Change of Spinel in High Ca Treament at 38CrMoAl Steel

Abstract

In this study, 38CrMoAl steel was treated with calcium under the pressure of 2 MPa, and four groups of high aluminum steels with the highest calcium content of 0.01% were obtained. Spinel inclusions with high MgO content, CaO and MgO inclusions with original spinel morphology were found by testing and analyzing the morphology towards composition of inclusions in the high Ca content steel. Combined with thermodynamic calculation, a Ca–Al spinel reaction model was proposed, the solid/liquid phase denaturation mechanism of calcium treated spinel was explored, and the influence of total oxygen content on phase equilibrium in denaturation process was analyzed. At last, it was obtained that the spinel denaturation products is MgO+Spinel, CaO–Al₂O₃–MgO and CaO–Al₂O₃ when the mass fraction of calcium was low; And the spinel denaturation products with high calcium content are MgO+Spinel, MgO and CaO.

1. Introduction

The use of calcium for microalloying in steel during smelting can be traced back to 1906 at the earliest,¹⁾ After more than 100 years of development, calcium treatment has become a relatively mature refining process for producing high-quality steel, which is mainly used to improve the purity of molten steel and reduce the amount of sulfide and Al₂O₃ inclusions in steel.²⁾ After a century of research and development, researchers have a clearer understanding of the changes of inclusions caused by Ca treatment in molten steel.

Among them, the most widely spread are³⁾ based on the binary phase diagram of CaO–Al₂O₃, the transition sequence of calcium treated Al₂O₃ inclusions is: Al₂O₃→CA₆→CA₂→CA→CA_x(liq), where C represents CaO and A represents Al₂O₃; and the CaS inclusions formed by the reaction of calcium and S in steel⁴⁾ towards Verma⁵⁾ Studied the viewpoint of calcium modification of magnesium aluminum spinel verified in laboratory. The modification of Al₂O₃ inclusions by calcium treatment has been clearly explained and demonstrated, on the contrary, the specific changes of spinel inclusions with Al₂O₃ inclusions during calcium treatment are controversial.

Scholars^6,7,8) believe that calcium treatment can denature spinel into liquid inclusions, The mechanism of action is that MgO component in spinel is preferentially reduced to dissolved Mg and enters molten steel, while the residual Al₂O₃ component in spinel will react with CaO obtained by reaction to form liquid inclusions. Three reaction paths as shown in Fig. 1⁹⁾ are proposed, and finally different types of inclusions will be formed. However, Hou¹⁰⁾ found that the ratio of MgO/Al₂O₃ in spinel changed from 0.3 before calcium treatment to 0.5 after calcium treatment in studying the influence of calcium treatment on inclusions in aluminum deoxidized steel. Similarly, Pretorius¹¹⁾ found in industrial experiments that the percentage of MgO components in spinel also increased during a period of time after calcium treatment. Therefore, the above theory can not to explain the causes of this phenomenon well, and the specific reaction pathway of calcium treatment on spinel denaturation process is still doubtful.

Fig. 1.

Possible mechanisms for the modification of MgO·Al₂O₃ inclusions by Ca treatment.⁹⁾ (Online version in color.)

Due to the limitation of addition method and smelting level, most of the current research^12,13) are studied in the range of low calcium mass fraction and small calcium composition in steel. However, how calcium affects stable spinel inclusions in high Mg and high Al environment in steel, and whether the increase of MgO component mass fraction in spinel inclusions in previous studies will continue to exist and change further, and what is the specific reason for its occurrence. The further research and demonstration are needed.

In order to solve the above problems, 38CrMoAl steel was treated with Fe-10%Ca alloy in different amount of addition under higher pressure, and finally the steel with larger range and higher calcium mass fraction was obtained. In this study, the composition of steel, morphology and composition change of inclusions in steel were analyzed theoretically from the perspective of thermodynamics. The denaturation paths of spinel inclusions in the range of higher Ca mass fraction and the causes of its formation were analyzed. The intermediate reaction of spinel inclusion denaturation reaction was proposed, and the specific reason for the increase of MgO component in spinel (MgO·Al₂O₃) inclusions during calcium treatment was found. It fills the blank in the field of spinel inclusion modification in steel with a large range of Ca mass fraction, and clarifies the specific reaction path of spinel in calcium treatment process.

2. Experimental

In this experiment, 38CrMoAl steel was smelted by different degrees of calcium treatment in a 25 kg pressurized induction furnace under a pressure of 2 MPa pressured by Ar(the pressure of 2 MPa is aims to get high Ca content in steel). The MgO crucible is used in smelting, and the steel composition ranges from C: 0.35–0.42. Si: 0.20–0.45; Mn: 0.30–0.60; Cr: 1.35–1.65; Mo: 0.15–0.25; Al: 0.70–1.10. In order to keep higher mass fraction and larger range of calcium in the steel, we smelted it in four furnaces, 1# is the original steel, 2# is added Fe-10%Ca to the original steel with 0.9% mass fraction of steel, 3# is added with 1.8% mass fraction of steel, and 4# is added with 3.6% mass fraction of steel.

Four furnaces of steel are the same smelting process: Industrial pure iron, chromium and molybdenum with known components are placed in MgO crucible, and high purity graphite block, industrial silicon, manganese, aluminium particle and Fe-10%Ca are placed in silo, then vacuumized to less than 7 Pa, and heated by electricity. After the raw materials are melted, kept filling Ar until the pressure arrived 2 MPa, then graphite, industrial silicon, manganese and aluminium particles are added in turn, and Fe-10%Ca blocks prepared by 30-ton 769YP-30T powder prototype machine are added after holding for 5 min. Keep the temperature for 5 min and then cast (1# also keep the same time), ensure that the pressure in the furnace is always 2 MPa within 20 min after casting. After the ingot is naturally cooled to a lower temperature, released the pressure and then sampled.

The mass fractions of Ca, P and S were determined by ICAP6300ICP-OES analyzer, N and O were determined by LecoTC500 N/O nitrogen and oxygen analyzer, and the rest were determined by ICP-prodigyXP direct reading spectrometer. After polishing the surface of the sample, ZEISS-EVO18 scanning electron microscope was used to analyze the composition distribution, morphology and element content of inclusions in the sample. Finally, Factsage 8.1 is used to carry out thermodynamic calculation and analysis under specific steel composition.

3. Results and Discussion

3.1. Change of Steel Composition

The composition of each element in the steel is obtained by testing each experimental steel as shown in Table 1. In order to compare the change trend of some elements more intuitively, the change of T.O, S and Mg mass fractions in the steel composition with [% Ca] was drawn as a line chart as shown in Fig. 2.

Table 1. Composition analysis of experimental steel (Mass fraction/%).

No	MPa	C	Si	Mn	Cr	Mo	Al	P	S	T.O	N	Ca	Mg
1	2.0	0.40	0.43	0.44	1.57	0.22	1.00	0.0091	0.0037	0.0009	0.0020	0.0003	0.0006
2	2.0	0.40	0.40	0.45	1.58	0.22	0.99	0.0088	0.0014	0.0008	0.0020	0.0034	0.0030
3	2.0	0.40	0.40	0.45	1.57	0.22	1.01	0.0086	0.0012	0.0006	0.0024	0.0071	0.0058
4	2.0	0.41	0.41	0.46	1.58	0.22	0.99	0.0067	0.0008	0.0004	0.0020	0.0100	0.0079

Fig. 2.

The mass fraction of T.O, S and Mg in steel varies with [%Ca]. (Online version in color.)

It can be clearly seen from Table 1 and Fig. 2 that Ca treatment under pressure can increase the mass fraction of Ca in steel to 0.01% at the highest. Moreover, with the increase of Ca content in steel, the content of S and T.O decreased, especially the content of S decreased most obviously, the lowest content of T.O decreased to 0.0004%, and the lowest content of S decreased to 0.0008%, which indicated that calcium had a good removal effect on O and S. However, the mass fraction of Mg increases proportionally with the increase of Ca, up to 0.0079%. It should be due to the reaction between Ca dissolved in steel and MgO crucible, which leads to the dissolution of Mg into molten steel.

3.2. Morphology and Composition Changes of Inclusions

In this study, SEM and EDS were used to analyze the inclusions of four experimental steels, and the oxide composition distribution in inclusions at 1873 K was marked in CaO–Al₂O₃–MgO ternary stable phase diagram, as shown in Fig. 3. The morphology and composition of typical inclusions are shown in Fig. 4.

Fig. 3.

Composition distribution of inclusions in experimental steel. (Online version in color.)

Fig. 4.

Morphology and composition changes of spinel and Al₂O₃ inclusions in 38CrMoAl steel during calcium treatment. (Online version in color.)

As can be seen from Fig. 3, in the 1# steel without calcium treatment, there is no calcium in the composition of steel, and the inclusions are also concentrated in Al₂O₃ to MgO, which is mainly pure Al₂O₃. The morphology and specific composition of the main inclusions are shown in 1# of Fig. 4 is spinel or Al₂O₃ inclusions with typical angular shape. However, in the 2# steel after partial calcium treatment, the inclusions in the steel are scattered in various regions of the phase diagram, including not only CaO–Al₂O₃–MgO inclusions, but also CaO–Al₂O₃ inclusions with different compositions, with little or no spinel composition. The morphology and specific composition of the main inclusions are shown in 2# in Fig. 4. Calcium reacts with spinel and Al₂O₃ to form liquid inclusions with higher CaO content. At the same time, the main inclusions have basically lost the typical angular spinel characteristics and become spherical-like inclusions.

It can be seen from Fig. 3 that the inclusion composition of 3# steel tends to be the higher MgO contained spinel, and a large number of purer CaO inclusions towards small amount of MgO inclusions appear at the same time. The morphology and specific composition of the main inclusions are shown in 3# in Fig. 4. The morphology of typical CaO–Al₂O₃–MgO inclusions is show in Fig. 5. At the same time, the mass fraction of MgO components in the inclusions is much higher than the original MgO mass fraction of spinel, and it has the original shape characteristics of angular spinel, but contains a small amount of residual CaO. Compared with 2# steel which content lower calcium, the mass fraction of CaO components in spinel inclusions of 3# steel is much lower than that of 2# steel, while the mass fraction of MgO components is much higher than that of 2# steel.

Fig. 5.

Typical inclusion morphology in steel after calcium treatment. (Online version in color.)

In the 4# steel with the highest content of Ca, the inclusions are concentrated in pure CaO, MgO and high MgO spinel, and there are no liquid inclusions. A large number of inclusions are pure CaO and MgO is less, while high MgO spinel also has a large number of inclusions. It is worth noting that most high MgO spinel does not contain CaO. The morphology and specific composition of the main inclusions are shown in 4# in Fig. 4 and MgO–Al₂O₃ in Fig. 5, the high MgO spinel does not contain Ca element instead of much higher Mg content. Compared with the mass fraction of MgO in inclusions of 3# steel, the mass fraction of MgO in 4# steel has been further increased, indicating that when the mass fraction of Ca reaches 0.01%, the spinel is modified by calcium to spinel with high MgO content, that is MgO+spinel. Similarly, a large number of pure MgO and CaO inclusions with a certain angular shape (indicated by arrows in the Fig. 4) appeared in the steel, among which CaO inclusions were the majority, indicating that they were denatured from spinel. The morphology of typical pure MgO, pure CaO and MgO+CaO inclusions is show in Fig. 5, The clear phase interface between them confirms that Ca will react with MgO and the products of the two reactions can not eutectoid, and will exist as independent CaO inclusions and MgO inclusions.

The quantity change of various types of inclusions in the experimental steel under different calcium treatment systems is shown in Fig. 6(a). When treated with a small amount of calcium, the proportion of Al₂O₃ inclusions and spinel inclusions with low MgO content decreased greatly, which indicated that they were transformed into CaO–Al₂O₃ inclusions, MgO–CaO–Al₂O₃ inclusions and a large number of CaO inclusions. When the mass fraction of calcium increased, the proportion of CaO, CaO–Al₂O₃ and MgO–CaO–Al₂O₃ inclusions decreased gradually and tended to disappear, while the proportion of spinel with high MgO content increased gradually. When the mass fraction of calcium was 0.071%, the proportion of pure MgO reached its peak, and then there was no obvious change.

Fig. 6.

Variation of (a) the number and percentage of inclusions and (b) the average composition of spinel inclusions with different calcium content. (Online version in color.)

The change of average composition in spinel inclusions with different calcium treatment systems is shown in Fig. 6(b). When treated with a small amount of calcium, the components of Al₂O₃, MgO and CaO decreased greatly. However, when the mass fraction of calcium increased to 0.0071%, MgO component and CaO component increased simultaneously, but the increase of CaO component slowed down. However, when the mass fraction of calcium in steel is further increased, the proportion of Al₂O₃ components remains unchanged, and the proportion of MgO components keeps increasing while the proportion of CaO components decreases.

In general, when the mass fraction of calcium in steel is less than 0.0034%, calcium treatment can effectively denature spinel into liquid inclusions containing calcium, and it can also reduce Al₂O₃ components and MgO components in spinel. The denaturation process follows the route shown in Fig. 1. However, when the mass fraction of calcium exceeds this range, the composition of MgO components in spinel will gradually increase, and its quantity ratio will further increase. Calcium will react with MgO and Al₂O₃ in spinel in the low composition range (0–0.0034%), but if the mass fraction of calcium in steel is further increased, calcium will only react with Al₂O₃, thus consuming Al₂O₃ components in spinel, resulting in an increase in the proportion of MgO components. Therefore, when considering the reaction pathway of modified spinel treated by calcium, we should not only consider the magnesium reaction between calcium and spinel, but also comprehensively consider the comprehensive reaction between calcium with MgO and Al₂O₃.

3.3. Thermodynamic Calculation and Analysis

It can be seen from the above analysis that the comprehensive reaction of MgO and Al₂O₃ should be considered for the denaturation of spinel by calcium, and the main equations involved are as follows.   

$$\mathrm{C}\mathrm{a}{\mathrm{O}}_{(\mathrm{s})}=\left[\mathrm{C}\mathrm{a}\right]+\left[\mathrm{O}\right]\hspace{1em}\hspace{1em}\hspace{1em}{K}_1={\displaystyle \frac{a_{\text{Ca}}\cdot a_{\text{O}}}{a_{{\text{CaO}}_{\text{(s)}}}}}$$(1)

  

$$\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{s})}=2\left[\mathrm{A}\mathrm{l}\right]+3\left[\mathrm{O}\right]\hspace{1em}\hspace{1em}\hspace{1em}{K}_2={\displaystyle \frac{a_{\text{Al}}^{\text{2}}\cdot a_O^{\text{3}}}{a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}}}$$(2)

  

$$\mathrm{M}\mathrm{g}\mathrm{O}\cdot \mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{s})}=\mathrm{M}\mathrm{g}{\mathrm{O}}_{(\mathrm{s})}+\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{s})}\hspace{1em}\hspace{1em}{K}_3={\displaystyle \frac{a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}\cdot a_{{\text{MgO}}_{\text{(s)}}}}{a_{\text{MgO}\cdot {\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}}}$$(3)

Where [M] denotes M element dissolved in steel, K₁, K₂ and K₃ are the equilibrium constants of Eqs. (1), (2) and (3) when the reaction reaches equilibrium, respectively. The thermodynamic data of the above reaction formula can be obtained by searching reference^{14,15,16,17,18,19,20,21,22,23,24)} as shown in Table 2.

Table 2. Equilibrium constants of reactions in relation to ca treatment.

Reactions	ΔGθ(J/mole)	logK	Ref.
Al2O3(s) = 2[Al] + 3[O]	1225000 − 393.8 T	20.57 − 64000/T	14)
	867370 − 222.5 T	11.62 − 45300/T	15)
	907580 − 235.9 T	12.32 − 47400/T	16)
CaO(s) = [Ca] + [O]	645200 − 148.7 T	7.76 − 33700/T	17)
	138240 + 63.0 T	−3.29 − 7220/T	18)
	370820	−10.34	19)
	297660	−8.3	20)
	264310	−7.37	21)
MgO·Al2O3(s) = MgO(s) + Al2O3(s)	20790 + 15.7 T	−0.82 − 1086/T	22)
	18828 + 6.3 T	−0.32 − 980/T	23)
	15734 + 13.42 T	−0.70 − 822/T	24)

However, considering that oxides do not only appear or react in the form of solid phase during calcium treatment, this basic reaction cannot be simply used for analysis. If we want to analyze the specific reaction between spinel and calcium, we need to clarify the main products formed by calcium and Al₂O₃ under the condition of certain calcium mass fraction, and then we can carry out subsequent analysis on this basis. Therefore, the subsequent analysis will be divided into two reaction processes: Al₂O₃ and spinel.

3.3.1. Reaction Equilibrium of Ca with Al₂O₃

The composite oxides formed by CaO and Al₂O₃ should be considered more during Ca treatment. The physical and chemical properties of different composite oxides including melting point are different. It can be seen intuitively from the binary phase diagram of CaO–Al₂O₃, which is shown in Fig. 7 (calculated by Factsage 8.1).

Fig. 7.

Binary phase diagram of CaO–Al₂O₃ (calculated by Factsage 8.1) C: CaO; A: Al₂O₃.

When the activity of oxides in the Eqs. (1) and (2) is replaced by the hypothetical pure liquid state as the standard state, the equilibrium constant becomes.   

$$K_4={\displaystyle \frac{a_{Ca}\cdot a_{\text{O}}}{a_{Ca{O}_{\text{(l)}}}}},\hspace{1em}\mathrm{S}\mathrm{o}\mathrm{\hspace{0.17em} }\mathrm{t}\mathrm{h}\mathrm{a}\mathrm{t}\mathrm{\hspace{0.17em} }{K}_1/K_4=a_{\mathrm{C}\mathrm{a}\mathrm{O}(\mathrm{l})}/a_{\mathrm{C}\mathrm{a}\mathrm{O}(\mathrm{s})}$$(4)

  

$$K_5={\displaystyle \frac{a_{\text{Al}}^{\text{2}}\cdot a_O^{\text{3}}}{a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}}},\hspace{1em}\mathrm{S}\mathrm{o}\mathrm{\hspace{0.17em} }\mathrm{t}\mathrm{h}\mathrm{a}\mathrm{t}\mathrm{\hspace{0.17em} }{K}_2/K_5=a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}/a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}$$(5)

Since the activity of solid phase in this standard state can be regarded as 1, the activity ratio in this standard state can be obtained only by calculating the activities of Al₂O_3(l) and CaO_(l) in liquid phase in molten steel at 1873 K. Considering that the mass fraction change of [O] and [Ca] in steel may affect its activity, the activity of Al₂O_3(l) and CaO_(l) in inclusion in 1873 K and 1823 K molten steel were calculated by Equlib module of Factsage 8.1. The result is shown in Fig. 8. Choose the Select FSstel, FToxid and FTsalt databases for calculation (Subsequent calculations use the same database without special explanation), all the elements composition used in the calculation are from the original 1# steel in Table 1 except Ca and O. In order to avoid the mutual influence of Ca and O, Ca in Fig. 8(a) is 0.0003% and O in Fig. 8(b) is 0.0004%, all of them are the lowest values involved in this study.

Fig. 8.

Activity of Al₂O₃ and CaO in slag varies with mass fraction of O (a) and Ca (b). (Online version in color.)

It can be seen from Fig. 8 that the activity changes of Al₂O_3(l) and CaO_(l) are limited at 1873 K and 1823 K, and the change of calcium mass fraction only has certain influence on them at lower temperatures. However, the change of oxygen content in molten steel has obvious influence on the activities of Al₂O_3(l) and CaO_(l), Therefore, the influence of Ca mass fraction and temperature on oxide activity can not be considered in the calculation, but the influence of oxygen mass fraction on activity should be fully considered, and different oxygen contents should be used for calculation. The first set of data is taken as an example for calculation.

The activity ratio of CaO and Al₂O₃ in standard state at 1873 K is   

$$a_{\mathrm{C}\mathrm{a}\mathrm{O}(\mathrm{l})}/a_{\mathrm{C}\mathrm{a}\mathrm{O}(\mathrm{s})}=0.0019;\hspace{1em}\hspace{1em}{a}_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}/a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}=0.2030$$

Substituting into Eqs. (4) and (5), can get   

$$K_4=K_1/0.0019\hspace{1em}\hspace{1em}{K}_5=K_2/0.2030$$

Take into account that following reaction occur during Ca treatment   

$$3\left[\mathrm{C}\mathrm{a}\right]+\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{l})}=3\mathrm{C}\mathrm{a}{\mathrm{O}}_{(\mathrm{l})}+2\left[\mathrm{A}\mathrm{l}\right]\hspace{1em}\hspace{1em}{K}_6={\displaystyle \frac{a_{\text{Al}}^{\text{2}}\cdot a_{{\text{CaO}}_{\text{(l)}}}^{\text{3}}}{a_{\text{Ca}}^{\text{3}}\cdot a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}}}$$(6)

At 1873 K, simultaneous Eqs. (4), (5) and (6), it can be deduced that   

$$K_6={\displaystyle \frac{a_{\text{Al}}^{\text{2}}\cdot a_{{\text{CaO}}_{\text{(l)}}}^{\text{3}}}{a_{\text{Ca}}^{\text{3}}\cdot a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}}}=K_4/K_{\text{5}}^{\text{3}}$$(7)

According to the binary phase diagram of CaO–Al₂O₃, the main products at higher CaO are C₁₂A₇, C₃A and CaO. Table 3 shows the activities of CaO and Al₂O₃ in the literature^25,26,27) and calculated by Factsage 8.1 in this study.

Table 3. Activities of CaO and Al₂O₃ in various types of calciumaluminates at 1873 K.

Type	mass%/CaO	a(Al2O3)	a(CaO)	Ref.
C12A7	50	0.0502	0.024	Present work
		0.027	0.53	25)
		0.05	0.45	26)
C3A	62	0.0205	0.173	Present work
		0.0087	1	27)
		0.012	0.6	30)
		0.011	0.81	12)
CaO	99.29	0.0204	1	Present work
		0.01	1	26)
		0.005	1	30)
		0.0065	1	12)
		0.0076	1	27)

That is to say, in order to form composite oxides of the above type, the activity of oxides in slag equivalents must meet the values shown in Table 3. By substituting this value into Eq. (7), a_Ca:a_Al can be obtained when different oxides are formed. The activity interaction coefficient can be calculated by the following equation   

$$\mathrm{l}\mathrm{g}{f}_{\mathrm{i}}=\sum e_i^j\left[\text{mass\%}j\right]$$(8)

Table 4 shows the activity interaction coefficients between different elements and the elements used in the calculation of this study.

Table 4. First-order interaction coefficients.^28,29)

j i	Al	Si	Mn	O	P	S	C	Cr	Mo	Ca
Al	0.045	0.0056	0.012	−6.6	0.05	0.03	0.091	0.025	/	−0.047
O	−3.9	−0.131	−0.021	−0.2	0.07	−0.133	−0.45	−0.04	0.0035	−271
Ca	−0.072	−0.097	0.0156	−780	/	−125	−0.34	0.02	/	−0.002

Therefore, the activity interaction coefficients of Ca and Al with different mass fractions of O can be calculated as shown in Table 5.

Table 5. Activity coefficients of elements in liquid steel at 1873 K.

Mass fraction of O/%	0.0004	0.0006	0.0008	0.0009
fCa	0.193	0.126	0.084	0.044
fAl	1.334	1.333	1.330	1.329

It can be seen from the calculation results in Table 5 that calcium treatment will make the oxygen and sulfur contents in steel change obviously, and then lead to the change of activity interaction coefficient, so it will have a great influence on the final calculation results. Considering the difference between dissolved calcium [Ca%]_e and dissolved aluminum [Al%]_e in steel and total calcium [Ca%]_T and total aluminum [Al%]_T in steel, this study assumes that [Ca%]_e = 0.7[Ca%]_T(The [Ca%]_T Measured in this study will also convert to [Ca%]_e according to this ratio), [Al%]_e = 0.9[Al%]_T. By synthesizing Eqs. (1), (2), (4), (5) and (7), the relationship between calcium and aluminum for C₃A and CaO products with different oxygen contents can be obtained as shown in Fig. 9.

Fig. 9.

Changes of CaO (a) and C₃A (b) under different oxygen contents at 1873 K.

It can be seen from Fig. 9 that the change curves of CaO and C₃A in steel before and after calcium treatment are quite different. When [Al%]_e in steel is 0.9, the required [Ca%]_e for complete denaturation to C₃A composite oxide has reached 0.016, while if it is completely denatured to CaO, it has reached an astonishing 0.077. However, with the decrease of oxygen and sulfur content in steel, the [Al%]_e required for the same [Al%]_e also decreases greatly. Therefore, we draw the oxide dominant region diagrams for two different conditions with T.O mass fractions of 9 × 10⁻⁶ and 8 × 10⁻⁶ at 1873 K, as shown in Fig. 10.

Fig. 10.

Relationship in Al and Ca for oxides formation under different T.O: (a) 9 × 10⁻⁶ and (b) 8 × 10⁻⁶.

It can be seen from Fig. 10 that if the influence of the change of oxygen and sulfur in steel is not considered, the calculation deviation will make the calculation result far deviate from reality. Therefore, in Fig. 10(b) after fully considering all factors, 2# steel is in the region where CaO and C₃A are produced together, while 3# and 4# steel are in the region where only CaO is produced. However, the inclusions are located between 2# steel and 3# steel, so it can be preliminarily speculated that the reaction between calcium and spinel is limited by whether liquid phase is formed in the product or intermediate product.

3.3.2. Variability of Spinel Inclusions during Calcium Treatment

Through the ternary stable phase diagram of CaO–Al₂O₃–MgO at 1873 K, the main oxides of CaO–Al₂O₃–MgO at different ratios can be seen intuitively, and its ternary stable phase diagram is shown in Fig. 11 (calculated by Factsage 8.1).

Fig. 11.

CaO–Al₂O₃–MgO ternary stable phase diagram (calculated by Factsage 8.1) C: CaO; A: Al₂O₃; Sp: Spinel.

It can be seen from Fig. 11 that at 1873 K, the common liquid oxides are concentrated on the CaO–Al₂O₃, so MgO can be regarded as solid in the reaction process, and the reaction can be concentrated on the reaction between the formed liquid phases. Equation (3) can be changed to   

$$\mathrm{M}\mathrm{g}\mathrm{O}\cdot \mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{s})}=\mathrm{M}\mathrm{g}{\mathrm{O}}_{(\mathrm{s})}+\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{l})}\hspace{1em}\hspace{1em}{K}_9={\displaystyle \frac{a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}\cdot a_{{\text{MgO}}_{\text{(s)}}}}{a_{\text{MgO}\cdot {\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}}}$$(9)

At the same time, there are   

$$K_9/K_3=a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}/a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}$$(10)

Is known, and simultaneous Eqs. (9) and (6) can obtain   

$$\begin{array}{l}3\left[\mathrm{C}\mathrm{a}\right]+2\mathrm{M}\mathrm{g}\mathrm{O}\cdot \mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{s})}=2\mathrm{M}\mathrm{g}{\mathrm{O}}_{(\mathrm{s})}+\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_{3(\mathrm{l})}+3\mathrm{C}\mathrm{a}{\mathrm{O}}_{(\mathrm{l})}+2\left[\mathrm{A}\mathrm{l}\right]\\ K_{10}={\left(K_9\right)}^2\cdot K_6={\displaystyle \frac{a_{{\text{MgO}}_{\text{(s)}}}^{\text{2}}\cdot a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}\cdot a_{\text{Al}}^{\text{2}}\cdot a_{{\text{CaO}}_{\text{(l)}}}^{\text{3}}}{a_{\text{MgO}\cdot {\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}\cdot a_{\text{Ca}}^{\text{3}}}}\end{array}$$(11)

Where $a_{\text{MgO}\cdot {\text{Al}}_{\text{2}}{\text{O}}_{\text{3(s)}}}$ and $a_{{\text{MgO}}_{\text{(s)}}}$ can be regarded as 1 in this standard state, while a_CaO(l) and $a_{{\text{Al}}_{\text{2}}{\text{O}}_{\text{3(l)}}}$ can be determined by forming different oxides from Table 3, from which a_Ca:a_Al can be calculated. Finally, the relationship between calcium-aluminum ratio and formed oxides at different T.O. is obtained, and the relationship diagram is drawn as shown in Fig. 12.

Fig. 12.

Relationship in Al and Ca for oxides formation under different T.O: (a) 9 × 10⁻⁶ and (b) 8 × 10⁻⁶.

It can be seen from Fig. 12 that with the change of oxygen and sulfur content, the calcium-aluminum ratio composition range of 38CrMoAl passes through the liquid phase region and liquid-solid miscible phase region, and finally falls into the complete solid phase region. The composition of 2# steel in Fig. 12(b) is located in the solid-liquid miscible region, calcium and spinel can transfer mass and react through the wrapped liquid phase formed on the surface, so there are more CaO–Al₂O₃–MgO and CaO–Al₂O₃ inclusions in the steel. 3# and higher Ca content 4# in Fig. 12(b) are located in the pure solid phase region, calcium and spinel can only react on the solid surface of inclusions, so CaO–Al₂O₃–MgO and CaO–Al₂O₃ inclusions in steel decrease and tend to disappear, while MgO and MgO+Spinel inclusions increase. Figure 13 shows the comparison between the calculation results of this study and the literature. It can be seen that although the solid-liquid miscible zone in the literature is small, which leads to the 2 # steel with low calcium entering the complete liquid phase zone, the 3 # steel and the 4 # steel with higher calcium are still located or close to the pure solid phase zone. This further confirms the viewpoint of this study.

Fig. 13.

Comparison between this study and different literature data.^12,30) (Online version in color.)

In addition, considering that the increase of Mg in steel will lead to the formation of MgO, which will contribute to the increase of MgO proportion in spinel. Previous studies³¹⁾ have shown that the increase of Mg will lead to the decrease of MgO formation temperature, and the lowest temperature is between 1550°C to 1600°C, so the correlation calculation for MgO is selected as 1550°C. Factsage was used to calculate the stable phase diagram of spinel in Fe–C–Si–Mn–Cr–Mo–Al–Mg–O system at 1550°C (considering the complex inclusion composition after Ca addition, no calcium was added in the calculation, as shown in Fig. 14(a)) and CaO–MgO in Fe–C–Si–Mn–Cr–Mo–Al–Mg–Ca–O system (Fig. 14(b)). It can be seen from Fig. 14(a) that with the decrease of T.O in steel, the formation areas of MgO and spinel decrease. Starting from the composition of 2 # steel, the composition of subsequent steels is located near the phase interface of MgO, indicating that there is a small possibility that MgO will be formed in steel. However, Fig. 14(b) shows that the change of T.O in steel has little effect on CaO phase region, and MgO phase region will decrease with the decrease of T.O. Under the condition of Ca and Mg components involved in this study, CaO is easier to form than MgO. Based on the above analysis, it is considered that the formation of MgO is extremely difficult under the condition of high calcium, so the increase of Mg mass fraction in steel is not the main reason for the formation of MgO+Spinel. The MgO found in Fig. 5 should also come from Ca and spinel reaction, rather than natural formation due to the increase of Mg.

Fig. 14.

Stable phase diagram of Spinel (a) and CaO–MgO (b) at 1550°C. (Online version in color.)

Based on the above analysis, three denaturation paths of spinel at low calcium and high calcium can be obtained, as shown in Fig. 15: in the low composition range, Ca reacts with spinel to form liquid calcium aluminate, and the formed liquid phase agglomerates around spinel, carries out mass transfer and reaction with molten steel and spinel, resulting in a decrease in the proportion of Al₂O₃ components, and the products may be MgO+spinel, CaO–Al₂O₃–MgO and CaO–Al₂O₃; However, when Ca content is high, that is, it is located in the completely solid CaO phase region, it can not form liquid phase agglomeration around inclusions, and calcium can only react directly with Al₂O₃ components in spinel, which leads to the decrease of Al₂O₃ components in spinel, while the increase of MgO components in spinel. The final reaction products are MgO+spinel and MgO or CaO with the original spinel morphology.

Fig. 15.

Denaturation mechanism of spinel inclusions during 38CrMoAl calcium treatment. (Online version in color.)

The denaturation process of spinel inclusions in high calcium steel is similar to the denaturation path of spinel in low calcium steel as shown in Fig. 1. Route 3 is the product of incomplete reaction of Al₂O₃ in spinel. Route 1 is the product only MgO remains in the original spinel when Al₂O₃ in spinel is completely reacted by calcium in steel, then MgO will continue to react with Ca to become pure CaO (CaO+MgO in Fig. 5 is the intermediate state of the reaction between the them) or remain in steel; CaO in Route 2 comes from two sources: one is the direct product of spinel reaction, and another is the reaction product in Ca and MgO (by-product of spinel).

4. Conclusion

In this study, the spinel inclusions with high Al and Mg content are easy to form under different degrees of calcium treatment. Combined with experimental results and theoretical analysis, the following conclusions are finally obtained.

(1) The content of Ca in 38CrMoAl steel can reach 0.01% under high pressure, and the content of O and S in steel can be effectively reduced. At the same time, Ca will react with MgO crucible, which will cause Mg to enter the molten steel and promote the formation of spinel.

(2) The reaction path of calcium-denatured spinel is the replacement of calcium with Al in spinel. When the ratio of Ca/Al is in the liquid phase formation zone, liquid inclusions are formed around spinel, and the liquid zone serves as the intermediate layer of mass transfer and reaction; When the ratio of Ca/Al is in the pure solid phase formation region, the reaction proceeds directly on the solid surface of spinel.

(3) The activity of Al₂O_3(l) and CaO_(l) in liquid inclusions will be greatly affected by the mass fraction of [T.O] in steel, and the activity interaction coefficient of Ca will also be changed. When the oxygen content is low, only a low Ca/Al ratio is needed to enter the pure solid phase formation zone from the liquid phase formation zone, which makes it more difficult for spinel to denature into liquid inclusions.

(4) At low mass fraction, calcium can react with spinel to form encapsulated inclusions containing calcium aluminate, and the products may be MgO+spinel, CaO–Al₂O₃–MgO and CaO–Al₂O₃. At high mass fraction, calcium can react with spinel to form MgO+spinel, MgO and CaO which has residual original spinel morphology.

Acknowledge

The authors are grateful for the financial support from National Natural Science Foundation of China (Grant Nos. 51434004/U1435205/52074075).

References

• 1)   O. P.  Watts: J. Am. Chem. Soc., 28 (1906), 1152. https://doi.org/10.1021/ja01975a004
• 2)   Z.  Deng and  M.  Zhu: Steel Res. Int., 84 (2013), 519. https://doi.org/10.1002/srin.201200250
• 3)   M.  Imagumbai and  T.  Takeda: ISIJ Int., 34 (1994), 574. https://doi.org/10.2355/isijinternational.34.574
• 4)   N.  Verma,  P. C.  Pistorius,  R. J.  Fruehan,  M.  Potter,  M.  Lind and  S. R.  Story: Metall. Mater. Trans. B, 42 (2011), 720. https://doi.org/10.1007/s11663-011-9517-2
• 5)   S. F.  Yang,  J. S.  Li,  Z. F.  Wang,  J.  Li and  L.  Lin: Int. J. Miner. Metall. Mater., 18 (2011), 18. https://doi.org/10.1007/s12613-011-0394-0
• 6)   M.  Jiang,  X.  Wang,  B.  Chen and  W.  Wang: ISIJ Int., 48 (2008), 885. https://doi.org/10.2355/isijinternational.48.885
• 7)   M.  Jiang,  X.  Wang,  B.  Chen and  W.  Wang: ISIJ Int., 50 (2010), 95. https://doi.org/10.2355/isijinternational.50.95
• 8)   Z.  Li,  C.  Kato and  Y.  Kobayashi: ISIJ Int., 61 (2021), 2340. https://doi.org/10.2355/isijinternational.ISIJINT-2021-082
• 9)   S.  Yang,  Q.  Wang,  L.  Zhang,  J.  Li and  K.  Peaslee: Metall. Mater. Trans. B, 43 (2012), 731. https://doi.org/10.1007/s11663-012-9663-1
• 10)   Z. W.  Hou,  M.  Jiang,  E. J.  Yang,  S. Y.  Gao and  X. H.  Wang: Metall. Mater. Trans. B, 49 (2018), 3056. https://doi.org/10.1007/s11663-018-1365-x
• 11)   N.  Verma,  P. C.  Pistorius,  R. J.  Fruehan,  M. S.  Potter,  H. G.  Oltmann and  E. B.  Pretorius: Metall. Mater. Trans. B, 43 (2012), 830. https://doi.org/10.1007/s11663-012-9660-4
• 12)   G.  Ye,  P.  Jönsson and  T.  Lund: ISIJ Int., 36 (1996), Suppl., S105. https://doi.org/10.2355/isijinternational.36.Suppl_S105
• 13)   G.  Xu,  Z. H.  Jiang and  Y.  Li: Metall. Mater. Trans. B, 47 (2016), 2411. https://doi.org/10.1007/s11663-016-0695-9
• 14)   G.  Li,  T.  Hamano and  F.  Tsukihashi: ISIJ Int., 45 (2005), 12. https://doi.org/10.2355/isijinternational.45.12
• 15)   H.  Itoh,  M.  Hino and  S.  Ban-ya: Tetsu-to-Hagané, 83 (1997), 773 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.83.12_773
• 16)   J. D.  Seo,  S. H.  Kim and  K. R.  Lee: Steel Res., 69 (1998), 49. https://doi.org/10.1002/srin.199801342
• 17)   G. K.  Sigworth and  J. F.  Elliott: Met. Sci., 8 (1974), 298. https://doi.org/10.1179/msc.1974.8.1.298
• 18)   H.  Itoh,  M.  Hino and  S.  Ban-ya: Tetsu-to-Hagané, 84 (1998), 85 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.84.2_85
• 19)   E. T.  Turkdogan: Steel Res., 62 (1991), 379. https://doi.org/10.1002/srin.199101314
• 20)   Q.  Han,  X.  Zhang,  D.  Chen and  P.  Wang: Metall. Trans. B, 19 (1988), 617. https://doi.org/10.1007/BF02659153
• 21)   T.  Arato,  T.  Uchida and  Y.  Omori: Trans. Iron Steel Inst. Jpn., 25 (1985), 326. https://doi.org/10.2355/isijinternational1966.25.326
• 22)   K.  Fujii,  T.  Nagasaka and  M.  Hino: ISIJ Int., 40 (2000), 1059. https://doi.org/10.2355/isijinternational.40.1059
• 23)   R. H.  Rein and  J.  Chipman: Trans. Metall. Soc. AIME, 233 (1965), 415.
• 24)   O.  Knacke,  O.  Kubaschewski and  K.  Hesselmann: Thermochemical Properties of Inorganic Substances, Springer, Berlin, (1991), 1136. https://doi.org/10.1007/978-3-662-02293-1
• 25)   K.  Mineura,  I.  Takahashi and  K.  Tanaka: ISIJ Int., 30 (1990), 192. https://doi.org/10.2355/isijinternational.30.192
• 26)   G.  Jing,  C.  Shu-Sen and  C.  Zi-Jian: ISIJ Int., 53 (2013), 2142. https://doi.org/10.2355/isijinternational.53.2142
• 27)   H. Y.  Zheng,  S. Q.  Guo,  M. R.  Qiao,  L. B.  Qin,  X. J.  Zou and  Z. M.  Ren: Adv. Manuf., 7 (2019), 438. https://doi.org/10.1007/s40436-019-00266-1
• 28)   J. X.  Chen: Data Manual of Common Charts in Steelmaking, Metallurgical Industry Press, Beijing, (1987), 114 (in Chinese). https://doi.org/10.13374/j.issn1001-053x.1987.s3.184
• 29)   J. H.  Park and  H.  Todoroki: ISIJ Int., 50 (2010), 1333. https://doi.org/10.2355/isijinternational.50.1333
• 30)   B.  Koroušić: Steel Res., 62 (1991), 285. https://doi.org/10.1002/srin.199101296
• 31)   W. C.  Jiao,  H. B.  Li,  H.  Feng,  Z. H.  Jiang,  H. J.  Wang,  H. C.  Zhu,  P. D.  Han and  W.  Wu: Metall. Mater. Trans. B, 53 (2022), 1196. https://doi.org/10.1007/s11663-022-02436-2