The emission of the platinum group elements Pt, Pd and Rh (PGE) from automobile catalytic converters has led to rapid increases in Pt, Pd and Rh concentrations in roadside media. The vast majority of previous research examining autocatalyst-derived PGE in the urban environment has been performed in Europe or North America. Although catalytic converters became mandatory on all new cars sold in Australia from 1986, no prior studies have focussed on urban platinum group element (PGE) concentrations in Australian environments. In general, the results of previous studies suggest a limited post depositional mobility of catalyst derived PGE. However, these findings are from research conducted in cool-temperate climate zones with regular rainfall and from environments where soils and sediments differ from the typically coarse grained, sandy soils with low levels of organic matter found in Perth. The relevance of European and North American findings to other regions with different climates and soils is therefore unclear and where the climate regime and properties of soils and sediments are not comparable to those previously studied, the potential exists for different geochemical behaviour of autocatalyst-derived PGE. Through investigations of spatial and temporal distribution and the identification of some of the main factors controlling transport and fixation, the principal aim of the research presented in this thesis was to elucidate aspects of the post depositional geochemical behaviour of autocatalyst derived PGE in selected roadside environments in Perth, Western Australia. The quality of some of the reported PGE data has been questioned by a number of workers. Possibly the most intractable diffculty in the determination of low concentrations of PGE in environmental samples by ICP-MS is the control of interferences from common matrix components. To ensure accurate and reliable data in this research, prior to the analysis of environmental samples, the optimal instrumental conditions for PGE determination and two commonly applied matrix separation methodologies (tellurium coprecipitation and ion-exchange) were investigated. The most effective matrix separation technique for the accurate determination of PGE in the environmental samples applicable to this study, such as road dusts and roadside soils, was found to be cation exchange. The lack of knowledge regarding urban PGE concentrations in an Australian context was addressed through examinations of PGE levels in road dusts, roadside soils and infiltration basin and wetland sediments. Data show significant elevation of all three PGE above local background and average upper crust values. PGE ratios in surface road dusts and soils were consistent with known catalytic converter compositions and while Pt and Rh concentrations are comparable with European studies, Pd levels were generally higher in these Australian samples. The effect of climate on PGE levels in roadside environments was investigated by repeat sampling of road dusts and roadside soils over a twelve month period. Both sample media exhibited seasonal variations. The presence of seasonal variability in PGE concentrations in roadside soils suggests that this environmental compartment does not represent a long term accumulative matrix for autocatalyst-derived PGE. Further examination of spatial distribution revealed that the PGE exhibit greater vertical mobility in the soils of Perth than has previously been reported, with elevations above local background concentrations occurring at depths of 14-20cm. Neither small scale spatial variability nor vertical mobilisation were of su±cient magnitude to explain the observed temporal variability. Based on the pattern of seasonal PGE distribution and that of rainfall, temporal fluctuations are attributed to transport by stormwater. The mobilisation of PGE by stormwater is thought to occur principally via the water-mediated transport of PGE bearing particulates. However, PGE fractionation leading to a greater post-depositional mobility of Pd may occur during transport through the urban stormwater system. In the urban environment of Perth, infiltration basin and wetland sediments represent a sink for autocatalyst-derived PGE. Based on the examination of PGE ratios, and the vertical distribution of PGE in infiltration basin sediments, Pt and Rh remain associated, whereas Pd may be differentially mobilised. For both soils and infiltration basin sediments, variation in pH was limited and does not show any correlation with vertical profiles of PGE, suggesting that pH does not act as a major control on PGE mobility. The role of organic matter is less clear, and although no straightforward relationships were apparent, where high levels of organic matter were present, profiles suggest an increased mobilisation of Pd. This differential mobilisation of Pd may therefore be caused by the formation of an organo-metallic species. Temporal fluctuations in PGE levels in road dusts and roadside soils indicate that inputs of PGE to aquatic environments are likely to occur as seasonal pulses. The routing of road runoff into urban wetlands therefore represents a major pathway by which aquatic ecosystems may be exposed to autocatalyst-derived PGE. The impact of such inputs is unclear, however, as other recent studies have shown that a portion of autocatalyst-derived PGE, and especially Pd is bioavailable, the potential for ecosystem degradation due to PGE contamination represents a major avenue for further research.

In this study, various techniques of hydrometallurgy and ultrasound-assisted solvent extraction were used to recover the platinum group metals (PGM¡¦s) from a composite sample of honeycomb-type autocatalysts. After they were removed from the converter casings, the autocatalyst substrates were first crushed and then ground by a ball mill. The recovery procedures employed are shown as follows: (1) dissolve PGM¡¦s from ground spent autocatalyst by aqua regia leaching; (2) separate PGM¡¦s from base metals in the aqua regia leachate by metal cementation using zinc powder so that PGM¡¦s can be precipitated out; (3) the PGM¡¦s precipitate was first dissolved by aqua regia, then proceed to remove nitrate and hydrochloride within. The residue was further dissolved in hydrochloride acid as a preparation step for solvent extraction; (4) the PGM¡¦s pregnant solution of hydrochloride acid was treated by solvent extraction and stripping to separate and purify each component of PGM¡¦s. Effects of ultrasound agitation on the efficiency of solvent extraction was also evaluated in this work. Results of aqua regia leaching experiments have shown that the quantity of dissolved PGM¡¦s increased as the solid-to-liquid ratio decreased. The maximum dissolved quantity of PGM¡¦s could be obtained by a 3-hr leaching time. At this stage, the PGM¡¦s recoveries are 80-90% for platinum and rhodium and greater than 99% for palladium. The result of a preliminary test has indicated that acetic acid can not effectively separate the PGM¡¦s and base metals. Thus, the method of cementation by zinc powder was employed to separate PGM¡¦s from base metals. Before cementation, the aqua regia leachate was diluted and pH-adjusted to greater than 2. In so doing, an almost complete cementation (>99%) could be obtained by the least quantity of zinc powder. In addition, the base metals occurred with the PGM¡¦s precipitate have been minimized except lead and zinc. While palladium was extracted by di-n-octyl sulfide (DOS), ultrasound assistance has rendered a complete extraction within a few minutes. At this stage, the extraction efficiency was found to be independent of the HCl concentration. It was found that platinum and rhodium were not extracted by DOS. When platinum was extracted by tri-n-octylamine (TOA) and assisted by ultrasound, rhodium will be extracted at the HCl concentration higher than 4M. Thus, TOA is not an effective chemical for selective extraction of platinum. TOA was then replaced by tributyl phosphate (TBP). Experimental results have indicated that the extraction of platinum using TBP was affected by the HCl concentration. The best result was obtained when the HCl concentration was 5M. Extraction by TBP was found to be fast. It took only 20-30 seconds to reach the equilibrium even with no ultrasound assistance. But multi-stage extractions are generally required to extract platinum completely. Rhodium was found to be not extracted by TBP. After palladium and platinum were extracted, only rhodium was remained in the reffinate. In summary, solvent extraction using DOS and TBP has made it possible to separate palladium, platinum, and rhodium effectively. In the palladium stripping solution almost no base metals was determined. However, zinc and lead were found in the platinum stripping solution and the rhodium-containing raffinate. These base metals should be removed to achieve a better purity for each precious metal. The TCLP (i.e., a leaching test for toxicity) result of the autocatalyst substrate after aqua regia leaching has found to be non-hazardous. However, several streams of wastewater and acid gas generated in the recovery process should be properly managed to avoid the secondary pollution.

Includes abstract. / Includes bibliographical references (leaves 74-79). / Historically the platinum group metals (PGMs) have been, and are still being dissolved by means of rather aggressive methods, e.g. aqua regia. Limited research has been conducted into the dissolution of the PGMs using different oxidizing agents. The dissolution of gold on the other hand has been afforded extensive research, and numerous papers and review articles have been published on the subject. The last number of years has seen the biggest application by volume of the PGMs as part of autocatalysts towards the degradation of harmful motor vehicle exhaust gases. This has subsequently sparked research into the recovery of specifically platinum, palladium, and rhodium from spent autocatalysts. Currently pyrometallurgical recovery of PGMs is being employed predominantly. A hydrometallurgical process on the other hand is, based on current technology, still a rather aggressive process that makes for high maintenance costs and an unpleasant environment. Gold has traditionally been dissolved by making use of cyanide, which is still the major route for gold dissolution. Due to environmental concerns lixiviants such as thiosulphate (S2O3 2-), thiourea (H2NCSNH2), and thiocyanate (SCN-) are gaining acceptance due to them being more environmentally friendly and giving good recoveries. These ‘softer’ alternatives have however not been tested on the PGMs. It is therefore the aim of this study to obtain an improved understanding of the leaching of the PGMs using lixiviants less aggressive than aqua-regia. These lixiviants include (i) SCN-, (ii) S2O3 2-, (iii) H2NCSNH2, and (iv) AlCl3/HCl. A thermodynamic study highlighted the fact that thermodynamic data for platinum-, palladium- and rhodium complexes are basically non-existent. To therefore obtain a clearer thermodynamic understanding of the leaching of the platinum group metals by means of these alternative lixiviants, future detailed speciation and thermodynamic investigations need to be conducted. An exploratory electrochemical investigation focusing on open circuit potentials and potentiodynamic scans, showed AlCl3 / HCl / NaOCl to be a good candidate for the leaching of the platinum group metals followed by SCN- / Fe3+ and CS(NH2)2 / Fe3+. Actual leach results, employing virgin autocatalysts as sample material, again highlighted the potential of AlCl3 / HCl / NaOCl as being a good lixiviant system. The surprise package, however, has been SCN- / Fe3+ that rendered very good results for Pd and Pt.