Time-Dependent Study of Inclusions in Bearing Steel Subjected to Rare Earth Treatment with Secondary Oxidation

Abstract

Due to the strong reducibility and chemical activity of rare earths, the diffusion behavior and secondary oxidation of rare earths in the steel liquid will also have a significant impact on the modified products when rare earths are added to bearing steel, resulting in poor control of distribution behavior. Therefore, this paper studies the influence of time factors on the evolution of rare earth inclusions. The inclusion evolution behavior at different times when the bearing steel was treated with rare earths and subjected to secondary oxidation was simulated at 1873 K (1600 °C). At a cerium content of 0.012% in steel and a secondary oxidation of 0.0025%, the cerium content in steel and the total oxygen (T.O.) content in steel were determined at the 30 s, 3 min, 5 min, and 7 min after the addition and the inclusions were characterized by automatic scanning electron microscopy. The results demonstrated the formation of a cerium-enriched zone after the addition of the cerium alloy to the steel. As time progressed, a considerable number of inclusions were generated in the cerium-enriched zone, which subsequently disappeared. The trend in the composition of the inclusions can be described as Al₂O₃ → Ce₂O₂S + CeS → Ce₂O₂S. The final composition of the inclusions matches the thermodynamic phase diagram. Following the addition of the transient oxidant Fe₂O₃ to the molten steel, an oxygen-enriched zone was formed. As time progressed, a considerable number of inclusions were generated in the oxygen-enriched zone and subsequently disappeared. The trend of inclusions composition was as follows: Ce₂O₃ + CeAlO₃ + Al₂O₃ → Ce₂O₃ + CeAlO₃ → Ce₂O₂S + CeAlO₃. The final inclusion composition coincides with the thermodynamic phase diagram.

1. Introduction

The utilization of rare earths to modify inclusions in steel [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20] was occasionally associated with generating heterogeneous [21,22] and unstable properties within the steel. In addition to the number of rare earth elements introduced, the diffusion behavior and secondary oxidation of rare earth [23] in the molten steel will also have a significant impact on the composition of the resulting modification products. When rare earth metals are added to the molten steel, they react with the oxygen and inclusions present in the molten steel. This reaction occurs too rapidly in the molten steel to allow for homogenization, thereby leading to the retention of the rare earth reaction products. During the actual production process, the initial pouring of molten steel from the furnace inevitably leads to secondary oxidation. It results in the formation of a significant number of oxidized inclusions, which in turn causes notable changes in their composition, quantity, and size, thereby affecting the uniformity of their distribution.

Zhang Yuexin et al. [24] examined the composition and morphology of inclusions in solid unoriented electrical steel before and after heating and analyzed the transformation of LaAlO₃ inclusions during heating. Ren Q et al. [25] investigated the transient effect of Ce on inclusions in aluminum sedum unoriented electrical steel and derived different modifications of rare earth inclusions corresponding to different Ce contents at different moments. Ren Qiang et al. [26] conducted laboratory experiments at 1873 K (1600 °C) to investigate the effect of Ce on inclusions in ultra-low carbon aluminum sedimented steel. The concentration of Ce in the steel ranged from 0 to 0.028%. The total oxygen (T.O), total nitrogen (T.N), total sulfur (T.S), total cerium (T.Ce), and dissolved aluminum ([Al]) contents were determined in steel samples at 1, 5, 10, and 30 min after the addition of Ce and the inclusions were characterized using an automated scanning electron microscope. The result shows that the composition of the inclusions exhibited a gradual transition from Al₂O₃ → CeAlO₃ → Ce₂O₂S → Ce₂O₂S + CeS as the Ce content increased. In a study by Li Bin et al. [27], the evolution of inclusions in high alumina steels was investigated following the addition of La. A series of laboratory experiments and thermodynamic calculations were conducted to consider the reaction time and the amount of La added. For La additions of less than 0.0041 wt%, the evolution pathway of inclusions in high alumina steel was Al₂O₃ → LaAl₁₁O₁₈ → LaAlO₃, with increasing reaction time. For high La additions, the evolution pathway of Al₂O₃ inclusions was as follows: Al₂O₃ → LaAl₁₁O₁₈ → LaAlO₃ → La₂O₂S → La₂S₃. Wang Yoguang et al. [28] conducted two industrial furnace tests to investigate the effect of secondary oxidation on the characteristics of rare earth aluminum sedum steel inclusions during casting. The tests employed RH, intermediate ladle, immersion spout, and slab sampling for stable casting and secondary oxidation. Consequently, the number of inclusions increased significantly and the near-spherical Ce₂O₂S inclusions were transformed into clustered CeAlO₃, which increased the average size of inclusions from 3.92 µm to 9.21 µm. Thermodynamic calculations indicated that with varying Ce additions, an increase in total oxygen content (T.O.) facilitated the transformation of Ce₂O₂S to CeAlO₃. However, there is a lack of studies on the uniformity of inclusion distribution. While most published research focuses on thermodynamic or experimental analyses, there are few studies that combine experimental data with thermodynamic analyses, especially those examining inclusions at various time points in bearing steel melts subjected to rare earth treatments and secondary oxidation processes. This study utilized thermal state experiments to simulate the modification of inclusions in bearing steel at various time points after rare earth treatment and secondary oxidation. Combined with thermodynamic calculations, the study analyzed the evolution mechanism and control conditions of inclusion modification over time. This analysis aims to offer theoretical guidance for enhancing the uniformity of inclusion distribution and the stability of steel properties.

2. Materials and Methods

2.1. Thermodynamic Simulation

The thermodynamic simulation was aimed at studying the evolution mechanism of inclusions in bearing steel after adding rare earth elements and the evolution mechanism of inclusions during the secondary oxidation process of bearing steel treated with rare earth elements. The equilibrium module of FactSage 8.2 software (FactSage is the result of over 20 years of collaborative efforts between Thermfact/CRCT (Montreal, QC, Canada) and GTT-Technologies (Aachen, Germany))was utilized alongside the FactPS, Ftoxide, and FTsteel databases. A thermodynamic model was established using pure solid and solution phases to calculate the equilibrium between the steel liquid and inclusions at 1873 K (1600 °C). Through rigorous data comparisons for Ce₂O₂S, information from the “Thermochemical data of elements and compounds” database was selected, which includes heat capacity, entropy, enthalpy change data, and corresponding calculated Gibbs free energy changes.

2.2. Vacuum Induction Furnace Experiment

The following two simulation experiments were carried out to study the formation mechanism of inclusions in rare earth-treated bearing steel.

(A) Experiments with instantaneous addition of 0.01% rare earth Ce content at an initial oxygen content of 0.001%.

(B) Based on the first experiment, Fe₂O₃ powder was used as an external O source to provide a secondary oxidation environment and increase the melt oxygen content to induce secondary oxidation. The specific addition amount was calculated according to the reaction equation (Fe₂O₃) = 2[Fe] + 3[O] and the secondary oxygen content was 0.0025%.

Experiment A: The melting process was as follows. (1) The bearing steel raw material was placed into the MgO crucible after removing the surface oil and iron oxide skin with a grinder. The rare earth ingot was polished with sandpaper, then weighed and treated with vacuum packaging to prevent oxidation. The rare earth Ce and Al grains were put into the container of the vacuum induction furnace and then prepared for melting. (2) Turn on the power supply and vacuum valve and evacuate the vacuum to 10 Pa; then fill the chamber with argon to protect the melt from oxidation. (3) Turn on the water and then the power and start to enter the melting period to ensure that the charge is fully melted. (4) After the melting period, the temperature reached 1873 K, the power supply was maintained at a stable level, and the melt was held for 10 min to ensure that a stable melt temperature was achieved. (5) After the end of the holding period, aluminum grains were added and held for 5 min, the sampling rod was lowered, and the first sample was taken using a glass tube sampler. The weighed rare earth was then added to the melt. At 30 s as well as 3, 5, and 7 min, the sampling rod was lowered to the same height and the fourth process sample was taken using a glass tube sampler. For each sampling process, the sampling tube was kept in the same orientation and depth from the bottom of the steel to ensure the same sampling position. In addition, to preserve the original composition and morphology of the inclusions in the steel and to prevent the steel from producing new precipitates during cooling and solidification, the molten steel samples were quenched in cold water immediately after removal from the molten steel.

The melting steps in experiment B were as follows: the pre-process was the same as (1), (2), (3), and (4) in experiment A. Then (5) after the end of holding, add aluminum particles, hold for 5 min, and drop the sampling rod using a glass tube sampler to take an experimental sample. Then, the weighed rare earth was added to the melt. After waiting for 10 min, the sampling rod was lowered to the same height and the first experimental sample was taken with a glass tube sampler. (6) The weighed Fe₂O₃ powder was added to the melt. At 30 s as well as 3, 5, and 7 min, the sampling rod was lowered to the same height and the fourth process sample was taken with a glass tube sampler. For each sampling, the sampling tube was held in the same direction and depth from the bottom of the steel to ensure the same sampling position. In addition, to preserve the original composition and morphology of the inclusions in the steel and to prevent the steel from producing new precipitates during cooling and solidification, the molten steel samples were quenched in cold water immediately after removal from the molten steel. The experimental setup is shown in Figure 1.

High carbon chromium-bearing steel (GCr15) from a steel mill with the composition shown in

Table 1. Chemical composition of bearing steel used in the experiment (mass fraction, %).

Heat Number	C	Si	Mn	P	S	O	Cr	Ni	Al	Ce
a	0.995	0.23	0.344	0.005	0.003	0.0010	1.418	0.018	0.014	0.012
b	0.991	0.25	0.331	0.003	0.003	0.0025	1.431	0.018	0.014	0.012

was used for the experiments. Pure aluminum was used as the deoxidizer during the experiment (A198%, Si0.6%, and Fe0.7%). The rare earth raw material was high-purity rare earth Ce with a purity of 99.98%, a melting point of 798 °C, and a boiling point of 3426 °C.

2.3. Testing Methods

The oxygen content was determined using an EMGA-820 oxygen and nitrogen analyzer, whereas the rare earth content in the bearing steel was assessed using a 5110 ICP-OES mass spectrometer. For the analysis and statistical assessment of inclusions larger than 1 μm in steel, the OPTON inclusions automatic analysis scanner was employed. The scanning field of view covered an area of 4,000,000 μm² and analysis was carried out by the GB/T 30834-2014 [29] standard, utilizing a magnification of 1000 times. The evolution of the composition, quantity, and size of the inclusions in different samples was investigated using SEM-EDS. This approach allowed observation of the morphology of the inclusions or clusters. Additionally, Image-Pro Plus 6.0 Image analysis and processing software were utilized to analyze the size of the inclusions in the captured images. These comprehensive analytical techniques offer detailed insights into variations in the composition, quantity, and size of inclusions across various samples.

3. Results and Discussion

3.1. Rare Earth Treatment of Bearing Steel

3.1.1. Thermodynamic Simulation of Rare Earth Treatments

Figure 2 illustrates the changes in thermodynamic equilibrium within the molten steel as the rare earth Ce content increases, with an initial oxygen content of 0.001% and a temperature of 1873 K. A specific experimental composition, Fe-0.995C-0.23Si-0.344Mn-1.42Cr-0.018Ni-0.25Cu-0.003S-0.005P-0.0015N-0.014Al-0.0033Ti, was chosen for simulation.

At a temperature of 1873 K, the inclusion curve in the figure shows that in the absence of rare earth additions, the inclusions in the steel were Al₂O₃. As the Ce content increased, the Al₂O₃ inclusion content decreased linearly, concurrently with the formation and gradual increase in rare earth inclusions, CeAlO₃ and CeAl₁₁O₁₈. With further increases in rare earth Ce content, CeAl₁₁O₁₈ reached its maximum at 0.0005% and then gradually decreased. Similarly, CeAlO₃ reached its maximum at 0.003% and then decreased. At a rare earth Ce content of 0.0025%, Ce₂O₂S inclusions began to form, transitioning the inclusions from CeAlO₃ to a combination of CeAlO₃ and Ce₂O₂S. At a Ce content of 0.0115%, CeS inclusions commenced, culminating in the transformation into a combination of Ce₂O₂S and CeS. The evolutionary pathway of inclusions was characterized by a transition from Al₂O₃ to CeAl₁₁O₁₈ + CeAlO₃, then to CeAlO₃ + Ce₂O₂S, followed by Ce₂O₂S, and ultimately to Ce₂O₂S + CeS.

3.1.2. Rare Earth Vacuum Induction Furnace Trials

When 120 ppm rare earth Ce was added, the sampled inclusions were analyzed and the results are shown in

Table 2. Types of inclusions at different times.

Time	0	30 s	3, 5 min	7 min
Typical inclusions	Al2O3	Ce2O2S, CeS	Ce2O2S, CeS	Ce2O2S

. At 0 min, the inclusions in the steel were mainly Al₂O₃ inclusions. The shape of the inclusions showed an irregular angular shape, black–grey color, and a large size in Figure 3, of about 1.5 μm. In the 30 s, the morphology and composition of the inclusions in the steel changed significantly and the inclusions in the steel were mainly Ce₂O₂S and CeS. After the addition of rare earth Ce, the inclusions were spherical or spheroidal and the size became fine, about 1 μm. At 3, 5, and 7 min, as shown in Figure 3, many CeS and Ce₂O₂S inclusions were formed and the fine inclusions had a high CeS content. As time increased, these two inclusions were still present in the steel but the number of CeS inclusions decreased and Ce₂O₂S dominated. It could be concluded that when rare earths were added to the steel, the modification law of localized inclusions in the steel increased with time as follows: Al₂O₃ → CeS + Ce₂O₂S → Ce₂O₂S + CeS → Ce₂O₂S, as shown in Figure 3.

3.1.3. Enrichment Zones Exist in Rare Earth Processing

It is noteworthy that, upon incorporating electron microscopy observations of intermediate samples from vacuum induction furnace experiments, it was discovered that following the addition of Ce, while the final phases remained largely consistent with the corresponding inclusions in the thermodynamic phase diagrams, a significant number of CeS inclusions appeared in the intermediate samples. This divergence resulted in a distinct phenomenon between the inclusion types observed in the process samples from the vacuum induction furnace experiments and those predicted by thermodynamic phase diagrams. Therefore, the limitations of thermodynamic calculations in predicting the type of Ce inclusions in steel can be elucidated by considering the dynamics of dissolution diffusion.

(1)

Rare earth Ce alloy dissolution and uniform distribution process

After rare earths were dispersed locally throughout the vessel, the addition of the rare earth-Ce alloy caused a significant increase in the local T.Ce content, indicating the formation of a Ce-enriched zone. Over time, the homogenization of Ce within the steel resulted in a decrease in T.Ce content within the fixed measurement region, as depicted in Figure 4. According to the Ce content curve, the peak at 30 s corresponds to the morphology and elemental distribution typical of inclusions observed in steel samples containing 0.0175% Ce, as illustrated in Figure 2. This corresponds to the emergence of CeS inclusions among the observed inclusions.

(2)

Number density, average size, and area density

The number density and average size of inclusions in steel, as depicted in Figure 5, increased after the addition of Ce and then gradually decreased with time. Based on the previous analysis, it was clear that a cerium-enriched zone formed initially after cerium addition. The average size of the inclusions subsequently increased over time.

The area density of inclusions in steel is shown in Figure 6, where the X-axis represents the horizontal direction of the photographed area, the Y-axis represents the vertical direction of the photographed area, and the Z-axis represents the area density of the inclusions. The higher the area density, the greater the number and size of inclusions and the worse the modification effect. After the addition of Ce, the area density of inclusions gradually increased with time.

3.2. Secondary Oxidation

3.2.1. Thermodynamic Simulation of Secondary Oxidation

Figure 7 shows the variation in equilibrium in Fe-0.99C-0.25Si-0.33Mn-1.43Cr-0.004 Al-0.002Ca-0.0005Mg-0.002S-0.012Ce steels with increasing total oxygen (T.O.) at 1873 K (1600 °C). The results indicated that the main inclusions are Ce₂O₂S and CeS at T.O. below 0.0016% and as the T.O. in the steel increases due to secondary oxidation, Ce₂O₂S is gradually converted to CeAlO₃.

3.2.2. Secondary Oxidation Vacuum Induction Furnace Experiments

When Fe₂O₃ was added, the sampled inclusions were analyzed and the results are shown in

Table 3. Types of inclusions at different times.

Time	30 s	3 min	5, 7 min
Typical inclusions	Ce2O3, CeAlO3, Al2O3	Ce2O3, CeAlO3	Ce2O2S, CeAlO3

. At 30, the morphology and composition of the inclusions in the steel changed significantly and the inclusions in the steel were mainly Ce₂O₃, CeAlO₃, and Al₂O₃. At 3 min, the inclusions in the steel were mainly Ce₂O₃ and CeAlO₃. The shapes of the inclusions in the steel all show obvious spherical or spherical-like shapes and the sizes were small, about 1.5 μm. At 5 and 7 min, the inclusions in the steel were mainly Ce₂O₂S and CeAlO₃ but the number of Ce₂O₂S gradually increased and the number of CeAlO₃ decreased. It can be concluded that when rare earths were added to steel, the modification law of local inclusions in steel increased with time as follows: Ce₂O₃ + CeAlO₃ + Al₂O₃ → Ce₂O₃ + CeAlO₃ → Ce₂O₂S + CeAlO₃, as shown in Figure 8.

3.2.3. Enrichment Zones Exist during Secondary Oxidation

The secondary oxidation process, similar to the rare earth treatment process for bearing steel, also manifested during the vacuum induction furnace experiments when sampling inclusions. Interestingly, discrepancies arose between the types of inclusions observed and those predicted by thermodynamic phase diagrams. While the final stage and thermodynamic phase diagrams aligned in terms of inclusion types, discrepancies emerged in the intermediate stages. This phenomenon can also be attributed to the kinetic aspects of dissolution diffusion.

(1)

Change the curve of oxygen content

Since the Fe₂O₃ powder was dispersed from the local area to the entire vessel after it was added to the steel, the T.O. content in the local area increased sharply when Fe₂O₃ powder was added to the steel, reaching a high value. Then, it gradually decreased as time elapsed. This indicated the presence of an oxygen-enriched region following the addition of Fe₂O₃. With time, the homogenization of O in the steel led to a decrease in the T.O. content in the fixed measurement region, as shown in Figure 9.

(2)

Number and area density

Initially, the number density of inclusions was extremely low. When Fe₂O₃ was added, the inclusion number density increased significantly due to the newly formed inclusions resulting from secondary oxidation. The number density rose from 1.2 to 36/mm². Three minutes after the addition of Fe₂O₃, the number density decreased to 23/mm². The decrease in the number density of inclusions was attributed to the diffusion effect within the steel, facilitating the upward removal and outward diffusion of inclusions. Over time, this led to a progressive reduction in the number density of inclusions. In the final sample, the number density of inclusions decreased to 17/mm², as depicted in Figure 10.

The variation in the area density of inclusions in steel is shown in Figure 11, which indicates that the number density and area fraction of inclusions increase in the initial stage (30 s) after the addition of Fe₂O₃ and then decrease with time. From the above analysis, it could be seen that an oxygen-enriched region was formed in the initial stage after the addition of Fe₂O₃.

(3)

Average size and parcel-type inclusions

With time, the average diameter of the inclusions initially increased and then decreased. Specifically, at 30 s after the addition of Fe₂O₃, the average diameter reached 4.6 μm, which was the maximum value. However, as time progressed, the diameter of the inclusions decreased significantly and the average size of the inclusions dropped from the maximum value to 3.52 μm, as depicted in Figure 12.

Following secondary oxidation, there were significant changes in the composition, quantity, and size distribution of the inclusions. A large number of complex inclusions of Ce₂O₂S encapsulated by CeAlO₃ were found in the samples, as shown in Figure 13. Together with the replacement of Ce₂O₂S by CeAlO₃ inclusions after secondary oxidation shown in Figure 8, these complex inclusions can be considered as an intermediate state in the transition from Ce₂O₂S inclusions to CeAlO₃ inclusions, which were formed in large quantities during the secondary oxidation process.

4. Conclusions

This study investigated the inclusion evolution behavior at various time points when simulated bearing steel was treated with rare earths and subjected to secondary oxidation. Based on the experimental results and analyses, the main conclusions are as follows:

(1)

When the bearing steel is subjected to rare earth Ce treatment, an enriched layer of Ce exists around the rare earth alloy. At a Ce content of 0.012%, the transformation path of inclusions over time is Al₂O₃ → CeS + Ce₂O₂S → Ce₂O₂S. After a certain period, the intermediate sample of inclusions in the vacuum induction furnace is finally in a state consistent with the thermodynamic phase diagram. With time, the number density increases and then decreases, the inclusion size gradually increases, and the area density gradually increases;

(2)

In transient secondary oxidation, local steel has an oxygen-rich layer. When the amount of secondary oxidation is 0.0025%, the transformation path of inclusions is Ce₂O₃ + CeAlO₃ + Al₂O₃ → Ce₂O₃ + CeAlO₃ → Ce₂O₂S + CeAlO₃. After a certain period, the intermediate samples of inclusions in the vacuum induction furnace are finally in a consistent state with the thermodynamic phase diagram. With time, the number density and size of the inclusions increase and then decrease and the area density gradually increases;

(3)

In the actual production process of bearing steel, when rare earth alloys are added for rare earth treatment, measures should be taken to reduce the influence of rare earth enrichment layers and oxygen enrichment layers.