Sensitive Influence of Floating State of Blast Furnace Deadman on Molten Iron Flow and Hearth Erosion

Abstract

Based on the survey of two Chinese large blast furnaces (BFs), it was found there were complex relationships between hearth sidewall and bottom temperature. The highest temperature positions for these two BFs were distinctly different, which was indicated that the positions of the most serious erosion were different. Therefore, the mathematical model of molten iron flow in BF hearth bottom was established and the influence of floating state of deadman on molten iron flow and wall shear stress was analyzed. The results showed that the status of molten iron flow was determined by the floating state of deadman so that the erosion position was influenced. The floating height of deadman of BF A was higher than that of BF B obviously, which led to the difference between hearth sidewall and bottom temperature and also the erosion migration. When deadman was sinking at bottom, the molten iron flow was faster and wall shear stress was larger near the bottom corners than that at other bottom positions so that the elephant-foot-type erosion occurred easily. When deadman was floating, the erosion position migrated from the hearth bottom corners to the corners between bottom of deadman and hearth sidewall. The permeability of deadman would also affect the erosion. Finally, the erosion profiles and their forming reasons were discussed at different floating heights of deadman.

1. Introduction

At present, the campaign life of an ironmaking blast furnace (BF) is mainly determined by the life of hearth bottom.^1,2,3,4) In recent years, the phenomenon of intensified hearth bottom erosion even burning out occurred frequently.⁵⁾ It was found that the erosion positions and profiles were different for different BFs.^5,6,7) What’s more, the relationships between hearth sidewall temperature and bottom center temperature were complex. For example, both of hearth sidewall temperature and bottom temperature were increased or decreased simultaneously for some BFs, whereas they were opposite for some other BFs, which are still no reasonable explanations. It has great significance to judge the erosion state of hearth bottom and the activity of deadman correctly by solving the above problems.

The floating state of deadman is decided by the force balance of deadman. The floating heights of deadman for different BFs are different.^8,9) Many workers have researched the phenomena of molten iron flow and erosion profile in hearth bottom, such as dissection investigation,^6,7) core sample analysis,^7,10) numerical simulation^11,12,13,14) and so on. Much efforts have also been made in study of influence factors and controlling measurements of erosion by calculation and production data.^{15,16,17,18,19,20)} However, the relationship between floating state of deadman and molten iron flow in hearth bottom as well as the erosion position is still confusing. The influence of floating height of deadman on molten iron flow is the key to solve this problem. To judge the molten iron flow and erosion state correctly is beneficial for guiding the reasonable operation and steady production of blast furnace. In this paper, taking two large Chinese BFs for example, the change rule and relationship between hearth sidewall temperature and bottom temperature were analyzed. Meanwhile, the molten iron flow and wall shear stress for different floating heights of deadman were calculated and the possible erosion positions and profiles were discussed, which were validated and illustrated by two small dissected BFs.

2. Survey of BFs

2.1. Parameter of BFs

Two large Chinese BFs were surveyed, which are called BF A and BF B, respectively. The effective volume of BF A and B are 4966 m³ and 3200 m³, respectively. The relevant parameters are listed in Table 1. Figure 1 presents the lining structure and thermocouple positions of hearth bottom for BF A and B. The bottom lining of BF A from bottom to top are graphite carbon brick, ordinary carbon brick, micropore carbon brick and ceramic pad. The hearth lining are hot pressing carbon brick NMD and NMA. Table 2 shows the physical and chemical properties of NMD and NMA.⁴⁾ The marked thermocouples in Fig. 1(a) are including B-1 (bottom center), H-1 and H-2 (below taphole with 1–1.5 m), H-3 and H-4 (near taphole). The bottom lining of BF B are similar to that of BF A, which are graphite carbon brick, super-micropore carbon brick and corundum-mullite brick from bottom to top. The hearth lining are NMD and NMA. The marked thermocouples in Fig. 1(b) are B-1~B-4 (bottom center) and thermocouples at the sixth layer of hearth.

Table 1. Parameters of BF A and B.

Parameters	Unit	BF A	BF B
Volume	m3	4966	3200
Hearth diameter	m	14.5	13.3
Throat diameter	m	10.8	8.0
Tuyere number		40	36
Depth of salamander	m	3.672	3.000
Blow-in date		2009.2	2010.3

Fig. 1.

Structure and erosion status of hearth bottom for BF A and B.

Table 2. Physical and chemical properties of NMD and NMA.⁴⁾

Item	Unit	NMD	NMA
Bulk density	kg/m3	1820	1620
Apparent porosity	%	16	18
Compressive strength (Room temperature)	MPa	30	33
Thermal conductivity (20°C)	W/(m·°C)	60	17

2.2. Hearth Bottom Temperature

Figure 2(a) presents the temperature change of thermocouples for BF A. The horizontal coordinate stands for time from January 2012 to July 2013. The positions of thermocouples are shown in Fig. 1(a). In the Fig. 2(a), it can be observed that the highest temperature of bottom thermocouple (B-1) and hearth thermocouples were about 300°C and 600°C, respectively. What’s more, the change trend of hearth temperature was contrary to bottom temperature. The hearth temperature increased with decreasing of bottom temperature. On the contrary, the hearth temperature decreased when bottom temperature increased.

Fig. 2.

Temperature of hearth bottom thermocouples for BF A and B.

Figure 2(b) shows the temperature curves of thermocouples for BF B. The time (horizontal coordinate) was from January 2012 to January 2014. The upper figure is the bottom thermocouples (B-1~B-4) and the lower figure is thermocouples at the sixth layer of hearth. The positions of thermocouples are shown in Fig. 1(b). It can be found from the Fig. 2(b) that the highest temperature of thermocouples at the sixth layer was low, which was less than 100°C. However, the highest temperature of bottom thermocouple B-4 was about 540°C. The hearth temperature and bottom temperature almost had the same tendency. The hearth temperature increased when bottom temperature increased. The hearth temperature also decreased with decreasing bottom temperature.

The erosion profiles of BF A and B were calculated by temperature data of thermocouples according to the method of inverse heat transfer problem,²¹⁾ which are shown in Fig. 1. It can be seen from Fig. 1(a) that the erosion at bottom of BF A was lighter than the erosion at hearth sidewall. The most serious erosion for BF A was not at the corners of hearth bottom, but near 1–1.5 m below the taphole central line. The temperature of thermocouples at this region was higher and the erosion would be more serious than that at other regions. In Fig. 1(b), the temperature of thermocouples at bottom was higher than that at hearth sidewall. Therefore, the erosion degree of bottom is greater than that of hearth sidewall. But there was still no serious erosion.

3. Model of Molten Iron Flow in BF Hearth Bottom

In order to investigate the reason of different temperature relationships and erosion profiles of BF A and B, based on the hearth bottom structure of BF B, a three-dimensional model of molten iron flow in hearth bottom was established and the influences of floating state of deadman on molten iron flow were analyzed.

3.1. Physical Model

Figure 3(a) shows the physical model of molten iron flow in hearth bottom. The hearth diameter and taphole diameter are 13.3 m and 0.08 m, respectively. The height of salamander is 3 m. For convenience of calculations, the deadman is regarded as cylinder. Considering with the influence of side wall effect²²⁾ and raceway, the porosity near hearth wall is generally bigger than that near hearth center. The shape and size of deadman is very different for different BFs. Many authors have studied the large coke free zone from the deadman to the wall.^20,23,24) In this paper, the authors aim to study the molten iron flow and erosion at different floating heights of deadman. Therefore, it is assumed that there is a coke free zone near the hearth wall and the thickness of the zone is 0.1 m, which is shown in Fig. 3(b). The different porosity near hearth sidewall will be also investigated in the following sections. The floating height of deadman is h_c m. Three kinds of float level (h_c=0, 0.4 and 1.5 m) are analyzed, which indicated that sitting deadman, slight floating deadman and heavy floating deadman, respectively. Three planes are selected as the focus of subsequent analysis, which are named as plane a (Y=0), plane b (X=6.55 m) and plane c (Z=0.2 m).

Fig. 3.

Physical model of molten iron flow in hearth bottom.

3.2. Mathematical Model

3.2.1. Governing Equations

The system to be analyzed consists of deadman (coke packed bed) and coke free zone, as shown in Fig. 3. Because of tapping operation continuously for large BFs, the mass flow rate of molten iron at taphole is approximately equal to the mass flow rate of molten iron into hearth. Therefore, a three-dimensional steady model of molten iron flow in hearth bottom was built in Cartesian coordinate system (X, Y, Z) as follows.

(1) Continuity equation:   

$$\nabla \cdot \left(\rho \overrightarrow{v}\right)=0$$(1)

where ∇ is hamilton operator. ∇= $\frac{\partial }{\partial x}\overrightarrow{i}$ + $\frac{\partial }{\partial y}\overrightarrow{j}$ + $\frac{\partial }{\partial z}\overrightarrow{k}$ . ρ and $\overrightarrow{v}$ are density and velocity vector, respectively.

(2) Reynolds averaged Navier-Stokes equation:   

$$\nabla \cdot \left(\rho \overrightarrow{v}\overrightarrow{v}\right)=-\nabla P+\nabla \cdot \left[{\mu }_{eff}\left(\nabla \overrightarrow{v}+\nabla {\overrightarrow{v}}^T\right)\right]+\rho \overrightarrow{g}+S_m$$(2)

  

$${\mu }_{eff}=\mu +{\mu }_t$$(3)

  

$${\mu }_t=\rho C_{\mu }\frac{k^2}{\epsilon }$$(4)

where P, $\overrightarrow{g}$ , μ_eff, μ_t, and μ are pressure, the acceleration of gravity, effective viscosity, turbulent viscosity and kinetic viscosity, respectively.

(3) Turbulent kinetic energy equation:   

$$\nabla \cdot \left(\rho \overrightarrow{v}k\right)=\nabla \cdot \left[\left(\mu +\frac{{\mu }_t}{{\sigma }_k}\right)\nabla k\right]+G_k-\rho \epsilon +S_k$$(5)

(4) Dissipation rate of turbulence kinetic energy equation:   

$$\nabla \cdot \left(\rho \overrightarrow{v}\epsilon \right)=\nabla \cdot \left[\left(\mu +\frac{{\mu }_t}{{\sigma }_{\epsilon }}\right)\nabla \epsilon \right]+\frac{\epsilon }{k}\left(C_1{G}_k-C_2\rho \epsilon \right)+S_{\epsilon }$$(6)

  

$$G_k={\mu }_t\nabla \overrightarrow{v}\cdot \left(\nabla \overrightarrow{v}+{\left(\nabla \overrightarrow{v}\right)}^T\right)$$(7)

where k and ε are turbulent kinetic energy and dissipation rate of turbulent kinetic energy, respectively. S_k and S_ε are the sink terms of momentum, respectively. C₁=1.44, C₂=1.92, σ_k=1.0, σ_ε=1.3, C_μ=0.09.

(5) Porous media

The deadman is a fixed bed, which is composed of coke particles. The deadman is modelled as a porous media and the Ergun equation is used to describe the flow resistance, which is including viscous loss term and inertial loss term, as shown in Eqs. (8), (9), (10).   

$$a=\frac{D_p^2}{150}\cdot \frac{{\epsilon }_d^3}{{\left(1-{\epsilon }_d\right)}^2}$$(8)

  

$$C_2=\frac{3.5}{D_p}\cdot \frac{1-{\epsilon }_d}{{\epsilon }_d^3}$$(9)

  

$$S_m=-\left(\frac{\mu }{a}\overrightarrow{v}+\frac{1}{2}{C}_2\rho \left|v\right|\overrightarrow{v}\right)$$(10)

where a and C₂ are permeability coefficient and inertial resistance coefficient, respectively. D_p and ε_d are average diameter of coke particle in deadman and porosity of deadman, respectively. The particle diameter of coke in the hearth is 0.03 m in the calculation.

3.2.2. Boundary Conditions

Boundary conditions are given as follows.

(1) Inlet boundary: inlet velocity of molten iron at top surface of the model is regarded as the inlet boundary condition. The inlet velocity is calculated by outlet flow mass of molten iron.

(2) Outlet boundary: taphole exit is kept at atmospheric pressure, which equals to 1.01×10⁵ Pa.

(3) At hearth sidewall and bottom wall surface: no-slip boundary conditions are employed.

In calculation, the density and viscosity of molten iron are 7000 kg/m³ and 0.00715 Pa·s, respectively.^3,19) The inlet velocity of molten iron flow is 7×10⁻⁵ m/s.

4. Results

4.1. Distribution of Molten Iron Flow in Hearth Bottom

Figure 4 presents the distribution of molten iron flow of plane a (hearth bottom profile) at different floating heights of deadman. The arrows of same length are used to show flow direction. The flow diagrams of molten iron are analyzed at floating height with 0 m, 0.4 m and 1.5 m, respectively. From the Fig. 4(a), it is found that the molten iron near hearth sidewall flows to the hearth coke free zone, whereas the molten iron near central line of hearth flows to the bottom. In Figs. 4(b) and 4(c), the flow paths in deadman are similar to that in Fig. 4(a). What’s more, the molten iron flows along the horizontal direction in bottom coke free zone.

Fig. 4.

Distribution of molten iron flow of plane a at different floating heights of deadman.

Figure 5 shows the flow field distribution of molten iron of plane b (X=6.55 m) at different floating heights of deadman. At the same tapping speed, the velocity of molten iron at the corners of hearth bottom with sitting deadman is larger than that with floating deadman. When deadman is floating at 0.4 m, the velocity of molten iron in bottom coke free zone is larger obviously than that when deadman is floating at 1.5 m. The velocity at the bottom of deadman changes a lot while deadman is floating. The velocity of molten iron near hearth sidewall increases sharply from bottom coke zone into hearth coke free zone. Therefore, the erosion will be easy to appear at hearth sidewall corresponding to the bottom of deadman.

Fig. 5.

Flow field distribution of molten iron of plane b at different floating heights of deadman.

Figure 6 shows that flow field distribution of molten iron of plane c (Z=0.2 m) at different floating heights of deadman. Because c plane is symmetrical along the center line of the taphole, only half plane c is shown. When deadman is sitting, the plane c (Z=0.2 m) includes two parts: hearth coke free zone and deadman. Because of large velocity difference between them, the flow distributions of them are shown respectively. It can be found from Figs. 6(a) and 6(b) that when deadman is sitting, the velocity of molten iron at the corners of hearth bottom is at least more than 10 times larger than that in deadman. It can be seen from Figs. 6(c) and 6(d) that in the plane c (h=0.2 m), the velocity of molten iron with floating deadman at 0.4 m is larger than that with floating deadman at 1.5 m, this is because the height of bottom coke free zone is small so that the velocity of molten iron flow is large. The global velocity of molten iron in bottom coke free zone will be smaller with higher floating height of deadman.

Fig. 6.

Flow field distribution of molten iron of plane c at different floating heights of deadman.

4.2. Flow Velocity of Molten Iron near Hearth Bottom Sidewall

Figure 7 presents that flow velocity distribution of molten iron at different floating heights of deadman. The hearth line AB is 0.05 m away from hearth wall and bottom line CD is 0.05 m away from bottom wall. The length of line AB is from bottom (point A) to taphole height (point B) and the length of line CD is from hearth center (point C) to hearth sidewall (point D). It can be observed from Fig. 7(a) that the velocity of molten iron at the corners of hearth bottom is large when deadman is sitting (h_c=0 m). However, the velocity of molten iron near the boundary between deadman and coke free zone is large when deadman is floating. And it increases gradually with increasing floating height. In addition, in the hearth coke free zone above 1.5 m height, the velocity of molten iron with heavy floating deadman (h_c=1.5 m) begins to increase. Therefore, the molten iron at the corners of deadman flows fast so that it is more likely to be eroded.

Fig. 7.

Flow velocity distribution of molten iron at different floating heights of deadman (h_c: floating height of deadman; ε_d: porosity of deadman).

It can be seen from Fig. 7(b) that the flow of molten iron at bottom center is slow when deadman is sitting. The velocity of molten iron increases suddenly at the corners of hearth bottom and a peak appears, so the elephant-foot type erosion is likely to be formed. However, the velocity of molten iron at the corners of hearth bottom with a floating deadman is smaller than that with a sitting deadman. The velocity of molten iron near bottom with a floating deadman at 0.4 m is larger than that with a floating deadman at 1.5 m, especially near hearth sidewall. This is because at the same tapping speed, a large bottom coke free zone can reduce the velocity of molten iron near bottom compared with a small bottom coke free zone.

4.3. Wall Shear Stress of Hearth Bottom

According to Fig. 3(b), the hearth wall shear stress (τ_hw) is given by Eq. (11) and the bottom wall shear stress (τ_bw) is given by Eq. (12), where v_h and v_b are the flow velocity parallel to the hearth sidewall and bottom wall, respectively.   

$${\tau }_{hw}=\mu {\left(\frac{\partial v_h}{\partial x}\right)}_{x=6.65}$$(11)

  

$${\tau }_{bw}=\mu {\left(\frac{\partial v_b}{\partial z}\right)}_{z=0}$$(12)

Figure 8 shows that wall shear stress corresponding to line AB and CD at different floating heights of deadman. In Fig. 8(a), when deadman is sitting at the bottom, the wall shear stress at the corners of hearth bottom is large. When deadman is floating slightly (0.4 m), the shear stress of hearth sidewall near bottom is also large. When deadman is floating heavily (1.5 m), the shear stress of hearth sidewall near bottom is the smallest in three cases. Therefore, the erosion is likely to be formed at the corners of hearth bottom when deadman is sitting or floating slightly. It can be found in Fig. 8(b) that when deadman is sitting at the bottom, the shear stress of bottom wall is the smallest, whereas the wall shear stress at the corners of hearth bottom is large. Hence, the erosion degree of bottom is the smallest when deadman is sitting. But the erosion might be serious at the corners of hearth bottom. When deadman is floating slightly, the shear stress of bottom wall is the largest. The washing and erosion from molten iron at bottom might be also serious. The calculated result of wall shear stress is less than the shear strength of hearth bottom brick, but the wall shear stress has obvious difference for different floating states of deadman. On the other hand, the results indicate that the brick erosion is not mainly caused by the wall shear stress for a short time, but also the chemical reaction, permeation, wearing, washing, alkalis damage and so on for a long period.

Fig. 8.

Wall shear stress of hearth bottom at different floating heights of deadman (h_c: floating height of deadman; ε_d: porosity of deadman).

4.4. Effect of Porosity of Deadman on Molten Iron Flow and Wall Shear Stress

It can be found from Figs. 7 and 8 that with decreasing porosity of deadman (ε_d) from 0.3 to 0.1, the velocity of molten iron at the corners of hearth bottom and shear stress of hearth sidewall increase when the deadman is sitting. When deadman is floating, the velocity of molten iron near the hearth sidewall and wall shear stress also increase with decreasing porosity of deadman. It shows that when the porosity of deadman is poor, the molten iron flow near hearth sidewall and wall shear stress of hearth bottom will be intensified and the hearth erosion might also be aggravated.

The porosity of coke bed near hearth sidewall can be different from that of deadman center. When porosity of coke bed near hearth sidewall decreases, it can also cause the change of molten iron flow. In order to study the influence of porosity of coke bed near hearth sidewall, the different porosity of coke bed near sidewall is analyzed. Figure 9 presents the influence of porosity of coke bed near hearth sidewall on flow velocity of molten iron. The calculation conditions are that deadman is sitting at the bottom and porosity of deadman center is 0.3 and thickness of coke bed near hearth sidewall is 0.1 m. It can be seen from Figs. 9(a) and 9(b) that the velocity of molten iron near bottom center decreases when porosity of coke bed near hearth sidewall increases. However, the velocity of molten iron at the corners of hearth bottom increases. When porosity of coke bed near hearth sidewall is smaller than or equal to porosity of deadman (porosity near hearth sidewall equals to 0.1 or 0.3 in the figure), the molten iron near hearth sidewall will flow slowly and the molten iron near bottom will flow fast. Therefore, if the porosity of deadman center is larger than the porosity of coke bed near hearth sidewall, the erosion of hearth sidewall will be relieved, whereas the erosion of bottom will be deepened. The bowl type erosion is likely to be formed.

Fig. 9.

Influence of porosity of deadman near hearth sidewall on flow velocity of molten iron.

5. Discussion

5.1. Verification of Model

The experimental apparatus of molten iron flow in hearth bottom had been established by Zhao.²⁵⁾ In this experiment, the hearth diameter and deadman diameter were 0.53 m and 0.51 m, respectively. The porosity of deadman was 0.37 and depth of salamander was 0.12 m. The taphole diameter was 5 mm and tapping speed was 2.8 m/s. The measured vertical plane in hearth coke free zone was 5 mm away from the hearth sidewall and its velocity distribution was measured by 3-D Laser Doppler Velocimetry (LDV), as shown in Fig. 10. Figures 10(a), 10(b) and 10(c) are the experimental results and 10(d), 10(e) and 10(f) are the calculating results at the conditions of the experiment. It can be found from Figs. 10(a) and 10(d) that the velocity of molten iron at corners of hearth bottom is generally larger than that in other regions below taphole. Figures 10(b) and 10(c) are the experimental results when the floating height of deadman are 0.03 m (quarter of the salamander’s height) and 0.06 m (half of the salamander’s height), respectively. The velocity at the corners of hearth bottom is smaller with a floating deadman than that with a sinking deadman. The phenomena of Figs. 10(e) and 10(f) have the similar results. Hence, the bottom coke free zone is beneficial for decreasing the velocity of molten iron at the corners of hearth bottom. Compared with experimental and computational results in this paper, the flow trends are basically similar, although some flow paths of molten iron are not identical. It further explains the influence of floating state of deadman on molten iron flow in hearth bottom.

Fig. 10.

Experimental results ((a)–(c)) and calculating results ((d)–(f)) of different floating heights of deadman.

5.2. Analysis of Hearth Erosion Migration

Figure 11 presents the migration of molten iron circulation and hearth erosion at different floating states of deadman. “ER” denotes where erosion is most likely to occur. In Fig. 11(a), there is almost no bottom coke free zone when deadman is sitting at the bottom. Therefore, the molten iron has to flow through deadman or along the hearth sidewall. When porosity of deadman is small, molten iron circulation will be strengthened so that elephant-foot type erosion can be possibly formed at the corners of hearth bottom. In Fig. 11(b), there is a small bottom coke free zone when deadman is floating slightly so that some molten iron can flow through bottom of deadman. The velocity of molten iron at corners of deadman bottom is also fast. Hence, the hearth sidewall corresponding to bottom of deadman is easy to be eroded when the permeability of deadman is poor. The erosion of BF B is this case. In Fig. 11(c), because deadman is floating heavily, the bottom coke free zone is large and velocity of molten iron at bottom is small. When the permeability of deadman is poor, the molten iron at the hearth sidewall corresponding to bottom of deadman will flow fast so that it is likely to form the erosion. The erosion of BF A is this case. That’s why erosion is obvious at the distance of 1–1.5 m under the taphole.

Fig. 11.

The migration of molten iron circulation and hearth erosion at different floating states of deadman.

The erosion profiles of blast furnace hearth bottom can be divided into bowl type erosion, elephant-foot type erosion and mumps face type erosion.^10,26) Actually, the erosion profiles are mostly decided by the floating state of deadman. Figure 12 shows the schematic diagram of erosion profiles at different floating states of deadman. Figure 12(a) presents the bowl type erosion. The position of the most serious erosion is at the bottom. This erosion is mainly due to the good permeability of deadman center, so that molten iron could flow to the bottom easily. Therefore, the bottom temperature is high and the stable skull of bottom is hard to be formed. However, there is bad permeability zone near hearth sidewall. The molten iron flows slowly and the skull can be formed. The hearth bricks don’t have serious erosion, whereas the bottom bricks erode seriously. Finally it forms the bowl type erosion. Figure 12(b) shows the elephant-foot type erosion, it is also called garlic type erosion. It is mainly caused by the concentration of molten iron in coke free zone and formed at the furnace bottom corners when deadman is sitting or floating slightly. It leads to high temperature and no stable skull at bottom corners because of strong circulation. Therefore, there is formed the elephant-foot type erosion. Figure 12(c) is the mumps face type erosion. The lining bricks in the middle portion of hearth bottom are eroded to a maximum degree. There is a large bottom free coke zone when deadman is floating heavily. The intermediate sidewall region between taphole and bottom is locally eroded, although molten iron flow at bottom corners is lessened. This is called mumps face type erosion.

Fig. 12.

The schematic diagram of erosion profiles at different floating states of deadman.

The hearth diameter of a Chinese BF with volume of 125 m³ is 3.2 m. The designed depth of the salamander is 0.304 m. Through calculation,⁸⁾ it’s found that the minimum depth of the salamander, which is required for floating deadman, is about 1.32 m at the early stage of campaign life. Therefore, deadman was sitting at the bottom at the early stage. The lining bricks at the furnace bottom corners were eroded quickly. As the erosion continuing, the depth of the salamander was deepening and the hearth diameter was also increasing. It was found from the dissection of BF that the depth of the salamander increased to 1.35 m after the blow down. It was indicated that the deadman could float at the later stage of campaign life. The dissection results also showed that there was the bottom coke free zone, that was to say, the deadman was floating. Figure 13(a) presents the salamander and erosion profile of hearth bottom after the dissection. It can be found clearly from the figure that the iron layer includes two parts. There is uniform distribution of coke particles in upper part, whereas there is no coke particle in the lower part, which means the deadman is floating. The appropriate floating height of deadman slowed down the erosion at the corners of hearth bottom. Considering the small hearth diameter of small BF, the permeability of deadman center is usually good and the bottom is easy to be eroded. It is formed the bowl type erosion eventually.

Fig. 13.

Erosion profiles of BF hearth bottom.

The hearth diameter of another Chinese BF with volume of 750 m³ is 6.9 m. The designed depth of the salamander is 1.316 m. It’s found by calculation that the minimum depth of the salamander, which is required for floating deadman is about 2.114 m. The deadman couldn’t float at the early stage of campaign life so that the bottom and corners were eroded seriously. However, the floating state didn’t change so much with the erosion going on. The body of BF was dissected after blow down. It was found the depth of the salamander was 2.316 m. The deadman could float slightly (the floating height was 0.202 m) after the later stage. Therefore, the deadman was generally sitting or floating slightly in the whole campaign life. It led to strong circulation and serious erosion at hearth sidewall and bottom corner. With the erosion going on, it formed the elephant-foot type erosion finally. Figure 13(b) shows the erosion profile of hearth bottom after the dissection. They are staves, carbon bricks and salamander from left to right in the figure. It can be observed clearly that shape of solidified iron revealed the obvious elephant-foot type erosion. The serious erosion happened at the bottom corner, where the thinnest thickness of the bricks is less than 200 mm.

In order to reduce blast furnace hearth erosion, the floating height of deadman should be judged accurately at first. The flow intensity of molten iron and erosion position are affected by the floating height of deadman. The current floating state and activity of deadman can be estimated through the temperature changes of hearth bottom thermocouples, which is beneficial for guiding the operation and maintenance. The permeability of deadman will influence the molten iron flow and the floating state of deadman. Good permeability of deadman can help reduce the washing and erosion from molten iron on hearth sidewall and maintain activity of hearth center.

6. Conclusions

Through the research of two large BFs in China, it was found that the change trends of bottom temperature and hearth sidewall temperature were different for BF A, whereas the change trends of them were almost the same for BF B. The most serious erosion positions were also different. A three-dimensional mathematical model of molten iron flow in hearth bottom has been established to analyze the reasons. The results are as following:

(1) The relationship between hearth sidewall temperature and bottom temperature is influenced by the floating states of deadman. The floating height of deadman for BF A is obviously higher than that for BF B. Therefore, the height of hearth erosion position of BF A is higher than that of BF B, where it is about 1–1.5 m below the taphole.

(2) The molten iron flow and wall shear stress of hearth bottom will change with the floating height of deadman. When deadman is sitting, the molten iron flow and wall shear stress at the corners of hearth bottom will be large. When deadman is floating slightly, the molten iron flow and wall shear stress at the corners of deadman and bottom is large. When deadman is floating heavily, the molten iron flow and wall shear stress between hearth sidewall and deadman is large, whereas the molten iron flow and wall shear stress at bottom is small. The decrease of porosity of deadman will also increase the velocity of molten iron and wall shear stress of hearth sidewall.

(3) The hearth erosion position will change with floating height of deadman. When deadman is sitting at the bottom, hearth erosion is easy to occur at hearth bottom corners. When deadman is floating, hearth erosion is easy to occur at the hearth sidewall corresponding to deadman corners. It is likely to form bowl type erosion when permeability of deadman center is good and velocity of molten iron near hearth sidewall is small. It will cause elephant-foot type erosion when deadman is sitting or floating slightly. It will form mumps face type erosion when deadman is floating heavily.

Acknowledgement

The authors express their thanks to National Natural Science Foundation of China for their kind financial support (No. 61271303).

References

• 1)   K.  Shibata,  Y.  Kimura,  M.  Shimizu and  S.  Inaba: ISIJ Int., 30 (1990), 208.
• 2)   L.  Shao and  H.  Saxen: Steel Res. Int., 83 (2012), 197.
• 3)   C. E.  Huang,  S. W.  Du and  W. T.  Cheng: ISIJ Int., 48 (2008), 1182.
• 4)   Z. J.  Liu,  J. L.  Zhang,  H. B.  Zuo and  T. J.  Yang: ISIJ Int., 52 (2012), 1713.
• 5)   S. R.  Zhang and  Z. J.  Yu: Blast Furnace Aberration and Accident Treatments, Metallurgical Industry Press, Beijing, (2012), 93.
• 6)   K.  Kanbara,  T.  Hagiwara,  A.  Shigemi,  S.  Kondo,  Y.  Kanayama,  K.  Wakabayashi and  N.  Hiramoto: Tetsu-to-Hagané, 62 (1976), 535.
• 7)   T.  Inada,  A.  Kasai,  K.  Nakano,  S.  Komatsu and  A.  Ogawa: ISIJ Int., 49 (2009), 470.
• 8)   J. F.  Zhu,  H. B.  Zhao,  S. S.  Cheng and  H. W.  Pan: J. Univ. Sci. Technol. Beijing, 31 (2009), 906.
• 9)   J. F.  Zhu,  S. S.  Cheng,  H. B.  Zhao and  L.  Zhang: J. Univ. Sci. Technol. Beijing, 31 (2009), 224.
• 10)   A.  Shinotake,  H.  Nakamura,  N.  Yadoumaru,  Y.  Morizane and  M.  Meguro: ISIJ Int., 43 (2003), 321.
• 11)   K.  Takatani,  T.  Inada and  K.  Takata: ISIJ Int., 41 (2001), 1139.
• 12)   B. Y.  Guo,  D.  Maldonado,  P.  Zulli and  A. B.  Yu: ISIJ Int., 48 (2008), 1676.
• 13)   J.  Torrkulla and  H.  Saxen: ISIJ Int., 40 (2000), 438.
• 14)   J.  Torrkulla and  H.  Saxen: ISIJ Int., 43 (2003), 1519.
• 15)   J.  Torrkulla,  J.  Brannbacka,  H.  Saxen and  M.  Waller: ISIJ Int., 42 (2002), 504.
• 16)   S. N.  Silva,  F.  Vernilli,  S. M.  Justus,  O. R.  Marques,  A.  Mazine,  J. B.  Baldo,  E.  Longo and  J. A.  Varela: Ironmaking Steelmaking, 32 (2005), 459.
• 17)   W. R.  Xu,  J. M.  Zhu,  X. M.  Mao,  L. L.  Zhang and  K. H.  Liu: Proc. AISTech 2014, AIST, Warrendale, PA, (2014), 891.
• 18)   M.  Swartling,  B.  Sundelin,  A.  Tilliander and  P.  Jonsson: Ironmaking Steelmaking, 37 (2010), 21.
• 19)   V.  Panjkovic,  J.  Truelove and  P.  Zulli: Ironmaking Steelmaking, 29 (2002), 390.
• 20)   B.  Desai,  R. V.  Ramna and  S. K.  Dash: ISIJ Int., 46 (2006), 1396.
• 21)   Y. L.  Li,  S. S.  Cheng and  C.  Chen: J. Iron Steel Res. Int., 22 (2015), 13.
• 22)   Subagyo,  N. Standish and G. A. Brooks: Chem. Eng. Sci., 53 (1998), 1375.
• 23)   A.  Preuer,  J.  Winter and  H.  Hiebler: Steel Res., 63 (1992), 139.
• 24)   K. C.  Zheng,  Z.  Wen,  X. L.  Liu,  Y. Q.  Ren,  W. F.  Wu and  H. B.  Qiu: ISIJ Int., 49 (2009), 1277.
• 25)   M. G.  Zhao: Doctoral Thesis, University of Science and Technology Beijing, (2007).
• 26)   H. B.  Zhao: Master’s Thesis, University of Science and Technology Beijing, (2006).