Geochemical characteristics of REE-enriched weathered anorthosite complex in Hadong district, South Korea

Abstract

The mineralogy and geochemistry of the weathered intercumulate-type anorthosite complex in the Hadong district of South Korea were investigated as a prospective area for rare earth resource exploration. In this area, gabbroic anorthosite was intruded by rare-earth elements and yttrium (REE)-bearing dioritic pegmatites (av. 242 ppm), which caused REE-enriched Fe–Ti orebody accumulation (av. 857 ppm) near the dioritic pegmatites. REE-bearing minerals, such as allanite, cerite, and xenotime, were identified from the pegmatite and orebody, which had undergone various degrees of weathering. The total REE (∑REE) in kaolin minerals-bearing clay fractions increased compared to those of the whole rocks. When treated with ammonium-salt solvents of different pH values, that is, 4.6 (NH₄Cl_(aq)), 5.0 ((NH₄)₂SO_4(aq)), and 7.8 (NH₄CO_3(aq)), 81%, 73%, and 70% of the exchangeable REE of the kaolin minerals-bearing clay fractions were recovered in the dioritic pegmatites. respectively. In contrast, in kaolin minerals-deficient clay fractions separated from the Fe–Ti ores, exchangeable REE were recovered less than 10% after treatment, regardless of the ammonium-salt solvent species. Therefore, the occurrence of REE-bearing primary minerals in the parent rock, the formation of kaolin minerals-bearing clay fractions, and the extraction rate of exchangeable REE with ammonium-salt solvents seem to be crucial to securing the ion-adsorption type REE resources.

Introduction

Rare earth elements are defined as lanthanides (La–Lu), yttrium (Y), and scandium (Sc) (Moeller and Kremers, 1945; Bau, 1996). They (i.e., lanthanides + Y + Sc; hereinafter referred to as REE) have been widely used as high-tech materials for modern technologies, such as electric cars, semiconductors, and wind turbines because they are extremely stable in chemical property (Balaram, 2019).

REEs associated with granitoids are produced by primary magma differentiation and hydrothermal processes (Estrade et al., 2014). The REEs are also gradually accumulated via media, such as kaolin minerals, during weathering processes (Wu et al., 1990; Murakami and Ishihara, 2008; Mukai et al., 2020). Several ion-adsorption type REE deposits have been reported in the Cathaysia Block of Southern China (Wu et al., 1990; Ishihara et al., 2008; Li et al., 2019; Mukai et al., 2020). Most orebodies are genetically associated with kaolin minerals, which are secondarily formed by parent rocks enriched in REEs (Bao and Zhao, 2008; Li et al., 2017). In ion-adsorption type REE deposits, REE³⁺ ions can be exchanged in the negatively imbalanced state of the layer surface of kaolin minerals (Zhenxiang et al., 2019). Additionally, the relative ease of mining weathered rocks and the convenience in the recovery of ion-adsorption type REE deposits attracted considerable attention (Feng et al., 2021). Weakly bonded exchangeable REE³⁺ ions on kaolin minerals can be extracted via ammonium-salt solvent treatments (Sanematsu and Kon, 2013). For example, Moldoveanu and Papangelakis (2012) extracted approximately 80% of the total REEs in kaolin mineral-bearing clay fractions using ammonium-salt solvents. Additionally, Chi et al. (2003) reported that specific pH values during the reaction with an ammonium-salt solvent can lead to the precipitation of RE(OH)₃. Therefore, inducing a suitable pH value to release exchangeable REE³⁺ ions from clay fractions is crucial (Chi et al., 2003; Moldoveanu and Papangelakis, 2012).

The intercumulate-type anorthosite is distributed over the Hadong district in South Korea and is composed of gabbroic anorthosite, dioritic pegmatite, and Fe–Ti ores (Jeong, 1987; Lee et al., 1999; Jung et al., 2010). REE minerals, such as allanite, monazite, and apatite, have been reported in the anorthosite complex (Choi and Kwak, 2012; Lee et al., 2021). Rare earth resource explorations have been conducted targeting of the Fe–Ti ores that occur in an intercumulated-type anorthosite (Kim et al., 1992; Kim and Jeong, 1993; Lee et al., 1999; Choi and Kwak, 2012). Although the specific REE reserves were not evaluated, approximately 7,400,000 tons of the Fe–Ti ore bodies were estimated together with allanite (Kim and Jeong, 1993). Additionally, the enrichment of exchangeable REE associated with kaolin minerals-bearing clay fractions formed by the weathered intercumulate-type anorthosite complex and the recovery of exchangeable REE using ammonium-salt solvents have not been sufficiently studied. Therefore, this study aimed to investigate REE distribution in the weathered intercumulate-type anorthosite complex, consisting of gabbroic anorthosite, dioritic pegmatites, and Fe–Ti ores, and to estimate the recovery rate of exchangeable REE for the total REEs in clay fractions after treatment with ammonium-salt solvents.

Geological background

The anorthosite complex (1860–1880 Ma; Lee et al., 2018) of the Hadong district in the southeastern Yeongnam Massif intruded the hornblende and leucocratic gneiss belonging to the Precambrian Jirisan metamorphic complex (1880–1920 Ma; Lee et al., 2017), after which the anorthosite complex was intruded by Triassic diorite from the south (227.3 ± 1.9 Ma; Moon et al., 2012) (Fig. 1). Subsequently, these complexes were unconformably covered by Cretaceous sedimentary rocks, such as conglomerate and sandstone, in the southeast, that is, the Wonji Formation belonging to the Gyeongsang system (Kim et al., 1992) (Fig. 1). Jeong et al. (1989) subdivided the anorthosite complex into four types based on mafic mineral contents and textural characteristics as follows: 1) massive type, which is a relatively pure anorthosite comprised of more than 95% plagioclase and some minor mafic minerals, such as hornblende, tremolite, and chlorite; 2) layered type, which shows a rhythmical layered texture, alternating with tens of centimeters of pure anorthosite and several centimeters of mafic minerals; 3) intercumulate type, which has cracks in early crystallized idiomorphic plagioclase filled with latterly crystallized mafic minerals; and 4) foliated type, which shows a mafic mineral assemblage arranged in a consistent direction. The anorthositic rocks developed in the north-south direction are ordered from east to west as follows: massive, layered, intercumulate, and foliated type in succession (Jeong et al., 1989) (Fig. 1).

Fig. 1.

Geological map of the Hadong district and sampling point in the study area (map revised from KIGAM, 2022). Abbreviations: GM = Gyeonggi Massif, TB = Taebaeksan Basin, OMB = Okcheon Metamorphic Belt, YM = Yeongnam Massif, GB = Gyeongsang Basin.

Material and Methods

Samples were collected from the weathered intercumulate-type anorthosite complex (Fig. 2), and petrographic studies were performed through microscopic observations of thin and ore sections. The magnetic susceptibility of each whole rock powder was measured five times based on the interpolation mode using a portable SM-30 instrument with a resolution of 10^–7 SI.

Fig. 2.

Outcrop images of the sampling area (A–D). (A) Foreground of the intercumulate-type anorthosite complex. (B) Gabbroic anorthosite comprising of plagioclase and amphibole. (C) Clay-rich zone developed in dioritic pegmatite. (D) Accumulation of Fe–Ti ores along the brecciated vein of dioritic pegmatite. Abbreviations: Amp = amphibole, Qtz = quartz, Pl = plagioclase, Mt = magnetite, Ilm = ilmenite.

The clay fraction (<2 μm) was separated by settling in deionized (DI) water. The size of the clay fraction was determined using a Shimadzu SALD-2300 laser diffraction-particle size analyzer (LD-PSA). The clay fraction powder (0.1 g) was ultrasonically dispersed in DI water for 10 min before analysis. The data were obtained as an average of 20 replicates per sample.

An electron microscope was employed to determine the mineral assemblage and texture of each parent rock. For the polished sections with a carbon coating, backscattered electron (BSE) imaging and energy dispersive spectroscopy (EDS) analyses were conducted at 15 kV and 20 nA, respectively, using a JEOL JXA-8530F PLUS housed at Gyeongsang National University, Jinju.

The mineral identification of the clay fractions was conducted using X-ray diffractometer (XRD, Rigaku MiniFlex600). The samples were analyzed at 40 kV and 15 mA using Ni-filtered Cu-Kα (λ = 1.5418 Å) radiation and the step-scan method (3–40° 2θ, 0.01 step, 10 s) at the Center for Research Facilities, Kongju National University. FT-IR spectra were analyzed for the clay fractions separated from the parent rocks. Spectra were obtained for the powder sample from 3800 to 3400 cm^–1 with a resolution of 4 cm^–1, averaging 32 scans, using PerkinElmer Spectrum 100 FT-IR spectrometer at the Center for Research Facilities, Kongju National University.

The major element compositions of the whole rock samples were measured using an X-ray fluorescence spectrometer (XRF, Philips PW2404) at the Korea Basic Science Institute, Seoul. The operating conditions were 40 kV and 30 mA, and the analytical precision was within ±0.01%. For trace and rare earth elements of whole rocks and clay fractions, samples were dissolved in an acidic mixture (HNO₃:HF:HClO₄ = 4:4:1) at 150°C for 3 h and then reacted with aqua regia while adding HF for 3 h. They were analyzed using inductively coupled plasma atomic emission spectroscopy (ICP-AES; JY ULTIMA 2, Jovin Yvon, France) and inductively coupled plasma mass spectrometry (ICP-MS; Agilent Technologies, USA) at the Korea Basic Science Institute, Ochang. with analytical precision of ±5 ppm and 1 ppb, respectively.

Ion exchange leaching tests were conducted using different 1M solutions of monovalent ammonium (i.e., NH₄Cl, (NH₄)₂SO₄, and NH₄CO₃) to measure the proportion of exchangeable REE for total REE in the clay fractions. One gram of the clay fraction was added to a 50 ml centrifuge tube with 1 M monovalent ammonium solution at a 1:2 solid-to-liquid ratio, and the solvent was stirred at 60 rpm for 1 h using a shaker. Subsequently, the clay fractions were filtered and dried overnight at 70°C, and finally assayed by ICP-MS. Recovery rate was described by Feng et al. (2021), as follows:

  

$$\text{Recovery (\%) =}100-\left[\frac{T\cdot t}{F\cdot f}\times 100\right]$$[1]

where T is the mass the of the tailing, t is the tailing concentration, F is the mass of the feed, and f is the feed assay.

Results

The magnetic susceptibilities of gabbroic anorthosite (av. 1.5 × 10^–3 SI) were similar to those of the dioritic pegmatite (av. 1.1 × 10^–3 SI), whereas that of the Fe–Ti ores (av. 44.8 × 10^–3 SI) were 30–40 times higher (Table 1). The particle sizes of the clay fractions (PSC) separated from the whole rocks are listed in Table 1. All PSCs had ranges typical for clay minerals (i.e., <2 μm). The PSC of the gabbroic anorthosite and the Fe–Ti have similar ranges with mean volumes of 1.3–1.4 and 1.4–1.5, respectively; however, the PSC of the dioritic pegmatites was slightly larger, with a mean volume of 1.8–2.0.

Table 1. The magnetic susceptibility of whole rocks, and particle size of clay fractions

Samples	Weathering degrees	M.S. (10–3 SI)		PSC (μm)
		Mean V.	STDV	Mean V.	STDV
Gabbroic anorthosite
BGA1	slightly to intermediately weathered	1.4	0.03	1.4	0.26
BGA2		1.4	0.03	1.3	0.23
BGA3		1.7	0.03	1.3	0.22
Dioritic pegmatite
BDP1	intermediately to extremely weathered	1.1	0.02	2.0	0.32
BDP2		1.2	0.01	2.0	0.34
BDP3		0.4	0.01	2.0	0.40
BDP4		1.6	0.03	1.8	0.27
BDP5		1.5	0.05	1.9	0.28
Fe–Ti ores
BFT1	slightly to intermediately weathered	38.8	0.59	1.4	0.29
BFT2		48.4	1.15	1.4	0.25
BFT3		47.2	0.36	1.5	0.23

Abbreviations: M.S. = magnetic susceptibility, PSC = particle size of clay fractions, Mean V. = mean volume, STDV = Standard deviation

In the gabbroic anorthosite, plagioclase occurs together with mafic minerals such as amphibole and pyroxene (Figs. 2B, 3A–C). Chlorite formed along the crystal boundary of the amphibole (Fig. 3A–C). The dioritic pegmatites intruded into the gabbroic anorthosite in multiple directions (Fig. 2A). Along the border between the dioritic pegmatite and the gabbroic anorthosite, clays formed via plagioclase weathering (Fig. 2C). The dioritic pegmatites were composed of plagioclase, quartz, muscovite, and amphibole (Fig. 3D–F). Amphibole in the vicinity of quartz was hydrothermally altered, and goethite precipitated along the cracks (Fig. 3D). Plagioclase was also hydrothermally altered and partly subjected to sericitization (Fig. 3E, F). Fe–Ti ores formed in the gabbroic anorthosite, in which the brecciated veins of the dioritic pegmatites were developed (Fig. 2D). Magnetite and ilmenite were accompanied by quartz veins (Fig. 3G, H) or occurred as small grains coexisting with amphibole, plagioclase, and apatite (Fig. 3I).

Fig. 3.

Microphotographs from polarized microscopy of representative samples. (A–C) Gabbroic anorthosite. (D–F) Dioritic pegmatites. (G–I) Fe–Ti ores. (A, B) Amphibole and pyroxene filled along the cracks in plagioclase. (A, C) Chlorite occurs on the border with hydrothermally altered amphibole and plagioclase. (D) Goethite precipitated in the vicinity of crystal boundary of hydrothermally altered amphibole or quartz. (E, F) Sericitization developed along the boundary of plagioclase and muscovite. (G) Magnetite together with ilmenite accompanied by quartz vein. (H) Large grains of magnetite partly replaced by ilmenite along the boundary. (I) Small grains of magnetite coexisting with amphibole, apatite, quartz, and plagioclase. Abbreviations: Goe = goethite, Chl = chlorite, Mu = muscovite, Ser = sericite, Apt = apatite. Refer to Fig. 2 for others.

The EDS mapping of the REE-bearing minerals is shown in Fig. 4. In the gabbroic anorthosite, rare earth minerals were hardly observed, except for ilmenite incorporating La and Ce (Fig. 4A). In the dioritic pegmatites, La- and Ce-enriched allanite occurred in the cracks of anorthosite or andesine (Fig. 4B, C), whereas Ce-enriched cerite occurred along the albite cracks (Fig. 4D). Additionally, Er is incorporated in iron oxide near the quartz (Fig. 3, Table 2). For the Fe–Ti ores, Y-rich mineral such as xenotime occurred in the quartz cracks (Fig. 4E) and apatite (Fig. 4F, Table 2).

Fig. 4.

Back scattered electron and elemental mapping images by energy dispersive X-ray spectrometry for representative REE-bearing minerals. (A) Gabbroic anorthosite. (B-D) Dioritic pegmatites. (E, F) Fe–Ti ores. (A) Ilmenite incorporating La and Ce in actinolite. (B) La- and Ce-enriched allanite observed in the boundary of quartz, andesine, and anorthite. (C) La- and Ce-enriched allanite observed with andesine. (D) Ce-enriched cerite in the crack of albite. (E) Y-enriched xenotime observed in the crack of quartz. (F) Apatite incorporating Yb coexists with actinolite and ilmenite. Abbreviations: An: anorthite, Act = actinolite, Ads = andesine, Ab = albite, Aln = allanite, Cer = cerite, Xtm = xenotime. Refer to Figs. 2 and 3 for others.

Table 2. Chemical compositions of REE-bearing minerals from the study area

Sample no. of spots.	Allanite (n = 14)		Cerite (n = 2)		Xenotime (n = 2)		Apatite (n = 7)		Anorthite (n = 2)		Iron oxide (n = 2)
oxides (%)	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Y2O3	0.79	—	—	—	48.45	1.09	—	—	—	—	—	—
La2O3	11.02	2.12	—	—	—	—	—	—	—	—	—	—
Ce2O3	7.81	0.96	65.36	0.27	—	—	—	—	—	—	—	—
Pr2O3	—	—	—	—	—	—	—	—	—	—	—	—
Nd2O3	1.56	0.31	—	—	—	—	—	—	0.93	—	—	—
Pm2O3	—	—	—	—	—	—	—	—	1.12	0.11	—	—
Eu2O3	1.11	0.08	—	—	—	—	—	—	—	—	—	—
Gd2O3	—	—	—	—	1.17	—	—	—	—	—	—	—
Dy2O3	—	—	—	—	4.17	0.25	—	—	—	—	—	—
Er2O3	—	—	—	—	3.46	0.17	—	—	—	—	2.90	0.83
Yb2O3	—	—	—	—	2.95	0.42	2.63	0.91	—	—	—	—
SiO2	34.91	0.86	10.35	2.23	6.90	3.88	—	—	44.31	0.80	18.57	3.61
TiO2	—	—	—	—	—	—	—	—	—	—	—	—
Al2O3	20.33	1.18	7.53	1.15	—	—	—	—	35.56	0.93	12.27	0.60
FeOT	12.01	0.36	5.71	1.02	—	—	—	—	—	—	65.41	5.48
MnO	—	—	5.81	0.54	—	—	—	—	—	—	—	—
MgO	0.88	0.11	—	—	—	—	—	—	—	—	—	—
CaO	12.43	1.20	0.60	0.03	1.23	0.04	52.54	0.64	17.60	1.20	0.22	0.06
Na2O	—	—	1.38	1.00	—	—	—	—	0.95	0.57	—	—
P2O5	—	—	3.11	1.07	30.52	0.20	42.12	0.29	—	—	0.51	0.19
SO3	—	—	0.32	—	—	—	—	—	—	—	—	—
Ir2O	—	—	—	—	3.49	—	—	—	—	—	—	—
F (wt.%)	—	—	—	—	—	—	2.70	0.31	—	—	—	—
a.p.f.u. based on 24 oxygen atoms
Y	0.07	—	—	—	4.97	0.34	—	—	—	—	—	—
La	0.69	0.14	—	—	—	—	—	—	—	—	—	—
Ce	0.49	0.06	6.49	0.01	—	—	—	—	—	—	—	—
Pr	0.07	—	—	—	—	—	—	—	—	—	—	—
Nd	0.10	0.01	—	—	—	—	—	—	0.05	—	—	—
Pm	—	—	—	—	—	—	—	—	0.06	0.01	—	—
Eu	0.06	0.01	—	—	—	—	—	—	—	—	—	—
Gd	—	—	—	—	0.08	—	—	—	—	—	—	—
Dy	—	—	—	—	0.26	0.03	—	—	—	—	—	—
Er	—	—	—	—	0.21	0.01	—	—	—	—	0.19	0.05
Yb	—	—	—	—	0.17	0.01	0.12	0.04	—	—	—	—
Si	5.94	0.04	2.81	0.59	1.32	0.69	—	—	6.18	0.13	3.82	0.59
Ti	—	—	—	—	—	—	—	—	—	—	—	—
Al	4.08	0.17	2.41	0.38	—	—	—	—	5.84	0.14	2.98	0.03
Fe	1.71	0.07	1.30	0.23	—	—	—	—	—	—	11.32	1.40
Mn	—	—	1.34	0.13	—	—	—	—	—	—	—	—
Mg	0.22	0.03	—	—	—	—	—	—	—	—	—	—
Ca	2.27	0.21	0.18	0.01	0.26	0.02	8.70	0.08	2.63	0.17	0.05	0.01
Na	—	—	0.72	0.52	—	—	—	—	0.26	0.15	—	—
P	—	—	0.72	0.25	4.98	0.20	5.52	0.03	—	—	0.09	0.03
S	—	—	0.06	—	—	—	—	—	—	—	—	—
Ir	—	—	—	—	0.19	—	—	—	—	—	0.02	—
F	—	—	—	—	—	—	1.32	0.15	—	—	—	—

FeO^T = total Fe, , a.p.f.u. = atoms per formula units, SD = standard deviation, — = not detected

The XRD results for the clay fractions separated from the representative rocks are shown in Fig. 5. In the gabbroic anorthosite, kaolin minerals (7.10 Å) and smectite (14.24 Å) were identified as clay minerals (Fig. 5A), and actinolite (8.41 Å), quartz (4.24 Å, 3.34 Å), anorthite (3.28 Å, 3.12 Å), and ilmenite (2.56 Å) were detected (Fig. 5A). In the dioritic pegmatite, clay fractions were primarily composed of kaolin minerals, such as kaolinite (7.28 Å) and halloysite (4.45 Å), while smectite (14.06 Å), quartz (3.34 Å), anorthosite (3.18 Å), and ilmenite (2.56 Å) were weakly developed (Fig. 5B). Meanwhile, in the Fe–Ti ores, actinolite (8.41 Å), quartz (3.34 Å), anorthite (3.28 Å, 3.12 Å), apatite (2.80 Å), ilmenite (2.75 Å, 2.56 Å), and magnetite (2.53 Å) were predominantly observed (Fig. 5C), and clay minerals, such as smectite and kaolin minerals (7.04 Å), were identified with weak reflection intensities (Fig. 5C). The FT-IR spectra for the clay fractions were characterized by typical absorptions of kaolin minerals in which OH-stretching occurred such as surface OH (3695 cm^–1, 3690 cm^–1) and inner OH (3620 cm^–1). The dioritic pegmatite sample predominantly showed OH absorption by kaolin minerals (Fig. 6).

Fig. 5.

XRD patterns of clay fractions separated from representative whole rocks. (A) Gabbroic anorthosite. (B) Dioritic pegmatites. (C) Fe–Ti ores. Abbreviations: Sm = smectite, K = kaolinite, H = halloysite. Refer to Figs. 2 and 3 for others.

Fig. 6.

FT-IR spectrum of representative clay fractions. Blue line: gabbroic anorthosite, Red line: dioritic pegmatites, Green line: Fe–Ti ores.

The XRF data for the whole rock samples are provided in Table 3. In the gabbroic anorthosite, CaO (av. 13.5 wt.%) contents were higher than those in the other rocks (Fig. 7C). The relatively high concentrations of K₂O (av. 1.5 wt.%) were observed in the dioritic pegmatite (Fig. 7D). In the Fe–Ti ores, Fe₂O₃^T (av. 28.3 wt.%), MgO (av. 5.3 wt.%), TiO₂ (av. 5.3 wt.%), and P₂O₅ (av. 3.4 wt.%) are markedly enriched (Fig. 7A, B, E, F). The trace elements of the whole rocks and clay fractions are shown in Fig. 8 and Table 4. In the gabbroic anorthosite, Sr was remarkably enriched (Fig. 7G). Additionally, Sr decreased with the SiO₂ content in the order of dioritic pegmatite and Fe–Ti ores. In the dioritic pegmatite, Ba contents gradually increased with the SiO₂ content, whereas they were relatively depleted in the gabbroic anorthosite and Fe–Ti ores (Fig. 7H). In the Fe–Ti ores, V contents were distinguishably higher compared to the other rocks (Fig. 7I).

Table 3. Major element composition of whole rock samples from the study area [wt.%]

Samples	SiO2	TiO2	Al2O3	Fe2O3T	MnO	MgO	CaO	Na2O	K2O	P2O5	Cr2O3	L.O.I	Total	CIA	CIW
Gabbroic anorthosite
BGA1	48.91	0.15	28.61	2.61	0.04	2.64	13.29	1.90	0.31	b.d.l.	0.05	1.47	99.99	65	65
BGA2	48.92	0.14	28.78	2.57	0.04	2.59	13.19	1.93	0.32	b.d.l.	0.05	1.45	99.99	65	66
BGA3	49.24	0.16	28.18	2.71	0.05	2.17	13.92	1.71	0.41	0.02	0.06	1.37	100.00	64	64
Dioritic pegmatite
BDP1	53.12	0.34	26.14	3.97	0.08	1.00	2.75	0.65	2.27	0.02	0.02	9.64	100.00	82	88
BDP2	58.94	0.27	24.34	2.55	0.03	0.69	1.21	1.51	1.66	0.03	0.02	8.75	100.01	85	90
BDP3	52.89	0.54	26.48	3.89	0.05	1.19	1.57	0.74	1.20	0.03	0.02	11.41	100.00	88	92
BDP4	65.58	0.49	16.53	4.14	0.06	0.33	4.98	1.92	1.20	0.27	0.03	4.48	100.00	67	71
BDP5	66.16	0.52	16.10	4.35	0.07	0.34	4.93	2.00	1.24	0.29	0.04	3.97	100.00	66	70
Fe–Ti ores
BFT1	40.05	3.99	6.47	26.32	0.31	5.15	12.28	0.38	0.21	3.56	0.01	1.27	100.01	33	34
BFT2	38.60	5.24	5.60	28.23	0.34	5.24	11.87	0.34	0.22	3.34	0.01	0.97	99.99	31	31
BFT3	37.10	6.54	4.29	30.28	0.36	5.52	11.66	0.31	0.21	3.15	0.01	0.57	100.00	26	26

Fe₂O₃^T = total Fe, CIA (chemical index of alteration): [Al₂O₃/(Al₂O₃ + CaO + Na₂O + K₂O) × 100], CIW (chemical index of weathering): [Al₂O₃/(Al₂O₃ + CaO + Na₂O) × 100], L.O.I. = loss on ignition, b.d.l. = below detection limit.

Fig. 7.

Harker variation diagram against SiO₂ contents for whole rock samples. (A) Fe₂O₃^T. (B) MgO. (C) CaO. (D) K₂O. (E) P₂O₅. (F) TiO₂. (G) Sr. (H) Ba. (I) V.

Fig. 8.

Chondrite-normalized REE (A, B) and trace elements (C, D) spidergrams for whole rocks (A, C) and clay fractions (B, D).

Table 4. Trace and rare earth element concentrations of whole rocks and clay fractions from the study area

Samples (ppm)	Gabbroic anorthosite						Dioritic pegmatite										Fe–Ti ores						Cation exchange of clay fractions
	Whole rocks			Clay fractions			Whole rocks					Clay fractions					Whole rocks			Clay fractions			BDP-1			CFT-1
	BGA-1	BGA-2	BGA-3	CGA-1	CGA-2	CGA-3	BDP-1	BDP-2	BDP-3	BDP-4	BDP-5	CDP-1	CDP-2	CDP-3	CDP-4	CDP-5	BFT-1	BFT-2	BFT-3	CFT-1	CFT-2	CFT-3	NH4Cl	(NH4)2SO4	NH4HCO3	NH4Cl	(NH4)2SO4	NH4HCO3
Ba	62.87	63.60	95.45	172.18	119.73	116.30	261.60	290.65	187.01	517.6	549.8	368.06	380.42	300.19	774.5	774.2	43.86	39.00	34.37	106.20	94.58	101.40	120.04	143.64	156.32	102.07	104.30	97.13
Rb	9.54	8.07	9.60	34.46	14.31	23.44	34.93	46.67	22.21	36.6	40.5	59.31	70.32	59.42	74.8	74.8	3.25	1.94	1.79	15.69	13.72	10.90	24.29	22.62	23.48	13.26	14.32	12.73
Cs	0.83	0.85	0.82	2.30	2.16	1.89	4.92	7.70	3.78	1.91	1.80	6.99	5.16	7.45	3.31	3.39	0.63	0.62	0.62	2.52	2.39	2.68	5.22	5.23	5.05	2.42	2.48	2.46
Th	b.d.l.	b.d.l.	b.d.l.	b.d.l.	0.63	b.d.l.	32.12	2.05	27.61	0.56	0.64	64.70	5.48	52.01	1.61	1.36	2.10	2.74	2.94	5.90	4.64	5.22	8.24	13.30	13.03	5.57	4.77	5.55
U	0.03	0.04	0.03	0.10	0.08	0.09	2.54	1.04	4.35	0.62	0.69	4.79	1.92	16.27	0.93	0.96	1.37	1.56	1.64	2.40	2.40	2.63	0.81	1.39	1.33	2.35	2.34	2.34
Zr	6.55	0.92	1.01	1.39	1.00	1.48	389.40	23.75	21.25	13.4	17.4	17.40	8.40	12.47	30.1	31.7	12.71	14.58	17.50	26.67	13.86	14.45	10.11	12.52	13.12	18.98	13.95	12.60
Nb	0.62	b.d.l.	0.71	0.87	1.77	0.47	10.15	7.64	20.38	16.1	17.7	9.76	7.62	20.27	18.2	17.8	33.15	47.11	48.98	23.06	26.06	23.03	8.36	7.90	8.40	18.53	17.86	21.53
Sc	6.95	6.78	5.57	7.28	7.55	6.88	6.43	8.65	18.03	4.97	5.63	10.96	8.84	21.58	10.1	9.80	71.04	70.83	75.12	80.20	79.66	71.03	2.06	2.94	2.61	70.12	73.68	68.11
V	47.68	42.48	35.26	45.25	44.98	42.93	32.48	26.47	34.19	11.3	12.1	28.36	24.85	54.17	22.8	25.3	95.28	90.88	96.03	102.30	94.66	98.00	22.09	23.43	22.29	88.72	97.89	83.05
Sr	258.23	255.17	293.35	302.11	268.38	270.45	115.49	149.25	132.03	146.9	165.9	138.80	193.06	145.10	188.3	191.6	80.65	69.00	57.21	111.37	101.86	98.67	28.31	36.82	50.00	102.66	108.53	100.78
Cr	392.79	390.70	425.82	175.09	163.00	143.44	153.55	122.75	172.04	222.3	263.1	89.75	58.67	58.17	67.4	62.7	86.24	78.09	80.34	69.30	64.38	58.00	83.65	78.40	84.29	57.09	66.72	50.11
Ni	31.81	25.79	21.07	17.36	16.90	15.93	15.50	7.77	13.98	11.0	13.5	8.98	7.37	12.67	7.39	6.85	14.36	15.36	15.15	17.25	17.97	19.25	8.84	8.12	7.99	16.79	17.41	16.74
Zn	21.02	14.92	2.14	12.88	16.36	24.88	13.52	21.15	34.71	29.0	30.1	32.60	52.88	36.80	84.0	75.2	298.00	323.02	350.63	378.27	368.68	395.69	19.34	15.98	30.83	357.83	353.64	338.90
Co	11.22	10.95	10.34	13.21	13.65	13.42	16.93	10.67	18.36	6.36	6.52	15.74	16.55	32.75	10.6	10.7	48.80	49.62	50.87	74.41	72.38	79.15	13.99	11.78	11.59	69.93	71.33	68.88
Mo	2.39	0.95	0.57	0.21	0.27	0.12	2.58	1.99	1.19	0.41	0.45	1.57	1.96	2.87	0.39	0.47	1.48	2.29	2.54	6.75	5.26	4.10	1.88	1.57	1.57	3.72	4.57	3.76
Cu	3.13	3.00	5.09	2.53	2.02	6.30	22.60	12.06	28.83	4.58	5.18	17.75	11.99	37.06	6.16	6.48	31.80	37.34	40.46	88.43	92.95	103.59	17.71	17.22	17.36	84.12	85.67	82.82
Pb	3.10	2.97	2.36	3.24	2.98	2.71	11.70	4.78	9.12	36.5	37.6	9.17	6.38	17.53	67.5	69.6	5.20	4.99	4.54	11.41	10.02	10.83	7.39	6.39	6.99	10.88	10.51	11.42
Ga	16.37	15.94	16.80	16.43	15.83	17.95	32.49	28.69	32.39	25.1	26.2	40.80	29.77	42.80	35.3	36.7	22.90	21.09	19.91	35.63	31.15	33.97	34.13	32.81	35.21	33.99	33.79	33.69
La	1.64	2.18	1.82	4.27	3.19	3.28	57.51	21.21	62.86	24.1	26.4	75.88	19.21	166.72	17.8	18.5	106.86	99.66	101.77	104.24	99.16	94.94	8.43	12.03	11.61	100.11	96.31	102.54
Ce	2.54	3.75	3.30	7.15	4.63	5.09	110.16	37.70	91.73	50.9	54.7	130.21	48.60	241.35	39.7	41.0	220.14	207.85	212.40	212.99	204.34	197.78	16.38	26.90	33.98	202.73	194.68	205.98
Pr	0.32	0.43	0.40	0.70	0.74	0.75	10.91	4.08	11.95	5.44	6.12	14.83	3.51	28.83	4.00	4.23	30.97	29.09	29.89	32.44	29.98	28.80	1.76	2.59	2.55	30.60	29.32	31.18
Nd	1.40	1.82	1.81	2.86	2.58	3.32	38.47	15.86	44.25	23.7	27.0	52.86	13.00	97.51	17.2	18.3	139.56	130.71	132.21	140.39	136.37	129.48	6.25	8.99	8.98	133.03	133.76	135.00
Sm	0.29	0.31	0.36	0.51	0.60	0.73	6.22	2.96	8.77	5.97	6.91	8.95	2.50	14.49	4.59	5.00	31.87	30.00	30.20	34.16	33.62	31.89	1.20	1.74	1.80	32.32	32.54	32.65
Eu	0.71	0.71	0.78	0.70	0.95	0.88	1.04	1.18	1.13	3.06	3.09	1.34	1.08	2.23	2.51	2.81	4.76	4.28	4.21	5.79	5.70	5.36	0.24	0.31	0.34	5.47	5.57	5.56
Gd	0.30	0.34	0.41	0.53	0.63	0.86	6.04	3.04	8.68	8.52	9.78	8.30	2.81	12.97	6.84	7.64	34.71	33.30	33.27	38.98	38.83	36.77	1.33	1.82	2.07	36.53	37.70	36.92
Tb	0.01	0.01	0.02	0.04	0.17	0.11	0.99	0.47	1.44	1.56	1.80	1.40	0.49	1.96	1.31	1.46	5.27	5.14	5.11	6.13	6.20	5.76	0.25	0.34	0.40	5.68	5.94	5.78
Dy	0.28	0.33	0.38	0.48	0.60	0.82	6.69	3.21	8.53	11.0	12.7	8.49	3.64	10.59	9.79	11.0	29.31	27.81	28.41	34.51	35.40	33.50	1.76	2.45	2.87	31.48	33.91	32.41
Y	1.93	1.99	2.36	4.69	3.20	3.20	38.21	15.90	35.56	66.5	77.4	44.16	16.19	31.70	45.9	54.0	160.36	162.45	162.75	190.12	177.41	172.07	8.21	10.78	13.57	176.72	171.50	180.40
Ho	0.04	0.05	0.05	0.08	0.19	0.16	1.62	0.70	1.70	2.59	2.94	1.85	0.83	2.07	2.25	2.51	5.95	5.71	5.81	6.98	7.29	6.77	0.41	0.56	0.69	6.45	6.95	6.53
Er	0.20	0.22	0.25	0.35	0.43	0.55	6.14	2.30	5.20	8.15	9.30	6.28	2.83	6.49	7.18	7.84	16.99	16.36	16.35	20.70	19.70	19.39	1.55	2.15	2.50	18.14	20.13	18.53
Tm	0.01	0.01	0.01	0.01	0.11	0.05	1.02	0.31	0.68	1.18	1.30	0.98	0.42	0.88	1.03	1.14	2.14	2.06	2.09	2.57	2.63	2.53	0.28	0.39	0.45	2.39	2.55	2.50
Yb	0.20	0.22	0.24	0.30	0.44	0.47	8.10	2.22	4.86	7.93	8.70	7.63	3.19	6.23	6.97	7.71	13.50	12.85	12.96	167.00	16.70	15.61	2.31	3.11	3.59	14.99	16.26	15.07
Lu	0.01	0.01	0.01	0.02	0.09	0.05	1.31	0.30	0.69	1.11	1.24	1.19	0.48	0.92	0.96	1.08	1.90	1.81	1.85	2.35	2.25	2.25	0.38	0.50	0.56	2.11	2.31	2.12
∑LREE	6.89	9.20	8.47	16.18	12.68	14.06	224.30	82.99	220.69	113.14	124.15	284.07	87.89	551.13	85.86	89.88	534.17	501.60	510.68	530.02	509.17	488.25	34.25	52.55	59.25	504.25	492.18	512.92
∑HREE	2.98	3.17	3.72	6.50	5.85	6.27	70.12	28.44	67.35	108.56	125.09	80.28	30.88	73.79	82.27	94.34	270.15	267.49	268.60	469.33	306.40	294.66	16.48	22.11	26.71	294.49	297.25	300.28
∑REE	16.83	19.14	17.76	29.97	26.08	27.21	300.85	120.09	306.06	226.67	254.87	375.31	127.60	646.50	178.26	194.02	875.35	839.91	854.40	1079.55	895.23	853.94	52.79	77.60	88.58	868.86	863.10	881.31
Ce/Ce*	0.85	0.93	0.94	1.00	0.73	0.78	1.06	0.98	0.81	1.08	1.04	0.94	1.43	0.84	1.14	1.12	0.93	0.93	0.93	0.89	0.91	0.92	1.03	1.17	1.51	0.89	0.89	0.88
Eu/Eu*	7.31	6.66	6.18	4.15	4.71	3.39	0.52	1.20	0.39	1.31	1.15	0.48	1.25	0.50	1.36	1.39	0.44	0.41	0.40	0.48	0.48	0.48	0.57	0.52	0.53	0.48	0.48	0.49

LREE = La–Eu, HREE = (Gd–Lu) + Y, REE = (La–Lu) + Y + Sc, (La/Yb)_n = (La/Yb)/(La/Yb)chondrite, Ce/Ce* = Ce_n/(La_n × Pr_n)^1/2, Eu/Eu* = Eu_n/(Sm_n × Gd_n)^1/2, b.d.l. = below detection limit

In our study, the recovery rate of exchangeable REE was estimated for the clay fractions separated from the REE-enriched representative rocks such as dioritic pegmatite and Fe–Ti ores (Table 4). In the clay fractions of the dioritic pegmatites, more than 80% of the exchangeable REEs were extracted when ammonium chloride (NH₄Cl) solutions were used, and the recovery rates were estimated for ammonium sulfate ((NH₄)₂SO₄, av. 73%) and ammonium bicarbonate (NH₄HCO₃, av. 70%). In contrast, in the clay fractions of the Fe–Ti ores containing the highest concentrations of ∑REE in our study, exchangeable REEs were recovered less than 10% from the clay fractions, regardless of the ammonium-salt solvent species.

Discussion

Chemical characteristics and REE enrichments

The gabbroic anorthosites in the study area have chemical compositions similar to those of a low-K tholeiitic intercumulate-type anorthosite as previously reported (Fig. 9A, B) (Ewart, 1982; Middlemost, 1994; Lee et al., 1999). Parent rocks with a low chemical index of alteration (CIA) (av. 65) and chemical index of weathering (CIW) (av. 65) values were associated with kaolinization (Table 3, Fig. 10B). Anorthite incorporated small amounts of REEs (Table 4) and formed clay minerals originating from weatherable anorthite (Fig. 5A). Additionally, the higher ∑REEs content of the clay fractions (av. 28 ppm) than that of the whole rocks (av. 18 ppm) suggest that the REEs were released from the parent rocks and then adsorbed on kaolin minerals (Table 4).

Fig. 9.

Total alkali versus silica (TAS) diagram (A) and K₂O vs. SiO₂ diagram (B) for whole rock samples.

Fig. 10.

Ternary plots for the index of lateritization (A) and chemical index of alteration (B) for whole rock samples.

The dioritic pegmatites are divided into gabbroic diorite, diorite, and granodiorite (Fig. 9A), belonging to the calc-alkaline and high-K calc-alkaline series, in contrast to the gabbroic anorthosite, which belonged to the low-K tholeiitic series (Fig. 9B). The studied samples contained REE-bearing minerals, such as allanite and cerite (Fig. 4B–D). Allanite primarily coexisted with anorthite and andesine, whereas cerite was observed with albite (Fig. 4D). For example, the clay fraction sample (i.e., BDP3) with abundant kaolin minerals (e.g., kaolinite and halloysite) showed a higher concentration of ∑REE (647 ppm) than the whole rock (306 ppm) (Fig. 5B, Table 4). Furthermore, these samples were commonly characterized by high CIA and CIW values of approximately 90 (Table 3). Thus, the REE³⁺ ions released from the REE-bearing weatherable allanite may have been adsorbed on the negatively charged layer surface of the kaolin minerals (Fig. 4B, C). CIA and CIW are generally interpreted as the measures of the conversion extent of feldspars to kaolin minerals in the lithosphere (Nesbitt and Young, 1984; Fedo et al., 1995; Price and Velbel, 2003). According to Fedo et al. (1995), the CIA and CIW values of fresh rocks mostly plot near 50; however, chemical indices of alteration and weathering seemed to be invalid for the Fe–Ti ores (Fig. 10B), because the chemical indices for the Fe–Ti ores in our study were less than 35 (Table 3). The abnormally low Al₂O₃ contents compared to the CaO contents in the bulk compounds of the Fe–Ti ores led to lower CIA and CIW values (Table 3), potentially due to the high proportion of Ca-rich actinolite in the parent rocks (Fig. 5C). The ∑REE contents of both whole rocks (av. 857 ppm) and clay fractions (av. 943 ppm) were higher than those of the dioritic pegmatites and gabbroic anorthosite, respectively (Fig. 8A, B). REE-bearing minerals, such as xenotime, allanite, and apatite, were frequently identified. Simultaneously, kaolin minerals contributed to REE migration from the parent rocks to clay fractions during the weathering process.

Relationship between Eu anomaly and REE enrichments

REE³⁺ has a relatively large ionic radius and preferentially forms complexes with hard ligands containing highly electronegative donor atoms, such as oxygen, sulfur, phosphorus, and fluorine (e.g., F^–, SO₄^2–, CO₃^2–, PO₄^3–, and OH^–) (Panahi et al., 2000; Kanazawa and Kamitatni, 2006). Additionally, the geochemical coherency of REE during the magmatic process, resulting from the characteristics of their electronic configuration, causes a particularly stable 3⁺ oxidation state, and gradually, they have a smaller ionic radius with increasing atomic number (Henderson, 1996). Exceptionally, Eu is different from other REEs, as it can exist in both Eu²⁺ state under reducing conditions and Eu³⁺ state under oxidizing conditions (Weill and Drake, 1973). In our study, the ∑REEs of the samples had a linear correlation with decreasing Eu/Eu* values (Fig. 11A). In the case of gabbroic anorthosite, the Eu/Eu* values of the clay fractions were lower than those of whole rocks. According to Henderson (1996), in plagioclase, Ca²⁺ can be a suitable substitution for the reduction of Eu³⁺ to Eu²⁺ in magmatic systems. Furthermore, Eu appeared to be gradually replaced by other REE³⁺ ions during weathering process (Fig. 11A).

Fig. 11.

Binary plots for the REE concentrations against (A) Eu/Eu* and (B) Ce/Ce* for whole rocks and clay fractions.

In this study, the clay fractions had higher ∑REEs concentrations than the whole rocks. For example, clay fractions containing kaolin minerals are secondarily formed by weatherable primary minerals such as anorthite. In this process, exchangeable REE³⁺ ions released from weatherable anorthite were probably adsorbed on the layer surface of kaolin minerals (Bao and Zhao, 2008; Dinali et al., 2019). In the dioritic pegmatite and the Fe–Ti ore, the Eu/Eu* values significantly decreased compared to those of the gabbroic anorthosite; however, the Eu/Eu* values were not clearly divided between their whole rocks and clay fractions. This may be attributed to insoluble rare earth minerals such as allanite containing Eu in the parent rocks (Table 2).

Relationship of Ce anomaly and REE enrichments

Generally, Ce fractionation occurs because of a difference in mobility between Ce⁴⁺ and other trivalent light rare earth elements (LREE), including Ce³⁺ (Bellot et al., 2018; Ichimura et al., 2020). For example, positive Ce anomalies have been found in weathered crusts, such as lateritic profiles, which were formed from amphibolites and gabbroic rocks (Braun et al., 1990; Bao and Zhao, 2008). These Ce anomalies can be attributed to the precipitation of Ce-rich minerals (Bao and Zhao, 2008), which were accumulated by the change of Ce³⁺ to Ce⁴⁺ during the supply of oxygen-rich rainfall in clay-rich weathered granitoids (Braun et al., 1990). In contrast, negative Ce anomalies have occasionally been found in some lateritic rocks that are enriched in Ti, Fe, P, and REE (Angélica and da Costa, 1993). For example, Ce⁴⁺ can be replaced by Ti⁴⁺; consequently, TiO₂-enriched rocks have a negative Ce anomaly (Angélica and da Costa, 1993). Additionally, hydrothermal Fe-rich sediments inherited from the mid ocean ridge (MOR) are characterized by negative Ce anomalies owing to the influence of oxygen-deficient environments (Kato et al., 1998, 2002). The whole rocks and the clay fractions from the study area have weakly negative Ce-anomalies (i.e., <1.0), except for several dioritic pegmatites, which show Ce anomalies of 1.0–1.4 (Fig. 11B). The positive Ce anomalies in these rocks could be attributed to the contribution of Ce-bearing minerals, such as allanite and cerite (Fig. 4B–D). On the other hand, the gabbroic anorthosites showing negative Ce anomalies were generated in oxygen-deficient reduced environment associated with mantle-origin anorthositic magma, which underwent assimilation with crustal material during crystallization (Kwon and Jeong, 1990). In case of the Fe–Ti ores, the negative Ce anomalies were attributed to the change of Ce⁴⁺ to Ti⁴⁺ (Fig. 4F) (Angélica and da Costa, 1993). Additionally, the relatively high ∑REEs in the Fe–Ti ores compared to those of the gabbroic anorthosites were probably attributed to the supply from the peripheral dioritic pegmatites.

Extraction of exchangeable REE on clay fractions by cation exchange

Ammonium cations (NH₄⁺) have been used to exchange REE³⁺ ions adsorbed on clay surfaces to resolve them in the solution (Chi et al., 2003; de Vasconcellos et al., 2008; Moldoveanu and Papangelakis, 2012; Nie et al., 2020; Feng et al., 2021; Luo et al., 2022). For example, the concentrations of leaching agents (e.g., ammonium-salt solvent) on the clay surface decrease because reaction consumption occurs on the surface layers, while the reaction products (i.e., exchangeable REE³⁺) gradually increase near the clay surface (Luo et al., 2022). The process of exchangeable REE³⁺ recovery using an ammonium-salt solvent is as follows:

  

$\left[\mathrm{C}\mathrm{l}\mathrm{a}\mathrm{y}\right]\cdot \mathrm{n}{\mathrm{R}\mathrm{E}\mathrm{E}}_{\left(\mathrm{s}\right)}^{3+}+3\mathrm{n}{\mathrm{N}\mathrm{H}}_{4\left(\mathrm{a}\mathrm{q}\right)}^{+}\rightleftharpoons \left[\mathrm{C}\mathrm{l}\mathrm{a}\mathrm{y}\right]\cdot {(\mathrm{N}\mathrm{H}}_4^{+}{)}_{3\mathrm{n}\left(\mathrm{s}\right)}+\mathrm{n}{\mathrm{R}\mathrm{E}\mathrm{E}}_{\left(\mathrm{a}\mathrm{q}\right)}^{3+}$[2]

where [Clay] is the clay fraction, s is the solid phase, and (aq) is the liquid phase (Luo et al., 2022).

Wieland and Stumm (1992) reported that proton-promoted dissolution at the edge surface of kaolinite starts under weakly-acidic pH conditions (<6.5). In this study, an ammonium chloride solvent with a pH value of 4.6 activated the dissolution of REE³⁺ ions from the kaolinite-bearing clay fractions more than ammonium sulfate solvent with pH of 5.0 (Fig. 12A). Ammonium bicarbonate is generally preferred as the recovery solvent for exchangeable REEs from the clays due to the advantage of being environmentally sustainable (Chi et al., 2003; de Vasconcellos et al., 2008; Anawati and Azimi, 2022). Of particular important factors in association with the leaching of exchangeable REEs from the clays are carbonate concentrations and pH change in solvent (Anawati and Azimi, 2022). Thus, significant releasing (about 70%) of exchangeable REEs by treatment with ammonium bicarbonate solution seems to be attributed by the interaction with carbonate ions (Fig. 12A). On the other hand, the REEs have rarely been preserved as exchangeable REE³⁺ ions in the Fe–Ti ores because only a small amount of kaolin minerals was identified from the XRD analysis (Figs. 5C, 12B). In contrast, most of the REEs seem to remain as clay size non-exchangeable particles derived from phosphate minerals, such as apatite and xenotime (Figs. 4E, 5C). Understanding the geochemical characteristics of REE-bearing weathered dioritic pegmatites in the Hadong district would help in the exploration of ion-adsorption type REE-bearing mineralization and in securing rare earth resources.

Fig. 12.

Relative residual concentrations of the REE for clay fractions from (A) dioritic pegmatite and (B) Fe–Ti ore after treatment with ammonium-salt solvents (upper graph) and each recovery rate of exchangeable REE (lower graph). Blue bar: bulk REE concentrations.

Conclusions

The geochemical characteristics and REE distribution of the weathered intercumulate-type anorthosite complex in Hadong district were investigated, and the contents of exchangeable REE from the clay fractions were estimated. The results led to the following conclusions:

1. REE minerals were rarely observed in the gabbroic anorthosites, except for some ilmenites incorporating LREEs. In contrast, LREE-rich minerals, such as allanite and cerites, were abundant in dioritic pegmatites, whereas heavy rare earth elements (HREE)-rich mineral, such as xenotime, was frequently observed in the Fe–Ti ores.

2. The ∑REEs of the whole rocks gradually increased with the contribution of REE-bearing minerals, in the order of gabbroic anorthosite (av. 18 ppm), dioritic pegmatite (av. 242 ppm), and Fe–Ti ores (av. 857 ppm). Additionally, the ∑REE of the clay fractions were more enriched, 28 ppm, 304 ppm, and 943 ppm, respectively, than those of the whole rocks because of the contribution of REE-bearing kaolin minerals.

3. For gabbroic anorthosite, the correlations between ∑REE and Eu/Eu* in both whole rocks and clay fractions suggested that they were controlled by weatherable feldspars. Negative Ce-anomalies of the whole rocks and clay fractions were attributed to cation substitution under oxygen-deficient reducing conditions.

4. For kaolin minerals-bearing clay fractions separated from the dioritic pegmatite, up to 80% of the exchangeable REEs were recovered using ammonium-salt solvents (i.e., NH₄Cl). In contrast, in kaolin minerals-deficient clay fractions separated from the Fe–Ti ores with the largest amount of ∑REEs, exchangeable REEs were recovered by less than 10% regardless of the ammonium-salt solvent species. Therefore, the geochemical characteristics of the REE-bearing weathered intercumulate-type anorthosite complex and cation exchange methods to recover exchangeable REEs from kaolinite-bearing clay fractions could be applicable in the exploration and extraction of ion-adsorption type REE deposits.

Acknowledgments

We would like to thank the members of the Korea Basic Science Institute for their kind assistance in analysis. This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Education (No. 2020R1I1A1A01058214) and the Ministry of Science and ICT (MSIT) (No. 2019R1A2C1085334, No. 2021M2E1A1099413, No. 2022R1A2C1008260).

References

•  Anawati,  J. and  Azimi,  G. (2022) Separation of rare earth elements from a South American ionic clay lixivium by sequential precipitation. Hydrometallurgy 213, 105946. https://doi.org/10.1016/j.hydromet.2022.105946
•  Angélica,  R. S. and  da Costa,  M. L. (1993) Geochemistry of rare-earth elements in surface lateritic rocks and soils from the Maicuru complex, Para, Brazil. J. Geochem. Explor. 47, 165–182. https://doi.org/10.1016/0375-6742(93)90064-S
•  Balaram,  V. (2019) Rare earth elements: A review of applications, occurrence, exploration, analysis, recycling, and environmental impact. Geosci. Front. 10, 1285–1303. https://doi.org/10.1016/j.gsf.2018.12.005
•  Bao,  Z. and  Zhao,  Z. (2008) Geochemistry of mineralization with exchangeable REY in the weathering crusts of granitic rocks in South China. Ore Geol. Rev. 33, 519–535. https://doi.org/10.1016/j.oregeorev.2007.03.005
•  Bau,  M. (1996) Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contrib. Mineral. Petrol. 123, 323–333. https://doi.org/10.1007/s004100050159
•  Bellot,  N.,  Boyet,  M.,  Doucelance,  R.,  Bonnand,  P.,  Savov,  I. P.,  Plank,  T. and  Elliott,  T. (2018) Origin of negative cerium anomalies in subduction-related volcanic samples: Constraints from Ce and Nd isotopes. Chem. Geol. 500, 46–63. https://doi.org/10.1016/j.chemgeo.2018.09.006
•  Braun,  J.-J.,  Pagel,  M.,  Muller,  J.-P.,  Bilong,  P.,  Michard,  A. and  Guillet,  B. (1990) Cerium anomalies in lateritic profiles. Geochim. Cosmochim. Acta 54, 781–795. https://doi.org/10.1016/0016-7037(90)90373-S
•  Chi,  R.,  Zhou,  Z.,  Xu,  Z.,  Hu,  Y.,  Zhu,  G. and  Xu,  S. (2003) Solution-chemistry analysis of ammonium bicarbonate consumption in rare-earth-element precipitation. Metall. Mater. Trans. B 34, 611–617. https://doi.org/10.1007/s11663-003-0031-z
•  Choi,  J. B. and  Kwak,  J. Y. (2012) Occurrence of REE-bearing allanite with Th-mineral (thorite) in Wolhoengri, Hadong, Korea. J. Miner. Soc. Korea 25, 295–304. (in Korean with English abstract). https://doi.org/10.9727/jmsk.2012.25.4.295
•  de Vasconcellos,  M. E.,  da Rocha,  S. M. R.,  Pedreira,  W. R.,  Queiroz,  C. A. da S. and  Abrão,  A. (2008) Solubility behavior of rare earths with ammonium carbonate and ammonium carbonate plus ammonium hydroxide: Precipitation of their peroxicarbonates. J. Alloys Compd. 451, 426–428. https://doi.org/10.1016/j.jallcom.2007.04.163
•  Dinali,  G. S.,  Root,  R. A.,  Amistadi,  M. K.,  Chorover,  J.,  Lopes,  G. and  Guilherme,  L. R. G. (2019) Rare earth elements (REY) sorption on soils of contrasting mineralogy and texture. Environ. Int. 128, 279–291. https://doi.org/10.1016/j.envint.2019.04.022
•  Estrade,  G.,  Salvi,  S.,  Béziat,  D.,  Rakotovao,  S. and  Rakotondrazafy,  R. (2014) REE and HFSE mineralization in peralkaline granites of the Ambohimirahavavy alkaline complex, Ampasindava peninsula, Madagascar. J. African Earth Sci. 94, 141–155. https://doi.org/10.1016/j.jafrearsci.2013.06.008
• Ewart, A. (1982) The mineralogy and petrology of Tertiary-recent orogenic volcanic rocks: with special reference to the andesite-basaltic compositional range. Andesites: Orogenic Andesites and Related Rocks (Thorpe, R. S., ed.), 25–95, John Wiley & Sons.
•  Fedo,  C. M.,  Wayne Nesbitt,  H. and  Young,  G. M. (1995) Unraveling the effects of potassium metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering conditions and provenance. Geology 23, 921–924. https://doi.org/10.1130/0091-7613(1995)023<0921:UTEOPM>2.3.CO;2
•  Feng,  X.,  Onel,  O.,  Council-Troche,  M.,  Noble,  A.,  Yoon,  R.-H. and  Morris,  J. R. (2021) A study of rare earth ion-adsorption clays: The speciation of rare earth elements on kaolinite at basic pH. Appl. Clay Sci. 201, 105920. https://doi.org/10.1016/j.clay.2020.105920
•  Henderson  P. (1996) The rare earth elements: Introduction and review. Rare Earth Minerals: Chemistry, Origin and Ore Deposits (Jones, A. P., Wall, F. and Williams, T. C., eds.), Chapman & Hall Publishing Ltd., New York, 17 pp.
•  Ichimura,  K.,  Sanematsu,  K.,  Kon,  Y.,  Takagi,  T. and  Murakami,  T. (2020) REE redistributions during granite weathering: Implications for Ce anomaly as a proxy for paleoredox states. Am. Mineral. 105, 848–859. https://doi.org/10.2138/am-2020-7148
•  Ishihara,  S.,  Hua,  R.,  Hoshino,  M. and  Murakami,  H. (2008) REE abundance and REE minerals in granitic rocks in the Nanling range, Jiangxi Province, southern China, and generation of the REE-rich weathered crust deposits. Resour. Geol. 58, 355–372. https://doi.org/10.1111/j.1751-3928.2008.00070.x
•  Jeong,  J. G. (1987) Magmatic differentiation of the anorthositic rocks in Hadong-Sancheong area. J. Geol. Soc. Korea 23, 216–228. (in Korean with English abstract).
•  Jeong,  J. G.,  Kim,  W. S. and  Watkinson,  D. H. (1989) Geological structure of Hadong anorthositic rocks and associated titanium orebody. J. Geol. Soc. Korea 25, 98–111. (in Korean with English abstract).
•  Jung,  J.-S.,  Kim,  J.-S.,  Cho,  H.-S.,  Song,  C.-W.,  Son,  M.,  Ryoo,  C.-R.,  Chi,  S.-J. and  Kim,  I.-S. (2010) Occurrence and deformation of Fe–Ti ores from the Proterozoic Hadong anorthosites, Korea. J. Petrol. Soc. Korea 19, 31–49. (in Korean with English abstract).
•  Kanazawa,  Y. and  Kamitani,  M. (2006) Rare earth minerals and resources in the world. J. Alloys Compd. 408–412, 1339–1343. https://doi.org/10.1016/j.jallcom.2005.04.033
•  Kato,  Y.,  Ohta,  I.,  Tsunematsu,  T.,  Watanabe,  Y.,  Isozaki,  Y.,  Maruyama,  S. and  Imai,  N. (1998) Rare earth element variations in mid-Archean banded iron formations: implications for the chemistry of ocean and continent and plate tectonics. Geochim. Cosmochim. Acta 62, 3475–3497. https://doi.org/10.1016/S0016-7037(98)00253-1
•  Kato,  Y.,  Kano,  T. and  Kunugiza,  K. (2002) Negative Ce anomaly in the Indian Banded Iron Formations: Evidence for the emergence of oxygenated deep-Sea at 2.9-2.7 Ga. Resour. Geol. 52, 101–110. https://doi.org/10.1111/j.1751-3928.2002.tb00123.x
• KIGAM (Korea Institute of Geoscience and Mineral Resources) (2022) Geologic map of Korea 1:50,000. https://www.kigam.re.kr/
• Kim, W.-S. and Jeong, J.-G. (1993) Rare metal (Ti.Ce.La.Nd.Pr.Sm.Y) occurrences within the anorthosite in the Hadong-Sanchong area. KOSEF 911-0501-016-2, 88 pp. (in Korean with English abstract).
•  Kim,  W.-S.,  Jeong,  J.-G.,  Lee,  K.-H. and  Watkinson,  D. H. (1992) Rare metal occurrences within the anorthosite in the Hadong-Sancheong area, Kyungnam Province, Korea. J. Miner. Soc. Korea 5, 14–21. (in Korean with English abstract).
•  Kwon,  S. T. and  Jeong,  J. G. (1990) Preliminary Sr-Nd Isotope Study of the Hadong-Sanchung Anorthositic Rocks in Korea : Implication for their Origin and for the Precambrian Tectonics. J. Geol. Soc. Korea 26, 341–349. (in Korean with English abstract).
•  Lee,  B. C.,  Park,  J. H.,  Oh,  C. W. and  Yi,  K. (2017) Metamorphic and magmatic evolution of the Paleoproterozoic gneisses in the Sancheong area, Yeongnam Massif, South Korea, and their implications to the tectonics in the Northeast Asia. Precambrian Res. 298, 439–461. https://doi.org/10.1016/j.precamres.2017.06.001
•  Lee,  J. M.,  Jeong,  J. G. and  Kim,  W. S. (1999) The Preliminary study on the evolution of Hadong anorthositic rocks and their genetic relations with ilmenite-bearing ore bodies, Korea. J. Geol. Soc. Korea. 35, 321–336. (in Korean with English abstract).
•  Lee,  Y.,  Cho,  M.,  Cheong,  W. and  Yi,  K. (2018) Prolonged high-temperature, low-pressure metamorphism associated with ~1.86 Ga Sancheong–Hadong anorthosite in the Yeongnam Massif, Korea: Paleoproterozoic hot orogenesis in the North China Craton. Precambrian Res. 307, 175–200. https://doi.org/10.1016/j.precamres.2018.01.018
•  Lee,  Y.,  Cho,  M.,  Kim,  T. and  Kim,  H. (2021) Incipient charnockite formation at the waning stage of Paleoproterozoic hot orogenesis, Yeongnam Massif, Korea. Precambrian Res. 365, 106388. https://doi.org/10.1016/j.precamres.2021.106388
•  Li,  M. Y. H.,  Zhao,  W. W. and  Zhou,  M.-F. (2017) Nature of parent rocks, mineralization styles and ore genesis of regolith-hosted REE deposits in South China: An integrated genetic model. J. Asian Earth Sci. 148, 65–95. https://doi.org/10.1016/j.jseaes.2017.08.004
•  Li,  M. Y. H.,  Zhou,  M. F. and  Williams-Jones,  A. E. (2019) The genesis of regolith-hosted heavy rare earth element deposits: Insights from the world-class Zudong deposit in Jiangxi Province, south China. Econ. Geol. 114, 541–568. https://doi.org/10.5382/econgeo.4642
•  Luo,  X.,  Zhang,  Y.,  Zhou,  H.,  He,  K.,  Luo,  C.,  Liu,  Z. and  Tang,  X. (2022) Review on the development and utilization of ionic rare earth ore. Minerals 12, 554. https://doi.org/10.3390/min12050554
•  Middlemost,  E. A. K. (1994) Naming materials in the magma/ igneous rock system. Earth-Sci. Rev. 37, 215–224.
•  Moeller,  T. and  Kremers,  H. E. (1945) The basicity characteristics of scandium, yttrium, and the rare earth elements. Chem. Rev. 37, 97–159.
•  Moldoveanu,  G. A. and  Papangelakis,  V. G. (2012) Recovery of rare earth elements adsorbed on clay minerals: I. Desorption mechanism. Hydrometallurgy 117–118, 71–78. https://doi.org/10.1016/j.hydromet.2012.02.007
• Moon, J. J., Moon, S. W. and Jwa, Y. J. (2012) SHRIMP zircon ages of the dioritic rocks from the Hadong area in the southeastern Yeongnam massif. Fall Conference Abstract of the Association of Korean Geoscience Societies, The Geological Society of Korea, 211. (in Korean).
•  Mukai,  H.,  Kon,  Y.,  Sanematsu,  K.,  Takahashi,  Y. and  Ito,  M. (2020) Microscopic analyses of weathered granite in ion-adsorption rare earth deposit of Jianxi Province, China. Sci. Rep. 10, 20194. https://doi.org/10.1038/s41598-020-76981-8
•  Murakami,  H. and  Ishihara,  S. (2008) REE mineralization of weathered crust and clay sediment on granitic rocks in the Sanyo Belt, SW Japan and the southern Jiangxi Province, China. Resour. Geol. 58, 373–401. https://doi.org/10.1111/j.1751-3928.2008.00071.x
•  Nesbitt,  H. W. and  Young,  G. M. (1984) Prediction of some weathering trends of plutonic and volcanic rocks based on thermodynamic and kinetic considerations. Geochim. Cosmochim. Acta 48, 1523–1534. https://doi.org/10.1016/0016-7037(84)90408-3
•  Nie,  W.,  Zhang,  R.,  He,  Z.,  Zhou,  J.,  Wu,  M.,  Xu,  Z.,  Chi,  R. and  Yang,  H. (2020) Research progress on leaching technology and theory of weathered crust elution-deposited rare earth ore. Hydrometallurgy 193, 105295. https://doi.org/10.1016/j.hydromet.2020.105295
•  Panahi,  A.,  Young,  G. M. and  Rainbird,  R. H. (2000) Behavior of major and trace elements (including REE) during Paleoproterozoic pedogenesis and diagenetic alteration of an Archean granite near Ville Marie, Québec, Canada. Geochim. Cosmochim. Acta 64, 2199–2220. https://doi.org/10.1016/S0016-7037(99)00420-2
•  Price,  J. R. and  Velbel,  M. A. (2003) Chemical weathering indices applied to weathering profiles developed on heterogeneous felsic metamorphic parent rocks. Chem. Geol. 202, 397–416. https://doi.org/10.1016/j.chemgeo.2002.11.001
•  Sanematsu,  K. and  Kon,  Y. (2013) Geochemical characteristics determined by multiple extraction from ion-adsorption type REE ores in Dingnan County of Jiangxi Province, South China. Bull. Geol. Surv. Japan 64, 313–330. https://doi.org/10.9795/bullgsj.64.313
•  Wieland,  E. and  Stumm,  W. (1992) Dissolution kinetics of kaolinite in acidic aqueous solutions at 25°C. Geochim. Cosmochim. Acta 56, 3339–3355. https://doi.org/10.1016/0016-7037(92)90382-S
•  Weill,  D. F. and  Drake,  M. J. (1973) Europium anomaly in plagioclase feldspar: Experimental results and semiquantitative model. Science 180, 1059–1060. https://doi.org/10.1126/science.180.4090.1059
•  Wu,  C.,  Huang,  D. and  Gao,  Z. (1990) REE Geochemistry in the weathered crust of granites, Longnan area, Jiangxi Province. Acta Geol. Sin. 3, 193–209. https://doi.org/10.1111/j.1755-6724.1990.mp3002006.x
•  Zhenxiang,  D.,  Lei,  Q.,  Guanshi,  W.,  Sihai,  L.,  Chenliang,  P. and  Qi,  L. (2019) Metallogenic process of ion adsorption REE ore based on the occurrence regularity of La in kaolin. Ore Geol. Rev. 112, 103022. https://doi.org/10.1016/j.oregeorev.2019.103022