ISIJ International
Online ISSN : 1347-5460
Print ISSN : 0915-1559
ISSN-L : 0915-1559
Fundamentals of Silico-Ferrite of Calcium and Aluminium (SFCA) Iron Ore Sinter Bonding Phase Formation: Effects of Titanium on Crystallisation during Cooling
Nathan A. S. WebsterMark I. PowncebyRachel PattelJames R. ManuelJustin A. Kimpton
Author information

2019 Volume 59 Issue 6 Pages 1007-1010


The effects of Ti on the crystallisation during cooling of complex Ca-rich ferrite iron ore sinter bonding phases SFCA and Fe-rich SFCA was investigated using in situ synchrotron X-ray diffraction. Cooling of a high temperature (T = 1623 K) assemblage comprising magnetite and melt in synthetic iron ore sinter mixtures showed that increasing the Ti concentration to 1, 3 and 6 mass% TiO2 resulted in the formation of SFCA to be favoured over Fe-rich SFCA. Fe-rich SFCA was not observed to form at all during cooling in the 6 mass% TiO2 mixture. The absence of Fe-rich SFCA in the high-Ti experiment was rationalised on the basis that the SFCA structure likely accommodates more Ti4+ than the Fe-rich SFCA structure, thereby stabilising SFCA relative to Fe-rich SFCA. Observation of Fe-rich SFCA in sinter may be an indicator that the localised Ti concentration within a sinter blend is likely to be low.

1. Introduction

‘SFCA’ (Silico-Ferrite of Calcium and Aluminium) phases are key bonding materials within industrial iron ore sinter.1) Sinter is utilised extensively worldwide in the production of steel from iron ore and typically constitutes more than 60% of the ferrous burden in modern blast furnaces in Japan and most of the blast furnaces in Europe.2) Increased understanding of (i) the compositional and thermal stability of SFCA phases, (ii) their formation mechanisms, and (iii) the effect of different processing parameters on (i) and (ii) has the potential to improve the efficiency of the sintering process, by being better able to predict optimal sintering conditions required to produce high-quality product, based on the chemical composition and physical characteristics of a given iron ore sinter blend.

The compositional, structural and/or textural characteristics of the ‘SFCA’ phases described in the literature – SFCA, SFCA-I (triclinic crystal structure), SFCA-II (triclinic) and Fe-rich SFCA – have been summarised recently.3) Mumme and Gable have also recently reported the crystal structures of the monoclinic variants of SFCA-I and SFCA-II.4) In terms of composition, SFCA has general formula M14O20, where M = 60–76 mass% Fe2O3, 13–16 mass% CaO, 3–10 mass% SiO2, 4–10 mass% Al2O3 and 0.7–1.5 mass% MgO.5,6) Mumme et al.7) reported that an SFCA-I (M20O28) phase in industrial sinter contained 84 mass% Fe2O3, 13 mass% CaO, 1 mass% SiO2 and 2 mass% Al2O3, and also synthesised Si-free SFCA-I (triclinic) material which had the composition 83.2 mass% Fe2O3, 12.6 mass% CaO and 4.2 mass% Al2O3. The Fe-rich SFCA reported in the literature had average composition 84.2 mass% Fe2O3, 9.6 mass% CaO, 1.8 mass% SiO2 and 5.7 mass% Al2O3 (synthetic sample), and 84.8 mass% Fe2O3, 9.4 mass% CaO, 2.2 mass% SiO2, 2.8 mass% Al2O3 and 1.8 mass% MgO (pot-grate sinter sample).3) The crystal structure details of Fe-rich SFCA are yet to be established, however electron probe microanalysis-based compositional results indicated that, similar to SFCA-I, Fe-rich SFCA contains a higher proportion of Fe2+ in its structure than SFCA.

A number of recent investigations8,9,10,11,12,13,14,15,16) have utilised in situ X-ray diffraction (XRD) in order to establish the formation mechanisms of SFCA phases under simulated sintering conditions from synthetic starting sinter mixtures. Conditions varied between 298–1623 K, under oxygen partial pressures (pO2) in the range 0.21 to 1 × 10−4 atm, and at basicity (B, = CaO:SiO2, mass ratio) values spanning 2.5–5. In one of these investigations the effects of basicity and Mg concentration on the crystallisation of Fe-rich SFCA, and SFCA, during cooling was established.16) Decreasing basicity from B = 4.0 to 2.5 resulted in the formation of an Fe-rich SFCA phase being suppressed, with SFCA becoming the only Ca-rich ferrite phase to crystallise from the melt during cooling at B = 2.5. An opposing effect was observed with increasing Mg addition, with Mg stabilising Fe-rich SFCA relative to SFCA, in a similar manner to the presence of Mg favouring the formation of SFCA-I compared to SFCA during heating.17) This was considered to be due to i) the similar sizes of the Mg2+ and Fe2+ cations (0.86 and 0.92 Å (0.086 and 0.092 nm) for Mg2+ and Fe2+ in octahedral coordination, respectively18)), ii) the Fe-rich SFCA (and SFCA-I) containing a higher proportion of Fe2+ than SFCA, and iii) Mg2+ substituting for Fe2+ in the crystal structures. Given that industrial sinter blends are typically fluxed to basicity < 2.0, it was inferred by Webster et al. that any observation via electron microscopy of the Fe-rich SFCA phase in industrial or pilot scale pot-grate sinter may be an indicator of localised high basicity, and/or, localised high Mg concentration in a sinter blend.16)

Given the effect of Mg on the crystallisation of Fe-rich SFCA and SFCA during cooling, and its rationalisation based on structural similarities between SFCA-I and Fe-rich SFCA when compared to SFCA (i.e. Fe-rich SFCA and SFCA-I containing a higher proportion of Fe2+), Webster et al.16) also hypothesised that increasing Ti content would suppress Fe-rich SFCA formation relative to SFCA formation during cooling since the SFCA crystal structure can accommodate more Ti than the SFCA-I crystal structure.13) The aim of the current paper is to test this hypothesis. The introduction of a minor amount of Ti-bearing sinter into the blast furnace burden is a strategy used by steelmakers to extend blast furnace operating campaigns,19) and Bluescope’s Port Kembla sinter plant incorporates 2–3 mass% fine grained (< 150 μm) New Zealand ironsand as a component of its iron ore sinter blend.20) The effect of TiO2 addition on sinter strength and reduction degradation index has been the subject of a number of investigations,21,22,23) and the effect of TiO2 on equilibrium sinter phases has also been studied.24) The current paper is the first to investigate the effect of TiO2 on Fe-rich SFCA.

2. Experimental

2.1. In situ XRD Sample Preparation

Table 1 shows the composition of each mixture (in terms of weighed mass% of oxides). The base mixture, SM4/5, with B = 4.0 and containing 5 mass% Al2O3 has also been utilised in a number of our previous investigations.8,9,11,12,16) The SM4/5-1TiO2, 3TiO2 and 6TiO2 mixtures, with 1, 3 and 6 mass% addition of TiO2 to SM4/5, respectively, were designed to examine the effect of Ti concentration. The mixtures were prepared from fine grained synthetic hematite, Fe2O3 (Acros Organics, 99.999%); calcite, CaCO3 (Thermo Fisher, 99.95%); quartz, SiO2 (Sigma Aldrich, 99.995%); gibbsite, Al(OH)3 (Alcan OP25 Super White, 99.9%); and a 94 mass% rutile/6 mass% anatase (TiO2) powder (Aldrich, 99.99%) Weighed powders were micronised in ethanol for 4 min g−1, centrifuged and dried at 333 K overnight. They were then remixed by hand using a mortar and pestle to ensure homogeneity.

Table 1. Nominal compositions, in mass% oxides, of the mixtures used in the in situ S-XRD experiments.
Concentration (oxide mass%)

2.2. In situ XRD Experimentation and Data Analysis

In situ synchrotron XRD (S-XRD) experiments were performed on the powder diffraction beamline at the Australian Synchrotron.25) An Anton Paar HTK 2000 high-temperature chamber employing a Pt resistance strip heater was fitted to the beamline. XRD data were collected over the range 5° ≤ 2θ ≤ 85.5° continuously throughout heating, with individual datasets collected for 60 sec, using an X-ray wavelength of 1.1061 Å (0.11061 nm) calibrated using LaB6 (NIST 660b line position standard).

For each in situ experiment a heating rate of 50 K min−1 was used from 298–1623 K. The rate was then reduced to 2 K min−1 during cooling between 1623–1273 K, and then increased to 50 K min−1 from 1273 to 298 K. The temperature was measured by a Pt/PtRh10% thermocouple connected to the underside of the platinum strip directly underneath the sample. The time–temperature profile used for these experiments, especially during cooling, is significantly slower than those encountered in industrial sintering. It was chosen so as to monitor phase crystallisation with reasonable temperature resolution in order to establish trends with compositional variation. The temperatures of phase formation/transformation given throughout the manuscript are those at the start – which is when the temperatures were automatically recorded – of the relevant data set. An error of ± 10 K is assigned to these values consistent with our previous work.8,16)

The decomposition of precursor phases and the formation of new phases as the experiments progressed was visualised by stacking the datasets to produce plots of accumulated data with temperature plotted vs 2θ, viewed down the intensity axis. The Anton Paar chamber was fed by a continuous flow of a 0.5% O2 in N2 gas mixture (pO2 = 5 × 10−3 atm) throughout heating and cooling. Hsieh and Whiteman concluded that this pO2 maximised the formation of Ca-rich ferrites while still producing mineral assemblages similar to those found in industrial sinters.26)

3. Results and Discussion

Figure 1 shows a plot of accumulated in situ S-XRD data, viewed down the intensity axis and with temperature plotted vs 2θ, for the experiment performed for the SM4/5-1TiO2 mixture. During heating the sequence of phase decomposition/formation events was as follows: decomposition of Al(OH)3; decomposition of CaCO3; formation of lime (CaO); formation of dicalcium ferrite (C2F, i.e. 2CaO.Fe2O3); and formation of SFCA. At the peak temperature the phase assemblage was Fe3O4 in a FeO–Fe2O3–CaO–Al2O3–SiO2 melt. During cooling the Fe-rich SFCA and SFCA phases began to crystallise from the melt at 1495 K.

Fig. 1.

In situ S-XRD data collected for the SM4/5-1TiO2 sample, viewed down the intensity axis, during heating and cooling in the range 298–1623 K and at pO2 = 5 × 10−3 atm. C2F = 2CaO.Fe2O3.

Table 2 shows the temperatures of Fe-rich SFCA and SFCA formation during cooling for each mixture. In the Ti-free sample (i.e. SM4/5), Fe-rich SFCA was observed to crystallise before SFCA and this is consistent with previous work.8,16) Increasing Ti concentration up to 1 mass% TiO2 for the SM4/5-1TiO2 mixture i) lowered the crystallisation temperature of both Fe-rich SFCA and SFCA significantly, and ii) resulted in both Fe-rich SFCA and SFCA crystallising simultaneously (at 1495 K). Increasing Ti concentration further to 3 mass% TiO2 resulted in SFCA crystallising before Fe-rich SFCA, and only a minor amount of Fe-rich SFCA was observed to form. For the SM4/5-6TiO2 mixture the Fe-rich SFCA was not observed. Figure 2 clearly shows the suppression of Fe-rich SFCA formation with increasing Ti content. There did not appear to be a significant change in the crystallisation temperature of SFCA as Ti increased from 1 mass% to 6 mass% TiO2, with any variation considered to be within experimental error.

Table 2. Temperatures of the Fe-rich SFCA and SFCA crystallisation events during cooling.
Temperature (K)
SampleFe-rich SFCASFCA
Fig. 2.

In situ S-XRD data collected between 1623 and 1323 K during cooling at 2 K min−1 and over the range 19° < 2θ < 20° for the (a) SM4/5, (b) SM4/5-1TiO2, (c) SM4/5-3TiO2 and (d) SM4/5-6TiO2 mixtures. The colours in the plots are proportional to X-ray intensity, with red and yellow more intense than green, and green more intense than blue.

The results, therefore, support the hypothesis that increasing Ti content suppresses Fe-rich SFCA formation relative to SFCA formation during cooling. This is due, presumably, to the SFCA crystal structure being able to accommodate more Ti than the Fe-rich SFCA structure, however this would need to be confirmed in future work.

Finally, the results for SM4/5 determined here (cooling rate = 2 K min−1), when compared with those determined recently by Webster et al. (cooling rate = 5 K min−1),16) suggest an effect of cooling rate on the crystallisation temperatures of Fe-rich SFCA and SFCA. Crystallisation was observed at temperatures ~80 K higher at the slower cooling rate. Future work will investigate further the effect of cooling rate on the crystallisation of Fe-rich SFCA and SFCA.

4. Conclusion

The effects of Ti concentration on the crystallisation of Fe-rich SFCA and SFCA phases during cooling of iron ore sinter mixtures heated and cooled in the range 298–1623 K and at pO2 = 5 ×10−3 atm have been investigated using in situ S-XRD. Increasing Ti concentration up to 6 mass% TiO2 resulted in the formation of Fe-rich SFCA being suppressed, with SFCA being the only Ca-rich ferrite phase to crystallise from the melt during cooling of the 6 mass% TiO2 mixture.

The observation of Fe-rich SFCA has until now only been reported in synthetic samples and a specimen of pot-grate sinter, and it is unknown whether Fe-rich SFCA is observed in industrial sinter. These results indicate, however, that if Fe-rich SFCA is observed then the localised Ti concentration within the sinter blend is likely to be low. Regardless of whether or not Fe-rich SFCA is actually present in industrial sinter, this study progresses the fundamental knowledge of iron ore sintering reactions, since Fe-rich SFCA may form as an intermediate phase during cooling even if it is not preserved in cooled sinter product.


This research was undertaken on the powder diffraction beamline (10BM1) at the Australian Synchrotron (part of ANSTO), Victoria, Australia, under beamtime award AS182/PD/13350.

© 2019 by The Iron and Steel Institute of Japan