The Neutralisation of acid mine drainage by fly ash

Thesis / Dissertation

2000

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
License
Series
Abstract
[Page i missing] South Africa is largely dependent of the combustion of coal for power generation. Approximately 22 Mt of ash was produced in 1999 by ESKOM, the largest producer of electricity in South Africa. Of this only 1.1 Mt was utilised, the rest being disposed of in landfills or ash dams. Various environmental impacts are associated with the disposal of fly ash including loss of usable land, dust pollution, water pollution, effects on vegetation and aesthetic impacts. The oxidation of pyrite exposed during the mining of coal results in the generation of large quantities of acid water with a high dissolved metal load. Current legislation requires this water to be collected and neutralised before being discharged. The costs involved in neutralising this effluent and safe storage of the resulting sludge can be enormous. In South Africa the power stations are built adjacent to the colliery which supplies the coal. This results in a unique situation where two waste stream of contrasting character are situated in close proximity to each other. The aim of this study is to investigate the reactions of acid mine drainage (AMO) and fly ash in order to evaluate the potential for their co-disposal. A simulated AMO containing 364 mg/I Al, 355 mg/I Fe, 2782 mg/I so/- as the major ions at pH 2.5 was prepared and used in all experiments. Fresh fly ash was collected from the Arnot power station and from Sasol. The mineralogy of the two fly ashes is similar. The mineralogical phases present in the fly ash are glass, quartz and mullite with accessory lime and hematite. Sasol fly ash has a lower lime and higher mullite concentration than the Arnot fly ash. The concentration of quartz is similar in the two ashes. The differences in the mineralogy of the fly ashes is reflected in their chemical composition, with the Sasol ash having a lower concentration of Cao (5.4 wt%) compared to 7.3 wt% in the Arnot fly ash. The Al 20 3 concentration in the Sasol - ash is higher (29.9 wt%) than the Arnot fly ash (22.1 wt%). The morphology of the Arnot fly ash is dominated by well rounded glass particles and spheres, indicating a high degree of melting in the furnace. The Sasol fly ash is dominated by blocky particles with a highly irregular morphology. This indicates higher melting temperatures as would be expected due to the lower Cao concentration. Batch neutralisation experiments were conducted with a series of fly ash: AMO ratios, ii and equilibrated on a reciprocating shaker for 72 hours. The results showed that the r------~----·---··· higher lime concentration of the Arnot fly ash results in a higher neutralisation potential than the Sasol fly ash. The titration curves of AMO and fly ash for the two ash samples have similar shapes, indicating that the neutralisation reactions are the same for both fly ash samples. The titration curves are characterised by four distinct regions. Mineralogical analysis by XRD and SEM and thermodynamic modelling of the solution chemistry enabled the identification of the mineral phases controlling the solution composition. At high pH values (>12.7) the solutions are in equilibrium with portlandite (Ca(OH)z). Between pH values of 12 and 10.3, ettringite (Ca6A'2(SO4)3(OH) 1226H2O) precipitates and acts as a weak pH buffer. Below pH 10.3 ettringite dissolves incongruently to form gypsum and amorphous Al(OH)3. Gypsum was identified by XRD to pH values of 5.6, although thermodynamic modelling indicates that the solution is in equilibrium with gypsum to pH value of 4.1. The presence of amorphous Al(OH)J could not be confirmed directly, but is inferred from the modelling of the solution chemistry. Very little buffering is observed between pH values of 10.3 and 4.5, where the solution composition is controlled by gypsum and amorphous Al(OH)3. Between pH values of 4.5 and 4.1 the system exibits a very strong buffering capacity, which is considered to be the result of the dissolution of the amorphous Al(OHh The mineralogy of the iron precipitates was identified by means of XRD to be 2 line ------ ferrihydrite. SEM examination of the reacted fly ashes showed the ferrihydrite to form coatings on the fly ash particles. The kinetics of the neutralisation reactions were investigated by monitoring the pH and electrical conductivity (EC) of three fly ash: AMO mixtures (1 :5, 1: 10, 1 :20) over a period of 72 hours. The kinetics of the neutralisation reactions are initially very rapid, with the pH increasing from 2.5 to 4 within 5 minutes for all fly ash: AMD ratios. This rapid neutralisation reaction is considered to be the result of the free lime present in the fly ash. The reaction rates decrease as equilibrium is approached. For both fly ash samples the pH had stabilised within 24 hours. Those experiments whose equilibrium pH was below pH 10 maintained a constant EC from the time the pH stabilised until the experiment was terminated. Where the equilibrium pH was higher than 10 the EC continued to decrease and had not stabilised when the experiments were ended. This slow decrease in the EC is considered to be due to the slower Ill kinetics of ettringite precipitation as equilibrium is approached. The co-disposal of AMD and fly ash shows potential as a means of neutralising the extreme pH of these waste streams. The Arnot fly ash has a greater neutralisation potential than the Sasol fly ash.
Description
Keywords

Reference:

Collections