Question 7 of 7: Backfill Preparation and Placement

Question 7: Backfill Preparation and Placement (15 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

7.1 — preparing mill backfill. Whole flotation/gravity tailings leaving the mill are either (a) cycloned to split coarse underflow (used as hydraulic fill, de-sliming the fines that would otherwise drain poorly) from fine overflow (sent to the tailings facility), or (b) thickened in a high-rate or deep-cone/paste thickener to raise solids content, optionally followed by pressure or vacuum filtration to push solids content still higher, producing a non-segregating paste. In either route the underground-bound stream is then blended with a binder (cement, or a cement/pozzolan blend) in a mix plant before being pumped or gravity-fed down a borehole/pipe range to the stope.

7.2 — when backfill is essential. Backfill is essential wherever the mining method itself depends on it for wall/back support — principally cut-and-fill (each lift's fill becomes both the support for the previous void and the floor for the next) and any method used for pillar recovery, where the fill must replace the structural function of the pillar being extracted. It also becomes necessary, even in methods that could otherwise leave an open void, wherever the host or ore rock is weak, highly stressed, or prone to rockburst at depth: filling the void limits convergence and seismic energy release, which an open-stope method in the same ground would not control.

7.3 — matching a continuous mill supply to intermittent stope demand. The mill's tailings stream is essentially continuous (it runs with the plant), while stope filling happens in discrete campaigns as each lift or panel becomes ready. This mismatch is absorbed with surge capacity: a paste plant typically includes agitated storage/holding tanks or silos sized to buffer several hours to a day of production, and a hydraulic-fill circuit similarly uses a fill/tailings pond or surge tank; when no stope is actively receiving fill, the surplus is diverted to the tailings storage facility, and when a stope calls for fill faster than current mill production, the buffer (plus, if undersized, a temporarily reduced fill rate) covers the gap. Good mine-mill coordination (a fill schedule shared between the two) minimises how often either constraint binds.

7.4 — additives and strength. Ordinary Portland cement is the default binder, added typically in the range of 3–7% by dry mass depending on the target strength; it is effective but, at remote sites, is often the single most expensive input to the fill (trucked or barged in, sometimes seasonally, at a landed cost well above a mill-gate price). Fly ash (a coal-combustion by-product) and ground granulated blast-furnace slag act as pozzolans, partially replacing cement at a lower unit cost and often improving long-term strength gain and reducing bleed water, but their availability is itself remote-site-dependent (they must come from a power plant or steel mill, so supply and freight cost vary regionally, and some remote mines have none reasonably available). Ground glass and similar waste pozzolans are used opportunistically where a local, cheap fine aggregate source exists, on the same substitution logic as fly ash. In all cases, strength increases with binder content and with the fineness/reactivity of the pozzolan blend, but every percentage point of cement or its substitutes is a direct operating cost, so the mine designs to the minimum strength the ground control plan actually requires rather than a blanket high-strength mix.

7.5 — paste fill: production, transport, placement and difficulties. In the mill, whole tailings are thickened to a high, non-segregating solids content in a paste thickener, then blended with cement (and any pozzolan) in an in-line or batch mixer immediately before transport; the mixed paste is pumped underground through a steel pipeline (positive-displacement piston pumps for high-pressure sections, sometimes assisted by gravity in a vertical borehole) to the stope, where it is placed behind an engineered barricade (typically brick, shotcrete or a fill fence with a drainage/filter layer) that retains the paste while it develops early strength. Anticipated difficulties span all three stages: at the plant, maintaining a consistent, pumpable rheology (too dry and the line plugs; too wet and the fill segregates and loses strength) requires continuous density/yield-stress monitoring; in the pipeline, abrasive wear at bends and pump seals, and the risk of a static "plug" forming during any stoppage, both threaten blockages that are costly and slow to clear; and at the barricade, an under-designed barricade can fail under the fresh paste's hydraulic head before it gains strength, while in cold, remote northern operations the paste itself and the barricade drainage can freeze, both in the pipeline and immediately behind the barricade, if not adequately heat-traced or scheduled around the cold season.