Abstract
Acid mine drainage (AMD) is one of the most important environmental issue arising from mining practices worldwide. It is caused by the disturbance, and then chemical weathering of pyrite rich rocks during mining operations. AMD is characterised by low pH (2-3) and increased metal concentrations, occasionally reaching toxic levels for humans. Extensive research has been undertaken world wide to prevent and combat AMD through proper identification of pyrite formation, oxidation, chemical weathering of other phases, and subsequent discharge chemistry.
In New Zealand, especially on the West Coast of the South Island, a rich history of
coal mining since the 1800s has caused environmental issues arising from AMD that have been studied by many authors, leading to remediation and prevention plans, and a better understanding of the constraints on AMD production.
Echo mine is a recently active coal mine that as of the end of this thesis, has moved into a “care and maintenance” mode. It produces coal mined from the acid forming Brunner coal measures. AMD is observed at the site, with elevated levels of Fe, Al, Mg, Ca, SO42- , and trace elements such as Ni and Zn. This thesis represents research undertaken to identify the main cause of AMD, how mine wastes change while stored, and the secondary phases responsible for any temporal and spatial changes in AMD chemistry.
Fieldwork at the mine revealed 3 main rock types, all of which contribute to mine wastes
- a white sandstone, a grey mudstone and a rare blue/green sandstone. Samples of these were collected and their mineralogy and geochemistry identified. The white sandstone was only potentially acid forming (PAF) rocks, while the mudstone provided a source of soluble iron in the form of a hydroxide matrix and minor alkalinity in the form of post-depositional cross cutting Mn-bearing ankeritic veins. The blue/green mudstone contains dolomite and Mn-ankerite, representing the main source of alkalinity at the site.
Samples of precipitates from the surface of rocks were collected and analysed to determine the phases present. Jarosite was frequently observed on the surface of the white sandstones, while amorphous iron hydroxides were present on the mudstones. These represent a considerable source of long term acidity storage at the site, and need to be taken into account when creating a long term remediation plan.
The site provided access to mine wastes of three different ages - 2, 5 and 12-15 years old. These were “autopsied” with horizons selected based on appearance and analysed to determine the phases responsible for the changes observed, titratable actual acidity
(TAA) and HCl soluble sulfur (S HCl), and the water soluble mineralogy and geochemistry determined. XRD patterns for both crushed and sieved samples returned identical patterns for all samples, suggesting differences between them was either below detection or amorphous in nature, therefore XRD was not of use. TAA levels progressively increased over time. S HCl levels were initially high at intermediate depths. Over time, concentrations increased at the surface and deeper in mine wastes. Saturation indices calculated from PHREEQC suggested schwertmannite, jarosite and rarely alunite and amorphous Fe(OH)3 were all over saturated, while most phases remained below saturation. Temporal changes were observed, with highest dissolved metals and lowest pH observed in the 5 year old wastes, with Mn, Ni, Zn and Mg sharply decreasing after 5 years, interpreted to be due to the epsomite solid-solution series having high solubility, coupled with a limited supply of these constituents.
During sample storage, a small hole developed on one of the sample bags, allowing cold, humid air to circulate around the sample. This produced an efflorescent bloom, which was analysed to determine the mineralogy. A combination of XRD and SEM analyses determined gypsum, epsomite, alunogen and mirabilite were present. This information was used as a basis for the changes in water soluble chemistry collected from horizon leaches.
The water soluble chemistry of the mine wastes were normalized and modelled using PHREEQC to determine the paragenesis of phases during evaporation. Most samples produced jarosite and gypsum, with schwertmannite over saturation ending after equilibration of the solution. Surprisingly, no epsomite, alunogen or mirabilite was formed, most likely due to the accidental nature of their precipitation.
Several trends were observed during evaporation modelling, with phases either increasing saturation, decreasing, or showing an increase then decrease. Increases were related to increased activity of dissolved species, while negative trends were attested to decreased [OH-]. The increase then decrease trend was not explainable, and is most likely an artifact from PHREEQC modelling.
Overall, recommendations can be made against the moving of mine wastes in the first 5 years of storage, due to the increased dissolved metal activity reaching a maximum at this point. However, since long terms of acidity migrate to shallower levels over time, the nature of sulfate controlled acidity changes from rapid to slow releases over time.