Is Near-zero Waste Production of Copper and Its Geochemically Scarce Companion Elements Feasible?

ABSTRACT

Copper ores, end-of-life electric and electronic equipment and car electronics can contain, besides Cu, substantial amounts of geochemically scarce companion elements. Geochemically scarce elements have an upper crustal abundance of <0.025 (weight)%. In view of resource conservation and reduction of pollution there is a case for near-zero waste processing. Improving the generation of ore concentrates and use of kinetic and thermodynamic data regarding smelting and converting can increase the production of geochemically scarce elements in the pyrometallurgical processing of copper ores. Reprocessing of copper ore processing residues can serve the generation of geochemically scare elements and the clean-up of matrix materials. Modularization of products, closed-loop take-back, including deposit-refund, systems for end-of -life products, and changes in the pre-processing thereof can be conducive to improved recovery of geochemically scarce elements from end-of-life electric and electronic equipment and car electronics. A comparatively large variety of geochemically scarce elements originating in end-of-life products can be recovered when in smelting lead and copper serve as collectors. Substantial research and development work is needed to optimize the co-production of geochemically scarce elements by hydrotechnology from copper ores and (mined) end-of-life products and to assess the potential of solvochemistry. There is technical scope for significant progress in the direction of near-zero waste processing in processing copper ores and (mined) end-of-life products, but for the realization of near-zero waste processing there are hurdles to be overcome related to marketability of outputs, safe handling of hazardous elements and company behavior. Also, the techno-economic potential of hydrotechnology and solvochemistry in extracting copper ores, copper ore processing residues and end-of-life products is uncertain. In view thereof, the feasibility near-zero waste production of copper and its geochemically scarce companion elements from copper ores and end-of-life electric and electronic equipment and car electronics is uncertain.

KEYWORDS:

• Copper
• geochemically scarce elements
• matrix materials
• pyrometallurgy
• hydrometallurgy

1. Introduction

Spooren et al. (2020) view near-zero waste processing as goal for handling low grade complex metal ores and wastes containing metals. They state that this goal is in line with current trends of process development. In view of the study of Spooren et al. (2020) one might formulate the hypothesis that near-zero waste processing is a feasible goal for handling metal ores and wastes containing metals.

Spooren et al. (2020) have discussed possibilities for near-zero waste processing for Polish and Greek laterites, complex ores that derive their economic value mainly from the presence of Ni, with scope for Co and Zn contributing to the future economic value of extracted metals. Spooren et al. (2020) also considered the valorization of matrix materials which are present in co-outputs of laterite processing. They furthermore discussed the scope for near-zero waste processing in the secondary production of metals from production wastes containing a variety of geochemically scarce metals (with an average upper crustal abundance of <0.025 (weight)%), such as Co, Cr, Cu, Ni and Zn.

The present review extends the work of Spooren et al. (2020) to copper ores and wastes containing copper. Copper in ores is often accompanied by other geochemically scarce elements (companion elements). Box 1 shows geochemically scarce elements that may have a substantial presence in copper ores. Copper ores show a wide variety as to the presence of these elements. Geochemically scarce elements may also be added to copper during copper life cycles for alloying, joining and coating. Table 1 presents these elements. Box 2 shows geochemically scarce elements that can become companions to copper in electric and electronic equipment and electronic components of cars. In part these are elements used for alloying, coating and joining copper present in electric and electronic equipment and electronic components of cars, such as Ag, Au, Be, Bi, In, Ni, Sb and Sn (cf. Table 1)

Box 1. Geochemically scarce elements, other than Cu, that may have a substantial presence in copper ores (Ayres, Ayres and Rade 2002; Sracek et al. 2010; Mudd, Weng and Jowitt 2013a,b; Izatt et al. 2014; Chmielewski, Wawzczak and Brykala 2016; Weidenbach, Dunn and Tao 2016; González et al. 2017; Safarzadeh et al. 2018; Crespo et al. 2018; Makuei and Senanayake 2018; Mikoda et al. 2019a; Araya et al. 2021).

Ag As Au B Bi Cd Co Cr Ga Ge Hg In Li Mo Ni

Pb Pt group metals (Ir, Os, Pd, Pt, Rh, Ru), Rare earths Dy, Er, Eu, Gd, Ho, La, Lu, Nd, Pm, Pr, Sc, Sm, Tb, Y, Yb) Rb Re Sb Se Sn Te Th U V W

Box 2. Geochemically scarce elements, other than Cu, associated with electric and electronic equipment and electronic components of passenger cars (Andersson, Söderman and Sandén 2019; Buechler et al. 2020; Cesaro et al. 2018; Widmer et al. 2015).

Ag As Au B Be Bi Cd Ce Cr Co Dy Eu Ga Gd Ge In Ir La

Li Lu Mo Nb Nd Pb Pd Pt Ru Sb Sc Sn Ta Tb W Y Zn

Box 3. Subjects requiring major research and development efforts to narrow gaps in present knowledge.

-Feasibility of (bio)hydrometallurgy-based primary processing technology producing geochemically scarce elements and cleaned-up matrix materials. This especially regards (bio)hydrometallurgy, at preferably low cost, for the extraction of copper ores, of copper ore tailings, and of dusts, slags and slag tailings from primary pyrometallurgy.

- Assessing the techno-economical scope for solvometallurgical processing of copper ores and copper ore processing residues for the production of a substantial range of geochemically scarce elements.

- Feasibility of (bio)hydrometallurgial and solvometallurgical production of a substantial range of geochemically scarce elements from mined complex end-of-life products.

- Reprocessing slags and dusts from secondary pyrometallurgical copper production for the extraction of geochemically scarce elements and the generation of sufficiently clean residual materials.

- Enabling the environmental acceptability of valorized matrix materials.

-Safe handling of hazardous elements such as Cd and As.

Table 1. Geochemically scarce functional elements that may become associated with copper by alloying, coating and joining. (Samuels and Méranger 1984; Marshakov 2005; Karpagavalli and Balasubramaniam 2007; Chen and Bull 2008; Sarver and Edwards 2011; Jolly 2013; Ruzic et al. 2013; Garza-Montes-de-Oca et al. 2014; Imamura et al. 2015; Kanlayasiri and Ariga 2015; Laws et al. 2015; Young and Dunand 2015; Forsén, Aromaa and Lindström 2017; Guarino et al. 2017; Noor, Zuhallawati and Radzali 2016; Lee 2018; Nagel 2018; Ulman et al. 2018).

Function	Relatively rare elements that may become functionally associated with copper
Alloying	Ag, B, Be, Bi, Cd, Co, Cr, Ni, Pb, Sb, Se, Sn, Te, Zn
Coating	Cr, Ni, Sn
Joining	Ag, Au, Bi, In, Ni, Pb, Sb, Sn, Zn

Pollution by copper and its geochemically scarce companion elements linked to the current handling of ores and wastes can impact the environment with consequences that may be detrimental for ecosystems and/or human health (e. g. Du et al. 2020; Fry et al. 2020; González-Castanedo et al. 2014; Kimball et al. 2016; Potysz et al. 2015; Rubinos et al. 2021; Sorooshian et al. 2012; Wiklund et al. 2018). Near zero-waste processing of ores and wastes might strongly reduce such pollution.

This paper will consider the question whether near-zero waste production of copper and its companion elements is a feasible goal for handling coper ores and waste electric and electronic equipment and car electronics. The focus of this paper will be on increasing geochemically scarce elements as product outputs and lowering of the presence of the same elements in production residues, which may improve the scope for valorization of matrix materials present in these residues and reduce pollution. In its focus on increasing geochemically scarce elements in product outputs and on the valorization of matrix materials present in ores, the present paper will be similar to the study of Spooren et al. (2020). Differently from the review of Spooren et al. (2020), the secondary production of metals from product wastes (specifically mined waste electric and electronic equipment and car electronics), rather than from production wastes, will be discussed. Due to the wide variety of geochemically scarce elements present in those wastes (cf. Box 2) their processing is challenging from a near-zero waste perspective. The present review will add to the review of Spooren et al. (2020), by considering the release of geochemically scarce elements from recycled matrix materials and proposals for preventing the release to the environment of hazardous companion elements, especially As, for which there is little demand. Also, hurdles to the feasibility of near-zero waste processing will be considered.

The co-production of Cu and other geochemically scarce elements from copper ores, and options for the improvement thereof, will be addressed in section 2. Both optimizing processing of copper ores and reprocessing of residues, to increase the recovery of copper and its geochemically scarce companion elements, will be discussed. This section will also deal with the valorization of matrix materials that are co-outputs of copper ore processing. The recovery of copper and its geochemically scarce companion elements from end-of-life electric and electronic equipment and car electronics, and options for the improvement thereof, will be addressed in section 3. Processing of mined end-of-life electric and electronic equipment and car electronics in secondary smelters, the pyrometallurgical co-processing thereof with copper ores, and the role of copper smelters will be discussed. Section 4 will consider hurdles for the feasibility of near-zero waste processing. These regard marketability of outputs, company behavior and safe handling of hazardous elements when potential supply exceeds demand. Section 5 will present the conclusions of this paper.

2. Production of geochemically scarce elements from copper ore and the valorization of matrix materials that are co-outputs of copper ore processing

The primary production of copper will be outlined in section 2.1. Technical options for increasing the production (recovery) of copper from ore will be presented, Section 2.1 also provides the necessary background for section 2.2. which considers the extraction of geochemically scarce companion elements from copper ores and options for the improvement thereof. Section 2.3 contains the concluding remarks regarding sections 2.1 and 2.2. Section 2.4 briefly discusses new technologies for copper extraction. Section 2.5 will consider the valorization of matrix materials that are co-outputs from copper ore processing.

2.1. Primary production of copper

Extraction of copper from ores is currently dominated by two types of processing: pyrometallurgical and hydrometallurgical, both combined with refining using electrometallurgy (electrolysis) (e. g. ICSG 2016; Norgate and Jahanshahi 2010). The main steps in both types of processing are in Figures 1a and 1b.

Figure 1. (a) Main steps in pyrometallurgical primary production of copper (cathodes). (b) Main steps in hydrometallurgical primary production of copper (cathodes).

Spooren et al. (2020), focusing on near-zero waste metallurgy, would seem to prefer hydrometallurgy and leaching or extraction with non-aqueous solvents (solvometallurgy), as these extraction technologies have a low impact on the physicochemical properties of the matrix materials present in concentrates or ores. However, there is also scope for valorization of the matrix materials that end up in copper slags, an important co-output of pyrometallurgy (Kinnunen et al. 2020).

Copper production from mining in 2020 is estimated at about 21 × 10⁹ kg (ICSG 2019), of which about 21% is relying on hydrometallurgical technology (ICSG (International Copper Study Group) 2016). Solvometallurgical leaching (leaching with non-aqueous solvents) of Cu from copper ores has been demonstrated, but commercial application thereof has not been reported (Binnemans and Jones 2017; Spooren et al. 2020). The application of electrochemistry in molten salts to the primary production of copper, in analogy to the primary production of aluminum in the Hall-Héroult electrolysis process, has been investigated (Allanore 2017; Daehn and Allanore 2020; Sahu, Chmielowiec and Allanore 2017; Sokhanvaran et al. 2016; Tan et al. 2016), but is as yet not commercially applied.

Pyrometallurgy regards sulfidic copper ores. Preceding pyrometallurgical processing, the copper content of ore-derived material is commonly increased. This is done by physico-mechanical treatment in concentrators that use crushing, grinding (milling) and flotation, occasionally combined with other technologies such as use of magnetic fields or centrifugation (Katwika et al. 2019; Lagos et al. 2018; Rönnlund et al. 2016; Spooren et al. 2020). Geometallurgical modeling may help in achieving improvement of copper recovery and copper concentrations in the concentrate (Boisvert et al. 2013; Rincon, Gaydardzhiev and Stamenov 2019). Electrical fragmentation, treatment with microwaves, sensor-based sorting, improved flocculation and machine learning have been suggested as technologies that can improve the composition of the concentrate (Spooren et al. 2020). Optimizing flotation time, particle size, pH and dosages of reagents such as collector and frother may improve copper recovery and/or copper grade (e. g. Asghari, Nakhaei and VandGhorbany 2018; Bakalarz 2019; Dhar, Thornhill and Kota 2019; Ghodrati et al. 2020). Copper ore concentrates may also serve hydrometallurgical processing (cf. section 2.2.3). The processing residues originating in concentrate production (tailings) are commonly dumped, landfilled or stored in dams and ponds. Worldwide, average Cu losses in mining and the generation of concentrates have been estimated at about 18% of the Cu input in primary production in 2010 (Glöser, Soulier and Espinoza 2013). Reducing this 18% loss may well be difficult when the downward trend of copper concentrations in ores continues (cf. Alcalde, Kelm and Vergara 2018). Increased exploitation of complex ores and increasing co-production of other geochemically scarce elements may add to the difficulty of reducing this percentage (e. g. Corin et al. 2017; Sousa et al. 2017). On the other hand, reprocessing of tailings from copper concentrate production may increase the recovery of Cu (e. g. Alcalde, Kelm and Vergara 2018; Bakalarz 2019; Brest et al. 2021; Drobe et al. 2021; Falagán, Grail and Johnson 2017; Lutandula and Maloba 2013; Shengo 2021).

In pyrometallurgy, firstly matte (containing substantial amounts of S and 45–85 (weight)% Cu) is generated in flash or bath smelters. In 2015 about 72% of copper ore smelting capacity regarded flash smelter-based production of matte (ICSG (International Copper Study Group) 2016). Flash smelting requires a dry input of particles <100 micrometer, whereas bath smelters can also process larger sized materials and have a less strict drying requirement (Forsén, Aromaa and Lindström 2017). A higher matte grade and a higher temperature in flash smelting increase solubility of Cu in slag (Forsén, Aromaa and Lindström 2017; Klaffenbach et al. 2021). Subsequently, through oxidative treatment in converters, matte is turned into blister copper with >90 (weight)% Cu (e. g. ICSG (International Copper Study Group) 2016). In the presence of tailored slag systems direct concentrate-to-blister copper smelting is an option (Voigt et al. 2017). Blister copper in turn is refined in an anode furnace (fire refining), to generate copper anodes with >99 (weight)% Cu (e. g. Dupont et al. 2016; Gregurek et al. 2018; Li et al. 2016). Subsequently, anode copper is subjected to electrolytic treatment. This implies dissolution of anode-Cu in the electrolyte and the generation of anode slime (mainly precipitates, also called anode mud). Cu is removed from the electrolyte, generating cathodes with high purity (>99.9 (weight)%) copper (Forsén, Aromaa and Lindström 2017; Norgate and Jahanshahi 2010; Rönnlund et al. 2016).

During smelting, converting and fire refining dusts and slags are generated. These dusts are partly discarded (Schreck 1999; Shibayama et al. 2010), but recovery of Cu from a part of smelter dusts by internal recycling is also often practiced commercially (Bakhtiari et al. 2011; Montenegro, Sano and Fujisawa 2013; Liu et al. 2018; Moats, Alagha and Awuah-Offei 2021; see also section 2.2.2.4). Furthermore, several primary smelters have been reported recover Cu from dusts by hydrometallurgy (Schlesinger et al. 2011). High-recovery extraction of Cu from dusts by bioleaching is also an option (e. g. Ebrahimpour et al. 2021): Smelting, converting and fire refining give rise to the co-production of slags. Published data about the Cu content of slags generated in matte production vary between about 0.3 and 2.74 (weight) % (Sridhar, Toguri and Simeonov 1997; Mikula et al. 2021; Tian et al. 2021). The partitioning of Cu to slags in converters is relatively high because the partial oxygen pressure in converters is relatively high, commonly leading to a Cu content in the order of 4–8 (weight)% (Bellemans et al. 2018). In an anode furnace, partial oxygen pressures are even higher than in converting: Selivanov et al. (2013) reported a copper content of about 33% in anode furnace slag. Slags are partly discarded (Li et al. 2018b; Sanchez et al. 2004) and partly the object of copper recovery (Coursol et al. 2012; Cusano et al. 2017; Guo et al. 2018; Han, Yu and Cui 2016; Hughes 2000; Khalid et al. 2019; Mikoda, Potysz and Kmiecik 2019b). Recovery of copper from slags is possible by a variety of processes. These are: internal recycling of slag concentrate (generated by milling and flotation), leaching and treatment in a slag cleaning furnace under reducing conditions (Chun et al. 2016; Coursol et al. 2012; Cusano et al. 2017; Guo et al. 2018; Han, Yu and Cui 2016; Khalid et al. 2019; Mikoda, Potysz and Kmiecik 2019b). Recovery of Cu in concentrate may be enhanced by modifying molten copper slag (Guo et al. 2016). Internal recycling of slag concentrate, and treatment in slag cleaning furnaces are practiced commercially (Coursol et al. 2012; Cusano et al. 2017).

After internal recycling of slag concentrate or treatment in a slag cleaning furnace, reported Cu concentrations in slag tailings vary from 0.2 to 1.2 (weight)% and are often in the range of 0.7–1 (weight)% (Carranza et al. 2009; Coursol et al. 2012; Cusano et al. 2017; Holland et al. 2019; Mikoda et al. 2019a; Muravyov et al. 2012). Slag tailings are often landfilled or used for infrastructural applications without further treatment (Khalid et al. 2019; Holland et al. 2019; Mikoda et al. 2019a, Sibanda et al. 2020).

Glöser, Soulier and Espinoza (2013) estimated the worldwide losses in 2010 of Cu to slags and dusts originating in primary copper smelting, converting and refining operations at about 3% of the Cu input in primary production. This percentage may be reduced by optimizing primary copper pyrometallurgy guided by thermodynamics and kinetics (e. g. Nakajima et al. 2011; Shihshin, Hayes and Jak 2018; Sineva et al. 2021) and by increasing the extraction of Cu from slags, slag tailings and dusts. There is technical scope for doing the latter by optimizing slag cleaning furnaces (e. g. Yang et al. 2018), by improving flotation of milled slag (Sibanda et al. 2020), by leaching slag tailings, whether or not after sulfation roasting (Carranza et al. 2009; Grudinsky et al. 2021; Mikoda, Potysz and Kmiecik 2019b; Muravyov et al. 2012) and by improved leaching of dusts whether or not after sulfation roasting (Liu et al. 2018; Priya et al. 2020; Sabzezari et al. 2019).

Current commercial hydrometallurgical processing mainly regards heap (bio)leaching of comminuted materials under acid conditions. This regards constructed heaps of low-grade sulfidic ores (oxidative acid bioleaching), and ores containing Cu-oxides and -carbonates (acid leaching) (e. g. Ghorbani, Franzidis and Petersen 2018; Petersen 2016; Scheffel, Guzman and Dreier 2016). For suitable copper ores (including chalcocite and covellite), and when effective measures preventing poor percolation (by improving agglomeration) and leakage of leachate are in place, copper recoveries of 60–90% have been reported (Domic 2007; Gentina and Acevedo 2016; Ghorbani, Franzidis and Petersen 2018; Norgate and Jahanshahi 2010; Petersen 2016; Ruan et al. 2013; Yin et al. 2018b). The optimization of heap (bio)leaching, which would be conducive to improved Cu recovery, has been discussed by Ghorbani, Franzidis and Petersen (2018). Options for improved Cu recovery include better heap permeability and aeration (Ghorbani, Franzidis and Petersen 2018). The Cu recovery from the main copper ore chalcopyrite by heap bioleaching tends to be poor due to the formation of a passivation layer on the mineral (e. g. Pradhan et al. 2019; Tanne and Schippers 2019). Options for improved Cu recovery from chalcopyrite include: better redox control, improved mixtures of bioleaching organisms, addition of chloride, the use of thermoacidophiles and high temperatures (Ai et al. 2019; Ghorbani, Franzidis and Petersen 2018; Ma et al. 2017, 2018; Watling 2006; Zhao et al. 2019).

Leachate obtained by heap leaching is commonly treated by solvent extraction. Reported estimated Cu extraction efficiencies are about 95% (Panda et al. 2016 or >95% (Hosseinzadeh, Azizi and Hassanzadeh 2021). Recycling the extracted leachate (raffinate) is common practice (Liu et al. 2020). However, the raffinate contains residues of solvent that may negatively impact microorganisms conducive to bioleaching (Zhang et al. 2019a). Moreover, metals (Cd, Cr, Fe, Ni, Zn) and As tend to accumulate in the raffinate which in practice can be considered problematic for ongoing recycling of raffinate (Liu et al. 2020). Chemical precipitation of metals from raffinate followed by landfilling is commonly practiced (Liu et al. 2020).

The solvent extract is stripped by sulfuric acid, which can be done with an efficiency >99% (Hosseinzadeh, Azizi and Hassanzadeh 2021). The resulting copper sulfate solution is treated by electrowinning (electrolysis) generating copper cathodes. Ion exchange technology can be more efficient than solvent extraction at low Cu concentrations and can also have other environmental benefits (Izatt et al. 2015; Sole, Mooiman and Hardwick 2017). Hawker et al. (2018) have proposed to precipitate copper from the leachate and to add the precipitate to the inputs for a copper smelter or- converter (the synergistic copper process). Hawker et al. (2018) mentioned as advantages of the synergistic copper process that it may reduce copper losses in pyrometallurgy and extend the use of low-grade copper ores. However, they did not compare the efficiencies in processing of leachate in copper production by the synergistic process with conventional (bio)hydrometallurgy based on heap leaching (cf. Figure 1b). It would seem uncertain whether the recovery of copper improves if compared with the conventional approach of solvent extraction, followed by stripping and electrowinning. Also, the exclusive focus of the synergistic copper process on copper production is at variance with developments in hydrometallurgy aimed at the co-production of geochemically scarce companion elements (discussed in section 2.2.3).

Heap leaching-based technologies for Cu extraction based on chloride chemistry (Ghorbani, Franzidis and Petersen 2018) and (bio)leaching in alkaline ammonia solutions (Yin et al. 2018a) have been demonstrated but their commercial application has not been reported.

In primary hydrometallurgical Cu production, there is furthermore commercial application of contained processes. These are: agitated tank leaching of copper ores and copper ore tailings and autoclave-based, pressurized oxidative acid leaching processes applied to processed copper ores. Agitated tank leaching of copper ores and copper ore tailings has been reported for Africa (Sole et al. 2019). In Cobre Las Cruces (Spain) an ore-derived input with 5.1% Cu is extracted with a 91.8% efficiency by pressurized oxidative acid leaching (Dreisinger 2015). In the Freeport plants in Arizona (USA) Cu is recovered from ore concentrates by pressurized oxidative leaching with an overall efficiency of about 98% (Dreisinger 2015; Gertenbach 2016; Green, Robertson and Marsden 2018). Pressurized oxidation and leaching of copper ore/pyrite mixture is practiced by Sepon Copper (Laos) (Dreisinger 2015). Solvent extraction and electrowinning are used to produce Cu from the leachate generated by pressurized oxidative leaching (Dreisinger 2015). Contained hydrometallurgy is also applied when, besides Cu, geochemically scarce companion elements are present in substantial concentrations (see section 2.2.3). Residues remaining after contained leaching can be reprocessed. This offers scope for increased recovery of Cu (Hubau et al. 2020; Shengo et al. 2021).

2.2. The co-production of Cu and geochemically scarce companion elements in copper ore processing

In primary production of copper from suitable ores, there is currently co-production of other geochemically scarce elements. Examples are the co-production of Cu and Mo (e. g. Izatt et al. 2014; Tabelin et al. 2021), Cu and precious elements (Schwartz et al. 2017) and Cu and Co (Sole et al. 2019). In the following part of this section the scope for increased co-production of geochemically scarce elements from copper ores will be discussed.

2.2.1. Geochemically scarce companion elements in copper ore concentration

Copper ore concentrators are often aiming at high concentrations of copper in concentrate for primary copper production while sacrificing as little copper recovery as possible (Alcalde, Kelm and Vergara 2018). A strategy optimized for Cu might well lead to relatively large losses of other valuable elements to concentrator residues or tailings (Agorhom et al. 2015; Zanin et al. 2009). For instance, in the case of Te present in copper ores, current worldwide average partitioning to tailings in concentrate production has been estimated at about 90% (Makuei and Senanayake 2018; Moats, Alagha and Awuah-Offei 2021). Similar losses to tailings are likely for Se and Bi (Moats, Alagha and Awuah-Offei 2021). Mikula et al. (2021) reported as to concentrator residues originating in Turkish, Chinese and Polish mines oxide concentrations for Zn up to 2.8%, for Cr up to 0.12%, and for Co up to 0.2%. Araya, Kraslawski and Cisternas (2020), (2021)) found substantial amounts of Co, Nb, rare earths, Sb, and V in copper mine tailings from the Antofagasta region, Chile. Hazardous substances are likely to be mobilized from stored tailings and even more so after mine tailings dam failures (Araya, Kraslawski and Cisternas 2020; Falagán, Grail and Johnson 2017; Mikula et al. 2021; Rubinos et al. 2021).

By changing concentrator processes, the recovery of other elements than Cu may be improved (Zanin et al. 2009; Agorhom et al. 2015; Katwika et al. 2019; Tabelin et al. 2021). Geometallurgical modeling may help in achieving improvement (Boisvert et al. 2013; Rincon, Gaydardzhiev and Stamenov 2019), though recovery of all elements of interest at high efficiency is difficult, if possible at all (Mudd, Jowitt and Werner 2017). Further technologies that may improve recoveries are sensor-based sorting, machine learning and selective flocculation (Spooren et al. 2020):

Other valuable elements than Cu may be channeled into other concentrates than copper concentrate, with some loss of Cu to those other concentrates (e. g. Li et al. 2013; Manouchehri 2018; Yin et al. 2017). There are commercial concentrator processes generating more than one concentrate. At Boliden (Sweden) a complex copper ore is processed into three outputs: concentrates of Cu, Pb and Zn (Manouchehri 2018). At Norilsk (Russia) ores are processed into three concentrates, of Cu, Ni and Co respectively (Fröhlich et al. 2017). Flotation may generate Cu and Co concentrates when substantial amounts of these elements are present in ores (Sracek et al. 2010); such technology has been applied in the African Copperbelt (Sole et al. 2019). Concentrates derived from Cu-Mo-Au-Re porphyry ore deposits can be converted to copper concentrates and molybdenite concentrates, with the latter containing most of the Re (Izatt et al. 2014; Mudd, Jowitt and Werner 2017).

The generation of more than one concentrate does not necessarily mean that the recovery of companion metals is optimized. There may be optimization detrimental to specific companion elements. In several cases of processing Cu-Ni-Co ores the presence of Cu and Ni in concentrates was reported to have been maximized to the detriment of Co (Mudd et al. 2013b).

As to the concentrator tailings there is a case for reprocessing, aiming at the generation of geochemically scarce elements and sufficiently clean matrix materials. In the case of old tailings with relatively high concentrations of other valuable geochemically scarce elements application of flotation technology may be useful in reprocessing, as shown by Parviainen, Soio and Caraballo (2020) focusing on Au and Cu. For rare earths, Abaka-Wood, Addai-Mensah and Skinner (2021) found that froth-flotation of Australian copper ore concentrator tailings might be conducive to a 90% recovery, and magnetic separation to a 86% recovery. (Bio)hydrometallurgy (cf. sections 2.1 and 2.2.3) is another option to exploit in reprocessing concentrator tailings. (Bio)hydrometallurgy applied to concentrator tailings can be conducive to the recovery of Ag, Au, Co, Cu, Ni, Pb, rare earths, Th, U, V and Zn, (Liu et al. 2007; Kondriat´eva et al. 2012; Lutandula and Maloba 2013; Chen et al. 2014; Chmielewski, Wawzczak and Brykala 2016; Falagán, Grail and Johnson 2017; Alcalde, Kelm and Vergara 2018; Altikaya et al. 2018; Shengo 2021; Sole et al. 2019; Araya, Kraslawski and Cisternas 2020; Fomchenko and Muravyov 2020; Drobe et al. 2021; Zhang et al. 2020). Particle sizes of the tailings may be at variance with successful conventional heap-leaching, which would imply heap leaching of agglomerated tailings or the use of relatively costly tank-based hydrometallurgy (e. g. Alcalde, Kelm and Vergara 2018; Ghorbani, Franzidis and Petersen 2018; Hao et al. 2017). Much research and development work would seem needed to optimize the production of geochemically scarce elements by reprocessing of concentrator tailings.

2.2.2. Distribution of Cu and other geochemically scarce elements during smelting and recovery of these elements

Applying thermodynamics and kinetics to smelting and converting can be used to optimize generating outputs of geochemically scarce elements by guiding the partitioning of geochemically scarce elements in smelting and converting (Chen, Zhang and Jahanshahi 2013; Jak et al. 2019; Nakajima et al. 2011; Shihshin, Hayes and Jak 2018; Shuva et al. 2016; Sineva et al. 2021; Van Schalkwyk et al. 2018).

In smelting, Cu and other geochemically scarce elements present in concentrates are distributed over de metal (matte) phase, the slag phase and the gas phase. Elements distributed to the gas phase largely partition to recoverable smelter dusts. Determinants of the distribution of Cu and other geochemically scarce elements over the metal, slag and gas phase are (Bellemans et al. 2018; Chen, Zhang and Jahanshahi 2013; Chen et al. 2006; Forsén, Aromaa and Lindström 2017; Moats, Alagha and Awuah-Offei 2021; Nakajima et al. 2011; Shihshin, Hayes and Jak 2018; Shuva et al. 2016; Sridhar, Toguri and Simeonov 1997; Wang et al. 2017b):

• the characteristics of the smelting operation (bath- or flash smelting; parameters such as temperature and partial oxygen pressure),

• the composition of concentrate and slagging materials, and

• the matte grade, that is the (weight)% of Cu in matte.

Jak et al. (2019) and Sukhomlinov et al. (2019) modeled the distribution of Ag, As, Au, Bi, Co. Cu, Ni, Pb, Pt group metals Sb, Sn and Zn over blister copper, slag and dusts, depending on converting conditions.

2.2.2.1. Geochemically scarce elements other than Cu partitioning to the metal phase in the production of matte and blister copper

In matte production and converting, noble metals, such as Ag, Au and Pt-group metals, partition to a large extent to the metal phase (Avarmaa et al. 2015; Dupont et al. 2016; Forsén, Aromaa and Lindström 2017; Schlesinger et al. 2011; Sineva et al. 2021; Sukhomlinov et al. 2019). Slag composition may be varied to increase the relative amounts of these elements ending up in the metal phase (e. g. Chen and Wright 2016). For several other geochemically scarce elements than the noble metals the percentages estimated to partition to matte are in Table 2. The ranges reflect differences in the smelting operations, the inputs therein and the matte grade.

Table 2. Estimated percentage of smelter input partitioning to matte for several geochemically scarce companion elements of copper (Moats, Alagha and Awuah-Offei 2021; Schlesinger et al. 2011).

Element	Estimated percentage of smelter input partitioning to matte
Bi	4–75%
Ni	70–80%
Sb	12–70%
Te	60–91%

A higher matte grade in flash smelting was reported to be associated with lower partitioning of As, Bi, Pb, Sn and Zn to matte (Forsén, Aromaa and Lindström 2017; Klaffenbach et al. 2021; Sineva et al. 2021). A higher Fe/SiO₂ ratio increases the partitioning of Pb, and lowers the partitioning of As, to matte (Klaffenbach et al. 2021). In bath smelting the partitioning of Bi, Ni and Sb to matte was estimated to be lower than in flash smelting (Moats, Alagha and Awuah-Offei 2021; Schlesinger et al. 2011). In bottom blown smelters (applying a variety of bath smelting), a higher matte grade is reportedly associated with lower partitioning to matte of Pb and Zn, and higher partitioning of As, Bi and Sb (Wang et al. 2017b). At Anglo Platinum, matte produced from ore concentrates, containing Cu, Ni and Pt group metals, is slowly cooled to form crystals containing Pt group metals and Cu-Ni alloy respectively. Thereafter, the matte is ground and subjected to magnetic separation (Fröhlich et al. 2017). The residue remaining after separation is treated to obtain Pt-group metals (Fröhlich et al. 2017). In flash converting of matte Se and Te decrease in the metal phase when temperature and CaO addition increase (Yu et al. 2021).

2.2.2.2. Geochemically scarce companion elements in anode slime and electrolyte

Ag, As, Au, Bi, Cu, Ni, Pb, platinum-group metals, Sb, Se, Sn and Te can be present in anode slime that originates in the electrolytic treatment of copper anodes generated in primary production (Dupont et al. 2016; Forsén, Aromaa and Lindström 2017; Liu et al. 2014). Anode slimes have been reported to contain the (weight) percentages for geochemically scarce elements presented in Table 3. The actual percentages are dependent on the characteristics of the pryrometallurgical process and the inputs therein.

Table 3. Reported (weight) percentages in anode slime for a variety of geochemically scarce companion elements (Hait, Jana and Sanyal 2009; Lu et al. 2015; Forsén, Aromaa and Lindström 2017; Jin, Hu and Hu 2018; Moats, Alagha and Awuah-Offei 2021).

Element	Reported (weight) percentage in anode slime
Ag	1.5–62%
As	0.4–6.8%
Au	−2.2%
Bi	−6.8%
Cu	0.5–53%
Ni	0.6–54.2%
Pb	1.7–31%
Pd	−0.35%
Pt	−0.08%
Sb	−16%
Se	0.2–46%
Sn	−5.3%
Te	−22%

A wide variety of technologies is available for recovering geochemically scarce elements from anode slime: hydrometallurgical, pyrometallurgical and combinations thereof. No comprehensive comparative evaluation regarding the resource efficiency of these technologies and combinations thereof could be found. Forsén, Aromaa and Lindström (2017) suggest that typical anode slime treatment currently includes: dissolving Cu, Ni and Te by oxidative acid leaching, roasting to recover Se by leaching, followed by smelting of residue and sequential electrolysis to recover the noble metals.

Substantial amounts of Ag and Au are currently commercially generated by anode slime processing (Forsén, Aromaa and Lindström 2017; Hedjazi and Monhemius 2014). Anode slime is currently also important as a source of Se and Te (e. g. Li et al. 2017a; Moats, Alagha and Awuah-Offei 2021; Xu et al. 2020b). For Te extraction from anode slime efficiencies of >99% have been reported in laboratory settings (Ma et al. 2015). For Se recoveries of up to 98% have been noted in laboratory settings (Hait, Jana and Sanyal 2009; Li et al. 2017a; Lu et al. 2015). The average Se recovery reported by 52 copper refiners was about 50% (Lu et al. 2015), suggesting that there is technical scope for increased co-production. Sb may also be produced from anode slime (Forsén, Aromaa and Lindström 2017; Moats, Alagha and Awuah-Offei 2021), but no significant production from this source has been noted (e. g. Moats, Alagha and Awuah-Offei 2021). The same holds for Bi (Moats, Alagha and Awuah-Offei 2021).

Apart from Cu, the elements Bi, Co, Ni, and Sb can dissolve in the electrolyte (Forsén, Aromaa and Lindström 2017). These elements can be removed from the electrolyte (Alam et al. 2007; Arroyo-Torralvo et al. 2017; Forsén, Aromaa and Lindström 2017). Ni and Co are reported to be commercially recoverable from electrolyte (Forsén, Aromaa and Lindström 2017; Tabelin et al. 2021). Sb and Bi are commercially recovered from electrolyte by several companies using ion-exchange, whereas in the case of Bi also molecular recognition technology is used commercially (Moats, Alagha and Awuah-Offei 2021).

2.2.2.3. Geochemically scarce companion elements present in slags of primary copper pyrometallurgy

Nakajima et al. (2011) and Sukhomlinov et al. (2019) have shown that, dependent on flux composition, in converting matte, the elements B, Cr, Ga, Ge, In, W and Zn may mainly, and Cd, Co, Ni, Re and Se may partially, partition to slags. Estimates for several geochemically scarce companion elements regarding the percentage of input with copper ore concentrate partitioning to slags are in Table 4. Table 5 presents reported concentrations of geochemically scarce companion elements in slags from smelters in a number of counties. The ranges reflect the differences in pyrometallurgical processing and the inputs therein.

Table 4. Estimates for several geochemically scarce companion elements as to the percentage of input with Cu concentrate partitioning to slags of primary copper pyrometallurgy (Ayres, Ayres and Rade 2002; Makuei and Senanayake 2018; Miganei et al. 2017; Moats, Alagha and Awuah-Offei 2021; Wang et al. 2017b).

Element	Estimated percentages of input with copper concentrate partitioning to slags
Bi	−30%
Pb	10–25%
Sb	5–71%
Se	5–30%
Te	7–30%
Zn	−64%

Table 5. Reported concentrations of geochemically scarce companion elements in slags from primary copper smelters in Canada, Chile, China, the Democratic Republic of Congo, Germany, Japan, Italy, Turkey, Uzbekistan and Zambia (Mikula et al. 2021; Tian et al. 2021).

Element	Range of reported concentrations in slags up to:
Co	2.0 (weight)%
Cr	0.57 (weight)%
Ni	0.48 (weight)%
Pb	1.02 (weight)%
Zn	8.0 (weight)%

As, Bi, Mo, rare earths and V have also been found in slags from primary copper smelters (Holland et al. 2019; Mikoda et al. 2019a; Tian et al. 2021). Though in practice slags are often discarded (Guo et al. 2016; Sanchez et al. 2004; Tian et al. 2021), reported concentrations in slags can offer commercial scope for reprocessing aimed at the recovery of geochemically scarce elements. Reprocessing of slags, aimed at increasing Cu yields, is commercially practiced in industry using flotation technology and slag cleaning furnaces (see section 2.1). Measurements regarding tailings of commercial slag processing found concentrations that may offer commercial scope for reprocessing generating geochemically scarce elements. Data regarding smelters in Finland, Namibia, Poland and Russia showed that slag tailings contained up to 0.08 (weight)% Co, up to 0.63 (weight)% Cu, up to 0.14 (weight)% Ni, up to 0.9 (weight)% Pb and up to 3.11 (weight)% Zn, whereas the presence of Bi, Mo, rare earths and Sb was also reported (Grudinsky et al. 2021; Holland et al. 2019; Mikoda et al. 2019a).

Data about the co-production of geochemically scarce companion elements from copper slags are patchy. In Zambia slags from primary smelters containing 1.5 (weight)% Cu and 1–3 (weight)% Co have been noted to be melted in an electrical furnace to obtain a metal containing Cu and Co (Sanchez et al. 2004). In Zimbabwe a top submerged lance smelter has been reported to be commercially used for the extraction of metals from a slag containing Co, Cu and Ni (Sanchez et al. 2004). The production of copper concentrate from slags has been reported to be at variance with the recovery of the companion elements Co and Ni (Shen and Forssberg 2003), but can be combined with the recovery of Mo (Sanchez and Sudbury 2013). Treatment in a slag cleaning furnace can be conducive to the recovery of geochemically scarce companion metals partitioning to the metal phase (e. g. Hughes 2000). Reductive sulfurizing smelting of slag with added CaO has been found conducive to Co recovery (Li et al. 2018b). Treatment in slag cleaning furnaces does give rise to dust formation. Such dust is likely to be landfilled but can also be treated by pyrometallurgy or hydrometallurgy to recover Pb and Zn (Perez-Moreno et al. 2018). Reduction of slag followed by magnetic separation offers scope for the recovery of Pb and Zn (Tian et al. 2021). Grudinsky et al. (2021) and Wan et al. (2021) studied sulfation roasting, followed by extraction with water, of respectively tailings of copper slag processing and copper slags and found technical scope for the extraction of Cu and other metals such as Co, Ni and Zn. Wan et al. (2021) concluded that a process using slag and waste sulfurdioxide might be commercially viable. If (bio)hydrometallurgy is applied to treatment of (sufficiently comminuted) slags or tailings from slag processing, there is scope for the recovery of Cd, Co, Cu, Mo, Ni, Pb, rare earths, V and Zn (Arslan and Arslan 2002; Banza, Gock and Kongolo 2001; Fomchenko and Muravyov 2020; He et al. 2021; Kaksonen et al. 2017; Kinnunen et al. 2020; Mikoda, Potysz and Kmiecik 2019b; Potysz and Kierczak 2019; Potysz, van Hullebusch and Kierczak 2018; Tian et al. 2021; Turan, Sari and Miller 2017). All in all, there is technical scope for substantially increasing the production of geochemically scarce elements from slags and slag tailings.

2.2.2.4. Geochemically scarce companion elements present in dusts of primary copper pyrometallurgy

5–10% of the mass entering smelters processing ore concentrates may end up in dusts from primary copper pyrometallurgy (Ha et al. 2015). Estimated partitioning of several geochemically scarce companion elements present in copper concentrates to dusts of primary copper pyrometallurgy is presented in Table 6. The ranges reflect differences in pyrometallurgical processing and the inputs therein.

Table 6. Estimated partitioning of several geochemically scarce companion elements present in copper concentrates to dusts of primary copper pyrometallurgy (Ayres, Ayres and Rade 2002; Chen, Zhang and Jahanshahi 2013; González-Castanedo et al. 2014; Wang et al. 2017b; Avarmaa et al. 2019; Makuei and Senanayake 2018; Moats, Alagha and Awuah-Offei 2021).

Element	Estimated partitioning of geochemically scarce companion elements present in copper concentrates to dusts of primary copper pyrometallurgy
As	35–90%
Bi	15–96%
Cd	−80%
Hg	−90%
Pb	−85%
Sb	5–72%
Se	−10%
Sn	−65%
Te	2–90%
Zn	−30%

When matte is low-grade, more of Sb tends to end up in smelter dust (Forsén, Aromaa and Lindström 2017). Sukhomlinov et al. (2020) have pointed out that Ag and Sn may to a large extent partition to dusts of anode furnaces. Dusts of primary copper pyrometallurgy have furthermore been reported to contain substantial amounts of Bi, Ge, In, Ni and Re (Ayres, Ayres and Rade 2002; Cusano et al. 2017; Dupont et al. 2016; Forsén, Aromaa and Lindström 2017; González et al. 2017; Moldabayeva et al. 2015; Montenegro, Sano and Fujisawa 2013; Selivanov et al. 2014). Some Li and Rb has also been found in dusts of primary copper pyrometallurgy (González et al. 2017). Best available technology (Cusano et al. 2017) should be used to prevent the emission of smelter dusts, as such dusts are a threat to human health (Fry et al. 2020; González-Castanedo et al. 2014; Sorooshian et al. 2012).

Published data about the current co-production of geochemically scarce elements from primary copper pyrometallurgy dusts are patchy. As pointed out in section 2.1, dusts can be internally recycled and such recycling is practiced in industry, to increase copper yields. However, dust recycling may lead to the accumulation of elements which may be considered unwanted such as As, Bi, Hg, Ni, Pb, Sb and Zn (Agorhom et al. 2015; Dupont et al. 2016; Ha et al. 2015; Montenegro, Sano and Fujisawa 2013). Efforts to prevent such accumulation have led to treatment of smelter dusts before recirculation (e. g. Montenegro et al. 2013; Liu et al. 2018). Dusts may be pretreated by roasting and centrifugation (e. g. Okanigbe et al. 2019). They can be treated by (bio)hydrometallurgy, pyrometallurgy and combinations thereof (Bakhtiari et al. 2011; Dupont et al. 2016; Gao et al. 2021; Grudinsky, Dyubanov and Kozlov 2019; Ha et al. 2015; Liu et al. 2018; Lucheva, Iliev and Kolev 2017; Montenegro, Sano and Fujisawa 2013; Morales et al. 2010; Sabzezari et al. 2019; Shibayama et al. 2010; Vitkova et al. 2011; Xu et al. 2020a; Zhang et al. 2019b). Such treatments provide technical scope for the co-production of Cu and companion elements such as Ag, Bi, Cd, Co, Ge, In, Li, Ni, Pb, Re, Sb, Se, Sn and Zn (Chen et al. 2012; Gao et al. 2020, 2021; González et al. 2017; Grudinsky, Dyubanov and Kozlov 2018, 2019; Ha et al. 2015; Morales et al. 2010; Ozberk et al. 1995; Paz-Gómez et al. 2020; Sabzezari et al. 2019; Vitkova et al. 2011; Xu et al. 2020a; Zhang et al. 2019b). Dusts may be treated at the same location (Government of Canada 2018; Ozberk et al. 1995) or be sold to facilities elsewhere, e. g. to smelters for Pb and/or Zn production (Cusano et al. 2017; Grudinsky, Dyubanov and Kozlov 2019; Ha et al. 2015). Re may be commercially produced from copper smelter dusts. One primary copper producer does this exploiting hydrometallurgy and ion exchange technology (Sole, Mooiman and Hardwick 2017). Sb may be commercially produced from copper smelter dusts too (Dupont et al. 2016; Dupont and Binnemans 2017). This is practiced at the Rönnskär plant of Boliden (Sweden) (Shuva et al. 2017), which has been reported to process an ore containing about 0.16 (weight)% Sb (Weidenbach, Dunn and Tao 2016). Such practice is so far apparently rare (Dupont and Binnemans 2017). There is technical scope for substantially increasing the generation of geochemically scarce elements from dusts of primary copper pyrometallurgy.

2.2.3. Geochemically scarce companion elements in hydrometallurgical processing of copper ores

Commercial heap leaching processes often focus on Cu recovery, and not on geochemically scarce co-products. Exceptions can be found in the African Copperbelt (Democratic Republic of Congo and Zambia), where both Cu and Co are reported to be recovered following heap leaching (Alexander et al. 2018; Sole et al. 2019). Co recovery from leachate is often by precipitation following the addition of hydroxides, but may also be by solvent extraction or (chelating, ion exchange) resins (Botelho, Dreisinger and Espinosa 2019; Sole et al. 2019). There is technical scope for the recovery of other companion elements. Rhenium (Re) can be found in heap leaching solutions (Sole, Mooiman and Hardwick 2017). Use of activated carbon and ion exchange have been proposed as extraction technologies for such Re (Sole, Mooiman and Hardwick 2017; Waterman 2013), but no commercial extraction of Re from heap leaching-based solutions has been reported (Sole, Mooiman and Hardwick 2017). Orrego, Hernández and Reyes (2019) noted the presence of substantial amounts of Mo and U in Chilean copper leaching solutions but did not report the commercial recovery thereof. One would furthermore expect that Cd, Cr, Ni, rare earths, V and Zn may be mobilized by bioleaching (Liu et al. 2020; Mikoda, Potysz and Kmiecik 2019b). As noted before, the geochemically scarce metals Cd, Cr, Ni, Zn tend to accumulate during the recycling in heap leaching of raffinate remaining after solvent extraction of Cu (Liu et al. 2020). This is likely to reduce the costs for the recovery of such metals from recycled raffinate using e. g. ion exchange technology or solvent extraction, if compared to recovery from virgin leachate. Supported liquid membranes ionic solvents, supported ionic liquid phases, and biosorption may also be used for recovering geochemically scarce co-products from leachates (e. g. Spooren et al. 2020), but no current commercial application thereof could be identified

Contained hydrometallurgical processes which may extract more than one metal from ore concentrates or fines from crushed ores have been developed (e. g. Akbulut et al. 2018; Alexander et al. 2018; Halinen 2015; Sole et al. 2019; Spooren et al. 2020: Vardanyan et al. 2019). In Africa there is widespread application of agitated tank leaching to co-extract Cu and Co (Sole et al. 2019). Also, stirred tank-based leaching is applied to copper-gold ore concentrates (Dreisinger 2016). Contained reductive leaching of oxidized copper-cobalt ores, using SO₂, is practiced in the African Copperbelt (Sole et al. 2019). Tshipeng, Tsamala-Kanaki and Kime (2017) have discussed the optimization of such leaching. Furthermore, pressurized oxidative acid leaching to co-produce Co and Cu is applied in the African Copperbelt (Alexander et al. 2018; Sole et al. 2019). There is also technical scope for the recovery of other geochemically scarce elements. Pressurized oxidative acid leaching processes can generate soluble sulfates from base metals other than Cu and Co, such as Ni, Pb and Zn, and may produce a residue from which precious metals Ag and Au, and possibly also Pt group metals, can be recovered (Dreisinger 2015; Jorjani and Ghahreman 2017; Safarzadeh, Horton and Van Rythoven 2018). Sequential solvent extractions can be used when more base metals are present in the leachate (Hedrich et al. 2018).

Pressurized oxidative leaching of copper ore concentrates can furthermore lead to co-leaching of Se and Te (Mokmeli, Dreisinger and Wassink 2015). This appears to have triggered efforts to reduce concentrations of Se and Te in Cu-electrowinning solutions, rather than efforts to generate Se and Te as co-products (Mokmeli, Dreisinger and Wassink 2015). Re is also solubilized in pressurized oxidative acid leaching and can be recovered (Muruchi et al. 2019), but no commercial recovery from leachate has been reported. All in all, there would seem to be technical scope for a substantial increase in the generation of geochemically scarce elements in the hydrometallurgical processing of copper ores and copper ore concentrates.

There is also technical scope for extracting geochemically scare elements from residues of hydrometallurgical processing. Hubau et al. (2020) found that stirred-tank bioleaching can extract Cu, Co, Ni and Zn from residues remaining after heap leaching at a sulfidic polymetallic mine in Finland. Shengo et al. (2021) have shown that additional Cu and Co might be extracted by reprocessing residues from commercial agitated tank leaching of Cu and Co. Wu, Ahn and Lee (2021) obtained Au by hydrometallurgical reprocessing of residue from commercial pressure oxidation extraction of Cu from chalcopyrite ore.

2.3. Concluding remarks regarding sections 2.1. and 2.2

There appears to be substantial technical scope for increasing the production of geochemically scarce elements from copper ores. Developing flexible minerals- and residues processing plants may help in optimizing the production of geochemically scarce elements (Reuter 2016). As it stands, technology for commercially producing geochemically scarce elements has been much better developed for pyrometallurgy-based processing than for (bio)hydrometallurgical processing of Cu ores (also: Izatt et al. 2014). A major research and development effort to widen the scope for commercial (bio)hydrometallurgical extraction technologies for geochemically scarce elements may help in optimizing (bio)hydrometallurgy-based processing technology (also: Guezennec et al. 2017; Mahmoud et al. 2017). This regards especially (bio)hydrometallurgy for the extraction of copper ores and tailings from copper ore concentrators and for the extraction of slag tailings and dusts, preferably at low cost.

2.4. New technologies for extraction of geochemically scarce companion elements of copper

The extraction of geochemically scarce companion elements from copper ores should be an important matter in the future development of new technologies for copper ore processing. As pointed out in section 2.1, extraction of copper from ores by non-aqueous solvents is under investigation. Research has been reported regarding manganese nodules investigating the recovery of copper, nickel and cobalt, using 5,8-diethyl-7-hydroxy-6 dodecanone oxime in kerosene as extractant (Binnemans and Jones 2017). Reported solvometallurgical recovery efficiencies after reductive treatment of manganese modules at 600°C were 69% for copper, 44% for Co and 71% for Ni under laboratory conditions (Binnemans and Jones 2017). Extraction of oxidized metals by deep eutectic solvents has been reported for Co, Cr, Ni, V and Zn, besides Cu, and also for Ag, Au and rare earths (Spooren et al. 2020). A major research and development effort would seem necessary to establish the techno-economical potential of solvometallurgical extraction of copper ores and copper ore processing residues. In research regarding electrolysis of CuS, MoS₂ and ReS₂ in molten sulfide salts Sahu, Chmielowiec and Allanore (2017) found that phase-separated grains of Mo and Re in liquid copper can be generated. This suggests there may be technical scope for the co-production of these elements in molten salt technology.

2.5. Valorization of matrix materials that are co-outputs of copper ore processing

Applications considered for valorization of matrix materials that are co-outputs of copper ore processing mainly regard building and construction materials in which matrix materials are used as substitutes for conventional materials such as gravel and sand. Most research has been done on matrix materials in tailings from copper ore concentrators and slags.

Valorizations of copper ore concentrator tailings by their application in construction and building materials (cement, geopolymers, bricks, paving stones, road base materials) have been proposed (Ahmari and Zhang 2012, 2013: Ince et al. 2020; Jian et al. 2020; Lam et al. 2020; Manjarrez and Zhang 2018; Wang et al. 2021). As the tailings contain geochemically scarce elements that after release may have negative environmental impacts (cf. section 2.2.1), comprehensive testing is necessary under a variety of conditions reflecting the conditions to which applied materials can be exposed. Testing should not only regard newly produced materials but should also include aging, wear, tear and recycling of these materials (e. g. Engelsen, van der Sloot and Petkovic 2017; Kosson et al. 2004). Also, radioactivity should be tested (e. g. Kubissa et al. 2020: Wang, Wang and Hang 2020).

Unfortunately, only very limited testing of leaching from construction and building products containing copper ore concentrator tailings has been reported. Ahmari and Zhang (2013) studied leaching from newly-produced bricks containing copper ore concentrator tailings-based geopolymer-bricks and found limited leaching of As, Co, Cu, Mo and Zn. Jian et al. (2020) studied leaching of Cu, Cr, Mo and Zn from newly-produced cement clinker derived from copper ore concentrator tailings and found limited leaching of these elements. In both cases leaching conditions did not fully reflect the conditions to which these materials can be exposed. Nor were aging, wear, tear and recycling considered.

No studies could be found regarding the valorization of matrix material in residues remaining after hydrometallurgy of copper ores or copper ore concentrates. There probably is scope for valorization of such residues as Komnitsas et al. (2019), (2021) have shown that, after acid Ni and Co extraction, hydometallurgical residues of polymetallic saprolitic laterites might be applicable in the construction sector.

Slags from primary copper ore pyrometallurgy contain matrix materials derived from copper ores and can also contain substantial amounts of geochemically scarce elements, including radioactive elements (cf. section 2.2.2.3; also, Kinnunen et al. 2020; Kubissa et al. 2020). Valorizations of slags by application in or as building and construction materials are practiced on a substantial scale (e. g. Duester et al. 2016; Potysz et al. 2016). Proposals regarding the application of comminuted slags in construction and building materials include: cement (mortar), concrete, asphalt mixes and fire resistant geopolymers (Kinnunen et al. 2020; Mitiadis et al. 2020; Murani, Siddique and Jain 2015; Prem, Verma and Ambily 2018; Raposeiras et al. 2021; Siddique, Singh and Jain 2020; Wang, Wang and Hang 2020).

In view of the presence in copper slags of geochemically scarce elements that may have negative environmental impacts (e. g. Potysz et al. 2015, 2016; Duester et al. 2017, comprehensive testing of leaching would seem needed under a variety of conditions reflecting the conditions to which applied materials can be exposed, and not only of newly produced materials but also regarding aging, wear. tear and recycling of these materials (e. g.; Engelsen, van der Sloot and Petkovic 2017; Kosson et al. 2004). Also, radioactivity should be tested (Kubissa et al. 2020).

Leaching studies regarding slags from primary copper production showing significant releases of hazardous substances have been published (Duester et al. 2017; Potysz et al. 2015, 2016), but due to the absence of legal limits these have not impacted applications of such slags in hydraulic constructions (Duester et al. 2016). Only very limited testing results regarding building and construction materials including powdered copper slags have been published. Wang, Wang and Hang (2020) found leaching of As, Cu, Ni, Pb and Zn from newly produced cement containing powdered slag from a copper ore smelter, and also found that high additions of slag powder might be linked to a radiological hazard. He et al. (2021) studied newly produced ordinary Portland cement composites with copper slags and reported limited leaching of Cd, Cu, Ni, Pb and Zn. Aging, wear, tear and recycling were not included in the studies of Wang, Wang and Hang (2020) and He et al. (2021).

(Bio)hydrometallurgical leaching of slags and slag tailings from primary copper production can provide scope for the extraction of Cd, Co, Cu, Mo, Pb rare earths, V and Zn, cf. section 2.2.2.3. Kinnunen et al. (2020) showed that acid leaching of ground slag can substantially reduce the concentrations of Cu, Co, Pb and Zn, which can be recovered, and that the leached material can be successfully included in cement. Kinnunen et al. (2020) concluded on the basis of their experiments that complete valorization of copper slag is feasible. However, studies also including aging, wear, tear and recycling cement with leached slag under real-world conditions relevant to exposure of organisms would seem necessary to assess whether such valorization is also environmentally acceptable. Expanding studies on extracting geochemically scarce elements present in slags may provide technical scope for safe valorization of slag-derived materials.

Dusts from pyrometallurgical copper ore processing contain matrix materials (e. g. Caplan et al. 2021) and can be subjected to extraction of geochemically scarce elements (see section 2.2.2.4). However, no study could be found about the valorization of dusts from which geochemically scarce elements were extracted. Also, no study could be found regarding the environmental acceptability of solvometallurgically extracted matrix materials.

All in all, there is technical scope for the valorization of matrix materials but much research and development is needed as to the environmental acceptability of valorized materials.

3. Co-production of copper and other geochemically scarce elements from end-of-life products, especially end-of-life electric and electronic equipment and car electronics

In dealing with products containing copper that are considered for disposal the first choice for efficiently and functionally conserving copper and other geochemically scarce elements is to focus on options such as refurbishment, remanufacture, reuse and cascading use (Graedel et al. 2019). When such options are exhausted, processing end-of-life products for materials recycling can be conducive to the recovery of Cu and, where applicable, other useful substances (e. g. Knapp 2018; Sun et al. 2016).

The worldwide collection rate of end-of-life copper as copper input for materials recycling is estimated at about 45% (Glöser, Soulier and Espinoza 2013; Henckens and Worrell 2020). Copper used in transporting electricity and data commonly has high purity and may account for > 50% of copper in current use (Henckens and Worrell 2020). Such copper may be recovered from cables and wires with high efficiency (Catinean et al. 2021; Li et al. 2017b: Lu et al. 2019; Tanabe et al. 2019; Xu et al. 2019; Zablocka-Malicka, Rutkowski and Szczapaniak 2015). Copper recovered from cables and wires may be subjected to fire refining to obtain conductive copper rods (Li et al. 2017b). Dedicating facilities for secondary pyrometallurgy to the processing of copper from high purity applications can be conducive to closed loop recycling of copper applied in electricity and data transport.

As outlined in the Introduction, copper applied in products may become closely associated with other geochemically scarce elements. Increasingly important is the recovery of Cu and other geochemically scarce elements from end-of-life or waste electrical and electronic equipment (WEEE) (Sun et al. 2016; Tesfaye et al. 2017). Outputs derived from processing (mining) WEEE (e-wastes), such as waste printed circuit boards containing relatively large amounts of geochemically scarce elements (Priya and Hait 2018a), are rather often used for the recovery of Cu and other geochemically scarce elements (e. g. Andersson, Söderman and Sandén 2019; Hagelüken 2015; Priya and Hait 2018a).

Printed circuit boards are also present in end-of-life cars, but they are currently hard to recover and they tend to be left in cars that can be shredded (Andersson, Söderman and Sandén 2019; Hagelüken 2015). Modular redesign of cars to make electronics easily recoverable from end-of-life cars and design of such modules for recycling is a necessary step in the direction of near-zero waste processing. The current copper output fraction from automobile shredders is used for copper production and residues of automobile shredders have been considered for the recovery of geochemical scarce elements (e. g. Andersson, Söderman and Sandén 2019; Gunaratne et al. 2020a; Jordao, Sousa and Cavalho 2016; Kodier, Williams and Dallison 2018; Lewis et al. 2011; Restrepo et al. 2017; Singh and Lee 2016). There are major geographical differences in the amount of Cu that is recovered after shredding. Soo et al. (2017) estimated the recovery of Cu in the copper output of vehicle shredders in Australia at 4.1% and in Belgium at 91.8%. This suggests that there is often scope for increased Cu recovery by vehicle shredders. When a fee is paid, there is a potential for application of residues from automobile shredders in a copper smelter, processing both primary and secondary raw materials, but the scope for such smelting can be limited by the presence of other metals (e. g. Al and Hg) (Gunaratne et al. 2020a). In practice, automobile shredder residues are often landfilled (Gunaratne et al. 2020a,b).

A major problem for recovering Cu and other geochemically scarce elements from end-of-life electric and electronic equipment is the collection thereof (Gorman and Dzombak 2020; Reck and Graedel 2012). It has for instance been estimated that that less than 10% of end-of-life mobile phones arrive at industrial recycling facilities (Navazo, Villalba Mendez and Talens Peiro 2014; Welfens, Nordman and Seibt 2016). Reck and Graedel (2012) estimated a retrieval rate for end-of-life electric and electronic equipment in the European Union in the order of 25–40%. Closed-loop take-back, including deposit-refund, systems for end-of life products can improve collection rates (Gong et al. 2021) and may also be conducive to design improvements aimed at the recovery of geochemically scarce elements.

In the following the focus will be on formal recycling of end-of-life products containing Cu and other geochemically scarce elements, especially of end-of-life car electronics and electric and electronic equipment, which is legal processing as it is practiced in industrialized countries. Formal recycling may be assumed to be more efficient than informal recycling in recovering rare elements (e. g. Parajuly et al. 2018; Sthiannopkao and Wong 2013). Formal recycling of WEEE has also been recently reviewed by Tabelin et al. (2021); here additional information is provided.

Often it is argued that in mining complex end-of-life products, physico-mechanical technologies should be applied, as in copper ore processing (e. g. Vidyadhar and Das 2013; Mäkinen et al. 2015; Kaya 2016; Ruan and Xu 2016; Zeghoul et al. 2016; Wang et al. 2017a; Jeon et al. 2018; Nekouei et al. 2018; Ding et al. 2019). Formal physico-mechanical pre-processing of WEEE and end-of-life cars is currently associated with substantial losses of geochemically scarce elements (Andersson, Söderman and Sandén 2017). The more materials are interlinked and particles are interwoven in WEEE, the worse are the results of physico-mechanical separation, and more sequential separations lead to larger losses (Hagelüken 2006a,b). Chancerel et al. (2009) estimated that in formal pre-processing of WEEE (depollution, shredding and sorting) on average about 40% of Cu is lost. The recovery of rare earths from formal pre-processing has been estimated at less than 2% (Marra, Cesaro and Belgiorno 2018a). In current formal recycling of WEEE losses of about 30–88% have been reported for Ag, Au, Pd, Pt and Rh (Bigum, Brogaard and Christensen 2012; Marra, Cesaro and Belgiorno 2018a; Wellmer and Hagelüken 2015). Changing formal WEEE pre-processing by selectively removing components or modules from end-of-life products which are relatively rich in geochemically scarce elements, would be conducive to increased recovery of these elements. Additional improvement may come from designing those modules for recycling. Also, there is the option of forgoing shredding. For instance, the Umicore plant (Belgium) uses as an input for its operation, after removal of batteries, relatively small devices such as mobile telephones, digital cameras and laptop computers (Hagelüken 2006b). A benefit thereof is that the losses of geochemically scarce elements due to physico-mechanical pre-processing are avoided (Ding et al. 2019; Hagelüken 2006b). In the case of precious metals this benefit is not invalidated when trade-offs linked to the presence of other metals are considered (Hagelüken 2006b). The Boliden Rönnskär plant also processes ‘coarse´ WEEE (Lennartsson et al. 2018). On the other hand, the Dowa plant at Kosaka uses crushing and separation as pre-treatment for WEEE (Shuva et al. 2017). Similarly, the Mitsubishi Naoshima smelter has been reported to use crushed e-waste and WEEE shredder output (Ariizumi et al. 2016; Kawasaki 2014),

It has been argued that hydrometallurgy (including biohydrometallurgy) enables ‘sound, metal-selective and cost-effective processing to recover metals from WEEE’ (Isildar et al. 2018). Sethurajan et al. (2019) discussing hydrometallurgical recovery of critical and precious elements from preprocessed e-waste stated that ‘leaching and recovery of heavy metals is the best solution to meet the growing critical raw materials demand and also to reduce the environmental impacts caused by the WEEEs in the environment.’ However, Li, Eksteen and Oraby (2018a) have reviewed current hydrometallurgical recovery of metals from waste printed circuit boards. In view of their composition these are relatively promising objects for extracting geochemically scarce elements, including Cu. Li, Eksteen and Oraby (2018a) found that extraction with acids, cyanide-based chemicals, thiourea, thiosulfate, thiocyanate, halides and ammonia did not meet economic, technical and environmental requirements. A comparative life cycle study about the recovery of Ag, Au, Cu and rare earths from electronic waste by current hydrometallurgy and pyrometallurgy suggested that there was no clear ‘winner’ as to the life cycle environmental burden (Li et al. 2019). So far, there is only limited application of hydrometallurgy, subsidiary to pyrometallurgy in the processing of outputs of mining complex end-of-life products (e. g. Karhu et al. 2018). However, further research and development in the field may improve the scope for its application.

There is a substantial amount of published laboratory-scale biohydrometallurgy research regarding the extraction of geochemically scarce elements from (comminuted) printed circuit boards by bacteria and fungi (e. g. Priya and Hait 2017; Kumar, Sain and Kumar 2018; Priya and Hait 2018b; Xia et al. 2018; Gu et al. 2019; Li, Eksteen and Oraby 2018a; Liu et al. 2019; Samal et al. 2019; Krishnamoorthy, Ramakrishnan and Dhandopani 2021). In some of these studies (Gu et al. 2019; Liu et al. 2019) biohydrometallurgy was combined with mechanical activation. The studies show that in principle, besides Cu, a substantial range of geochemically scarce elements may be extracted with varying recoveries. These include, Ag, Au, Cd, Co, Ni, Pb, rare earths, V and Zn. There is also limited research on biohydrometallurgical extraction of shredder dust (Marra et al. 2018b). Biohydrometallurgical studies have been judged encouraging, when reviewed by Priya and Hait (2017) and this judgment remains defensible in view of the studies published after 2017. But a large concerted research and development effort would seem to be needed to assess whether industrial biohydrometallurgical production of a substantial range of geochemically scarce elements from mined complex end-of-life products is feasible (also: Priya and Hait 2017). Such an effort is also needed to assess the potential for the application of solvometallurgy.

Electrochemical extraction of Ag, Au, base metals and rare earths from electronic waste has been suggested. One process applies a mild oxidant (Fe³⁺) (Diaz, Clark and Lister 2017; Li et al. 2019). Another process is suspension electrolysis, generating metals (Yun et al. 2019). There is as yet no reported commercial application of such electrochemical technologies.

A further option is processing of mined WEEE and car electronics at elevated temperature. Substantial research has been published about treatment of printed circuit boards by pyrolysis (e. g. Williams 2010; Flandinet et al. 2012; Bidini et al. 2015; Riedewald and Sousa-Gallagher 2015; Kim et al. 2018; Khanna et al. 2018; Ulman et al. 2018). No commercial implementation of such pyrolysis-based treatments has been reported. Panda et al. (2020) presented laboratory-scale pyrolysis of printed circuit boards, followed by roasting the metal-rich pyrolysis residue in the presence of excess NH₄Cl, leading to the formation of metal chlorides to be dissolved in water. High extraction yields were reported for the base metals Cu, Ni, Pb and Zn, whereas the maximum recovery of Au and Ag was reported to be 42–43% (Panda et al. 2020). The relatively low recovery of the noble metals is a problem as the prices to be realized for extracted Ag and Au are important determinants of the commercial feasibility of a recovery process (Fizaine 2020).

Copper smelting at about 1200–1350 °C, followed by converting and fire refining, is widely used for the commercial recycling of geochemically scarce elements present in complex end-of-life electric and electronic equipment. The presence of relatively valuable elements (such as Au, Ag and Pt-group metals) in the outputs of mining complex end-of-life products can be conducive to the recovery of other geochemically scarce elements by pyrometallurgy (Zeng, Mathews and Li 2018). A potential trade-off between the recoveries in pyrometallurgy, when more than one element is recovered, has been reported for Pd and Ge (Shuva et al. 2017). Potential trade-offs also apply to Cu and the precious metals Ag and Au (Ghodrat et al. 2017) and to Au and Pd and on the other hand Ag (Avarmaa et al. 2019). There is presently no recycling technology that can simultaneously recover Ag and Ta (Ueberschaar et al. 2017). A solid basis of thermodynamic and kinetic data may help optimizing recoveries of geochemically scarce elements in pyrometallurgy (e. g. Shuva et al. 2017; Van Schalkwyk et al. 2018).

Outputs of mining complex end-of-life electric and electronic equipment containing copper can be fed to secondary smelters. The main steps in secondary pyrometallurgy of copper starting with end-of-life products containing copper are in Figure 2. Secondary copper smelters generate a copper output with about 80 (weight)% Cu (‘black copper’) (Ghodrat et al. 2019; Schlesinger et al. 2011). Secondary smelters include electric furnaces, submerged lance furnaces and bath melting reactors (Steinacker and Antrekowitsch 2017; Tabelin et al. 2021). Black copper can be added to converters producing rough copper or blister copper (Forsén, Aromaa and Lindström 2017; Ghodrat et al. 2017; Schlesinger et al. 2011). Rough copper can be added to anode furnaces (e. g. Ghodrat et al. 2017; Gregurek et al. 2018). Copper anodes can be subjected to dissolution and electrolysis to generate copper cathodes, as described in section 2.1.

Figure 2. Main steps in pyrometallurgical secondary production of copper (cathodes) from end-of-life products containing copper.

There are secondary copper smelters involved in co-producing a wide range of geochemically scarce elements (cf. Table 7). Ag, Au, Bi, Cu, In, Ir, Ni, Pb, Pd, Pt, Rh, Ru, Sb, Se, Sn, and Te are recovered in standard runs by the Umicore (Belgium). This company processes industrial residues and (mined) end-of-life products. Umicore applies copper, lead and nickel smelting, electrochemistry and hydrometallurgy (Hagelüken 2015; Hagelüken and Corti 2010; Kaya 2016; Khaliq et al. 2014; Stamp, Althaus and Wäger 2013). The Umicore plant is flexible, as it can recover e. g. Co, Ga or rare earths in special runs (Hagelüken 2015). But the Umicore plant is e. g. unable to recover tantalum (Ta) (Hagelüken 2015). Aurubis (Germany), where primary and secondary copper production share a refinery and a lead smelter is also present, is reported to recover Ag, Au, Bi, Ni, Pb, platinum group metals, Sb, Se, Te and Zn as co-products, using pyrometallurgy, electrochemistry and hydrometallurgy (Aurubis 2017; Karhu et al. 2018; Rocchetti, Amato and Beolchini 2018; Zschiesche, Ayhan and Antrelkowitsch 2018).

Table 7. Reported relatively rare elements produced by plants co-processing mined complex end-of-life products containing copper (references in text).

Plant	Reported output of relatively rare elements
Aurubis (Hamburg)	Ag, Au, Bi, Ir, Ni, Os, Pd, Rh, Ru, Pt, Sb, Se, Te, Zn
Boliden (Rönskär)	Ag, Au, Cu, In, Ni, Pb, Pd, Pt, Sb, Se, Zn
Dowa (Kosaka)	Ag, Au, Bi, Cu, Ir, Ni, Os, Pb, Pd, Pt, Rh, Ru, Sb, Se, Sn, Te
Glencore (Horne)	Ag, Au, Ir, Ni, Os, Pd, Pt, Rh, Ru, Se, Te
Mitsubishi (Naoshima)	Ag, Au, Cu, Pb, Pt, Se
Umicore (Belgium)	Standard run: Ag, Au, Bi, Cu, In, Ir, Ni, Pb, Pd, Pt, Rh, Ru, Sb, Se, Sn, Te Special runs: Co, Ga, rare earths

Alternatively, outputs of mining complex end-of-life electric and electronic products can be fed to the large pyrometallurgical plants processing copper ore concentrates (Forsén, Aromaa and Lindström 2017). Here the focus will be on plants that generate a relatively large number of geochemically scarce elements.

The Boliden Rönnskär plant (Sweden) co-processes secondary sources of Cu, including coarse and fragmented WEEE (Lennartsson et al. 2018; Rocchetti, Amato and Beolchini 2018). At the Rönnskär plant, e-waste is first processed in a secondary smelter; black copper, with or without slag, is subsequently added to the converter of the primary copper smelter (Lennartsson et al. 2018). Also, at the Rönnskär plant a lead smelter serves the production of WEEE-derived geochemically scarce elements (Lennartsson et al. 2018). The Boliden Rönnskär plant is reported to recover Ag, Au, Cu, Ni, Pb, Pt-group metals, Se and Zn (Andersson, Söderman and Sandén 2019; Khaliq et al. 2014). Dust from this plant is processed to recover Sb, In and Cu (Shuva et al. 2017). A 2017 report about the Rönnskär smelter stated that 28% of Ag, 38% of Au, 27% of Cu and 72% of Zn produced by the smelter originated in secondary raw materials (Boliden 2017). The Dowa Kosaka plant (Japan), co-processing Cu ore concentrate and e-waste, is able to recover Ag, Au, Bi, Cu, Ni, Pb, Pt-group metals, Sb, Se, Sn and Te (Shuva et al. 2017). In the case of Dowa, Bi, Pb, Sb and Sn are outputs of a lead smelter and refinery processing copper smelter residues (Mitsune and Satoh 2007). The Glencore Horne copper pyrometallurgical plant, co-processing copper ore concentrates and outputs of mining end-of-life products containing copper (including e-waste), co-produces Ag, Au, Ni, Pt-group metals, Se and Te (Glencore 2018). The Naoshima smelter of Mitsubishi (Japan) has been reported to co-process copper concentrate, crushed e-scrap and WEEE shredder output, and to recover Ag, Au, Cu, Pb, Pt and Se (Ariizumi et al. 2016; Kawasaki 2014).

The Aurubis, Boliden and Dowa plants are characterized by the presence of both copper and lead processing, with copper and lead functioning as collectors for other geochemically scarce elements (also: UNEP 2013; Van Schalkwyk et al. 2018). Umicore in addition also uses a nickel. smelter. The systematic use of copper and lead as collectors for geochemically scarce elements may be expanded by cooperation. It is e. g. not uncommon that pyrometallurgical plants processing copper ore outsource processing of copper smelter dusts to lead smelters (see section 2.2.2.4). It is also not uncommon that co-outputs of lead smelters (crusts, dusts, matte, speiss) are processed in pyrometallurgical plants processing copper (e. g. Clement, Wettlaufer and Scott 1986; Ha et al. 2015; Ibragimov et al. 2019; Kotarski et al. 1999).

As indicated in Table 7, processes used by companies such as Aurubis (Germany), Boliden (Sweden), Dowa (Japan) and Umicore (Belgium) do relatively well in generating geochemically scarce elements from WEEE. They should be considered for wider application. They, however, are limited in their performance if compared with the range of companion elements in electric and electronic equipment and car electronics presented in Box 2. They also might be improved as to their emissions. For instance, Sutliff-Johansson et al. (2021) found that processing of WEEE scrap by the Rönnskar plant of Boliden is associated with substantial Ta emissions. The Umicore plant processing WEEE in Belgium is a point source for As, Cd and Pb emissions to air which should be reduced (Vlaamse Milieu Maatschappij 2018).

All in all, there appears to be substantial technical scope for increasing the production of geochemically scarce elements from end-of-life electric and electronic equipment and car electronics.

Only a few studies could be found specifically addressing the extraction of geochemically scarce elements from residues generated in the secondary production of copper. Ahmed, Nayl and Daoud (2016) and Xia et al. (2020) found that much Cu and Zn might be recovered from smelting slag of waste brass by leaching with sulfuric acid and ammonium chloride respectively. Wijenayake and Sohn (2020) studied the extraction of ZnO, for application in tires, from flue dust of a secondary smelter. Other geochemically scarce metals present in the flue dust (Cd 3.9 weight%, Cu 13.0 weight% and Pb 6.7 weight%) were not recovered in the study of Wijenayake and Sohn (2020). Valorization of residues emerging from the treatments studied by Ahmed, Nayl and Daoud (2016), Wijenayake and Sohn (2020) and Xia et al. (2020) was not addressed. In view thereof the valorization of residues from secondary copper smelters would require much additional research and development work.

4. Hurdles for the feasibility of near-zero waste production

Co-producing companion elements from copper ores and co-processing copper concentrates with wastes can have economic benefits. If there are substantial amounts of marketable companion elements present in copper ores (e. g. Mudd, Weng and Jowitt 2013a,b), there may be a financial case to consider ores containing copper ores as ores from which more geochemically scarce elements than copper may be extracted. Improving the case for such extraction may occur when copper concentrations in copper ores continue to decrease (Batterham 2013; Sverdrup, Olafsdottir and Ragnarsdottir 2019), and when interest in exploiting complex low-grade polymetallic ores increases (e. g. Mahmoud et al. 2017; Spooren et al. 2020). Co-production of companion elements from copper ores may have some protective effects against the fluctuations of individual metal prices (e. g. Afflerbach et al. 2014; Zemlick et al. 2017). Though there are, or can be, trade-offs as to yields in such co-production (e. g. Li et al. 2013; Reuter 2016; Shuva et al. 2016; Yin et al. 2017), financial benefits may well increase when more than one element is produced from copper ore (e. g. Dreisinger 2016; Lasheen et al. 2015; Sole et al. 2019). Pyrometallurgical co-processing of copper concentrates with mined end-of-life electric and electronic equipment can increase financial returns (Knapp 2018; Zeng, Mathews and Li 2018). Generating extracted matrix materials from ore processing residues which can be applied in construction and building materials may have financial benefits (e. g. Kinnunen et al. 2020; Spooren et al. 2020).

However, there are examples of company behavior that are at variance with progressing in the direction of near-zero waste primary copper production. Recoveries of geochemically scarce elements may well be lower than recoveries achievable by the application of best available commercial technologies (Mudd et al. 2013b; Pazik et al. 2016; Reuter and Kojo 2014; Sole et al. 2019). Furthermore, many smelters of copper ores restrict inputs of ‘penalty elements´. Such elements may cause hazardous outputs, copper anode passivation or excessive slag viscosity, or may negatively impact the properties of the final copper product (e. g. Agorhom et al. 2015; Dupont et al. 2016; Lane et al. 2016; Tabelin et al. 2021; Weidenbach, Dunn and Tao 2016). This, in turn, tends to lead to extraction and landfilling of penalty elements (e. g. Weidenbach, Dunn and Tao 2016). But the penalty elements Bi, Ni, Pb, Sb, Se, Te and Zn can also be marketable outputs of copper production. Handling companion elements such as Bi, Ni, Pb, Sb, Se, Te and Zn as potential co-products rather than as penalty elements, for instance by reprocessing extracted penalty elements, is likely to be conducive to the co-production of these elements.

A further problem for near-zero waste processing would seem that there may well be insufficient demand for potential product outputs of near-zero waste processing. Generating outputs that might be marketed entails costs. As shown in the previous sections there are cases in which the costs are such that the outputs can be commercially sold. This is not necessarily so for potential product outputs that are technologically possible within the framework of near-zero waste processing. In the following hurdles will be presented based on available studies regarding the following potential outputs: valuable geochemically scarce elements (section 4.1), outputs of matrix materials (section 4.2) and hazardous elements with potential supplies that exceed demand (section 4.3).

4.1. Valuable geochemically scarce elements

In view of life cycle losses, quantitatively important sources of geochemically scarce elements are concentrator tailings and complex end-of-life products, such as WEEE and end-of-life car electronic components. Geochemically scarce elements extracted from such sources are currently undifferentiated commodities: ´copper is copper and selenium is selenium´ (Casarin, Lazzarini and Vassolo 2020; Olvera 2021). So, to be commercially viable in current economic settings, production costs have to be competitive with those of conventionally produced elements. In practice this may be a hurdle. There are several cases in which Cu extraction by reprocessing of copper ore concentrator tailings was shown to be probably profitable at current copper prices (Chen et al. 2014; Drobe et al. 2021; Falagán, Grail and Johnson 2017). But there are also cases that commercial viability of the extraction of valuable elements from concentrator tailings has not been categorized as probable. Parviainen, Soio and Caraballo (2020) concluded that in the case of reprocessing (old) Au-Cu tailings from the Haveri mine in Finland Au recovery would probably be commercially viable, but that commercial viability was uncertain for Cu and Co recovery. Araya, Kraslawski and Cisternas (2020) suggested that heap leaching of the tailings in the Antofagasta region of Chile to recover V was likely to be commercially viable, whereas there would be a marginal commercial case for the recovery of rare earths (Araya, Kraslawski and Cisternas 2020). Araya, Kraslawski and Cisternas (2020) also concluded and that recovery of Co, Nb and Sb was not commercially viable. In addition, Araya et al. (2021) studied extraction of the rare earth Sc by hydrometallurgy with an estimated recovery of 80%. It was found that estimated capital costs of such extraction would probably be too high for the production of Sc at a competitive price. Araya et al. (2021) concluded that changing extraction technology in a way that capital costs are lowered would be necessary to make extracted Sc a viable product.

Current prices used in establishing the commercial viability of generating geochemical scarce elements by reprocessing may be at variance with future prices. Both higher or lower prices may occur. Lower prices, which would negatively impact commercial viability, may originate in large increases of production linked to reprocessing. This may be exemplified by Se and Te. From data presented by Moats, Alagha and Awuah-Offei (2021) one might estimate that extraction of Se and Te from copper ore concentrator tailings with an efficiency of about 80% may increase the worldwide production of Te by about a factor 100 and for Se by about a factor 50. In section 4.3 it will emerge that there is limited stockpiling for future use of As and Cd. It is conceivable to also stockpile Se end Te for future use. But in view of the wide disparity between potential supply and current demand the feasibility thereof is uncertain. Strongly increasing demand for Se and Te and cutting back the primary production of copper would be other options. But, again, the feasibility thereof is uncertain.

As to WEEE and electric and end-of-life car electronics, the concentrations of valuable geochemically scarce elements in mined wastes may well be too low for commercially viable recovery (Andersson, Söderman and Sandén 2017, 2019; Fizaine 2020). Improved recoveries of geochemically scarce elements from end-of-life electrical and electronic equipment, and cars, may be achieved by redesign of products and by selectively removing components or modules from end-of-life products which are relatively rich in specific geochemically scarce elements (e. g. Charles et al. 2020; Tabelin et al. 2021) and subjecting these to pyrometallurgical processing in smelters selected on the basis of component or module composition (also: Reuter, van Schaik and Ballester 2018). Such an approach may however still have its limitations, as it may well be that for specific elements (e. g. Ga, In, lanthanides, Ta, Y) concentrations remain too low (e. g. Andersson, Söderman and Sandén 2019; Liu and Keoleian 2020). This problem may perhaps be overcome by subjecting these components or modules to technologies that can upgrade and concentrate geochemically scarce elements, such as cryo-jacking, microwave ashing and treatment in solder baths (Charles et al. 2020), but applying these technologies adds to production costs which can negatively affect the competitive position of geochemically scarce elements generated from WEEE and electronic parts of end-of-life cars. If the valuable geochemically scarce elements produced from copper ores and end-of-life products remain undifferentiated commodities (Casarin, Lazzarini and Vassolo 2020; Olvera 2021), the hurdles presented in this section would seem formidable.

4.2. Matrix materials

As indicated in section 2.5, matrix materials derived from tailings and slags are substitutes for conventional construction and building materials such as sand and gravel. No characteristic of matrix materials could be found which would currently justify a relatively high price of matrix materials in their competition with conventional construction and building materials. Relatively low production costs of matrix materials would benefit their competitive position, but clean-up costs of matrix materials may well be at variance with low production costs. On the other hand, local scarcity of conventional construction and building materials would favor the competitive position of unconventional materials such as matrix materials (cf. Torres et al. 2021; Ulsen et al. 2021).

If compared with valuable elements as potential product outputs of processing copper ores, matrix materials are characterized by a much larger mass and by a relatively low price, with transport being a major factor in costs. Apart from exceptional cases such as the imports of construction and building materials by Dubai and Singapore, transport distances for gravel and sand are typically <100 km −300 km (Ioannidou et al. 2017; Torres et al. 2021; Ulsen et al. 2021). In view of such transport distances and the large volumes of matrix materials that are currently non-product outputs, large markets within 300 km would seem important for the viability of matrix materials as product outputs for application in construction and building. This is a substantial hurdle in cases that primary processing is remotely located and/or in thinly populated areas.

4.3. Hazardous elements

As and Cd are two hazardous elements that can occur in inputs used for the primary and secondary production of copper (Arslan, Djamgoz and Ukün 2016; Kammel, Göktepe and Oerlmann 1987; Sakoor et al. 2017; Schwartz, Omaynikova and Stocker 2017; Shi et al. 2020; Vermeulen et al. 2015; Yang et al. 2013). In both cases the potential supply of these elements as emerging from primary and secondary metallurgy is higher than demand (Karhu et al. 2018; Mudhoo et al. 2011; US Geological Survey 2020). Against this background in principle two options may be considered: stockpiling for future use and indefinite sequestration. Some stockpiling for future use is an option that has been chosen for Cd in the European Union (Karhu et al. 2018). When Cd is considered to be an element of strategic or critical importance for the economy (e. g. Randive and Jawadand 2019), this can be conducive to such stockpiling. Some stockpiling of As for future use has been reported for China (US Geological Survey 2020). In the case of sequestration in stable substances to be stored indefinitely in landfills, there is still a waste, but one might consider risk of storage in landfills to be removed. Long-term stability of sequestered cadmium has not been much studied. Tian et al. (2020) found stability of Cd sequestration by suspended ferrous sulfide nanoparticles for 717 days under aerobic and anaerobic conditions with pH kept at 7. However, it is unlikely that the condition of pH 7 can be indefinitely met in landfills. Sequestration has been much discussed for As. However, finding stable substances to indefinitely sequester As appears far from easy. Traditionally, Fe³⁺ and arsenate were co-precipitated in pools for this purpose. Precipitates with a high Fe/As ratio are stable under oxidizing conditions at pH 4 and in the presence of heavy metals (Riveros, Dutizac and Spencer 2013). It is highly unlikely that such conditions can be indefinitely met in landfills. Li et al. (2020) have advocated immobilization of As using copper slag. This gives rise to the formation of silica gel coated-FeAsO₄. (Li et al. 2020). However, silica gel is not indefinitely stable when exposed to water (Sögaard et al. 2018). It has furthermore not been proven that the stabilization of As by interaction with sludge from a paper factory, as proposed by Morales et al. (2010), provides sequestration of As in case of indefinite storage in landfills.

A crystalline octagonal variety of FeAsO₄(.2 H₂O), scorodite, is considered to have a greater thermodynamic stability than traditional Fe/arsenate co-precipitate, and has often been proposed as a candidate for the stable sequestration of As (e. g. Coudert et al. 2020; Okibe et al. 2014; Riveros, Dutizac and Spencer 2013; Rong et al. 2020; Schwartz, Omaynikova and Stocker 2017; Sun et al. 2018). However, exposure to natural waters or atmospheric weathering can degrade scorodite to goethite inducing the release of As. The stability of scorodite in landfills is limited to aerobic conditions and a pH of 2–5 (e. g. Nazari, Radzinski and Ghahreman 2017; Yuan et al. 2017). Under reducing condition and at a pH<2 or >5 there is substantial solubility of As from scorodite (Langmuir, Mahoney and Rowson 2006; Nazari, Radzinski and Ghahreman 2017; Paktunc and Bruggeman 2010; Yuan et al. 2017, 2016). Reducing conditions and pHs <2 or >5 may well occur over indefinite periods of storing scorodite in landfills. Moreover, As may be released from scorodite by iron reducing bacteria such as varieties of Shewanella and Desulfururomonas (Papassiopi, Vaxevanidou and Paspaliaris 2003; Revesz, Fortin and Pactunc 2015, 2016; Wang et al. 2016).

In view of the oxic conditions and the limited pH range required for scorodite stability, a variety of coatings have been proposed to reduce the release of As by scorodite under anaerobic conditions and at pH >5 and <2. These include coatings of ferric hydroxide sulfate, of Al or Ca phosphate, of Al (OH)₃/AlOOH, and of silicate-derived gels (Adelman, Elouatik and Demopoulos 2015: Coudert et al. 2020; Guo and Demopoulos 2018; Ke and Liu 2018; Ke, Song and Liu 2018; Lagno et al. 2010; Leetma et al. 2016; Nazari, Radzinski and Ghahreman 2017). Silicate-derived gels are not indefinitely stable when exposed to water (Sögaard et al. 2018). Moreover, Adelman, Elouatik and Demopoulos (2015) noted that the silicate-derived gel they studied (at pH 6) only marginally affected, or even increased, the release of As from scorodite due to ion exchange. The As containment by Al or Ca phosphate coatings may not remain protective in the long run due to loss of phosphate (Lagno et al. 2010; Leetma et al. 2016). Reduced leaching of As from scorodite coated with ferric hydroxide sulfate seems linked to preferential solution of the latter (Ke and Liu 2018), which strongly suggests that the reduction of As leaching from scorodite by this coating will not be indefinite. Coudert et al. (2020) pointed out that the long-term effect of coating of scorodite with Al(OH)₃/AlOOH should be further investigated. Arsenical natroalunite has been proposed as a more stable alternative to scorodite (Xu et al. 2020a). But testing such stability (Xu et al. 2020a) has so far been very limited and has not covered all conditions that may be associated with indefinite storage.

Other options for sequestration of As that have been suggested include vitrification, generating a glass, and cement-based solidification/stabilization technology. Solidification/stabilization technology is at risk of As release after long periods of time (ZZhao et al. 2017). Glass is metastable, subject to coarse-graining over time (e. g. Brito and Wyart 2006), and thus also appears to be at risk of As release over long time periods.

So far, the feasibility of indefinite sequestration of Cd and As for storage in landfills has not been demonstrated. A major research and development project aimed at the safe handling of hazardous element such as As and Cd would seem to be needed.

4.4. Valuing the hurdles

In view of the hurdles presented here, the feasibility near-zero waste production of copper and its geochemically scarce companion elements from copper ores, WEEE and end-of-life car electronics is uncertain.

5. Conclusions

This paper considered the feasibility of near-zero waste production of copper and its companion elements. It can be concluded that there is technical scope for significant progress in toward near-zero waste processing of copper ores and end-of-life electric and electronic equipment and car electronics. The generation of concentrates can be improved using e. g. geometallurgical modeling, sensor-based sorting, improved flotation and machine learning. Use of kinetic and thermodynamic data regarding smelting and converting can assist optimizing the co-production of geochemically scarce elements in the pyrometallurgical processing of copper ore. There is substantial technological scope for increasing the production of geochemically scare elements and cleaning-up matrix materials by improved processing and reprocessing of residues from concentrators and pyrometallurgical processing of copper ore. Developing flexible minerals- and residues processing plants may help in optimizing the co-production of companion elements. There is a case for narrowing gaps in knowledge which may be conducive to progress toward near-zero waste production. The research and development efforts to narrow these gaps are highlighted in Box 3. A major research and development effort to widen the scope for commercial (bio)hydrometallurgical extraction technologies may be conducive to optimizing (bio)hydrometallurgy-based primary processing technology producing geochemically scarce elements and cleaned-up matrix materials. This especially regards (bio)hydrometallurgy, at preferably low cost, for the extraction of copper ores, of copper ore tailings, and of dusts, slags and slag tailings from primary pyrometallurgy. Assessing the techno-economical scope for solvometallurgical processing of copper ores and copper ore processing residues for the production of a substantial range of geochemically scarce elements also requires a major research and development effort. Much research and development work is furthermore needed as to the environmental acceptability of valorized matrix materials. In developing new copper extraction technologies, the co-production of other geochemically scarce elements should be an important matter.

Well-considered modularization of products, design of modules for recycling and changes in the pre-processing of end-of-life products can help in improving the generation of geochemically scare elements from end-of life car electronics and electric and electronic equipment. A comparatively large variety of geochemically scarce elements may be recovered when both lead and copper serve as collectors in smelting mined car electronics and end-of life electric and electronic equipment. Process design aimed at obtaining outputs which can be efficiently processed by others into usable elements and guidance of smelting and converting by data about kinetics and thermodynamics can be conducive to improved recovery of geochemically scarce elements. The same holds for flexible smelters dealing with the outputs of mining WEEE and end-of-life electric car electronics. Slags and dusts from secondary pyrometallurgical copper production should be reprocessed for the extraction of geochemically scarce elements and the generation of sufficiently clean residual materials. This requires much additional research and development. A concerted research and development effort is needed to assess whether industrial (bio)hydrometallurgial and solvometallurgical production of a substantial range of geochemically scarce elements from mined complex end-of-life products is feasible.

Financial benefits of increasing the output of geochemically scarce elements and of matrix materials are known and the downward trend of Cu concentrations in copper ores and the increased consideration of exploiting polymetallic ores may be conducive to an increased interest of primary copper producers in the co-production of other geochemically scarce elements and of marketable matrix materials. However, to achieve substantial progress toward near-zero waste processing there are also hurdles to be overcome. Such hurdles regard marketability of outputs, the safe handling of hazardous elements and company behavior. In view of these hurdles and of the uncertainties about the techno-economical scope for improved processing, the feasibility near-zero waste production of copper and its geochemically scarce companion elements from copper ores and end-of-life car electronics and electric and electronic equipment is uncertain.

Acknowledgments

The comments by three anonymous reviewers are gratefully acknowledged.

Disclosure statement

The author declares that he has no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Funding

The was no funding from outside the University of Amsterdam.