Upgrading of PGM-rich leach residue by high pressure caustic leaching

Date
2015-12
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: There is a lack of clear understanding of the rate of selenium (Se), arsenic (As) and sulphur (S) dissolution during caustic (NaOH) batch leaching of PGM-rich leach residue in the presence of oxygen. This has been a limitation in the optimisation of hydrometallurgical processes for the upgrading of PGM concentrates before refining the precious metals. Conditions to improve the rate of leaching of amphoteric elements while minimizing PGM losses were examined to enhance the performance of the leaching process. Development of intrinsic leaching rate equations represent the core of the overall batch leaching model developed in this study. The robustness of the model was assessed by its ability to accurately simulate the effects of changing operating parameters on the reaction extents. The effects of the interfacial oxygen mass transfer rate and temperature on the leaching rates were therefore also included in the overall model. The first part of the experimental program focussed on the interfacial oxygen mass transfer rate in the test autoclave. This enabled an accurate mathematical description of the interfacial mass transfer rate of the primary oxidant, diatomic oxygen (O2) molecule from the gas to the liquid phase. Mass transfer tests were conducted using the sodium sulphite method at 60°C, 100 kPa oxygen partial pressure and agitation speed of between 500 to 1000 rev/min. Cobalt(II) was used as the catalyst with a concentration range of 1 to 5 mg/L. Oxidation of amphoteric elements was investigated by leaching of PGM-rich leach residue (residue from sulphuric acid leaching of converter matte) in caustic solution. The test work was conducted to determine the intrinsic leaching rates in 0.125, 0.25 and 0.5 mol/L NaOH solutions in the 160° to 190°C temperatures range over a period of 6 hours. Oxygen partial pressure was maintained at 11 atm in the factorial experiments. The effect of oxygen partial pressure was quantified by conducting tests with oxygen partial pressure ranging from 7 to 16 atm. The intrinsic rate constants and activation energies derived from this test work were incorporated in the overall kinetic model to simulate the batch leaching profiles under real plant conditions. During the caustic pressure oxidation of amphoteric elements, the rate of oxidation was rapid during the first 10 minutes and decreased steadily over the course of experiment. The experimental results suggest that the oxidation kinetics are controlled by product layer diffusion with sulphur, selenium and arsenic (Arrhenius) activation energies of 31.8 kJ/mol, 26.1 kJ/mol and 10.7 kJ/mol respectively over the temperature range of 160 to 190°C. The reaction mechanism, as well as the observed kinetic behaviour, is most likely due to the base metal/PGMs hydroxide layer that formed as a result of precipitation. An increase in temperature increased the sulphur and arsenic reaction rates. The selenium reaction rate also increased as the temperature was increased from 160 to 175°C. A further increase in temperature above 175°C did not yield a significant increase in the reaction rate. An increase in the caustic concentration increased the reaction rates of all the elements. Increased oxygen partial pressure also improved the reaction rates, with the most significant change observed for sulphur oxidation; the extent of sulphur oxidation increased from 75 to 85% when oxygen partial pressure was increased from 7 to 16 atm. Reaction orders of 0.25, 0.12 and 0.21 with respect to hydroxide concentration and 0.37, 0.29 and 0.36 with respect to dissolved oxygen concentration were obtained for sulphur, selenium and arsenic respectively. Kinetic models were developed for sulphur, selenium and arsenic extraction. The sulphur and selenium simulation gave better agreement between the experimental and model predicted values, while the arsenic simulation gave a relatively poor prediction of the extractions. The caustic concentration had a notable effect on the dissolution of the PGMs. An increase in the caustic concentration increased the dissolution of platinum, palladium and ruthenium. Ruthenium dissolution also increased with an increase in temperature. To the contrary, platinum and palladium dissolution decreased with an increase in temperature. Rhodium and iridium precipitated and did not report in the solution phase while osmium could not be traced. The oxygen partial pressure did not have a significant effect on the dissolution rate of platinum, palladium and ruthenium.
AFRIKAANSE OPSOMMING: Daar is ‘n tekort aan die begrip van die tempo van seleen (Se), arseen (As) en swawel (S) oplosbaarheid gedurende bytsoda (NaOH) enkelladingsloging van platinum groep metaal (PGM)-ryk oorskot materiaal in die teenwoordigheid van suurstof. Hierdie inligting word benodig wanneer die optimisering van tipiese hidrometallurgiese prosesse wat PGM oorskot materiaal opgradeer verlang word. Hierdie bytsoda druklogingsproses vind tipies voor raffinering van die PGM metale plaas. Kondisies wat die tempo van amfoteriese element-loging verbeter, terwyl die PGM verliese geminimaliseer word, was in hierdie werk geondersoek om sodoende die effektiwiteit van die logingsproses te verbeter. Die ontwikkeling van intrinsieke logingtempo vergelykings vorm die kern van die algemene enkelladingsloging model wat ontwikkel was. Die robuustheid van hierdie model word geevalueer op sy vermoë om akkuraat die effekte van veranderende bedryfstelsel parameters op die logingstempo van betrokke reaksies te simuleer. Die effekte van suurstof tussenvlak massaoordrag en temperatuur was ook in die algehele model ingesluit. Die eerste deel van die eksperimentele program het gefokus op die suurstof tussenvlak massaoordrag in die outoklaaf. ‘n Akkurate wiskundige model wat die massaoordrag van die primêre oksidant, diatomiese suurstof (O2), van die gas fase na die vloeistof fase beskryf, was gebruik om die suurstof oordragtempo te kwantifiseer. Suurstof massaoordrag toetse het van die natrium sulfiet metode gebruik gemaak by 60°C, 100 kPa suurstof parsiële druk en tussen 500 en 1000 rev/min roerspoed. Kobalt(II) het gedien as katalis wat tussen 1 tot 5 mg/L gevarieer was. Die amfoteriese element oksidasie was volgende ondersoek deur die PGM-ryk oorskot materiaal te loog met bytsoda (wat stroomop onderwerp was aan swawelsuur loging van omskakelaar mat). Die toetswerk wou die intrinsieke logingtempo’s met 0.125, 0.25 en 0.5 mol/L NaOH oplossings by temperature 160 en 190°C oor 6 uur residensie tyd vasstel. Die suurstof parsiële druk was konstant gehou op 11 atm in hierdie faktoriale eksperimente. Die effek van suurstof parsiële druk was apart vasgestel, deur die suurstof parsiële druk te varieër vanaf 6 tot 16 atm. Die intrinsieke tempokonstantes en aktiveringsenergieë wat in hierdie toetswerk afgelei is, was in ‘n algehele kinetiese model ingekorporeer wat die enkellading logingsprofiele gesimuleer het by aanleg kondisies. Die tempo van oksidasie was vinnig in die eerste 10 minute en het geleidelik afgeplat, gedurende die bytsoda druk oksidasie van amfoteriese elemente. Die eksperimentele resultate suggereer dat produklaagdiffusie die oksidasie kinetika beheer met swawel, seleen en arseen (Arrhenius) aktiveringsenergieë as volg bereken in die temperatuur interval 160 tot 190°C: 31.8 kJ/mol, 26.1 kJ/mol en 10.7 kJ/mol. Die reaksie meganisme en kinetiese gedrag word hoogs waarskynlik veroorsaak deur die onedelmetaal/PGM hidroksied laag wat deur middel van presipitasie gevorm het. Temperatuur toename het die swawel en arseen se reaksietempo’s verhoog. Met seleen het die reaksietempo met temperatuur toename tussen 160 en 175°C ook verhoog, maar afplatting het by 175°C opwaarts plaasgevind. Oor die algemeen het die bytsoda konsentrasie die amfoteriese elemente se reaksietempo’s verhoog. Die verhoging van die suurstof parsiële druk het ook die reaksietempo’s verhoog. Swawel oksidasie het van 75 tot 85% verhoog vanaf 6 tot 16 atm, wat die mees noemenswaardige verandering was. Swawel, seleen en arseen reaksieordes van 0.25, 0.12 en 0.21 met hidroksied konsentrasie en 0.37, 0.29 en 0.36 met opgeloste suurstof konsentrasie het die beste paslyn op die eksperimentele data tot gevolg gehad. Hierdie data was gebruik om die kinetiese modelle van swawel, seleen en arseen ekstraksie te ontwikkel. Terwyl swawel en seleen ‘n goeie paslyn vir die eksperimentele data tot gevolg gehad het, kon passing van arseen ekstraksie nie ‘n goeie model oplewer nie. Varierende bytsoda konsentrasie het ‘n noemenswaardige effek op die PGM ontbinding gehad. Wanneer die bytsoda se konsentrasie vermeerder word, los daar meer platinum, palladium en rutenium op. Rutenium ontbinding het tydens ‘n temperatuur toename verhoog. In kontras het platinum en palladium ontbinding velaag tydens temperatuur toename. Rodium en iridium het gepresipiteer en was nie ontbind nie. Osmium kon nie gemeet word nie. Die suurstof parsiële druk het nie ‘n noemenswaardige effek op platinum, palladium en rutenium ontbinding gehad nie.
Description
Thesis (MEng)--Stellenbosch University, 2015.
Keywords
Leaching, Hydrometallurgy, UCTD
Citation