Enrichment Regularity of Indium in the Dulong Mineral Processing Plant, Yunnan Province, China

Abstract

The Dulong deposit in Wenshan, southeastern Yunnan Province, is rich in zinc, tin, and copper resources, accompanied by rare metals such as indium and silver. It is a particularly important indium production base, with reserves of approximately 7000 tons, ranking first globally. Enrichment and recovery of indium-bearing minerals are mainly achieved through mineral processing technology. However, the recovery rate of indium in the Dulong concentrator remains relatively low, and there is an insufficient understanding of its occurrence state and distribution characteristics, resulting in marked indium resource wastage. Here, we conducted a systematic process mineralogy study on indium-bearing polymetallic ore in the Dulong concentrator. The average grade of indium in the ore is 43.87 g/t, mainly occurring in marmatite (63.63%), supplemented by that in silicate minerals (23.31%), chalcopyrite (7.84%), and pyrrhotite (4.22%). The indium has a relatively dispersed distribution, which is inconducive to enrichment and recovery. The substitution mechanism of indium in marmatite was investigated using laser ablation inductively coupled plasma mass spectrometry. This revealed a positive correlation between indium and copper, allowing us to revise the substitution relationship to: ${\mathrm{Zn}}_x\mathrm{S}+{\mathrm{Cu}}^{+}+{\mathrm{In}}^{3+}\to {\mathrm{Zn}}_{\mathrm{x}-2}\mathrm{CuInS}+2{\mathrm{Zn}}^{2+}$ or ${\mathrm{Zn}}_{\mathrm{x}-1}\mathrm{FeS}+{\mathrm{Cu}}^{+}+{\mathrm{In}}^{3+}\to {\mathrm{Zn}}_{\mathrm{x}-2}\mathrm{CuInS}+{\mathrm{Zn}}^{2+}+{\mathrm{Fe}}^{2+}$. Electron probe microanalysis revealed the presence of roquesite (CuInS₂), an independent indium mineral not previously reported from this deposit. Our detailed investigation of the Dulong concentrator mineral processing technology showed that the recovery rate of indium from marmatite is currently poor, at only 48.01%. To improve the comprehensive utilization rate of indium resources, it will be necessary to further increase the recovery rate from marmatite and explore the flotation recovery of indium from chalcopyrite and pyrrhotite.

1. Introduction

Indium has good ductility and plasticity, a low melting point, high boiling point, low resistance, and good corrosion resistance [1,2]. It is extensively used in electronics, semiconductor, and aerospace industries, thereby playing a critical role in national security and economic development [3,4]. The rapid expansion of information technology and new energy industries has driven substantial growth in global indium consumption [5]. Consequently, the price of indium metal (≥99.9% purity) has risen from 200 to 400 US$/kg, and it is expected that the global supply of primary indium will reach 4000–7700 t/year by 2050 [6]. Faced with escalating demand, nations are increasingly prioritizing indium resource production and recycling strategies.

The average content of indium in Earth’s crust is 0.056 ppm [7], with a relatively dispersed distribution; no rich ore bodies have been identified to date. Such a rare and scattered status quo poses a great challenge to the beneficiation and recovery of indium resources. To date, a total of 19 indium-bearing mineral species have been identified globally, of which five have been reported from China; the most notable representatives are roquesite (CuInS₂), indite (FeInS₄), sakuraiite ((Cu, Zn, Fe)₃(In, Sn)S₄), and dzhalindite [In (OH)₃] [8]. Indium occurs most commonly as a trace element in sphalerite (indium content of 0.0001%–0.1%), associated with Cu, Fe, and Sn sulfides [9,10]. Indium is primarily obtained through two routes. (1) Extraction from ores via flotation enrichment of indium-containing minerals such as sphalerite and chalcopyrite; for example, enrichment can be achieved by using xanthate or ethonium as the collector and X-1 as the activator [11]. (2) Recovery from secondary sources such as discarded liquid crystal display panels and indium tin oxide industrial waste [3,12].

Indium-rich countries include China, Peru, the United States, Canada, and Russia, which collectively account for 80.6% of global indium reserves. Notably, China holds the largest indium reserves (72.7% of the global total) and has served as the primary producer and consumer of indium, contributing 46% of global primary indium production between 2000 and 2018 [13,14]. Indium resources in China are mainly concentrated in Yunnan, Guangxi, Inner Mongolia, and Hunan (Table 1). Indium reserves in the Dulong deposit amount to approximately 7000 tons, making it the most important indium production base in China. The mineral genesis, enrichment characteristics, and substitution mechanism of indium in this region have been research hotspots for some time. Previous studies have shown that the Dulong deposit extends approximately north–south, with an approximate area of 10 km² (Figure 1). From north to south, it comprises the Tongjie, Manjiazhai, Lazizhai, Jinshipo, and other minor ore segments. The deposits are cassiterite sulfide ores with relatively high mineralization temperatures; indium has a very high correlation with copper, tin, iron, and cadmium [15,16,17]. All ore bodies are hosted by Cambrian marble and quartz mica schist, and are controlled by N–S-trending faults. The Laojunshan granite, predominantly comprising two-mica monzogranite, is mainly exposed in the northern part of the ore field.

Table 1. Types, distribution, and reserves of typical large-scale indium-bearing ores in China.

Mine	Vein Type	Carrier Minerals	Grade/ppm	Reserves/t	References
Inner Mongolia, dajing	tin-polymetallic veins	copper and zinc sulfides and sulfosalts	0.5–296	768	[18,19]
Inner Mongolia, Meng’entaolegai	quartz-sulfide veins	sphalerite	9–295	>500	[20]
Guangxi dachang	Volcanic polymetallic massive sulfides	Marmatite	0.2–673	>5000	[21]
Yunnan dulong	Sn-Zn polymetallic ore	Marmatite and chalcopyrite	0.74–4572	7000	[8]
Yunnan gejiu	Zn-enriched sulfide ores	Sphalerite and chalcopyrite	198.3–1570	>4000	[15]
Hunan qibaoshan	quartz-sulfide ores (a Sn-Poor Skarn Deposit)	sphalerite-pyrite ores	28.9–203	585	[22]
Hunan yejiwei	skarn-type Cu–Sn ores	sphalerite, chalcopyrite, and stannite	2.3–214	>806	[23]
Fujian zhongjia	polymetallic sulfide-magnetite ore	sphalerite, Galena and Magnetite	22–1346	569	[24]

Table 1. Types, distribution, and reserves of typical large-scale indium-bearing ores in China.

Mine	Vein Type	Carrier Minerals	Grade/ppm	Reserves/t	References
Inner Mongolia, dajing	tin-polymetallic veins	copper and zinc sulfides and sulfosalts	0.5–296	768	[18,19]
Inner Mongolia, Meng’entaolegai	quartz-sulfide veins	sphalerite	9–295	>500	[20]
Guangxi dachang	Volcanic polymetallic massive sulfides	Marmatite	0.2–673	>5000	[21]
Yunnan dulong	Sn-Zn polymetallic ore	Marmatite and chalcopyrite	0.74–4572	7000	[8]
Yunnan gejiu	Zn-enriched sulfide ores	Sphalerite and chalcopyrite	198.3–1570	>4000	[15]
Hunan qibaoshan	quartz-sulfide ores (a Sn-Poor Skarn Deposit)	sphalerite-pyrite ores	28.9–203	585	[22]
Hunan yejiwei	skarn-type Cu–Sn ores	sphalerite, chalcopyrite, and stannite	2.3–214	>806	[23]
Fujian zhongjia	polymetallic sulfide-magnetite ore	sphalerite, Galena and Magnetite	22–1346	569	[24]

Figure 1. (a) Geological map of the Dulong polymetallic deposit. (b) Geological cross-section along exploration line 125 (quoted from [25]).

Figure 1. (a) Geological map of the Dulong polymetallic deposit. (b) Geological cross-section along exploration line 125 (quoted from [25]).

However, most previous studies have been concentrated in professional fields such as geology and mineralogy [26,27,28]. There are very few studies in the field of mineral processing related to the enrichment and recovery of indium resources; this has resulted in insufficient understanding of the textural characteristics and substitution mechanism of indium in various minerals and its enrichment characteristics during the beneficiation process, leading to low recovery rates and high losses of indium resources. Therefore, in this study, we clarified the embedding fineness, distribution patterns, and intergrowth of indium with the components of carrier minerals for indium-bearing polymetallic minerals of the Duolong concentrator. This was achieved using process mineralogy combined with laser ablation inductively coupled plasma mass spectrometry (LA–ICP–MS), plus high-resolution imagery and mineral phase identification via automatic software analysis. This provides a scientific basis for optimizing the beneficiation process of indium resources in complex minerals and fills a research gap related to indium carrier minerals in the field of mineral processing.

2. Materials and Methods

2.1. Sample Handling and Experimental Procedures

Ore samples were collected from the Manjiazhai ore segment of the Dulong deposit. Because the mine adopts a production method of mixed mining and mixed selection, the obtained ore samples were mixed raw ores mined from this ore segment. To ensure the representativity of the samples, 50 kg of ore samples was randomly selected from the stockpiling plant of the mine and crushed to −75 μm accounting for 57.15%, and then thoroughly mixed and sample compositing until the material became homogeneous. A 500 g portion of sample was split out and ground to −75 μm for X-ray fluorescence (XRF), chemical multi-elemental analysis (CMEA), and X-ray diffraction (XRD) analysis. A 2 kg portion of sample was used for single-mineral separation analysis, and 100 g of sample was used for mineral liberation analysis (MLA) and electron probe microanalysis (EPMA), the latter after being inlaid with resin. The sample preparation process is shown in Figure 2.

The preparation of single minerals, including marmatite, chalcopyrite, magnetite, pyrrhotite, arsenopyrite, and pyrite, was conducted through hand-picking, selective dissolution to remove impurities, and chemical analysis to ensure purity.

In the field of mineral processing technology, the three core indicators used to assess the processing efficiency and economic benefits of ores are grade, yield, and recovery. Hence, we carried out a statistical analysis of these mineral processing indicators for the Dulong mineral processing plant over a period of six months. Based on process mineralogy, the distribution pattern of indium in various mineral processing products can be clearly delineated.

2.2. Analytical Methods

2.2.1. Chemical Multi-Element Analysis

Chemical analyses of the bulk sample and all size fractions were performed via XRF and inductively coupled plasma optical emission spectroscopy (ICP–OES), with XRF performed using the fused glass slice method (AxiosMAX; Panalytical, Almelo, The Netherlands). Major and trace element concentrations were determined via ICP–OES using a Thermo Scientific iCAP 7000 Series instrument from Massachusetts, USA. Prior to analysis, 0.1 g of finely ground sample material was accurately weighed and subjected to acid digestion using a mixture of HNO₃, HF, and HClO₄ in a closed Teflon vessel. The digested solution was then diluted with ultrapure water to a fixed volume.

2.2.2. Mineral Characterization

Mineral liberation analysis was performed via scanning electron microscopy combined with energy-dispersive X-ray spectroscopy (SEM–EDS; Quanta 650 and MLA650 instruments from Oregon, Hillsboro, OR, USA), X-ray microanalysis (EDAX Apollo X from Mahwah, NJ, USA), and automated process mineralogy testing software (JKTech Pty Ltd, Indooroopilly, QLD, Australia). Backscattered electron (BSE) imaging, standard X-ray spectroscopy, XRD, surface scanning X-ray spectroscopy, and mineralogical composition analysis were used in combination with image analysis techniques for data computation and processing to obtain mineralogical parameters including composition, elemental valence and state of occurrence, dissociation, and embedding of the ore, and images of typical mineral particles [29]. Working conditions of MLA were an accelerating voltage of 25 kV, beam current of 40 μA, and beam diameter of 6.5 µm.

2.2.3. Electron Probe Microanalysis

High resolution X-ray elemental intensity mapping was conducted at the laboratory of Guangzhou Tuoyan Analytical Technology Co., Ltd., Guangzhou, China, using a JEOL JXA-iSP100 electron probe microanalyzer equipped with five wavelength-dispersive spectrometers. The operating conditions for X-ray mapping were an accelerating voltage of 15 kV, beam current of 100 nA, step size of 1 μm, and dwell time of 30 ms. A total of five EPMA mapping analyses were carried out for marmatite.

2.2.4. Laser Ablation Inductively Coupled Plasma Mass Spectrometry

Trace elements within marmatite were determined using an NWR 193 nm ArF Excimer laser ablation system coupled to an iCAP RQ (ICP–MS) at the Guangzhou Tuoyan Analytical Technology Co., Ltd., China. The ICP–MS was tuned using NIST 610 standard glass sample to yield low oxide production rates. The carrier gas was fed into the cup at a rate of 0.7 L/min, and the aerosol was subsequently mixed with 0.89 L/min Ar gas. The laser fluence was 3.5 J/cm², with a repetition rate of 6 Hz, spot size of 30 μm, and analysis time of 45 s, followed by a 40 s background measurement. Trace element compositions were calibrated against various reference materials (NIST 610 and MASS-1) without using an internal standard [30]. Sulfide pellet standards SRM 612 and BCR-2Ga were used as unknown samples to evaluate data quality [31]. Raw data were reduced using the 3D trace element data reduction scheme [32] within the IOLITE V4 software package. In IOLITE, user-defined time intervals were established for the baseline correction procedure to calculate session-wide baseline-corrected values for each element. A total of 27 LA–ICP–MS spot analyses and five mapping analyses were carried out for marmatite.

3. Results and Discussion

3.1. Elemental and Mineralogical Analysis

3.1.1. Ore Major Element Composition

Semi-quantitative elemental analysis of the ore was carried out via XRF, with the results shown in

Table 2. X-ray fluorescence spectroscopy results.

Element	Fe	Si	Ca	Mg	Al	Zn	S	K	F	As	Mn
Content/%	11.00	10.00	7.00	5.00	4.00	2.00	2.00	0.80	0.80	0.30	0.30
Element	Cu	Sn	Ti	P	Rb	Cl	Bi	Sr	Zr	Ga	Y
Content/%	0.20	0.10	0.10	0.03	0.02	0.008	0.005	0.005	0.005	0.002	0.002

. The main elements comprising the ore were Fe, Si, Ca, Mg, Al, Zn, and S, with small amounts of K, F, As, Mn, Cu, Sn, and Ti.

Chemical multi-element analysis of the ore was based on the semi-quantitative XRF results (

Table 3. Results of chemical multi-element analysis.

Compound	In *	SiO2	Fe2O3	CaO	Al2O3	MgO	ZnS	CuS	SnO2	K2O	MnO
Content/%	43.87	28.20	17.97	11.10	7.49	7.26	4.04	0.23	0.22	1.36	0.39

* In is given in g/t.

). The indium content of the ore was 43.87 g/t, and the other main components were SiO₂ (28.20%), Fe (17.97%), CaO (11.10%), Al₂O₃ (7.49%), MgO (7.26%), S (5.01%), Zn (2.83%), and K₂O (1.36%), with small amounts of As, F, Mn, Cu, and Sn. According to Chinese national standard GB/T 25283-2010, the Zn and Sn contents in the ore met the minimum industrial grade for Zn and Sn deposits, and the Cu and In contents met the evaluation standard for associated components in the ore.

3.1.2. Ore Mineral Composition

Mineral particles were analyzed via MLA to determine the mineralogical composition of the ore, with the results shown in

Table 4. Mineral composition and abundance.

Serial Number	Mineral	Molecular Formula	Content/%	Serial Number	Mineral	Molecular Formula	Content/%
1	Marmatite	(Zn, Fe) S	5.14	16	Biotite	K{(Mg < 2\3, Fe > 1\3)3[AlSi3O10](OH)2}	3.77
2	Pyrrhotite	Fe1-xS	6.12	17	Chlorite	(Mg, Fe, Al)3(OH)6{(Mg, Fe, Al)3[(Si, Al)6O10(OH)2]}	11.47
3	Chalcopyrite	CuFeS2	0.66	18	Tremolite	Ca2Mg5[Si4O11]2(OH)2	1.63
4	Pyrite	FeS2	1.02	19	Sanidine	K[AlSi3O8]	3.43
5	Arsenopyrite	FeAsS	0.98	20	Andradite	Ca3Fe2[SiO4]3	5.42
6	Galena	PbS	0.02	21	Epidote	Ca2FeAl2[SiO4][Si2O7]O(OH)	2.36
7	Magnetite	Fe3O4	11.01	22	Fluorite	CaF2	1.44
8	Cassiterite	SnO2	0.28	23	Ferroactin-olite	Ca2(MgFe)5[Si4O11]2(OH)2	0.53
9	Calcite	Ca[CO3]	11.02	24	Antigorite	Mg6[Si4O10](OH)8	0.37
10	Dolomite	CaMg[CO3]2	9.07	25	Albite	Na[AlSi3O8]	0.27
11	Quartz	SiO2	11.12	26	Sphene	CaTiSiO5	0.25
12	talc	Mg3[Si4O10](OH)2	3.30	27	Zircon	Zr[SiO4]	0.05
13	Minnesotaite	Fe3[Si4O10](OH)2	1.96	28	Ilvaite	CaFe3(SiO4)2OH	0.46
14	Muscovite	K{Al2[AlSi3O10](OH)2}	5.21	29	Apatite	Ca5(PO4)3(F, Cl, OH)	0.15
15	Hlogopite	K{(Mg > 2\3, Fe < 1/3)3[AlSi3O10](OH)2}	1.39	30	Rutile	TiO2	0.10

and Figure 3 and Figure 4.

The ore consisted of 30 different minerals, with the principal metallic minerals being magnetite (11.01%), pyrrhotite (6.12%), and marmatite (5.14%). The ore also contained trace amounts of chalcopyrite (0.66%), pyrite (1.02%), and arsenopyrite (0.98%). The oxide minerals were predominantly calcite (11.02%), dolomite (9.07%), quartz (11.12%), chlorite (11.47%), talc (5.26%), and mica (10.37%).

3.1.3. Grain Size Distribution of Major Minerals

The grain size distributions of the main minerals in the ore were measured via MLA, using the raw ore sample as the source material. The results are shown in

Table 5. Particle size characteristics of raw ore and main metallic minerals.

Particle Size (μm)	Raw Ore		Marmatite		Pyrrhotite		Magnetite		Chalcopyrite		Arsenopyrite		Pyrite
	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)	Content (%)	Cumulative (%)
+300	1.33	1.33	—	—	—	—	—	—	—	—	—	—	—	—
−300 + 212	5.26	6.58	—	—	—	—	—	—	—	—	—	—	—	—
−212 + 150	10.50	17.08	—	—	2.00	2.00	2.18	2.18	—	—	—	—	—	—
−150 + 106	12.41	29.49	6.25	6.25	2.17	4.17	4.82	7.00	7.06	7.06	10.73	10.73	4.35	4.35
−106 + 75	13.36	42.85	11.53	17.78	14.80	18.97	8.95	15.95	15.60	22.66	7.13	17.86	16.33	20.69
−75 + 53	13.50	56.35	16.47	34.25	15.86	34.83	16.84	32.79	22.28	44.94	9.86	27.72	13.47	34.16
−53 + 38	11.88	68.23	15.31	49.57	14.92	49.75	16.68	49.47	6.96	51.90	12.54	40.26	19.87	54.03
−38 + 27	10.54	78.77	16.18	65.75	14.50	64.26	17.01	66.48	15.37	67.26	17.36	57.62	15.14	69.17
−27 + 19	7.96	86.72	12.61	78.36	12.71	76.97	12.70	79.18	9.14	76.40	16.41	74.03	11.46	80.63
−19 + 13.5	5.39	92.11	8.79	87.15	9.39	86.36	9.14	88.32	8.48	84.88	10.76	84.79	8.87	89.50
−13.5 + 9.6	3.67	95.78	5.36	92.51	6.55	92.91	6.03	94.34	6.59	91.46	6.50	91.29	4.19	93.68
−9.6 + 6.8	2.42	98.20	4.09	96.61	4.06	96.97	3.54	97.88	4.65	96.11	4.93	96.23	3.54	97.22
−6.8 + 4.8	1.44	99.64	2.59	99.19	2.33	99.31	1.65	99.53	2.93	99.04	2.99	99.21	2.05	99.28
−4.8 + 3.4	0.34	99.97	0.73	99.93	0.61	99.91	0.41	99.94	0.86	99.90	0.69	99.90	0.63	99.91
−3.4	0.03	100.00	0.07	100.0	0.09	100.0	0.06	100.0	0.10	100.0	0.10	100.0	0.09	100.0

and

Table 6. Particle size characteristics of main oxide minerals.

Particle Size (μm)	Quartz		Muscovite		Chlorite		Calcite Garnet		Dolomite		Calcite
	Cont-Ent (%)	Cumulative (%)	Cont-Ent (%)	Cumulative (%)	Cont-Ent (%)	Cumulative (%)	Cont-Ent (%)	Cumulative (%)	Cont-Ent (%)	Cumulative (%)	Cont-Ent (%)	Cumulative (%)
+300	5.70	5.70	—	—	—	—	—	—	—	—	—	—
−300 + 212	11.94	17.64	—	—	1.35	1.35	4.49	4.49	1.74	1.74	6.74	6.74
−212 + 150	13.85	31.49	5.76	5.76	11.91	13.26	10.23	14.72	8.95	10.69	12.55	19.29
−150 + 106	13.09	44.58	12.71	18.46	10.74	24.00	13.98	28.70	11.37	22.06	14.72	34.02
−106 + 75	11.85	56.42	12.79	31.26	15.02	39.01	13.16	41.87	13.32	35.38	13.06	47.08
−75 + 53	14.95	71.37	12.69	43.95	12.71	51.72	10.78	52.64	19.18	54.57	12.90	59.98
−53 + 38	9.81	81.18	11.26	55.21	11.17	62.89	12.67	65.32	14.50	69.06	11.86	71.84
−38 + 27	7.34	88.52	12.46	67.67	10.75	73.65	11.53	76.85	12.12	81.18	9.73	81.56
−27 + 19	4.76	93.28	11.28	78.95	9.17	82.81	7.73	84.58	8.56	89.74	6.66	88.22
−19 + 13.5	3.26	96.54	8.14	87.09	6.45	89.26	5.58	90.16	4.62	94.37	4.66	92.88
−13.5 + 9.6	1.78	98.33	6.12	93.21	4.70	93.96	4.25	94.41	2.70	97.07	3.29	96.18
−9.6 + 6.8	1.02	99.34	3.76	96.97	3.27	97.23	3.15	97.56	1.68	98.75	2.22	98.40
−6.8 + 4.8	0.53	99.87	2.33	99.30	2.14	99.38	1.82	99.38	0.99	99.74	1.28	99.68
−4.8 + 3.4	0.12	99.99	0.60	99.90	0.55	99.93	0.54	99.92	0.24	99.98	0.30	99.98
−3.4	0.01	100.00	0.10	100.00	0.07	100.00	0.08	100.00	0.02	100.00	0.02	100.00

. The cumulative distribution in the raw ore was as follows: +300 μm particles, 1.33%; +150 μm particles, 17.08%; +75 μm particles, 42.85%; +38 μm particles, 68.23%; +19 μm particles, 86.72%; and +9.6 μm particles, 95.78%. The main metallic minerals were dominantly fine-grained, except for chalcopyrite (+75 μm cumulative distribution of 22.66%); the other metallic minerals were in the range of 16%–20%. The main oxide minerals were dominantly coarse-grained, except for calcite (+75 μm cumulative distribution of 47.08%); the other oxide minerals were in the range of 31%–41%. On this basis, it would be difficult to separate chalcopyrite, marmatite, pyrite, and other metallic minerals because their fine-grained nature is inconducive to flotation separation.

3.2. Content and Distribution of Indium in Carrier Minerals

The main indium-bearing single minerals are shown in Figure 5; these included marmatite, chalcopyrite, magnetite, pyrrhotite, arsenopyrite, pyrite, silicate minerals, carbonate minerals, quartz, and other gangue minerals [33]. Indium was enriched in marmatite (543 g/t) and chalcopyrite (521 g/t), with lower levels in pyrrhotite (86.33 g/t), arsenopyrite and pyrite (22.35 g/t), and silicate minerals (24.45g/t) (

Table 7. Indium content in single minerals.

Mineral	Marmatite	Chalcopyrite	Pyrrhotite	Arsenopyrite and Pyrite
Content of In (g/t)	543.00	521.00	86.33	22.35
Mineral	Silicate Minerals	Carbonate Minerals	Quartz and Other Veinstone	Magnetite
Content of In (g/t)	24.45	0.00	0.00	0.00

). Magnetite, carbonate minerals, quartz, and other gangue minerals did not contain indium.

The distribution of indium across the respective host minerals is shown in

Table 8. Distribution of indium in different minerals.

Mineral	Content/%	In Grade of Carrier Minerals (g/t)	Distribution of In in Carrier Minerals/%
Marmatite	5.14	543.00	63.62
Chalcopyrite	0.66	521.00	7.84
Magnetite	11.01	0.00	0.00
Pyrrhotite	6.12	86.33	4.22
Arsenopyrite, pyrite	2.00	22.35	1.02
Silicate minerals	41.82	24.45	23.30
Quartz	11.12	0.00	0.00
Carbonates	20.09	0.00	0.00
Other	2.70	0.00	0.00
Raw mineral	100.00	43.87	100.00

. Indium mainly occurred in marmatite (63.62%) supplemented by that in silicate minerals (23.31%), chalcopyrite (7.84%), pyrrhotite (4.22%), and arsenopyrite and pyrite (1.02%). The formula for distribution is

$$\mathrm{distribution}={\displaystyle \frac{\gamma ·\beta }{100·\alpha }}\times 100\mathrm{\%}$$

(1)

where $\gamma$ is the mineral content (%), $\beta$ is the In grade of the mineral (g/t), and $\alpha$ is the In grade of the raw mineral.

3.3. Textural Characteristics and Substitution Mechanism of Indium-Bearing Marmatite

3.3.1. Textural Characteristics of Indium-Bearing Marmatite

Marmatite was found to be the main carrier mineral of indium; hence, we conducted a systematic investigation into the occurrence state and enrichment mechanism of indium within marmatite. Our EDS analysis (Figure 6 and

Table 9. Energy-dispersive X-ray spectroscopy analysis of marmatite.

Number	S/%	Fe/%	Zn/%	Number	S/%	Fe/%	Zn/%
1	34.3	12.6	53.1	12	33.8	12.1	54.1
2	31.8	10.9	57.3	13	34.1	11.1	54.8
3	33.9	10.7	55.4	14	33.8	12.4	53.9
4	33.8	11.9	54.2	15	34.2	10.9	54.9
5	34.7	9.5	55.8	16	34.4	10.3	55.3
6	34.2	9.0	56.8	17	34.2	9.6	56.2
7	33.3	6.5	60.2	18	34.5	7.8	57.6
8	35.8	13.4	50.8	19	32.9	11.1	56.1
9	34.0	11.7	54.3	20	34.7	13.7	51.7
10	36.2	11.4	52.5	21	33.8	10.1	56.1
11	33.9	11.1	55.0	average value	34.1	10.8	55.1

) showed that the marmatite contained 55.1% Zn, 10.8% Fe, and 34.1% S, on average. Combined with MLA (Figure 7) and SEM–BSE (Figure 8) data, the textural characteristics of marmatite can be summarized as follows: some marmatite occurs as single dissociated particles (Figure 8a); some marmatite has a close intergrowth relationship with metallic minerals such as pyrrhotite, chalcopyrite, pyrite, and magnetite, with the intergrowth mainly consisting of inlaid distribution and adjacent intergrowth (Figure 8c–e); some marmatite exhibits intergrowth with oxide minerals (calcite, dolomite, quartz, chlorite, and biotite), with the intergrowth characterized by wrapping in an oxide mineral or being inlaid with other minerals (Figure 8b,f–i); and the grain size distribution of marmatite has a wide range, primarily distributed between 0.005 and 0.15 mm (Figure 7).

3.3.2. Substitution Mechanism of Indium in Marmatite

To further clarify the substitution mechanism of indium in marmatite from the Dulong deposit, a 300 × 300 μm area of the surface of a marmatite grain was subjected to EPMA, with a focus on analyzing the correlation between indium and copper, iron, zinc and sulfur. We found a positive correlation between indium and copper (Figure 9b,c), consistent with the findings of an existing study [34]. However, three elongated bright zones (Figure 9a) exhibited marked enrichment in copper, indium, and sulfur, with negligible iron and zinc content (Figure 9b–f); these are interpreted as roquesite (CuInS₂). This constitutes the first documented observation of an independent indium mineral species within the Dulong mine.

Laser ablation inductively coupled plasma mass spectrometry is a high-efficiency analytical technique capable of simultaneous in situ measurement of precious metals and critical trace elements in mineral phases, while achieving low detection limits [27,35,36]. Here, we used LA–ICP–MS to obtain higher-precision data in relation to indium-bearing marmatite (Figure 10). These data revealed that Cu and In contents in the marmatite ranged from 100 to 700 ppm, exhibiting a strong positive correlation (correlation coefficient = 0.9869) with a slope of 1.04 (Figure 10b). This is consistent with the coupled substitution mechanism 2 Zn²⁺ $\boxed{\leftrightarrow }$ ∂ Cu⁺ + In³⁺ [26,35,37,38]. Additionally, Fe and In exhibited a moderate positive correlation (Figure 10d; correlation coefficient = 0.7691), suggesting that the presence of Fe facilitates In substitution, whereas S did not correlate with In (Figure 10c). Further elemental mapping visually corroborated the positive correlation between Cu and In (Figure 11 and Figure 12). According to the relevant literature [17], X-ray photoelectron spectroscopy analysis indicates that the valence states of Zn, S, Fe, and In in indium-rich sphalerite of the Dulong deposit are Zn²⁺, S²⁻, Fe²⁺, and In³⁺, respectively. This, in conjunction with the findings of our study, means that the substitution formula for indium in marmatite can be revised to

$${\mathrm{Z}\mathrm{n}}_x\mathrm{S}+{\mathrm{C}\mathrm{u}}^{+}+{\mathrm{I}\mathrm{n}}^{3+}\to {\mathrm{Z}\mathrm{n}}_{x-2}\mathrm{C}\mathrm{u}\mathrm{I}\mathrm{n}\mathrm{S}+2{\mathrm{Z}\mathrm{n}}^{2+}$$

(2)

or

$${\mathrm{Z}\mathrm{n}}_{x-1}\mathrm{F}\mathrm{e}\mathrm{S}+{\mathrm{C}\mathrm{u}}^{+}+{\mathrm{I}\mathrm{n}}^{3+}\to {\mathrm{Z}\mathrm{n}}_{x-2}\mathrm{C}\mathrm{u}\mathrm{I}\mathrm{n}\mathrm{S}+{\mathrm{Z}\mathrm{n}}^{2+}+{\mathrm{F}\mathrm{e}}^{2+}$$

(3)

3.4. Enrichment Regularity of Indium in the Dulong Mineral Processing Plant

In the Dulong mineral processing plant, indium, as an important by-product of zinc concentrate, has considerable economic value [11,39]. However, because the ore is a complex polymetallic deposit, the recovery rate of indium in the beneficiation process is low, and the recovery methods are single. Above, we have clarified the enrichment characteristics and substitution mechanisms of indium, but important indicators, such as the yield, grade, and recovery rate of the beneficiation products, remain unclear. Hence, the beneficiation process of the Dulong beneficiation plant must be investigated.

We statistically analyzed the beneficiation indicators of five concentrates products, one middlings products, and one tailings product (Figure 13). The main beneficiation product minerals are listed in

Table 10. Main minerals in different mineral processing products.

Product	Main Mineral
Zinc Concentrate	Marmatite
Copper Concentrate	Chalcopyrite
Iron Sulfur Concentrate	Pyrrhotite
Tin Concentrate	Cassiterite
Iron Concentrate	Magnetite
Tin Middling	cassiterite
Tailings	Carbonates, Magnetite Silicate minerals, Quartz

, and the distributions of indium and other major elements in the beneficiation products are presented in

Table 11. Distribution of indium and other major elements in beneficiation products.

Indicator	Element	Products
		Zinc Concentrate	Copper Concentrate	Tin Concentrate	Tin Middling	Iron Sulfur Concentrate	Iron Concentrate	Tailings	Raw Ore
Grade/%	Zn	46.73	6.97	0.08	0.11	0.22	0.05	0.18	2.58
	Cu	0.73	16.58	0.02	0.03	0.05	0.01	0.02	0.18
	Sn	0.07	0.14	38.22	2.60	0.17	0.16	0.12	0.23
	Fe	13.93	20.47	12.93	16.99	60.25	66.22	11.62	17.26
	S	33.01	24.68	3.93	5.15	9.62	2.33	2.07	4.61
	In	483.10	319.42	53.65	47.61	28.31	34.92	25.35	51.11
	Ag	60.17	771.08	3.25	4.87	8.59	3.47	3.57	12.59
	As	0.30	0.21	0.17	0.38	0.06	0.01	0.29	0.27
	MgO	1.04	5.76	1.55	5.34	2.11	1.24	7.15	6.33
Recovery/%	Zn	92.74	1.99	0.01	0.01	0.84	0.02	5.91	100.00
	Cu	20.78	67.56	0.03	0.06	2.62	0.04	11.39	100.00
	Sn	1.67	0.47	43.80	3.72	7.20	0.74	43.99	100.00
	Fe	4.12	0.87	0.20	0.32	33.59	3.98	55.78	100.00
	S	36.62	3.94	0.22	0.37	20.08	0.52	37.24	100.00
	In	48.01	4.57	0.28	0.30	5.30	0.71	40.83	100.00
	Ag	24.44	45.14	0.07	0.13	6.57	0.29	23.49	100.00
	As	5.61	0.58	0.16	0.46	2.05	0.03	88.35	100.00
	MgO	0.84	0.67	0.06	0.28	3.21	0.20	93.69	100.00
Yield/%	-	5.11	0.74	0.26	0.33	9.62	1.04	82.90	100.00

Grade unit: “%” for Zn, Cu, Sn, Fe, S, Ag, As and MgO, “ppm” for In.

. The zinc concentrate had the highest indium recovery rate (48.01%), and this was the only recoverable indium in production. Further enhancement of the recovery rate of the zinc concentrate could concurrently increase the recovery rate of indium. The recovery rate of indium in the tailings was as high as 40.83%; this portion of indium had the lowest grade at 25.35 g/t, and flotation recovery was particularly challenging, with almost no economic value [40]. The copper and iron–sulfur concentrates had relatively high indium grades and recovery rates (total indium distribution of the two concentrates was 9.87%), consistent with the results shown in Table 8. Considering the difficulty of flotation, these indium-bearing sulfide minerals are easier to concentrate and possess higher recovery potential. The efficient utilization of multiple carrier minerals will be crucial in enhancing the overall utilization rate of indium resources during beneficiation and smelting processes.

4. Conclusions

The Dulong ore comprises over 30 minerals, within which the average grade of indium is 43.87 g/t. In addition to its enrichment in marmatite through isomorphic substitution, indium occurs within the independent indium mineral roquesite (CuInS₂). The distribution rate of indium in carrier minerals is ordered as follows: marmatite > silicate minerals > chalcopyrite > pyrrhotite > arsenopyrite and pyrite.

We studied the substitution mechanism of indium in its main carrier mineral, marmatite, and found that the substitution of indium correlated positively with copper. This allowed us to determine a revised substitution formula: ${\mathrm{Z}\mathrm{n}}_x\mathrm{S}+{\mathrm{C}\mathrm{u}}^{+}+{\mathrm{I}\mathrm{n}}^{3+}\to {Zn}_{x-2}\mathrm{C}\mathrm{u}\mathrm{I}\mathrm{n}\mathrm{S}+2{\mathrm{Z}\mathrm{n}}^{2+}$ or ${\mathrm{Z}\mathrm{n}}_{x-1}\mathrm{F}\mathrm{e}\mathrm{S}+{\mathrm{C}\mathrm{u}}^{+}+{\mathrm{I}\mathrm{n}}^{3+}\to {Zn}_{x-2}\mathrm{C}\mathrm{u}\mathrm{I}\mathrm{n}\mathrm{S}+{\mathrm{Z}\mathrm{n}}^{2+}+{\mathrm{F}\mathrm{e}}^{2+}$.

Our investigation of process mineralogy and the beneficiation process at Dulong revealed that zinc concentrate is the most important carrier ore of indium, and flotation research in relation to this ore should be further strengthened. Meanwhile, copper and iron–sulfur concentrates contain high-grade indium, with a combined share of 9.87%, and thus have great recovery potential. These should be regarded as future foci for improving the recovery rate of indium resources.