Red scale and copper brittle defects are common defects in the production process of hot-rolled strip steel, especially the red scale defect of high-silicon steel grades, which has always been one of the important reasons that plague the surface quality of strip steel. In recent decades, scholars from Japan and other countries have made some research progress on red scale and copper brittleness, sorted out the methods, content, and technical ideas of related research, and compiled them into an article for readers’ reference.
Keywords: Hot-rolled strip production
The production process of plate and strip is accompanied by high-temperature oxidation of steel materials. Oxidation defects caused by this can account for one-third of the surface quality problems of plate and strip products. Among them, red scale and copper brittle (also known as hot brittle) defects are the most common defects caused by material oxidation. In 1994, Japanese scholar Fukagawa Chiki et al.[0] proposed the classic formation mechanism of red scales. In the past 30 years since then, the mechanism has been widely accepted and cited in various literatures. At the same time, the research on the effect of Si element on red scale defects is also flourishing, and a relatively mature theoretical system has been formed so far. However, when this defect explains the actual phenomenon on the spot, there are also doubts. The author found this doubtful point in the production practice of studying the red scale defect, and verified it. At the same time, some of the author’s research conclusions coincide with the research results of Chinese Taiwan Sinosteel scholars. Whether the red scale defect is necessarily caused by the heating furnace–the sub-oxidized iron scale is not clean, it is worth further investigation at present.
Copper brittle defects are also one of the common defects of hot-rolled thin strip steel. This paper introduces the research of Japanese scholars on the mechanism of copper embrittlement, and clarifies the deep-seated reasons for the measures to control copper embrittlement in the industry. In fact, Japanese scholars often carry out some very interesting researches that may not seem to have the conditions for large-scale promotion at present, such as the suppression of copper embrittlement,
1. Hot-rolled strip red scale defect
1.1 The traditional theory of red scale formation mechanism
In the 1990s, some Japanese scholars proposed the formation mechanism of red scale defects in hot-rolled strip steel. This mechanism was widely recognized and was repeatedly cited by other researchers in the past 30 years.
Tomoki FUKAGAWA et al studied the role of Si element in the formation mechanism of red scale defects in hot-rolled strip steel with two steels of 0.09%-0.54%Si-1.46%Mn and 0.05%C-0.15%Mn as objects . The slab is heated at 1220 C for 2 hours, the heating atmosphere is 77.1%N2+14.3%H2O+8.2%CO2, and the generation of primary oxide scale is controlled by whether to use a stainless steel protective cover.
It was found that even in Si-free steel, incomplete descaling before hot rolling can lead to the formation of red scale. The steel is heated at 1000°C in the air, and the oxides formed include FeO, Fe3O4. and Fe2O3, and the volume ratio is about 95:4:1. If FeO is broken during the rolling process, the broken FeO particles will be supplemented by matrix Fe2+ and reduce, but the contact area with O2 in the air will increase, so the reaction of FeO→Fe3O4→Fe2O3 will be accelerated. The mechanism of red scale formation is shown in Figure 1 shown. Studies have shown that the color depth of red scale defects is determined by the proportion of Fe2O3 particles below 2 μm.
The schematic diagram of the formation mechanism of red scale in hot rolling of Si-containing steel is shown in Fig. 2. Since the red scale defect of Si-containing steel only appears on the prefabricated furnace-grown primary scale slab, researchers believe that the occurrence of red scale defects is closely related to the furnace-grown primary scale. Further analysis shows that the FeO/Fe2SiO4 eutectic compound penetrates into the FeO grains and steel matrix to form a so-called “anchor structure”, which is the direct cause of deteriorating descaling properties.
The melting point of Fe2SiO4 is 1173°C. When heated at 1220°C, Fe2SiO4 is in the liquid phase, and it wraps FeO particles to form a FeO/Fe2SiO4 eutectic compound, creating an anchor-like morphology. When the Fe2SiO4/FeO eutectic compound is at 1163°C (just solidified, melting point 1173°C), it can also undergo plastic deformation, but when it is at 1107°C, the eutectic compound cannot be deformed, and its hardness is much higher than that of FeO, so in the subsequent rolling deformation , FeO particles are broken and further oxidized to Fe2O3. This high-strength eutectic compound makes it difficult to remove by descaling even if the temperature is only lower than the solid temperature of 1173 °C. It can be seen that Fukagawa Chichi et al. support the view that “descaling of eutectic compounds in liquid state helps to remove iron oxide scale”.
Fig. 3 is the situation of the oxidation interface obtained when the differential thermal analyzer is used for oxidation simulation. The steel used in the test is DP590 steel (w(Si)=0.45%), the volume fraction of O2 is controlled to 2%, the rest is hydrogen, and the oxidation time is 100 min. It can be clearly seen that solid Si-rich particles (SiO2 + Fe2SiO4) are formed at 1100 °C; while at 1300 °C, Fe2SiO4 liquefies and penetrates into the Fe matrix to form an anchor structure, and at the same time invades FeO and forms a tortoise shell network along the grain boundary. Morphology; this morphology is considered to be the reason for the difficult removal of iron oxide scale.
1.2 Factors affecting the formation of red scale defects
1.2.1 The influence of alloying elements Si and Ni
Among the effects of alloying elements on the high temperature oxidation behavior of iron and steel materials, the role of Si element is the most important.
Added Si elements with a mass fraction of 0.1% to 1.5% to industrial pure iron, and investigated the influence of temperature and Si content on the high-temperature initial oxidation of Fe-Si alloys. Si mass fractions reaching 1.5% basically belong to the Si dosage range of advanced high-strength steel for plates. Studies have shown that at 1100°C, the enriched layer of Si plays a protective role, and the amount of oxidation decreases with the increase of Si content; at 1200°C, due to the formation of Fe2SiO4 liquid phase, the amount of oxidation increases with the increase of Si content . The highlight of this study is that according to the theoretical phase diagram, Fe2SiO4 in steel with a Si mass fraction of 0.5% at 1150 C has not yet reached the liquidus temperature, but from the measured oxidation behavior, a liquid phase is indeed formed at this time. The study believes that this is due to the local overheating caused by the surface oxidation reaction to form a local liquid phase, and the formation of the liquid phase further accelerates the oxidation reaction, and the heat released by the oxidation reaction intensifies the formation of the liquid phase. Oxygen present in the liquid phase is observed at the temperature of the liquidus point
behavior.
Tata steel explores the effect of the optimal composition (0.06%C-1.6%Mn-0.15%Si-1%Ni) chosen to achieve the properties of the steel on the formation of oxides in the heating furnace and its primary Influence of oxide scale descaling performance. The heating atmosphere simulates the actual heating furnace exhaust gas, mainly 8.8%CO2-1.7%O2-18%H2O (steam)-71.5%N2. This study focuses on the role of Ni element in the formation of red scale defects. After Si-Ni composite, the interfacial entanglement will be strengthened, which has a particularly serious impact on the descaling performance. Ni-rich metal particles are distributed around the pores of FeO scale, mainly near Fe2SiO4. This type of interface structure will increase the probability of red scale defects.
1.2.2 Effect of hot rolling and descaling conditions on red scale defects
Hikaru OKAD, Tomoki FUKAGA-WA, etc. used a three-stand simulated rolling mill to study the effects of hot rolling and descaling conditions on common red scale defects in high-Si hot-rolled strip steel.
The research shows that no matter whether the steel contains Si element or not, as long as the thickness of the oxide scale on the surface of the hot-rolled strip is greater than 20 μm before rolling and the rolling temperature is lower than 900 °C, red scale defects will form. This is mainly due to the fact that when hot rolling below 900 degrees Celsius, the surface part of the oxide scale (mainly FeO) is broken into powder, and the powder is oxidized during the cooling process to form red Fe2O3. With the decrease of rolling temperature, the content of powder in the iron oxide scale gradually increased (the amount of powdering was evaluated by detecting the weight loss of the iron oxide scale after ultrasonic cleaning). During the heating process of the slab, the thick iron oxide scale formed by low-Si steel can be completely removed by descaling. However, due to the strong adhesion of Fe2SiO4 on high-silicon steel, FeO remains, and the residual iron oxide scale forms red scale after rolling and cooling. defect. According to the analysis, the oxygen formed after heating the high Si steel slab, and after descaling of the iron scale, there is still 40~180 um thick FeO. The relationship between rolling temperature, oxide scale thickness and chroma a* is shown in Figure 4.
Considering the deformation characteristics of FeO, it is necessary to adopt effective descaling and high temperature finishing rolling process for low Si steel. For high-Si steel, the descaling after slab heating is more important, and a higher descaling pressure is required, and the descaling temperature must be kept above the melting point of Fe2SiO4 at 1173°C. For high Si (0.5%) steel, the slab can use a heating temperature lower than 1173 ℃ (such as 1150 ℃) to completely suppress the liquefaction of Fe2SiO4, or use an extremely high heating temperature (such as 1300 ℃) to ensure instant Fe2SiO4 descaling Still maintain a liquid state, can reduce the red scale area of the final product. But obviously, the effect of high temperature heating regime decreases with the prolongation of heating time. As shown in Figure 5, when the heating time of high Si steel at 1300 °C is extended from 60 min to 120 min, the area of red scale increases from 15% to 60%, which is difficult to meet the requirements of actual production for surface quality.
It is worth noting that this study still emphasizes that if the slab is heated in a stainless steel heat preservation cover to ensure that no primary scale is formed, but only secondary scale is prefabricated, no matter what the Si content is, red scale defects will not occur. This is because the temperature of the prefabricated secondary oxide scale in the experiment is much lower than 1 173 C, and the formed secondary oxide scale can be easily completely removed after descaling.
Twenty years later (2015), Hikaru Okada continued to study in depth the conditions for the deformation of scale and red scale defects during the rolling process of strip steel. The study takes 0.05%C-0.15%Mn low-carbon steel as the object, uses N2 or stainless steel box as a protection method to heat the slab, and then prefabricates the original oxide scale with different thickness by staying on the roller table for different times, and The deformation behavior of the oxide scale during the subsequent rolling process was investigated.
The research shows that when the iron oxide scale is thin before rolling, the iron oxide scale after rolling is smooth and the thickness is basically uniform. However, when the iron oxide scale reaches 10 μm or more before rolling, the iron oxide scale will produce cracks that can be distinguished by naked eyes after rolling. The thicker the iron sheet before rolling and the higher the rolling temperature, the easier it is to produce such cracks on the oxide sheet after rolling, as shown in Figure 6. It is worth mentioning that the compression deformation of scales with different thicknesses is basically the same as that of the matrix.
In fact, the properties and interface state of the oxide scale after rolling deformation have always been the difficulty of research. Yuan Kenichiro et al. used the method of spraying glass powder on the surface of the deformed steel plate to protect the scale and interface state, and then studied the relationship between the rolling deformation and the deformation of the scale. The study takes low-carbon steel as the object. The experimental steel plate is rolled at 1000°C on a two-roll experimental rolling mill, and glass powder is quickly sprinkled for protection after rolling. The schematic diagram of the experiment is shown in Figure 7. The slab heating process is protected by Ar gas, and the rolling deformation is 10% to 40%. The research shows that when the rolling deformation is less than 20%, the deformation of the scale is relatively uniform. If the deformation is higher than 30%, the iron oxide scale will be broken, and the base steel will be extruded to the outermost layer through the cracks in the iron oxide scale, and the surface will be exposed from the cracks in the iron oxide scale, which will deteriorate the surface properties of the steel plate. It can be seen that the deformation should be controlled below 30% when the rolling temperature is low.
Further research by Yuan Kenichiro and others showed that at 900 °C and the oxidation time was 40s, the oxide film indented on the base steel, and the interface between the oxide film and the base steel appeared obvious roughness.
At 1000℃, the oxide scale breaks, and the base steel is extruded to the outermost layer through the crack. At 1100℃, cracks appear in the thickness direction of the oxide scale, but the interface between the oxide scale and steel is relatively smooth. It is believed that the thick iron oxide scale introduces relative slip between the roll and the base steel, which makes the oxide scale easy to break and reduces the shear deformation, so the rolling load is reduced.
As shown in Figure 8. The longer the oxidation time before hot rolling, the greater the thickness of the oxide scale, the worse the ductility, and the higher the probability of irregular deformation of the oxide scale.
1.2.3 Effect of slab mold flux adhesion on red scale defects
Jorge RA MIREZ-CUELLAR from Mexico and others studied the formation mechanism of primary oxide defects in CSP production lines. The study shows that the slab entering the tunnel furnace
The combination of time and temperature is the decisive factor for the formation of primary oxide scale on strip steel. When the slab is vibrating or the rolls are not moving, the possibility of primary scale on the slab increases. This conclusion is consistent with
The standard understanding is relatively consistent. After the iron scale is pressed in once, it will gradually evolve into red scale defects in the subsequent rolling production. Figure 9 shows the appearance of the primary oxide scale indentation defect after pickling, the overall shape is “boat shape”.
It is worth noting that this study explored the relationship between mold flux and the causes of primary scale defects in the furnace, and found that exogenous elements in oxides appeared in primary scale defects, such as Si and Ca, which are the main factors for the strong adhesion of the original oxides in the strip. Sprinkling mold mold slag powder on the slab at the entrance of the tunnel furnace, although it cannot completely reproduce the defect shape of the primary scale indentation, it can confirm that the Si and Ca elements come from the mold slag.
In fact, it is very common for slabs to adhere to mold powder. For ordinary hot continuous rolling slabs, after pickling, it will be found that there are a large amount of white hardening in the crystallizer shock mark, which is the mold powder adhering to the surface of the slab, as shown in Figure 10. This mold slag will indeed increase the risk of impure descaling of the primary oxide scale. However, the production practice shows that most of the mold slag adhered to the surface can be removed with the primary oxide scale generated by the furnace, and rarely cause the primary scale indentation defects of ordinary hot continuous rolling production lines.
1.3 Doubts about the formation mechanism of red scale defects and defect control
1.3.1 Doubts about the formation mechanism of red scale defects
Due to the existence of Si element, the furnace-grown primary oxide scale is difficult to be removed by high-pressure water descaling, so it gradually evolves into red scale defects in the subsequent rolling process. The theory is widely accepted.
However, in fact, there are some doubts in this theory that are difficult to explain. Taking DP590 (0. 07%C-0.45%Si) steel as the object, a small sample was prepared, placed on the normal production slab and entered into the production line heating furnace for heating (1220 ℃, 200 min, λ=1. 05~1.30 ), and then remove the sample from the surface of the returned billet (the sample of the back billet after heating is shown in Figure 11), and observe the interface morphology between the oxide scale and the substrate of the sample after water quenching.
Figure 12 shows the interface morphology between the oxide scale and the substrate of the back blank sample. It can be found that the tortoise shell network morphology formed after the liquefaction of Fe2SiO4 is obvious (Fig. 12(a)), and the anchor-like morphology is wedged into the matrix (Fig. l2(b)); this is consistent with the laboratory simulation results (Fig. 3 Show). It is worth noting that there is also a layer of fine oxide particles on the surface of the substrate, and the thickness of the inner oxide layer is about 15-25 μm.
Figure 13 shows the physical analysis results of the hot-rolled DP590 strip. There is indeed a large area of red scale defects on the surface of the strip, and the enrichment of Si element is also found at the interface between the oxide scale and the matrix. The physical analysis of the steel strip shows that no matter in the red scale area or the normal area, there is a Si-rich layer at the interface between the oxide scale and the substrate; the difference is that the thickness of the oxide scale in the red scale area is significantly greater than that in the normal area. However, comparing Figure 13(b) with Figure 12(b), we can find suspicious points, the anchor structure and the inner oxide particle layer have disappeared.
The rough rolling process of hot rolling is actually a process of peeling off the surface of the slab layer by layer. Generally, after the slab comes out of the furnace, it will go through 1 pass of initial descaling, and then enter the rough rolling mill for about 3 passes of rough descaling, that is, the surface layer of the slab is in the process of oxidation-stripping-re-oxidation-re-stripping. It can be considered that this peeling removes all the inner oxide particle layer of the surface layer. It can be clearly seen from Figure 12(b) that the depth of the oxidized particle layer is greater than the thickness of the anchor structure. If the oxidized particle layer has been removed, the anchor structure will obviously not be preserved to the final product. Then, how to determine that the red scale defect on the surface of the hot-rolled strip comes from the FeO+ Fe2SiO4 analysis structure generated by the heating furnace (the formation mechanism of the red scale defect shown in Figure 2)?
1.3.2 Control of red scale defects – banded oxide scale defects of Si-containing steel
For the cause of red scale defects in Si-containing steel, Chun-Chao Shih et al. proposed the formation mechanism of striped iron scale when studying the banded scale defects of strip steel with a Si mass fraction of about 0.2%. In this study, steels with Si mass fractions of 0.20% and 0.02% were used as the research objects, only the cooling water between the first and second racks was turned on, and the flow rates were 5%, 10% and 15%. The research results show that there is almost no striped iron scale defect under 5% water volume; there are slight stripe defects under 10% water volume; under 15% water volume, relatively obvious equidistant striped iron scale defects appear, and the distance between the striped iron oxide scale and The spacing of the finishing descaling nozzles is consistent.
The cross-sectional analysis of the oxide scale is shown in Figure 14.
The average thickness of normal oxide scale is about 4.8 μm, and the surface is smooth, as shown in Figure 14(a). The average thickness of the slightly black striped iron scale is about 8.2 μm, which is larger than the normal scale area. On the surface of the oxide scale, some parts are crushed (white dotted line box), while other parts are intact; on the whole, except that the scale/substrate interface becomes more undulating than the normal area, the black striped oxide remains well Integrity, as shown in Figure 14(b). In the cross-sectional morphology of the obviously banded black stripe oxide, the iron oxide scale on the surface of the black stripe area is completely broken into particles with a particle size of less than 2 μm. In addition to the broken part (white dotted line box), the iron oxide scale also peeled off. The oxide scale thickness in most areas of the strip defect position exceeds 11 μm, and the oxide scale thickness in some areas exceeds 20 μm. Due to the inhomogeneity of scale thickness, the distortion of the scale/substrate interface becomes severe, as shown in Fig. 14(c). All types of iron oxide scale are composed of Fe3O4, FeO and a-Fe, and do not contain Fe2O3.
TEM studies show that Fe2SiO4 is formed at the iron sheet/substrate interface. Mn behaves as a solid solution element in FeO. Therefore, Mn is distributed throughout the scale, and there is no obvious enrichment at the scale/matrix interface.
It is worth noting that the study found that even the obvious striped iron scale shown in Figure 14(c) still appears “black”, but it does not mean that this striped iron scale is not a “red scale” defect. Generally speaking, the surface of α-Fe2O3 powder steel plate with particle size below 2 μm looks red. After the iron sheet is broken, it will be transformed into Fe2O3 at an accelerated rate, which is the element that forms the “red scale” defect. The surface of the iron oxide scale shown in Figure 14(c) in this study has been completely broken into particles with a particle size of less than 2 μm, which fully meets the conditions for further oxidation to form red powder. It’s just that under the experimental conditions, the subsequent oxidation time is not very sufficient, so only broken Fe3O4 was found. In addition, studies have shown that the powder content must be greater than 15% to achieve the “red” that can be distinguished by the naked eye. In production practice, it is found that the steel strip with a Si mass fraction of 0.45% is basically dark red in this equidistant oxide scale; therefore, whether the surface of the strip is red or not is also related to the Si content in the steel.
The overlapping area of fine descaling will cause the surface of the strip to be overcooled, and the temperature drop in the overlapping area can reach tens of degrees Celsius, as shown in Figure 15(a). Theoretically, due to the lower temperature, the thickness of the newly formed scale is slightly thinner in the overlap/banding region. However, the research results of Chun-ChaoShih and other scholars are just the opposite. This is because, once the new scale is formed, due to the direct contact between the scale and the components of the rolling mill, the interaction between them has to be considered. When the strip enters the finishing stands, the temperature in the strip zone may not be fully recovered due to the strip contacting the work rolls of each stand and being cooled by others
System cooling (such as cooling water between racks), the strip surface temperature will be further reduced. Through these interactions, the temperature in the strip zone on the strip surface is likely to drop by 100-200°C. Therefore, the influence of heat transfer inside the material on the scale formation process must be considered. When the strip cools, its surface temperature can be recovered by heat flow from the base material. Figure 15(f)~(h) shows the situation of 0.02% Si steel, at this time. The surface of the strip forms a uniform thickness of oxide scale. For Si-containing steel, the compound on its surface has a great influence on heat transfer. Si has a strong affinity for O, and Fe2SiO4 must be formed at the oxide scale/matrix interface during high temperature oxidation. Studies have shown that the thermal conductivity of Fe2SiO4 is only 1/3 of that of FeO, that is, Fe2SiO4 can hinder heat transfer.
During the finish rolling process, the oxide scale on the strip surface is mainly composed of newly formed FeO. FeO can undergo plastic deformation at 700 °C, but steady deformation can only be observed above 1000 °C. The starting temperature of finish rolling is usually 1000~1050°C. If the cooling flow rate of the rolling mill is high, the surface temperature of the low temperature area caused by the descaling overlapping area will drop sharply, even down to 700°C. Since the Fe2SiO4 formed at the interface will hinder the heat transfer of the matrix, the temperature of the supercooled zone in the oxide scale may not be fully restored before the subsequent rolling; as the strip surface temperature decreases, the deformation ability of FeO decreases. When the rolling force is applied, the scale with low deformability in the low-temperature zone extends longitudinally less than the adjacent normal zone, and finally the scale with low deformability is rolled into the matrix to form pits, and the scale pits Formation leads to the difference in the thickness of the oxide scale between the overlapping area of fine descaling and the adjacent area. When the cooling flow rate of the rolling mill is large, a large number of iron sheet pits spread along the rolling direction in the overlapping area of fine descaling, thus forming black iron oxide belts, as shown in Figure 15(c)~(e), the distance between the black belts Corresponds to the distance between the overlapping areas of the descaling nozzles. Due to the low surface temperature of the strip steel, the iron oxide scale is relatively brittle, and due to the compound action of rolling force and friction force during the rolling process, part of the iron oxide scale is broken and broken, and the enlarged version of this process is the DP590 steel type. Extensive red scale defects.
The above process can also prove from another perspective that the primary iron oxide scale in the heating furnace is not the only cause of the red scale defect.
