An experimental investigation into the trapping model core pillars with reinforced fly ash composites

CIM Bulletin, Vol. 3, No. 4, 2008

M.K. Mishra and U.M. Rao Karanam

In the room-and-pillar mining methods, pillars form an important load-bearing element of the system, controlling the stability of the near field domain.  The effective performance of a pillar support system is related to the depth of working, dimensions of the opening and the pillars, as well as the extraction percentage of coal/ore. One of the historic methods of containing the ground failures over mine voids is through backfilling of the mine voids by river bed sand or mill tailings. However, fly ash applications as an alternate material for backfilling mine voids is fast proving its potential. This paper presents investigations relating to the development of fly ash composite materials, as well as that of the load-deformation characteristics of model core pillars when confined by the wire mesh reinforced fly ash composite materials. Pure anhydrous chemical grade lime and gypsum were added in various proportions to the class F fly ash.  Lime contents were 15% and 20% and gypsum was 5% by weight of the fly ash.  Reinforcement was provided using commercially available galvanized iron wire 2 square mesh per centimetre and 0.9 mm thick.  Model core pillars 57 mm in diameter and 200 mm in length were made of various ratios of cement and sand. The engineering properties of the model core pillars, as well as that of the fly ash composite materials, were determined as per the recommendations of ISRM and ASTM. The length to diameter ratios of the final trapped models were between 1.33 and 2.0. Unconfined compressive strength and Brazilian indirect tensile strength tests were performed on a large number of samples for 28 and 56 days of curing periods, as well as for different annular thickness of the confining materials. The strength of fly ash composite changed with the addition of lime and gypsum, as well as the curing period. Experimental investigations have revealed that the percentage increase in the strength of the trapped model core pillar varied with the type of composite material, curing period and ratio of the annular thickness of fill area to the model core pillar radius. The seven-day strength of fly ash composites substantially improved with additives. The slake durability indices for the first and second cycles were more than 90% and 80%, respectively.  The 28-day curing period increased the strength dramatically.  An addition of 15% lime improved the strength of the composite to 5.45 MPa for a 28-day curing period, about 185% more than that for a seven-day curing period; however, the strength gain dropped to 56% beyond 28 days.  Similarly, a gain of 205% compared to the strength values for a seven-day curing period was noticed after 28 days, while the strength gain dropped to 46% for a 56-day curing period with 5% more lime.  Although an addition of 5% gypsum increased the strength value, it suffered reduced gain percentages for both curing periods. These observations confirm that the addition of excess lime to fly ash composites is not beneficial.  A maximum strength gain of 14% was achieved with model cores of a cement-sand ratio of 1:2.5 for fly ash composite containing 15% lime and 5% gypsum, as well as for fly ash composite material with 20% lime only.  Shear failure pattern was predominant in almost all the trapped model pillars, indicating that the reinforced fly ash composite materials offered significant radial confinement to the core pillars and induced the core pillars to fail in a ductile form.  The experimental investigations reveal that the brittle failure of the model cores could be changed to ductile failure through a suitable confining material, which has some strength of its own. The trapping of pillars could improve the post-peak strength characteristics of the structure.  It is also concluded that suitable fly ash composite materials, reinforced with wire ropes, can substantially enhance the strength of the load-bearing element and also significantly change the post peak characteristics of trapped cores.