I would like to thank my colleagues Peter Campbell of Huckleberry Mines, Shannon Shaw of Robertson GeoConsultants and Brian Soregaroli , formerly of Klohn Crippen for helpful discussions during the preparation of this page.
The interpretation of kinetic testwork results varies from straightforward to extremely difficult, depending upon the mineralogy of the samples tested, their origin and their history. For all but the simplest cases valid interpretation depends upon solid knowledge of sample petrology and mineralogy, size distribution and liberation characteristics, ABA data, and geological, hydrological, meteorological, and other relevant data. For an extensive review of kinetic geochemical processes and their relevance to the assessment of waste rock ARD potential, see Perkins et al. (1995)
In order to demonstrate the importance of some of the factors noted above, I have selected kinetic data from four mine projects. Each case illustrates one or more interpretative problem that may be roughly categorized as follows:
In all cases the data interpretation is my own and may or may not coincide with that of the original authors.
Previously Oxidized Material
When kinetic tests are conducted on previously oxidized material such as waste rock or ore stockpiles, or tailings, it must be assumed that particle surfaces, including cracks and fissures, may contain water soluble oxidation products such as sulphates and/or hydroxy sulphates. These compounds will dissolve as the kinetic test proceeds and will be recorded as cations and (usually) sulphate ions generated by the sample. While the cations and sulphate are the products of ARD, they are not the products of sulphide oxidation during the kinetic test itself since they may have accumulated over a period of many years. Also, since most of these products accumulate at the particle or grain boundaries of sulphide minerals, they will tend to protect, to varying degrees, the sulphide mineral grains from further oxidation. Dagenais and Poling (1997) have stated that (for work conducted at the Island Copper mine in British Columbia):
"......kinetic testing conducted on previously oxidized material spent many weeks of expensive testwork leaching reaction products from the material before rates of sulphide oxidation could be calculated"
Kinetic tests on previously oxidized material will, therefore, usually generate results that potentially show both the dissolution of oxidation products and the generation of oxidation products. Depending upon reaction kinetics these two separate phenomena may generate either overlapping or separate cation/anion production rate versus time curves. It is possible, therefore, that dissolution rates from oxidation products may mask generation rates from sulphide oxidation.
The following two examples are from data obtained from column testwork on waste rock dumps at United Keno Hill Mines in the Yukon Territory as reported by Access Mining Consultants (1996).
Figure 1 shows sulphate, calcium, magnesium and zinc leaching rates and pH for a waste dump quartzite sample containing 0.14% Sulphide Sulphur and less than 0.01% Sulphate Sulphur with AP = 4.4 kg/tonne CaCO3 , NP = 3.2 kg/tonne CaCO3 , NNP = -1.2 kg/tonne CaCO3 and NPR = 0.7. This is a sample that would be expected to be acid generating.
Figure 1: Metal and Sulphate Leaching Rates and pH for Sub-Aerial Column Test on Keno Hill Waste Dump Quartzite (data from Access Mining Consultants Ltd., 1996)
The sample clearly is acid-generating, but only after a period of about 14 weeks. During these initial 14 weeks sulphate generation, paralleled by calcium and magnesium generation, is probably from accumulated gypsum. It might be argued that the calcium and magnesium could equally well be produced from acid generation and neutralization until, after 14 weeks, acid generation exceeded neutralization capability. Nevertheless, calcium and magnesium generation clearly continues at a pH-independent rate from Week 14 to Week 28. In addition, zinc generation initially falls, independently of pH, suggesting the dissolution of residue, and then rises with pH, suggesting primary mineral reaction. It is likely that the curves of Figure 1 represent a case of overlapping residue dissolution and acid generation as the source of dissolved ions.
Figure 2 shows sulphate, calcium, magnesium, copper, iron and zinc leaching rates and pH for a mixed vein, quartize and schist waste dump sample containing 0.38% Sulphide Sulphur with AP = 11.9 kg/tonne CaCO3 , NP = 0 kg/tonne CaCO3 , NNP = -11.9 kg/tonne CaCO3 and NPR = <0.1. The sample was from a waste dump more than twenty years old in which acid generation and neutralization had clearly taken place to the exhaustion of available neutralization capability. Based on ABA data alone. this is a sample that would be expected to be acid generating from the oxidation of the remaining sulphide sulphur.
Figure 2: Metal and Sulphate Leaching Rates and pH for Sub-Aerial Column Test on Keno Hill Waste Dump Mixed Vein, Quartzite & Schist (data from Access Mining Consultants Ltd., 1996)
All of the metal and the sulphate leaching rates are initially high and steadily fall by one to two orders of magnitude over 27 weeks, showing some stability between Weeks 12 and 27 (sulphate, calcium and magnesium) and Weeks 20 and 27 (iron, copper and zinc). There appears to be a significant fall in leaching rates for all metals and sulphate after Week 27. pH increases slowly, but steadily, from pH 1.9 to pH 2.5 at Week 30. The low pH values are the result of metal ion hydrolysis (see, for example, Hughes and Maloney, 1964) in the sulphate environment, not from sulphide oxidation, and the pH increase with time reflects the decrease in metal ion concentration with time. The pH values agree well with predictions generated by the computer geochemical equilibrium speciation model MINTEQA2 (Allison et al., 1991) using leachate cation and anion aqueous concentrations.
Overall the curves of Figure 2 are indicative of the dissolution of accumulated oxidation products. These products might include gypsum, CaSO4 , melanterite, FeSO4.7H2O, copiapite, Fe5(SO4)6(OH)2.20H2O, roemerite, Fe3(SO4)4.14H2O, kornelite, Fe2(SO4)3.7H2O, coquimbite, Fe2(SO4)3.9H2O, rhomboclase, HFe(SO4)2.4H2O, halotrichite, FeAl2(SO4)4.22H2O, mallardite, MnSO4.7H2O , brochantite, Cu4(SO4)(OH)6 , chalcanthite, CuSO4.5H2O, boothite, CuSO4.7H2O, zinc-melanterite, (Zn,Cu,Fe)SO4.7H2O, smithsonite, ZnCO3 , and hydrated oxides of iron such as ferrihydrite, 5Fe2O3.9H2O and goethite, FeO(OH).
Since the sample has no NP, there can be no formation of gypsum by acid neutralization and soluble calcium, and corresponding sulphate, must derive from existing gypsum dissolution. However, gypsum alone would not account for the very high initial sulphate generation rate, and the difference in shape of the sulphate and calcium curves through approximately Week 14 tends to confirm this. Some other soluble sulphate minerals containing iron, manganese, copper and zinc are noted above.
The continuous shape of the sulphate curve is not indicative of acid generation by sulphide oxidation. It would be expected that oxidation of the 0.38% sulphide sulphur in the sample would begin only after oxidation products at sulphide grain surfaces were sufficiently depleted to allow oxygen and water diffusion to the sulphide grains, as can be seen in the latter weeks of Figure 1. For Figure 2, it is clear that after 30 weeks there is still dissolution of oxidation products.
In order to "by-pass" the stage of oxidation product dissolution in kinetic testing of potentially acid-generating, previously oxidized material (and to thus reduced the length of time for kinetic testing), Dagenais and Poling (1997) examined the effects of sample pre-treatment with distilled water and sulphuric acid acidulated water of pH 4 and 2. They concluded that (for the Island Copper waste rock), pre-treatment with sulphuric acid at pH 2 removed 70% of the accumulated oxidation products from their samples. Price and Kwong (1997) refer to a procedure for oxidation product removal using a dithionite-citrate solution buffered with sodium bicarbonate (CBD).
Material Containing Soluble Sulphate Minerals
Materials containing soluble sulphate minerals such as gypsum and anhydrite require special consideration in kinetic testwork. Mineralogical testwork and an ABA program in which both sulphate sulphur and sulphide sulphur are determined will clearly identify the presence and mode of occurrence of these minerals. Unlike sulphates and hydroxy-sulphates in previously oxidized material, natural occurrences of gypsum and anhydrite do not occur specifically on sulphide mineral surfaces, but as discrete grains and aggregates, often in cracks and veinlets. However, from start-up of the kinetic test the minerals will dissolve to give calcium and sulphate ions in solution, and the concentration of these ions is likely to be large enough to completely mask the concentration of any calcium or sulphate produced by sulphide mineral oxidation/neutralization. In addition, since sulphate ion is a product of both sulphide mineral oxidation and gypsum or anhydrite dissolution, the concentration of sulphate ion from mineral dissolution may be high enough to suppress sulphide oxidation to sulphate because of the "common ion effect".
It may take many weeks or months for all of the gypsum or anhydrite to dissolve, and during this period the onset of sulphide oxidation and acid production will almost certainly be delayed, even for samples that are clearly potentially acid generating based on ABA testwork.
The following two examples are from testwork on materials (crushed drill core) from the Huckleberry Mine in British Columbia. This data is not in the public domain, but was generously supplied by Peter Campbell, Manager Environmental Affairs, Huckleberry Mines Ltd. The testwork was performed by Dr. Richard Lawrence, University of British Columbia. The mineralogy of the Huckleberry deposit in relation to ARD has been discussed by Mills (1995).
Figure 3 shows humidity cell results for a sample of low grade ore over a period of 125 weeks. This low grade ore contained 4.15% Total Sulphur and 2.98% Sulphide Sulphur. The difference, Sulphate Sulphur, accounted for 1.17% which was contributed by gypsum and anhydrite. ABA data for the sample was AP = 93 kg/tonne CaCO3 , NP = 72 kg/tonne CaCO3 (Modified Sobek), NPR = 0.77. This sample would be expected to be acid generating.
Figure 3: Metal and Sulphate Leaching Concentrations and pH for a Humidity Cell Test on Huckleberry Mine Low Grade Ore (data courtesy of Peter Campbell, Huckleberry Mines)
Despite the potential for acid generation, Figure 3 clearly shows that the cell leachate remained at or about pH 7 for the 125 weeks of the test. Calcium and sulphate leaching rates fell slightly over the test period, while maintaining an almost constant ratio corresponding to the calcium:sulphate ratio of gypsum or anhydrite. Figure 3 is, therefore, a demonstration of gypsum and anhydrite dissolution from the low grade ore. If any acid generation/neutralization took place, it is masked by sulphate mineral dissolution. The relatively constant and low level of copper leaching (Huckleberry is a copper ore) is consistent with the dissolution of gypsum and anhydrite, rather than acid generation/neutralization. 125 weeks is a much longer period for humidity cell testwork than is normally used, but even after this period a clearly potentially acid generating sample was still neutral and dominated by gypsum/anhydrite dissolution.
Figure 4 shows the results of 24 months of column testwork to evaluate the sub-aqueous disposal of potentially acid generating volcanic waste rock with NPR = 0.2. In addition to sulphides, this sample also contained significant gypsum and anhydrite.
Figure 4: Metal and Sulphate Leaching Concentrations and pH for a Sub-Aqueous Column Cell Test on Huckleberry Mine Volcanic Waste Rock (data courtesy of Peter Campbell, Huckleberry Mines)
Figure 4 differs markedly from Figure 3 in that it shows a significant decrease in calcium and sulphate leaching rates beginning at about Month 11 and levelling off at about Month 17. Between Months 11 and 17 these leaching rates fell by two orders of magnitude, and this is interpreted as coinciding with the practical depletion of liberated gypsum and anhydrite in the sample. With an NPR of 0.2 this sample would be expected to oxidize and generate acid after gypsum and anhydrite depletion. However, this test was conducted to simulate sub-aqueous deposition with water as the oxygen barrier, and the absence of oxidation and acid generation is demonstrated by the relatively constant pH 7 of the system throughout the entire 24 months of the test.
Material with Low Carbonate Content
Material with Complex Neutralizing Mineralogy
ŠThe contents of this web page are protected by copyright law. Please contact the authors for permission to re-use the contained information.
Access Mining Consultants Ltd. (1996), United Keno Hill Mines Limited: Geochemical Testing Report - Progress Report No. 1, December 16, 1996.
Allison, J.D., Brown, D.S., and Novo-Grada, K.J. (1991), MINTEQA2/PRODEFA2, A Geochemical Assessment Model for Environmental Systems, version 3.11, EPA/600/3-91/021, 1991, Office of Research and Development, U.S. Environmental Protection Agency, Athens, GA.
American Society for Testing and Materials (1996), ASTM Designation: D 5744 - 96 - Standard Test Method for Accelerated Weathering of Solid Materials Using a Modified Humidity Cell, ASTM, West Conshohocken, PA, 13p.
Dagenais, P.J. and Poling, G.W. (1997), An Investigation into the Geochemical History of a Waste Rock Dump and its Effect on Water Quality of the Flooded Open Pit at Island Copper Mine, Port Hardy, British Columbia, Proceedings 4th International Conference on Acid Rock Drainage, Vancouver, p1709-1726.
Hughes, D.E.P. and Maloney, M.J. (1964), Advanced Theoretical Chemistry, Chatto & Windus, London, p139-140.
Mills, C. (1995), Technical Review of the Acid Rock Drainage (ARD) and Metal Leaching Aspects of the Metallurgical Testwork, Milling Practices and Tailings Monitoring for the Huckleberry Project, report (draft) to BC Ministry of Energy, Mines and Petroleum Resources, Vancouver, 34p.
Perkins, E.H., Nesbitt, H.W., Gunter, W.D., St-Arnaud, L.C. and Mycroft, J.R. (1995), Critical Review of Geochemical Processes and Geochemical Models Adaptable for Prediction of Acidic Drainage from Waste Rock, MEND Report, No. 1.42.1, MEND, Ottawa, ON, 120p.
Price, W.A. and Kwong, Y.T.J. (1997), Waste Rock Weathering, Sampling and Analysis: Observations from the British Columbia Ministry of Employment and Investment Database, Proceedings 4th International Conference on Acid Rock Drainage, Vancouver, p31-45.
Rescan Environmental Services Ltd. (1997), Tulsequah Chief Project Report, Volume IV.
Soregaroli, B.A. and Lawrence, R.W. (1998), Update on Waste Characterization Studies, paper presented at Mine Design, Operations & Closure Conference, Polson, MT, 10p.
Return to ARD at Enviromine