Thermodynamic Analysis of the Reduction of Vanadium-titanium Magnetite by Gasification Gas

Abstract

Vanadium-titanium magnetite (VTM) has a high comprehensive utilization value. The gas-based reduction process combined with the electric furnace melting process has the advantages of a high recovery rate of precious metal elements in VTM, low cost, high-energy efficiency, and low environmental pollution. Gasification gas provided by coal gasification technology is a potential reducing agent. Through FACTSAGE calculations and experiments, the reduction thermodynamic properties of VTM oxidized pellets by gasification gas were systematically studied to illustrate the feasibility of the system before practical experiments. The reaction mechanism of the direct reduction of VTM by gasification gas was proved, and the conversion behavior of iron and titanium in the reduction system was also clarified. This work clearly describes the reduction process of VTM oxidation pellets and clarifies the essence of the direct reduction of gas groups of VTM oxidized pellets.

1. Introduction

Vanadium-titanium magnetite (VTM) is a composite polymetallic ore composed of titanium (Ti), iron (Fe), and vanadium (V).^1,2) VTM is widely distributed in China, the United States and India.^3,4) The key to using VTM is to effectively separate and extract the valuable metals, such as iron, vanadium, and titanium.⁵⁾ Currently, the blast furnace process is the primary technology for utilizing VTM.^6,7) Because a solvent is introduced in this process, not only the grade of titanium in the blast furnace slag is very low, but TiO₂ reacts with the solvent (CaO) to form perovskite, which makes it difficult to extract the titanium resources from VTM. More importantly, although the current blast furnace process dominates the global steel industry, it has obvious disadvantages. First, this process consumes a large amount of coke,⁸⁾ and the current lack of coke resources cannot support the long-term sustainable development of the blast furnace process. In addition, the blast furnace ironmaking process consumes a massive amount of energy and causes serious pollution.⁹⁾ Non-blast furnace processes, especially gas-based reduction processes,^10,11) have great potential.^12,13) The gas-based reduction process combined with the electric furnace melting process does not require the use of coke or the addition of calcium oxide and has the advantages of a high recovery rate of precious metal elements in iron ore, low-cost, high-energy efficiency, and low environmental pollution.^14,15) Natural gas is currently the main gas-reducing agent used in the gas-based reduction process around the world.¹⁶⁾ However, China’s natural gas reserves are insufficient, and resources are unevenly distributed,¹⁷⁾ which severely limits the development of China’s gas-based reduction. Fortunately, China has abundant coal resources.¹⁸⁾ As shown in Fig. S1 (Supporting Information), combining China’s abundant coal resources and increasingly mature gasification technology, our research group proposed a new process for the integration of flue gas gasification and the direct reduction of VTM that opens an effective way to develop gas-based direct reduction technology. We conducted a feasibility study on the production of gasification gas from coal and biomass^{19,20,21,22,23)} and found that the main components of gasification gas are carbon monoxide (CO) and hydrogen (H₂). In this process, gasification gas was used as the reducing agent, and the flue gas after the VTM reduction was used as the gasification agent for coal gasification. The gas-based direct reduction of VTM does not add solvent and coal as the reducing agent, therefore, the product is relatively pure.

A thermodynamic analysis of the Gibbs free energy changes was conducted for a feasibility analysis of the system before practical experiments.^24,25) Some key reactions of VTM reduction were carried out by this method, and our research group also conducted exploratory research.¹⁴⁾ However, as a complex symbiotic mineral of iron, titanium, vanadium, and oxygen, the reaction process of VTM is very complex. Moreover, the effects of the reaction temperature, gasification gas flow rate, composition on the phase composition, and reduction degree of the product need to be further explored. Therefore, in this work, based on the analysis of FactSage software (www.factsage.com), a detailed thermodynamic study of the gas-based direct reduction of VTM was carried out. First, the phase composition and proportion of the VTM were determined by XRD, XRF, and XPS analyses. The effects of reduced temperature, total amount of reducing gas, gasification gas composition, and reaction pressure on the direct reduction of VTM oxidized pellets were systematically studied. Furthermore, the reduction characteristics of the VTM were studied by VTM reduction experiments. The reduction degree and phase composition of the VTM were confirmed. In this work, the reaction mechanism of the gas-based direct reduction of VTM oxidized pellets was determined, and the transformation behavior of substance types in the system was also clarified.

2. Materials and Methods

2.1. Materials

The VTM ore was obtained from Panzhihua, China. As shown in the XRD pattern of the VTM ore in Fig. 1, the main mineral phases of the VTM ore are titanomagnetite (TTM, the solid solution formed by magnetite as the main component and ilmenite spar) and ilmenite (FeTiO₃). The VTM oxidized pellets were prepared as we have described in our published paper.²⁶⁾ The VTM in the subsequent calculation simulations and experiments in this article refers to VTM oxidized pellets. The chemical compositions of VTM oxidized pellets are listed in Table 1. Figure 1 indicates that TTH (the solid solution formed by hematite as the main component and ilmenite) and pseudobrookite (Fe₂TiO₅) are the main mineral phases in the VTM oxidized pellets. As shown in Fig. 2, the spatial feature is the close symbiosis of Fe₂O₃—TiO₄—MgO·Al₂O₃—FeO·TiO₂. Moreover, MgO and VO replace a part of the FeO, which greatly increases the difficulty of reduction.

Fig. 1.

XRD patterns of VTM ore and VTM oxidized pellets. (Online version in color.)

Table 1. Composition of VTM oxidized pellet, mass%.

Fe2O3	TiO2	SiO2	Al2O3	MgO	CaO	Na2O	V2O5	MnO	K2O	S
65.0	13.4	8.42	6.54	3.28	1.40	0.633	0.54	0.32	0.063	0.037

Fig. 2.

Schematic diagram of the lattice transformation of the main phases during the VTM oxidation roasting process: a(TTM); b(TTH); c(FeTiO₃); d(Fe₂TiO₅). (Online version in color.)

2.2. Simulation Methods

FactSage is a commercially available software that was introduced in 2001 as the fusion of the F*A*C*T/FACT-Win (Thermfact, Canada) and ChemSage (GTT-Technology, Germany) thermochemical packages. FactSage (8.1 version) uses the ChemSage Gibbs energy minimization routine and has been trusted and utilized by researchers in the field of materials and metallurgy.²⁷⁾ The equilibrium compositions were calculated using the Equilib module of FactSage²⁸⁾ based on the conservation of mass and energy and minimization of the total Gibbs free energy. FactPS, FToxid and FTsalt of FactSage are selected for the simulated database. The amount of VTM is 200 g, which is converted into molar amount and input into FactSage. The reducing gas takes the volume as the variable, which is also converted into the molar quantity and input into FactSage. The ratio between H₂ and CO is volume ratio/molar ratio. In the experimental section, the reducing gas is continuously fed into the reactor at a given flow rate. We extended the reaction time to 240 min at a certain temperature to achieve a sufficient reaction. In the thermodynamic calculation section using FactSage, we convert the reducing gas flow to the total amount of reducing gas. We calculate the total amount of reducing agent by multiplying the reducing gas flow by 240 min. Then we use this calculated total amount of reducing gas as the input parameter of the FactSage, and the total amount of reducing gas in a single calculation is taken as the variable.

2.3. Experimental Methods

To investigate the direct reduction feasibility of VTM, reduction experiments were performed. As shown in Fig. 3, the experimental device consisted of a gasification gas flow control cabinet, tube furnace, refractory alloy steel material reactor, gas analyzer, electronic balance, and computer. The VTM oxidized pellets were evenly spread on the corundum balls. The tube furnace power was turned on, and the heating program was started. During the heating process, N₂ was passed into the reactor as a protective gas. The volume ratio of N₂ in reducing gas is 25%. At the specified temperature, the gasification gas was cut for reduction. During the reduction process, the electronic balance continuously recorded the weight loss of the pellet to obtain the reduction degree of the VTM. The gas analyzer continuously collected exhaust gas data. After the reduction, the N₂ was removed. In order to maintain the airtightness of the reaction system, we placed the thermocouple outside the tube during the reduction experiments. Therefore, before we carried out the reduction experiments, we carried out temperature calibration experiments to estimate the actual temperature of the sample through the thermocouple outside the furnace tube, and the experimental device is shown in Fig. S2.

Fig. 3.

Schematic diagram of the experimental apparatus. (1-gas cylinder, 2-flow control cabinet, 3-gas mixing chamber, 4-reactor, 5-fever zone, 6-alundum tube, 7-corundum ball, 8-oxidized VTM pellets, 9-thermocouple, 10-tube furnace, 11-temperature controlling cabinet, 12-wash bottle, 13-drying bottle, 14-coal gas analyzer, 15-electronic balance, 16-computer.) (Online version in color.)

2.4. Characterization

The patterns of X-ray diffraction (XRD) were collected by a powder X-ray diffractometer (Rigaku Dmax-2400, RIGAKU, Japan) with Cu Kα radiation. Scanning electron microscopy (SEM) was performed on a field emission SU8010N instrument, operating at an accelerating voltage of 10 kV. The surface compositions and valence states of Ti and Fe in VTM were characterized by an X-ray photoelectron spectroscopy (AXIS Supra, Kratos Analytical Ltd, UK) with Al Kα radiation.

3. Results and Discussion

3.1. Reactions of the VTM Reduction System

Owing to the low vanadium content of VTM in this raw material, thermodynamic research on the vanadium phase was not conducted in this work but has been reported in many previous studies.^14,29,30) Therefore, this work focuses on the iron and titanium phases in VTM. The reduction reaction of ordinary iron ore under CO or H₂ atmospheres can be described by Eqs. (1)–(8), as shown in Table 2. Similarly, the most likely reaction in the VTM oxidized pellets in a reducing atmosphere of CO or H₂ is described by Eqs. (9)–(18). The VTM reaction system should also include reactions of the gasification gas Eqs. (19)–(24). It can be seen that because of the close symbiosis of titanium and iron, the reduction of VTM is more complicated than that of ordinary iron ore.

Table 2. Reduction reaction equation in the reduction system of VTM.

NO.	Reaction equations
1	3Fe2O3 + CO = 2Fe3O4 + CO2
2	3Fe2O3 + H2 = 2Fe3O4 + H2O
3	Fe3O4 + CO = 3FeO + CO2
4	Fe3O4 + H2 = 3FeO + H2O
5	1/4Fe3O4 + CO = 3/4Fe + CO2
6	1/4Fe3O4 + H2 = 3/4Fe + H2O
7	FeO + CO = Fe + CO2
8	FeO + H2 = Fe + H2O
9	Fe2TiO5 + CO = Fe2TiO4 + CO2
10	Fe2TiO5 + H2 = Fe2TiO4 + H2O
11	Fe2TiO4 + CO = FeTiO3 + Fe + CO2
12	Fe2TiO4 + H2 = FeTiO3 + Fe + H2O
13	2FeTiO3 + CO =FeTi2O5 + Fe + CO2
14	2FeTiO3 + H2 = FeTi2O5 + Fe + H2O
15	FeTiO3 + CO = TiO2 + Fe + CO2
16	FeTiO3 + H2 = TiO2 + Fe + H2O
17	3/5FeTi2O5 + CO = 2/5Ti3O5 + 3/5Fe + CO2
18	3/5FeTi2O5 + H2 = 2/5Ti3O5 + 3/5Fe + H2O
19	2CO = CO2+C
20	CO+3H2=CH4+H2O
21	CO+H2=C+H2O
22	CO+H2O=CO2+H2
23	CH4=C+2H2
24	CH4+3CO2 = 4CO (g) + 2H2O

Figure S3 shows the Δ_rG_θ of the main reaction in the VTM reduction system at 273–2073 K as a function of temperature. For most reactions, as the temperature increases, the Gibbs free energy decreases to less than zero. In the range of 400–1600 K, the Δ_rG_θ of the reaction in Eq. (1) is smaller than that in Eq. (9). Therefore, in the range of 400–1600 K, under a CO atmosphere, Fe₂TiO₅ is more difficult to be reduced than Fe₂O₃. In the reaction in Eq. (7), Δ_rG_θ is smaller than in Eqs. (11), (13), (15), and (17). Therefore, Fe₂TiO₄, FeTiO₃, and FeTi₂O₅ are more difficult to be reduced than FeO in the range of 400–1900 K, and VTM is more difficult to be reduced than the equivalent ordinary iron ore. In addition, the value of Δ_rG_θ shown in Fig. S3 indicates that Fe₂TiO₅ is more easily reduced in the VTM oxidation pellets than Fe₂TiO₄.

Through thermodynamic calculations, the dominant regions and reduction curves of each phase in the VTM were plotted. Figure 4(a) shows the change in CO equilibrium concentration with temperature during the reduction of iron oxides, and Fig. 4(b) shows the change in H₂ equilibrium concentration with temperature during the iron oxide reduction process. At the stage of FeO reduction to metallic iron, the equilibrium concentration of CO was approximately 72% at 1400 K, and the theoretical thermodynamic conversion of CO is approximately 28% at 1400 K, which is the case for ordinary iron concentrates. However, when the FeTiO₃ in VTM was reduced to metallic iron, the equilibrium concentration of CO increased sharply to 91% at 1400 K, the theoretical thermodynamic conversion rate of CO was only approximately 9% at 1400 K, and the maximum gas utilization rate was only approximately 6%. It is difficult to achieve thermodynamic equilibrium in the actual process, and the gas utilization rate will be lower. This leads to a lower efficiency in the reduction process and a significant increase in production costs. This is also the primary reason why VTM is more difficult to reduce than ordinary iron concentrates. The Δ_rG_θ of VTM in 973 K–1373 K are shown in Tables S1–S4.

Fig. 4.

Relationship between the equilibrium concentration of the gasification gas and temperature during the reduction of VTM in (a) CO; and (b) H₂. (Online version in color.)

3.2. Thermodynamic Calculation: Effects of Temperature on Equilibrium Compositions

First, we explored the effect of the reduction temperature. The other reaction conditions are listed in Table S5 and were fixed. We analyzed the equilibrium composition under the five reduction temperatures of 973, 1073, 1173, 1273, and 1373 K. The effect of the reduction temperature on the equilibrium yield of the solid compounds and gas, is shown in Tables S6 and S7. Figure 5 shows the effect of the reduction temperature on the composition of the equilibrium system. As the reduction temperature increased, the amount of metallic iron produced gradually increased. When the reduction temperature increased to 1373 K, the amount of metallic iron stabilized. In addition, the mass of the residue decreased with increasing temperature. When the reduction temperature increased to 1373 K, the mass of the residue tended to stabilize. It should be noted here that when the reduction temperature was 973 K, some carbon was generated. When the reduction temperature increased to 1073 K, carbon was no longer generated. This is due to the reaction that generates carbon. Eqs. (19) and (21) represent an exothermic reaction, although the reaction to generate carbon from the cracking of CH₄ Eq. (23) is endothermic. However, the reaction Eq. (20) that generates CH₄ is exothermic. As the reduction temperature increased, the amount of CH₄ generated gradually decreased. Therefore, the reaction to generate carbon was suppressed as the reduction temperature increased.

Fig. 5.

Effects of temperature on equilibrium compositions. (Online version in color.)

When the reduction temperature was 973 K, a large amount of Fe₂TiO₄ was produced. At this temperature, MgTiO₃ reacted with FeO to form MgO and FeTiO₃. The generated MgO reacted with SiO₂ and Al₂O₃ to form Mg₂SiO₄ and Mg₄Al₁₀Si₂O₂₃. At the same time, the generated FeTiO₃ was converted into Fe₂TiO₄. On the one hand, Fe₂O₃ reduction takes precedence over FeTiO₃ reduction. The Δ_rG_θ of the side reactions Eqs. (19), (21), (23), such as C and CH₄, is smaller than that of the FeTiO₃ reduction Eq. (15), resulting in FeTiO₃ not being reduced. On the other hand, the reduction of Fe₂O₃ was carried out in steps, and the Δ_rG_θ of the side reactions Eqs. (19), (21), (23), such as C and CH₄, is also smaller than that of FeO reduction Eqs. (7) and (8), which led to a backlog of FeO. In addition, FeSiO₃ existed in the time-equilibrium product. When the reduction temperature rose to 1073 K, the degree of FeO backlog reduced, and FeSiO₃ was no longer generated. More importantly, at this temperature, the Δ_rG_θ of the Fe₂TiO₄ formation reaction is very small. This series of factors provided the conditions for the formation of Fe₂TiO₄. When the reduction temperature increased to 1073 K, the side reactions were suppressed. Not only was Fe₂TiO₄ not formed, but the FeTiO₃ was partially reduced. However, TiO₂ did not exist stably and reacted with MnO to form MnTiO₃. When the reduction temperature further increased to 1173 K, FeTiO₃ was completely reduced. When the reduction temperature increased to 1273 K, MgTi₂O₅ was generated. When the reduction temperature increased to 1373 K, Mg₂SiO₄ no longer formed, but the amount of MgTi₂O₅ increased. For vanadium, V₂O₃ could react with FeO to form FeV₂O₄ at low temperatures. As the reduction temperature increased, FeV₂O₄ no longer formed, and vanadium existed as V₂O₃. This is mainly because the reaction to form FeV₂O₄ is exothermic.

Figure 5 also shows the effect of the reduction temperature on the equilibrium yield of the gaseous substances. Because the generation of CH₄ is an exothermic reaction, the mole fraction of CH₄ in the gas phase decreases with increasing temperature. When the temperature increased to 1273 K, the mole fraction of CH₄ was only 0.00006. When the reaction temperature was 973 K, even if some H₂O was generated, the mole fraction of H₂O was still less than that of CO₂. This shows that the reduction ability of CO is stronger than that of H₂ at 973 K. When the temperature increased to 1173 K, the mole fraction of H₂O was greater than that of CO₂, indicating that at this temperature, the reduction ability of H₂ is stronger than that of CO.

3.3. Thermodynamic Calculation: Effects of the Gasification Gas Flow Rate on Equilibrium Compositions

We also explored the effect of the gasification gas flow. The other reaction conditions are listed in Table S8 and have been fixed. Figure 6 shows the relationship between the gasification gas flow rate and the metal iron production, residue mass, and gas phase composition. The equilibrium production of metallic iron increases with the increase in gasification gas flow rate, until the flow rate reaches 5 L/min, after which the production amount tends to be stable. The mass of the residue decreases with the increase in gasification gas flow rate, until the reducing degree of the pellets becomes stable after the flow rate increases to 5 L/min. The effect of the gasification gas flow rate on the equilibrium production of solid compounds is also shown in Table S9. When the gasification gas flow rate is 2 L/min, MgTiO₃ is completely consumed. Because Fe₂O₃ is reduced to form FeO, and MgTiO₃ reacts with FeO to form MgO and FeTiO₃, the generated MgO reacts with SiO₂ and Al₂O₃ to form Mg₂SiO₄ and Mg₄Al₁₀Si₂O₂₃.

Fig. 6.

Effects of gasification gas flow on equilibrium compositions. (Online version in color.)

When the reduction temperature is 1273 K, the Δ_rG_θ of the FeTiO₃ being reduced to TiO₂ is smaller than that of the FeTiO₃ being reduced to FeTi₂O₅. According to the Δ_rG_θ of each reduction reaction of Fe₂O₃, it can be concluded that the reduction of Fe₂O₃ is not a one-step process but a step-by-step process. The specific process occurs because the Δ_rG_θ of the Fe₂O₃ reduction to Fe₃O₄ is smaller than the Δ_rG_θ of Fe. Fe₂O₃ is first reduced to Fe₃O₄. Then, Fe₃O₄ is reduced to FeO instead of Fe, and FeO is finally reduced to Fe. When the reduction temperature was 1273 K, whether H₂ or CO was used as the reducing agent, the Δ_rG_θ at each stage of the Fe₂O₃ reduction was smaller than the Δ_rG_θ of FeTiO₃ reduction. Therefore, when the gasification gas flow was small and the gasification gas was insufficient, Fe₂O₃ was preferentially reduced. When the gasification gas flow was increased to 3 L/min, a small amount of FeTiO₃ was reduced to obtain TiO₂ and Fe. However, TiO₂ does not exist stably but reacts with MnO to form MnTiO₃. This is because the Δ_rG_θ of TiO₂ reacting with MnO is less than zero, and the reaction occurs easily. When the flow of gasification gas was increased to 4 L/min, FeTiO₃ was completely reduced, and the amount of TiO₂ increased. MgTi₂O₅ and TiO₂ were formed in the equilibrium product. Because the Δ_rG_θ of MnTiO₃ was less than that of MgTi₂O₅, MnTiO₃ preferentially formed. When the flow of the gasification gas was small, the V₂O₃ reacted with FeO to form FeV₂O₄. As the flow of gasification gas increased, FeV₂O₄ was no longer generated.

3.4. Thermodynamic Calculation: Effects of the Gasification Gas Composition on Equilibrium Compositions

We also explored the effect of the gasification gas composition at gas flow rates of 3 L/min and 5 L/min, respectively. The reaction conditions are listed in Table S10. This work analyzes the effect of the gasification gas carbon-hydrogen ratio (n (CO) /n (H₂)) on the VTM reduction equilibrium system, under the conditions of gasification gas flows of 3 L/min and 5 L/min.

When the gasification gas flow rate is 5 L/min, the effect of the carbon-hydrogen ratio (the ratio of the moles of CO to the moles of H₂ in the gasification gas, n (CO) /n (H₂)) on the equilibrium production of iron and the composition of the solid phase material are shown in Table S11. When the gasification gas flow rate was 5 L/min, with the gasification gas carbon-hydrogen ratio changing, there was no change in the amount of iron produced. This phenomenon indicates that when the flow rate is 5 L/min, the gasification gas is sufficient. Similar to the results of studies by Jeongseog,³¹⁾ under the condition that the gasification gas is sufficient, the difference between the reducing capacities of H₂ and CO is not the influencing factor in the reduction reaction; the carbon-hydrogen ratio has no detailed effect on the reaction.

Therefore, we reduced the gasification gas flow rate under the calculation conditions to 3 L/min. Figure 7 shows the effect of the carbon-hydrogen ratio on the composition of the equilibrium system. As the carbon-hydrogen ratio increases, the equilibrium production of metallic iron decreases. This is because H₂ has a stronger reducing ability than CO when the reduction temperature is 1273 K. The effect of changes in the carbon-hydrogen ratio on the equilibrium yield of solid compounds when the gasification gas flow rate was 3 L/min is shown in Table S12. FeTiO₃ cannot be completely reduced when the H₂ content in the gasification gas is low. As the H₂ content increased, more FeTiO₃ was reduced. Therefore, to enhance the reducing ability of the gasification gas, the carbon-hydrogen ratio should be reduced as much as possible, that is, the H₂ content should be increased.

Fig. 7.

Effects of gasification gas composition on equilibrium compositions. (Online version in color.)

We investigated the influence of reaction pressure (1–6 atm) on the reaction system. The reaction conditions are listed in Table S13. The effects of the reaction pressure on the equilibrium compositions of the solid compounds and gas products are shown in Tables S14 and S15. The reaction pressure has almost no effect on the equilibrium production of iron and the quality of the residue. This is because, whether H₂ or CO is used as the reducing agent, Fe₂O₃ and FeTiO₃ reduction reactions are all equal volume reactions; therefore, the reaction pressure does not affect the equilibrium of the reaction.

3.5. Experimental Analysis

3.5.1. Effects of Temperature on the Composition of Reaction Products and Reduction Degree

Figure 8(a) shows the final reduction degree and iron-containing product composition for reducing VTM oxidized pellets at different temperatures for 240 min, and the experimental conditions are shown in Table S16. Figure 8(b) is the XRD spectrum of the reduced product. As the reduction temperature increases, the final reduction degree of VTM continues to increase. When the reduction temperature reaches 1273 K, the final reduction degree of VTM tends to be stable, which is consistent with the thermodynamic calculation results. The phase of the reduced product also changes at different temperatures. For example, the main mineral phases at lower temperatures are Fe₃O₄ and Fe₂TiO₄. As the temperature increases, the content of FeO increases. The FeO step is reduced to Fe, and the FeTi₂O₄ exists stably. When the temperature reaches 1073 K, Fe₂TiO₄ is reduced to FeTiO₃. As the temperature further increases to 1173 K, the FeTiO₃ phase disappears, and the main mineral phases of the reduction product are Fe and FeTi₂O₅. With increasing temperature, the content of FeTi₂O₅ gradually decreases, which is consistent with the reaction path of the thermodynamic analysis of Fig. 4. Based on these experimental results, it can be concluded that a high temperature is beneficial for the reduction of VTM oxidized pellets. Based on this, we recommend that the reduction temperature of VTM oxidized pellets is set to 1273–1373 K in actual industrial production. The photos of the VTM metalized pellets after reduction from 973 K to 1373 K are shown in Figs. 9(a)–9(e). As can be seen from Fig. 9(a), when the reduction temperature is 973 K, the outer surface of the VTM metalized pellets is heavily wrapped in black powder. As can be seen from Fig. 9(b), when the reduction temperature increases to 1073 K, the amount of black powder generated is reduced but still exists. When the reduction temperature increases to 1173 K, the black powder is no longer generated and is not tightly adhered to the surface of the pellets. The XRD pattern of the powder is shown in Fig. 9(f), and carbon was found in the black powder. The carbonation reaction occurred at the reduction temperatures of 973 and 1073 K. The result of the simulation calculation is that carbon is generated only at the reduction temperature of 973 K. There are some discrepancies with the experimental research results, which may be because the reactor used in the actual reduction is made of steel. Steel has a catalytic effect on the carbon precipitation reaction. It was further verified that the reduction reaction is carried out progressively from the outer layer of the pellet to the core. The uneven reduction rate in each layer of the pellet causes stress in the pellet. As the reduction temperature increases, the reduction rate is accelerated, and the difference between the reduction rate of the outer layer of the pellet and that of the core increases. The resulting stress is enhanced; therefore, as the reduction temperature increases, the cracks in the VTM metalized pellets increase and become deeper. At the same time, as the reduction temperature increases, the density of the VTM metal pellets decreases, and the structure becomes increasingly looser. When the reduction temperature is 1373 K, the shell of the VTM metalized pellets begins to show obvious flaking.

Fig. 8.

Reduction of VTM oxide pellets at different temperatures for 240 min (a) Final reduction degree and product composition; (b) XRD pattern. (Online version in color.)

Fig. 9.

(a–e) Photos of VTM metalized pellets at 973 K, 1073 K, 1173 K, 1273 K, and 1373 K, (f) XRD patterns of black powder products at 973 K (red line), and 1073 K (blank line). (Online version in color.)

Figure 10 shows the morphology of the VTM metalized pellets at the reduction temperature of 973–1373 K. As the reduction temperature increases, the VTM metalized pellets gradually exhibit a metallic luster. It can be seen from Fig. 10 that when the reduction temperature is 973 K, the metallic iron crystals in the reduced VTM pellets gradually precipitate as a sheet-like structure, and the metallic iron wafers overlap each other with a certain degree of crystalline connection. When the reduction temperature rises to 1073 K, the metallic iron precipitates as fibrous iron whiskers. When the reduction temperature is increased to 1173 K, more metallic iron crystals are precipitated, which causes the fibrous metallic iron whiskers to fuse with each other, and the fibrous iron whisker structure disappears. When the reduction temperature is increased to 1273 K, the VTM metalized pellets have softened to a certain extent. When the reduction temperature is increased to 1373 K, the metallic iron aggregates in the metallization of the VTM and exhibits a granular structure, and the connection between the grains is weak.

Fig. 10.

SEM image of VTM metallized pellets (a) 973 K; (b) 1073 K; (c) 1173 K; (d) 1273 K; (e) 1373 K.

3.5.2. Effects of the Flow of Gas on the Composition of Reaction Products and Reduction Degree

Figure 11(a) shows the phase changes and final degree of reduction in the VTM reduction process at different gasification gas flows, and the experimental conditions are shown in Table S17. It can be seen from the figure that as the gasification gas flow increases, the final reduction degree of VTM increases. When the gasification gas flow rate increased to 5 L/min, the degree of reduction became stable. The thermodynamic calculation results showed that the reduction degree stabilized after the gasification gas flow rate reached 4 L/min. The difference is because the reduction process is also affected by kinetics. Figure 11(b) shows the XRD pattern of the VTM reduction products. FeTiO₃ is present in the VTM metalized pellets when the gasification gas flow rate is 2 L/min and 3 L/min, and the gasification gas flow continues to increase.

Fig. 11.

Reduction of VTM pellets under different gasification gas flows for 240 min (a) Final reduction degree and iron-containing product composition; (b) XRD pattern. (Online version in color.)

3.5.3. Effects of the Gasification Gas Composition on the Composition of Reaction Products and Reduction Degree

As with the thermodynamic calculations, we explored the effect of the carbon-hydrogen ratio of the gasification gas when the flow rate of the gasification gas was 3 L/min and 5 L/min. When the reaction gas flow rate was 5 L/min, the reduction degree of VTM oxidized pellets was not affected by the carbon-hydrogen ratio, which was consistent with the results of thermodynamic calculations.

Figure 12(a) shows the effect of the carbon-hydrogen ratio of the gasification gas on the phase composition and the final degree of reduction of the VTM oxidized pellet reduction system at a flow rate of 3 L/min, and the experimental conditions are shown in Table S18. The final reduction degree of the pellets indicates that the reducing ability of H₂ is stronger than that of CO, which is also consistent with the results of the thermodynamic calculations. Figure 12(b) shows the XRD pattern of the VTM oxidized pellets reduced by gasification gas with different gasification gas carbon-hydrogen ratios when the gasification gas flow rate was 3 L/min. It can be seen from the figure that when the carbon-hydrogen ratio is relatively large, titanium in the VTM metalized pellets exists in the form of FeTiO₃; when the carbon-hydrogen ratio is reduced to 2:1, FeTiO₃ is reduced to form FeTi₂O₅, which is consistent with the results of the thermodynamic calculations of Fig. 4.

Fig. 12.

Reduction of VTM pellets under different gasification gas composition for 240 minutes (a) Final reduction degree and iron-containing product composition; (b) XRD pattern. (Online version in color.)

3.5.4. Dynamic Study of the Reduction Process by XPS

At 1273 K, the dynamic process of the reduction of VTM oxide pellets was verified by XPS at different reaction times. The flow rate of the gasification gas was 5 L/min, the carbon-hydrogen ratio was 1:1, and the diameter of the VTM oxidized pellets was 10–12 mm. The following six samples were subjected to XPS testing: VTM oxidized pellets, VTM oxidized pellets reduced for 10 min, VTM oxidized pellets reduced for 50 min, VTM oxidized pellets reduced for 90 min, and VTM oxidized pellets completely reduced (240 min). In this way, the valence changes of Fe and Ti during the reduction of vanadium-titanium magnetite oxide pellets were explored.

Figure 13 show the valence state distribution of Fe on the VTM (Fig. 13(a)) and oxidized VTM (Fig. 13(b)), which are reduced by gasification gas for different time (10 min (Fig. 13(c))\ 50 min (Fig. 13(d))\ 90 min (Fig. 13(e))\ 240 min (Fig. 13(f)), and the results of peak fitting. According to the electronic binding energy comparison table of Fe, the electron binding energy of the 2p_3/2 orbitals of Fe²⁺ is approximately 710 eV, and the electron binding energy of the 2p_3/2 orbitals of Fe³⁺ is approximately 713 eV. The relative concentration of Fe²⁺ on the surface of the raw material was higher than that of Fe³⁺, and the Fe²⁺/Fe³⁺ ratio was approximately 1.93 (Table 3). After oxidation, the relative content of Fe³⁺ increased significantly, and the Fe²⁺/Fe³⁺ ratio decreased to 0.77, which also conformed to the distribution law of oxidized Fe valence. As the reduction reaction progressed, the Fe²⁺/Fe³⁺ ratio gradually increased, the relative content of Fe²⁺ continued to increase, and the relative content of Fe³⁺ continued to decline. During this process, the VTM was gradually reduced to elemental iron. Figure 14 show the valence state distribution of Ti on the VTM (Fig. 14(a)) and oxidized VTM (Fig. 14(b)), which are reduced by gasification gas for different times (10 min (Fig. 14(c))\50 min (Fig. 14(d))\90 min (Fig. 14(e))\240 min (Fig. 14(f)), and the results of peak fitting. According to the electronic binding energy comparison table of Ti, the electronic binding energy of the 2p_3/2 orbital of Ti⁴⁺ is approximately 458 eV. It can be seen from the figure that the fine Ti spectrum of the raw material has more noise signals. That is because the relative content of the Ti element on the surface is low; therefore, the detected signal intensity is very low, and as the reduction reaction proceeds, the Ti fine spectrum is shown as a single peak (there is a slight interference from the O₂ spectrum at 459 eV), and the Ti element always exists in the form of Ti⁴⁺ on the surface, which is consistent with the results of the thermodynamic calculations. During the entire reduction process, the Ti⁴⁺ oxides are always stable.

Fig. 13.

XPS pattern of Fe in (a) raw materials of VTM; (b) oxidized VTM; the oxidized pellets are reduced by gasification gas for (c) 10 min; (d) 50 min; (e) 90 min; and (f) 300 min. (Online version in color.)

Table 3. Changes in the concentrations of Fe and Ti in VTM during the reduction process.

Samples	Relative concentration of elements /%		Valence distribution of Fe
	Fe	Ti	Fe2+	Fe3+	Fe2+/Fe3+
VTM	0.85	0	6725.16	3488.81	1.93
VTM oxidized pellets	6.15	3.45	12039.45	15627.08	0.77
10 min reduced	5.16	3.14	8939.47	9796.65	0.91
50 min reduced	4.49	2.08	14229.99	11094.93	1.28
90 min reduced	3.46	5.20	13116.87	9493.87	1.38
Complete reduction	1.13	5.60	10333.22	5115.32	2.02

Fig. 14.

XPS pattern of Ti in (a) raw materials of VTM; (b) oxidized VTM; the oxidized pellets are reduced by gasification gas for (c) 10 min; (d) 50 min; (e) 90 min; and (f) 300 min. (Online version in color.)

Table 3 shows the relative concentrations of Fe and Ti on the surface of each sample during the reduction process. The surface of the raw VTM was covered with more impurities. The relative concentration of Fe was very low, while Ti was present. After VTM was sintered by oxidation, the relative concentrations of Fe and Ti on the surface increased significantly. During the reduction of the VTM, as the reduction time increased, the concentration of Fe on the surface gradually decreased. This is due to the continuous progress of the reduction reaction, and the iron oxide in the VTM was gradually reduced to metallic iron. Therefore, the relative concentration of Fe on the surface decreased. In the process of the reduction reaction, the relative concentration of the Ti element decreased first and then increased. This can be understood in that when the Fe element was partially reduced, Ti remained in the eutectic structure with Fe, and the relative concentration of Ti gradually decreased. When the Fe was reduced to a certain extent, the Ti element broke away from the eutectic structure with Fe, and the relative concentration of Ti on the surface increased significantly.

4. Conclusions

Non-blast furnace processes, especially gas-based reduction processes have great potential. The integration of coal gasification and the direct reduction of VTM that opens an effective way to develop gas-based direct reduction technology. In this process, gasification gas was used as the reducing agent. In this work, the direct reduction of VTM oxidized pellets by the gasification gas was systematically analyzed through experimental research and thermodynamic calculations.

(1) We give all possible reactions of the system in which VTM is reduced by gasification gas and the corresponding Δ_rG_θ.

(2) VTM oxidized pellets are more difficult to reduce, and the reduction process is more complicated. The reduction process of VTM oxidized pellets are clearly described.

(3) With an increase in the reduction temperature, the reduction degree of VTM oxidized pellets increases gradually, and the reduction system approaches equilibrium when the reduction temperature rises to 1273 K.

(4) This study lays a theoretical foundation for the integration of flue gas gasification and the direct reduction of VTM. The reduction temperature should be approximately 1273–1373 K, steam should be used for coal gasification as much as possible to obtain hydrogen-rich gasification gas.

Supporting Information

Integration of flue gas gasification and the direct reduction of VTM (Fig. S1); Gibbs free energy change curves of reactions in VTM (Fig. S2); The gas-based direct reduction reaction of VTM (Table S1–S4); Calculation condition settings (Table S5, S8, S10, S13); Effect of reduction temperature/ gas flow/ gas composition/ reaction pressure on equilibrium yield (Table S6–S7, S9, S11–S12, S14–S15); Experimental conditions (Table S16–S18).

This material is available on the Journal website at https://doi.org/10.2355/isijinternational.0_ISIJINT-2022-182.

Acknowledgements

This research was funded by the National Key Research and Development Plan of China (2018YFC1901503 and 2018YFC1901505), Shanxi Unveiling Bidding Project (20191101007), National Natural Science Foundation of China (51672006), and the Ministry of Land and Resources Public Welfare Industry Research Project (201511062-02).

References

• 1)   W.  Li,  G.  Fu,  M.  Chu and  M.  Zhu: Ironmaking Steelmaking, 48 (2021), 33. https://doi.org/10.1080/03019233.2020.1721955
• 2)   L.  Shi,  Y.  Zhen,  D.  Chen,  L.  Wang and  T.  Qi: ISIJ Int., 58 (2018), 627. https://doi.org/10.2355/isijinternational.ISIJINT-2017-515
• 3)   W.  Zhao,  M.  Chu,  H.  Guo,  Z.  Liu,  B.  Yan and  P.  Li: ISIJ Int., 61 (2021), 146. https://doi.org/10.2355/isijinternational.ISIJINT-2020-309
• 4)   Z.  Pang,  X.  Lv,  Y.  Jiang,  J.  Ling and  Z.  Yan: Metall. Mater. Trans. B, 51 (2020), 722. https://doi.org/10.1007/s11663-019-01756-0
• 5)   G.  Cheng,  X.  Xue,  T.  Jiang and  P.  Duan: Metall. Mater. Trans. B, 47 (2016), 1713.
• 6)   K.  Gao,  K.  Jiao,  J.  Zhang and D. E: ISIJ Int., 60 (2020), 1902. https://doi.org/10.2355/isijinternational.ISIJINT-2020-004
• 7)   W.  Zhao,  M.  Chu,  H.  Wang,  Z.  Liu,  J.  Tang and  Z.  Ying: ISIJ Int., 58 (2018), 1989. https://doi.org/10.2355/isijinternational.ISIJINT-2018-175
• 8)   S. B.  Kuang,  Z. Y.  Li,  D. L.  Yan,  Y. H.  Qi and  A. B.  Yu: Miner. Eng., 63 (2014), 45. https://doi.org/10.1016/j.mineng.2013.11.002
• 9)   M.  Zhou,  T.  Jiang,  X.  Ding,  S.  Ma,  G.  Wei and  X.  Xue: J. Therm. Anal. Calorim., 136 (2019), 885. https://doi.org/10.1007/s10973-018-7687-8
• 10)   R.  Beheshti,  J.  Moosberg-Bustnes,  M. W.  Kennedy and  R. E.  Aune: Ironmaking Steelmaking, 43 (2016), 394. https://doi.org/10.1080/03019233.2015.1104072
• 11)   H.-b.  Zuo,  C.  Wang,  J.-j.  Dong,  K.-x.  Jiao and  R.-s.  Xu: Int. J. Miner. Metall. Mater., 22 (2015), 688. https://doi.org/10.1007/s12613-015-1123-x
• 12)   H.  Long,  T.  Chun,  P.  Wang,  Q.  Meng,  Z.  Di and  J.  Li: Metall. Mater. Trans. B, 47 (2016), 1765. https://doi.org/10.1007/s11663-016-0644-7
• 13)   W.  Yu,  Q.-y.  Tang,  J.-a.  Chen and  T.-c.  Sun: Int. J. Miner. Metall. Mater., 23 (2016), 1126. https://doi.org/10.1007/s12613-016-1331-z
• 14)   J. W.  Chen,  Y.  Jiao and  X. D.  Wang: Int. J. Miner. Metall. Mater., 26 (2019), 822. https://doi.org/10.1007/s12613-019-1795-8
• 15)   D.  Wang,  J.  Xu,  K.  Ma,  Y.  Xu,  J.  Dang,  M.  Kou,  X.  Lv and  L.  Wen: Int. J. Hydrog. Energy, 42 (2017), 14047. https://doi.org/10.1016/j.ijhydene.2017.04.065
• 16)   X.  Jiang,  F.  Long,  L.  Wang,  Y.  Zhou,  H.  Zheng,  Q.  Gao and  F.  Shen: ISIJ Int., 61 (2021), 129. https://doi.org/10.2355/isijinternational.ISIJINT-2020-303
• 17)   J.  Dai,  D.  Dong,  Y.  Ni,  F.  Hong,  S.  Zhang,  Y.  Zhang and  L.  Ding: J. Nat. Gas Geosci., 5 (2020), 169. https://doi.org/10.1016/j.jnggs.2020.07.004
• 18)   J.  Xie,  L.  Xin,  X.  Hu,  W.  Cheng,  W.  Liu and  Z.  Wang: J. Clean. Prod., 250 (2020), 119487. https://doi.org/10.1016/j.jclepro.2019.119487
• 19)   J.  Chen,  W.  Chen,  Y.  Jiao and  X.  Wang: Energies, 12 (2019), 496. https://doi.org/10.3390/en12030496
• 20)   J.  Chen,  W.  Chen,  R.  Ji,  Y.  Jiao and  X.  Wang: Int. J. Green Energy, 16 (2019), 1144. https://doi.org/10.1080/15435075.2019.1653881
• 21)   Y.  Sun,  J.  Chen and  Z.  Zhang: Energy, 167 (2019), 688. https://doi.org/10.1016/j.energy.2018.11.019
• 22)   Y.  Sun,  S.  Seetharaman and  Z.  Zhang: Energy, 149 (2018), 792. https://doi.org/10.1016/j.energy.2018.02.119
• 23)   Y.  Sun,  Q.  Liu,  H.  Wang,  Z.  Zhang and  X.  Wang: Bioresour. Technol., 223 (2017), 1. https://doi.org/10.1016/j.biortech.2016.10.028
• 24)   A. K.  Shukla: Trans. Indian Inst. Met., 72 (2019), 767. https://doi.org/10.1007/s12666-018-1529-z
• 25)   M.  Sun,  J.  Zhang,  K.  Li,  G.  Wang,  H.  Wang and  Q.  Wang: Metall. Res. Technol., 115 (2018), 411. https://doi.org/10.1051/metal/2018064
• 26)   W.  Chen,  Z.  Dong,  Y.  Jiao,  L.  Liu and  X.  Wang: Crystals, 11 (2021), 188.
• 27)   Y.  Harada,  N.  Saito and  K.  Nakashima: Metall. Mater. Trans. B, 52 (2021), 968. https://doi.org/10.1007/s11663-021-02069-x
• 28)   C. W.  Bale,  E.  Bélisle,  P.  Chartrand,  S. A.  Decterov,  G.  Eriksson,  A. E.  Gheribi,  K.  Hack,  I. H.  Jung,  Y. B.  Kang,  J.  Melançon,  A. D.  Pelton,  S.  Petersen,  C.  Robelin,  J.  Sangster,  P.  Spencer and  M. A.  Van Ende: Calphad, 55 (2016), 1. https://doi.org/10.1016/j.calphad.2016.07.004
• 29)   R. H.  Sievwright,  J. J.  Wilkinson,  H. S. C.  O’Neill and  A. J.  Berry: Contrib. Mineral. Petrol., 172 (2017), 62. https://doi.org/10.1007/s00410-017-1385-6
• 30)   R. B.  Wanty and  M. B.  Goldhaber: Geochim. Cosmochim. Acta, 56 (1992), 1471. https://doi.org/10.1016/0016-7037(92)90217-7
• 31)   J.  Oh and  D.  Noh: Fuel, 196 (2017), 144. https://doi.org/10.1016/j.fuel.2016.10.125