Maud Creek Gold Project — 2016 Technical Report

Figure 4-7: Simplified process diagram – NT environmental assessments

The report presents a preliminary economic assessment of processing Maud Creek sulphide, oxide, and transitional gold mineralization through a modified flotation circuit integrated with the existing Union Reefs Processing Plant.

Report context

The Maud Creek Gold Project 2016 Technical Report is dated 13 May 2016 and addresses the processing of gold mineralization from the Maud Creek deposit. The study and process design focus on the Maud Creek sulphide resource, which is refractory to direct cyanidation but responds well to flotation. The deposit also includes approximately 300 kt of oxide material and 230 kt of transitional material. The oxide mineralization is generally amenable to direct cyanidation and will be treated through the existing Union Reefs Processing Plant. The transitional mineralization is less amenable to conventional cyanide leaching and may be treatable by flotation, potentially with a controlled potential sulphidization (CPS) pretreatment process. The process design philosophy is to add a flotation circuit to the existing facilities at the Union Reefs Processing Plant, producing a gold sulphide concentrate for sale or possible in-house processing.

Processing route

Background and metallurgical testwork history

The main body of metallurgical testing and development work was conducted from 1994 to 1998 for Kalmet Resources NL and Kilkenny Gold NL. Minor work was done for Harmony Gold Operations Ltd in 2003, and further testing and piloting were undertaken for Terra Gold Mining Ltd in 2006. John W MacIntyre and Associates Pty Ltd supervised the earlier testing starting in late 1996, culminating in a detailed report “Metallurgical Evaluation of the Maud Creek Project” in September 1998. This review draws heavily on the MacIntyre report and the original metallurgical testwork reports, and also considers later testwork where appropriate. The present mine plan is considerably different to that envisaged in the late 1990s so the original design recommendations must be treated with caution.

Process flowsheet selection

Several earlier studies were used in conjunction with the testwork reports for process flowsheet selection. These included the 1998 Signet Engineers “Process design Criteria 3081-G-00-F-001 Rev A” (Draft), the 1997 “Review of the Metallurgy, Capital Cost and Operating Cost for the Maud Creek Project” by Signet Engineering, and the 1998 “Metallurgical Evaluation of the Maud Creek Project” by John MacIntyre and Associates Pty Ltd. Initially a simple single stage crushing and closed circuit SAG milling circuit, followed by centrifugal style gravity concentration, a simple rougher/scavenger/scavenger cleaning flotation circuit, and concentrate dewatering was chosen as the base case flowsheet. Once the decision was made to utilise the spare capacity at Union Reefs Processing Plant, the front end crushing and grinding circuit design reverted to the existing three stage crushing and closed milling circuit configuration at Union Reefs.

The Union Reefs Process Plant was originally designed to process free milling but predominantly fresh rock ores with a BBMWi of 16-18 kWh/t at a grind size of 106 microns. Annual processing rates of 2.8 Mtpa were achieved between 1999 and 2003. These ores typically had an average gold head grade of 1.5 g/t recovering 92-94%, of which 30% was recovered by gravity. The plant currently treats approximately 750-850 ktpa and has excess crushing capacity and two underutilised grinding mills.

The selected grind size is 80% passing 75 microns, chosen following 1996 optimisation testing. Earlier pilot programs used 125 microns (1996) and 100 microns (1997). The evaluation conducted by J MacIntyre in 1998 determined 100-110 microns as the optimum grind size, but a P80 of 75 micron was judged more appropriate for design purposes. Several important design parameters have changed since that evaluation: circuit tonnage was 300 ktpa now 500 ktpa; ore grade was 7.4 g/t now approximately 4.38 g/t; downstream processing was bio-oxidation now direct concentrate sale; gold price was AUD482/oz now greater than AUD1,500; power cost was AUD0.125/kWh now AUD0.22/kWh.

The selected flotation circuit includes rougher, scavenger, and scavenger cleaner stages, but no flash flotation. Some testing showed marginal benefit from flash flotation. A cleaner flotation stage will increase concentrate grade but possibly marginally reduce recovery. The flotation reagent scheme consists of sodium isobutyl xanthate (SIBX, collector), copper sulphate (CuSO4, activator), and frother, with no pH adjustment.

Existing processing facilities utilised

Crushing and screening circuit: Run of mine material is reclaimed from stockpiles by a wheel loader and tipped into a ROM bin. A vibrating feeder transports material to a C140 Nordberg Single Toggle Jaw Crusher. Jaw crusher product is transported to a double deck banana 3.1 m x 7 m Nordberg Product Screen with a 40 mm aperture top deck and a 14 mm aperture bottom deck. Screen undersize (minus 14 mm) is conveyed to a Fine Ore Bin/Stockpile of 3000 tonnes live capacity. Screen oversize (plus 40 mm) is transported to a Nordberg Omni-Cone 1560 secondary cone crusher, which returns product to the Product Screen. Intermediate product of minus 40 mm plus 14 mm is transported to a Nordberg HP500SX tertiary cone crusher, which also returns product to the Product Screen. No changes are envisaged to the existing crushing circuit.

Milling circuit: Fine crushed ore is reclaimed from the fine ore bin by a slot type belt feeder. The mill feed conveyor discharges to a Mill Feed hopper with a split discharge onto two variable speed ball mill feeder conveyors, one feeding the 3 MW, 4.7 metre by 8.2 metre ANI No.1 Ball Mill, the other feeding the 4 MW, 5 m x 9.1 m ANI No.2 Ball Mill. Each mill is in closed circuit with separate hydrocyclone clusters. No changes to the existing milling circuit are envisaged.

Gravity recovery circuit: Each ball mill hydrocyclone cluster is fitted with a direct off-take on the hydrocyclone feed pot, directing a bleed stream of ball mill discharge to a Nordberg 1.2 metre by 1.5 metre scalping screen to remove coarse scats. Fine screen product reports to one of two automatic discharge 30 inch Knelson concentrators. Rough Knelson concentrate is automatically discharged to a secured hopper located within the gold room. On a batch basis, rough Knelson concentrates are transferred to an Acacia intensive leach reactor (ILR). No changes to the existing gravity circuit are envisaged.

Proposed new flotation and concentrate handling circuits

Flotation circuit: Cyclone overflow is first screened to remove trash, then flows to the conditioning tank where copper sulphate, frother, and SIBX (xanthate collector) are added. Conditioned concentrate overflows to the rougher flotation cells. Rougher concentrate reports directly to the final concentrate thickener. Rougher tails flow to the scavenger cells. Scavenger cells flotation concentrate is pumped back to the cleaner flotation cells, while scavenger tails are pumped to the tailings thickener. Cleaner concentrate reports to the final concentrate thickener. Cleaner tailings are usually returned to the rougher cells, but may also be returned to the scavenger cells. The next stage of design will also enable tailings to be pumped to the CIL circuit to allow for low recovery Maud Creek transitional mineralization.

Concentrate thickening and filtration circuit: Flotation concentrate is first settled in a high rate thickener. The thickener underflow is further dewatered in a filter press. It is dumped to ground and contained in a storage shed where it can be bagged, then loaded into sea containers and onto trucks. Filtrate and thickener overflow are returned to the process water tank. A separate process water system has been allowed for at this stage of study to eliminate concerns with reagent contamination between the CIL and the flotation circuits.

Tailings circuit: Flotation tailings are thickened and the underflow pumped to the tailings storage facility. The overflow is returned to the process water tank. The next stage of design will also enable tailings to be pumped to the CIL circuit to allow for low recovery Maud Creek transitional mineralization.

Flotation concentrate characteristics

Flotation concentrate composition from pilot plant runs averaged 46.5 g/t Au, 19.6% S, 3.6% As, and 21.3% Fe. Additional assays from the 1996 campaign indicated Ag at 20 g/t, Cu at 1600 ppm, total C at 2.79%, organic C at 0.04%, and SG of 3.2. XRD analysis on a combined concentrate showed approximately 40% pyrite, 25% arsenopyrite, 3% chalcopyrite, 3% marcasite, 10% quartz, 10% muscovite/sericite, 2% chlorite, and 6% carbonate (dolomite). A concentrate grade of 45 g/t Au has been selected for design purposes and transportation costs, considered to be conservative. It is expected cleaning of the concentrate, flash flotation, and/or a finer grind will improve the concentrate grade at similar recovery.

Downstream processing options considered

The report reviewed multiple downstream processing options for the gold concentrate: ultrafine grinding (UFG) which was found to be relatively ineffective with median gold extraction rates of 61% increasing only to 65% at 100% minus 20 microns; bio-oxidation which achieved extraction rates up to 95% in Signet tests but with challenges regarding high carbonate and arsenic levels; pressure oxidation (POX) with extraction of 98% but no available POX autoclaves for gold concentrate in Australia; ultrafine grinding followed by moderate pressure leaching (Albion/Activox/CESL) which are capital cost inhibitive due to short life of mine and low tonages; GEOCOAT process with gold extraction of 81% and poor performance on other grounds; and roasting which is not an option due to roaster closures and transport costs. Preliminary techno-economic assessment shows the terms offered by Chinese processors for direct smelting of the concentrate, including allowance for arsenic content, to be the most favourable option. The likely market for the gold concentrate is China, selected as the base case.

Key reported parameters

Variable Unit Value Basis
LoM tonnes Mt 3.8 Design assumption
Mineralization type % 96% Fresh, 4% Oxide/Transitional Resource estimate
Mill Tonnes tpa 500,000 Design capacity
Plant utilisation % 90 Design assumption
Milling throughput tpa 70 Design basis
Gold Feed Grade g/t 4.38 Resource estimate
Gravity Recovery (fresh) % 20 Testwork based
Flotation Recovery (fresh) % 75 Testwork based
Total Recovery (fresh) % 95 Testwork based
Total Recovery (oxide/transitional) % 85 Testwork based
Recovered Gold oz/a 66,890 Calculated from above
Concentrate Mass Recovery % 7.3 Testwork based
Concentrate Production tpa 36,500 Calculated from above
Concentrate Gold Grade g/t 45 Design basis (conservative)
Concentrate moisture % 10 Design assumption
Tailings tpa 463,514 Calculated from above
Water Consumption (plant) GL/a 0.35 Design estimate
Reagent Frother g/t 20 Testwork based
Reagent Collector (SIBX) g/t 125 Testwork based
Reagent Activator (CuSO4) g/t 50 Testwork based
Reagent Flocculant g/t 70 Testwork based
SAG Mill Grinding Media g/t 2810 Testwork based

Project website: https://geoscience.nt.gov.au/gemis/ntgsjspui/handle/1/93306

Project website: https://miningdataonline.com/property/3053/Maud-Creek-Project.aspx

Technical qualifications

The process design presented is based on available information considered by SRK to be appropriate for this preliminary economic assessment. Some minor gaps remain and further development and optimisation work is recommended. The present mine plan is considerably different to that envisaged in the late 1990s so the original design recommendations must be treated with caution. A detailed integration study has not been undertaken into the Union Reefs processing option and will be undertaken at the next phase of the study. Certain assumptions have been made based on incomplete knowledge of the metallurgy and market conditions. Metallurgical samples were biased to shallower samples. There is a lack of information on flotation testwork performed on the transitional mineralization. Development of feed and concentrate grade versus recovery relationships needs to be developed, potentially requiring testing on some lower grade samples (testing has been biased to higher grades). Additional review of the oxide mineralization behaviours is required although it makes up a relatively small proportion of the overall feed. The target accuracies for this stage of study work is plus/minus 30% for capital cost and plus/minus 20% for operating cost. There is potential for costs to rise or fall in future in line with future moves in exchange rates, equipment, material and labour prices. The report notes that the Union Reefs plant is currently operated on a 9 days on 5 days off roster, with the crushing circuit utilisation lower again and typically only one of the two mills operating.

Source: Maud Creek Gold Project , 2016 Technical Report, Sections 17 Recovery Methods, 17.1 Background, 17.2 Processing Plant Basis of Design, 17.2.1 Process Flowsheet Selection, 17.2.2 Plant Throughput Selection, 17.3 Engineering Deliverables, 17.3.1 Processing Description, 17.3.2 Crushing and Screening Circuit, 17.3.3 Milling Circuit, 17.3.4 Gravity Recovery Circuit, 17.3.5 Flotation Circuit, 17.3.6 Concentrate Thickening Filtration Circuit, 17.3.7 Tailings Circuit, 17.4 Reagents and Services, 17.4.1 Reagents, 17.5 Flotation Concentrate, 17.5.1 Flotation Concentrate Grade, 17.6 Process Processing Risks and Opportunities, 17.7 Toll Treatment Options for Product Concentrate, 17.8 Stand-alone Processing Plant Option

Mineral processing basics

Scroll to Top