Authors: Grace Inman, Ikenna C. Nlebedim, Denis Prodius (Critical Materials Institute, Ames Laboratory (US DOE) Ames, USA

1. Battery Metals

The following metals to be discussed play a vital role in rechargeable battery technologies, of which the market is rapidly expanding in conjunction with the popularity of Electric Vehicles (EVs). Most EVs are powered by Li-ion batteries. The lithium-ion cell consists of an anode, cathode, current collectors, separator, and a nonaqueous electrolyte. The anode is usually granite, and the cathode is a lithium-based oxide, such as LiCoO

, LiMn

O

, Li

MnO

, and LiNiMnCo

. Therefore Li, Mn, Co, and Ni markets have been evaluated to account for battery material consumption.

According to Olivetti et al. [48], most lithium-ion materials are expected to meet near future demands. However, concern over the rapid adoption of electric vehicles may strain the supply of some battery-grade materials. In 2019, 100% of the U.S. manganese and 78% of cobalt supplies were imported, leaving the metal supply vulnerable to geopolitical factors, especially with the Democratic Republic of Congo producing 60% of the global supply of cobalt [49].

Lithium-ion batteries are popular due to their lightweight, high-energy density, and long-life cycles [49,50]. However, the technology is still improving, and the future of Li-ion batteries may require lower Ni content to optimize energy density. As a result, the Global EV Outlook 2020 estimates that the amount of electric vehicle batteries retired by 2030 will be roughly equivalent to the current annual production of 100–120 GWh. In addition, they emphasize the need for battery reuse and recycling to decrease environmental liability and promote sustainable end-of-life practices [49].

1.1. Lithium

Primary sources of lithium include pegmatites and brines, with salt lakes being the top exploitable resource. However, lithium in salt lake brines is always accompanied by magnesium, and separating the two elements poses quite a challenge due to the elements’ similar chemical properties. Current Li/Mg separation methods are time and reagent intensive, generate large amounts of waste, and have low Li recovery efficiency. IL systems are being developed to improve lithium recovery process.

Li et al. showcased the use of binary IL extractants for Mg removal to cut down on acid and base consumption [50]. [A336][V10] was synthesized from Aliquat 336 and Versatic Acid (a mixture of carboxylic acids) to be used in a three-stage counter-current extraction that boasted almost complete removal of magnesium and only 10% co-extraction of Li. Versatic acid was used because of its ability to extract magnesium impurities from concentrated lithium solution in a previous study [50]. Accordingly, [V10]

acted as the de-protonated acidic extractant molecule.

The binary IL was diluted with p-cymene, a bioderived solvent, and proved to be more efficient on original brine than concentrated brine in an acid-based extraction where the reactions are driven by pH. Whereas with binary extractants, the reactions are guided by the common ion effect and the ion, which is chloride in this case. The binary extractants dependency on the common ion effect has been the major obstacle for implementing binary extractants because the salt concentration of most feeds, including seawater, is not sufficient to drive extraction. However, salt lake brines have high salt concentrations; hence, they present a desirable scenario for binary extractants.

Recent findings published in Angewandte Chemie (Applied Chemistry) [51] led to the developing of novel macromolecules for targeted complex coordination and selective extraction. For example, the Sessler group reported a hemispherand-strapped calix[4]pyrrole ligand for selective binding to Li

salts. Further research demonstrated that the molecule could operate successfully in solid-liquid or liquid-liquid phase extraction.

The Sessler process was mentioned in a review on metal separation for recycling, and the authors’ only critique was the use of toxic solvents, such as chloroform [52]. The main accomplishment of ILs is as a greener substitute for traditional organic solvents. Combining ILs with this novel extractant would make the perfect combination. Sadly, there are likely many scenarios where ILs could be applied for exceptional process optimization, but the relative novelty of the field has prevented it thus far.

The study of ILs is often paired with supported liquid membrane (SLM) systems to improve efficiency, as seen by Zante et al. in their efforts to recover Li from brine selectively. Supported ionic liquids feature an organic phase immobilized into the pores of polymeric support placed between the feed and stripping aqueous solutions. In this work, a hydrophobic porous polyvinylidene fluoride (PVDF) membrane was impregnated with a mixture of [C

min][Tf

N] and TBP. The benefits of an SLM include a continuous permeation situation where extraction and stripping overlap. However, the lack of stability of SLM currently limits their scale-up potential [53].

As exhibited by SLMs and the earlier magnetic nanoparticles, ILs can be placed in or on many different structures. Liu et al. attempted a similar approach with TBP [C

min][Tf

N] [54]. Rather than impregnating a SLM, they used an electrodialysis system with a sandwiched liquid membrane consisting of the organic lithium-ion carrier and introduced two cation exchange membranes. The inner membranes acted as physical barriers between the feeding brine and receiving solution[54].

This system displays a great perspective on the application for Li recovery from Mg/Li brines and offers potential minimal impact on the environment with excellent Li-ion selectivity. Although, in both cases, a non-fluorinated, cheaper IL with similar efficiencies could have been used. The electrodialysis can also be applied towards lithium extraction from seawater, which, in the past, has not been considered feasible due to the low salt concertation [54].

1.2. Cobalt

Cobalt (Co) is the most critical element in lithium-ion batteries due to its vulnerable supply chain. Due to this, many papers have been published focusing on cobalt recovery from battery cathodes and mixed cobalt and nickel waste.

Most leaching processes, at least for nickel laterites and sulfides, extract both cobalt/nickel and result in a mixed Co/Ni solution that needs to be separated [60]. A favored industrial process for the selective solvent extraction of cobalt from nickel is the use of phosphinic acid derivatives, such as Cyanex 272. However, Cyanex 272 tends to extract impurities, such as Cu, Zn, Fe, Cd, Ca, Mg, and Mn, along with cobalt [60], and thus there is a need for more selective processes.

1.3. Nigurcke 1

) ^[7].

Fil and Manguranese 1. Schematic diagram of the electrodialysis device. (A) An assembled cell and (B) the placement of electrodes and CEMs in the cell. Reprinted from Desalination, 474, G. Liu, Z. Zhao, L. He, Highly selective lithium recovery from high Mg/Li ratio brines, 114185, Copyright (2020), with permission from Elsevier.

This system displays a great perspective on the application for Li recovery from Mg/Li brines and offered potential minimal impact on the environment with excellent Li ion selectivity. Although, in both cases, a non-fluorinated, cheaper IL with similar efficiencies could have been used. The electrodialysis can also be applied towards lithium extraction from seawater, which, in the past, has not been considered feasible due to the low salt concertation ^[7].

1.2. Cobalt

Cobalt (Co) is the most critical element in lithium-ion batteries due to its vulnerable supply chain. Due to this, there have been many papers published focusing on cobalt recovery from battery cathodes and mixed cobalt and nickel waste.

Most leaching processes, at least for nickel laterites and sulfides, extract both cobalt, nickel and result in a mixed Co/Ni solution that needs to be separated ^[8]. A favored industrial process for the selective solvent extraction of cobalt from nickel is the use of phosphinic acid derivatives, such as Cyanex 272. However, Cyanex 272 tends to extract impurities, such as Cu, Zn, Fe, Cd, Ca, Mg, and Mn along with cobalt ^[8], and thus there is a need for more selective processes.

1.3. Nickel and Manganese

Cobalt, nickel, and manganese are also components of lithium battery cathodes. The next few studies have taken a more holistic approach to lithium-ion battery recycling, attempting to separate all potential critical metals from simulated end-of-life batteries. Othman et al. ^[9] created a feed solution consisting of Cu, Mn, Ni, and Li, as reflecting typical cathode material. Tetraoctylphosphonium oleate [P₈₈₈₈][Oleate], a functionalized fatty-acid-based IL with selective extractive ability influenced by pH, was chosen due to the hydrophobic nature of both the cation and anion.

The bulky nature of the oleate-ion means that even when deprotonated, the ion remains very hydrophobic, which prevents the loss of the anion to the aqueous phase, a common problem in IL chemistry. At pH 5, cobalt, manganese and nickel were extracted with nearly 100% efficiency through interaction with the oleate ions, and Li was not extracted at all. At pH values lower than zero, Co and Mn maintained their high extraction efficiencies, but Ni was no longer extracted.

This phenomenon is due to the formation of tetrachloro-anion complexes in the presence of excess Cl

Cobalt, nickel, and manganese are also components of lithium battery cathodes. The following few studies have taken a more holistic approach to lithium-ion battery recycling, separating all potential critical metals from simulated end-of-life batteries. Othman et al. [65] created a feed solution consisting of Cu, Mn, Ni, and Li, as reflecting typical cathode material. Tetraoctylphosphonium oleate [P₈₈₈₈][Oleate], a functionalized fatty-acid-based IL with selective extractive ability influenced by pH, was chosen due to the hydrophobic nature of both the cation and anion.

The bulky nature of the oleate-ion means that even when deprotonated, the ion remains very hydrophobic, which prevents the loss of the anion to the aqueous phase, a common problem in IL chemistry. At pH 5, cobalt, manganese, and nickel were extracted with nearly 100% efficiency through interaction with the oleate ions, and Li was not extracted at all. Co and Mn maintained their high extraction efficiencies at pH values lower than zero, but Ni was no longer extracted.

This phenomenon is due to the formation of tetrachloro-anion complexes in the presence of excess Cl

ions. Mn and Co were separated using mixed salt regeneration solutions, and Ni was selectively precipitated from Li in the form of NiCO

. Overall, [P₈₈₈₈][Oleate] successfully facilitated the separation and recovery of Co, Mn, Ni, and Li from simulated HCl-based Li-ion leachate [65]. Although this method was acidic and reagent intensive, it transformed a rather messy mixture of metals and separated them with one IL based on pH dependence. It would be interesting to more thoroughly understand the pH dependence of ILs as they present a better atom economy than a multi solvent-based system.

Another IL-based Mn, Co, Ni, Li separation system [66] involved the following: first, manganese extraction via N,N,N′,N′-tetra(n-octyl) diglycolamide (TODGA) dissolved in imidazolium-based ILS. TODGA is an extractant commonly used in metal recovery and has an affinity for trivalent metals. Next, cobalt removal using a phosphonium-based IL, such as [P

][Cl]. Lastly, nickel/lithium separation with a DES made from carboxylic acids and lidocaine (DecA:Lid [2:1]). Although this process used significantly less acid than the previously mentioned process, it still required large quantities of solvents and extractants and needs optimization and compactization.

2. Metals for Electronic Devices and Optics

Indium, gallium, and germanium are used in technologies, including light-emitting diodes (LEDs), fiber optics, infrared optics, and solar cells. Indium and germanium are often grouped due to their similar uses and because both are recovered as byproducts from the same ore. Demand for these metals has increased in the past few years as new telecommunication networks compete to have the fastest transmission speeds. Indium is used mainly to produce tin-doped indium oxide used in flat-panel displays and touchscreens, as well as in the form of indium phosphide for lasers and receivers, used in fiber optic networks application [67].

The consumption of Gallium Arsenide (GaAs) has also increased due to the growing manufacturing of integrated circuits and optoelectronic devices [43]. In addition, germanium is mainly used in optics due to its transparency to part of the infrared electromagnetic spectrum, high refractive index, and low chromatic dispersion. Germanium also can catalyze the polymerization of PET resin without undesirable coloring.

2.1. Indium

Some ILs studied for the leaching and extraction of In(III) from aqueous solution include Cyphos

IL 101 [68], Aliquat

336 [69], [A324H+][Cl−] in Solvesso 100 [70], and [Hbet][Tf

N] in ascorbic acid [71]. The experiment using [Hbet][Tf

N] [71] was performed on crushed liquid crystal displays (LCDs). Ascorbic acid was added to decrease the distribution rate of iron and promote selective indium extraction. Overall, extraction of indium from this process reached up to 98.63%.

Indium was also selectively extracted from Ni(II) and Zn(II), by tri-butyl-phosphate (TBP) in n-hexyl-trimethylammonium bis(trifluoromethylsulfonyl)amide [N1116][TFSA] and then recovered by direct electrodeposition of the loaded organic phase [72]. The aqueous solution consisted of Fe(II), Ni(II), Zn(II), In(III), and Sn(II) ions in H[TFSA]. Nitric acid was added to oxidize Sn(II)/Sn(IV) and Fe(II)/Fe(III), which were then precipitated as hydroxide and oxide, respectively.

TBP, a neutral extractant soluble in ILs, was contacted with the aqueous phase to extract indium via a cation exchange mechanism. The electrodeposition of [In(TBP)

]/In(0) under different potentiation conditions was then performed. The reduction peak at −1.0 V was assigned at the [In(TBP)

]/In(0) couple and resulted in the electrodeposition of most pure In metal [72].

Binnemans’ group recently published studies on the electrodeposition of indium from ILs. The electrochemical deposition of indium from aqueous solutions has the unfortunate consequences of simultaneously releasing hydrogen gas. Ionic liquids present a way to circumvent hydrogen evolution by offering a wider electrochemical window than water. In the case of electrodeposition from an IL, the IL acts as a liquid cathode, which is advantageous over a solid cathode because liquid cathodes offer a constant electrode surface and promote more regular crystal growth. The electrodeposition of In in Cyphos IL 101 involved two steps: the reduction of In(III) to In(I) followed by the reduction of In(I) to In(0). The electrodeposition of InCl

and In(Tf

N)

in DME or PEG400 was also tested. A combination of these electrolytes seemed to form an electroactive complex [73].

Indium can be found in goethite residue from zinc production. However, indium is only present in small concentrations, which means other higher concentrated elements, such as iron, are often leached together with indium. Therefore, Supported Ionic Liquid Phase (SILP) extractions have also been examined for recovery of indium, as well as germanium, from iron-rich solutions.

SILPs consist of a thin IL layer covalently anchored or impregnated onto a solid support, usually a porous resin. SILPs hold great potential for industrial-scale operations due to the ease in which the solid phase can be separated from the aqueous phase after the reaction. Both indium and germanium were selectively recovered from iron-rich solutions using Aliquat336- based ionic liquids impregnated on Amberlite XAD-16N, with SF

equal to 5400 and SF

equal to 34,400 [69,74]. Aliquat 336 is a commercial liquid anion exchanger that consists of a quaternary ammonium salt and is a popular choice for IL cations.

2.2. Germanium

As aforementioned, one of the main sources of Ge used in industrial processes is a by-product from zinc ore processing. Another major source of Ge is coal fly ash from power plants. Coal ash can be leached with water; however, the resulting leachate contains many elements and germanium in the form of GeO

and GeS

, therefore, solvent extraction is a popular method for separation [75]. Extraction agents Alamine 336, Aliquat 336, and Cynaex 923 have been used for Ge separation, with the aid of tartaric acid and oxalic acid complexing agents [76]. Aliquat 336 has also been used in conjugation with a flat sheet and hollow fiber supported ionic liquid membrane (HFSLM) system for selective germanium(IV) extraction [77].

Since 2016, at least 30% of the germanium in the world has been provided through recycled germanium materials, such as electronic devices and optical fibers. According to Ruiz, Sola, and Palmerola, the recycling efficiency for fiber-optics scrap is higher than 80% [78].

After extraction of germanium from end-of-life products, it can be reprocessed through electrodeposition, which guarantees high purity and the ability to form unique structures, such as nanoparticles and nanowires (for semiconductors).  For example, one study reported the deposition of germanium from GeCl

saturated [BMIM][PF

] onto gold [79]. Another study reported electrochemical deposition of gallium nanoparticles and germanium nanostructures from [EMIm]Tf

N using pulsed laser assistance [80].

2.3. Gallium

Gallium can be found in semiconductors in the forms of GaN and GaAs, accompanied by InAs. Despite containing valuable sources of gallium and indium, semiconductors and LEDs are often not recycled due to recycling difficulties. Plastic casings and dismantling small parts are examples of some challenges. Acid can be used to effectively leach semiconductors; however, the simultaneous formation of lethal arsine (AsH

) gas poses a significant health and safety hazard.

ILs offer an alternative leaching method that is significantly safer. Trihalide ILs are capable of dissolving metals and alloys without forming any gases. van de Bossche et al. used [P

][Br

] to demonstrate its effectiveness in leaching semiconductor compounds and real LEDs [81]. After the metals were saturated in the organic phase, Arsenic was stripped using NaBr salt, gallium was stripped with water, and indium was recovered using Cyphos 101 to form indium(III) hydroxide precipitate. Unfortunately, GaN could not be dissolved through this process.

Gallium does not readily form minerals in which it is the major component; consequently, most gallium is recovered as a by-product from processing other raw materials. Like germanium, gallium can be found in zinc ore residue, coal ash, bauxite residues, and iron mine tailings. Bauxite is the primary aluminum ore. During the Bayer Process, the bauxite is digested by sodium hydroxide, and aluminum is crystallized out.

The leftover Bayer leach liquor contains a mixed cocktail of gallium, vanadium, silica, residual ammonium, and other impurities. In the 1990s, the extraction of gallium with Kelex 100 was explored but determined to be impractically slow to be employed on an industrial scale. However, Raigual et al. recently studied the behavior of Kelex 100 in 1,2,3- triazolium IL diluents [82].

ILs were superior to kerosene systems because they equilibrated faster, bonded to gallium stronger, and avoided the use of volatile and flammable organic solvents. At optimal conditions, 4-ethyl-5-methyl-1,3-dihexyl-1,2,3-triazolium bis(trifluoromethylsulfonyl)imide [HHT23][Tf

N] reached Ga/Al distribution ratios as high as 1000 and yielded a pure gallium product. This process shows promise for industrial application, especially since it does not require decomposing the Bayer liquids before extraction.

Kinsmen et al. [83] attempted to recover gallium from iron mine tailings using methyltriocytlammonia [MTOA][I] (where [MTOA] = [A336]), which serves the dual purpose of reducing iron(III) to a less soluble state and extracting gallium. Using [MTOA][I] as a reductant and reactant proved to be operationally simple and cost-effective. This experiment exemplifies Green Chemistry principles with environmentally viable reagents, a regenerable organic phase, and reduction of waste streams compared to traditional methods. [MTOA][I] was diluted in toluene, and water was used as a stripping agent. The next steps would be using pure [MTOA][I] without a diluent as was proven possible in the experiment done with water-saturated [HTOA][NO

] extraction of Pd mentioned earlier in this paper [45,83].