Thesis (MScEng)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: A 2 stage heap leach process to extract base and precious metals from the Platreef ore is currently being investigated industrially. A first stage bioleach is used to extract the base metals. In the 2nd stage, cyanide is used as the lixiviant at high pH to extract the platinum group metals and gold. By analogy with current gold recovery practices, the present study investigates the preferential and quantitative adsorption of precious metals (Pt, Pd, Rh and Au) over base metals (Cu, Ni and Fe) from an alkaline cyanide medium, by means of granular activated carbon. Experiments were designed statistically to optimise the process parameters using synthetic alkaline cyanide solutions close in composition to those expected from plant leach solutions. The statistical approach allowed the development of a reliable quantitative approach to express adsorption as a response variable on the basis of a number of experiments. A 2IV(7-2) fractional factorial design approach was carried out in a batch adsorption study to identify significant experimental variables along with their combined effects for the simultaneous adsorption of Pt(II), Pd(II), Rh(III) and Au(I). The adsorbent was characterized using SEM-EDX, and XRF. Precious metals adsorption efficiency was studied in terms of process recovery as a function of different adsorption parameters such as solution pH, copper, nickel, free cyanide ion, thiocyanate, initial precious metal (Pt, Pd, Rh and Au) ion and activated carbon concentrations. It was shown that adsorption rates within the first 60 minutes were very high (giving more than 90% extraction of precious metals) and thereafter the adsorption proceeds at a slower rate until pseudo-equilibrium was reached. Among the different adsorption parameters, at 95% confidence interval, nickel concentration had the most influential effect on the adsorption process followed by the adsorbent concentration. Adsorption of Ni was found to proceed at approximately the same rate and with the same recovery as the precious metals, showing a recovery of approximately 90% in two hours. The kinetics of Cu adsorption were slower, with less than 30% being recovered at the 120 minute period. This suggests that the co-adsorption of Cu can be minimised by shortening the residence time. Adsorption of Fe was found to be less than 5%, while the recovery of Rh was negligibly small. The effect of thiocyanate ion concentration was not as important as the effect of free cyanide ion concentration but still had some influence. The correlation among different adsorption parameters was studied using multivariate analysis. The optimum experimental conditions resulted in a solution with pH of 9.5, [Cu(I)] of 10 ppm, [Ni(II)] of 10 ppm, [CN ] of 132.44 ppm, [SCN ] of 98.95 ppm, [PMs] of 2.03 ppm and [AC] of 10 g/L. Under these conditions, predicted adsorption percentages of Pt, Pd and Au were approximately 98, 92 and 100%, at the level of 95% probability within two hours as an effective loading time. The negative values of ΔG° for all ions under optimum conditions indicate the feasibility and spontaneous nature of the adsorption process. Chemisorption was found to be the predominant mechanism in the adsorption process of Pt(II), Pd(II) and Au(I). Based on their distribution coefficients, the affinity of activated carbon for metal ions follows the selectivity sequence expressed below. Au(CN) > Pt(CN) > Pd(CN) > Ni(CN) > Cu(CN) Finally, it is important that additional research and development activities in the future should prove the economic viability of the process. Future work is also needed to investigate the adsorption of precious metals (PMs) by comparing the efficiencies and kinetics of adsorption when using sodium hydroxide (in this study) or lime, respectively, in order to control the pH. / AFRIKAANSE OPSOMMING: ʼn Tweefasige hooploogproses vir die ontginning van basis- en edelmetale van die Platrif-erts word tans industrieel ondersoek. ʼn Eerstefase-bioloog word gebruik om die basismetale te ontgin. In die 2de fase word sianied gebruik as die uitloog by hoë pH om die platinum-groepmetale en goud te ontgin. Na analogie van hedendaagse goudherwinningspraktyke het die huidige studie die voorkeur- en kwantitatiewe adsorpsie van edelmetale (Pt, Pd, Rh en Au) bo basismetale (Cu, Ni en Fe) vanuit ʼn alkaliese sianiedmedium met behulp van korrelrige geaktiveerde koolstof ondersoek. Eksperimente is op statistiese wyse ontwerp om die parameters van die proses te optimaliseer deur van sintetiese alkaliese sianiedoplossings wat in hulle samestelling nou ooreenstem met dié wat van oplossings van plant-loog verwag word, gebruik te maak. Die statistiese benadering het die ontwikkeling van ʼn betroubare kwantitatiewe benadering om adsorpsie as ʼn responsveranderlike op grond van ʼn aantal eksperimente uit te druk, moontlik gemaak. ʼn 2IV(7-2) -Fraksionele faktoriale ontwerp-benadering is tydens ʼn lot-adsorpsiestudie gevolg om beduidende eksperimentele veranderlikes tesame met hulle gekombineerde uitwerkings vir die gelyktydige adsorpsie van Pt(II), Pd(II), Rh(III) en Au(I) te identifiseer. Die adsorbeermiddel is met behulp van SEM-EDX en XRF gekenmerk. Adsorpsiedoeltreffendheid van edelmetale is bestudeer ten opsigte van proseskinetika en herwinning as ʼn funksie van verskillende adsorpsieparameters soos oplossing-pH, koper, nikkel, vry sianiedioon, tiosianaat, aanvanklike edelmetaal (Pt, Pd, Rh en Au)-ioon en geaktiveerde koolstofkonsentrasies. Daar is aangetoon dat adsorpsietempo‟s binne die eerste 60 minute baie hoog was (het meer as 90% ekstraksie van edelmetale opgelewer) en daarna het die adsorpsie teen ʼn stadiger tempo voortgegaan totdat pseudo-ekwilibrium bereik is. Onder die verskillende adsorpsieparameters, by 95%-vertroubaarheidsinterval, het nikkel-konsentrasie die grootste invloed op die adsorpsieproses gehad, gevolg deur konsentrasie van die adsorbeermiddel. Daar is bevind dat die adsorpsie van Ni teen nagenoeg dieselfde tempo en met dieselfde herwinning as die edelmetale voortgegaan het, wat ná twee uur ʼn herwinning van nagenoeg 90% getoon het. Die kinetika van Cu-adsorpsie was stadiger, met minder as 30% wat teen die 120-minute-tydperk herwin is. Dit dui daarop dat die ko-adsorpsie van Cu tot die minimum beperk kan word deur verkorting van die verblyftyd. Daar is bevind dat die adsorpsie van Fe minder as 5% is, terwyl die herwinning van Rh onbeduidend klein was. Die uitwerking van die konsentrasie van die tiosianaatione was nie so belangrik as die uitwerking van die konsentrasie van vry sianiedione nie maar het steeds ʼn mate van invloed gehad. Die korrelasie tussen verskillende adsorpsieparameters is met behulp van meerveranderlike analise bestudeer. Die optimale eksperimentele toestande het gelei tot ʼn oplossing met ʼn pH van 9.5, [Cu(I)] van 10 dpm, [Ni(II)] van 10 dpm, [CN] van 132.44 dpm, [SCN] van 98.95 dpm, [EM‟e] van 2.03 dpm en [AC] van 10 g/L. Onder hierdie toestande was die voorspelde adsorpsiepersentasies van Pt, Pd en Au nagenoeg 98, 92 en 100%, op die vlak van 95%-waarskynlikheid binne twee uur as ʼn doeltreffende laaityd. Die negatiewe waardes van ΔG° vir alle ione onder optimale toestande dui op die uitvoerbaarheid en spontane aard van die adsorpsieproses. Daar is bevind dat chemiesorpsie die deurslaggewende meganisme by die adsorpsieproses van Pt(II), Pd(II) en Au(I) is. Gebaseer op hulle distribusiekoeffisiënte volg die affiniteit van geaktiveerde koolstof vir metaalione die selektiwiteitsvolgorde soos hieronder voorgestel. Au(CN) > Pt(CN) > Pd(CN) > Ni(CN) > Cu(CN) Laastens, dit is belangrik dat addisionele navorsing en ontwikkelingsaktiwiteite in die toekoms die ekonomiese haalbaarheid van die proses bewys. Werk in die toekoms is nodig om die adsorpsie van edelmetale (EM‟e) te ondersoek deur vergelyking van die doeltreffendhede en kinetika van adsorpsie wanneer natriumhidroksied (in hierdie studie) of kalk, onderskeidelik, gebruik word ten einde die pH te beheer