Feasibility Study for Lac des Iles Mine Incorporating Underground Mining of the Roby Zone

The 2018 feasibility study for the Lac des Iles mine describes the existing processing facility’s design capacity, flowsheet, and metallurgical performance as the basis for incorporating underground ore from the Roby Zone.

Report context

This feasibility study, dated October 2, 2018, evaluates the incorporation of underground mining of the Roby Zone into the existing Lac des Iles Mine (LDIM) operation. The processing section of the report documents the existing mill flowsheet, capacity, and operating parameters that would be used to process the blended ore feed. The report presents historical operating data from the mill, testwork results used to establish process design targets, and proposed configurations for continuous operation.

Processing route

Crushing and grinding circuit

The run-of-mine ore feed is crushed through a primary 54” x 75” gyratory crusher driven by a 447kW (600 hp) motor. The crusher discharge is conveyed to a live stockpile. Apron pan feeders feed the conveyor between the stockpile and the main grinding circuit, with the capability of diverting a variable proportion of coarse ore to secondary crushing. The secondary crushing circuit uses a standard HP800 cone crusher that discharges back onto the mill feed conveyor.

The grinding circuit consists of a SAG mill, two ball mills, and a pebble crusher. The SAG mill has dimensions of 9.14m diameter by 4.27m equivalent grinding length (EGL) and is driven by a 6,400kW (8,500 hp) motor. The SAG mill can be operated in autogenous or semi-autogenous configurations. Mill output passes over a vibrating screen where oversized particles are directed to a short head cone HP800 pebble crusher, with the product returned to the grinding mill. The pebble circuit is configured to allow oversized particles to bypass the pebble crusher and return directly to the SAG mill feed.

Undersized particles from the SAG mill screen are directed to a ball mill diverter box that splits the stream between two ball mills. Each ball mill has dimensions of 6.10m diameter by 10.36m EGL and is fitted with a 6,400kW (8,500 hp) motor. Each ball mill operates in closed circuit with a hydrocyclone cluster. Underflows from the hydrocyclones report back to the ball mills and overflows report to a collection tank. The primary hydrocyclones were upgraded to GMAX 20” and the ball mill media was reduced to 1” high chrome balls to produce the current P80 of 55μm.

Flotation circuit

The rougher conditioning tank receives feed from the primary hydrocyclone overflow. The conditioner feeds two flotation lines, each with a rougher flotation cell of 50m³ capacity. Concentrate from the rougher flotation cells reports to the 1.7m rougher column cleaner cell (column cell C). Tailings from rougher flotation move forward to a series of scavenger flotation cells of 130m³ each. Tailings from scavenger flotation are final tailings pumped to the tailings management facility (TMF). Scavenger concentrates from cells 1A, 2A, 1B, 2B, 1C, and 2C are combined and pumped to the cleaner A feed bank. Scavenger concentrates from cells 1D, 2D, 1E, 2E, 1F, 2F, 1G, and 2G are combined and pumped to the cleaner B feed bank.

The cleaner circuit has two main divisions: rougher cleaning (column cells A, B, C) and scavenger cleaning (four conventional banks of cells called Cleaners A, B, C and Denvers). Rougher concentrate reporting to the 1.7m column C is cleaned once to produce final concentrate. Tailings from this column are split and fed into two 1.3m columns (columns A and B) operating in parallel. Concentrate from these columns goes to final concentrate while tailings enter the scavenger cleaner circuit at the Denvers cleaner bank feed.

The scavenger cleaner circuit comprises four banks of conventional cleaners. Cleaners A, B, and C each employ 3 x 38m³ cells, while the Denvers consist of 8 x 2.8m³ cells. Cells operate in counter current mode with concentrate from each bank pumped to the next higher cleaner and tailings flowing by gravity through successively lower cleaners (Cleaner C to Cleaner B to Cleaner A to Denvers). Entry points are at Denvers (from column A and B tails), at Cleaner A (from scavengers A, B, C on each line), and at Cleaner B (from scavengers D, E, F, G on each line). Outputs of the scavenger cleaner circuit are final concentrate from the Denvers and Cleaner C tailings circulated back to the main circuit and split between scavenger cells 1A and 1B.

Concentrate handling

Final concentrate streams are combined and pumped to the 17m high-rate concentrate thickener. Thickener underflow is pumped into a 4.3m by 7m storage tank prior to filtration. Two pressure filters (PF-19) are operated to reduce concentrate moisture to 10%. Concentrate is shipped via trucks to the smelter.

Water management and tailings

Final plant tailings containing rougher scavenger tails and cleaner 1 scavenger tails are pumped to the tailings management facility (TMF) for permanent impoundment at approximately 30% solids content. Tailings are spigoted into containment cells within the TMF, where solids settle and water reports to a central water management facility (WMF1) for recycling to the concentrator. Surplus water from precipitation is clarified at the concentrator to meet discharge limits and released to the environment.

A 33.5m tailings thickener designed by Westpro was under construction at the time of the report. Once completed, the thickener is designed to produce an underflow containing solids above 50% and an overflow containing less than 400 NTU solids for recirculation directly to the concentrator.

Plant capacity basis

The plant capacity of 4,470,000 tonnes per annum is based on grinding rates demonstrated to achieve the target flotation feed grind size. The plant targets a flotation feed grind size of P80 = 55μm, based on testwork by XPS (May 2011) that showed improved Pd recovery possible by reducing P80 to as low as 38μm. The selected option increased grinding efficiency and decreased grinding rates to reach the selected target. Improvement was achieved by reducing the transfer size at the SAG mill discharge screen and replacing cyclones with a more efficient model. The ball mill charge was changed to 2.5cm high-chrome grinding media and the ball load was increased to the motor limit. Throughput was reduced to 555 tph to achieve the new grind size. Since 2014, the plant has produced a significantly finer grind of P80 of 55μm, resulting in increased Pd recovery.

The plant throughput at 55μm flotation feed size is governed primarily by the capacity of the SAG mill, ball mill, and pebble crusher circuit. At a throughput rate of 554.5 tph and 92% utilization, the plant capacity is 4,470,000 tpa. The 92% utilization is projected based on 365 days available, less 18 days of planned maintenance, less 2.5 days of peak power management, and 97.5% reliability.

Consumables

The SAG mill uses 12.5cm forged steel balls and the two ball mills use 2.5cm high-chrome balls. A trial of 4cm balls was conducted in the ball mills during 2017. The flotation process uses potassium amyl xanthate (PAX) as the primary collector, sometimes augmented with dithiophosphate (Aero 3477) collector, methyl isobutyl carbinol (MIBC) as frother, and CMC as a talc depressant. Concentrate is thickened with anionic polyacrylamide and filtered with addition of CuSO₄. Surplus site water is treated through the mill clarifier using Floquat as a cationic coagulant and Magnafloc E-10 as a high molecular weight anionic flocculant.

Recovery models

Palladium, platinum, gold, and copper are the pay metals present in the concentrate. Recovery models were developed from plant data. The relationship between Pd in feed and Pd in plant tails is: Pd Grade in Tails = 0.1293 x Pd Grade of Feed + 0.137. The relationship for concentrate is: Pd Grade in Concentrate = 43.703 x Pd Grade of Feed + 155.38. These models were evaluated over the observed range of Pd head grades between 2 gpt to 5 gpt by comparing predicted tailings grade against actual reconciled balances from January 2014 through December 2016. The 2017 data was added and models were confirmed to remain valid.

Recovery equations for other metals developed from regressions of plant data are: Pt Recovery to Concentrate = Pt Grade in feed x 83.433 + 57.177; Au Recovery to Concentrate = Au Grade in feed x 71.449 + 63.44; Cu Recovery to Concentrate = Cu Grade in feed x 283.43 + 68.328. Validation against monthly reconciled metallurgy from January 2014 through December 2017 showed average differences and standard deviations. The Pt model generated the highest difference and slightly over-predicted recovery, but an adjustment was not warranted as Pt occurs in the ore at only 1/12th that of Pd.

Regular grade ore stockpile testing

A regular grade ore (RGO) stockpile at site has a stated head grade of 0.97 gpt Pd, 0.12 gpt Pt, 0.08 gpt Au, 0.030% Cu and 0.060% Ni. Calculated recoveries from this ore using the equations are 73% Pd, 67.2% Pt, 69.2% Au, 76.8% Cu and 30.5% Ni into a concentrate grading 198 gpt. Since the dataset did not span head grades as low as 1 gpt, a limited one-day plant trial was performed on April 5, 2016. During the trial, observed head grade was 1.00 gpt Pd, 0.21 gpt Pt, 0.108 Au, 0.029% Cu and 0.048% Ni. The plant achieved a concentrate grade of 170 gpt Pd with recoveries of 74% Pd, 86% Pt, 75% Au, 78% Cu and 23% Ni. The higher Pt and Au recovery observed in the concentrate were discounted due to low feed grades and trial duration.

Key reported parameters

Technical qualifications

The recovery models were developed from plant data over the period January 2014 through December 2017 and are stated to be valid for Pd feed grades between 2 gpt and 5 gpt. The models for Pt, Au, and Cu were developed from regressions of plant data over the 48-month period from January 2014 through December 2017. The regular grade ore stockpile testing was a limited one-day plant trial performed on April 5, 2016, and the report notes that higher Pt and Au recovery observed during the trial were discounted due to low feed grades and short test duration. The report states that when RGO is combined with underground ore and the combined head grade is maintained above 2.0 gpt Pd, the general equations will apply. The projected plant capacity of 4,470,000 tpa and 92% availability are forward-looking estimates based on historical data combined with industry norms, not sustained actual performance at the time of the report, as the plant had been operating on reduced schedules (14 days on, 14 days off) through 2016 and 2017 and began ramping to continuous 24/7 operation late