Softening-melting Properties of High-chromium Vanadium–titanium Magnetite Pellets with Different Basicity

Abstract

High-chromium vanadium-titanium magnetite (HVTM) is a critical polymetallic ore resource, and its large-scale utilization is considered feasible by smelting flux pellets in blast furnaces. This study investigated the softening-melting properties of pure HVTM pellets (100% HVTM pellets) and HVTM with 30% conventional iron pellets (70% HVTM pellets) with different basicity (R). The results indicated that the softening-melting properties of 100% HVTM pellets deteriorated when R increased from 0.2 to 1.8, and the properties of 70% HVTM pellets first deteriorated and then improved. Moreover, 100% HVTM pellets exhibited superior softening-melting properties for R<1.0, while 70% HVTM flux pellets were better at R>1.6. The gas permeability of 100% and 70% HVTM pellets decreased with increasing basicity, although the permeability of 70% HVTM pellets was higher. Notably, the key components of acid slag are Mg-bearing anosorite and pyroxene, whereas those of basic slag are perovskite and melilite. With the increase of basicity, the content of chromium in dropping iron decreased and that of vanadium increased.

1. Introduction

Vanadium-titanium magnetite is a complex polymetallic iron ore resource with comprehensive utilization potential.^1,2,3) and has attracted attention for multi-stage utilization. The process involving blast furnace ironmaking, blast furnace slag titanium extraction, converter steelmaking, converter slag vanadium extraction has been rationalized and adopted in Panzhihua Iron and Steel Company, China.⁴⁾ High-chromium vanadium-titanium magnetite (HVTM) is a chromium-containing vanadium-titanium magnetite ore found in the Hongge area of Sichuan, with a chromium content of more than 0.8%.^1,5,6) The reserves of this mineral amount to more than 3 billion tons, and their exploitation and utilization have been planned. Owing to the massive size of HVTM reserves and its potential for multi-level utilization, the blast furnace process is a feasible choice.

As basicity burden, flux pellets have more stable properties, consume less energy, and have lower pollutant emissions compared with basicity sinter.^7,8,9,10,11) To achieve China’s goals of energy conservation and emission reduction, the proportion of flux pellets added to blast furnace has been gradually increased in recent years.¹²⁾ Therefore, the use of flux pellets for the smelting of HVTM should be considered.

Softening-melting property is a critical metallurgical property of iron-bearing burden in blast furnaces.^13,14,15) Iron-bearing burden is introduced into a blast furnace from the top and moves downward under gravity. The burden reduces under the action of temperature and reduction atmosphere. The reduction is accompanied by heating, which results in the softening and melting of the burden. Consequently, the burden forms a semi-molten slag-iron mixture bonded to coke. The zone where the burden behavior transitions from melting to dropping is called the cohesive zone.¹⁶⁾ The cohesive zone is the most crucial area through which a gas passes in a blast furnace, and its relative position and thickness has notable influence on the blast furnace smelting process.¹⁷⁾ The fundamental properties of the cohesive zone represent the softening and melting properties of the burden. The lower the position of the cohesive zone and the narrower the interval, the more conducive it is to the blast furnace operation.^18,19,20,21) Furthermore, the higher the gas permeability, the more conducive it is to the smooth flow of the burden and the improvement of the blast furnace operation.^{22,23,24,25,26)}

Our previous studies analyzed the oxidation consolidation mechanism and phase transformation behavior of HVTM flux pellets.²⁷⁾ To further improve the utilization of HVTM flux pellets, their softening and melting properties should be investigated. To determine the optimum basicity of HVTM pellets with different iron ore structures and to achieve effective burden operation, the present study investigated the significance and influence of the addition of a conventional ore to HVTM in a blast furnace. The results may provide theoretical support for the further development and utilization of HVTM flux pellets.

2. Experimental

2.1. Materials

The HVTM concentrate used in this study was obtained from Panzhihua, Sichuan Province, China. Typical conventional hematite concentrate was supplied by the Panzhihua Steel Company. The key constituents of the two concentrates are listed in Table 1. The X-ray diffraction (XRD) pattern of the HVTM concentrate is displayed in Fig. 1. The micromorphology and energy dispersive spectrometry (EDS) results of the HVTM concentrate are illustrated in Fig. 2. The obtained HVTM was constituted by magnetite (Fe₃O₄), ilmenite (FeTiO₃), titanomagnetite (Fe_2.75Ti_0.25O₄), chromohercynite (FeCr₂O₄), and ferrovanadium spinel (FeV₂O₄). As depicted by the results in Figs. 1 and 2, magnetite and titanomagnetite had disproportionate dissemination, and the primary gangue contained silicate and magnesia-alumina spinel. Pure CaCO₃ (Sinopharms Chemical Reagent Co.) was used as a calcium-containing additive to prepare flux pellets. Furthermore, bentonite was used as the binder. The key physical properties of bentonite are summarized in Table 2. The HVTM concentrate and conventional iron concentrate were ground to 80% concentration with particle size <74 μm.

Table 1. Compositions of HVTM concentrate and conventional hematite concentrate (mass%).

	TFe	Fe2O3	FeO	SiO2	CaO	V2O5	TiO2	MgO
HVTM Concentrate	51.81	44.41	26.64	4.15	0.92	0.54	12.7	3.98
Common hematite concentrate	57.94	63.1	17.7	6.93	0.96	0.088	1.2	1.45

	Al2O3	Cr2O3	K2O	Na2O	S	P
HVTM Concentrate	2.81	1.03	0.08	0.12	0.16	<0.01
Common hematite concentrate	1.37	–	–	–	0.573	–

Fig. 1. X-ray diffraction pattern of HVTM concentrate. (Online version in color.)

Fig. 2. SEM-EDS analysis of HVTM concentrate. (Online version in color.)

Table 2. Key physical properties of bentonite.

Methylene blue adsorbed/g/100 g	Colloid index/%	Swelling capacity/ml/g	pH	< 74 μm/%	< 46.9 μm/%
35.70	62.0	42.0	10.44	97.79	64.72

2.2. Procedures

The experimental procedures included pelletizing, oxidation roasting and softening-melting test (S-M test). HVTM, a conventional hematite ore, bentonite and CaCO₃ were mixed according to the basicity scheme, and the roasted pellets were obtained after pelletizing, drying, and roasting. The ingredient scheme is shown in Table 3. The pelletizing and roasting parameters were described in previous studies.²⁷⁾ The roasted pellets were used for the S-M test according to GB/T 34211–2017 (Iron ores–Method for determination of iron reduction softening droppinger performance under load). Figure 3 displays a schematic of the softening–melting experimental device. The device consisted of a heating system, loading system, gas supply system, cooling system, and data recording system. The diameter and height of the graphite crucible were 75 and 180 mm, respectively. The bottom of the crucible had 19 holes with a diameter of 8 mm for airflow and droplet dropping.

Table 3. Experimental conditions of HVTM pellets (mass%).

Scheme	HVTM	Conventional iron concentrate	CaCO3	Bentonite
I-0.19	98.00	–	0.00	2
I-0.4	95.68	–	2.32	2
I-0.6	93.68	–	4.32	2
I-0.8	91.76	–	6.24	2
I-1.0	89.89	–	8.11	2
I-1.2	88.09	–	9.91	2
I-1.4	86.34	–	11.66	2
I-1.6	84.65	–	13.35	2
I-1.8	83.01	–	14.99	2
II-0.21	68.60	29.40	0.00	2
II-0.4	67.09	28.75	2.16	2
II-0.6	65.55	28.09	4.36	2
II-0.8	64.07	27.46	6.48	2
II-1.0	62.64	26.85	8.51	2
II-1.2	61.27	26.26	10.48	2
II-1.4	59.94	25.69	12.37	2
II-1.6	58.66	25.14	14.20	2
II-1.8	57.42	24.61	15.97	2

Fig. 3. S-M test equipment and charging system. (Online version in color.)

At the bottom of the crucible, 40 g of coke was placed, and its height was measured using a thickness gauge. Then, 500 g of dry pellets were placed onto the crucible and covered with 40 g of coke. Thereafter, the crucible-containing coke and pellets were transferred to the furnace. The ballast device was started with a load of 1.0 kg/cm², and the column’s original height was recorded. Subsequently, the thermocouple and pressure sensor were loaded, and the device’s air tightness was determined. The data collection was completed using automatic recording software.

Upon loading, the crucible began to heat up. The furnace temperature increased from 20°C to 900°C at a rate of 10°C/min, to 1100°C at a rate of 2°C/min, and finally to 1600°C at a rate of 5°C/min. At 200°C, N₂ was introduced at a rate of 3 L/min; at 500°C, reduction gas (mixture of 70% N₂ and 30% CO) was introduced into the furnace at a total gas flow rate of 10 L/min. At the end of the reaction, the crucible was cooled down in the furnace under N₂ flowing at a rate of 3 L/min. The heating system and atmosphere are depicted in Fig. 4. Table 4 lists the key parameters and their significance in the S-M test.

Fig. 4. Heating profile and atmosphere of the S-M test. (Online version in color.)

Table 4. Characteristic parameters and their significance in the softening and melting of HVTM pellets.

Symbol	Parameters	Significance	Unit
Δt	Shrinkage of the sample	Ratio of the sample height to the original height during the experiment	%
T10	Softening start temperature	Temperature of burden shrinkage reaching 10%	°C
T40	Softening finishing temperature	Temperature of burden shrinkage reaching 40%	°C
Ts	Melting start temperature	Temperature of pressure drop first reaching 0.5 kPa	°C
Td	Dropping temperature	Temperature of burden dropping 2 g from crucible	°C
ΔT1	Softening interval	Temperature interval between T10 and T40	°C
ΔT2	Soft melting interval	Temperature interval between Ts and Td	°C
ΔP	Pressure drop	Pressure drop of the burden in the softening-melting experiment	kPa
ΔPmax	Maximum pressure drop	Maximum pressure drop of the burden in the softening-melting experiment	kPa
S	Permeability index	Integration of pressure drop and temperature $\text{S}=\int {}_{{\text{T}}_{\text{s}}}^{{\text{T}}_{\text{d}}}\mathrm{\Delta }\text{P}\cdot {\text{d}}_{\text{T}}$	kPa·°C

2.3. Characterization

The phase composition of HVTM was analyzed through XRD using an X’PERT PRO MPD/PW3040 (PANalytical B. V. Corp., Netherlands) with Cu Kα radiation and a PIXcel 1Dsuper energy detector. The HVTM concentrate to be measured was ground to a 100% concentration with particle size >74 μm. The scanned range was 2θ = 10° to 90°. The scanning time and speed were 10 min and 8°/min, respectively. Microscopic observation and analysis of the element distribution in the samples were conducted through scanning electron microscopy (SEM, TESCAN VEGA III) equipped with EDS (INCA Energy350). An electronic voltage of 20 kV and a sample current of 20 μA were used for analysis. All components were identified using their atomic ratios obtained through EDS analysis. The phase diagram and theoretical viscosity of slag generated in the S-M test were obtained using Factsage 8.3 thermodynamic software.

3. Results and Discussion

3.1. Effects of Basicity on Softening Properties

Figure 5 displays the shrinkage versus temperature curves of 100% HVTM and 70% HVTM pellets. The softening start temperature (T₁₀) and softening finishing temperature (T₄₀) under different schemes were determined based on the curves. The softening intervals (ΔT₁) of the 100% HVTM pellets and 70% HVTM pellets are depicted in Fig. 6. At approximately 900°C, the shrinkage curves under all schemes slightly shifted upward because of the increase in volume due to the reduction of hematite to magnetite and iron whiskers. Subsequently, the shrinkage rate gradually increased with the increase in temperature, and softening began. When the basicity ranged from 0.2 to 1.8, T₁₀ of the 100% HVTM pellets decreased from 1071°C to 990°C, and T₄₀ decreased from 1174°C to 1077°C. The softening properties of the 70% HVTM pellets exhibited different trends. When the basicity ranged from 0.2 to 1.0, T₁₀ of the 70% HVTM flux pellets decreased from 1045°C to 998°C and then increased to 1061°C, and T₄₀ decreased from 1146°C to 1087°C and then increased to 1145°C. When the basicity increased from 1.2 to 1.8, T₁₀ of the 70% HVTM flux pellets increased from 1053°C to 1089°C and then decreased to 1028°C, and T₄₀ increased from 1121°C to 1166°C and then decreased to 1111°C. These results indicated that when basicity ranged from 0.2 to 0.8, the 100% HVTM pellets exhibited higher softening start and finishing temperatures. The properties of the 100% and 70% HVTM flux pellets were similar when the degree of basicity was between 1.0 and 1.2. When the basicity was increased from 1.4 to 1.8, the 70% HVTM flux pellets exhibited better softening properties.

Fig. 5. Temperature vs. shrinkage curves of (a) 100% HVTM pellets and (b) 70% HVTM pellets with different basicity. (Online version in color.)

Fig. 6. Softening intervals of the 100% HVTM pellets and 70% HVTM pellets with different basicity. (Online version in color.)

With the increase of basicity, the pellets exhibited better reducibility, the indirect reduction reaction was enhanced, and the FeO content in the slag system increased, thereby reducing the melting point of the primary slag²⁸⁾ and the softening temperature. With the gradual improvement of reducibility, the indirect reduction reaction was carried out thoroughly, and the content of iron increased, thereby increasing the deformation resistance of the iron-bearing burden and the softening temperature. Because of the increase in the amount of perovskite under the higher basicity scheme (1.0 < R < 1.8), the softening interval increased. Perovskite is brittle and has adverse effects on the compressive strength of pellets, as demonstrated by studies.²⁷⁾ On the premise of poor compressive strength of pellets, the partial reduction of flux pellets does not ensure good deformation resistance, and the softening start temperature decreased. The softening interval of the 100% HVTM pellets with higher basicity shifted upward. By contrast, the softening interval of the 70% HVTM pellets did not shift because of lower perovskite content. Upon the addition of 30% conventional iron ore to neutralize the content of TiO₂, the production of perovskite was better than that with 100% HVTM pellets, indicating the importance of perovskite.

In the bulk zone of the blast furnace, the increase in the melting point of the primary slag is more beneficial with the decrease in the FeO content because of the reduction of the pellets; this is conducive to increasing the softening temperature and shifting the cohesive zone downward. The indirect reduction should be carried out more thoroughly to considerably increase the FeO content in the primary slag. This kind of primary slag with low melting temperature and good fluidity lands in the lower part of the blast furnace at higher speeds. The direct reduction of a large amount of FeO increases the heat consumed in the lower part, thereby increasing the coke ratio and decreasing production.

3.2. Effects of Basicity on Melting Properties

The relationship between the decrease in pressure and temperature with different basicity of the 100% HVTM pellets and 70% HVTM pellets is illustrated in Fig. 7. Accordingly, specific values of the melting interval were obtained, as depicted in Fig. 8. When the basicity was within the range from 0.2 to 1.8, the melting start temperature (T_s) of the 100% HVTM pellets decreased from 1221°C to 1129°C, and the dropping temperature (T_d) decreased from 1505°C to 1474°C. When the basicity ranged from 0.2 to 1.8, T_s of the 70% HVTM pellets decreased from 1216°C to 1149°C, and T_d first decreased from 1494°C to 1459°C and then increased to 1507°C. In the basicity range from 0.2 to 1.0, 100% HVTM pellets exhibited a better melting performance. By contrast, in the basicity range from 1.2 to 1.8, the 70% HVTM flux pellets exhibited a higher dropping temperature. The melting intervals (ΔT₂) of both pellets first exhibited a decreasing trend and then an increasing trend.

Fig. 7. Temperature and pressure drop curves for the (a) 100% HVTM pellets and (b) 70% HVTM pellets with different degrees of basicity. (Online version in color.)

Fig. 8. Melting interval of 100% HVTM pellets and 70% HVTM pellets with different basicity. (Online version in color.)

With the increase in the degree of basicity, the reducibility of the pellets gradually improved, and more Wüstite entered the primary slag, thereby reducing the melting point of the slag phase. At the same temperature, the primary slag liquid phase volume under the high basicity scheme was greater and further increased the resistance of the reducing gas through the iron-bearing burden, thereby decreasing T_s. The phase diagram of the CaO–TiO₂–SiO₂–MgO–Al₂O₃ system with isotherms is displayed in Fig. 9. With the increase of basicity, the melting point of the theoretical slag system under the two schemes gradually decreased first and then increased. The trend of the dropping temperature remained the same. Although the principal component of the droplet was iron, slag may be formed under some basicity schemes. A correlation was observed between the melting point of the slag system and dropping temperature. Upon melting, an exchange occurred between the slag and iron, considerably improving the separation kinetics between them. A favorable slag–iron reaction facilitated the mass transition and element diffusion, thereby accelerating the carburization of iron and enabling iron dropping.

Fig. 9. Phase diagram of CaO–SiO₂–TiO₂(-Al₂O₃–MgO)(P=1 atm,P_O2 = 1.0 × 10⁻¹²). (Online version in color.)

The content of TiO₂ in the 100% HVTM pellets was higher. FeO generated through indirect reduction enabled the generation of a FeO·TiO₂ solid solution with TiO₂.²⁰⁾ The melting point of the FeO·TiO₂ solid solution (1669°C) was higher than that of the fayalite phase (1200°C), indirectly increasing the melting point of the primary slag. Therefore, T_s of the 100% HVTM pellets were slightly higher than that of 70% HVTM pellets. As illustrated in Fig. 9, the melting point under Scheme II was considerably lower than that under Scheme I. Therefore, T_d of the 100% HVTM pellets was higher than that of the 70% HVTM pellets when the basicity was higher than 1.0, as revealed by the S-M test. However, this did not explain the better melting performance of the 70% HVTM flux pellets when basicity was higher than 1.6. Section 3.4 presents a comprehensive analysis of this phenomenon.

3.3. Effects of Basicity on the Gas Permeability of Pellets

The gas permeability of an iron-bearing burden can be characterized using gas permeability index (S index) and the maximum pressure drop (ΔP_max), which are primarily reflected in the cohesive zone. With the increase in the S index and ΔP_max, the gas flow resistance increased and the gas permeability of the burden decreased. The gas permeability indices under different schemes are presented in Fig. 10. The changes in the S index and ΔP_max with basicity are depicted in Fig. 11. The S index of the 100% and 70% HVTM pellets exhibited an increasing trend. The S index of the 100% HVTM pellets increased from 2776.56 to 5971.92 kPa·°C with increasing basicity. The S index of the 70% HVTM pellets increased from 1513.72 to 4817.54 kPa·°C and then slightly decreased to 3800.69 kPa·°C. The gas permeability of the 100% HVTM pellets decreased with increasing basicity, whereas that of the 70% pellets was the lowest at II-1.6 and then increased. ΔP_max reflected the maximum gas resistance during the softening- melting of the pellets. As depicted in Fig. 11(b), ΔP_max of the 100% HVTM pellets increased from 20.8 to 29.42 kPa with increasing basicity (0.2–1.8), and ΔP_max of the 70% HVTM pellets increased from 12.66 to 26.48 kPa and then decreased to 21.39 kPa. ΔP_max of the 100% HVTM pellets was higher than that of the 70% HVTM pellets except when basicity was 0.4. ΔP_max of the 100% HVTM pellets increased with basicity, and ΔP_max of the 70% HVTM pellets decreased with increasing basicity and was maximum at II-1.6. Under each basicity scheme, the gas permeability of the 70% HVTM pellets was higher than that of the 100% HVTM pellets.

Fig. 10. Permeability index of the (a) 100% HVTM pellets and (b) 70% HVTM with different degrees of basicity.

Fig. 11. (a) Gas permeability index (S index) and (b) maximum pressure drop of the 100% HVTM pellets and 70% HVTM pellets. (Online version in color.)

The gas permeability of the iron-bearing burden was directly related to the viscosity and amount of liquid in the slag system. The theoretical viscosity calculation based on Factsage 8.3 was carried out under the condition of final slag. The viscosity–temperature curves of the slag are displayed in Fig. 12, wherein slag viscosity decreased with the increase of basicity. The increase of basicity promoted the dissociation of the Si–O and Al–O bonds and reduced slag viscosity.^29,30) The viscosity of the slag simultaneously decreased with the increase in the temperature. The liquid volume of the slag calculated at the dropping temperature is depicted in Fig. 13. With the increase of basicity, the area of the liquid phase region gradually increased with increasing basicity and temperature. At the same temperature, the liquid phase regions under schemes with higher basicity were larger, indicating that the amount of liquid generated in the S-M process increased with the increase of basicity. Therefore, the increase of basicity resulted in increased generation of the liquid phase with lower viscosity, thereby further facilitating the flow of the slag system. Meanwhile, the amount of slag increased with the increase of basicity. The coke layer forms a virtual channel called the coke slit through which the reducing gas passes in a blast furnace. However, when the slag volume is large and viscosity is low, the slag easily flows to the coke layer, wets and blocks the coke slit, thereby blocking the reducing gas and increasing the S index and ΔP_max.

Fig. 12. Viscosity calculations of the S-M test slag for the (a) 100% HVTM and (b) 70% HVTM with different degrees of basicity. (Online version in color.)

Fig. 13. Liquid regions of the CaO–SiO₂–TiO₂–Al₂O₃–MgO slag in the S-M test. (Online version in color.)

The S index and ΔP_max of the 70% HVTM pellets were small owing to the low TiO₂ content. A comparison between the curves in Figs. 12(a) and 12(b) revealed that the viscosity of the 70% HVTM pellet slag system was smaller at the same basicity. The low TiO₂ content reduced the production of Ti(C,N), and the small particle size and high melting point of Ti(C,N) directly caused an increase in the viscosity of the slag system.³¹⁾ Controlling the production of Ti(C,N) was one of the technical challenges to the utilization of vanadium–titanium magnetite in blast furnaces.

3.4. Effects of Basicity on Micromorphology and Elemental Distribution of Slag and Iron

The cross-section of the crucible and droplet after the S-M test is depicted in Fig. 14. The residue after the reaction was constituted by flooding slag, residual slag and iron, and dropping slag and iron. The flooding slag, residual slag and iron, and dropping slag and iron for different degrees of basicity were analyzed through micromorphology and EDS analyses.

Fig. 14. Macromorphology of residuals and droppings after the S-M test. (Online version in color.)

3.4.1. Flooding Slag

Figure 15 displays the morphology and EDS analyses results for the flooding slag of the 100% and 70% HVTM pellets with basicity degrees of 0.2, 1.0, and 1.8. At the basicity degree of 0.2 (Figs. 15(a) and 15(b)), the flooding slag consisted of a dark substrate and gray dendrite. As revealed by the EDS analysis and composition analysis of the reference titanium-bearing electric furnace slag,^21,32) he gray dendrite was composed of Mg-bearing anosorite (Mg_xTi_3−xO₅), and the dark substrate was composed of pyroxenes (Ca(Mg,Fe,Al)[(Si,Al)₂O₆]). In Scheme II-0.21, TiO₂ content was low and reduced the content of Mg-bearing anosorite, thereby changing the morphology from dendritic to strip-like. Moreover, the orientation of dendrite growth may have been different because of the different grinding planes and the composition revealed by the EDS analysis was not substantially different. When the basicity was increased to 1.0, the morphology of the flooding slag changed (Figs. 15(c) and 15(d)). In the flooding slag of I-1.0, skeletal crystals of perovskite (Fig. 15(c-2)) were observed. Meanwhile, the Mg-bearing anosorite decreased, and the calcium content of the pyroxene substrate increased, changing it to melilite (Ca₂(Mg,Al)[(Si,Al)O₇]). Under Scheme II-1.0, because of a decrease in titanium content, the content of Mg-bearing anosorite decreased and the size of perovskite dendrites increased. At basicity of 1.8 (Figs. 15(e) and 15(f)), no Mg-bearing anosorite was observed under either scheme, and the morphology of perovskite changed to agglomerate (Figs. 15(e-2) and 15(f-2)). The melilite formed under Scheme I-1.8 (Fig. 15(e-1)) contained more magnesium, whereas that formed under Scheme II-1.8 (Fig. 15(f-1)) contained more calcium. Moreover, the melting point of magnesium melilite was lower than that of calcium melilite; thus, the dropping temperature of the 70% HVTM flux pellets was higher when the basicity was 1.6. The morphology and composition of the slag indicated the better performance of the 70% HVTM flux pellets with high basicity (R > 1.6), as described in Section 3.2.

Fig. 15. Morphology and EDS analyses results for the flooding slag of the 100% HVTM pellets and 70% HVTM pellets under (a) Scheme I-0.19, (b) Scheme II-0.21, (c) Scheme I-1.0, (d) Scheme II-1.0, (e) Scheme I-1.8, and (f) Scheme II-1.8. (Online version in color.)

3.4.2. Residual Slag and Iron

Figure 16 displays the morphology and EDS analysis results for the residual slag and iron of the 100% and 70% HVTM pellets with basicity degrees of 0.2, 1.0, and 1.8. In the residual slag and iron (Figs. 16(a) and 16(b)), a large amount of Mg-bearing anosorite in the 100% HVTM scheme flooded the slag phase, almost covering the substrate when the basicity was 0.2. Under Scheme II, because of a low Ti content, magnesium silicate and pyroxene phases were observed. Meanwhile, slag and iron coexisted in the residual matter. When the basicity was increased to 1.0 (Figs. 16(c) and 16(d)), the slag phase composition was constituted by Mg-bearing anosorite, perovskite, and melilite. When the basicity was 1.8 (Figs. 16(e) and 16(f)), Mg-bearing anosorite disappeared and the grain shape of perovskite was similar to that of the flooding slag. However, the content of perovskite in the residual slag was lower than that in the flooding slag because the flooding slag was squeezed out during the formation of intermediate slag from the primary slag. Meanwhile, the slag formation time of the flooding slag was short, and the ash in the coke was not absorbed. Therefore, the basicity of the flooding slag was greater than that of the residual slag. However, the residual slag continuously absorbed ash with the melting loss of coke. At this time, the slag formation time of flooding slag was short, and the ash in coke was not absorbed. Therefore, the basicity of flooding slag was generally more remarkable than that of residual slag. However, the residual slag continuously absorbed ash with the melting loss of coke. Furthermore, the contact between the flooding slag and iron was limited, and the exchange of the elements was incomplete.

Fig. 16. Morphology and EDS analysis results for the residual slag of the 100% HVTM pellets and 70% HVTM pellets under (a) Scheme I-0.19, (b) Scheme II-0.21, (c) Scheme I-1.0, (d) Scheme II-1.0, (e) Scheme I-1.8, and (f) Scheme II-1.8. (Online version in color.)

The elements in the residual slag with different degrees of basicity were clearly observed in the scanning map displayed in Fig. 17. When the basicity was low, vanadium was primarily concentrated in the Mg-bearing anosorite. With the increase of basicity, perovskite began to appear in the slag, and vanadium primarily existed in perovskite. However, vanadium and titanium were always symbiotic regardless of the degree of basicity. The distribution law of the remaining elements was consistent with the aforementioned micromorphology and EDS results.

Fig. 17. Scanning map of the residual slag and iron under (a) Scheme I-0.19, (b) Scheme I-1.0, and (c) Scheme I-1.8. (Online version in color.)

The basicity and the content of TiO₂ affected the morphology and composition of the slag. The increases of basicity basicity resulted in a decrease in the content of Mg-bearing anosorite, and the substrate transformed from pyroxene to melilite in the slag. The melting point of the Mg-bearing anosorite was high (~1690°C), and the melting point of the slag thus decreased during the process. With the increase of basicity, perovskite appeared in the slag as an independent phase, and the melting point of perovskite was approximately 1980°C, thus increasing the melting point of the slag. The trend was the same, as described in Fig. 9, and changes in the softening and melting properties were reflected by changes in the phase diagram and slag morphology.

The results of the S-M tests indicated that the grain size in the slag of the 100% HVTM pellets was smaller than that of the 70% HVTM pellets under both acidic and basic schemes. At the same degree of basicity, the decrease in TiO₂ content resulted in decreased generation of titanium-bearing minerals and an increase in grain size. The wettability between the particles and the slag increased with increasing dispersion of refractory particles in the slag. Furthermore, the viscosity of the slag increased with the increasingly even dispersion of the particles.¹⁸⁾ This was consistent with the theoretical calculation results of viscosity of the slag (Fig. 12). The increase in the viscosity and liquid content of the slag reflected the trend of gas permeability with the increase of basicity in the S-M test.

3.4.3. Dropping Slag and Iron

The morphology and EDS analyses results for the dropping slag and iron for different degrees of basicity are illustrated in Fig. 18. The schemes with basicity degrees of 0.2 or 1.8 did not result in dropping slag, unlike the scheme with a basicity of 1.0 because the melting point of slag first decreased and then increased with the increase of basicity (Fig. 9). Meanwhile, the viscosity of the slag system was suitable and thus decreased simultaneously with iron. The dropping slag exhibited prominent rapid cooling characteristics and was glassy. Meanwhile, the composition of the droplet slag was similar to that of the residual slag, and the skeleton crystalline perovskite was distributed on the pyroxene substrate.

Fig. 18. Morphology and EDS analyses results for the dropping slag and iron of the 100% HVTM pellets and 70% HVTM pellets under (a) Scheme I-1.0 iron, (b) Scheme II-0.21 iron, (c) Scheme I-1.0 iron and slag, (d) Scheme II-1.0 iron and slag, (e) Scheme I-1.8 iron, (f) Scheme II-1.8 iron. (Online version in color.)

A comparison of the components of dropping iron with different basicity revealed that Si, V, Ti, Cr, and other metals constituted the dropping iron. Because of the limitations of EDS analysis, C content should be determined more accurately. With the increase of basicity, the content of Cr decreased, and the content of V increased; this was consistent with the distribution coefficient law of the slag. The droplet iron in the soft melting experiment corresponded to the dropping zone in the simulated blast furnace, and the iron did not react through the slag–iron interface of the slag–iron accumulation zone. Thus, the dropping iron was not comparable to the iron in blast furnace production.

4. Conclusions

(1) When basicity was increased from 0.2 to 1.8, the softening-melting properties of the 100% HVTM pellets deteriorated, whereas those of the 70% HVTM pellets deteriorated first and then improved. The 100% HVTM pellets exhibited superior softening-melting performance at R < 1.0, whereas the 70% HVTM flux pellets were better at R > 1.6. Thus, conventional iron ore should be added in HVTM flux pellets with high basicity.

(2) When the basicity was increased from 0.2 to 1.8, the S index of the 100% HVTM pellets increased from 2776.56 to 5971.92 kPa·°C. The S index of the 70% HVTM pellets increased from 1513.72 to 4817.54 kPa·°C and then slightly decreased to 3800.69 kPa·°C. With the increase of basicity, the gas permeability of the 100% and 70% HVTM pellets decreased; however, the gas permeability of the 70% HVTM pellets was higher.

(3) The main components of the acidic slag were Mg-bearing anosorite and pyroxene, which were converted to perovskite and melilite with the increase of basicity.

(4) With the increase of basicity, the content of Cr decreased and that of V increased in the dropping iron. Therefore, appropriate basicity should be selected to facilitate slag–iron interface reactions.

Conflict of Interest Statement

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Acknowledgement

This research was financially supported by the Programs of the National Natural Science Foundation of China (Nos. 52174277 and 52204309) and China Postdoctoral Science Foundation (2022M720683).

References

• 1)   W.  Li,  G.  Fu,  M.  Chu and  M.  Zhu: Powder Technol., 343 (2019), 194. https://doi.org/10.1016/j.powtec.2018.11.027
• 2)   S.  Wang,  Y.  Guo,  T.  Jiang,  F.  Chen,  F.  Zheng and  L.  Yang: JOM, 71 (2019), 1858. https://doi.org/10.1007/s11837-018-2983-0
• 3)   W.  Li,  G. Q.  Fu,  M. S.  Chu and  M. Y.  Zhu: Ironmak. Steelmak., 44 (2017), 294. https://doi.org/10.1080/03019233.2016.1210751
• 4)   J.  Wen,  T.  Jiang,  X.  Zheng,  J.  Wang,  J.  Cao and  M.  Zhou: Sep. Purif. Technol., 230 (2020), 115881. https://doi.org/10.1016/j.seppur.2019.115881
• 5)   W.  Li,  N.  Wang,  G.-q.  Fu,  M.-s.  Chu and  M.-y.  Zhu: Int. J. Miner. Metall. Mater., 25 (2018), 391. https://doi.org/10.1007/s12613-018-1583-x
• 6)   G.  Fu,  W.  Li,  M.  Chu and  M.  Zhu: Metall. Mater. Trans. B, 51 (2020), 114. https://doi.org/10.1007/s11663-019-01734-6
• 7)   F. R.  Silva,  L. R.  Lemos,  P.  de Freitas Nogueira and  M.  Bressan: Metall. Mater. Trans. B, 52 (2021), 69. https://doi.org/10.1007/s11663-020-01917-6
• 8)   R.  Wang,  J.  Zhang,  Z.  Liu,  Y.  Li and  C.  Xu: Powder Technol., 390 (2021), 496. https://doi.org/10.1016/j.powtec.2021.06.001
• 9)   P.  Prusti,  K.  Barik,  N.  Dash,  S. K.  Biswal and  B. C.  Meikap: Powder Technol., 379 (2021), 154. https://doi.org/10.1016/j.powtec.2020.10.063
• 10)   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
• 11)   D.  Zhu,  T.  Chun,  J.  Pan and  J.  Zhang: Int. J. Miner. Process., 125 (2013), 51. https://doi.org/10.1016/j.minpro.2013.09.008
• 12)   K.  Bai,  L.  Liu,  Y.  Pan,  H.  Zuo,  J.  Wang and  Q.  Xue: Ironmak. Steelmak., 48 (2021), 1048. https://doi.org/10.1080/03019233.2021.1911770
• 13)   P. F.  Nogueira and  R. J.  Fruehan: Metall. Mater. Trans. B, 35 (2004), 829. https://doi.org/10.1007/s11663-004-0077-6
• 14)   P. F.  Nogueira and  R. J.  Fruehan: Metall. Mater. Trans. B, 36 (2005), 583. https://doi.org/10.1007/s11663-005-0049-5
• 15)   P. F.  Nogueira and  R. J.  Fruehan: Metall. Mater. Trans. B, 37 (2006), 551. https://doi.org/10.1007/s11663-006-0038-3
• 16)   M.  Hayashi,  K.  Suzuki,  Y.  Maeda and  T.  Watanabe: ISIJ Int., 55 (2015), 1223. https://doi.org/10.2355/isijinternational.55.1223
• 17)   M.  Iljana,  E.-P.  Heikkinen and  T.  Fabritius: Metals, 11 (2021), 1515. https://doi.org/10.3390/met11101515
• 18)   Q.  Hu,  D.  Ma,  K.  Zhou,  Y.  Liu,  Y.  You,  Z.  You and  X.  Lv: Powder Technol., 396 (2022), 710. https://doi.org/10.1016/j.powtec.2021.11.035
• 19)   X.  Fan,  J.  Zhang,  K.  Jiao,  J.  Zhang,  H.  Ma,  S.  Gao and  R.  Wang: Ironmak. Steelmak., 49 (2022), 626. https://doi.org/10.1080/03019233.2022.2036085
• 20)   C.  Yin,  S.  Zhang,  X.  Yang,  W.  Yuan,  W.  Yu,  L.  Wen,  T.  Li and  C.  Bai: Metall. Mater. Trans. B, 52 (2021), 4096. https://doi.org/10.1007/s11663-021-02331-2
• 21)   Y.  Du,  J.  Gao,  X.  Lan and  Z.  Guo: Ceram. Int., 48 (2022), 7918. https://doi.org/10.1016/j.ceramint.2021.11.339
• 22)   C.  Yin,  S.  Zhang,  Y.  Guo,  L.  Wen and  C.  Bai: ISIJ Int., 62 (2022), 1777. https://doi.org/10.2355/isijinternational.ISIJINT-2022-071
• 23)   Y.  Xiao,  K.  Zhu,  S.  Ye,  Z.  Xie,  Y.  Zhang and  X.  Lu: Int. J. Hydrog. Energy, 47 (2022), 31129. https://doi.org/10.1016/j.ijhydene.2022.07.035
• 24)   B.  Wang,  W.  Zhao,  X.  Zhang,  S.  Hu,  H.  Guo and  M.  Chu: Minerals, 12 (2022), 842. https://doi.org/10.3390/min12070842
• 25)   K.-i.  Ohno,  Y.  Kitamura,  S.  Sukenaga,  S.  Natsui,  T.  Maeda and  K.  Kunitomo: Tetsu-to-Hagané, 108 (2022), 703 (in Japanese). https://doi.org/10.2355/tetsutohagane.TETSU-2022-025
• 26)   L.  Niu,  J.  Zhang,  Y.  Wang,  J.  Huang,  B.  Feng and  Z.  Liu: JOM, 74 (2022), 2733. https://doi.org/10.1007/s11837-022-05301-4
• 27)   B.  Chen,  J.  Wen,  T.  Jiang,  L.  Li,  G.  Yang and  T.  Zhao: Metall. Mater. Trans. B, 53 (2022), 178. https://doi.org/10.1007/s11663-021-02353-w
• 28)   H. C.  Chuang,  W. S.  Hwang and  S. H.  Liu: Mater. Trans., 50 (2009), 1448. https://doi.org/10.2320/matertrans.MRA2008372
• 29)   C.  Feng,  J.  Tang,  L.  Gao,  Z.  Liu and  M.  Chu: ISIJ Int., 59 (2019), 31. https://doi.org/10.2355/isijinternational.ISIJINT-2018-444
• 30)   J.  Ma,  W.  Li,  G.  Fu and  M.  Zhu: ISIJ Int., 60 (2020), 2408. https://doi.org/10.2355/isijinternational.ISIJINT-2020-244
• 31)   T.  Jiang,  D.-m.  Liao,  M.  Zhou,  Q.-y.  Zhang,  H.-r.  Yue,  S.-t.  Yang,  P.-n.  Duan and  X.-x.  Xue: Int. J. Miner. Metall. Mater., 22 (2015), 804. https://doi.org/10.1007/s12613-015-1137-4
• 32)   Z.  Wang,  H.-y.  Sun and  Q.-s.  Zhu: Int. J. Miner. Metall. Mater., 26 (2019), 1120. https://doi.org/10.1007/s12613-019-1830-9