PRECONCENTRATION AND ORE SORTING
capable of measuring a wider array of metals and lighter elements, a back calculation of mineralogy or gangue can be completed to enable‘ value add decisions’ according to metallurgical attributes. Elements such as Cu, Fe, S, Si, Al, Mg, Ca, K, Na, Mn( plus more) can be measured accurately to allow for discrimination for ore destinations such as a leach pad or a sulphide mill.”
The Thermo Scientific GS Omni slurry PGNAA analyser extends the same philosophy into the processing plant. Using a labyrinth flow cell and multiplexed sample handling, GS Omni( Gamma Slurry) measures multiple slurry streams in real time, capturing both major metals and key gangue or deleterious light elements that drive recovery and reagent consumption.
“ Together, CB Omni Agile on the conveyor and GS Omni in the plant provide a consistent sensing backbone from pit to product,” Rutter explains.“ They allow miners to apply bulk ore sorting and downstream process control using continuous, representative PGNAA data rather than sparse sampling alone – and to adjust the‘ modifying factors’ of BOS( SDU size, cut-off thresholds, diversion logic) as ore types, market conditions and operating strategies evolve.”
NextOre refines magnetic resonance proposition
As NextOre Ltd CEO Chris Beal settles into his new environment within Australian deeptech company MagnaTerra Technologies Ltd he remains convinced of the ore sensing and sorting value case he and his team have been proving up since the magnetic resonance-based technology now under NextOre’ s guise was spun out of CSIRO in 2017.
NextOre and MRead Ltd completed a merger and A $ 11 million($ 7.7 million) fundraising last year, forming MagnaTerra, with the intent of expanding the two business’ MR detection capabilities for explosives, critical minerals for bulk sorting and iron ore sensing applications.
“ The strategy for the new company relies on both divisions continuing at a great pace and, ideally, accelerating the pace that they can go at,” Beal told IM.“ Boosting sales in the minerals space is an essential ingredient to MagnaTerra’ s success, as is fostering the opportunities within explosives detection.”
MR sensing has, to date, been proven in copper operations, with several on-conveyor installations showcasing its ability to provide grade estimates in seconds, with no material preparation requirement. This helps deliver run of mine grade readings, providing“ complete transparency” for tracking downstream processing and allowing operations to selectively reject waste material, according to the company.
The technology pulses radiofrequency waves into ore, which are tuned to the signature resonant frequency of target minerals. Response strength corresponds to mineral content, so it is a direct, quantitative and penetrative measurement technique, according to the company.
One of several interesting case studies to surface over the last year was a paper presented at Procemin-Geomet 2025, in Santiago, Chile. This paper, authored by Minera Candelaria’ s Jose Luis Bello, as well as Beal and his colleagues Jesus Caro and Bernardo Ossandon, showcased how NextOre’ s real-time grade monitoring can provide up- and downstream benefits from its normal captive position – on a conveyor belt just after the primary crusher.
“ The interesting point to start with at Candelaria is the fact that it was the geological team that had the reason to install the analyser in the first place, and they continue to‘ own it’,” Beal explained.
Candelaria, an operating copper mine in the Atacama region of Chile that Lundin Mining owns 80 % of, produces between 150,000 t / y and 180,000 t / y of metal from some 78,000 t / d of mined ore. The mine sources this tonnage from one open-pit mine( Candelaria), two underground mines( Santos and Alcaparrosa) and five distinct intermediate feed stockpiles.
The operation, prior to installing the MR analyser, had been struggling to reconcile the modelled mining areas with the annual tonnage-weighted grade. Between 2018 and 2021, there was a persistent negative error between the two, producing less metal despite achieving mined tonnage targets. Mine personnel sought to improve on conventional geological feedback systems and implement a system using modern instrumentation that would allow real-time and precise comparison of extracted grades against resource and grade control models.
This is where NextOre and its MR analyser came in, with a unit installed at the primary crushing discharge end. Thanks to the system’ s analysis speed and detection accuracy, the operation experienced a notable reduction in variance against forecast of approximately 50 %, going from-4.1 % to-1.8 %, or the equivalent of over $ 30 million of copper.
But the MR analyser has done more than improve the reconciliation process between the model and the actual processed grade.
Now seeing some 6,500 t / h of material across the conveyor belt, the system’ s outputs are being integrated with other data sources such as grade control, information from the fleet management system( FMS) and XRF-based online analysers positioned after milling.
“ We’ re now being told that the geologists are assisting operations with retrospectively selecting digging equipment and digging directions based on the granular data they are receiving from the conveyor belt, which, because it is real time, is having a significant impact,” Beal says.
He continued:“ In general, it is tricky to convince people of how valuable our MR sensor data is going to be for the operation, but the Candelaria example shows how valuable it is – and that is even without having a diverter fitted on the conveyor belt!”
It appears this type of on-conveyor value case is one the company wants to replicate going forward.
The open geometry developments NextOre has achieved are notable – the OG3 being a 3-m diameter MR sensor that could, in theory, be placed in an underground mine for truck tray grade detection and the OG7 7-m diameter MR sensor being used for open-pit truck tray analysis – yet they require mining operations to reconsider their operational status quo. This could be pausing for 20-60 seconds to take grade readings of a truck-load full of material or diverting off the normal haul road to find a fixed MR sensor to conduct such analysis.
“ The mining companies we have engaged with on such projects are normally very enthused by the technology’ s potential, but adding steps to the standard operating procedure on site can dampen that enthusiasm,” Beal explained.“ But what you can see at Candelaria is a‘ virtual truck analyser’; it gives you the same real-time data in the pit with a delay that allows you to make calls about where to dig and how to process that tonnage in the future.”
It is this philosophy that the company is taking forward into its newest target market: lithium.
The MagnaTerra announcement last year hinted that the company would be heading in this new direction, leveraging all its experience in copper while tailoring the MR hardware to analyse lithium.
Beal was able to share more in early February.
“ We have partnered with a major group and CSIRO, Australia’ s National Science Agency, on the development of a direct detection system for lithium,” he said.“ It is on a conveyor again and it will target hardrock lithium applications. While there is a pretty dramatic change in the MR hardware, we think it will still preserve the benefits of the conventional system we have had for copper.”
CSIRO completed a series of lab tests with an advanced, bulk-scale laboratory prototype in 2025, and now the partners are nearing the point of industrialising the system.
“ Testing has been carried out at scale on
64 International Mining | MARCH 2026