One of the environmental concerns associated with mining waste is the contamination of soil. This study addresses the decontamination of soil, particularly of Co and Ni using Berkheya coddii (B. coddii). B. coddii is a hyperaccumulater plant that is able to decontaminate Co and Ni from the contaminated land. The use of B. coddii to decontaminate soil or waste must be based on a cognizance of the complicated, integrated effects of pollutant sources and soil-plant variables. Phytoextraction pot trials using B. coddii were carried out under green house condition, with controlled watering. A contaminated metallurgical waste residue known as Rustenburg Base Mine Refineries waste (RBMR waste soil) collected from Rustenburg while a serpentine (native) soil (N soil) where B. coddii grows naturally was collected from Mpumalanga. The experiment involved the addition of sulfur doses to both soils in order to test whether acidification and higher sulfur availability could enhance the uptake of both Co and Ni by B. coddii. The results indicate that the addition of sulfur from 2.0 to 8.0 g per kilogram decreased pH in both substrates. RBMR waste soil pH was found to have decreased from 7.8 to 7.4 while the N soil pH was found to have decreased from 6.4 to 4.7. The reduction oxidation potential (redox potential) in both substrates was observed to have decreased along with the increase in sulfur dosage. The mean redox potential for RBMR waste soil was found to be 350 mV and 506 mV for the N soil after the addition of sulfur. Conductivity increased along with the increase in sulfur dosage in both substrates. The mean conductivity for the N soil was found to be 961 μS/cm while that of the RBMR waste soil was found to be 1453 μS/cm after the addition of sulfur. The decrease in soil pH was significant (p = 0.00115) in the N soil than RBMR waste soil. Despite the increase in sulfur dosage and decrease in soil pH in both substrates, B. coddii observed growing. Although it was evident that B. coddii is able to grow in the RBMR waste soil, it was observed that the RBMR waste soil limits the root depth of the B. coddii, reducing chances for the roots to penetrate into the ground especially when dry. The RBMR waste soil becomes more compacted than the N soil when dry. It is therefore crucial to ensure that there is enough moisture to allow for the B. coddii being able to survive effectively in the RBMR waste soil. B. coddii plant height in the RBMR waste soil after four months was observed to be in the range of 190 to 200 mm tall. This was found to be less than the height observed for the B. coddii planted in the N soil, which was in the range of 350 to 400 mm. Nonetheless, plants grown in both substrates were able to absorb Ni and Co into their tissues. More Co and Ni were found to have accumulated into the leaf tissues than in other parts of the plant. This could be an advantage since one would harvest only the leaf part or the canopy (shoots) and allow B. coddii to resprout in order to continue taking up more Co and Ni from the same waste substrate to remediation levels that could be stipulated by Government as desirable for the ecosystem and the protection of human health. Although the accumulated Ni and Co can be recovered from biomass, this alone might not provide sufficient economic justification for phytoextraction due to the low concentrations that could be recovered. B. coddii was found to absorb higher concentrations of Co and Ni from the N soil than from the RBMR waste soil. However, the results found in this study may not be conclusive. This could be due to many variables that could control metal uptake which were not investigated. These include mycorrhizal fungi and metal forms in the soil. Moreover, this study was performed in a green house and not in the outdoor environment. Ni is generally toxic to most plants, hyperaccumulators (i.e. B.coddii) contain elements that nullify the toxic effect of nickel, and in this case the accumulated nickel is bound to malate to form a harmless nickel complex which could be absorbed by the plants as nutrients.

As pollution becomes an ever-increasing threat to the global environment pressure is being placed upon industry to "clean-up" its act, both in terms of reducing the possibility of new pollution and cleaning up already contaminated areas. It was with this in mind that Amplats embarked on a phytoremediation project to decontaminate nickel-polluted soils at one of their mine sites in Rustenburg, using the nickel hyperaccumulating plant, Berkheya coddii, which is endemic to the serpentine areas near Barberton, Mpumalanga. Besides the applied aspects pertaining to the development of the phytoremediation process we were also interested in more academic aspects concerning the transport and storage of nickel within the plant tissues. In order that the progress of nickel could be followed through the plant, a radio-tracer of ⁶³nickel was placed in the soil and its movement within the plant followed by analysing the plant material, at set intervals, using a liquid scintillation counter. From these studies it was found that the nickel appeared to be transported from the roots to the leaves of the plant via the xylem. It appeared that the nickel was not confined to the leaf to which it was initially transported and so movement of nickel within the phloem also appears to occur in B. coddii. As nickel is generally toxic to most plants, hyperaccumulators contain elements that nullify the toxic effect of nickel. In the case of Berkheya coddii it is thought that the accumulated nickel is bound to malate to form a harmless nickel complex. With this in mind an assay for L-malic acid was developed in order that any effect on L-malic acid, caused by growing Berkheya coddii on soils containing various concentrations of nickel, could be determined. This method also enabled comparisons of L-malic acid concentrations to be made between hyperaccumulators and non-hyperaccumulators of various plant species. From the L-malic acid comparisons it was found that the nickel concentration within soils affected the levels of L-malic acid within B. coddii and that the levels of L-malic acid within B. coddii were greater than that of a closely related non-hyperaccumulator, suggesting that L-malic acid is indeed involved in the hyperaccumulation mechanism within B. coddii. B. coddii was chosen as the tool in nickel phytoremediation at Rustenburg Base Metal Refineries as it was found to accumulate up to 2.5% nickel in the dry biomass, it grows rapidly and has a large above-ground biomass with a well developed root system, and it is perennial and so does not need to be planted each season. Earlier work had shown that the nickel levels in the roots were comparatively low (up to 0.3% nickel in the dry material) and thus, for ease of harvesting and to ensure the continued vegetative growth of the plant on the planted sites, it was decided that the leaves and stems of the plants would be harvested at the end of each growing season. The plant was also found to accumulate low levels (0.006 - 0.3 %) of precious metals, including platinum, palladium and rhodium, within its above ground biomass, making it attractive for the remediation of certain soils that contain low levels of these metals. Before B. coddii could be introduced to the Rustenburg area a comparison of the climatic and soil conditions of Barberton, the area to which B. coddii is endemic, and Rustenburg needed to be made to ensure that the plant would be able to survive the new conditions. These comparisons showed that Rustenburg receives on average, 484 mm less rain per year than Barberton, indicating that irrigation was required when the Rustenburg sites were planted out with B. coddii, in order to reduce water stress. Rustenburg was also found to be, on average, 4.6°C warmer than Barberton, but as B. coddii growth responds to wet/dry seasons, as opposed to hot/cold seasons, it was not felt that this temperature difference would have a negative effect on the growth of the plants. The soil comparisons showed the contaminated Rustenburg sites to be serpentine-like in nature, with respect to Barberton, again giving confidence that the plant would adapt to the conditions occurring at the contaminated sites. However, to ensure optimal growth, nutrient experiments were also performed on B. coddii to ascertain the ideal macronutrient concentrations required, without inhibiting nickel uptake. These trials indicated that the individual addition of 250 mg/l ammonium nitrate, 600 mg/l calcium phosphate, 2 000 mg/l calcium chloride, 600 mg/l potassium chloride and 250 mg/l magnesium sulphate enhanced plant growth and nickel uptake, suggesting that, for phytoremediation purposes, these nutrients should be added to the medium in which the plants are growing. The growth-cycle of naturally occurring B. coddii plants in Barberton was also studied in order that seedlings could be germinated, in greenhouses, at the correct time of year so that the plants could be sown as the naturally occurring plants were germinating. From this information the seeds of the plants could be collected at the correct time of year and the above ground biomass harvested when the nickel concentrations were at their highest. It was found that the plants began to germinate as the first rains fell, which was generally at the beginning of September, and plant maturity was reached at about five months, after which flowers were produced. Seeds were produced from the flowers and these matured and were wind-dispersed one month to six weeks after full bloom, usually during February. The plants then started to die back and dry out and dormancy was reached about nine months after germination, generally in about mid- to late- May. It was found that the nickel concentration was at its highest about one month after the plants had begun to dry out and thus it was decided that the above ground biomass would usually be harvested at the end of April each season, in order to achieve maximum nickel recovery. Finally, in order that the plant's potential for use in phytoremediation could be fully assessed, field trials at the contaminated sites in Rustenburg were performed. Germination procedures were developed for the mass production of B. coddii and it was found that, although fully formed plants could be propagated in tissue culture, it was cheaper and faster to germinate the seeds in speedling trays, containing a zeolite germination mix, in greenhouses. It was also found that the seeds had a low germination rate, due to dehydration of the embryos and thus, in order to obtain the number of plants required, four to five times the amount of seeds needed to be sown. The two-month-old seedlings were transferred to potting bags, containing a mixture of potting soil and RBMR soil, and grown up in the greenhouse for a further three months. This growth period allowed B. coddii to adapt to the RBMR soil and also ensured that the plants were relatively healthy when transplanted into three prepared sites at RBMR. The plants were allowed to grow for the entire season after which the above ground biomass, comprising the leaves and stems, was harvested, dried and then ashed in an ashing vessel designed by the author, with the help of Mr K Ehlers. The ashed material was acid-leached with aqua regia in order that the base metals (mainly nickel) and precious metals could be removed from the silicates and carbonised material. The acid solution was then neutralised, causing the base metals (mainly nickel) and precious metals to be precipitated. This precipitate was then smelted with a flux in order that nickel buttons could be formed. Thus, from all the phytoremediation trials it was found that this process is highly successful in employing B. coddii for the clean-up of nickel-contaminated sites. This constitutes the first time that such a complete phytoremediation process has ever been successfully developed with B. coddii as the phytoremediation tool. It also appears to be the first time that phytoremediation has been performed "commercially" to produce a saleable metal product. The success of this project has stimulated Amplats to continue with, and expand it, to include more studies on phytoremediation as well as in the biomining of certain areas containing very low levels of precious metals which, with conventional techniques, were previously not worth mining. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1998.