Analysis of Blast Furnace Hearth Sidewall Erosion and Protective Layer Formation

Abstract

The blast furnace campaign life is largely determined by the damage to the hearth area. Bricks and skulls in the hearth were sampled and investigated after the blowing-out of one commercial blast furnace. A variety of techniques, such as chemical analysis, scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS), and X-ray diffraction (XRD), were applied in characterizing the erosion and protection mechanism of the hearth sidewall. The upper area of the hearth was heavily eroded. Large amounts of alkali chlorides were identified in the area, which could be the main contribution to the serious erosion. The taphole area was protected by the formation of the skull, which consisted of slag with CaZnSi₃O₈, Zn₂SiO₄ and CaMg₂Al₂O₇ phases. The circulation of the hot metal in the hearth is the major cause of brick erosion in the hearth bottom area, which leads to a ring shaped wear. Controlling peripheral hot metal flow across the hearth wall using long taphole and coke with high degradation resistance may slow the carbon brick erosion. Increasing the carbon content in the hot metal can help reducing the dissolution of the carbon bricks and enhance the formation of a graphite layer to protect the hearth lining.

1. Introduction

In recent years there has been increasing emphasis on extending blast furnace (BF) campaign life.^1,2) Procedures have been pursued and BF campaign lives have been steadily increasing.^3,4) Many studies on the relationship between inner hearth condition and BF operating conditions, such as heat load, molten iron flow, the lower boundary level and permeability distribution of deadman in hearth have been carried out by using numerical simulation method.^5,6) Considering the complexity and variety of phenomena and reactions occurring in the BF, our current knowledge of the inner condition of the BF is still limited. The inner erosion condition of BF can only be observed by the dissection of the furnace after blow-out. Dissection investigations have shown many features in the erosion of hearth refractory.

The skull, a layer of metal iron and slag attached on the inner surface of the hearth lining, provides a barrier between carbon brick and molten materials such as liquid iron or slag. As long as the skull exists, the hearth erosion is limited. However, a key question is how a skull layer formed can be maintained during the BF operation. A number of studies have investigated the hearth erosion of the BF, but the information on the composition and the microstructure of lumped skull lining material are still very limited.^{7,8,9,10,11,12,13,14,15,16)} An improved understanding of the skull formation and the wear mechanisms of the refractories are important for optimizing the furnace operation and extending the BF campaign life.

In this paper, the refractory erosion of a dissected BF was investigated. The profile of the erosion of the hearth was described. The skulls formed on the hot surface of the refractory were sampled and analyzed using chemical analysis, X-ray diffraction (XRD), and scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS). The chemical composition and phase information of the skull were obtained and the possible erosion mechanisms occurring in the hearth were proposed.

2. Experiment

2.1. Investigation of the Refractory Damage in the BF Hearth

The studied BF had an inner volume of 1350 m³ and two tapholes located in east and west sides. The furnace bottom was lined with ceramic cup, micro carbon bricks and semi-graphite bricks, while the ceramic cup and micro carbon bricks were used for the hearth sidewall. The BF was started up on November 9, 2012, and shut down on August 16, 2015 for maintenance. Erosion of the hearth lining was investigated during the repair of the BF and serious erosion was observed in the hearth. It was found that the residual thickness of the refractory lining was generally in the similar range in the circumferential direction, while in the vertical direction, the erosion condition of the refractory lining varies with the height.

The inner view and conditions of the furnace hearth is shown in Fig. 1(a). It can be seen that the ceramic cup was completely disappeared and the carbon brick was also eroded. At the level of layers 14 to 17 (Fig. 1(b)) of carbon brick, the erosion of the brick was serious. The remaining thickness of the brick was only 475 mm (the total thickness of the brick lining was 1100 mm. It also shows that the erosion of the brick at layer 13 is much less serious, so the residual thickness of the brick at layer 13 is much larger than that of layer 14 and the difference of the residual thickness was about 300–400 mm. As illustrated in Fig. 1(b), a platform appeared at the interface of the bricks at layer 13 and 14. The ceramic cup at the level of layers 10 and 11 has been completely eroded, which leaves the carbon bricks in a quite good condition at the same level. The carbon bricks at the layers 7 to 9 were found to be seriously eroded into a bowl shape. In contrast, there is little erosion of the carbon bricks at the bottom and the original shape of the brick was kept almost unchanged.

Fig. 1.

Erosion profile of the blast furnace hearth. (a) Inner view of the furnace hearth. (b) Schematic of the hearth bottom erosion profile.

2.2. Sampling

After blow-out and cool-down of the furnace, the furnace hearth was dissected and the skull, a layer of solid materials attached to the refractory were found in various locations. Some skull samples were collected manually. In this study, the circumferential positions of the collected samples were close to the west taphole of the furnace while the vertical positions: P1 at layer 15, P2 at layer 11 and P3 at layer 8 were marked in the Figs. 1(a) and 1(b). It needs to be mentioned that the skull samples were taken from the hot surface of the carbon brick and were retained in their original shape as much as possible.

2.3. Analysis Methods

The phase information, chemical composition and microstructure of the sampled skulls were analyzed using XRD and SEM–EDS. The XRD analysis was conducted using a Rigaku diffractometer (DMAX-RB 12 kW; Rigaku Corporation, Tokyo, Japan). The Cu Kα radiation (30 kV, 30 mA) was used as the X-ray source; the scanning angles were in the range of 10 to 90 deg (2θ) at the scanning rate of 10 deg/minute. Small pieces were cut from the selected samples under dry conditions and mounted in resin. The polished samples were coated with Au and examined using a Quanta 250 Environmental scanning electron microscope (SEM) equipped with an energy-dispersive X-ray spectrometer (EDS) for chemical analysis and element mapping.

3. Results and Discussion

3.1. Characterization of the Samples

3.1.1. Sample P1 in the Hearth

A large number of carbon bricks at layers 14–17 were found to be loose and brittle, and could be easily removed manually. As shown in Fig. 2(a), a lot of SiO₂ particles of various morphologies and sizes and pores were observed in the carbon bricks. Figure 2(b) shows the size and morphology of alkali chlorides on the hot surface of the carbon brick. Alkali chlorides were present in various morphologies including cylindrical and acicular shapes. The cylindrical morphology had a diameter of about 6 μm and a height of about 10 μm (Fig. 2(c)). The EDS results confirmed that the cylindrical shaped crystals were KCl (Fig. 2(d)) and the acicular shaped crystals were NaCl (Fig. 2(e)). Large amounts of KCl was identified on the hot surface of the carbon brick (Fig. 2(f)), which agrees with the findings from a number of post mortem studies of commercial BF hearths,^17,18) in which the existence of an alkali rich zone located in the intermediate lining area was reported. The XRD analysis of the selected skull was shown in Fig. 3. The alkalis mainly consisted of KCl and distributed on the upper area of the furnace hearth, which was different from the previous findings^17,18) that K existed as K₂O in the carbon brick in the hearth.

Fig. 2.

Microstructure of the Sample P1 in the hearth.

Fig. 3.

XRD analysis of the sample P1 in the hearth.

3.1.2. Sample P2 in the Hearth

Figure 4 shows the microstructure of the cold surface of sample P2 near the taphole in the hearth. The identified major phases are carbon matrix (black area), metallic iron (white area) and SiO₂ (grey area) (Fig. 4(a)). The light grey area on the edge of SiO₂ grain (red circle in Fig. 4(b)) is potassium nepheline (K₂O·Al₂O₃·2SiO₂) and leucite (K₂O·Al₂O₃·4SiO₂) phases which formed through the reactions of alkali with silica and alumina in the brick (Fig. 4(c)). The formation of nepheline and leucite phases has been confirmed in several laboratory experiments concerning alkaline attack on carbon refractories.^19,20,21)

Fig. 4.

Microstructure of the cold surface of sample P2 near the taphole in the hearth.

The microstructure of SiO₂ grain is shown in Fig. 5. The SiO₂ grain was wrapped in a layer of Zn₂SiO₄ which suggests that the Zn vapor diffused into carbon brick along the pores and reacted with SiO₂(in carbon brick) to form a compound.

Fig. 5.

Microstructure of the SiO₂ grain of the Sample P2 in the hearth.

As shown in Fig. 6, some calcium and zinc rich phases and silicon were identified on the hot surface of sample P2, which suggests that CaO and ZnO in the slag transferred to the slag/brick boundary and reacted with SiO₂ and Al₂O₃ in the refractory and formed complex silicates and aluminates compounds. The XRD analysis of the sample showed that the phases were mainly CaZnSi₃O₈, Zn₂SiO₄ and CaMg₂Al₂O₇ (Fig. 7). The reactions between slag and refractory, the formation and possible suspension of the solid phases in the slag could lead to the increased viscosity of the slag layer close to the refractory. This can help forming a layer of viscous slag on the carbon brick surface and protect the brick.

Fig. 6.

Microstructure of the hot surface of Sample P2 near the taphole in the hearth.

Fig. 7.

XRD analysis of the sample P2 in hearth.

3.1.3. Sample P3 in the Hearth

Figure 8 shows the microstructure of sample P3 in the hearth. The major phases identified were graphite and metal iron. The iron and the graphite were separated into different layers and the clear boundaries could be identified. As shown in Fig. 8, the width of the iron layer was 1000–1800 μm while the graphite width was 300–400 μm. Due to the large size of the graphite and the iron, the formation of the skull is thought to be formed during the furnace normal operation, not during the furnace shut-down and cooling. Different to the SEM results, the XRD results in Fig. 9 show that Fe exist as magnetite, which may be due to the oxidation of metal iron during the preparation of the powder sample for XRD analysis.

Fig. 8.

Microstructure of the sample P3 in the hearth.

Fig. 9.

XRD analysis of the sample P3 in the hearth.

3.2. Erosion Mechanism of the Refractory

The hearth is an important component of BF, and the campaign life of a BF is largely determined by the life of the refractory lining in the hearth. Refractories in the BF hearth are subjected to the following erosion mechanisms: oxidation, alkali attack, CO attack, thermal stress erosion and dissolution due to hot metal and slag flow.^3,13) Through dissection investigations of blast furnaces, the hearth erosion was found to be in various shapes, such as “bowl-shaped” erosion and “elephant foot shaped” erosion.²²⁾ It is not clear which is the dominant mechanism for causing the hearth erosion. The characterization of the samples in previous section show that the erosion profile varies with the position in the furnace, which is also different from what has been reported in other studies.^22,23)

3.2.1. Upper Area in the Hearth

A large number of alkali were found in the upper region of the BF hearth, which confirms that alkali was circulated and accumulated inside the BF. Figure 10 is a schematic of the circulation of alkali in the BF. It has been widely reported that a high alkali load can cause scaffold formation on the BF shaft wall. However, alkali chloride was also found in the furnace hearth in the present study. The alkali in the raw materials such as iron ores, coke and flux, exists in the form of sodium and potassium chlorides and complex silicates. The alkali chlorides evaporate at high temperature and then deposit on the refractory hot surface at lower temperature. Alkali silicates decompose and are reduced by carbon in bosh and the hearth regions to produce elemental potassium and sodium vapors by the following equations:^24,25)   

$${\text{2K}}_{\text{2}}{\text{SiO}}_{\text{3}}+\text{2C}=\text{4K(g)}+{\text{2SiO}}_{\text{2}}+\text{2CO}$$(1)

  

$${\text{2Na}}_{\text{2}}{\text{SiO}}_{\text{3}}+\text{2C}=\text{4Na(g)}+{\text{2SiO}}_{\text{2}}+\text{2CO}$$(2)

Fig. 10.

Schematic of the circulation of alkali in the blast furnace.

Hydrogen chloride in the injected coal volatilizes in the tuyere. Alkali chloride can be formed by the following equations:   

$$\text{2K}+\text{2HCl(g)}=\text{2KCl}+{\text{H}}_{\text{2}}$$(3)

  

$$\text{2Na}+\text{2HCl(g)}=\text{2NaCl}+{\text{H}}_{\text{2}}$$(4)

The high temperature in the tuyere and the hearth area, indicates that the formed alkali chlorides exist in the gaseous state (The boiling point of KCl and NaCl are 1420 and 1465°C, respectively), which diffuse into the pores of the refractory and deposit and solidify as the temperature decrease to below their melting point in the carbon brick (the melting point of KCl and NaCl are 770 and 801°C, respectively). Therefore, the formation of alkali chlorides could be a major cause for the serious erosion of the brick in the upper location in the hearth. Reducing the input of the alkali and chloride into the furnace may be an effective way for slowing the erosion of the upper hearth. The further investigation of the effect of alkali chlorides on the refractory erosion is an interesting subject for future study.

3.2.2. Taphole Area in Hearth

As shown in Fig. 1(b), the ceramic cup was lined by the carbon blocks in the furnace hearth. It was found that the ceramic cup was fully eroded in the taphole area and the erosion of the ceramic cup was irreversible. After tapping the hot metal, the slag is in contact with the brick at the taphole level and a skull could be formed. The formed skull separates the carbon brick from the hot metal, so the carbon brick is protected. The Zinc, alkali and silicon deposits were also found in the skull, which suggests that those harmful elements diffuse into and through the ceramic cup to the brick and formed deposits due to the low temperature in the area. Considering the short life (2 years) in this BF, the elements are thought to have accelerated the erosion of the ceramic cup. Figure 11 is the schematic of the formation of the skull. The formed skull layer may also decrease due to the fluctuations of operating conditions. Therefore, a stable operation of the BF is very important for slowing the refractory erosion.

Fig. 11.

Schematic of the formation of skull.

3.2.3. Interface of the Hearth Bottom

The hearth wall is usually protected by a skull formed on its surface during operation. However, when the hot metal flow becomes very turbulent, refractory sections are eaten away, which leads to the ring shaped wear. The circulation of the hot metal in the hearth is the major cause of the erosion of hearth bottom area. The hot metal in the hearth is not carbon saturated.^26,27) The carbon dissolves into the hot metal when the hot surface temperature of the refractory is higher than the carbon saturation temperature Tc. The erosion mechanism of the hearth bottom refractory due to the metal flow is shown in Fig. 12. There is a temperature gradient in the carbon brick due to the circulation of hot metal (α₁) and cooling water (α) on both sides of the brick. When the hot surface temperature of carbon brick (T₁) is higher than the carbon saturation temperature (T_C), the carbon brick will be eroded by dissolving into hot metal. Increase of carbon content in the hot metal results in the increase of carbon saturation temperature Tc. When the hot surface temperature equals to the carbon saturation temperature, the brick dissolution stops. In the case of when the circulation of hot metal is reduced (α₂) or the water cooling is enhanced (α₃), the hot surface temperature will decrease to T₃, which will be lower than carbon saturation temperature Tc, and graphite starts to precipitate out of the carbon saturated hot metal and deposit on the brick hot surface to form a protective layer. Hot surface temperature is the most important factor affecting the protective layer formation. The quantity of the precipitated graphite is mainly determined by the hot surface temperature. The thickness of the graphite protective layer changes with the variation of the hot surface temperature and reaches a dynamic balance. The heat flux increases with decreasing of brick thickness during erosion. After the formation of the graphite layer, the heat flux decreases and becomes stable, leading to lower heat loss and lower temperature in the hearth. In addition to the temperature, the carbon content in the hot metal has significant effect on the brick erosion and the graphite protective layer formation. Increase carbon level in hot metal will reduce the dissolution erosion of the brick and enhance the formation of a graphite layer for protecting the hearth lining.

Fig. 12.

Erosion mechanism of the hearth bottom refractory due to the metal flow.

Therefore, it is important, from an operation of view, to control peripheral hot metal flow across the hearth wall, which accounts for the ring shaped wear. In practical terms, using a longer taphole to tap the metal from an area further away from the sidewalls can help reducing peripheral flow against the hearth refractory lining. The use of lump coke with high degradation resistance is expected to be an efficient way to increase hearth permeability, slow the peripheral hot metal flow and therefore reduce the hearth lining wear rate.

4. Conclusions

After the blowing-out of a commercial BF, bricks and skulls in the hearth were sampled and analyzed. The findings are summarized as follows.

The erosion shape of the refractory varies with the position in the BF hearth. The upper area in the hearth was eroded seriously. The taphole area was protected by the formation of the skull. The hearth bottom area was presented in a ring shaped wear.

The upper area in hearth was eroded seriously and large amount of alkali chlorides were identified in the area, which could be the main contribution to the serious erosion. The taphole area was protected by the formation of the skull, which consisted of viscous slag with CaZnSi₃O₈, Zn₂SiO₄ and CaMg₂Al₂O₇ phases.

The circulation of the hot metal in the hearth is the major cause of brick erosion in the hearth bottom area, which leads to a ring shaped wear. Controlling peripheral hot metal flow across the hearth wall using long taphole and coke with high degradation resistance is expected to slow the carbon brick erosion. Increasing the carbon content in the hot metal can help reducing the dissolution erosion of the brick and enhance the formation of a graphite layer for protecting the hearth lining.

Acknowledgements

This work was financially supported by the National Science Foundation for Young Scientists of China (51304014), the Key Program of the National Natural Science Foundation of China (No. U1260202) and the 111 Project (No. B13004). Support from the CSIRO Mineral Resources is gratefully acknowledged.

References

• 1)   T. J.  Yang: Ironmaking, 33 (2014), 1.
• 2)   Z. J.  Liu,  J. L.  Zhang and  T. J.  Yang: ISIJ Int., 55 (2015), 1146.
• 3)   Z. J.  Liu,  J. L.  Zhang,  T. J.  Yang and  H. B.  Zuo: ISIJ Int., 52 (2012), 1713.
• 4)   M. E.  Barbes,  E.  Marinas and  E.  Brandaleze: ISIJ Int., 48 (2008), 134.
• 5)   K.  Takatani,  T.  Inada and  K.  Takata: ISIJ Int., 41 (2001), 1139.
• 6)   K. C.  Zhang,  Z.  Wen and  X. L.  Liu: ISIJ Int., 49 (2009), 1277.
• 7)   A.  Shinotake,  H.  Nakamura,  N.  Yadoumaru,  Y.  Morizane and  M.  Meguro: ISIJ Int., 43 (2003), 321.
• 8)   T.  Inada,  A.  Kasai,  K.  Nakano,  S.  Komatsu and  A.  Ogawa: ISIJ Int., 49 (2009), 470.
• 9)   M. S.  Song and  Z. Q.  Zou: Proc. 2004 Natl. Iron Production & Technology and Annual Ironmaking Conf., CSM, Beijing, (2004), 585.
• 10)   K.  Narita,  M.  Maekawa,  T.  Onoye,  Y.  Satoh and  M.  Miyamoto: Tetsu-to-Hagané, 62 (1976), 525.
• 11)   W. D.  Zhang,  L. J.  Ren,  H. B.  Shen and  Z. M.  Guo: Ironmaking, 29 (2010), 11.
• 12)   Y. F.  Zhao,  F.  Dong,  W.  Louis,  Y.  Chen,  C. Q.  Zhou and  G.  Grindey: Steel Res. Int., 85 (2013), 891.
• 13)   S. N.  Silva,  F.  Vernilli and  S. M.  Justus: Mater. Corros., 64 (2012), 1032.
• 14)   K. X.  Jiao,  J. L.  Zhang,  Z. J.  Liu and  F.  Liu: Int. J. Miner. Metall. Mater., 22 (2015), 1017.
• 15)   K. X.  Jiao,  J. L.  Zhang,  Z. J.  Liu and  M.  Xu: J. Miner. Metall. Sect. B, 51 (2015), 143.
• 16)   Y.  Li,  Y. Q.  Li and  R. J.  Fruehan: ISIJ Int., 41 (2001), 1417.
• 17)   H. W.  Pan,  S. S.  Cheng and  S.  Yu: Iron Steel, 3 (2011), 13.
• 18)   H.  Jiang and  J. J.  Cai: China Metall., 24 (2014), 57.
• 19)   S. N.  Silva,  F.  Vernilli and  S. M.  Justus: Ironmaking Steelmaking, 32 (2005), 459.
• 20)   K. J.  Li,  K.  Rita,  J. L.  Zhang,  Z. J.  Liu,  S.  Veena,  T. J.  Yang and  D. W.  Kong: Fuel, 133 (2014), 194.
• 21)   H. B.  Zhao,  S. S.  Cheng and  M. G.  Zhao: J. Iron. Steel Res. Int., 14 (2007), 6.
• 22)   Z. Y.  Xiang: China Metallurgy, 23 (2013), 1.
• 23)   K. X.  Jiao,  J. L.  Zhang,  Z. J.  Liu,  L. S.  Liang and  F.  Liu: Int. J. Miner. Metall. Mater., 23 (2016), 1.
• 24)   R.  Altland,  B.  Beckmann and  K.  Stricker: Stahl Eisen, 11 (1999), 53.
• 25)   Q. D.  Zhou and  X. G.  Bi: Scand. J. Metall., 16 (1987), 57.
• 26)   F.  Neumann,  H.  Schenck and  W.  Patterson: Carbon Act. Satur. Point Cast Iron, 47 (1960), 25.
• 27)   R.  Khanna,  R.  Sakurovs and  H.  Sun: Metall. Mater. Trans. B, 37 (2006), 623.