IM 2016 June 2016 | Page 68

PASTE- GOLDER_proof 25/05/2016 09:36 Page 1 Paste Supplement Cemented tailings versus paste Thomas Skocir and Isaac Ahmed of Golder Associates in Mississauga, Ontario, Canada; and Glenn Barr of Twin Metals Minnesota, outline how continued binder testing led to process optimisation and potential savings in backfill costs for the TMM project investigating different strategies for using mill tailings as mine backfill material. One proposed mining method only requires low strength (150 kPa), non-liquefiable backfill material to fill the mined out stopes because each stope is independent from another, and mining will not take place beside exposed backfill material. Binder selection and optimisation One of the main options investigated as a backfill material was cemented tailings (CTT). Extensive laboratory testing was performed, investigating binder types, binder blends, binder addition quantities, and tailings slurry Cast cylinder before (left) and after (right) UCS testing, CTT with 8% binder at 14 days curing time win Metals Minnesota LLC (TMM) is currently pursuing the development of an underground mine in northeast Minnesota. Situated between the towns of Ely and Babbitt, the proposed project will produce two concentrate products, containing copper, nickel, gold, platinum and palladium. As part of the project development and engineering stage, TMM is T Figure 1: Unconfined compressive strengths for two different binders at varying concentration at a 7 in slump paste blend. Results for the use of FA are not shown P8 International Mining | JUNE 2016 Supplement density. Samples were cast into cylinders and allowed to cure for specific periods, most typically 7, 14 and 28 days. After being cured to the prescribed time period, the cylinders were removed from their moulds, and compressed to failure in a load frame, resulting in an unconfined compressive strength (UCS) measurement of the fill, which provided a relative means of evaluating the variables under consideration. Binder types tested with the TMM tailings sample were: n 100% Normal Portland Cement (NPC) n 50% NPC/50% Fly Ash (FA) n 70% NPC/30% Fly Ash n 80% Blast Furnace Slag (BFS)/20% NPC n 90% BFS/10% NPC. Initial testing was undertaken with 7 in slump tailings (79% solids), consistent with a typical ‘paste’ material. Figure 1 presents the results of the 7 in slump paste test work with NPC and BFS binders. A comparison between the 90% BFS/10% NPC binder (red) and the straight NPC binder (blue) tests show that significantly higher strength material is achieved with BFS binder than with NPC binder. In order for mill tailings to reach a paste consistency, two stages of dewatering and one stage of repulping are required: mill tailings are thickened to 73 wt% solids, dewatered (filtered) to 83% solids, and then repulped to a 7 in slump slurry. Based on the favourable results obtained using 90% BFS/10% NPC, and a continued desire to simplify the backfill process, a second round of test work was undertaken, with the goal of determining the material strength that could be achieved with addition of BFS binder directly to the thickened tailings. The results can be seen in Figure 2. The results were better than expected. Binder contents on tested samples only overlap at 1 wt%; at this value there is only a marginal decrease in strength, in the 10-20% range at the 28 day break. Although