Effect of Al₂O₃ Content and Particle Size of Iron Ore on the Assimilation Characteristics of Sintered Ore

2.1 Preparation of raw materials

Generally, the design of the blending process for sintered ore is based on the representative chemical composition. Table 1 shows the results of analyzing the chemical composition and composition variation according to the particle size of the Pisolite series iron ore of Australian company B, which was used in this study. Even within the same iron ore brand, the contents of the chemical components differed depending on the particle size. In particular, differences in chemical composition were observed in iron ores smaller than 1 mm. Additionally, the Al₂O₃ content increased relatively with decreasing particle size. It is believed that the alumina content is higher on the smaller grain sizes due to a higher clay content.

Table 1.

Chemical composition and composition variation according to the particle size

In this study, iron ore with particle sizes of less than 1 mm was used as the attached particles in the process of granulating fine iron ore and additive materials in the sintering process [11,18-19]. To evaluate the assimilation properties of the iron ore, the calcium ferrite that was reacted with the iron ore contained 26 wt% CaO and 74 wt% Fe₂O₃ [15,17]. The calcium ferrite was prepared using reagent grade Fe₂O₃ (Sigma–Aldrich, ≥99.9%) and CaCO₃ (Thermo Fisher, 99.5%). CaCO₃ powder was calcined at 1173 K for 8 h to obtain CaO, after which the CaO was mixed with Fe₂O₃ to prepare the calcium ferrite. The fully mixed calcium ferrite powder was homogenized by pre-melting in a muffle furnace at 1773 K for 12 h in an Ar (99.99 vol%, 5 slm) atmosphere. The treatment time of 12 h was considered sufficient for homogenization because previous works have shown that that the molten calcium ferrite attains an equilibrium state when the holding time is more than 2 h [17,20-21].

2.2 Reaction experiment for assimilation between iron ore and calcium ferrite

Fig 2 shows a schematic of the experimental procedure. Ore fines were pressed into cylindrical tablets with diameters of 10 mm and heights of 5 mm. To understand the assimilation between the iron ore and calcium ferrite during sintering, a pre-melted CaO and Fe₂O₃ composite to produce calcium ferrite (CaFe₂O4) was also shaped into a cylindrical tablet with a length of 8 mm and height of 3 mm and placed on top of the ore tablet. The iron ore reaction depth was determined by measuring the calcium ferrite/ore reaction zone 10 times to derive the mean value (Fig 2(a)). The heat treatment was carried out in an Ar atmosphere. The temperature was increased at a rate of 10 °C/min and the temperature was maintained at 723 and 1553 K for 30 min, as shown in Fig 2(b). After the heat treatment, the samples were allowed to cool naturally and polished for microstructural observation of the bonding zones between the iron ore and minor elements. Microstructure and composition were analyzed using optical microscopy, field emission scanning electron microscopy, and field electron probe microanalysis (FE-EPMA, JXA-iHP200F, JEOL). The thermodynamic calculation program FactSage 8.3 was used to calculate the temperature of the emerging liquid melt phase resulting from the reaction between the iron ore and additive materials.

Fig. 2.

Experimental schematics and conditions. (a) A couple of tablets simulating and measuring positions of melt penetration depth and (b) experimental temperature and atmosphere

3.1 Assimilation characteristics according to the Al₂O₃ content of iron ore

Fig 3 shows the cross-sectional microstructure after the reaction with calcium ferrite, for each iron ore particle size. The assimilation characteristics of the iron ore and calcium ferrite could be distinguished by the difference in image contrast under an optical microscope. To quantify and compare the assimilation characteristics in detail, the previously developed Iron ore Reaction Index (IRI; Fig 4) was used to measure the assimilation characteristics [17]. The assimilation area of the calcium ferrite was measured at the interface reference, indicated by a red straight line along the cross-sectional microstructure. The IRI indicates the tendency of the initial melt and iron ore to react with each other and provides an index of the melt penetration depth. The IRI was calculated using equation (1):

Fig. 3.

Macrostructure of the vertical cross section according to particle size after the assimilation characteristic tests: (a) 1–0.5 mm, (b) 0.5–0.25 mm, (c) 0.25–0.15 mm, and (d) <0.15 mm

Fig. 4.

IRI according to iron ore size

The value of this index is 5 when the iron ore and the initial melt are fully penetrated and reacted, and 0 when the iron ore and the initial melt do not penetrate and react. If at least one point is fully penetrated, the IRI is considered to be 5. In this study, the reaction area varied depending on the sample size and reaction time. Thus, the IRI was used for accurate comparison with previous results [18,22-23].

When the iron ore particle size was 1–0.5 mm, the IRI was 2.23, indicating that ~45% of the iron ore tablets reacted with the calcium ferrite. As the particle size of the iron ore decreased, the IRI decreased, and at a particle size of 0.15 mm or less, the value was 0.46, indicating little assimilation with calcium ferrite. This generally runs counter to the theory that the smaller the particle size, the greater the reactivity due to the wider reaction surface area. To interpret this phenomenon, the correlation between the chemical components and IRI was investigated.

Fig 5 shows the correlation between the Al₂O₃ content in the iron ore and the IRI. It is evident that the IRI decreases with increasing Al₂O₃ content in iron ore because of the increase in viscosity as well as the decreased ore particle size, and this trend is the same as that observed in previous studies [11,17-19].

Fig. 5.

Relationship between the IRI and Al₂O₃ content in the iron ore

This phenomenon can be explained according to the change in the melt formation behavior as a function of Al₂O₃ in the melt. During melt formation, the Al₂O₃ acts as a melt inhibitor and decreases the melt surface tension. Fig 6 shows IRI the phase diagram of the Fe₂O₃–CaO–Al₂O₃ ternary system derived using the FactSage thermodynamics program; the liquidus line can be confirmed from the phase diagram. In the Fe₂O₃–CaO–Al₂O₃ ternary system, the temperature at which the liquid phase appears changes depending on the composition, and the temperature at which the liquid phase appears increases as the Al₂O₃ content increases. As a result, the amount of liquid phase at 1553 K decreases, and the amount of the solid phase increases depending on the Al₂O₃ content (Fig 7).

Fig. 6.

Ternary phase diagram of the Fe₂O₃-CaO-Al₂O₃ system illustrating liquidus lines

Fig. 7.

Variation in solid and liquid phase ratios as a function of the Al₂O₃ content

Therefore, when the iron ore is a fine particle, the observed decrease in reactivity with additives can be attributed to a higher liquid phase appearance temperature, which is caused by increased Al₂O₃ content, coupled with a reduced bonding area stemming from a smaller amount of liquid phase. Therefore, it can be inferred that the assimilation characteristics of iron ore and calcium ferrite are more dependent on the Al₂O₃ content within the iron ore rather than on the enlarged reaction surface area resulting from reduced particle size.

3.2 Effect of particle size and Al₂O₃ content of iron ore on assimilation characteristics

To investigate the effects of particle size and Al₂O₃ content on the assimilation properties of the iron ore, the chemical composition of the iron ore was fixed by blending iron ores with different particle sizes. The designed blends are presented in Table 2, and the chemical compositions of the blends are presented in Table 3. To minimize the influence of the chemical components, the design goal was to achieve a uniform chemical composition across the samples, with an increased ratio of iron ore of 0.15 mm or smaller, which had the highest Al₂O₃ content.

Table 2.

Size distribution of iron ore blending ratio

Table 3.

Chemical composition of iron ore blends

Fig 8 displays the cross-sectional microstructures for each blending case post-assimilation. Case 1 exhibited the most significant assimilation among the blends, with the degree of assimilation of iron ore and calcium ferrite being similar in Cases 2–5. Consequently, Case 1 exhibited the highest IRI at 2.82, while Cases 2–5 showed comparable levels of around 1. In Case 1’s blend, the Al₂O₃ content was 2.33, higher than the IRI in the 1-0.5 mm particle size sample, with 2.11 wt. % Al₂O₃, as mentioned in Section 3.1. However, in Cases 2–5, despite an increase in the ratio of iron ore with high Al₂O₃ content (0.15 mm or smaller) and low Al₂O₃ content (1 to 0.5 mm), the IRI remained low (roughly 1) due to the increased ratio of iron ore with high Al₂O₃ content. This confirms the effect of blended iron ore particle size on assimilation and provides a guide to the use range of -0.15 mm particle size, even with similar Al₂O₃ content. As more care is taken to match the average chemical composition of the iron ore, a situation may arise in which the content of fines increases. Therefore, this research result means that even if the average particle size and the chemical composition of the iron ore satisfy the specifications, the Al₂O₃ content range of -0.15mm should be considered.

Fig. 8.

Macrostructure of the vertical cross section post-assimilation tests for various blending cases: (a) Case 1, (b) Case 2, (c) Case 3, (d) Case 4, (e) Case 5

Fig. 9.

IRI according to blending cases

Fig 10 presents the SEM images of Cases 1 and 5, and the EPMA mapping results for the cross-sectional microstructures at the top, middle, and bottom of the reaction zone. As shown in Fig 10(a), a broad reaction area between the calcium ferrite and iron ore was observed in Case 1, indicating complete assimilation of calcium ferrite with iron ore. EPMA mapping revealed an even distribution of the Al component across the reaction area, which directly affects the assimilation of iron ore. In contrast, Al was concentrated in the top layer of the sample. This suggests that assimilation occurs when the flow of the calcium ferrite liquid phase and the movement of ore particles happen concurrently, promoting assimilation and leading to a smoothly formed melt with an expanded melting area. Conversely, Fig 10(b) shows there was limited assimilation in Case 5, where the Al component is present only in the iron ore portion (Fig 11(b)-(c)) and in some assimilated areas (Fig 10(b)-(b)). In the calcium ferrite area, only Fe and Ca components were observed, indicating that assimilation with the iron ore was not achieved in this case.

Fig. 10.

EPMA analysis of samples post-reaction for different blending cases: (a) Case 1, (b) Case 2

Fig. 11.

Schematic illustrating the assimilation mechanism across different particle sizes: (a) Large assimilation and (b) small assimilation

A schematic of the reaction is shown in Fig 11 to explain this phenomenon. As the temperature increases, iron ore with a low aluminum content of 0.25 to 0.15 mm is preferentially melted to produce a liquid phase. This process is facilitated by the latent heat produced as the melt volume grows. Notably, it is hypothesized that iron ore particles smaller than 0.15 mm, which possess a relatively high Al₂O₃ content, will melt more readily due to this latent heat (Fig 11(a)). Conversely, when the proportion of iron ore with a lower Al₂O₃ content (particle sizes between 0.25 to 0.15 mm) is reduced in favor of iron ore with particle sizes less than 0.15 mm, the initial melting temperature tends to be higher. This observation holds even when the proportion of iron ore with the largest particle size (1 to 0.5 mm) and the lowest Al₂O₃ content among the mix increases. In such cases, the reaction tends to be slower due to the smaller reaction surface area, attributed to the larger particle size (Fig 11(b)). Consequently, differences in the assimilation characteristics of calcium ferrite and iron ore can arise for the same reaction time.

In conclusion, the findings of this study underscore the significant impact of chemical composition on the assimilation process. However, the influence of particle size distribution is also non-negligible. Ultimately, it becomes evident that both the composition and particle size distribution must be carefully considered when determining the blending ratio of iron ore. This is crucial to ensure the strength of the sintered ore.Top of Form