The Recovery of TiO₂ from Ilmenite Ore by Ammonium Sulfate Roasting–Leaching Process

Abstract

TiO₂ production is a key part of Ti metallurgy and Ti recycling, and the process itself has turned out to be energy-consuming and material-consuming. New technologies are needed to utilize complex Ti ores, such as ilmenite, and reduce the carbon footprint of TiO₂ extraction. Ammonium sulfate roasting has been revealed as an efficient way to carry out phase transformations of complex minerals. A low-temperature sulfation roasting approach was studied to chemically breaking down the crystal structure of ilmenite and generate metal soluble sulfates simultaneously. These roasted products were introduced to water leaching, then the residue of the water leaching was leached by diluted HCl acid, and the TiO₂ product was enriched in the leaching residue. The effects of roasting temperature, roasting time, ilmenite-to-ammonium sulfate mass ratio, ilmenite particle size, and second-stage roasting on iron removal and titanium loss leaching efficiency were systematically studied. The results show that the optimum roasting conditions were a roasting temperature of 500 °C, a roasting time of 210 min, an ilmenite-to-(NH₄)₂SO₄ mass ratio of 1:7, and an ilmenite particle size of below 43 µm. Under optimized conditions, the TiO₂ grade in the obtained synthetic rutile reached 75.83 wt.%. Furthermore, the phase transformation and reaction mechanism during roasting are discussed and interpreted.

1. Introduction

Titanium dioxide (TiO₂) is one of the most important white pigments, and is widely used in plastics, rubber, paints, paper, food, cosmetics, medicine, ceramics, and textile industries [1,2,3]. It has high potential in high-technology applications such as semiconductors, biomedical devices, and catalysts [4,5]. TiO₂ is the primary source for manufacturing Ti metal and Ti chemicals, which are used in aerospace and military applications [6]. Titanium dioxide is originally found in three crystalline forms: rutile, anatase, and brookite. The three crystals are basically pure titanium dioxide but contain slight amounts of impurities like iron, which gives it a black color [7]. TiO₂ pigment is non-toxic and offers the highest opacity, brightness, and whiteness, so it excels in the industry as a white pigment. When the titanium dioxide particle size is reduced to less than 100 nm, unfamiliar optical properties appear. This is due to its high UV absorption and high visibility in visible light [8]. Other significant features of TiO₂ pigments are their high thermal stability, superior resistance to chemical attack, and resistance to ultraviolet degradation. According to the United States Geological Survey (USGS) statistics of 2023, titanium pigment and sponge production worldwide were 9.4 and 0.35 million tons, respectively [9]. China was the largest producer of titanium pigment and sponge, estimated at 5 and 0.18 million tons, respectively [9]. Therefore, prices of TiO₂ pigment are predicted to increase in the next few years due to the rising costs of chlorine, caustic soda, coke, and energy. Additionally, progress in the aerospace and defense industries will definitely increase titanium metal demand.

Titanium dioxide is mainly produced from rutile and ilmenite. Rutile is considered the most titanium-rich mineral, and it is used directly as a raw material for TiO₂ [10]. On the other hand, ilmenite is the most abundant titanium mineral in nature. Deposits exist on coastal beaches as sand placers and in mines as igneous rocks. Although the total world resources of ilmenite, anatase, and rutile were estimated at more than 2 billion tons [9], natural rutile reserves are limited in the earth’s crust. They only represent one-tenth of ilmenite reserves [9,11]. The lack of natural rutile has led to research into the conversion of ilmenite into synthetic rutile [12,13]. There is no doubt that ilmenite will continue to dominate the titanium industry in the future as the most important titanium resource.

Ilmenite industrial practice falls into three types: smelting, chloride, and sulphate processes. These processes differ in their raw material requirements and chemistry [14]. In the smelting process [15], ilmenite is upgraded by a carbothermic process, yielding TiO₂-rich slag as the primary product and pig iron as a by-product. More than 65% of ilmenite is smelted using electric furnaces worldwide and 85% in China, where it is the primary production method for ilmenite [16]. But the smelting process produces large amounts of CO₂ gas, which pollutes the environment and requires excessive energy consumption. Berkovich developed the chloride process to dissolve at least 80% of the titanium and iron in ilmenite ore through direct leaching with concentrated hydrochloric acid [17,18,19]. This method has a long process flow, low product quality, poor recovery, and high cost and waste emissions. The sulphate process was the first commercial process for producing titanium pigment from ilmenite. In this process [20,21], sulfuric acid is combined with ilmenite (40–60% TiO₂) or high-titania slag (72–87% TiO₂) in digestion to form intermediate products (titanyl sulfate and iron sulfate). The biggest challenge to developing titanium dioxide using the sulfuric acid process is waste emissions, including acid hydrolysis residue, waste acid, and acid waste water. Therefore, introducing environmentally clean technologies is necessary. In addition, there are commercialized or proposed processes to produce TiO₂ pigments from ilmenite, such as the Murso process [22], in which fluidized beds are used for thermo-conversions. Ilmenite is pre-oxidized in a fluidized bed at 900–950 °C, and another fluidized bed is used to reduce the ferric ions in the hot oxidized ore with a reducing agent like H₂ gas. Then, the product from the second fluidized bed is digested with about 20% HCl at a temperature range of 108–110 °C. In this process, it is easier to recycle HCl than in the sulfate process. In the Laporte process [23], the ore undergoes pre-oxidation in a fluidized bed at 950 °C, and it is followed by reductive roasting using coke in a rotary kiln at about 900 °C. Using coal as a reductant, ferrous oxide is formed from the ferric content in ilmenite. The roasted ore is leached with 18% HCl at atmospheric pressure for 3.5 h to eliminate the ion content, yielding synthetic rutile with high-purity TiO₂. The Kataoka process in Japan [24] consists of trivalent iron content of ilmenite ore being reduced to ferrous oxide by reduction roasting, then leaching by sulfuric acid. The rate of precipitation of titanium salts is increased by adding hydrated TiO₂ as seeds to the leaching solution in order to improve the removal of iron species and accelerate the sedimentation of titanium salts. As a result, the Kataoka process is able to separate iron from ilmenite at lower temperatures and using less sulfuric acid, thus making it more sustainable and economically viable. TiO₂ recovery reaches more than 95%. In the Austpac process [25], ilmenite is roasted at temperatures between 800 and 1000 °C. Magnetic separation can remove gangue minerals. HCl (25%) removes iron and impurities by leaching. The produced residue is filtered, washed, and calcined. Magnetic separation is used to obtain synthetic rutile > 97% TiO₂. The Benelite process [26] utilizes carbon thermo-reduction for converting other iron forms to ferrous states via leaching with 18–20% HCl.

Nowadays, sulfation roasting has been widely used in the metallurgical processing of complex minerals. The principle of sulfation roasting involves converting metal oxides in complex minerals into metal sulfate salts. These salts can then undergo a leaching process to produce the valuable metals [27]. Sulfuric acid can be used in sulfation roasting, but the sulfuric acid mass fraction is high. This leads to significant corrosion of the equipment used [28]. Roasting using ammonium sulfate is similar to sulfuric acid roasting; however, sulfation roasting with ammonium sulfate is not as corrosive to the equipment. The NH₃ gas produced during the sulfation process can be recycled as ammonium sulfate again [29]. Moreover, ammonium sulfate decomposes at a lower temperature (250 °C), which is considered an advantage from an energy requirement standpoint [30]. Several years ago, some studies investigated the extraction of metals from different minerals by using the ammonium sulfate roasting method. These minerals included vanadium slag [31], bauxite ore [32], spent lithium-ion batteries [33], manganese ore [34], and nickel sulfide ores [35]. Some studies have been conducted on ilmenite ore using ammonium sulfate roasting. Malaysian ilmenite was roasted using ammonium sulfate in an N₂ atmosphere. The separation process was performed by adjusting the pH of the solution and adding rutile as well as iron scrap seeds [36]. However, the flowsheet was long and the reaction mechanism was not involved in the roasting process. Panzhihua ilmenite was roasted using ammonium sulfate in an N₂ atmosphere in two steps: The first step was the sulfation of ilmenite at a lower temperature, and in the second step, the optimum sulfated sample was thermally decomposed at a higher temperature [37]. However, the study focused on the factors that affect the leaching process, with little consideration given to the parameters that affect roasting.

Hereby, the present work proposes a route for TiO₂ recovery from ilmenite ore via an ammonium sulfate roasting process. This is to achieve the mineral phase transformation of metal oxides to soluble metal sulfate salts. The obtained roasted product is leached with distilled water and acid leaching followed by calcination to produce synthetic rutile. This proposal focuses on enhanced selective TiO₂ recovery and maximized leaching efficiency of iron from roasted ilmenite. The effects of roasting factors including roasting temperature, roasting time, mass ratio of ilmenite to ammonium sulfate, and ilmenite particle size on the leaching efficiency of iron removal and titanium loss are investigated in detail. Previous articles have demonstrated that the complexity of the chemical structure of ilmenite ore prevents selective TiO₂ recovery and increases energy requirements. Therefore, in this study, the reaction mechanism of the roasting process is discussed in depth. It has been reported that few studies have investigated the effect of ilmenite particle size on leaching efficiency during roasting. Additionally, the effect of second-stage roasting on leaching efficiency is investigated. Therefore, the study aims to cover all of these factors that have not been reported previously. The process is simple, short, easy to implement, and environmentally friendly.

2. Materials and Methods

2.1. Materials

A black sand ilmenite ore was received from the Nuclear Materials Authority in Egypt. the chemical composition, phase composition, and morphology of the ore are presented in

Table 1. Chemical composition of the black sand ilmenite ore (wt.%).

Component	TiO2	Total Fe	SiO2	CaO	MnO	MgO	Al2O3	V2O5
Content wt.%	43.13	36.45	2.35	1.55	1.16	1.16	0.81	0.32

and Figure 1.

The X-ray diffraction (XRD) analysis of the ilmenite is shown in Figure 1a, which indicates that the major mineral phases are FeTiO₃ and Fe₂O₃. Figure 1b indicates the morphology of the raw material of the ilmenite ore supported by energy dispersive spectroscopy (EDS) analysis. The main components are two different grains; the light grain is the FeTiO₃ phase with a complex crystal structure between iron and titanium oxides. The other light grain with dark shades from SiO₂ and metal silicate impurities in Figure 1c was observed at high magnification. The chemical composition of the ilmenite ore was determined by inductively coupled plasma optical emission spectrometry (ICP-OES). Table 1 indicates that the main metal oxides are iron oxides with a total Fe of 36.45 wt.% and TiO₂ of 43.13 wt.%, in addition to SiO₂ of 2.35 wt.% and CaO of 1.55 wt.% as the major impurities. Other impurities include traces of Mn, Mg, Al, and V oxides. These traces form shades of metal silicates with Si. During the experiments, the −74 μm fraction of the ore was utilized. The chemicals, including (NH₄)₂SO₄ and HCl acid, were analytical reagent grade purchased from Sinopharm Chemical Reagent Co. Ltd, Shanghai, China.

2.2. Methods

The ilmenite ore was dried for 6 h in a drying oven. The dried sample of 3 ± 0.005 g of ilmenite was mixed with ammonium sulfate in an agate mortar to obtain a homogeneously mixed powder. The sample was placed in an alumina crucible for each roasting experiment, as shown in Figure 2.

The crucible with the sample was pushed into the hot zone in a horizontal electric tube furnace (SG-GL1200, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Science, Shanghai, China), and a vacuum pump was used to remove the air from the quartz tube by inserting argon gas for 20 min before switching on the furnace to heat it up to the required temperature. The heating rate was 8 °C/min, and the argon gas flow rate was maintained at 50 mL/min via a flow controller. The sample was left in the hot zone for roasting at the target temperature. As soon as the roasting experiment was completed, the crucible was slowly pulled out to the cool zone using metallic wire, in the presence of argon flow gas, in order to cool it quickly. Some parameters that could have influenced the roasting experiment were investigated, such as roasting temperature (350–600 °C), roasting time (60–240 min), mass ratio of ilmenite ore to ammonium sulfate (1:3 to 1:11), and ilmenite particle size.

The roasted product was weighed in each experiment to determine the relative weight loss, and a sample of 4 ± 0.005 g was leached after grinding it with distilled water in a water bath (leaching temperature = 50 °C, liquid–solid ratio = 10 mL/g, magnetic stirring speed = 150 rpm, leaching time = 1 h). After water leaching, the solid residue from the water leaching was leached using a diluted HCl acid solution of 2.5 wt.% (leaching temperature = 98 °C, liquid–solid ratio = 50 mL/g, magnetic stirring speed = 150 rpm, leaching time = 2 h). The optimum acid leaching residue was thoroughly rinsed with water and calcined at 1000 °C for 1 h to obtain synthetic rutile. The leaching efficiency is mainly used to evaluate the roasting process efficiency, and the leaching efficiency is calculated as follows:

$$\mathsf{\eta }=\frac{{\mathrm{c}}_1\times {\mathrm{v}}_1\times {10}^{-3}}{{\mathrm{m}}_1\times {\mathrm{w}}_1}\times 100$$

(1)

η: iron removal efficiency, titanium loss (wt.%), c₁: Fe or Ti concentration in the solution (g·L⁻¹), v₁: volume in the leaching solution (mL), m₁: mass of roasted sample (g), and w₁: Fe or Ti mass percentage in ilmenite (wt.%).

2.3. Analytical Methods

Thermogravimetric analysis and differential scanning calorimetry (TG-DSC, NETZSCH, STA 449 F3, and Germany) were conducted to investigate the thermal composition behaviors of ammonium sulfate and a mixture of ilmenite and ammonium sulfate. Phase identification for the products was conducted using an X-ray powder diffractometer (XRD, Rigaku Company, Tokyo, Japan, D/max-2500) to determine their qualitative mineralogical composition, and a copper-Ka anode target, a wavelength λ of 1.54 Å, a scanning range of 10°–80°, and a scanning speed of 10°/min were employed. Elemental chemical composition of the materials and metal concentrations in the leachate were determined by inductively coupled plasma optical emission spectrometry (ICP-OES) (ICAP7400 Radial, Thermo Fisher, Waltham, MA, USA). Products from the experiments were mounted in epoxy, ground, and polished to observe their morphology using a scanning electron microscope (SEM, Tescan, Brno, Czech Republic) and energy dispersive spectroscopy (EDS, Thermo Fisher Scientific, Waltham, MA, USA).

3. Results and Discussion

3.1. Sulfation Roasting

3.1.1. Effect of Roasting Temperature

The influence of roasting temperature on the leaching efficiency of iron removal and titanium loss in the leaching solution was studied from 350 °C to 600 °C under a mass ratio of ilmenite to (NH₄)₂SO₄ of 1:5, a roasting time of 120 min, and an ilmenite particle size of <74 µm. In order to evaluate the roasting process, the leaching efficiency of iron removal is an important indicator. The roasted products obtained under different conditions were leached in water followed by diluted HCl treatment of the residue of the water leaching. In other words, the sulfation process breaks down the chemical structure of ilmenite metal oxides and converts metal oxides into soluble metal sulfates. As shown in Figure 3a, the leaching efficiency of iron and titanium loss was 10.3% and 7.1%, respectively, at an operating temperature of 350 °C. By increasing the temperature to 500 °C, iron removal and Ti loss efficiency were increased to 34.4% and 11.9%, respectively. Up to 500 °C, iron removal and Ti loss were decreased to 18.8% and 0.8%, respectively. This decrease is due to the decomposition of metal sulfate or ammonium metal sulfate salts into oxides at higher temperatures.

Figure 4a shows the color change of roasted products after they are roasted. It is shown that the roasted product was still black at 350 °C, which can be attributed to the sulfation process in the beginning stage. The color of the roasted product became whitish with increasing temperature. At higher temperatures up to 550 °C, the color became brown. The color change of the roasted products agreed well with the weight loss curve after roasting at different temperatures. This is shown in Figure 3a, where the weight loss at 600 °C reached 83%. It is not recommended to roast at higher temperatures because the metal sulfate salts or ammonium metal sulfate will decompose into their respective oxides during roasting.

The XRD pattern of roasted products at roasting temperatures ranging from 350 °C to 550 °C is shown in Figure 5. It is clear from the roasting process that the temperature significantly affected conversion. At a roasting temperature of 350 °C, the (NH₄)₃H(SO₄)₂ phase appeared due to ammonium sulfate decomposition. In addition, some peaks of the (NH₄)₃Fe(SO₄)₃ phase formed due to the reaction of (NH₄)₃H(SO₄)₂ with ilmenite ore. Additionally, diffraction peaks of (NH₄)₂SO₄ were observed, which indicates that the decomposition was not completed. At a roasting temperature of 400 °C, all of the (NH₄)₂SO₄ decomposed into the (NH₄)₃H(SO₄)₂ phase, which reacted directly with ilmenite to produce (NH₄)₃Fe(SO₄)₃ and FeSO₄.4H₂O. When the temperature was increased to 450 °C, the diffraction peaks of the (NH₄)₃H(SO₄)₂ phase completely disappeared, indicating that all amounts of (NH₄)₃H(SO₄)₂ reacted with the ilmenite ore. The main phase was NH₄Fe(SO₄)₂, which was directly converted from of the (NH₄)₃Fe(SO₄)₃ phase. By increasing the temperature to 500 °C, Fe₂(SO₄)₃, TiOSO₄, and traces of Fe₂O₃ phases appeared and the peak intensity of NH₄Fe(SO₄)₂ decreased. By increasing the temperature to 500 °C, TiO₂ and Fe₂O₃ appeared due to the transformation of ammonium metal sulfates and metal sulfates salts into their respective oxides. This study converted the metal oxides in the ore to ammonium metal sulfate and metal sulfates because these compounds easily dissolve in water. In order to save energy and guarantee high leaching efficiency, the roasting temperature of 500 °C was selected as an optimal operating temperature for studying the other factors.

The morphology of the products after roasting at different temperatures was examined by SEM-EDS analysis. Figure 6a shows that the product roasted at 400 °C obtained surface roughness, which could have resulted from the excess of produced gases, indicating that the roasting reaction had begun, which is in good agreement with the XRD results at 400 °C. As shown in Figure 6b, the micrograph at 500 °C showed non-uninform particle size distribution and less porosity than at 400 °C. Fe and Ti sulfate phases appeared in dark colors. At high magnification, as shown in Figure 6d, the ilmenite phase (FeTiO₃) had a very complex crystal structure due to the interaction of iron and titanium oxides in the crystal structure. It was observed that ammonium sulfate reacted with the outer surface of the ilmenite phase to form metal sulfates, which appeared in a grey color on the outer layer of the ilmenite phase, and the FeTiO₃ phase is indicated by the white areas, which show that there were many particles of the ilmenite phase that were not sulfated yet. The micrograph at 550 °C, as shown in Figure 6c, is different than the sample roasted at 500 °C. This is because there were fewer residuals of ammonium sulfate, the roasted particles became smaller, and a small fusion was detected. Furthermore, the micrograph of the residue after leaching at 500 °C is shown in Figure 6e. The residue had two crystalline phases, FeTiO₃ and TiO₂, but the ilmenite phase was still found in the residue in high amounts. These findings indicate that the roasting process did not occur completely.

The thermal decomposition of pure ammonium sulfate and the mixture of ilmenite and ammonium sulfate were investigated using TG/DSC at a heating rate of 5 °C/min to 800 °C in the argon atmosphere (Figure 7). As shown in Figure 7a, (NH₄)₂SO₄ showed better stability below 250 °C, and the first stage of weight loss was 13.9% from 250 to 305.3 °C, which may match the NH₃ release in Equation (2). In the second stage, a weight loss of 7.8% was observed between 305.3 and 344.8 °C, which corresponds with the occurrence reaction in Equation (3) (theoretical weight loss of 6.81%). In the final stage, a big weight loss step appeared in the range of 344.8 to 413.7 °C, with a weight loss of 74.09%, which agrees with the process of (NH₄)₂S₂O₇ decomposition into NH₃, N₂, SO₂, and H₂O according to Equation (4) [38,39]. This indicates that ammonium sulfate decomposition was completed.

$${\left({\mathrm{NH}}_4\right)}_2{\mathrm{SO}}_4\to {\mathrm{NH}}_4{\mathrm{HSO}}_4+{\mathrm{NH}}_3\left(\mathrm{g}\right)$$

(2)

$$2{\mathrm{NH}}_4{\mathrm{HSO}}_4\to {\left({\mathrm{NH}}_4\right)}_2{\mathrm{S}}_2{\mathrm{O}}_7+{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(3)

$$3{\left({\mathrm{NH}}_4\right)}_2{\mathrm{S}}_2{\mathrm{O}}_7\to 2{\mathrm{NH}}_3\left(\mathrm{g}\right)+2{\mathrm{N}}_2\left(\mathrm{g}\right)+6{\mathrm{SO}}_2\left(\mathrm{g}\right)+9{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(4)

The overall reaction is:

$$3{\left({\mathrm{NH}}_4\right)}_2{\mathrm{SO}}_4\to 4{\mathrm{NH}}_3\left(\mathrm{g}\right)+{\mathrm{N}}_2\left(\mathrm{g}\right)+3{\mathrm{SO}}_2\left(\mathrm{g}\right)+6{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(5)

The TG-DSC curves of the mixture of ilmenite and ammonium sulfate from 100 to 800 °C are shown in Figure 7b. The mixture property did not change below 250 °C, indicating that no reaction occurred at this stage. When the temperature rose to 300 °C, the mixture had a weight loss rate of about 15.4%. Based on the TG-DSC curves of (NH₄)₂SO₄ and the above analysis results, it can be concluded that (NH₄)₂SO₄ was decomposed into (NH₄)₂S₂O₇, NH₃, and H₂O at this stage. In addition, (NH₄)₃H(SO₄)₂ could have been formed according to Equations (6) and (7). When the temperature increased to 380 °C, a big weight loss step appeared, and the weight loss rate was 34.55%. This resulted from the reaction in Equations (8) and (9) between ilmenite and (NH₄)₃H(SO₄)₂ to generate the corresponding sulfates. As the temperature was raised to 462 °C, the mixture lost 12.7% weight as a result of the reactions listed in Equations (10) and (11), during which the ammonium metal sulfate phase transforms into the metal sulfate phase. These results agree with the XRD pattern at the same temperature. In the last stage, from 462 to 600 °C, the weight loss was about 11.38% due to the reactions listed in Equations (12) and (13), in which the metal sulfate phases transform into metal oxides. This indicates that the reaction finished at this stage.

Based on the XRD, SEM-EDS, and TGA-DSC data, the expected reaction mechanism during the roasting process was investigated at different temperatures in the following reactions [38,40,41].

$${\left({\mathrm{NH}}_4\right)}_2{\mathrm{SO}}_4+{\mathrm{NH}}_4{\mathrm{HSO}}_4\to {\left({\mathrm{NH}}_4\right)}_3\mathrm{H}{\left({\mathrm{SO}}_4\right)}_2$$

(6)

$$2{\left({\mathrm{NH}}_4\right)}_2{\mathrm{SO}}_4\to {\left({\mathrm{NH}}_4\right)}_3\mathrm{H}{\left({\mathrm{SO}}_4\right)}_2+{\mathrm{NH}}_3\left(\mathrm{g}\right)$$

(7)

$$2{({\mathrm{NH}}_4)}_3\mathrm{H}{({\mathrm{SO}}_4)}_2+{\mathrm{FeTiO}}_3\to {\left({\mathrm{NH}}_4\right)}_3\mathrm{Fe}{\left({\mathrm{SO}}_4\right)}_3+{\mathrm{TiOSO}}_4+3{\mathrm{NH}}_3\left(\mathrm{g}\right)+2{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(8)

$$6{({\mathrm{NH}}_4)}_2{\mathrm{SO}}_4+{\mathrm{Fe}}_2{\mathrm{O}}_3\to 2{\left({\mathrm{NH}}_4\right)}_3\mathrm{Fe}{\left({\mathrm{SO}}_4\right)}_3+6{\mathrm{NH}}_3\left(\mathrm{g}\right)+3{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(9)

$${\left({\mathrm{NH}}_4\right)}_3\mathrm{Fe}{\left({\mathrm{SO}}_4\right)}_3\to {\mathrm{NH}}_4\mathrm{Fe}{\left({\mathrm{SO}}_4\right)}_2+2{\mathrm{NH}}_3\left(\mathrm{g}\right)+{\mathrm{SO}}_3\left(\mathrm{g}\right)+{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)$$

(10)

$$4{\mathrm{NH}}_4\mathrm{Fe}{\left({\mathrm{SO}}_4\right)}_2\to 2{\mathrm{Fe}}_2{\left({\mathrm{SO}}_4\right)}_3+2{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)+4{\mathrm{NH}}_3\left(\mathrm{g}\right)+2{\mathrm{SO}}_2\left(\mathrm{g}\right)+{\mathrm{O}}_2\left(\mathrm{g}\right)$$

(11)

$${\mathrm{Fe}}_2{\left({\mathrm{SO}}_4\right)}_3\to 2{\mathrm{Fe}}_2{\mathrm{O}}_{3 }+6{\mathrm{SO}}_2\left(\mathrm{g}\right)+3{\mathrm{O}}_2\left(\mathrm{g}\right)$$

(12)

$$2{\mathrm{TiOSO}}_4\to 2{\mathrm{TiO}}_2+2{\mathrm{SO}}_2\left(\mathrm{g}\right)+{\mathrm{O}}_2\left(\mathrm{g}\right)$$

(13)

3.1.2. Effect of Roasting Time

The effect of roasting time, ranging from 60 min to 240 min, on the leaching efficiency of iron and titanium loss was investigated under the mass ratio of ilmenite to (NH₄)₂SO₄ of 1:5, a roasting temperature of 500 °C, and an ilmenite particle size of <74 µm. Figure 3b shows that at a roasting time of 60 min, the leaching efficiency of Fe removal and Ti loss were only approximately 26.1% and 21.1%, respectively. By increasing the roasting time to 210 min, an increased leaching efficiency of iron of approximately 44.1% was reached, whereas Ti loss in the solution decreased to 1.5%. By further expanding roasting time to 210 min, the leaching efficiency of iron was still steady at 44.4% and Ti loss in the solution decreased slightly to 1%. The weight loss of the roasted products after roasting at different roasting times ranged from 58% to 77%, as shown in Figure 3b. In addition, there was a slight change in the color of the roasted products, as shown in Figure 4b, because the difference in the weight loss at different roasting times was not as large compared with the weight loss in the previous parameter.

Figure 8 shows the XRD pattern of the roasted products at roasting times ranging from 60 min to 210 min. The roasting time clearly affected the phase transformation. At 60 to 90 min of roasting time, NH₄Fe(SO₄)₂ was the primary phase. By increasing the roasting time from 120 to 180 min, the intensity of the diffraction peaks of NH₄Fe(SO₄)₂ decreased. New phases of Fe₂(SO₄)₃, TiOSO₄ and traces of Fe₂O₃ appeared. This means that by increasing roasting time, the ammonium iron sulfate phase transformed into ferric sulfate according to Equation (11). At a high roasting time of 210 min, the diffraction peaks of ammonium iron sulfate completely disappeared beside the existing Fe₂(SO₄)₃, TiO₂, and Fe₂O₃ phases. Another FeSO₄ phase appeared, possibly due to converting some of the Fe₂(SO₄)₃ into FeSO₄. It seems that the XRD patterns of various roasting times aligned well with those of XRD patterns at different roasting temperatures.

The morphology of the roasted product and the residue of acid leaching at a roasting time of 210 min were examined by SEM-EDS analysis. Figure 9a depicts the roasted product at 210 min, showing a non-sulfated ilmenite phase surrounded by a metal sulfate-phase layer. The residue after leaching at 210 min was investigated by SEM-EDS, as shown in Figure 9b. The residue had two main crystalline phases, FeTiO₃ and TiO₂, and the ilmenite phase was still found in the residue in smaller amounts than in Figure 6e. From the SEM-EDS images, we concluded that the non-sulfated ilmenite phases remained in the residue after leaching, which indicates that the roasting process did not occur completely. To study the next factors, 210 min was selected as the optimal roasting time.

3.1.3. Effect of Ilmenite-to-(NH₄)₂SO₄ Mass Ratio

The dosage of (NH₄)₂SO₄ required for high iron leaching is a critical factor for the roasting process. The effect of the ilmenite ore-to-(NH₄)₂SO₄ mass ratio on the leaching efficiency of iron and titanium loss was investigated under the following optimal conditions: roasting temperature of 500 °C, roasting time of 210 min, and ilmenite particle size of <74 µm. Figure 3c shows that increasing the mass ratio of ilmenite to ammonium sulfate promoted the leaching efficiency. However, increasing the ratio further may have affected the leaching efficiency. At a mass ratio of 1:3, the leaching efficiency of Fe removal and Ti loss was only approximately 37.3% and 0.9%, respectively. Increasing the mass ratio to 1:7 further enhanced the iron leaching efficiency by 45.6%, whereas Ti loss was 1%. The leaching efficiency of iron and Ti loss was still fixed at around 45.2% and 1%at a mass ratio from 1:7 to 1:11, respectively. When more (NH₄)₂SO₄ was added, a large amount of foam formed in the roasted sample. This foam was formed due to the large amounts of gas products released during roasting. Due to the difference in density between ilmenite particles and foam, the reactants were divided into two layers. In this reaction, ilmenite sunk into the lower layer, whereas ammonium bisulfate floated in the upper layer, which caused irregular mixing of the ilmenite and (NH₄)₂SO₄, leading to lower leaching efficiency. This was obvious at a mass ratio of 1:11; the weight loss of the roasted sample reached approximately 85%, as shown in Figure 3c. In addition, it was observed that the color change of the roasted samples turned whitish by increasing the ammonium sulfate dosage, as shown in Figure 4c.

The XRD pattern of the roasted samples at an ilmenite-to-(NH₄)₂SO₄ mass ratio of 1:3 to 1:11 is shown in Figure 10. At a mass ratio of 1:3, the main phases in the roasted sample were FeSO₄, Fe₂(SO₄)₃ and TiO₂. By increasing the dosage of ammonium sulfate from a mass ratio of 1:5 to 1:11, NH₄Fe(SO₄)₂ and Fe₂O₃ phases appeared. Based on these XRD patterns, the last two parameters also matched. Therefore, the mass ratio of 1:7 was chosen as an optimum operating mass ratio to study the next parameters.

3.1.4. Effect of Ilmenite Particle Size

The influence of ilmenite particle size on the leaching efficiency of iron and titanium loss was studied under the following conditions: roasting temperature of 500 °C, roasting time of 210 min, and ilmenite-to-(NH₄)₂SO₄ mass ratio of 1:7. Figure 3d shows the relationship between iron leaching efficiency and titanium loss with ilmenite particle size. The leaching efficiency of Fe removal and Ti loss was only approximately 34.2% and 0.9%, respectively, at an ilmenite particle size of 150 to 98 µm. The leaching efficiency of Fe increased significantly with the ilmenite particle size decreasing from 150 to less than 43 µm, so the optimum leaching efficiency of iron was 49.6% at an ilmenite particle size of less than 43 µm. Furthermore, the Ti loss in the leaching solution decreased to 0.1% by reducing the ilmenite particle size. When the ilmenite particle size was finer, its crystal structure was more easily broken down and iron oxide was released more easily. Energy consumption was high when grinding the ore particles to smaller sizes, although high leaching efficiency was achieved.

3.2. Effect of Second-Stage Roasting

The effect of second-stage roasting was studied to enhance the iron removal leaching efficiency at the following roasting conditions: roasting temperature of 500 °C, roasting time of 210 min, roasted sample-to-(NH₄)₂SO₄ mass ratio of 1:4, and ilmenite particle size of <43 µm. Figure 11a shows that second-stage roasting improved the iron removal efficiency, where the iron removal efficiency increased from 49.6% to 53.3%. In addition, Ti loss increased slightly to 0.3% in the leaching solution. The color change of the roasted products after roasting more than one time was clear, as shown in Figure 4d. It changed from whitish at the first stage of roasting to brown at the fourth stage of roasting. This is due to the thermal decomposition of ammonium metal sulfate and metal sulfates into metal oxides. Figure 11b shows the morphology of the roasted product after second-stage roasting, where the dark grains of Fe and Ti sulfates appear in agglomerates and the white grains are non-sulfated ilmenite.

The residue of second-stage roasting after acid leaching was washed with pure water, dried, and calcined at 1000 °C for 1 h. Figure 11c shows that the main phase of the final product was rutile (TiO₂), and a small diffraction peak of pseudobrookite (Fe₂TiO₅) resulted from the oxidation of non-sulfated ilmenite in the residue during calcination. Based on Figure 11d, the morphology of the final product is consistent with the XRD analysis, which showed that the main gray phase was rutile and the small light areas were pseudobrookite.

Table 2. Chemical composition of synthetic rutile after calcination (wt.%).

Component	TiO2	Total Fe	SiO2	CaO	MnO	MgO	Al2O3	V2O5
Content wt.%	75.83	16.53	4.26	0.06	0.27	0.11	0.30	0.92

shows the chemical composition of the final product, determined by ICP-OES, where the TiO₂ grade reached 75.83 wt.%.

3.3. Reaction Mechanism

Figure 12 illustrates the conversion mechanism of ilmenite ore during the sulfation roasting process. Firstly, (NH₄)₂SO₄ gradually decomposes at temperatures greater than 250 °C to produce (NH₄)₃H(SO₄)₂, NH₃, and H₂O, as shown in Equations (2), (6) and (7), respectively. The formed (NH₄)₃H(SO₄)₂ reacts with ilmenite ore. By increasing the roasting temperature, the metal oxides in the ore convert to NH₄Fe(SO₄)₂ and TiOSO₄ according to Equations (8) and (9), respectively. By increasing the roasting temperature, NH₄Fe(SO₄)₂ transforms to Fe₂(SO₄)₃ and Fe₂O₃, but TiOSO₄ transforms to TiO₂, as shown in Equations (10)–(13), respectively. The roasted sample is leached in water to remove water-soluble sulfate salts. The residue of the water leaching is leached by diluted HCl solution to remove some of the Fe₂O₃ formed as a result of the iron sulfate decomposition. The final product consists of TiO₂ phase and some ilmenite particles that are not sulfated during the roasting process.

3.4. Flow Sheet

A proposed flow sheet with optimal conditions for recovering TiO₂ from ilmenite based on this study’s results is shown in Figure 13. The ilmenite is first roasted with ammonium sulfate. The roasted sample is thus leached in water to dissolve the water-soluble sulfate. The residue of the water leaching is followed by diluted HCl to remove some of the iron oxides formed by iron sulfate decomposition during the roasting process. The residue of acid leaching at optimum conditions is calcined to obtain synthetic rutile. The NH₃ and SO₃ gases released during roasting can be efficiently absorbed by water. Then, ammonium sulfate can be acquired through evaporation and crystallization. Additionally, solution containing iron can be recycled by separating iron chloride through evaporation. Via pyrolysis, iron chloride is transformed into hydrochloric acid and iron oxide is produced. The HCl acid can be reused again, and the iron oxide can be used as a feedstock for cement production, steel production, or other purposes. This not only reduces environmental pollution but also reduces raw material consumption.

4. Conclusions

• The recovery of TiO₂ from titanium ores such as ilmenite via traditional methods like smelting or acid leaching processes has disadvantages. Theses disadvantages include high energy consumption, CO₂ emissions, and acid waste emissions. Therefore, there is an urgent need for an efficient, economical, clean, and sustainable technology for the recovery of TiO₂ from ilmenite ore.
• An ammonium sulfate roasting method was proven to be an effective way to convert metal oxides in ilmenite ore to soluble sulfates at a low temperature. (NH₄)₂SO₄ acted as a sulfation agent to produce NH₄Fe(SO₄)₂, Fe₂(SO₄)₃, TiOSO₄, and Fe₂O₃ during the roasting process. The roasted products were leached by water, followed by diluted HCl acid.
• The optimal iron leaching efficiency was 49.6% at the following roasting conditions: roasting temperature of 500 °C, roasting time of 210 min, ilmenite-to-(NH₄)₂SO₄ mass ratio of 1:7, and ilmenite particle size of less than 43 µm. The second-stage roasting enhanced the iron leaching efficiency from 49.6% to 53.3%.
• The obtained synthetic rutile after calcination had a TiO₂ grade of 75.83 wt.%. According to the XRD analysis, the main phase was rutile-TiO₂, with small diffraction peaks of Fe₂TiO₅. The iron removal efficiency was not high due to the non-sulfated ilmenite particles shown in the XRD and SEM-EDS analysis. These remaining non-sulfated ilmenite particles, which were surrounded by a metal sulfate layer, demonstrate the complexity of the ilmenite ore crystal structure. Therefore, more attention should be paid to the crystal structure breakdown of ilmenite in future studies.
• During this process, ammonium sulfate and hydrochloric acid can be recycled, as well as iron that is wasted in the solution. Therefore, the process is environmentally friendly.