An Integrated Separation Approach for Beneficiation of Low-Grade Iron Ore

Abstract

Iron ore beneficiation is crucial for sustainable utilization of low-grade iron ores. Conventional separation methods often fail to efficiently upgrade the iron ores containing complex mineral phases, such as hematite, goethite, ankerite, and limonite. This study evaluated an integrated roasting–leaching–magnetic separation approach applied to an iron ore. Roasting at 800 °C for 90 min significantly enhanced the Fe(II) content and improved magnetic susceptibility, facilitating superior iron-containing minerals separation efficiency. Furthermore, sulfuric acid leaching effectively eliminated non-magnetic impurities such as calcite, improving iron recovery. Davis tube recovery experiments confirmed that this combination markedly enhanced weight recovery (R_w) and iron recovery (R_g), outperforming traditional methods. The findings highlight the synergistic effect of roasting and leaching for refining iron ores and propose an optimized beneficiation strategy. This approach offers an effective method for processing complex iron ores, particularly those with low Fe grade and challenging impurities, improving beneficiation efficiency and mineral extraction.

1. Introduction

Considering the importance of iron metal in the current world, the processing of iron ore is of great significance. Therefore, this topic has been continuously studied for technological development and advancement. One of the most crucial parameters for evaluating the application of iron ore and its processing methods is the iron grade and type of iron-bearing minerals. In some cases, conventional iron enrichment methods such as magnetic [1,2,3,4,5,6,7] and gravity [6,7,8,9,10,11,12] separation alone are not sufficient to upgrade the iron grade. For example, in a study conducted by Ostadrahimi et al. [7] on an iron ore sample from Chah Bacheh (Yazd, Iran), increasing the iron grade using gravity and magnetic methods was not feasible, which was attributed to the presence of multiple iron-bearing minerals such as hematite, goethite, ankerite, and limonite in the sample. Therefore, before processing these types of iron ores to produce marketable iron ore products, it is necessary to modify the nature of the iron minerals. The roasting process is one of the practical methods for this purpose. In this method, iron ores that cannot be upgraded through magnetic separation undergo roasting to alter their mineral composition and enhance their magnetic properties. For example, hematite transforms into magnetite during the roasting process, increasing its magnetic susceptibility [13].

3Fe₂O₃ + CO → 2Fe₃O₄ + CO₂

Various materials such as coal [14,15,16,17,18] and hydrogen gas [18,19,20,21] are often used during the roasting process. For example, Yu et al. [15] used coal to roast iron ore at 800 °C before a magnetic separator, and they achieved a weight recovery of about 65%. In another study, powdered coal with a size of 210 μm was used for selective roasting and conversion of hematite to magnetite [17]. According to the industrial production data, energy consumption in the roasting process is less than 35 kg of coal per ton of raw ore [16]. Additionally, Zhu et al. [20] proposed a new hydrogen-based reduction roasting technology for processing iron ore tailings. They found that under a gas composition of CO:H₂ of 1:3 and the optimal roasting conditions at a reduction temperature of 520 °C, most of the weakly magnetic hematite was converted into strong magnetite during the reduction process [20].

Although application of the roasting process is common for minerals such as hematite, goethite, and siderite, research has been conducted for other iron ores such as limonite ore [21], banded quartzite iron ore [22], and high-phosphorus iron ores [23]. The fast reductive roasting (FRR) technique is another method of roasting iron ore. In this method, weakly magnetic materials with a grain size <0.30 mm are converted to strong magnetic materials by FRR for several to tens of seconds [24]. Recently, a method for microwave fluidized magnetic roasting based on hydrogen has been proposed for selective and rapid heating, providing a high efficiency in the heat and mass transfer processes [25].

Hydrometallurgical methods have been effectively utilized for the extraction of valuable elements associated with iron in iron ore tailings [26,27]. However, leaching has been applied in a limited capacity to remove undesirable impurities in the iron ore processing. This method is particularly useful for low-grade iron ores or those containing problematic impurities. In a study conducted by Xia and colleagues, hydrochloric acid was used to remove phosphorus from high-phosphorus iron ore [28]. In another study, the feasibility of combining fluidized reduction and a melt-separation process was studied [29]. In this study, the effect of acid leaching on the reducing properties of high-phosphorus iron ore, as well as the mechanism of phosphorus conversion and transport during the carbon-free melt separation process, was also investigated [29]. In a study conducted by Deng et al. [30], a combination of reductive roasting and alkaline leaching was applied to iron ore tailings. This approach enabled the separation of silica and alumina-bearing minerals, resulting in an increase in the iron grade of the concentrate from 60% to 65% [30].

In this study, a novel approach to iron ore processing, including integrating roasting, leaching, and magnetic separation, was investigated. Given the limitations of conventional enrichment methods, this research explores a combined process to enhance the magnetic properties of low-grade iron ores and remove undesirable impurities. The significance of this study lies in its alternative strategy for processing iron ores with complex and unconventional mineral compositions. Unlike previous studies which were focused on single beneficiation techniques, this research investigates the interaction between thermal and chemical processes, offering a novel perspective on optimizing iron ore processing that could be practically implemented in industrial applications.

2. Review of Previous Study

According to previous quantitative X-ray diffraction (QXRD) studies conducted on the Chah Basheh iron ore sample (Yazd, Iran), the iron-bearing minerals included hematite, goethite, ankerite and limonite. The most significant non-iron minerals in the sample were identified as pyrolusite, dolomite, calcite, and quartz.

Table 1. Specification and quantity of minerals in the sample [7].

No.	Mineral	Weight (%)	Formula	No.	Mineral	Weight (%)	Formula
1	Calcite	21	CaCO3	6	Ankerite	15	Ca (Fe,Mg)(CO3)2
2	Dolomite	15	CaMg(CO3)2	7	Goethite	14	FeO(OH)
3	Barite	5	BaSO4	8	Hematite	16	Fe2O3
4	Quartz	4	SiO2	9	Amorphous	3	Fe2O3(H2O)n
5	Clay mineral	1		10	Pyrolusite	5	MnO2

specifies the minerals and their quantities in the sample, showing that approximately 48% of the sample consists of iron-bearing minerals. Additionally, microscopic studies revealed that the iron-bearing minerals are composed of nodules ranging from 10 to 20 µm, formed predominantly in a calcareous sedimentary environment and were metamorphosed in later stages (Figure 1). It can be concluded that the sample is low-grade iron ore, and it contains manganese [7]. In addition, lime impurities are mainly associated with carbonate minerals such as calcite (CaCO₃) and dolomite (CaMg(CO₃)₂), which were identified in the QXRD analysis (Table 1). These phases are distributed in the gangue matrix and occasionally interlocked with iron-bearing nodules, explaining their effective removal by sulfuric acid leaching.

Gravity and magnetic separation tests were conducted on the sample, but the results were unsatisfactory. In fact, the gravity method worked oppositely, as the Fe grade in the concentrate and tailings was found to be 19% and 30%, respectively. Moreover, the weak magnetic properties of minerals such as ankerite and goethite were problematic in the magnetic separation, and the maximum Fe grade reached only 32% [7].

3. Materials and Methods

To evaluate the effect of roasting on enhancing the magnetic properties of iron ore samples from the Chah Basheh mine, re-sampling was conducted from two distinct operational fronts of the site (named as “L” and “H” in this work). The 50 kg samples of each stream were first crushed using a cone crusher and then ground in a ball mill, reducing their particle size to a maximum of 1 mm.

The Fe and Fe(II) contents of the samples were determined by titration according to the standard ASTM method [31].

The roasting process was conducted in a laboratory muffle furnace at a temperature of 800 °C for 90 min, employing coke at a 5% weight ratio to facilitate oxidation. The roasting temperature and duration (i.e., 800 °C for 90 min) were selected based on preliminary experiments and supported by previous studies, resulting in efficient conversion of hematite and goethite to magnetite under similar conditions. Yu et al. [15] achieved effective beneficiation of hematite at 800 °C, while Kukkala et al. [17] confirmed the suitability of this temperature for low-grade hematite fines. These reports, along with our experimental observations, justify the chosen condition.

To assess the magnetic absorption of the roasted samples, Davis tube recovery (DTR) equipment was utilized. The DTR experiments were conducted with a water flow rate of 0.5 L/min, an oscillatory movement speed of 60 rpm, a duration of 5 min, and magnetic field strengths of 1500 and 5000 G. In addition, a Slon high-gradient magnetic separator (Slon-100 cyclic model, Slon Magnetic Separator Ltd., Ganzhou, China) was employed to evaluate the effect of roasting on the magnetic properties of the samples. The operating conditions were set at a magnetic intensity of 10,000 G, a solids concentration of 20%, and a pulsation frequency of 200 rpm. To enhance the oxidation environment during the roasting process, sulfuric acid was applied at a volume ratio of 15%. In these experiments, the solid content was maintained at 30%, with a leaching time of 15 min. The solid phase was then separated from the solution using a vacuum filtration apparatus. In both magnetic separation methods, the feed material was ground to a particle size of less than 60 µm, and two products were collected: magnetic concentrate and non-magnetic tailings.

All experiments were repeated at least twice to ensure reproducibility. While formal statistical validation was not performed in this study, the consistency of repeated measurements supports the reliability of the reported trends.

To calculate the weight recovery (R_w), enrichment ratio (ER), and iron recovery (R_g), Equations (1), (2), and (3), were respectively used.

$$R_w={\displaystyle \frac{\mathrm{C}}{F}}\times 100$$

(1)

$$ER={\displaystyle \frac{\mathrm{c}}{f}}\times 100$$

(2)

$$R_g={\displaystyle \frac{C\times c}{F\times f}}\times 100$$

(3)

where F is the initial sample weight, C is the weight of the material absorbed by the DTR, f is the Fe grade of the initial sample, and c is the Fe grade of the DTR product.

The experiments were carried out at the Iran Minerals Research Centre and the Iranian Mineral Processing Research Centre. The experiment’s flow sheet is shown in Figure 2.

4. Results and Discussion

4.1. Sample Analysis

The chemical analysis results showed Fe grades of 17.9% and 27.4% for Samples L and H, respectively. In both cases, the Fe(II) content was found to be relatively low, i.e., 2.8% and 2.3% for samples L and H, respectively.

4.2. Roasting

Following the roasting process, the samples were weighed and analyzed. The results indicated an increase in Fe(II) grade for both samples, accompanied by a decrease in weight (

Table 2. Grade and weight ratio of samples after roasting.

Sample	Fe (%)	Fe(II) (%)	WR/W (%)
LR	23.0	4.3	83.2
LSR	25.2	12.5	71.2
HR	31.5	3.7	88.2
HSR	32.8	14.1	78.9

). For example, in the case of the LR sample, the Fe(II) grade increased by 154%, while its weight decreased by 83% compared to pre-roasting values. It also appears that as the Fe grade of the sample increases, the weight loss after roasting decreases. For example, comparing the HSR sample with the LSR, it is observed that the sample’s weight loss is approximately 10% less.

The reduction in the sample weight after roasting is primarily attributed to the loss of volatile compounds, the evaporation of moisture contained within the material structure, and chemical oxidation reactions. During the roasting process, minerals may decompose or transition into simpler oxides, resulting in a reduction of the overall weight of the sample. Additionally, using coal or coke as reducing agents facilitates thermal reactions that produce gases such as CO and CO₂, and it further contributes to the sample’s weight loss [32].

The increase in Fe(II) grade after roasting can be attributed to the reduction in higher oxidation state iron oxides, such as Fe(III) (hematite or goethite), to lower oxidation state oxides like Fe(II). This transformation occurs due to the thermal decomposition and partial reduction reactions facilitated by the presence of reducing agents such as, coke or coal, during the roasting process. These reactions enhance the proportion of Fe(II) within the roasted samples, thereby increasing its grade [33].

It is noteworthy that the amount of Fe(II) increases, and the weight loss of leached samples (LSR, HSR) was much greater compared to the non-leached samples (LR, HR). The average increase in the Fe(II) after roasting in the leached samples was 330% greater than in the non-leached samples. The observed weight reduction in leached samples (12.0% for L and 9.3% for H, compared with their roasted counterparts) can be attributed mainly to the dissolution of carbonate gangue minerals, particularly calcite (CaCO₃) and dolomite (CaMg(CO₃)₂), which were identified in the mineralogical study (Table 1). The increase in Fe grade in the solid residue indicates that iron dissolution into the leach liquor was negligible, confirming that sulfuric acid leaching selectively removed lime-bearing phases.

The sulfuric acid leaching process was applied to remove impurities such as lime from the iron ore sample. This will facilitate the liberation of iron minerals and enhance the magnetic properties of the sample. This process resulted in weight reduction due to the dissolution of non-iron impurities, while the iron content increased in the solid phase, indicating minimal iron dissolution into the solution. The notable rise in Fe(II) can be attributed to the acidic environment, which promotes the reduction of Fe(III) to Fe(II).

4.3. Magnetic Separation

To investigate the increase in the magnetic property of roasted samples and their absorption capacity, tests were conducted at magnetic intensities of 1500 and 5000 G using DTR. According to the results, the absorbed fraction was much higher for the leached samples, but its grade was significantly reduced (

Table 3. Magnetic absorption value of the roasted sample.

Sample	WDTR/WR (%)			Fe (%)
	1500 G	5000 G	10,000 G	1500 G	5000 G	10,000 G
LR	10.9	18.5	84.8	48.9	48.3	24.9
LSR	51.2	61.7	96.2	32.1	31.6	25.9
HR	21.6	44.4	88.8	51.0	47.6	34.4
HSR	50.9	66.4	91.6	42.5	40.4	33.8

). For example, for sample L, the magnetic absorption after leaching (LSR) increased by about 5 times compared to the non-leached sample (LR), but its grade decreased by about 17%. Of course, the intensity of the changes was less for sample H.

Since sulfuric acid dissolved some non-magnetic impurities (such as calcite) before roasting, the magnetic separation efficiency of iron-bearing minerals (present as micronodules in the sample) improved. However, due to incomplete liberation of iron-bearing minerals, some gangue minerals are also reported to the magnetic product. Moreover, in the LSR and HSR samples, the increased Fe(II) content enhanced the overall magnetic susceptibility [34], which likely indicates a partial transformation of hematite into magnetite [13]. As a result, the DTR was able to capture additional low-grade iron-bearing minerals (such as ankerite and limonite). This phenomenon decreased the iron grade of the concentrate but increased the overall recovered fraction.

It should be noted that the inference of incomplete liberation in this study was based on mineralogical characterization (Table 1, Figure 1) and recovery–grade behavior, but no direct SEM or QEMSCAN evidence was obtained. Future work will require microstructural analysis to quantitatively confirm the extent of the minerals liberation.

The absorbed fraction increased with increasing magnetic intensity, and the product iron grade decreased. For example, the absorbed fraction at a magnetic intensity of 1500 G for samples LR and HR was approximately 11% and 22%, respectively, while at an intensity of 5000 G, it increased to 18% and 42%, respectively. The Fe grade of the DTR product decreased with an increase in magnetic intensity, such that for sample HR, it showed a reduction of over 6% when transitioning from 1500 G to 5000 G. At a magnetic field intensity of 10,000 G, the majority of the roasted samples were transferred to the magnetic fraction. This high weight recovery, exceeding 80% in most cases, is attributed to the enhanced magnetic properties of the roasted phases, indicating that some gangue minerals were also recovered alongside the iron-bearing phases.

With the increase in magnetic field intensity, particles with weak magnetic properties are separated, leading to a decrease in the product’s grade [35].

4.4. Process Evaluation Parameters

Figure 3 illustrates the concentration ratio (CR, defined as the ratio of the Fe grade in the concentrate to the Fe grade in the feed), weight recovery (R_w), and iron recovery (R_g) of the product obtained from the Davis tube recovery, compared with the initial samples (L and H). According to the results, the leached samples (LSR and HSR) showed a lower CR but higher Rw and Rg compared to the non-leached samples (LR and HR). For example, at a magnetic intensity of 1000 G, the concentration ratio of the CR sample was about 52% higher than the LSR sample, while R_w and R_g were about 75% and 60% lower, respectively. The leaching process removed some impurities from the sample, resulting significant increase in the Fe(II) content of the sample during roasting. This resulted in a decrease in iron grade but an increase in the weight recovery.

Furthermore, the CR was higher in the low-grade samples compared to the high-grade ones. This phenomenon typically occurs in lower-grade ores due to the more pronounced effect of grade enhancement in the concentration ratio calculation. Additionally, with increasing magnetic intensity, the CR decreased while R_w increased, owing to the recovery of particles with lower iron content.

5. Conclusions

This study highlights the limitations of conventional gravity and magnetic separation methods for processing low-grade iron ores, primarily due to their complex mineralogy and the weak magnetic properties of phases such as goethite, ankerite, and limonite. To address these challenges, an integrated process combining roasting, leaching, and high-intensity magnetic separation was investigated.

Roasting at 800 °C for 90 min with coke as a reducing agent significantly increased the Fe(II) content, thereby enhancing the magnetic susceptibility of iron-bearing minerals. Using the Slon high-intensity magnetic separator at 10,000 G, more than 80% of the sample weight was recovered in the magnetic fraction. Increasing the magnetic field intensity improved weight recovery but reduced iron grade, reflecting the typical trade-off between recovery and grade. For example, in sample L, the iron content decreased from 49% at 1500 G to 25% at 10,000 G, while weight recovery increased by nearly eightfold.

Sulfuric acid leaching effectively removed soluble non-magnetic impurities such as calcite and lime, further promoting Fe(II) formation and improving iron recovery as compared with roasting alone. However, some low-grade iron minerals remained in the product, resulting in a slight decrease in the iron grade; for instance, in sample H, the iron content declined from 48% to 40% at 5000 G.

Overall, the proposed roasting–leaching–magnetic separation process improved both weight recovery (R_w) and iron recovery (R_g) in low-grade iron ores. These results highlight the importance of careful process optimization to balance enhanced recovery with potential grade reduction. While technically promising, additional steps, such as acid leaching, may increase operational costs, and further studies are needed to assess the economic feasibility and environmental sustainability of large-scale industrial implementation.