Effect of Molybdenum Content on the Combined Effect of Boron and Molybdenum on Hardenability of Low-Carbon Boron-Added Steels

Abstract

Addition of Mo to boron-alloyed steel improves the hardenability by suppressing the precipitation of Fe₂₃(C, B)₆; this is known as the combined Mo-B effect. However, the maximum Mo content for the combined effect to occur is still unclear because previous studies on this effect mainly investigated steels with a Mo content of less than 0.80%. Therefore, in this study, 0.15% C steels containing more than 0.80% Mo were investigated to determine the maximum content required for the occurrence of the combined Mo-B effect. The combined effect increased with increasing Mo content up to 0.75%, after which it decreased. The optimum B content decreased from 12 to 11 ppm with increasing Mo content from 1.0% to 1.5%. In 1.0%Mo-20 ppm B steel and 1.5%Mo-20 ppm B steel, Mo₂FeB₂ precipitated instead of Fe₂₃(C, B)₆. Thermodynamic calculations revealed that the temperature at which Mo₂FeB₂ precipitation started increased with increasing Mo content in 20 ppm B steel. Moreover, Mo₂FeB₂ could precipitate even at a reheating temperature of 950°C. Thus, it is suggested that the maximum Mo content for the combined Mo–B effect on hardenability is determined by the precipitation of Mo₂FeB₂ mainly during reheating.

1. Introduction

During the fabrication of high-tensile steels, it is important to utilize B because the addition of this element even in small amounts (several ppm) dramatically increases the hardenability of steel, which in turn considerably decreases the alloying cost.^1,2) It is well known that the segregation of solute B atoms at the austenite grain boundaries increases the hardenability of steels;^3,4,5) however, the precipitation of borides such as BN, Fe₂₃(C, B)₆, or Fe₂B decreases the hardenability.^6,7,8,9,10) Therefore, various methods for preventing the precipitation of these borides have been intensely investigated to ensure the maximum use of solute B atoms. Al or Ti is added to fix the solute N atoms and thus prevent the precipitation of BN,^11,12) while Nb and Mo are added to prevent the precipitation of Fe₂₃(C, B)₆ and Fe₂B.^10,13,14,15) In particular, the improvement of the hardenability of B steels by the addition of Mo is called the “combined Mo-B effect”.¹⁴⁾

The features of the combined Mo-B effect in terms of the dependence of Mo and B contents have been investigated. Desalos et al.¹⁶⁾ investigated two types of Mo steels, 0.4%Mo and 0.8%Mo, and reported that the presence of Mo strongly enhances the effect of B and has a synergistic effect on both the nucleation and growth of ferrite. Ueno et al.^17,18) investigated the hardenability of B steels containing 0% to 0.8% Mo, and proposed a prediction formula for hardenability that is applicable to Mo-B combined added steels. The prediction formula indicates that the combined effect is observed for up to 0.8% Mo. In contrast, Asahi¹⁴⁾ investigated the B content dependence of the combined Mo-B effect and showed that the addition of 0.5% Mo to B steels increases the optimal B content required to increase the hardenability from 5 ppm to 13 ppm. However, these studies were mainly focused on the combined effect with Mo content below 0.8%, and the maximum Mo content for the combined effect has not yet been investigated in detail.

Another issue of interest concerning the combined Mo-B effect is the determination of the role of Mo in suppressing the precipitation of Fe₂₃(C, B)₆. There are mainly two hypotheses. Hara et al. proposed a model wherein Mo-C clusters kinetically retard the diffusion of C to the grain boundaries, thereby suppressing the precipitation of Fe₂₃(C, B)₆.¹⁵⁾ To understand the role of Mo based on the kinetic aspects, the non-equilibrium segregation behavior of B,^{19,20,21,22,23,24)} which is described as rapid segregation through the formation of a B-vacancy pair, needs to be taken into account because it has been reported that the combined Mo-B effect disappears at high austenitization temperatures over 1100°C.¹⁴⁾ In contrast, through phase calculations using the Thermo-Calc software, Han et al. demonstrated that Mo affects not only the kinetic behavior but also the thermodynamic behavior,^25,26) which indicates that the role of Mo on the hardenability of B steel is possibly ascribed to deterioration of the phase stability of Fe₂₃(C, B)₆.

In this study, we revealed the maximum Mo content for the combined Mo-B effect and clarify the role of Mo, particularly in regions with high Mo and B contents of over 0.8% and 14 ppm, respectively. We investigated the effect of Mo and B contents on the hardenability with 0% to 1.5% Mo and 0 to 25 ppm B; these concentration ranges are wider than those used in previous studies. We also observed borides in the austenite phase and estimated the precipitation temperature of borides by thermodynamic calculations to evaluate the effect of Mo on the precipitation behavior of B. The experimental and theoretical results are discussed for determining the maximum Mo content for the combined Mo-B effect.

2. Experimental Procedure

2.1. Material

The chemical compositions of the test steels used in this study are given in Table 1. Each of the steels is designated based on the Mo and B contents; for example, 0.50% Mo steel with 14 ppm B is designated as 05Mo14B. To prevent the precipitation of BN, Ti was added to all the steels to fix N as TiN. These steels were prepared in laboratory facilities, wherein they were melted in a vacuum induction furnace and casted into 50 kg ingots. The ingots were reheated at 1250°C for 3600 s and hot-rolled into 35-mm-thick plates, from which specimens were machined.

Table 1. Chemical compositions of test steels.

(mass%, *ppm)
No.	C	Si	Mn	P*	S*	Ti	Al	Mo	B*	N*	O*
(0, 10, 20)B	0.15	0.27	1.31	<20	19	0.020	0.020	0.00	0, 10, 20	7	<10
05Mo(0, 5, 10, 12, 14, 18, 20, 25)B	0.14	0.27	1.29	<20	20	0.020	0.017	0.50	0, 5, 9, 11, 14, 16, 20, 22	8	<10
07Mo(0, 20)B	0.14	0.27	1.28	<20	20	0.020	0.017	0.76	0, 20	8	<10
10Mo(0, 5, 10, 12, 14, 18, 20, 25)B	0.14	0.28	1.27	<20	20	0.020	0.017	1.01	0, 5, 10, 12, 14, 17, 20, 25	13	<10
15Mo(0, 5, 10, 12, 14, 18, 20, 25)B	0.14	0.28	1.26	<20	20	0.020	0.017	1.52	0, 4, 9, 11, 14, 16, 20, 23	18	<10

2.2. Transformation Behavior and Hardenability Tests

Transformation behaviors were investigated using a heat cycle simulator. The specimen dimensions were 3 mm in diameter and 10 mm in length. After austenitization at 950°C for 20 s, the specimens were cooled at various cooling rates from 0.01°C/s to 100°C/s, and the dilatation curves were measured during cooling to evaluate the transformation start temperature. In this study, the temperature at which the transformation rate corresponds to 10% was employed as the transformation start temperature.

The ordinary Jominy end-quench test was conducted to measure the hardenability of the steels. The dimensions of the specimens were 25 mm (diameter) ×100 mm (length). All the specimens were water quenched at one end after austenitization at 950°C for 2700 s. The hardness distribution from the quenched end was measured on two pieces of flat ground 1 mm deep by a Rockwell hardness tester. In this study, the critical cooling rate, V_c-90 (°C/s), at which the hardness corresponds to 90% martensite structure, was employed as the hardenability index.^17,18) The reported data were used as the hardness corresponding to 90% martensite structure.²⁷⁾

2.3. B Precipitate Analysis

Figure 1 shows the heat patterns of the specimens for the observation of B precipitation. The specimen dimensions were 8 mm (diameter) ×12 mm (length). The specimens were reheated at 1200°C for 600 s for solid solution treatment and then quenched using He gas. Then, the specimens were austenitized at 950°C for 20 s and quenched by water directly or by He gas after cooling to 650°C or 550°C, by He gas blowing to preserve the state of the B precipitates in the austenite phase. The cooling rates after austenitization were 30°C/s and 5°C/s at a quenching temperature of 650°C, and 0.5°C/s at a quenching temperature of 550°C. After these heat treatments, the amount of B precipitates was measured by quantitative chemical analysis of the electrolytic extraction residues. The precipitates were separated as residues from the electrolysis solution obtained after the controlled-potential electrolysis. A nuclepore membrane with a pore diameter of 0.2 μm was used for filtration. The precipitates obtained by filtering were dissolved using mixture of sulfuric and phosphoric acids, and the amount of B in the precipitates was measured by inductively coupled plasma mass spectrometry after distillation10). The distribution of B precipitates was observed by secondary ion mass spectrometry (SIMS),²⁸⁾ and the B precipitates were identified by transmission electron microscopy (TEM), which can reveal the electron beam diffraction pattern, and energy-dispersive X-ray spectroscopy (EDS). The TEM sample used in this observation was prepared by the extraction replica method after electrolytic etching.

Fig. 1.

Heat patterns of the specimens for analyzing boron precipitates.

3. Results

3.1. Effect of Mo and B on Hardenability

3.1.1. Transformation Start Temperature Test

In order to investigate the transformation behavior of the steels, we measured the 10% transformation temperature as the transformation start temperature for different cooling rates between 0.1 and 100°C/s. Figure 2 shows the transformation start temperature, and Fig. 3 shows the optical micrographs of the test steels cooled at 30, 5, and 0.5°C/s. In Fig. 2, the values of A_e3 temperature were calculated by Thermo-Calc software using the thermodynamic database, TCFE8, with 0.15% C, 1.3% Mn, 0.002% B, and 0, 0.5, and 1.5% Mo (mass percent). The values of B_s (bainite transformation start temperate) and M_s (martensite transformation start temperature) were calculated by the experimental prediction formula.²⁹⁾ Figures 2(a) and 3(a) show the results for the 0B, 10B, and 20B steels. Clearly, the addition of B to the 0B steel decreased the transformation start temperature, and the effect of B increased as the cooling rate increased. The microstructures changed upon B addition from ferrite-bainite to martensite at the cooling rate of 30°C/s and from ferrite-perlite to ferrite-bainite at the cooling rate of 5°C/s. Although the effect of B on the transformation start temperature at the cooling rate of 0.5°C/s was small and the microstructures of 0B and 10B steels were ferrite-perlite, B seemed to affect the ferrite transformation even at a low cooling rate of 0.5°C/s. The diameters of the ferrite grains of the 10B steel were larger than that of the ferrite grains of the 0B steel, which implies that B decreases the nucleation rate of ferrite when cooled at 0.5°C/s, because previous studies indicate that B does not significantly change the growth rate of ferrite.^5,30) The transformation temperature of the 20B steel was slightly lower than that of the 10B steel, with little difference in the microstructures. Figures 2(b) and 3(b) show the results for the 05Mo0B, 05Mo10B, and 05Mo20B steels. In the 05Mo0B steel, ferrite transformation at cooling rates over 5°C/s was suppressed as compared to that in the 0B steel, which implies the effect of Mo on the transformation start temperature. The addition of 10 ppm B to the 05Mo0B steel significantly suppressed not only bainite transformations at a cooling rate over 5°C/s, but also ferrite transformation below 5°C/s, unlike the case of Mo-free steel. The microstructure of the 05Mo0B steel cooled at 0.5°C/s changed from ferrite-perlite to full-bainite, implying that Mo maintains the effect of B on transformation in the low-cooling-rate region. Moreover, in contrast to the relation between 10B and 20B steels, the bainite transformation of 05Mo20B steel was promoted as compared to that of the 05Mo10B steel. Figures 2(c) and 3(c) show the results for the 15Mo0B, 15Mo10B, and 15Mo20B steels. The bainite transformation was suppressed due to the increasing Mo content from 0.5% to 1.5% in the B-free steel; that is, the experimental B_s temperature of the 05Mo0B steel was 620°C at the cooling rate of 5°C/s against the B_s temperature of 631°C, whereas that of the 15Mo0B steel was 485°C at the cooling rate of 5°C/s against the B_s temperature of 548°C. In case of 1.5% Mo steels, there was little difference between the transformation temperatures for the 15Mo10B and 15Mo20B steels at each cooling rate, similar to the case of the 10B and 20B steels. These experiments suggest that the effect of B on the transformation behavior is different for each of Mo content. Therefore, we next evaluated the hardenability of the test steels that contained various combinations of Mo and B contents by the Jominy test.

Fig. 2.

Transformation start temperature. (a) 0B, 10B, and 20B steels; (b) 05Mo0B, 05Mo10B, and 05Mo20B steels; (c) 15Mo0B, 15Mo10B, and 15Mo20B steels.

Fig. 3.

Optical micrographs of test steels. (a) 0B, 10B and 20B steels, (b) 05Mo0B, 05Mo10B and 05Mo20B steels, (c) 15Mo0B, 15Mo10B, and 15Mo20B steels.

3.1.2. Jominy Test

Figure 4 shows the hardness distribution of the Jominy specimen for the 10Mo0B steel. By using the distance from quenched end, which corresponds to 90% martensite hardness, we can obtain the hardenability index, V_c-90.^18,31)

Fig. 4.

Typical hardness distribution of a Jominy specimen.

First, we investigated the effect of B content on hardenability for each Mo content. Figure 5 shows the effect of B content on the hardenability index, V_c-90, determined by the Jominy test for 0.5%, 1.0%, and 1.5% Mo. In Fig. 5, the predicted values of V_c-90 for the B steels calculated using the formula proposed by Ueno et al.¹⁸⁾ and Asahi’s data¹⁴⁾ for 0.15C-0.02Ti-added steels with 0.5% Mo are plotted as reference. V_c-90 of the 0.5% Mo-added steels in this study showed a clear local minimum around 10 ppm B content and decreased hardenability over 10 ppm B content. This behavior agrees with Asahi’s data; that is, Mo increases the optimal content of B on hardenability from 5 ppm to 13 ppm.

Fig. 5.

Effect of B content on hardenability (V_c-90) influenced by Mo content. (Online version in color.)

In contrast, the B dependence of V_c-90 in the 1.0% and 1.5% Mo steels was different from that in the 0% and 0.5% Mo steels. The V_c-90 value of the 1.5% Mo steels decreased with increasing B content up to around 12 ppm and became almost constant or gradually increased at B contents exceeding 13 ppm. Moreover, the optimum B content affecting V_c-90 decreased from 12 to 11 ppm despite the increase in Mo content from 1.0% to 1.5%. This result indicates that over 1.0% Mo content, the combined Mo-B effect decreases with increasing Mo content.

Therefore, we next investigated the effect of Mo content on hardenability for 0, 5, 10, and 20 ppm B contents. Figure 6 shows the effect of Mo content on V_c-90 of Mo steels and Mo-B steels. The data for Mo-0B and Mo-10B in Fig. 6 are the same as those in Fig. 2 of Ref. 32, and the values of V_c-90 predicted using the formula proposed by Ueno et al.¹⁸⁾ are shown in Fig. 6. Note that we plotted the calculated value of V_c-90 for the 5 ppm B steel as the experimental value for Mo-05B. In B-free steels, V_c-90 decreased with increasing Mo content. However, in 20 ppm B steels, V_c-90 showed a constant value for Mo contents over 0.75%, although V_c-90 decreased up to 0.75% Mo following the calculated result predicted based on Ueno’s formula.¹⁸⁾ This result indicates that the effect of B is diminished above 0.75% Mo content in 20 ppm B steels, contrary to the case of B steels with Mo contents below 0.75% (the combined Mo-B effect).

Fig. 6.

Effect of Mo content on hardenability (V_c-90) of B-free steel and B-added steel. (Online version in color.)

3.2. Precipitation Behavior

We determined the maximum Mo content for the combined effect on hardenability, i.e., the effect of B decreases with increasing Mo content over 0.75% in 20 ppm B steels. These behaviors are contrary to the effect of Mo below 0.75% Mo content, known as the combined Mo-B effect. To understand the origin of the change in the effect of Mo, it is important to investigate the borides, mainly at the prior austenite grain boundary. In this experiment, the total amount of borides was too small to be determined. For example, the precipitated B content in 15Mo20B, as detected by chemical analysis of the electrolytic extraction residues, was 1 ppm.

Table 2 shows the results of TEM and SIMS observations of borides in the grain boundary in the 10 ppm B steels. As mentioned in a previous report, the precipitation of Fe₂₃(C, B)₆ is suppressed by the addition of 0.5% Mo at the cooling rates of 30°C/s and 5°C/s. Figures 7(a) and 7(b) show the SIMS images of the B-related secondary ion maps of the sample that was cooled at 5°C/s. The discrete granular signals in Fig. 7(a) denote Fe₂₃(C, B)₆, and the continuous line signals in Figs. 7(a) and 7(b) denote the segregated solute B in the prior austenite grain boundary. In the 15Mo10B steel, no B precipitation was observed even at the lowest cooling rate of 0.5°C/s. Note that we cannot estimate the amount of segregated solute B atoms at the prior grain boundaries from the intensity of the B-related secondary ion obtained by SIMS measurements. This is because SIMS is not suitable for quantitative measurements, as the yield of the B-related secondary ion is strongly dependent on the surface chemistry.²⁸⁾

Table 2. Observation results of the precipitation along grain boundaries (10 ppm B added steels).

10 ppm B	30°C/s	5°C/s	0.5°C/s
10B (Vc-90=41°C/s)	Fe23(C,B)6 Size: 0.1–0.5 μm Site: G.B.	Fe23(C,B)6 Size: 0.3–1.5 μm Site: G.B. (Fig. 7(a))	–
05Mo10B (Vc-90=4.2°C/s)	No boride	No boride (Fig. 7(b))	–
07Mo10B	–	–	–
15Mo10B (Vc-90=1.6°C/s)	–	–	No boride

Fig. 7.

SIMS images of boron related secondary ion maps for different heat patterns. (a) 00Mo10B (5°C/s), (b) 05Mo10B (5°C/s), (c) 05Mo20B (5°C/s), (d) 07Mo20B (0.5°C/s), (e) 15Mo20B (0.5°C/s), (f) 15Mo25B (water quenched just after reheating).

Table 3 shows the results of TEM and SIMS observations of 20 ppm B steels. The precipitation behavior of the 20 ppm B steels was different from that of the 10 ppm B steels: precipitation of Fe₂₃(C, B)₆ was observed at the prior austenite grain boundary even in the 05Mo20B steel containing 0.5% Mo. Figures 7(c) and 8 show the Fe₂₃(C, B)₆ phase in 05Mo20B. In contrast, there was no B precipitation in 07Mo20B cooled at 30°C/s. This indicates that in the 20 ppm B steel, 0.75% Mo is enough to suppress the precipitation of Fe₂₃(C, B)₆ at a cooling rate of 30°C/s.

Table 3. Observation results of precipitation along grain boundaries (20 ppm B added steels).

20 ppm B	30°C/s	5°C/s	0.5°C/s
20B (Vc-90=41°C/s)	–	–	–
05Mo20B (Vc-90=9.5°C/s)	Fe23(C,B)6 Size:0.1–0.5 μm Site: G.B.	Fe23(C,B)6 Size: 0.1–1.0 μm Site:G.B. (Fig. 7(c), 8)	–
07Mo20B (Vc-90=2.4°C/s)	No boride	–	Mo2FeB2 Size: 0.05–0.1 μm Site: G.B. (Fig. 7(d))
15Mo20B (Vc-90=2.5°C/s)	–	–	Mo2FeB2 Size: 0.1–0.5 μm Site: G.B., W.G. (Figs. 7(e), 9, 10)

Fig. 8.

(a) Extraction replica TEM photograph, (b) Electron diffraction pattern, (c) EDS spectra of boron precipitate in 05Mo20B steel cooled at 5°C/s from reheating temperature of 950°C. (Online version in color.)

In 07Mo20B and 15Mo20B steels, however, at the lowest cooling rate of 0.5°C/s, different types of boride and Mo₂FeB₂ were observed. Figures 7(d) and 7(e) show the SIMS images of 07Mo20B and 15Mo20B. It is clear that the number of discrete granular signals from the borides increases with increasing Mo content. Figure 9(a) shows the extraction replica TEM photograph of Mo₂FeB₂ in the 15Mo20B steel. The sizes of the precipitates along the prior austenite grain boundary were approximately 0.5 μm. Figure 9(b) shows the electron diffraction pattern, and Fig. 9(c) shows the EDS spectra of Mo₂FeB₂. The precipitates shown in Fig. 9(a) were determined as Mo₂FeB₂ by electron diffraction analysis. The intensities of the Mo peaks in the EDS spectra were almost twice as large as that of the Fe peak, which is consistent with the diffraction analysis. It is noted that Mo₂FeB₂ exist not only at the prior austenite grain boundary, but also at the intragranular sites in the 15Mo20B steel. In contrast to the Mo₂FeB₂ phase at the grain boundary, Mo₂FeB₂ in the intragranular sites precipitated in combination with (Ti, Mo)(C, N). Figure 10(a) shows the extraction replica TEM photograph of intragranular Mo₂FeB₂; the intragranular Mo₂FeB₂ is approximately 0.1 μm in size and is smaller than Mo₂FeB₂ in the grain boundary. Analysis of the electron diffraction pattern in Fig. 10(b) confirmed that there is no special crystallographic orientation relationship. The Mo₂FeB₂ phase was observed even in the 15Mo25B steel specimen, which was water-quenched just after reheating at 950°C, as shown in Fig. 7(f). This suggested that Mo₂FeB₂ precipitated during reheating at 950°C.

Fig. 9.

(a) Extraction replica TEM photograph, (b) Electron diffraction pattern, (c) EDS spectra of boron precipitate along grain boundaries in 15Mo20B steel cooled at 0.5°C/s from reheating temperature of 950°C. (Online version in color.)

Fig. 10.

(a) Extraction replica TEM photograph, (b) Electron diffraction pattern, (c) EDS spectra of intragranular boron precipitate in 15Mo20B steel cooled at 0.5°C/s from reheating temperature of 950°C. (Online version in color.)

It is also noted that Mo₂FeB₂ has been reported in many studies, especially in the field of Maraging Steels³³⁾ and cermets produced by the sintering method.³⁴⁾ Karlsson et al. reported that M₃B₂-type borides containing Cr and Mo precipitate in stainless steels with 40 ppm B, 2.6% Mo, and 17.5% Cr.³⁵⁾ However, this is the first report that reveals the role of Mo₂FeB₂ in determining the maximum Mo content for the combined Mo-B effect on hardenability.

4. Discussion

4.1. Relation between Hardenability and Precipitation Behavior

The experimental results, hardenability, and type of borides in the Mo-B steels are summarized in Fig. 11. Mo increases the hardenability of B steels by suppressing the precipitation of Fe₂₃(C, B)₆. This effect appears in the “no boride” region in Fig. 11. In 0.5% Mo steels, the hardenability decreases rapidly for more than 13 ppm B content, with the precipitation of Fe₂₃(C, B)₆, which is shown as the “Fe₂₃(C, B)₆” region in Fig. 11. In this study, we first observed a decrease in the effect of B on hardenability by Mo addition for Mo contents more than 0.75%, wherein Mo₂FeB₂ precipitated instead of Fe₂₃(C, B)₆, which is shown as the “Mo₂FeB₂” region in Fig. 11. In contrast to 0.5% Mo steels, in 1.0% and 1.5% Mo steels, the hardenability gradually decreased for B contents over 13 ppm. To clarify the difference between the effects of two types of borides on the hardenability, we need to understand the difference between their precipitation behaviors. Therefore, we conducted thermodynamic calculations using Thermo-Calc software to estimate the precipitation temperature of each boride.

Fig. 11.

The relation between the property of hardenability and the B precipitates in Mo-B steels. (Online version in color.)

4.2. Conditions for Thermodynamic Calculation

In this calculation, considering the experimental results, we used a thermodynamic database, which included M₃B₂ and M₂₃C₆ phases. To focus on the precipitation behaviors of the borides, only boride phases, M₂₃C₆, M₃B₂, Mo₂BC, and M₂B were included in the calculation, and FCC was chosen as the bulk phase with 0.15% C, 1.3% Mn, 0.002% B, and 0–1.5% Mo.

As mentioned in the introduction, Han et al. have already demonstrated by thermodynamic calculations that Mo deteriorates the phase stability of the M₂₃(C, B)₆ phase, in addition to the kinetic effect retarding the carbon diffusivity.^19,20) However, the thermodynamic effect of Mo on the precipitation behavior of M₃B₂ is not clear. Therefore, we mainly discuss M₃B₂ in this section.

4.3. Effect of Mo and B Contents on Precipitation Temperature of M₃B₂ and Amount of Solute B

Figure 12 shows the phase diagram of Mo-B steel, and Fig. 13 shows the B content in the B precipitates, calculated by the Thermo-Calc software. In the case of 0% Mo shown in Fig. 13(a), all the B atoms are completely dissolved in FCC over 1180°C. M₂B starts precipitating below 1180°C, and M₂₃C₆ stably exists below 600°C. The precipitation region of M₂₃C₆ is shown as region 1 in Fig. 12. The precipitation start temperature of M₂₃C₆, i.e., 600°C in this calculation, does not change at 0.1% Mo content. From 0.1% Mo to 0.8% Mo, only M₂B and Mo₂BC exist as the stable boride phases instead of M₂₃C₆. M₃B₂ starts to precipitate over 0.8% Mo content, which is shown as region 2 in Fig. 12, and the precipitation start temperature of M₃B₂ increases from 870°C to 1050°C with increasing Mo content from 0.8% to 1.5%. For 1.5% Mo content shown in Fig. 13(b), M₃B₂ can precipitate at 930°C and 1050°C. This calculation suggests that Mo₂FeB₂ observed in this study precipitated mainly during reheating at 950°C. The calculated Gibbs free energy supports the precipitation of Mo₂FeB₂ during reheating at 950°C. Figure 14 shows the effect of the site fraction of Mo in the M₃B₂ phase on the Gibbs free energy. The Gibbs free energy decreases when the ratio of the Mo to Fe contents is almost 2:1, which indicates that Mo₂FeB₂ is the most stable structure among the M₃B₂-type borides at the reheating temperature of 950°C. These calculations are consistent with the experimental results because Mo₂FeB₂ is observed in the 15Mo25B steel specimen, which is water-quenched just after reheating at 950°C.

Fig. 12.

Phase diagram of the Mo-B steel calculated by the Thermo-Calc software assuming intragranular contents in austenite: C = 0.15%, Mn = 1.3%, Mo = 0%–1.5%, B = 0.002%, in mass percent. (Online version in color.)

Fig. 13.

Calculated B content in B precipitates: C = 0.15%, Mn = 1.3%, B = 0.002%, (a) 0%Mo, (b) 1.5%Mo in mass percent. (Online version in color.)

Fig. 14.

Effect of the site fraction of Mo on the Gibbs free energy of M₃B₂.

4.4. Effect of Mo on Hardenability of B Steels

Considering all these results, we can schematically summarize the effect of Mo addition on the precipitation behaviors of Mo₂FeB₂ and Fe₂₃(C, B)₆, as shown in Fig. 15. Mo suppresses the precipitation of Fe₂₃(C, B)₆ during cooling by the thermodynamic effect and/or kinetic effect below the Mo content of 0.75%. The solute B content at the austenite grain boundaries before transformation increases by suppressing Fe₂₃(C, B)₆, which is known as the combined Mo-B effect on hardenability.³⁴⁾ In contrast, Mo promotes the precipitation of Mo₂FeB₂ mainly during reheating at Mo content over 0.75%. This is supported by thermodynamic calculations. The calculated B content in the B precipitates suggested that the precipitation of Mo₂FeB₂ decreases the solute B content during reheating, which leads to a decrease in the combined Mo-B effect over 0.75% Mo content. Note that the decrease in the solute Mo content by the precipitation of Mo₂FeB₂ has little effect on the hardenability because the amount of solute Mo consumed as M₃B₂ is below 0.02%, considering the B content is 20 ppm; this value is too small compared with the bulk Mo contents.

Fig. 15.

Schematic image of effect of Mo addition on precipitation behavior of Mo₂FeB₂ and Fe₂₃(C, B)₆.

Lastly, we mention some differences between the experimental results and calculations. Although the evaluated precipitation behavior of M₃B₂ during reheating qualitatively agrees well with the experimental results, there are some discrepancies between the experiment and calculation. It must be noted that the calculation cannot perfectly explain the precipitation behavior of Mo₂FeB₂ in the 07Mo20B steel. Although our calculation suggests that Mo₂FeB₂ can precipitate over 0.8% Mo, Mo₂FeB₂ in the 07Mo20B steel precipitated at the prior austenite grain boundaries only at the cooling rate of 0.5°C/s but not at the cooling rate of 30°C/s. The experimental results indicated that Mo₂FeB₂ precipitated during cooling from 950°C. Thus, we assume that the discrepancy probably originates from the assumptions made in the calculation. In this study, thermodynamic calculation was conducted by assuming only the bulk contents of the elements. However, considering that the dominant precipitation site of Mo₂FeB₂ in the 07Mo20B steel was at the prior austenite grain boundaries, the calculations should include the amounts of each element at the austenite grain boundaries during cooling. Therefore, to predict the grain boundary precipitation of borides, not only Mo₂Feb₂, but also Fe₂₃(C, B)₆ and the amount of each element at the grain boundaries must be evaluated using STEM-EELS³⁶⁾ or 3DAP.^31,37) We aim to examine this in detail in future.

Another discrepancy is that M₂B and Mo₂BC type borides were not observed in the experiment, although the thermodynamic calculation suggests that these borides exist in the Mo-B steel. This discrepancy may be ascribed to the kinetic effect. As for M₂B, Fujishiro et al.¹⁰⁾ reported that the precipitation amount of Fe₂B increases after 1000 s at 650°C in B steels (0% Mo). In addition, they reported that Mo retards the precipitation of not only M₂₃(C, B)₆, but also M₂B. This suggests that the absence of M₂B in this experiment is due to the lack of time for the precipitation of M₂B during reheating and cooling. On the other hand, the precipitation kinetics of Mo₂BC has not been reported and needs to be investigated.

5. Conclusions

• The optimum B content affecting the hardenability decreases from 12 to 11 ppm with increasing Mo content from 1.0% to 1.5%.

• When the Mo content < 0.75%, Mo suppresses the precipitation of Fe₂₃(C, B)₆ during cooling. When the Mo content > 0.75%, the precipitation of Mo₂FeB₂ is promoted, mainly during reheating, which reduces the effect of B and determines the maximum Mo content for the combined Mo-B effect.

• The thermodynamic calculation adequately describes the intragranular precipitation behavior of Mo₂FeB₂.

• The amount of each element at the prior austenite grain boundaries and around the borides needs to be measured using STEM-EELS or 3DAP to reveal the relationship among the solute B, borides, and hardenability.

References

• 1)   M. A.  Grossman: Trans. AIME, 150 (1942), 227.
• 2)   M.  Ueno and  K.  Itoh: Tetsu-to-Hagané, 74 (1988), 910.
• 3)   R. A.  Grange and  T. M.  Garvey: Trans. Am. Soc. Met., 37 (1946), 136.
• 4)   C. R.  Simcoe,  A. R.  Elsea and  G. K.  Manning: J. Met., 8 (1956), 984.
• 5)   S.  Yoshida,  K.  Ushioda,  Y.  Abe and  J.  Ågren: Tetsu-to-Hagané, 98 (2012), 482.
• 6)   M.  Ueno and  T.  Inoue: Trans. Iron Steel Inst. Jpn., 13 (1973), 210.
• 7)   G. F.  Melloy,  P. R.  Slimmon and  P. P.  Podgursky: Metall. Trans., 4 (1973), 2279.
• 8)   K.  Yamanaka and  Y.  Ohmori: Tetsu-to-Hagané, 62 (1976), 895.
• 9)   Y.  Ohno,  Y.  Okamura,  S.  Matsuda,  K.  Yamamoto and  T.  Mukai: Tetsu-to-Hagané, 73 (1987), 1010.
• 10)   T.  Fujishiro,  T.  Hara and  G.  Shigesato: Tetsu-to-Hagané, 101 (2015), 300.
• 11)   R.  Habu,  M.  Miyata,  S.  Sekino and  S.  Goda: Tetsu-to-Hagané, 60 (1974), 1470.
• 12)   B. M.  Kapadia,  R. M.  Brown and  W. J.  Murphy: Trans. AIME, 242 (1968), 1689.
• 13)   H.  Tamehiro,  M.  Murata,  R.  Habu and  M.  Nagumo: Tetsu-to-Hagané, 74 (1986), 458.
• 14)   H.  Asahi: ISIJ Int., 42 (2002), 1150.
• 15)   T.  Hara,  H.  Asahi,  R.  Uemori and  H.  Tamehiro: ISIJ Int., 44 (2004), 1431.
• 16)   Y.  Desalos,  Ph.  Maitrepierre and  J.  Rofes-Vernis: Mem. Sci. Rev. Metall., 73 (1976), 735.
• 17)   M.  Ueno and  K.  Itoh: Tetsu-to-Hagané, 74 (1988), 918.
• 18)   M.  Ueno and  K.  Itoh: Tetsu-to-Hagané, 74 (1988), 1073.
• 19)   L.  Karlsson,  H.  Norden and  H.  Odelius: Acta Metall., 36 (1988), 1.
• 20)   L.  Karlsson and  H.  Norden: Acta Metall., 36 (1988), 13.
• 21)   L.  Karlsson: Acta Metall., 36 (1988), 25.
• 22)   X. L.  He,  Y. Y.  Chu and  J. J.  Jonas: Acta Metall., 37 (1989), 2905.
• 23)   G. M.  Carinci: Appl. Surf. Sci., 76–77 (1994), 266.
• 24)   X. L.  He,  Y. Y.  Chu and  J. J.  Jonas: Acta Metall., 37 (1989), 147.
• 25)   F.  Han,  B.  Hwang,  D. W.  Suh,  Z.  Wang,  D. L.  Lee and  S. J.  Kim: Met. Mater. Int., 14 (2008), 667.
• 26)   B.  Hwang,  D. W.  Suh and  S. J.  Kim: Scr. Mater., 64 (2011), 1118.
• 27)   J. M.  Hodge and  M. A.  Orehoski: Trans. Metall. Soc. AIME, 167 (1946), 627.
• 28)   T.  Tanaka,  T.  Fujishiro,  G.  Shigesato and  S.  Hayashi: Surf. Interface Anal., 46 (2014), 297.
• 29)   W.  Steven and  A. G.  Haynes: J. Iron Steel Inst., 183 (1956), 349.
• 30)   J. E.  Morral and  T. B.  Cameron: Boron in Steel: Proc. Int. Symp. on Boron Steels 1979, ed. by S. K. Banerji and J. E. Morral, TMS of AIME, Warrendale, PA, (1980), 19.
• 31)   T.  Akutagawa: J. Jpn. Soc. Mech. Eng., 59 (1956), 452.
• 32)   J.  Takahashi,  K.  Ishikawa,  K.  Kawakami,  M.  Fujioka and  N.  Kubota: Acta Mater., 133 (2017), 41.
• 33)   K.  Kuribayashi and  R.  Horiuchi: Maraging Steels: Recent Development and Application, ed. by R. K. Wilson, TMS, , Warrendale, PA, (1988), 157.
• 34)   Y.  Kondo,  T.  Watanabe,  S.  Ohira,  K.  Takagi and  T.  Ide: Bull. Jpn. Inst. Met., 26 (1987), 302.
• 35)   L.  Karlsson and  H.  Norden: Acta Metall., 36 (1988), 35.
• 36)   G.  Shigesato,  T.  Fujishiro and  T.  Hara: Mater. Sci. Eng., A556 (2012), 358.
• 37)   Y. J.  Li,  D.  Ponge,  P.  Choi and  D.  Raabe: Scr. Mater., 96 (2015), 302.