Biomining in Finland: Commercial Application of Heap and Tank Bioleaching Technologies for Nickel Recovery

Abstract

The use of tank and heap bioleaching technologies for the extraction of nickel and cobalt from sulfide ores and concentrates has found commercial application in Finland. At the Terrafame mine (previously Talvivaara), the complexity of the black schist orebody precluded the production of a suitable concentrate and a decision was made to bioleach the crushed ore in heaps. To date, it is the only industrial-scale mine to utilise heap bioleaching for nickel production. In the case of Mondo Minerals, a sulfide concentrate, generated as a by-product from talc mining operations, contains significant quantities of nickel and cobalt, but also a small but significant quantity of arsenic. Previously the concentrates had been sold to toll smelters, but the arsenic content has made this option less attractive. Tank bioleaching technology was identified as the most suitable option for the recovery of nickel and cobalt from this sidestream and led to the first commercial implementation of nickel sulfide concentrate bioleaching. This chapter will briefly cover developments at the Terrafame mine since the previous edition of this book, while the main emphasis will be on the technical development of the Mondo Mineral tank bioleaching process from bench- to commercial-scale application.

12.1 Introduction

Bioleaching is a well-established process used in the extraction of base metals and refractory gold from sulfide ores and concentrates. The process has been commercially applied in a wide variety of locations for the pre-oxidation of refractory gold concentrates and for the recovery of copper from secondary sulfide ores (Gericke et al. 2009). Current estimates are that biomining accounts for around 10–20% of global copper production, ~1% of gold production, and smaller percentages of other metals such as zinc, nickel, and uranium (Chap. 1).

Nevertheless, the application of bioleaching has remained a niche, rather than a mainstream, technology in the mining sector. Some unique benefit is typically required for bioleaching to be selected as the process option of choice. Examples, where bioleaching processes do show significant advantages over conventional mineral extraction processes, include: (i) treatment of refractory gold ores, (ii) recovery of metals from low-grade ores and tailings, (iii) treatment of sulfide concentrates containing significant quantities of impurities that would incur smelter penalties, and (iv) under conditions where production of a suitable flotation concentrate is problematic (Gericke et al. 2009; Johnson 2018).

In this chapter, two Finnish case studies are described where bioleaching has been selected as the technology of choice for the treatment of nickel-sulfide-containing ores and concentrates. At the Terrafame mine (previously Talvivaara), the complexity of the polymetallic black schist ore body precluded the production of a suitable concentrate and the decision was made to bioleach the crushed ore in heaps. Nickel production commenced in 2008 and it is still the only industrial-scale mine to utilise heap bioleaching for nickel production (Riekkola-Vanhanen 2007). The bench-scale development of the heap leach process applied at the Terrafame mine was covered in a previous edition (Puhakka et al. 2007). The developments at Terrafame since then are described in this chapter.

In the case of Mondo Minerals, a sulfide concentrate, produced as a by-product from their talc mining operations, contains valuable quantities of nickel and cobalt, but also a small but significant amount of arsenic. While sales to smelters had been the long-established method of commercially dealing with the concentrate, Mondo chose to create a value-added nickel product to enhance its revenue and profitability streams and avoid environmental liabilities. Tank bioleaching technology was identified as the most suitable option for the recovery of nickel and cobalt from this side stream. The commercial plant was commissioned in 2015 and it is the first commercial implementation of nickel sulfide concentrate bioleaching (Neale et al. 2015). The technical development of the process from bench-scale test work to piloting and commercial application are highlighted in this chapter.

12.2 The Talvivaara/Terrafame Project

The Talvivaara deposit, located in Sotkamo, Finland, is the largest known nickel sulfide deposit in Europe (Riekkola-Vanhanen 2007). The deposit comprises two polymetallic orebodies hosted by a black schist, Kuusilampi and Kolmisoppi, and the size of the resource is estimated to be 1550 million tonnes (Heikkinen and Korte 2019). The complex ore has average grades of 0.27% nickel, 0.56% zinc, 0.14% copper, 0.02% cobalt, 10.3% iron, 8.4% sulfur, and 7.2% carbon. The main minerals present are pyrrhotite, pyrite, pentlandite, sphalerite, violarite, chalcopyrite, and graphite (Riekkola-Vanhanen 2007).

Owing to the complex and low-grade nature of the Talvivaara ores, exploitation of the deposits was not economically viable using conventional processes. The application of bioleaching for the treatment of the Talvivaara ores was studied extensively in laboratory-scale and column leaching test work over a period of two decades, leading to the construction of a 17,000-tonne demonstration heap in 2005. Over a period of 500 days, 92% nickel, 82% zinc, 14% cobalt, and 2.5% copper were recovered. Copper is present as chalcopyrite, which explains the low copper recoveries achieved (Riekkola-Vanhanen 2007).

In February 2007, the primary demonstration heap was reclaimed and restacked to a secondary heap with the aim of enhancing copper and cobalt recoveries. Transferring the material to a secondary leaching stage has the added advantage of enhancing the recovery of metals from those parts of the primary heaps where the leaching solution has poor contact with the ore particles. Such areas include the slopes of the heaps, and areas within the bulk of the heap where solution flow was not ideal (Riekkola-Vanhanen 2011). The acid consumption was 15 kg t⁻¹ in the primary leaching stage and 2 kg t⁻¹ in the secondary leaching stage (Riekkola-Vanhanen 2011). The final recoveries after an additional 21 months of secondary leaching were 99% nickel, 99% zinc, 35% cobalt, and 22% copper (Riekkola-Vanhanen 2011).

12.2.1 Commercial Implementation

Based on the successful operation of the demonstration heap, commercial implementation of the bioheap leaching process was initiated in July 2008 and the first metal sulfides were produced in October 2008. The mining method is large-scale open-pit mining. The heap leaching process is divided into two steps, a dynamic primary leaching and multi-lift secondary leaching. The process flowsheet and the challenges experienced during commercial implementation of the heap leach operation have been described in detail in literature (Riekkola-Vanhanen 2007, 2011, 2013; Saari and Riekkola-Vanhanen 2012; Ahoranta et al. 2018).

The ore is crushed and screened to P₈₀ = 8 mm (80% of particles <8 mm) followed by agglomeration with pregnant leach solution (PLS). The ore is then stacked on the primary heaps which are irrigated with an acidic solution (pH around 2) and aerated. Most of the acid required is generated in the heaps. The residence time on the primary leaching pads is approximately 13–14 months, after which the ore is reclaimed and restacked to permanent secondary heaps to continue the bioleaching process. After secondary leaching for a further three and a half years, the barren ore remains permanently in the secondary heaps. The secondary leaching pads are constructed on top of waste rock dumps which reduces earthwork quantities, the final footprint of the operation, and the rehabilitation costs. The secondary pads are planned to be stacked with four 15 m lifts, and the 60 m high heap to be eventually covered and revegetated.

In the metals recovery process, the metals are precipitated in stages from the PLS using gaseous hydrogen sulfide. The resulting intermediate products are transported to various refineries for further processing (Riekkola-Vanhanen 2007, 2011, 2013). After the target metals have been recovered, the solution is further purified to remove unwanted metals and returned to irrigate the heaps. During removal of residual metals, the pH of the PLS is raised to 9–10 with lime slurry, leading to the precipitation of residual metals (mostly manganese, magnesium, and iron) as hydroxides, together with gypsum and calcium carbonate. The resulting slurry is thickened and the thickener underflow is directed to gypsum waste ponds (Ahoranta et al. 2018; Tuovinen et al. 2018).

Challenges with crushing and the aeration systems at the beginning of industrial-scale leaching delayed the increase in metal recovery, but after the first two operational years the leaching results have improved significantly. The planned nickel production of 50,000 t y⁻¹ was anticipated to be reached in 2012, with additional production targets of 90,000 t y⁻¹ of zinc, 15,000 t y⁻¹ of copper, and 1800 t y⁻¹ of cobalt. Due to several challenges that were experienced, considerably lower metal production was achieved. By 2011, the production volumes achieved were 16,087 t y⁻¹ of nickel and 31,815 t y⁻¹ of zinc (Riekkola-Vanhanen 2013).

12.2.2 Talvivaara Becomes Terrafame

By 2014, the Talvivaara operation was facing serious operational and environmental challenges. The production delays combined with several leaks of metal-contaminated tailings which threatened local waterways, drove the company to bankruptcy. The asset was acquired by Terrafame in 2015, after which it went into a new commissioning phase (Arpalahti 2017; Ahoranta et al. 2018).

The fortunes of the operation have improved substantially in the years after the acquisition by Terrafame. Several practical factors have been addressed to improve the operability of the plant. One of the major complications was that the material hardens significantly during primary leaching, so much so that it fuses together and loses its granularity entirely, making re-mining of the ore after primary leaching difficult. To solve this issue, Terrafame developed mobile surface mining as a more feasible solution for the second round of reclamation. Several other practical aspects of the operation were also optimised and improved. These included repositioning the aeration and drainage pipes, and optimising the movement and stacking of the ore after re-mining of the primary leach pad, to minimise downtime (Arpalahti 2017).

In the period between September 2015 and early 2017, the third complete dynamic cycle of re-mining and re-stacking of the leach pads was completed (Arpalahti 2017). As a result, in 2017 20,864 t of nickel and 47,205 t of zinc were produced. Annual increases in production were recorded and nickel production of 27,468 t and zinc production of 55,222 t were achieved during 2019 (Neale 2020).

12.2.3 Future Developments

There are currently two strategic developments underway at the Terrafame operation, namely uranium production and the establishment of a battery chemicals plant (Ahoranta et al. 2018). The Finnish government granted a uranium extraction permit to Terrafame in February 2020, but the decision has been appealed, and it is expected that it will take a further 2 years before a final decision is made (Neale 2020).

Terrafame also announced a plan to invest in a battery chemicals plant to produce nickel and cobalt chemicals to be used in the electric vehicle (EV) industry. The decision was based on forecasts of growing demand for EV batteries, coupled with an indication that the share of nickel in battery applications is also increasing. The plant is intended to have an annual production capacity of around 150,000 tonnes of nickel sulfate and 5000 t of cobalt sulfate, making Terrafame one of the largest nickel sulfate producers in the world (Ahoranta et al. 2018). The plant is currently under construction, and commercial production is planned to commence in 2021 (Neale 2020).

12.3 The Mondo Minerals Tank Bioleaching Project

Finland hosts some of the world’s largest talc deposits and is a significant global producer of this mineral (Mg₃Si₄O₁₀(OH)₂₎. Mondo Minerals (acquired by Elementis plc in 2018) is the world’s second-largest talc producer with mining operations at two sites in Finland, Sotkamo, and Vuonos. At both sites, a nickel-rich sulfide concentrate, containing pyrrhotite (Fe_(1-x)S (x = 0 to 0.2)), pentlandite ((Fe,Ni)₉S₈), pyrite (FeS₂), gersdorffite (NiAsS), magnesite (MgCO₃), and talc, is produced as a by-product of the flotation process. Nickel is the main metal of interest, but the concentrates also contain a small amount of cobalt and a small but significant quantity of arsenic (Table 12.1).

Table 12.1 Composition of the Sotkamo and Vuonos concentrates

Previously, the concentrates had been sold to toll smelters but the arsenic contents have made this option less attractive. Mondo, therefore, elected to produce a value-added nickel product to enhance its revenue and profitability streams and avoid environmental liabilities. Mondo tested and evaluated several processing options before selecting Mintek’s proprietary bioleaching technology as the most suitable for the recovery of nickel and cobalt from these side streams (Neale et al. 2016).

It is recognised that bioleaching is particularly suited to the treatment of concentrates containing problematic elements such as arsenic. In this project, those circumstances are present—the concentrate contains arsenic, which makes it increasingly more expensive to treat the material via smelting, and stockpiling of the material is undesirable as it would create an unacceptable environmental liability.

A 2-year-long test work programme was undertaken which developed and successfully demonstrated the application of Mintek’s technology to treat the by-products from Mondo’s talc production process. The study formed the basis for a feasibility study that showed that bioleaching, followed by a nickel- and cobalt-precipitation process, was an economically viable option for Mondo Minerals to derive value from the by-product. An important aspect of the process is that it includes the production of a stable arsenic-bearing waste, suitable for impoundment.

12.3.1 Metallurgical and Pilot-Scale Studies

The main objectives of the laboratory test work conducted by Mintek were to confirm the technical suitability of bioleaching to recover nickel and cobalt from the Sotkamo and Vuonos concentrates and to define the optimum bioleach operating parameters. The programme conducted incorporated regrinding of the concentrates, bioleaching, iron and arsenic precipitation, stability testing of the product, and nickel–cobalt (or mixed) hydroxide (MHP) precipitation. The bench-scale facilities used in the test work programme provided for close control and monitoring of the bioleaching process, allowing simulation of near-commercial scale conditions and control.

The bioleach test work was conducted on 1:1 blends of the Sotkamo and Vuonos concentrates. The typical chemical and bulk modal analyses of the concentrates used in the test work program are summarised in Table 12.1.

A moderately thermophilic culture from Mintek’s culture collection was used in the bioleach test work programme. The culture has been adapted over a number of years to tolerate high soluble nickel and iron concentrations and has an optimum growth temperature of 45 °C. The dominant organisms present in the microbial consortium include Acidithiobacillus caldus, Leptospirillum ferriphilum, Sulfobacillus benefaciens and Sulfobacillus thermosulfidooxidans. Three phases of laboratory test work were performed, as described below.

12.3.1.1 Phase 1: Laboratory Amenability Test Work

Bioleach amenability and optimisation tests were carried out on the blended concentrate (without pyrrhotite rejection) in multi-stage laboratory-scale, continuously operated reactors with a total working volume of 7 L.

The effect of process parameters such as residence time, feed solids concentration, and grind size of the concentrates on metal extractions was evaluated. The initial results of this phase of the test work programme showed that nickel and cobalt extractions of 95% could be obtained in a five-stage continuous plant, at an overall residence time of 7 days, a feed grind size of P₈₀ < 30 μm, and a feed solids concentration of 15%.

Optimisation of the feed solids concentration to 17.5% resulted in a marginal decrease in the nickel and cobalt extractions, to 93.5% and 94.0%, respectively. Further reduction of the feed grind size to P₈₀ = 20 μm improved the nickel and cobalt recoveries to 97% and 95%, respectively. Under these operating conditions, the pH level in the first stage reactor was not controlled and could be maintained at pH 1.5 without acid addition. Stable operation was demonstrated in the presence of around 45 g L⁻¹ total soluble metals and 120 g L⁻¹ sulfate in the first stage growth reactor. There was a linear relationship between sulfide oxidation and nickel dissolution obtained in the bioleach process, indicating that high nickel recoveries were dependent on achieving a high degree of sulfide oxidation (Gericke et al. 2014).

12.3.1.2 Phase 2: Mini-Pilot Plant Scale Test Work

The results of the Phase 1 tests were used to define the bioleach operating parameters for Phase 2, which was performed in a continuously operated bioleach mini-plant with a total operating volume of 120 L. In addition, the product from the mini-plant was collected and used to assess the operating parameters for iron and arsenic removal from the bioleach slurry with minimal nickel loss, to confirm the stability of the neutralised iron- and arsenic-bearing precipitates and to demonstrate the production of a high-grade mixed hydroxide precipitate with a combined nickel and cobalt content of more than 40%.

The results from this phase of the test work demonstrated that nickel and cobalt extractions of 94% could be obtained in a four-stage continuous plant, at an overall residence time of 7.2 days, a feed grind size of P₈₀ = 20 μm, and a feed solids concentration of 17.5%. The slightly lower extractions in this test were ascribed to the mini-plant system having four and not five stages, and a greater potential for some short-circuiting of solids in the gravity overflow system used in the mini-plant. (The laboratory-scale reactor system used in Phase 1 employed pumps to transfer the pulp between the reactors).

A continuous mini-plant test was undertaken for the iron/arsenic precipitation step, employing a recycle for seeding and limestone (CaCO₃) as the neutralising agent. Iron and arsenic removal of over 95% could be achieved in a six-stage neutralisation plant, operated at a temperature of 35 °C, with the pH level being controlled at between 3.0 and 3.5. No nickel and cobalt losses were observed. The precipitated product was tested for stability using the European Standard EN 12457–3 procedure (EN 12457–3, 2002) and since no nickel, iron, cobalt, or arsenic were released, the precipitate could be classified as regular waste (Gericke et al. 2014).

MHP production was evaluated in a batch test using magnesia (MgO) as the neutralising agent. A MHP containing around 42% nickel, 2.4% cobalt, and between 1.8 and 2.0% magnesium could be produced at a precipitation pH of between 7.0 and 7.8. A two-stage approach for MHP production was recommended, in which between 80 and 90% of the nickel and cobalt would be precipitated in the first stage, to minimise contamination of the MHP precipitate with unreacted MgO, and therefore to maximise the nickel grade. Following a solid/liquid separation step, the nickel and cobalt remaining in the barren liquor would be recovered by precipitation with lime (CaO), and the resultant solids recycled to the iron/arsenic precipitation process, where the target metals would be re-solubilised (Neale et al. 2015).

12.3.1.3 Phase 3: Test Work on the Upgraded Concentrate Blend

It was subsequently determined that rejection of the pyrrhotite contained in the concentrates would result in a significant reduction in the size and therefore the cost of the bioleach plant. Pyrrhotite rejection was undertaken by magnetic separation, with minimal loss of nickel. Additional upgrading of the concentrate by flotation, to reject talc and magnesite from the concentrate, was also undertaken. The rejected pyrrhotite had a low nickel content, and was a potential saleable product.

The suitability of bioleaching to recover nickel and cobalt from the upgraded concentrate after pyrrhotite rejection was demonstrated in laboratory-scale, continuously operated reactors, and a continuously operated bioleach mini-plant. Additional test work was also conducted on a synthetic solution to confirm the design of the iron/arsenic precipitation process, and the solid–liquid separation process that followed. This was deemed necessary because the upgraded concentrate had a significantly higher nickel and iron content, and so the concentrations of these metals in the bioleach product were also higher.

Nickel and cobalt extractions of 97% and 98%, respectively, were obtained at an overall residence time of 7 days, a feed grind size of 80% < 20 μm, and a feed solids concentration of 15%. Very high soluble metal concentrations were measured and the total metal concentration (of iron, nickel, cobalt, and arsenic) approached 65 g L⁻¹ in the first-stage growth reactor. Under these conditions, the redox potential in the first-stage bioleach reactor did not exceed +550 mV (vs Ag/AgCl), and so the maximum recommended feed solids concentration for the process was set at 15%. Under these operating conditions, the pH level in the first-stage bioleach reactor was controlled at 1.6, resulting in an acid consumption of 120 kg t⁻¹ (Gericke et al. 2014).

At very high iron (60 g L⁻¹) and nickel (30 g L⁻¹) tenors in the feed to the iron/arsenic precipitation unit, efficient agitation of the slurry was not possible for pH values greater than 1.5, since the slurry viscosity increased significantly. The introduction of a seed recycle was able to decrease the viscosity of the slurry, owing to the dilution of the iron tenor in the feed solution. Iron removal of greater than 99% could be achieved in a five-stage neutralisation plant, with minimal nickel and cobalt losses. By implementing a seed recycling, viscosity effects were eliminated, the slurry could be agitated efficiently, and the product could be thickened and filtered (Neale et al. 2015).

The overall conclusion reached at the end of the metallurgical test work programme was that a process consisting of concentrate regrinding, magnetic separation, flotation, bioleaching, iron/arsenic removal by lime precipitation, metal precipitation to produce a MHP, and tailings neutralisation, had been successfully demonstrated and these results provided the process design specifications for the commercial-scale plant design.

12.3.2 Process Design Criteria

Based on the outcomes of the metallurgical test work programme, a set of process design criteria, a process flowsheet, and a mass balance were developed; these formed the basis for a feasibility study that was commissioned by Mondo Minerals and executed by Tenova Mining & Minerals (Neale et al. 2015).

A simplified block diagram of the process flowsheet is shown in Fig. 12.1.

Fig. 12.1

General process flowsheet of the Mondo Minerals plant

The sulfide treatment plant is modest in size and is required to treat 35 t d⁻¹ of sulfide concentrate at a nickel production rate of 1000 t y⁻¹.

The concentrate preparation section includes the following stages:

• A regrinding circuit to grind the concentrate to a P₈₀ of 20 μm.

• Magnetic separation to remove some of the pyrrhotite from the concentrates.

• A flotation circuit to upgrade the nonmagnetic fraction further by removal of some of the remaining gangue materials, predominantly magnesite and talc, and some additional pyrrhotite and pyrite. The flotation circuit design includes:

  • 1 Flotation train

  • 4 Flotation cells per train

  • 3-stage rougher flotation configuration

  • Target of 99.9% pentlandite and 65.8% gersdorffite recovery

The magnetic separation and flotation processes achieved an upgrading of almost 50%, reducing the quantity of concentrate requiring bioleaching to 18 t d⁻¹ compared with 35 t d⁻¹ being fed to the mill (Neale et al. 2015).

A summary of the process design criteria that were developed for the bioleach, iron/arsenic precipitation, and metal recovery sections is presented in Table 12.2. This is by no means an exhaustive list of the design criteria, but merely highlights some of the key features of the design.

Table 12.2 Target bioleach and metal recovery design criteria (Neale et al. 2015)

12.3.3 Process Economics

The feasibility study included an estimation of the capital and operating costs of the plant, which were used to derive measures of profitability (Neale et al. 2015). Costs and prices prevailing in October 2013 were used.

The capital cost was developed as a Class 2 estimate, with approximately 10–15% accuracy. The estimated capital cost for the engineering development, procurement, and construction of the processing facility was in the range of €13–15 million, comprising 80% direct and 20% indirect costs. With the addition of a contingency and escalation, the capital cost estimated was in the range of €15–16 million.

The operating cost estimate included elements of varying accuracy, and was considered accurate to −12/+20%. The annual operating cost was estimated to be approximately €2.5–4.0 million, based on various operating parameters.

A detailed financial evaluation was undertaken, taking into account metal price forecasts, market potential, and market competition. A nickel price of US$20,000 t⁻¹ and a discount rate of 8% were used for the base case, and sensitivity analyses were based on that rate.

The project was evaluated by considering various inflation scenarios for the prices of energy, consumables and metals, and labour rates. For the base case—with no inflation—the net present value at a discount rate of 8%, or NPV(8), was approximately €36 million, and the internal rate of return (IRR) was 19.8%. For the various inflation scenarios that were investigated, the NPV(8) ranged between €32 million and €41 million, and the IRR varied between 19% and 22%.

Sensitivity analyses showed that the nickel price was the most significant determinant of project economics. For a lower case nickel price of ±US$14,000/t, the NPV(8) was around €15 million and the IRR was about 9%. For an upper case nickel price of ±US$30,000/t, the NPV(8) rose to approximately €77 million and the IRR to about 35%.

The project economics determined in the financial evaluation were deemed acceptable by Mondo, and the project was approved.

12.3.4 Process Description

The nickel sulfide treatment plant was constructed on the site of the existing Vuonos talc concentrator plant. The plant was fully integrated with the existing plant in terms of labour, services, and utilities. The location of the plant in eastern Finland meant that it would experience very low temperatures in winter. The minimum ambient external temperature specified in the design criteria was −15 °C, with the capacity to operate at short-term temperatures as low as −30 °C. For this reason, the existing concentrator and most of the new treatment plant are located indoors, in order to shield the plant and its operators from the harsh winter conditions. Only the bioleach tanks, which are heat generating, are located outside of the existing buildings (Neale et al. 2015).

12.3.4.1 Concentrate Preparation

The sulfide treatment plant was designed with a feed rate of 35 t d⁻¹, typically comprising a 50:50 blend of concentrates derived from Mondo’s two talc production plants at Sotkamo and Vuonos. The concentrate preparation circuit comprises stockpiling of concentrate, repulping, milling, magnetic separation, and flotation.

The stockpiled Sotkamo concentrate is transferred by a pipe conveyor into a repulp tank. The repulp tank also receives fresh concentrate as slurry directly from the Vuonos plant. The combined feed slurry is then pumped to the mill where the concentrate is ground to P₈₀ = 20 μm. The regrind mill is an ultrafine vertical grinding mill operating in an open circuit. The mill discharge is then pumped to the magnetic separator, where a concentrate that contains 3 to 5% nickel is produced. The concentrate is filtered on a disc filter and the nonmagnetic fraction is pumped to flotation for further upgrading.

Flotation is conducted in three rougher cells and one scavenger cell, with a conditioner ahead of the first stage. Flotation tailings are returned to the existing Vuonos flotation circuit tails disposal tank and pumped to the tailings facility. The flotation concentrate is thickened to reduce the level of flotation reagents in the bioleach feed slurry. The thickener underflow is pumped to the bioleach feed tank where the density is reduced from 65% to 50% solids using recycled water, and nutrients to support microbial growth are added (Neale et al. 2015).

12.3.4.2 Bioleaching

The slurry discharged from the bioleach feed tank is diluted in-line to 15% solids, from where it is distributed to the three primary bioleach reactors. The bioleach circuit originally consisted of seven 112 m³ tanks, with an overall residence time of 7 days at the design flow rate (Fig. 12.2). Three of the tanks were configured as primary oxidation reactors, followed by four tanks configured as secondary oxidation reactors. However, a fourth primary reactor was added at a later stage to provide additional capacity in the plant, if and when required. Each tank is fitted with an agitator for dispersing air, suspending solids and maintaining homogeneity, an air sparge ring for injecting air supplied by a blower, a separate pipe for adding gaseous carbon dioxide (CO₂), and a number of vertical cooling coils which perform a dual function as baffles (Fig. 12.3).

Fig. 12.2

The bioleach reactors at the Mondo Minerals plant

Fig. 12.3

The dual P4/P3 impeller system in the three primary bioleach reactors at the Mondo Minerals nickel sulfide plant

The concentrate is fed into the primary reactors operating in parallel. The secondary oxidation stage comprises four tanks in series, which provides sufficient residence time for the target level of nickel recovery to be achieved and sufficient reactor units to minimise short-circuiting of the slurry (Neale et al. 2016).

The main services for the bioleach plant include concentrated sulfuric acid (H₂SO₄), blower air, and CO₂, which is stored in a pressurised tank. Carbon dioxide is required in the bioleach process for microbial growth and is supplied by adding CO₂ gas directly to the primary reactors.

The pH level in the reactor bank is allowed to vary as the reaction proceeds and typically remains within a range of 1.2 to 1.6, with the higher pH values in the primary oxidation reactors, and declining as the level of sulfide oxidation increases in the secondary oxidation stages. Provision is made for pH control, if needed, by the addition of either H₂SO₄ or CaCO₃.

The oxidation reactions are exothermic and the slurry temperature inside the bioleach reactors tanks is maintained at around 46 °C by banks of cooling coils located in each tank. It has been determined that the operating temperature should not exceed 49 °C, as this will have a deleterious effect on the performance of the moderately thermophilic bioleaching consortium. The cooling coils also act as baffles in each tank, and are supported by brackets mounted on the tank walls.

The agitators in the bioleach reactors were supplied by Afromix. Each agitator in the four primary bioleach reactors is fitted with a dual impeller system, similar to that shown in Fig. 12.3, which is a relatively new innovation in bioleach reactor design.

The lower impeller, known as the P4, is a downward-pumping, four-blade, high-solidity-ratio hydrofoil impeller of the type that has become the standard in bioleach reactors. It is designed to disperse high volumes of air while also maintaining solids in suspension and promoting heat transfer. Such impellers are characterised by their ability to operate at high gas volumes without flooding. The upper impeller is the innovative aspect: it is an upward-pumping, three-blade medium-solidity-ratio impeller, known as the P3, which was originally designed for high-viscosity applications, but has now found application in three-phase (gas-liquid-solid) mixing systems. Improved mass-transfer performance is achieved through surface air induction created by the top impeller, and enhanced gas hold-up from the specific mixing pattern that is created by the dual-impeller configuration.

The four secondary bioleach reactors, which have considerably lower gas dispersion duties than the primary reactors, are fitted with single P4 impeller systems (Neale et al. 2015). The slurry from the final oxidation tank is pumped to the iron/arsenic removal circuit.

12.3.4.3 Iron and Arsenic Precipitation

In the iron and arsenic removal circuit, iron and arsenic are eliminated from the pregnant solution in a six-stage precipitation circuit, comprising one conditioning and five precipitation tanks, a thickener, and a horizontal filter press. Iron and arsenic are precipitated in aerated tanks with CaCO₃ to form a stable ferric arsenate. Slurry from the final iron/arsenic removal tank is pumped to a thickener, and the thickener overflow is pumped to a cartridge filter, which acts as a clarifier and the clarified solution proceeds to the metal precipitation circuit for nickel and cobalt recovery. The filtered ferric arsenate cake is re-pulped with water and pumped via the neutralisation circuit to the tailings pond (Laukka et al. 2018).

12.3.4.4 Metal Precipitation

The objective of the metal precipitation circuit is to precipitate a mixed nickel and cobalt hydroxide product with a nickel content of typically 42%, bagged for sale. The metal precipitation circuit is a five-stage precipitation circuit followed by thickening and precipitation. In the primary stage of precipitation, the pH is raised to a level of 8.5 by addition of MgO. The precipitated product is thickened and the thickener underflow is filtered in two horizontal filter presses. The filter cake from the filter presses is bagged and sold. The thickener overflow and the filtrate from the filter presses are directed to a secondary precipitation and filtration stage, where slaked lime (Ca(OH)₂) and MgO are used to further increase the pH to a level of 10. The filtrate from the second-stage filter can either be used as recycled water (for example, as wash water for the iron/arsenic filter or as repulp water for the iron/arsenic filter cake), or directed to the tailings stream (Laukka et al. 2018).

12.3.4.5 Recycle Water Treatment

Neutralised and clarified water from the metal recovery circuit is used as a process water source for the plant. However, it has too high a magnesium sulfate level for reuse in the bioleaching process or for safe environmental discharge and is therefore processed in the recycle water treatment circuit. The water is fed to a neutralisation tank, where CaO is used to raise the pH to a level of around 9.5. The resulting precipitated hydroxides and gypsum are then removed from the solution via a thickener. The thickener underflow is pumped to tailings, while the water, now containing low levels of sulfates and magnesium, is reused in the process (Neale et al. 2016).

12.3.4.6 Tailings Neutralisation

The tailings from the nickel sulfide treatment plant consist of repulped iron/arsenic filter cake and tailings water from the metal recovery circuit, which usually contains in the range of 0.01–0.1 g nickel L⁻¹. These streams are neutralised in a three-stage neutralisation circuit (comprising two neutralization tanks and a pumping tank with Ca(OH)₂ addition), where the pH target is 10.3–10.5 for optimal nickel precipitation.

The neutralised tailings stream is pumped to a separate area in the plant’s tailings storage facility in order to minimize the impact of the discharge of sulfates to the existing tailings dam, from which the talc concentrator and talc refinery derive their process water. Clarified water from this area is combined with the water stream from the talc operation’s tailings (Laukka et al. 2018).

12.3.5 Inoculation and Commissioning of the Commercial Plant

The build-up of the microbial inoculum for the bioleach plant was conducted in several phases. Initially, an inoculum was prepared at Mintek’s facilities in South Africa and air-freighted in the form of a filter cake to the plant site. The on-site inoculum buildup comprised a number of successive stages, starting with a small-scale five-stage continuous mini-plant and progressing through batch operation in 1, 3, and 6 m³ reactors.

The inoculation and start-up of the production plant commenced in late September 2015. The first two attempts failed, and this was attributed to the highly reactive pyrrhotite in the Sotkamo concentrate (which was used in the commissioning phase) reacting with acid in the absence of aeration and producing hydrogen sulfide (H₂S) gas, which poisoned the inoculum as soon as it was introduced into the vessels. This behaviour was not encountered during the metallurgical test work programme, for several reasons: the scale of operation in the laboratory- and pilot-scale test work prevented reducing conditions from forming, and pre-acidification was conducted under aerated (and therefore oxidising) conditions. The inoculation method was adjusted, and on the third attempt, successful inoculation of one of the primary bioleach reactors was achieved. From there, the other bioleach reactors were filled, allowing feeding of the bioleach plant (Neale et al. 2016).

During commissioning, several practical challenges, mainly equipment related, were encountered that needed to be addressed. These included a build-up of agglomerated lumps and small pebbles in the conically bottomed concentrate repulp tank, which then passed into the suction of the peristaltic pump that fed the regrind mill feed tank, causing regular blockages of the pump. This was fixed by the installation of a makeshift filter basket in the repulp tank, and lifting the withdrawal point (which was close to the bottom of the tank) above the conical section of the tank (Neale et al. 2016). A screen was also subsequently installed in the regrind mill feed tank (Laukka et al. 2018).

Initially, the upgrading circuit—comprising magnetic separation and flotation—was not commissioned, which resulted in the appearance of a persistent foam in the bioleach reactors, considered to be caused by the talc that occurs in the non-upgraded concentrates. To counter this problem, the flotation section of the upgrading circuit was designed to remove almost all of the talc in the concentrates, and foaming was combated by the use of an antifoaming agent.

The greatest challenge in the iron/arsenic precipitation circuit has been to find the correct parameters and operational methods for iron–arsenic filtration. This included issues such as the selection and blinding of the filter cloth material, finding the correct flocculant type and dosage, and the correct handling of the non-flocculated fines from the thickener. The commissioning of the hydroxide precipitation circuit was straightforward without major issues (Neale et al. 2016).

One of the main features of the design of this bioleach plant is that it needs to withstand the harsh Finnish winter. This aspect of the design was fully tested in January 2016, when the temperatures were particularly low for a sustained period. Temperatures as low as −30 °C were experienced and the impact on the plant was quite severe, with all of the pipework and instrumentation above the bioleach reactors becoming enveloped in ice. During the coldest week, the regrind mill feed tank pump failed, which in turn resulted in the bioleach feed pump being stopped and caused the bioleach feed line to freeze. This caused the exothermic sulfide oxidation reactions in the bioleach reactors to slow down, and the pulp temperatures to drop. This then caused the cooling water flow rates to decrease, resulting in the cooling water supply lines also freezing. Steam generators were brought to site, the lines were defrosted, and normal operation resumed. The bioleach pulp temperatures began rising once feeding was re-instituted, and the cooling water flow rates returned to the expected levels with no long-term impact on the process or microbial performance (Neale et al. 2016).

12.3.6 Operational Performance

During April and May 2016, a mass balance sampling campaign was instituted over the bioleach plant. At the time, the design throughput had not been reached yet and the bioleach plant was operating at a feed solids concentration of 11% and a grind size of 80% < 50 μm. The average nickel and cobalt extractions attained in the bioleach plant over this period were 97.4% and 98.4%, respectively. Although the plant was not operating under full design load, these results provided confidence in the robustness, stability, and efficiency of the bioleaching section of the plant.

The residue from the iron–arsenic precipitation circuit was subjected to environmental stability testing, which showed that the neutralised product met the requirements for classification as a regular waste. The metal precipitation plant was also successfully commissioned and at the time a product containing around 42 to 43% nickel hydroxide and about 1% cobalt hydroxide was produced (Neale et al. 2016).

Over the following 2 years, the complete circuit was commissioned and production was ramped up, although a few bottlenecks remained which prevented the attainment of the design throughput. A mass balance campaign conducted during 2018 at a feed solids concentration of 17% and an overall residence time of 9 days, indicated nickel and cobalt recoveries of 87.8% and 90.7%, respectively, in the bioleaching circuit. An overall sulfide oxidation level of 94.9% was estimated. The P₈₀ of the feed material was 58.7 μm and it is anticipated that metal extraction would improve with further optimisation of the regrind mill circuit.

The overall nickel recovery achieved in the treatment plant was 80% of the MHP product from the bioleach feed. It was thought that further improvement to a desired range of 85–90% could be possible, particularly with optimisation of pH control in the precipitation circuits and of the iron/arsenic filter wash sequences.

The quantity of the mixed hydroxide precipitate being produced was of excellent quality, typically containing 47% nickel and 2% cobalt. The MHP produced was sold and used for the production of battery-grade nickel and cobalt sulfates (Laukka et al. 2018).

Further developments during this time included investigation of the possibility to derive additional value from the gold and platinum group metals occurring in low concentrations in some of the nickel concentrates that form the feed to the nickel bioleaching plant. This led to the development and piloting of a process to recover gold and platinum group metals from the bioleach residue. These metals can successfully be recovered through the production of an upgraded gold-containing concentrate derived from the bioleach residue which is suitable for sale to refiners. The process, developed with the Geological Survey of Finland (GTK), utilises Knelson concentrators and a strong acid wash (Laukka et al. 2018).

In the last quarter of 2018, Mondo Minerals was acquired from US private equity firm Advent International by the British speciality chemicals company Elementis plc. In view of the decline in the nickel price during the second half of 2018, the new owners took the decision to suspend operations, and the plant is currently being kept in “care and maintenance” mode, though this halt in production is considered to be temporary. Prior to the cessation of the operation, good progress had been made in ramping up the production, and the throughput was just below the design value (Neale 2020).

12.4 Conclusions

The suitability of bioleaching as an extraction process for nickel and cobalt sulfides has been successfully demonstrated at commercial scale. However, the application of mineral processing biotechnologies remains relatively limited and some unique feature of the feed material is typically required for bioleaching to be selected as the technology of choice. In the case of the Talvivaara/Terrafame project, the complexity of the orebody was the overriding factor that led to the selection of a heap leaching process, while at Mondo Minerals, the presence of arsenic in the sulfide material was the reason for the selection of a tank bioleaching treatment process.

Although the demand for nickel is driven mainly by the world stainless steel market, it is anticipated that the demand for high-quality nickel and cobalt sulfate will surge as the trend towards electric vehicles and the development of battery-based technologies for large-scale energy storage increases. It is assumed that, if demand grows to the extent that is forecast, exploration may lead to the discovery of new nickel- and cobalt-containing sulfide deposits, which may increase the potential for future applications of bioleaching technologies, particularly since they are ideally placed to enable integration with downstream value addition.