Ore Sorting: Pre-concentration Before the Mill

What is ore sorting?

Ore sorting is a pre-concentration process that examines individual rocks or small batches on a conveyor and physically removes barren particles before fine grinding. Sensor-based sorting uses one or more sensor types—X-ray transmission, optical, near-infrared, or electromagnetic—to measure a distinguishing property, then triggers air jets or mechanical paddles to eject the selected fraction. This isn't a replacement for downstream flotation or gravity separation; it's a front-end waste rejection step that upgrades the mill feed. A typical installation sits after primary crushing and screening, where particles are large enough for reliable ejection. In a well-matched ore, the sorter rejects rock that would otherwise consume grinding energy and steel media for no metal return.

The sensor families: XRT, optical, NIR, and EM

X-ray transmission (XRT) sorting measures atomic density differences across a particle. It sees through surface coatings, so it is used to separate sulfide ore from quartz, barite from limestone, or diamond-bearing kimberlite from waste. Optical sorting uses visible-spectrum cameras and image analysis to classify particles by colour, brightness, or surface pattern; it's common for industrial minerals where colour contrast is strong. Near-infrared (NIR) sensors detect mineral-specific reflectance signatures in the short-wave infrared band. They can identify clays, talc, carbonates, and some lithium minerals like spodumene. Electromagnetic (EM) sensors respond to conductivity and magnetic susceptibility, separating conductive sulfides or magnetic iron oxides from non-conductive gangue. Each family has a detection envelope, and the right choice depends on the contrast between your ore and the waste you're trying to reject. You don't need a sensor that sees everything; you need one that sees the difference that matters.

Amenability testwork decides feasibility

You can't just install a sorter and expect it to work. Amenability testing follows a staged path. First, sample ore and waste zones separately. Second, measure the physical properties a sensor could read—atomic density by XRT, colour variation by optical, conductivity by EM, or spectral response by NIR. Third, run bulk pilot sorting on representative size fractions and compare mass pull, waste rejection, and metal recovery against a baseline. The critical question is not whether the sensors can see a difference; it's whether that difference occurs at the particle size you can feed and at the throughput your plant needs. If contrast only appears below a few millimetres, sorting may not pay. Xinhai's mineral processing test center can assess this, and the company reports that its test work covers more than 70 ore types. For a new project, you'll want to see a sorter screening pilot result before committing capital.

How pre-concentration changes mill throughput and energy

Pre-concentration alters the mass balance before the grinding circuit. When a sorter rejects barren rock, the mill receives a smaller tonnage of higher-grade feed. That means you can either maintain the same metal output by feeding more ROM tonnes to the sorter, or you can downsize the grinding circuit for a new project. Since grinding usually consumes the largest share of energy in a hard-rock concentrator, rejecting waste before that stage cuts specific energy per tonne of product. It also reduces wear on liners and media and lowers water consumption because less fine tailings are generated. The effect scales with waste rejection. Even a modest mass pull to reject can shift a marginal project into viability, particularly when the orebody has broad zones of barren dilution. For pegmatite lithium ores, the incentive is strong because, as the USGS National Minerals Information Center reports, batteries account for 88% of lithium end use, so any pre-concentration that lifts spodumene feed grade helps meet a high-demand market. The Canadian Institute of Mining, Metallurgy and Petroleum (CIM) also notes that comminution is often the largest energy consumer in a mineral processing plant.

What testwork should deliver

Amenability studies should produce four outputs: a mass pull estimate, a waste rejection curve, a metal recovery curve, and a throughput recommendation. Mass pull is the percentage of feed sent to the sorting reject stream. Waste rejection is the percentage of barren rock removed. Metal recovery is the percentage of valuable units retained in the accept stream. These numbers compete; you can reject more waste but often lose more metal alongside it. Testwork helps you find the balance. You'll also get a particle size specification—typically a narrow size band that the sensor can read efficiently—and an ejection accuracy figure. The last one matters because air jets sometimes hit the wrong particle; if the sorter misplaces too many high-grade rocks, the economics erode. Xinhai's pilot facilities can run these campaigns, and the company reports that its test program covers more than 70 ore types, giving the design team a basis to recommend a sorter configuration.

Sensor-based sorting versus other pre-concentration methods

Ore sorting is one of several pre-concentration tools. Dense medium separation uses heavy liquid or ferrosilicon media to separate by density; it works well for coarse coal, diamonds, and some base metals but requires media handling and cleaning. Gravity concentration in jigs or spirals separates by specific gravity but usually works on finer sizes and lower capacities per unit. Magnetic separation removes strongly magnetic minerals but only when the valuable or waste fraction is magnetic. Sensor-based sorting is different because it can respond to properties other than density or magnetism—colour, conductivity, atomic number, spectral signature—so it opens pre-concentration to ores that lack density contrast. The trade-off is that sorting is usually capacity-limited at fine sizes and requires a monolayer presentation. You'll choose among these methods after testwork shows which physical property is most reliable for your ore.

Integrating a sorter into your flowsheet

A sensor-based sorter fits between primary crushing and fine grinding. The usual sequence is primary crush, screen to a narrow size range, wash or dry the surface, present particles as a monolayer to the sensor, and eject rejects with air jets. For coarse sorting, particle sizes in the tens of millimetres are typical, but the actual range depends on the sensor and ejection speed. Fine particles below a few millimetres are usually not sorted because the ejection becomes inefficient and the mass per particle is low. This is why the sorter works alongside the screening circuit. You'll often need a separate belt or chute for accepts and rejects, plus a means to return the rejects to the waste dump or to a secondary crusher if they contain occasional ore. Xinhai's engineers can evaluate the flowsheet integration during design, using pilot data to position the sorter where it protects the mill without losing metal.

Is ore sorting right for your ore?

Sensor-based sorting works best when ore and waste have a measurable contrast and the waste occurs as discrete particles large enough to eject. If your deposit has fine-grained dissemination or the valuable mineral is uniformly distributed through all rock, sorting won't help. If your orebody has barren dykes, low-grade stockwork, or dilution bands, it probably will. The decision should come from testwork, not vendor claims. Xinhai reports more than 600 EPC+M+O projects and serves mines in over 100 countries, according to the company's published figures. That experience often starts with bench-scale amenability tests and pilot campaigns before a sorter is included in a project design. The goal isn't to add a new machine; it's to remove the cost of grinding rock that carries no value.

Frequently asked questions

What is ore sorting in mineral processing?

Ore sorting is a pre-concentration process that uses sensors to identify and reject barren rock particles before they enter the grinding circuit. It upgrades mill feed and reduces energy spent on waste.

Which sensor type should I choose for my ore?

The choice depends on the contrast between your ore and waste. XRT works on atomic density, optical on colour, NIR on spectral reflectance, and EM on conductivity or magnetism. Testwork determines which property is most reliable.

Does ore sorting always improve plant economics?

No. It works only when ore and waste have a measurable contrast and waste occurs as discrete particles large enough to eject. If the valuable mineral is finely disseminated and uniform, sorting may not help.

How does pre-concentration affect grinding energy?

By rejecting barren rock before grinding, pre-concentration reduces the tonnage entering the mill, cutting specific energy consumption and wear while raising the effective capacity of the existing grinding circuit.