8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
1/203
THE CHEMISTRY OF THE CARBON-IN--PULP PROCESS
Michael David Adams
A Thesis submitted to the Faculty of ScienceUniversity of the Witwatersrand, Johannesburg
in fulfilment of the requirements for thedegree of Doctor of Philosophy
Johannesburg 1989
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
2/203
ABSTRACT
Several conflicting theories of the adsorption ofaurocyanide onto activated carbon presently exist. Toresolve the mechanism, adsorption and elution ofaurocyanide are examined by several techniques, includingMossoauer spectroscopy, X-ray photoelectron spectroscopy, X-ray diffractometry, Fourier Transform Infraredspectrophotometry, ultraviolet-visible spectrophotometryand scanning electron microscopy.
The evidence gathered indicates that, under normal plantconditions, aurocyanide is extracted onto activated carbon
in the form of an ion pair M n* [Au(CN) 2 3 n, and eluted byhydroxide or cyanide. The hydroxide or cyanide ions reactwith the carbon surface, rendering it relativelyhydrophilic with a decreased affin ity for neutral species.Additional adsorption mechanisms are shown to operateunder other conditions of ionic strength, pH, andtemperature. The poor agreement in the literatureregarding the mechanism of adsorption of aurocyanide ontoactivated carbon is shown to be due to the fact thatdifferent mechanisms operate under different experimentalconditions. The AuCN produced on the carbon surface by acidtreatment is shown to react with hydroxide ion via thereduction of AuCN to metallic gold with formation ofAu (CN) 2 , and the oxidation of cyanide to cyanate. Other
species, such as An(CN)5 and Ag(CN)g adsorb ontoactivated carbon by a similar mechanism to that postulatedfor Au(CN) 2 .
Ion association of MAu(CN ) 2 salts in aqueous solution isdemonstrated by * aans of potentiometric titration andconductivity measurements, and various associatedspecies of KAu(CN), salts are shown to occur in organic
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
3/203
solvents by means of infrared spectrophoteaietric anddistribution measurements.
A kinetic model was developed for elut ion of aurocyanidefrom activated carbon and was found to predict gold elutionperformance successfully using the Zadra procedure.
The influence of the surface chemistry and structure ofactivated carbon on adsorption of aurocyanide wasinvestigated by characterization of activated carbons thatwere synthesized or oxidized under various conditions.Synthetic polymeric adsorbents with characteristicssimilar to activated carbons were also studied. The
evidence suggests that a large nicropore volume isimportant in providing suitable active sites foradsorption. Another important factor is the presence ofbasic functional groups within the micropore, which act assolvating agents for the ion pair.
The aim is to provide a self-consistent adsorptionmechanism that accounts for all observations presented inthe literature. Interprrtation of results in terras ofpreconceived ide..s, and neglect of observations of otherauthors has greatly contributed to current disagreement inthe literature.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
4/203
Declaration
I dec lare that this Thesis is ray own, unaided work. It isbeing submitted for the degree of Doctor of Philosophy. Ithas not been submitted before for any degree or examinationin any other University.
2- 3 i day of , 1989
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
5/203
Acknowledgements
I wish to express my sincere gratitude and appreciation to
my supervisors. Prof R.D. Hancock and Prof R.G.Copperthvaite for their enthusiasm and help;
my wife, Jenni, for her support ;
my colleagues, Dr C.A. Fleming, Dr R.L. Paul, Dr M.J.Niccl, Dr B.P. Green, Dr P.J. Harris amongst othe rs , forthe many hours of useful discussions;
Prof Dr F.E. Wagner of the Physik-Depart-ment, TechnischeUniversitat Hunchen, for running the Mossbauer spectraand for engaging in a useful collaboration regarding thistechnique;
Mr P. Ellis and Mrs I. Klingbiel, for their patience withrunning the Scanning Electron Microscope;
Mr D.E. Im es , for running the X-ray Photoelectronspectra;
Mr P.W. Wade, for his time and help with the MolecularMechanics calculations;
Mrs M. Arinto, Mrs M.R. Hazell and Ms T. McArthur, fortheir assistance in the typing of this manuscript;
the Council for Mineral Technology (Mintek), forallowing me to undertake this research during ray periodof employment with them;
my parents, for their encouragement through the years.
i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
6/203
Contents
pageCHAPTER 1 INTRODUCTION 11.1 The Adsorption of Aurocyanide onto Activated
Carbon - An Historical Perspective 11.2 The Carbon-in-pulp Process - Process
Chemistry and Unknown Areas 3a; Leaching of Gold from the Ore 3b) Adsorption of Aurocyanide onto Activated
Carbon l
c) Carbon Acid Washing and Elution 7d) Gold Electrot'inning or Precipita tion 8e) Carbon Reactivation 8f) Other Features of the CIP Process 9
1.3 The Structure and Chemistry of ActivatedCarbon 10
a) Physical Structure of Activated Carbon 10b) Chemical Stricture of Activated Carbon li1.4 The Chemistry of Gold - An Introduction 131.5 Current Processes for the Recovery of Gold 17
CHAPTER 2 EXPERIMENTAL PROCEDURE 192.1 Reagents and Chemicals 192.2 Activated Carbon Adsorption studies 20a) Equilibrium Adsorption 20b) Elution of the Carbon 21c) Precipi tation of AuCN 21d) Decomposition of AuCN 22e) Oxygen Effect Experiments 22
f) Gold(XII) Cyanide Experiments 23g) Silver and Mercury Experiments 23h) Miscellaneous Techniques 24
(i) X-ray diffractometry 24
(ii) Scanning electron microscopy 24
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
7/203
(iii) Fourier Transform infraredSpectrophotometry 24
(iv) 19 7 Au Mossbauer spectroscopy 25
(v) x y photoelectron spectroscopy (XPS) 332.3 Solution Studies 35a) Pctantiometric Titration 35by Conductivity Measurements 35c) Infrared Spectrophotometry 36d) Distribution Experiments 36e; Determination of Chloride Concentration in
Organic Phases 37Determination of Dielectric Constant 37
g) Determination of Water Content by Automatic
Karl-Fischer Titration 382.4 Molecular Mechanics Calculations 38a) The Force Field 38by Conformations of Polyether in Organic
Solvents 39
2.5 Activated Carbon Elution Studies 40a) Elution Mechanism Experiments 40by Kinetic Experiments 41
2.6 Activated Carbon Surface Chemistry andStructure Studies 4 5
ay Carbons and Adsorbents Used 45by Synthesis of Activated Carbons 45cy Oxidation of Activated Carbons 47dy Synthesis of Polyxanthene and Polyquinone 47ey Adsorption of Aurocyanide 48fy Techniques for the Characterization of
Physical Properties 49gy Techniques for the Characterization of
Chemical Properties 51
CHAPTER 3 THE MECHANISM OF ADSORPTION OFAUROCYANIDE ONTO ACTIVATED CARBON 53
3.1 The Mechanism of Adsorption of Aurocyanideonto Activated Carbon - A Literature Review 53
ay Summary of Factors Influencing Adsorp-
vii
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
8/203
b!
c,
3.2
a)
b;
c)
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
9/203
a) The Mechanism of Adsorption of Au(CN) ̂
onto Activated Carbon 145b) The Mechanicms of Adsorption of Ag(CN) 2
and Ag* onto Activated Carbon 154c) The Mechanisms of Adsorption of Hg(CN) 2 and
HgCl2 onto Activated Carbon 172d) The Mechanism of Adsorption of AuCl ~ 4 onto
Activated Carbon 1823.4 Summary and Conclusions 182
CHAPTER 4 ION ASSOCIATION OF MAti(CN) , SALTS INVARIOUS SOLVENTS
4.1 Ion Association of HAu(CN) 7 Salts in Various
Solvents - A Literature Survey4.2 The Study of Ion Association - A Literature
Survey4.3 Ion Association of MAu(CN ) 2 Salts in Aqueous
SolutionPotsntiometric i'itration of Aqueous Solution of HAufCN ;2 Conductimetric Study of Aqueous Solutions of
KAu(CN)? and NaAu (CN) 2
Ion Association of MAu(CN ) 2 Salts in Organic
SolventsInfrared Spectra cf MAu(CN) 2 Salts in Organic Solvents
Distribution of MAu(CNj 2 Salts betweenAqueous Solutions and Organic SolventsEffect of Solvent Effect of Diluent Effect of Cation
Extraction of Aurocyanide Ion Pairs by
Poly(oxyethylene) ExtractantsSolvent Extraction of M#Au (CN) 2 Ion Pairs by Triton X-10C
Molecular Mechanics Calculations of the Complexation of Alkali Metal Cations by Polyethers
190
a)
h)
aj
a;b)
c)
4.6
h)
195
212224234238
251
258
1%
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
10/203
4.7 Summary and Conclusions 266
CHAPTER 5 THE MECHANISM OF ELUTION OF
AUROCYANIDE FROM ACTIVATED CARBON5.1 Introductiona) Processes for Elution of Aurocyanide
frcm Activated Carbon
b) Chemistry and Mechanism of Elution of
Aurocyanide from Activated Carbon
5.2 Novel Studies on the Mechanism of Elution ofAurocyanide from Activated Carbon
5.3 Kinetics of Elution of Aurocyanide fromActivated Carbon
a) Effect of Finite, Constant Concentration
of Gold in Eluant Feed
b) Effect of Variable Gold Concentration inEluant Feed
c) Elution of Aurocyanide from Carbon in aPacked Column
5.4 Factors Influencing Elution of Gold fromActivated Carbon
a) Effect of Temperature
b) Effect of Ionic Strength
c) Effect of Cyanide and Hydroxide
Concentrations
d) Effect of Organic Solvents
5.5 Oth er Aspects of the Elution Processa) Effect of Acid washing Prior to Elution
b) Effect of Cyanide Decomposition onGold Elution Efficiency
c) Effect of Removal of Cyanide from Eluant Solution
d) Selective Elution of Copper and Mercury5.6 Conclusions
CHAPTER 6 THE INFLUENCE OF ACTIVATED CARBONSURFACE CHEMISTRY AND STRUCTURE ONTHE ADSORPTION OF AUROCYANIDE
269269
269
270
272
278
286
291
293
298298301
301303306307
?08
311
311316
319
X
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
11/203
6-1 Influence of Activation Conditions andStarting Material on the Surface Chemistry,Structure and Gold Adsorption Activity of
Activated Carbona; Gold Adsorption Activity b) Activated Carbon Structure
c) Activated Carbon Surface Chemistry
6.2 Effect of Surface Oxidation on theSurface Chemistry, Structure and GoldAdsorption Activity of Activated Carbon
a) Gold Adsorption Activity b) A. rivated Carbon Structure
c) Activated Carbon Surface Chemistry
6.3 Synthesis and Characterization of PolymericModels for Activated Carbon: Polyxanthenesand Pol uinones
a} Gold Adsorption Activity
b) Polymer Structure
c) Polyxanthene and Polyquinone Surface Chemistry
6.4 Summary and Conclusions
CHAPTER 7 CONCLUSIONS AND RAMIFICATIONS7.1 The Mechanism of Adsorption of Aurocyanide
onto Activated Carbon7.2 Ion Association of MAu{CN)z Salts in
Various Solvents7.3 The Mechanism of Elution of Aurocyanide from
Activated Carbon7.4 Influence of Activated Carbon Surface
Chemistry and Structure on the Process7.5 Ramifications for Existing Carbon-in-Pulp
Operationsa) Cyanide-free Elution
b) Acid Treatment of Loaded Carbon Followed by Cyanide-fzee Elution
c) Effects of Oxygen and Ionic Strength
d) Kinetics of Elution
319320322337
347347349351
353357364
36737 a
375
375
379
380
382
383383
384384384
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
12/203
7.6 Ramifications for Future Process Optionsa) Modification of Dielectric Constant within
the Ion-exchange Resin
b) Alternative Functional Groups and Matrices for Ion-exchange Resins
APPENDICES1. Some Physical and Chemical Properties of the
Organic Solvents Studied2. Calculation of Estimated Effective Ionic
Radii for Au(CN) ~2 and Ag(CN);3. Distribution and Associa tion Data for
MAu(CN) 2 in Various Solvents
4. Reduction Potentials of AuCN and AgCN5. List of Publicat ions Aris ing from this
Thesis6 . Directory of Analytical Data Located on
Microfiche7. Directory of Computer Programs Developed
During this Research, Located on Microfiche
REFERENCES
?S4
385
385
386
387
389
390
391
393
395
396
x l i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
13/203
List of Symbols
A Onsagar constanta interionic distance§ ionic radius in modified Born equationa degree of dissociationBV bed-volume of eluant solution8 XRD peak width at £ maximum intensityB3 stability constant of two-coordinate complexC concentration of gold on the carbon at time tC e initial concentration of gold on the carbonc concentration of gol d in the aqueous phasec hy velocity of lightc 8 concentration of gold in the solvent phaseD distribution ratioDp pore diameter of adsorbentDN solvent donor numbaraHhyj hydration enthalpy
solvation enthalpy
aHg electrostatic contribution to tnesolvation enthalpyneutral contribution to the solvationenthalpy
5 solubility parameterE single-pass electrowinning efficiencyEa activation energyE fc binding energy of electronE; energy of incident Mossbauer radie'jjnEp kinetic energy of photoelectronEuf work functionEy energy of 7-radiation from stationary Mossbauer
source@ ti electronic charge< bulk dielectric constant of solvent< % dielectric constant of primary solvation shell
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
14/203
F Faraday constantfs mean activity coefficient in organic phasef+ activity coefficientK equilibrium constantK k association constant
ionic extraction constantKjj ion-pair extraction constantKj® T total extraction constantk intrinsic rate constantk' measured first order rate constant for gold
elutionki measured first order rate constant for cyanide
decompositionkg Bol tzmann’s constantt conductivi tyL Avogadro1s number
microcrystallite diameterLc microcrystalli te heightA equivalent conductanceA0 equivalent conductance at infinite dilutionX wavelength of X-radiationX* equivalent cationic conductance at infinite
dilutionXg equivalent anionic conductance at infinite
dilutionM e mass of carbon in the elution columnM, mass of solution in the elut ion columnn aggregation numberi) solution viscosityR eras constant
Ra ratio of carbon mass to solu tion mass in theelution column
r solvent radiusrc erystal lographic ionic radiusreff effective ionic radius of asymmetrical ion
j radial distance from centre of ionic charger s Stokes radiusS eut concentration of gold in the eluate solution
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
15/203
exiting the carbon column cr bed at t ime tconcentration of gold in the eluant feed solutionto the carbon column or bed at time tincident angle corresponding to XRD peak maximumflowrata of eluantflowrate of eluant solution from carbon bed intosimulated electrowjnning cellflowrate of diluent solution into simulatedelectrowinning cellvelocity applied to Mossbauet sourcemole fraction of water in the organic phaseionic charge
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
16/203
List of Tables
1.1 Stability Constants for a Selection ofComplexes of Gold(I) and Gold(III)
2.1 Mossbauer parameters for MAuC1 4 salts
2.2 Conditions of preparation of activatedcarbons studied
3.1 Analysis for potassium, gold, and nitrogenon carbons loaded from 0,1M potassiumchloride and 0 ,1M hydrochloric acid
3.2 Elution of aurocyanide from activated carbonwith 0,1 M sodium hydroxide at 90°C
3.3 Analyses of sodium hydroxide eluatesfor Au (CN )2 and Au after elution
3.4 Effect of pre-treatment of carbon withhydrochloric acid on the rate of elution ofgold with sodium hydroxide
3.5 Mossbauer spectral parameters for goldspecies on activated carbons and goldcompounds
3.6 Debye-Waller factors for some gold compounds
(after Cohen et a l. 88)
3.7 Decomposition of AuCN in aqueous solution
page
16
30
46
5C
60
61
69
72
74
80
3.8 N/Au ratios for activated carbons loadedwith aurocyanide under various conditions
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
17/203
and dried at different temperatures 8 6
3.9 Mossbauer spectral parameters for goldspecies on activated carbons and goldcompounds
3.10 Changes in solution pH during the adsorptionof aurocyanide onto activated carbon undervarious conditions
3.11 Changes in pH of a 0 ,1M KC1 solution incontact with activated carbon, in theabsence of aurocyanide
3.12 Concentrations of nitrogen and gold on^ carbons
3.13 Concentrations of anions in gold adsorptionsolutions
3.14 Concentrations of chloride on carbons
3.15 Au(4f 7/2) binding energies (eV) of goldcompounds and adsorbed species
3.16 Na(2p), K(2p) and H(ls) binding energies ofgold compounds and adsorbed species withreferencing normalized to C(ls) of activated
carbon at 284,4 eV
3.1" Au(4f 7/2) binding energies (eV) of goldcompounds and adsorbed species with
referencing normalized to C(ls) of activatedcarbon at 284,4 eV
3.18 Infrared spectral data for solid aurocyanidesalts
97
105
106
107
107
108
117
119
123
133
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
18/203
3.19 Infrared spectral data for aurocyanidespecies in solution and adsorbed on ion-exchange resins
3.20 Infrared spectral data for aurocyanidespecie? adsorbed on activated carbon
3.21 Extraction of Au(CN); and Au(CN); byactivated carbon
3.22 Extraction constants for complex cyanidesfrom aqueous solution at 25 eC bytetrahexylarasoniua eidmannate in hexone(After Irving and Damodaran138)
3.23 Analyses of Au(CN)^ loaded carbons
3.24 Values of isomer shift (IS) and quadrupolesplitt ing (QS) for Mossbauer spectra ofgold samples
3.25 Analyses of K, Ag and N on activated carbonloaded with silver from KAg(CN) 2 solution
3.26 r(CN) stretching frequencies for silver
cyanide species
3.27 Effect of acid and base treatment on thestoichiometry of the adsorbed Ag(CN)%species
3.28 Elemental analyses of carbons loaded from
various mercuric cyanide solutions
3.29 Standard reduction potentials of mercury-
containing species
3.30 y(CN) stretching frequencies for mercuric
x v i i i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
19/203
cyanide species 179
3.31
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
Solubilities of mercury compounds
Values of pKa and a (degree of dissociation)for titration data in Figure 4.1
Stability of HAu(CN )2 solutions with time
Concentrations and corresponding equivalentconductances for MAu(CH ) 2 salts in aqueoussolution
Constants derived using Equation 4.3
Specific ionic conductances and ionic radiifor some anions in aqueous solution
Comparison of estimated effective ionic radiiwith Stokes radii for Au(CN)% and Ag(CN),in aqueous solution
Infrared spectral data for LiAu(CN)z species
in tetrahydrofuran (THF) and methyl ethylketone (MEK)
Association and extraction constants forKAu(CN) 2 in several isodielectric solvents
Association and extraction constants forKAu(CN) 2 in diethyl ether and nitrobenzene
Aggregation numbers and relevant propertiesof solvent mixtures
Ionic radii and absolute hydrationenthalpies of some ions
180
197
200
202
202
205
206
211
232
233
237
244
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
20/203
4.12 Solubility of aurocyanide salts in aqueoussolution
4.13 Efficiencies of extraction of M*Au(CN),ion pairs (M* = Li*, >.*, Cs*) by TritonX— 100 dissolved in various solvents
4.14 Solubilities of water in the various solventphases
4.15 Strain energies of conformations ofuncomplexed poly(oxyethylene) in media oflow dielectric constant (€=r;j), highdielectric constant (t=4rij) and infinitedielectric constant
4.16 Strain energies of metal complexes ofpoly(oxyethylene) in benzene (low polari tymedj’-’ij € = r|j)
4.17 Strain energies of metal complexes ofpoly(oxyethylene) in nitrobenzene (high
polarity medium? < = 4r;j)
5.1 Extraction of aurocyanide and sodium ionsfrom aqueous solutions by polymericadsorbents
5.2 Extraction of sodium and hydroxide ions fromsodium hydroxide solution by activatedcarbon at 25°C
5.3 Comparison between observed and calculatedvalues for k* and the intercept forEquations (5.19) and (5.20)
5.4 Gold concentrations on the carbon after 16hours elution for simulated electrolytic
xx
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
21/203
292
306
308
316
322
326
332
335
337
348
350
350
extractions of differing single-passefficiency
5.5 Elution of gold by organic solvents at 25®C(after Nicol efc a l 251)
5.6 Effect of acid treatment on the elution rate
5.7 Elution of gold, silver and copper withsodium cyanide at 20eC (after Fleming andNicol68)
6.1 Gold adsorption activities of synthetic andcommercial activated carbons
6.2 Structural properties of synthetic andcommercial activated carbons
6.3 Conductivities of synthetic and commercialactivated carbons
6.4 Microcrystallite dimensions of synthetic andcommercial activated carbons, from XRD peakdata
6.5 Chemical characte sties of synthetic andcommercial activated carbons
6 .6 The effect of surface oxidation on the goldadsc rption activity of Le Carbone G210activated carbon
6.7 Structural properties of oxidized Le CarboneG210 activated carbon
6.8 The effect of surface oxidation on theconductivity and microcrystallite dimensionsof Le Carbone G210 activated carbon
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
22/203
£.9 The effect of surface oxidation on thechemical characteristics of Le Carbone G210
activated carbon 352
6.10 Extraction of aurocyanide by powderedpolyrars, adsorbents and activated carbonsunder various conditions 356
6.11 Extraction of aurocyanide by polyxantheneand polyquinone under various conditions 363
6.12 Structural properties of powdered polymers,adsorbents and activated carbons 365
6.13 Conductivities of powdered polymers,adsorbents and activated carbons 366
6.14 Chemical characteristics of powderedpolymers, adsorbents and activated carbons 367
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
23/203
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
24/203
gold cyanide compounds 29
2.3 Correlation of isomer shifts and quadrupolesplittings for gcld(l) and gold(III)compounds, for different chemicalenvironments. Isomer shift values arerelative to Au in Pt source. On y atomsbonded to gold are indicated. (After Cashionet al 3 9) 31
2.4 Plot of IS and QS for 197Au Mossbauerspectra of MAuCl 4 salts against cationicradius 32
2.5 Laboratory eluti m apparatus 42
2.5 Simulation of the Zadra process by dilution 44
2.7 Apparatus for the measurement ofconductivity of carbons 50
3.1 Effect of duration of acid pre-treatment on
the efficiency of the elution of gold fromactivated carbon with a sodium hydroxidesolution. (Conditions: 4 per cent (m/m)hydrochloric acid pre-treatment at 95°Cfollowed by elution with 0, 1 mol /1 sodiumhydroxide at 90°C) 63
3.2 SEM micrographs of activated carbon surfacesafter(a) acid pre-treatment for 6 hours (4 per
cent hydrochloric acid at 95°C), and(b) acid pre-treatment for 6 hours followed
by elution with sodium hydroxide
(0,1 sol/1) at 90°C 65
3.3 SEM micrographs of activated carbon surfaces
xxiv
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
25/203
after pre-treatipent for 30 minutes with 4per cent hydrochloric acid at 95 °C followedby elution with sodium hydroxide ( 0,1 mol/ 1 )
at 90eC(a> Gold crystals concentrated in a
ma cr pore, and(b) a magnified port ion of (a), showing the
dendritic structure of the crystals 66
3.4 SEM micrographs of activated carbonparticles after(a) pre-treatment for 6 hours with 4 per
cent hydrochloric acid at 95*C followedby elution with sodium hydroxide ( 0, 1 mol/1) at 90°C, and
(b) acid pre-treatment for 30 minutesfollowed by elution with sodiumhydroxide 67
3.5 |e?Au Mossbauer spectrum of aurocyanideloaded onto carbon and subsequently boiledin 4% hydrochloric acid for 3 hours. (Solid
lines indicate fitted curves? circlesindicate data points) 71
3.6 l*1Au Mossbauer spectrum of the activatedcarbon sample shown in Fig. 3.5 that wassubsequently boiled in 0,1M sodium hydroxidesolution for 5 hours 75
3.7 i 97 Au Mossbauer spectrum shown in Fig.3.6, showing detail of quadrupole assigned
to Au(CN)g 76
3.8 Kinetics of AuCN precipitation from aqueoussolution at various temperatures 78
3.9 Arrhenius plot of data for the precipitation
xxv
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
26/203
of AuCN from aqueous solution 79
3.10 Ultraviolet spectra of aqueous solutions incontact with polymeric AuCN under variousconditions 8 3
3.11 X-ray diffractograms of solid residuesresulting from the contact of polymeric AuCNwith aqueous solutions under variousconditions. (Numbers on the 2 9 axisindicate peak position in degrees) 33
3.12 X-ray diffractograms of activated carbonsloaded with aurocyanide .n the presence of0 , 1M KC1, after heating at varioustemperatures 87
3.13 SEM micrograph of an activated carbonsurface loaded with aurocyanide in thepresence of 0,1M KC1, after heating at 340aC 89
3.14 SEM micrograph of an activated carbon
surface loaded with aurocyanide in thepresence of 0,1M KC1, after heating at 270°C 90
3.15 SEM micrograph of an activated carbonsurface loaded with aurocyanide in thepresence of 0 , 1 M KCl, after heating at 3OO0O 91
3.16 SEM micrograph of the activated carbonsurface shown in fig. 3.15, showing thedistribution of gold particles 92
3.17 X-ray diffractograms of activated carbonsloaded with aurocyanide in the presence of0,1M HCl , after heating at varioustemperatures 94
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
27/203
3.18 19 7 Au KossXauer spectra of activatedcarbon loaded with aurocyanide in thepresence of Q,?.M XC1, and f isequently
heated to various temperatuies 95
3.19 Expanded view of ls?Au Mossbauer spectrashown in Fig. 3.18, showing detail of peaksdue to minor species 96
3.20 Effects of oxygen and nitrogen bubbling onthe adsorption of aurocyanide onto activatedcarbon from solution containing no additives(Carbon mass l, 0 g; solution volume 250 ml;initial gold concentration in solution300 mg/1) 101
3.21 Effects of oxygen and nitrogen bubbling onthe adsorption of aurocyanide onto activatedcarbon from 0,1M KCl solution. (Carbon massl,0g; solution volume 250 ml; initial goldconcentration in solution 300 mg/1) 102
3.22 Effects of oxygen ari nitrogen bubbling onthe adsorption of aurocyanide onto activatedcarbon from 0,1M KOH solution. (Carbon massl,0g? solution volume 250 ml; initial goldconcentration in solution 300 mg/1) 103
3.23 Effects of oxygen and nitrogen bubbling onthe adsorption of aurocyanide onto activatedcarbon from 0,1M HCl solution. (Carbon massl,0g; solution volume 250 ml; initial gold
concentration in solution 300 mg/1) 104
3.24 Au(4f) photoelectron spectrum of goldspecies on activated carbon contacted with a
solution of AuCi; ina) 1,OH HCl
x x v l i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
28/203
b) aqua regia 125
3.25 Au(4f) photoelactron spectrum of HAuC1 4 salt
in physical contact with activated carbon 127
3.26 C(ls) photoelectron spectra ofa) n2X0 activated carbon,b) activated carbon contacted with a
solution of AuCi; in 1 , 0 M HCl, andc) activated carbon in physical contact with
HAUC1, salt 129
3.27 0(ls) photoelectron spectri of
a) 6210 activated carbonb) activated carbon contacted with a
solution of AUCI 4 in l,0M HCl, andc) activated carbon in physical contact with
HAuCl4 salt 130
3.28 Infrared spectra showing the CN stretchbands of KAu (CN) 2 on activated carbon atvarious concentrations. (Drying conditions:25°C, in vacuo) 132
3.29 Infrared spectra showing the CN stretchbands of CsAu(CN)? on activated carbonafter drying under various conditions.(18,9 per cent Au on carbon) 137
3.30 Infrared spectra showing the CN stretchbands of CsAu(CN) t on activated carbon atvarious concentrations.
a) Dried at 25°C, in vacuo 138b) Dried at 120°C 139
O ;3.31 Infrared spectra showing the CN stretch
i bands of LiAu(CN ) l on activated carbonafter drying with various conditions.
o
xxvili
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
29/203
(19,7 per cent Au on carbon) 140
3.33
3.34
3.35
3.36
3.37
3.38
Infrared spectrum showing the CN stretch
bands of LlAu(CN), on activated carbon atvarious concentrations. (Drying conditions:25°C, in vacuo) 141
Infrared spectra showing the C?f stretchbands of LiAu(CN ) 2 in methyl ethyl ketone atvarious concentrat ions 143
Infrared spectra showing the CN stretchbands of Ca[Au(CN )2]2 on activated carbonafter drying under various conditions.(21,3 per cent Au on carbon) 144
1 9 7Au Mossbauer spectrum of activatedcarbon loaded from Au(CN)5 solutioncontaining 0,1M KCl 151
197 Au Mossbauer spectrum of activatedcarbon loaded from Au(CN); solutioncontaining 0,1M HCl 152
197 Au Mossbauer spectrum of activatedcarbon loaded from Au(CN)J solution andsubsequently boiled in 4% hydrochloric acidfor four hours 153
Infrared spectra showing r(CN) for silvercyanide species adsorbed on activatedcarbons from 0,1M KCl solution after various
drying treatments 158
Effect of cyanide concentration on theext.action efficiency of activated carbonfor silver 161
xxix
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
30/203
3.40 Distribution of silver species in cyanidesolution with increasing total cyanideconcentration. (pH - 11; unadjusted) 162
3.41 SEM micrograph of an activated carbonsurface loaded with argentocyanide andsubsequently boiled in 4% HCl for 5 h.(520 k magnificat ion) 165
3.42 SEM micrograph of the activated carbonsurface in Fig. 3.41, subsequently boiledin 0,1 M NaOH for 5 h.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
31/203
stoich
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
32/203
4.6 Infrared spectra showing the CN stretchbands of NBu4Au(CN )2 in THF at various
concentrations
4.7 Infrared spectra showing the CN stretchbands of NBu4 *u (CN )2 in MEK at variousconcentrations
4.8 Infrared spectra showing the effect of wateron the CN stretch bands of LiAu(CN), andNB u 4A u (CN) 2 in THF
4.9 Infrared spectra showing the effect of wateron the CN stretch bands of LiAu(CN)2 andNBu4Au(CN ) 2 in MEK
4.10 Total gold concentration in the organicphase versus that in the aqueous phase forthe distribution of KAu(CN)% between TBP and0,1 M KCl solution
4.11 Total gold concentration in the organicphase versus that in the aqueous phase forthe distribut ion of KAu(CN)2 between1-pentanol and 0,1 M KCl solution
4.12 Total gold concentration in the organicphase versus that in the aqueous phase forthe d istribution of HAu(C>i)2 between1-pentanol and 0,1 M HCl solution
4.13 Effect of water content of solvent on K* forKAu{CN)2 in water-saturated solvents
4.14 Effect of water content of solvent on K%%*for the distribution of KAu(CN)2 betweenvarious solvents and 0,1 M KCl solution
213
214
215
216
218
220
222
225
226
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
33/203
4.15 Effect of water content of solvent on k J® b for the distribution of KAu(CN)2 between
various solvents and 0,1 M KCl solution 227
4.16 Effect of water content of solvent onfor the distribution of KAu(CN)2 betweenvarious solvents and 0,1 M KCl solution 228
4.17 Plot of log K r versus log t for KAu(CN)3 inwater-saturated solvents. (Solid line is thetheoretical Bjerrum relation (4.8) for 1:1salts, after Bockris and Reddy11) 231
4.18 Total gold concenti cion in the organicphase versus that in the aqueous phasefor the distribution of KAu(CN)2 oetween1-pentanol/benzene mixtures (molar ratio1:1) and 0,1 M KCl solution 235
4.19 Total gold concentration in the organicphase versus that in the aqueous phase forthe distribution of KAu(CH)2 between1-pentanol/cyclohexane mixtures (molar ratio1:1) and 0,1 M KCl solution 236
4.20 Effect of cationic radius on forthe distribution of MAu(CN)2 between varioussolvents and 0,1 M MCI solution 239
4 21 Effect of cationic radius or. K%%* forthe distribution of MAu(CN)8 between various
solvents and 0,1 M MCI solution 240
4.22 Effect of cationic radius on xj{ forthe distribution of MAu(CN )2 between varioussolvents and 0,1 M MCI solution 241
x x x l i i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
34/203
4 . 2 4
4.25
4.26
4.27
4.28
4.29
4.30
Effect of cationic radius on Ke for MAu(CN)2in various water-saturated solvents 242
Effect of water content of solvent on thedistribution and association constants forHA u (CN) 2 between various solvents and 0,1 MHCl solution 247
Effect of water content of solvent on thedistribution and association constants forLiAu(CN)2 between various solvents and 0,1 MLiCl solution 248
Effect of water content of solvent on thedistribution and association constants forCsAu(CN)2 between various solvents and 0,1 MCsCl solution 249
Effect of water content of solvent on thedistribution and association constants forNEt4Au(CN ) 2 between various solvents and0,1 M NEt 4Cl solution 250
Effect of concentration of potassiumchloride on the extraction of M*Au(CN)5ion pairs by solvents, polyethers andion-exchangers 256
Minimum-energy structures ofCH3 (CH 2OCH5):zCH 3 and its Cs* complex.
Dark atoms represent oxygen, light atomsrepresent carbon, and shaded atoms
represent metals 260
Minimum-energy structures of the alkali-
metal complexes of CH3 (CH2OC H 2 ) i ahigh-dielectric environment. Dark atomsrepresent oxygen, light atoms represent
x x x i v
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
35/203
carbon, and shaded atoms represent metals 265
Extraction of sodium ions by activated
carbon at 25eC in the presence of cyanide orhydroxide ions
Equilibrium isotherm for the distributionof aurocyanide between activated carbon anda solution containing 0,2M NaOH and 0,2MNaCN at 35°C
Rate of elution of gold from activatedcarbon at 95°C
Variation of the rate of elution with themass of carbon and flowrate of the eluant
Rate of elution of gold from activatedcarbon by the AARL procedure
Effect of aurocyanide concentration in theeluant feed, on the rate of elution at 95 6C
Plot of mg gold eluted against time, forelutions with various constant aurocyanideconcentrations in the eluant feed
Elution of aurocyanide from a 5ml carbon bedby the Zadra method, with a single-passextraction efficiency of 0,42 .(The solid linerefers to the theoretical concentrationprofile calculated using Equations (5.20)
to (5.22))a) Plot of gold concentration on the carbon
against timeb) Plot of gold concentration in the eluate
solution against time 294
280
283
284
285
287
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
36/203
295
297
299
300
302
304
305
310
312
Elution of aurocyanide from a 25 m 1 carboncolumn with no pre-equilibration prior toelution. (The solid line n.fers to thetheoretical concentration profile calculatedaccording to the model)
Elution of aurocyanide from a 25 ml carboncolumn equilibrated with eluant solutionprior to elution. (The solid line refers tothe profile calculated according to themodel)
Elution of aurocyanide from a 5ml carboncolumn pre-equi1ibrated with eluant solutionat various flowrates
Variation of the rate of elution withtemperature
The effect of the ionic strength of theeluant on the rate of elutior at 956C
The effects of the concentration of cyanideor hydroxide on the rate of elution at 956C
Variation of the rate of elution with eluantconcentration at a constant ionic strengthof 1,2 mol/kg at 956C
Effect of cyanide decomposition on theelution of aurocyanide from activated carbonat 956C in a batch reactor
Effect of cyanide concentration on the rateof elution at 95*C, in the presence of0,2M NaOH
Distribution of copper species at pH 10 with
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
37/203
314
315
321
323
324
325
329
331
333
339
340
343
variation in cyanide concentration
Distribution of copper species at pH 7 with
variation in cyanide concentraLion
Effect of pyrolysis temperature of S-761polymeric adsorbent on aurocyanide capacity(after Adams et a l* 2)
Effect of pyrolysis temperature of S-761polymeric adsorbent on skeletal density
Effect of pyrolysis temperature of S-761polymeric adsorbent on micropore volume
Effect of pyrolysis temperature of 6-761polymeric adsorbent on pore sizedistribution
Effect of pyrolysis temperature of S-761polymeric adsorbent on surface area
Effect of pyrolysis temperature of S-761polymeric adsorbent on conductivity
X-ray diffractogram of a typical activatedcarbon (Le Carbone G210)
Effect of pyrolysis temperature of S-761polymeric adsorbent on reduction potential
Variation of measured reduction potentialfor R03515 activated carbon with time
Effect of pyrolysis temperature of S-761polymeric adsorbent on phenol activity
Infrared spectra of pyrolysis products of
x x x v i i
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
38/203
S-761 polymeric adsorbent after activationat different temperatures
6.12 Infrared spectra of commercial activatedcarbons studied
6.13 Infrared spectra of Le Carbone G210activated carbon after different oxidationtreatments:(a) None(b) Boiled in 1M Naocl, ih(c) Boiled in 1M HN03, ih(d) Boiled in 1:1 HN03:H2S0e, 2h(e) Soaked in 20% H a0 2 for I week at 2f°C
6.14 X-ray diffractograms of grapnite before andafter contact with an aurocyanide solutioncontaining 0,1M HCl
6.15 Infra \ spectrum of graphite after contactwith ar, aurocyanide solution containing0,1M HCl
6.16 Scanning electron micrograph of graphiteafter contact with an aurocyanide solutioncontaining 0,1M HCl
6.17 Infrared spectrum of polyxanthene product(a) untreated(b) after contact with a 0,1M HCl solution
for 24 hours(c) after contact with a 0,1M NaOH solution
for 24 hours(d) after contact with a 0,1M NaCN solutionfor 24 hours
6.18 Infrared spectrum of polycuinone product(a) untreated
x x x v i i i
o
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
39/203
(b) after contact with a 0,114 HCl solutionfor 24 hours
(c) after contact with a 0 ,1M NaOH solutionfor 24 hours
(d) after contact with a 0,1M NaCN solutionfor 24 hours 372
M X Is
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
40/203
CHAPTER 1 INTRODUCTION
The carbon-in-pulp (CIP) process for extraction of goldfrom cyanide leach liquors, which has undergone widespreadapplication in recent yea rs1 # 2, is generally the preferredmethod for recovery of gold in all new gold plant s.However, many aspects of the chemistry involved at variousstages of the process remain poorly understood. Even themechanism of adsorption of the gold species onto activated
carbon, the key step in the process, is not fullyunderstood. The aim of this Thesis is to resolve thesequestions through a fundamental study of these unknownareas.
1.1 The Adsorption of Aurocyanide onto ActivatedCarbon - An Historical Perspective
Carbon has been used as an adrorbent throughout history,
with the ancient Hindus using it for water purification2.Several centuries ago, it was used extensively in theremoval of colours from solutions and for adsorption ofgases. It was not until 1847 that its adsorptiveproperties fcr gold were first discovered3 . In 1880,Dav is4 patented a procsss for recovery of gold fromchlorination leach liquors ueing wood charcoal.
Several years later, in 1890, MacArthur and the Forrestbrothers discovered cyanide to be a good lixiviant for
go ld 5, and soon thereafter, Johnson* in 1894 patented theuse of wood charcoal for the recovery of gold from cyanide,solutions. Cya.'idation eventually became the standardtechnique for leaching gold, which was subsequentlyrecovered by cementation onto zinc. This remained the
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
41/203
process of choice until the 1970's.
During this lengthy period, carbon was never usedextens ively for gold recovery, except for a brief perio d7at the Yuanmi Mine, Australia, in 1917. The unpopularityof the process during these years is partly due to the pooradsorptive properties cf charcoals produced in those days,as compared with the properties of modern activatedcarbons. Another factor w»s lack of a suitable method forthe elution of gold from the activated carbon afteradsorption.
A viab le e lution process was developed by Zadra8 in 1952,
that paved the way for the first large-scaleimplementation1 of carbon-in-pulp (CIP) , at the HomestakeMine, U.S.A., in 197: .
Carbon-in-pulp for gold recovery has greatly increasedsince tha t time, aided by the advances ma de 1'10 by Mintek(Council for Mineral Technology) in engineering aspects of
the process. Activated carbon has been used for severalyears for recovery of gold from clear solutions; however,the simplicity of CIP, in which carbon granules are added
direc tly to the cyanided pulp and recovered by screening,has been the most popular new process over the lastdecade11.
CIP epitomises the type of innovative technology used inthe extractive metallurgy of gold, evidenced by currentvariat ions on the conventional process, such as carbon-inleach11 (CIL), carbon-in-pu lp- in-column12 (CIPIC), andother innovations, such as heap leaching1 and resin-in-pulp 13'14 (RIP). Engineering and technological aspectsof carbon adsorpt ion processes for gold recovery have beensuccessfully developed and, for the most part,commercialized. However, after a hundred years of
development, the fundamental chemistry involved in manyaspects of the process, remains largely unknown.
2
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
42/203
1.2 The Carbon-in-pulp Procesr - Process Chemistry and Unknown Areas
CIP relies on one fact - activated carbon exhibits aremarkable ability to adsorb gold as the aurocyanide anion.
Au(CN): This forms he basis of the concentration,separation and recovery of gold on gold plants world wide.
A typical flowsheet for CIP is shown schematically inFigure 1.1. It comprises the following major unitoperations:
a) leaching,b) carbon adsorption,c) carbon acid-washing and elution,d) gold electrowinning or precipitation, and
e) carbon reactivation.
A description of each of these unit operations follows,with emphasis on the chemistry involved, and unknownaspects thereof.
a* Leaching of Gold from the Ore
Crushed ore is contacted with a so lution of sodium cyanide,typically about 0,01 mol/1, in the presence of oxygen, andfor a period of about twenty hours. Gold is dissolved insolution as the soluble aurocyanide ion, Au(CN)j , by anelectrochemical process described by an anodic dissolution
reaction:
Au + 2CN“ * Au(CN ) 2 + e (l.l)
in conjunction with a cathodic reduction of oxygen:
3
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
43/203
Ore
Air Cyanide
{ " Leaching
Adsorption
CarbonCarbon
Loadedcarbon
Carbon
Carbon
Tailingsto waste
Cyanide
CausticElution j-
Carbon Eluate
Barrensolution
Fiectrowinning
- j ~Screening jGold and silver
Screening
Screening
Adsorption
Screening
Reactivation
torefinery
Fig. 1 .1 Typical flowsheet of a CIP plant for the recoveryof gold (after McDougall and Fleming21)
4
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
44/203
0 2 + 2H20 + 4e" * 40H' (1 . 2 )
These reactions result in a "mixed potential", EB , at which
leaching will occur, this is illustrated schematically inFigure 1.2.
Overall features of t 2 reactions involved here are wellestabli shed15. However, certain details of the reactionmechanisms are uncertain, and the reaction rate can beslowed considerably by a diverse range of factors:
(i) In some instances, this slowing of the leaching ratecan be directly attributed to the depletion of cyanide andoxygen in the pulp, due to reaction with variousconstituents of the pulp.
(ii) Carbonaceous material present in the pulp has beenfound to adsorb aurocyanide ions. This effect can beeradica id by oxidation with nitric acid16 orchlor ine17'18 ; however, the reason for this effect is notyet clear.
(iii) It is generally accepted that dissolution of gold incyanide solution occurs via the formation of an adsorbedAuCN species:
Au + CN"* = AuC Na 8 + e" (1.3)
AUCN,4,, t CN" * AU(CN); (1.4)
A passivating layer of AuCN has been found by severalworkers15 to occur at a potential of about -0,4 V, depending
on the presence of heavy metal impurities in the solution.Passivation is not normally observed in practice? however,when sulphides are present, it has been postulated tooccur. U l m e r 19 suggests this to be due to an AuCN layer(Equations 1.3 and 1.4) . Fink and Putnam20 ascribe it toan A u 2S layer. The possibility also exists that sulphide
5
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
45/203
Cur ren tAu + 2CN" ^ Au ICN) - + e -
Anodic
Ca thod ic2 0H
Poten t ia l
Fig. 1.2 simplified schema ic diagram of the mixedpotential model for the dissolution of gold incyanide solutions (after Nicol, et al15)
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
46/203
merely depletes the cyanide concentration in the pulp, viaEquation 1.5, for example:
2S*- 4- 2CN" + 0 2 + 2 H 20 * 2SCN" + 40H" (1.5)
Several other lixiviants for gold, apart from cyanide, havebeen considered for specific applications21. Thesereagents include, amongst others, thiourea, thiosulphate,chlor ine , bromine and iodine. Hone of these have yetachieved any widespread use, however.
to) Adsorption of Aiirocyanidp onto Act iva ted Carbon
From the leach tanks, cyanided pulp is pumped to a cascadingseries of adsorption tanks containing about 25 g/1 granularactivated carbon, that is retained in the tanks by screensthrough which pulp passes freely. Carbon is periodicallymoved countercurrent to the flow of pulp.
Residence time in each stage and number of stages employedare variables that are usually built into the design for aparticular target performance of the plant.
The current disagreement regarding the mechanism ofadsorption of aurocyanide onto activated carbon is a keyissue, and will be dealt with in detail in this Thes is . Thechemical nature of the gold cyanide adsorbate has neverbeen unequivocally established, but recent wo rk 2 2 suggeststhat the formation of M n +[Mi(CN) 2 ] n ion pairs may beimportant. Little is known at present regarding the ionassociation behaviour of aurocyanide, however.
c) Carbon Acid Washing and Elution
Gold-leadad carbon from the first adsorption tank istransferred to an elution column, where it is contacted
7
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
47/203
1
with a strong caustic cyanide eluant solution of variablecomposition, depending on the elution procedure employed.The tempera ture is also dependent on this factor, but in
general, high temperatures (> 90°C) are employed. Priorto elution, loaded carbon is often contacted with a hot acidsolution (c. 10 % (m/m) HCl) to remove precipita ted CaC03,fines and slimes from the carbon pores. The chemistry ofthe elution process has never been addressed in any deta il,and no elut ion mechanism has yet been postulated that cansatisfactorily account for all observed facts. Moreover,elution of gold from activated carbon by sodium sulphidesolution has been demonstrated2 3 to be potentiallyefficient at room temperature, but a fundamental
understanding of this procedure has yet to be reached.
d) Gold Electrovinning or Precipitation
Gold recovery from pregnant eluate solution can be achievedby electrowinning, either in a single-pass of eluatethrough the ce ll , or in a continuous circuit with theelution column. Alternative /, the gold can be recoveredby zinc cementation. Both of these are well established
procedures. The chemistry is well known; for example, thecementat ion of gold onto zinc proceeds via reactions (1.6)and (1.7):
A u (CN) j
Zn + 4CM"
= Au + 2 o r
Zn(CN)j
(1 6)
(1-7)
Carbon Reactivation
In practice, it has been found that activated carbonreturning to the c.dsorption circuit after elution is apoorer adsorbent for gold than the fresh carbon. Thereason for this is the poisoning of the carbon by various
8
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
48/203
organic compounds present in the pulp. Humates, fulvates,hydrocarbons and many other organic compounds exhibitstrong adsorpt ion onto activated carbon2 4, thereby
blocking sites for gold adsorption.
This poisoning has necessitated a thermal reactivationtreatment of the carbon when necessary, before beingreturned to the adsorption circuit. Typically, the carbonis heated to about 750®C in the presence of steam for half anhour, but once again, whe parameters used seem to varydepending on the whims of the individual operator. Thistreatment essentially involves vaporization of volatileadsorbates and burn-off of non-volatile adsorbates.
f) Other Features of the CIP Process
Apart from the aspects of the unit operations discussedabove, there are various other features of CIP of whichlittle is known regarding the fundamental chemistryinvolved. These include, amongst others, the chemistry ofthe decomposition of cyanide throughout the CIP flowsheet,the effect of hot acid treatment on the loaded carbon, and
the mechanisms of adsorption of other complexes ontoactivated carbon, either present in the leach pulp, oruseful for elucidation of the aurocyanide adsorptionmechanism. The influence of activated carbon surfacechemistry and structure on the process is another importantaspect. The rurface chemistry of activated carbon iscomplex24 - even the identities of the surface functionalgroups are uncertain.
The current commercial success of the carbon-in-pulpprocess is largely due to decreased capital and operatingcosts. The improved efficiency of gold recovery by CIPover the zinc precipitation process is most marked whentreating calcines and other materials that are difficult tofilter. Carbon-in-pulp is by no means the ideal process.
9
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
49/203
however - the harsh conditions and relatively long timesrequired to elute the gold from the carbon, the need forhigh-temperature thermal regeneration and the fouling ofactivated carbon by organic and inorganic materials in thepulp, are some of the more practical factors that could bedrastically improved upon.
1.3 The structure and Chemistry of Activated Carbon
Activated carbon is a generic term that encompasses afamily of highly porous, amorphous carbonaceous materials,none of which can be characterized by a structural formulaor chemical analys is25. Modern activated carbons are farsuperior in terms of abrasion resistance and adsorptionactivities, as compared with the wood charcoals used in thepast. They are used for the adsorption of a range ofcompounds2*'2*'22, both organic and inorganic, gaseous andin solution. Activated carbons also have numerousapplications as vutalyst supports28.
a) Physical Structure of Activated Carbon
The single structural factor that results in the adsorptiveproperties of activated carbon is the extensive porestructure. Most commercial activated carbons havespecific surface areas of about 800 to 1200 m 2/g, which ispredominantly contained with? n micropores less than 2 nra indiameter. The material contains a complex pore networxthat is normally categorized as follows:
macropores : 30 < Dp < 10 000 nmmesopores : 2 < D p < 30 nramicropores : Dp < 2 nra,
where Dp is the pore diameter.
1 0
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
50/203
The pore size distribution depends on the type of rawmaterial and the conditions of manufacture. Activatedcarbons can be made from virtually any carbonaceous
material, such as coconut shells, peach pits, bituminouscoals and peat. The raw material is first carbonized at atemperature not exceeding 700°c, and this charring processis followed by activation at temperatures between 800 and110 l °C in the presence of suitable oxidizing gases, such assteam, air or carbon dioxide. The activation process isresponsible for creating the pore structure, by virtue ofburn-off of more reactive components of th* char. Thematerial also acquires a disordered structure of graphiticplatelets known as microcrystallites, shown schematicallyin Figure 1.3. Also shown is the structure of puregraphite.
b) Chemical Structure of Activated Carbon
The edges Df the microcrystal lites formed during theactivation process contain unfilled carbon valencies, thatreact with the oxidizing gas to form a wide variety ofoxygen-containirg functional groups. The basal plane ofthe microcrystall ite is relatively inert, so the surface ofactivated carbon is heterogeneous in nature, with bothhydrophilic and hydrophobic domains. This propertyexplains the wide variety of compounds that are stronglyadsorbed onto activated carbon.
Activated carbons used for gold recovery are generallyprepared by the high-temperature process outlined above.Such carbons are basic in character, by virtue of their
ability to strongly adsorb acids, and are popularlyclas sif ied2f as H-carbons. Carbons prepared under lowtemperature conditions are acidic in character, and areclassified as L-carbons. These acid-base properties ofactivated carbons result from the preponderance offunctional groups of either acidic or basic character on
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
51/203
S
Fig. 1.3a Schematic representation of the structure of
graphite. The circles denote the positions ofcarbon atoms, whereas the horizontal lines
represent carbon-to-carbon bonds
o--
,o~
o-
Fig. 1.3b Schematic representation of the structure ofactivated carbon. Oxygen-containing organicfunctional groups are locatad at the edges ofbroken graphitic ring systems (after Mattsonand Mark 2 4)
12
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
52/203
the surface. The fact that all carbons adsorb both acidsand alkalis to some extent shows that both types offunctional groups are normally present.
The precise identities of the functional groups presentremain the topic of some debate, but ome of the most likelycandidates are shown in Figures 1.4 and 1.5.
Activated carbons are known30 to be electrical conductors,and this property i s due to the re latively high degree ofcondensation of aromatic rings, resulting in an expensivedeloca lized electron cloud. Electron Spin Resonance(BSR) studies3 1 indicate the presence of free radicals.Thc.se properties, together with the presence of functionalgroups that can undergo electrochemical reactions, such asquinones, impart a reduction potential to the activatedcarbon. Different carbons art found32 to exhibi t a rangeof different reduct ion potentials, betv. en +0,06 V and+0,40 V (relative to the Standard Hydrogen Electrode) .Hence, activated carbon exhibits both electrochemical andcatalyt ic properties. For example? it can readily effectreduction of AuCl% and Ag+ to the respective metal:;, and
can also act as an oxidation catalyst for Fe 2 * to F e3* .
t The Chemistry of Gold - An Introduction
The key to the natural occurrence of gold in the metallicform, as well as its many practical uses in jewellery and inindustry, is its nobility. It is the only metal that doesnot corrode in air in the presence o£ either oxygen orsulphur.
The aqueous chemistry of gold is of great importance co theextractive metallurgist, and an introduction to this isuseful in this Thesis. The important oxidation states arethe aurous (+1) and the auric (+3) . The free uncomplexedions are thermodyn tically unstable in aqueous solution.
13
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
53/203
C a r b o x y l i c g r o u p P h e n o l i c h y d r o x y l g r o u p
Q u i n o n e - t y p e c a r b o n y l g r o u p N o r m a l l a c t o n e g r o u p
F l u o r e s c e i n - t y p e l a c t o n e g r o u p C a r b o x y l 1 c a c i d a n h y d r i d e g r o
C y c l i c p e r o x i d e g r o u p
Fig. 1.4 Acidic functional groups postulated to bepresent on the surface of activated carbon(after Cookson26)
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
54/203
C h r o m e n e R e n z o p y r y l i u m io n( C a r b o n i u m io n)
C O - “ o > - - a > -
B e n z o p y r / H u m d e r i v a t i v e s
0 --"-P y r o n e - l l k e s t r u c t u r e s .Fig. 1.5 Basic functional groups postulated to be present
on the surface of activated carbon (after
Cookson26)
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
55/203
and are reduced by water to meta llic gold. There areseveral ligands that form stable complexes wi th gold, and afew of those of importance in gold metallurgy have already
been mentioned. Table 1.1 lists the revelant stabilityconstants of these complexes.
Table l.l
Stability Constants* for a Selection of Complexes of
Gold(I) and Gold(III)
Gold(I) Gold(III)
Complex 8 , Complex 64
A u (CN); 2 x lo3e Au(CN ) 4 1 0 ''Au(S 2O3 )|~ 5 X 10 !8 a u i ; 5 X lO”
Au(CS(NH 2 ) 3)2 2 x 1 0 ” Atl(SCN); 1 0 *'
A u I 2 4 x 1 0 ” AuBr,' 1 0 3 !a u ( s c n ); 1 , 3 X 1C 17 AUCI; 1 0 ,e
AuBr; 1 0 iaa u c i ; 1 0 *
*Values are taken from Nicol, et al15.
Gold(I) complexes have the 4 fl45d10 electronicconfiguration and are diamagnetic. The preferredcoordination number of gold(I) is two, forming linearcomplexes. Vary stable complexes are formed33 with scftligands', such as cyarfde (AufCN) ̂ -* 8; = 2xl038) 5- InCwxoride solution, where a hard ligand is involved, theAuCl 2 species is somewhat unstable, tending todisproportionate according to the following reaction:
3AuCl g * AuCi; 4- 2Au° + 2 01" (1.8)
Gold(III) complexes have the 4fI45d8 electronicconfigurat ion, and are also diamagnetic. In this case.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
56/203
more stable complexes are formed with hard ligands, -such aschloride (AuCl̂ ; S 4 - 1026)E. The preferredcoordination number for gold(III) is four, with squar?
planar complexes usually being formed.
Other aspects of the cbenistry of gold, both well known andnovel, wil l be discussed later in this Thesis. It will beshown how these features of the chemistry of gold areimportant in determining its extractive metallurgicalcharacteristics, and in particular the remarkableselectivity of activated carbon for the adsorption ofAu(CN ) 2 over other metal cyanide anions.
1.5 Current Processes for the Recovery of Gold
In the past, gold has been collected from placer (surface)deposits by gravity techniques, due to its high specificgravity, sometimes aided by mercury amalgamation. Suchprocessing techniques are relatively unimportant today -ii the United States in 1986, 94% of the 3,733 million
ounces of gold produced was processed by cyanidation34.
The two most prevalent gold processes in use today arecarbon-in-pulp and zinc cementation, and both of these havebeen discussed earlier in this chapter. GIF processes arethe most popular on all new gold plants; hence theimportance of a fundamental study of tha processesinvolved.
There are alternative configurations to the typical CIPprocess. Carbon-in-leach (CIL) and carbon-in-solut ion(CIS) are two versions that find many applications. Theformer involves the placing of carbon in the leach tanks toinstil more efficient leaching, and this variation is mostappropriate when the pulp contains carbonaceous ("preg-robbing") materials that also have an affinity for goldadsorption. The latter version involves passing a
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
57/203
clarified leach liquor through a /.ad or column of carbon?this version is most appropriate when treating leachliquors from heap leach operations.
Heap leaching is without doubt the most obvious and simpleway to deal with certain lean ores and wastes, and involvesthe percolation of cyanide solution through a large heap ofceea. Pregnant solution is allowed to run out from thebottom of the heap, and the gold is then recovered.
Oeaqpite the current popularity of the CIP process, there isauch room "or improvement, and the search is on foralternative configurations, such as the carhon-in-pulp-
in-coluar. (CIPIC) configuration proposed by MintekresearcherF12, and the horizontal flow variation, proposedby Davy-McKee workers35.
The search is also on for alternative adsorbents toactivated carbon. The use of convention; 1 anion-exchangeresins in a res in-in-pulp (RIP) variation are reported36 tobe employed in full-scale plants in the Soviet Union. Thefirst reports13'14 of such an innovation being operatedsuccessfully in the western world have recently been made.It is evident that for certain applications, res in-in-pulpis more viable a process than carbon-in-pulp.
The relatively recent advent of carbon-in-pulp on acommercial scale is probably the single most importantinnovation in gold metallurgy since the introduction ofcyanidation in 1890. It is the aim of this Thesis toaddress the many unknown areas still associated with thisprocess, in particular the mechanism of adsorption ofaurocyaride onto activated carbon.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
58/203
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
59/203
derived from coconut shells, with an average particlediameter of about 2 mm. The surface area as determined bythe N, BET method is 1100 to 1200 B !/g.
The polymeric adsorbents S-763 and S-862 were supplied byDuolite. S-761 is a phenol-formaldehyde matrix , and S-862 a polystyrene matrix. The polymeric adsorbent XAD-8,which comprises an acrylic ester matrix, was supplied byRohm and Haas. These adsorbents have surface areas ofabout 450 m 2/g.
The soluble poly(oxyethylene) used in this study was thenon-ionic surfactant Tri ton X-100, which has the structure
(CH,),CCH;C(CH,),- -0-(CH2CH 20 ) io h .
2.2 Activated Carbon Adsorption studies
a) Equil ibrium Adsorption
For the experiments in section 3.2a, 1,0 g of carbon wasequilibrated in a rolling bottle for 3 days with 400 ml of asolution containing 50 to 700 mg of gold, and 0,1 mol ofhydrochloric acid or potassium chloride, per litre.Samples of the solution were analysed for gold by atomic-absorption spectrophotometry (AAS). The carbon wasanalysed for gold by X-ray fluorescence spectroscopy(XRFS) , for potassium, by AAS, and for nitrogen by a Heraeuselemental analyser, as well as a LZCO nitrogen analyser.The estimated error in all analyses is about 5 per cent.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
60/203
b) Elution of the Carbon
Prior to some elutions 3 ,0 g of loaded carbon was washed in a
rolling bottle with 250 ml of a hydrochloric acid solution(0,1 mol/1) for 24 hours. No gold was detected in the washsolution. In other instances, 1,0 g of loaded carbon wasboiled under reflux in 250 ml of solution containing 4 percent hydrochloric acid (by mass) for varying periods oftime. No gold was detected in the pre-treatment solution.The carbon was subsequently washed to neutrality withdeionized water.
The carbon was then eluted as follows: approximately 1 g ofloaded carbon was placed in an insulated, jacketed columnthat was heated by hot (906C) water circulating through the
jacket. A fresh solution of sodium hydroxide (0,1 mol/1)was siphoned slowly through the carbon bed until analysisof the eluate solution showed the gold content to be below0,1 mg/1. The carbon was analysed for gold k>y XRFS beforeand after the elution.
Eluates from reveral of the elutions were analysed forAu(CN) 2 by UV spectrophotometry. Absorbances weremeasured at 239,5 nm on a Beckman Acta MIVSpectrophotometer, and were found to follow Beers law, thusyielding a standard curve.
c) Precipitation of AuCN
Fresh 250 ml solutions of KAu(CN) 2 containing 1000 mg goldper litre, in 0,1 mol/1 hydrochloric acid were placed in a
magnetically-stirred , round-bottomed flask with water-cooled condenser, and heated to the relevant temperatureusing an oil bath. The solution was analysed periodicallyfor gold by atomic-absorption spectroscopy.
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
61/203
d) Decomposition of AuCN
AuCN that was produced in the previous experiment waswashed well with ethanol and water and dried under vacuumfor twenty four hours. AuCN (0, 05 g) was placed either in arolling bottle or under reflux as above, and contacted with20 ml of de-ionized water adjusted with hydrochloric acidor sodium hydroxide to pH values in the range 1,0 to 13,0.Solutions were then analysed for gold by atomic absorptionspectroscopy, and for cyaniae and cyanate by ionchromatography. The solution was further characterized byultraviolet (UV) spectrophotometry on a Beckman Acta MIVspectrophotometer, and the solid residue by X-ray
diffractometry (XRD) using a Philips PW1050 X-raydiffractometer,
e) Oxygen Effect Experiments
Typically, l,0g of carbon was contacted with 250ml ofsolution containing the appropriate additive. Highpurity gas (oxygen or nitrogen) was bubbled through thesolution vigorously enough to achieve efficient mixing.
For each experiment, the temperature was kept at a constanto,5 ± 0,016C by means of a Hetotherm PFS23 thermostat inconjunction with a Grant cooling unit, to minimizeevaporation losses. Samples of solution were analysedfor gold by atomic absorption spectrophotometry (AAS), andthe pH value was measured using a Labion Model 15 pH/mVmeter in conjunction with an Orion combinatibn pHelectrode. Carbon was analysed for gold by X-ray
fluorescence spectroscopy (XRFS), for nitrogen using aHeraeus elemental analyser as well as a LEGO nitrogenanalyser, ard for chloride using silver ni trate turbidity.Solutions were analysed for NH{, HCOj and CNO" by ionchromatography.
22
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
62/203
f) Gold (1X1) Cyanide ScparLmnntm
KAu(CN) 4 war, prepared by the method of Smith, et al37.KAu(CN) . was dissolved ?.»i water and reacted with a slight,excess of t roaine, with m e excess being re!F jvea by heatingto 80‘C. lae y.Vj.'Nj Pr, lit the resultant solution wasthen reacted wi’.H » stclchionetric pr;tasf»iu» cyanidesolution, to yiei i KAu(C*! ( in solution. In theadsorption experiments, 0,25 g of carbon was equilibratedin a rollinc battik for tw n t y hours vith 100ml of solutioncontaining JOrag of gold pe: litre 'is Kftu(CH), crKAu(CN) 2), as well as patassim, ch ’o -idu calcium chlorideor hydrochloric acid. Samples of solution and car ban vereanalyzed for gold, potassitn and nitrogen as inscribedabo ve. A carbon that; was loaded with KAu(CK) , wan alsoboiled in 250ml of 4't hydrochloric acid for four bourr;,washed to neutrality and dried xr. a vacuum desiccator, torsubsequent investigation.
gj Silver and Myrci.ry Sxperiients
Carbon ( 1,0 g) was equilibrated in a rolling bottle for 72
hours with 400 ml of a solution contain ing 31.0 ec of silverper litre (as KAg(CN) ,), *8 well as 0,1 mcl oi potassiumchloride or hydrochloric acid per litre. In the case ofmercury, the solution contained 700 ag of mercury per litre(as Hg(CN) j) e ither alone, or with the molar equivalent ofCa (C Nj . The carbon was analysed for silver, mercury,potassium and nitrogen as described above. In a SeconIseries of experiments, 0,25 g of carbon was equil ibrated ina rolling bottle for 24 hours with 100 ml of a solutioncontaining either 150 mg of silver or 300 mg of mercury, as
well as 0 to 1,0 mol of potassium cyanide per litre. Theionic strength was kept at a constant 1,0 mol/kg, byaddition of the required amounts of potassium chloride.Samples of solution were analysed for silver by atonic-abisorption spectrophotometry ( M S ) . Silver was also
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
63/203
loaded onto the carbon (0,25 g) from silver nitratesolution (0, 0255 M AgM03 , 20 ml) in a roll ing bottle for 24hours, and mercury from a 0,01 M HgCl, solution.
In the investigation of the effects of acid and basetreatment, loaded carbon (0,5 g) was boiled under reflux in250 ml of solution containing 4 per cent hydrochloric acid(by mass) for five hours. The carbon was subsequently
washed to neutrality with deionized water. A portion ofthe carbon was then boiled in 250 ml of 0,1 M sodiumhydroxide for a further five hours. The carbon was then&nalyF»d for silver and nitrogen as above.
h) Miscall aiieous Techniques
(1) X-ray diffractometry
X-ray diffractograms of loaded and treated carbons wereobtained using Cu K tt radiation, on a Philips PW1050/25 X-ray diffractometer.
(i V Scanning electron microscopy
Scanning electron micrographs of loaded and treatedcarbons were obtained using a Hitachi scanning electronmicroscope (3EM), as well as a Jeol 840A SEM.
(lii) Fourier Transform infraredSpectrophotometry
Samples of activated carbon for infrared spectrophotometrywere prepared by contacting 0,5 g of carbon with 25 ml ofsolution containing typically 0,003 moles of gold asKA u (CN) % and 0,03 moles of either LiCl, KC1, CsCl or CaCl2in a rolling bottle for 24 hours. In the cases of Ag and Hg
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
64/203
a similar procedure was followed.
Samples were crushed using a mortar and pestle prior todrying. Approximately 0,8 mg of dried carbon powder wasmixed with 200 mg of dry KBr by shaking manually. Themixture was made into a disk by placing it under 10 tonnes ofpressure for about 10 seconds.
Infrared spectra were obtained using a Perkin-Elmer 1725XFourier Transform Infrared Spectrophotometer. For eachspectrum, 200 scans were made in the transmission mode, at aresolution of 4 cm- 1. No mathematical smoothing functionswere performed on the spectra; however the ordinate of eachspectrum was normalized to facilitate comparisons betweenspectra. All spectra were baseline-flattened to enhancespectral detail.
(iv) 197Au Mossbauer spectroscopy
This technique is not commonly used, which necessitates amore detailed discussion of the underlying principles. Itis cnly in the last ten yea rs38 that sufficient development
has been made in the field of 1 * 7Au Mossbauer spectroscopyto facilitate application of the technique to practicalproblems. The technique is a powerful one, providinginformation both on oxidation state and chemicalenvironment of the gold atom, provided suitable referencedata are available. Recent applications have been somepreliminary work39,40 on the nature of the auroc^anideadsorbate on activated carbon, and a study of the chemicalstate of gold in gold ores41.
The principle of the Mossbauer effect ir? essentiallysimilar to the adsorption of infrared radiation in infrared(IR) spectrophotometry; however in the case of Mo$$sbauerspectroscopy, 'V-rays are used. These ir-rays are emittedduring an isomeric transition of a nucleus in a radioactive
25
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
65/203
source, and are subsequently absorbed by a nucleus of thesame isotope, causing it to undergo a transi tion from itsground state to the respective excited state, and thisphenomenon is called whe Mossbauer effect.
To observe a Mossbauer effec t in I,7Au, the radioactivesource used is l,7Pt, prepared from isotopically enrichedl96Pt that has been irradiated with neutrons in a nuclearreactor. The decay scheme for the l97Pt isotope is shownin Figure 2.1. The relatively short half-li fe of 18 hoursnecessitates prompt measurement of the Mossbauer effect,which may take several days to yield a statistica lly usefulresult, depending on the gold concentration in the sample.Some measurements may thus require two or three sources inorder to provide a reasonable spectrum. Figure 2.1 showsthat 90 percent of the 1’7Pt decays to the 77,3 keV excitedstate of 197Au, which decays to the ground state withemission of Tf-radiation of energy 77,3 keV. When thisradiation reaches the gold sample of interest in theabsorber holder, some of the radiation is absorbed by 1 ’ 7Auatoms in the samp la undergoing the transition to theexcited state, and measurement of the intensity of theradiation exiting from the absorber yields the percentage
transmission, which has a similar meaning to that for IRspectrophotometry.
The resonant absorption of 7-radiation by l97Au nuclei canonly take place if they do not lose energy due to recoilduring emission or absorption, i.e. the atoms must beembedded in a solid matrix. Moreover, the recoil-freefraction, f, decreases with increasing o âa energy andwith increasing temperature. The relati J y high 77,3 keVgamma energy of 1 9 7Au makes it necessary to jool both source
and absorber down to about 4 K in a liquid hel ium cryostat.
The 7-radiation produced is very monochromati c42 , ?nd inorder to provide a spectrum, a variation in frequency, andhence energy, of the radiation passing through the sample.
26
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
66/203
18 h
1,89
Fig. 2 .
^ P tT T -
90% 10%
ns
1 D"cy mch«. for th. :"pt l.otop. uwd .# asource for 1’ 7au Mossbauer spectroscopy
2 7
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
67/203
must be produced. This is achieved via the Doppler effect.The source is given a velocity that results in a slightchange in energy of the radiation:
Ey - E 4 (1 + v/cliy>, (2.1)
whure E| is the energy of incident radiation, is theenergy of 7-radiation from the stationary source, v is theapplied velocity, and c hr is the velocity of light. Only asmall energy range is required for this application. Thedata are normally plotted in units of velocity.
Mossbauer spectra of gold compounds normally consist of a
doublet, or quadrupole, as shown in Figure 2.2, which showsthe Mossbauer spectra for the gold cyanide compounds AuCN,KAu(CN) 2 and KAu(CN)4 . The quadrupole splitting, QS,which is the separation between the two absorption peaks ofthe doublat, is a measure of the asymmetry of distributionof electronic charge about the gold nucleus,intermetallic systems usually show zero or very smallvalues of QS, due to the high symmetry of their structures.An example, that of metallic gold, is also shown in Figure2 . 2 .
Another important parameter in a Mossbauer spectrum is theisomer shift, IS, the position of the centroid of thedoublet. The isomer shift is a measure of total electrondensity at the gold nucleus. An increase in IS is alsocharacteristic of an increase in covalency in bonds toneighbouring atoms. Metallic gold has the lowest electrondensity of the Au nuclei, and hence exhibits the lowestisomer shift of all known gold compounds and alloys, viz. -1,23 mm/s.
A comparison between the Mossbauer fcpectra of differentgold compounds in Figure 2.2 suggests that identificationof unknown gold species is possiole, since the IS and QSparameters vary significantly from one gold species to the
2 8
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
68/203
Fig. 2
Q M
096
0,95
KAutCN).
Q90
1.00
0,98
0,96
I 04 0 04 0.8D o p p l er v e l o c i t y, c m / s
•2 1,7A u Mossbauer spectra of representativegold cyanide compounds. (After Faltens andShirl ey4 3)
29
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
69/203
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
70/203
10-1
(a).
-8
- aV‘5.
-1 0
-1 2
- 2 - 1 0 1 2 3 4 S 0 7
Isomer Shift (mm.s’-1)
Fig. 2.3 Correlation of isomer shifts and quadrupolesplittings for gold(l) and gold(IH)compounds, with gold in different chemicalenvironments. Isomer shift values are
j relative to Au in Pt source. (After Cashionj O , et al 3 9)
!
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
71/203
Fig. 2.4 Plot of IS and QS for l17Au Mossbauerspectra of MAuCl„ salts against cationicradius
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
72/203
free fraction, f , which was mentioned earlier. It is onlyfrom a knowledge of the value of f for the relevant speciesunder the experimental conditions, that any quantitative
data can be obtained. For example, the va lue47 of f (whichis sometimes also called the Debye-Waller factor) formetallic gold is 0,19 , whereas that for crystallineKAu(CN)2 is 0,015.
Samples for Mossbauer spectroscopy were prepared bycontacting 20g of carbon with 200 ml of solution containingtypically 2000 mg of gold per litre, rolling in a bottle fortwenty hours and washing with cold de-ionized water. Onesample was subsequently boiled for twenty hours in 250 ml of
hydrochloric acid solution containing 40% HC1 (mm) andwashed with cold de-ionized water. Half of this carbon wasthen boiled in 250 ml of 0,3M NaOH solution for five hours,cooled and washed with cold deionized water. Both fractionswere then dried in a vacuum desiccator over silica gel.All MoFxibauer spectra were deconvoluted at the TechnischeUnivc ^itat Hunchen, using computer methods.
(v) X-ray photoelectron spectroscopy (XPS)
In this technique, the surface of the sample is irradiatedwith X-rays, and the energy distribution of the ejectedelectrons is measured. The common acronyms for thetechnique are ESCA (Electron Spectroscopy for ChemicalAnalysis) and XPS (X-ray Photoelectron Spect roscopy).Analysis of the resultant spectrum of electron countsagainst photoelectron kinetic energy, can provide usefulinformation regarding chemical and electronic environmentof elements in the material, i.e. valence state, bindingenergies of electrons, etc.
The kinetic energy of the photoelectron produced, Ep, isrelated to the energy of the incident photon, hr, and thebinding energy of the electron in its particular shell, Eb ,
A 33
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
73/203
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
74/203
Surface analysis was performed in a multitechnique VacuumGenerators "Solar 300" UHV chamber, which was fitted with aVG "ctAM 100* (Combined Lens Analyser Module) comprisingXPS and AES facilities. The main chamber was kept at a basepressure of 5 x 10" 10 mbar with an Edwards £04 oil diffusionpump frunning on polyphenyl ether fluid) with a watercooled baffle fitted with a VG CCT150 liquid nitrogencryostat and an Edwards EDM12 rotary backing--pump. The X-ray source made use ot magnesium Ka radiation. Samples forXPS wrre mounted as powders on double-sided sticky tape.Samples of pure compounds were mixed with powderedactivated carbcn by physical shaking, and binding energies
were referenced against the carbon (Is) peak in activatedcarbon, which was assigned as 284,4eV. Deconvolution ofspectral data was achieved by use of a modified version ofPARAFIT4 8 , a parametric least squares curve-fit tingsubroutine, on an Apple lie microcomputer.
2.3 Solution Studies
a> Potentiometric Titration
A 50 ml aliquot of 0,005 mol/1 HAu(CN) % which was preparedas described above, was titrated against 0,01062 mol/1 NaOHsolution previously standardized against 0,0100 mol/1 HC1.The titration was monitored using a radiometer PHM 64research pH meter with an Orion combination pH electrode.The solution was stirred magnetically throughout thetitration
bj Ccnductivity Measurements
Measurements were made using a Wayne-Kerr B642 Autobalancebridge in conduction with a Radiometer CDC 344 Immersion
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
75/203
Conductivity cell comprising three non-platinizedplatinum ring electrodes with a cell cons tant of 3,318 cm.The temperature was kept at a constant 25,00 ± 0 ,019C by
means of a Hetotherm 02PF 623 UO thermostat/bathcombination in conjunction with a Grant cooling unit.
All solutions were made up using water of conductance lowerthan 1,5 x 10*® S/cm. Measurements were made by a serialconcentration method, starting off with pure water andadding weighed aliquots of the relevant salt solution.
c) Infrared Spectrophotometry
samples of LiAu(CN)2 and NBu4 (CN)2 were placed in pyrexcontainers and dried in a vacuum oven overnight at 85°C.The containers were subsequently stoppered and allowed tocool in a vacuum desiccator over dried silica gel. Themethyl ethyl ketone (MEK) and tetrahydrofuran (THF)solvents used in this study were dried over molecularsieves (4 8), that had been dried for four days i t 140°C andcooled in vacuo. Infrared spectra of the dried solventsdisplayed no bands at ~ 3300 cm* : that would be attributable
tr the presence of water. Solutions were made up rapidly toensure the shortest possible exposure time to theatmosphere. Spectra were recorded within a few hours ofmaking up the solutions. Solutions were stored in adesiccator when not in use. IR spectra were recorded on aBeckman IR4260 Research Infrared Spectrophotometer. A cellwith CaF; windows and path-length adjustable to 0,221 mm,supplied by Research and Industrial Instruments Company,England, was used.
d) Distribution experiments
In these experiments, 20 ml of water-saturated organicphase was equilibrated with 20 ml of aqueous phase in a 100
3 6
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
76/203
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
77/203
however, the results were considered sufficiently accuratefor the present purposes.
g; Determination of Water Content by Automatic Karl-Fischer Titration
The solubilit ies of water in several solvents, mixtures andsolutions of Triton X-1C0 in the solvents, were determinedby automatic Karl-Fischer titration. A radiometer TTA 80automatic titration assembly was linked to an Apple liemicrocomputer. The titration was checked by comparison ofthe measured soluble ity of water in nitrobenzene with that
reported in the literature4 9 . Good agreement was obtained.
2.4 Molecular Mechanics Calculations
a) The Force Field
Molecular mechanics calculations on alkali metal complexesof poly(oxyethylene) ligands were implemented on a version
of MOLBLD-3 due to Boyd50, modified to reproduce the unitedatom force-field AMBER5 1. AMBER has been used extensivelyto explain the propert ies of prote ins52, nucleic at Ids83,the alkali metal complexes of spherands5 4 and, appropriateto the present application, the alkali metal complexes of18C6, a cyclic poiyether55. AMBER uses a potentialfunction of the form:
r *.(>->,)'->■ Z Ti l +»N- 2
” W 7)] * E I Ve " - -W* + — 1‘I Vv J
8/18/2019 Adams Michael David 1989-001-todo carbon.pdf
78/203
where the nonbonded terms are summed over all atom puirsseparated by more than three bonds♦ A "united atom"approximation56 was used, In which aliphatic Cff2 groups arerepresented as single atoms. This has the advantage of
substantially reducing the number of interactions, andspeeding up computer refinement. Van der Waals tarms wereinput into the modified M0Lt)D3 as 12/6 Lennard Jonesfunctions. Egg - A/rj j12 - r/ r,} 6, where the coeffieents A and C were derived from group and atomic polarizabilities,effective united atom atomic numbers, and effective ionic
radii by substitution into the Slater-KirJewood equationsas described by Scott and Sheraga51, Nonbonded parametersfor the CH, group, the 0 atcn* and the K* a?«d Cs* ions were
taken from Wipff et a.ls 5 , a..d those for Li* were taken fromKollman et a l 54. Ethereal torsion parameters were takenirom Wipff efc a l 55, and bond streching and angle bandingparameters were taken from Kollman et 5 3. The partialatomic charge on complexed oxygen was taken by Wipff et a l 5 5to be - 0,6. Carbon atoms bended to the oxygen had acompensating charge of +0.3 to mainte/n the ligandelectrical neutrality. The cutoff distan' for nonbondedinteractions