Formation of [Mg_1-x,Fe_x]O·Fe₂O₃ in Solid-state Reactions between MgO and Fe₂O₃ in the Fe₂O₃-rich System

Abstract

Aiming to better understand the effect of MgO on sintering process of iron ores, the formation of [Mg_1-x,Fe_x]O·Fe₂O₃ in solid-state reactions between MgO and Fe₂O₃ was studied. Experiment was carried out in air from 873 K to 1573 K by MgO mixing with Fe₂O₃. X-ray diffraction, optical microscope, scanning electron microscopy and energy-dispersive spectroscopy were used to characterize the phase change of the sintered samples. The content of ferrous ion in the sintered samples was determined by potassium dichromate titration for distinguishing the MgO·Fe₂O₃ (x=0) and the Fe₃O₄ (x=1). Thermogravimetric and differential scanning calorimeter test was carried out in air by MgO mixing with Fe₂O₃ to investigate the thermal decomposition of Fe₂O₃. The results show that the reactions between the Fe₂O₃ and MgO in air formed first the magnesium ferrite at 1073 K, subsequently magniferous magnetite appeared at 1173 K, resulting that the thermal decomposition of Fe₂O₃ was carried out at a lower temperature than that of its own self. The following conversion of the magniferous magnetite to the magnesium ferrite was also observed with the temperature increasing to 1482 K. It has been deduced that the [Mg_1-x,Fe_x]O·Fe₂O₃ is formed by the reaction between the prior formed MgO·Fe₂O₃ and the Fe₂O₃ in the heating-up process. It was obtained that amount of ferrous ion formed in sintering process is mainly related by the amount of MgO in raw materials and it’s diffusion rate. Therefore, adding MgO appropriately into raw material can be beneficial to improve the low temperature reduction degradation of iron ore sinter.

1. Introduction

Iron ore sintering is a process to form the bonding matrix by reaction between the iron ore and flux. In current metallurgical technology, an appropriate content of MgO is usually selected to provide a blast furnace slag for good fluidity and desulphurization properties.^1,2) In addition, the secondary hematite in iron ore sinter induces easily the reduction degradation, which can be markedly reduced with increasing of MgO content.^{3,4,5,6,7,8,9)} So, MgO has been a desirable constituent of blast furnace flux. However, excess MgO in iron ore sinter reduces the content of silico-ferrite of calcium and aluminium (SFCA)¹⁰⁾ and increases the total FeO content in iron ore sinter,¹¹⁾ the strength and reducibility of iron ore sinter will be also deteriorated. Therefore, the magnesioferrite plays an extremely important role in the development of an acceptable sinter product as a primary product of the reactions between MgO and Fe₂O₃ in solid state.

As early as 1931, Roberts had investigated the decomposition characteristics of Fe₂O₃ in the phase diagram of MgO–Fe₂O₃ system in air. He found that Fe₂O₃ can be decomposed into magnetite while the sintering temperature is below the decomposition temperature of Fe₂O₃ itself when blended MgO.¹²⁾ Then, the similar phenomena had been also found in the study of Richards and White,^13,14) and they hypothesized that the formation reaction of ferrous ion in the Fe₂O₃-rich system is forced by lowering of the free energy of magnetite formation due to the same spinel crystal structure between magnetite and MgO·Fe₂O₃. Although these explanations are no doubt thermodynamically correct, it is felt that they do not provide an approving evidence of decomposition kinetically. Meanwhile, no decomposition was detected at a low temperature for MgO·Fe₂O₃ by itself in the study of Blackman,¹⁵⁾ and he pointed out that the ferrous iron is formed in the system of MgO–Fe₂O₃ at relatively lower temperatures due to the preferential diffusion of iron ion.

MgO·Fe₂O₃ is formed by the counter diffusion of magnesium ion and ferric ion through a relatively rigid oxide lattice that was pointed by Carter,¹⁶⁾ who used pores as inert markers. However, the diffusing ion might not be ferric ion but ferrous ion was pointed out by Zubets based on (Mg, Fe)O formation in MgO side by diffusion coupling method.¹⁷⁾ Moreover, he also pointed that there is no oxygen diffusion through the gas phase, and the formation of ferrous ion is possibly attributed to partial electron compensation for the charge on the magnesium ion as a result of counter-diffusion. The reaction rate is decided by the initial percentage of iron in the mixture at sintering temperature,¹⁶⁾ based on that the concentration of vacancies increases with the amount of iron in the samples, where three magnesium ions would be replaced by two ferric ions and a vacancy. However, Paik¹⁸⁾ pointed out that the diffusion of magnesium ion in the rigid MgO lattice is a rate controlling reaction in solid-state reactions between MgO and Fe₂O₃.

On the other hand, the low-temperature reduction degradation (RD) of iron ore sinter can be improved by the addition of MgO-bearing materials as a basic flux constituent into raw materials in sintering or induration process of iron ore fine due to formation of magnesioferrite.^{11,19,20,21,22,23,24,25,26,27,28)} The mechanism of magnesioferrite formation is explained by that Mg²⁺ from MgO enters into the magnetite lattice to form the magnesioferrite by displacing Fe²⁺ ion in the magnetite, in where it has some discrepancy because the magnetite is not stable thermodynamically at the temperature.

Although a large number of researches indicated that the trace of ferrous ion had been found in solid-state reactions between MgO and Fe₂O₃ while the sintering temperature is below the decomposition temperature of Fe₂O₃ itself.^{11,12,13,14,15,17,20,21)} However, the formation mechanism of [Mg_1-x,Fe_x]O·Fe₂O₃ as the magnesioferrite has not been in-depth studied, for example how to form kinetically, and there is not much information available in literatures about the change of x in [Mg_1-x,Fe_x]O·Fe₂O₃ during the reaction. It is particularly important to study the formation of [Mg_1-x,Fe_x]O·Fe₂O₃ in solid-state reactions between MgO and Fe₂O₃, providing a better understanding as well as controlling it over the quality of iron ore sinter.

In present work, TG-DSC (Thermogravimetric and differential scanning calorimeter) was used to follow the mass and energy changes for investigation of MgO effect on the thermal decomposition of Fe₂O₃ in heating-up process, simultaneously XRD (X-ray diffraction), OM (optical microscope) and SEM-EDS (scanning electron microscopy and energy-dispersive spectroscopy) were used to determine the structure change of samples with various contents of MgO in constant temperature processes for understanding the effect of MgO on the [Mg_1-x,Fe_x]O·Fe₂O₃ formation. Furthermore, PDT (potassium dichromate titration) was used more to quantify the content of ferrous ion in the samples with various contents of MgO in sintering process for distinguishing of magnetite, magniferous magnetite and MgO·Fe₂O₃ as reaction products.

2. Experimental

2.1. Reaction Temperature Determination (TG-DSC)

In order to investigate the effect of MgO on the thermal decomposition of Fe₂O₃ in the heating-up process, different proportions of α-Fe₂O₃ and MgO (0, 5, 10, 20 wt pct MgO) were mixed to ensure thorough homogenization for TG-DSC samples. The chemical agents employed in the present work were of analytical grade supplied by Sinopharm Chemical Reagent Co. Ltd. The chemical agents were pre-sintered at 1473 K for 120 min respectively to remove the volatile impurities. After that, the pre-sintered chemical reagent was crushed into the small particles with diameter less than 50 μm. Each group of mixtures was homogenized by grinding in anhydrous ethanol for approximately 30 minutes using an agate mortar. The TG-DSC analysis was carried out by using Netzsch STA 449C (Netzsch-GmbH, Selb, Germany). The rate of heating was kept at 20 K/min from room temperature to 773 K, and then 5 K/min to 1673 K. The measurements were carried out in air with a constant flowrate of 100 cm³/min.

2.2. Sample Preparation

In order to investigate the formation of minerals in the samples, the content of MgO was selected within 0–20 wt pct for constant temperature sintering. To begin, the mixtures were pressed into tablets (~15 mm in diameter and ~2 mm in thickness) by using a briquetting machine at a constant pressure of 5 MPa for 2 min. Then, the pellets were sintered in Pt crucible in range of 873–1573 K in air with a flowrate that is maintained at 500 cm³/min for different time. Finally, the samples were quenched by liquid nitrogen after sintering.

2.3. Phase Determination (XRD, SEM and OM)

A part of each quenched samples were ground into fine powders with an agate mortar and pestle and then the powder XRD analysis was carried out using a Rigaku UItima IV X-ray diffractometer (Rigaku Corporation, Tokyo, Japan). The patterns were run with Cu Kα at 40 kV and 40 mA with a scanning speed of 10 degree in 2θ per minute. To consider that the XRD patterns can´t distinguish the magnetite (Fe₃O₄), magnesium ferrite (MgO·Fe₂O₃) and magniferous magnetite ([Mg_1-x,Fe_x]O·Fe₂O₃, 0<x<1) due to that they are the same crystal structure, so the magnesioferrite ([Mg_1-x,Fe_x]O·Fe₂O₃) is used in advance to express them in which the magnetite, magniferous magnetite and magnesium ferrite are corresponding x=1, 0<x<1 and x=0 respectively.

The obtained samples were also polished by mounting them into the ethylenediamine-doping epoxy resin for optical microscope to observe the mineral phase and microstructure. Then, SEM-EDS was performed using a scanning electron microscope (MLA250, FEI Hong Kong Co. Ltd., US), and the accelerating voltage is 15 kV, to characterize the phase change of the sintered samples.

2.4. Ferrous Ion Determination (PDT)

In order to investigate the changes of ferrous ion in the samples, the sample for XRD analysis was used simultaneous for quantitative analysis by PDT. To begin, m (approximate 0.2000 g) of the sample, 1 g of NaF and 2 g of NaHCO₃ were putted into 250 mL conical flask and blended homogeneously; Secondly, 15 mL of hydrochloric acid (ρ=1.19 g/mL) was added, and conical flask was immediately tightly plugged by a plug with a hose that connecting to the water for cutting off from air, and the conical flask was putted into a water bath (343 K) until the powder was completely dissolved; Thirdly, the plug was removed and 100 mL of NaHCO₃ saturated solution and 20 mL of mixed acid (V_H2SO4: V_H2PO4: V_H2O=1:1:5, ρ_H2SO4=1.84 g/mL and ρ_H2PO4=1.69 g/mL) were added immediately; After that, 8–10 drops of diphenylamine sulphonic acid sodium salt as an indicator solution were added; Finally, the solution was immediately titrated with potassium dichromate standard titration until it changed into purple, and the consumption of potassium dichromate standard titration was recorded. Each analysis of powder was repeated three times and then averaged the results. The content of ferrous ion in samples can be calculated according to the equation as following,   

$$w=\frac{V*H}{m}\text{*}100\text{\%}$$(1)

w, m and V are the mass percent of ferrous ion in samples (pct), the mass of the samples (g) and the titration volume (mL) respectively. In addition, H is a titer (g/mL), which indicates that the mass of ferrous ion could be oxidized by the potassium dichromate standard solution per milliliter.

3. Results

3.1. Effect of MgO on Formation Process of Minerals

3.1.1. Change of Mineral Composition

The XRD patterns and optical microstructures of the samples with various contents of MgO sintered at 1473 K for 60 min are shown respectively in Figs. 1 and 2. The XRD patterns show that with increasing of MgO from 4 wt pct to 20 wt pct, the intensity of magnesioferrite peaks increased and that of hematite peaks decreased until the content of MgO reached 20 wt pct completely disappeared, which agrees with the optical microstructures results as shown in Fig. 2. Meanwhile, although the peaks of magnesium oxide is hard to find in the powder XRD patterns of samples, due to the diffraction peaks of the magnesium oxide overlapping with that of magnesioferrite or hematite, a small amount of magnesium oxide peak was found in the XRD patterns of the sample with 20 wt pct MgO by magnifying the XRD patterns as shown in the right side of Fig. 1. And the magnesium oxide could be also found from microstructure photograph of the sample with 20 wt pct MgO.

Fig. 1.

XRD patterns of the samples with various contents of MgO sintered at 1473 K for 60 min.

Fig. 2.

Optical microstructures of the samples with various contents of MgO sintered at 1473 K for 60 min. (a) +4 wt pct MgO, (b) +6 wt pct MgO, (c) +10 wt pct MgO, and (d) +20 wt pct MgO. H-Hematite, M-Magnesioferrite or magnetite, O-Magnesium oxide, P-Pore.

3.1.2 TG-DSC Analysis

TG-DSC test was used to follow the mass and energy changes for investigating the thermal decomposition of Fe₂O₃ in heating-up process, as shown in Fig. 3. It can be found from the TG curves that the sample of Fe₂O₃ powder began to lose the mass only until the temperature is higher than 1662 K in air, but the mass loss of the sample with MgO was started from 1318 K. Moreover, the sample with MgO lost the mass rapidly at the early stage with increasing of temperature and then slightly later. It is worth nothing that the mass loss of the sample with 20 wt pct MgO terminated at about 1482 K, and thereafter the mass regained during the temperature increased to 1673 K.

Fig. 3.

TG-DSC curves of the sample with various contents of MgO. a-Pure Fe₂O₃, b-+5 wt pct MgO, c-+10 wt pct MgO, and d-+20 wt pct MgO.

There were no obvious exothermic and endothermic peaks in the DSC curves of the sample with various contents of MgO as shown in Fig. 3, which evidenced that exothermic or endothermic is not obvious in solid-state reactions between MgO and Fe₂O₃. The strong endothermic peak at 1662 K was found in the DSC curve of pure Fe₂O₃ that might correspond to thermal decomposition of Fe₂O₃ yielding magnetite. The DSC curves of the samples with 5 wt pct, 10 wt pct and 20 wt pct MgO have similar characteristics. The DSC curve of the sample with 20 wt pct MgO began into exothermic when the temperature is higher than 1170 K, while it transformed into endothermic over 1400 K, and into exothermic again when the temperature is higher than 1580 K. The reasons for this change will be explained later.

DTG (differential thermal gravity) curves of the sample with various contents of MgO are shown in Fig. 4. Firstly, the peak temperature of the DTG curve, which is defined as T_m (a temperature at the maximum change rate of mass loss) of sample, and the temperature of the start reaction (T_s) is expressed as a temperature of the intersection between the horizontal line of initial mass and the tangent of the TG curve on the T_m point. The T_m and T_s shifted to lower value when MgO increased from 0 to 20 wt pct as shown in Table 1. Therefore, it can be deduced that the T_m and T_s decrease with increasing of MgO in the sample in heating-up process. It proves that the MgO-added promoted the decomposition of Fe₂O₃ into magnetite, which accords with the binary phase diagram of MgO–Fe₂O₃.²⁹⁾ Finally, the DTG of sample with 20 wt pct MgO transformed into negative number when the temperature is higher than 1482 K, proving that the sample can regain the mass at the temperature range.

Fig. 4.

DTG curves of the sample with various contents of MgO.

Table 1. The T_s and T_m of the sample with various contents of MgO.

specimen	Pure hematite	+5 wt pct MgO	+10 wt pct MgO	+20 wt pct MgO
Ts/K	>1662	1521	1437	1373
Tm/K	>1673	1595	1550	1435

3.2. Effect of Sintering Time on the Change of Mineral Composition

The XRD patterns and optical microstructures of the samples with 20 wt pct MgO sintered at 1473 K for different time are shown respectively in Figs. 5 and 6. The XRD patterns show that with the prolongation of sintering time from 10 min to 60 min, the intensity of magnesioferrite peaks increased and that of hematite peaks decreased until they disappeared completely when the sintering time reached 60 min, but the intensity of magnesium oxide peaks decreased with the prolongation of sintering time, which agrees with the optical microstructure results as shown in Fig. 6. Then continuing to prolong the sintering time, the peaks of magnesium oxide disappeared completely and only that of magnesioferrite existed as shown in the XRD patterns.

Fig. 5.

XRD patterns of the samples with 20 wt pct MgO sintered at 1473 K for different time.

Fig. 6.

Optical microstructures of the samples with 20 wt pct MgO sintered at 1473 K for different time. (a) 10 min, (b) 20 min, (c) 40 min, and (d) 60 min. M-Magnesioferrite or magnetite, O-Magnesium oxide, P-Pore.

3.3. Effect of Temperature on Change of Mineral Composition

The XRD patterns of the sample with 20 wt pct MgO which were sintered at different temperatures for 60 min are shown in Fig. 7. Figure 7 shows that only hematite and magnesium oxide peaks were observed in samples sintered at 873 K and 973 K. However, there were obvious magnesioferrite peaks in the sample sintered at 1073 K. It also shows that with increasing of temperature from 1073 K to 1473 K, the intensity of magnesioferrite peaks increased and that of hematite peaks decreased. However, the intensity of magnesioferrite peaks changed no longer and hematite peaks had not been observed when sintered at 1473 K and 1573 K. It is worth noting that the intensity of magnesium oxide peaks decreased with the increase of temperature but had not disappeared completely until sintering at 1573 K for 60 min, which agrees with the optical microstructures results as shown in Fig. 8. It indicates that magnesium ion had continued entering into the magnesioferrite after the hematite disappeared, whose reason will be explained later.

Fig. 7.

XRD patterns of the samples with 20 wt pct MgO sintered at different temperatures for 60 min.

Fig. 8.

Optical microstructures of the samples with 20 wt pct MgO sintered at different temperatures for 60 min. (a) 1373 K, (b) 1423 K, (c) 1473 K, and (d) 1573 K. M-Magnesioferrite or magnetite, O-Magnesium oxide, P-Pore.

3.4. Change of Ferrous Ion in Samples

To consider that the XRD patterns, OM and SEM could not distinguish the magnesium ferrite, magnetite and magniferous magnetite ([Mg_1-x,Fe_x]O·Fe₂O₃, 0<x<1) due to them with the same crystal structure, in which the magnesium ferrite is corresponding with x=0. Potassium dichromate titration (PDT) was used to quantify the content of ferrous ion in sintered samples, as shown in Figs. 9 and 10. It can be seen from Fig. 9 that the ferrous ion was not detected when the sintering temperature is lower than 1073 K. The content of ferrous ion increased with the temperature from 1073 K to 1473 K, and then decreased while continue increasing the temperature.

Fig. 9.

Change of ferrous content in the samples with 20 wt pct MgO sintered at different temperatures for 60 min.

Fig. 10.

Change of ferrous content in the samples with various contents of MgO sintered at 1473 K for different time.

It can be found from Fig. 10 that the ferrous ion was influenced greatly by the content of MgO. With the content of MgO increasing from 0 to 10 wt pct, the ferrous ion gradually increased until reached the maximum value, then decreased while continue increasing the content of MgO. It reveals that the formation of magnesioferrite increases with the increase of MgO in the initial stage of sintering, but with the prolongation of sintering time, x in the magnesioferrite is determined by the ratio of MgO to Fe₂O₃ in raw materials. It also can be found that the content of ferrous ion in the samples with 10 wt pct MgO is much higher than that in 20 wt pct MgO samples. It reveals that formation of the magnesium ferrite is contributed while the amount of MgO approaches to 20 wt pct or more. On the contrary, the magniferous magnetite is contributed when the molar amount of MgO is far less than Fe₂O₃ in Fe₂O₃-rich system.

The results of the sample with 20 wt pct MgO when prolonging sintering time at 1473 K in air as shown in Fig. 11. It shows that the content of ferrous ion in the sample with 20 wt pct MgO decreased with prolonging sintering time until sintered 2880 min (48 h) completely disappeared to formation of MgO·Fe₂O₃.

Fig. 11.

Content of ferrous ion in the samples with 20 wt pct MgO sintered at 1473 K for different time.

The changing trend of ferrous ion quantitative result by PDT is as shown in Fig. 12, very consistent with that from Wilshire et al. The content of ferrous ion in samples from 4.45, 6.62, 7.76 to 3.52 molar pct (corresponding 2.30, 3.64, 4.81 to 2.61 wt pct) increased firstly and then decreased with increasing of MgO contents from 14.3, 20.3, 30.8 to 50 molar pct (corresponding 4, 6, 10 to 20 wt pct), though the reactions at 1473 K for 60 min reached to the equilibrium no enough except that in the sample with 20 wt pct MgO for 2880 min in the MgO–Fe₂O₃–FeO system.²⁹⁾

Fig. 12.

Comparison between the results of this work and the results from J. C. Wilishee et al.

3.5. Change of Element Content in Minerals

SEM image and EDS patterns of the samples with 20 wt pct MgO sintered at different temperatures for 60 min are presented in Fig. 13, and the element content based on the EDS results are given in Table 2. The SEM image of the samples sintered at 1373 K and 1423 K show that the hematite, megnesioferrite and magnesium oxide were observed. However, only two phases as megnesioferrite and magnesium oxide were observed when sintered at 1473 K and 1573 K, which indicates that the hematite reacted completely with magnesium oxide to form the megnesioferrite when sintered at 1473 K and 1573 K, agreeing with the results of XRD and optical microstructures. The EDS result shows that the atomic ratio of Fe to Mg in megnesioferrite was very closed, increased and then decreased with increasing of the temperature from 1373 K to 1573 K. Furthermore, the value of x in magnesioferrite was calculated from 0.050 to 0.080 and then to 0.069 with increasing of the temperature. With increasing of temperature from 1373 K to 1573 K, the content of magnesioferrite increased from 72.71 pct to 97.92 pct, which can be calculated by according to the quantitative result in Fig. 9. The magnesium ferrite and magnetite could not be distinguished when they exist at the same time, but the value of x will decrease, it can be deduced that the magniferous magnetite reacts with MgO and O₂ to form MgO·Fe₂O₃, through replacement of Fe²⁺ by Mg²⁺ in the lattice of magniferous magnetite.

Fig. 13.

Changes of SEM images and EDS patterns of samples with 20 wt pct MgO sintered at different temperatures for 60 min. (a) 1373 K; (b) 1423 K; (c) 1473 K; (d) 1573 K.

Table 2. Element content in minerals of 20 wt pct MgO samples sintered at different temperatures for 60 min.

Sintering temperature, K	Position points	Element compositions, at pct			Mineral	Atom ratio of Fe to Mg	x calculated according to the atom ratio of Fe to Mg
		Fe	Mg	O
1373 K	A1	38.92	17.70	43.38	[Mg1-x,Fex]O·Fe2O3 (72.71 pct)*	2.16	0.050
	A2	38.72	17.97	43.31
	A3	38.89	17.87	43.24
	A4	38.98	18.28	42.74
	A5	38.52	18.14	43.34
	B	53.49	0.17	46.35	Fe2O3	–	–
	C	1.99	68.41	29.60	MgO	–	–
1423 K	A1	40.71	18.70	40.59	[Mg1-x,Fex]O·Fe2O3 (86.11 pct)	2.21	0.065
	A2	40.02	18.30	41.68
	A3	40.56	17.58	41.86
	A4	39.99	18.26	41.75
	A5	39.13	17.68	43.19
	B	54.09	0.38	45.53	Fe2O3	–	–
	C	6.39	60.83	32.78	MgO	–	–
1473 K	A1	41.08	18.25	40.67	[Mg1-x,Fex]O·Fe2O3 (91.86 pct)	2.26	0.080
	A2	40.79	18.59	40.62
	A3	41.80	18.03	40.17
	A4	42.50	18.60	38.90
	A5	41.60	18.44	39.96
	C	5.18	60.21	34.61	MgO	–	–
1573 K	A1	42.11	18.79	39.10	[Mg1-x,Fex]O·Fe2O3 (97.92 pct)	2.22	0.069
	A2	41.07	18.64	40.29
	A3	40.93	18.41	40.66
	A4	41.72	18.79	39.49
	A5	41.38	18.67	39.95
	C	8.49	54.83	36.68	MgO	–	–

*The  amount of magnesioferrite ([Mg_1-x,Fe_x]O·Fe₂O₃) was calculated by the quantitative analysis for ferrous ion.

4. Discussions

The reaction between Fe₂O₃ and MgO can form the magnesium ferrite in air at lower temperature in the heating-up process as calculated thermodynamically by Eq. (2),³⁰⁾ which is an exothermic process. Simultaneously, the exothermic process may be also the reaction of magnetite, MgO and O₂ at a low temperature to form the magnesium ferrite as calculated by Eq. (3).³⁰⁾   

$$\begin{array}{l}{\text{Fe}}_2{\text{O}}_3+\text{MgO}=\text{MgO}\cdot {\text{Fe}}_2{\text{O}}_3\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\mathrm{\Delta }{\text{G}}^{\varnothing }=-43\mathrm{\hspace{0.17em} }865+36.62T,\mathrm{\hspace{0.17em} }\text{J}\end{array}$$(2)

  

$$\begin{array}{l}4{\text{Fe}}_3{\text{O}}_4+6\text{MgO}+{\text{O}}_2(\text{g})=6\text{MgO}\cdot {\text{Fe}}_2{\text{O}}_3\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\mathrm{\Delta }{\text{G}}^{\varnothing }=-974\mathrm{\hspace{0.17em} }444+427.33T,\mathrm{\hspace{0.17em} }\text{J}\end{array}$$(3)

As shown in Fig. 3, it was found that the DSC curve began into exothermic when the temperature is higher than 1170 K, it could be caused by the formation of MgO·Fe₂O₃ by according to the chemical equation of Eq. (2). The results of XRD and ferrous ion content as shown in Figs. 7 and 9 confirmed that MgO·Fe₂O₃ was formed at 1073 K due to that the XRD peak of the magnesioferrite had appeared while the ferrous ion content equal to zero.

The forming temperature (FT) of MgO·Fe₂O₃ obtained from DSC result is higher than that from XRD result due to the sample needs physically endothermic in heating-up process but the MgO·Fe₂O₃ formation is exothermic reaction. Meanwhile, a sharp change of the mass appeared at 1482 K from the TG curve of the sample with 20 wt pct MgO (the endothermic process was transformed into exothermic when the temperature was higher than 1570 K from the DSC curve) as shown in Fig. 3. It is possible that the magniferous magnetite reacts with MgO and O₂ to form MgO·Fe₂O₃ according to Eq. (3), though the magnesium ion entered into the magnetite to form the magniferous magnetite as a solid solution, increasing the difficulty of the reaction.

The MgO·Fe₂O₃ lattice is an inverse spinel crystal structure,^31,32) in which the ferric ion exists simultaneous in the tetrahedral and octahedral interstices of oxygen ions, it results in the diffusion rate of the ferric ion more than the magnesium ion in MgO·Fe₂O₃ lattice^16,33,34) though that is different with result of Zubets et al. based on (Mg, Fe)O formation in MgO side.¹⁷⁾ Therefore the final product of reactions between MgO and Fe₂O₃ are magnesioferrite and magnesium oxide or magnesioferrite when the MgO content in samples wt pct is less than or equal to 20 wt pct.

It is interesting to note that the mass loss of sample with 20 wt pct MgO terminated at about 1482 K, and thereafter the mass regained during the temperature increased to 1673 K, as shown in Fig. 3. Meanwhile, the ferrous ion increased with the temperature from 1073 K to 1473 K and then decreased as shown in Fig. 9. However, the results of XRD patterns (as seen Fig. 7) show that only the magnesioferrite and the magnesium oxide peaks were observed when sintering at 1473 K and 1573 K. Therefore, the increase of sample mass can be considered that magnesioferrite was oxidized at the temperature. The reason for the oxidation reaction may be that the magnesium ion continues entering into the magnesioferrrite, to carry out the replacement of ferrous ion by magnesium ion, and simultaneous transformation of ferrous ion to ferric ion, finally forming the magnesium ferrite as an equilibrium phase because the magnesium ion, the ferrous ion and the ferric ion that can exist simultaneously in the octahedral interstice of [Mg_1-x,Fe_x]O·Fe₂O₃ lattice on the crystal structure.

The formation mechanism of [Mg_1-x,Fe_x]O·Fe₂O₃ in solid-state reactions between MgO and Fe₂O₃ in the Fe₂O₃-rich system can be suggested as shown in Fig. 14.

Fig. 14.

Schematic diagram for formation mechanism of [Mg_1-x,Fe_x]O·Fe₂O₃ by MgO–Fe₂O₃ solid-state reactions in the Fe₂O₃-rich system.

(1) Solid-state reactions between MgO and Fe₂O₃ are carried out in heating-up process, forming MgO·Fe₂O₃ first at interface between MgO and Fe₂O₃ at a low temperature about 1073 K.

(2) According to the Wagner mechanism,³⁵⁾ then dissociations of 3MgO and Fe₂O₃ had appeared respectively at interfaces between MgO·Fe₂O₃ and MgO, Fe₂O₃ respectively to form 3Mg²⁺ and 3O²⁻, and 2Fe³⁺ and 3O²⁻, simultaneous they have the counter diffusion of 3Mg²⁺ and 2Fe³⁺ ions across MgO·Fe₂O₃ layer. Because the ferric ion diffuse more rapidly than that of the magnesium ion in MgO·Fe₂O₃ layer, the partial magnesium ion is replaced by the ferric ion, simultaneous the electronic compensation are carried out, forming the ferrous ion and the vacancies, i.e. [Mg_1-x,Fe_x]O·Fe₂O₃ solid solution would be formed near the side of Fe₂O₃; on the other hand, the ferric ion diffuses at the interface between the MgO·Fe₂O₃ and MgO, and reacts with MgO to form the MgO·Fe₂O₃.

(3) With the occurrence of the reaction, the [Mg_1-x,Fe_x]O·Fe₂O₃ layer has become thick, the interface between the MgO·Fe₂O₃ and MgO moves towards the side of MgO until MgO layer consumed completely.

(4) The ferric ion continues diffusing into the [Mg_1-x,Fe_x]O·Fe₂O₃ solid solution, which results in increase of x value through replacement of the magnesium ion by the ferric ion until the MgO·Fe₂O₃ layer disappeared and the [Mg_1-x,Fe_x]O·Fe₂O₃ reaches an equilibrium state thermodynamically.

According to the above mentioned results, it can be understood that MgO is added into the raw materials of iron ore sintering, which is beneficial to form the ferrous ion more in iron ore sinter by formation of MgO·Fe₂O₃ to drop the forming temperature, i.e. [Mg_1-x,Fe_x]O·Fe₂O₃ is formed easily, which will improve the RDI of iron ore sinter. However, the diffusion of magnesium ion is a rate-controlling factor on formation of [Mg_1-x,Fe_x]O·Fe₂O₃. The temperature difference from FT is a motive potential to form the magnesioferrite in heating-up process, which indicates that after the MgO added into the Fe₂O₃ the magnesioferrite stability is more than that of magnetite at a low temperature, it will be to restrain formation of the secondary hematite in cooling process.

5. Conclusions

In the present work, TG-DSC, XRD, OM, SEM-EDS and PDT were used to follow the changes of mass, energy, minerals, and microstructure in samples in heating-up and thermostatic processes for investigating the formation mechanism of [Mg_1-x,Fe_x]O·Fe₂O₃ in solid-state reactions between MgO and Fe₂O₃. The results are summarized as follows:

(1) The Fe₂O₃ reacts with the MgO to form MgO·Fe₂O₃ at about 1073 K in air. Subsequently, magniferous magnetite appears at 1173 K, resulting that the thermal decomposition of Fe₂O₃ occur at lower temperature than that of Fe₂O₃ itself.

(2) MgO·Fe₂O₃ is formed easily while the amount of MgO approaches to 20 wt pct. Simultaneous, the formation also contributes to form magniferous magnetite when the amount of MgO is far less than Fe₂O₃.

(3) Because magnesium ion and iron ion exist simultaneous in the octahedral interstice of MgO·Fe₂O₃ lattice as reverse spinel crystal structure, it is essential to form the [Mg_1-x,Fe_x]O·Fe₂O₃. The amount of magnesioferrite formed in sintering process is mainly related to the amount and diffusion rate of MgO in raw materials.

Acknowledgments

The authors would like to thank the National Natural Science Foundation of China (grants nos. U1460201 and No. 51374017) for financially supporting this work.

References

• 1)   U. S.  Yadav,  B. D.  Pandey,  B. K.  Das and  D. N.  Jena: Ironmaking Steelmaking, 29 (2002), 91.
• 2)   F. M.  Shen,  X.  Jiang,  G. S.  Wu,  G.  Wei,  X. G.  Li and  Y. S.  Shen: ISIJ Int., 46 (2006), 65.
• 3)   T.  Inazumi,  K.  Shinada and  M.  Kawabe: Tetsu-to-Hagané, 68 (1982), 2207.
• 4)   N.  Sakamoto,  H.  Fukuyo,  Y.  Iwata and  T.  Miyashita: Tetsu-to-Hagané, 70 (1984), 512.
• 5)   I.  Shigaki,  M.  Sawada,  M.  Maekawa and  K.  Narita: Trans. Iron Steel Inst. Jpn., 22 (1982), 838.
• 6)   L. H.  Hsieh and  J. A.  Whiteman: ISIJ Int., 33 (1993), 462.
• 7)   L.  Lu,  R. J.  Holmes and  J. R.  Manuel: ISIJ Int., 47 (2007), 349.
• 8)   G.-H.  Li,  Z.-K.  Tang,  Y.-B.  Zhang,  Z.-X.  Cui and  T.  Jiang: Ironmaking Steelmaking, 37 (2010), 393.
• 9)   T.  Umadevi,  A.  Brahmacharyulu,  R.  Sah,  P. C.  Mahapatra and  M.  Prabhu: Ironmaking Steelmaking, 41 (2014), 270.
• 10)   L.  Yang,  C. E.  Loo and  J.  Ostwaid: Trans. Inst. Min. Metall. Sect. C, 99 (1990), C80.
• 11)   S. C.  Panigrahy,  P.  Verstraeten and  J.  Dilewijns: Metall. Mater. Trans. B, 15B (1984), 23.
• 12)   H. S.  Roberts and  H. E.  Merwin: Am. J. Sci. (5th Series), 21 (1931), 145.
• 13)   R. C.  Richards and  J.  White: Trans. J. Br. Ceram. Soc., 53 (1954), 423.
• 14)   D.  Woodhouse and  J.  White: Trans. J. Br. Ceram. Soc., 54 (1955), 333.
• 15)   L. C. F.  Blackman: J. Am. Ceram. Soc., 42 (1958), 143.
• 16)   R. E.  Carter: J. Am. Ceram. Soc., 44 (1960), 116.
• 17)   A. V.  Zubets: Inorg. Mater., 38 (2002), 718.
• 18)   J. G.  Paik,  M. J.  Lee and  S. H.  Hyun: Thermochim. Acta., 425 (2005), 131.
• 19)   J. J.  Friel and  E. S.  Erickson, Jr.: Metall. Mater. Trans. B, 11B (1980), 233.
• 20)   S. C.  Panigrahy,  P.  Verstraeten and  J.  Dilewijns: Ironmaking Steelmaking, 11 (1984), 17.
• 21)   S. C.  Panigrahy,  B. C.  Jena and  M.  Rigaud: Metall. Mater. Trans. B, 21B (1990), 463.
• 22)   U. S.  Yadav,  B. D.  Pandey,  B. K.  Das and  D. N.  Jana: Ironmaking Steelmaking, 29 (2002), 91.
• 23)   S.  Dwarapudi,  T. K.  Ghosh,  V.  Tathavadkar and  M. B.  Denys: Int. J. Miner. Process., 112–113 (2012), 55.
• 24)   S.  Dwarapudi,  T. K.  Ghosh,  A.  Shankar,  V.  Tathavadkar,  D.  Bhattacharjee and  R.  Venugopal: Int. J. Miner. Process., 96 (2010), 45.
• 25)   S.  Dwarapudi,  T. K.  Ghosh,  A.  Shankar,  V.  Tathavadkar,  D.  Bhattacharjee and  R.  Venugopal: Int. J. Miner. Process., 99 (2011), 43.
• 26)   S.  Dwarapudi,  P. K.  Banerjee,  P.  Chaudhary,  S.  Sinha,  U.  Chakraborty,  C.  Sekhar,  T.  Venugopalan and  R.  Venugopal: Int. J. Miner. Process., 126 (2014), 76.
• 27)   D.-Q.  Zhu,  T.-J.  Chun,  J.  Pan and  J.-L.  Zhang: Int. J. Miner. Process., 125 (2013), 51.
• 28)   C. E.  Loo,  L. C.  Mackey and  B. M.  England: Trans. Inst. Min. Metall. Sect. C, 105 (1996), C175.
• 29)   J. C.  Wilishee and  J.  White: Trans. J. Br. Ceram. Soc., 66 (1967), 541.
• 30)   I.  Barin and  O.  Knacke: Themochemical Properties of Inorganic Substances, Springer, Berlin, (1973), Supplement, 1997.
• 31)   H.  Knoch and  H.  Dannheim: Phys. Status Solidi A, 37 (1976), K135.
• 32)   J. P.  Singh,  S. O.  Won,  W. C.  Lim,  I. J.  Lee and  K. H.  Chae: J. Mol. Struct., 1108 (2016), 444.
• 33)   V. P.  Pashchenko,  G. A.  Potapov and  O. G.  Litvinova: Sov. Powder Metall. Met. Ceram., 7 (1976), 560.
• 34)   D. L.  Fresh: Proc. Inst. Radio Eng., 44 (1956), 1303.
• 35)   J. S.  Armijo: Oxid. Met., 11 (1969), 171.