Optimization of high alumina slag practice in blast furnace ironmaking: an industrial approach (PART 1: fundamental aspects)

Abstract

The productivity of the blast furnace Ironmaking process is significantly dependent on the slag chemistry. The presence of compounds like Al₂O₃ in high quantities can make the hearth slag highly viscous and difficult to tap. The objective of part-1 of this study is to provide a comprehensive analysis of the role of high alumina (Al₂O₃ > 15%) compositions in the blast furnace slags. Slag viscosity, liquidus temperature, activation energy, etc., are critically reviewed with respective solutions to operate under the desired conditions. Flow parameters of the slag are examined to understand the melting-softening phenomena which directly affect the slag tap-to-tap time. The addition of certain additives to the slag system has also been studied with the aim to improve the target properties of the slag. A range of slag compositions and their optimization techniques are discussed in this part to provide greater insight into the topic.

Introduction

Blast furnace Ironmaking is the primary route of hot metal production in an integrated steel-making process with more than 70% stake among all other routes of production throughout the world [1]. The techno-economic impact of the blast furnace ironmaking process is substantial with an estimated turnover of 900 billion USD per year accounting for 1.7 billion tonnes of steel produced annually [2]. The production is predicted to reach 2.8 billion tonnes of crude steel worldwide by the year 2050 [2]. India is estimated to be the second largest steel producer and consumer, with domestic crude steel production accounting for 124.7 million tonnes in the year 2022 [3]. National Steel Policy estimates 200 million tonnes of steel production in India by the year 2025 with more than 500 million tonnes of Iron ore requirement to be met by that time period [4]. Iron and steel production in India also accounts for more than 10 million tonnes of slag production per year [5]. This slag has the potential to serve as raw material for foundry and cement industries as a replacement for costly silica sand (SiO_2, Al₂O₃, Fe₂O₃) [5]. Proper utilization of slag produced along with the optimization of production techniques in a fast and efficient manner is, therefore, the key to scaling up future production [6]. The efficiency of the Ironmaking process is strongly influenced by the quality of slag generated in the lower region of the blast furnace, i.e. hearth slag. The fluidity of the slag affects the productivity of the blast furnace operation, and effective slag formation is essential to that operation. The viscosity and flow characteristics of the slag are influenced by its internal structure as well as the chemical composition of the slag system [7]. Changes in slag’s chemical composition significantly impact its overall properties.

Limited reserves of high-quality grades of Iron ore have led to a surge in the adoption of ores with high alumina (Al₂O₃) content [8] and therefore a substantial need to incorporate corresponding changes to maintain production standards. Figure 1 shows the surge in alumina content in Iron ore slime from Indian mines over the two decades from 2000 to 2020, which in turn indicates the subsequent presence of a higher amount of Al₂O₃ content in the burden and the slag. The high Al₂O₃ content in low-grade iron ores primarily results in the high viscosity of slag produced in the hearth region of the blast furnace. To minimize the ill effect of high Al₂O₃, higher flux consumption is required, resulting in higher slag volume, leading to a constriction of productivity and an increase in fuel consumption. Hence, the cost of hot metal production has increased while the rate of production has substantially decreased. It is therefore essential to look at the underlying structure and property correlation for blast furnace hearth slag. Acidic, basic, and amphoteric oxides are common components of metallurgical slags [9]. The structure and corresponding properties of slags are defined in terms of the network-modifying behaviour of basic components (e.g. CaO, MgO, Na₂O, K₂O, FeO, etc.) and the network-forming behaviour of acidic components (e.g. SiO₂), with a notable anomaly with amphoteric oxides (e.g. Al₂O₃) [10]. Viscosity, activation energy, and liquidus temperature of slag are some of the governing quantities for blast furnace operation, so their influence on the efficiency of the blast furnace is of primary interest [11]. The blast furnace slag forms as a low-density liquid phase, formed at a temperature range of 1600K–1700K [12]. The tapping process removes the molten metal and slag gets collected as one of the smelting products in the blast furnace hearth [13]. The properties of slag decide the ease of slag tapping which is essential to the productivity of a blast furnace. Improper tapping could leave a large amount of leftover slag in the hearth area, which would increase the blast pressure. The frequency of slag tapping and permeability in the hearth zone, are two essential characteristics that are determined by the softening and melting behaviour of slag [14]. The slag and molten metal have various discharge patterns [13] which depend on the fluidity of both along with the number of tapholes. For continuous tapping of slag and molten metal in huge blast furnaces having many tapholes, slag, and metal discharge simultaneously [15]. The molten metal on account of higher density and lower viscosity can easily flow to the taphole through a ‘coke-free’ path [16] while the slag has to go through the hearth coke bed and experience additional flow resistance. This phenomenon might result in slag accumulation below the taphole area even if the taphole appears dry enough. A porous coke bed is therefore essential to prevent slag depletion and promote proper tapping practice [17]. Slag tapping should be accomplished in a short period of time and at a faster pace than slag formation [13]. The slag tap-to-tap time is directly associated with slag viscosity which highlights the importance of always maintaining a slag with high fluidity for improving productivity and fuel consumption. Slag viscosity is also described as a property that links the slag flow behaviour to the reaction kinetics and the degree of reduction of the final slag [16]. The slag viscosity impacts gas permeability, heat transmission, and the reduction of SiO₂, FeO, and other elements [18]. It is preferable to look for slag systems that can maintain good fluidity at low temperatures. The lower the viscosity of the slag, the higher will be the slag-metal separation rate and hence the yield of the blast furnace would be higher [19]. Reducing the viscosity at a higher temperature is comparatively easier than at lower temperatures because of the increase in mobility of slag components at higher temperatures [20]. However, increasing the temperature can significantly increase energy expenditure, so a proper combination of low temperature and low viscosity is a key to blast furnace productivity. Because of the high Al₂O₃ concentration in slag, the pressure differential in the bottom section of the blast furnace will increase, the column's permeability will deteriorate, and the temperature will rise. Entrapment losses of metallic Iron are highly likely when viscous slag is formed [14].

Figure 1.

Alumina content in Iron ore slime of Indian mines [29–36].

Slag liquidus temperature is an essential property in determining the requirement of energy for tapping out of slag. The lower the liquidus temperature, the lower will be the viscosity at comparatively lower temperatures [21]. Many systems have strong temperature-dependent Arrhenius activation energies [22] for viscous flow, which describe their viscosity-temperature characteristics [22]. The activation energy for the viscous flow of slag also dictates both the values of slag liquidus temperature and slag viscosity [23]. Therefore, reducing the activation energy for viscous flow and thereby increasing slag fluidity are the main objectives of excellent slag production. The slag’s specific heat capacity (Cp) is another crucial measurement that closely matches the specifications of blast furnace slag. A drop in the value of Cp will result in a rise in blast furnace temperature, which will result in decreased slag viscosity for a constant heat input into the blast furnace [24]. Enthalpy change of slag formation, slag melting and phase transformation are measured to determine the endothermic capacity of the respective processes and subsequently link it with energy requirement. A rise in the value of basicity can lower the enthalpy of the system. This is owing to the presence of a high content of CaO that can assist in network modification of silicate and aluminate structures and would subsequently reduce the endothermic need of the system for network breaking. It can therefore help in maintaining a uniform temperature [25]. Another crucial characteristic of the slag is its surface tension, which is related to how reactive its top surface is at the slag-metal interface. The elements with the lowest surface tension (referred to as ‘surfactants’) will preferentially occupy the surface layer (e.g. S in steels and surfactants like B₂O₃ in slags). Surface concentration is determined by surface tension as well as the chemical activity of the component which is governed by the Gibbs adsorption isotherm [26]. The sulphide capacity and desulphurization potential are two other important characteristics necessary for maintaining a reasonable level of operation [27]. The sulphur partition between slag and hot metal is also based on the optical basicity and MgO/Al₂O₃ ratio [28].

On the basis of the above-mentioned factors, this paper highlights different techniques used in blast furnaces and their consequent applications in the modern industrial ironmaking process with an aim to incorporate possible solutions to operate with the high Al₂O₃ regime of slag.

The alumina anomaly

Al₂O₃ acts as an amphoteric oxide in the blast furnace slag system on account of different concentrations and operating conditions [37]. Due to a borderline electronegativity value of 1.5 [38], Al₂O₃ can show both acidic and basic behaviour. It has been observed that in the CaO–SiO₂–MgO–Al₂O₃ type of slags, Al₂O₃ acts as a network former up to 10 wt-% concentration and consequently acts as a network modifier for composition greater than 10 wt-% [39]. The network-forming ability accounts for an increase in viscosity value via the incorporation of [AlO₄]^- tetrahedra into the [SiO₄]^- units. At the same time, an increase in the number of [AlO₆]^- octahedral units at higher concentrations, triggers the network modifying ability. The Si_xO_y ^z- and Al_xO_y ^z- in the aluminosilicate slag are substantially larger than the cations, therefore their movement also requires a significant amount of energy [40]. These composite anions then turn into the slag’s main structural components and restrict its movement. Fourier Transform Infrared (FT-IR) studies have been used to experimentally verify the amphoteric nature of Al₂O₃ which is subsequently supported by thermodynamic calculations [41–45]. At around the basicity of 1.0–1.2 (which is common for blast furnace slags), an increase in Al₂O₃ and/or MgO content increases the area of spinel (Al₂O₃.MgO) in the phase diagram, resulting in a high value of liquidus temperature [40]. It has been deduced using the excess stability function [46] that at a composition of X_Al2O3 = 0.11–0.12, the function shows a peak value indicating a sharp change in the chemical stability of Al₂O₃ leading to a behaviour that confirms its amphoteric nature. In one study [47] for high Al₂O₃ slags, it was observed that the viscosity of slag increased up to 20 wt-% concentration of Al₂O₃ and then showed a decreasing trend. Depending on the basicity value and wt-% MgO in alumino-silicate slags, the network-forming ability of Al₂O₃ can reach up to 21.6 wt-% concentration of Al₂O₃ [48] which is a case of concern for industries operating anywhere between 18 and 21 wt-% Al₂O₃. For a number of slag systems, keeping the temperature fixed, increasing the Al₂O₃ content up to 24 wt-% increases the viscosity, thereafter which the viscosity decreases [49]. The relative movement resistance of Al–O having a bond length of 1.87 Å is weaker than that of the Si-O bond length of 1.60 Å.

Effect of high alumina on thermophysical and thermochemical properties of slag

Effect on density and surface tension

Using the principle of buoyancy, it has been observed that at constant basicity and low concentration, Al₂O₃ reduces slag density, whereas, at a higher basicity value and high Al₂O₃ content, the trend is reversed. Density increases with increasing Al/Si ratio and larger degree of polymerization (DOP ratio) [40]. The composite anions become more tightly linked as the degree of polymerization rises, increasing density [50]. It was discovered using a similar method that uses the maximum bubble pressure (MBP) method [51] that the surface tension increased as the Al2O3 content and the ratio of (wt-% Al2O3)/(wt-% SiO2) increased [40]. The increase in the magnitude of these quantities is closely related to dynamic properties like fluidity and consequently resistance to flow mobility, i.e. viscosity. Higher surface tension may not imply a higher viscosity but the cohesive force between molecules might be an indication of the inherent tendency of the system to resist deformation and hence indicate higher viscosity. So, an increase in the Al₂O₃ content below the inflection point can result in higher density, surface tension, and consequently higher viscosity.

Effect on sulphide capacity

Blast furnace ironmaking aims for a reasonable amount of desulphurization. The sulphur partition between slag and hot metal is dependent on the sulphide capacity, optical basicity, CaO/SiO₂, and MgO/Al₂O₃ [52,53]. Thermophysical properties like viscosity along with thermochemical properties like sulphide capacity are strongly dependent on the underlying slag structure. Observations made on the role of different compositional adjustments [40] on high Al₂O₃ slag predict that sulphide capacity in CaO–MnO–MgO–SiO₂–Al₂O₃ slag with high CaO and MnO concentration is unaffected by the substitution of Al₂O₃ with SiO₂. Al₂O₃ has a better desulphurization capacity than SiO₂ but is less desulphurizing than MnO, CaO, MgO, and FeO [54]. The sulphide capacity of the slag increases significantly when the (wt-% Al₂O₃)/ (wt-% SiO₂) is increased from 0.47 to 0.79 for a certain CaO concentration and very low when it exceeds 0.79 [40].

Industrial impact of high alumina burden in the blast furnace

The increase in Al₂O₃ content in Indian blast furnace slag systems is depicted in Figure 2. The increase in Al₂O₃ content from 3.5 to 4.5 wt-% can cause a significant volume change of 70 kg/THM of slag leading to a significant drop in productivity. The concern for a high Al₂O₃ burden is significantly higher for Indian ores because of their lower grade.

Figure 2.

Rate of increase of slag volume as a function of Al₂O₃ with time predicted by Tata Steel.

Figure 3 shows the comparative content of Al₂O₃ and SiO₂ in iron ore for seven countries that are major producers of crude steel and the Miko deposits of Uganda. It can be observed that though the silica content is on the moderate side for Indian ores, the Al₂O₃ content is very high, which poses a major challenge for industries to cope within the rest of the world. On the other hand, even industries in developed countries like Japan are significantly importing high alumina grade ores for blast furnace operation [55]. This is significantly affecting productivity and increasing the cost of crude steel. The high amount of Al₂O₃ in the burden affects the gas permeability at the lower part of the furnace as a result of increased slag hold-up in the pores of the burden [55]. The motivation for using low-grade iron ores containing high Al₂O₃ is primarily the lower cost of these ores [56]. However, the trade-off between the lower cost of raw materials and the higher technical cost of operation is yet to be balanced in choosing suitable techniques that can modify the aluminate network. The primary melting temperature, slag hold-up, and slag viscosity are all controlled by the amount of Al₂O₃, therefore industries must continually find ways to deal with the high Al₂O₃ burden. Apart from the slag tapping difficulties, the presence of high Al₂O₃ content in the ore is also affecting the sintering plant and its productivity [57]. High sintering temperature and longer sintering time have been the need to meet the poor reactivity of the sinter owing to high Al₂O₃ content. Hence, processes like the selective granulation process [58] having increased adhesion of CaO with the sinter are being adopted to minimize the above-mentioned effects. The sinter basicity (CaO/ Al₂O₃) is being increased to counteract the aluminate network [59].

Figure 3.

Al₂O₃ vs SiO₂ content in different countries (A: Australia, B: Brazil, C: China, I: India, R: Russia, U: U.S.A, and Ug: Uganda) [60].

In addition to that, the process involving the separation of slag and hot metal is also affected owing to the high Al₂O₃ content in the slag [61]. The liquidus temperature is significantly increased owing to higher Al₂O₃ content which increases the temperature requirement of the blast furnace. Therefore, both the cost of operation and working conditions are affected in the industries. In the following section, a relationship between the slag’s physical properties and the process parameters of the blast furnace is discussed to highlight possible reasons and methods to deal with the high alumina burden.

Structure-based studies on alumina bonding

Alumina exists in several stable and metastable polymorphic forms like α [62], β [63], γ [64], δ [65] and others. The α-alumina phase is stable at temperatures above 1300 K [62] with an ordered lattice. Using a neutron diffraction pattern [66], it has been revealed that amorphous alumina contains densely packed oxygen atoms with an Al–O coordination number of 4.66 [66]. Diffraction patterns have predicted linear dependence of Al–O coordination number with temperature according to the relation: n_Al-O = 4.41–[1.25 × 10⁻⁴](T- T_m) [67]. Al₂O₃ recovered from the blast furnace slag can be employed in various applications depending on its structure, coordination number, and economic significance [68]. γ-alumina is a commercially important phase for application in manufacturing industries whose structure is revealed to be a defective cubic-spinel by XRD and neutron diffraction studies [69]. δ-alumina on the other hand has been studied using microscopy imaging and XRD to reveal its two intergrowth phases- δ₁ and δ₂- alumina [70]. δ-alumina along with other metastable phases are employed in catalytic applications in industries [71]. Using Magnetic angle spinning (MAS) and Nuclear magnetic resonance (NMR) [72] studies, the AlO₄ and AlO₆ structural units have been detected which have variable concentrations depending on the type of phase. The distribution of aluminium ions at different sites- tetrahedral and octahedral – leads to the formation of these structural units. Amorphous alumina shows yet another kind of Penta-bonding state of AlO₅ which can exist at ambient temperatures [73]. By the combined alumina-silica MAS-NMR analysis, the amorphous nature of silica has been observed to be maintained at annealing temperatures whereas alumina has been observed to crystallize at temperatures greater than 1273 K [74]. Through the FT-IR study [75], the presence of Si–O–Si and Si–O–Al bonds in alumino-silicate slags has been detected in different orders of magnitude. The longer Al–O bonds are less mobile and therefore contribute to the slag's increased viscosity [50]. The dominant presence of Al tetrahedra units is further supported by NMR analysis [75].

Increased Al₂O₃ content in blast furnace slag has been shown to produce high-crystallinity and high melting point minerals like spinel and Akermanite [76]. The slag viscosity is predominantly increased owing to these crystalline phases being present at high temperatures. The process parameters have to be adjusted in an optimal manner to decrease the relative amounts of these crystalline phases from a thermodynamic point of view.

Studies on measurement of slag properties

Viscosity measurement

The rotating cylinder method at elevated temperatures is the most widely used technique for estimating viscosity [77]. This technique involves the use of a viscometer that consists of a crucible and a bob. The viscosity is calculated using Newton’s law of viscosity at different rotational speeds in the crucible filled with molten slag. Another viscosity testing method involves an oscillating crucible which estimates the viscosity based on the oscillation period [78]. Many studies have been done for measurements of viscosities of MgO–Al₂O₃–CaO–SiO₂ quaternary slags at various compositions and temperatures [79–82] to determine a proper operating regime for the respective slag systems. On the basis of measurements done by various studies, viscosity-temperature graphs are plotted for slag systems with Al₂O₃> 18%. The graphs are classified in the range of 15–20%, 20–25%, and 25–30% Al₂O₃.

Figure 4 has 11 slag systems in which the Al₂O₃ concentration is in the range of 15–20% with other slag constituent compositions as mentioned in Table 1. Series-7 has the lowest viscosity value because the Al₂O₃ content is lower than any of the other systems [56]. According to the Al₂O₃ content, the highest viscosity should be for series-11 as the alumina content is greater than any other system, but owing to the effects of MgO and C/S ratio on the viscosity, series-4 has the highest viscosity values [83]. In series-8, the TiO₂ content is 10% while series-9 has 10% Fe₂O₃ which has a greater effect on viscosity than TiO₂ concentration. Therefore, the viscosity of series-9 is less than series-8.

Figure 4.

The effect of temperature on the viscosity of slag systems with an alumina percentage of between 15 and 20%.

Table 1.

Chemical compositions of slag systems for 15–20% Al₂O₃ (wt-%) [20,56,96].

Series	Al₂O₃	MgO	C/S	TiO₂	Fe₂O₃
1	17.3	8.9	1.1	–	–
2	17.2	5.8	1.10	–	–
3	17.2	3	1.09	–	–
4	19	5	1.0	–	–
5	18	10	1.0	–	–
6	17	15	1.0	–	–
7	16	20	1.0	–	–
8	18	–	1.0	10	–
9	18	–	1.0	–	10
10	18	9	1.1	–	–
11	20	11	1.1	–	–

Figure 5 shows the viscosity-temperature relationship of 8 slag systems with Al₂O₃ concentration in the range of 20–25% with other slag constituent compositions as mentioned in Table 2. It can be observed that series-1 has the lowest viscosity in comparison to all other slag systems. In series-1 and series-2, Al₂O₃ content is the same but series-1 has the presence of TiO₂ which plays a major role in decreasing viscosity [84]. In the case of series-6, series-7 and series-8, the concentration of Al₂O₃ is 20% but the viscosity of series-7 is the highest. In comparing series-6 and series-7, the MgO content has more effect on viscosity as compared to the C/S ratio [85]. The viscosity of series-8 is less than that of series-6. Series-4 and series-3 have approximately the same values of viscosity and the rate of change in viscosity with temperature is more in series-4 compared to series-3.

Figure 5.

The effect of temperature on the viscosity of slag systems with an alumina percentage of between 20 and 25%.

Table 2.

Chemical compositions of slag systems for 20-25% Al₂O₃ (wt-%) [20,104,81].

Series	Al₂O₃	MgO	C/S	TiO₂
1	21.2	4.8	1.231	2.17
2	21.2	5.01	1.15	–
3	22.2	4.66	0.93	–
4	23.7	4.94	0.87	–
5	23.3	4.7	0.85	1.82
6	20	–	1	–
7	20	5	0.69	–
8	20	5	0.875	–

Figure 6 contains 15 slag systems based on their Al₂O₃ content in the range of 25–30% with other slag constituent compositions as mentioned in Table 3. In the case of series-1 and series-4, the MgO content is 12% in series-1 while it is 16% in series-4, yet the viscosity of series-1 is less than that of series-4. For the temperature of 1800K, series-8 has the highest viscosity value. For temperatures above 1800K, series-5 has the highest viscosity value. In the case of series 10-12, series-12 has the comparatively lowest viscosity. As the C/S and TiO₂ content in series-12 is high, the viscosity is lower. Series-10 has lower viscosity values than series-11 because the C/S ratio is above 1 in the case of series-10 whose effect is greater than the effect of TiO₂ in series-11. In the case of series 13-15, series-13 has the lowest viscosity followed by series-14 and series-15. The composition of series-13 and series-14 are similar, only the C/S ratio of series-13 is higher resulting in lower viscosity. In the case of series-14 and series-15, the C/S ratio of series-15 is higher but TiO₂ content in series-14 is higher than in series-15. So, the viscosity of series-14 is less compared to series-15.

Figure 6.

The effect of temperature on the viscosity of slag systems with an alumina percentage of between 25 and 30%.

Table 3.

Chemical compositions of slag systems for 25-30% Al₂O₃ (wt-%).

Series	Al₂O₃	MgO	C/S	TiO₂	B₂O₃
1	25	12	1.2	–	–
2	25	16	1.6	–	–
3	25	8	0.8	–	–
4	25	16	0.8	–	–
5	25	8	6.97	–	4
6	25	8	5.92	–	4
7	25	8	5.99	–	4
8	28.5	2.28	0.72	–	–
9	26.1	1.98	0.94	–	–
10	25.3	2.16	1.08	–	–
11	25.8	2.11	1	1.86	–
12	25.2	2.14	1.19	2.28	–
13	25	8	1.05	2	–
14	25	8	0.95	2	–
15	25	8	1.05	1	–

Liquidus temperature measurement

High-Temperature Microscopy is the most widely used technique to determine the liquidus temperature of a given slag system. In the microscopy assembly, the sample is shaped into a cube and then heated in an electric furnace. When the sample’s form is altered, it is photographed with the use of a camera. The four temperature properties of the slag are identified using grid division. In some studies [86], a 5 mm cube is used for the measurement of flow characteristics of the slag while in others [87,88,89], a 3 mm cube is used for the corresponding measurements. The grid division that is shot at the same time as the sample, as well as the temperature where the sample is heated, may be used to identify four temperature characteristics [90]. Initial deformation temperature (IDT) is measured when the sample begins to distort [91]. When the sample decreases by one division or softens, it is classified as softening temperature (ST) which is below the solidus line. ST is measured below the solidus line where the sample is in a completely solid state but has softened to a high degree. The temperature is classified as hemispherical temperature (HT) when the sample takes on the shape of a hemisphere with a height-to-diameter ratio of 1:2 [92]. The sample’s height is lowered to one-third of the sample’s height at HT, and the result is Flow temperature (FT) [82]. The softening and corresponding melting takes place for the phase with the lowest melting point and that subsequently dissolves the other phases around it. The determination of characteristic temperatures done in previous studies can be analysed to understand their impact on slag properties.

Activation energy measurement (Arrhenius)

The viscosity and activation energies of Al₂O₃–MgO–TiO₂–CaO–SiO₂ slag at binary basicity of 1.07–1.50 has been determined for MgO concentrations of 7% and 11% [49]. The activation energies were calculated to be 178 and 232 kJ/mol, respectively. For Al₂O₃ concentrations of 12% and 18%, the activation energies were 273 and 360 kJ/mol, respectively. In yet another study on the CaO–SiO₂(C/S = 1)–Cr₂O₃–Al₂O₃ slags [93] for finding the effect of Al₂O₃ on the viscosity, critical temperature and activation energies were determined for viscous flow. The critical temperature is also referred to as the breakpoint temperature at which the burden starts softening and getting into a mushy zone before melting. For 17% Al₂O₃, the critical temperature was recorded to be 1612 K and the activation energy was 181 kJ/mol. For 29% Al₂O₃, the critical temperature was recorded to be 1668 K and the activation energy was 210 kJ/mol. Viscosity measurement of CaO–SiO₂–Cr₂O₃–MnO–Al₂O₃–MgO type slags was done in a study [94] where the C/S ratio was taken as 1.01 and MnO varied from 0 to 3%. For 0% MnO, the breakpoint temperature (T_B) was recorded to be 1665 K and the viscous activation energy was calculated to be 218 kJ/mol. T_B is defined as a temperature where the viscosity dependence on temperature does not strictly follow the Arrhenius equation [95]. The measurement of T_B can help in the measurement of viscosity below the liquidus temperature of the slag that characterizes the disappearance of the liquid phase and the appearance of large-scale solidification. For 3% MnO, T_B was recorded to be 1632 K and the viscous activation energy was seen to be a minimum with the value of 186 kJ/mol. The viscosities and activation energies for Al₂O₃–MgO–CaO–SiO₂–TiO₂–FeO slags were calculated under the conditions of 10% to 18% Al₂O₃ with binary basicity of 1.3 [96]. The lowest activation energy was observed for 12% Al₂O₃ which was 85 kJ/mol and the highest was observed for 18% Al₂O₃ which was 116 kJ/mol. For the calculation of melting temperature and viscosity, the rotating cylinder method was employed in one of the studies [97] where the activation energies for the viscous flow of FeO–MnO–Al₂O₃–MgO–TiO₂–CaO–SiO₂ slag were determined by varying the Al₂O₃ content from 15 to 24%. For 15% Al₂O₃, activation energy was calculated to be 157 kJ/mol and the melting temperature of slag was recorded to be 1630 K. Similarly, for 24% Al₂O₃, activation energy was calculated to be 172 kJ/mol and the melting temperature was 1658 K.

Slag property prediction models

Viscosity-prediction models

The models as mentioned in Table 4 with their relative performances can give a rough idea about the viscosity of high alumina slag systems at different compositions and temperatures. The models and their improved versions with appropriate parameters can be used to determine a suitable composition of blast furnace slag to be engineered for a given operational regime. The above RMSE and RAE values are representative of fitting the models into a number of slag systems but deviations can be expected with application-specific calculations. It is advisable to go through the reference for calculating viscosity and comparing the model performances for a given set of coordinates one is working with.

Table 4.

Different Models for Viscosity Calculation [98].

Sr. no.	Model	Equation	Relative average Error [16]	Root mean square error (RMSE) [16]
1	Urbain Model		1.958	1.597
2	Urbain and Mills		0.285	0.424
3	Kondratiev & Jak	m = m_Ax_A + m_Cx_C + m_Fx_F +m_Sx_S	5.132	4.799
4	Forsbacka		0.242	0.490
5	Ray and Pal Model		0.305	0.641
6	Mills and Sridhar Model		0.605	1.796
7	Shankar Model	Ln η = ln A +	–	–
8	KTH Model		0.447	0.310
9	Iida Model	η (Pa.s) = A×η₀×exp(E/B_i)	0.302	0.391
10	Zhang Model		0.259	0.497
11	Lei Gan Model		0.211	0.239

Prediction of liquidus temperature characteristics

The liquidus temperatures- ST, FT and HT, as discussed in section 6, are modelled for the Al₂O₃–CaO–SiO₂–MgO–TiO₂ type of slag [21]. The model equations are found to be:

Here, R is the basicity, M is the MgO concentration, and T is the TiO₂ concentration. The validation of the model indicates that the range in which the model can be applied is: C/S (0.9–1.4), MgO (6–12%) and TiO₂ (0.1–1%).

Simulations using Factsage

Factsage calculations can give a relative idea of the effect of blast furnace constituents on their thermophysical properties. However, this may not have the values corresponding to actual blast furnace conditions and are to be used only for reference. The impact of Al₂O₃, MgO, and C/S ratio on the viscosity of the slag system is depicted in Figures 7(a–c). The MgO concentration and C/S ratio were set to 8 weight per cent and 1, respectively, for the effect of Al₂O₃, and the viscosity values were anticipated at three different temperatures. Similarly, for the effect of the MgO and C/S ratio on the viscosity of the slag system, the Al₂O₃ composition was fixed at 18 wt-%. For all the calculations, the TiO₂ value was fixed at 1 wt-%.

Figure 7.

(a) Effect of Al₂O₃ on viscosity (b) Effect of MgO on viscosity using Factsage (c) Effect of C/S ratio on viscosity using Factsage.

The variation of liquidus temperature was determined for the similar variation in Al₂O₃, MgO and C/S ratio in Figure 8(a–c) respectively, which indicates that the concentration of basic oxides can only be increased to a certain value beyond which for a target reduction in viscosity, there will be an associated increase in the liquidus temperature. So, the optimization of both properties has to be done to ensure a valid state of variables for the operation.

Figure 8.

(a) Effect of Al₂O₃ on liquidus temperature using Factsage (b) Effect of MgO on liquidus temperature using Factsage (c) Effect of C/S on liquidus temperature using Factsage.

Observations on compositional changes of high alumina slag

The choice of slag system for the review has been done around the regime of blast furnace slag containing high Al₂O₃ (>15%) with other slag systems being included that have a close resemblance with the system of interest. The slag systems have been identified and analysed based on the following key steps:

Identification of slag system based on % Al₂O₃. The system is considered to be a high alumina slag system if Al₂O₃ > 15% and vice-versa.

Identification of actual slag composition based on %Al₂O₃, C/S ratio, A/S ratio, %MgO, %TiO₂, %MnO and other minor constituents of the slag.

Determination of slag properties like viscosity and liquidus temperature from the respective research paper/article/thesis, under standard operating conditions.

Conclusion on a solution to operate the respective slag system by adjusting operating parameters and controlling slag composition.

The observations made using the research methodology are listed in the following section with their respective compositional data and solutions to operate them under a high alumina slag environment.

Influence of MgO adjustment on high alumina slag

In the blast furnace, increasing the MgO concentration of the molten slag is a way for ensuring its stability and fluidity as shown in Figure 9 [83, 99]. When the MgO level is greater than 8% by weight, slag has improved fluidity and thermal stability [103]. In a study conducted at an operating temperature of 1723–1773K for composition no.1 in Table 5, the slag viscosity showed a range of 0.48–1.19 Pa.s. With the addition of MgO, the slag viscosity decreased constantly at a fixed temperature of 1773K. For proper functioning, the high alumina slag with 20% Al₂O₃ required a composition of around 10% MgO [99].

Figure 9.

Influence of MgO on High Alumina Blast Furnace slag viscosity [99].

Table 5.

Chemical compositions of slag systems (wt-%) for studying effect of MgO [21,97,100,101,102].

No.	Al₂O₃	C/S	MgO	Others
1	20	1–1.2	5–13	-
2	25	0.65	4–10	1–4
3	12–18	1.07–1.5	7–11	5
4	18.25	1.04	9	1
5	10–18	1.3	8–12	15
6	20	0.9–1.4	6–12	2.7

For composition no.2 of Table 5, the slag viscosity was estimated to be 0.50 Pa.s at a temperature of 1823K for 10% MgO. Slag viscosity and activation energy for viscous flow can be decreased by increasing the MgO % from 4% to 10% [100].

In yet another study, using composition no.3 of Table 5, slag viscosity was estimated to be less than 3 poise for 11% MgO and C/S = 1.17 at a temperature of 1773K. The solution to operating with this system involves increasing the MgO content up to 11% which significantly brings down the viscosity of the slag. By increasing the value of basicity of the slag from 1.07 to 1.50, the MgO content has to be kept fixed at 8% so that viscosity decreases with an increase in basicity at an optimum rate [101]. For composition no.4 of Table 5 (Al₂O₃ = 18.25%), the liquidus temperature was reported to be 1623 K and viscosity to be 2.84 poise at an operating temperature of 1798K. For 22% Al₂O₃, the liquidus temperature is reported to be 1623 K and viscosity to be 2.83 poise at 1798K. The liquidus temperature and viscosity of the slag are lowered when the MgO concentration in the slag is raised to 14% while the Al₂O₃ content is maintained at 18.25%. On increasing the MgO concentration beyond a value of 14%, the liquidus temperature of the slag system increases. The Blast furnace (BF) can be operated with 22% Al₂O₃ slag without the use of an external fluidizer if the slag basicity is kept at 1.0 and the MgO content is kept at 12.5% [102]. For slag composition no.5 in Table 5, viscosity was determined to be in the range of 0.65–0.70 Pa.s for 8% MgO and 0.25–0.30 Pa.s for 10% MgO. Viscosity can be lowered by increasing the MgO concentration up to 10% at 1698 K for high alumina slag. When 15% Al₂O₃ is used, the viscosity decreases faster than when 18% Al2O3 is used. When the slag composition reaches 10% MgO and 12% Al₂O₃, which means that MgO serves to improve the slag viscosity while the increase in Al₂O₃ concentration decreases the viscosity, this is an uncommon situation [97]. The MgO and TiO₂ concentrations of the BF slag are both involved in affecting the slag's characteristic temperatures. A proper percentage of these two in the form of additives or burden adjustment can help in optimizing the required properties [21].

Influence of CaO/SiO₂ ratio on high alumina slag

For the study done using composition no.1 of Table 6, at a standard operating temperature of 1723K–1773K, the slag viscosity showed a range of 0.4–0.65 Pa.s. Increasing the slag basicity from 1.0 to 1.3 reduces slag viscosity by 0.4 Pa.s to 0.2 Pa.s at a given and high MgO % (e.g. 14%) [19]. For BF slags using the new definition of basicity (Optical basicity of basic component/ Optical basicity of acidic component), a basicity value of 1.4-2.5 is estimated to be in the optimum range for BF slag [104]. In another study done with composition no.3 of Table 6, slag viscosity was minimum for a C/S ratio of 5.04 and 30% Al₂O₃ (viscosity = 1.51–1.7 Pa.s at 1803K and 1743K respectively). In the temperature range of 1773–1803K, the slag system showed high viscosity for low basicity values, i.e. 9.4–26.4 Pa.s. With an increase in basicity, the viscosity value dropped to a range of 1.5–6.1 Pa.s, indicating the need for higher basicity in the case of high Al₂O₃ slag [105]. In the case of CaO–SiO₂–Al₂O₃–MgO–FeO slag, the Al₂O₃ acts as an acidic oxide, so the relative amount should be lowered by introducing a higher basicity ratio [48]. The viscosity of high Al₂O₃ blast furnace slags decreases with basicity, with basicity ranging from 0.46 to 0.8. Even with a little amount of TiO₂ (2%) added to the slag at high basicity (0.8), the slag viscosity falls [104]. At an operating temperature of 1873K the slag viscosity is reported to be 8.6 poise and the liquidus temperature to be 1783K for composition no.6 of Table 6 [106]. At the same operating temperature, the slag viscosity was reported to be 3.5 poise and liquidus temperature as 1788K for composition no.7 of Table 6 [106]. A lower Al₂O₃ percentage is required to reduce viscosity, which can be aided by increased lime addition in the blast furnace, which will also result in a drop in silica concentration in hot metal. However, excess lime addition can increase the liquidus temperature, so an optimal lime composition and basicity are required to operate with high Al₂O₃ slag [106].

Table 6.

Chemical compositions of slag systems (wt-%) for studying effect of CaO/SiO₂ [19,48,104,105,106].

No.	Al₂O₃	C/S	MgO	Others
1	20	1–1.3	5–14	-
2	18–27	0.8–1.2	0–10	0–5.5
3	15–30	5–8	8	4
4	24.42	1.27	7.73	19.23
5	21.2–28.5	0.46–0.8	1.98–5.1	-
6	30.3	0.664	-	-
7	21.4	0.681	-	-

Influence of TiO₂ on high alumina slag

Figure 10 shows the dependence of viscosity on TiO₂ content for different temperatures. For composition no.1 of Table 7, at an operating temperature of 1773K–1823K, the slag viscosity was reported to be 5.40 dPa.s at 1% TiO₂ and it decreased to a value of 3.81 dPa.s at 5% TiO₂. The viscosity of the slag reduced as the TiO₂ level (1–5 weight %) increased. The modest TiO₂ level in high alumina BF slag continues to be helpful in lowering the Arrhenius viscous activation energy. The critical point for viscosity is at a composition of 5% TiO₂. Above this point, there is no trend in viscosity [107].

Figure 10.

Influence of TiO₂ on slag Viscosity [109].

Table 7.

Chemical compositions of slag systems (wt-%) for studying effect of TiO₂ [97,107,108].

No.	Al₂O₃	TiO₂	C/S	MgO
1	22–28	1–5	0.92–1.27	9
2	23	0–6	1	8
3	10–18	2–14	1.3	8
4	20–30	0–2	0.8–1.2	0–8

In another study done with composition no.2, at an operating temperature of 1723K and with a TiO₂ content of 6%, slag viscosity was observed to be 0.50 Pa.s. At 6% TiO₂, the activation energy for the viscous flow of the slag was found to be 183 kJ/mol, while the critical temperature was reported to be 1620 K. The slag viscosity decreased with an optimum TiO₂ composition of 6% by weight which brought down the viscosity from 0.8 Pa.s at 1723K (with 0% TiO₂) to approximately 0.5 Pa.s with 6% TiO₂. The activation energy and slag critical temperature could be brought down using the composition of 6% TiO₂ [109]. Slag viscosity can be decreased by increasing the Al₂O₃ content from 15% to 18% along with a continuous addition of TiO₂ from 0 to 6%. Weak Si–O–Al linkages become increasingly prevalent when Al₂O₃ exceeds 15%, resulting in a drop in viscosity. Al₂O₃ with 18% concentration by weight and 6% TiO₂ is one of the compositions to achieve the lowest critical temperature (CT) and minimum viscosity in case of high alumina and TiO₂ bearing slags [97]. For composition no.4 of Table 7, the value of viscosity was estimated to be in the range of 0.6–0.8 Pa.s at a temperature of 1750K. Slag viscosity reduced as the basicity increased. TiO₂ has no great influence on viscosity at low levels of basicity (e.g. 0.5), whereas viscosity reduces with TiO₂ addition at higher levels of basicity (0.8).

Influence of MgO/Al₂O₃ (M/A) ratio on adjustment on high alumina slag

The viscosity variation with MgO/Al₂O₃ ratio at different temperatures is depicted in Figure 11. For composition no.1 of Table 8, the slag melting temperature was reported to be 1599 K and the slag critical viscosity was reported to be 1.61 Pa.s. For MgO/Al₂O₃ < 0.8, with increasing MgO/Al₂O₃ ratio, slag viscosity and melting temperature showed a decreasing trend. However, MgO/Al₂O₃ = 0.8 is the inflection point above which melting temperature and viscosity increase with increasing MgO/Al₂O₃ ratio. Slag with MgO/Al₂O₃ = 0.78 has the lowest melting temperature [82]. For Al₂O₃= 18.93% and MgO/Al₂O₃= 0.4, the viscosity is estimated to be 0.39 Pa.s. There is no constraint on the MgO/Al₂O₃ ratio if Al₂O₃ concentration is less than 14 weight %, and the amount of MgO in slag may be estimated based on hearth temperature and desulphurization. When the Al₂O₃ level is between 15 and 17 weight %, the appropriate MgO/Al₂O₃ ratio is 0.40–0.50. When the Al₂O₃ concentration is between 18 and 20% by weight, the appropriate MgO/Al₂O₃ ratio is 0.45–0.55 [93].

Figure 11.

Influence of MgO/Al₂O₃ on slag critical viscosity and critical temperature [110].

Table 8.

Chemical compositions of slag systems (wt-%) for studying effect of MgO/Al₂O₃ ratio [82,93,110].

No.	Al₂O₃	MgO	M/A	C/S
1	16.63	12–14.5	0.75–0.90	1.02
2	14.84–19.84	8.87–17.06	0.39–0.86	1.14
3	20	3.83–13.33	0.2–0.8	1–1.2

Influence of Al₂O₃/SiO₂ ratio adjustment on high alumina slag

In a study performed using composition no.1 of Table 9, the slag viscosity was estimated to be approximately 0.2 Pa.s (minimum) for an A/S ratio of 0.92 (all other compositions fixed and Al₂O₃ > 24%). Al₂O₃ acts as a network modifier above 24 weight % and hence changing the A/S ratio from 0.47 to 0.92 can result in achieving a minimum viscosity. The viscosity of the slag increases with the A/S ratio. A favourable operation area may be obtained above the liquidus temperature if the Al₂O₃ concentration is greater than 24 weight % and the CaO content is between 35 and 45 weight % [111]. Because the degree of polymerization of [SiO4]- tetrahedra reduces with increasing A/S ratio, viscosity can be reduced for a given high alumina slag with high C/S ratio and increasing A/S ratio up to 0.56. When spinel structures emerge, viscosity quickly decreases if A/S > 0.56 [49].

Table 9.

Chemical compositions of slag systems (wt-%) for studying the effect of Al₂O₃/SiO₂ ratio [49,111].

No.	Al₂O₃	A/S	C/S	MgO	TiO₂
1	16–30	0.47–1.5	1.2	9	1
2	5.5-25	0.11–0.82	0.59–0.97	15	-

Mixed compositional adjustment on high alumina slag

For composition no.1 of Table 10, the apparent activation energy was estimated to be 181 kJ/mol and the critical temperature of slag to be 1612 K. The viscosity increased gradually with an increase in the Al₂O₃ content within the range of 17–29%. The critical temperature also increased from 1612K to 1668 K with increases in the Al₂O₃ content from 17–29% [115]. For composition no.2, at a temperature range of 1748K–1848K, the slag viscosity was lowest for composition: Al₂O₃= 25%, MgO = 16%, and C/S = 1.6. The [AlO4]- unit depolymerizes when the C/S ratio and MgO concentration increase, resulting in reduced viscosity [43]. For the slag with composition: C/S = 2.5 and Al₂O₃ = 25%, the liquidus temperature was reported to be 1792K [113]. Slags with a fixed basicity value had identical viscosity values over the temperature range, regardless of the Al₂O₃/SiO₂ ratio. For composition no.4, the slag melting temperature was estimated to be 1630 K and the viscous flow activation energy to be 157 kJ/mol. In the case of CaO/SiO₂ = 1.18, TiO₂< 4%, and Al₂O₃ > 15%, slag viscosity increases with increased Al₂O₃ concentration, and the melting temperature rises from 1630 K to 1658 K, demonstrating a linear rising trend [112].

Table 10.

Chemical compositions of slag systems (wt-%) for studying mixed compositional adjustment [43,112,113,114,115].

No.	Al₂O₃	MgO	C/S	Cr₂O₃	TiO₂	FeO
1	17–29	8	1	1	-	-
2	25–30	8–16	0.8–1.6	-	-	-
3	30–44	-	2.5–6	-	-	-
4	15–24	8.43–9.33	1.18	-	3.54–3.92	0.66–0.71

Effect of additives

Operating the blast furnace with high alumina slag (Al₂O₃ > 15%) comes with several challenges including the need to cope with high viscosity, melting temperature, etc. In addition to the compositional variation as discussed in section 8, certain additives can be used to influence and optimize the target properties of the slag. The following additives with their respective structure–property correlation are described in detail.

Influence of TiO₂

TiO₂ acts as a network breaker for silicate and alumina slags which is the primary mechanism for reducing the viscosity of the blast furnace hearth slag. It has been observed that for 17% Al₂O₃, a combination of high temperature and higher concentration of TiO₂ can help in achieving the low target viscosity of the slag [109]. At a temperature of 1748K and TiO₂ = 5%, the viscosity of the slag system (Al₂O₃ = 17%, MgO = 10%, CaO/SiO₂ = 0.8) is approximately 0.3 Pa.s which reduces to 0.25 Pa.s on increasing the temperature to 1773K. Similarly, on increasing the TiO₂ level to 10%, the viscosity dropped to a value of 0.27 at 1748K and 0.24 at 1773K. On increasing the TiO2 level from 0 to 6%, the Arrhenius activation energy of the slag, dropped from 209 to 183 kJ/mol respectively [116].

Influence of B₂O₃ on slags

B₂O₃ has a similar effect on reducing the viscosity of high alumina slag as that of TiO₂. For a slag system (Al₂O₃= 11.22–13.11%, MgO = 7.4–8.65%, CaO/SiO₂ = 1.06), the addition of B₂O₃ from 0 to 4% has a more pronounced effect than the addition of TiO₂ for reducing the viscosity of the slag [115]. Minor addition of B₂O₃ in form of chemical grade in slag can help in reducing the slag viscosity to slightly lower levels. On increasing the B₂O₃ level to 4% along with an increase in temperature of slag from 1423K to 1523 K, the viscosity of copper converter slag was reported to be lower than 0.5 Pa.s [41].

Influence of MnO

MnO is an additive that can be added in powder form to reduce the viscosity and activation energy of the slag. It has been observed that at 0% MnO, for a slag system (Al₂O₃= 22.8%, MgO = 8.6%, CaO/SiO₂= 1.01 and Cr₂O₃ = 1%), the slag viscosity was around 1 Pa.s at 1700K. At the same temperature, adding up to 3% MnO reduced the slag viscosity to 0.5 Pa.s. The activation energy also dropped from 217 to 186 kJ/mol on increasing the MnO level from 0% to 3% at 1700K [41].

Influence of CaF₂ (Fluorspar)

Traces of additives in powder or chemical agglomerate form can be used to do small modifications in viscosity and other properties of slag. Additives like CaF₂ can be used in combination with the primary additives to further decrease the viscosity, liquidus temperature and activation energy of the slag. The viscosity of the slag system (Al₂O₃= 25%, MgO = 2%, [CaO + MgO]/[SiO₂+ Al₂O₃] = 0.66) was reported to be 0.6 Pa.s at a temperature of 1750K with no CaF₂ addition. Increasing the CaF₂ level to 3%, decreased the viscosity of the system to 0.5 Pa.s. [90]. CaF₂ is also reported to have reduced the viscosity and optimized the properties of Ruhrstahl Heraeus (RH) refining slag in the secondary steel-making process. However, the use of CaF₂ is reduced because of the more critical environmental hazards as compared to the technical requirements.

Influence of Na₂O and BaO

For the slag system (Al₂O₃ = 9%, CaO = 38.67%, SiO2 = 36.83%, MgO = 8.0%, BaO = 2.5%), viscosity was reported to be in range of 0.6–0.7 Pa.s at a temperature range of 1680–1700K with no Na₂O addition. On increasing the Na₂O concentration to 5% in the same temperature range, the slag viscosity dropped to 0.4 Pa.s [77]. This indicates that the addition of Na₂O in powder form can help in decreasing the slag viscosity and optimize other properties. The presence of BaO first increases the viscosity owing to Ba²⁺ ions forming the network. Further addition of BaO decreases the viscosity as SiO₂ content decreases as a result of excess O^2- ions.

Conclusion

The blast furnace slag is key to the overall productivity of the system and the composition of slag determines important properties like viscosity, activation energy and liquidus temperature. The presence of high alumina makes the slag viscous and hence difficult to tap. Adjusting the composition according to the system of interest can ensure easy operation for which increasing the C/S ratio, %MgO, and adjusting the A/S ratio are some of the techniques. The CaO–MgO–Al₂O₃–SiO₂ quaternary slag systems were analysed along with minor constituents like FeO, MnO, B₂O₃, etc. In addition, the influence of minor constituents on the viscosity-temperature relationship was also studied in this paper with the help of graphs plotted from viscosity measurements done in the literature. A comprehensive solution set is provided according to different slag systems which can be incorporated into the blast furnace operation supported by experimental validation.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the author(s).

ORCID

Sujan Hazra

References

Ahmed

New trends in the application of carbon-bearing materials in blast furnace iron-making. Minerals. 2018;8(12):561.

Mousa

Modern blast furnace ironmaking technology: potentials to meet the demand of high hot metal production and lower energy consumption. Metall Mater Eng. 2019;25(2):69–104.

World Steel Association . (2022). Home. https://www.worldsteel.org/.

Srivastava

Chikhale

Challenges of Indian iron ore industry – geology, mining and processing perspective, p. 812–827. In: 43°.

Sadarang

Nayak

Panigrahi

Challenges and future prospective of alternative materials to silica sand for green sand mould casting: a review. Trans Indian Inst Met. 2021;74(12):2939–2952.

Smith

MP.

Blast furnace ironmaking–a view on future developments. Procedia Eng. 2017;174:19–28.

Mukherjee

Prakash

On dependence of high temperature rheological behaviour of blast furnace slag on its network structure. IFAC Proc Vol. 2013;46(16):312–320.

Rachappa

Prakash

Iron ore recovery from low grade by using advance methods. Proc Earth Planet Sci. 2015;11:195–197.

Duffy

Ingram

Sommerville

ID.

Acid–base properties of molten oxides and metallurgical slags. J Chem Soc Faraday Trans 1: Phys Chem Condens Phases. 1978;74:1410–1419.

10.

Min

Tsukihashi

Recent advances in understanding physical properties of metallurgical slags. Met Mater Int. 2017;23(1):1–19.

11.

Ivanov

Savov

Janke

Alkali capacity and physical properties of blast furnace type slags. Steel Res Int. 2004;75(7):433–441.

12.

Siddique

Cachim

Waste and supplementary cementitious materials in concrete: characterisation, properties and applications. Cambridge (UK): Woodhead Publishing; 2018.

13.

Nishioka

Fujiwara

Ohno

et al. Modeling of dripping behavior in particles packed bed filled with immiscible fluid. ISIJ Int. 2010;50(7):1016–1022.

14.

Sunahara

Nakano

Hoshi

et al. Effect of high Al2O3 slag on the blast furnace operations. ISIJ Int. 2008;48(4):420–429.

15.

Kim

Ozturk

RJ.

Slag-metal separation in the blast furnace trough. ISIJ Int. 1998;38(5):430–439.

16.

Shao

Xiao

Zhang

et al. Dead-man behavior in the blast furnace hearth—a brief review. Processes. 2020;8(11):1335.

17.

Kondratiev

Jak

Hayes

PC.

Predicting slag viscosities in metallurgical systems. Jom. 2002;54(11):41–45.

18.

Lee

Min

Jung

et al. Influence of basicity and FeO content on viscosity of blast furnace type slags containing FeO. ISIJ Int. 2004;44(8):1283–1290.

19.

Han

Chen

Zhang

et al. Viscosity model for iron blast furnace slags in SiO2–Al2O3–CaO–MgO system. Steel Res Int. 2015;86(6):678–685.

20.

Saito

Hori

Nakashima

et al. Viscosity of blast furnace type slags. Metall Mater Trans B. 2003;34(5):509–516.

21.

Sahoo

Tiwari

Mishra

et al. Prediction of flow characteristics of Al2O3–CaO–MgO–SiO2–TiO2-type blast furnace slag and Its evaluation. Arab J Sci Eng. 2019;44(7):6393–6401.

22.

Dienes

GJ.

Activation energy for viscous flow and short-range order. J Appl Phys. 1953;24(6):779–782.

23.

Deng

Zhang

et al. Effect of CaF2 on viscosity, structure and properties of CaO-Al2O3-MgO-SiO2 slag glass ceramics. J Non-Cryst Solids. 2018;500:310–316.

24.

Fang

Pang

Xing

et al. Thermodynamic properties, viscosity, and structure of CaO–SiO2–MgO–Al2O3–TiO2–based slag. Materials (Basel). 2020;14(1):124.

25.

Zhang

Jiao

Zhang

et al. Thermal stability of molten slag in blast furnace hearth. ISIJ Int. 2021;61(8):2227–2236.

26.

Mills

The estimation of slag properties. South Afr Pyrometall. 2011;7(3):35–42.

27.

Liu

Chen

Wang

et al. Sulfide capacity of CaO–SiO2–MgO–Al2O3–BaO–Na2O slag at 1773K. J Sust Metall. 2021;7(3):1169–1177.

28.

Shatokha

Slag parameters and sulphur partition in blast furnace hearth: Ukrainian case and international comparison. Ironmak Steelmak. 2022;49(1):60–69.

29.

Yadav

Das

Baijal

AD.

(2002). Sinter quality at Tata steel: past, present and future. In Proceedings of the Seminar on Raw Material Preparation for Metallurgical Industries: Problems and Solutions, April 8-9 2002 (p. 187). Allied Publishers.

30.

Srivastava

Pan

Prasad

et al. Characterization and processing of iron ore fines of kiriburu deposit of India. Int J Miner Process. 2001;61(2):93–107.

31.

Maulik

Bhattacharyya

KK.

(2004). A beneficiation scheme for reduction of alumina in iron ores from Barsua.

32.

Upadhyay

Venkatesh

AS.

Current strategies and future challenges on exploration, beneficiation and value addition of iron ore resources with special emphasis on iron ores from eastern India. Appl Earth Sci. 2006;115(4):187–195.

33.

Roy

Das

Mohanty

MK.

Feasibility of producing pellet grade concentrate by beneficiation of iron ore slime in India. Sep Sci Technol. 2007;42(14):3271–3287.

34.

Upadhyay

Venkatesh

Roy

Mineralogical characteristics of iron ores in Joda and Khondbond areas in Eastern India with implications on beneficiation. Res Geol. 2010;60(2):203–211.

35.

Roy

Das

Recovery of valuables from low-grade iron ore slime and reduction of waste volume by physical processing. Part Sci Technol. 2013;31(3):256–263.

36.

Kumar

Jain

Rai

Can carboxymethyl cellulose be used as a selective flocculant for beneficiating alumina-rich iron ore slimes? A density functional theory and experimental study. Miner Eng. 2018;121:47–54.

37.

Jiang

et al. Performance and transition mechanism from acidity to basicity of amphoteric oxides (Al2O3 and B2O3) in SiO2–CaO–Al2O3–B2O3 system: a molecular dynamics study. Ceram Int. 2021;47(9):12252–12260.

38.

Trasatti

Work function, electronegativity, and electrochemical behaviour of metals: II. Potentials of zero charge and “electrochemical” work functions. J Electroanal Chem Interfacial Electrochem. 1971;33(2):351–378.

39.

Tiwary

Sarkar

Mishra

et al. Structural aspects of blast furnace slag. Emerg Mater Res. 2013;2(3):152–162.

40.

Yan

High-Temperature physicochemical properties of high alumina slag. In: High temperature physicochemical properties of high alumina blast furnace slag. Singapore: Springer; 2022. p. 77–102.

41.

Zhang

Fan

et al. Effect of MnO on high-alumina slag viscosity and corrosion behavior of refractory in slags. ISIJ Int. 2017;57(11):1887–1894.

42.

Chargui

Hamidouche

Belhouchet

et al. Mullite fabrication from natural kaolin and aluminium slag. Boletín de la sociedad española de cerámica y vidrio. 2018;57(4):169–177.

43.

Talapaneni

Yedla

Pal

et al. Experimental and theoretical studies on the viscosity–structure correlation for high alumina-silicate melts. Metall Mater Trans B. 2017;48(3):1450–1462.

44.

Zhang

Liu

et al. Effect of high alumina-based solid waste on efflorescence behavior of alkali-activated steel slag. Constr Build Mater. 2022;349:128804.

45.

Wang

Jiang

Guo

et al. Effects of basicity and Al2O3 content on viscosity and crystallization behavior of super-high-alumina slag. Crystals (Basel). 2022;12(6):851.

46.

Park

Kim

Min

DJ.

Novel approach to link between viscosity and structure of silicate melts via Darken’s excess stability function: focus on the amphoteric behavior of alumina. Metall Mater Trans B. 2008;39(1):150–153.

47.

Sun

Liu

et al. Influence of Al2O3 and MgO on the viscosity and stability of CaO–MgO–SiO2–Al2O3 slags with CaO/SiO2 = 1.0. ISIJ Int. 2017;57(6):978–982.

48.

Wang

Sun

Sridhar

et al. Effect of Al2O3 on the viscosity and structure of CaO-SiO2-MgO-Al2O3-FetO slags. Metall Mater Trans B. 2015;46(2):537–541.

49.

Liao

Zhang

Sridhar

et al. Effect of Al2O3/SiO2 ratio on the viscosity and structure of slags. ISIJ Int. 2012;52(5):753–758.

50.

Wang

Zhang

Liu

et al. Thermal conduction mechanism based on microstructural transformations of molten slag: the role of calcium oxide. Int J Heat Mass Transfer. 2020;160:120167.

51.

Mysels

KJ.

The maximum bubble pressure method of measuring surface tension, revisited. Colloids Surf. 1990;43(2):241–262.

52.

Shankar

Görnerup

Seetharaman

et al. Sulfide capacity of high alumina blast furnace slags. Metall Mater Trans B. 2006;37(6):941–947.

53.

Wang

et al. Raman structure investigations of CaO-MgO-Al2O3-SiO2-CrO x and its correlation with sulfide capacity. Metall Mater Trans B. 2016;47(1):10–15.

54.

Guo

Shen

Zheng

et al. Desulfurization ability of blast furnace slag containing high Al2O3 at 1773K. Crystals (Basel). 2021;11(8):910.

55.

Hino

Nagasaka

Katsumata

et al. Simulation of primary-slag melting behavior in the cohesive zone of a blast furnace, considering the effect of Al2O3. Fe t O, and basicity in the sinter ore. Metall Mater Trans B. 1999;30(4):671–683.

56.

Yao

Ren

Wang

et al. Effect of Al2O3, MgO, and CaO/SiO2 on viscosity of high alumina blast furnace slag. Steel Res Int. 2016;87(2):241–249.

57.

JR.

Effects of alumina on sintering performance of hematite iron ores. ISIJ Int. 2007;47(3):349–358.

58.

Umadevi

Deodar

Mahapatra

et al. Influence of alumina on iron ore sinter properties and productivity in the conventional and selective granulation sintering process. Steel Res Int. 2009;80(9):686–692.

59.

Umadevi

Sah

Mahapatra

PC.

Influence of sinter basicity (CaO/SiO2) on low and high alumina iron ore sinter quality. Min Proc Extractive Metall. 2014;123(2):75–85.

60.

Muwanguzi

Karasev

Byaruhanga

et al. Characterization of chemical composition and microstructure of natural iron ore from Muko deposits. Int Sch Res Notices. 2012;2012:174803. 9 pages.

61.

Agrawal

Das

Singh

et al. Means to cope with the higher alumina burden in the blast furnace. Ironmak Steelmak. 2020;47(3):238–245.

62.

Dynys

Halloran

JW.

Alpha alumina formation in alum-derived gamma alumina. J Am Ceram Soc. 1982;65(9):442–448.

63.

Feret

Roy

Boulanger

Determination of alpha and beta alumina in ceramic alumina by X-ray diffraction. Spectrochim Acta Part B. 2000;55(7):1051–1061.

64.

Rotole

Sherwood

PM.

Gamma-alumina (γ-Al2O3) by XPS. Surf Sci Spectra. 1998;5(1):18–24.

65.

Kiyohara

Santos

Coelho

ACV

et al. Structure, surface area and morphology of aluminas from thermal decomposition of Al (OH)(CH3COO) 2 crystals. Anais da Academia Brasileira de Ciências. 2000;72:471–495.

66.

Hashimoto

Onodera

Tahara

et al. Structure of alumina glass. Sci Rep. 2022;12(1):1–9.

67.

Shi

Alderman

Berman

et al. The structure of amorphous and deeply supercooled liquid alumina. Front Mater. 2019;6:38.

68.

Dippenaar

Industrial uses of slag (the use and re-use of iron and steelmaking slags). Ironmak Steelmak. 2005;32(1):35–46.

69.

Stuart

Sohlberg

The microstructure of γ-alumina. Energies. 2021;14(20):6472.

70.

Kovarik

Bowden

Genc

et al. Structure of δ-alumina: toward the atomic level understanding of transition alumina phases. J Phys Chem C. 2014;118(31):18051–18058.

71.

Chandran

Kirschhock

Radhakrishnan

et al. Alumina: discriminative analysis using 3D correlation of solid-state NMR parameters. Chem Soc Rev. 2019;48(1):134–156.

72.

Kaur

Khanna

Kaur

et al. Synthesis and structural characterization of alumina nanoparticles. Phase Trans. 2020;93(6):596–605.

73.

Tavakoli

Maram

Widgeon

et al. Amorphous alumina nanoparticles: structure, surface energy, and thermodynamic phase stability. J Phys Chem C. 2013;117(33):17123–17130.

74.

O’Dell

Savin

Chadwick

et al. Structural studies of silica-and alumina-pinned nanocrystalline SnO2. Nanotechnology. 2005;16(9):1836.

75.

Song

Kim

Park

et al. Preparation of silica-alumina nanoparticles via blast-furnace slag dissolution in low-concentration acetic acid for carbonation. Minerals. 2017;7(11):206.

76.

Wang

Jiang

Guo

et al. Effects of basicity and Al2O3 content on viscosity and crystallization behavior of super-high-alumina slag. Crystals (Basel). 2022;12(6):851.

77.

Zhang

Han

et al. Effect of BaO and Na2O on the viscosity and structure of blast furnace slag. Ironmak Steelmak. 2018;47:168–172.

78.

Siafakas

Matsushita

Hakamada

et al. Measurement of viscosity of SiO2-CaO-Al2O3 slag in wide temperature range by aerodynamic levitation and rotating bob methods and sources of systematic error. Int J Microgravity Sci Appl. 2018;35(2):350204.

79.

Chebykin

Heller

Dubberstein

et al. Viscosity measurement of slags using rotating bob and vibrating finger viscometer. ISIJ Int. 2017;57(8):1319–1326.

80.

Saito

Hori

Nakashima

et al. Viscosity of blast furnace type slags. Metall Mater Trans B. 2003;34(5):509–516.

81.

Tang

Zhang

Guo

et al. Viscosities behavior of CaO-SiO2-MgO-Al2O3 slag with low mass ratio of CaO to SiO2 and wide range of A12O3 content. J Iron Steel Res Int. 2011;18(2):1–6.

82.

Yao

Ren

Liu

et al. Study of the effect of the MgO/Al2O3 ratio on viscosity of blast furnace slag. Metall Res Technol. 2015;112(6):602.

83.

Bian

Gao

Influence of Al2O3, CaO/SiO2, and B2O3 on viscous behavior of high alumina and medium titania blast furnace slag. J Chem. 2017;2017:6895928. 6 pages.

84.

Fukutake

Okabe

KO.

Experimental studies of slag flow in the blast furnace hearth during tapping operation. Trans Iron Steel Inst Jpn. 1976;16:309.

85.

Sunahara

Nakano

Hoshi

et al. Effect of high Al2O3 slag on the blast furnace operations. ISIJ Int 2008;48(4):420–429.

86.

Lee

Min

Jung

et al. Influence of basicity and FeO content on viscosity of blast furnace type slags containing FeO. ISIJ Int 2004;44(8):1283–1290.

87.

Han

(2017). Viscosity studies of high-temperature metallurgical slags relevant to ironmaking process.

88.

Feng

Chu

Tang

et al. Effect of CaO/SiO2 and Al2O3 on viscous behaviors of the titanium-bearing blast furnace slag. Steel Res Int. 2016;87(10):1274–1283.

89.

Dienes

GJ.

Activation energy for viscous flow and short-range order. J Appl Phys. 1953;24(6):779–782.

90.

Deng

Zhang

et al. Effect of CaF2 on viscosity, structure and properties of CaO-Al2O3-MgO-SiO2 slag glass ceramics. J Non-Cryst Solids. 2018;500:310–316.

91.

Fang

Pang

Xing

et al. Thermodynamic properties, viscosity, and structure of CaO–SiO2–MgO–Al2O3–TiO2–based slag. Materials (Basel). 2020;14(1):124.

92.

Tang

Zhang

Guo

et al. Viscosities behavior of CaO-SiO2-MgO-Al2O3 slag with low mass ratio of CaO to SiO2 and wide range of A12O3 content. J Iron Steel Res Int. 2011;18(2):1–6.

93.

Jiang

Zhang

Zheng

et al. Three-segment control theory of MgO/Al2O3 ratio based on viscosity experiments and phase diagram analyses at 1500° C. J Iron Steel Res Int. 2020;27(6):624–630.

94.

Zhang

Wang

Jiao

et al. Effect of BaO and MnO on high-temperature properties and structure of blast furnace slag. J Non-Cryst Solids. 2021;571:121066.

95.

Park

Kim

et al. Effect of TiO2 on the viscosity and slag structure in blast furnace type slags. steel Res Int. 2012;83(2):150–156.

96.

Chen

Zhang

Kou

et al. Viscosities of iron blast furnace slags. ISIJ Int. 2014;54(9):2025–2030.

97.

Sun

Song

et al. Influences of Al2O3 and TiO2Content on viscosity and structure of CaO–8% MgO–Al2O3–SiO2–TiO2–5% FeO blast furnace primary slag. Metals (Basel). 2019;9(7):743.

98.

Gan

Lai

A general viscosity model for molten blast furnace slag. Metall Mater Trans B. 2014;45(3):875–888.

99.

Kim

Sohn

et al. The effect of MgO on the viscosity of the CaO-SiO2-20 wt% Al2O3-MgO slag system. steel Res Int. 2010;81(4):261–264.

100.

Xing

Pang

Zheng

et al. Effect of MgO and K2O on high-Al silicon–manganese alloy slag viscosity and structure. Minerals. 2020;10(9):810.

101.

Yan

Zhang

et al. Influence of MgO, Al2O3 and CaO/SiO2 on the viscosity of blast furnace type slag with high Al2O3 and 5 wt-% TiO2. Can Metall Q. 2016;55(2):186–194.

102.

Das

Agrawal

Reddy

et al. Factsage studies to identify the optimum slag regime for blast furnace operation. Trans Indian Inst Met. 2021;74(2):419–428.

103.

Chang

Jiao

Zhang

et al. Experimental investigation of relationship between enthalpy change and viscosity in blast furnace type slags. ISIJ Int. 2019;59(1):39–45.

104.

Shankar

Görnerup

Lahiri

et al. Estimation of viscosity for blast furnace type slags. Ironmak Steelmak. 2007;34(6):477–481.

105.

Babenko

Istomin

Protopopov

et al. Viscosity of CaO-SiO2-Al2O3-MgO-B2O3 slags. Steel Transl. 2014;44(2):110–112.

106.

Ghosh

Krishnamurthy

Sankaranarayanan

SR.

Application of optical basicity to viscosity of high alumina blast furnace slags. J Min Metall Sect B. 2010;46(1):41–49.

107.

Yan

Pang

et al. Effect of TiO₂ on the viscous behavior of high alumina blast furnace slag. In: TMS annual meeting & exhibition Cham: Springer; March. p. 725–733.

108.

Jiao

Zhang

Wang

et al. Melting features and viscosity of TiO2-containing primary slag in a blast furnace. High Temp Mater Processes. 2018;37(2):149–156.

109.

Zhang

et al. Influence of TiO2 on the viscosity of a high alumina slag and on carbon brick corrosion. Steel Res Int. 2018;89(3):1700353.

110.

Zhang

Jiang

Xue

et al. Influence of MgO/Al2O3 ratio on viscosity of blast furnace slag with high Al2O3 content. Steel Res Int. 2016;87(1):87–94.

111.

Yan

Liang

et al. Transition of blast furnace slag from silicates-based to aluminates-based: viscosity. Metall Mater Trans B. 2017;48(2):1092–1099.

112.

Liu

Qin

Yang

et al. Influence of Al2O3 content on the melting and fluidity of blast furnace type slag with low TiO2 content. J Chem. 2018;2018:1–6.

113.

Siafakas

Matsushita

Jarfors

AEW

et al. Viscosity of SiO2–CaO–Al2O3 slag with low silica–influence of CaO/Al2O3, SiO2/Al2O3 ratio. ISIJ Int. 2018;58(12):2180–2185.

114.

Wang

et al. Effect of Al2O3 on the viscosity of CaO-SiO2-Al2O3-MgO-Cr2O3 slags. Int J Miner Metall Mater. 2021;28(5):797–803.

115.

Zhang

Wang

et al. Influence of Cr2O3 and B2O3 on viscosity and structure of high alumina slag. steel Res Int. 2017;88(4):1600241.

116.

Gao

Bian

Liang

Influence of B2O3 and TiO2 on viscosity of titanium-bearing blast furnace slag. steel Res Int. 2015;86(4):386–390.

Optimization of high alumina slag practice in blast furnace ironmaking: an industrial approach (PART 1: fundamental aspects)

Abstract

Introduction

The alumina anomaly

Effect of high alumina on thermophysical and thermochemical properties of slag

Effect on density and surface tension

Effect on sulphide capacity

Industrial impact of high alumina burden in the blast furnace

Structure-based studies on alumina bonding

Studies on measurement of slag properties

Viscosity measurement

Liquidus temperature measurement

Activation energy measurement (Arrhenius)

Slag property prediction models

Viscosity-prediction models

Prediction of liquidus temperature characteristics

Simulations using Factsage

Observations on compositional changes of high alumina slag

Influence of MgO adjustment on high alumina slag

Influence of CaO/SiO2 ratio on high alumina slag

Influence of TiO2 on high alumina slag

Influence of MgO/Al2O3 (M/A) ratio on adjustment on high alumina slag

Influence of Al2O3/SiO2 ratio adjustment on high alumina slag

Mixed compositional adjustment on high alumina slag

Effect of additives

Influence of TiO2

Influence of B2O3 on slags

Influence of MnO

Influence of CaF2 (Fluorspar)

Influence of Na2O and BaO

Conclusion

Footnotes

Disclosure statement

ORCID

References

Influence of CaO/SiO₂ ratio on high alumina slag

Influence of TiO₂ on high alumina slag

Influence of MgO/Al₂O₃ (M/A) ratio on adjustment on high alumina slag

Influence of Al₂O₃/SiO₂ ratio adjustment on high alumina slag

Influence of TiO₂

Influence of B₂O₃ on slags

Influence of CaF₂ (Fluorspar)

Influence of Na₂O and BaO