Advances in Analytical Techniques and Applications in Exploration, Mining, Extraction, and metallurgical Studies of Rare Earth Elements

1. Introduction

At present there is an intense exploration activity going on the world over for critical elements like rare earth elements (REE), lithium, gold, platinum, palladium, copper, and cobalt because of their wide applications in several green technologies required to transition to a low-carbon economy. In fact, currently, these elements are fulfilling thousands of different needs in our technology-powered society. REE group of elements consists of the 15 lanthanide elements (La to Lu) plus Y and Sc. Based on atomic numbers, they are divided into two groups. The lower atomic weight elements from La to Sm, the most abundant ones, with atomic numbers 57-62 are referred to as light REE (LREE); while Eu to Lu, and the least common and the most valuable with atomic numbers 63-71, are known as heavy REE (HREE). Despite their low atomic weights, Y and Sc are included in the HREE subgroup because of their co-occurrence, similar ionic radii, and closer behavioural properties to HREE than LREE (Figure 1). Because of their unique physical, chemical, electronic, optical, mechanical, catalytic, and magnetic properties, they are being extensively used to make different high-technology devices such as computers, televisions, smartphones, catalysts for fuel cells, light emitting diodes, hard drives for computers, corrosion inhibitors, and magnets for wind turbines, and other power generating systems. Table 1 presents REE concentrations in different earth materials. Concentrations of REE in different earth materials provide critical information about the origin and evolution of the Earth and especially Ce and Eu are extremely sensitive to changes in atmospheric conditions with two different oxidation states each (Ce³⁺/Ce⁴⁺ and Eu²⁺/Eu³⁺), hence, these two redox pairs are used to understand oxygen fugacity (fO₂) in different geological environments [1,2]. Because of these similarities, all these elements are usually studied as a group in several research and development studies.

The demand for REE, Y, and Sc is increasing day by day especially because of their utility in green technology applications as mentioned above. As a result, there has been a significant surge in the exploration, mining, and extraction activities for these elements worldwide. European Commission and the US government declared REE group as an economically critical elements [4,5] and currently, there is intense exploration activity going on the world over for discovering new economically viable deposits. Even the mining, efficient extraction of REE from the ore materials, and their metallurgy studies assume a lot of importance at present. There is a need to have rapid, ecological, and cost-efficient analytical techniques during exploration, geochemical mapping, mining, and ore processing operations, and even during metallurgical works [6,7]. In all these activities, it is essential to analyse several types of geological and industrial materials for REE, Y, and Sc. Usually, it is very complicated to determine them by classical methods such as gravimetry, titrimetric, and spectrophotometry because of close similarities in their physical and chemical properties and particularly so when a selected REE element among them has to be determined in the mixture of the other REE, because of numerous interferences and coincidences. Moreover, it is extremely difficult to determine them at crustal levels of concentrations in various geological materials. But astonishing advances have taken place during the last three decades in the analytical techniques for the detection and determination of different elements including REE not only in terrestrial materials but also in faraway planets such as Moon and Mars. Recently eight elements including one of the REE, Tb were detected in the exoplanet's (KELT-9 b's) atmosphere using high-resolution spectrographs [15]. Exoplanets are planets that are in other solar systems than our own. Usually, sophisticated instrumental analytical techniques like instrumental neutron activation analysis (INAA) and different forms of inductively coupled plasma mass spectrometry (ICP-MS) including tandem-ICP-MS, ICP-time-of-flight-MS (ICP-TOF-MS), high resolution-ICP-MS (HR-ICP-MS), Multi-collector-ICP-MS (MC-ICP-MS) and Mattauch–Herzog geometry-ICP-MS (MH-ICP-MS) are commonly used for REE determination in different kinds of materials because of their multi-element capability, high sensitivity wide linear dynamic range, fewer interferences, and ease of operation. Dai et al. [16] in a study related to the understanding of modes of occurrences of different elements including REE, used from simple AAS to complex SHRIMP instrument. Recent developments in various instrumental analytical techniques including the introduction of portable techniques and some new analytical techniques and their applications in the determination of REE in various geological materials during exploration studies, mining, extraction, and metallurgical processes form the focal point of this review. In addition, advances in sample preparation/sample dissolution, and quality control protocols for obtaining accurate data are also discussed. Some of the important analytical techniques and their utility towards REE elemental, isotopic, and mineral analyses are considered and discussed in the following.

2. Instrumental Analytical Techniques

Recent advances especially in ICP-MS technology led to the development of several types of ICP-MS instruments with amazing performance characteristics such as rapid elemental as well as isotopic analyses, very high sensitivity, and fewer interference effects. Recent developments in microelectronics and computer technologies helped some of the analytical techniques like XRF and LIBS shrink in size and became hand-held helping the exploration scientist to take them directly to fields to generate data rapidly. These portable techniques have together made the analysis of REE in different types of geological and industrial materials rapid, and easy. Latest ICP-MS instruments like ICP-MS/MS and HR-ICP-MS provide more choices and effective solutions for separating spectral interferences in REE analysis while taking the detection limits down to sub-pg/g levels. Internal standardisation only minimizes the drift effects because the drift is usually non-linear, depends on mass, and changes frequently when analysing over large mass ranges and many times offline data reduction procedures were helpful for the correction of drift [17]. Figure 2 Schematic representation of different major analytical techniques used for elemental, isotopic, and mineralogical studies related to REE in geological materials.

2.1. UV-Vis Spectrophotometry

UV/Vis. spectrophotometry is one of the oldest techniques that can be used to determine extremely low concentrations of several metals including REE. If the desired constituent is not self-coloured, the addition of a selective ligand selectively binds to metal to produce a coloured complex with a higher molar absorptivity to enable their sensitive determination. There are a number of spectrophotometric methods available in the literature for the determination of different REE. For example, Pu et al. [18] developed a spectrophotometric method that involves microcolumn online preconcentration for the determination of ∑ REE which is highly sensitive, selective, and accurate. Saputra et al. [19] developed a fast quantitative analytical method by combining ultraviolet-visible spectroscopic and multivariate analysis for the determination of Sm, Eu, Gd, Tb, and Dy, and successfully applied it to the quantitative analysis of these REE in monazite samples. Though spectrophotometry is an established technique for the determination of several metals including REE in geological materials, most methods are complex and suffer from several interferences from the matrix components, not very sensitive, some procedures are lengthy and many times it is difficult to determine one REE in the presence of other because of close chemical similarities. Moreover, a lot of other instrumental multi-element analytical techniques are available currently. As a result, spectrophotometry methods are not very popular, especially during exploration, mining, ore processing, and metallurgy studies.

2.2. X-ray Fluorescence Spectrometry (Both WD-XRF & ED-XRF)

X-ray fluorescence spectrometry (XRF) is one of the most frequently used non-destructive analytical techniques for determining the elemental composition of different geological materials like rock, sediment, soil, and ore samples. The technique can be classified into wavelength-dispersive X-ray fluorescence spectrometry (WD-XRF), and energy-dispersive X-ray fluorescence spectrometry (ED-XRF) depending on the methods of excitation, dispersion, and detection [20]. Both forms of X-ray fluorescence techniques are very popular for the determination of REE in geological and environmental materials. XRF works on the principle that when a sample is bombarded with high-energy X-rays, the atoms in the sample get ionized by losing electrons. The resulting electron hole in the inner shell is then filled by an outer-shell electron accompanied by the release of energy in the form of a photon which is known as fluorescence. The energy of the emitted radiation reflects the energy difference between the two shells involved which is also characteristic to the element present in the sample and its intensity is directly proportional to the concentration. When the instrument is calibrated with the known concentration of a particular element, it is possible to determine the exact concentration of the same element in an unknown sample. XRF is a reliable non-destructive and multi-element analytical technique for several trace elements analyses at the µg/g level although the technique is relatively insensitive to the REE. In general, XRF offers very high detection limits for REE and is not suitable for many types of geological materials hence several determination procedures involve separation and preconcentration of REE for their accurate determination [21]. For example, Juras et al. [22] separated all REE from other constituents by an ion-exchange procedure and determined the different types of geological materials ranging from ultramafic rocks to rhyolites. De Vito et al. [21] developed a separation and preconcentration method for REE using (o-[3,6-disulfo-2-hydroxy-1-naphthylazo]–benzenearsonic acid which is retained on a polyamide membrane by a chemo-filtration process. The membrane containing the REE is measured by XRF and this preconcentration method allows the detection of very low concentrations of REE. A comprehensive report on the applications of XRF spectroscopy in the Chinese REE industry was given by Wu et al [23]. These applications involve the XRF analysis of REE in rocks, minerals, ores, soils, and concentrates, raw metals, alloys, functional materials, along with a fast and online separation process. De Pauw et al. [24] determined REE in geological materials by the WD-XRF technique which was designed to measure the L-lines of REE between 4.5 and 7 keV with a sensitivity down to the µg/g level. The method could detect REE in the inclusions in deep Earth diamonds with detection limits lower than 0.50 µg/g for characteristic L-lines which were 10 times lower than those of regular K-lines. Adeti et al. [25] demonstrated the limitations of the tube-based XRF for the determination of REE in geological materials. The major reason was found to be the interference between the K-series X-ray emission from the transition metals and the relatively low intensities of the L-series lines of the REE. Carbonatite tailings from an Australian mine were analysed by XRF and ICP-MS for assessing the REE recovery potential. The carbonatite-related tailings were found to contain about 9% REE [26]. ED-XRF is capable of measuring from Na through U including elements like sulphur in a variety of geological matrices for a wide range of applications including mineral exploration and quality control. Wenqi et al. [28] summarised various XRF applications of REE in China. Table 2 presents selected REE determined in an inhouse monazite CRM by ED-XRF. In general, XRF techniques are not sensitive enough to detect and determine low concentrations of REE in rocks and usually, separation and preconcentration methods are employed to make these trace elements detectable. Table 3 presents a few more applications in REE research studies. Figure 3 presents a schematic diagram showing REE analysis by ED-XRF.

2.3. Instrumental Neutron Activation Analysis (INAA)

INAA is a highly sensitive and versatile multi-element analytical technique for the accurate determination of the concentration of major, minor, and trace elements including REE in a variety of geological materials. Usually, small amounts of samples (~5 to 100 mg) in polythene bags along with standards are subjected to a neutron flux in a nuclear reactor. The stable nuclei absorb neutrons during irradiation and become radioactive, and the resultant radioactive nuclides decay with the emission of particles or, more importantly gamma rays, which are characteristic of the elements from which they are emitted. The nuclide is identified by its gamma-ray energy and the intensity of the gamma-ray is directly proportional to its concentration, and by comparing it to that of the known standard the concentration of a particular nuclide/element can be calculated. Both scintillation and semiconductor-type detectors are normally used for quantitative measurement. As nuclear reactions and decay processes are virtually not affected by the chemical and physical structure of the material during and after irradiation and the composition of the matrix has little influence on the induced activity, matrix effects are minimum with several types of geological materials and the results are very accurate. These methods are very popular in places where nuclear reactors are available for sample irradiation. Although INAA is a nondestructive method and detects all REE at ng/g levels (Table 3) in several rock types, it is difficult to make accurate measurements when the matrix radionuclides with higher radioactivity led to a high background in the gamma ray spectrum. Sometimes after sample irradiation, the matrix is separated by methods such as coprecipitation, ion exchange, and solvent extraction, and the measurement is performed by radiochemical NAA (RNAA). However, several studies ranging from pure geochemical studies to REE exploration studies employed INAA techniques for the determination of REE in different earth materials [40,46,47]. Table 4 presents REE concentrations determined by INAA in zircon concentrate CRM OREAS 100a together with certified values. Depending upon the background intensity, and half-life, different elements were determined by two different techniques, namely INAA and ED-XRF. More details on the methodology followed are provided by Silachyov [48]. Ravisankar et al. [49] used INAA to measure REE in beach rock samples using a single comparator method. Samples along with SRM 1646a (estuarine sediment) were irradiated using a thermal neutron flux of ~10¹¹n cm^-2 s^-1 at 20 kW power using the Kalpakkam mini reactor, IGCAR, Kalpakkam, Tamil Nadu, India. REE were determined in 15 samples using high-resolution gamma spectrometry. Figure 4 depicts the INAA analysis set up for Geoanalysis.

In order to understand the distributions of REE and their source in mangrove surface mangrove sediments of Juru River, West Coast of Peninsular Malaysia, Krishnan and Saion [50] used INAA. Samples along with CRM, SL-1, and blank samples were irradiated together before measurements. Ahmed et al. [51] analyzed REE in soil samples collected from gold-mining sites by INAA to understand REE concentrations in gold-mining areas in Sudan. The accuracy of REE data was checked using SRM–NIST 2586. Kin et al [52] made a comparative study on the determination of REE by INAA and ICP-MS and found that INAA is more powerful with very good precision and accuracy though all REE cannot be determined by INAA. El-Taher at al. [53] presented an overview of the capabilities of different nuclear techniques such as INAA, radiochemical NAA (RNAA), WD-XRF, ED-XRF, and total reflection XRF (TXRF) for the determination of REE in different kinds of materials including geological materials such as rocks, minerals, sediments, and soils, along with some case studies. Adeti et al. [25] used four nuclear analytical techniques for the determination of Sc, La, Ce, Nd, Sm, Eu, Tb, and Lu in volcanic rock specimens along with an IAEA-Soil-7 CRM. These techniques are: i) Am-241 excitation-based XRF, ii) Ag-anode X-ray tube XRF, iii) INAA, and iv) regular XRF. Except for the Ag-anode X-ray tube XRF method, all other three methods gave comparable results, with the best results coming from Am-241 excitation-based XRF. The measurement is faster and the sample preparation is minimal and the overall analysis time is shorter than with other techniques. Though INAA is popular for highly sensitive multi-element analysis, it certainly has some limitations. INAA needs matrix-matching reference materials for calibration to allow for the emitted x-rays which are subjected to self-attenuation in the sample under the same counting conditions. Hence synthetic standards prepared by mixtures of REE are not suitable. It is also time-consuming, not independent, and involves longer cooling times for certain elements. On the other hand, some nuclides are very short-lived and cannot be determined with reasonable accuracy. Moreover, the method is not independent, and requires a reactor nearby.

2.4. Indirect Measurement of REE by the Radiometric Method

Ghannadpour et al. [54] used methods based on radioactivity and radiation measurement for the exploration studies of REE in the Central Sangan iron ore mine, NE of Iran. These authors found a direct correlation between the concentration of radioactive elements (U, Th, and K) and REE mineralisations, and thus this is a useful exploration technique for REE and presents a novel viewpoint to decision-makers of the exploration industry. Drill core samples in the field were directly analyzed for radioactive elements, U, Th, and K separately by the handheld gamma-ray spectrometer. Individual REE were determined by ICP-MS, and it was observed that radioactivity in the study area is entirely affected by uranium and ∑REE showed acceptable correlation with radioactivity and consequently uranium. This is some kind of indirect determination of ∑REE using a handheld gamma-ray spectrometer [55]. In order to understand the radioactivity levels of radionuclides of ²³⁸U, ²³²Th, ²²⁶Ra, ⁴⁰K, near the REE processing plant in Guangdong, China, the natural radioactivity of these radionuclides in the surrounding soil samples was determined using a high-purity Germanium energy spectrometer.

2.5. Atomic Absorption Spectrometry (Both Flame-AAS & GF-AAS)

Flame atomic absorption spectrometry (F-AAS) is based on the principle that the amount of light absorbed by the atoms in the flame is a measure of the concentration of a particular analyte at a particular wavelength [56]. By comparing the absorption of a known standard with that of an unknown, the concentration of a particular element can be computed in an unknown sample. On the other hand, graphite furnace AAS (GF-AAS) is a technique that offers better sensitivity and involves the injection of a small amount of sample solution to be analyzed into a small graphite tube [57] and thus is suitable for the analysis of metals at ultra-trace levels (ng/ml) [58]. F-AAS methods are well suited for single-element analysis and useful only when very high concentrations are being analyzed particularly from the REE analysis point of view. Because of the low cost of the equipment and low running cost, F-AAS instruments are being used even now by many industries. Determination of REE in geological samples by flame-AAS is complicated by less sensitivity of the technique for these group of elements coupled with severe spectral interferences by other REE, Si, Na, Fe, Al, hydrofluoric acid, phosphates, etc., even after using high-temperature acetylene-nitrous oxide flame. Individual REE were successfully determined using flame-AAS after adding excess potassium to the sample solution to suppress ionization interferences. Currently because of the availability of other powerful techniques such as ICP-MS, the use of AAS techniques for the determination of REE is very limited, particularly at low concentration levels.

2.6. Microwave Plasma Atomic Emission Spectrometry (MP-AES)

Microwave atomic emission spectrometry (MP-AES) is relatively a new analytical technique that was introduced in 2011 and is an attractive alternative to flame-AAS and ICP-OES [59,60]. Figure 5 presents a schematic diagram that explains the principle of MP-AES, and REE analysis. This technique utilizes a relatively new design of plasma torch, utilizing nitrogen gas for generating high-temperature microwave plasma, and proved to be attractive for the analysis of different types of geological samples, including industrial effluents, water, sediments, soils, rocks, and ores [61,62,63]. Helmeczi et al. [64] developed a novel and rapid digestion procedure for the dissolution of REE ores and determined REE by MP-AES and the REE results obtained by MP-AES were comparable to those obtained by well-established ICP-MS thus demonstrating the capability of MP-AES for the analysis of REE in geological materials. A comparative study of the direct determination of REE in water samples by ICP-OES and MP-AES showed that MP-AES is an attractive alternative to ICP-OES because of several advantages like favourable detection limits for REE, and low running cost [65].

2.7. Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES)

The ICP-OES is another very important and very popular multi-element analytical technique for the determination of not only REE but also several other elements in geological materials. When a sample solution is aspirated through the sample introduction system, the liquid sample is converted to an aerosol and transported to the high-temperature ICP. In the plasma, the sample undergoes desolvation, vaporization, atomization, and ionization, and the electrons within the atoms, and ions absorb energy from the plasma which causes the electrons to move from one energy level to another. When the electrons fall back to the ground state, the atoms/ions emit light of wavelengths which is specific to each element. By comparison of the measured emission intensities to the intensities of known concentrations, the respective elemental concentrations in an unknown sample are obtained [66,67]. The technique has a wide linear dynamic range with high sensitivity, and can simultaneously measure up to 60 elements in a variety of samples including geological materials [68,69]. Makombe et al. [70] determined REE in sediment sample solutions dissolved by both four-acid digestion (HCl, HF, HNO_3, and HClO₄), and lithium metaborate fusion digestion by ICP-OES and compared well with those obtained by ICP-MS. The results were also found to be within the confidence limits of the results for reference material. The detection limits obtainable by ICP-OES presented in Table 5 are in µg/ml range. Low sensitivity, and spectral, and non-spectral interferences of ICP-OES make it difficult to detect and determine low concentrations of REE in most rocks. However, separation and preconcentration methods can reduce detection limits as well as help in removing most interferences [71]. The complex REE spectra have long been a challenge for analysis by ICP-OES because of the hundreds of possible orbital transitions and corresponding emission wavelengths. Spectral overlaps and near-overlaps, especially in complex geological matrices, have made trace-level REE analysis nearly impossible. But the use of alternate wavelengths can help to avoid potential interferences. In addition, interference removal techniques such as separation methods such as solvent extraction, ion exchange, and precipitation methods allow the user to remove a few elements which cause spectral interferences in the rock matrices. Pradhan and Ambade [72] developed a solvent extraction method for the separation and preconcentration of REE in geological materials such as rock, soils, grabs, stream sediments, beneficiation products, minerals, and core samples and determination by ICP-OES. Amaral et al. [73] determined all REE including Sc and Y in geological samples after the samples are digested using microwave digestion technique by ICP-OES using both 'radial view mode' and 'dual view mode' with minimum matrix interference effects. Dual-view mode gave better detection limits than the radial view mode.

2.8. Laser-Induced Breakdown Spectroscopy (LIBS)

LIBS is a rapidly developing technique for the chemical characterization of different materials for element, molecule, and isotope analysis in real-time, with minimum destruction to the sample in use, and is extremely popular for the geochemical studies, monitoring of production processes in an industry, and

remote analysis of materials even in a hostile environment [81,82]. Figure 6 depicts the working principle of LIBS. The technique uses an intense, highly focused laser pulse fired at a sample to be analyzed, creating a small plume of plasma consisting of electronically excited atoms and ions. The most common laser used is excimers and Nd: YAG lasers that create a very short plasma - typically 5 - 20 nanoseconds long. The emitted characteristic spectral lines are analysed to obtain qualitative and quantitative information about different elements in a sample. This technique is an in situ analytical technique that requires little to no sample preparation. It is not easy to determine REE by LIBS in geological samples due to higher detection limits (Table 5), significant spectral interferences, and relatively low concentrations encountered in the majority of the geological materials LIBS has been widely used for the detection and quantification of elements in materials irrespective of their states. Several types of samples can be directly analysed by LIBS as the techniques require no or minimal sample preparation. Analysis of liquid samples by LIBS is hampered by inherent drawbacks of the method such as splashing, surface ripples, quenching of emitted intensity, and a shorter plasma lifetime. But Alamelu et al. [83] analysed Sm, Eu, and Gd simultaneously in aqueous solutions with an accuracy of about 5%. Abedin et al. [84] identified for the first time REE elements such as Ce, La, Pr, Nd, Sm, Gd, Dy, Yb, Y, and several associated trace elements in monazite sands using LIBS. Bhatt et al. [85] used LIBS to determine La, Ce, and Nd in some geological samples which were found to be comparable to those obtained by ICP-MS analysis. LIBS technique is used for the quantitative analysis of doped La and Nd in phosphors and the analytical performance was verified by the analysis of certified reference material [86]. Matrix effects and signal uncertainty are the two major drawbacks of LIBS for quantitative analytical applications. Long et al. [87] recently proposed a data selection method for the reduction of matrix effects and signal uncertainty and to improve repeatability during LIBS analysis. LIBS was used for the detection of REE and associated elements including nonmetals like Si, and P in the beach sands of southern Bangladesh [88]. LIBS can be used for automatic coal-rock recognition with 97% accuracy when combined with an artificial neutral network in coal quality testing during unmanned coal mining operations [89]

LIBS technique is not a very highly sensitive technique for REE determination. Especially the detection of trace amounts of REE is more difficult because weak signals are very often not detectable because of the strong interference from other major, and minor elements in the rock matrix. Gaft et al. [90] detected all REE using both atomic and ionic lines to enable effective detection. Simultaneous elemental and molecular LIBS combined with plasma-induced luminescence (PIL) is extremely sensitive for Eu, Sm, Gd, and Tb and can be used to determine them very accurately with substantially longer acquisition times (10 to 100 times). In LIBS analysis, spectral interference and the weaker emission lines of low levels of REE are key challenges for complex REE emission spectra. As emission intensity is dependent on laser wavelength, Afgan et al. [91] used IR (1064 nm), visible (532 nm), and UV (266 nm) irradiation and obtained higher signal intensity with even low concentrations of REE in monazite sample.

2.9. Inductively Coupled Plasma Mass Spectrometry Techniques (All Forms of ICP-MS, ICP-MS/MS, ICP-TOF-MS, HR-ICP-MS, MH-ICP-MS & MC-ICP-MS with Both Solution and Direct Solid Sampling by Laser Ablation)

Currently, ICP-MS occupies an invaluable position in the modern geochemical laboratory due to its multi-element capability, simplicity, limited interferences, excellent sensitivity, precision, and accuracy not only for REE but also for several other groups of elements in geological samples. Several instrumental advancements have been incorporated during the last four decades into different forms of ICP-MS instruments for overcoming interferences and improving accuracy for elemental as well as isotopic determinations. The following sections present brief accounts of different types of ICP-MS instruments like the quadrupole-ICP-MS, tandem ICP-MS (ICP-MS/MS), time of flight ICP-MS (ICP-TOF-MS), high-resolution ICP-MS (HR-ICP-MS), simultaneous -ICP-MS or Mattauch–Herzog geometry (MH-ICP-MS), and multi-collector ICP-MS (MC-ICP-MS), and their applications in the elemental and isotopic analysis of REE in different types of geological materials.

2.9.1. Inductively Coupled Mass Spectrometry (ICP-MS)

The successful linking of ICP to a quadrupole mass spectrometer, more than four decades back presented a most valuable addition to the range of analytical techniques already available for elemental as well as isotopic analysis [92]. When the sample in the form of a solution as an aerosol after passing through a nebulizer and a spray chamber/desolvation is introduced into ICP, where it is heated to a temperature of approximately 9000 K resulting in a series of processes involving desolvation, vaporization, dissociation, atomization and ionization within the ICP. Thereafter, the singly-charged positive ions enter the quadrupole mass spectrometer through a differentially pumped interface. At any given time, depending upon the applied RF/DC potentials, the quadrupole mass spectrometer transmits only ions of a particular mass-to-charge ratio. Any neutral species accompanying the plasma gases would be pumped away by the vacuum pumps. After mass isolation, the ions pass into the detector system usually consisting of an electron multiplier which is used in pulse counting mode to register and generate a signal. In addition to the solution nebulization, different other sample introduction systems such as laser ablation, and different chromatography techniques can be coupled to the instrument allowing direct analysis of solids and speciation analysis capability respectively. Before ICP-MS was invented, other analytical techniques such as F-AAS, and ICP-OES required REE to be separated from the rock matrix for their determination to avoid interferences from matrix elements, by using methods such as coprecipitation, solvent extraction, or solid phase extraction. On the other hand, for REE analysis by ICP-MS, the samples can be directly nebulised without using any separation or preconcentration procedures. Over the last 40 years, ICP-MS has established itself as a powerful analytical tool that is being used for the analysis of more diverse applications including geological materials. Even INAA methods are also very slow when compared with those of ICP-MS for the determination of REE in geological samples, because of the sample irradiation, and cooling requirements. The detection limits offered by quadrupole ICP-MS are in the ng/ml range (Table 5). Despite the fact that ICP-MS has very good performance characteristics for the determination of REE, interferences by various spectral and matrix interferences hamper the accurate determination of REE in geological materials. However, there are ways to minimise these interferences by using methods such as internal standardisation, use of matrix-matching calibrations, removing interfering matrix elements by methods such as ion exchange, and solvent extraction methods, and the use of collision cell/reaction chemistry, in addition to the use of other standard methods such as standard addition, isotope dilution, and by the use of integrated sample introduction and aerosol dilution systems [30,93].

Exploration samples very often contain highly refractory phases and such samples are to be effectively digested by high-temperature alkaline flux fusion, and REE can be determined by ICP-MS which offers low detection limits, no sample pre-concentration, and is very rapid [94]. Beach placer deposit soils from Odisha, India were analysed for REE by ICP-MS [95], and Lin et al. [96] used ICP-MS to determine REE in both organic and inorganic phases of the coal sample from the central Appalachian basin coal region, USA. The results showed that 25% of the REE concentration is associated with organic matter. Nguyen et al. [97] developed a simple and rapid HPLC-ICP-MS method for the determination of ultra-trace REE impurities in high pure Eu₂O₃ and Yb₂O₃ due to the formation of ¹⁵³Eu¹⁶O⁺ and ¹⁷⁴YbH⁺ which directly overlap with the monoisotopic ¹⁶⁹Tm⁺ and the most abundant isotope of Lu (¹⁷⁵Lu⁺), respectively. The method is rapid and takes only 3.5 minutes, and is suitable for routine analysis of trace levels of all REE impurities, including Tm and Lu in high-purity europium and ytterbium oxides. Several studies reported the use of REE determination in a variety of rock and ore samples such as Ni laterites by ICP-MS [74]. Barium is a major element in several types of geological samples, and spectroscopic interferences caused by Ba and its oxide, and hydroxide compounds in barite-rich samples (with BaSO₄ ranging from 18.87 wt% to 42.41 wt%) hamper the determination of REE in geological samples. Sometimes it becomes necessary to use separation methods such as ion exchange chromatography for REE purification by eliminating Ba. Liu et al. [98] developed an ion exchange separation method to eliminate Ba for the precise measurement of REE contents in Ba-rich samples by ICP-MS. Determination of REE in natural waters, including surface and groundwaters as well as rainwater, and ice is not a smooth afire because of insufficient sensitivity of the technique, spectral and matrix interferences, potential contamination, and lack of water CRMs. Wysocka [99] in a review on this subject, discussed these issues comprehensively and offered solutions for obtaining the most reliable REE results in water samples. With analytical strategies such as mathematical corrections, multi-element calibrations, and internal standardisation, accurate REE results can be obtained in different types of water samples with precisions in the range of 5–15% (RSD). In studies related to the investigation of the economic aspects of REE in coal-by products, ICP-MS was used for the determination of REE [100]. REE and several other trace elements including U and Th were determined by ICP-MS to understand the nature of the parent sediments from which these rocks were derived and to create an elemental fingerprint in phosphate rocks from the El-Sibayia and El-Hamrawein mines in Egypt [101]. ICP-MS analysis of fly ash and bottom ash generated at power stations the Collie Basin in Western Australia revealed very high concentrations of REE and demonstrated that Australian coal ash may represent a promising resource of REE [103]. REE and several other trace elements in lake bottom sediments, surface water samples in the vicinity of mines, and processing area of rare metal ores in the Lovozero tundra, Kola Subarctic, Russia, were determined by ICP-MS to understand the anthropogenic inputs [104]. Wencai et al. [105] used ICP-MS to determine REE in marine sediments by using ¹¹⁵In and ¹⁸⁵Re as internal standards to overcome matrix and signal drift effects. It is very difficult to accurately measure REE in seawater by ICP-MS as their concentrations are very low and the high salt content (~3.5%) makes it difficult because of a number of spectroscopic and matrix interferences, signal drift, and frequent clogging of the sample introduction system. Li et al. [106] coupled ICP-MS with a fully automatic separation system (ELSPE-2 Precon system) which eliminates the high-salt matrix within the seawater and enables the measuring of REE in seawater very accurately. Using a similar method, Wysocka et al. [107] determined REE concentrations in natural mineral waters from wells or mineral water bottles from the market with high elements recoveries (98% ± 4%). Li et al. [108] used crab shall particles for the separation and preconcentration of REE from seawater and determined by ICP-MS with extremely low detection limits ranging from 0.6 to 8.8 pg/ml. The concentrations of REE in seawater were found to be in the concentration range from 0.025 ng/ml to 0.172 ng/ml. The recovery of REE were between 95.3 and 104.4%, proving the efficiency of this proposed extraction process.

2.9.2. ICP-Tandem Mass Spectrometry (ICP-MS/MS)

Due to very close physical, chemical, and spectral properties, and associated inter-elemental spectral interferences, the determination of REE is not very easy by a classical quadrupole ICP-MS instrument though highly efficient high-temperature ICP provides nearly 100% ionization efficiency for all REE. Figure 7 presents a schematic diagram of a) a typical ICP-MS/MS instrument. For example, spectral interferences from polyatomic oxide and hydroxide species such as ¹³⁷Ba¹⁶O⁺ and ¹³⁶Ba¹⁶O¹H⁺on ¹⁵³Eu⁺, and the interference of ¹³⁹La¹⁶O⁺ on ¹⁵⁵Gd⁺are a couple of the many such examples. The spectral interferences problem can be minimized by i) procedural blank subtraction, ii) temperature-controlled sample introduction system, iii) mathematical interference correction, iv) off-line and online sample pre-treatment methods, v) internal standardization, vi) standard-addition method, vii) isotope-dilution method, viii) use of mixed plasmas, ix) matrix-matching calibration, x) use of cool plasma technology, xi) use of high-resolution ICP-MS, and xii) the use of collision/reaction technology [109]. However, recent studies by several authors demonstrated that the use of tandem ICP-MS or ICP-MS/MS with a collision/reaction cell (operated with gases like NH₃, O_2, and N₂O) sandwiched by two quadrupole analysers, and measuring an REE either on-mass mode or mass-shift mode provided the best results with detection limits in sub-pg/ml levels (Table 5) for different REE [102,110]. Some studies indicated that NH₃ as a cell gas was found to react with many of the polyatomic ions that interfere with the REE and are useful in eliminating/minimizing interferences. However, NH₃ also reacts with some of the REE leading to reduced sensitivity. On the other hand, Zhu [110] found that N₂O as a reaction gas gave more accurate values compared to the use of O₂ gas when REE were determined by ICP-MS/MS when both N₂O and O₂ were used as reaction gases for the removal of interferences during the determination of the whole set of REE in river water reference material. Especially the use of N₂O as the reaction gas helped to suppress Ba-related spectral interferences, and also yielded lower detection limits. Santoro et al. [111] used Na₂O₂ fusion coupled to ICP-MS/MS to rapidly screen quartz-rich geological samples for REE contents. During the analysis the most abundant isotopes were selected, and four different acquisition modes: no cell gas, He or H_2, and indirect mass-shift mode using O₂ gas were used. The suitability and accuracy of the method were checked by analysing the geological CRM, QLO-1 (USGS), and also by comparing with the results obtained by INAA (Table 6).

Lancaster et al. [112] made investigations on the use of N₂O as a reaction gas for the determination of 73 elements including REE. However, the authors did not conduct any application studies on geological materials. The Na₂O₂ fusion/ICP-MS/MS method was found to be useful for the rapid screening of samples from quartz-rich geological areas for REE content during exploration studies. The tandem ICP-MS technique is very powerful and competes with the detection limits obtainable by even HR-ICP-MS (Table 5). Zhu [75] accurately determined REE concentrations in seawater by ICP-MS/MS after adopting the coprecipitation with magnesium hydroxide separation and preconcentration method. Mass-shift mode was used for their measurement by permitting each isotope to pass through the first quadrupole and make it to react with oxygen to form a monoxide of the metal which was permitted to pass through and then measured by the detector. The Gakara REE deposit, Burundi in Africa is one of the world’s highest grades REE deposits the formation of this deposit is linked to a carbonatitic magmatic-hydrothermal activity. U-Th-Pb geochronology of monazite and bastnaesite grains from the Gakara deposit was studied by in-situ LA-ICP-MS/MS using an ESI NWR193UC Excimer laser directly in thin sections. The ages obtained on bastnaesite and monazite were 602 ± 7 Ma and 589 ± 8 Ma respectively demonstrating that bastnaesite crystallization was rapidly followed by monazite alteration, and reinforcing [113]. This work demonstrates the usefulness of ICP- MS/MS with LA sampling even for in situ dating studies.

2.9.3. ICP-TOF-MS

The development of ICP time-of-flight MS (ICP-TOF-MS) brought yet another dimension to the REE determinations in geological materials [114,115,116]. ICP-TOF-MS offers extremely high data-acquisition speeds (~30,000 scans/sec), high ion transmission, and quasi-simultaneous measurement of all masses in a packet extracted from the ion source leading to better detection limits than those offered by particularly quadrupole ICP-MS (Table 5). Dick et al. [117] determined REE in Antarctica ice using ICP-TOF-MS with recovery rates of ~103%, and precisions of ~3.4% RSD. This study demonstrated that the ICP-TOF-MS technique meets the demands of restricted sample mass, and the results are in very good agreement with those obtained by ICP-MS as well as HR-ICP-MS. LA-ICP-TOF-MS was used for the determination of REE in zircon crystals collected from Himalayan orogen. The technique was also used to determine U-Pb ages at sampling depths in the range of 0.59–0.66 μm. [118]. Peng et al. [119] used high-resolution mapping by LA-ICP-TOF-MS for the visualization of critical metals including REE in manganese nodules in deep oceans for understanding redox conditions with the help of Ce concentrations during their growth process, with advantages such as time-saving as a result of high-speed stage movement, and high lateral resolution at the micro-scale. The seemingly-simultaneous detection capability of LA-TOF-MS allows rapid multi-element (including REE) analysis of very fast transient signals (i.e., laser ablation imaging in low dispersion laser ablation cells) and is also ideally suited to 2D and 3D imaging of geological materials and will likely replace LA-ICP-MS for such applications in future [120].

2.9.4. Magnetic Sector or High Resolution -ICP-MS (HR-ICP-MS)

The magnetic sector field or high-resolution inductively coupled plasma mass spectrometer (HR-ICP-MS) combines an ICP source with a double-focusing magnetic analyzer to perform trace/ultra-trace metal analysis and/or isotope ratio measurements [121]. The HR-ICP-MS has several advantages such as higher transmission combined with very low background levels over conventional quadrupole ICP-MS. As a result, the detection limits are much superior over not only ICP-MS but also other popular analytical techniques (Table 4). Several common oxide, and hydroxide interferences can be eliminated. The instrument is normally operated in three resolution settings: low (300 R), medium (3000R), and high (1000 R) for the elimination of different types of interferences.

Since this is a very highly sensitive technique, the laboratory environment has to be extremely clean for obtaining optimum performance. In fact, this technique has gained confidence among the geological community due to its capability to clearly resolve several spectroscopic interferences and offer extremely high sensitivity. Charles et al. [123] noticed very little use of the high resolving power of the technique during the determination of REE in manganese nodule-certified reference materials (Table 7). Balaram et al [124] designed and developed a procedure for the determination of REE and Y in seawater samples by HR-ICP-MS after preconcentration using bis-2-ehtylhexyl phosphoric acid (HDEHP) complexing agent. Gao et al. [125] used HR-ICP-MS for the determination of REE contents in the coals (∑REE 15.17 μg/g) and silty mudstones (∑REE 210.57 μg/g) of Ordos Basin, North China, and found that REE are mainly derived from felsic and intermediate rocks in a continental island arc setting. Table 8 presents REE data (ng/ml) determined by HR-ICP-MS in different water samples drawn from a drill hole, a spring, a couple of lakes, and some rivers, from the Ladakh region in India. An analytical procedure was devised to determine REE impurities in praseodymium oxide (Pr₆O₁₁) after the REE impurities were separated by using HPLC and determining by HR-ICP-MS with recoveries ranging from 85 to 100% and with measurement precisions varying between 3 and 5% RSD [126]. 2001). Analysis of REE in seawater is difficult even when a highly sensitive technique like HR-ICP-MS is used because of their extremely low concentration levels and high total dissolved salts (TDS).

Soto-Jiménez et al. [127] used a 4-step method consisting of i) filtration and acidification (pH<2) of the sample; ii) matrix separation by a separation and preconcentration system (seaFAST-SP3™ system), iii) off-line injection of the eluted sample; and 4) analysis by HR-ICP-MS. Estuarine water CRM, SLEW-3 was used for validation of the results. Table 9 presents the analytical results of REE in water CRM, SLRS-4 obtained by different well-established ICP-MS-based analytical techniques including HR-ICP-MS, and ID-HR-ICP-MS from different laboratories, in comparison with certified values.

2.9.5. MH-ICP-MS

An ICP-MS instrument that fits into a compact Mattauch–Herzog geometry (MH-ICP-MS) with a permanent magnet and a large spatially resolving semiconductor ion detector which has the capability to detect a number of elements/isotopes simultaneously, has been introduced. MH-ICP-MS instrument will have a single 4800-pixel element detector permitting the simultaneous detection of isotopes over the full relevant inorganic mass spectrum from ⁶Li to ²³⁸U including REE in aqueous samples using only sample amounts as little as 1-4 ml [133]. REE composition in 66 freshwater samples determined by MH-ICP-MS, from various rivers and lakes in Upolu Island, Samoa, South Pacific Ocean are presented in Table 10. The detection limits obtainable by MH-ICP-MS for REE are presented in Table 5. It can be seen that the detection limits offered by different ICP-MS techniques varied significantly and they are also element and technique dependent. Obviously, HR-ICP-MS and ICP-MS/MS techniques offer the lowest detection limits.

3. Isotopic Studies

The study of isotopic abundances (radiogenic as well as stable) and their distribution provides clues for understanding various geological processes that these elements/isotopes underwent during the crustal evolution. Particularly, radiogenic isotope signatures are used to understand the long-term evolution and the origin of sources of volcanic rocks and the evolution of the Earth. Stable isotopes have a lot of applications in exploration, water and soil management, and in tracer studies. On the other hand, radiogenic isotope pairs like ¹⁴⁷Sm-¹⁴³Nd, and ¹⁷⁶Lu-¹⁷⁶Hf are used for dating applications. to determine the ages of rocks and minerals, but also to evaluate the nature and evolution of their source regions. REE isotopes in geological materials are applied very extensively in geological sciences to obtain new insights into several natural processes, ranging from crustal formation to chemical weathering and ocean circulation. Different mass spectrometric techniques such as TIMS, MC-ICP-MS, SIMS, and SHRIMP are used for the precise determination of isotopic abundances and isotopic ratios both in liquid as well as solid materials. The following sections deal with a little more about each of these techniques for different REE isotopic applications.

3.1. Multi-Collector ICP-MS (MC-ICP-MS)

A basic MC-ICP-MS instrument contains an ICP source, an energy filter, a double-focusing mass spectrometer of Neir-Johnson geometry, and an array of detectors, typically Faraday cups for the simultaneous detection of a number of isotopes of interest [135,136]. Among REE, Nd, for example, is highly incompatible and refractory and has seven isotopes. The variations in Nd isotopes can be caused by both mass-dependant and mass-independent effects which are used to understand a number of geological processes [137]. However, their precise determination is very challenging by utilizing instruments like TIMS because of severe interference effects. But the advent of MC-ICP-MS together with advanced sample preparation methods and separation methods such as the use of a two-column ion-exchange separation procedure allowed the high precision determination of isotope ratios of various REE [136,138]. In addition to the precise isotope ratios, MC-ICP-MS is also used for the determination of elemental concentrations. Lee and Ko [139] determined REE concentrations in a natural river water CRM, SLRS-6, very accurately and precisely after group separation by 2-hydroxyisobutyric acid (HIBA). As isotope-dilution methods are highly accurate and robust, Kent et al. [140] used isotope dilution-MC-ICP-MS (ID-MC-ICP-MS) to accurately determine eleven REE abundances in several international geological CRMs (USGS, NIST, GSJ) using rock powders and glass discs made of these rock powders, and made a comparative study of the determination of REE concentrations by both ID-MC-ICP-MS and ID-TIMS. ID methods cannot be used for monoisotopic elements like Tb as the method requires a minimum of two isotopes. Replicate analysis of both rock powder and glass materials agreed very well with analytical uncertainty is typically S 1%. Pourmand et al. [141] devised a novel extraction chromatography method after reducing the REE impurities in the fusion flux and used the MC-ICP-MS technique for rapid analysis of REE, Sc, and Y in primitive chondrites using a desolvating nebulizer and standard-sample bracketing technique. Baker et al. [142] developed a method for a high-precision analysis of REE by ID-MC-ICP-MS which is superior in terms of the analytical reproducibility or rapidity of analysis, on smaller sample amounts compared with ID-ICP-MS or with ID-TIMS. Bastnäsite is one of the important REE minerals and is the end member of a large group of carbonates–fluoride minerals which represents the major economic LREE deposits related to carbonatite and alkaline intrusions. Yang et al. [143] used LA-MC-ICP-MS for the determination of age by the U-Th-Pb geochronology method and Sr-Nd isotopic composition. Monazite and bastnasite are two principal REE ore minerals in the most famous Bayan Obo REE deposit in China. Sm-Nd chronometer and LA-MC-ICP-MS were used to understand the primary crystallization timing of the REE deposit at 1293 ± 48 Ma [144].

3.2. Thermal Ionization Mass Spectrometry (TIMS)

For isotope dilution analysis by TIMS, usually chemical separations of REE into two or more factions to separate LREE and HREE in order to reduce the inter-elemental interferences during the analysis. It requires a minimum of two isotopes for isotope dilution and it is not possible to analyse monoisotopic elements like Pr, Ho, and Tm. The new ages of 296.9 ± 1.65 Ma and 296 ± 4.2 Ma were obtained by U–Pb zircon dating by TIMS from tonstein layers interbedded with coal seams from the Candiota coalfield in the southern Paraná Basin, Brazil [145]. The Pope's Hill REE deposit, Labrador, Canada was studied by LA-ICP-MS and ID-TIMS U-Pb geochronology and in situ Sm-Nd isotopes using LA-MC-ICP-MS in monazite from the ore and host rock to understand the timing of deposit formation and determine the source of the REE [146]. Ramesh et al. [147] determined REE and heavy elements in surficial sediments of the Himalayan River system by TIMS to understand the behaviour of REE during weathering and transport in a secondary sedimentary environment. In addition, the decay systems of the radioactive REE isotopes ¹³⁸La, ¹⁴⁷Sm, and ¹⁷⁶Lu are used to establish ages of a range of geological events, starting from the first steps of planetary formation to some of the younger events like Deccan volcanism, and TIMS technique was proved to be very valuable in such studies. Uranium isotope (²³⁴U/²³⁸U, ²³⁵U/²³⁸U) ratio data generated by TIMS was used to understand the occurrence and provenance of Kanyakumari beach placer deposits of REE in India [148].

3.3. Sensitive High-Resolution Ion Micro Probe (SHRIMP)

SHRIMP is a high-resolution and highly sensitive instrument developed by William Compston at the Australian National University primarily used for geological applications, especially in-situ U–Pb dating of the mineral apatite, as well as zircon [149]. SHRIMP dating methods have been extensively applied for the dating of a number of REE deposits the world over. REE deposits such as carbonatite deposits can be dated very precisely using the decay systems of the radioactive REE isotopes ¹³⁸La, ¹⁴⁷Sm, and ¹⁷⁶Lu not only for understanding the age of REE deposits but also a range of events, starting from the first steps of planetary formation to younger steps of geodynamic development [150]. Thus, the abundance of all REE occurring in a large range of concentrations as well as precise isotope ratios must be analysed in different geomaterials for such studies. Campbell et al. [151] dated zircon from the Bayan Obo Fe–Nb–REE deposit to understand carbonatite-related magmatism and REE mineralization events. These SHRIMP dating studies indicated that Bayan Obo zircon cores represent part of a magmatic, intrusive, carbonatitic protolith suite that crystallized at 1,325 ± 60 Ma. Bhunia et al. [152] used SHRIMP to date the carbonatite-hosted REE deposit of Kamthai, N W India. This is one of the well-studied carbonatite REE deposits in India which gave an age of 68.4 ± 1.8 Ma which is linked to the carbonatite emplacement to the Deccan Large Igneous Province which triggered the mass extinctions near the K-Pg boundary. SHRIMP is capable of producing high transmission at high-resolution data of REE abundances in minerals like zircon and apatite [153]. SHRIMP-RG was used to determine REE to understand their distribution in coal, as the knowledge of the REE distribution is essential to design cost-effective REE extraction procedures from coal and its by-products (Kolker et al. [154]. Carbonatite–alkaline complexes in N W Pakistan were dated using SHRIMP U-Pb dating technique of zircons Koga syenites which are spatially associated with the carbonatite deposit which gave mean ages of 283.6 ± 1.7(1σ) providing a useful timing alkaline magmatism in that area [155]. Such in-depth studies on well-proven REE deposits in general shed light on the full origin of these world-class mineral deposits.

4. Mineralogical Studies and In-Situ Analytical Techniques

Mineralogical characteristics of different REE ores and resources are extremely important as the leaching process depends on their mineralogical composition and morphological distribution to determine the best leaching/extracting process during the extraction of REE from different ores. Mineral characterization techniques such as XRD, scanning electron microscopy - energy dispersive spectrometry (SEM-EDS), and laser ablation -ICP-MS (LA-ICP-MS) are popular for the qualitative and quantitative identification of different REE minerals. During exploration studies pXRD and Raman spectrometer are used to understand the mineralogy of a geological formation in the field itself for planning future courses of action. Some of the important analytical tools for understanding the mineralogy are discussed below.

4.1. X-Ray Diffractometry (XRD)

Identification of different indicator minerals and their concentrations in rocks, ores, and soils is extremely important in mineral exploration studies. XRD is one of the most powerful methods used to identify and quantify minerals in different earth materials [156]. XRD was effectively used in the identification of different minerals in the clay fractions of REE-enriched weathered anorthosite complex in Hadong district, South Korea [157]. In order to understand the differences between the mineralogy and mineral chemistry of REE in phosphate and carbonates in the Bahoruco karsr bauxites, Dominican Republic, Villanova-de-Benavent et al. [158] utilised XRD, SEM-EDS, and EPMA. This study revealed that phosphates are mostly enriched in Y and HREE (Gd, Dy), on the other hand, carbonates display a wide range of compositions in terms of LREE (La, Ce, Nd, SM) and HREE (Gd).

4.2. Electron Probe Micro Analyser (EPMA)

EPMA can be used for the detection and determination of high concentrations of REE in rocks and also to understand the REE distribution patterns in different minerals [159]. It is a non-destructive technique and uses EDS to accurately determine the chemical composition of small amounts of solid materials by comparing it with the standards of similar and known composition. EPMA and other analytical techniques such as ICP-MS and scanning electron microscopy (SEM) were used to study the occurrence and distribution of REE in a fly ash sample from the Qianxi coal-fired power plant in Guizhou province in China. Discrete REE mineral particles distributed throughout were seen directly by SEM and EPMA and these studies revealed that the ∑REE concentration in coal fly ash was 630.51 µg/g and wet-grinding followed by acid-leaching was found to liberate more REE during the extraction process [160]. Many times, such multi-disciplinary studies would help in devising economically viable and environment-friendly extraction procedures for REE from coal fly ash which are found to be promising alternative sources for REE [161]. Sano et al. [162] determined all REE in oceanic basalt glasses by EPMA. The concentrations measured by EPMA compared favourably with those obtained by ICP-MS. The same authors also developed a method to determine REE in glass samples by SHRIMP with a detection limit of ~ 6 ng/g.

4.3. Ion-Microprobe (SIMS)

Secondary ion mass spectrometry (SIMS) or ion-microprobe is one of the best techniques for the analysis of REE in geological materials [162]. The principle of SIMS analysis involves the use of a primary ion beam (consisting of O^- or Cs⁺) to strike the sample surface of a polished rock or a thin section, and the sputtering process produces secondary ions which are extracted into a mass spectrometer and analysed. This technique provides a unique combination of extremely high sensitivity for all elements from hydrogen to uranium including all REE with a detection limit down to ng/g level both with extremely high spatial resolution. High resolution-SIMS (HR-SIMS) with large-diameter, and double-focusing SIMS offer > 5000R, and can measure the isotopic and elemental abundances in minerals at a 10 to 30 μm-diameter scale. and with a depth resolution of 1–5 μm. This instrument can also be used as ion microscope for providing elemental distribution imaging maps. Zinner and Crozaz [163], used ion microprobe for the quantitative measurement of REE in phosphates. REE working curves were found to be linear over a wide range. The detection limits were found to be > 50 ng/g for LREE and ~ 200 ng/g for HREE. It is possible to determine REE concentrations in individual mineral grains such as zircon and monazite using the SIMS technique.

Shi et al [164] developed a method to analyse all REE in silicate glasses and zircon minerals using a high lateral resolution SIMS (Nano SIMS) with a high mass-resolving power of 9,400R. Table 11 presents REE concentrations in AS3 zircon together with reference values. Ling et al. [165] used high-resolution SIMS (HR-SIMS) to date bastnaesites to understand the mineralization time of the Himalayan Mianning–Dechang REE deposits in China. Bastnaesite is a major economic REE mineral which is normally considered to be a promising geochronological tool for determining mineralization time by U–Pb dating method. For example, the SIMS bastnaesite Th–Pb dating of the Dalucao deposit yielded ages of 11.9 ±0.2 Ma and 11.8 ± 0.2 Ma. Sahijpal et al. [166] devised analytical procedures for the measurement of REE in two primitive carbonaceous chondrite samples using an ion probe with uncertainties of ~ 10 to 15%. In these studies, REE abundances measured provided insights to understand even the microscopic details of the early solar system.

4.4. Scanning Electron Microprobe (SEM-EDS)

SEM-EDS combines the capabilities of a scanning electron microscope and an ED-XRF for material characterization and proved to be very useful for mineral characterization [167]. In a study related to the leaching recovery of REE from coal fly ash, Pan et al [168] used SEM-EDS and XRD techniques to prove that monazite, apatite, scheelite, and aluminosilicate were the major phases to be REE carriers. Such studies help in devising cost-effective procedures for the leaching and recovery of REE from different ores. Jason et al. [103] used SEM-EDS and Integrated Mineral Analysis (TIMA) to study fine <5μm monazite grains in the coal fly ash which were typically bound to Al/Si-rich phases. These studies helped to devise more efficient extraction procedures using aqua regia. In a study to find out the potential of the carbonatite tailings and to identify the major mineral phases for the REE recovery, by using XRD and SEM-EDS, Sarker et al. [26] found that monazite and florencite were the main REE-bearing minerals. The SEM-EDS-based liberation analysis indicated that REE-minerals were found to be primarily associated with goethite and were locked within the larger particle sizes over 100 µm and but when the particle size was reduced to 50 µm, most REE mineral grains were liberated. These findings suggested that grinding the ore material to 63 µm would potentially liberate the most REE-minerals for the subsequent separation processes such as gravity, magnetic, and floatation [26]. Li et al. [169] detected anomalously high concentrations of REE in acid mine drainage of a coal mine in northern Guizhou, China. SEM-EDS analysis revealed two types of REE-bearing minerals. Bastnaesite and monazite, were detected in the claystone samples and they were found to be adsorbed by a large amount of clay minerals, mainly kaolinite. Careful documentation of mineralogical and geochemical variations of REE deposits is extremely important for a clear understanding of difficulties in mineral processing and to arrive at an optimal process for mineralogical beneficiation. SEM-EDS and XRD were effectively utilised to understand the mineralogy of Bear Lodge REE deposit (Wyoming, US) and classify them into four types: i) fluor-carbonates (bastnaesite, parisite, synchysite), ii) phosphates (monazite, xenotime, florencite, rhabdophane, churchite), iii) cerianite, and iv) ancylite, and the distribution of REE was found to be very heterogeneous throughout the deposit [170].

5. Laser Ablation-ICP-MS (LA-ICP-MS)

The determination of elemental concentrations and isotope ratios directly in solid samples has been an attractive frontier in the applications of these in situ techniques in earth science studies. Gray [171] was the first to demonstrate the feasibility of analysing direct solids by ICP-MS using laser ablation sampling (LA-ICP-MS). Compared with solution nebulization-ICP-MS for the bulk analysis of geological samples, LA-ICP-MS analysis has several advantages such as very low background, lower oxide and hydroxide interference levels, a simpler sample preparation procedure, faster analyses, and cost-effectiveness for the determination of REE concentrations and isotope compositions. However, they are affected by interferences caused by polyatomic, oxide, and hydroxide ion species formed in the plasma. In addition, the fractionation effects leading to non-stoichiometric behaviour also affect isotope ratio measurements. LA-ICP-MS is one of the most powerful microbeam techniques for the direct analysis of minerals, and mineral inclusions generally using a spot size of 20-50 µm for the determination of major, minor, and trace elements. Recently LA-ICP-MS was utilised for the direct analysis of different elements including REE in micrometer-scale ilmenite lamellae in titanomagnetite using LA-ICP-MS with high precision (>10% RSD) and accuracy [172]. Guo et al. [173] determined REE concentrations in scheelite minerals from scheelite-quartz veins in the Honghuaerji scheelite deposit, Inner Mongolia, Northeast China, using LA-ICP-MS which showed that the scheelite samples contain elevated REE concentrations with ΣREE + Y contents in the range of 3339 to 6321 µg/g. This information was useful in understanding how these minerals got crystallized from relatively reduced hydrothermal fluids during their formation. The application of femtosecond lasers instead of the more common nano-second lasers in LA-ICP-MS systems reduces laser-induced fractionation, and matrix effects [136]. Mohanty et al. [174] analysed detrital zircon grains from the river banks and beach placers of coastal Odisha, India, using LA-ICP-MS and found that Hf (mean = 11270 µg/g) and Y (mean = 1064 µg/g) were the two most abundant trace elements found within zircon grains as compared to other trace elements. Such studies are useful for understanding the resource potential and delineating the resource regions. Jiu et al. [175] determined REE, Li, Ga, and Nb in coal from Jungar coalfield, Ordos Basin, Chinaby using in-situ LA-ICP-MS, in order to understand the distribution and modes of occurrences of these elements in coal. The determination of these elements in the strong organic makes this technique very challenging nevertheless possible. The majority of the La and Ce are found to be concentrated in La- and Ce-enriched minerals, including monazite, bastnaesite, and lanthanite in the coal. Major, minor, trace elemental analysis including REE by LA-ICP-MS was used to map the barren sandstone distal to mineralized areas, for obtaining the signature of hydrothermal REE mineralization in the Proterozoic Athabasca Basin, Canada. Quantitative evaluation of the element correlations, together with SEM-EDS studies, suggested that most of the elevated U and REE are hosted in aluminium phosphate sulfate minerals [176]. LA-ICP-MS is especially valuable for the direct analysis of materials like barite, tourmaline, and corundum minerals for chemical analysis which are difficult to dissolve completely. Oostingh [177] determined REE in barites by LA-ICP-MS. The author also developed a method of fusion of barite with Na₂CO₃ in the ratio of 1:10 and dissolving the resultant fused material in 6N HCl. Most of the Ba-ions (99%) were removed from the solution by ion exchange chromatography procedure using 2(ethylhexyl)-orthophosporic acid as binding fluid. However, both solution as well as LA sampling methods suffered from problems such as PrO interference on Gd, and Ce contamination in the fusion flux. In this context, probably the LA-ICP-MS/MS is the best method to effectively eliminate all these interferences. Liu et al. [178] summarised the applications of LA-ICP-MS for the micro-geochemistry and for bulk analysis of whole-rock and soil samples. Various related issues such as sample preparation, matrix interferences, sensitivity drift correction, calibration, and the use of matrix-matched CRMs were discussed in depth. Maruyama et al. [179] developed the LA-ICP-MS method of measuring 58 major, minor, and trace elements including all REE in volcanic glass shards using femtosecond laser and demonstrated that this technique is a viable alternative to techniques such as SIMS and EPMA for chemical characterisation of rhyolitic and basaltic glasses.

5.1. LA-ICP-MS/MS

Interestingly LA-ICP-MS/MS was successfully applied to in situ Lu–Hf dating of Paleozoic-Precambrian xenotime, apatite, and garnet. NH₃–He was used as a reaction gas mixture and matrix-matched reference materials were utilised during the study. Lu was detected by mass-shift mode. The accuracy of the single-spot ages is generally better than 1.5%, for xenotime was comparable to those obtained by in situ U-Pb analysis [180]. Ham-Meert et al. [181] demonstrated that by using CH₃F/He as the reaction gas, the spectral overlaps can be overcome in a mass-shift approach Sr isotopic analysis of medieval stained glass with elevated Rb and REE concentrations using LA-ICP-MS/MS, and this can provide a viable alternative when a non-destructive analysis is required.

5.2. Laser Ablation Split Stream (LASS) Technique

The laser ablation split stream (LASS) technique, an innovative idea was first described by Yuan et al. [182], where the ablated aerosol (a common sampling event) is split and shared by two spectrometers (MC-ICP-MS and any other ICP-MS system). This combination makes simultaneous elemental as well as isotope ratio determinations possible from a single sampling event [136]. Using this technique Qian and Zhang [183] simultaneously analysed REE on the HR-ICP-MS and Nd isotopic composition on the MC-ICP-MS in apatite standards. Kylander-Clark et al. [184] gave an account of how the laser ablation split-stream technology works using dual MC-ICP-MS and HR-ICP-MS simultaneously for high-speed, high spatial-resolution, simultaneous isotopic and elemental analysis enables petrochronology at a new level. These authors also gave a few important examples of how this novel technique is helpful in understanding the evolution of rocks and even more complex geological events.

6. Portable Miniatured Analytical Techniques

Portable devices in general offer fast, accurate, and cost-effective analysis of geological samples with little to no sample preparation, especially in the field during exploration studies and also in the actual mines while the excavation and transportation of the ore material is in progress. One of the most useful features of portable analytical techniques is the possibility to perform on-site measurements without or with minimal sample preparation. Modern geologists have a line-up of portable digital tools at their disposal such as portable XRF, portable LIBS, GPS, and laptop computers providing access to imagery, maps, and on-site real-time geochemistry.

6.1. Portable XRF (pXRF or µXRF)

Portable XRF technology can be used in a wide range of geological, mineralogical, exploration, mining, and extraction applications. pXRF is extremely useful to locate REE deposits containing < ∑REE 300 especially La, Ce, Pr, and Nd in the field during exploration studies. ∑REE 300 µg/g is the industry cut-off grade for several types of REE deposits, and earlier general observations revealed that the Y concentration typically must be present in the order of 25 - 30 µg/g to indicate an ∑REE concentration ~ 300 µg/g. Rock samples can be directly analysed in the field, or using the sample preparation accessories, representative sample units can be powdered and made a pressed pellet. In the case of rock samples, since a rock sample is a heterogeneous unit, it is necessary to take analytical measurements at multiple points for obtaining a representative result. Phosphate deposits are one of the major sources of REE [161]. Simandl et al. [185] used pXRF to determine major, minor, and trace element concentrations including REE (La, Ce, Pr, Nd, Y) in phosphate rocks from the Fernie Formation, British Columbia, Canada with acceptable precision, and accuracy. With suitable sample preparation and by following proper calibration procedures, pXRF can be used to identify zones of REE-enriched phosphate rocks in the field for large-scale exploration projects. For improving the detection of REE at lower concentrations by pXRF, an advanced 55 kV X-ray tube is available commercially which can effectively analyse all of the light REEs including La, Ce, Pr, Nd, Sm, and also a couple of heavy REE namely Eu, and Gd. Regular commercial instruments which use industry typical 50 kV, are being used currently for the determination of Y. Yang et al. [143] presented a case study from anthracite-associated clays from north-eastern Pennsylvania, US by determining only Y contents by a pXRF and proved that Y can serve as an indicator element for the estimation of ∑REE. The use of pXRF is a cost-effective way in REE exploration studies. Fajber and Simandl [186] during studies related to the evaluation of REE-enriched sedimentary phosphate deposits in south-eastern British Columbia, Canada, obtained acceptable quantitative and semi-quantitative results for Nd, Pr, Ce, La, and Y using the hand-held XRF instrument. As pXRF allows the screening of hundreds of samples quickly and reliably during exploration studies, few samples can be selected and taken to the laboratory for cross-checking the results by more reliable analytical techniques like ICP-MS. Sukadana et al. [187] used μXRF to identify REE minerals like monazite, zircon, and several other minerals in Adang Volcanic Complexes, Mamuju Area, West Sulawesi, Indonesia in order to understand the role of carbonatite magma during the hydrothermal process of mineralization.

6.2. pLIBS

For pLIBS analysis, pressed pellets of samples are usually made using a hydraulic press to remain consolidated during the ablation process. Gibaga et al. [188] determined selected REE in ion absorption clays using both pLIBS and ICP-MS and found that there is a very good agreement between the results obtained by both techniques (Table 12) which proves that pLIBS can be an alternative technique to ICP-MS for a rapid determination of REE in real-time. In fact, pLIBS can be used directly for field applications which is even more useful for a cost-effective analysis in the field. Bellie et al. [189] developed a new low-cost portable LIBS, with a gross weight of 3 kg, for instant element identification with provision for mobile access of emission spectra and online system monitoring using the cloud for use by the mining industry.

6.3. pXRD

Hydrometallurgical processes are used for better extraction yield and variable recovery rates particularly when REE ores are being processed. However, there is a limited understanding of the microscale phenomena controlling the extraction of REE, for example, from fly ash. In order to better understand these processes, fly ash sample phases present at various stages of the processes were characterized by XRD. This study reviled that the morphology and the elemental makeup of the ash matrix play a critical role in the accessibility of REE [190]. Although pXRD devices are larger and heavier than handheld pXRF devices, they can be operated in the field. The ability of pXRD to identify minerals in situ is useful for more accurate geological logging and can provide valuable insights into geological systems.

6.4. Portable Raman Spectrometer

The ability of laser Raman spectroscopy to optically identify mineralogy in situ makes it an ideal instrument not only for terrestrial field applications but also for extreme environments that prevail in places such as other planets and the deep ocean. The DORISS instrument [191] is the first step in using Raman spectroscopy in the deep ocean. Moroz et al. [192] used a micro-Raman spectrometer to detect carbonate and phosphate minerals with REE associations. These observations confirmed the presence of biogenic presence of the cyanobacteria mat and its impact on the formation of the unique Nb-REE Tomtor deposit in Russia.

6.5. Fourier Transform Infrared (FTIR) Spectrometry

The FTIR spectrometer continuously scans the surface in the NIR range and analyses the absorption and intensity. The analysis from the scans is averaged over one-minute intervals and the result is returned to the control software. This technique is very popular as an online analyser for following various industrial processes in different industries such as in bauxite/alumina, coal, and phosphate industries with several benefits such as installation entirely above the conveyor belt, and sampling is not necessary. Ye and Bai [195] used microscopy, FTIR spectroscopy, Raman microprobe spectroscopy, and UV–Vis. spectroscopy to study the spectral characteristics, mineral structure, and mineral composition, and fluid inclusion and REE geochemistry to understand the metallogenic mechanism of the Nanlishu fluorite deposit in Nanlishu, Jilin Province in China.

7. Hyperspectral Remote Sensing Techniques (Handheld, Drone, and Satellite-Based)

Hyperspectral remote sensing techniques have the potential to identify the REE minerals based on their sharp and distinctive absorption features in the visible near-infrared (VNIR) and shortwave infrared (SWIR) electromagnetic spectral profiles [156]. This is especially valuable with high spatial resolutions where the spectral response is dominated by mineralogy rather than lithology. Here, the intensity and wavelength position of some specific REE³⁺related absorption bands are sensitive to different crystal structures and not solely related to varying contents of specific REE. Nd is one of the most abundant LREE which can provide distinctive absorption features in the VNIR region of the electromagnetic spectrum even at a concentration of ~ 1000 µg/g in the Sillai Patti carbonatite complex, Pakistan [194]. Figure 8 presents a hyperspectral remote sensing imaging spectrometer system based on acousto-optic tuneable (AOTF) technology [193], and an example of the absorption feature of Nd at wavelength 810.25 nm in a carbonatite sample. In addition to satellite mounting, such systems can also be mounted on a drone apart from their use as handheld spectrometers on the ground. This modern remote sensing technique has become a novel tool for detecting and quantifying REE (Nd) in geological materials and can offer an efficient, non-destructive, and relatively less expensive tool. Boesche et al. [197] used multitemporal hyperspectral images consisting of two sensors one operating in the visible and near-infrared (350 nm–1000 nm) and the other operating in the shortwave infrared (1000 nm–2500 nm) to detect Nd-enriched surface materials for hyperspectral REE mapping of outcrops in Norway recognized for their REE-bearing igneous carbonate rocks. European Space Agency (ESA) launched Sentinel-2 which contains a multispectral instrument (MSI) that can target the REE (Nd³⁺)-bearing minerals and demonstrate the capability of MSI data. The remote sensing data generated in the visible to shortwave infrared (VNIR-SWIR) region, showed the most prominent absorption features of Nd which was a key pathfinder for REE in identifying REE-rich Esfordi phosphate deposit, Iran [198]. Booysen et al. [199] utilized data acquired by hyperspectral sensors onboard drones (lightweight unmanned aerial vehicles -UAVs) to detect REE (Nd) with low detection limits (< 200 µg/g for Nd) in carbonatite complexes of Marinkas Quellen, Namibia, and Siilinjärvi, Finland, and introduced a rapid technology for exploration of REE deposits. This kind of study promotes the use of cheaper technology of deploying drones for hyperspectral imaging in exploration studies. The advantage is these techniques can cover larger areas in a limited time and the accuracy of results can be verified by laboratory-based techniques like, XRD, and ICP-MS. Brazil has some of the largest uranium-phosphate deposits in which very high concentrations of U and REE have been reported besides phosphorous, which allows for the co-mining of these three ores. In these areas near-infrared (NIR) spectroscopy was used to investigate REE in the soil of the uranium-phosphate deposit of Itataia, Brazil. The concentrations of REE were later verified using ICP-OES. These results show the promise of NIR spectroscopy as a tool for mapping the concentrations of REE in topsoil [200]. Wang et al. [196] investigated the potential of Vis-NIR reflectance spectroscopy for the simple and rapid method of determining REE (La, Ce, Pr, Nd, Sm, and Eu) in soil (Figure 9). Turner et al. [201] studied the REE-bearing silicate minerals using reflectance spectroscopy in the VSWIR regions (500 to 2500 nm). Some specific REE-related absorptions, such as an Er³⁺ and Yb³⁺related absorption near 978 nm and Nd³⁺-related absorptions near 746, 803, and 875 nm, were used to identify different REE. Later on, these results were verified by SEM and EPMA techniques.

8. Electrochemical and Biosensors for the Detection of REE

A terbium-specific biosensor was used to quantify 3 ng/ml of Tb directly from acid mine drainage at pH 3.2 in the presence of a 100-fold excess of other REE and a 100 000-fold excess of other metals. The presence of Tb in acid mine drainage would indicate higher concentrations of other important and expensive REE like Sm and Dy. Successful application of this kind of biosensors leads to the development of rapid and inexpensive tools for selective sensing of individual REE [202]. Gold electrode modified with 2-pyridinol-1-oxide was used as a novel electrochemical sensor for the rapid detection of Eu in different environmental samples including river water [203]. This method requires no sample pre-treatment.

9. Miscellaneous Analytical Techniques

Fluorite (CaF₂) is one of the most common hydrothermal minerals in many deposits which has relatively high REE contents. Shehu and Bagudo [204] used XRF, FTIR and UV-Vis. spectrometer for obtaining mineralogical information on fluorite samples. REE in the fluorite samples was to be found responsible for its color. Hirose et al. [80] determined La, Pr, Nd, Gd and Tb in metallic REE matrices by glow discharge mass spectrometry (GD-MS) with ng/g detection limits and precisions <5%. Xiong, et al. [205] analyzed REE in an aqueous solution using a new analytical technique called microwave plasma torch coupled with the linear ion trap mass spectrometry (MPT-MS) with detection limits ranging from 0.04 to 0.574 ng/ml for different REE. Subsequently, Yuan et al. [206] termed this technique microwave plasma torch mass spectrometry (MPT-MS) which can offer higher sensitivity with detection limits at sub-ng/ml level and was used to determine several trace elements including REE in the water samples in local Poyang Lake, China. The authors are of the opinion that the separation of individual REE is necessary for more accurate analysis. Near-infrared spectroscopy is a valuable technique for the qualitative identification of some REE like Sc, Y, Pr, and Sm in soils and sediments. However, to confirm their presence and determine exact concentrations, standard analytical techniques like ICP-OES and ICP-MS have to be utilized. Therefore, near-infrared spectroscopy is a viable tool for rapid monitoring REE and other trace elements in soil and bed sediments in a river basin [207].

A rapid determination of major REE minerals such as bastnäsite, monazite, and xenotime enables the efficient exploration of REE mines. In this context, Imashuku [208] used cathodoluminescence (CL) imaging to qualitatively identify bastnäsite in mineral ores and determine a rough estimate of bastnäsite content rapidly for subsequent verification by more precise quantitative analysis by a technique like electron-probe micro analyzer (EPMA). The isolation of individual REE has always been a challenging task because of their similar physical and chemical properties. Zeolites such as ferrierite, faujasite, and the Linde type L are well-known for their ion exchange capabilities that would be potentially applicable for the REE separation. Duplouy [209] characterized them by XRD, ED-XRF, and attenuated total reflectance FTIR (ATR-FTIR). It was noticed that the uptake sequence followed La > Nd > Dy, which indicated the decrease of capacity with increased atomic number. Almost all the REE could be leached off from the REE-loaded zeolites at a pH 1.51 with nitric acid solution. These studies help in understanding their ion exchange behavior and devising a suitable separation process. Borst et al. [211] used synchrotron X-ray absorption spectroscopy for understanding the distribution and local bonding environment, with Y and Nd, as proxies for heavy and light REE in the regolith-hosted ion absorption clay deposits. Such knowledge is important to devise the best extraction methods for REE from ores. Obhodaš et al. [212] proposed a couple of nuclear techniques for the in-situ measurement of ¹⁷⁶Lu radioactivity, and a device that utilizes the neutron sensor for the measurement of Gd by thermal neutron capture. From Lu and Gd concentrations by both techniques, the ∑REE can be calculated as both Lu and Gd show a strong correlation to the ∑REE. These techniques in a remotely operated vessel (ROV) can be used for the in-situ measurement of these elements from a ship on the ocean floor during deep ocean exploration for REE deposits [213]. In order to understand REE binding environments and the association on various mineral phases in coal and coal-by products, it is necessary to have speciation information of REE in coal and coal-by products. Stuckman et al. [214] used synchrotron microscopy and spectroscopy to obtain such information on coal and coal by-products. The REE results from all three techniques were found to be in good agreement. Jochum et al [215] utilised isotope dilution spark source mass spectrometry (ID-SSMS) for the quantitative determination of 35 elements including REE in geological materials with detection limits ranging from 0.001 to 0.05 µg/g and with precisions and accuracies ranging ~ 5%. This technique is valuable when many elements are needed on a very small amount of sample.

10. Online analysis of La by In-Situ Gamma Spectrometry during the Industrial Extraction Process

During industrial extraction or metallurgical processes, quick feedback on the concentration(s) of a certain element(s) is important to enabling to have appropriate control of the processes and also for the successful completion of the required industrial process. However, offline analytical techniques such as XRF, ICP-OES, and ICP-MS may struggle to achieve real-time feedback. In order to meet such demands for direct analysis of La during the lanthanum extraction process, Zhao et al. [210] successfully designed a device based on in-situ and high-sensitivity detection of La (Figure 10) at 1435.8 keV through online germanium gamma spectrometry with minimum detectable concentration ranging from 4.66 to 4.99 µg/g. ¹³⁸La, is an isotope with stable abundance (0.089%), emitting characteristic gamma-ray 1435.8 keV (65.5%) during the decay process, it is practicable to determine lanthanum via detecting the characteristic gamma-ray.

11. Comparison of Different Analytical Techniques for REE Analysis

Adeti et al. [25] compared the performances of ICP-MS, INAA, tube-based XRF, and ²⁴¹Am excitation-based XRF, for the detection of Sc, La, Ce, Nd, Sm, Eu, Tb, and Lu in volcanic rock specimens from Ghana. Folkedahl et al. [216] conducted a Round-Robin interlaboratory study involving 13 laboratories on REE in geological materials to understand lab-to-lab and method-to-method variability in analysing REE in geological materials like coal, fly ash, shale, and mine waste. Different dissolution/sample preparation procedures and instrumental techniques such as INAA, ICP-OES, and ICP-MS were utilised. Although variability was relatively higher, exceeding 13% RSD, three of the five labs reported excellent performance in terms of repeatability, and reproducibility in the whole exercise.

12. Sample Preparation Methods for REE Studies (Acid, Fusion, and Microwave)

In general, REE occur in low concentrations in most rocks and hence it is essential to employ suitable sample preparation methods and choose measurement techniques that allow for reliable quantification. For example, to obtain the best results by ICP-MS techniques, effective sample preparation involving complete digestion of the sample including refractory phases and minerals, such as zircon, garnet, chromite, and tourmalines is necessary. Use of high-pure acids and other reagents, to ensure low blank values, separation of interfering element(s) before analysis, and elimination of molecular interferences (e.g., oxides and hydroxides) by the use of tandem ICP-MS, if possible, ensures reliable REE data.

12.1. Acid Dissolution Methods

Since most analytical techniques offer very low detection limits even to REE, and as a result, the sample volumes/weights taken for chemical analysis have reduced considerably compared to earlier times. Hence acid dissolution methods are more popular and reasonably effective for decomposing a large variety of geological materials [217]. Complete dissolution of refractory mineral phases can be assured by using sample preparation involving HF-HCl-HNO₃-HClO₄ acid mixtures and either a microwave digestion system or high-pressure asher, and a fusion procedure [94]. Fusion procedures by using fluxing agents like use of lithium borate normally produce homogeneous synthetic glasses which can be used by both XRF and LA-ICP-MS. Proper dilution of the sample solution, the use of proper internal standards, application of matrix matching rock and mineral-certified reference materials (CRMs), and a clean laboratory environment are prerequisites for accurate and precise REE concentration and isotope ratio determination in geological and industrial materials [109,150]. Red mud is a potential REE resource that has very high concentrations of many matrix elements such as Al and Fe which cause interferences to the determination of REE by ICP-OES. For dissolution, the red mud sample was melted with NaOH and the contents were extracted with hot water. Al and Fe were removed by the extraction with a triethanolamine solution. After complexing Ca and Mg, and other elements by EDTA disodium solution, REE hydroxide precipitate was dissolved by hydrochloric acid and determined by ICP-OES with recovery rates between 85 and 105 %. The accuracy of the REE results in bauxite obtained by ICP-OES was verified by those obtained by ICP-MS and found to be in good agreement [218].

12.2. Fusion Dissolution Methods

Acid digestions very often result in low recoveries of REE for certain types of rocks and minerals. In order to overcome this problem, Leitzke et al. [78] used alkali fluxes like lithium borate to produce homogeneous synthetic glass disks for both XRF and LA-ICP-MS analysis of several elements including REE. Schramm [219] used a ratio of 1:1 for rock to lithium tetraborate flux in a platinum crucible. The beads obtained after fusing at 1,000°C in a muffle furnace were utilised for XRF analysis. The precisions obtained for Sm, Gd, Eu, Y and Th range 0.005–1 %RSD, and for the rest of REE, the range is 0.02–25 % RSD. Lett and Paterson [220] made a comparative study of different sample preparation and an instrumental analytical method for the accurate determination of REE in geological materials. He used lithium metaborate-tetraborate pressed pellets (sample and flux in 1:5 ratio), 0.2 g sample with 0.8 g of sodium peroxide for 1 hour at 480°C sintering in a closed nickel crucible, and fusion of 0.2 g sample with lithium metaborate-lithium tetraborate flux at 980°C in a graphite crucible, followed by the dissolution of the fused bead with weak hydrofluoric and hydrochloric acids, and 0.5 g sample in Teflon test tubes with HF- HClO₄-HNO₃-HCl acids, for ICP-MS analysis. The authors found that sodium peroxide sinter-ICP-MS, lithium metaborate fusion-ICP-MS, INAA, and XRF are preferred methods for geochemical research because they produce the most accurate and precise data for REE. Dissolution of REE ores is difficult and often only partially completed with low recoveries because REE oxides and accessory minerals zircon are refractory and resistant to most of the common sample dissolution procedures. Zircon, for example, is a recurrent accessory mineral in many rocks that frequently incorporates P, Hf, REE, Th, and U into its crystal lattice [221]. Over the last three decades, many investigations used four acid dissolution procedures for the digestion of REE ores over fusion procedures as fusion procedures increase TDS which some instrumental analytical techniques cannot tolerate [217]. Tupaz et al. [74] developed a modified digestion procedure and accurately determined REE in mafic and ultramafic rock powders and validated the results using CRMs. The precisions achieved were generally below 5% RSD with minor modifications in the digestion procedure including changes in the concentrations of different acids and drying temperatures. Udayakumar et al [222] used a microwave digestion procedure for the dissolution of a Malaysian monazite sample by using both acid dissolution in a mixture of H₂SO₄, HNO₃, and HF, and a fusion procedure by lithium tetraborate, and metaborate flux followed by acid attack. Both procedures gave REE values that are in close agreement and also exhibited consistency with the XRF data. He et al. [223] dissolved uranium ore samples with NH₄F and REE were separated using an ion exchange procedure and their concentrations were determined by ICP-MS.

12.3. Microwave, Ultrasound-Assisted, High-Pressure Digestion, and Infrared Heating Methods

Kasar et al. [224] digested geological reference materials and soils by microwave, and Savillex digestion procedures to evaluate the best quality control for the analysis of the REE. Microwave digestion was found to be more effective and faster. At present, most hotplate digestions are replaced by microwave digestions and infrared heating as the former gives low recoveries many times. Microwave digestions provide complete dissolution of REE ore and mineral samples, and also faster. Most advanced microwave digestion systems currently have the capability to digest as many as 40 samples at a time thereby contributing to high productivity of the laboratory. Helmeczi et al [64] used focussed infrared radiation to expedite the acid digestions of geological materials and achieved efficient digestion in a shorter time. Zuma et al. [225] used microwave-assisted ashing and ultrasound-assisted extraction methods for the determination of REE in South African coal by ICP-OES with precisions <5%.

13. Quality Assurance and Quality Control during Analysis

Since REE have become very important currently, many numbers of studies are going on worldwide for the production of concentration anomaly maps, finding new deposits, mining, extraction of REE from different ores using different methods, and finally for the production of pure metals or metal oxides for industrial use. During all these studies generation of precise and accurate REE data is important [226]. The use of very dilute sample solutions, internal standardisation, and matrix matching calibrations in addition to other quality control protocols such as off-line data reduction algorithms ensure data quality. On several occasions, the accuracy of the REE determination by instrumental techniques such as LA-ICP-MS is hampered by the lack of matrix-matched reference materials. Zhang et al. [227] synthesized REE-doped NaY(WO₄)₂ single crystal and employed it as a candidate reference material apart from using ⁸⁹Y as the internal standard for the quantitative determination of the REE in certain geological materials. During the determination of REE (Ce, La, Nd, Pr, and Sm) by LIBS, due to significant overlapping of REE emission lines, a high pairwise correlation between REE contents in certified reference materials (CRMs) of natural origin, decreasing the multivariate model prediction capacity, are used [228]. Verplanck et al. [229] developed two water CRMs, with elevated REE levels (0.45 to 161 ng/ml) collected from a mine site. The `round-robin' analysis program was participated by seventeen international laboratories and contributed data for all REE, Y, and Sc. The most probable values were determined using robust statistical procedures. Even after one year, no change in concentration has been observed.

14. Conclusions

Analytical techniques play an important role in all aspects of REE research and production such as exploration, mining, extraction, and metallurgy. At every stage of these activities, different types of materials have to be analysed for elemental, isotopic, and mineralogical concentrations using different analytical techniques. Out of all described methods, ICP-MS/MS with solution nebulisation or laser ablation sampling is the best and more cost-effective laboratory analytical technique currently available. Even the detection limits for different elements including REE by ICP-MS/MS approach the detection limits obtainable by more expensive HR-ICP-MS. XRF is still popular for samples having high concentrations, because of its low sensitivity. Since pXRF offers reasonable sensitivity for Y and the direct correlation of Y concentration to the total REE, portable or handheld XRF instruments are valuable, especially during exploration studies. In general, currently, there are a number of field portable analytical techniques such as pXRF and pLIBS which are making REE exploration and other related activities faster and cost-effective. Hyperspectral remote sensing techniques including handheld, drone, and satellite-based techniques have become very popular in REE exploration studies worldwide because of their ability to cover larger areas in a limited time thus becoming very cost-effective. Microanalytical devices/sensors mounted in remotely operated vehicles (ROV) are being successfully utilized in detecting REE-rich deposits in the deep oceans. Finally, some nuclear techniques are also being applied to monitor and control extraction processes online in the REE industry.