Northern’s dysprosium hopes with Brown

Published: Wednesday, 29 October 2014

Andrew Scogings, IM Correspondent, looks into Northern Minerals’ Brown project in Western Australia and discusses the rare earths markets it is targeting.

Northern Minerals Ltd is focussed on becoming a globally significant producer of the heavy rare earth (HRE) dysprosium1. Northern has large landholdings in Western Australia and the Northern Territory which are considered highly prospective for this element.

Northern’s flagship project is the Browns Range project, where it has a number of deposits and prospects (Figure 1) containing high value dysprosium and other HREs, hosted in xenotime mineralisation. Dysprosium is an essential ingredient in the production of NdFeB (neodymium iron-boron) magnets used in clean energy and high technology solutions. The project’s xenotime mineralisation facilitates the use of a relatively simple and cost effective processing flowsheet to produce a high grade dysprosium rich mixed rare earth oxide.

Northern claims that one of the key features of the project is the simplicity of the rare earth mineralogy, dominated by xenotime (yttrium rare earth element (REE) phosphate) which is only one of a few rare earth minerals that has a track record of successful separation. The xenotime mineralisation and mainly silica host rock offers a key advantage as it allows the ore to be readily upgraded through the beneficiation process by a factor of 30 times to achieve a 20% total rare earth oxide (TREO) mineral concentrate.

Northern announced an updated JORC compliant mineral resource estimate for the Browns Range project on 26th February 2014. Resource delineation and estimation has been completed on four deposits, known as Wolverine, Gambit, Gambit West and Area 5 (Figure 2). The deposits remain open at depth and have further exploration potential. Several additional prospects displaying similar geological characteristics to Wolverine have also been discovered at Browns Range which, according to Northern, supports the potential for further dysprosium discoveries.

Over 74,000 metres of drilling underpins the current total mineral resource at the Browns Range project which is estimated at 6.48 m tonnes at 0.74% TREO containing 4,000 tonnes dysprosium within 47,997 tonnes TREO using a cut-off grade of 0.15% TREO (Table 1). At the Wolverine deposit the total mineral resource is estimated at 4.46 m tonnes at 0.86% TREO containing 3,300 tonnes dysprosium within 38,269 tonnes TREO using a cut-off grade of 0.15% TREO. Of the total mineral resource, 66% is classified as indicated resource, with the remainder in the inferred resource category.

A maiden ore reserve for the project of 3.4m tonnes of ore containing 2,048 tonnes dysprosium and 23,595 tonnes TREO was reported on 24 June 2014. The ore reserve statement is based on the outcomes of the Browns Range pre-feasibility study (PFS) and the mineral resource estimate announced on 26 February 2014. The ore reserve is classified as 100% probable ore reserve (Table 2).

The PFS is based on a conventional mining operation involving both open cut and underground operations and a relatively simple processing flowsheet with all infrastructure located on site. The project includes a base case production rate of 279 tpa of dysprosium, contained within 3,200 tonnes of high purity mixed rare earth oxide (Table 3).

Northern’s managing director George Bauk said that, “The completion of the PFS is another major milestone for Browns Range, and has confirmed the project is robust and well positioned to become a significant global dysprosium supplier. We have a quality ore reserve supporting a solid mining operation, with a key feature being the project’s xenotime mineralisation.”

“In particular, the PFS has reinforced that the xenotime mineralisation at Browns Range is our key competitive advantage - its richness in dysprosium and predictable processing allow us to significantly concentrate the ore through the beneficiation process and has delivered a competitive cost estimate. The positive results from the PFS will support the project’s continuation to feasibility study (FS), as we move toward a 2016 production target. Once in full production, the project will deliver $173m average annual operating free cash flow, which will be an outstanding result for shareholders,” Bauk said.

Exploration history

Interest in the exploration potential of the area was generated by reconnaissance mapping carried out by the Bureau of Mineral Resources (BMR) in the late 1950s. The first record of commercial exploration was in the early 1960s by New Consolidated Goldfields, with the area attracting various phases of gold, base metals and uranium exploration between 1960 and 2010. Anomalous rare earth elements in outcrop were first identified through evaluation of radiometric targets by PNC Exploration Australia Pty Ltd during their uranium exploration programmes between 1987 and 1992.

In 2010, preliminary exploration work by Northern identified high grade rare earth mineralisation at the Wolverine and Gambit prospects. An inaugural drilling programme was completed in 2011 at the Wolverine, Gambit, Area 5 and Area 5 North prospects.


The deposits can be characterised as breccia-hosted hydrothermal systems with the dominant mineralisation being the rare earth phosphate mineral xenotime, which is a rich source of the heavy rare earth element dysprosium, plus yttrium (YPO4). Overall, the mineralisation at Wolverine and Gambit has a high heavy rare earth oxide (HREO) proportion to the TREO, being between 85% and 90%.

Locally, at Wolverine, the host rocks are a sequence of meta quartz-lithic arkosic arenites and conglomerates with minor interbedded schists. The host rocks in the mineralised zone are silicified and brecciated along structures trending between east-west and 290 degrees (Figure 2), and dipping steeply to the north (Figures 3 and 4). Hematite and sericite alteration is associated with mineralisation.

Xenotime is associated with varying degrees of veining and brecciation (Figure 5); from 1mm to 2mm crackle vein selvages to matrix infill in 5 metres wide zones of chaotic breccia. There are open spaced textures, vugs and minor cross-cutting quartz, pyrite or barite veins that are interpreted to post-date mineralisation.

Minerals of the florencite ((Nd,La,Ce)Al3(PO4)2(OH)6) - goyazite (Sr Al3(PO4)2(OH)5.H 2O) series are the only other rare earth element minerals (besides xenotime) recognised to date.

Exploration drilling at Wolverine

Diamond core drill holes accounted for 40% of the drill metres within the mineralisation and comprised NQ and HQ core. Reverse circulation (RC) drilling accounts for the remainder with diameters of either 115mm or 140mm.

Drilling has been completed on a nominal 25 metres in easting by 25 metres in northing grid spacing, although this increases to broader spacing at the lateral extremities of the deposit. The spacing of down-hole intercepts of the mineralisation varies from the nominal collar spacing due to deviation of drill holes. Prior to October 2013, resource drilling was exclusively conducted at -60 degrees to the south. From October 2013, diamond drilling was completed using casing wedges and directional drilling, resulting in variable intersection angles to the Wolverine deposit.

Sampling techniques at Wolverine

Diamond core was cut in half using an electric core saw. Sample intervals were selected on the basis of lithological and structural features, together with indicative results from hand held XRF measurements. Drill core was sampled at a nominal one metre interval, but constrained within geological intervals.

RC samples were collected from the drill rig by either riffle splitting or using a static cone splitter. All samples were collected dry with a minor number being moist due to ground conditions or excessive dust suppression. RC drill holes were sampled at one metre intervals exclusively and split at the rig to achieve a target 2-5 kg sample weight.

Field QAQC procedures included the field insertion of certified reference materials (standards), blanks and duplicates. Earlier drilling (2011 to July 2012) did not include the insertion of standards as suitable materials were not sourced. Blanks were developed from local host rock following chemical analysis. Field duplicates were collected by either a second sample off the splitter (RC) or by quarter core samples of the original half core sample (diamond) and separate submission and analysis at the laboratory. Insertion rates averaged 1:20 for duplicates, blanks and standards, with increased frequency in mineralised zones.

Determinations of bulk density were completed by a combination of core immersion techniques and downhole density surveys with values typically in the range 2.10 g/cm3 to 3.40 g/cm3.

Sample analysis method

Diamond and RC samples were dried, crushed, split and pulverised by Genalysis Laboratories in Perth prior to analysis of rare earth element suite using ICP-MS. The sample preparation techniques employed for the diamond and RC samples follow industry best practice.

Samples assayed by Genalysis for rare earth elements were fused with sodium peroxide within a nickel crucible and dissolved with hydrochloric acid for analysis. Fusion digestion ensures complete dissolution of the refractory minerals such as xenotime.

The digestion solution, suitably diluted, was analysed by ICP Mass Spectroscopy (ICP-MS) for the determination of the rare earth elements (La - Lu) plus Y, Th, U, Sr, W and As.

Mineral resource and reserve

Resource classification is based upon continuity of geology, mineralisation and grade, using drill hole and density data spacing and quality, variography and estimation statistics (such as number of samples used, estimation pass, and slope of regression). Parts of the estimate poorly supported by drilling have not been classified as mineral resource.

The ore reserve (Table 2) is based entirely on the mineral resource (Table 1), as released, in accordance with the JORC 2012 Code, by Northern on 26 February 2014. The quantity and classification of this mineral resource is stated below in Table 1. The mineral resource is inclusive of the ore reserves.


AMC Consultants Pty Ltd (AMC) prepared a PFS level technical mining study on behalf of Northern and advised that the mining aspects that underpin the Browns Range ore reserve are technically feasible and economically viable. As part of this study AMC:

1 Identified suitable open pit designs for Wolverine, Gambit, Gambit West and Area 5 deposits.

2 Selected a suitable underground mining method for Wolverine and Gambit West.

3 Estimated costs for mining operations under assumption of contractor mining.

4 Reviewed and validated all other inputs provided by Northern (including general and administrative costs, processing costs and recoveries, other overhead costs, metal pricing) as being suitable to support the ore reserve estimate.

Mine design and optimisation

The open cut mine designs are based on pit shells (Figure 6) produced by AMC using GEOVIA Whittle 4X software. Several pits were generated at increasing revenue factors. All pit shells selected were conservative when compared against the Whittle maximum cash flow pit.

The proposed mining technique at the Wolverine underground mine is sub-level open stoping (Figure 7) with post extraction backfill using a mixture of cement stabilised aggregate and run-of-mine waste. The proposed mining technique at the Gambit West underground mine is narrow bench and fill stoping with post extraction backfill using run-of-mine waste.

Proposed production process

The combination of xenotime with a silica host rock offers a key advantage in simple and cost-effective beneficiation, allowing the ore to be readily upgraded by a factor of 30 times to achieve a 20% TREO mineral concentrate.

The proposed mining fleet will transport the run-of-mine ore to the run-of-mine pad where the ore will be stockpiled and blended to the desired grade before transferring to the crushing circuit which consists of a primary jaw crusher, followed by a 750kW 4.8 metres diameter x 2.4 metres SAG mill and then a 1700kW 4.2 metres diameter x 6.2 metres ball mill that grinds the ore down to a size of 80% passing 63 metres.The ground ore is fed to a wet high gradient magnetic separator (WHGMS) that produces two products, a magnetic concentrate rich in xenotime and iron oxide, and a non-magnetic stream containing largely silica and mica which is rejected as tailings. The magnetic concentrate is then fed to a flotation circuit where selective reagents are used to collect the xenotime material in the froth and reject unwanted gangue material in the tailings. The flotation tailings are combined with the WHGMS circuit tailings and thickened before being comingled with the hydrometallurgical tails and pumped to the tailing storage facility. The 20% TREO flotation concentrate is thickened, filtered and stored in bunkers prior to being fed into the hydrometallurgical plant.

The 20% TREO mineral concentrate is reclaimed from bunkers with a bobcat, fed into a live bottom bin and screw conveyed into a dryer. The dry concentrate is fed into an acid mixer and then the kiln. The sulphation bake is performed at 275-300 oC which cracks the xenotime mineral and the rare earths are readily leached in water. Following the water leach step, the leach residue is washed, filtered and separated from the pregnant leach solution (PLS). The PLS undergoes a series of purification steps where the pH of the solution is steadily increased with lime and magnesia to reject impurities such as phosphate, iron, aluminium, thorium and uranium. The purification residue is separated from the PLS by thickening and filtering, and the PLS is passed through an ion exchange column to remove any residual uranium. The purification residue is repulped and mixed with the repulped water leach residue before being comingled with the beneficiation tailings and pumped to the TSF. Following purification and ion exchange, the PLS is contacted with oxalic acid which selectively precipitates the rare earths. The mixed RE oxalate is thickened, filtered and washed before being calcined to a high purity, mixed RE oxide (Figure 8).

Hydrometallurgy - pilot scale testing

Northern Minerals has been conducting a series of continuous pilot scale testing of the project’s hydrometallurgical processing plant at ANSTO in New South Wales (NSW). The third and final five-day continuous pilot plant run during April 2014 achieved the best recovery results of the test work to date, including 92% for dysprosium and 92.6% for TREO. These recoveries have improved from the results from the second test run, which were 90% for dysprosium and 88.2% for TREO. This increase in recovery will result in an additional six tonnes of dysprosium and 157 tonnes of TREO being produced on average per annum. Using the PFS input data, this recovery improvement equates to around $21m of additional net present value (NPV) for the project.

The improvement is due to the increased bake residence time and, most importantly, the steady state continuous operation, with the kiln running uninterrupted for five consecutive days. Approximately 120kg of rare earth carbonate was produced in the third run for customer validation. The material, which typically contained a mixed rare earth carbonate grading 48%, again featured a high proportion of dysprosium.

Marketing landscape

According to Northern the key market driver for dysprosium is the growing demand for NdFeB permanent magnets. The permanent magnet sector’s forecasted growth from 2014 to 2020 is expected by Northern to be 8-12% per annum, which could increase as secure sources of dysprosium supply, such as Browns Range, come online. While current dysprosium prices are lower than in recent years, global producers are forecasting prices to steadily increase over the next six years due to this demand pressure, with Chinese production continuing to consolidate and global demand increasing for environmentally responsible HRE sources.

Environmental approvals

WA’s Environmental Protection Authority (EPA) has advised that it considers that the project can be managed to meet the EPA’s environmental objectives subject to the EPA’s recommended conditions being adopted. The company has also commenced preliminary planning and drafting work on the secondary approvals required for the proposed mining operation. These approvals will be considered by the relevant decision making authority following the Minister’s determination, and will include:

- Mining proposal and project management plan from the WA Department of Mines and Petroleum.

- Works approval and licences from the WA Department of Environment Regulation.

- Licences to construct bores and take water from the WA Department of Water.

In addition, the Federal Government Department of Environment has assessed Browns Range as a “not controlled action” which means the project does not require assessment and approval under the EPBC Act 1999 before it can proceed.

Feasibility study progress

Following the release of a positive PFS in June, the company has the project’s FS underway which it is aiming for completion at the end of 2014. The process design criteria for the FS have been completed and Northern has appointed a number of key consultants and consultant engineers. DRA Global has been appointed to undertake the design of the beneficiation plant, process infrastructure and also manage the overall co-ordination of the FS. Engineering and project management firm Lycopodium has been engaged to undertake the design of the hydrometallurgical plant. Global management consultant Accenture will be responsible for project controls and procurement for the FS.

1 In this report dysprosium is to be read as dysprosium oxide (Dy2O3) unless otherwise stated.


The management of Northern Minerals Ltd is sincerely thanked for permission to publish this report.