Clarke, NC 2024, 'Backfilling tailings above an active cave mine', in AB Fourie & D Reid (eds), Paste 2024: Proceedings of the 26th International Conference on Paste, Thickened and Filtered Tailings, Australian Centre for Geomechanics, Perth, pp. 475-494, https://doi.org/10.36487/ACG_repo/2455_38
(https://papers.acg.uwa.edu.au/p/2455_38_Clarke/)
Abstract: The potential to backfill tailings above an active cave mining operation has been studied for the model case of a deep, near vertical, porphyry copper orebody in the humid tropics. 
The concept has major social and environmental advantages, particularly in leaving behind a rehabilitated land surface capable of beneficial use and in reducing the area of the subsidence cone. Surface water flows could be better managed to ameliorate community concerns. Underground water management would be simplified and the risks from mud rushes at drawpoints reduced, and if iron sulphides are present in the orebody, costs for acid drainage control and treatment would be greatly reduced. A separate tails storage facility would still be required but would be reduced in volume by up to 50%.  
Despite the advantages, the perceived risks have meant that backfilling tailings above an active cave mine appears not to be currently practiced. The major perceived safety risk is of a mud inrush with fatal consequences, as at Mufulira in 1970. Other risks are that ore may be contaminated by tailings, either by bulk flow through the broken rock, by mixing, or carried in percolating water; or that the tailings may prevent development of the cave or impede flow through drawpoints and ore passes. 
This paper seeks to identify the causes of each risk and argues through a combination of small-scale test work and analysis of documented fatalities caused by mud rushes that all these risks can be effectively mitigated by appropriate design and management practices. The key requirement is that tailings be placed in dewatered form, either thickened to near-paste (yield stress >30 Pa) or filtered, and that a self-draining surface is always maintained, with no ponding of water permitted. 
The test work conducted initially investigated migration of solids in seepage and bulk flow of near-paste copper tailings through a high voidage pebble bed. The tailings were observed to rapidly dewater, penetrating the bed to less than one pebble diameter. Liquefaction by shaking for two minutes resulted in flow to a maximum of three pebble diameters. There was almost no entrainment of solids in seepage. The potential for bulk flow of tailings through rock at high pressures, as the depth of tailings increases, was also tested using two different materials. With an applied head equivalent to 174 m of tailings, tailings penetrated the bed to a maximum depth of two particle diameters (20 mm). 
A physical scale model of a caving operation was constructed and thickened tailings deposited in layers, followed by drawdown of ore. The test showed no mixing of tailings and rock, even with funnel flow of rock lying immediately below the tailings to the drawpoints. It appears the cohesiveness and lower bulk density of the tailings impede mixing.
Larger scale and project specific testing would be required before implementing the concept presented, but the results provide strong support for considering tails backfill for a new cave mine.

Keywords: backfill, paste tailings, filtered tailings, cave mining, tropics