2021 Volume 61 Issue 3 Pages 865-870
The heat absorption method (HAM) was proposed to improve the quality of a large steel ingot. The CaO–CaF2 inorganic material rods was used to reduce 5 K superheat of the 6-ton GCr15 molten steel. The quality of the steel ingots with and without the HAM was compared. In comparison with conventional casting, the HAM not only significantly alleviated the A- and V-segregation and the segregation levels of carbon and sulfur, but also reduced the number of inclusions and the shrinkage porosity zone in the 6-ton steel ingot. The simulation results demonstrated that the molten steel could quickly be cooled from within by the inorganic material rods, which is the main reason for the reduction of macrosegregation. Additionally, the majority of inclusions could be absorbed and removed when the liquid inorganic material floated up in the molten steel.
The large steel ingots have been widely used in heavy industries such as power, aerospace, metallurgy and heavy machinery. Nevertheless, the various solidification defects, such as macrosegregation1,2,3,4) and shrinkage porosity,5) always restrict the production of the large steel ingots. It has been realized that natural convection is one of the main reasons for the formation of macrosegregation, such as A-segregation (channel segregation)6,7) and V-segregation8,9) in large steel ingots. To alleviate macrosegregation, various methods have been proposed such as composition control,10) electromagnetic stirring,11) multi-pouring,12) etc. Nevertheless, these methods cannot soundly solve segregation problems during practical production of large steel ingots. Additionally, the shrinkage porosity and the large inclusions frequently appear in large steel ingots, both of which degrade performance of steel ingots as well. Essentially, the formation of macosegregation originates from the relative motion of solute-rich or poor liquid and solid phases during slow solidification.13) The critical thermal conditions for the formation of A-segregation showed that the increase of the solidification rate is a crucial factor for alleviating macrosegregation.6) This means that the enhancement of cooling can reduce the flow time and thus macrosegregation. Unfortunately, there has been no effective method to raise the cooling rate of large steel ingots. Following the above idea, one of attempts is that the solid steel balls were added to the molten steel to enhance its cooling.14) It was demonstrated that this method could reduce macrosegregation and refine microstructures. However, the steel balls are difficult to guarantee complete melting and thus probably form new kinds of severe defects in an ingot. Up to now, how to manufacture a high-quality large steel ingot still is a great challenge.
In this work, a heat absorption method (HAM) is proposed to improve the quality of the large steel ingot. In principle, the inorganic material is added to the molten steel and will absorb the latent heat during its melting. The liquid inorganic material then moves up to the surface of the molten steel owing to a lower density in comparison with the molten steel. This quickly dissipates heat in the interior of the molten steel and accelerates its solidification. The experimental results showed that the HAM could significantly reduce the extent of macrosegregation, shrinkage porosity and the inclusion index in the 6-ton steel ingot.
The GCr15 bearing steel was used and its chemical composition was C 0.99, Si 0.24, Mn 0.34, S 0.001, P 0.006, Cr 1.49, Ni 0.01, Mo 0.01, Al 0.031, Ti 0.006 and Fe in balance (in wt%). The 6-ton steel ingots were cast in the steel plant of Jiangsu Yonggang Group CO. LTD., China. The dimension of the cast-iron mold and the ingot is shown in Fig. 1. The 50-ton molten steel was poured into a group of 8 molds from the bottom. One of these ingots was used for comparison. In a symmetrical mold, the HAM was employed. Four rods of the inorganic material were fixed in the mold (Fig. 1). The rods with size of Φ60 mm × 300 mm were fabricated by solidifying the mixture of 1:4 CaO–CaF2 (molar ratio), whose liquidus temperature is 1693 K.15) Based on the latent heat of melting, the amount of the inorganic material was determined to reduce 5 K superheat of the 6-ton molten steel. The pouring temperature of the molten steel was 1783 K. The mold filling time was 840 s. After the inorganic material rods were completely melted in the molten steel, the support rods were removed. It took 913 s from the beginning of pouring to the removal of the rods.
Characteristics of 6-ton steel ingot for the HAM (left) and positions for analysis of inclusions and C/S (right). (Online version in color.)
The steel ingots were sectioned along the vertical centerline. After milling, the shrinkage porosity of the ingots was observed using a high-resolution camera. Further, the halves of the ingots, which were etched for 180 s at 333 K using 50%HCl solution, was used to observe macrosegregation. From the other halves, a series of samples, which were drilled along longitudinal and transversal directions of the ingots at the interval of 50 mm, were used for chemical analysis (Fig. 1). Distributions of carbon and sulfur in the ingots were measured using a C/S analyzer (Leco CS844). The species, size and content of inclusions in the samples with dimension of 10 mm × 10 mm × 5 mm were automatically analyzed using the ASPEX Explorer. For inclusion analysis, the following operating parameters were used: the accelerated voltage was 20 KeV, the dwelling time was 1 s, the minimum inclusion size was 1 μm and the scanned area was 5 mm × 5 mm.
A series of simulations were performed to compare the temperature fields in the large steel ingots with and without the HAM using the commercial software ProCAST 2018. Some assumptions for simulation are made: (1) the mold filling process is not included. (2) The molten steel is free convection. (3) Flotation of the liquid inorganic material is not considered. The initial conditions for simulation are listed as follows: the initial temperatures of the molten steel, the mold and the inorganic material rods are 1783 K, 333 K and 333 K, respectively. The thermal boundary conditions are given below: the interfacial heat transfer coefficient is 2000 W/m2·K between the molten steel and the mold, 10 W/m2·K between the mold and air, 550 W/m2·K between the molten steel and the rod and 2.5 W/m2·K between powder and air.
Figure 2 compares the macrostructures in two 6-ton GCr15 ingots with and without the HAM. For conventional casting, the ingot exhibits severe V-segregation and A-segregation (Fig. 2(a)), both of which can frequently been observed in large steel ingots.8) When the HAM is used, the extent of macrosegregation is significantly reduced (Fig. 2(b)).
Comparison of the etched sections of the 6-ton GCr15 ingots with and without the HAM, (a) reference ingot, (b) ingot with HAM. (Online version in color.)
Further, the carbon and sulfur segregation along the centerline and at different heights are compared (Fig. 3). For carbon segregation, irrespective of applying the HAM, a general law can be observed,16,17) i.e. the positive segregation at the top and the negative segregation at the bottom of the ingots (Fig. 3(a)). However, one can see that there is an obvious fluctuation of centerline carbon concentration in the reference ingot, several times from positive to negative segregation, especially in the intermediate part of the ingot. In comparison, the centerline carbon concentration fluctuates much less when the HAM is applied. The change in centerline sulfur segregation is similar with that of carbon. The sulfur segregation range and the concentration fluctuation are also significantly reduced (Fig. 3(b)). For transversal segregation, Fig. 3(c) shows a distinct decrease in maximum carbon concentration and an increase in minimum concentration in the case of application of the HAM. The segregation range is reduced from 0.94–1.23 to 0.98–1.05. The sulfur segregation also shows less fluctuation along the transversal direction when applying the HAM (Fig. 3(d)). It follows that the application of the HAM can effectively reduce the extent of macrosegregation.
Profiles of carbon and sulfur contents in the ingots, (a, b) along centerlines on longitudinal sections, (c, d) at 1.5 m height on transversal section. (Online version in color.)
Furthermore, the amount and size of inclusions are compared in the two conditions, as displayed in Fig. 4. Two obvious features can be seen, i.e. the number of the large size inclusions (≥10 μm) decreases and the inclusion index, which is defined as the ratio of the inclusion area to the measured area, decreases when the HAM is used. In the case of applying the HAM, the number of all size inclusions is obviously less at the height of 1 m (Figs. 4(a) and 4(b)). At the other heights, the amount of the large size inclusions is also reduced though the number of small size inclusions increases (Figs. 4(c), 4(d), 4(e), and 4(f)). Additionally, the inclusion index is distinctly reduced by the HAM in comparison with the conventional casting. For example, the inclusion indexes at the 1/2R position of the heights of 1 m, 1.4 m and 1.8 m are decreased from 2.808%, 1.069% and 0.534% with the HAM to 0.255%, 0.145% and 0.297% under conventional casting conditions, respectively.
Distribution and size of inclusions at 1/2 radius positions of different heights in the two ingots without (a, c, e) and with (b, d, f) the HAM, (a, b) 1 m, (c, d) 1.4 m, (e, f) 1.8 m. (Online version in color.)
The shrinkage porosity and cavities were also observed. In reference ingot, there is a porosity region with length of about 80 cm in the center and the maximum size of the shrinkage cavity is about 15 mm. In contrast, the porosity region in the ingot with HAM disappears and the maximum size of the isolated shrinkage cavities is reduced to about 4 mm, as shown in Fig. 5.
Shrinkage porosity zones of the reference ingot and the ingot with HAM, (a) Reference ingot, (b) Ingot with HAM. (Online version in color.)
The above experimental results demonstrate that the HAM can significantly alleviate macrosegregation and reduce the number of inclusions and the shrinkage porosity zone. It is well known that the solidification rate of the large steel ingot is extremely slow owing to a large volume. Long time thermal and solute convection leads to severe macrosegregation.9) In this work, the inorganic material absorb much heat in the molten steel to be liquid and further move up to the surface owing to a lower density (the densities of the molten liquid and the inorganic material at 1773 K are 6.995 × 103 kg/m3 and 2.562 × 103 kg/m3,18) respectively). This process enhances the cooling of the molten steel from within and accelerates solidification. Figure 6 compares the temperature fields at different times with and without the HAM. Obviously, when the HAM is used, the molten steel near the rods are more rapidly cooled at the initial stage and the cooling of the internal molten steel is enhanced. As solidification advances, the profiles of the temperature fields with and without the HAM gradually tend to be consistent since the thermal boundaries in the both cases become nearly the same. It can be imagined that the cooling action of the inorganic material rods increases the viscosity of the molten steel and thus reduces the time of convection which is beneficial to alleviate macrosegregation (Figs. 2 and 3). Additionally, based on the recent findings that the occurrence of the channel segregation was driven by the inclusion flotation,19) it is reasonable to speculate that the decrease in the number of inclusions necessarily reduces the amount of the segregation channels.
Predicted temperature fields at different times with (upper) and without (lower) the HAM. (Online version in color.)
It is worth noting that the HAM affects the macrosegregation level and the amount of inclusions in the whole ingot, not only at the local region near inorganic material rods. This is because that the convection of the molten steel leads to transport of heat and solutes as well as inclusions. On the one hand, although the inorganic material rods cooled the local region in the molten steel, the action of convection and thermal conduction leads to a fall in temperature of the whole molten steel. On the other hand, the liquid inorganic material attracts the inclusions along the ways the droplets of the inorganic material float up. Owing to convection, the inclusions from the other regions are possible to meet the liquid inorganic material. Therefore, the whole molten steel can be cleaned. That’s why the amount of inclusions can also be reduced in the region below the inorganic material rods when the HAM is applied (Fig. 4).
In comparison with the technique employed by Li et al.,14) the advantage of the HAM in this work is explicit. In their technique, the un-melted steel balls in the steel ingot form new inclusions, which is an acute problem for large steel ingots. However, the liquid inorganic material can easily move up to the surface of the molten steel owing to a much lower density. That is to say, the HAM does not lead to the formation of new inclusions. Inversely, the liquid inorganic material can clean the molten steel, because the liquid inorganic material can attract a great deal of inclusions in the molten steel and take them to the surface of the molten steel. This is why the inclusion index was significantly reduced by the HAM. Additionally, the heat absorption due to melting of the inorganic material can reduce the difference between the temperatures in the center and at the edge of the ingot and thus the temperature gradient. According to Niyama criterion (G/R<constant, where G and R are the temperature gradient and cooling rate, respectively),20) this cooling effect will be beneficial for elimination of shrinkage porosity, which is in good agreement with experimental observation (Fig. 5).
It should be pointed out that it is necessary for the HAM to control the amount of the inorganic material for a given steel ingot and to avoid the excessive cooling in the center of the ingot. Otherwise, if the inorganic material droplets do not move up to the surface in time, they will form new inclusions. The detailed studies concerning the float kinetics of the liquid inorganic material in the molten steel are in progress.
A new method, i.e. HAM, was proposed to improve the quality of the large steel ingot. The experimental results demonstrated that the HAM could significantly reduce the extent of macrosegregation, the amount of inclusions and the shrinkage porosity. This is because, on the one hand, the HAM significantly enhances the cooling of the molten steel from within and thus increases the solidification rate. On the other hand, the liquid inorganic material, which attract inclusions, take them to the surface of the molten steel. Thus, the technique is expected to be used to fabricate a high-quality large steel ingot.
This work was supported by National Key R&D Program of China [grant number 2019YFA0705303], National Science and Technology Major Project “Aeroengine and Gas Turbine” [grant number 2017-VII-0008-0102], Shanghai Pujiang Talents Program [grant number 18PJ1403700], Shanghai Science and Technology Committee [grant number 19DZ1100704].
