PRECONCENTRATION AND ORE SORTING
Challenging the low-grade narrative
The mining sector should stop striving for lab-based precision and focus on making staged orebody knowledge improvements with these technologies, Dan Gleeson hears
Current industry perception is that declining feed grades are an unavoidable consequence of ore deposit geology and mass mining technologies for increasingly mature mining operations, or the“ new generation” of low-grade operations. While true at the front“ digging” end of mining operations, the old adage of“ grade is king” is resurfacing, according to Jon Rutter, Business Development Manager for Mining and Mineral Processing at Thermo Fisher Scientific. This is where applying the lever of bulk ore sorting( BOS) can impact the economics of an operation.
Rutter explains:“ The grade in the ground may be lower, but the opportunity exists to significantly enhance the delivered grade to the already installed( or planned) capacity of the recovery plant by exploiting the retained heterogeneity.”
An industry focus on throughput as the main driver of revenue has led to a bulk average mentality with respect to in-situ cut-off grades. In many cases, average grades used to define bench- or stope-scale processing destination decisions such as mill, dump leach, waste, etc include significant sub-volumes of material outside cut-off specifications, Rutter says. An averaging approach ignores potentially exploitable grade heterogeneity below the scale of the minimum mining unit even though significant localised grade heterogeneity is a dominant characteristic of many base and other metal deposits and ore types.
“ My experience in deploying BOS solutions in mining operations has demonstrated that adequate heterogeneity is retained in most orebody systems despite the mixing and’ smoothing’ effect of the blast, load, haul and crush interface,” he told IM.
Many detractors of BOS will claim that adequate heterogeneity does not exist in bulk mining deposits, but Rutter’ s experience has so far shown that to not be the case.
“ The ultimate test is presented in many porphyry copper processing plants,” he explains.“ Despite being a generally disseminated‘ bulk-style’ mineralisation, mined in 15-m benches, several hundred thousand tonne blasts, 80-100 t shovel buckets, 300 t trucks, 2,000-8,000t / h crushers, blended and crushed to circa 100 microns in 160,000 t / d milling and flotation plants, a significant heterogeneity signature is still present in the online sampling analysers!”
A key focus question for BOS is‘ what’ are you trying to achieve and‘ where / how’ do you wish to operate? The focus word is in the title: bulk.“ We do not need assay lab precision,” Rutter says.“ We need to‘ measure’( or more accurately‘ sense’) the metal content of the pod of material adequately enough to be‘ right’ far more often than we are wrong. The measure of‘ rightness’ needs to incorporate the error bar of the entire system. This can
The NextOre MR unit installed at the primary crushing discharge end of the process plant at Candelaria Mine
include precision of the sensor, integration time required( commonly linked to precision), the metals( or non-metals) being targeted and the method or location of deployment.”
This commonly distils down to what the planned ore loss and dilution in the BOS system is, with the decision( accept or reject) being above or below a threshold – commonly being either grade or a geometallurgical attribute.
“ What is not well discussed or recognised in the industry is that these factors are not‘ fixed’,” Rutter says.“ They are customisable and can be modified to an operation’ s needs.”
They are similar to“ modifying factors” in the mine planning and scheduling sequence, whereby mine planning selects a SMU( smallest mining unit) with a chosen ore loss and dilution suitable to the mineralisation style and scale of mining.
In BOS you can select a SDU( Sensor Diversion Unit), which could allow the same 15,000 t dig block selected as the SMU to have smaller SDUs dependant upon the heterogeneity and the sensors themselves. These modifying factors can include:
• Sensor type:
o Elemental – Prompt Gamma Neutron Activation Analysis( PGNAA), X-ray Fluorescence( XRF), Laser Induced Breakdown Spectroscopy( LIBs), etc;
o Mineralogical – hyperspectral imaging( HSI), magnetic resonance( MR);
• Requirement for surface, partial or fullypenetrative measurements;
• Base metal mines( Cu, Ni, Pb, Zn, etc) versus bulk commodities( cement, phosphate, manganese, iron ore, etc);
• Precious metals( eg: arsenic);
• Precision required( eg: 0.1 % Cu +/- 0.01 %);
• Integration time( eg: 15 seconds, 30 seconds, 1 minute, 5 minutes, etc); and
62 International Mining | MARCH 2026