Keywords

Introduction

The cohesive zone inside blast furnace, where the ferrous burden softens and melts, significantly affects the productivity of the blast furnace [1]. A Reduction Degree (RD) of a ferrous burden is typically above 50% in the cohesive zone area of a BF [2, 3]. At this point the iron bearing burden consists of metallic iron (Fe), wustite (FeO) and other oxides (CaO, SiO2, MgO, Al2O3 etc.) which form a slag phase. The phase compositions are dependent on the chemistry of the iron bearing burden [4,5,6,7,8]. Many researchers have found that the softening and melting properties of ferrous materials are controlled by reduction degree, basicity, slag viscosity, gangue content and their distribution in the microstructure [9,10,11,12].

However, though various studies are conducted to research the properties of blast furnace slags, no satisfactory knowledge has been acquired about gangue on of the formation of FeO-bearing primary-slags process in blast furnace. Because the mixture between sinter and lump ores in blast furnace was not simply physical mixing, it is of great significance to study the influences of CaO-/SiO2-/MgO-/Al2O3 on the softening and melting properties of FeO respectively [13,14,15]. In the present work, the softening and dripping experiments under the blast furnace conditions are designed to explore the influences of respective CaO-/SiO2-/MgO-/Al2O3 on the softening and melting properties of FeO. The chemical compositions, liquidus temperature of primary-slags were also studied.

Experimental

Experimental Method

The samples were prepared from CaCO3 (≥99.9%), SiO2(≥99.9%), MgO(≥99.9%), Al2O3(≥99.9%) and FeC2O4·2H2O (≥99.9%) and the powdery CaCO3/SiO2/MgO/Al2O3 were uniformly mixed with FeC2O4·2H2O respectively with a 70:10:10:5:5 (FeO:CaO:SiO2:MgO:Al2O3) mass ratio and then pressed to cylindrical shape samples. The samples were finally roasted to 1173 K (900 °C) with a triple-fired furnace in the inert gas atmosphere (N2, 99.9%). This temperature was held for 1 h, in order to make sure the complete decomposition of FeC2O4·2H2O. Finally, the samples were cooled to the room temperature in the inert gas atmosphere. Both the above mentioned stages were accomplished in the softening-melting equipment with the schematic diagram shown in Fig. 1.

Fig. 1
figure 1

Schematic diagram of softening-melting equipment

Samples having a layer thickness of 50 mm were charged in a graphite crucible and cokes with a layer thickness of 15 mm were placed over and below the samples. The iron ore samples are 10–15 mm in diameter and the cokes are 6–10 mm in diameter. Subject to the condition of constant layer thickness, the charge of the iron ore samples weighted different due to the difference of density. The inner diameter of the graphite crucible is about 50 mm.

Experimental conditions for the softening and dripping behavior are shown in Fig. 2. The heating up rate is 10 °C/min below 900 °C and 5 °C/min over 900 °C to make sure the iron ore samples are adequately reduced after 900 °C. Gas flow is 5 L/min of N2 below 900 °C and 10 L/min of reducing gas over 900 °C, the composition of reducing gas is CO:N2 in mole fraction = 40:60(%). The load of 1 kg/cm2 is added to the samples in the whole process. The experiment was stopped when pressure drop reached the maximum. The samples are cooled down to room temperature by N2 with the flow rate 5 L/min. In this work, a series of five experiments to determine the impact of CaO-/SiO2-/MgO-/Al2O3 to the softening and melting properties of samples, the chemical compositions of the samples for softening and melting experiment as listed in Table 1.

Fig. 2
figure 2

Experimental conditions for softening and dripping experiment

Table 1 Chemical compositions of samples

Results and Discussion

Softening and Melting Properties of Base Case

A typical set of softening and dripping test results is shown in Fig. 3, in which pressure drop, temperature and shrinkage rate are conducted to evaluate the primary slags formation behaviors of iron ore samples. It is difficult to observe the behavior directly as the internal of the softening-melting equipment is not visible. Then some indexes are conducted to evaluate the primary-slags formation behavior of iron ores. T10% is the temperature when the shrinkage rate of the samples reaches 10%, indicating the samples start to soften, softening end temperature (T40%) is the temperature when the shrinkage of the samples reaches 40%. The temperature interval (T40%–T10%) represents the softening zone of samples. Tm is the temperature when the pressure drop of the samples reaches 0.98 kPa, meaning the samples begin to melt. Tp is the temperature when the pressure drop of the samples reaches the highest (ΔPmax), meaning the primary–slags are totally produced. Td is the temperature at which dripping starts and the temperature interval (Td–T10%) reflect the thickness of the cohesive zone.

Fig. 3
figure 3

A typical set of softening and dripping experiment

To study the whole process of the high temperature behaviors of iron ores and gain the primary-slags at Tp, every test contains two steps. Firstly, the whole process of samples from room temperature to dripping finished is conducted. Secondly, stop heating the samples at Tp and cool down the samples to room temperature under the protection of pure N2.

The results of the softening and melting properties of base case were shown in Table 2. The results show that when sample is pure FeO, the softening zone is 111 °C, thickness of the cohesive zone is 652 °C, and the dripping temperature is 1436 °C.

Table 2 Softening and melting properties of base case

Effects of the Samples Composition on Softening and Melting Properties

In order to study the impact of CaO/SiO2/MgO/Al2O3 on softening and melting properties of iron ores, the samples composition was changed as listed in Table 1.

Table 3 shows the softening and melting properties of samples. For the respective samples, the starting and finishing softening temperature of sample B and C higher than base case and the the softening temperature interval are also wider than base case, of which the sample B temperature is the highest. Figure 4 shows the starting softening temperature of sample D is lower than pure FeO, but the finishing softening temperature is higher than base case, and softening temperature interval is extremely wide. For the sample A, both the starting and finishing softening temperature are lower than base case. While the dripping start temperature of sample B is higher (1525 °C) than base case, indicated that addition of SiO2 increase the softening temperature and dripping start temperature as well as the dripping temperature. Although the maximum pressure drop of the case with the addition of SiO2 decreases to 1.5 kPa, the Tp is the highest among all the cases. The addition of CaO, MgO and Al2O3 have no dripping temperature.

Table 3 Softening and melting properties of samples
Fig. 4
figure 4

Softening temperature intervals of samples

To further study the influence of CaO–SiO2–MgO–Al2O3 on the formation behavior of FeO-bearing primary-slags , the primary-slags samples of A, B, C and D were analyzed through X-Ray diffraction. Figure 5 shows the diffraction patterns of primary-slags of samples. The main primary-slags phases of CaO additions are CaO·Fe3O4 (liquidus temperature 1104 °C) and FeO, and that with SiO2, MgO, Al2O3 additions are fayalite (2FeO·SiO2, 1889 °C), magnesioferrite (MgO·FeO, 1791 °C), hercynite (FeO·2Al2O3, 1780 °C), respectively. The results of the liquidus temperature of the slags were obtained from FactSage6.2. The changes in the softening temperature of the slags are owing to the new compounds generated when compositions changed. Therefore, the decreasing softening temperature of the primary slag is mainly caused by its high CaO. While the slag has a higher SiO2 and higher MgO content, the softening temperature is increased, because it will produce high liquidus temperature substances.

Fig. 5
figure 5

X-ray diffraction analysis of samples

On the other hand, there was no doubt that FeO-bearing primary-slags was circumscribed, because the environment of blast furnace is complexed. The above analysis results inspired to design more experiments to study the impact of slag phase on the softening and melting properties.

Conclusions

The influences of CaO/SiO2/MgO/Al2O3 on the formation behavior of FeO-bearing primary-slags is experimentally demonstrated, and the following conclusions are obtained.

  1. 1.

    Additions of CaO or Al2O3 decrease the starting softening temperature and no dripping behaviors are found in comparison with the base case results when only FeO is used.

  2. 2.

    The addition of SiO2 or MgO rises the starting and ending softening temperature, as well as the dripping temperature. In addition, the addition of SiO2 decreases the maximum pressure drop to as low as 1.5 kPa while increases Tp to as high as 1403 °C.

  3. 3.

    According to the phase diagram and XRD analysis results, the initial phase with CaO addition in the primary-slags should be CaO·Fe3O4 and that with SiO2, MgO, Al2O3 additions are fayalite (2FeO·SiO2), magnesioferrite (MgO·FeO), hercynite (FeO·2Al2O3), respectively.