The Direct Leaching of Nickel Sulfide Flotation Concentrates – A Historic and State-of-the-Art Review Part I: Piloted Processes and Commercial Operations

ABSTRACT

The extraction of nickel (Ni) from sulfide resources commences with flotation to produce a concentrate which is then smelted to produce a nickel-enriched phase called matte, and further refined to produce pure Ni products as well as by-products, such as cobalt (Co), copper (Cu) and precious metals. However, the traditional concentrate smelting-matte refining process, whilst technologically robust, is capital intensive and suffers from several environmental and technical issues such as sulfur dioxide emissions, poor recovery of cobalt and difficulty processing concentrates high in magnesia and arsenic without appropriate blending with high grade concentrates to dilute the concentration of these species. The direct hydrometallurgical processing of nickel sulfide concentrates and whole ores may be a remedy to these issues and hydrometallurgy offers several advantages over pyrometallurgy such as potentially lower capital costs, the ability to process lower grade materials and produce marketable metals or compounds directly from ore/concentrate. Despite the advantages of hydrometallurgy over traditional base metal sulfide smelting, the hydrometallurgical processing of nickel concentrates has been limited to a small handful of commercial operations, some of which only produce an upgraded intermediate for pyrometallurgical processing. In Part I of this three-part series, a comprehensive review of piloted processes for direct hydrometallurgical processing of nickel sulfide concentrates is presented, followed by a survey of industrial operations which have carried out direct leaching of nickel sulfide concentrates. A review of research activities and challenges/opportunities in the direct hydrometallurgical processing of nickel sulfide concentrates are presented in Part II and Part III of this series.

KEYWORDS:

• Nickel
• cobalt
• pentlandite
• nickel sulfides
• hydrometallurgy

1. Introduction

Nickel (Ni) is an essential metal for society with stainless steels being its major end use, though high purity nickel sulfate, a key raw material for production of lithium-ion battery (LIB) cathodes essential for electric vehicles (EVs) is a rapidly growing market (McRae 2020a). Nickel demand has traditionally been driven by stainless steel production, either in the form of class 2 (<99.8% Ni) nickel products such as nickel oxide, ferronickel and nickel pig iron (NPI), or higher purity class 1 (>99.8% Ni) nickel such as nickel powder, briquettes or electrowon cathode (Campagnol et al. 2017). Campagnol et al. (2017) have indicated that demand for class 1 nickel in the form of high purity nickel sulfate for lithium-ion batteries is expected to reach 570,000 tonnes (Ni metal basis) in 2025.

Terrestrial nickel resources are contained in sulfide and lateritic deposits, with 30–40% of the land-based nickel resources being contained in sulfide deposits with the balance in laterite deposits (Dalvi, Bacon, and Osborne 2004; Hoatson, Jaireth, and Jaques 2006; Mudd 2010). Major nickel sulfide deposits are in Australia, Canada, Russia and Southern Africa whilst most nickel laterite deposits are located in Indonesia, New Caledonia, Brazil and the Philippines (Hoatson, Jaireth, and Jaques 2006). Despite laterites being more abundant than sulfide deposits, Ni production has historically been derived from sulfides due to these resources being more amenable to beneficiation and less challenging to process, resulting in lower costs relative to Ni production from laterites (Elias 2002; Mudd 2010). However, recent data from the United States Geological Survey (USGS) show that Ni production from laterite ores has overtaken sulfide resources over the 2011–2016 period (McRae 2018, 2020b). The key driver appears to be Chinese stainless-steel producers using NPI rather than class 1 nickel to reduce manufacturing costs (Campagnol et al. 2017); Ni production data from China show that approximately 60% of Ni production in 2016 was in the form of ferronickel and NPI (McRae 2020b). Since 2014, NPI production has also expanded in Indonesia, coinciding with the ban by the Indonesian government on the export of direct shipping nickel laterite ores (McRae 2020b).

Lateritic ores are not typically amenable to beneficiation and therefore direct processing of whole ore using hydrometallurgical or pyrometallurgical techniques is necessary with the choice of processing method being dependent on the type of ore (Dalvi, Bacon, and Osborne 2004; Elias 2002; Marsh, Anderson, and Gray 2013). High pressure acid leaching (HPAL) is the preferred method for the extraction of Ni and Co from oxide (limonite) and clay silicate type ores to produce a diverse range of products (i.e. mixed sulfide precipitate, mixed hydroxide precipitate, metal powder, cathode) whilst saprolitic ores are generally processed via pyrometallurgy to produce either ferronickel or sulfide matte (Dalvi, Bacon, and Osborne 2004; Elias 2002; Marsh, Anderson, and Gray 2013). Nickel laterite projects employing hydrometallurgical or pyrometallurgical processing have often suffered from high capital expenditure and cost blowouts, have required long ramp up times, suffered from technical issues relating to commissioning and ramp up, and in the case of HPAL, have at times failed to achieve nameplate capacity (Jones et al. 2010; McDonald and Li 2020; Oxley, Smith, and Caceres 2016). The extraction of Ni from sulfide ores is much simpler relative to laterites and is dominated by pyrometallurgical processing. This typically involves comminution and flotation to produce a nickel concentrate which is subsequently smelted and converted to produce a sulfur-deficient, metal-enriched phase termed matte, which is then refined to produce pure Ni, Cu, Co metal products and precious and platinum group metals (PGMs) where applicable (Crundwell et al. 2011b). A simplified flow chart of the flotation-smelting-matte refining process for treatment of nickel sulfide ores is presented in Figure 1. There is great deal of variation in the standard flow sheet with respect to flotation practice (i.e. mineralogy, liberation size, reagents, etc.) and converter matte refining, however concentrate smelting is dominated by either flash smelting or roasting-electric furnace smelting technology whilst matte converting is almost exclusively done using Peirce-Smith converters (Moskalyk and Alfantazi 2002; Warner et al. 2007).

Figure 1. Simplified traditional process flowsheet for the extraction of Ni, Cu and Co from nickel sulfide resources.

The objective of concentrate smelting is to remove part of the sulfur and iron present in the concentrate by oxidation; sulfur is oxidized to sulfur dioxide (SO₂) whilst iron forms iron oxide which is fluxed by reaction with silica to form an iron silicate slag (Crundwell et al. 2011f). This iron silicate slag also removes any non-sulfide gangue present in the concentrate which dissolves into the molten slag (Crundwell et al. 2011f). Typical base metals content of concentrates utilized for smelting contain a combined Ni, Cu and Co grade of 8% to above 20%, though lower grade concentrates having a combined base metal concentration of 3–4% are smelted in Southern Africa due to the high PGM content of these materials (Warner et al. 2007). Concentrate smelting is carried out either by roasting-electric furnace smelting or via flash smelting and the key process steps in both processes are presented in Figure 2; the key principles of both methods are covered in detail elsewhere (Crundwell et al. 2011c, 2011f). The production details of both pyrometallurgical processing routes are presented in Table 1 including typical Ni and Cu grades in the concentrates, mattes and slags. It can be seen from Table 1 that either pyrometallurgical route is very efficient at recovering Ni, Cu and precious metals to the matte. Base metal contents in slags produced from flash smelting are generally higher and recoveries are lower relative to the electric furnace route due to the more oxidizing conditions employed, however slags are processed to recover base metals through slag cleaning operations. Cobalt recoveries are poor by either route (Matousek 1982; Warner et al. 2007).

Figure 2. Process steps during the pyrometallurgical processing of nickel sulfide concentrates to produce nickel converter matte via flash smelting or the roasting-electric furnace melting route. Adapted from Crundwell et al. (2011f).

Table 1. Production details of nickel concentrate smelting via the roasting-electric furnace smelting route and the flash smelting route. Data for roasting-electric furnace smelting and flash smelting adapted from Crundwell et al. (2011c, 2011f).

Smelting process	Roasting-electric furnace melting^a	Flash smelting^b,c (Outotec & Inco)
Typical concentrate grades	5–14% Ni, 0.3–4.5% Cu, 0.2–0.5% Co, 31–43% Fe	5–15% Ni, 0.3–12% Cu, 0.2–0.4% Co, 30–43% Fe
%Ni recovery to matte	98	95
Ni grade in matte (wt%)	13–36	17–65
Ni grade in slag (wt%)	0.1–0.4	0.2–4
%Co recovery to matte	50–80	26–70
Co grade in matte (wt%)	0.7–1	0.4–0.8
Co grade in slag (wt%)	0.1–0.2	0.1–0.5
%Cu recovery to matte	98	80–93
Cu grade in matte (wt%)	1–11	1.5–25
%Cu grade in slag (wt%)	–	0.1–1.3
Recovery of precious metals (%)	98–99	95

Based on production details of three Ni smelters – Falconbridge (Canada), Thompson (Canada) and Norilsk (Russia).

Based on production details of seven Ni smelters – Kalgoorlie (Australia), Selebi Phikwe (Botswana), Gansu (China), Norilsk (Russia), and Harjavalta (Finland), and two Ni smelters in Sudbury (Canada).

Figures do not consider slag cleaning which is typically performed either ex-situ in a separate furnace or as an extension of the flash smelter.

The final product from nickel concentrate smelting by either route is a molten nickel sulfide enriched phase (matte) which typically contains 15–40% Ni, 20–40% Fe, and 20–25% S as well as Cu, Co and any PGMs (Crundwell et al. 2011a). The molten matte is transferred to a converter where iron is oxidized using air or oxygen and removed from the matte as an iron silicate slag; the final product, termed converter matte, typically contains 40–70% Ni, 0.5–4% Fe, 21–24% S as well as Cu and Co (Crundwell et al. 2011a; Warner et al. 2007). There is a great deal of variation with regards to the processing of nickel converter matte to recover base and precious metals, with direct hydrometallurgical processing of granulated converter matte being practiced at some refineries (Warner et al. 2007). In other instances, such as mattes arising from smelting of nickel-copper concentrates and concentrates high in PGMs, slow cooling of the converter matte is carried out to enable copper and nickel phase separation and the formation of coarse sized crystals of chalcocite (Cu₂S), heazlewoodite (Ni₃S₂) and a metallic alloy phase enriched in precious and platinum group metals; the cooled matte is beneficiated to produce separate Cu and Ni-sulfide concentrates and a PGM-enriched concentrate for further processing (Bulatovic 2007; Crundwell et al. 2011e; Warner et al. 2007). Magnetic separation is carried out at Anglo American Platinum’s refinery in Rustenburg, South Africa, on slow-cooled mattes to recover the PGMs, with the non-magnetic fraction (nickel-copper matte) being processed via hydrometallurgy in a dedicated base metals refinery producing copper cathode and metallic nickel and cobalt powders (Crundwell et al. 2011d). Processes for the smelting of nickel flotation concentrates to produce matte followed by matte refining to produce Ni, Co and Cu products are well established. The primary advantages of nickel sulfide concentrate smelting are the following (Crundwell et al. 2011b, 2011c, 2011d, 2011f):

• Combustion of part of the sulfur and iron in the concentrate can supply most of the energy requirements for smelting (flash smelting only) and converting, with the added advantage of removing a large part of the iron and sulfur from the concentrate as slag and sulfur dioxide (SO₂), respectively, simplifying downstream matte refining,

• Starting from a nickel flotation concentrate of grade 5–20% Ni, smelting and converting produces a nickel matte containing 40–70% Ni, resulting in a considerable reduction in size of downstream leaching and refining equipment relative to direct leaching of flotation concentrates,

• Precious metals report to the matte and are more concentrated relative to the original flotation concentrates

Against these advantages are several drawbacks such as:

• Difficulty processing nickel concentrates high in arsenic (As) and magnesia (MgO), which can render many concentrates, particularly those produced from disseminated sulfide ores and talc-carbonate ores, unsuitable if they cannot be blended with higher quality concentrates to reduce the concentration of the deleterious species,

• Poor recovery of cobalt to the matte phase during smelting and converting, where typically 30–50% of cobalt is lost to the metallurgical slag (Matousek 1982; Warner et al. 2007),

• High capital cost associated with concentrate smelting and matte converting, including the capture of SO₂ and its conversion to sulfuric acid. The latter can only be justified where there is a market for the sulfuric acid that is produced (Lu et al. 2000),

• Plant hygiene issues relating to the use of ladles and launders during pyrometallurgical processing (Dutrizac 1992).

Elevated MgO grade (as talc, orthopyroxene etc.) in nickel smelter feeds cause operational problems for the smelter due to MgO raising the slag liquidus temperature and viscosity, leading to poor slag-matte separation and higher metal losses due to entrainment (Eksteen, Oraby, and Nguyen 2020; Lotter et al. 2008). This necessitates furnace operation at higher temperature, which shortens the life of the furnace refractories. The MgO content in the flotation concentrate(s) must be limited to below 7–10% to prevent operational problems during smelting (Feng et al. 2012; Senior and Thomas 2005). Arsenic in the smelter feed is also problematic both from an environmental and operational perspective. Toguri, Babaie, and Sridhar (1995) has shown that the presence of As complicates the refining of Cu-Ni mattes, leading to a decrease in grain size of chalcocite and alloy phases during cooling and altering the morphology of chalcocite (from dendritic to spherical in the presence of As). Whilst As can sometimes be rejected at the concentrator, this can be an issue when the As is present as niccolite (NiAs) and gersdorffite (NiAsS), which will result in nickel losses to the mill tailings.

Other situations where smelting may be unjustified include if the size of the resource to be processed is too small to justify the high capital cost associated with a dedicated smelter and matte refining facility, or when the transport cost to ship nickel concentrate(s) to a toll smelter is prohibitively high (Fleming et al. 2000). Under these circumstances, direct hydrometallurgical processing of nickel sulfide concentrates can be a viable alternative, and many of the technical, economic and environmental challenges imposed on smelting have spurred research and development activities around hydrometallurgical processes for the extraction of Ni, Co and other base metals (Cu, Zn) from sulfide concentrates. Direct hydrometallurgical processing of whole nickel sulfide ores has also received considerable attention, particularly bioleaching, which was the subject of a review by Watling (2008). Despite the level of research and development over the years on nickel concentrate hydrometallurgy, the adoption of hydrometallurgical processes for the extraction of Ni, Co and Cu from nickel flotation concentrates remains low, particularly given several operations utilize some form of hydrometallurgical processing to produce an upgraded sulfide product for smelting, or the operation originally built for processing sulfide concentrates has switched over to other nickel sulfide feedstocks.

The purpose of Part I of this three-part series is to review and summarize piloted leaching processes as applied to the direct hydrometallurgical processing of nickel sulfide flotation concentrates and secondly, to survey commercial nickel operations which have implemented hydrometallurgical processing. An overview of the geology and mineralogy of nickel sulfide deposits, as well as current beneficiation practice is also provided. Following an examination of pilot plant developments and commercial operations concerning the hydrometallurgy of nickel sulfide concentrates, laboratory research activities concerning direct leaching of nickel sulfide concentrates are reviewed in Part II and Part III of this series and the challenges and research opportunities relating to nickel sulfide concentrate hydrometallurgy are discussed.

2. Nickel sulfide deposits

Nickel sulfide deposits are found on most continents, however major deposits are found in Russia, Australia, Canada, and Southern Africa (Hoatson, Jaireth, and Jaques 2006). Canada and Russia possess the largest resources of nickel sulfides, followed by Australia and South Africa; however, Australia has the largest nickel resources when laterites are also considered (Hoatson, Jaireth, and Jaques 2006). All nickel sulfide deposits are magmatic in origin and are broadly characterized as either sulfide rich (>5%) or sulfide poor (<5%); for sulfide rich deposits, Ni and Cu are of primary economic interest whilst for sulfide poor deposits, generally PGMs are of economic importance (Naldrett 1999; Song, Wang, and Chen 2011). The major nickel sulfide deposits including the nickel grade, contained nickel metal and year of discovery are presented in Table 2; typical Ni and Cu concentrations in magmatic Ni-Cu sulfide deposits are 0.7–3% Ni and 0.2–2% Cu (Eckstrand and Hulbert 2007); the deposits of the Bushveld Igneous Complex (South Africa) are lower grade but possess high concentrations of PGMs for which they are primarily mined. The variation in Ni grade among some of the major deposits is indicative of the proportion of high-grade massive sulfide ore to low grade disseminated sulfide ore (Barnes, Barnes, and Perring 2007). It can be seen from Table 2 that the Sudbury complex (Canada) and the Noril’sk-Talnakh deposits (Russia) are the largest sulfidic nickel deposits, being far larger than the other listed deposits based on contained Ni metal alone. Whilst Australia possesses significant reserves of sulfidic nickel and is ranked after Canada and Russia, the deposits are much smaller.

Table 2. Nickel grade, contained nickel metal (in tonnes) and year of discovery of major nickel sulfide deposits across the globe; data adapted from Hoatson, Jaireth, and Jaques (2006) unless stated otherwise.

Deposit	Nickel grade (% Ni)	Contained Ni metal (tonnes)	Year of discovery
Australia
Kambalda, WA	3.30%	1,388,820	1966
Honeymoon Well, WA	0.75%	1,011,000	1973
Perseverance, WA	1.05%	2,480,640	1971
Mt Keith, WA	0.57%	3,411,420	1968
Yakabindie, WA	0.58%	1,682,000	1970
Nebo-Babel, WA*	0.47%	1,200,000	2000
Canada
Voisey’s Bay	1.59%	2,173,530	1993
Thompson	2.32%	3,486,960	1956
Sudbury	1.20%	19,776,000	1856
Russia
Noril’sk – Talnakh	1.84%	23,128,800	Noril’sk discovered in 1926, Talnakh in 1960
Pechenga	1.18%	4,000,200	1912
South Africa
Merensky reef, Bushveld complex	0.15%	6,313,500	1924
Platreef, Bushveld complex	0.41%	6,547,700	1925
China
Jinchuan, Gansu province	1.06%	5,459,000	1958
USA
Duluth complex, Minnesota	0.2%	8,000,000	1948

*Based on mineral resource and ore reserves estimate by OZ Minerals (2020).

2.1. Sulfide mineralogy

A list of economically important nickel sulfide and arsenide minerals, as well as common sulfidic gangue in nickel sulfide deposits is presented in Table 3. Pentlandite is the most important nickel sulfide mineral and commonly occurs with pyrrhotite and chalcopyrite. Other nickel sulfides listed in Table 3 can be important where alteration has taken place. Violarite is a supergene alteration product of pentlandite and typically occurs in pyrite-violarite and transition zones above the primary zone (massive pentlandite-pyrrhotite ore) (Marston et al. 1981; Nickel, Ross, and Thornber 1974) and can be economically important during the mining of some massive nickel sulfide ores. Millerite forms as an alteration product of pentlandite (Bide, Hetherington, and Gunn 2008; Holwell et al. 2017) and is an important nickel sulfide mineral in some deposits such as Mt Keith, Black Swan and the Otter Shoot, Kambalda (Dowling et al. 2004; Grguric et al. 2007; Keele and Nickel 1974). Niccolite and gersdorffite can occur in nickel sulfide deposits as a result of hydrothermal alteration; their presence in nickel concentrates is undesirable due to the deleterious effects of As during smelting and typically requires dilution to an acceptable level by blending with low As concentrates (Grguric et al. 2007). Pyrrhotite, an iron sulfide, is the primary gangue sulfide that occurs in nickel sulfide deposits though it typically contains Ni, either in solid solution or as fine pentlandite intergrowths (Rezaei et al. 2017; Toguri 1975). Typically, the Ni concentration in solid solution in pyrrhotite is 0.4–0.6% and the presence of micron-size inclusions of pentlandite can raise the nickel content even higher (Rezaei et al. 2017; Toguri 1975). Hence during beneficiation of nickel sulfide deposits, nickel loss due to pyrrhotite rejection is inevitable.

Table 3. Economically important nickel sulfide and arsenide minerals and their mode of occurrence in nickel sulfide deposits listed in terms of increasing Ni content (wt%) and sulfidic gangue typically present in nickel sulfide deposits.

Name	Chemical formula	Ni content (wt%)	Mode of occurrence
Nickel sulfide minerals
Pentlandite	(Fe, Ni)9S8	34.21	Mafic intrusions
Violarite	FeNi2S4	38.93	Supergene alteration of pentlandite-pyrrhotite assemblages
Millerite	NiS	64.67	Replacement product of pentlandite
Nickel arsenide minerals
Gersdorffite	NiAsS	35.42	Hydrothermal replacement of pentlandite
Niccolite (nickeline)	NiAs	43.93
Sulfidic gangue
Pyrrhotite	Fe(1-x)S	0.4–0.6	Mafic intrusions
Pyrite	FeS2	Variable, secondary pyrites can have Ni contents as high as 12%*. Primary pyrites can have Co contents as high as 2.37%† and can be an important host for this element.	Mafic intrusions (primary), supergene alteration of pentlandite-pyrrhotite assemblages (secondary)

*Nickel, Ross and Thornber (1974).

†Watmuff (1974).

2.2. Ore textures

Nickel sulfide ores are broadly classified into the following categories: massive sulfide ore, matrix ore (also termed net-textured ore) and disseminated sulfide ore, the distinction being based on the sulfide content and silicate texture (Barnes and Lightfoot 2005). Massive sulfide ores consist of 75–100% sulfides as pyrrhotite, pentlandite and chalcopyrite, with or without minor pyrite (Barnes et al. 2018; Barnes, Barnes, and Perring 2007). Where silicate inclusions are present in massive ores, they are generally dispersed and isolated (Barnes et al. 2017; Jones 1996). Matrix sulfide ores comprise silicate inclusions in a matrix of sulfide, the sulfide making up 30–70% of the ore (Barnes et al. 2017). Disseminated sulfide ores contain between 1% and 33% sulfides dispersed in a silicate matrix, the sulfides are not interconnected and are present as patches in the silicate interstices or as globules (Barnes and Lightfoot 2005). The presence of multiple ore textures/types in a single deposit is not uncommon; for example, the Jinchuan deposit (Gansu province, China) contains both disseminated and matrix ore (Naldrett 2004) whilst all three ore textures are found at the Noril’sk-Talnakh camp (Barnes et al. 2018; Eckstrand and Hulbert 2007). The massive sulfide ores at Noril’sk-Talnakh are high grade, containing 3.42% Ni and 5.38% Cu, but are of lower tonnage (88.7 Mt) relative to the lower grade disseminated ore (1706.3 Mt) which grades 0.51% Ni and 1.02% Cu (Eckstrand and Hulbert 2007). Consequently, different flotation regimes have been employed for processing the various types of Ni-Cu sulfide ores present at the Noril’sk-Talnakh camp (Kozyrev et al. 2002).

2.3. Beneficiation practice

Nickel sulfide ores are subjected to mineral beneficiation to reject sulfidic and non-sulfidic gangue and produce a nickel concentrate grading from 5% to 25% Ni suitable for smelting (MäKinen and Taskinen 2008). Beneficiation is accomplished through comminution and froth flotation (refer to Figure 1); in some instances, magnetic separation is employed before or after flotation to reject monoclinic pyrrhotite which is ferromagnetic (Agar 1991; Bulatovic 2007). The choice of unit operations and operating parameters is largely dictated by the type of ore being processed i.e. massive sulfide ore vs disseminated sulfide ore which subsequently influences the liberation size and choice of flotation collectors and depressants for separation of the values from the gangue. Other factors to consider include if the ore has been altered, which can bring about mineralogical and textural changes; the effects of supergene alteration and serpentinization on nickel sulfide flotation have been well documented (Eltham and Tilyard 1973; Kerr 2002; Mani et al. 1997; Mishra, Viljoen, and Mouri 2013; Sizgoric 1981). Flotation practice differs by geographical location. Bulk sulfide flotation is generally carried out on Western Australian nickel ores which tend to be low in copper and contain substantial concentrations of magnesium bearing minerals and hydrophobic gangue such as talc (Bulatovic 2007; Eltham and Tilyard 1973; Pietrobon et al. 1997). Bulk sulfide flotation is also carried out on ores from the Merensky Reef and UG-2 chromitite in South Africa due to the association of PGMs with base metal sulfides in these ores (Xiao and Laplante 2004); the concentrates are lower grade (3–4% combined base metals) but possess high concentrations of PGMs (100–400 g/t) (Warner et al. 2007). Flotation practice in Russia and Canada differs from Western Australia where the deposits tend to be massive copper-nickel sulfide ores. In these ores, pyrrhotite is the major gangue mineral and flotation is carried out to reject pyrrhotite and produce separate copper and nickel concentrates. Magnetic separation is generally incorporated in the flowsheets to remove monoclinic pyrrhotite (Agar 1991; Bulatovic 2007). Pyrrhotite rejection is necessary as the smelting of pyrrhotite generates large quantities of SO₂ but yields very little nickel (Senior, Shannon, and Trahar 1994). Pyrrhotite rejection however must be balanced against Ni losses during beneficiation since pyrrhotite contains Ni in solid solution or as exsolved pentlandite that is too fine to liberate (Bulatovic 2007; Kerr 2002; Senior, Shannon, and Trahar 1994). Sequential flotation is employed to process massive sulfide ore from the Noril’sk-Talnakh deposit in Russia and produce separate copper, nickel and pyrrhotite concentrates (Bulatovic 2007; Kozyrev et al. 2002); the pyrrhotite concentrate has a high concentration of Ni, Cu and PGMs and is processed further in a dedicated hydrometallurgical facility to recover the contained metals (refer to Section 3.2.2).

3. Piloted processes and commercial operations

In response to the technical and environmental hurdles presented by the traditional concentrate smelting-matte refining route, the direct leaching of nickel sulfide flotation concentrates has been intensely researched and a number of these processes have been piloted and, in some cases, have seen commercial implementation. Several nickel processing operations have practiced hydrometallurgical processing of nickel sulfide flotation concentrates or whole ore at commercial scale and these are summarized in Table 4. It can be seen in Table 4 that there is wide variability in terms of the feedstocks processed, finished nickel products and leaching technology employed. A number of these operations are now closed or no longer process flotation concentrates, having undergone modifications to the original refineries to accommodate changes in feedstock (i.e. Kwinana, WA and Fort Saskatchewan, Canada). Other operations such as the Nadezhdin mill (Russia) and the Cosmic Boy nickel concentrator (West Australia) carry out hydrometallurgical processing to produce upgraded intermediates which are blended with flotation concentrates destined for smelting. The processing of nickeliferous pyrrhotite concentrates is practiced at the Nadezhdin mill (Russia) whilst historic production took place in Canada (refer to Table 4). The Canadian operations are unique in that roast-leach type processes were utilized to extract base metals from the pyrrhotite and produce an iron oxide residue suitable for production of iron ore pellets destined for steel making. The purpose of this section is to summarize piloted processes that have been developed to extract Ni directly from flotation concentrates and review several commercial operations that have practised direct hydrometallurgical processing. While a number of these operations are now closed or no longer process flotation concentrates, a description has been provided for historical purposes.

Table 4. Commercial nickel operations that have processed nickel-bearing sulfide concentrates via hydrometallurgical methods.

Operation	Location	Operator	Leaching process employed	Feed to refinery	Final Ni products	Operational status	Reference(s)
Ammoniacal pressure leaching
Fort Saskatchewan	Canada	Sherritt International Limited	Sherritt-Gordon ammonia pressure leach	Concentrate from Lynn Lake; 1954 –1970s, mix of concentrate and matte; 1970–1980s mixed sulfide precipitate from Moa Bay (Cuba); 1991– present	Nickel briquettes and powder	Still operating	Forward (1953); Forward and Mackiw (1955); Kerfoot (1989); Kerfoot and Cordingley (1997)
Kwinana	Western Australia	BHP Ni West (formerly owned by WMC Resources)	Sherritt-Gordon ammonia pressure leach	Concentrate(s); 1970 – 1975, Concentrate(s) + flash smelter matte; 1975–1985, matte only; 1985 – present	Nickel briquettes and powder; plan to produce battery grade nickel sulfate after completion of brownfield expansion	Still operating	Deng (1993), Wishaw (1993); Woodward (2014); Woodward and Bahri (2007); Anon (2021)
Sulfuric acid pressure oxidation
Garfield cobalt refinery	USA	Calera Mining Co.	Sulfuric acid pressure oxidation	Cobaltite concentrate from the Blackbird mine, Idaho (USA)	Co-Ni granules grading 95.6% Co and 3.9% Ni	Plant shut in 1959 due to shortage of ore	Fassell (1962); Meddings and Evans (1971); Berezowsky (2000)
Fredericktown refinery	USA	National Lead Co.	Sulfuric acid pressure oxidation	Pyritic concentrate produced as a by-product of copper-lead mining	Nickel powder	Plant shut in 1960 due to shortage of feed	Fassell (1962); Meddings and Evans (1971); Mitchell (1957); Berezowsky (2000)
Long Harbor	Canada	Vale Limited	Sulfuric acid pressure oxidation preceded by atmospheric chlorine leach	Concentrate from Voisey’s Bay (Canada)	Nickel cathode	Still operating	Crundwell et al. (2011b); Stevens et al. (2009)
Kokkola works	Finland	Outokumpu Oy	Sulfuric acid pressure oxidation (HIKO)	Ni concentrate from the Hitura mine (Finland, now closed)	Nickel chemicals	Site is operating as a cobalt refinery by Umicore. Status of the pressure leach plant unknown.	Nyman et al. (1992), Berezowsky (2000), Cooke (2021); Törmänen and Tuomela (2021)
Nadezhdin mill	Russia	Nornickel (Formerly known as Noril’sk Mining and Metallurgy Corporation)	Sulfuric acid pressure oxidation	Pyrrhotite concentrates from the Talnakh deposit(s) grading 2–5% Ni, 1–2% Cu and up to 15 g/tonne PGMs	Non-ferrous metals precipitated from solution after leaching and a Ni-Cu-Co-PGM concentrate is recovered for smelting	Still operating	Deng (1992, 1993); Berezowsky (2000); Kitay, Mechev, and Volkov (1991)
Bioleaching
Talvivaara	Finland	Terrafame Ltd	Heap bioleach	Black schist ore from Sotkamo deposit, grading 0.27% Ni, 0.56% Zn, 0.14% Cu	Mixed nickel-cobalt sulfide precipitate; MSP is being processed further to produce Ni and Co sulfates	Still operating	Riekkola-Vanhanen (2007, 2013); Watling (2008); Törmänen and Tuomela (2021)
Vuonos	Finland	Elementis (formerly Mondo Minerals)	Stirred tank bioleaching	Nickel concentrate by-product from talc mining at Sotkamo and Vuonos	Ni-Co mixed hydroxide precipitate (MHP)	Plant only periodically operational according to Törmänen and Tuomela (2021)	Laukka et al. (2018); Neale et al. (2016, 2015), Törmänen and Tuomela (2021)
Cosmic Boy nickel concentrator	Western Australia	Western Areas Limited	Stirred tank bioleaching (BioHeap^TM)	Cleaner tailings stream from flotation grading 18.3% Ni	Ni recovered from solution by precipitation as a sulfide and blended into final flotation concentrate	Still operating	Fewings and Seet (2014); Fewings et al. (2015, 2016)
Roast-leach
Falconbridge iron ore plant	Canada	Falconbridge Ltd.	Sulfation roasting	Pyrrhotite concentrate (1.13% Ni) from Falconbridge operations	Ni (& Cu, Co) recovered from solution by precipitation as a sulfide	Plant shut in 1972	Rezaei et al. (2017), Thornhill (1954, 1961)
Inco iron ore recovery plant (IORP)	Canada	Inco Ltd. (now Vale Ltd)	Modified Caron process	Pyrrhotite concentrate (0.75% Ni) from Copper Cliff or Clarabelle operations	Nickel oxide product grading 77% Ni, 0.15% Co and 0.15% S	Reduction roast-leach section of the plant shut in 1982. Oxidizing roaster closed in 1991.	Conard (2013)’; Inco Staff (1956); Rezaei et al. (2017); Queneau, Sproule, and Illis (1949)
Kokkola works	Finland	Outokumpu Oy	Sulfation roasting	Pyrite-pyrrhotite concentrate from Outokumpu copper mine grading 0.67% Co and 0.37% Ni	Mixed Cu-Ni-Zn sulfide precipitate fed to the smelter at Harjavalta (Finland); cobalt metal produced	Roaster operated between 1966–1987. Co and Ni production continued after roaster shutdown	Deng (1992); Palperi and Aaltonen (1971); Sauni et al. (2010)

3.1. Ammoniacal leaching processes

The ability of ammonia to selectively dissolve Cu, Ni, Co and Zn from their ores and concentrates has formed the basis for several commercially developed leaching processes. The first commercial operations utilizing ammonia-ammonium carbonate solution as a lixiviant were deployed to extract Cu from mill tailings by Kennecott in Alaska, and Calumet and Heca in Michigan, USA (Forward and Mackiw 1955; Meng and Han 1996). Anaconda’s Arbiter process and BHP’s Escondida process were developed to extract copper from sulfide flotation concentrates utilizing ammonia-ammonium sulfate as a lixiviant (Duyvesteyn and Sabacky 1993; Kuhn, Arbiter, and Kling 1974). Ammonia-ammonium lixiviants have been utilized for the extraction of Ni and Co from lateritic ores via the Caron process, which involves reduction roasting followed by ammonia-ammonium carbonate leaching (Caron 1950) and was in use at the Yabulu refinery in Queensland for processing laterite ores (now in care and maintenance) and by Inco (now Vale) until 1982 for treating pyrrhotite concentrates via a modified process (see Section 3.6.2). The only commercialized ammonia leaching process to date for the direct treatment of nickel sulfide concentrates is the Sherritt-Gordon ammonia pressure leaching process which is currently in use at Fort Saskatchewan, Canada, and Kwinana, Western Australia. The Sherritt-Gordon process is essentially a pressure oxidation process carried out in aqueous ammonia-ammonium sulfate at elevated temperature using air as an oxidant; the key operating characteristics and features of the process are listed in Table 5 and compared against other piloted and commercialized processes. The leaching technology was developed to process nickel concentrate from Lynn Lake (Canada) grading 10–16% Ni, 1–2% Cu and 0.3–0.4% Co, where a hydrometallurgical process was deemed to be more cost effective than concentrate smelting and matte refining (Forward 1953; Forward and Mackiw 1955; Kerfoot 1989).

Table 5. Summary of hydrometallurgical processes developed for the direct leaching of nickel sulfide flotation concentrates and status of the technology.

Process	Developer	Typical Ni grade in feed	Lixiviant	Operating Pressure (absolute)	Operating temperature	Total oxidation of sulfide to sulfate (Y/N)	Commercial status	Remarks	Reference(s)
Pressure oxidation processes
Sherritt-Gordon	Sherritt International	10–16%	NH3-(NH4)2SO4-air	801–1101 kPa	70–95°C	Y	In operation at Fort Saskatchewan (Canada) and Kwinana (WA)	Both refineries no longer treat flotation concentrates	Deng (1993); Forward (1953); Forward and Mackiw (1955)
CESL	Teck Resources Limited	2–20%	H2SO4-O2 +8–12 g/L chloride & 5–10 g/L Cu(II)	1481 kPa	115–175°C	N	Has not been commercially deployed for processing Ni concentrates	Feed ground to a P80 of 45 µm	Defreyne et al. (2006); Jones (1996); Jones, Hestrin and Moore (1998); Jones, Mayhew and O’Conner (2009); Jones et al. (2010); Jones, Moore and Stocker (2003)
Activox®	Western Minerals Technology Pty Ltd	4–12%	H2SO4-O2 +4 g/L chloride	500–1100 kPa	80–110°C	N	Demonstrated at the Tati Nickel Project (Botswana), full scale plant never built	POX preceded by ultra-fine grinding (P80 of 5–15 µm)	Adams and Johnson (2001), Corrans, Johnson and Angove (1993); Johnson and Evans (1998), Johnson, Evans and Turner (2000); Palmer and Johnson (2005); Robles, Cronje and Nel (2009)
PLATSOL^TM	International PGM Technologies Ltd	3–15%	H2SO4-O2 +5–20 g/L NaCl	790 kPa	225°C	Y	Developed to treat polymetallic concentrates from Duluth complex (Minnesota); has not been commercially deployed	Flotation concentrate ground to P80 of 15 µm	Dreisinger et al. (2005); Ferron and Fleming (2001); Ferron and Fleming (2002); Ferron et al. (2000, 2002); Fleming et al. (2000)
Kell Process	Lifezone Limited	–	H2SO4-O2	500–3000 kPa	170–230°C	Y	Process developed to treat PGM-bearing sulfide concentrates; has not been commercially deployed	PGM-enriched leach residue processed further via a roast-chlorine leach process	Adams, Liddell and Holohan (2011); Adams, Liddell, and Smith (2015); Liddell (1999); Liddell and Adams (2012, 2013); Liddell et al. (2011, 2019)
Voisey’s Bay Process	Vale Limited	5–25%	1^st stage; Cl2-O2 2^nd stage; H2SO4-O2	1^st stage – atmospheric 2^nd stage – 601–1101 kPa	1^st stage; 40–90°C 2^nd stage; 100–150°C	N	In operation at Long Harbor (Canada)	Concentrate ground to 20 µm prior to leaching	Crundwell et al. (2011b); Kerfoot et al. (2001); Stevens et al. (2009)
HIKO	Outokumpu	5–7%	H2SO4-O2	601 kPa	<110°C	N	Developed to treat Hitura concentrate (Finland)	–	Nyman et al. (1992), Berezowsky (2000)
IGO Process	IGO Limited	–	H2SO4-O2 +1–5 g/L chloride	800–1200 kPa O2 pressure (Low temp. POX) 300 kPa O2 pressure (High temp. POX)	100–120°C (Low temp. POX) T > 200°C (High temp. POX)	N (Low temp. POX) Y (High temp. POX)	Has not been commercially deployed	Main feature of IP is a method for production of nickel sulfate hydrate from leach liquor via SX. Ultra-fine grinding for low temperature POX	Grassi et al. (2019)
Bioleaching processes
BioNIC®	BHP Limited	3–15%	Mesophiles in acid-ferric sulfate media using air as the oxidant	Atmospheric	30–45°C	N	Successfully demonstrated but not commercially deployed	Stirred tank bioleach process	Cole et al. (1997); Dew et al. (1999); Gilbertson (2000); Heinzle, Miller and Nagel (1999); Norton, Coetzee, and Barnett (1998); Rorke, Basson, and Miller (2001)
BacTech	BacTech Limited	3–7.5%	Thermophiles in acid-ferric sulfate media using air as the oxidant	Atmospheric	45–55°C	N	Successfully demonstrated but not commercially deployed	Stirred tank bioleach process	Fewings and Wilathgamuwa (1998); King and Budden (1996); Miller et al. (1999); Miller (1997); Taylor, Fairley, and Winby (1997)
BioHeap^TM	BioHeap Limited (subsidiary of Western Areas Limited)	–	Proprietary bacterial culture in a sulfate medium above pH 2	Atmospheric	15–95°C (Temp. range microbes are tolerant to)	N	Process is in use at Western Areas Limited’s Cosmic Boy nickel concentrator to treat flotation cleaner tailings	Bacterial cultures can be used for heap, column and stirred tank leaching of whole ores and concentrates	Fewings and Seet (2012); Fewings and Seet (2014); Fewings et al. (2015, 2016)
Atmospheric pressure chemical leaching processes
HydroNic	Outotec	13–32%	1^st stage; CuCl2-HCl-brine 2^nd stage; HCl-brine-O2 or Cl2	Atmospheric	90–110°C	N	Has not been commercially deployed	Either O2 or Cl2 can be used as the oxidant in the second stage leach	Haavanlammi et al. (2006), Karonen, Tiihonen, and Haavanlammi (2009)
Intec Nickel Process	Intec Limited	3–6%	1st stage; H2SO4-brine-air 2nd stage; brine-BrCl2-	Atmospheric	1^st stage; 80–90°C 2^nd stage; 100°C	N	Successfully tested on a Ni-Cu-PGM concentrate, not commercially deployed	Concentrate ground to P80 of 25 µm. Brine in both leach stages contains Cl^− & Br^−	Moyes and Houlis (2003)
Albion Process^TM	Glencore (formerly Xstrata)	–	O2 + H2SO4 ± ferrous and ferric sulfate	Atmospheric	93–98°C	Sulfide oxidation can be controlled by varying pH	Process in use for processing zinc, gold and copper concentrates; has not been commercially applied to the processing of nickel sulfide concentrates	Concentrates ground to P80 of 10–12 µm using an IsaMill^TM	Hourn and Turner (2012); Hourn, Turner, and Holzberger (1996), Stieper (2018)

In the Sherritt-Gordon Process, base metals are dissolved from the sulfide mineral particles by reaction with oxygen, water and ammonia; ferrous iron is oxidized and precipitated as hydrous ferric oxide (Forward 1953; Forward and Mackiw 1955). Sulfidic sulfur (S^2-) is oxidized to thiosulfate (S₂O₃^2-) which then undergoes further oxidation by dissolved oxygen in solution to ultimately form sulfate (SO₄^2-) and sulfamate (NH₂SO₃^2-) via a series of sequential reactions involving intermediates such as trithionate (S₃O₆^2-) (Forward 1953; Forward and Mackiw 1955). Dissolved copper in solution is believed to play a catalytic role in oxidation of thiosulfate to trithionate, sulfate and sulfamate (Forward 1953). Batch leaching tests by Forward (1953) showed that the nickel dissolution rate was influenced by factors such as temperature, oxygen partial pressure and free ammonia concentration; 95% of Ni could be extracted in under an hour by leaching at 220°F (104.4°C). Higher leaching temperatures and oxygen partial pressures, and longer retention times favor the oxidation of thiosulfate and trithionate to sulfate and sulfamate, however, a key aspect of the Sherritt-Gordon process is the retention of a portion of the unsaturated sulfur containing ions in solution for subsequent copper removal (Forward 1953; Forward and Mackiw 1955). Copper removal is achieved by boiling the solution (termed “copper boil”), where Cu(II) ions react with thiosulfate and trithionate in solution, precipitating copper as covellite (CuS) and chalcocite (Cu₂S) (Forward 1953; Forward and Mackiw 1955). Precious and platinum group metals present in nickel matte are reported to be partially extracted during ammoniacal pressure leaching and are subsequently precipitated during the “copper boil” stage, reporting to the copper sulfide precipitate (Wishaw 1993). The copper-free solution produced from the ‘copper boil’ is treated in an autoclave at 245°C and air pressure of 4100 kPa to destroy residual unsaturated sulfur compounds and ammonium sulfamate present in solution (termed oxydrolysis) prior to metallic nickel recovery by pressure hydrogen reduction (Forward 1953; Wishaw 1993). The oxydrolysis step is necessary to prevent sulfur contamination in the final nickel product and to remove sulfamate, which is a herbicide and would not be permissible in the final ammonium sulfate by-product if used as a fertilizer (Wishaw 1993). The final products from the Sherritt-Gordon process are metallic nickel powder or briquettes, cobalt metal powder, briquette or sulfide precipitate, and ammonium sulfate (Forward 1953; Wishaw 1993).

The operations at Fort Saskatchewan and Kwinana have demonstrated the adaptability of the Sherritt-Gordon ammonia pressure leach technology to processing different nickel sulfide feedstocks (flotation concentrates, matte or mixed sulfide precipitate). A significant advantage of the leach process for concentrates is its selectivity toward Ni, Cu and Co, with iron present in pentlandite, chalcopyrite and pyrrhotite being converted to a hydrous ferric oxide residue (Forward and Mackiw 1955). Pyrite present in the concentrate does not react in the ammoniacal solution and reports to the residue; this does however mean that any Ni and Co present in pyrite are not dissolved during leaching (Forward and Mackiw 1955). Other advantages of the Sherritt-Gordon ammonia leach process are that metallic nickel can be recovered directly from solution by hydrogen reduction, and part of the ammonia can be recovered for re-use in the leaching stage. The main disadvantages of the Sherritt-Gordon ammonia leach process are the high cost of ammonia relative to sulfuric acid, though this is partially offset by recovery and recycling, the generation of ammonium sulfate [(NH₄)₂SO₄] as a by-product, which requires disposal if there is no market for it, and cobalt recovery and refining is considerably more complex relative to acid-based leaching processes. Cobalt refining at Fort Saskatchewan involves complex multi-stage processing of cobalt intermediates (MSP or Co(III) hexamine sulfate) via leaching, recrystallization and hydrogen reduction; by contrast cobalt recovery is simpler in acid sulfate systems where cobalt separation from nickel is well established and is achieved by conventional solvent extraction and electrowinning (SX-EW).

3.1.1. Fort Saskatchewan, Canada

The ammonia pressure leach process was originally developed to treat Ni-Cu-Co flotation concentrate from Sherritt-Gordon Mines’ Lynn Lake concentrator in Canada (Deng 1994; Forward 1953; Forward and Mackiw 1955; Kerfoot 1989). Once the Lynn Lake Mine became exhausted (mid 1970s) the refinery continued to operate, processing pentlandite concentrates from across Canada and nickel mattes from overseas (Kerfoot and Cordingley 1997). The Fort Saskatchewan nickel refinery continues to operate to this day, still employing ammonia pressure leaching; however, since 1990 mixed Ni-Co sulfide precipitate produced as an intermediate from nickel laterite processing at Moa Bay (Cuba) is the primary feed to this refinery (Kerfoot and Cordingley 1997; Kofluk and Freeman 2006). The flowsheet for the original refinery operation at Fort Saskatchewan for concentrate processing is presented in Figure 3 and involved ammonia pressure leaching in a train of autoclaves operating at a temperature range of 80–95°C and used compressed air as the oxidant at a pressure of 800 kPa (Kerfoot and Cordingley 1997). Nickel, cobalt and copper are leached into solution as their ammine complexes whilst sulfur is oxidized to sulfate, sulfamate and unsaturated sulfur compounds such as trithionate, thiosulfate etc (Forward 1953; Forward and Mackiw 1955). Iron present in the concentrate is oxidized and precipitates as a hydrous ferric oxide residue (Forward and Mackiw 1955). Copper is removed from the leach liquor by precipitation as its sulfide by reaction with unsaturated sulfur compounds present in the liquor (Forward 1953; Forward and Mackiw 1955). An interesting feature of the Sherritt-Gordon ammonia leach process is control of the degree of sulfur oxidation in the autoclaves to ensure there is sufficient unsaturated sulfur compounds present in the liquor for subsequent copper sulfide precipitation. Following on from copper precipitation, further removal of residual unsaturated sulfur compounds and sulfamate in the Ni-Co leach liquor is carried out by boiling in an autoclave (Forward 1953; Forward and Mackiw 1955). The process chemistry of the copper removal and oxydrolysis stages are discussed in detail by Forward (1953) and Forward and Mackiw (1955).

Figure 3. Simplified process flowsheet of the Sherritt-Gordon ammonia pressure leach process for leaching of Ni sulfide concentrates and recovery of Ni via pressure hydrogen reduction; refining of the mixed sulfide precipitate has been omitted. Typical operating conditions of some unit operations in the Sherritt-Gordon process are presented along with overall metal recoveries during processing of a nickel concentrate grading 10–14% Ni. Adapted from Forward (1953), Forward and Mackiw (1955), Kerfoot and Cordingley (1997) and Wishaw (1993).

Metallic Ni powder is produced via hydrogen reduction of the copper-free solution in an autoclave at elevated temperature and pressure and the resultant powder is briquetted (Forward 1953; Forward and Mackiw 1955). The solution from Ni reduction is treated with H₂S to precipitate Ni and Co as a mixed sulfide precipitate (MSP) and the barren solution is then processed further to recover ammonium sulfate as a by-product (Forward 1953; Forward and Mackiw 1955; Mackiw et al. 1958). The MSP product is further processed in a cobalt refinery, the main process steps of which are depicted in Figure 4; leaching is carried out via sulfuric acid pressure oxidation in an autoclave at 121°C and pressure of 790 kPa (absolute) using air as an oxidant (Mackiw et al. 1958). The leach solution is purified to remove dissolved iron by adjusting the pH to 4.9–5.1 using ammonia and sparging air to oxidize any ferrous iron to the ferric state (Mackiw et al. 1958). After iron removal, solid ammonium sulfate is added to increase the concentration of (NH₄)₂SO₄ in solution to the level required for cobalt reduction, and to remove Ni from solution as nickel ammonium sulfate which is recycled to the nickel reduction circuit (Forward 1953; Mackiw et al. 1958). Cobalt is recovered from solution by hydrogen reduction at elevated temperature and pressure in an autoclave to form cobalt powder (Mackiw et al. 1958).

Figure 4. Cobalt recovery from mixed sulfide precipitate (MSP) as originally practiced at Fort Saskatchewan (Canada) prior to 1990. Adapted from Forward (1953) and Mackiw et al. (1958).

The refinery at Fort Saskatchewan has undergone significant modifications to the leaching and cobalt refining circuit due to changes in the feed, with the refinery switching over to the processing of MSP from Moa Bay, Cuba. The MSP grades 55 wt% Ni and contains high concentrations of cobalt (5 wt%) and could not be processed satisfactorily in the original refinery (Kerfoot and Cordingley 1997). Mixed sulfide precipitate from Moa Bay comprises 80% of the feed to the refinery, the balance being nickel matte and nickel sulfate by-product from copper refineries (Kerfoot and Cordingley 1997). Nickel and cobalt are dissolved from the MSP by pressure leaching in aqueous ammonia using air as an oxidant; nickel matte (70 wt% Ni) is dissolved in a separate autoclave train also via ammonia pressure leaching (Kerfoot and Cordingley 1997). The combined Ni-Co leach liquor is subjected to Ni-Co separation by precipitating cobalt as a Co(III) hexamine salt, which has a low solubility in saturated ammoniacal solution, through the addition of anhydrous ammonia (Kerfoot and Cordingley 1997). The Ni-rich solution is processed in the exact same sequence of operations previously described i.e. copper sulfide precipitation, oxydrolysis and hydrogen reduction (Kerfoot and Cordingley 1997). Copper sulfide precipitation is however achieved through the addition of elemental sulfur and SO₂, rather than relying on unsaturated sulfur compounds present in the leach liquor (Kerfoot and Cordingley 1997). The cobalt(III) hexamine sulfate is processed in a separate circuit which involves repulping in water to dissolve any precipitated nickel hexamine sulfate; the purified cobalt(III) hexamine sulfate is re-dissolved in ammonium sulfate solution followed by Co(III) reduction to Co(II) using metallic cobalt powder and finally hydrogen reduction in an autoclave at high temperature and pressure to produce high-grade metallic cobalt powder (Kerfoot and Cordingley 1997). A portion of this metallic cobalt powder is recycled to the Co(III) reduction stage with the balance being briquetted (Kerfoot and Cordingley 1997).

3.1.2. Kwinana, Western Australia

The Sherritt-Gordon ammonia pressure leach process is employed at BHP Nickel West’s refinery in Kwinana, WA (formerly owned by WMC Resources) and was originally built to process nickel flotation concentrates from Kambalda. The plant was commissioned in 1970, however, with the commissioning of the Kalgoorlie nickel smelter in 1973, co-processing of matte and concentrate commenced, with the refinery switching to an all-matte feed in 1985 (Deng 1994; Wishaw 1993). Some advantages of switching to all matte feed were stated to be higher Ni tenors in the leach liquor, an increase in Ni throughput, reduced cooling requirements in the autoclaves due to the lower sulfur content of the matte feed, and generation of a smaller volume of leach residue (Deng 1993, 1994; Wishaw 1993). The operating conditions of the autoclave train at the Kwinana refinery processing flash furnace smelter matte grading 66 wt% Ni (ground to a P₉₀ of 106 µm), are a temperature range of 85–95°C and a pressure of 750–1000 kPa using compressed air as the oxidant (Wishaw 1993; Woodward 2014; Woodward and Bahri 2007). The Ni-Cu-Corich ammoniacal solution is processed in a similar manner to the original Sherritt-Gordon process at Fort Saskatchewan which was previously described (refer to Figure 3). The primary difference is that further processing of the MSP product for recovery of cobalt values does not take place (Wishaw 1993). The refinery at Kwinana has undergone an expansion with the addition of a plant to produce 100,000 tonnes per annum of nickel sulfate (Anon, 2021) , a key ingredient in LIB cathodes used in electric vehicles.

3.2. Sulfuric acid pressure oxidation processes

Sulfuric acid pressure oxidation processes have been commercially employed since the 1980s for processing of Ni-Cu mattes, zinc concentrates and refractory gold concentrates (Berezowsky et al. 1991) and have seen limited use in the processing of nickel sulfide concentrates. The sulfuric acid required for oxygen pressure leaching can be introduced through recycling of raffinates from solvent extraction-electrowinning circuits and/or can be generated in-situ during leaching from hydrolysis of ferric iron and oxidation of part or all the elemental sulfur. Oxygen pressure leaching processes are characterized according to the operating temperature at which leaching is performed which has an effect on sulfur speciation and oxygen consumption during leaching; low-temperature oxidation (100–110°C) takes place below the melting point of elemental sulfur (113–119°C), intermediate temperature oxidation (120–180°C) takes place above the melting point of elemental sulfur but below a temperature at which majority of the sulfur is further oxidized to sulfate whilst high-temperature pressure oxidation is typically carried out at 180–220°C where sulfur is fully oxidized to sulfate (Berezowsky 2000; Honey, Muir, and Hunt 1997). Further details of the process chemistry during oxygen pressure leaching of nickel sulfide concentrates are discussed by Berezowsky (2000). Sulfuric acid pressure oxidation processes for the treatment of nickel flotation concentrates are by far the most numerous and advanced with respect to pilot plant development and commercial deployment. The key features of the piloted processes are presented in Table 5. Several nickel processing operations have practiced pressure oxidation which are presented in Table 4. Two of these operations have been closed (Fredericktown and Garfield, both in the USA) but have been briefly described for historical purposes.

3.2.1. Piloted pressure leach processes

3.2.1.2 Activox®

The Activox® process was originally developed by Western Minerals Technology (WMT) for oxidative leaching of base metal sulfide concentrates (Johnson, Evans, and Turner 2000) and has also been applied to treatment of refractory gold ores (Corrans, Johnson, and Angove 1993). The Activox® process is essentially a low temperature (80–110°C), pressure oxidation process utilizing oxygen gas as the oxidant, whereby base metals are leached from the concentrate whilst sulfide is oxidized to elemental sulfur, thus reducing the oxygen requirement for leaching versus high temperature, high pressure oxidation processes (Corrans, Johnson, and Angove 1993; Johnson, Evans, and Turner 2000). Elemental sulfur reports to the leach residue and since the leaching process operates below the melting point of sulfur (approximately 120°C), there is no need for the addition of a sulfur dispersing agent during leaching. The leaching media is an acid-sulfate system, with the sulfuric acid being generated in situ by oxidation of sulfide minerals in the feed (Johnson, Evans, and Turner 2000). The concentrate to be leached is subjected to ultra-fine milling down to a P₈₀ of 10 µm, thereby improving leaching kinetics in subsequent pressure oxidation (Corrans, Johnson, and Angove 1993; Johnson, Evans, and Turner 2000; Palmer and Johnson 2005). Chloride addition during Activox® leaching has been stated to be beneficial for copper extraction (Adams and Johnson 2001) which is consistent with the findings by Subramanian and Ferrajuolo 1976) and McDonald and Muir (2007). Typical residence times for leaching nickel sulfide concentrates are reported to be 1–2 hours (Johnson, Evans, and Turner 2000), chalcopyrite being refractory requires longer times (up to 5 hours) (Corrans, Johnson, and Angove 1993). Precious metals are not leached in the process and report to the leach residue from which they are recovered by conventional means (Johnson, Evans, and Turner 2000; Palmer and Johnson 2005). Iron is initially leached from the concentrate but undergoes hydrolytic precipitation, forming goethite and hematite (Corrans, Johnson, and Angove 1993; Palmer and Johnson 2005). There is no guarantee of goethite formation and the study by McDonald and Muir (2007) on chalcopyrite leaching under Activox® conditions (108°C 700 kPa O₂ pressure) found that goethite formation only occurred during leaching with 10 g/L Cl⁻. Leaching with sulfuric acid and chloride resulted in formation of natrojarosite and poorly crystalline iron oxyhydroxides such as 2-line ferrihydrite (McDonald and Muir 2007). Metal recoveries during the leaching of a range of sulfide concentrates using the Activox® process have been reported to be greater than 95% for Ni and Co, and 75–90% for Cu, with 40–80% conversion of sulfidic sulfur to elemental sulfur (Palmer and Johnson 2005). Activox® leaching technology has been applied to the treatment of a variety of nickel sulfide concentrates from Western Australia and Africa (Corrans, Johnson, and Angove 1993; Johnson, Evans, and Turner 2000), and was the chosen leaching technology for the Yakabindie Nickel Project, Western Australia, and the Tati nickel project, Botswana (Johnson, Evans, and Turner 2000; Palmer and Johnson 2005; Robles, Cronje, and Nel 2009). The technology has not been applied commercially for the processing of nickel flotation concentrates.

3.2.1.2. CESL process

The CESL process was originally developed by Cominco (now Teck Resources Limited) for the hydrometallurgical processing of copper sulfide concentrates, as an alternative to smelting, but was also adapted for the processing of nickel sulfide concentrates (Jones, Hestrin, and Moore 1998). The CESL process is a medium temperature pressure oxidation process taking place under an oxygen pressure of 1379 kPa (gauge) and temperature above the melting point of sulfur (120°C) but below 160°C to limit oxidation of elemental sulfur to sulfate. The first step in the CESL process is re-grinding of flotation concentrate(s) in a ball mill to a P₉₅ of 45 µm to liberate the sulfide minerals and improve leaching kinetics (Jones, Hestrin, and Moore 1998; Mayhew et al. 2013). Oxidative dissolution of concentrates takes place in an acidic sulfate solution containing up to 60 g/L H₂SO₄, 12 g/L chloride and 5–10 g/L Cu(II), the role of chloride being to limit oxidation of elemental sulfur to sulfate (Jones, 1996; Jones, Hestrin, and Moore 1998). Brown and Papangelakis (2005) have also shown that chloride ions have a dispersing effect on elemental sulfur and it is possible that chloride in the CESL process serves an additional function as a sulfur dispersing agent during leaching. In the absence of any suitable sulfur dispersing agent, sulfide mineral dissolution kinetics would likely be inhibited due to encapsulation of unreacted sulfides by molten elemental sulfur (Brown and Papangelakis 2005; Tong and Dreisinger 2009). The specific role of copper ions in the CESL process is not stated, however WMC Resources previously recognized the benefits of copper ions in combination with chloride during pressure oxidation of nickel sulfide concentrates which formed the basis for a patent (Crossley et al. 1998). It may be that in the presence of sufficient chloride ions in solution, the Cu(II)/Cu(I) couple functions as a redox mediator during oxygen pressure leaching. Iron sulfides in the concentrate(s) are oxidized and precipitated as hematite; in the case of pyrrhotite, oxidation also produces elemental sulfur whilst sulfide in pyrite is oxidized to sulfuric acid (Defreyne et al. 2006). Whilst only 50% to 75% of the pyrite is reported to undergo oxidation in the CESL process, its presence in concentrates is stated to be undesirable as it increases oxygen consumption and heat generation during pressure oxidation and generates sulfuric acid which must be neutralized downstream (Defreyne et al. 2006). The heat generation due to pyrite oxidation may not be an issue if the process can be operated autogenously whilst the acid generation from pyrite might also not be a problem if a use can be found for the acid such as in a heap leach operation.

Once the base metals have been leached into solution, the autoclave leach slurry is subjected to solid–liquid separation and the leach liquor is neutralized with limestone to remove iron and aluminum from solution (Jones et al. 2010; Mayhew et al. 2013). Copper can be recovered as cathode via solvent extraction and electrowinning (SX-EW) whilst Ni and Co can be recovered as a mixed hydroxide precipitate (MHP), mixed sulfide precipitate (MSP), or separated via SX to produce individual products (Jones et al. 2010; Jones, Mayhew, and O’Conner 2009; Mayhew et al. 2013). The CESL process has been tested on a range of nickel concentrates ranging in Ni grade from 2.1 to 20.4 wt% at bench and pilot plant scale, including concentrates high in MgO arising from the flotation of disseminated Ni sulfide ores which are not suitable for smelting (Jones et al. 2010; Jones, Hestrin, and Moore 1998; Jones, Mayhew, and O’Conner 2009; Mayhew et al. 2013). High Ni and Co recoveries have been reported for a range of concentrates leached using the CESL process, being above 95% for both elements (Jones, Hestrin, and Moore 1998; Jones, Mayhew, and O’Conner 2009). Despite being tested at bench and pilot plant scale, the CESL process has not been deployed commercially for the processing of nickel sulfide concentrates.

3.2.1.3. Platsol^TM

PLATSOL^TM leaching technology was originally developed by International PGM Technologies Limited at SGS Lakefield Research (Canada) for processing bulk polymetallic sulfide flotation concentrates containing Cu, Ni, Co, gold (Au) and PGMs from the NorthMet deposit (Duluth Gabbro, Minnesota), owned by Polymet Mining Corporation (Dreisinger et al. 2005; Ferron et al. 2002; Fleming et al. 2000). The bulk sulfide concentrates were considered uneconomical for smelting and hence a hydrometallurgical process was an attractive alternative to development of a dedicated smelting facility or shipping flotation concentrate to toll smelters (Fleming et al. 2000). The leaching technology was also tested on a range of different base metal sulfide concentrates containing a range of Ni grades from 3 to 15 wt% as well as Co, Cu and PGMs (Ferron et al. 2000, 2002; Fleming et al. 2000). The leaching process is essentially a high-temperature pressure oxidation process carried out at 225°C utilizing oxygen gas as the oxidant in sulfuric acid media containing dissolved chloride (Dreisinger et al. 2005; Ferron and Fleming 2001); base and precious metals are dissolved from concentrates under the conditions employed and total oxidation of sulfidic sulfur to sulfate takes place (Fleming et al. 2000). Most of the iron (>90%) in the concentrate reports to the leach residue as hematite (Ferron et al. 2000; Fleming et al. 2000). Key features of the process are the addition of an ultra-fine milling stage (P₈₀ 15–20 µm) and chloride addition (5–20 g/L NaCl) which facilitates dissolution of Au and PGMs during leaching via formation of soluble chloro complexes (Ferron et al. 2000; Fleming et al. 2000). Base metals recovery during leaching was reported to be unaffected by fine grinding or chloride addition, however recoveries of Au, Pt and Pd were improved both when ultra-fine milling and chloride addition were employed (Fleming et al. 2000). Gold recoveries during leaching were also reportedly affected by the grinding media, with ceramic media giving the best results during leaching versus steel grinding media, which was believed to be due to cementation of dissolved gold onto iron powder produced during attrition grinding using steel media (Fleming et al. 2000). At the high temperatures employed, leaching kinetics were very rapid with more than 95% of the Ni, Cu and Co being leached from concentrates in the first 20 minutes with longer retention times (up to 180 min) being necessary to dissolve the precious metals (Dreisinger et al. 2005; Ferron and Fleming 2002; Ferron et al. 2000). A demonstration plant utilizing PLATSOL^TM leaching technology was built at SGS Lakefield for processing bulk sulfide flotation concentrate produced from the NorthMet deposit (Ferron and Fleming 2001, 2002); however, the leaching technology does not appear to have been industrially implemented.

3.2.1.4. Kell process

The Kell process is a multi-step hydrometallurgical process that has been developed for the extraction of base metals, gold, silver and PGMs from flotation concentrates as an alternative to smelting. The first stage in the Kell Process is high-temperature pressure oxidation (170–230°C) in a sulfate medium using oxygen gas to dissolve base metals and oxidize sulfidic sulfur to sulfate; precious and platinum group metals are retained in the leach residue (Liddell and Adams 2012; Liddell et al. 2019). By comparison, PGMs are intentionally dissolved in the PLATSOL^TM process, also utilizing high-temperature pressure oxidation, via the addition of 5–20 g/L NaCl. The leach liquor arising from the pressure leaching stage (Kell process) is processed to recover base metals whilst the leach residue is further processed to recover Au, Ag and PGMs (Liddell and Adams 2012; Liddell et al. 2019). The leach residue may be subjected to a HCl leach at ambient pressure to dissolve silver and residual base metals retained in the residue followed by roasting of the residue at 500–1000°C to convert PGM minerals to a readily leachable form (Liddell et al. 2019). Test work performed by Adams, Liddell, and Holohan (2011) found that reducing conditions during roasting are ideal for subsequent PGM recovery. Recovery of precious and platinum group metals from the roasted leach residue takes place via an ambient pressure oxidative leach in a chloride medium using chlorine gas as an oxidant (Liddell et al. 2019). The process has been tested on a range of PGM bearing sulfide concentrates, primarily from Southern Africa and average base metals (Ni, Cu, Co) recoveries are 96–98% whilst Pt and Pd recoveries are 98% (Liddell et al. 2019).

3.2.1.5. IGO process^TM

IGO Limited, an ASX listed nickel miner, has recently patented a process for production of high purity/battery grade nickel sulfate directly from nickel sulfide concentrates via oxygen pressure leaching, solution neutralization and purification, and solvent extraction to produce nickel sulfate hydrate crystals (Grassi et al. 2019). The main feature of the invention is the stripping of nickel from the loaded organic phase using a sulfuric acid solution saturated in Ni such that nickel sulfate hexahydrate crystals are immediately precipitated from solution (Grassi et al. 2019). Nickel dissolution from the concentrate is proposed to take place either via low temperature (100 – 120°C) or high temperature (>200°C) pressure oxygen leaching. For the former case, ultra-fine grinding (P₈₀ 6.6 µm) improved nickel leaching from the concentrate (Grassi et al. 2019) just like in Activox®. Grassi et al. (2019) reported near complete extraction of base metals within 60 min during high-temperature pressure leaching; however, the resultant leach liquors contained high concentrations of dissolved iron (up to 22 g/L total Fe, mostly as Fe(III)) and free acid (up to 78 g/L H₂SO₄) which would increase limestone consumption in downstream neutralization. The high iron and acid concentrations in the leach liquor is not necessarily an issue if a use can be found for them, such as leaching nickel sulfide flotation tailings or nickel oxide ores, though this is not mentioned in the invention. The IGO Process^TM has not been commercially deployed; however, IGO Limited has indicated that they are assessing opportunities for partnership/collaboration to leverage the technology (IGO Limited 2020).

3.2.2. Commercial operations

3.2.2.1. Garfield Cobalt Refinery, USA

The Garfield cobalt refinery in Utah, USA, operated by Calera Mining Company was the first application of sulfuric acid pressure oxidation to the processing of nickel-bearing sulfides (Berezowsky 2000). The plant was operational from 1953 to 1959 and shut down due to a shortage of ore (Meddings and Evans 1971). The plant processed a sulfide concentrate from the Blackbird Mine in Idaho (USA) grading 17.5% Co, 1.0% Ni and 0.5% Cu; major sulfide minerals in the concentrate were cobaltite (CoAsS) and pyrite with minor gersdorffite, chalcopyrite and arsenopyrite (Berezowsky 2000). The pressure leaching process in use at the refinery involved total pressure oxidation at 375°F (190.6°C) and a total pressure of 3600 kPa using air as the oxidant in a lead and brick lined autoclave (Berezowsky 2000). Under these conditions, 95–97% of the Co was leached (Berezowsky 2000); the extraction efficiencies for Ni and Cu have not been stated. Majority of the iron and arsenic in the concentrate was precipitated as basic iron sulfate and ferric arsenate (Meddings and Evans 1971). The leach slurry from the autoclave was neutralized to remove residual arsenic and iron, and copper was recovered by cementation with cobalt powder (Meddings and Evans 1971). Cobalt and Ni recovery was via pressure hydrogen reduction to produce an impure product after adjustment of the solution with ammonia to produce the desired ammonia to metal ratio in solution (molar ratio of 2) for reduction (Meddings and Evans 1971). Construction and operation of the plant was rushed, and no piloting of the process occurred, as such the plant was plagued with numerous issues due to the abrasive and corrosive nature of the slurry (Berezowsky 2000).

3.2.2.2. Fredericktown metals refinery, USA

The Fredericktown refinery in Missouri, USA, operated by National Lead Company was the second commercial plant to utilize sulfuric acid pressure oxidation in the processing of nickel-bearing sulfide concentrates (Berezowsky 2000). The plant was operational between 1954 and 1960 and like the Garfield refinery was closed due to lack of feed (Meddings and Evans 1971). The feed to the leach plant was a flotation concentrate grading 5.0% Ni, 4.2% Co and 4.8% Cu, as siegenite and chalcopyrite, produced as a by-product of copper-lead mining (Berezowsky 2000; Meddings and Evans 1971). The pressure leaching process in use at the refinery involved total pressure oxidation at 232°C and a total pressure of 5300 kPa using air as the oxidant (Berezowsky 2000). Most of the iron was precipitated in the autoclave as basic iron sulfate and further removal of iron post-pressure leaching took place through pH adjustment and aeration to oxidize residual ferrous iron (Meddings and Evans 1971). Metallic Ni, Cu and Co products were recovered by pressure hydrogen reduction (Meddings and Evans 1971).

3.2.2.3. Kokkola refinery, Finland

There has been a long history of Ni/Co refining at Kokkola by Outokumpu Oy, starting with Ni and Co extraction from pyrite-pyrrhotite concentrate from the Outokumpu copper mine via sulfation roasting between 1966 and 1987 (refer to Section 3.6.3). From December 1991, the Kokkola refinery began processing a low grade, high magnesia (10% MgO) nickel sulfide flotation concentrate from the Hitura mine (Finland) grading 5–7% Ni. The plant used a pressure oxidation process (HIKO process – refer to Table 5) in autoclaves at temperatures below the melting point of sulfur (<120°C) with oxygen gas at 500 kPa pressure (Berezowsky 2000; Nyman et al. 1992). Additional details of the HIKO process such as acid addition to the leach charge, concentrate feed particle size, leaching time and degree of sulfide conversion to elemental sulfur are unknown (Berezowsky 2000). The leach solution was subjected to iron precipitation using limestone followed by copper precipitation as a sulfide using H₂S (Nyman et al. 1992). Following on from iron and copper removal, the solution was subjected to SX using di-2-ethylhexyl phosphoric acid (D2EHPA) to remove calcium followed by separation of cobalt (and magnesium) from nickel using Ionquest 801, an organophosphorus-based extractant (Nyman et al. 1992). The raffinate from Co/Mg SX was then sent to Ni SX; the final products were purified Ni and Co chemicals (Nyman et al. 1992); no details however are given of the Ni SX circuit i.e. type of extractant used, nor the Co refining circuit after Co/Mg SX. The status of the pressure leaching plant is unknown though the refinery has changed ownership a number of times and is currently a cobalt refinery operated by Umicore (Cooke 2021; Törmänen and Tuomela 2021) and the nickel mine that supplied concentrate to the leaching plant has been closed since 2015 (Björkman 2019). A simplified process flow diagram of the HIKO process used at the Kokkola refinery is presented in Figure 5.

Figure 5. Simplified process flow diagram of the Outokumpu HIKO process used at the Kokkola refinery (Finland) for processing of nickel sulfide concentrates via low temperature pressure oxidation and recovery of nickel and cobalt chemicals. Operating conditions of the leaching process and metal extractions during leaching for a given nickel concentrate are presented. Adapted from Nyman et al. (1992) Honey, Muir, and Hunt (1997) and Berezowsky (2000).

3.2.2.4. Long Harbour, Canada

Hydrometallurgical processing of nickel flotation concentrate is carried out by Vale Limited at Long Harbor using nickel concentrate produced from the Voisey’s Bay nickel deposit; the first shipment of metallic nickel produced from the plant occurring in 2015 (Anon 2015). A simplified flowsheet of the Voisey’s Bay Process is presented in Figure 6 and the key features of the leaching stages are summarized in Table 5. The process commences with regrinding of the flotation concentrate to below 20 µm followed by a first stage leach in sulfuric acid (return electrolyte from Ni-electrowinning) at ambient pressure using a mixture of chlorine gas and oxygen as the oxidant (produced from the Ni-electrowinning circuit). This is followed by a second stage leach at elevated pressure using oxygen as the oxidant in an autoclave (Crundwell et al. 2011b; Kerfoot et al. 2001; Stevens et al. 2009). The clarified leach solution produced from the two-stage leach process undergoes a first stage of iron removal where 95% of the iron is removed by raising the pH to 3 using limestone and aerating with oxygen to oxidize any ferrous iron in the solution to the ferric state (Stevens et al. 2009). Following on from the first iron removal stage, copper is recovered from the solution via solvent extraction-electrowinning (SX-EW) using LIX 84, a commercially available hydroxyoxime extractant (Stevens et al. 2009). The raffinate from Cu SX undergoes a second stage of iron removal, followed by cadmium removal (by sulfide precipitation) and finally an impurity SX stage using D2EHPA to remove calcium, magnesium and other impurities (Fe, Pb, Zn, Mn) prior to cobalt solvent extraction (Crundwell et al. 2011b; Stevens et al. 2009). The purified raffinate is then subjected to Co SX-EW using Cyanex 272, an organophosphinic acid extractant. The raffinate from Co SX is sent to the Ni electrowinning circuit; the final products produced at Long Harbor are nickel cathode, copper cathode and cobalt rounds (Crundwell et al. 2011b; Stevens et al. 2009).

Figure 6. Simplified process flow diagram for the hydrometallurgical processing of nickel concentrate from Voisey’s Bay (Canada) at Vale’s Long Harbour operations to produce Ni, Cu and Co cathode via a combination of pressure oxidation, solvent extraction (SX) and electrowinning (EW). Some unit operations such as hydroxide re-leaching after solid/liquid separation and cadmium removal prior to impurity SX have been omitted for simplicity. Adapted from Crundwell et al. (2011b), Kerfoot et al. (2001) and Stevens et al. (2009). Conditions utilized in atmospheric pre-leach and pressure oxidation and extraction efficiencies are based on the original patent filed by Kerfoot et al. (2001) and may not reflect actual conditions in current use.

3.2.2.5. Nadezhdin mill, Russia

Flotation of massive sulfide ores from the Noril’sk-Talnakh ore field (Siberia) produces three products, copper concentrate, nickel concentrate and pyrrhotite concentrate (Bulatovic 2007; Kozyrev et al. 2002). The pyrrhotite concentrates produced typically contain 2–5% Ni, 1–4% Cu and up to 15 g/t of PGMs (Berezowsky 2000; Deng 1992, 1993; Klyushnikov et al. 2021). Nickeliferous pyrrhotite concentrates are processed to recover the base and platinum group metals content via a combination of autoclave pressure leaching and residue flotation at the Nadezhdin mill, operated by Noril’sk Mining and Metallurgy Corporation (now Nornickel) (Deng 1992, 1993). The autoclaves were commissioned in 1979 and involved leaching pyrrhotite concentrates at 125–135°C and pressure of 1.5 MPa, using 80% O₂ as the oxidant with addition of 4 kg/t of lignin sulfonic acid salts as a sulfur dispersing agent (Berezowsky 2000; Deng 1992, 1993). The leaching process results in oxidation of 96% of the pyrrhotite in the concentrate and conversion of 65% of sulfidic sulfur in the feed to elemental sulfur; 67% of the Ni is initially leached into solution but is subsequently precipitated from the leached slurry as a sulfide (Deng 1992, 1993). Precipitation of dissolved base metals is carried out by treating the leach pulp with metallic iron at 90°C which together with elemental sulfur present in the leach slurry precipitates the dissolved metals as their sulfides (Berezowsky 2000).

The precipitated base metal sulfides along with unreacted sulfides and PGMs are separated from elemental sulfur and iron oxides via sequential flotation (Deng 1992, 1993). Firstly, flotation is carried out to reject iron oxides formed during pressure oxidation and recover an elemental sulfur-metal sulfide flotation concentrate (Berezowsky 2000). The concentrate is then treated in an autoclave at 110–130°C and pH 9–10 with addition of 5–7 g/L Na₂S to disintegrate the concentrate into a mixture of elemental sulfur and metal sulfides; the pulp is then subjected to flotation to produce a sulfide concentrate and an elemental sulfur product (Berezowsky 2000). The final nickel grade in the concentrate is 8–10% with an overall recovery of base and platinum group metals to the final concentrate of 80–85% (Berezowsky 2000). The base and platinum group metal enriched product is then combined with the nickel concentrate produced from the beneficiation circuit and processed further via flash smelting (Berezowsky 2000; Klyushnikov et al. 2021). A simplified process flow diagram is presented in Figure 7. The pressure oxidation autoclaves at the Nadezhdin mill do not have sufficient capacity to process all the nickeliferous pyrrhotite concentrate produced and as such, a large proportion of the concentrate is diverted to stockpile (Klyushnikov et al. 2021; Moskalyk and Alfantazi 2002).

Figure 7. Simplified process flow diagram for the hydrometallurgical processing of nickeliferous pyrrhotite concentrates from Talnakh (Russia) at the Nadezhdin mill via a combination of pressure oxidation, non-ferrous metal sulfide precipitation and flotation to recover a sulfide concentrate for flash smelting and elemental sulfur. Operating conditions of major unit operations and overall nickel recovery and final nickel product grade during processing of a pyrrhotite concentrate grading 2.2% Ni are presented (from Berezowsky 2000). Adapted from Kitay, Mechev, and Volkov (1991) and Berezowsky (2000).

3.3. Atmospheric sulfuric acid leaching

The only processes developed to date employing chemical leaching of nickel concentrates in a sulfate medium at atmospheric pressure are the Intec Nickel Process developed by Intec Limited, and the Albion Process^TM developed by Glencore (formerly Xstrata). Neither of these processes have been deployed commercially for the processing of nickel sulfide concentrates though the Albion Process^TM has been commercially deployed for processing gold, copper and zinc concentrates and may therefore be suitable for leaching of nickel flotation concentrates. The key features of both processes are presented in Table 5 and are briefly described below.

3.3.1. Intec Nickel Process

The Intec Nickel Process is a two-stage atmospheric leaching process developed for processing base metal sulfide concentrates containing precious metals and PGMs. Much of the development work for the Intec Process focused on copper extraction from copper sulfide concentrates (Moyes and Houlis 2003; Palmer and Johnson 2005). The first stage leach involves carrying out oxidative dissolution of reground flotation concentrate (ground to P₈₀ 25 µm) at 80°C in a sulfuric acid solution containing halides (Cl-, Br-) using air (21% O₂) as an oxidant (Moyes and Houlis 2003). Nickel, copper and cobalt dissolution efficiencies in the first leaching stage were reported to be greater than 95%, with approximately 50% of the Au and PGMs dissolving in this leaching stage (Moyes and Houlis 2003); the residence time for the first leaching stage was not stated. The second leaching stage, termed the “Halex leach”, involves the extraction of un-leached Au, Pt, Pd and remaining base metals from the first stage leach residue. Leaching is carried out at 100°C using spent electrolyte from the Cu- and Ni-electrowinning circuits, where electrogenerated BrCl₂- serves as the oxidant (Moyes and Houlis 2003). Base metal recoveries in the second stage leach were again reported to be greater than 95%, with 86% of the Au, and 87% of the Pd also being leached; Pt recoveries were poor however with only 54% extraction (Moyes and Houlis 2003). Recovery of precious metals post-leaching is by precipitation using either sodium hydrogen sulfide (NaHS) or pyrrhotite-rich nickel concentrate, and copper and nickel are recovered in separate electro-winning circuits as cathode.

3.3.2. Albion Process^TM

The Albion Process^TM is an atmospheric oxidative leaching process which is preceded by an ultra-fine grinding stage. The process was developed to treat a range of sulfide concentrates containing base and precious metals and was originally developed by MIM Holdings/Xstrata (Hourn and Turner 2012; Hourn, Turner, and Holzberger 1996) with Glencore Technology (a subsidiary of Glencore) being the licensor of the process/technology after the purchase of Xstrata. The original patent for the Albion Process^TM covered the ultra-fine grinding of base metal sulfide concentrate or ore to a P₈₀ of 20 µm or less followed by leaching in a sulfuric acid medium containing dissolved ferric and ferrous iron with sparging of air, oxygen-enriched air or pure oxygen at temperatures between 60°C and the boiling point of the solution (Hourn, Turner and Holzberger 1996). The first step in the Albion Process^TM is the fine grinding of the sulfide-bearing material in an IsaMill^TM to a P₈₀ of 10–12 µm followed by leaching of the milled material in specially designed agitated leaching tanks at 93–98°C with oxygen injection using Hypersparge™ technology which can deliver oxygen at supersonic velocities (450–550 m.s⁻¹) (Hourn and Turner 2012). No external heating or cooling requirements are required for the leaching process, the heat being generated from oxidation of sulfidic sulfur and cooling being achieved via water evaporation. Sulfidic sulfur is oxidized to elemental sulfur and sulfate, with the degree of elemental sulfur oxidation to sulfate being controlled by the pulp pH. Processing of base metal sulfide concentrate via the Albion Process^TM is carried out under acidic conditions whilst the processing of refractory gold concentrates is carried out under neutral conditions (pH 5–7) to oxidize elemental sulfur to sulfate. The Albion Process^TM has been commercially deployed for the processing of sphalerite, refractory gold concentrates (Hourn and Turner 2012) and most recently, chalcopyrite concentrate at Sable Zinc (Zambia) (Stieper 2018). The process has not been industrially applied to the processing of nickel sulfide concentrates and information on processing of nickel sulfide concentrates is limited; the original patent by Hourn, Turner, and Holzberger (1996) only had one application of the Albion Process^TM to the processing of a low-grade nickel concentrate (1.7% Ni as pentlandite) which was milled to a P₈₀ of 5 µm prior to leaching at atmospheric pressure.

3.4. Chloride-based leaching processes

Chloride leaching involves the dissolution of base metal sulfide concentrates, ores or mattes exclusively in chloride media as opposed to sulfate-media with the addition of halide ions such as in the CESL, Voisey’s Bay or Intec processes which were previously described. Advantages of chloride lixiviants have been stated to be a high recovery of metals during leaching, ability to regenerate and recycle the lixiviants, inertness of pyrite in chloride media and reduced formation of sulfate during leaching (Dutrizac 1992). Against these advantages however are several limitations such as the lack of selectivity of chloride lixiviants during leaching which can complicate downstream liquor purification and lixiviant regeneration/recycling and the corrosivity of chlorides toward many common materials of construction (Dutrizac 1992; Forward and Warren 1960; Honey, Muir, and Hunt 1997). Despite the disadvantages, chloride-based leaching processes have been successfully utilized for processing nickel mattes with these processes in current use at the Falconbridge Nikkelverk refinery (Norway), the Niihama nickel refinery (Japan) and Sandouville-Le Havre (France) (Kerfoot and Weir 1988; Makino et al. 1996; Stensholt, Zachariasen, and Lund 1986). A common theme to all three matte leach processes is the use of chlorine gas as an oxidant in conjunction with either Cu(II) (Falconbridge, Niihama) or Fe(III) (Sandouville) (Muir and Ho 2006).

Despite the successful application of chloride leaching in nickel matte refining, the only chloride-based leaching process developed at scale for processing nickel sulfide concentrates is the HydroNic process developed by Outotec Oyj (Karonen, Tiihonen, and Haavanlammi 2009) and key details of the process are summarized in Table 5. HydroNic employs a two-stage leaching process to dissolve Ni, Co and Cu from base metal sulfide concentrates. In the first stage, nickel is leached from pentlandite by reaction with copper(II) chloride dissolved in a sodium chloride solution recycled from the second leach stage, with Ni and Fe being leached into solution (as their chlorides) whilst copper is precipitated as copper(I) sulfide (Cu₂S). This leaching stage is non-oxidative and is a cementation process. In the second stage of leaching, the concentrate is leached under oxidative conditions using oxygen or chlorine gas as the oxidant. Iron is hydrolyzed and precipitated from solution during leaching whilst sulfide in the concentrate is oxidized to elemental sulfur and sulfate. The solution from the second leaching stage is recycled to the first stage to remove copper from solution via cementation as Cu₂S. After copper removal, the solution is treated further with caustic (NaOH) and limestone to precipitate iron (as oxyhydroxide) and sulfate (as gypsum) from solution. After iron/sulfate removal, Ni and Co are precipitated from solution as a mixed hydroxide by neutralization with MgO. The patent for the HydroNic process was filed by Outotec Oyj in 2006 (Haavanlammi et al. 2006). Interestingly, another patent was filed by Outotec Oyj in the same year for a chloride leach process for treatment of nickel sulfide ore and concentrates, key differences in this process were the absence of a non-oxidative leaching step, copper(II) was precipitated from the leach liquor as atacamite (Cu₂(OH)₃Cl) which was returned to the leaching stage and copper recovery after nickel hydroxide precipitation in the form of cuprous hydroxide (CuOH) from which copper metal was recovered via hydrogen reduction (Krebs et al., 2006). The hydronic process has been tested on a range of nickel concentrate grades; however, it does not appear to have been implemented at an industrial scale.

3.5. Bioleaching processes

Bioleaching generally involves the dissolution of sulfide ores/concentrates in an acidic sulfate medium where ferric ions act as an oxidant, oxidizing the sulfide minerals to elemental sulfur with release of metal ions into solution (Watling 2006). The microorganisms in bioleaching serve a dual purpose of regenerating ferric ions by oxidation of ferrous iron and generation of sulfuric acid via oxidation elemental sulfur (Watling 2006). Some reasons for considering bioleaching for the processing of sulfide ores/concentrates stated by Miller (1997) are simple equipment requirements and flexibility of bioleach processes with respect to variation in concentrate/ore head grade and mineralogy (Miller 1997). Potential disadvantages of bioleaching are the slow reaction kinetics and sensitivity of the microorganisms to temperature, mineral pulp density, agitation rate and chloride concentration in process water (Astudillo and Acevedo 2008; Bailey and Hansford 1993; Honey, Muir, and Hunt 1997; Nemati and Harrison 2000). Many of these limitations are dependent on the bacterial strain and in the case of chloride concentration, can be addressed through the development of halotolerant bacterial strains such as in the BioHeap^TM process described below (Section 3.5.2.3). A summary of piloted bacterial leaching processes and commercial operations employing bioleaching for processing of nickel sulfide flotation concentrates is described. After sulfuric acid pressure oxidation, bioleaching processes are second with respect to the number of developed technologies and industrial deployment. The key features of the piloted processes are presented in Table 5 whilst a summary of commercial operations utilizing bioleaching technologies are presented in Table 4.

3.5.1. Piloted processes

Three bacterial leaching processes have been developed for the processing of nickel sulfide bearing material; the processes are BioNIC® (BHP Limited), BacTech (BacTech Limited) and the BioHeap^TM process (BioHeap Limited, a subsidiary of Western Areas Limited). The BioNIC® and BacTech processes are stirred tank bacterial leaching processes that have been successfully piloted on a range of nickel concentrates and a review of these processes and pilot plant operations has been covered by Watling (2008) in a review on the bioleaching of nickel-copper sulfides. Neither of these processes progressed beyond pilot plant scale. The BioHeap^TM process developed by BioHeap Limited, a subsidiary of Western Areas Limited, is in current use at Western Areas Limited’s Cosmic Boy nickel concentrator in Western Australia and the process and operations are described below.

3.5.2. Commercial bioleach operations

3.5.2.1. Talvivaara heap bioleaching operation, Finland

The operation at Talvivaara is the only Ni sulfide heap bioleaching operation, processing a black schist ore from the Sotkamo deposit (Finland), containing 0.23–0.27% Ni, 0.5–0.56% Zn, 0.13–0.14% Cu and 0.02% Co (Crundwell et al. 2011b; Riekkola-Vanhanen 2013; Watling 2008). The principal sulfide minerals in the ore are pyrrhotite, pyrite, sphalerite, pentlandite, violarite and chalcopyrite whilst non-sulfidic gangue phases are graphite, quartz, mica, anorthite and microcline (Riekkola-Vanhanen 2007, 2013; Watling 2008). Ore from the mine is subjected to four stages of crushing to a P₈₀ of 8 mm followed by agglomeration with bacteria and sulfuric acid, stacking of the agglomerates to form heaps and leaching of the heaps with a sulfuric acid solution at pH 1.8 (Crundwell et al. 2011b; Riekkola-Vanhanen 2007, 2013; Watling 2008). Metal values (Zn, Ni, Co, Cu) are recovered from the pregnant leach solution by sequential precipitation as their sulfides using H₂S (Crundwell et al. 2011b; Riekkola-Vanhanen 2013). A simplified process flow diagram of the Talvivaara heap bioleaching operation is presented in Figure 8. The mine is still operational and is currently being run by Terrafame Limited, a Finnish state-owned mining company which purchased the operation from the parent company (Talvivaara Mining Company) after declaring bankruptcy (Sairinen, Tiainen, and Mononen 2017). Recently, Terrafame plans on exploiting another nearby black schist ore body (Kolmisoppi) which is currently undergoing environmental assessment (Törmänen and Tuomela 2021). Terrafame intends to begin production of battery grade Ni and Co sulfates starting from the MSP produced from the heap leach operation, with pressure oxidation of the MSP being the first step in the process (Törmänen and Tuomela 2021). Based on a recent company announcement, production ramp-up of the battery chemicals plant started in June 2021 and first delivery of products to customers in July 2021 (Anon 2021).

Figure 8. Simplified process flow diagram for the hydrometallurgical processing of black schist ore from the Sotkamo deposit at the Talvivaara heap bioleaching operation (Finland) including expected leaching recoveries for the base metals and product grades. a copper product had not been produced at the time of reporting by Riekkola-Vanhanen (2013). Adapted from Riekkola-Vanhanen (2013).

3.5.2.2. Mondo nickel project, Finland

Talc processing operations at Sotkomo and Vuonos in Finland operated by Mondo Minerals (now Elementis) produce nickel sulfide concentrate grading 8.70% Ni as a by-product (Neale et al. 2015). The primary sulfide minerals in order of decreasing abundance in the concentrate are pyrrhotite, pentlandite, pyrite and gersdorffite (Neale et al. 2015). In the past the concentrate was typically sold to toll smelters, however the high arsenic content (1.56% As) had made this option less feasible (Neale et al. 2015). To produce a more valuable product and increase revenue, a hydrometallurgical plant was built to process 12,000 tonnes per annum of concentrate and recover nickel and cobalt as a mixed hydroxide precipitate (Neale et al. 2015); a simplified process flow diagram of the hydrometallurgical process in use is depicted in Figure 9. Nickel concentrate from both operations is combined and processed in a beneficiation circuit where feed is re-ground to a P₈₀ of 30–40 µm followed by magnetic separation to reject pyrrhotite, and flotation to remove talc and magnesite, to produce an upgraded concentrate containing 12.4% Ni (Laukka et al. 2018; Neale et al. 2015, 2016). The nickel concentrate is subjected to a bioleaching process which was developed in conjunction with Mintek; the leaching conditions are a temperature range of 41–49°C pH range of 1.3–1.6 and a pulp density of 15–17% solids (Laukka et al. 2018; Neale et al. 2015). The overall residence time in the bioleaching circuit is 9.5 days and nickel and cobalt extractions are reported to be 87.8% and 90.7%, respectively (Laukka et al. 2018). The leach slurry is neutralized to pH 3.5 using limestone to precipitate iron and arsenic from solution followed by solid–liquid separation (Laukka et al. 2018; Neale et al. 2015). The nickel-rich solution is subsequently treated with magnesia (MgO) slurry to precipitate Ni and Co as a mixed hydroxide precipitate (MHP) (Neale et al. 2016). The raffinate produced from MHP precipitation is treated with lime (CaO) to precipitate residual Ni and the precipitate is recycled to iron-arsenic precipitation circuit to re-dissolve residual Ni (Neale et al. 2016). Törmänen and Tuomela (2021) have indicated that the bioleach plant has only been operational periodically with the nickel sulfide concentrate by-product presumably being sold on the market at other times.

Figure 9. Simplified process flow diagram for the bioleaching of nickel concentrate produced as a by-product of talc mining at Mondo Minerals (now Elementis) Vuonos talc concentrator operations in Finland. the beneficiation circuit has been omitted for simplicity. Operating conditions of major unit operations and final nickel product grade during processing of nickel concentrate grading 10.49% Ni are presented. Adapted from Neale et al. (2015), Neale et al. (2016).

3.5.2.3. Cosmic boy nickel concentrator, Western Australia

The Cosmic Boy nickel concentrator in Western Australia is owned and operated by Western Areas Limited (Fewings and Seet 2014). The concentrator batch processes high grade nickel sulfide ore from the Flying Fox (5.1% Ni grade) and Spotted Quoll (5.6% Ni grade) underground nickel mines in Western Areas Limited’s Forrestania operations (Fewings and Seet 2014; Fewings et al. 2015). The output from the concentrator is sold to nickel smelters; however, the processing of ore from the Spotted Quoll mine presents challenges as the ore has a high arsenic content and the arsenic is present as niccolite and gersdorffite (Fewings and Seet 2014; Fewings et al. 2015). Niccolite and gersdorffite are rejected during flotation using cyanide as a depressant at pH 9.8 to 10.5, though this results in a loss of Ni to the tailings (Fewings et al. 2016). The cleaner tailings arising from processing Spotted Quoll ore grade 18.3% Ni and 6.4% As (Fewings et al. 2016). To recover Ni values from the cleaner tailings, Western Areas Limited has implemented a bioleach process based on proprietary BioHeap^TM technology, whereby Ni is leached from the flotation cleaner tailings whilst As and Fe are rejected as an environmentally stable precipitate (Fewings and Seet 2014; Fewings et al. 2015, 2016).

A key feature of the BioHeap^TM process is the use of proprietary bacterial cultures which are tolerant to arsenic and high concentrations of total dissolved salts (TDS) (up to 200 g/L TDS) with a wide temperature range of operation (15–95°C) (Fewings and Seet 2012). Leaching using proprietary BioHeap^TM cultures is reported to take place at elevated pH (above pH 2) (Fewings and Seet 2012) unlike the more common acid ferric sulfate leaching processes utilizing mesophilic and thermophilic microorganisms. The reported benefits of bioleaching above pH 2 is lower acid consumption and simpler downstream neutralization due to reduced dissolution of gangue minerals and rejection of iron during leaching (Fewings and Seet 2012). The operating conditions of the leach circuit are reported to be a pH between 2.8 and 3.5 at 55°C in stirred tank reactors using treated process water containing 50 g/L TDS (Fewings et al. 2016); the residence time in the bioleaching circuit is reported to be 5–7 days (Fewings et al. 2015). The process water is treated with Caro’s acid (H₂SO₅) to destroy cyanide which was reported to cause some retardation of Ni leaching (Fewings et al. 2015). The Ni is subsequently recovered from the leach liquor by precipitation using sodium sulfide (Na₂S) to produce a nickel sulfide precipitate grading 45–65% Ni, which is blended with the final flotation concentrate (Fewings et al. 2016). A process flow diagram of the BioHeap^TM leach process implemented at the Cosmic Boy nickel concentrator is presented in Figure 10.

Figure 10. Simplified process flow diagram for bacterial leaching (BioHeap^tm) of high arsenic nickel flotation cleaner tailings at the Cosmic Boy nickel concentrator, Forrestania Nickel Project (western Australia), owned and operated by Western Areas Limited. Adapted from Fewings et al. (2016).

3.6. Roast-leach processes

Roasting of nickel bearing sulfide concentrates followed by leaching has been practised at two operations in Canada, the Falconbridge iron ore plant and the Inco iron ore recovery plant, both treating nickeliferous pyrrhotite concentrates from their mill operations and one operation in Finland, the Kokkola refinery, treating pyrite-pyrrhotite concentrates. It appears there was an economic incentive at the time for these operations to extract iron and base metal values from iron sulfide concentrates produced as a by-product of copper/nickel mining, rather than diverting them to tailings. The Falconbridge and Inco pyrrhotite processing plants have been closed since 1972 and 1982, respectively (Rezaei et al. 2017), whilst the pyrite roaster at the Kokkola refinery in Finland shut in 1987 (Sauni et al. 2010), however brief descriptions are provided for historical purposes. Since the closure of the Canadian plants, nickeliferous pyrrhotite is rejected to the tailings at Canadian mill operations and there is a considerable amount of pyrrhotite in tailings storage areas, where it is a potential source of acid mine drainage but is also of potential economic value due to the contained Ni (Peek, Barnes, and Tuzun 2011; Rezaei et al. 2017).

3.6.1. Falconbridge iron ore plant

The Falconbridge iron ore plant was operated between 1955 and 1972 and processed pyrrhotite concentrate via a sulfation roast-leach process to recover Fe in the form of a hematite product and base metals (Ni, Cu, Co) as a sulfide precipitate which is subsequently smelted (Rezaei et al. 2017). Pyrrhotite concentrate was subjected to oxidative roasting in the presence of Na₂SO₄ (4–5% by weight of the pyrrhotite feed) at 680°C (Rezaei et al. 2017; Thornhill 1961) which converted the base metals to their respective sulfates and iron to hematite. There were no provisions to capture SO₂ from the roasters which was discharged directly into the atmosphere (Rezaei et al. 2017). The calcine from the roasters was leached in water to dissolve the base metal salts and recover an iron oxide leach residue which was dried and pelletized, producing an iron oxide product grading 66% Fe, 0.1% Ni, 2.2% SiO₂ and 0.5% S (Rezaei et al. 2017). The leach liquor is treated with pyrrhotite concentrate to reduce ferric iron in the liquor to the divalent state, followed by precipitation of base metals as their sulfides via addition of elemental sulfur, Na₂S and iron borings which produces a sulfide precipitate typically grading 20.28–20.91% Ni, 1.85–2.30% Cu and 0.44–0.49% Co at an overall Ni recovery of 82.2% (Thornhill 1961). The plant was shut in 1972 due to the introduction of environmental regulations which imposed annual limits on SO₂ emissions from smelters (Potvin and Negusanti 1995). Further operational details of the Falconbridge iron ore process are provided by Thornhill (1961) whilst technical details of the roasting process and sulfide precipitation method are provided in the relevant patents (Thornhill 1954, 1960; Thornhill and Coulter 1966).

3.6.2. Inco iron ore recovery plant

Inco (now Vale) operated a modified Caron process between 1956 and 1982 and extracted iron, sulfur, Ni and Cu from pyrrhotite concentrate recovered from the Copper Cliff or Clarabelle mills in the form of iron oxide pellets, sulfuric acid, nickel oxide (NiO) and copper sulfide (CuS) (Conard 2013; Rezaei et al. 2017). The reduction roast/leach sectio n of the plant was shut in 1982 amid an economic downturn whilst the pyrrhotite roasters continued to operate supplying sulfuric acid to the market until 1991 (Rezaei et al. 2017). Conard (2013) indicated that production of NiO continued until the mid-1990s, though with the shutdown of the reduction kiln, the feed was metallic Ni powder from Inco’s Copper Cliff nickel refinery.

The first stage in the process was an oxidizing roast of the pyrrhotite concentrate at 760°C to produce an iron oxide calcine and SO₂; the SO₂ stream was sent to a sulfuric acid plant whilst the calcine was transferred to the reduction kiln (Conard 2013). The Ni present in pyrrhotite was oxidized to nickel ferrite (NiFe₂O₄) (Conard 2013; Rezaei et al. 2017). In the reduction kiln, Ni in nickel ferrite was reduced to a mixture of Fe-Ni alloy and nickel sulfide whilst hematite in the calcine was reduced to magnetite using partially combusted natural gas (Conard 2013; Rezaei et al. 2017). The reduced calcine was leached in aerated ammonia-ammonium carbonate lixiviant at 55°C to dissolve Ni and other base metals in the calcine (Conard 2013). The iron oxide leach residue was sent to a pelletizing plant to produce hematite pellets grading 67% Fe (Conard 2013). Copper (along with some Co) was recovered from the leach liquor by precipitation as a sulfide using sodium hydrosulfide (NaHS) with the precipitate being sent to a smelter (Conard 2013). Nickel was recovered from the leach liquor after Cu removal by precipitation as a basic nickel carbonate using soda ash (Na₂CO₃) and calcination at 550°C to produce an acid-soluble nickel oxide product grading 77% Ni and 0.15% Co (Conard 2013).

3.6.3. Kokkola refinery, Finland

The Kokkola refinery operated by Outokumpu Oyj processed pyrite-pyrrhotite flotation concentrate recovered as a by-product from the Outokumpu copper mine for recovery of base metals between 1966 and 1987 (Palperi and Aaltonen 1971; Sauni et al. 2010). The pyrite-pyrrhotite concentrate graded 0.67% Co, 0.37% Ni, 0.28% Cu and 0.4% Zn (Palperi and Aaltonen 1971). A two-stage roasting process was employed whereby part of the concentrate was roasted to produce an oxide calcine (termed dead roasting) which was then combined with fresh sulfide concentrate and sodium sulfate in the second roasting stage where sulfation of the base metals takes place at 680°C (Deng 1992; Palperi and Aaltonen 1971). The rationale for this two-stage roasting approach was (a) finely divided ferric oxide produced from dead roasting catalyses SO₂ oxidation to SO₃ which reacts with metal oxides to form their sulfates and (b) addition of fresh concentrate to the second stage maintained a desirable atmosphere for sulfation and supplied heat for roasting (Palperi and Aaltonen 1971). The SO₂ produced from roasting was sent to a sulfuric acid plant whilst the iron oxide leach residue (termed purple ore) grading 64.3% Fe (Palperi and Aaltonen 1971) was presumably sold. Nickel, Cu and Zn were recovered from the leach liquor by precipitation as their sulfides, with the sulfide precipitate being smelted and refined at Harjavalta (Finland) (Palperi and Aaltonen 1971). Metallic cobalt was recovered by hydrogen reduction of a cobalt-ammonium sulfate solution (Palperi and Aaltonen 1971). The sulfation roasting plant reportedly closed in 1987 (Sauni et al. 2010), ahead of the closure of the Outokumpu copper mine in 1989, though the production of cobalt products at Kokkola continues to this day and a cobalt refinery is currently operated by Umicore (Törmänen and Tuomela 2021).

4. Summary

A summary of the advantages, disadvantages, and commercial progress of each of the various leaching systems that have been employed for direct extraction of nickel from flotation concentrates is presented in Table 6. It can be seen from Table 6 that sulfate-based leaching technologies, particularly pressure oxidation and bioleaching, are the most advanced with regards to the number of processes that have been piloted and commercially deployed at nickel sulfide processing operations. Both technologies are well established, having enjoyed success in other industries and therefore pose a low technical risk relative to other less established leaching technologies. Other factors to consider in selection of a process/technology for direct leaching of nickel sulfide concentrates besides the track record/experience in other industries is the flexibility/adaptability of the process to changes in feed composition and the types of finished nickel products that can be produced i.e. intermediates such as MHP or MSP, or class 1 nickel (>99.8% Ni) products such as electrowon cathode, metallic Ni briquettes or powder, or high-purity nickel sulfate. Some of the processes are summarized in Table 6 such as nitric acid leaching have not progressed beyond laboratory scale investigation and therefore remain unproven at large scale. It is interesting to note that despite the success of chloride-based leaching processes in the commercial processing of nickel mattes, chloride-based leaching has not seen use for direct processing of nickel sulfide concentrates. Despite the number of technologies that have been developed for direct leaching of nickel sulfide concentrates, the adoption of direct leaching processes for treatment of nickel sulfide concentrates/ores remains low. However, the expected demand for class 1 nickel in lithium-ion batteries and electric vehicles could provide renewed interest in direct leaching of nickel sulfide concentrates for the following reasons; firstly, Ni extraction from sulfidic resources is likely to be lower cost relative to lateritic resources and secondly, production of class 1 nickel or intermediates such as MHP could enable nickel sulfide miners to better capture the value of their sulfide concentrates through hydrometallurgical processing.

Table 6. Comparison of leaching processes that have been evaluated for the direct leaching of nickel sulfide flotation concentrates and commercial progress.

Leaching system	Advantages	Disadvantages	Commercial progress
Ammoniacal leaching: - Sherritt-Gordon process (ammoniacal pressure leaching)	- Highly selective for metals which form soluble amine complexes (Ni, Co, Cu & Zn) - Iron, aluminum, magnesium & calcium rejected during leaching as insoluble residues - Fast leaching kinetics - Ni metal powder can be produced directly from the resultant leach liquor - Pyrite is inert, though any Ni/Co associated with pyrite is not extracted	- Ammonia is expensive, though it can be partly recovered for reuse - Ammonium sulfate by-product requires disposal if there is no market for it. Where it is to be used for fertilizer applications, must meet specifications with respect to metals and sulfamate content, the latter being herbicidal. - Cobalt refining more complex relative to other processes	- Sherritt-Gordon Process originally developed to treat flotation concentrates and is in use at two locations (Fort Saskatchewan, Canada & Kwinana, Western Australia); however, both operations no longer process nickel flotation concentrates. Fort Saskatchewan is processing MSP and Kwinana is processing nickel matte.
Acid sulfate leaching: - Pressure oxidation -Co-processing of Ni-laterites & sulfides - Atmospheric leaching	- Pressure leaching is conducted by well-established technologies - Pressure oxidation processes can be operated autogenously i.e. no external heat input - Sulfuric acid is a low cost lixiviant, can also be generated in situ from oxidation of part or all the elemental sulfur produced during leaching or recycling SX and EW raffinates to the leaching stage - High degree of flexibility with respect to type of products that can be produced (cathode, MHP, MSP, sulfate salts) - Depending on the chosen leach conditions, oxidation of sulfide to elemental sulfur can be favored (which reduces the amount of oxidant required) - Pyrite is generally inert (depends on leach conditions), though any Ni/Co associated with pyrite may not be extracted - Arsenic can be managed by precipitation as stable ferric arsenate - Iron hydrolysis products are typically well-crystalline and readily filterable (depends on leach conditions)	- Autoclaves and oxygen gas represent significant capital and operating costs for pressure oxidation processes - Recovery of Au, Ag and PGMs from the leach residue can be difficult depending on the chosen leach conditions - Iron precipitation products generally unsuitable for steel making due to sulfur & Zn/Cu content - Ca/Mg-bearing silicate minerals can pose problems with regards to acid consumption & silica gelling, and dissolved Mg ions can be problematic for Ni/Co SX and effluent treatment - Potentially slow dissolution kinetics at ambient pressure for stirred tank leaching processes unless ultra-fine grinding is utilized such as in the Albion Process^TM or through the addition of halide salts such as in the Intec Nickel Process	- Numerous pressure oxidation processes have been piloted (Activox®, PLATSOL^TM, CESL, Kell process) - Pressure oxidation has been commercially deployed at the Kokkola refinery (Finland), Long Harbor (Voisey’s Bay Process, Canada), the Nadezhdin mill (Russia), Fredericktown and Garfield refineries (USA). The US operations are now closed but were the first commercial operations to utilize pressure oxidation to process nickel bearing sulfide concentrates. - Co-processing of laterites and sulfides appears promising but is currently limited to laboratory/pilot investigations - Albion Process^TM (atmospheric leaching process) expected to be suitable for nickel sulfide processing based on commercial success in processing of gold, zinc and copper concentrates
Chloride leaching: - Ferric & cupric chloride leaching - Gaseous oxidants (oxygen, chlorine, and their mixtures)	- Fast leaching kinetics - Oxidation of sulfide to elemental sulfur favored, and As(III) to As(V) which is precipitated as ferric arsenate phases - Elemental sulfur formed at temperatures below melting point is porous, facilitating more complete leaching of base metals - Leaching can be carried out at atmospheric pressure & lixiviants can be regenerated for reuse - Precious metals and PGMs are soluble in chloride lixiviants though rejection to the leach residue is possible through control of the ORP - Pyrite is inert, though any Ni/Co associated with pyrite is not extracted - Ni & Co electrowinning from chloride media is well established	- Chloride ion is corrosive and therefore exotic materials of construction are required for the leaching vessels - Impurity buildup can occur from recycling of chloride lixiviants, necessitating their removal once the concentration becomes intolerable - Use of chlorine gas as an oxidant for leaching poses a hazard to humans and the environment. ‘Wet’ chlorine gas also corrosive, requiring exotic materials of construction for leaching vessels - Recovery of Au and PGMs from the leach residue can be difficult where the residues are contaminated with iron hydrolysis products and elemental sulfur -Advantages of chloride media such as producing sulfur textures ideal for facilitating base metal sulfide dissolution, and limited oxidation of sulfur to sulfate can be equally addressed in sulfate/sulfuric acid systems with addition of chloride salts (Activox®, CESL process)	- Chlorine leach processes have been successfully employed for extraction of Ni and Co from mattes at Nikkelverk (Falconbridge process), Sandouville-Le Havre and Niihama (MCLE process) but have not been utilized for processing nickel flotation concentrates - Atmospheric chloride leach processes have been patented (Outotec, Jaguar Nickel Inc.) but have not seen industrial deployment
Nitric acid leaching	- Nitric acid serves as both the lixiviant and the oxidant, simplifying reagent requirements for leaching. It is also a strong oxidizing agent - Nitric acid can be regenerated from hydrolysis of ferric nitrate solutions, scrubbing of NOx gases evolved during leaching and thermal decomposition of nitrate salts such as Mg(NO3)2 if recovered - Hematite produced from hydrolysis of nitrate solutions is ‘cleaner’ and more marketable i.e. low sulfur, base metal contaminants compared to sulfate-based routes - Stainless steels can be used as materials of construction for leaching vessels - Arsenic can be managed by precipitation as stable ferric arsenate	- Sulfur oxidation to SO4^2- undesirable as it consumes excess nitric acid and the sulfate must be subsequently removed from the leach liquor for nitric acid regeneration - Metals recovery by electrowinning or H2 reduction likely to require transfer of the dissolved metals from a nitrate medium into a sulfate or chloride system - Potential for degradation of organic extractants due to residual nitric acid in leach liquors - Environmental limits on discharge of nitrate & NOx gases means recovery and re-use of reagent must be rigorous -Benefits of nitric acid can be addressed in sulfate systems by adding of small quantities of nitrates/nitrites to function as an oxidation catalyst i.e. Nitrogen Species Catalyzed (NSC) pressure leaching, or used exclusively as an oxidant	- Direct Nickel Process (DNi Process^TM) has been piloted for processing nickel laterite ores - Very few published studies on leaching of Ni-sulfide concentrates in nitric acid solutions
Bacterial leaching: - Mesophiles (T < 45° C) - Moderate thermophiles and thermophilic archaea (55–85° C)	- Processing equipment is simpler relative to other competing technologies - High degree of flexibility with respect to type of products that can be produced (cathode, MHP, MSP, sulfate salts) - Bioleaching processes are stated to be flexible with regards to concentrate head grades and mineralogy - Bioleaching is suitable for the processing of polymetallic concentrates containing gold, silver and PGMs in addition to base metals	- Slow leaching kinetics which translates to long residence times; 5–10 days being typical for continuous stirred tank bioleaching - Microbes may be sensitive to chloride content in water which poses problems for regions where only saline waters are available. Microbes may also be sensitive to pulp density and shear rate which affects throughput and equipment sizing. Finally, microbes may be sensitive to leaching temperature which affects throughput and cooling requirements for bioleach processes - Iron precipitation product may be poorly crystalline & voluminous with poor settling & filtration properties; base metal losses could occur through adsorption & entrainment - Poor rejection of As during leaching; As is solubilized as As(III) which must be oxidized to As(V) for precipitation from solution	- Three bioleaching processes for treatment of nickel sulfide concentrates have been piloted: BioNIC® (BHP Limited) BacTech (BacTech Limited) and BioHeap^TM (Western Areas Limited); the latter having been implemented at a nickel sulfide concentrator - Bioleaching of nickel sulfide concentrates has been implemented at two operations for treating low volumes of material; in both cases the concentrates being processed were high in arsenic (as niccolite and gersdorffite) - Bio-heap leaching is in use in Finland for processing of a base metals (Ni, Cu, Zn, Co) bearing black schist
Roast-leach processes: - Oxidizing roast & Caron-type processes - Chloridizing roasting - Sulfation roasting - Reduction roasting	- Iron conversion to an insoluble iron oxide during oxidizing roasting processes can result in minimal iron dissolution in subsequent leaching stages & could also enable the recovery of iron oxide as a by-product - Conversion of base metal sulfides to sulfates or chlorides during salt and sulfation roasting means water or dilute acids can be used as the lixiviant in leaching - Reduction roasting converts many ‘refractory’ sulfides to a more reactive form amenable to subsequent leaching via non-oxidative dissolution	- Gaseous emissions from roasting processes requires emissions abatement equipment or a sulfuric acid plant which significantly increases capital cost. Arsenic may be volatilized during roasting - Nickel ferrites if formed during oxidizing roasting processes limit Ni recovery during subsequent leaching processes - Use of alkali sulfates or chlorides for roasting could be problematic for disposal of effluents - Roasting processes can be energy intensive - Reduction roasting followed by non-oxidative leaching generates H2S, a toxic gas, whose emissions must be managed such as by the Claus process or other suitable technologies - Non-oxidative leaching of reduced sulfide concentrates require high concentrations of mineral acids which can require the use of exotic materials of construction, thereby increasing capital cost. Regeneration and recycling of the lixiviant would also be required for this approach to be economical	- A modified Caron process was in use by Inco Ltd. at their Copper Cliff Operations for recovery of iron and base metals from pyrrhotite concentrates. The plant is no longer operational - The Falconbridge iron ore plant in Canada and the Kokkola works in Finland utilized sulfation roasting to recover iron and base metals from pyrrhotite and pyrite-pyrrhotite concentrates. Both plants are no longer operational - Chloridizing roasting and reduction roasting + leaching processes have only been tested at laboratory scale for the processing of nickel sulfide concentrates

Acknowledgements

The authors are grateful for the support provided by CSIRO Mineral Resources to Nebeal Faris through the ResearchPlus CERC Postdoctoral Fellowship program. The authors would like to thank Dr. Robbie McDonald and Mr. David McCallum of CSIRO Mineral Resources for taking time to review and provide valuable feedback during the preparation of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the CSIRO [Research Plus CERC Postdoctoral Fellowship program].