‘Prediction is very difficult, especially if it’s about the future’ (Nils Bohr).
‘The future ain’t what it used to be’ (Yogi Berra)
In the run-up to the mid-20th century, the technology of mining had not changed significantly over several decades or even centuries but the development of bulk-minable lower grade orebodies, the use of bigger equipment and the rise of global mining companies led to a step change in the last fifty years. The use of robotics and big data stands to create a step change in how mining works into the 21st century.
The clearest example is the development of the Pilbara iron ore operations in Western Australia, which is now close on a billion tonne per year operation between the major players so that the region is now the biggest source of iron ore on the planet. In effect, the leading companies like Rio Tinto have created a massive logistical operation whereby there is a mine at one end in Australia and the customer in China, Taiwan, South Korea or Japan at the other. In Rio Tinto’s case, 15 mines and a rail network linked to four ports permit a mining, transportation, blending and shipping operation that matches the individual requirement of the customer and can do this for less than US$15 per tonne, net of capital costs. In response to the competitive advantage of dominating the lowest cost quartile (and thereby remaining profitable under any price pressure in a cyclical market – which mining is, thanks to the tendency to oversupply in response to higher prices created by demand), automation is being rolled out to more than 50% of the mines in the region. This is achieved by automated drilling, autonomous haulage systems (AHS™), the roll-out of automated trains (Autohaul™) and bulk blending and ship loading systems. All of this can be controlled through pit supervisors in the mine and through an overall control in an air-conditioned building in Perth.
While the expectation is that automation will gradually become normal in open pit mining operations, the pits themselves are getting deeper with age. The ‘low-hanging fruit’ of shallow deposits is becoming a thing of the past as such orebodies can only be accessed by underground mining. The constraints on open pit mines are simply that it costs more to go deeper, waste mining increases as the ratio of waste material increases compared with ore in order to maintain slope stability. This cannot always be achieved, despite all the measures in place to combat creep and potential slope failure – as demonstrated so vividly at Bingham Canyon in Utah in 2013. Bulk underground mining will gradually replace many open pits as the deeper portions of the orebodies are accessed. The most effective tool that has been in use on a number of mines in recent years is block caving, whereby a reticulated series of access tunnels and extraction slots are developed at some 300m depth intervals. Once the network is established, a blast loosens the rock immediately above the drawpoints and continuous extraction causes the overlying rock to ‘fail’ and replace the void that has been created. This requires a very strong knowledge of rock mechanics and not all orebodies are amenable to block caving.
An impressive example of long-term mine development is Oyu Tolgoi in southern Mongolia, where a deep block cave is being created over the next five years to partially replace depletion of the main open pit. The project’s scale is impressive. Over the next 16 years, the development requires some 203km of tunnelling, five shaft systems, and more than 2000 drawpoints over an orebody footprint of 2000m by 280m and at a depth of 1300m. Such developments can only be justified where the ore grades are high enough. The deep ‘Hugo North’ deposit reserves are estimated at 499 million tonnes at a grade of 1.66% copper and 0.35 grammes per tonne gold within a resource shell (JORC-measured and indicated) of a further 495 million tonnes at 1.26% copper and 0.25 grammes per tonne gold. The whole deposit amounts to well over 5 billion tonnes of potentially-exploitable copper. it is increasingly common for such orebodies to be exploited by a degree of automation, which keeps the operators away from the working face and potential danger.
The recovery of mined materials is a perpetual challenge and, while some 20% of copper can be recovered by heap leach solution extraction and electrowinning (SXEW) and 12% of gold by heap leach cyanidation, the recovery from sulphide ores requires crushing, comminution, flotation and drying with the resulting concentrates having to be transported to smelters. This is energy-intensive and self-evidently costlier. The ‘Holy Grail’ of mineral processing remains the discovery of a sulphide leaching method. The nearest approximation so far is the ammonia leach method AmmLeach™ developed by Mintek and Alexander Mining. However, while it works at the pilot plant scale it has not yet succeeded at the production level. Meanwhile, major companies pare costs by troubleshooting processing glitches at various recovery plants in real time by the use of robotic XRD analysis and the use of big data and feeding the information to consultant process engineers wherever they happen to be. That way, the processes suffer fewer failures.
Exploration’s challenge is to find the higher grade deposits, given that the ‘low-hanging fruit’ might have all been picked. The productive geology often lies beneath more recent ‘cover’ rocks and the need is to utilise and refine geochemical and geophysical data and techniques to make up for the decline in exploration success in the last few years during which expenditure has increased enormously. Cost-effective exploration techniques being developed include the use of cloud computing public domain data (where it exists – mainly countries like the US, Canada, Australia and most of western Europe) and the development of geophysical methodologies like 3D seismic processing (to understand how major orebodies sit in the global tectonic environment), magneto-tellurics to search beneath ‘cover’ or extend known orebodies to depth, and airborne gravity to an accuracy that allows deep, dense features like ore deposits to manifest themselves. In the last-mentioned case, the desired accuracy of 1 Eö is likened to an Olympic archer in Rio de Janeiro aiming for a 10cm bullseye in Melbourne whereby, operating at cryogenic temperatures on a highly balanced beam, measuring mass movements less than the width of a hydrogen atom via superconductivity on a platform moving at >60 metres per second within 100 metres of surface (with turbulence) is not easy. But, remarkably, it is increasingly likely that this can be achieved.
The companies that are most adaptable to change will be the fittest in the coming years and decades. That is the challenge that mining faces.