Chemical, optical and mineralogical analyzes of the sinter samples were carried out to determine the sinter structures in the sinter samples. Table 4 shows the results of the sinter laboratory studies. The sintered compound's construction varies depending on the particle size, chemical composition of iron ore, fluxes, and the extent of the reactions (Goldring et al. 1989; Lu and Ishiyama 2015). Sinter production consists of many phases and especially SFCA and SFCA-I compositions. These different phases occur depending on the operational and processual conditions of the sinter. The mineralogical formation has a significant effect on softening and melting temperature (Pal et al. 1998).
Effects of pellet dust on sinter phase composition.
The mineralogical structure of the sinter was investigated using the dust diffraction technique (rietveld method). Quantification of calcium and aluminum silicoferrites (SFCA) and calcium silicates (larnite) determines the strength of sinter materials and the return rate of fine dusts. Table 5 shows the mineralogical compounds and results of several produced sinter samples.
Table 5
Quantitative XRD analysis of sinter productions (%).
Phases | Sin-1 | Sin-2 | Sin-3 | Sin-4 | Sin-5 |
Larnite | 7.65 | 7.27 | 7.42 | 6.70 | 6.83 |
Hematite | 25.43 | 28.63 | 24.04 | 23.08 | 28.52 |
Magnetite | 27.89 | 26.60 | 27.54 | 32.67 | 27.22 |
SFCA-I | 16.20 | 15.39 | 16.01 | 14.36 | 10.30 |
SFCA | 22.84 | 21.17 | 24.99 | 22.17 | 27.13 |
Total SFCA | 39.04 | 36.56 | 41.00 | 36.53 | 37.43 |
SFCA-I/SFCA | 0.71 | 0.73 | 0.64 | 0.65 | 0.38 |
Wuestite | 0.00 | 0.96 | 0.00 | 1.02 | 0.00 |
When iron ore (A) was replaced with iron ore (A) pellet dust, the columnar SFCA phase in the sinter somewhat improved and reached a maximum value of 27.13% (Sin-5) (Fig. 5). The pellet dust in the sinter blend creates regions of high density, resulting in higher assimilation ability (Honeyands et al. 2019). The sintering mixture containing 13.53% pellet dust fraction (Sin-3) probably reached the maximum assimilation reaction rate and produced higher SFCA and SFCA-I values. When the particle size of the pellet dust is over 13.53%, the reduction of the reaction area results in poor assimilation reactions and thus poor values of SFCA-I.
The sinter pot test revealed the SFCA-I/SFCA correlation comparatively. Correlation is crucial for sintering carried out with fine iron ore. The SFCA-I/SFCA value ratio decreased from 0.71 (Sin-1) to 0.38 (Sin-5), depending on the iron ore (A) pellet dust fraction in the sinter mixture (Fig. 6). It can be said that SFCA-I compulsion shapes with similitude of pellet grains. Hida et al. (1983) declared that, meanwhile another models of SFCA are consisted of the fusing stage.
As seen in Fig. 7, while the hematite phase in the sinter formation increases, the SFCA + SFCA-I phases decrease. There is a complete contrast between the hematite and the other phases in the sinter matrix.
Replacing fine iron ore with pellet dust particles differed the sintering rate and SFCA-I. The sinter blend with 0% fine iron ore fraction (Sin-1) had the highest sintering rate and SFCA-I (25.25 mm/min, 16.20%, respectively), while the sinter mix with 21.86% fine iron ore fraction (Sin-5) has the lowest sintering rate and SFCA-I (21.20 mm/min, 10.30% respectively). However, the SFCA value increased from 22.17–27.13% due to the rising fine iron ore in the sinter mixture (Fig. 8). This situation suggests that the replacement of coarse ore with fine iron ore reduces airflow rate and slows the flame front across the sinter bed (FFS: flame front speed). With the reduction of the average particle size of the iron ore, the airflow velocity in the sintered body decreased. As a result, it was observed that the sintering efficiency and sintering rate decreased. The higher the coarse ore content in the sinter bed, the greater the sintering rate.
Figure 9 shows the variation in sintering time depending on the amount of SCA and SFCA-I. Replacing fine iron ore with pellet dust particles increased sintering time and SFCA. Sinter mix with 0% fine iron ore fraction (Sin-1) has the lowest sintering time and SFCA (19.80 min, 22.84%, respectively), while 21.86% fine fraction mix (Sin-5) has the highest sintering time and SFCA (25.90 min, 27.03%, respectively). However, the SFCA-I value decreased from 15.36–10.30% due to the increase in fine iron ore in the sinter mix (Fig. 9). This situation shows that the replacement of coarse ore with fine iron ore increases the sintering time and quantity of SFCA in the sinter.
Optical microscopy analysis of the sinter formation
The sintering process contains a lot of chemical reactions during sinter production. These reactions form several compounds in the sintered material (Dawson 1993). The temperature distribution varies in the sintered material, and some cracks occur around the magnetite compound. These cracks and pores affect the physical properties of the sintered product (Webster et al. 2013; Hida et al. 1983; Dawson et al. 1984; Honeyands et al. 2017). Different compounds occur in different amounts depending on the basicity of the mixing material and the cooling rate. Through the distinct reflections obtained from sinter phases, microscopic analysis revealed that the sinter matrix usually consisted of secondary magnetite and hematite grains (settled from primary sinter melt), flux grains, and complex calcium ferrites known as Silicoferrite Calcium and Aluminum (SFCA), glass and silicate (e.g., larnite, Ca2SiO4). Figure 10 shows optical micrographs of the major sinter phases. In the microscope research on sinter material, hematite, magnetite, and SFCA phases are seen among other sinter phases with different reflections. A glowing compound may be hematite and found right next to the gray SFCA phase. These phases can be in heterogeneous textures within the industrial sinter material.
The replacement of fine iron ore (A) fractions with iron ore (A) pellet dust resulted in the sinter formation with more magnetite, hematite, SFCA, and less larnite and SFCA-I. Pellet dust fractions are more reactive than the iron ore particle and provide more softening in solid grains.
Figure 10(a) shows the sintered microstructure containing 0% pellet dust (Sin-1). It has been observed that sinter reactions occur in the SFCA bond phase formation. The sin-1 sample comprises fine, acicular tissue of SFCA and hematite. Here, fine magnetite crystal grains and several interlaced hematite growths are seen. The Sin-2 sample, a sinter blend containing 9.11% iron ore (A) pellet dust, could be argued to start the recovery with hematite, magnetite, SFCA phases, and SFCA-I appearing in the sintered sample. The sin-3 sintered sample containing 13.53% pellet dust fraction produced the best phase result in hematite, magnetite, and especially SFCA and SFCA-I (24.99% and 16.01%, respectively).
As shown in Fig. 10, when the pellet dust ratio rose above 17.88% in Sin-4 and Sin-5 samples, the hematite, magnetite, and SFCA phase increased in the sinter, but SFCA-I decreased (Figs. 10d-10e). This situation may be due to increased cracking and stresses in contact with pellet dust particles during the compaction and reaction of the sinter mixture. Figures 10a, 10b, and 10e (Sin-1, Sin-2, and Sin-5, respectively) show optical micrographs of SFCA and multiporous SFCA-I phases.
It can be observed more hematite, magnetite, and different SFCA types. Sinter samples have no clear differentiation between flat and blocky SFCA. Most of the bonding structure is considered blocky SFCA. Recent studies revealed that acicular SFCA could appear flat and be easily confused with blocky SFCA in these phases. Therefore, the correct classification of the different SFCA types requires utilizing other analytical methods (Tonžetić and Dippenaar 2011).
SEM study on the sinter formation
SEM analyzes were performed on O, Mg, Al, Si, K, Ca, Mn, and Fe elements to identify sinter patterns. Larnite-C2S, wuestite, hematite, magnetite, and SFCA phases were analyzed and identified through optical microscopy. Approximately 510 SEM-EDS analyzes were performed on equally shining parts of the five different sinter samples to identify and learn the chemical composition of the different phase structures. Results were collected according to elemental analysis and then compared with the literature to reveal the chemical composition of the phases (especially for the different types of SFCA phases).
After a detailed investigation of iron ore sintering, Pownceby et al. (2016) declared that general chemical analysis of sinters could not be performed by only characterizing microstructures, considering each phase in the sinter might have a significant variation in the chemical composition (Goldring 1989). Each phase may appear in different morphologies. Figures 11(c), 11(d), and 11(e) show typical micrographs of SFCA and SFCA-I. Columnar SFCA has less microporosity than acicular SFCA in these micrographs. Columnar SFCA appears to be more associated with other melt phases, such as hematite, magnetite, and glass (Figs. 11(a), 11(c), 11(e)). Acicular SFCA and columnar SFCA phases have formed at a lower temperature through a solid-state reaction between fine iron ore grains and pellet dust particles. SEM analysis has shown that columnar SFCA has the same chemical composition as acicular SFCA but different morphology.
Figure 11 shows the morphological structure of the different phases in the sintered material. Hematite, magnetite, and SFCA phases were recognized in the sinter. The segregation between several types of SFCA was not clear. This situation might have stemmed from very few chemical differences between the SFCA phases. The composition of the SFCA phases was similar to each of SFCA and SFCA-I. The morphology of the sintered material is mainly associated with the composition mode and is related to its specific chemical composition, heating, and cooling rate. Therefore, comparing the chemical compound with those in the literature can help to reveal various SFCA structures.
While SFCA develops slowly and persists longer at low temperatures, it develops and disappears rapidly at high temperatures (Lu 2015; Egundebi and Whiteman 1989; Bai et al. (2019); Turriff 2007). The authors explained that the length of time at sintering is a significant factor in the amount of SFCA produced. Most scientists suggest that slower sintering at lower temperatures will yield more SFCA than sintering at higher temperatures. EDS analyses from stoichiometry calculations of the analyzed SFCA phases did not allow for an obvious distinction between the two SFCA types based on M14O20 (SFCA) and M20O28 (SFCA-I) phases (Honeyands et al. 2017; Tonžetić et al. 2011).
Morphological features of SFCA
As shown in Fig. 12, SFCA can be divided into three types of morphological structures, depending on their character traits: blocky type (SFCA), acicular type (SFCA-I), and dendritic type (SFCA-II). Dendritic type SFCA was not in the area of interest of this study. The chemical composition and morphology of two types of SFCA were investigated. The blocky type SFCA was observed to have lower Fe2O3 content than the platy type but higher Ca and Si. There was no significant difference in the amount of Al2O3 between the two structures. SFCA of the columnar type had a lower MgO than the acicular type.
In conclusion, increasing the liquid phase content in the softening zone and developing a bonding structure chiefly composed of blocky and platy SFCA seems effective in increasing the strength.
As shown in Fig. 13, EDS analyses of SFCA structures were performed on the Al2O3-CaO-Fe2O3 triple-phase. This phase diagram was used to position hematite, magnetite, SFCA, and SFCA-I mineral compounds in the sintered material. The chemical compounds of the SFCA phases appeared to be close to SFCA and SFCA-I. Figure 13 shows SFCA, SFCA-I, and SFCA-II in a series of solid solutions of SFCA. The results of the EDS analysis on the SiO2-CaO-Fe2O3 and Si-Ca-Fe phase diagrams are shown in Figs. 12, 13, and 14. It was also observed that the replacement of iron ore (A) pellet dust with Iron Ore (A) gradually transferred most of the chemical compounds of SFCA toward the SFCA-I region (Figs. 13 and 14). This situation may confirm the results of XRD spotting that high pellet dust fractions in the sinter support SFCA formation rather than platy SFCA-I (Ji et al. 2019; Wang et al. 2019; Liao and Guo 2019; Zhang et al. 2019).
The shaded region in Fig. 14 shows that most SFCA compounds are intertwined in industrial sintering. Pownceby and Clout (2003) reported that SFCA and SFCA-I form in a range of solid solutions of SFCA (Kahlenberg et al. 2021; Zöll et al. 2018). Different relative ratios of SFCA models can coexist depending on the sinter parameters (e.g., oxygen potential) and physical properties (particle size, porosity, etc.), and the chemical composition of the raw materials.
In this study, approximately 510 SEM-EDS analyzes were performed on five sinter samples. All compositions of sinter samples were marked with dots representing microprobe data. Figure 15 shows the color distribution of Si/Ca/Fe elements, which are necessary for determining the SFCA and SFCA-I phase structures in the sintered microstructure. The key element map of SFCA showed the sintered particle color distribution as Si, Ca, and Fe. Figure 15 shows the relevant map of elements using the Thermo Scientific-FEI Apreo S equipped with an Ultra Dry EDS Detector and Quasor II EBSD. This image shows all the recognized phases in the sintered material containing the SFCA type.
Figure 16(a) shows microprobe map analyses displaying phase mineralogy and textures developed from sinter samples. Here, the first cluster (blue) represented a mixture of hematite and magnetite phases, the second cluster (red) the SFCA-I phase structure, and the third cluster (green) the SFCA phase formation. The region between (1) one number (blue) and (2) two number (red) might be a different phase structure in the sintered material (Fig. 16(a)). Magnetite and SFCA-I were close to each other in the (2) two number region because of their similar chemical composition and color texture. Therefore, it was difficult to distinguish them from one another by the microscope and the SEM-EDS method. The chemical morphologies of the SFCA models were characterized. The results showed that the SFCA structures were similar to chemical compounds for which they were identified in the literature. Figure 16(b) is the key element map of SFCA showing the distribution of sinter particle color of Si, Ca, and Fe. The phase patched map showed the distribution of all phases in the sintered material. SFCA was chiefly related to magnetite, while SFCA-I was mainly to hematite.
Chemical analysis of Ca-Si-Al-Fe elements in total sinter samples
SEM and EDS spectrum analyzes of sintered samples were performed. In order to determine the chemical composition of SFCA phase types, 602 SEM-EDS analyzes were performed on five sinter samples. Among the O, Mg, Al, Si, K, Ca, Mn, and Fe elements, Fe, Si, Ca, and Al were selected from Figs. 17 to 20 show the analysis results.
Figure 18 shows that the phases containing 0.5–5% Si could be SFCA and SFCA-I. The ones with less than 0.5% Si may be hematite, magnetite, etc., while the phases containing over 5% Si are possibly larnite phase formations. In this investigation, approximately 79% of the total SFCA phases in the sinter samples included Si in proportions ranging from 2.5–5%. Roughly 85% of the SFCA-I phases in the sinter contained Si ratios ranging from 0.5–2.5%.
Figure 19 shows that the phases with 6–12% Ca can be SFCA and SFCA-I. Phases containing less than 6% Ca may be hematite, magnetite, etc. Phases containing more than 12% Ca may be the larnite phase formation. In this study, almost 75% of the total SFCA phases in the sinter had Ca between 9–12%. Approximately 80% of the SFCA-I phases in the sinter samples possessed Ca in proportions ranging from 6–9%.
Figure 20 shows that the phases containing 0–1% Al will exhibit hematite and magnetite structure rather than SFCA and SFCA-I. Phases containing 1–3% Al will demonstrate an SFCA structure rather than SFCA-I. In this study, almost 66% of the total SFCA phases in the sinter contained Al in proportions ranging from 1.5–3%. Approximately 54% of the SFCA-I phases in the sinter had Al content ranging from 0–1.5%.
Particle size of phases in the sintered material
SFCA phase structures in sinter were investigated by optical microscope and SEM analysis. Elemental values of the phases were determined with the EDS analysis method. The grain size of the phases in the sinter was performed using the Image J program. Depending on the size of the crystals, SFCA (silico-ferrite of calcium and aluminum) phases can develop as dendritic SFCA, flat SFCA, and columnar SFCA. As a result of the tests performed in this study were determined that more than 32% of the total SFCA consisted of platy SFCA (SFCA-I). The relationship between fine iron ore particle size and the formation of sinter phases during the sinter tests is summarized in Fig. 21.
In Fig. 21, considering the grain size smaller than 26 microns, the phases should be SFCA and SFCA-I. Especially, grain sizes smaller than 16 microns may contain more SFCA-I phase. Grain sizes larger than 26 microns may have more hematite, magnetite, etc.