Smelting high purity 55SiCr automobile suspension spring steel with different refractories

2.1 Experimental apparatus and procedure

Table 2 lists the target composition of 55SiCr spring steel and the amount of added elements. C, Si, Mn, Cr, Ni, and V were added into molten steel during the steelmaking process, while [Al]_s, P, and S were brought in by industrial pure iron and alloy used in the steelmaking. The chemical compositions of the selected crucibles, Al₂O₃, MgO·Al₂O₃, and MgO–CaO, are listed in Table 3. The corresponding experimental heated samples were abbreviated as Cru-A, Cru-MA, and Cru-MC. All crucibles were pressed using hydraulic press. In detail, the oxide powers and binder were mixed together and then pressed into a mold with a pressure of 20 MPa. After that, the rough crucibles were sintered with an inert atmosphere at 1,783 K (1,500°C) for 10,800 s. It should be noted that MgO–CaO refractory should be packed with vacuum plastic bags to prevent it from absorbing moisture in the air and becoming invalid. The section diagram of experimental equipment is shown in Figure 1.

Figure 1

Section diagram of experimental equipment.

The experimental procedure was implemented in detail as follows: industrial pure iron was placed into the selected crucible with a graphite crucible on the outer side. Afterward, both crucibles were put together in the MoSi₂ furnace and heated slowly. When the furnace temperature reached 673 K, cooling was provided with circulating water. In addition, high purity argon (99.999 vol%) was introduced into the furnace at a rate of 3–5 NL·min⁻¹ when the temperature increased to 873 K. The automatic temperature control program was started, and the molten steel sample was held at 1,873 K for at least 10 min. Then, high purity oxygen was introduced into molten steel at a 2 NL·min⁻¹ rate and sustained for 10 s. Next, high purity silicon wafers were added into molten steel as a deoxidizer, and C, Mn, Cr, Ni, and V alloying elements were added stepwise. The molten steel was stirred using a graphite bar to provide uniform mixing after the addition of each alloying element. The time when the alloying elements were added is set as the initial time and the molten steel is held at 1,873 K for 5,400 s to ensure its full interaction with the crucible. Finally, the experimental steel samples were cooled down with the furnace to room temperature with the protecting argon atmosphere and cooling water.

2.2 Analysis methods

The testing methods and equipment used to detect experimental steel inclusions and determine their chemical composition were described in our previous article in detail [28].

In addition, the electrolysis of a small amount of sample was done to extract inclusions from the steel samples. The experimental device and flowchart are shown in Figure 2.

Figure 2

Schematic diagram of the experimental device for electrolytic inclusions: (a) steel samples for electrolysis; (b) electrolytic device.

The examination of scanning electron microscope-energy-dispersive X-ray spectrometric (SEM-EDS) inclusions was performed after the following sample preparation: (1) powder samples containing inclusions extracted using electrolysis were dispersed on a white paper and carefully stirred with a knife to avoid accumulation. Then, the powder sample was taken from the white paper with conductive adhesive and gently pressed. (2) The sample was blown with a blowing pump (3X550-1100; Shangma Industrial Zone, Shitang Town, Wenling City) and put into the ion-sputtering instrument (JS-1600, maximum vacuum ≤4 × 10⁻² mbar; Beijing Hetong venture Technology Co., Ltd.) for gold sputtering. (3) The samples were examined further using SEM-EDS.

3.1 Mass fraction of elements in experimental steel samples

The mass percentages of each element in the experimental steel are shown in Table 4. The crucible material exhibits a significant influence on the amount of each element in steel, having the most apparent effect on C, Si, Mn, Ca, T.O, P, S, and [Al]_s.

In Cru-A, the Si content is significantly lower than the added amount, while the [Al]_s fraction is in the highest position among all of the heated samples. In Cru-MA and Cru-MC, the contents of all of the elements coincide perfectly with targeted values. In particular, the contents of P, S, [Al]_s, and T.O in Cru-MC steel samples are sharply reduced to 56 × 10⁻⁶, 10 × 10⁻⁶, ≤0.0005 × 10⁻⁶, and 4 × 10⁻⁶, respectively, although the content of Ca increased to 19 × 10⁻⁶.

3.2 Inclusions in steel samples

3.2.3 The composition of inclusions

The statistical results of the composition of inclusions in each experimental steel are shown in Figure 3. In Cru-A, the center of composition distribution of inclusions is located in the middle of the SiO₂-(MnO + MgO + CaO)-Al₂O₃ ternary phase diagram. On the other hand, there are some inclusions with a high Al₂O₃ content (higher than 50 wt%) or even some pure Al₂O₃ inclusions. The results of Cru-MA are similar to those of Cru-A. The only exception is that the fraction of inclusions with a high Al₂O₃ content (higher than 50 wt%) becomes lower. In Cru-MC, the content of Al₂O₃ in most of the inclusions was lower than 10 wt%.

Figure 3

Composition of inclusions in experimental steel samples.

3.2.4 The results of two-dimensional (2D) detection of inclusions

The 2D detection results of typical inclusions in each sample are shown in Figure 4. In Cru-A and Cru-MA, the main types of inclusions were multiphase composite oxide inclusions with an MnS layer wrapped around the edge. The size of inclusions is larger than 5 μm. The morphology of inclusions was irregular, as shown in Figure 4(a–f). In Cru-MC, most of the inclusions have a smaller size and spherical shape as shown in Figure 4(g–i).

Figure 4

Typical inclusions in steel samples: (a–c) Cru-A, (d–f) Cru-MA, and (g–i) Cru-MC.

In particular, the size and morphology of MnS inclusions in Cru-MC are quite different from those in all the other experimental heated samples. Particularly, in Cru-MC, MnS inclusions are nearly spherical with a diameter of less than 2 × 10⁻⁶ m since the S content of these steel samples is only 10 × 10⁻⁶. By contrast, the diameter of MnS inclusions in other heated samples is larger than 10 × 10⁻⁶ m, and some of them appear as long strips (Figure 4(g–i)).

The three-dimensional morphology of inclusions is shown in Figure 5.

Figure 5

Typical inclusions in steel samples obtained using electrolysis: (a–c) Cru-A, (d–f) Cru-MA, and (g–i) Cru-MC.

3.3 Refractories

3.3.1 Al₂O₃ crucible

The micro-morphology of the Al₂O₃ crucible is shown in Figure 6. There was no apparent reaction layer on the inner surface of the Al₂O₃ crucible. However, the EDS analysis results at points A, B, and C of the crucible indicate that the permeable layer of refractory is enriched in Si, Ca, Mn, and C elements, as shown in Tables 7 and Figure 7.

Figure 6

SEM picture of cross section at 1/2 height of Al₂O₃ refractory.

Table 7

Element mass fraction in selected point at crucible (wt%)

Point	O	Al	Si	Ca	Mn	C
A	51.42	42.62	5.96
B	38.92	13.18	23.61	5.35	6.77	12.17
C	43.96	13.77	13.77	4.09	8.56	10.36

Table 8

Composition of crucible (wt%) at selected points

Point	O	Mg	Fe	Si	Al	Ca
A	57.30	17.94	—	3.23	18.76	2.78
B	39.40	23.67	—	17.30	1.05	18.58
C	—	—	100	—	—	—

Figure 7

Element mapping results of Al₂O₃ crucible.

3.3.2 MgO·Al₂O₃ crucible

The micro-morphology of the MgO·Al₂O₃ refractory is shown in Figure 8. There was no apparent reaction layer on the inner surface of MgO·Al₂O₃. Furthermore, the EDS results at points A and B of the crucible indicate that the crucible's inner surface is enriched with Si, Ca, Mg, and Fe elements, as shown in Table 8. In addition, the white spherical particles at point C consist of pure iron beads (Table 8). The elemental mappings of the crucible as shown in Figure 9 distinctly represent these observations.

Figure 8

SEM picture of cross section at 1/2 height of MgO·Al₂O₃ refractory.

Figure 9

Element mapping results of MgO·Al₂O₃ crucible.

3.3.3 MgO–CaO refractory

The micro-morphology of the MgO–CaO refractory is shown in Figure 10. The reaction layer cannot be found on the MgO–CaO refractory. The mass fractions of elements at points A and B are listed in Table 9 in detail. At point B, the mass fractions of both Mg and Ca are smaller than those at point A, while it is the opposite for the contents of Al and Si (Figure 11).

Figure 10

SEM picture of cross section at 1/2 height of MgO–CaO refractory.

Table 9

Composition of crucible (wt%) at selected points

Point	O	Mg	Al	Si	Ca
A	41.98	10.97	—	2.50	44.54
B	54.47	4.53	21.12	14.56	5.33

Figure 11

Element mapping results of MgO–CaO crucible.

3.4 Phase analysis of MgO·Al₂O₃ refractories

The phase composition of refractory after smelting was analyzed using mineral automatic analyzer (MLA). In the test process, the closely connected mineral phases can be distinguished using SEM and EDS: through intensive management, the component information can be collected, and the mineral phases can be distinguished by different components. Then, the energy spectrum of the mineral phase collected using EDS was compared with the spectrum of the database, and the type of mineral phase can be determined. The results of the MLA test and Factsage calculation are compared and verified with each other, to determine the interaction mechanism between liquid steel and refractory.

MgO·Al₂O₃ crucible is selected for testing in this experiment, and the results are shown in Figure 12 and Table 10. MgO is the main phase of the MgO·Al₂O₃ crucible. In detail, the mass percentages of the MgO and MgO_ phase are 47.12 and 7.38%, respectively; the Spinel phase is also a very important component of the MgO·Al₂O₃ crucible, containing 26.49%. The white bright spot at the boundary in the backscattered electron image is detected as Fe and infiltrated by liquid steel. This shows that liquid steel will penetrate into the refractory through pores during the process of steelmaking.

Figure 12

(a) MgO·Al₂O₃ crucible backscattered electron image; (b) MgO·Al₂O₃ crucible mineral phase diagram.

4.1 Self-decomposition of the refractories

In the present research, the main oxides that compose the crucibles were Al₂O₃, MgO + Al₂O₃ + MgO·Al₂O₃, and MgO + CaO. The dissolution reaction equations of oxides of these refractories are listed in equations (1)–(4) and can be expressed by the general equation (5). It could be speculated whether these reactions can take place by combining the activities of elements in liquid steel and ΔG = ΔG ^Θ + RT lnJ. In addition, the a [ M ] x ∗ a [ O ] y values of Al₂O₃, MgO·Al₂O₃, MgO, and CaO reckoned by equation (6) are 4.41 × 10⁻¹⁴, 4.76 × 10⁻¹⁵, 1.52 × 10⁻⁶, and 7.53 × 10⁻¹¹, respectively. The activities and activity coefficients can be calculated by lg f i = ∑ j e i j [ % j ] and a i = f i [ % i ] , severally. The interaction coefficients are listed in detail in Table 11. For MgO and CaO, the values of a [ M ] x ∗ a [ O ] y are higher than those for Al₂O₃ and MgO·Al₂O₃. Therefore, the possibilities for MgO and CaO decomposition are much higher than those for Al₂O₃ and MgO·Al₂O₃ at 1,873 K.

Furthermore, when the temperature of 1,873 K is selected, the ΔG values in equations (1)–(4) are 428.76, −7.63, 65.92, and 158.94 kJ·mol⁻¹, respectively. Thus, only the ΔG ₍₂₎ values are below zero, so the MgO·Al₂O₃ crucible may self-decompose.

4.2 The stability of crucible materials

Factsage 7.2 software was selected to predict the temperature influence on the change of phase composition of the refractories, as shown in Figure 13.

Figure 13

Phase composition of refractories: (a) Al₂O₃ crucible, (b) MgO·Al₂O₃ crucible, and (c) and (d) MgO·CaO crucible.

In Cru-A, the refractory stability is excellent due to the absence of phase transitions in the entire temperature range. In Cru-MA, three phases exist when the temperature reaches 1,873 K, that is, slag phase, MgO, and MgO·Al₂O₃. With the temperature increase by degrees, the MgO·Al₂O₃ mass fraction slowly decreases, while the transitions of slag and MgO phases behave differently. In Cru-M, there are MgO and MO (M stands for a metal element) phases at 1,873 K, indicating quite good stability of MgO refractory. In Cru-MC, there are slag, MgO, and CaO phases at 1,873 K, and the mass fraction of the slag phase is only ∼1.5 wt%.

4.3 The influence of Al₂O₃ refractory

There are four possible reactions between liquid steel and crucible material as listed in equations (7)–(10) in Cru-A. It could be speculated whether these reactions can take place by combining the activities of elements in liquid steel and ΔG = ΔG ^Θ + RT ln J. The activities and activity coefficients can be calculated using lg f i = ∑ j e i j [ % j ] and a i = f i [ % i ] , respectively. The interaction coefficients are listed in detail in Table 11. Equation (7) was bound to occur at steelmaking temperature (1,873 K), causing a higher Ca mass fraction in the refractory than at other positions, as shown in Table 7. Equation (8) can also occur due to ΔG = −345.64 kJ·mol⁻¹ < 0. Therefore, the Si content in experimental steel samples decreases sharply from 1.6 wt% (the added amount) to 1.47 wt% (the final content). Finally, neither equation (9) nor (10) can occur since their ΔG values are 162.37 and 138.45 kJ·mol⁻¹, respectively. This result indicates that the chemical stability of the Al₂O₃ crucible is very good. Similar behavior was reported in refs [34–37].

To sum up, the mass fraction of [Al]_s quickly rises to 68 × 10⁻⁶ as the reactions between [Ca] and [Si] in liquid steel and the Al₂O₃ refractory proceed. Furthermore, the Al₂O₃ content in inclusions is much higher than in other heated samples, and even some pure Al₂O₃ inclusions can be found, as shown in Figure 4 [30,31,32,33,38].

4.4 The influence of MgO·Al₂O₃ refractory

The main reactions between liquid steel and refractory in Cru-MA are shown in equations (11) and (12). The P mass fraction in final steel samples is smaller than that in Cru-A according to equation (11). In addition, the mass fraction of [Al]_s will slightly rise according to equation (12) [39,40].

4.5 The influence of MgO–CaO refractory

In Cru-MC, CaO can also play the role of a desulfurizer, and it can remove Al₂O₃ inclusions according to equations (13)–(18). The contents of S, Al, and T.O reduce quickly as these reactions proceed. Finally, the quantity and equivalent diameter of inclusions significantly decrease, as well as the Al₂O₃ content in inclusions [41,42,43,].