Recent Uses of Ionic Liquids in the Recovery and Utilization of Rare Earth Elements

Abstract

The importance of rare earth elements as a basis for the development of new technologies or the improvement of existing ones makes their recovery from raw and waste materials necessary. In this recovery, hydrometallurgy and its derivative solvometallurgy play key roles due to their operational characteristics, which are emphasized with the use of ionic liquids. This manuscript reviews the most recent advances (2023 and 2024) in the use of ionic liquids in unit operations (leaching and separation technologies) aimed at the recovery of these valuable and strategic metals. Moreover, a comprehensive review is presented of the use of these chemicals in the development of advanced materials containing some of these rare earth elements.

1. Introduction

Rare earth elements (in which yttrium and scandium are included) consist of 17 elements presenting similar and unique properties that, recently, have gained enormous interest. Traditionally, these elements are divided into light rare earths (lanthanum, cerium, praseodymium, neodymium, promethium, samarium, and europium) and heavy rare earths (gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, lutetium, and yttrium). Compared with other metals, these elements cannot be found in nature in their zero-valent form, but they form compounds such as phosphates, silicates, carbonates, oxides, and halides. Their total abundance in Earth’s crust is about 169 ppm, which is divided into nearly 138 ppm for light rare earth elements and 31 ppm for heavy rare earth elements. Thus, they are not rare per se, since their abundance is comparable to that of other metals (copper, tin, nickel, etc.). It has been described [1] that the concept of “rare” emerged because the deposits containing them are distributed around the world in low concentrations.

The strategic importance of these elements is due to their unique properties, which are the key factors in the development of different high-tech fields and military affairs, such as electronics, permanent magnets, extremely powerful magnets (holmium), medical, technological, and renewable energy applications, etc. These various uses have led to the consideration of the recovery of these valuable metals not only from raw materials but also from secondary resources, and thus, to the recycling of these metals; among these secondary resources, phosphogypsum, e-waste, red mud, coal, coal fly ash, and mine tailings are included.

It is worth considering that not all the world’s concern for rare earth elements is as smooth as it seems, and in real terms, there are serious, though still scarce, concerns about environmental problems and health issues related to rare earth element mining and processing [2].

As mentioned above, the recovery of REEs (rare earth elements) includes the treatment of raw materials as well as secondary or waste materials. Their high demand and utilization in smart technologies means that the recovery of these elements from these secondary and waste resources has gained interest in recent years [2,3,4,5,6]. The recovery of these strategic metals from one or another resource is accomplished either by hydrometallurgical or pyrometallurgical processes, although in some cases, both types of processing are coupled to fulfill the task [7,8,9]. Solvometallurgy, which can be considered a third type of processing, has evolved from hydrometallurgy. In the former, the conventional aqueous medium is substituted for solvents with green credentials [10,11,12,13,14], and even traditional extractants and diluents used in liquid–liquid extraction are substituted with these green chemicals in order to obtain a more environmentally friendly technology and it is here that ionic liquids (ILs) enter in the metal recovery field.

Ionic liquids are chemicals exclusively composed of a cation and an anion. Both types of ions can be of organic or inorganic nature, though the common features are ILs formed by an organic cation and organic anion or an organic cation and inorganic anion. The other properties of these chemicals include being liquid at temperatures below 100 °C (mandatory), having low vapor pressure, thermal and radioactive stability, non-flammability, etc., whereas their uses encompass a wide spectrum of applications: additives in lubricants, drug design, wastewater treatment, adsorbents for gases, etc. Despite its characteristics, its green credentials have been questioned for some time [15,16,17,18,19,20] since it has been demonstrated that ionic liquids are harmful to living organisms [21].

Thus, these ILs have seen widespread use in the recovery of REEs and other metals [22], and recent reviews [23,24,25] have described the utilization of ILs in hydrometallurgical processes as an alternative to classical extractants for liquid–liquid extraction. In a description of the different methodologies aimed at the recovery of REEs from magnets used on wind generators [26], several ILs alone (Cyphos IL101, a phosphonium salt) or mixed with traditional extractants (Cyanex 923, a phosphine oxide) are used for the recovery of dysprosium or neodymium. Other procedures using polyamides or epoxy derivatives dissolved in ILs are also mentioned in this work. Some ILs in the following groups are available for REE extraction [27]: carbamoylmethylphosphine oxide (CMPO), diglycolamides (DGAs), phenantroline-dicarboxamides, hydroxypyridine, etc.; and, whereas most of these ILs are extracted REEs by a cation-exchange mechanism, some (ammonium+DHDGA (dihexyldiglycoamide) or ammonium+carboxylate) responded to an ion-association mechanism. These two last systems necessitated the use of diluents such as hexane or sulfonated kerosene plus n-octanol, respectively.

An important secondary source for REE recovery is electronic waste, and here, ILs also play a key role in the processing of these otherwise polluting residues [28]. Another waste of interest because of its recyclability is spent Ni-MH batteries, which contain about 10 wt.% of rare earth elements (La, Ce, Nd, and Pr), showing the potential of ILs to recover La and Ce from different aqueous (real and synthetic) solutions [29]. A third source for REE recovery is NdFeB scraps, and though a recent review [30] does not mention the utilization of ILs in dissolution or separation steps, it makes claims about the use of ILs in the electrochemistry of these scraps.

Liquid membranes in various configurations (supported liquid membranes (SLMs) and polymer inclusion membranes (PIMs)) are used to recover rare earth elements from various sources [31]. These liquid membranes contain different ionic liquids that act both as sole carriers (Cyphos IL104, phosphonium salt) or in combination with other extractants to form synergic mixtures to aid REE transport: TOPO (tri-octyl phosphine oxide), DEHPA (di-(2-ethylhexyl) phosphoric acid), and TBP (tri-n-butyl phosphate). Mixtures of the ionic liquid [C₆mim⁺][NTf₂⁻] (1-hexyl-3-methylimidazolium bis((trifluoromethyl)sulfonyl)imide) with DEHPA or TBP; TOPO and TBP; and DEHPA and TBP are used to investigate the transport of Ce, Nd or Pr, and Gd or Y.

Despite all the green properties associated with ILs and the benefits of their usage, it has recently been stated [32] that there are very few examples in which ILs performed better than conventional aqueous systems. As mentioned above, the first drawback of the utilization of ILs is that an important number of IL systems need traditional organic diluents to decrease their inherent viscosity, which increases as the IL phase becomes loaded with the extracted metal. Thus, the extractant+diluent pair of the traditional liquid–liquid extraction is not completely avoided with the use of ILs. The solubility of these ILs due to the mechanisms involved in the extraction of metals must be considered, and this is an odd point about its practical uses. Very often, these extraction mechanisms closely resembled those shown by conventional extractants [33]. As cleverly described in [32] about the use of ILs to recover metals, there is a mismatch between what scientists have investigated and proposed and what the industry needs.

Nevertheless, interest in the utilization of ILs in the various steps involved in the recovery of REEs (and metals in general) from primary and secondary sources remains amid proposals about the use of these chemicals. Thus, this work reviews very recently published investigations about the involvement of ILs in the various steps of REE dissolution–separation–purification. Also, recent references about the use of ILs and REEs to develop some materials of interest are reviewed.

2. Using Ionic Liquids in the Recovery of Rare Earth Elements

To establish better environmental regulations about the use of different chemicals in hydrometallurgical operations, ionic liquids have been suggested as candidates to be used in the different hydrometallurgical–solvometallurgical operations aimed at the recovery of these rare earth elements for different resources. Next, recent literature about the use of ionic liquids in these operations is reviewed.

2.1. Leaching

Betaine hydrochloride ([Hbet⁺]Cl⁻) solution was used to dissolve permanent magnet waste containing Ce, Nd, Gd, Pr plus Fe, Al, and B [34]. Results indicated that the best leaching conditions were time 8 h, temperature 200 °C, ionic liquid concentration 0.1 M, and solid–liquid relationship 1 g/150 mL. Under these conditions, the percentage of REEs leached was 56.24 Ce, 97.05 Nd, 95.51 Gd, 99.81 Pr, and 0.20 Fe, with iron oxide remaining in the residue. The REEs were dissolved by the next reaction:

$${\mathrm{R}\mathrm{E}}_2{\mathrm{O}}_3+8\left[\mathrm{H}{\mathrm{b}\mathrm{e}\mathrm{t}}^{+}\right]{\mathrm{C}\mathrm{l}}^{-}+{\mathrm{H}}_2\mathrm{O}\to [{\mathrm{R}\mathrm{E}}_2(\mathrm{b}\mathrm{e}\mathrm{t}{)}_8({\mathrm{H}}_2\mathrm{O}{)}_4{]}^{6+}+2{\mathrm{H}}^{+}+8{\mathrm{C}\mathrm{l}}^{-}$$

(1)

Once the REEs were dissolved, they were extracted with N,N-di-2-ethylhexyl diglycolamic acid (D2EHDGAA) at pH 2.57. The percentage of extraction for these elements is shown in

Table 1. Approximate percentage of REEs extraction with D2DEHDGAA.

Medium	Ce	Nd	Gd	Pr
HCl	45	82	99	68
[Hbet+]Cl−	44	95	99	69

From [34].

using two media: HCl and [Hbet⁺]Cl⁻.

It can be seen that the change in the medium hardly affects the extraction of the metals, except in the case of neodymium. The authors claimed that the REEs could be separated using the extractant; however, data presented in Table 1 shows that in both media, all the REEs were co-extracted. Moreover, no data about the stripping step were given in the published manuscript.

An ionic liquid also based on betaine, such as betainium bis(trifluoromethylsulfonyl)imide [Hbet⁺][Tf₂N⁻] and water in a 1:1 ratio, was used to recover the REEs from end-of-life mobile-phone speakers [35]. The experiments carried out under various experimental conditions utilized binary (Nd₂O₃–Pr₂O₃) or ternary (Nd₂O₃–Pr₂O₃–Fe₂O₃) systems, as the oxides of light REEs were selectively leached by the IL. In this case, the dissolution reaction was expressed as:

$$\mathrm{R}{\mathrm{E}}_2{\mathrm{O}}_3+6\left[{\mathrm{H}\mathrm{b}\mathrm{e}\mathrm{t}}^{+}\right][{\mathrm{T}\mathrm{f}}_2{\mathrm{N}}^{-}]\to 2[\mathrm{R}\mathrm{E}{\left(\mathrm{b}\mathrm{e}\mathrm{t}\right)}_3][{\mathrm{T}\mathrm{f}}_2\mathrm{N}{]}_3+6{\mathrm{H}}_2\mathrm{O}$$

(2)

with optimum leaching conditions as: 1 M NaNO₃, 60 °C, 250 rpm, and 30 min. In the binary system, Nd₂O₃ was leached preferably to the praseodymium oxide, and together with the above oxides, the oxides of La, Eu, and Gd were dissolved, whereas Y₂O₃, Tb₄O₇, and Yb₂O₃ were not. A complete hydrometallurgical process for the recovery of REEs as oxides from as-received waste speakers was developed. This process included pretreatment, leaching, precipitation, and calcination.

The same ionic liquid as in the previous reference was used for the recovery of REEs from various coal fly ash (CFA) solids [36]. In this case, the leaching reaction was described as:

$${\mathrm{R}\mathrm{E}\mathrm{E}}^{3+}+3\left[{\mathrm{H}\mathrm{b}\mathrm{e}\mathrm{t}}^{+}\right]\left[\mathrm{T}{\mathrm{f}}_2{\mathrm{N}}^{-}\right]\to \left[\mathrm{R}\mathrm{E}\mathrm{E}{\left(\mathrm{b}\mathrm{e}\mathrm{t}\right)}_3\right][{\mathrm{T}\mathrm{f}}_2\mathrm{N}{]}_3+3{\mathrm{H}}^{+}$$

(3)

The REEs in the organic phase were removed as dissolved salt using a mild acid-stripping step, which regenerated the ILs. It was described that an alkaline pretreatment converted the REEs in the CFA to their respective oxides, and afterward, the complete recovery sequence consisted of seven steps:

• Leaching at 85 °C (IL:aqueous phase mass ratio 1:1, 1 M NaCl, solid:total liquid ratio 15 mg/g, 3 h),
• Cooling at 25 °C,
• Removal of CFA residue and addition of ascorbic acid plus betaine (25 mM ascorbic acid/g aqueous phase and 10 mg betaine/g aqueous phase),
• Heating at 85 °C for 1.5 h,
• Cooling at 25 °C and keeping at 4 °C overnight, and separation of aqueous and IL (containing REEs) phases,
• Addition of 1.5 M HCl and heating at 85 °C (IL:HCl mass ratio of 1, 1.5 h),
• Cooling at 25 °C and keeping at 4 °C overnight, followed by separation of an acidic phase containing the REEs and an IL regenerated phase. In all the steps, the different aqueous phases were located above the IL due to the higher density of this phase. The recovery of total REEs varies in the 44%–66% range, with Sc, Y, Nd, Sm, Gd, Dy, and Yb extracted preferably into the IL phase, with recovery efficiencies in the 53.8%–66.2% range, while the recovery of other REEs greatly depends on the different CFA sample used in the investigation.

The recovery of REEs and Sc either by direct leaching through H₂SO₄ and ion exchange or by direct leaching with task-specific ionic liquids, i.e., betainium bistriflimide and subsequent stripping with acidic HCl solutions, has been described [37].

The next reference [38] investigates the separation of Nd and Pr by the selective dissolution of Nd₂O₃ from their oxide mixtures (Nd₂O₃ + Pr₆O₁₁) using as leaching medium a mixture of the ionic liquid 1-butyl-3-methylimidazolium-bis(trifluoromethylsulfonyl) imide ([C_4mim⁺][NTf₂⁻]), 2-thenoyltrifluoro acetate (HTTA) and 0.1% water. The procedure allowed the yield of a separation factor (β_Nd/Pr) > 500, demonstrating that Pr₆O₁₁ remains practically undissolved. Nd was stripped with 0.5 M nitric acid solution and precipitated as Nd₂O_3.

Previous literature showed that ionic liquids based on betainium cation [Hbet⁺] are the most used leachants for the recovery of REEs from various solid materials. It was the composition of these solid materials that made the overall process more or less complicated in terms of REE recovery and separation.

2.2. Electrodeposition

It was shown that the electrodeposition of Nd(III) in ionic liquids media is dependent on the presence of water in the system. In fact, the presence of water catalyzed the electrodeposition of the rare earth. Thus, the introduction of water into the coordination sphere of the complex formed by this REE and bis(trifluoromethanesulfonyl)imide [NTf₂⁻] anions, and their influence on the metal deposition, was investigated [39]. The addition of water produced various effects in the Nd³⁺ [NTf₂⁻] complexes, affecting the Nd³⁺ electrochemical reduction. These effects can be summarized as cis/trans transformation of the anion, change from bidentate to monodentate forms in the coordination mode of the anion, and preferably coordination of water with the transition metal.

This reference investigated the electrochemical (co)deposition of La and Fe-group transition metals from solutions in 1-butyl-1-methyl-pyrrolidinium dicyanamide [BMP⁺][DCA⁻] with controlled water content [40]. The presence of Fe-group metals produced a positive shift in the onset potential of La³⁺ reduction, with lanthanum being co-deposited with the respective Fe-group metal. This electrodeposition occurred at room temperature and E ≤ −1.75 V; however, the ionic liquid tended to degrade. Also, the presence of water up to 3.092 M produced a positive shift of La³⁺ electrodeposition. This ionic liquid containing the [BMP⁺] group had better stability against cathode degradation when compared with ILs with imidazolium-type cations. Lanthanum was deposited in both zero-valent and oxidized states.

2.3. Ion-Exchange Resins

The performance of different ion-exchange resins (tetra-n-alkyl diglycolamide) in [C_4mim⁺][Tf₂N⁻] ionic liquid on the uptake of La(III), Ce(III), Pr(III), Nd(III), Sm(III), Gd(III) and Dy(III) from nitric acid solutions were investigated [41]. Experimentation was carried out both in batch and continuous (column) operations. The ion-exchange resins contained different n-alkyl groups (

Table 2. Composition and maximum and minimum uptakes (mg/g) of the resins.

Resin	Maximum	Minimum
TPDGA: n-pentyl	Dy: 41.7	Sm: 36.8
THDGA: n-hexyl	Dy: 40.9	Pr: 33.1
TODGA: n-octyl:	Dy: 38.4	Pr: 29.2
TDDGA: n-decyl	Dy: 32.6	La: 19.6

From [41].

), and the uptake order was Dy > Gd > Sm > Nd > Pr > Ce > La, increasing this uptake from 0.3 M to 3 M nitric acid, with little variation in the 3–6 M nitric acid concentration range. Table 2 shows the maximum and minimum uptakes of the resins.

It can be seen that with all resins, maximum uptake occurred in the case of dysprosium, and that uptake decreased with the increase of the length of the alkyl chain. As somewhat expected, for all the resins, the breakthrough volume followed the same order as above. Prior to the elution step (0.05 M EDTA in 1 M guanidine carbonate), the column was washed with 3 M nitric acid solution.

2.4. Membranes

PVDF or chitosan-based membranes were investigated for the extraction of lanthanides (Yb, Er, Dy, Sm, Gd, Nd, and Ce) from 10⁻⁴ M Ln solutions [42]. The ionic liquids used in the investigation were; 1-n-Butyl-3-methylimidazolium hexafluorophosphate [Bmim⁺][PF₆⁻], trihexyltetradecylphosphonium bis(trifluoromethylsulfonyl)imide [P₆₆₆₁₄⁺][NTf₂⁻], Butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [C_4mim⁺][NTf₂⁻], and 1-Butyl-3-methylimidazolium-di(2-ethylhexyl)-oxamate [C_4mim⁺][DEHOX⁻]. In the case of PVDF membranes, the best extraction was produced when the membrane was soaked in [C_4mim⁺][DEHOX⁻], with an extraction order as Yb > Er > Dy = Sm > Gd > Nd > Ce; when the membrane was soaked with the other three ionic liquids, the lanthanides were not extracted. In the case of chitosan membranes, all the membranes containing the above ionic liquids extracted all the REEs in an appreciable form, though the chitosan membrane formulated with [C_4mim⁺][DEHOX⁻] presented the best results. In any case, the best extraction results were obtained when the chitosan membranes contained sucrose/citric acid and were irradiated at 5kGy. In all the above experiments, the stripping phase was a 0.5 M nitric acid solution.

2.5. Liquid–Liquid Extraction

Two ionic liquids, derived from N,N-dioctyl diglycol amic acid DODGA and quaternary ammonium salts, were fabricated and formulated as [N₁₈₈₈⁺][DODGA⁻] and [N₄₄₄₄⁺][DODGA⁻], were used in the separation of europium from a europium-uranium binary system [43]. Maximum Eu(III) extraction was obtained at 4–5 M nitric acid, with N₄₄₄₄-based IL performing better than N₁₈₈₈ as their respective distribution ratios were assessed: 13 versus 9. Also, the europium(III)/uranium(VI) separation factors were better in the case of N₄₄₄₄ IL: 60.35 versus 55.44. Despite these results, there was a continuous loss of extraction efficiency under continuous use: 85% (1st cycle) to 45% (5th cycle) and 95% (1st cycle) to 65% (5th cycle) for N₁₈₈₈ and N₄₄₄₄-based ILs, respectively. This work uses a kinetics investigation approach: pseudo-first and pseudo-second kinetics models, which were inadequate for a liquid–liquid extraction investigation but for adsorption or ion exchange with adsorbents/resins investigations.

Real scrap NdFeNi magnets from computer hard disks were used to investigate the recovery of Nd(III) [44]. Using Aliquat 336 (quaternary ammonium salt) and Cyanex 572 (phosphonic acid), an ionic liquid formulated as [R₄N⁺][Cy⁻] and diluted in kerosene was used for extraction studies to separate Nd(III) from Ni(II). Nd(III) extraction was maximum at 0.1 M acidic concentrations (HCl, HNO₃, or H₂SO₄), decreasing the extraction as the acid concentration increased from 0.1 to 1 M in the aqueous phase. Nd(III) extraction performed better in the nitric acid medium; however, the highest (26.3) Nd/Ni separation factor was obtained at 0.2 M HCl. Thus, a process was defined using 0.2 M HCl solution to leach the magnets, and the extraction occurred at pH 1.23, accordingly with the equilibrium:

$$\mathrm{N}{\mathrm{d}}_{\mathrm{a}\mathrm{q}}^{3+}+3{\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+3({\mathrm{R}}_4{\mathrm{N}}^{+}{\mathrm{C}\mathrm{l}}^{-}{)}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{N}\mathrm{d}{\mathrm{C}\mathrm{l}}_3·3({\mathrm{R}}_4{\mathrm{N}}^{+}{\mathrm{C}\mathrm{y}}^{-}{)}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(4)

The extraction of nickel(II) responded to the formation in the organic phase of a compound with NiCl₂ · 2(R₄N⁺Cy⁻) stoichiometry. Both Nd(III) and Ni(II) were extracted by a solvation mechanism. Stripping was accomplished in a 0.5–2 M HCl medium, and after stripping, Nd was precipitated with oxalic acid, followed by calcination to produce Nd₂O₃.

An industrial solution containing Ho, Er, Tm, Yb, Lu, and Y was used to investigate the separation of yttrium from the rest of the REEs [45]. The separation was carried out by liquid–liquid extraction with the ionic liquid trioctylmethylammonium (2,6-dimethylheptyl propionate ([N₁₈₈₈⁺][POPA⁻]) dissolved in sulfonated kerosene. The extraction of yttrium responded to the next equilibrium:

$${{\mathrm{Y}}_3\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+3{[\mathrm{N}}_{1888}^{+}][\mathrm{P}\mathrm{O}\mathrm{P}{\mathrm{A}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow {[\mathrm{N}}_{1888}{]}_3\mathrm{Y}{[\mathrm{P}\mathrm{O}\mathrm{P}\mathrm{A}]}_3{\mathrm{C}\mathrm{l}}_3$$

(5)

The best extraction results were reached at pH 5, whereas stripping was performed in an HCl medium. The use of this ionic liquid had selectivity for Y against the presence of heavy rare earths (Ho–Lu), which can be stripped from the organic phase using deionized water. A countercurrent experiment consisting of nine extraction stages and six scrubbing stages allowed the yield of a solution in which yttrium was purified near 99 wt.%.

The extraction performance of the task-specific ionic liquid trihexyltetradecylphosphonium 3-hydroxy-2-naphthoate against the presence of La, Ce (IV), Nd, Ho, and Lu was investigated [46]. Using single metal-bearing solutions, it was demonstrated that the extraction was pH-dependent, with the best results yielded at pH 2.5, 23 °C, and 6 h of contact between both aqueous and organic phases. The extraction of these elements responded to the equilibrium:

$$\mathrm{L}\mathrm{n}{(\mathrm{N}\mathrm{O}}_3{)}_{5\mathrm{a}\mathrm{q}}+2{[\mathrm{P}}_{66614}^{+}][{\mathrm{H}\mathrm{N}\mathrm{A}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow {[\mathrm{P}}_{66614}{]}_2[\mathrm{L}\mathrm{n}{(\mathrm{N}\mathrm{O}}_3{)}_5{]}_{org}+2\mathrm{H}\mathrm{N}{\mathrm{A}}_{\mathrm{a}\mathrm{q}}^{-}$$

(6)

where Ln = La, Nd, Ho, and Lu. Cerium(IV) was extracted by a similar reaction, but the extracted species presented the [P₆₆₆₁₄]₂[Ce(NO₃)₆] stoichiometry. Metals loaded onto the organic phase can be stripped with a 0.5 M nitric acid solution. Despite the apparent good characteristics of the ionic liquid, extraction experiments showed that the ionic liquid dissolved in the raffinate and during the stripping stage to the acidic stripping solution. It is worth noting here that the authors of this investigation considered leaching to be the process of IL dissolution into the different aqueous phases.

Tri-n-octylamine (R₃N) and octanoic acid (HA) were the precursors of an ionic liquid [R₃NH⁺][A⁻], which, when dissolved in n-octanol, was used for the extraction of REEs (Eu, Y, Tb, Gd) from a nitrate medium [47]. Two types of equilibria were responsible for REE extraction: (i) when the initial IL concentration in the organic phase versus the REE concentration in the organic phase was 3, the reaction was:

$$\mathrm{L}{\mathrm{n}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{N}\mathrm{O}}_{3\mathrm{a}\mathrm{q}}^{-}+3{\mathrm{R}}_3\mathrm{N}{\mathrm{H}}^{+}\mathrm{C}{\mathrm{l}}_{\mathrm{o}\mathrm{r}\mathrm{g}}^{-}\leftrightarrow {\mathrm{L}\mathrm{n}\mathrm{A}}_{3\mathrm{o}\mathrm{r}\mathrm{g}}+{3\mathrm{R}}_3{\mathrm{N}\mathrm{H}\mathrm{N}\mathrm{O}}_{3\mathrm{o}\mathrm{r}\mathrm{g}}$$

(7)

and (ii) when the above relationship was 2, the equilibrium responded to the reaction:

$${\mathrm{L}\mathrm{n}}_{\mathrm{a}\mathrm{q}}^{3+}+{\mathrm{N}\mathrm{O}}_{3\mathrm{a}\mathrm{q}}^{-}+{2\mathrm{R}}_3{\mathrm{N}\mathrm{H}}^{+}{\mathrm{C}\mathrm{l}}_{\mathrm{o}\mathrm{r}\mathrm{g}}^{-}\leftrightarrow \mathrm{L}\mathrm{n}{(\mathrm{N}\mathrm{O}}_3){\mathrm{A}}_{2\mathrm{o}\mathrm{r}\mathrm{g}}+{2\mathrm{R}}_3{\mathrm{N}\mathrm{H}\mathrm{N}\mathrm{O}}_{3\mathrm{o}\mathrm{r}\mathrm{g}}$$

(8)

These REEs were best extracted at pH 5 and in the presence of NaNO₃ in the aqueous solution. At this pH value, the extraction order was Tb = Eu > Gd > Y. Stripping can be accomplished with the use of distilled water, and under five extraction–stripping cycles, the extraction efficiency slightly decreased, but the stripping one remained almost constant.

The ionic liquid formed by trihexylamine di-2-ethyl hexyl phosphate ([TAHAH⁺][DEHP⁻]) was utilized in the liquid–liquid extraction of yttrium(III) from nitrate solutions [48]. As in other studies, the ionic liquid was dissolved in an organic diluent, with n-hexane as the diluent used here. Best extraction results were obtained using 0.1 M ionic liquid concentration in the organic phase, 0.2 M sodium nitrate in the aqueous phase, and pH 3 yttrium extraction, represented by the next equilibrium:

$${\mathrm{Y}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{N}\mathrm{O}}_{3\mathrm{a}\mathrm{q}}^{-}+{2[\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{H}}^{+}][{\mathrm{D}\mathrm{E}\mathrm{H}\mathrm{P}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow {\mathrm{Y}(\mathrm{N}\mathrm{O}}_3{)}_3{2\left[\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{H}\right]\left[\mathrm{D}\mathrm{E}\mathrm{H}\mathrm{P}\right]}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(9)

Yttrium stripping from the Y-loaded organic phase was carried out with 0.1 M nitric acid solution. The extraction of this metal was greatly dependent on the organic diluent used to dissolve the ionic liquid (

Table 3. Influence of the organic diluent on the percentages of Y(III) or Nd(III) extraction by trihexylamine di-2-ethyl hexyl phosphate.

Diluent	a Y	b Nd
n-hexane	73	81
n-heptane	13	86
chloroform	76	41
carbon tetrachloride	96	35
kerosene	97	99

^a In 0.2 M NaNO₃. ^b In 2 M NaNO₃. Other experimental conditions: 0.1 M IL in the respective diluent and pH 3. From [48,49].

). After continuous cycles, this IL also lost its extraction performance: 100% in the first cycle versus 29% in the fifth cycle.

The same ionic liquid as in the previous reference was diluted in kerosene and used in the solvent extraction of Nd(III) [49]. Under different experimental conditions, results indicated that the extraction was pH-dependent and increased with the increase of the pH value from 1 to 3 and in the presence of a salt (i.e., NaNO₃) in the aqueous feed phase. Neodymium extraction responded to the formation of species with Nd(NO₃)₃ · 3IL stoichiometry in the organic phase. Stripping was best performed with diluted (2 × 10⁻³ M) sulfuric acid solutions. The extraction ability of this ionic liquid also changed with the different diluents used to dissolve it (Table 3). After four cycles of continuous use, the extraction performance of the ionic liquid considerably decreased: 100% in the first cycle and 45% in the fourth cycle.

From the results presented in this table, it is worth mentioning the use of two toxic (carcinogenic) diluents, chloroform and carbon tetrachloride.

The ionic liquid 1-Butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide ([C_4mim⁺][Tf₂N⁻]) was used to dissolve tri-n-octylphosphine (TOPO) extractant and used in the liquid–liquid extraction of Nd(III) [50]. The extracted species presented a 1:6 Nd:TOPO stoichiometry, with a 45% increase in the extraction percentage in the presence of the IL, with respect to extractions performed only with TOPO. No stripping data were found in the published manuscript.

This reference presented a system in which Zr(IV) can be separated from Eu(III)/Am(III) as elements representative of lanthanides and actinides, respectively [51]. The ionic liquid tri(hexyl)tetradecylphosphonium nitrate ([P₆₆₆₁₄⁺][NO₃⁻] was used as an extractant, whereas nitric acid solutions formed the feed phase. Zr(IV) was extracted preferably to Eu(III) and Am(III) in the 1–5 g/L initial metal concentrations range and 3 M nitric acid medium. It was mentioned that 1,3-diisopropyl benzene (another toxic diluent) was used to dilute the ionic liquid unless the Abstract of the manuscript did not mention this.

This reference investigated the separation of Th(IV) from uranium and REEs using the ionic liquid tetraphenylethane-1,2-diylbis(phosphoramidate) diluted in chloroform [52]. Thus, this is a manuscript that should have never been authorized to be published due to the utilization of the toxic organic diluent (chloroform) in the experimentation.

Nd(III) extraction was investigated using P204 (DEHPA) extractant and the IL tributylhexylphosphonium bis(trifluoromethanesulfonyl)imide [P₄₄₄₆⁺][NTf₂⁻] [53]. Under the various experimental conditions, it was found that extraction in nitric acid medium favored metal extraction over the use of HCl solutions; the extraction rate increased with the increase of the aqueous pH value, reaching a maximum efficiency at pH 4; an increase of the temperature also increased the extraction percentage, reaching a maximum at 60 °C. The apparent role of the IL was to combine with the protons generated during the metal-extraction process, allowing the continuity of the extraction process. The manuscript did not present stripping results, and some contradictions appeared: i.e., the authors concluded that at pH 4, molar ratio (P204:Nd(III)) greater than 5:1 and 20 °C, the extraction rate of the system in both HCl and HNO₃ media reached 99.99%, whereas previous results contradicted the above.

The separation of thorium(IV) from cerium(III), (IV) and lanthanum(III) in nitrate solutions was investigated using the quaternary ammonium chloride salt Cyphos^® IL-101 [54]. Experimental results showed that at 50 mg/L of the metals in the aqueous solution and pH 2, the extraction order was Th(IV) > La(III) > Ce(III) with Th/Ce and Th/La separation factors of >5000 and >1000 for the respective systems. At 110 mg/L metal concentration, the extraction order was Th(IV) > Ce(III) > La(III). Cerium(IV) was extracted much better than Ce(III), with results comparable to that of Th(IV). Both Th(IV) and Ce(IV) were extracted by the formation of a species with [PR₄⁺] · (M(OH)₃(NO₃)₂)⁻ stoichiometry in the organic phase. The extracted metals can be stripped using an EDTA + 0.5 M NaCl solution. Experiments carried out on a simulated REE solution demonstrated that the extraction order was Th >>> Sm > Eu > Nd > Gd > La > Ce(III). The extractant suffered a considerable loss of thorium(IV) extraction efficiency after continuous use: 99% in the first cycle and 46% in the fifth cycle.

The next reference also used liquid–liquid extraction with ionic liquids as a procedure to eliminate impurities from an REE solution [55]. In this case, the impurity was aluminum, and the REEs were La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, and Y. The composition of the ILs and some of the results derived from this investigation are shown in

Table 4. Composition of the ILs and Al and REE percentages of extraction and separation factors (SF).

Ionic Liquid	Al	REEs	SFAl/REEs
Methyltrioctylammonium-neodecanoic acid [N1888+][NDA−]	80	<1	250
Methyltrioctylammonium-decanoic acid [N1888+][DA−]	65	<1	175
Methyltrioctylammonium-palmitic acid [N1888+][PA−]	50	<1	50
Methyltrioctylammonium-lauric acid [N1888+][LA−]	50	<1	100

The ILs were dissolved in n-hexane. From [55].

.

The best aluminum separation from the REEs was reached at an equilibrium pH value of 4.8. After the extraction step, the organic phase was stripped with sulfuric acid solutions, with aluminum completely stripped with the use of a 7 × 10⁻² M sulfuric acid solution at an O:A ratio of 1. Furthermore, the IL phase was regenerated with a 1 M NaOH solution. The process was demonstrated on a real leach solution (REEs + Al + Ca + Mg).

Cyanex 272 extractant (bis(2,4,4-trimethylpentyl)phosphinic acid) was utilized together with Alamine 336 and Aliquat 336 extractants (trioctylamine and N-methyl-N,N-di-octyloctan-1-ammonium chloride, respectively) to form ionic liquids [ALA336⁺][C272⁻] and [Ali⁺][C272⁻], which dissolved in a kerosene-type diluent, were used to extract Tb(III) from HCl solutions and containing organic acids (see

Table 5. Percentages of Tb(III) extraction in the presence of organic acids by [ALA336⁺][C272⁻] and [Ali⁺][C272⁻] ILs.

Organic Acid	[ALA336+][C272−]	[Ali+][C272−]
Formic acid	80	10
Lactic acid	100	10
Fumaric acid	75	75
Maleic acid	90	90

From [56].

) [56]. The presence of these organic acids served to complexify terbium(II) and helped to buffer the solution. In all cases, the percentage of terbium extraction increased with the increase of the pH of the feed solution, reaching the best extraction results at pH 5 (Table 5).

As can be seen from these results, the presence of formic or lactic acids in the aqueous solution had a different influence on Tb(III) extraction. With the IL derived from the tertiary amine Alamine 336, metal extraction was greatly reduced with respect to the presence of fumaric or maleic acids in the solution, an effect that is not found when the IL derived from the quaternary ammonium salt Aliquat 336 was used to extract the REE. In fact, with this last IL, maximum Tb(III) extraction was yielded when lactic acid was present in the aqueous solution. These results can be attributed to the fact that, together with the metal, the organic acids can be extracted by the ILs, and whereas the extraction of formic and lactic acids is less than 10% by [Ali⁺][C272⁻], these acids were extracted at an approximate rate of 80% in the case of [ALA336⁺][C272⁻]. Against the above, fumaric and maleic acids were extracted at 90% or higher rates by both ILs. It was evident that the co-extraction of the organic acids increased the extraction of terbium(III). At 0.5 M, initial acid concentrations and for both ILs, the acid extraction efficiency followed the sequence: maleic > fumaric > lactic > formic. Again, the published manuscript did not include data about the terbium-stripping step.

Several ILs ([N_118H⁺][DEHP⁻] (N,N-dimethyloctylamine bis(2-ethylhexyl)phosphate), [N_1110H⁺[DEHP⁻] (N,N-dimethyldecylamine bis(2-ethylhexyl)phosphate), and [N_1112H⁺][DEHP⁻] (N,N-dimethyldodecylamine bis(2-ethylhexyl)phosphate)) were formulated and used to separate Nd(III) from Pr(III), La(III) and Ce(III) from aqueous solutions [57]. Prior to the extraction step, the ILs were pre-saturated with water to decrease the risk of their solubilization in the aqueous phases. Best Nd(III) extraction efficiency was reached at pH 4, with 10 min of contact between the organic and aqueous phases, and at 30 °C; at this pH value, optimal Nd–REE separation can be reached (

Table 6. REE distribution coefficients (D) and Nd–REE separation factors (SF).

REE	D	SF
Nd(III)	768	-
Pr(III)	286	2.7
Ce(III)	111	6.9
La(III)	56	13.7

Extractant: [N_118H⁺][DEHP⁻]. O:A ratio: 1/3. From [57].

).

It was described that the extraction of Nd(III) responded to the next equilibrium:

$${\mathrm{N}\mathrm{d}}_{\mathrm{a}\mathrm{q}}^{3+}+3{\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+3\left[{\mathrm{N}}_{118\mathrm{H}}^{+}\right][\mathrm{D}\mathrm{E}\mathrm{H}{\mathrm{P}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow {[\mathrm{N}}_{118\mathrm{H}}^{+}{]}_3\mathrm{N}\mathrm{d}[\mathrm{D}\mathrm{E}\mathrm{H}{\mathrm{P}}^{-}{]}_3{\mathrm{C}\mathrm{l}}_{3\mathrm{o}\mathrm{r}\mathrm{g}}$$

(10)

Quantitative Nd(III) stripping from metal-loaded organic phases can be reached using a 0.16 M HCl solution. The extractant maintained an extractant efficiency of about 98% after eight cycles.

This reference investigated the same problem—the separation of Nd(III) from La(III)-Ce(III)-Pr(III)—as the previous one, but using different ionic liquids [58]. In this case, they were based on bis(2-ethylhexyl)phosphate moiety, but here known as [DOP⁻], whereas the cation moiety was based on 1-methyl-3-ethylhexylimidazole [O_min⁺], 1-methylpyrrole [H_mpy⁺] and the previously used N,N-dimethyloctylamine [N_118H⁺]. Using feed solutions containing 0.017 M NdCl₃, 20 min of contact time, 20 °C and an IL/aqueous phases volume ratio of 1:5, results indicated that at pH 4, Nd(III) extraction efficiency followed the sequence O_mim > H_mpy > N_118H. Thus, REEs were extracted by the formation of [O_mim⁺]₃Nd[DOP⁻]₃Cl₃ species into the organic phase, whereas the distribution coefficients were D(Nd) = 369, D(Pr) = 48, D(Ce) = 27, D(La) = 19, leading to separation factor values of 7.7, 13.7, and 19.4 for the Nd/Pr, Nd/Ce, and Nd/La pairs. Stripping can be accomplished by HCl solutions; at 0.3 M HCl, the stripping order was La = Ce > Pr > Nd, but at 0.6 M HCl, the stripping efficiency was almost equal (100%) for the four REEs. The ionic liquid showed an extraction efficiency of about 97% after seven cycles. No data were included about what to do with the stripped solution to recover Nd(II) and the purity of the final product.

Waste generated by ion-adsorbed rare earth ore (IATREO) separation plants contained thorium(IV), which needed to be eliminated due to environmental implications. Thus, different ionic liquids (

Table 7. Ionic liquids used in the extraction of Th⁴⁺ from the waste-residue leaching solution.

Ionic Liquid	Acronym
Bis(3,3-dimethyl-2-oxobutyl)dioctylammonium sec-octyl-phenoxyacetate [OB2DTA+][CA12−]	IL1
Bis(1-ethoxy-1-oxooct-2-yl)dioctylammonium sec-octyl-phenoxyacetate [EO2DTA+][CA12−]	IL2
(3,3-dimethyl-2-oxobutyl)trioctylammonium sec-octyl-phenoxyacetate [OBTA+][CA12−]	IL3
(1-ethoxy-1-oxooct-2-yl)trioctylammonium sec-octyl-phenoxyacetate [EOTA+][CA12−]	IL4

From [59].

) dissolved in kerosene were used for the elimination of this REE from the waste-residue leaching solution [59]. In a simulated solution of pH 3, Th⁴⁺ extraction followed the IL1 = IL2 > IL3 > IL4 order. It was also noted that the addition of tributylphosphate (TBP) or isooctyl alcohol to the system improved phase separation. A 0.2 M HCl solution stripped thorium, and again, the presence of the above two modifiers improved the stripping efficiency in the case of IL1 and IL 2 extractants. From the real waste leach solution, thorium was separated after two extraction stages with minimal loss of the rest of the REEs of about 15 mg/L. From the raffinate, these REEs can be recovered by oxalate precipitation to form RE₂(C₂O₄)₃ salts, whereas from the stripped solution and after pH adjustment, thorium was recovered by precipitation as Th(OH)₄.

Thorium(IV) was again the target to be extracted by liquid–liquid extraction, in this case, using dimethyl distearyl ammonium chloride [DMDSA⁺][Cl⁻] (an ionic liquid itself) and N,N-dialkyl-succinamide acid (R-SCA, alkyl chains: n-butyl, n-hexyl or n-octyl), which formed different [DMDSA]⁺[R-SCA]⁻ ionic liquids [60]. These ILs were dissolved in different diluents, and the extraction of Th⁴⁺ followed the sequence n-hexadecane > n-tetradecane > n-dodecane > n-decane > kerosene > n-decylalcohol. The presence of sodium nitrate in the aqueous solution improved the extraction of REEs. After the extraction at 8 M nitric acid, nearly 99% of thorium (IV) immediately aggregated into a self-assembly solid at the interface between the ionic liquid and raffinate phases, though apparently, in the extraction process, some thorium was released to the raffinate as Th(NO₃)₅⁻. The extraction responded to the formation of [SCA · Th(NO₃)₄⁻] species. Using a 0.01 M [DMDSA⁺][DOSCA⁻] (dimethyl distearyl ammonium N,N-dioctyl-succinamidate) in n-hexadecane solution and 8 M HNO₃ concentration in the aqueous feed phase, the separation factor values of 2516, 1885, 558, and 1512 were experimentally found for the Th/U, Th/La, Th/Lu, and Th/Eu pairs. There was no mention of how thorium was recovered from the solid or the fate of the ionic liquid in this recovery.

The purification of Gd(III) from Al(III) impurities is of interest, and liquid–liquid extraction was used to perform this purification [61]. In this case, different ionic liquids, diluted in sulfonated kerosene and based on the methyltrioctylammonium ([N₁₈₈₈⁺] cation and various carboxyl groups (

Table 8. The various carboxylic acids used in the extraction of Al(III) from Al/Gd solutions.

Carboxylic Acid	Derived Anion	a Al	a Gd
Cyclopentaneacetic acid	C7H11O2−	75	11
Heptanoic acid	C7H13O2−	60	18
3-cyclopentylpropionic acid	C8H13O2−	60	10
n-octanoic acid	C8H15O2−	55	25
3-cyclohexanepropionic acid	C9H15O2−	57	30
Nonanoic acid	C9H17O2−	58	20

^a Percentages of extraction. From [61].

) as counterions, were investigated to perform this task.

Table 8 also presents some results relative to the extraction of these elements from the binary Al/Gd solution. At the initial conditions of 0–1 M ionic liquid, 0.01 M (each) Al and Ga, pH 1.50, and 25 °C, highest Al/Gd separation factor was obtained with [N₁₈₈₈⁺][C₇H₁₁O₂⁻] ionic liquid, though the increase of the temperature in the 25–75 °C range was accompanied by a decrease in the separation factor value. Both elements (M) were extracted by the same speciation of 1:3.3:3 (M:N₁₈₈₈:C₇H₁₁O₂:Cl). Stripping was accomplished using HCl solutions. It is worth noting here that after all this investigation, a subsection in the published manuscript mentioned the formation of a third phase (or second organic phase) in the system. This third phase was eliminated by the addition of iso-octanol to the organic phase; however, this addition sharply decreased aluminum extraction efficiency (5% alcohol resulted in 36% efficiency, 15% alcohol in 9.55% efficiency). The authors of the manuscript did not mention if all the previous investigations were performed in the presence of this alcohol. If not, and all the experimentation was carried out under conditions of third-phase formation, the data do not have any value because a liquid–liquid extraction system presenting a third phase is simply useless, sic: its (third phase) presence must be avoided like the plague [62].

Gd(III) was extracted by 4-benzoyl-3-methyl-1-phenyl-2-pyrazolin-5-one (HP) using [C₁C_nim⁺][NTf₂⁻] (n = 4 or 10) ionic liquid [63]. The investigation compared the conventional liquid–liquid extraction procedure (aqueous and organic phases) with the use of two immiscible organic phases. Under this system, Gd³⁺ appeared in the more polar phase (MP, ethylen glycol), whereas the phase containing the ligand and the ionic liquid formed the less polar phase (LP). In this system, the extraction of the REE at an initial pH value of 3.3 responded to the equilibrium:

$${\mathrm{G}\mathrm{d}}_{\mathrm{M}\mathrm{P}}^{3+}+{\displaystyle \frac{1}{2}}\mathrm{H}{\mathrm{P}}_{\mathrm{L}\mathrm{P}}+\mathrm{n}[\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}{]}_{\mathrm{L}\mathrm{P}}\leftrightarrow \mathrm{G}\mathrm{d}(\mathrm{P}{)}_{{\displaystyle \frac{1}{2}}}^{2+}·(\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}{)}_{\mathrm{n}\mathrm{L}\mathrm{P}}+{\mathrm{H}}_{\mathrm{M}\mathrm{P}}^{+}$$

(11)

with no transfer of [C₁C_nim⁺] to the ethylen glycol phase. If the extraction occurs in a conventional Gd(III)-aqueous and organic-phase liquid–liquid extraction system, the extraction equilibrium responds to the formation of GdP²⁺ · (NTf₂⁻)_n species in the organic phase, releasing protons and [C₁C_nim⁺]to the aqueous phase or raffinate. In the ethylen glycol medium, neither La(III) nor Ce(III) was extracted into the organic phase, whereas Lu(III) was extracted with an efficiency greater than 50%, and Eu(III and Gd(III) only in the presence of the ionic liquid in this LP phase. In the system involving aqueous and organic phases, Ce(III) was not extracted, and La(III) had an efficiency lower than 50% and in the presence of the ionic liquid, and Eu(III, Gd(III), and Lu(III) had efficiencies greater than 50%. No stripping data were included in the published manuscript.

Europium(III) was extracted by mixtures of the acidic chelating compounds thenoyltrifluoroacetone (HTTA), 4-benzoyl-3-methyl-1-phenyl-pyrazol-5-one (HP) or 4-bemzoyl-3-phenyl-isoxasol-5-one (HPBI) and meso-hexamethylpropyleneamine oxime (HM-PAO) or its bis-imine precursor (pre-HM-PAO) [64]. Also, the effect of the presence of different ionic liquids as [C₁C_nim⁺][NTf₂⁻] (n = 4,6,8,10) or [C₁C₄pyr⁺][NTf₂⁻] (1-butyl-1-methylpyrrolidinium bis (trifluoromethylsulfonyl) imide) on the extraction of the REE was investigated. When the organic phase contained HTTA, a strong synergism effect on Eu(III) extraction was observed, and HM-PAO was more effective than pre-HM-PAO. Using the HPBI ligand, the nature of the IL favored the extraction of the metal in the series: n = 10 > 8 > 6 > 4 > C₁C₄pyr⁺. The manuscript did not include stripping data.

The amide derivative 4-chloro-N-(1-methyl-1H-pyrazol)-3-yl)picolinamide (L) dissolved in two ionic liquids: 1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([C_6mim⁺][NTf₂⁻]) or 1-hexyl-3-methylimidazolium bromide ([C_6mim⁺][Br⁻]) was used to extract Eu³⁺ from nitric acid solutions [65]. Preliminary tests showed that in the presence of the ionic liquid containing the [NTf₂⁻] anion, the L ligand was more effective with respect to the extraction of the rare earth element. Thus, further investigation used only this IL to dissolve the amide derivative. This REE was extracted by a cation-exchange mechanism and formation of Eu(L)₃(NO₃)₂⁻ species in the organic phase, and the same species also coordinated with one water molecule. In the extraction process, the ionic liquid cation [C_6mim⁺] was transferred to the aqueous solutions for charge compensation. The increase in the temperature decreased the IL viscosity, but extraction efficiency also decreased. Stripping was accomplished by 0.05 EDTA and 1 M guanidine carbonate solutions. The manuscript did not mention how the IL was regenerated and what to do with the europium-loading stripped solution.

These series of references [66,67,68] investigated the extraction of lanthanides using the extractants shown in the next

Table 9. Several ligands used in the extraction of lanthanides.

Ligand	IL	Remarks	Reference
1,2-bis(diphenylphosphinyl)-ethane or 1,2-bis(diphenylphosphinyl)benzene	[C4mim+][NTf2−]	Use of toxic H2CCl2 as diluent	[66]
tetraoctyldiglycolamide	[N1888+][NTf2−]	a Use of 1-octanol or n-nonane and 1.1 M TBP as diluents	[67]
tetraoctyldiglycolamide	[TOAH+][NTf2−]	b Use of dodecane:1-octanol (9:1% v/v) mixture as diluent	[68]

[C_4mim⁺][NTf₂⁻]: 1-butyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]imide. [N₁₈₈₈⁺][NTf₂⁻]: methyltrioctylammonium bis(trifluoromethanesulfonul)imide. [TOAH⁺][NTf₂⁻]: {trioctylammonium bis[(trifluoromethyl)sulfonyl]imide. ^a Other diluents used: nitrobenzene, chloroform, 1,2-dichloroethane. ^b Other diluents used: chloroform, 1,2-dichloroethane. From [66,67,68].

. Studying these references, it can be concluded that the data do not have practical interest since the diluent used in the investigation is highly toxic. The same can be extended to the results derived from the utilization of nitrobenzene, chloroform, and 1,2-dichloroethane using TODGA plus ILs in the extraction of the lanthanides. In the case of the system formed by 0.1 M TODGA plus 0.01 M [N1888⁺][NTf₂⁻], the best extraction results, in comparison with the n-nonane plus TBP mixture, were obtained when 1-octanol was used as a diluent of the organic phase, with the extraction sequence being Er,Tm,Yb,Lu > Ho > Dy > Th >> Eu = Gd > Sm > Nd > Pr > Ce > La. The extraction responded to a cation-exchange mechanism with the formation of Ln(TODGA)₆(NTf₂)₃ species in the organic phase and the release of N1888⁺ ions to the aqueous phase. It was mentioned that stripping was performed with 0.1 M 1-hydroxyethane-1,1-diphosphinic acid aqueous solution, although no more data were given in the manuscript. Also, it was not mentioned how the lanthanides were recovered from the stripped solution.

In the case of the system using TODGA and [TOAH⁺][NTf₂⁻], the extraction sequence was the same as that shown above. In this system, the extraction responded to the formation of the Ln(TODGA)₃(NO₃⁻)(NTf₂⁻)₂ species in the organic phase and the formation of the [TOAH⁺][NO₃⁻] ionic liquid in this same phase. At high nitric acid concentrations, the formation of this new ionic liquid was favored, and the extraction of all the lanthanides except for lutetium decreased with respect to the values obtained from dilute nitric acid solutions. The role of TBP and 1-octanol in the respective previous references was to act as modifiers to avoid the formation of the undesirable third phase after phase disengagement in the extraction step. No stripping data were included in this last reference.

Column experiments were developed to investigate the influence of an ionic liquid on the extraction of dysprosium [69]. The organic phase was formed by 1% v/v D2EHPA (di-(2-ethylhexyl) phosphoric acid) in tributylphosphate (TBP) and the ionic liquid. [C_6mim⁺] [NTf₂⁻]. The presence of the ionic liquid in the organic phase enhanced the physical properties of the system in comparison to when this ionic liquid was not present in the organic phase. Furthermore, the presence of the ionic liquid diminished the interfacial tension, resulting in a decrease in Sauter’s mean diameter and an increase in the holdup of the dispersed phase, as well as the extraction of dysprosium. The most effective conditions for extracting Dy(III) ions from an aqueous solution were observed at a rotation speed of 359 rpm, with a solvent stream rate (Qd) of 30 L/h and an aqueous solution stream rate (Qc) of 28 L/h. Dysprosium was extracted by a cation-exchange reaction with D2EHPA.

Utilizing trihexylammonium octonoate and trioctylammonium myristrate pseudo-protic ionic liquid as extractants, this investigation showed the impact of the Hofmeister effect and the Van ‘t Hoff–LeChatelier effect on the extraction of cobalt and neodymium from aqueous phases [70]. A high driving force is required from the Hofmeister effect or both effects simultaneously to successfully increase the extraction of cobalt (more than ten-fold). Lower driving forces are sufficient to increase the extraction of neodymium (up to 100%). No stripping data was included in the work.

Quaternary ammonium-based ILs (QA-ILs), i.e., [N₃₃₃MeOAc⁺][NTf₂⁻] (N,N,N,-tripropyl-1-methylacetate ammonium bistriflimide) and [N₄₄₄MeOAc⁺][NTf₂⁻], (N,N,N,-tributyl-1-methylacetate ammonium bistriflimide) have been synthesized and dissolved in chloroform for selective Sc(III) extraction from other REEs [71]. Both ILs are selective with respect to the separation of Sc(III) from other REEs, although the selectivity of Sc/Lu and Sc/Yb is greater when [N₄₄₄MeOAc⁺][NTf₂⁻] is used. Specifically, at pH 4.5 and 5, the extraction efficiency of Sc(III) using [N₃₃₃MeOAc⁺][NTf₂⁻] was 73.6% and 83.4%, respectively, and it was up to 90.5% and 95.9%, respectively, using the former extractant. These ILs can be recycled with a high stripping efficiency of up to 86% using a 4 M H₂SO₄ solution. The above information has no practical value since the ILs are dissolved in harmful chloroform.

An organic phase containing N,N,N′,N′-tetra-(2-ethylhexyl)diglycolamide (T2EHDGA) diluted in the ionic liquid trihexyl(tetradecyl)phosphonium nitrate (Cyphos nitrate: [P₆₆₆₁₄⁺][NO₃⁻]) was used to investigate the extraction of Eu(III) from aqueous solutions [72]. Results showed an increase in the REE extraction with an increase in the aqueous acidity followed by the attainment of a plateau region at higher acid concentrations (

Table 10. Effect of nitric acid concentration in the feed phase on Eu(III) extraction.

[HNO3], M	% Extraction
0.5	50
1	85
3	95
5	99
6	99
8	99

Feed phase: nitric acid spiked with ^(152+154)Eu(III) tracer. Organic phase: 0.1 M T2EHDGA and Cyphos nitrate. Equilibration time: 1 h. Temperature: 25 °C. O:A ratio: 1. From [72].

).

In the absence of T2EHDGA, the ionic liquid did not extract the rare earth. Thus, the extraction of europium(III) was due to the next equilibrium:

$${\mathrm{E}\mathrm{u}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{N}\mathrm{O}}_{3\mathrm{a}\mathrm{q}}^{-}+{2\mathrm{T}2\mathrm{E}\mathrm{H}\mathrm{D}\mathrm{G}\mathrm{A}}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow {\mathrm{E}\mathrm{u}(\mathrm{N}\mathrm{O})}_3·{2\mathrm{T}2\mathrm{E}\mathrm{H}\mathrm{D}\mathrm{G}\mathrm{A}}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(12)

The mixture of T2EHDGA/[P₆₆₆₁₄⁺][NO₃⁻] was implemented on the extraction of Fast Reactor Simulated High-Level Liquid Waste (FR-SHLLW) solution spiked with Am(III), and the results were promising with respect to the efficacy of the proposed IL phase. All the trivalent Ln(III) and Am(III) were selectively separated from some of the co-extracting fission products using a suitable holding agent in the feed phase. The present investigation uses dilute nitric acid (pH 2) as the stripping agent.

Two hydrophobic carboxylic acid ionic liquids, namely lauryl betaine bis(trifluoromethanesulphonyl)imide ([Laur⁺][NTf₂⁻]) and cocamidopropyl betaine bis(trifluoromethanesulphonyl)imide ([Coca⁺][NTf₂⁻]), were synthesized and used in the extraction of Sc(III) [73]. The best extraction conditions were 40 min (equilibrium), initial Sc(III) concentration of 0.001 M, sodium nitrate concentration of 0.5 M, and pH of 3. The extraction efficiency for the above ILs could even exceed 98.7% and 96.0%, respectively. Both ILs extracted Sc(III) by a cation-exchange mechanism, though the stoichiometry of the extracted species differed from one IL to the other:

$${\mathrm{S}\mathrm{c}}_{\mathrm{a}\mathrm{q}}^{3+}+{3([\mathrm{L}\mathrm{a}\mathrm{u}\mathrm{r}}^{+}][\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}{])}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{S}\mathrm{c}(\left[\mathrm{L}\mathrm{a}\mathrm{u}{\mathrm{r}}^{+}\right]\left[\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}\right]{)}_{3\mathrm{o}\mathrm{r}\mathrm{g}}+{3\mathrm{H}}_{\mathrm{a}\mathrm{q}}^{+}$$

(13)

$${\mathrm{S}\mathrm{c}}_{\mathrm{a}\mathrm{q}}^{3+}+{2([\mathrm{C}\mathrm{o}\mathrm{c}\mathrm{a}}^{+}][\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}{])}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{S}\mathrm{c}(\left[\mathrm{L}\mathrm{a}\mathrm{u}{\mathrm{r}}^{+}\right]\left[\mathrm{N}\mathrm{T}{\mathrm{f}}_2^{-}\right]{)}_{2\mathrm{o}\mathrm{r}\mathrm{g}}+{2\mathrm{H}}_{\mathrm{a}\mathrm{q}}^{+}$$

(14)

From the respective organic phases (Equations (13) and (14)), Sc(III) is completely stripped with 0.1 M and 0.2 M HNO₃ solutions. The extraction differences at different pH levels made it possible to separate Sc(III) from other REEs using both ionic liquids. The best Sc/Lu separation factor (238) is obtained with the lauryl betaine-bearing IL and pH 2, whereas in the case of the pair Sc/La, this IL also presented the best separation since La(III) is not extracted into the organic phase. Comparatively, the separation factors using [Coca⁺][NTf₂⁻] are 30.5 and 41.4 for the pairs Sc/Lu and Sc/La, respectively.

A series of bifunctional ionic liquids with methyltrioctylammonium as the cation and o-alkoxybenzoic acetate as the anion were synthesized [74] and dissolved in sulfonate kerosene to investigate the separation of REE and transition metal for NdFeB waste recycling. It was found that 0.12 M methyltrioctyl ammonium o-octyloxybenzoic acetate ([N₁₈₈₈⁺][OOB⁻]) solutions in kerosene could preferentially extract REEs in preference to Fe and Co. The extraction sequence was Dy > Nd > Pr >> Fe > Co, and although the extraction increased with the increase of the initial pH value in the 2.02–2.92 range (equilibrium pH values in the 4.20–4.96 range), at the limiting pH value of 2.92, the extractability follows the sequence Dy >> Nd > Pr > Co > Fe. The extraction mechanism for REE with the IL responded to an ion association in a molar ratio of 2:1 (IL:REE):

$${\mathrm{R}\mathrm{E}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+{2\mathrm{I}\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{R}\mathrm{E}{\mathrm{C}\mathrm{l}}_3·2\mathrm{I}{\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(15)

The REE loaded onto the organic phase is selectively stripped with deionized water. Contrary to what was mentioned in the title of the published manuscript, all the investigation was carried out using synthetic solutions and not solutions derived from the leaching of solutions of salt. For further references, see manuscripts related to NdFeB magnets.

Various functionalized ionic liquids were formulated to investigate their performance in the separation of yttrium from heavy REEs [75]. Extractions were carried out, dissolving the ionic liquid in kerosene. The results show that [N16MOP⁺][HDA⁻] is the best extractant to separate Y from the other REEs, achieving a Ho/Y separation factor of 2.82. The formation of a third phase (or second organic phase) makes it necessary to use a modifier. Both iso-octanol and tributylphosphate are investigated in this role, and although the presence of the alcohol tends to decrease metal extraction, the presence of the phosphoric ester derivative in the organic phase eliminated the third phase and enhanced the metal extraction. This enhancement is attributable to a synergistic effect. Taking Ho(III) as an example, the extraction mechanism responded to an ion-pair association:

$${\mathrm{H}\mathrm{o}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+2({[\mathrm{N}}_{16}\mathrm{M}\mathrm{O}{\mathrm{P}}^{+}][\mathrm{H}\mathrm{D}{\mathrm{A}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow ({[\mathrm{N}}_{16}\mathrm{M}\mathrm{O}{\mathrm{P}}^{+}]\left[\mathrm{H}\mathrm{D}{\mathrm{A}}^{-}\right]{)}_2\mathrm{H}\mathrm{o}\mathrm{C}{\mathrm{l}}_{3\mathrm{o}\mathrm{r}\mathrm{g}}$$

(16)

Complete stripping of the loaded REEs can be achieved using 0.1 M HCl solution. Furthermore, an industrial solution adjusted at pH 4 (La-Dy (23 mg/L), Y (1 g/L), Ho (35.8 mg/L), Er (0.11 g/L), Tm (27.2 mg/L), Yb (0.16 g/L), Lu (19 mg/L), Fe (0.4 mg/L), and Cu (0.6 mg/L), Ca (0.2 mg/L)) was extracted with an organic phase containing 0.013 M ionic liquid + 15% v/v TBP in kerosene. After the extraction, the values of the separation factors 2.52, 3.31, 4.22, 6.43, and 5.05 were obtained for the respective pairs Ho/Y, Er/Y, Tm/Y, Yb/Y, and Lu/Y.

An ionic liquid ([N₁₈₈₈⁺][NA⁻]) was synthesized from methyltrioctylammonium chloride ([N₁₈₈₈⁺]Cl⁻) and industrial naphthenic acid (NA) for the purification of Al³⁺ from GdCl₃ solution [76]. Under the experimental conditions, the extraction properties of diluted ionic liquid in kerosene diluted NA and saponified NA were comparatively investigated, and the separation factors Al/Gd were 23.0, 3.5, and 1.3, respectively. In the case of the ionic liquid, the extraction responded to the next equilibrium:

$${\mathrm{M}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+{\mathrm{I}\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{M}\mathrm{C}{\mathrm{l}}_3·\mathrm{I}{\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(17)

Near quantitative aluminum, stripping efficiency is obtained using 0.3 M NaOH or HCl solutions as strippants. On an industrial GdCl₃ feed, aluminum impurity was lowered from 889 mg/L to 5 mg/L. The removal efficiency of Al was 99.4%, and the purity of Gd was 99.998%.

The separation of Nd from Co and Ni using a pseudo-protic ionic liquid formed from the reaction of lauric acid (LA) and the primary amine N1923 (RNH₂) was investigated [77]. This ionic liquid ([RNH₃⁺][RCOO⁻]) was diluted in sulfonated kerosene, and again, the use of a modifier (iso-octanol) was mandatory to avoid the formation of a third phase. The increase in iso-octanol concentration in the 5–25% v/v range slightly decreased the extraction of the REE. Effective separation of Nd from Co and Ni can be achieved, with separation factors of 1503 and 2762 for the pairs Nd/Co and Nd/Ni, respectively. The next equilibrium was responsible for neodymium extraction:

$$\mathrm{N}{\mathrm{d}}_{\mathrm{a}\mathrm{q}}^{3+}+{3\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+2\mathrm{I}{\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{N}\mathrm{d}{\mathrm{C}\mathrm{l}}_3·2\mathrm{I}{\mathrm{L}}_{\mathrm{o}\mathrm{r}\mathrm{g}}$$

(18)

Na₂C₂O₄ and water were used as strippants with respect to the first and second steps, respectively. Furthermore, the Nd–oxalate solid was roasted, yielding a product of 99.97% wt.% Nd purity. In the treatment of leachate (1 g/L La, 92.9 mg/L Ce, 18.8 mg/L Pr, 0.3 g/L Nd, 0.25 g/L Co, 1.9 g/L Ni, and 0.21 g/L Mn) derived from the treatment of NiMH battery alloy, REEs are extracted preferably to base metals (

Table 11. Metal distribution ratios (D) in the extraction of a NiMH battery alloy leachate with [RNH₃⁺][RCOO⁻].

Element	D
La	15
Ce	quantitative extraction
Pr	quantitative extraction
Nd	125
Co	0.049
Ni	0.070
Mn	0.028

From [77].

).

The ionic liquid [N-methyl-N,N-dioctyloctan-1-aminium][2-(bis(2-ethylhexyl)amino)-2-oxoacetate] ([N₁₈₈₈⁺][DEAO⁻]) dissolved in kerosene was used to investigate the separation of thorium from a series of REEs [78]. The best separation of Th/REEs was achieved at a pH of 2.4 and in the absence of salt (NaCl) in the feed solution. With a maximum loading capacity of 1.02 g/L Th in a 0.04 M ionic liquid in the kerosene organic phase, the extraction equilibrium was represented by the equation:

$${\mathrm{T}\mathrm{h}}_{\mathrm{a}\mathrm{q}}^{4+}+{4\mathrm{C}\mathrm{l}}_{\mathrm{a}\mathrm{q}}^{-}+4([{\mathrm{N}}_{1888}^{+}][\mathrm{D}\mathrm{E}\mathrm{A}{\mathrm{O}}^{-}{]}_{\mathrm{o}\mathrm{r}\mathrm{g}}\leftrightarrow \mathrm{T}\mathrm{h}\mathrm{C}{\mathrm{l}}_4·([{\mathrm{N}}_{1888}^{+}][\mathrm{D}\mathrm{E}\mathrm{A}{\mathrm{O}}^{-}{]}_{4\mathrm{o}\mathrm{r}\mathrm{g}}$$

(19)

A mixture of saturated NaC₂O₄ and 1.5 M NaOH solution was used as a strippant.

3. Using Ionic Liquids and Rare Earth Elements in the Development of Advanced Materials

Different investigations utilized the properties of the ILs as a medium in the manufacture of REE materials with diverse uses. Next, some of these recent usages are reviewed.

The couple formed by Er(III) and 1-Butyl-methylimidazolium trifluorometahne sulfonate ionic liquid ([B_mim⁺][TFO⁻]) was used for the separation of the ethane/ethylene pair (C₂H₆/C₂H₄) [79]. However, the system needed the addition of dimethylformamide (DMF) to decrease both the density and the viscosity of the IL. At the same time, this DMF not only served as a diluent but it was also found that a strong interaction existed between Er³⁺ ions and DMF, resulting in a greater amount of Er³⁺ being complexed with ethane. The best ethane/ethylene separations were obtained in the [B_mim⁺][TFO⁻]/Er(TFO)₃ plus 20 wt.% DMF system at 1 + 1 molar ratio. Gas desorption was carried out at 30 °C and 0.1 KPa for one hour, and continuous use showed constant ethane adsorption after five cycles.

Lanthanum-doped graphene oxide functionalized with imidazole- or imidazolium-based benzothiadiazole ligands were used as fluorescent biosensors for the detection of cholesterol by colorimetric procedures [80]; for the above task, ChOx (cholesteroloxidase) enzyme was immobilized onto the various materials. Results showed that not all the compositions performed equally, as summarized in

Table 12. Response in cholesterol detection of GO-La-based biosensors.

Response	Biosensor + ChOx	Sensitivity, a.u.d. L/mg
Best	GO-La:BTDIm (1:1)	0.0649
Poor	GO-La:BTDAc (1:1)	<0.0602

Linear range of detection: 0.06–5.83 mg/dL range of cholesterol concentrations: BTDIm: 4,7-Bis-(imidazol-1-yl)-2,1,3-bezothiadiazole. BTDAc: 4,7-bis-[1-carboxymethyl-(imidazol-3-ium)]-2,1,3-benzothidiazole chloride. From [80].

, with the limiting detections yielded by some of these derivatives.

The next reference investigated the formation of neutral betaine species during the formation of protonated betaine bis(trifluoromethylsulfonyl)imide ([HBet⁺][[NT₂⁻]) ionic liquid [81]. This IL was widely used in the extraction of metals (including REEs), being synthesized by an anion-exchange reaction between [HBet⁺][Cl⁻] and [Li⁺] [NTf₂⁻] (two ionic liquids by themselves); during this synthesis, the following reaction occurred:

$$\mathrm{H}\mathrm{B}\mathrm{e}{\mathrm{t}}_{\mathrm{a}\mathrm{q}}^{+}+{\mathrm{H}}_2\mathrm{O}\leftrightarrow {\mathrm{B}\mathrm{e}\mathrm{t}}_{\mathrm{a}\mathrm{q}}+{\mathrm{H}}_3{\mathrm{O}}^{+}$$

(20)

and thus, a neutral zwitterionic betaine was formed due to the deprotonation of the carboxylic group of betaine. The above resulted in a loss of effectiveness in extracting metals. It was concluded that due to the reaction shown in Equation (20), acid must be added after the ion-exchange process responsible for the formation of the IL. This acid addition serves to promote the complete protonation of betaine, allowing maximum REE (and metal) extraction.

Al₂O₃/FeAl coatings doped with Er₂O₃ were prepared on CLAM steel by electrochemical deposition in ionic liquids and ion injection [82]. The procedure to fabricate these materials was based on CLAM steel samples working as anodes being electrochemically etched and aluminum-coating deposited in a 1-ethyl-3-methylimidazolium chloride ionic liquid and aluminum chloride (1:2 molar ratio) medium. Afterward, the erbium atoms were injected into the aluminum layer at room temperature. Following the corrosion experiments in liquid Pb-15.7Li at 550 °C and treatment for 1000 h, and contrary to the results observed from the steel without the presence of the REE, it was found that the steel with the surface of the coating doped with erbium oxide was smooth and uniform with a minor quantity of corrosion products.

In all-solid-state lithium-ion batteries with high energy densities, the substitution of the organic liquid electrolytes for perovskite-type Li_0.33La_0.56TiO₃ (LLTO) presented a series of advantages [83]. To improve the properties of this LLTO, 1-butyl-3-methylimidazolium tetrafluoroboarte ionic liquid ([BMIM⁺][BF₄⁻]) was used. The presence of the ionic liquid enhanced the total conductivity of the hybrid electrolyte, with a maximum of 4–7 × 10⁻⁴ S/cm at an IL concentration of 12.5 wt.%. This characteristic was attributable to the coupling of the ionic liquid with Li⁺, improving pathways for the migration of these ions across the boundaries, resulting in a decrease in the activation energy.

Ring-opening copolymerization (ROCOP) of epoxides and cyclic anhydrides to produce the aliphatic polyesters need carboxylic salts to actuate as active catalysts [84]. The presence of lanthanide salts improved the actuation of these catalysts. Thus, from the use of trihexyltetradecylphosphonium chloride ionic liquid [H3DP⁺][Cl⁻], several compositions of IL-LnCl₃ (Ln = Gd, Nd, Ho, Y) were tested. Two routes for the synthesis of these products were followed: (i) at 95 °C and wet formation of LnCl₃ · 6H₂O · 4[H3DP⁺][Cl⁻], and (ii) at room temperature and using water and dichloromethane with a dry formation of LnCl₃ · 4[H3DP⁺][Cl⁻]. After the synthesis, all the products were dried at 50 °C prior to its use. The best results, with respect to the cycloaddition process, were obtained when dry dichloromethane monomers were used and in the presence of YCl₃.

Several lanthanide-bearing paramagnetic ionic liquids, [C₁C_nim⁺]₃[LnCl_3×3³⁻] (n = 4, 6, 8; Ln = Dy, Gd, and Ho; X = Cl or Br), were synthesized and characterized [85]. All these PILs presented the properties expected from their conditions of ILs plus significant magnetic susceptibility (

Table 13. Volume magnetic susceptibility of some of the synthesized PILs.

PILs	Χ × 10−4	PILs	Χ × 10−4
[C1C8im+]3[HoCl63−]	1.759	[C1C4im+]3[HoCl3Br33−]	1.752
[C1C8im+]3[DyCl63−]	1.425	[C1C4im+]3[DyCl3Br33−]	1.650
[C1C8im+]3[GdCl63−]	1.014	[C1C4im+]3[GdCl3Br33−]	1.011

From [85].

).

Generally, susceptibilities followed the order Ho > Dy > Gd. However, in the case of LnCl₃Br₃³⁻ bearing PILs, the corresponding susceptibility depended on the alkyl chain of the PIL cation and decreased in the order C₄ > C₆ > C₈. These PILs presented potential applications in diverse areas, including biomedicine, mass transfer, magnetic resonance imaging, and environmental remediation.

A β-diketone, (1,1,1,2,2,3,3-heptafluoro-7,7-dimethyl-4,6-octanedione (HFOD)) was used to investigate its coordination with Ln(III) (Ln = Nd, La, and Eu) in the presence of different methylimidazolium-based ionic liquids [C_nmim⁺][NTf₂⁻] (n = 4, 6, 8) [86]. Two complexes were formed with Ln(HFOD)₂⁺ and Ln(HFOD)₄⁻ stoichiometries (Ln = Nd, Eu) and characterized. Furthermore, density functional theory calculation was carried out to optimize structural parameters such as bond angles and distances, and energies associated with the formation of respective complexes. The investigations concluded that the formation of the complexes responded to an anion-exchange reaction, which released NTf₂⁻ species to the aqueous phase.

A series of phosphors doped with REEs, Lu₇O₆F₉ (REEs = Eu, Gd, Yb, Er), were synthesized in a 1-octyl-3-methylimidazolium tetrafluoroborate ([O_mim⁺][BF₄⁻]) ionic liquid medium [87]. In the process of the growth of the precursors, water penetrated the IL, and BF₄⁻ was hydrolyzed according to

$$\mathrm{B}{\mathrm{F}}_4^{-}+{\mathrm{H}}_2\mathrm{O}\to \mathrm{B}{\mathrm{F}}_3·{\mathrm{H}}_2\mathrm{O}+{\mathrm{F}}^{-}$$

(21)

and fluoride ions reacted with the REEs to form REEF₃ compounds. It was described that the role of the ionic liquid was to act as a solvent, fluorine source, shape control agent, and crystal phase regulator. Lu₇O₆F₉:10%Yb³⁺,1%Eu³⁺ phosphors presented intense pure red luminescence, whereas Gd³⁺, Eu³⁺ codoped phosphors showed bright emissions, high thermal stability and paramagnetic properties.

Using [C_nmim⁺][NO₃⁻] and RE nitrate salts (RE = Ho, Er, Tm, Yb, Lu) as precursors, several ionic liquids with [C_nmim⁺]₂[RE(NO₃)₅²⁻] formulation were formed. The formation of C₁₂H₂₂N₉O₁₅RE crystals was also described [88]. These various ionic liquids presented properties that can be used in the development of soft materials with tunable optical transparency and light scattering.

In this reference [89], 1,3-bis(1-dicarboxylatoethyl)imidazolium bromide [H2bcei⁺][Br⁻] ionic liquid was used for the formulation, via a solvothermal procedure, of two lanthanide metal–organic frameworks (LnMOFs): Er₂[(C₉N₂O₄H₁₁)₃(H₂O)Br]Br · Cl}_n and Yb₂[(C₉N₂O₄H₁₁)₃(H₂O)_2]Br·Cl₂}_n. The synthesis of these compounds was dependent on the amount of NaOH used, temperature, type of organic solvents, and their volume ratios. These variables influenced the morphology and the single-crystal growth of the corresponding MOF. [bcei⁺] ligand adopted one bridging bidentate coordination mode to bond to Er and Yb. At 700 °C, both MOFs decomposed, and Er₂O₃ and Yb₂O₃ were formed. These two MOFs possessed stability, water solubility, and various properties, which made them candidates to be used in catalysis and biomedicine. Moreover, Yb–MOF presented antiferromagnetic interaction.

To improve a sustainable chemical industry, the direct synthesis of dimethyl carbonate (DMC) from renewable carbon dioxide (CO₂) and methanol (MeOH) was studied using CeO₂ nanoparticles modified with 1-butyl-3-methylimidazolium hydrogen carbonate ([BMIm⁺][HCO₃⁻]) ionic liquid (IL) [90]. The synthesized CeO₂@IL catalyst presented various DMC yields and methanol conversions depending on the temperature and pressure (

Table 14. Effect of temperature and pressure on MeOH conversion and DMC yield.

Temperature, °C	% Conversion	% Yield	Pressure, MPa	% Conversion	% Yield
80	3	2	1	2	<1
100	8	5	2	6	2
120	15	8	4	11	7
130	16	10	5	16	10
140	16	10	6	16	10

Operational conditions: 50 mg catalyst. 26.5 mM methanol. 6 h. Temperature experiments at 5 MPa CO₂. Pressure experiments at 130 °C. From [90].

), with the optimal conditions described as a CO₂ pressure of 5 MPa and temperature of 130 °C.

The catalytic behavior of CeO₂@IL suggests that the synergy between the two catalytic components originating from increased surface oxygen vacancies boosts the overall catalytic performance. After several recycling tests, the catalyst demonstrated no significant reduction in DMC yield and methanol conversion.

One approach to enhance the catalytic activity of metal–organic frameworks (MOFs) in CO₂ cycloaddition reactions is to create open metal sites within MOFs [91]. Thus, the amino-functionalized REE Gd-MOF (Gd-TPTC-NH₂) and its ionic liquid composite catalysts (Gd-TPTC-NH-[B_MIM⁺][Br⁻]) were synthesized using 2′-amino-[1,1′:4′,1″-terphenyl]-3,3″,5,5″-tetracarboxylic acid (H₄TPTC-NH₂) as the ligand. Under the optimized reaction conditions of temperature and catalyst dosage, Gd-TPTC-NH-[B_MIM⁺][Br⁻] can effectively catalyze the cycloaddition reaction of a variety of epoxide substrates with promising yields of cyclic carbonate products. Comparatively, epichlorohydrin and epibromohydrin, which possess halogen substituents, promote higher yields of cyclic carbonates due to the electron-withdrawing nature of Cl⁻ and Br⁻ substituents.

CeO₂ (a widely used abrasive) can be synthesized via coprecipitation, hydrothermal, and solvothermal methods. To overcome some issues encountered within these syntheses, an ionothermal synthesis approach for high-quality CeO₂ particles using different ionic liquids without a template was investigated [92]. The ionic liquids used in the investigation were: 1-ethyl-3-methylimidazolium tetrafluoroborate ([E_mim⁺][BF₄⁻]), 1-butyl-3-methylimidazolium tetrafluoroborate ([B_mim⁺][BF₄⁻]), and 1-hydroxyethyl-3-methylimidazolium tetrafluoroborate ([HOEt_mim⁺][BF₄⁻]). Based on the scanning electron microscopy and particle size distribution, CeO₂ spherical and crystalline particles, synthesized at a reaction temperature of 210 °C for 16 h, displayed sizes ranging between 0.1 and 0.3 μm. The results of chemical mechanical polishing (CMP) experiments indicated that the abrasives could reduce surface roughness from 1.63 to 0.29 nm (scanning area is 5 μm × 5 μm). The obtained synthesized CeO₂ particles demonstrated promising performance in Si polishing, achieving removal rates of up to 255.5 nm/min.

In the next reference [93], a terbium-based ionic liquid, vinylimidazolium tetranitroterbium ([V_im⁺][Tb(NO₃)₄⁻]), and the corresponding polymer, poly(vinylimidazolium tetranitroterbium) (P[V_im⁺][Tb(NO₃)₄⁻]), are synthesized by introducing terbium into counterions. Magnetic property tests show that the monomer and polymer are paramagnetic, and the susceptibility of P[V_im⁺][Tb(NO₃)₄⁻] is higher than that of the former. The magnetic monomer and polymer show bright fluorescence under ultraviolet excitation, while there is no color under sunlight. [V_im⁺][Tb(NO₃)₄⁻] exhibits stronger fluorescence than the polymer form in solution, while pure P[V_im⁺][Tb(NO₃)₄⁻] exhibits stronger fluorescence than pure monomer. Thus, the polymer seems to be more suitable for anti-counterfeiting applications and droplet manipulation.

Zero-valent scandium, zirconium, hafnium, and manganese nanoparticles are prepared by the reduction of ScCl₃, ZrCl₄, HfCl₄, and MnCl₂ with lithium or sodium naphthalenide in a one-pot, liquid-phase synthesis [94]. Small-sized monocrystalline nanoparticles are obtained with diameters of 2.4 ± 0.2 nm (Sc), 4.0 ± 0.9 nm (Zr), 8.0 ± 3.9 nm (Hf) and 2.4 ± 0.3 nm (Mn). Therefore, Zr⁰ and Hf⁰ nanoparticles are shown for the first time. To probe the reactivity and reactions of the as-prepared zero-valent REE nanoparticles, they are reacted in the liquid phase (e.g., THF, toluene, and ionic liquids: ethylmethylimidazolium chloride ([EMIm⁺][Cl⁻]) or butylmethylimidazolium chloride ([BMIm⁺][Cl⁻])) with different sterically demanding, monodentate to multidentate ligands mainly comprising O-H and N-H acidic alcohols and amines. These include isopropanol (HOiPr), 1,1′-bi-2-naphthol (H2binol), N,N′-bis(salicylidene)ethylenediamine (H2salen), 2-mercaptopyridine (2-Hmpy), 2,6-diisopropylaniline (H2dipa), carbazole (Hcz), triphenylphosphane (PPh3), N,N,N′,N′-tetramethylethylenediamine (tmeda), 2,2′-bipyridine (bipy), N,N′-diphenylformamidine (Hdpfa), N,N′-(2,6-diisopropylphenyl)-2,4-pentanediimine ((dipp)2nacnacH), 2,2′-dipydridylamine (Hdpa), and 2,6-bis(2-benzimidazolyl)pyridine (H2bbp). As a result, 22 different compounds are obtained, which frequently exhibit a metal center coordinated only by the sterically demanding ligand. However, sometimes the final compound did not contain the REE:

$$4\mathrm{S}{\mathrm{c}}^0+12\left(\left[\mathrm{E}\mathrm{M}\mathrm{I}{\mathrm{m}}^{+}\right]\left[\mathrm{C}{\mathrm{l}}^{-}\right]\right)+12\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}\to 3(\left[\mathrm{L}{\mathrm{i}}_4{(\mathrm{C}}_6{\mathrm{H}}_{10}{\mathrm{N}}_2{)}_4{\mathrm{C}\mathrm{l}}_4\right])+4{\mathrm{S}\mathrm{c}\mathrm{C}\mathrm{l}}_3+{6\mathrm{H}}_2$$

(22)

Most compounds show the metal coordinated by the sterically demanding ligand:

$$\mathrm{S}\mathrm{c}+5\left(\left[\mathrm{B}\mathrm{M}\mathrm{I}{\mathrm{m}}^{+}\right]\left[\mathrm{C}{\mathrm{l}}^{-}\right]\right)+{3\mathrm{A}\mathrm{l}\mathrm{C}\mathrm{l}}_3\to {[\mathrm{B}\mathrm{M}\mathrm{I}{\mathrm{m}}^{+}]}_5{[\mathrm{A}\mathrm{l}\mathrm{C}\mathrm{l}}_4^{-}][{\mathrm{S}\mathrm{c}\mathrm{C}\mathrm{l}}_6^{4-}]+{\mathrm{A}\mathrm{l}}^0$$

(23)

$${3\mathrm{Z}\mathrm{r}}^0+3\left(\left[\mathrm{B}\mathrm{M}\mathrm{I}{\mathrm{m}}^{+}\right]\left[{\mathrm{C}\mathrm{l}}^{-}\right]\right)+{4\mathrm{A}\mathrm{l}\mathrm{C}\mathrm{l}}_3+3\mathrm{N}\mathrm{a}\mathrm{C}\mathrm{l}\to 3\left(\left[\mathrm{B}\mathrm{M}\mathrm{I}{\mathrm{m}}^{+}\right]\left[\mathrm{N}\mathrm{a}\mathrm{Z}\mathrm{r}{\mathrm{C}\mathrm{l}}_6^{-}\right]\right)+4{\mathrm{A}\mathrm{l}}^0$$

(24)

Options and restrictions for the liquid-phase syntheses of novel coordination compounds using the oxidation of base-metal nanoparticles near room temperature are evaluated.

Garnet composites with ionic liquid dispersion have been prepared for electrolytic applications. Various ionic liquids in small amounts were added to garnet-type Li_6.75Al_0.25La₃Zr₂O₁₂ (LALZO) to improve its ionic conductivity and electrode–electrolyte interfacial compatibility [95]. The optimal composition with near 6 wt.% [EMIm⁺][BF₄⁻] in LALZO showed a high ionic conductivity of 6 × 10⁻⁴ Ω⁻¹cm⁻¹ at room temperature, which is almost two orders of magnitude higher than the pristine garnet pellet. Such a high conductivity is attributed to the alteration of the ionic liquid–garnet interface by a weak non-uniform chemisorption. It is concluded that these materials can be used for energy storage applications.

The luminescent amphiphilic polyoxometalates (POM) complex, namely surfactant-encapsulated POM hybrid (SEP), was prepared mainly through electrostatic interaction from polyanionic lanthanide-containing POM, Na₉(EuW₁₀O₃₆) · 32H₂O (EuW10) and cationic surfactant, 1-octadecyl-3-methylimidazolium bromide ([C₁₈mim⁺][Br⁻] (OB)) [96]. Next, highly luminescent lamellar microfiber aggregates were assembled by SEP in a polymerizable imidazolium-IL solvent, i.e., 1-vinyl-3-butyl imidazolium terafluoroborate ([Vbim⁺][BF₄⁻]). After morphology and structure characterizations for soft aggregates in detail, a lamellar aggregation arrangement mechanism was proposed. Furthermore, through in situ photopolymerization for such aggregate matrices, flexible and organized films with enhanced luminescent properties were obtained. The investigation shows that the rational design of polymerizable organized aggregate matrices in ionic liquids improves the photophysical properties of such luminescent soft materials and broadens their practical applications. These results indicate that hybrid co-assembly from lanthanide-containing polyoxometalates and surfactants optimizes the formed aggregate luminescent properties and structural diversity. Moreover, compared to conventional solvents, aprotic ionic liquids are useful as assembly media and act as remarkable solvents for luminescent lanthanide compounds.

A strategy of lanthanum manganese mixed metal oxides (LMMO) regulated via an ionic liquid-assisted hydrothermal method was proposed to yield multifunctional catalysts, which exhibited excellent catalytic performance in the selective aerobic oxidation of cyclohexane [97]. Up to seven different ionic liquids were tested, with LMMO-[B_mim⁺][HSO₄⁻]-1.0 wt.% having the highest cyclohexane conversion (8.9%) and 90% KA oil selectivity, against the above, LMMO-[B_mim⁺][BF₄⁻]-1.0 wt.% presenting the lowest values of conversion (4%) and selectivity (59.2%). The investigation indicated the structure-directing and template effect of the ionic liquids on the multifunctional catalysts.

Ascorbic acid (AA) is widely used as an antioxidant in several industries, e.g., food. The concentration of this essential chemical can be influenced by different factors: time, temperature, and pH value during food preparation and storage, leading to its instability in content. Thus, the accurate detection of AA content is of paramount importance to avoid health issues. Based on the above and using the ionic liquid 1-aminopropyl-3-methylimidazolium bromine ([APMIm⁺][Br⁻]), the formulation of an imidazole-based ionic liquid-functionalized carbon dot (CD) hybrid APM-CD was investigated [98]. A photo-functional hybrid probe Eu³⁺@APM-CDs was obtained via an ion-exchange reaction between APM-CD and lanthanide complex Eu(DPA)₃. (DPA = 2,6-pyridinedicarboxylic acid). The results showed that the fluorescent intensity of Eu³⁺@APM-CDs had a good linear response to AA concentrations in the range of 25–500 μM, with a detection limit of 80 nM. The fluorescence probe is highly selective with respect to ascorbic acid and in the presence of analogs such as aspartic, glutamic, and citric acids, glycine, threonine, alanine, arginine, lysine, and fructose.

As mentioned in the Introduction, the green credentials for ionic liquids are arguable. Thus, investigations about the degradation compounds of these chemicals are of interest. In the next reference [99], La and Ce-codoped TiO₂@PAM (polyacrylamide) composite microspheres were formed to investigate the degradation of three imidazole ionic liquids. The experimental results showed that the composite La (0.25%) and Ce (0.15%)-codoped TiO₂@PAM composite microspheres with calcination temperature of 450 °C presented a photocatalytic activity of 97.4%, 91.2%, and 88.5% for 1-butyl-3-methyl imidazolium hexafluorophosphate, 1-hexyl-3-methyl imidazolium hexafluorophosphate, and 1-octyl-3-methyl imidazolium hexafluorophosphate, respectively, and using a concentration of 10 mg/L and 90 min of reaction time. In all three cases, the various intermediates finally degraded to CO₂ and water. Though the catalyst degrades with the number of cycles (1–6), this degradation is less than that shown by the TiO₂P₂₅ catalyst (

Table 15. Photodegradation of imidazolium-based ionic liquids at various cycles.

Catalyst	1-butyl	1-hexyl	1-hexyl-3-methyl
La/Ce-based	1st(95%)–6th(90%)	1st(90%)–6th(80%)	1st(85%)–6th(80%)
TiO2P25	1st(70%)–6th(50%)	1st(55%)–6th(40%)	1st(50%)–6th(35%)

Temperature: 25 °C. Dosage: 40 mg of the catalyst. Experiments under xenon lamp irradiation. From [99].

). In all three cases, the various intermediates finally degraded to CO₂ and water.

Again, in relation to the food industry, the determination of bisphenol A is of the utmost necessity to ensure food safety. The next reference [100] developed an electrochemical sensor based on a modified carbon paste electrode (CPE) to measure this bisphenol A. The sensor was developed using CeO₂ nanoparticles and the ionic liquid 1-butyl-3-methylimidazolium hexafluoro phosphate (BMI⁺)[PF₆⁻] to modify the carbon-pass electrode. Under the optimum detection conditions, the CeO₂/IL/CPE sensor showed good linearity in the detection of bisphenol A within the range of 0.02–460.0 µM with a limit of detection (LOD) of 0.01 µM. There were no observed important interferences in the presence of a number of metal ions (Pb²⁺, Li⁺, Ag⁺, Zn²⁺), anions (CN⁻, SO₄²⁻, SCN⁻), and other chemicals (ascorbic acid, glucose, urea, fructose, sucrose).

The next investigation [101] combined the use of europium-doped diamond-like carbon (Eu-doped DLC) with varying atomic concentrations of the dopant element (1.7 at% and 2.4 at%) and gadolinium-doped diamond-like carbon (Gd-doped DLC) with different atomic concentrations of the dopant element (1.7 at% and 2.3 at%) together with a pure DLC coating, incorporating an ionic liquid (1-ethyl-3-methylimiazolium diethylphosphate) additive in a tribological block-on-ring system. In the investigation, solutions at 1% wt.% of the ionic liquid in polyalphaolefin 8 were used. With respect to the tribological properties of these systems, the lubricant based on europium presented the best results. It was demonstrated that the solution of the ionic liquid in polyalphaolefin 8 induced a significant reduction in wear for all the investigated coatings. A composite lubricating system that combines solid and liquid lubrication created a synergistic system that takes advantage of the strengths of both types of lubricants.

Since anhydrous REE chlorides have a series of practical applications in various fields (organic metallic chemistry, catalysts, etc.), it is of interest to investigate approaches to the established synthesis methods. This reference [102] uses the reaction of RE₂O₃ in [BMIm⁺][Cl⁻] · nAlCl₃ at 175 °C to fabricate pure anhydrous RECl₃, where [BMIm⁺][Cl⁻] stands for 1-butyl-3-methylimidazolium chloride. The synthesis procedure allowed for the obtaining of these pure salts in the case of La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, and Dy.

4. Conclusions

Ionic liquids are used to accomplish various roles in the recovery of REEs from raw or secondary (waste) resources. They are used in various unit operations involved in hydrometallurgical–solvometallurgical processing taken from leaching and exchange operation with solid ion-exchange resins and electrodeposition. Even laboratory operation-supported liquid membranes used this type of compound.

Table 16. Comparison of the various technologies.

Technology	Current Situation	Status
Leaching	Investigations on the recovery of REEs from various sources	Laboratory
Electrodeposition	Investigations on synthetic solutions	Laboratory
Ion-exchange resins	Investigations on synthetic solutions	Laboratory
Membranes	Investigations on synthetic solutions	Laboratory
Liquid–liquid extraction	Investigations mainly performed on synthetic solutions and batch operation mode. Sometimes, the stripping stage was not investigated	Laboratory

summarizes the status of these different technologies; all investigations have not gone beyond the laboratory.

As somewhat expected due to the tendency toward environmentally friendly practices in which we are living, most of the investigations are related to the use of these chemicals in liquid–liquid extraction studies. However, the reality shows that in many cases, the use of the ILs is accompanied by the use of classical organic diluents. Thus, the apparent green advantage of the utilization of these ILs versus traditional liquid–liquid extraction is lost. It is worth noting here that a series of investigations used and proposed applications of ILs in systems that utilized highly toxic (and forbidden) organic diluents and that the corresponding reviewers (and editors) do not realize and criticize this issue.

This review also shows that in a series of these proposals, researchers do not present data about the recovery of REEs from stripped solutions and liquid–liquid extraction, a two-step operation: extraction and stripping. Moreover, some publications do not present data on how these REEs are finally recovered from the solutions.

Also, in a series of systems, ILs take an active part in the recovery of REEs from the solutions, sometimes releasing part of the organic moiety from the IL to the corresponding raffinate, causing a probable environmental problem, together with the destruction of the IL, with little attention from the authors to the above and the regeneration of the IL.

At present, the industrialization or even scaling up, from laboratory to pilot plant, of technologies using ILs is far from being real, requiring more fundamental data on the efficiency of the involved processes and investigations of the recovery and separation of REEs from real wastes.