about us | careers | terms & conditions | intranet | extranet | sitemap | contact us
Skip Navigation Links
Skip Navigation Links
Knowledge Hub
Skip Navigation Links
Skip Navigation Links
Resources & Tools
Skip Navigation Links
Skip Navigation Links
Skip Navigation Links
Skip Navigation Links
News & Media
Skip Navigation Links
FET Water
Skip Navigation Links
Login | Register
Go Search
Towards passive treatment solutions for the oxidation of sulphide and subsequent sulphur removal from acid mine water
Expanded Title:Sulphide can be biologically oxidised under anaerobic conditions by photosynthetic bacteria and denitrifying organisms and in the presence of oxygen by colourless sulphur bacteria. These organisms have the potential to remove sulphide in passive acid rock drainage (ARD) treatment systems, such as the integrated managed passive (IMPI) treatment system, where sulphide oxidation is facilitated by a floating sulphur biofilm (FSB) in the linear flow channel reactor (LFCR). A scaled-up version of the LFCR was included in a demonstration scale plant, based on the IMPI process, which has been constructed on the Middelburg mine site. However, the LFCR was not sufficiently robust and still required improvements in design and operation. The fundamental microbiology and chemistry of biological sulphide oxidation is relatively well understood, but the integration of this information with fundamental engineering principles in the context of a treatment system, particularly the LFCR, required further work and underlies the rationale behind this project. Laboratory-scale LFCR reactors were constructed at the University of Cape Town (UCT) and Golder Associates Research Laboratories (GARL) and were used to characterise the hydrodynamics using tracer studies and assess the effect of sulphide loading and hydraulic residence time on sulphide oxidation performance. Analytical techniques to quantify intermediate reduced sulphur species were developed and allowed the sulphur mass balance across the LFCR. Molecular biology was used to characterise the microbial populations in the bulk phase and FSB. The data were used to inform a refined conceptual model to describe biological sulphide oxidation in the FSB. The information was used to suggest modifications to and improvements in the management of the demonstration plant. The most important results and their implications for the operation and management of the overall process are summarised below. The laboratory-scale unit at GARL was operated successfully to assess the impact of feed rate (33, 66 and 132 L/d) and inter-harvest period (1 and 2 d) on reactor performance. The results suggested that complete, or near complete, sulphide removal was achieved, although recovery as elemental sulphur was relatively low. It was speculated that sulphide loss as gaseous hydrogen sulphide could be significant. Different biofilm phases were observed at GARL and the implications of the physical properties on the ease of harvest investigated. Optimum harvesting was achieved when the biofilm was in the thick, brittle phase. The time taken to achieve thick, brittle biofilm was dependent on the sulphide loading rate. The laboratory-scale reactors at UCT were specifically designed to facilitate improved process control, analysis and assessment of hydrodynamics. The low Reynolds number (Re < 18) indicated very little turbulence and predominantly laminar flow, later confirmed by tracer studies that indicated significant stratification within the reactor. The hydrodynamic study clearly showed that the earlier assumption of plug flow was incorrect and that significant short circuiting occurred when the outlet point was near the base of the reactor. A preliminary model for oxygen requirement was developed from first principles. The model allowed the prediction of the oxygen required to achieve different levels of sulphide oxidation. The model assumed no significant inhibition of gas liquid mass transfer through the biofilm. This is a simplification and the model will be further refined in the future. Dissolved organic carbon was required in the LFCR feed for good FSB formation. Under carbon-limited conditions slow or incomplete FSB formation resulted in significant oxidation of sulphide to sulphate. Addition of sodium acetate (1 g/L) to the LFCR at the beginning of an experimental run significantly improved FSB formation and system performance. The threshold organic carbon level required for good FSB formation still needs to be determined. Sulphide conversions to sulphur in excess of 90% were achieved with the majority of product reporting to the biofilm. The optimised analytical techniques allowed the sulphur mass balance to be closed to within 5% and confirmed that the loss of sulphide as hydrogen sulphide gas was insignificant. Molecular biology results indicated the population was dominated by autotrophic sulphur oxidisers belonging to the genera Chlorobium and Chromatium. This was consistent with a system with limited organic carbon. A refined conceptual model was developed. Efficient operation of the LFCR depends on rapid biofilm development, is facilitated by heterotrophic organisms rapidly converting dissolved organic carbon to extracellular polymers, which support the sulphide oxidisers and the sulphur product. The biofilm bars oxygen mass transfer, which allows the creation of a reaction space within the FSB where the pH and redox conditions are conducive to partial oxidation of sulphide. Sulphide must be delivered to the reaction space consistently. This is dependent on the hydrodynamic flow and the relative density of the feed and bulk liquid. The density of the bulk must be lower than that of the sulphide in the feed.
Date Published:01/04/2012
Document Type:Research Report
Document Subjects:Mine water - Impact prediction, Mine water - Mine water treatment
Document Format:Report
Document File Type:pdf
Research Report Type:Standard
WRC Report No:1834/1/12
ISBN No:978-1-4312-0249-2
Authors:Van Hille RP; Mooruth N; Raja S; Heath RA
Organizations:University of Cape Town; Golder Associates (Africa) Pty Ltd
Document Size:61 415 KB
Copyright 2015 - Water Research Commission Designed By: Ceenex