Assessment and modelling of chromium release in minerals processing waste deposits

Petersen, Joachim

Assessment and modelling of chromium release in minerals processing waste deposits

Doctoral Thesis

1998

Publisher

University of Cape Town

Department

Department of Chemical Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

The minerals processing industry is by far the largest generator of mineral solid wastes, which are commonly stored in large scale landfill deposits. The potential environmental impact of these is directly linked to the time-dependent process of leachate generation within these deposits. Rainwater draining through the porous matrix of a deposit creates a slowly moving aqueous environment within the deposit. Heavy metal species that may be contained in trace amounts in the waste material can be mobilised into the aqueous phase by various chemical reactions and be transported by mechanisms of diffusion and convection to the base of the deposit and from there further into the surrounding environment. Laboratory assessment methods aim to provide indicators to the leachate generation potential of a particular waste material, often based on "worst case" assumptions, but generally fail to offer a meaningful appreciation of the time-dependent leach behaviour of the material in a full scale deposit. This is to a large part due to the lack of a thorough description - in terms of a rigorous mathematical model - of the leachate generation process itself. Such a model is developed in the present work, building on an existing model for heap leaching, which, conceptually, is very similar to the leachate generation process. The model is based on the continuity equation formulated for reaction-diffusion processes at the level of an individual porous particle and for convection-dispersion transport at the bulk level. This is combined with a number of reaction models, both kinetic rate expressions and thermodynamic equilibrium models, to describe the release process of individual species at the solid liquid interface and also within the aqueous phase. The model has been translated in the WASTESIM computer code within which waste iv Abstract material and disposal scenario are characterised by a number of parameters, such as those describing reaction modes and constants, particle size and pore diffusion effects as well as bed transport and saturation. The program was found to be a versatile tool for modelling a wide range of multi-species, multi-reaction deposit and batch leach scenarios. However, for modelling real waste materials the model parameters have to be established from a systematic laboratory investigation. An assessment methodology is proposed which aims to combine lysimeter studies with bench scale leach and physico- chemical characterisation experiments to enable determination of all model parameters entirely on the basis of laboratory experiments and validate them at this level against the results from independent lysimeter studies with the modelling tool. It is argued that, if all model parameters are validated at the laboratory scale in this way, modelling of full scale scenarios involving the same waste material can be conducted with some confidence. This approach has been put to the test with two waste materials from the ferro-alloy industry - a furnace emission control dust and a smelter slag. The contaminant species of particular interest for both these materials was chromium, especially Cr(VI), and therefore it was the release behaviour chromium on what much of the work presented herein has focused. The aqueous and environmental chemistry of chromium is extensively reviewed and, as a side aspect, the long-term atmospheric oxidation of Cr(III) to Cr(VI) has been positively identified by experimental work with a third chromium-containing waste material. The two test materials have been subjected to intensive characterisation in terms of column and batch leach experiments, adsorption studies, column tracer studies and physical characterisation experiments. The results are carefully interpreted with a view to establishing a complete set of parameters to simulate the leachate generation behaviour with respect to chromium species in a deposit scenario. It is demonstrated Abstract V that the modelling tool can in fact also be used for the interpretation of batch leach data through curve fitting exercises. For both materials the WASTESIM code, calibrated with parameters established entirely through the laboratory experimentation, has been used to simulate the leach curves of two independent lysimeter experiments, which are then compared to the measured data. In both cases the modelled and measured curves compared reasonably well and in most regards discrepancies can be explained by insufficient characterisation in the bench-scale experiments. The overall approach is therefore seen as valid in principle, but it is acknowledged that further experimental work and model development would be needed to take account of the remaining discrepancies. Two aspects were found to be particularly significant. The first relates to slow reaction mechanisms, which may go unnoticed in short-term laboratory experiments, but may become significant in full scale deposits given their long life-span. The slow atmospheric oxidation of chromium is a point in case. The second aspect relates to the hydro-dynamic characterisation of flow through unsaturated beds. Both model and laboratory assessment methods are insufficiently developed to account for effects such as dead pore diffusion and a distribution of flows. Recommendations for further development work should focus on these two aspects and on expansion of the approach to heavy metal species other than chromium. It is hoped that the modelling and assessment methodology will ultimately find welcome application in the environmental risk assessment of mineral processing waste disposal operations.

Keywords

Chemical Engineering

Reference:

Collections

PhD / Doctoral

Full item page