AMD Prediction and Implications for Mine Water Quality_TonyJong.pptx

AMD Prediction and Implications for Mine Water Quality_TonyJong

1. AMD Prediction & Implications for Mine Water Quality Dr Tony Jong

2. 2 Introduction Coal is deposited within reducing environments that have potential to produce sulfides (most commonly pyrite (FeS2) ). Waste generated through mining in the form of overburden and coal processing (rejects and tailings) can expose these sulfides to air and water, resulting in their oxidation . Oxidation of pyrite can be represented by the following overall reaction : As acid (H 2 SO 4 ) water migrates through a site (e.g. tailings or waste rock), it further reacts with other minerals and may dissolve a range of metals and salts.  

3. Introduction Potentially lead to the generation of acid and metalliferous drainage (AMD ). AMD can be: Acidic drainage  characterised by low pH and elevated dissolved metals. Neutral drainage acid completely neutralised by dissolution of common carbonate minerals (such as calcite, dolomite and magnesite ), leading to precipitation and thus removal of metals such as Al, Cu and Pb . Near-neutral but metalliferous  elevated in metals that are soluble at neutral pH conditions such as Zn, As, Ni and Cd. Saline drainage  no dissolved metal residues remain, but contain high ( sulfate ) salinity. 3

4. 4 AMD P rediction A number of procedures have been developed to assess the acid generating characteristics of mine waste materials. Overall acid generating potential of a sample is mainly evaluated by its Acid Base Account (ABA) and the Net Acid Generation (NAG) test. These are static test procedures (i.e. single measurement in time). The ABA is a theoretical balance between a samples’ maximum capacity to generate acid (MPA ) relative to its acid neutralising capacity (ANC ). The NAG test represents a direct measurement of the net amount of acid generated by the sample.

5. AMD Prediction The Net Acid Production Potential (NAPP) and the ANC/MPA ratio are two measures of the ABA. NAPP > 0  positive net acid producing potential ( PAF) NAPP ≤ 0  non-acid forming (NAF) or potentially acid consuming (AC). T otal sulfur is commonly used to calculate MPA, but chromium reducible sulfur (CRS) provides a better estimate of pyritic sulfur and thus NAPP. The ANC/MPA ratio provides an indication of the relative margin of safety (or risk) within a material to generate acid. Typically this is between 1.5 and 3.   5

6. AMD Prediction The ANC/MPA ratio provides an indication of the relative margin of safety (or risk) within a material to generate acid. Typically, this ratio is ≥ 2. NAPP and NAG values provide an indication of the potential for acid generation from a sample; however, additional test work is required to predict the potential for metalliferous or saline drainage . Typically done by conducting leachability tests ( 1:5 deionised water to solid extractions ). 6

7. Geochemical Classification NAG test commonly used in conjunction with the NAPP to reduce the risk of misclassifying the acid-generating potential of the mineral waste sample. Comparison between NAPP and NAG results will help identify uncertainties that require follow up. Typical geochemical classification scheme based on NAPP and NAG results: 7 Geochemical Classification NAPP (kg H 2 SO 4 /t NAGpH Potentially Acid Forming (PAF) >10 a <4.5 Potentially Acid Forming – Low Capacity (PAF-LC) 0 to 10 a <4.5 Non-Acid Forming (NAF) -100 to <0 ≥4.5 Acid Consuming (AC) <-100 ≥4.5 Uncertain (UC) b >0 ≥4.5 <0 <4.5 a Site -specific but typically in the range 5 to 20 kg H 2 SO 4 /t. b Further testing required to confirm material classification.

8. Geochemical Classification 8

9. Potential Acid Risk 9

10. Metals and Salinity Composite Number S1 S2 S3 S4 S5 S6 S7 S8 S8 No. of Sub-samples 5 4 2 6 2 3 2 4 2 Parameters Livestock Drinking Water a Siltstone # Claystone # Carbonaceous Claystone # Sandstone # Sandstone/siltstone Conglomerate Siltstone/ Claystone/ Sandstone Shale Mudstone Ca 1000 b 1.4 1.0 2.7 1.3 <1 <1 <1 <1 <1 Mg 2000 c 1.0 1.0 1.2 1.0 <1 <1 <1 <1 <1 SO 4 2- 1000 d 137 75 79 75 11.6 22.4 23.3 4.6 10.2 Al 5 0.75 0.18 0.16 0.50 0.04 0.04 0.04 0.08 0.11 As 0.5 to 5 e 0.096 0.026 0.009 0.090 0.006 0.005 0.089 0.003 0.004 B 5 0.10 0.03 0.01 0.06 <0.1 <0.1 <0.1 <0.1 <0.1 Cd 0.01 0.0003 0.0001 0.0001 0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Cr 1 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Co 1 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Cu 0.4 to 5 f 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Pb 0.1 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Hg 0.002 0.0003 0.0001 0.0001 0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Mo 0.15 0.043 0.015 0.014 0.029 0.002 0.010 0.010 0.002 0.009 Ni 1 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Se 0.02 0.02 0.01 0.01 0.01 <0.01 <0.01 <0.01 <0.01 <0.01 U 0.2 0.001 0.001 0.001 0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Zn 20 0.005 0.003 0.003 0.005 <0.005 <0.005 <0.005 <0.005 <0.005 10 All values in mg/L. # Mean value; where values were less than the limit of reporting (LOR), the LOR value was used for calculation purposes. a ANZECC & ARMCANZ (2000). b Stock should tolerate concentration if calcium is the dominant cation and dietary phosphorus levels are adequate. c Insufficient information is available to set trigger value; however, concentrations up to 2000 mg/L have been found to have no adverse effects on cattle. d No adverse effects to stock are expected if the concentration does not exceed 1000 mg/L. e May be tolerate if not provided as a food additive and natural levels in the diet are low. f Dependent on livestock species.

11. Water Quality Prediction Water quality prediction based on geochemical modelling: The Geochemist's Workbench ( Bethke et al., 2008) PHREEQC (USGS, 2011). Modelling conducted by calculating chemical equilibrium distributions and typically does not take into account kinetic reaction rates. For GWB, “general” thermodynamic database can model solutions: Up to 3 molal (~96,000 µS/cm). Temperatures from 0 to 300  C. 11

12. Water Quality Prediction Analyte ANZECC ( 2000) 1 Wet Season Water Conceptual Mixtures 20% Model 1 Water 40% Model 1 Water 60% Model 1 water 80% Model 1 water 80% Creek 60% Creek 40% Creek 20% Creek pH 6.0 to 8.0 a 5.72 5.99 6.09 6.14 6.17 Al 27 b 18 0.18 0.19 0.21 0.23 Cd 0.06 0.1 0.42 0.75 1.08 1.40 Co ID 0.39 1.49 2.60 3.71 4.82 Cr(VI) 0.01 1 0.01 0.01 0.01 0.02 Cu 1.0 2 1.97 1.98 1.98 1.98 As 1.8 c 1 0.02 0.04 0.05 0.07 Mn 1200 0.65 136 271 406 541 Ni 8 2.36 6.41 10.5 14.6 18.6 Pb 1.0 0.65 0.91 1.18 1.44 1.71 Se 5 1 0.01 0.01 0.01 0.01 Zn 2.4 60 258 456 655 854 Results for equilibrated drainage water mixed with receiving Creek water Cd , Co, Pb and Zn concentrations exceed the ANZECC (2000) trigger values for 99% species protection. 12 All values in μ g/L unless otherwise stated. 1 For 99% species protection. a Lowland rivers ( ≤ 150 m altitude); b pH > 6.5; b Limit for total Cr; c Sum of As(III) and As(V); ID = Insufficient data to derive a reliable trigger value.

13. Water Quality Prediction 13 Mineralisation/precipitation as AMD is treated with hydrated lime (i.e. increasing pH). Ag, As, Sb , Se, U and Sn are not removed by simple precipitation with hydrated lime.

14. Conceptual AMD Risk Map 14

15. Conclusion The ABA and NAG test provides a useful screening of AMD potential, and can delineate areas that need more detailed investigation. Static tests provides a snap-shot of the mine waste materials’ potential to cause or alleviate AMD. Leachability tests combined with geochemical modelling can provide an initial indication of the potential water quality impacts and/or AMD treatment efficiency. 15

16. Questions?