Ore sorting — using sensors to identify and reject waste rock before it enters processing — has existed for decades but remained niche technology. Recent advances in sensor technology, computing power, and handling systems have broadened potential applications. Several Australian operations now use ore sorting in production, with varying results.

The technology is straightforward in principle: scan material, identify ore versus waste based on some detectable property, and physically separate the streams using compressed air jets or mechanical diverters. In practice, economic viability depends on ore characteristics, processing costs, and capital investment requirements in ways that make generalizations difficult.

How Ore Sorting Works

Multiple sensing technologies identify ore-bearing material:

X-ray transmission (XRT) detects density differences between ore and waste. High-density sulphide minerals in waste rock can be identified and rejected. XRT works for base metal deposits where ore minerals have significantly different density than gangue minerals.

X-ray fluorescence (XRF) detects specific elements directly. This enables sorting based on actual metal content rather than proxies like density. XRF is particularly useful for deposits where visual or density characteristics don’t reliably predict grade.

Near-infrared and visible light spectroscopy identifies minerals based on their reflectance properties. This works well for minerals with distinctive optical properties — certain iron oxides, carbonates, and silicates can be distinguished optically.

Electromagnetic sensors detect conductivity differences. Useful for sorting conductive sulphide minerals from non-conductive waste.

Modern sorting systems combine multiple sensor types to improve discrimination accuracy. Companies like TOMRA and Steinert offer multi-sensor systems that can identify material based on several properties simultaneously.

The Economic Value Proposition

Ore sorting creates value through several mechanisms, though not all apply to every operation:

Waste rejection before milling. Removing waste rock before grinding reduces energy consumption and increases effective mill throughput. For operations that are mill-constrained, sorting can increase metal production without expanding milling capacity. The value here scales with energy costs and the cost of additional milling capacity.

Grade improvement. Increasing head grade to the mill improves recovery rates in some processing circuits and reduces reagent consumption per unit of metal produced. The benefit varies significantly by mineralogy and processing method.

Tailings volume reduction. Less material through the mill means less tailings to store. For operations approaching tailings storage limits, this can defer expensive tailings facility expansions.

Lower-grade material treatment. Sorting can make marginal ore economic by upgrading it to processable grade before milling. Material that would otherwise be stockpiled or sent to waste can contribute to production.

When Ore Sorting Works Well

Certain conditions favor ore sorting economics:

High contrast between ore and waste. If ore minerals are strongly distinguishable from waste — large density difference, distinctive elements, clear visual contrast — sorting accuracy improves and false reject/accept rates decline. High sorting accuracy directly translates to better economics.

Coarse ore fragments. Sorting works best on material in the 50-300mm size range. Finer material becomes difficult to handle mechanically, and sensor resolution becomes limiting. Operations where ore can be sorted at relatively coarse sizes avoid the energy cost of crushing to finer sizes.

High milling and processing costs. The value of waste rejection scales with avoided processing costs. Deep underground operations in remote locations with expensive power and reagent logistics see the highest value from reducing mill throughput.

Mill or tailings capacity constraints. If an operation is constrained by milling capacity or approaching tailings storage limits, ore sorting can expand production or extend mine life without the capital cost of expanding those facilities. The value of deferred capital can be substantial.

When Ore Sorting Doesn’t Work

Several factors undermine sorting economics:

Low ore-waste contrast. If ore and waste minerals are similar in density, elemental composition, and optical properties, sorting accuracy suffers. High false rejection rates (discarding ore as waste) and high false acceptance rates (treating waste as ore) both reduce economic benefit.

Fine-grained ore. Deposits where ore minerals occur as fine disseminations within waste rock don’t sort well — individual rock fragments contain both ore and waste in proportions that don’t differ much between fragments. Sorting requires that ore minerals be concentrated in specific fragments that can be separated.

Low processing costs. Operations with cheap power, low reagent costs, and unconstrained capacity may find that the capital and operating cost of ore sorting exceeds the value of reduced processing tonnage. The economics tilt against sorting when avoided costs are low.

Complex mineralogy. Deposits with multiple ore minerals requiring different processing conditions, or where deleterious elements must be managed carefully, can become more difficult to process if sorting changes feed characteristics unpredictably.

Capital and Operating Costs

Ore sorting systems aren’t cheap. A production-scale sorting plant processing 500-1000 tonnes per hour costs $15-40 million AUD including sensors, conveyors, diverters, and supporting infrastructure. Operating costs include power, compressed air, maintenance, and control system monitoring.

The payback calculation is straightforward: does the value of reduced processing costs, increased throughput, or deferred capital exceed the cost of the sorting system over its operating life (typically 10-15 years)?

For many operations, the answer depends critically on mine life. Sorting systems require several years to recover capital investment. Short remaining mine life reduces the benefit window and makes economics marginal.

Implementation Considerations

Installing ore sorting isn’t just buying equipment — it requires integration with existing operations:

Material handling modifications. Sorting requires crushing material to appropriate size, presenting it to sensors at consistent speed and spacing, and handling the separated ore and waste streams. This often means significant conveyor and crushing plant modifications.

Feed variability management. Sorting performance depends on consistent feed characteristics. Changes in ore type, hardness, moisture content, and size distribution affect sorting accuracy. Operations with highly variable ore require more sophisticated control systems and may see less consistent benefits.

Downstream process adjustment. Changing mill feed grade affects optimal grinding, flotation, or leaching conditions. Process control systems must adapt to higher head grades, and operators need training on managing the modified feed characteristics.

Case Examples

Several Australian operations demonstrate where ore sorting succeeds:

Coarse gold operations. Sorting to recover coarse gold particles at early stages prevents gold losses in subsequent crushing and reduces security requirements for gold in process streams.

Nickel sulphide operations. High-contrast density differences between sulphides and silicate gangue enable effective XRT sorting. Several Australian nickel mines use sorting to upgrade low-grade ore to mill-grade material.

Iron ore. Large-scale iron ore sorting based on XRF and optical sorting upgrades lower-grade material and reduces processing plant feed tonnage.

These successful applications share common characteristics: clear ore-waste distinction, economic value from waste rejection or grade upgrading, and capital available for the required investment.

The Future of Ore Sorting

Technology continues advancing. Faster processors enable higher throughput. Better sensors improve discrimination. Machine learning algorithms optimize sensor interpretation and sorting decisions in real-time based on operational data.

The economic sweet spot for ore sorting is expanding as technology improves and as mines pursue marginal ore bodies where small grade improvements determine viability. Operations that dismissed ore sorting five years ago based on marginal economics might find the calculation has changed.

But ore sorting isn’t a universal solution. It’s a specialized tool that works excellently in the right circumstances and poorly in others. The key is honest assessment of whether specific ore characteristics and operational economics fit the technology’s strengths, rather than assuming that new technology automatically improves outcomes.

Ore sorting will become more common in Australian mining as technology matures and as operations seek efficiency improvements without major capital expansion. Understanding where it works and where it doesn’t is the difference between valuable investment and expensive disappointment.