [CN"] = 0.05 M and 3.5 at [CN"] = 3.5 M , a Tafe l slope o f about 0.06 V decade" 1 was observed over the potential range 0.1 - 0.25 V vs . S C E . A second Tafe l slope o f about 0.17 -0.20 V decade" 1 was noted over the h igher potent ial range. T h i s change is related to the change i n the dis t r ibut ion o f copper cyanide species w h i c h i n turn depends on the total cyanide concentration. The current is propor t iona l to the concentrat ion o f cuprous tetracyanide and almost independent o f the total cyanide concentrat ion. p H has li t t le effect o n cyanide oxida t ion . C u ( C N ) 4 3 " is d ischarged at the electrode. In the midd le potential reg ion (roughly 0.4 to 0.6 V vs . S C E ) , copper ox ide is precipi tated o n the electrode. Copper cyanide is o x i d i z e d to copper ox ide and cyanate. The potential for the precipi ta t ion o f copper oxide is dependent o n C N : C u mo le ratio and temperature. The higher the mole ratio o f cyanide to copper, the higher the potent ial for the precipi ta t ion o f copper ox ide . H o w e v e r , w h e n cyanide concentrat ion was h i g h and hydrox ide concentrat ion was l o w , no copper oxide was precipitated, but cyanogen gas was evo lved . The current decreases w i t h decreasing hydroxide concentrat ion and the rate-con t ro l l ing step involves hydroxide . The catalysis o f copper ox ide precipi tated decreases w i t h increasing copper cyanide concentration. The anodic behaviour o f copper cyanide c o u l d be compared to that o f sulphite and the mixture o f sulphite and copper cyanide to understand h o w sulphite can l i m i t the ox ida t ion o f cyanide. 158 8. ANODIC OXIDATION OF MIXED COPPER CYANIDE AND SULPHITE IN ALKALINE SOLUTION The anodic behaviour o f sulphite and copper cyanide has been discussed i n Chapters 6 and 7 w h e n they are i n the solut ion separately. The anodic behaviour o f m i x e d sulphite and copper cyanide solut ion is presented i n this Chapter. The objective o f this study has been to understand h o w sulphite is o x i d i z e d as a sacr i f ic ia l species w h i l e protect ing the cyanide f rom oxida t ion . The study was conducted us ing the rotating disk technique. 8.1 Experimental Apparatus and Set-up The graphite rotating disk was the same as described i n Sec t ion 6.3. The electrode treatment was the same as i n Chapter 7. Graphi te rod hav ing 12- and 2 4 - m m diameters was fashioned as rotating disks for coulometr ic measurements. The w o r k i n g electrode (anode) was separated f rom the counter electrode (the cathode) to m i n i m i z e the effect o f the change i n the C N : C u mole ratio due to copper deposi t ion at the cathode. The v o l u m e o f the catholyte was on ly about 1-2 c m 3 and the in i t i a l concentration o f hydroxide was ten t imes that i n the anolyte. The evo lu t ion o f hydrogen buil t a h igh concentrat ion o f hydrox ide w h i c h can be transported to the anode compartment to mainta in the concentrat ion o f hydrox ide i n the anolyte. The p H o f the anolyte was moni tored. The rotating disk electrode system was an E G & G m o d e l 636 Elec t rode Rotator . A S O L A R T R O N 1286 E lec t rochemica l Interface was used as the potentiostat. Excep t as noted, the po la r iza t ion curves were generated at a scanning rate o f 1 m V s"1. The exper imental set-up was the same as shown i n F igure 6-3. The l i q u i d j u n c t i o n potential was not considered since the concentrat ion o f hydrox ide is not very h igh and the mobi l i t i es o f the ions o f sulphate, sulphite and copper cyanide species are close to that o f the sod ium ion . The thermal l i q u i d j unc t i on potential was measured us ing two ca lome l reference electrodes w h i c h were p laced o n the two sides o f an electrolyte bridge. Samples were taken for cyanide analysis ( A p p e n d i x 2) and sulphite analysis ( A p p e n d i x 4). The copper concentrat ion was measured by o x i d i z i n g copper cyanide to cupr ic 159 nitrate us ing concentrated ni t r ic ac id and titrating w i t h E D T A (see A p p e n d i x 3). The analysis o f copper i n the anodic precipitate was conducted by d i s so lv ing the precipitate i n n i t r ic ac id and titrating w i t h E D T A . Reagent grade chemicals were used throughout a l l the experiments. 8.2 Anodic Behaviour of Mixed Sulphite and Copper Cyanide Solution 8.2.1 Anodic Behaviour of Dilute Copper Cyanide Solution with Sulphite The anodic ox ida t ion o f m i x e d sulphite and copper cyanide has been studied as a funct ion o f temperature, the mole ratio o f cyanide to copper, sulphite concentrat ion and hydrox ide concentrat ion. F igure 8-1 shows the polar iza t ion curves o f the solu t ion w i t h 0.05 M C N " , a C N : C u mole ratio o f 3, 0.4 M N a ^ , 0.25 M N a O H and 1 M N a 2 S 0 4 . A t 25 and 40 ° C , the current first increased and then decreased sharply to a m i n i m u m value w i t h the format ion o f copper oxide on the anode. W i t h further increase i n potential , the current increased again. A t a potential > about 0.8 V vs. S C E , some gas bubbles were observed o n the anode. T h e y were be l ieved to be due to o x y g e n evolu t ion . The pass iva t ion is probably due to the precipi ta t ion o f copper oxide and the adsorption o f oxygen . A very th in layer o f copper oxide was precipitated o n the graphite but not on the outer insulator. W h e n on ly copper cyanide was present i n the solut ion, copper ox ide was precipi tated both o n the graphite and the outer insulator w i t h the amount o f copper ox ide be ing m u c h larger. Therefore sulphite can reduce cupric ions to cuprous ions and decrease the extent o f copper ox ide format ion. A t 50 ° C , the polar iza t ion curves (Figure 8 - l c ) became different. A t 100 r p m , the current increased to a l i m i t i n g value, w h i c h was approximate ly the s u m o f copper cyanide and sulphite l i m i t i n g currents when they are present separately i n the solut ion. A t 400 and 1600 r p m , the current first increased and then decreased to a m i n i m u m value w i t h the precipi ta t ion o f copper oxide . A t a potential > 0.64 V vs. S C E , the current rose sharply to a l i m i t i n g value and the electrode surface was reactivated. A t a potential > 1.0 V vs . S C E , bubbles were observed and the current decreased sharply. O x y g e n evo lu t ion passivated the electrode surface. 160 A t 60 ° C , the anodic behaviour for 100 and 400 r p m is s imi la r to that for 100 r p m at 50 °. H o w e v e r , at 1600 rpm, the polar iza t ion curve was s t i l l s imi la r to that at 50 ° C . T h i s dependence o f the anodic behaviour o n the rotational speed is due to the difference i n the compos i t i on at the electrode surface at different rotational speeds. The difference i n the composi t ions o f the electrolyte can affect the precipi ta t ion o f copper ox ide and evo lu t ion o f oxygen and f ina l ly the e lect rochemical properties. F r o m Figure 8-2, it can been seen that the pass ivat ion decreased w i t h increasing potential scan rate. A t 5 raV/s, the current increased to a m a x i m u m and decreased w i t h the precipi ta t ion o f copper ox ide f ina l ly increasing to a l i m i t i n g value. A t 10 and 20 m V / s , the current increased cont inuously to a l i m i t i n g value. Th i s current was related to the ratio o f the precipi tated copper ox ide to copper hydroxide . F r o m the polar iza t ion curves o n the electrode w i t h and wi thout pre-coated copper oxide (Curves 1 and 2 i n F igure 8-3) i n the solut ion containing both copper cyanide and sulphite, the copper oxide had an inh ib i t ing effect on the ox ida t ion o f copper cyanide and sulphite. H o w e v e r , i n compar ing the polar iza t ion curves containing on ly sulphite (Curves 3-5) i n F igure 8-3, the copper cyanide oxide d i d not show a large inh ib i t ing effect o n the ox ida t ion o f sulphite. Therefore the passivat ion might be caused by the adsorpt ion o f copper cyanide species i n the presence o f sulphite or concomitant effect o f copper cyanide and sulphite. In compar ing three polar iza t ion curves respect ively for (1) m i x e d sulphite and copper cyanide , (2) sulphite and (3) copper cyanide (Figure 8-4), it can been seen that copper ca ta lyzed the ox ida t ion o f sulphite. The anodic behaviour for 0.2 M and 0.1 M N a 2 S 0 3 was shown i n A p p e n d i x 8 (Figures A - 3 7 and A - 3 8 ) is different f rom that for 0.4 M N a 2 S 0 3 . M o r e copper ox ide was formed and more oxygen evolved . In the potential range 0.6 -1.0 V vs. S C E , the current d i d not change s ignif icant ly w i t h decreasing sulphite concentrat ion f rom 0.4 to 0.1 M . The decrease i n the sulphite concentration resulted i n an increase i n o x y g e n evolu t ion . W h e n the mole ratio o f cyanide to copper increased f rom 3 to 4 ( [Cu + ] decreased f rom 0.0167 to 0.0125 M ) at [CN"] = 0.05 M , the polar iza t ion curves were different (Figures A - 3 9 to A - 4 0 i n A p p e n d i x 8). The difference is due to the change i n the d is t r ibut ion o f copper cyanide species. 161 The precipi ta t ion o f copper oxide affected the anodic ox ida t ion o f sulphite and copper cyanide. The concentrat ion o f hydroxide was decreased to 0.05 M f rom 0.25 M to see its effect on the anodic behaviour o f sulphite and copper cyanide. F igure 8-5 shows the po la r iza t ion curves for the solut ion w i t h 0.05 M C N - , 0.0167 M C u + ( C N : C u = 3), 0.4 M N a 2 S 0 3 , 0.05 M N a O H and 1 M N a 2 S 0 4 . The current first increased smooth ly w i t h increasing potential . W h e n the potential was larger than a certain value (dependent o n the rotat ional speed), it rose rapid ly to a l i m i t i n g value and then decreased s l igh t ly w i t h increasing potential . A t the potential > about 0.70 V vs. S C E , the current increased s l ight ly and became stable around 1.0 V vs. S C E . N o v i s ib le copper ox ide was formed. F r o m Figure 8-6, at a potential < about 0.30 V vs. S C E , the sulphite ox ida t ion d i d not seem to be cata lyzed by the ox ida t ion o f copper cyanide. H o w e v e r , at a potential > about 0.3 V vs. S C E , the current increased rapid ly and the sulphite ox ida t ion was ca ta lyzed by the ox ida t ion o f copper cyanide. The potential for the sharp increase i n the current for m i x e d sulphite and copper cyanide is almost the same as that for copper cyanide. A t a potential > about 0.9 V vs. S C E , the current d i d not increase as expected f rom the further ox ida t ion o f sulphite species ( H S 0 3 " or S 0 2 ) poss ib ly because the ox ida t ion o f S 0 3 2 " , but not H S 0 3 2 " and S 0 2 , was ready to be catalyzed by the ox ida t ion o f copper cyanide and the electrode surface was passivated for the ox ida t ion o f H S 0 3 " a n d S 0 3 2 . W h e n the sulphite concentration decreased f rom 0.4 M to 0.2 M , copper ox ide and hydrox ide was precipitated on the electrode. Hence the anodic behaviour (see F igure A - 4 2 i n A p p e n d i x 8) became quite different. The anodic behavior for 0.05 M C N " , 0.0125 M C u + ( C N : C u = 4) , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 (Figure A - 4 3 i n A p p e n d i x 8) was s imi la r to that for C N : C u = 3 (F igure 8-5). The current first increased smoothly and then rose rapid ly to a m a x i m u m value. W h e n the concentrat ion o f sulphite was decreased to 0.2 M , the anodic behaviour (Figure A - 4 4 i n A p p e n d i x 8) was s imi la r to that for 0.4 M N a 2 S 0 3 (Figure A - 4 4 ) . H o w e v e r , w h e n the concentrat ion o f sulphite was decreased to 0.1 M , the anodic behaviour ( A - 4 5 i n A p p e n d i x 8) was different due to the format ion o f copper oxide . F r o m Figure 8-7, it appears that sulphite ox ida t ion was cata lyzed by the oxida t ion o f copper cyanide at a potential > about 0.35 V vs . S C E . 162 800 700 (M E 600 < £< 500 '55 c 400 a> TJ *-» C 300 £ Cu 200 100 1600 rpm 400 rpm . . . . 100 rpm Precipitation of copper oxide _ 0 -I— 0.0 Evolution of oxygen 0.2 2500 0.4 0.6 0.8 Potential vs. SCE / V 1600 rpm — 400 rpm . . . 100 rpm Precipitation of copper oxide 0.2 0.4 0.6 0.8 Potential vs. S C E / V 7000 6000 5000 4-4000 3000 2000 1000 4-0 (a) 25 ° C 0.0 1600 rpm . — 400 rpm . . . 100 rpm Precipitation of copper oxide Evolution of oxygen 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V 1.2 9000 j 8000 CM 7000 --E < 6000 --5000 --in c o> •a 4000 --c £ 3000 3 o 2000 1000 --(b) 40 ° C 1600 rpm hvoiution ot oxygen 400 rpm . . . . 100 rpm Precipitation of copper oxide J i r \ — \ \ \ \ V V 0.0 0.2 0.4 0.6 0.8 Potential vs. S C E / V 1.0 1.2 (c) 50 ° C (c) (d) 60 ° C Figure 8-1 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 0.0167 M C u + ( C N : C u mole ratio = 3), 0.25 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 163 9000 Potent ia l v s . S C E Figure 8-2 Effect o f potential scanning rate on the anodic behaviour o f m i x e d sulphite and copper cyanide at 4900 r p m and 60 ° C . Elec t ro ly te : 0.05 M C N " , 0.0167 M C u + ( C N : C u m o l e ratio = 3), 0.25 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 . 4000 Potent ia l v s . S C E / V Figure 8-3 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0167 M C u + and 0.4 M N a 2 S 0 3 , (2) the same compos i t ion as (1), the electrode coated w i t h copper oxide at 0.5 V vs. S C E for 10 minutes i n the same solut ion. (3) 0.4 M N a 2 S 0 3 , (4) 0.4 M N a 2 S 0 3 o n the electrode coated w i t h copper ox ide i n the same solut ion as (1), and (5) 0.4 M N a 2 S 0 3 o n the electrode coated w i t h copper ox ide f rom 0.05 M C N " and 0.0167 M C u + at 400 r p m and 60 ° C . Suppor t ing electrolyte: 0.25 M N a O H and 1 M N a 2 S 0 4 . 164 4000 0.0 0.2 0.4 0.6 0.8 1.0 Potent ia l v s . S C E / V Figure 8-4 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0167 M C u + ( C N : C u mo le ratio = 3) and 0.4 M N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 and (3) 0.05 M C N " and 0.0167 M C u + at 400 r p m and 60 ° C . Suppor t ing electrolyte: 0.25 M N a O H and 1 M N a 2 S 0 4 . 165 3500 0.2 0.4 0.6 0.8 1.0 Potential v s . S C E / V 4500 j 4000 --CN E 3500 --< 3000 --nsity 2500 nsity a •u 2000 | 1500 O 1000 500 4900 rpm . — 1600 rpm - - - 400 rpm . - -100 rpm 0.0 0.2 0.4 0.6 0.8 Potential vs. S C E / V 1.0 1.2 (a) 25 ° C (b) 40 ° C 5000 4500 CN 4000 E < 3500 3000 co c 2500 CD •D •4—1 2000 c a> 1500 3 o 1000 500 0 —4900 rpm - 1600 rpm . - 400 rpm . -100 rpm 0.0 0.2 0.4 0.6 0.8 Potential vs. S C E / V 6000 5000 CN E < 4000 CO c 3000 CD •o 4900 rpm — 1600 rpm _ 400 rpm - -100 rpm 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V (c) 50 ° C (d) 60 ° C F igure 8-5 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 0.05 M C N " , 0.0167 M C u + ( C N : C u mole ratio = 3), 0.05 M N a O H , 0.4 M N a 2 S 0 3 and 1 M N a 2 S 0 4 . 166 3500 3000 --0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential vs. S C E / V Figure 8-6 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0167 M C u + ( C N : C u mo le ratio = 3) and 0.4 M N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 and (3) 0.05 M C N " and 0.0167 M C u + at 400 r p m and 60 ° C . Support ing electrolyte: 0.05 M N a O H and 1 M N a 2 S 0 4 . 3500 -r 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential vs. S C E / V Figure 8-7 Po la r i za t ion curves for (1) 0.05 M C N " , 0.0125 M C u + and 0.4 M N a 2 S 0 3 , (2) 0.4 M N a 2 S 0 3 , (3) 0.05 M C N " , 0.0125 M C u + and 0.2 M N a 2 S 0 3 , (4) 0.2 M N a 2 S 0 3 and (5) 0.05 M C N " and 0.0125 M C u + at 400 r p m and 60 ° C . Support ing electrolyte: 0.05 M N a O H and 1 M N a 2 S 0 4 . 167 8.2.2 Anod ic Behaviour of Concentrated Copper Cyan ide Solution with Sulphite The polar iza t ion curves for the solut ion w i t h 3 M C N " , 1 M [Cu + ] and 0.5 M N a 2 S 0 3 and 0.25 M N a O H are shown i n F igure 8-8. A t 25 ° C , the current first increased and then decreased s l ight ly w i t h the precipi ta t ion o f copper oxide . A t a potential > about 0.52 V vs. S C E , the current increased to a peak value and decreased rapid ly . The second pass iva t ion is probably due to oxygen adsorption. A t 40 ° C , the polar iza t ion curves at 400 and 1600 r p m were s imi la r to those at 25 ° C . Howeve r , at 100 rpm, the current reached a l i m i t i n g value and became independent o f potential . The oxide formed i n the potential range 0.38 to 0.5 V vs. S C E was d i sso lved w h e n the current was at its l i m i t i n g value. T h i s is w h y the current d i d not decrease w i t h potential after the current sharply increased to a l i m i t i n g value. T h i s dependence o f the anodic behaviour on the rotational speed is related to the compos i t i on o f the reactive species on the surface o f the electrode. A t a potential > about 0.5 V vs. S C E , the current increased sharply w i t h increasing potential and was almost independent o f the rotational speed. Therefore the concentrat ion o f hydrox ide o n the electrode surface decreased w i t h decreasing rotational speed. A t 100 rpm, the concentrat ion o f hydrox ide was so l o w that the format ion o f copper oxide was not favored. E v e n copper ox ide was more readi ly reduced by sulphite ions and d isso lved . Therefore a second pass ivat ion was not observed. A t 400 and 1600 rpm, the concentrat ion o f hydrox ide o n the surface was s t i l l h i g h and the format ion o f copper oxide was s t i l l favored. W i t h increas ing potential , the second pass iva ion appeared probably due to the adsorption o f oxygen . A t 50 ° C and 100 rpm, the current increased cont inuously to a l i m i t i n g value and no copper oxide was formed on the electrode. A t 400 rpm, the anodic behaviour o f current vs . potential was s imi la r to that at 100 r p m and 40 ° C . A t 1600 rpm, the anodic behaviour was s t i l l s imi l a r to that at 50 ° C . A t 60 ° C and a rotational speed o f 100 or 400 rpm, the current increased cont inuously to a l i m i t i n g value and became independent o f the potential . The anodic behaviour at 1600 r p m was s imi la r to that at 100 r p m and 40 ° C . The anodic ox ida t ion o f sulphite and cyanide increases w i t h increasing temperature m u c h faster than the dif fus ion o f the hydroxide ion . Therefore even at a higher rotating speed, the hydrox ide concentrat ion on the surface o f anode is so l o w that no copper hydrox ide was formed and the current reached a l i m i t i n g value. 168 Figure 8-9 shows the plots o f the current vs. t ime at different potentials at 400 r p m . A t 25 ° C and 0.4 V vs. S C E , the current first decreased rap id ly and then s l o w l y and finally became stable. A th in layer o f copper oxide was precipitated on the anode. A t 0.60 V vs . S C E , the current increased to a certain value and then became stable. N o copper ox ide was precipi tated on the electrode. A t 0.80 V vs. S C E , the current decreased to a l i m i t i n g value and became stable w i t h no copper oxide appearing on the anode. It should be noted that at 25 ° C and 0.80 V vs. S C E , the current densities i n F igure 8-9a do not match those i n F igure 8-8a. T h i s can be expla ined by: (1) the current obtained i n F igure 8-9a was obtained us ing the cont ro l led potential method. W h e n the potential was appl ied, the instantaneous current reached a value where the concentration o f hydroxide on the electrode surface was l o w so that copper ox ide was not formed and the current was s tabi l ized at a l i m i t i n g value; (2) the current i n F igure 8-8a was generated by a potential scan at 1 m V s"1 and so the current never reached a value at w h i c h copper oxide was readi ly reduced and d isso lved . Hence it passivated the electrode surface. A t 40 ° C (Figure 8-9b), the results are s imi la r to those at 25 ° C (8-9a). A t 50 ° C and 0.3 V vs. S C E , the current decreased and became stable. A t 0.4 V vs. S C E , the current density increased and then decreased and was f ina l ly s tabi l ized. The current was m u c h higher than those i n Figures 8-8 b. The reason for this behaviour is the same as discussed for 25 ° C . A t a potential > 0.60 V vs. S C E , the current is the same as that obtained us ing a potential scan rate o f 1 m V s"1. A t 60 ° C and 0.2 or 0.3 V vs. S C E , the current decreased, then increased to a certain value and was stabil ized. A t a potential > 0.4 V vs. S C E , the current decreased or increased to a l i m i t i n g value and became stable. The current i n F igure 8-9d is the same as that i n F igure 8-8d. The precipi ta t ion o f copper oxide affected the anodic behaviour . Hence the concentrat ion o f hydrox ide was decreased to investigate the effect o f p H o n the anodic behaviour. F igures 8-10 and 8-11 show the polar iza t ion curves for the solu t ion conta in ing 0.1 and 0.05 M N a O H . The passivat ion d i d not appear because there was no precipitate o n the anode. F igure 8-12 shows the polar iza t ion curves for m i x e d sulphite and copper cyanide solut ion, sulphite so lu t ion and copper cyanide solut ion w i t h 0.25 M N a O H at 60 ° C . Sulphi te ox ida t ion appears to be catalyzed by copper cyanide oxida t ion . The ox ida t ion o f copper 169 cyanide also seems to be affected by sulphite. F r o m Figure 8-13, w h e n the hydrox ide concentrat ion decreased to 0.05 M , the ox ida t ion o f copper cyanide and sulphite was s ignif icant ly cata lyzed by each other. The increase i n sulphite concentrat ion f rom 0.2 to 0.4 M resulted i n the increase i n the current. H o w e v e r , its l i m i t i n g value is m u c h smal ler than that expected f rom the increase i n the concentration poss ib ly because the p H o n the electrode was so l o w that the speciat ion o f sulphite shifted f rom S 0 3 2 " to H S 0 3 " and S 0 2 w h i c h were less active. W h e n the concentrat ion o f cyanide increased f rom 3 to 3.5 M and the concentrations o f the other species were maintained constant, the polar iza t ion curves (Figure 8-14) became different and no pass ivat ion was observed. A t 25 ° C , the current increased w i t h increasing potential and then reached a l i m i t i n g value and became independent o f the potent ial . A t 40 , 50, and 60 ° C , there was no l i m i t i n g current and no passivat ion. A t 50 and 60 ° C , w h e n the potential exceeded 0.4 V vs. S C E , a significant amount o f bubbles was observed at 100 and 400 rpm. The bubbles were rapid ly d isso lved i n two seconds after turning o f f the current. The graphite was not corroded. A t such a h igh current, sulphite on ly l imi t ed a part o f the cyanide ox ida t ion and p H o n the electrode surface was so l o w that the rate o f the p roduc t ion o f ( C N ) 2 was higher than the rate o f the reaction between ( C N ) 2 and O H " . Therefore ( C N ) 2 bubbles were evo lved . F igure 8-15 shows the polar iza t ion curves for m i x e d sulphite and copper cyanide solut ion, copper cyanide and sulphite. The current for m i x e d copper cyanide and sulphite is higher than that for copper cyanide or sulphite. So the ox ida t ion o f both sulphite and copper cyanide contr ibuted to the total anodic current. F igure 8-16 shows the polar iza t ion curves for the solu t ion w i t h 4 M C N " , 1 M C u + , 0.5 M N a 2 S 0 3 and 0.25 M N a O H . The current increased cont inuously w i t h increasing potential . W h e n the current exceeded a certain value (depending o n the rotat ional speed), a layer o f bubbles was formed on the graphite. W i t h increasing potential , the bubbles became larger and had a significant effect o n the mass transfer. Thus the current increased s ignif icant ly . D u e to the format ion o f the bubble layer, the I R drop was even larger than 1 V . F igure 8-17 shows the polar iza t ion curves for m i x e d sulphite and copper cyanide solut ion, copper cyanide and sulphite. The current for m i x e d copper cyanide and sulphite was 170 a litt le higher than that for copper cyanide. So the ox ida t ion o f sulphite d i d not contribute very m u c h to the total anodic current. 900 800 1" 700 E < 600 I 5 0 0 c * 400 S 300 O 200 100 0 1600 rpm . — 400 rpm . . . 100 rpm 0.0 2500 0.4 0.6 0.8 1.0 Potential v s . S C E / V 0.4 0.6 0.8 Potential vs. S C E / V 1.2 (a) 25 ° C (b) 40 ° C 6000 5000 CN E < 4000 3000 i-(0 c T3 § 2000 3 1000 0 -I t — . 1600 rpm t 1 400 rpm II . . . . 100 rpm /J fl 1 I 0.0 0.2 0.4 0.6 0.8 Potential vs. S C E / V 1.0 18000 16000 4 CM 14000 ^ 12000 £< 10000 1.2 1600 rpm 400 rpm . . . . 100 rpm 0.2 0.4 0.6 0.8 1 Potential v s . S C E / V 1.2 (c) 50 ° C (b) 60 ° C Figure 8-8 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 3 M C N " , 1 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 . 171 E < 3500 3000 2500 2000 in c "O 1500 • c P 500 0.4 VSCE 0.6 VSCE 0.8 VSCE 50 100 Time / s 150 200 6000 5000 CM E < 4000 Hsu 3000 TJ C 2000 3 o 1000 0 50 -0.4 V vs. SCE -0.6 V vs. SCE • 0.8 V vs. SCE 100 Time Is 150 200 (a) 25 ° C 400 rpm 100 Time / s 200 10000 9000 ^ 8000 J 7000 ^ 6000 C 5000 4000 3000 2000 1000 0 (b) 40 ° C 0.2 V vs. SCE 0.3 Vvs.SCE 0.4 Vvs.SCE 0.5 V vs. SCE 0.6 V vs. SCE 0.8 V vs. SCE 100 200 Time Is 300 400 (c) 50 ° C (d) 60 ° C Figure 8-9 Current density vs. t ime at constant potential , 400 r p m and different temperatures. Elec t ro ly te : 3 M C N " , 1 M C u + , 0.25 M N a O H , 0.5 M N a 2 S 0 3 . 172 9000 8000 «* 7000 -E < 6000 --£ 5000 --in c ® 4000 § 3000 j O 2000 1000 -j-0 0.0 -4900 rpm 1600 rpm . 400 rpm -100 rpm 12000 0.2 0.4 0.6 O.J Potential vs. S C E / V 10000 E < in c TJ O 8000 S 4000 2000 0.0 -4900 rpm 1600 rpm . 400 rpm -100 rpm 0.2 0.4 0.6 Potential vs. S C E / V (a) 25 ° C (a) (b) 40 ° C 25000 0.2 0.4 0.6 Potential vs. S C E / V 0.2 0.4 0.< Potential vs. S C E / V (c) 50 ° C (d) 60 ° C Figure 8-10 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 3 M C N ' , 1 M C u + , 0.1 M N a O H , 0.5 M N a 2 S 0 3 . 173 8000 -7000 -<N 'E 6000 -< 5000 -in c 4000 -01 •D 3000 . C £ 3 2000 -o 1000 -0 --4900 rpm 1600 rpm . 400 rpm -100 rpm 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 -4900 rpm 1600 rpm . 400 rpm -100 rpm 0.4 0.6 Potential vs. S C E / V 1.0 0.2 0.4 0.6 O.t" Potential v s . S C E / V 1.0 (a) 25 ° C (b) 60 ° C 20000 18000 16000 14000 12000 10000 8000 4-6000 4000 2000 0 0.0 4900 rpm 1600 rpm . . . .400 rpm — - -100 rpm 0.2 0.4 0.6 0.£ Potential vs. SCE / V (c) 50 ° C (d) 60 ° C F igure 8-11 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 3 M C N " , 1 M C u + 0.05 M N a O H , 0.5 M N a ^ . 174 7000 -, Figure 8-12 Po la r i za t ion curves for (1)3 M C N " , 1 M C u + ( C N : C u mo le ratio = 3), 0.25 M N a O H and 0.5 M N a 2 S 0 3 , (2) 0.5 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 and (3) 3 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 4 at 400 r p m and 60 ° C . F igure 8-13 Po la r i za t ion curves at 400 r p m and 25 ° C for (1) 3 M C N " + 1 M C u + + 0.4 M N a 7 S 0 3 + 0.1 M N a 2 S 0 4 (2) 3 M C N " + 1 M C u + + 0.2 M N a 2 S 0 3 + 0.3 M N a 2 S 0 4 , (3) 0.4 M N a 2 S 0 3 + 1 M N a 2 S 0 4 , (4) 0.2 M N a 2 S 0 3 + 1 M N a 2 S 0 4 (5) 3 M C N " + 1 M C u + + 0.5 M N a 2 S 0 4 at [ N a O H ] = 0.05 M N a O H . 175 -1600 rpm 400 rpm . 100 rpm 0.0 0.2 0.4 0.6 0.8 1.0 Potential vs. S C E / V E < in c TJ c £ 3 o 1.2 30000 25000 20000 15000 10000 5000 1600 rpm 400 rpm . . . . 100 rpm 0.0 0.2 0.4 0.6 O.J Potential vs. S C E / V 1.0 < 20000 4-0.0 (a) 25 ° C I ' / / 1600 rpm / / 400 rpm / / . . . . 100 rpm / / / / / > / /" 1 J 1 • / / 1 / / f yt . -• 0.2 0.4 0.6 Potential vs. S C E / V 30000 0.8 (b) 40 ° C 0.2 0.4 0.6 Potential vs. S C E / V (c) 50 ° C (d) 60 ° C F igure 8-14 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 3.5 M C N " , 1 M C u + , 0.25 M N a O H , 0.5 M N a ^ . 176 Figure 8-15 Po la r i za t ion curves for (1)3.5 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 3 , (2) 3 M C N " , 1 M C u + , 0.25 M N a O H and 0.5 M N a 2 S 0 4 and (3) 0.5 M N a 2 S 0 3 , 0.25 M N a O H at 400 r p m and 60 ° C . 177 30000 , 25000 4-CN E < 20000 --i;su 15000 TJ C 0) 10000 --L_ 3 o 5000 o -J 1600 rpm 400 rpm . . . . 100 rpm 0.0 0.2 0.4 0.6 Potential vs. S C E / V 30000 , 25000 4-<N E < 20000 Pr. i;su 15000 --TJ C 0) 10000 --k_3 o 5000 - -0.8 1600 rpm 400 rpm - . . . 100 rpm 0.2 0.4 0.6 Potential vs. S C E / V 0.8 (a) 25 ° C (b) 40 ° C 30000 , 25000 E < ID C a TJ <*« C 3 o 20000 4-15000 10000 5000 - 1600 rpm 400 rpm -100 rpm 0.2 0.4 0.6 Potential vs. S C E / V 30000 25000 CN E < 20000 i« I 15000 a> TJ g 10000 o 5000 1600 rpm 400 rpm . . . . 100 rpm 0.8 0.2 0.4 0.6 Potential vs. S C E / V 0.8 (c) 50 ° C (d) 60 ° C Figure 8-16 Po la r i za t ion curves at different temperatures. Elec t ro ly te : 4 M C N " , 1 M C u + 0.25 M N a O H , 0.5 M N a 2 S 0 3 . 178 30000 -, CM 25000 - 2 E < 20000 3 £ 15000 -CD T3 c 10000 -O 5000 -0 0 0.2 0.4 C Potential vs. S C E / V 0.6 0.8 Figure 8-17 Po la r i za t ion curves for (1) 4 M C N " + 1 M C u + + 0.25 M N a O H + 0.5 M N a 2 S 0 3 , (2) 0.5 M N a 2 S 0 3 + 0.25 M N a O H + 1 M N a 2 S 0 4 and (3) 4 M C N " + 1 M C u + + 0.25 M N a O H + 0.5 M N a 2 S 0 4 at 400 r p m and 60 ° C . 8.3 Coulometric Measurements The coulometr ic measurements were conducted us ing cont ro l led potential and cont ro l led current methods to investigate the anodic current eff iciencies o f cyanide and copper ox ida t ion i n the presence o f sulphite. The results obtained us ing the cont ro l led potential method are l is ted i n Table 8-1. Tests 1-4 show the anodic current efficiencies o f cyanide and copper for the so lu t ion w i t h 0.05 M C N " , 0.0167 M ( C N : C u mole ratio = 3), 0.4 M N a 2 S 0 3 , 0.25 M O H " and 1 M N a 2 S 0 4 . In the presence o f 0.4 M N a 2 S 0 3 , the current eff ic iency decreased f rom 86 % to about 10 % for cyanide ( C N " -> C N O " ) and f rom 13 % to about 3 % for copper ( C u + -> C u O or C u ( O H ) 2 ) . Th i s means that sulphite can effectively l i m i t the anodic ox ida t ion o f copper cyanide. The anodic current eff iciency for cyanide at 0.5 V vs. S C E was a li t t le bit higher than that at 0.3 V vs. S C E because at 0.5 V S C E , the current for sulphite and copper cyanide was closest to a l i m i t i n g value and sulphite was less efficient i n l i m i t i n g the ox ida t ion o f copper cyanide. The anodic current eff ic iency at 60 ° C is a s l ight ly higher than that at 50 ° C . Tests 5-8 show the anodic current efficiencies o f cyanide and copper for the so lu t ion w i t h 0.05 M C N " , 0.0125 M ( C N : C u mole ratio = 4), 0.4 M N a 2 S 0 3 , 0.25 M O H " and 1 M N a 2 S 0 4 . In the presence o f 0.4 M N a 2 S 0 3 , the anodic current eff ic iency decreased f rom 90 % 179 to about 12 % for cyanide and f rom 10 % to about 2 % for copper (I). S i m i l a r to Tests 1-4, the anodic current efficiencies o f copper cyanide at 0.60 V vs. S C E are s l ight ly higher than those at 0.4 V vs. S C E . The anodic current eff ic iency o f cyanide i n Tests 5-8 is higher than that i n Tests 1-4, the anodic current eff ic iency o f copper (I) i n Tests 5-8 is lower than that i n Tests 1-4 poss ib ly because the speciat ion o f copper cyanide shifted to C u ( C N ) 4 3 " and more free cyanide was present i n the solut ion. Table 8-1 Current efficiencies f rom copper cyanide us ing contro l led potential cou lomet r ic measurements (supporting electrolyte: 1 M N a 2 S 0 4 ) Test Composition Controlled Temperature Rotational Current Current No. potential ( ° Q speed efficiency efficiency (V vs. SCE) (rpm) for CN (%) for Cu (%) 1 0.05 M CN", 0.3 50 100 9 2 2 CN:Cu = 3 0.3 60 100 8 2 3 0.4 MS0 3 2 - 0.5 50 100 12 3 4 0.25 M OH- 0.5 60 100 13 3 5 0.05 M CN", 0.4 50 100 11 1 6 CN:Cu = 4 0.4 60 100 12 1 7 0.4 M S032- 0.6 50 100 13 2 8 0.25 M OH- 0.6 60 100 14 2 Table 8-2 lists the anodic current efficiencies o f cyanide and copper (I) for the solutions w i t h different compos i t ion us ing the contro l led current method. Tests 1-6 l ist the anodic current eff iciencies o f cyanide and copper for the solut ion w i t h 0.05 M C N " , 0.0167 M ( C N : C u = 3), 0.4 M N a 2 S 0 3 , 0.25 M O H " and 1 M N a 2 S 0 4 at different current densities and rotat ional speeds. The rotational speeds (100-1600 rpm) and the current densities (250- 500 A m" 2) do not s ignif icant ly affect the current efficiencies o f cyanide and copper (I). S i m i l a r results for the solutions w i t h 0.05 M C N " , 0.0125 M C u + , 0.4 M N a 2 S 0 3 , 0.25 M N a O H and 1 M N a 2 S 0 4 (Tests 7-12) were obtained. The difference is that almost no copper ox ide was formed on the anode due to the h igh mole ratio o f cyanide to copper. The current ef f ic iency o f cyanide d i d not change very m u c h when the cyanide concentrat ion was increased f rom 0.05 M to 0.4 M (Tests 13-16) and 1 M (Tests 17-20) and the concentrations o f the other species were kept constant. F r o m Tests 1-8 and Tests 21-26, at C N : C u mole ratio = 3, w h e n the concentrat ion o f hydrox ide decreased f rom 0.25 M to 0.05 M and the concentrations o f the other species were kept at constants, the anodic current eff ic iency o f cyanide decreased w h i l e the anodic current eff ic iency o f copper (I) decreased to almost zero. T h i s means that sulphite was more efficient 180 i n l i m i t i n g the ox ida t ion o f copper cyanide at l o w hydroxide concentration. H o w e v e r , at C N : C u mole ratio = 4, the decrease i n the concentrat ion o f hydrox ide f rom 0.25 to 0.05 M d i d not affect the anodic current eff iciency. T h i s m a y be related to the d is t r ibut ion o f copper cyanide species. Table 8-2 Current efficiencies f rom copper cyanide us ing cont ro l led current coulomet r ic measurements (supporting electrolyte: 1 M N a 2 S 0 4 ) Test Composition Controlled current Temperature Rotational Current Current No. (A m"2) ( ° Q speed efficiency efficiency (rpm) for CN (%) for Cu (%) 1 0.05 M CN\ 250 50 1600 13 3 2 CN:Cu = 3 250 60 1600 10 2 3 0.4 MS0 3 2 - 250 50 100 14 2 4 0.25 M OH" 250 60 100 11 3 5 500 50 100 13 3 6 500 60 100 12 3 7 0.05 M CN\ 250 50 1600 14 0 8 CN:Cu = 4 250 60 1600 11 0 9 0.4 M S O 3 2 - 250 50 100 13 0 10 0.25 M OH- 250 60 100 11 0 11 500 50 100 15 -12 500 60 100 12 -13 0.4 M CN", 500 50 100 14 0 14 CN:Cu = 3 0.4 M S O 3 2 -0.25 M OH" 500 60 100 9 15 0.4 M CN", 500 50 100 17 0 16 CN:Cu = 4 0.4 M S032-0.25 M OH-500 60 100 15 17 1 M CN", CN:Cu 500 50 100 15 2 18 = 3 0.4 M S O 3 2 -0.05 M OH" 500 60 100 10 0 19 1 M CN", CN:Cu 500 50 100 18 20 = 4 0.4 MSCy2-0.25 M OH" 500 60 100 14 21 0.05 M CN", 250 50 1600 10 22 CN:Cu = 3 250 60 1600 6 23 0.4 M S O 3 2 - 500 50 1600 9 24 0.05 M OH" 500 60 1600 8 25 500 50 100 8 26 500 60 100 7 27 0.05 M CN", 250 50 1600 14 0 28 CN:Cu = 4 250 60 1600 13 0 29 0.4 M S O 3 2 - 500 50 1600 14 0 30 0.05 M OH" 500 60 1600 12 0 Table 8-3 lists the anodic current efficiencies o f cyanide and copper w h e n the cyanide concentrat ion was increased to 3 or 4 M . F r o m Tests 1-4, i n the presence o f 0.5 M N a 2 S 0 3 , at [CN"] = 3 M and [Cu + ] = 1 M , the anodic current eff ic iency o f cyanide decreased to around 12 % f rom about 82 % i n the absence o f sulphite and the anodic current ef f ic iency o f copper (I) decreased to about 2.5 % from 13.6 % i n the absence o f sulphite. Thus sulphite can l i m i t the ox ida t ion o f copper cyanide. The decrease i n the concentrat ion o f hydrox ide f rom 0.25 to 0.10 M (Tests 5-8) or 0.05 M (Tests 9-12) resulted i n a slight decrease i n the anodic current eff ic iency. There was no precipi ta t ion o f copper ox ide w h e n the other composi t ions were the same. So sulphite more eff icient ly l imi t s the ox ida t ion o f copper cyanide at a l o w p H . F r o m Tests 1-4 and Tests 13-24, the current eff ic iency o f cyanide increased w i t h increas ing concentrat ion o f cyanide. F r o m Tests 9-12 and 25-28, the anodic current e f f ic iency for cyanide increased by about 7-8 % w i t h decreasing sulphite concentrat ion f rom 0.5 to 0.3 M . Table 8-3 Current efficiencies f rom copper cyanide us ing contro l led current coulomet r ic measurements at 100 r p m (0.5 M N a 2 S 0 4 ) Test Composition Controlled Temperature Current efficiency Current efficiency No. current (A m"2) (°C) for CN (%) for Cu (%) 1 3 M CN", l M C u + 250 50 13 2.4 2 (CN:Cu = 3) 250 60 12 2.6 3 0.5 M S032" 500 50 12 2.0 4 0.25 M OH 500 60 11 2.5 5 3 M CN", l M C u + 250 50 13 0 6 (CN:Cu = 3) 250 60 12 0 7 0.5 M S032" 500 50 12 0 8 0.10 MOH" 500 60 11 0 9 3 M CN", l M C u + 250 50 10 0 10 (CN:Cu = 3) 250 60 9 0 11 0.5 M S032" 500 50 12 0 12 0.05 M OH" 500 60 11 0 13 3.2 MCN", l M C u + 250 50 13 0 14 (CN:Cu = 3.2) 250 60 12 0 15 0.5 M S032" 500 50 15 0 16 0.25 M OH" 500 60 14 0 17 3.5 MCN", l M C u + 250 50 19 0 18 (CN:Cu = 3.5) 250 60 18 0 19 0.5 M S032" 500 50 22 0 20 0.25 M OH 500 60 21 0 21 4 MCN", l M C u + 250 50 40 0 22 (CN:Cu = 4) 250 60 39 0 23 0.5 M S032" 500 50 45 0 24 0.25 M OH" 500 60 46 0 25 3 MCN", l M C u + 250 50 17 0 26 (CN:Cu = 3) 250 60 16 0 27 0.3 M S032" 500 50 18 0 28 0.05 M OH" 500 60 16 0 182 In the above tests, the anodic current efficiencies o f cyanide and copper (I) were obtained f rom the analysis o f the cyanide concentration and the amount o f the copper ox ide precipi tated o n the anode. F r o m the anodic current eff iciencies o f cyanide and copper (I), we cannot predict the amount o f o x i d i z e d sulphite because sulphite can be o x i d i z e d to sulphate (two electrons process) and dithionate (one electron process) and there are possible side reactions such as o x y g e n evolut ion. Therefore the amount o f o x i d i z e d sulphite was determined di rect ly . Table 8-4 lists the anodic current efficiencies o f cyanide , copper (I) and sulphite (assuming sulphite was o x i d i z e d to sulphate). F r o m Table 8-4, the sum o f the anodic current eff iciencies o f cyanide , copper (I) and sulphite is very close to 100 % and so sulphite was o x i d i z e d to sulphate. Table 8-5 lists the current eff iciency for cyanide , copper and sulphite us ing cont ro l led potential method. A t a higher potential , the current was at a l i m i t i n g value and the anodic current eff ic iency for sulphite was l o w . Table 8-4 Current eff ic iency for copper cyanide us ing control led current coulomet r ic measurements (supporting electrolyte: 1 M N a 2 S 0 4 for Tests 1 and 2) at 100 r p m * Test Composition Controlled Temperature Current Current Current No. current (°Q. efficiency efficiency efficiency (A m"2) for S032" (%) for CN (%) for Cu + (%) 1 0.05 M CN", 500 50 86 12 1.6 2 0.0167 M C u + 500 60 89 10 1.8 0.4 M S032-0.25 M OH" 3 3 M C N \ 1 M 250 50 83 14 2.2 4 Cu + (CN:Cu = 3) 250 60 86 13 2.4 5 0.5 M SCy 500 50 84 15 2.5 6 0.25 M OH" 500 60 86 12 2.6 7 3 M C N \ 1 M 250 25 88 15 0 8 Cu + (CN:Cu = 3) 250 40 88 14 0 9 0.5 M S032" 250 50 89 10 0 10 0.05 M OH" 250 60 87 09 0 * F o r Tests 3-10, the in i t i a l concentration o f sulphite was 0.6 M . The amount o f e lectr ic i ty passed decreased the concentration o f sulphite to 0.4 M assuming 85 % for the anodic current eff ic iency o f sulphite. 183 Table 8-5 Current eff ic iency for copper cyanide us ing control led potential cou lomet r ic measurements (supporting electrolyte: 1 M N a 2 S 0 4 for Tests 1 and 2) at 100 r p m * Test Composition Controlled Temp. Current Current Current No. potential (°C) efficiency efficiency efficiency (V vs. SCE) for CN (%) for Cu + (%) for S032" (%) 1 0.05 MCN", 0.0167 M C u + 0.3 60 11 2.2 87 2 0.4 M S032" 0.25 M OH" 0.5 60 13 2.9 84 3 0.05 MCN", 0.0125 M C u + 0.4 60 10 1.3 85 4 0.4 M S032" 0.25 M OH" 0.6 60 13 2.8 83 5 0.05 MCN", 0.0167MCu + 0.25 60 9 0 89 6 0.4 M S032" 0.05 M OH- 0.6 60 11 0 90 9 0.05 M C N , 0.0125 M C u + 0.25 60 11 0 89 10 0.4 M SO,2" 0.05 M OH" 0.6 60 13 0 88 11 3 M CN", 1 M Cu + 0.3 60 13 2.3 87 12 0.5 M S0 3 2 \ 0.25 M NaOH 0.6 60 52 0 47 13 3 M CN", 1 M Cu + 0.3 60 11 0 87 14 0.5 M S032", 0.05 M NaOH 0.6 60 48 0 57 F o r Tests 11-14, the in i t i a l concentration o f sulphite was 0.6 M . The amount o f electr ic i ty passed decreased the concentration o f sulphite to 0.4 M assuming 85 % for the anodic current eff ic iency o f sulphite. 8.4 Possible Anodic Reactions The anodic behaviour o f m i x e d copper cyanide and sulphite so lu t ion is a funct ion o f hydrox ide , sulphite and cyanide concentrations, the mole ratio o f cyanide to copper, temperature and rotational speed. The current for m i x e d copper cyanide solu t ion was not just the sum o f the currents o f copper cyanide and sulphite when they are present separately i n the solut ion. Sulphi te ox ida t ion was affected s ignif icant ly by the ox ida t ion o f copper cyanide. Copper cyanide ox ida t ion was also affected by sulphite ions. C o m p a r i n g F igures 8-12, 15 and 17, the higher the mole ratios o f cyanide to copper , the less the effect o n the ox ida t ion o f copper cyanide and sulphite. Th i s may be related to the dis t r ibut ion o f copper cyanide species. P robab ly the discharge o f C u ( C N ) 4 3 " is less affected by sulphite. So sulphite also has a smaller effect o n the ox ida t ion o f cyanide. The ox ida t ion o f C u ( C N ) 3 2 " is more affected by sulphite. So sulphite has a greater effect o n the ox ida t ion o f copper cyanide . One effect o f sulphite is to reduce the precipi ta t ion o f copper oxide and so affect the ox ida t ion o f copper cyanide. A t a concentrat ion o f hydroxide b e l o w a certain l eve l , sulphite comple te ly suppresses the precipi ta t ion o f copper oxide . Therefore the probable anode reactions are: C u ( C N ) „ - ( n - ° C u ( C N ) n " ( n - 2 ) + e (n= 2, 3, 4) (8-1) 184 S 0 3 2 " + 2 0 H " -> S 0 4 2 ~ + H 2 0 + 2e (8-2) 2 C u ( C N ) n " ( n " 2 ) + 2CN~ -> 2C«(C/V)„~ ( "~ 1 ) + (CN)2 (8-3) (CN)2 + 20H~ -> CN~ + CNOr + H20 (8-4) 2 C u ( C N ) n " ( n - 2 ) + S 0 3 2 " + 2 0 H -> 2 C u ( C N ) n ' ( n " 1 ) + S 0 4 2 " + H 2 0 (8-5) C u ( C N ) „ ~ ( n _ 2 ) + 2 0 H " C u ( C N ) n " ( n _ 1 ) + C u ( O H ) 2 ( o r C u O + H 2 0 ) (8-6) Reac t ion 8-5 m a y undergo the f o l l o w i n g steps s imi la r to the react ion between ferr icyanide and sulphite [337]: C u ( C N ) n - ( n - 2 ) + S 0 3 2 " -> C u ( C N ) n - ( n - 2 ) S 0 3 2 " (8-7) C u ( C N ) „ " ( n " 2 ) S 0 3 2 " + Cu(CN)x-°'-2) -> Cu(CN);{"-2)SO; + Cu(CN);("-l) (8-8) C u ( C N ) n - ( n - 2 ) S 0 3 " + 2 0 H ~ -> C u ( C N ) n " ( n _ , ) + S 0 4 2 " + H 2 0 (8-9) Reac t ion 8-1 is cata lyzed by sulphite ions w h e n n = 3. S 0 3 2 " m a y be bound to C u ( C N ) 3 2 " and fo rm C u ( C N ) 3 2 " S 0 3 2 ' w h i c h m a y be discharged faster than C u ( C N ) 3 2 \ So the ox ida t ion o f sulphite and copper cyanide is s ignif icant ly catalyzed. W i t h increasing mo le ratio o f cyanide to copper, the concentration o f C u ( C N ) 3 2 " is decreased and so it is less affected by sulphite. 8.5 Summary The anodic behaviour o f m i x e d sulphite and copper cyanide is not just the sum o f sulphite and copper cyanide w h e n they are present separately i n the solut ion. Sulphi te ox ida t ion is enhanced by the presence o f copper cyanide. The effect o f sulphite o n l i m i t i n g the ox ida t ion o f copper cyanide decreases w i t h increasing mo le ratio o f cyanide to copper. Th i s is related to the shift i n the discharged species f rom C u ( C N ) 3 2 " to C u ( C N ) 4 3 " w i t h increasing mo le ratio o f cyanide to copper. Sulphite ions affect the discharge o f C u ( C N ) 3 2 " more than that o f C u ( C N ) 4 3 " . Sulphi te is o x i d i z e d to sulphate. A t [Cu] = around 1 M , C N : C u mo le ratio = 3 -3.2, [OH"] = 0.05-0.25 M , [S0 3 2 "] = 0.4-0.6 M and the temperature = 50-60 ° C , the anodic current eff ic iency o f sulphite ox ida t ion reached 80-90%. The above condi t ions are suitable for 185 obtaining a good copper deposi t ion current eff ic iency and therefore w o u l d be suitable for industr ial appl ica t ion. 186 9. CONCLUSIONS The f o l l o w i n g are the p r inc ipa l conclus ions resul t ing f rom the study o f the thermodynamics o f copper cyanide, direct copper e lec t rowinning f rom dilute cyanide solut ion, copper e lec t rowinning f rom concentrated copper cyanide solu t ion us ing alternative anodic reactions, the anodic ox ida t ion o f sulphite, the anodic ox ida t ion o f copper cyanide and the anodic ox ida t ion o f m i x e d sulphite and copper cyanide solut ion. (1) The distr ibutions and the equ i l i b r ium potentials o f copper cyanide species, ca lcula ted us ing rel iable stabil i ty constants, are shown to be functions o f the mo le ratio o f cyanide to copper, total cyanide concentration, p H and temperature. W i t h increasing C N : C u mole ratio, the dis t r ibut ion o f copper cyanide species shifts more comple te ly to the h igh ly coordinated complex ( C u ( C N ) 4 3 " ) at a h i g h cyanide concentrat ion than that at a l o w cyanide concentration. W i t h increasing C N : C u mo le ratio, the e q u i l i b r i u m potential for Cu(I ) / C u decreases rapid ly at a C N : C u mole ratio < about 4 and more s l o w l y at a C N : C u mole ratio > about 4. Increasing p H is s imi la r to increas ing free cyanide concentration. Increasing temperature results i n decreasing the stabil i ty constants. Therefore the dis t r ibut ion o f copper cyanide shifts to the l o w l y coordinated complexes . The potential measurements have conf i rmed the va l id i ty o f the calculated results. In the p H - potential diagrams. C u C N , C u ( C N ) 2 \ C u ( C N ) 3 2 - and C u ( C N ) 4 3 ' can predominate i n the different p H regions. (2) The current eff ic iency o f copper deposi t ion o n a graphite felt electrode decreases w i t h increasing mo le ratio o f cyanide to copper. Due to the l o w conduct ivi t ies o f the so lu t ion and the graphite felt, the potential and current d is t r ibut ion o f copper throughout the 3-d imens iona l electrode are not un i form. The accumula t ion o f deposited copper o n the graphite felt as the p la t ing proceeds s ignif icant ly improves the conduct iv i ty o f the graphite felt and increases the specif ic surface area benefit ing copper deposit ion. Coppe r can be eff icient ly deposited o n the graphite felt f rom solut ions o f l o w concentrat ion (0.5 g -2 g L " 1 C u ) at a h igh mole ratio o f cyanide to copper ( C N : C u = 3-9). 187 The energy requirement for copper deposi t ion was as l o w as 1-2 k w h / k g C u (1000-2000 kwh/ tonne C u ) i n the current range 30-100 A m " 2 . These values compare favorably w i t h that obtained i n convent ional copper e lec t rowinning f rom sulphuric acid-copper sulphate solutions. (3) O f four sacr i f ic ia l species (sulphite, methanol , thiocyanate and ammonia) , on ly sulphite can effect ively l i m i t the ox ida t ion o f cyanide. W h e n the compos i t ion o f the electrolyte was cont ro l led at 50-60 g L " 1 Na2S03, 70 g L " 1 C u , C N : C u = 3-3.2, the anodic current e f f ic iency o f cyanide decreased f rom about 100 % to 10-20 % i n the current range 250-500 A m" 2 and the temperature range 50-60 ° C . Unde r the above condi t ions , the copper depos i t ion current eff ic iency was 90-96 % and the energy consumpt ion was 0.76-1.0 k W h / k g C u . The anodic current eff ic iency o f cyanide increased f rom about 15 % to 56 % w i t h increas ing C N : C u mo le ratio f rom 3 to 4.5 at [Cu] = 70 gL" 1 . W i t h increasing the current density, the anodic current eff ic iency o f cyanide decreases greatly at a current density > 500 A m" 2 and s l igh t ly at a current density < 500 A m" 2 . The anodic current eff ic iency o f cyanide decreases s l igh t ly w i t h increasing temperature. The copper deposi t ion current ef f ic iency decreases w i t h increasing C N : C u mole ratio and decreasing temperature. The presence o f thiocyanate increases the copper deposi t ion current eff ic iency at C N : C u mole ratio > 4.5. (4) A t l o w potentials ( e.g. < 0.25 V vs. S C E at 25 °C) , the react ion order for the ox ida t ion o f sulphite is b e l o w 1 and decreases w i t h increasing sulphite concentrat ion. The Tafe l slope is 0.060 -0.065 V decade" 1. A t h igh potentials (> 0.4 V vs. S C E ) , the react ion order w i t h respect to sulphite ions is 1 up to 0.4 M and the Tafe l slope is 0.19 - 0.21 V decade" 1. The react ion order w i t h respect to hydroxide ions is close to zero. The act ivat ion energy for the kinet ic current decreases f rom 85.2 kJmol" 1 at 0.2 V vs . S C E to 45.3 kJmol" 1 at 0.6 V vs. S C E . The dif fusion coefficients o f sulphite ions were 5.6, 8.6, 9.99 and 12.4 x 10" 1 0 m 2 s"1 respect ively for 25, 40 , 50 and 60 ° C . Sulphi te ox ida t ion i n a lkal ine solut ion appears to undergo a radical -e lec t ron mechan i sm. A t l o w potentials, the adsorbed sulphite ox ida t ion is dominant and at h i g h potentials, the sulphite ions are o x i d i z e d direct ly on the electrode surface. 188 (5) Copper has a significant catalytic effect on cyanide oxida t ion . A t l o w potentials ( roughly 0 to 0.4 V vs. S C E ) , cuprous cyanide is o x i d i z e d to cupric cyanide complexes w h i c h produce cyanogen, w h i c h i n turn reacts w i t h hydroxide to fo rm cyanate. In the midd le potential reg ion (roughly 0.4 to 0.6 V vs. S C E ) , cuprous cyanide is o x i d i z e d to cupr ic ox ide and cyanate. The potential for the precipi ta t ion o f copper oxide increases w i t h increasing mo le ratio o f cyanide to copper. In the h i g h potential region (> about 0.60 V vs. S C E ) , o x y g e n is evo lved at the electrode. H o w e v e r , w h e n the concentrat ion o f cyanide was h igh and the concentrat ion o f hydrox ide was l o w , no copper oxide was precipitated and but cyanogen gas was evo lved . The e lect rochemical k inet ic behavior is dependent o n C N : C u mo le ratio, p H and total cyanide concentrat ion. A t C N : C u = 3 and [OH"] = 0.25 M , the Tafe l slope is about 0.12 V decade" 1 and the react ion order w i t h respect to C u ( C N ) 3 2 " is one. C u ( C N ) 3 2 " is d ischarged o n the electrode. The current and Tafe l slope decrease w i t h decreasing hydrox ide concentrat ion and so hydrox ide is i n v o l v e d i n the rate-determining step. W i t h increasing mole ratio o f cyanide to copper, the anodic behaviour o f copper cyanide changes. W h e n the mole ratio o f cyanide to copper is larger than a certain value w h i c h depends o n the total cyanide concentration, e. g. about 4 at [ C N ] = 0.05 M and 3.5 at [CN"] = 3.5 M , a Tafe l slope o f about 0.06 V decade" 1 was observed over the potential range 0.1 - 0.25 V vs . S C E . A second Tafe l slope o f about 0.17 -0.20 V decade" 1 was noted over the higher potential range. T h i s change is related to the change i n the d is t r ibut ion o f copper cyanide species. The current is proport ional to the concentrat ion o f tetracyanide and almost independent o f the total cyanide concentration. p H has litt le effect on cyanide ox ida t ion and the Tafe l slopes do not change w i t h p H . In the potential r eg ion where copper oxide was precipitated, the current at a constant potent ial decreases w i t h decreasing hydroxide concentration and the ra te-control l ing step invo lves hydrox ide . The catalysis o f copper oxide is l im i t ed w i t h increasing copper cyanide concentrat ion and temperature. (6) The anodic behaviour o f m i x e d sulphite and copper cyanide is not just the sum o f sulphite and copper cyanide w h e n they are present separately i n the solut ion. Sulphi te ox ida t ion is ca ta lyzed by the ox ida t ion o f copper cyanide. It also affects the ox ida t ion o f copper cyanide . 189 The effect o f sulphite o n the oxida t ion o f copper cyanide decreases w i t h increasing mo le ratio o f cyanide to copper. Th i s is related to the shift i n the discharged species f rom C u ( C N ) 3 2 " to C u ( C N ) 4 3 " w i t h increasing mole ratio o f cyanide to copper. Sulphite ions affect the discharge o f C u ( C N ) 3 2 - more than that o f C u ( C N ) 4 3 \ A t [Cu] = around 1 M , C N : C u = 3 -3.2, [ O H ] = 0.05-0.25 M , [S0 3 2 "] = 0.4-0.6 M and the temperature = 50-60 ° C , the anodic current eff ic iency o f sulphite reached 80-90%. In relat ion to the recovery o f copper f rom cyanide go ld leach solut ion, it has been shown that i n the e lec t rowinning step, it is poss ible to l i m i t the ox ida t ion o f cyanide by us ing the ox ida t ion o f sulphite as an alternative anode react ion w i t h an electrolyte hav ing a compos i t ion s imi la r to that indicated above. A t a current densi ty o f 250 to 500 A m " 2 , copper can be e lec t rowon at a current eff ic iency o f 95 % w i t h a energy requirement o f about 0.8 k W h / k g C u . 190 10. RECOMMENDATIONS M a n y important aspects have not been investigated due to the t ime constraint. Regard ing the fundamental aspects and the pract ical appl ica t ion o f copper e lec t rownning , the f o l l o w i n g areas need to be studied i n the future. The morpho logy and dis tr ibut ion o f the copper deposit should be studied to better understand the effect o f the copper deposit on the copper deposi t ion current ef f ic iency. The B E T method needs to be used to measure the real surface area o f the graphite fibre w i t h and wi thout a copper deposit. The objective w o u l d be to understand h o w the copper deposits so as to improve the current eff iciency o f copper deposi t ion. Po la r i za t ion curves should be measured to better understand copper deposi t ion on the graphite f rom cyanide solut ion. The measurement should be carr ied out for current passing i n the same and opposite direct ions to the electrolyte f l ow. The effect o f temperature, C N : C u ratio, support ing electrolyte, deposi ted copper and f l o w rate should be studied. H y d r o g e n evo lu t ion o n the graphite fibre w i t h and wi thout deposited copper cou ld be studied us ing steady-state po lar iza t ion measurements. There are some reports on the effect o f thiocyanate [92, 94-96, 98] and sulphite [90] i n copper cyanide p la t ing baths. H o w e v e r , there is a lack o f fundamental w o r k about h o w thiocyanate and sulphite affect the copper deposi t ion process. The condi t ions used i n p la t ing m a y not be the same as those employed i n copper e lec t rowinning . F o r example , the deposi t ion t ime for e lec t rowinning is m u c h longer than that for p la t ing. Some phenomena occur r ing i n e lec t rowinning may not be observed i n plat ing. So the effect o f thiocyanate and sulphite o n the e lec t rowinning process needs to be studied careful ly. Comprehens ive experiments should be conducted us ing a small-scale p i lo t c e l l operating condi t ions to op t imize the e lec t rowinning condi t ion. H y d r o x i d e and sulphite catalyze the discharge o f C u ( C N ) 3 2 " on the anode. H o w e v e r , it is not clear h o w they affect the anodic ox ida t ion o f copper cyanide. It is possible that hydrox ide and sulphite are bound to C u ( C N ) 3 2 " . Spectroscopic studies such as R a m a n , U V , nuclear magnet ic resonance might be useful i n leading to an understanding o f the above phenomena. 191 11. REFERENCES 1. M. C. Jha, "Refractory of Certain Gold Ores to Cyanidation: Probable Causes and Possible Solutions", Mineral Processing and Extractive Metallurgy Review", 2, 1987, pp. 331-352. 2. H. V. Michaelis, "Gold Processing Update, The Prospects for Alternative Leach Reagents, Can Precious Metals Producers Get along without Cyanide?", Engineering & Mining Journal, 1987, pp. 43 - 47. 3. J. A. King, and D. B. Dreisinger, "Autoclaving of Copper Concentrates", Proc. Copper'95, edited by W. C. Cooper, D. B. Dreisinger, J. E. Dutrizac, H. Hein and G. Ugarte, Can. Inst. Min. Met., Montreal, 1995, pp. 511-534. 4. N. W. Hanf and C. G. Schmidt, "The Roasting of Witwatersrand Pyrite Concentrates", J. S. Afr. Inst. Min. Metall., Vol. 79, 1979, pp. 365 - 371. 5. J. O. Marsden, J. G. Mansanti and J. G. Sass, "Innovative Methods for Precious Metals Recovery in North America", Mining Engineering, Vol. 45, 1993, pp. 1144-1151. 6. C. D. McDoulett and G. W. Reschike, "Metal Leaching and Recovery Process", U.S. Patent 5364444, Nov. 15,1994. 7. R. Poulin and R. W. Lawrence, "Economic and Environment Niches of Biohydrometallurgy", Minerals Engineering, Vol. 9, pp. 1996, pp. 799 -810. 8. C. L. Brierley, "Gold and Copper Ore/Concentrate Bioleaching", Conference of Impurities Control and Disposal in Hydrometallurgical Processes", Toronto, 1994, Canadian Institute of Minning, Metallurgy and Petroleum (Canada), pp. 371-379. 9. G. Deschenes, "Literature Survey in the Recovery of Gold from Thiourea Solutions and the Comparison with Cyanide", C. I. M. Bulletin, Vol. 79, 1986, pp. 76-83. 10. R. Y. Wan, "Importance of Solution Chemistry for Thiosulphate Leaching Gold", Conference - World Gold '97, Australasian Institute of Mining and Metallurgy, 1997, pp. 159-162. n. A. Dadgar and J. Howarth, "Advances in Bromide Gold Leaching Technology", EPD Congress '92, edited by J. P. Hager, TMS, pp. 123 - 137. 12. S. Vukcevic, "The Mechanism of Gold Extraction and Copper Precipitation from Low Grade Ores in Cyanide Ammonia Systems", Minerals Engineering, Vol. 10, 1997, pp. 309-326. 13. J. S. Scott, "An Overview of Gold Mill Effluent Treatment", The Proceedings of Gold Mining Effluent Treatment Seminars", Vancouver, British Columbia, Feb. 15-16, 1989, pp. 1-22. 14. V. McNamara, "The AVR Process for Cyanide Recovery, and Cyanide Control for Barren Recycle and Barren Bleed", The Proceedings of Gold Mining Effluent Treatment Seminars, Vancouver, British Columbia, Feb. 15-16, 1989, pp.199-239. 15. N. L. Piret and H. J. Schippers, "Cyanide Destruction versus Cyanide Regeneration - Evaluation of the Processes for Optimum Mill Effluent", Extraction Metallurgy'89, London, UK, 10-13 July 1989, The Institution of Mining and Metallurgy, pp.1041 - 1080. 16. M. J. Kitney, "Cyanide Regeneration from Gold Tailings - Golconda Beaconsfield's Experience", The Proceedings of Gold Mining Effluent Treatment Seminars, Vancouver, British Columbia, Feb. 15-16, 1989, pp.389-399. 192 17. H. Soto, F. Nava and J. Jara, "Cyanide Regeneration and Copper Recovery from Cyanide Solutions", Global Exploration of Heap Leachable Gold Deposits , edited by D. M. Hausen, TMS, 1997, pp. 151-160. 18. A. Bergmann, A. Guzman and G. M. Potter, "Regeneration of Cyanide, A Case Study: Pilot Cyanidation of a High Copper Ore from Punitiaqui, Chile", Conference on Cyanide and the Environment, Tucson, Arizona, Dec. 1984, pp.457-467. 19. E. L. Coltrinari, "Method for Recovery of Cyanide from Waste Streams", U.S. Patent No. 4,708,804, November 24, 1987. 20. S.A. Patent 866,244 (1986), "A Process for Recovering Cyanide from Effluent Solutions". 21. G. W. Lower, "Electrolytic Recovery of Copper from Copper Cyanide Leaching Solutions", U.S. Patent 3,463,710, Aug. 26, 1969. 22. Orocon Inc Technology Report, "CELEC Cyanide Regeneration System". 23. ECO Corporation report, "ECO-Metal Recovery System". 24. HAS REACTORS Limited Technology Report, "Copper Recovery/Cyanide Regeneration". 25. D. G. Dickson, "Electrochemical Reactor for Copper Removal from Barren Solution", U. S. Patent 4,911,804, Feb. 1, 1988. 26. C. A. Fleming, W. G. Grot and J. A. Thorpe, "Hydrometallurgical Extraction Process", U. S. Patent, 5,411,575, May 2, 1995. 27. G. R. Maxwell, J. A. Thorpe, K. M. Schall and K. A. Brunk, "ACCS Technology from Lab to Feasibility Study", Global Exploration of Heap Leachable Gold Deposits , edited by D. M. Hausen, TMS, 1997, pp. 141-149. 28. D. B. Dreisinger, J. Ji, B. Wassink, and J. King, "The Solvent Extraction and Electrowinning Recovery of Copper and Cyanide Using XI 7950 Extractant and Membrane Cell Electrolysis, The Proceedings of Randol Gold Forum - Perth' 95, 1995, pp. 239-244. 29. A. G. Sharpe, "The Chemistry of Cyano Complexes of the Transition Elements", Academic Press, London, 1976. 30. W. H. Vigoe, "The Chemistry of Copper Cyanides", Eng. Min. J., Vol. 77(20), 1901, pp. 795-796. 31. S. Glasstone, "Studies of Electrolytic Polarization. Part VIII, Complex Cyanides: (b) Copper", J. Chem. Soc, Vo.132, 1929, pp. 702-713. 32. M. G. Vladimirova and I. A. Kakovsky, "The Physicochemical Constants Characteristic of the Formation and Composition of the Lowest Cuprous Cyanide Complex", Journal of Applied Chemistry of the USSR, Vol. 23, 1950, pp. 615 - 632. 33. R. A. Penneman and L. H. Jones, "Infrared Absorption of Aqueous Complex Ions II Cyanide Complexes of Cu (I) in Aqueous Solutions", J. Chem. Phys., Vol. 14 (2), 1956, pp. 295-296. 34. G. W. Chantry and R. A. Plane, "CN Stretching Bands in the Raman Spectra of Some Group lb and Group lib Complex Cyanides", J. Chem. Phys., Vol.33(2), 1960, pp. 736-740. 35. A. Brenner, "Determination of the Composition of Complexes and Their Instability Constants by Calorimetry 1. The Cuprocyanide Complexes", J. Electrochem. Soc, Vol. 112, 1965, pp. 611- 621. 193 36. D. Cooper and R. A. Plane, "Raman Study of Complex Cyanides of Copper(I)", Inorganic Chem., Vol. 5 (1), 1966, pp.16-20. 37. D. Cooper and R. A. Plane, "Mixed Complexes of Copper(I) Cyanide", Inorganic Chem., Vol. 5 (12), 12, 1966, pp.2209-2212. 38. R. M. Izatt, H. D. Johnston, G. D. Watt and J. J. Christensen, "Thermodynamics of Metal Cyanide Coordination VI. Copper(I)- and Silver(I)- Cyanide Systems", Inorganic Chemistry, Vol.6 (1), 1967, pp. 132-135. 39. E. A. Simpson and G. M. Waind, "The Ultraviolet Absorption Spectra and Stability Constants of Cuprous Cyanide Complexes", J. Chem. Soc, 1958, pp. 1746-1749. 40. J. S. Coleman, R. Gorge, L. Allaman and L. H. Jones, "Stepwise Formation of Cyanide Complexes of Copper (1) in Anion Exchangers", J. Phys. Chem., Vol. 72 (7), 1968, pp.2605-2608.\ 41. H. L. Noblitt, "Complex Ions of Copper and Cyanide", Mines Branch Research Report R268, Department of Energy, Mines and Resources, Ottawa, Canada, 1973. 42. H. P. Rothbaum, "The Composition of Copper Complexes in Cuprocyanide Solutions", J. Electrochem. Soc, Vol. 104 (10), 1957, pp. 862-686. 43. T. Yamamoto, H. Haraguchi and S. Fujiwara, "Copper Nuclear Magnetic Resonance Study of Cyanocuprate(I) Ions in Solution. Formation of Polynuclear Species and Mixed Complex", J. Phys. Chem., Vol. 74 (25), 1970, pp. 4369-4373. 44. P. Duby, "The Thermodynamic Properties of Aqueous Inorganic Copper Systems", The International Copper Research Association, New York, 1977, pp. 64-65. 45. G. G. Simulin, S. B. Lebed, V. D. Domashich and V. V. Kosenko, "IR Spectroscopic Investigation of a Ferrocyanide Electrolyte for Copper-Plating Steel", Elektrokhimiya, Vol. 8 (6), 1971, pp. 791-794. 46. R. D. Hancock, N. P. Finkelstein and A. Evers, "Stabilities of the Cyanide Complexes of the Monovalent Group IB Metal Ions in the Aqueous Solutions", J. Inorg. Nucl. Chem. Vol. 34, 1972, pp. 3747 - 3751. 47. R. Yu. Bek , B. D. Zhukov, "Electrodeposition of Copper from Cyanide Electrolytes Pt. II, Formation Constants for Copper Cyanide Complexes", IZV. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk, (4), 1973, pp. 52-56. 48. C. Kappenstein and R. Hugel, "Cyanocuprate (I) in Neutral and Acidic Aqueous Solution", J. Inorg. Nucl. Chem., Vol. 36, 1974, pp. 1821-1825. 49. R. Izatt, J. Christensen , R. Pack and R. Bench, "Thermodynamics of Metal -Cyanide Coordination. I. Pk, AH 0 , AS0 Values as a Function of Temperature for Hydrocyanic Acid Dissocation in Aqueous Solutions", Inorg. Chem., Vol.1, 1962, pp.828-831. 50. J. S. Solis, P. M. May and G. Hefter, "Cyanide Thermodynamics Part 4, Enthalpies and Entropies of Cyanide Complexation of Cu+, Ag +, Zn 2 + and Cd 2 + ", J. Chem. Soc. Faraday Trans., Vol. 92, 1996, pp. 641-644. 51. A. R. Taylor, M. H. Brown and E. G. Taylor, "Heat of Formation of Cuprous Cyanide and Its Heat Capacity from 10 to 400 K", U. S. Bureau of Mines, Report of Investigation 7499, 1972. 52. M. T. Beck, "Critical Survey of Stability Constants of Cyano Complexes", Pure & Appl. Chem., Vol. 59, 1987, pp. 1703-1720. 194 53. B. D. Zhokov, L. I. Borodikhina, V. M. Shchekotikhin, N. P. Poddubny and R. Yu. Bek, "Investigation of the Electrodeposition of Copper from Cyanide Electrolytes III Composition of Cuprocyanide Solutions", IZV. Sib. Otd. Akad. Nauk, (4), 1973, 1966, pp. 57-60. 54. I. Banyai, J. Blixi, J. Glaster and I. Toth, "On the Dissociation of Hydrogen Cyanide in Aqueous Solutions Containing Different Ionic Media. A Combined Potentiometric and Carbon-13 NMR Study", Acta Chem. Scand., Vol. 46, 1992, pp. 138-141. 55. J. S. Solis, P. M. May and G. Hefter, "Cyanide Thermodynamics Part 3, Enthalpies and Entropies of Ionization of Water and Hydrogen Cyanide", Aust. J. Chem., Vol. 49, 1996, pp. 651-657. 56. V. Gaspar and M. T. Beck, " The Influence of the Ionic Strength on the Dissociation Constant of Hydrogen Cyanide", Acta Chim. Acad. Sci., Vol. 110 (4), 1982, pp.425-427. 57. P. Verhoeven, G. Heffer and P. M. May, "Dissociation Constant of Hydrogen Cyanide in Saline Solutions", Mineral & Metallurgical Processing , 1990, pp. 185 - 188. 58. J. H. Baxendale and D. T. Westcott, "Kinetics and Equilibria in Copper (II) - Cyanide Solutions", J. Chem. Soc, 1959, pp. 2347-2351. 59. G. Hefter, P. M. May and P. Sipos, "A General Method for the Determination of Copper (I) Equilibria in Aqueous Solution", J. Chem. Soc. Chem. Commun., 1993, pp. 1704 - 1706. 60. D. Cooper and R. A. Plane, "Cyanide Complexes of Copper with Ammonia and Ethylenediamine", Inorg. Chem., Vol. 5, 1966, pp.1677 - 1682. 61. "Gmelin Habdbuch" Cu[B] 60 , Verlag Chemie Gmbh (1965), p. 1447. 62. T. Chitani, "Muki Kagaku", Sangyo Tosho, Tokyo, 1964, p. 479. 63. A. Glasner and K. R. S. Asher, "A Violet Water-Soluble Copper Cyanide", J. Chem. Soc, 1949, p. 3296-3299. 64. W. P. Griffith, "Cyanide Complexes of Transition Metals", Quart. Rev. Chem. Soc (London) Vol. 16, 1962, pp. 188-207. 65. F. R. Duke and W. G. Courtney, "Complexes in Oxidation-Reduction Reactions. The Copper(II) - Cyanide Reaction", J. Phys. Chem. Vol. 56, 1952, pp. 19-21. 66. N. Tanaka, M. Kamada and T. Murayma, "Kinetics of the Reaction between Copper(II) -Ethylenediamine-tetra-acetate Complex and Cyanide", Bull. Chem. Soc. Japan, Vol. 31, 1959, pp. 895-900. 67. A. Longo and T. Buch, "Complex Cyanides of Copper(II)", Inorganic Chemistry, Vol. 6(3), 1967, pp. 556-5621. 68. O. Monsted and J. Bjerrum, "Spectrophotometric and Reaction Kinetic Studies of the Tetracyanocuprate (II) Complex in Methanol - Water Mixture at Low Temperatures", Acta Chem. Scand. Vol. 21, 1967, pp. 116-118. 69. R. Paterson and J. Bjerrum, "Stability of the Tetracyanocuprate(II) Complex from E.M.F. Measurements in a Methanol - Water Solvent at Low Temperature", Acta Chem. Scand. Vol. 19, 1965, pp. 729-734. 70. A. Katagiri, S. Yoshimura and S. Yoshizawa, "Formation Constant of the Tetracyanocuprate(II) Ion and the Mechanism of its Decomposition", Inorg. Chem., Vol. 20, 1981, pp. 4143 - 4147. 195 71. A. Katagiri, S. Yoshimura and S. Yoshizawa, Addition and correction for "Formation Constant of the Tetracyanocuprate(II) Ion and the Mechanism of Its Decomposition", Inorg. Chem., Vol. 25, 1986, pp. 2278. 72. G. Nord and H. Matures, "A Stopped-flow Kinetics Study of the Copper (II) Cyanide Reaction in Water and in Aqueous Methanol", Acta Chem. Scand. A, Vol. 28, 1974, pp. 13-19. 73. S. Yoshimura, A. Katagiri, Y. Deguchi and S. Yoshizawa, "Studies of the Anodic Oxidation of the Cyanide Ion in the Presence of the Copper Ion. IV. The Kinetics and Mechanism of the Decomposition of the Intermediate Tetracyanocuprate (II) Ion", Bull. Chem. Soc. Japan, Vol. 53, 1980, pp. 2437-2442. 74. R. R. Tessier, "Cyanide Removal from Waste Waters by Catalytic Oxidation with Hydrogen Peroxide and Copper Sulphate Catalyst", Canadian Patent CA 1,257,019, July 4, 1989. 75. G. H. Robbins, "Historical Development of the INCO S02/AIR Cyanide Destruction Process", CIM Bulletin, Vol. 89, 1996, pp. 63-69. 76. Y. Chen, C. You and W. Ying, "Cyanide Destruction by Catalytic Oxidation", 46th Purdue Industrial Waste Conference Proceedings, 1992, Lewis Publishers, Inc., Chelsea, Michigan, pp.539 - 545. 77. R. Shantz and J. Reich, "A Review of Copper Cyanide Metallurgy", Hydrometallurgy, Vol. 3, 1978, pp.99-109. 78. J. E. Clennell, "The Cyanide Handbook", 2nd Edition, McGraw-Hill Book Co., New York, 1915. 79. C. P. Richmond, "Electrolytic Precipitation of Cyanide Solutions", Eng. Min. J., Vol.83(11), 1907, pp. 512-515. 80. F. Passal, "Copper Plating during the Last Fifty Years", Plating, Vol. 46(6), 1959, pp. 628-638. 81. A. G. Gray, "Modern Electroplating", John Wiley & Sons, New York, 1953, pp.194-224. 82. F. A. Lowenheim, "Modern Electroplating", 3 r d Edition, John Wiley & Sons, New York, 1974, pp. 165-182. 83. J. V. Petrocelli, "A Study of the Electrolysis of Sodium Cuprocyanide Solution", Trans. Am. Electrochem. Soc, Vol. 78, 1940, pp.133-144. 84. G. H. Clevenger, "The Electrolytic Precipitation from Cyanide Solutions", Trans. Am. Electrochem. Soc, Vol. 28, 1916, pp. 263 - 306. 85. G. H. Clevenger, "Electrolytic Precipitation from Cyanide Solution", Engineering and Mining Journal, Vol. 102, No. 14, 1916, pp. 579-582. 86. O. C. Watts and A. Brann, "The Evolution of Hydrogen from Cyanide Plating Bath", Trans. Am. Electrochem. Soc, Vol. 31, 1917, pp. 303-310. 87. G. M. Smith and J. M. Breckenridge, "Cathode Potentials and Electrode Efficiencies of Copper in Copper Cyanide Solutions", Trans. Am. Electrochem. Soc, Vol. 56, 1930, pp. 397-408. 88. S. Glasstone, "The Limiting Current Density in the Electrodeposition of Noble Metals", Trans. Am. Electrochem. Soc, Vol. 59, 1931, pp. 277-285. 89. H. L. Benner and C. J. Wernlund, 'The High Efficiency Cyanide Copper Bath", Trans. Am. Electrochem. Soc, Vol. 80, 1942, pp. 355-365. 196 90. A. K. Graham and H. J. Read, "Rochelle Copper Plating", Trans. Am. Electrochem. Soc., Vol. 80, 1942, pp. 341-354. 91. I. C. Pan, "Concentrated Copper Cyanide Solutions", Trans. Am. Electrochem. Soc, Vol. 68, 1936, pp. 471-482. 92. J. Horner, "A Study of the Effect of Some Variables on the Speed and Distribution of Deposits from Cyanide Copper Plating Solutions", Technical Proceedings of the American Electroplaters' Society, Vol. 51, 1964, pp. 71-80. 93. E. Ivaskevic, "Electrodepsoition of Copper from Cyanide Solutions Containing Additives. II. Effect of Selenite and Selenide Additives on Electrode Process", Liet. TSR Moksu Akad. Darb., Ser. B, No. 5, 1989, pp24-30. 94. E. H. Hadley and K. A. Van Lente, "Copper(I) Thiocyanate Sodium Cyanide Electroplating Baths", Plating, Vol.60(3), 1973, pp.256-260. 95. C. J. Wernlund, "Bright Copper Plating", US Patent 2,347,448, April 25, 1944. 96. C. J. Wernlund, N. Tonawanda, and H. L. Benner and R. R. Bair, "Copper Plating", US Patent 2,287,654, June 2, 1942. 97. G. W. Jernstedt and J. D. Patrick, "Electroplating of Copper From Cyanide Electrolyte", U S Patent 2,636,850, April 28, 1953. 98. C. J. Wernlund "Addition Agent for Copper Plating", US Patent 2,774,728, Dec. 18, 1956. 99. G. N. Shivirin and E. M. Shivirina, "Electrode Processes in Copper Cyanide and Thiocyanate Solutions", Izvestiia Vysshikh Uchebnykh Zavedentii. Tsvetnaia Metallurgiia Tsvetnaia Metallurgiia, No. 4, 1977, pp. 18-22. 100. G. N. Shivirin and E. M. Shivirina, "Kinetics of the Electrochemical Reduction of Copper from Copper Cyanide and Thiocyanate Solutions", Izvestiia Vysshikh Uchebnykh Zavedentii. Tsvetnaia Metallurgiia, No. 4, 1977, pp.23-27. 101. R. Jagmines and B. Girdauskas, "Electrodeposition of Cu in Cu (I) Cyanide Solutions Containing a Thicyanate Additive 1. Electrodeposition under Steady-State Conditions of Electrolysis", Chemijia, (4). 1994, pp. 74-79. 102. G. R. Rogers and K. J. Taylor, "The Electrodeposition of Copper from Alkaline Cyanide Solution - I the Effect of Trace Selenide", Electrochimica Acta, Vol. 20, 1975, pp. 695-702. 103. E. Ivaskevic, I. Zitkeviciute and A. Steponavicius, "Electrodeposition of Copper from Cyanide Solution Containing Additives. 11. Effect of Selenate Additives on Electrode Processes", Liet. TSR Mokslu Akad. Darb., Ser. B, (4), 1989, pp. 15-21. 104. E. Ivaskevic, I. Zitkeviciute and A. Steponavicius, "Electrodeposition of Copper from Cyanide Solution Containing Additives. 11. Effect of Selenite and Selenide Additives on Electrode Processes", Liet. TSR Mokslu Akad. Darb., Ser. B, (5), 1989, pp. 24-30. 105. B. S. Krasikov and A. M. Gvozd, "Investigation of a Thiocyanate Electrolyte for Copper Plating", Journal of Applied Chemistry of the USSR, Vol. 30(6), 1957, pp. 1012-1016. 106. S. I. Uspenskii and M. A. Shluger, "The Influence of Ultrasonics on the Electrodeposition of Copper II Cyanide Electrolyte", Elektrokhimiya, Vol. 2, 1964, pp. 326-329. 197 107. U. Bertocci and D. R. Turner, "Copper", Encyclopedia of Electrochemistry of the Elements, Vol. 2, edited by A. J. Bard, Marcel Dekker Inc. New York, 1974, pp. 383-497. 108. M. L. Blanc and K. Schick, "Electrolysis with AC", Z. Phys. Chem., Vol. 46, 1903, pp. 213-243. 109. N. Pangarrov, I. Nenov and I. Khristova, "Predominant Orientation of Electrodeposited Copper and Nickel", IZV. INST. FIZIKOKIM. BURLGAR. AKAD. NAUK, Vol. 3, 1963, pp. 133-140. 110. M. Costa, "Kinetics of the Reaction of the Copper-Potassium Copper Cyanide Electrode", J. Rech. Centre Natl. Sci. Lab. Bellevue(Paris), No. 64, 1963, pp. 285-315. in. F. D. Lowenheim, Electroplating, McGraw-Hill, New York, 1978, p. 126. 112. R. Raub and K. Muller, Fundamentals of Metal Deposition, Elsevier Publishing Co., Amsterdam, 1967, pp 64-65. 113. R. Yu. Bek , B. D. Zhukov, "Mechanism and Kinetics of Copper Electrodeposition from Cyanide Electrolytes. Pt I, Hydrogen Evolution of Copper from Alkali Metal Cyanide Solutions", IZV. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk, (4), 1972, pp. 37-40. 114. R. Yu. Bek , B. D. Zhukov, "Mechanism and Kinetics of Copper Electrodeposition from Cyanide Electrolytes. Pt. 4, Concentration Polarization in Cyanide Cu-Plating Bath", IZV. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk, (6) (14), 1974, pp. 35-40. 115. R. Yu. Bek , B. D. Zhukov, "Electrodeposition of Copper from Cyanide Electrolytes Pt. V, Mechanism of the Process", IZV. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk, (6), 1974, pp. 40-44. 116. R. Yu. Bek and B. D. Zhukov, "Mechanism and Kinetics of Copper Electrodeposition from Cyanide Electrolytes", Elektrokhimiya, Vol. 129 (9), 1975, pp. 1291-1295. 117. A. A. Dikchyus, "Electrodeposition of Cu from Cyanide Solutions Ptl, Formation of Poor-Solubility Compounds on the Surface of a Cu Cathode", TRUDY AKADEMII NAUK LITOVSKOI SSR. SERIIA B., Vol.3 (100), 1977, pp. 23-28. 118. R. E. Sinitski, V. Srinivski and R. Haynes, "Electrode Kinetics of Copper Deposition from Copper Cyanide Solution", J. Electrochem. Soc, Vol.127 (1), 1980, pp. 47-51. 119. A. Juozenas, "Study of Surface of Copper Electrode in Cyanide Solutions", TRUDY AKADEMII NAUK LITOVSKOI SSR. SERIIA B., Vol.2 (147), 1985, pp. 16-22. 120. O. Dzilbute, "Influence of Conditions and Time of Electrolysis on Structure of Copper Deposits from Cyanide Solutions", TRUDY AKADEMII NAUK LITOVSKOI SSR. SERIIA B, Vol. 1 (152), 1986, pp. 30-42. 121. O. P. Koshcheev, V. I. Kichigin and V. V. Kamelin, "Electrodeposition of Copper on Highly Porous Cellular Materials", J. Appl. Chem. USSR, Vol. 65 (7), 1992, pp. 1251-1255. 122. D. Chu and P. S. Fedkiw, "The Electrochemistry of a Cuprous Cyanide Strike-Plating Bath", J. Electroanal. Chem., Vol. 345, 1993, pp. 107-120. 123. P. G. Hatherley, K. G. Watkins and M. McMahon, "The Properties of Copper Cyanide Electrolysis: a Comparison of Sodium and Potassium Formations", Trans. Inst. Met. Fin., Vol.70(4), 1992, pp. 177-183. 124. P. G. Hatherley and P. J. Carpenter, "Cathode Efficiency in Copper Cyanide Plating", Trans. Inst. Met. Fin., Vol.73(3), 1995, pp. 85-90. 198 125. R. Yu Bek, A. F. Zherebilov, "Copper Electrodepostion on Carbon Fibres", Zh. Prikl. Khim., Vol. 68(9), 1995, pp. 1470-1473. 126. D. A. Dudek and P. S. Fedkiw, "A Model of Mass-Transport Limited Copper Electrodeposition from Cyanide Electrolyte", The 83rd Ann. Tech. Conf, American Electroplaters and Surface Finshers Society, 1996, pp. 803 - 818. 127. A. Steponavicious, B. Radziuniene, E. Ivaskevic and B. Girdauskas, "On the Mechanism of the Cathodic Reaction in Cu(I) Cyanide Solution", Chemija, 1996, pp. 48-54. 128. A. Katagiri, H. Inoue and N. Ogure, "Cathodic Polarization of Copper Electrode in the CuCN-KCN solutions and the Current Distribution for Copper Deposition on Grooved Substrates", J. Appl. Electrochem., Vol. 27, 1997, pp. 529-538. 129. Y. J. Hsu and M. J. Tran, "Electrochemical Study on Copper Cementation from Cyanide Liquors Using Zinc", Electrochemimica Acta, Vol. 44 (10), 1999, pp. 1617-1625. 130. A. C. Halferdahl, "The Problem of Treating Cupriferous Precious Ores by the Cyanide Process", Eng. Min. Journal", Vol. 128(9), 1929, pp. 350-357. 131. A. T. Kuhn, "Electrolytic Decomposition of Cyanides, Phenols and Thiocyanates in Effluent Streams- A Literature Review", J. Appl. Chem. Biotechnol., Vol.21, 1972, pp. 29-34. 132. A. T. Kuhn, "The Role of Electrochemistry in Enviromental Control" in Modern Aspects of Electrochemistry, Vol. 8, edited by J. O'M. Bockris and B. E. Conway, Plenum Press, New York, 1972. 133. G. H. Clevenger and M. I. Hall, "The Electrolysis of Aqueous Solutions of the Simple Alkaline Cyanide", Trans. Am. Electrochem. Soc, Vol. 24, 1914, pp. 271-296. 134. D. Lay, "Electrical Precipitation of Cyanide Solution", Engineering and Mining Journal, Vol. 110, No.2, 1920, pp. 58-62. 135. L. B. Sperry and M. R. Caldwell, "Destruction of Cyanide Copper Solution by Hot Electrolysis", Plating, Vol. 36, 1949, pp. 343-347. 136. Von Hass Schmidta and Hass Meinert, "Electrolysen von Cyaniden in Wabrigen Losungen", Zeit. Fur Anorgan. Allg. Chem, Vol. 293, 1957, pp. 214-227. 137. Yu. Yu. Lure and V. E. Genkin, "Electrochemical Purification of Waste Waters Containing Cyanide Compounds", J. Appl. Chem., USSR, Vol. 33 (2), 1960, pp. 379-384. 138. D. T. Sawyer and R. J. Day, "Electrochemical Oxidation of Cyanide Ion at Platinum Electrodes", J. Electroanal. Chem., Vol. 5, 1963, pp. 195-203. 139. M. C. Dart, J. D. Jentles and D. G. Renton, "Electrolytic Oxidation of Cyanide Wastes", J. Appl. Chem., Vol. 13, 1963, pp. 55-64. 140. J. Drogon and L. Pasek, "Continuous Electrolytic Destruction of Cyanide Waste", Plating and Surface Finishing, Vol. 18, 1965, pp 310-313. 141. J. K. Easton, "Electrolytic Decomposition of Concentrated Cyanide Plating Waste", Plating, Vol. 53, 1966, pp 1340-1342. 142. J. J. Byerley and K. Enns, "Laboratory Study of Continuous Electrooxidation of Dilute Cyanide Waste", U.S.N.I.T.S., A.D. Rep., pp 1-49, 1974. 199 143. T. Arikado, C. Iwakura, H. Yoneyama and H. Tamura, "Anodic Oxidation of Potassium Cyanide on the Graphite Electrode", Electrochimica Acta, Vol. 21, 1976, pp. 1021-1027. 144. S. Yoshimura, A. Katagiri and S. Yoshizawa, "Studies on the Anodic Oxidation of Cyanide Ion in the Presence of Copper Ion I. Behavior of Cyano Complexes of Copper(I) as a Catalyst in the Anodic Oxidation of Cyanide Ion", Deki Kagaku Oyobi Kogyo Butsuri Kagaku, Vol. 47(8), 1979, pp. 360-366. 145. S. Yoshimura, A. Katagiri and S. Yoshizawa, "Studies on the Anodic Oxidation of Cyanide Ion in the Presence of Copper Ion II. Behavior of Cyano Complexes of Copper(I) as a Catalyst in the Anodic Oxidation of Cyanide Ion", Deki Kagaku Oyobi Kogyo Butsuri Kagaku, Vol. 47(6), 1979, pp. 488-491. 146. Z. Alaune, "Anodic Oxidation of Oxalic Acid and Cyanide Ions during Electrodeposition of Gold from Cyanide-Oxalate Solutions", Liet. TSR Mokslu Akad. Darb. Ser. B, (6), 1980, pp. 11-18. 147. S. Yoshimura, A. Katagiri, Y. Deguchi and S. Yoshizawa, "Studies of the Anodic Oxidation of the Cyanide Ion in the Presence of the Copper Ion. III. The Kinetics and Mechanism of the Decomposition of the Intermediate Tetracyanocuprate (II) Ion", Bull. Chem. Soc. Japan, Vol. 53, 1980, pp. 2434-2436. 148. S. S. Telepnev, "Modern State of Waste Purification from Cyanide in Gold Extraction Plants", Tsvetnye Metally, (9), 1980, pp. 102-106. 149. S. Yoshimura, A. Katagiri and S. Yoshiawa, "Reaction Products of Anodic Oxidation of Cyanide Ion in the Presence of Copper Ion", Nippon Kagaku Kaishi, No. 9, 1980, pp. 1327-1333. 150. A. Katagiri, S. Yoshimura, Y. Deguchi and S. Yoshizawa, "Catalytic Effect of Copper Ion on the Anodic Oxidation of Cyanide Ion", Proceedings of the Symposium on Electrocatalysis, (PV 82-2), Edited by W. E. O'Grady, P. N. Ross, Jr., and F. G. Will, The Electrochemical Society, Softbound proceedings series, Pennington, NJ (1982), pp. 336-346. 151. R. J. Daubaras, "Waste Treatment for Plating in the Soviet Union", Plating and Surface Finishing, Vol. 68, 1981, pp. 62-64. 152. S. Ehdaie, M. Fleischmann and R. E. W. Jansson, "Application of the Trickle Tower to Problems of Pollution Control III. Heavy Metal Cyanide Solutions", J. Appl. Electrochem., Vol.12, 1982, pp. 75-80. 153. S. Ehdaie, M. Fleischmann and R. E. W. Jansson, " Application of the Trickle Tower to Problems of Pollution Control. II. The Direct and Indirect Oxidation of Cyanide", J. Appl. Electrochem., Vol.12, 1982, pp. 69-73. 154. S. Ehdaie, M. Fleischmann and R. E. W. Jansson, " Application of the Trickle Tower to Problems of Pollution Control. I. The Scavenging of Metal Ions", J. Appl. Electrochem., Vol.12, 1982, pp. 59-67. 155. F. Kitamura, M. Takahashi and M. Ito, "Oxidation of the Cyanide Ion at a Platinum Electrode Studied by Polarization Modulation Infrared Reflection Absorption Spectroscopy", Chemical Physics Letters, Vol. 130, 1986, pp. 181-184. 156. F. Hine, M. Yasuda, T. Iida and Y. Ogata, "On the Oxidation of Cyanide Solutions with Lead Dioxide Coated Anode", Electrochimica Acta, Vol. 31, No.l 1, 1986, pp. 1389-1395. 157. J. Y. Hwang, Y. Y. Wang and C. C. Wan, "Electrolytic Oxidation of Cuprocyanide Electroplating Waste Waters under Different pH Conditions", J. Appl. Electrochem., Vol. 17, 1987, pp. 684-694. 158. F. Kitamura, M. Takahashi and M. Ito, "Anodic Oxidation of Cyanide and Cyanate Ions on a Platinum Electrode", Chemical Physics Letters, Vol. 136, No.l, 1987 pp. 62-66. 200 159. B. Wells and D. C. Johnson, "Electrocatalysis of Anodic Oxygen Transfer Reactions: Oxidation of Cyanide at Electrodeposited CoppefOxide Electrodes in Alkaline Media", J. Electrochem. Soc, Vol. 137, 1990, pp. 2785-2791. 160. T. C. Tan, W. K. Teo and O-T. Chin, "Electrochemical Destruction of Complex Ions", Chem. Eng. Commun., Vol. 38, 1985, pp. 125-133. 161. G. H. Kelsall, "Cyanide Oxidation at Nickel Anodes II. Voltammetry and Coulometry of Ni/CN-H 20 Systems at 298 K ", J. Electrochem. Soc, Vol. 138, No. 1, 1991, pp. 117-116. 162. M. L. Lin, Y. Y. Wang and C. C. Wan, "A Comparative Study of Electrochemical Reactor Configurations for the Decomposition of Copper Cyanide Effluent", J. Appl. Electrochem. Vol. 22, 1992, pp. 1197-1200. 163. C. S. Hofseth and T. W. Chapman, "Indirect Electrochemical Processes at a Rotating Disk Electrode: Catalytic Alkaline Cyanide Oxidation", J. Electrochem. Soc, Vol. 139 (9), 1992, pp. 2525-2529. 164. C. D. Zhou and D. T. Chin, "Copper Recovery and Cyanide Destruction with a Plating Barrel Cathode and a Packed-Bed Anode", Plating and Surface Finishing, Vol. 80 (6), 1993, pp. 69-78. 165. A. Socha, E. Kusmierek and E. Chrzescijanska, "Electrochemical Oxidation of Cyanides at Carbon Fibre and at Electrodeposited Copper Oxide Electrode", Polish Journal of Chemistry, Vol. 67(3), 1993, pp. 517-527. 166. C. W. Hwang, Y. Y. Wang and C. C. Wan, "Anodic Kinetics of Cyanide Ions in an Copper Cyanide Solution", J. Chin. Inst. Chem. Eng., Vol. 24, 1993, pp. 349 - 353. 167. L. Szpyrkowiz, F. ZilioGrandi, S. N. Kaul and S. RingoniStern, "Electrochemical Treatment of Copper Cyanide Wastewaters Using Stainless Steel Electrodes", Water Science and Technology, Vol. 38, 1998, pp. 261-268. 168. C. S. Hofseth and T. W. Chapman, "Electrochemical Treatment of Copper Cyanide Wastewaters Using Stainless Steel Electrodes", J. Electrochem. Soc, Vol. 146 (1), 1999, pp. 199-207. 169. M. N. Hughes, "General Chemistry", Chemistry and Biochemistry of Thiocyanic Acid and its Derivatives, edited by A. A. Newman, Academic Press, London, 1975, pp. 2-61. 170. J. E. Clennell, "Electrolytic Cyanide Regeneration", The Engineering and Mining Journal, May 27, 1911, pp. 1064-1066. 171. W. J. Crook, L. E. Booth and A. Thiel, "Electrolysis of Alkaline Solutions of Potassium Sulphate", Metallurgical and Chemical Engineering, Vol. 14, No. 10, 1916, pp. 587-591. 172. E. F. Kern, "The Electrolysis of Cyanide Solutions", Trans. Am. Electrochem. Soc, Vol. 24, 1924, pp. 241-270. 173. H. Kerstein and R. Hoffmann, "The Electrolytic Production of Free Thiocyanate from Alkali Thiocyanates", Ber. Vol. 57, 1924, pp. 491-496. 174. R. Gauguin, "Potentiometric Study of the Reducing Properties of the Thiocyanate Ion", J. Chem. Phys., Vol.42, 1945, pp. 136-148. 175. R. Gauguin, "Reducing Properties of Thiocyanate and Cyanide Ions - Oxidation-Reduction Potentials of Irreversible Systems", Ann. Chim 1949, pp. 832-835. 176. R. Gauguin, "Electrochemical Oxidation of the Thiocyanate Ion. Application to Determination and the Study of Reactions", Analytica Chimica Acta, Vol. 5, 1951, pp. 200-214. 201 177. A. Rius and S. T. Alonso, "Anodic Oxidation of Thiocyanate I", Anales Real Soc. Espan. Fis. Y Quim, Vol. 44B, 1948, pp. 1234-1250. 178. A. Rius and S. T. Alonso, "Anodic Oxidation of Thiocyanate II. Polarization Study in an Alkaline Medium", Anales Real Soc. Espan. Fis. Y Quim, Vol. 44B, 1948, pp. 1251-1260. 179. A. Rius and S. T. Alonso, "Anodic Oxidation of Thiocyanate II. Polarization Study in an Alkaline Medium", Anales Real Soc. Espan. Fis. Y Quim, Vol. 45B, 1949, pp. 359-366. 180. M. M. Nicholson, " Voltammetry of Thiocyanate Ion at the Stationary Platinum Electrode", Analytical Chemistry, Vol.31, No. 1, 1959, pp. 128-132. 181. A. J. Calandra, M. E. Martins and A. J. Arvia, "Kinetics of Anodic Reactions of Molten Potassium Thiocyanate on Platinum: Formation and Growth of Anodic Film", Electrochimica Acta, Vol. 16, 1971, pp. 2017-2080. 182. A. J. Arvia, A. J. Calandra and M. E. Martins, "Electrochemical Kinetics of Molten Potassium Thiocyanate on Platinum. Mechanisms of Anodic Film Formation and its Cathodic Dissolution", Electrochimica Acta, 1972, Vol.17, pp. 741-761. 183. C. Martinez, A. J. Calandra and A. J. Arvia, "The Anodic Oxidation of Thiocyanate Ion Dissolved as KSCN in Dimethylsulphoxide", Electrochimica Acta, 1972, Vol. 17, pp. 2153-2179. 184. R. Piereiro, A. J. Arvia and A. J. Calandra, " Kinetics of the SCN7(SCN)2 Couple on Platinum in Acetonitrile", Electrochim Acta, 1972, Vol. 17, pp. 1723-1734. 185. D. A. Holtzen and A. S. Allen, "Kinetics Parameters for the Anodic Oxidation of Thiocyanate at the Glassy Carbon Electrode", Anal. Chim. Acta, Vol. 69, 1974, pp. 153-160. 186. O. V. Belyi, L. M. Belaya and N. E. Karalovskaya, "A Potential and Galvanostatic Study of the Electrooxidation of the Thiocyanate Ion on a Stationary Platinum Electrode", Elektrokhimiya, Vol. 11, No. 3, 1975,pp.356-359. 187. L. Yu. Matulyaskene, "Behaviour of Thiocyanate Ions in Cyanide Bath for Electroplating of Copper", Lietuvos Tsr Mokslu Adademitijos Darbai. Serig B, Vol. 3 (100), 1977, pp. 9-15. 188. L. M. Tyrina and A. F. Morozov, "Electrochemical Treatment of Industrial Waste Water Containing Copper Cyanide and Thiocyanate", Khimiia Tekhnologiia Vody, Vol. 4 (5), 1982, pp.462-464. 189. V. K. Varentsov and Z. T. Belyakova, "Oxidation of Cyanide and Thiocyanate Ions in Effluent on Flow Type Plate and Fiber Anodes", Tsvetnye Metally, Vol. 24, No. 3, 1983, pp. 121-122. 190. V. K. Varentsov and Z. T. Belyakova, "Influence of Process Parameters on Oxidation of Cyanide and Thiocyanate Ions on Oxide Anode", Tsvetnye Metally, Vol. 25, No. 2, 1984, pp. 105-106. 191. V. K. Varentsov and Z. T. Belyakova, "Use of Titanium-Based Oxide Anodes in the Electrolytic Purification of Cyanide-Thiocyanate Solutions", Tsvetnye Metally, Vol. 25, No. 10, 1984, pp. 21-22. 192. E. Itabashi, "Identification of Electrooxidation Products of Thiocyanate Ion in Acidic Solutions by Thin-layer Spectroelectrochemistry", J. Electroanal. Chem., Vol. 177, 1984, pp. 311-315. 193. J. J. Byerley and K. Enns, "Electrochemical Regeneration of Cyanide from Waste Thiocyanate for Cyanidation', CIM Bulletin, 1984, pp. 87-93. 202 194. J. J. Byerley and K. Enns, "Processes for the Recovery of Cyanide from Aqueous Thiocyanate Solutions and Detoxication of Aqueous Thiocyanate Solutions", US Patent 4,526,662, July 28, 1985. 195. J. J. Byerley and K. Enns, "Processes for the Recovery of Cyanide from Aqueous Thiocyanate Solutions and Detoxication of Aqueous Thiocyanate Solutions", US Patent 4,519,880, May 28, 1985. 196. John K. Foley and Stanley Pons, "Fourier Transform Infrared Spectroelectrochemical Studies of Anodic Processes in Thiocyanate Solutions", Langmuir 1985, Vol. 1, pp. 697-701. 197. Zuana Tocksteinova and Fantisek Opekar, "The Electrochemical Generation of Small Amounts of Hydrogen Cyanide", Talanta, Vol. 33, No.8, 1986, pp. 688-690. 198. James A. Cox and Thomas J. Gray, "Controlled Potential Electrolysis of Bulk Solutions at a Modified Electrode: Application to Oxidation of Cysteine, Cystine, Methionine and Thiocyanate", Anal Chem, Vol. 62, 1990, pp. 2742-2744. 199. Eugene Y. Cao, Ping Gao, John Y. Gui, Frank Lu, Donald A. Stern and Arthur T. Hubbard, "Adsorption and Electrochemistry of SCN": Comparative Studies at Ag(l 11), and Pt(l 11) Electrodes by means of AES, CV, HREELS and LEED", J Electroanal. Chem., Vol. 339, 1992, pp. 311-325. 200. A. Zulauskaite, "Electrochemical Characteristics of System Containing Thiocyanate Complexes of Copper and Tin of Various Oxidation States", Chemija , No. 2, 1991, pp. 31-41. 201. P. Krishnan and V. G. Gurjar, "Electrochemical Thiocyanation by Two Phase Electrolysis", J Appl. Electrochem., Vol. 23, 1993, pp. 268-270. 202. P. Krishnan and V. G. Gurjar, "Electrochemical Oxidation Thiocyanate in a Two-Phase Electrolyte", J. Appl. Electrochem. Vol. 25, 1995, pp. 792-796. 203. T. Loucka and P. Janos, "Adsorption and Oxidation of Thiocyanate on a Platinum Electrode", Electrochimica Acta, Vol. 41, No. 3, 1996, pp. 405-410. 204. I. G. Casella, M. R. Guascito and G. E. DeBenedetto, "Electrooxidation of Thiocyanate on the Copper-Modified Gold Electrode and Its Amperometric Determination", Analyst, Vol. 123 (6), 1998, pp. 1359-1363. 205. N. N. Greenwood and A. Earnshaw, "Chemistry of the Elements", Chapter 15, Pergamon Press, New York (1990), pp. 851-853. 206. S. I. Zhdanov, "Sulphur", in Encyclopedia of the Electrochemistry of the Elements, A. J. Bard, ed., Vol.4, Marcel Dekker, New York, 1975, pp. 333-335. 207. F. Foerster and A. Friessner, "The Kinetics of Electrolytic Oxidation of Sulphite and Synthesis of Dithionate", Ber., Vol. 35, 1902, pp. 2525-2519. 208. V. A. Friessner, "Electrolytic Oxidation of Sulphite and Synthesis of Dithionate", Z. Electrochemie, Vol. 17, 1904, pp. 266-291. 209. O. Essin, "The Electrolytic Formation of Dithionate", Z. Elektrochem. Angew. Physik. Chem.,