International Journal of Minerals, Metallurgy and Materials V olume 20 , Number 8 , August 2013 , P age 733 DOI: 10.1007/s12613-013-0791-7

Design of a low-alloy high-strength and high-toughness martensitic steel Yan-jun Zhao1,2) , Xue-ping Ren2) , Wen-chao Yang1) , and Yue Zang2) 1) School of Materials Science and Engineering, Guangxi University, Nanning 530004, China 2) School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing 100083, China (Received: 23 October 2012; revised: 4 January 2013; accepted: 8 January 2013)

Abstract: To develop a high strength low alloy (HSLA) steel with high strength and high toughness, a series of martensitic steels were studied through alloying with various elements and thermodynamic simulation. The microstructure and mechanical properties of the designed steel were investigated by optical microscopy, scanning electron microscopy, tensile testing and Charpy impact test. The results show that cementite exists between 500◦ C and 700◦ C, M7 C3 exits below 720◦ C, and they are much lower than the austenitizing temperature of the designed steel. Furthermore, the Ti(C,N) precipitate exists until 1280◦ C, which refines the microstructure and increases the strength and toughness. The optimal alloying components are 0.19% C, 1.19% Si, 2.83% Mn, 1.24% Ni, and 0.049% Ti; the tensile strength and the V notch impact toughness of the designed steel are more than 1500 MPa and 100 J, respectively. Keywords: high strength steel; martensitic steel; alloy design; thermodynamics; alloying elements; microstructure; mechanical properties

1. Introduction The high-strength low-alloy (HSLA) steel, combined with high strength, high toughness, low alloying element content and sustaining ability in severe service conditions, is a widely used steel in the world. The low carbon martensitic alloy steel is a common HSLA steel. Through quenching and tempering, it is strengthened by martensitic phase transformation and ε-carbide precipitation. However, the strength decreases with increasing toughness [1-3]. This finding induced extensive studies on the relationship between the mechanical properties and microstructure of the martensitic alloy steel [4-6]. Grain refinement of austenite attracted much attention due to the Hall-Petch relationship [5, 7-11]. It was found that the yield strength increased 235 MPa and the Charpy U-notch impact energy at 77 K improved more than eight times in a 17CrNiMo6 steel when the austenite grain size was refined from 199 to 6 µm [5]. On the other hand, the yield strength and toughness of the martensitic steel strongly depend on the microstructure of lath martensite, such as martensite packet size, block width, and dislocation density [12-15]. Corresponding author: Yan-jun Zhao

This study aims at developing a new HSLA steel, and specially focuses on the effect of the existing temperature range and constituting elements of the precipitates. That is because the two factors greatly affect the austenite grain size, strength, and toughness of the HSLA steel. The software Thermo-Calc [16] with Gibbs free energy database was used to simulate the effect of alloying, microstructure, optimization, austenite grain refinement, and precipitate formation. In particularly, the effect of alloying with Mn, Si, and Ni was investigated. The microstructure optimization, such as precipitate formation and stabilization of MX, M7 C3 , and cementite, was explored as well. Finally, the microstructure and mechanical properties of the designed HSLA steel were experimentally studied to verify the design concept.

2. Experimental The steel used in the work was prepared by vacuum induction melting, continuous casting into ingots, and then forging. The forged pieces were annealed at 680◦ C for 4 h and then normalized at 860◦ C for 30 min. The steel

E-mail: [email protected]

c University of Science and Technology Beijing and Springer-Verlag Berlin Heidelberg 2013 

734

Int. J. Miner. Metall. Mater., V ol. 20 , No. 8 , Aug. 2013

treated by the above procedures was austenized at 900◦ C for 40 min with oil-quenching, air-cooling and then tempering at 200◦ C for 2 h with air-cooling. The samples were polished and etched using supersaturated carbazotic acid solution for austenite grain characterization by using a DMM-660 optical microscope. The samples were etched using 4% nital for S-3400N scanning electron microscopy (SEM) observation. Tensile specimens were machined into a φ5 mm cylinder, and their gauge length was 25 mm. Tensile tests were conducted using an MTS810 materials testing system at a strain rate of 10−2 s−1 . Impact specimens were machined into a 10 mm × 10 mm × 55 mm block with a V-shape notch. Impact tests were conducted using a JB-30B pendulum machine based on the standard GB/T229-1994.

3. Results and discussion 3.1. Design of the steel The target of the developed martensitic steel is that the tensile strength is above 1500 MPa and the V notch imTable 1. Steel PG1 PG2 PG3 PG4 PG5 PG6

C 0.20 0.19 0.19 0.23 0.18 0.21

Si 1.02 0.23 1.19 0.94 0.20 0.55

Mn 2.26 1.46 2.83 2.93 1.76 2.04

Ni 0.062 1.030 1.240 0.100 0.018 0.026

pact toughness is above 100 J, so that the steel can sustain in severe conditions, such as discontinuous and high-speed circulating bump. Lath martensite consists of high-density dislocations, which make the steel have high strength and ductility. Therefore, lath martensite with high-density dislocations should be obtained in the steel. In addition, the developed martensitic steel should have good weldability and sufficient hot workability. It is necessary to increase the carbon content of the steel to achieve the high strength. However, higher contents of carbon would decrease the amount of lath martensite and increase the amount of plate martensite, which decreases the plasticity and toughness of the steel. When the carbon content is lower than 0.2wt%, austenite transforms into lath martensite; while the carbon content is 0.2wt%-1.0wt%, mixed microstructures with lath martensite and plate martensite form. Therefore, the carbon content of the steel should not exceed 0.25wt% to achieve a combination of high toughness and strength. The contents of carbon are nearly the same in the developed PG1-PG6 steels (Table 1).

Chemical compositions of the steels Cr 0.620 0.330 0.031 0.059 1.200 0.610

Ti 0.020 0.017 0.049 0.073 0.017 0.018

V 0.012 0.015 0.013 0.015 0.014 0.015

The atom radius of Mn, Si, Ni, and Fe is nearly equal, so they are selected as the alloying elements in the study to achieve the high strength and toughness. Specifically, the alloying contents of Mn and Si should be more than that of Ni to lower costs. In addition, the Mn content should be kept as high as possible to achieve the high tensile strength, above 1500 MPa, under the condition of less than the 0.25wt% carbon content. However, the weldability of the developed steel deteriorates with the Mn content increasing. For this reason, the Mn content should be below 3.0wt%. Moreover, the Si content should remain below 1.0wt%-1.5wt% to improve the impact toughness when the tempering is at 200◦ C [7]. For the content of Ni, it is believed that Ni increases the hardenability, strength, and toughness. However, the content of Ni should be controlled at about 1.0wt% due to cost issue. The elements Ti and V are added for grain refinement and the formation of carbonitride. As shown in Table 1, PG3 and PG4 were used to reveal the effect of V-Ti complex carbonitride on the austenitic grain growth.

3.2. Thermodynamic equilibrium phases and precipitates Thermodynamic simulation using Thermol-Calc has been performed for the tested 3Mn-1Si-1Ni-Cr-V-Ti steel

P 0.015 0.014 0.015 0.015 0.013 0.013

S 0.0052 0.0064 0.0045 0.0059 0.0110 0.0068

wt% N <0.0027 <0.0027 <0.0027 <0.0027 <0.0027 <0.0027

O <0.0015 <0.0015 <0.0015 <0.0015 <0.0015 <0.0015

Al <0.010 0.024 <0.010 — — <0.010

Fe Bal. Bal. Bal. Bal. Bal. Bal.

to study the alloying effect on the phase precipitation behavior, which is used to optimize the composition of the steel and control the precipitates. Figs. 1 and 2 show the effect of temperature on the matrix phases and precipitates of PG1 and PG3 steels. The austenitizing temperature and precipitates of the six steels are shown in Table 2. Thermodynamic calculations indicate that the matrix phases are ferrite and austenite, and the main equilibrium precipitates are M7 C3 carbides, MX carbonitrides (M represents metal and X represents N or C), and cementite in the six tested steels. A small amount of impurity phases, such as AlN and MnS, should exist as well. Moreover, Figs. 3-5 show the mass fraction of main elements in the equilibrium precipitates. Among them, the main elements of M7 C3 carbides are C, Fe, Mn, and Cr. The main elements of MX carbonitrides are V, Ti, N, and C, which are V-Ti complex carbonitrides. The main elements of cementite are C, Fe, Mn, and Cr. Table 2 shows the beginning and complete austenitizing temperatures of the six developed steels. Interestingly, both the beginning (about 600◦ C) and complete austenitizing temperatures (about 775◦ C) of PG3 steel are the lowest among the six developed steels, mainly because of their different alloying elements and contents. If the aust-

735

Y.J Zhao et al., Design of a low-alloy high-strength and high-toughness martensitic steel

Fig. 1.

Relationship between equilibrium phases and temperature in PG1 steel: (a) matrix phases; (b) precipitates.

Fig. 2.

Relationshipsbetween equilibrium phases in PG3 steel and temperature: (a) matrix phases; (b) precipi-

tates.

Table 2.

Austenitizing temperature and precipitates’ data in the designed steels

Steel

Beginning austenitizing temperature/ ◦C

Complete austenitizing temperature/ ◦C

Mass fraction of V-Ti complex carbonitride at 900◦ C/%

PG1 PG2 PG3 PG4 PG5 PG6

690 650 600 650 700 660

810 785 775 785 800 800

0.040 — 0.100 0.100 0.030 0.025

Complete dissolution temperature of V-Ti complex carbonitride/ ◦C 1160 — 1280 1310 1110 1125

Complete dissolution temperature of M7 C3 /◦ C

Mass fraction of M7 C3 at 400◦ C/%

Temperature range of cementite/ ◦C

750 650 520 680 780 720

1.9 1.9 2.1 2.3 1.8 2.2

— 580-680 500-650 580-680 — 680-700

736

Fig. 3.

Int. J. Miner. Metall. Mater., V ol. 20 , No. 8 , Aug. 2013

Relationships between temperature and the mass fraction of elements in MX carbonitrides: (a) PG3; (b)

PG5; (c) PG6 (the phases do not exist in the temperature range of the dashed line).

enizing temperature is lower, the tendency of austenitic grain growth is lower. Therefore, alloy elements in PG3 steel, such as Ni, Si, and Mn, are the most promising austenite forming elements. On the other hand, the solid solubility of Ti(N,C) or V(N,C) in the steel increases with the increase of Mn content. Because of higher Mn contents, V-Ti complex carbonitrides in PG3 and PG4 steels dissolve completely at about 1280 and 1310◦ C, respectively, whose contents are the highest, approximately, 0.1wt%. The VTi complex carbonitrides undissolved still exist at about 1280◦ C in PG3 steel, which can greatly restrain austenitic grain growth. The dispersed carbonitrides inside the matrix are often generated by the combination reactions of alloy elements and C or N. They would deter austenitic and ferrite grain growth and grain boundary migration, which refines the microstructure and increases the strength and toughness [17]. Therefore, the average austenite grain size of PG3 steel is the smallest among the six tested steels. The main elements of M7 C3 are C, Fe, Mn, and Cr, which constitute the complex carbide (Fe,Mn,Cr)7 C3 (Fig. 4). The contents of (Fe,Mn,Cr)7 C3 are similar at 400◦ C in PG1-PG6 steels, which is approximately 2.0% (Table 2). The contents of Cr in PG1, PG5, and PG6 steels are higher than that in PG3 steel, and so, the diffusion of Cr in M7 C3 needs a higher temperature (Table 1). Then, the completely dissolved temperatures of (Fe,Mn,Cr)7 C3 in PG1, PG5 and PG6 steels are about 720◦ C, while in PG3 steel, the temperature is only 520◦ C, which improves the Cr homogeneity in austenite during the subsequent heating up. As seen in Fig. 5, the main elements of cementite are C, Fe, Mn, and Cr, which constitute the complex cementite (Fe,Mn,Cr)3 C. The complex cementite in PG3 steel exists in a range of temperature, from 500◦ C to 650◦ C, in PG2 and PG4 steels from 580◦ C to 680◦ C, and in PG6 steel from 680◦ C to 700◦ C. When the temperature is above 500◦ C, the elements Mn and Cr in the complex cementite will gradually dissolve out. The contents of Mn in PG2,

PG3, and PG4 steels are higher than those in the other steels, while Cr is lower (as shown in Table 1). The content of Cr is very high in PG6 steel. This content makes the complex cementite in PG6 steel stable and only existing in a narrow temperature range of 680-700◦ C, however, in PG3 steel, existing in a wide temperature range of 500650◦ C. From the discussed above, it indicates that cementite exists between 500◦ C and 700◦ C, and M7 C3 exists below 720◦ C. Also, the Ti(C, N) precipitate exists until 1280◦ C. When the steels are austenized at 900◦ C, cementite and M7 C3 are disappeared, which cannot deter austenite grain growth.

3.3. Microstructure and mechanical properties Figs. 6 and 7 show the microstructure and precipitates of PG3 steel after being water quenched from 870◦ C and 990◦ C, respectively. The microstructure of the steel is lath martensite. The MX carbonitride was determined to be Ti(C, N) by energy spectra, corresponding to the ThermoCalc calculations. In addition, the MX carbonitride shows rectangular or irregular shape, and its particle size is between 2.0 µm and 4.0 µm. Also, the MX carbonitride nucleates not only at the austenite grain but also on the boundaries of AlN phase at 990◦ C. Fig. 8 shows the mor phologies of austenite grains in the tested steel after being water quenched from 900◦ C. Obviously, the austenite grain sizes of PG3 and PG4 steels are the smallest among the six tested steels, in coincidence with the above Thermo-Calc simulations. The mechanical properties of the steels at room temperature are shown in Table 3. The alloy compositions were selected according to the Thermo-Calc calculations and experimental results, respectively. PG3 steel can obtain the optimal combination of strength and toughness, the tensile strength is more than 1500 MPa, and the V notch impact toughness is more than 100 J. PG3 steel is named as 20SiMn3NiA steel, and its main chemical compositions are listed in Table 4.

Y.J Zhao et al., Design of a low-alloy high-strength and high-toughness martensitic steel

Fig. 4.

737

Relationships between temperature and the mass fraction of elements in M7 C3 : (a) PG1; (b) PG2; (c)

PG3; (d) PG4; (e) PG5; (f ) PG6 (the phases do not exist in the temperature range of the dashed line).

738

Fig. 5.

Int. J. Miner. Metall. Mater., V ol. 20 , No. 8 , Aug. 2013

Relationships between temperature and the mass fraction of elements in cementite: (a) PG2; (b) PG3; (c)

PG4; (d) PG6 (the phases do not exist in the temperature range of the dashed line).

Fig. 6.

SEM image and energy spectrum of the MX carbonitride in PG3 steel water-quenched from 870◦ C.

Y.J Zhao et al., Design of a low-alloy high-strength and high-toughness martensitic steel

Fig. 7.

739

SEM micrograph and energy spectra of the MX carbonitride nucleating on AlN in PG3 steel water-

quenched from 990◦ C.

Fig. 8.

Optical microscopy photos of austenite grains in the tested steels water-quenched from 900◦ C: (a) PG1;

(b) PG2; (c) PG3; (d) PG4; (e) PG5; (f ) PG6.

740

Int. J. Miner. Metall. Mater., V ol. 20 , No. 8 , Aug. 2013 Table 3.

Steel PG1 PG2 PG3 PG4 PG5 PG6

Tensile strength / MPa 1510 1380 1570 1640 1450 1500

Table 4.

Mechanical properties of the six tested steels at room temperature Yield strength / MPa 1170 1150 1260 1350 1120 1150

Elongation rate / % 13.0 14.5 14.0 12.0 12.5 12.5

Reduction of area / % 55.5 60.0 54.0 50.0 58.5 52.5

Chemical composition of 20SiMn3NiA steel

wt% C 0.18-0.24

Si 1.0-1.3

Mn 2.8-3.0

Ni 1.0-1.3

Ti ≤ 0.06

Fe Bal.

4. Conclusion (1) The optimal alloy contents of the developed steel, named as 20SiMn3NiA, are 0.19% C, 1.19% Si, 2.83% Mn, 1.24% Ni, and 0.049% Ti. The tensile strength and V notch impact toughness of 20SiMn3NiA steel are more than 1500 MPa and 100 J, respectively. (2) Cementite and M7 C3 cannot change the tendency of austenitic grain growth. Cementite exists between 500◦ C and 700◦ C, M7 C3 exists below 720◦ C, and they are much lower than the austenitizing temperature of the designed steel. (3) Ti(C, N) exists until 1280◦ C after the designed steel was austenized, which refines the microstructure and increases the strength and toughness.

[6]

[7]

[8]

[9]

[10] [11]

Acknowledgement This work was financially supported by the Scientific Research Foundation of Guangxi University (No. XBZ110407).

[12]

References

[13]

[1] G. Krauss, Deformation and fracture in martensitic carbon steels tempered at low temperatures, Metall. Mater. Trans. B, 32(2001), p. 205. [2] M. Saeglitz and G. Krauss, Deformation, fracture, and mechanical properties of low-temperature-tempered martensite in SAE 43xx steels, Metall. Mater. Trans. A, 28(1997), p. 377. [3] L.Y. Li, Y. Wang, T. Han, and C.W. Li, Microstructure and embrittlement of the fine-grained heat-affected zone of ASTM4130 steel, Int. J. Miner. Metall. Mater., 18(2011), No. 4, p. 419. [4] Y.W. Gao, T.F. Jing, G.Y. Qiao, J.K. Yu, T.S. Wang, Q. Li, X.Y. Song, S.Q. Wang, and H. Gao, Microstructural evolution and tensile properties of low-carbon steel with martensitic microstructure during warm deforming and annealing, J. Univ. Sci. Technol. Beijing, 15(2008), No. 3, p. 245. [5] C.F. Wang, M.Q. Wang, J. Shi, W.J. Hui, and H. Dong, Effect of microstructure refinement on the strength and

[14]

[15]

[16]

[17]

Impact toughness/ J 137 108 130 93 111 117

Heat treatment 900◦ C, 40 min oil quenching + 200◦ C, 120 min air cooling after tempering

toughness of low alloy martensitic steel, J. Mater. Sci. Technol., 23(2007), p. 659. S. Morito, H. Yoshida, T. Maki, and X. Huang, Effect of block size on the strength of lath martensite in low carbon steels, Mater. Sci. Eng. A, 438-440(2006), p. 237. L.J. Wang, Q.W. Cai, H.B. Wu, and W. Yu, Effects of Si on the stability of retained austenite and temper embrittlement of ultrahigh strength steels, Int. J. Miner. Metall. Mater., 18(2011), No. 5, p. 543. J. Chakraborty, P.P. Chattopadhyay, D. Bhattacharjee, and I. Manna, Microstructural refinement of bainite and martensite for enhanced strength and toughness in highcarbon low-alloy steel, Metall. Mater. Trans. A, 41(2010), p. 2871. M.Y. Liu, B. Shi, C. Wang, S.K. Ji, X. Cai, and H.W. Song, Normal Hall-Petch behavior of mild steel with submicron grains, Mater. Lett., 57(2003), p. 2798. H.J. Rack, Age hardening-grain size relationships in 18Ni maraging steels, Mater. Sci. Eng. A, 34(1978), p. 263. R. Ishibashi, H. Arakawa, T. Abe, and Y. Aono, Tensile strength of austenitic stainless steels with grain refinement by mechanical milling, ISIJ Int., 40(2000), Suppl., p. S169. B. Hwang, T.H. Lee, and S.J. Kim, Effects of deformationinduced martensite and grain size on ductile-to-brittle transition behavior of austenitic 18Cr-10Mn-N stainless steels, Met. Mater. Int., 16(2010), No. 6, p. 905. A. Di Schino and J.M. Kenny, Grain size dependence of the fatigue behaviour of a ultrafine-grained AISI 304 stainless steel, Mater. Lett., 57(2003) , p. 3182. C.Y. Zhang, Q.F. Wang, J.X. Ren, R.X. Li, M.Z. Wang, F.C. Zhang, and Z.S. Yan, Effect of microstructure on the strength of 25CrMo48V martensitic steel tempered at different temperature and time, Mater. Des., 36(2012), p. 220. Z.M. Shi, K. Liu, M.Q. Wang, J. Shi, H. Dong, J. Pu, B. Chi, Y.S. Zhang, and J. Li, Effect of tensile deformation of austenite on the morphology and strength of lath martensite, Met. Mater. Int., 18(2012), No. 2, p. 317. P. Berthod, P. Lemoine, and L. Aranda, Experimental and thermodynamic study of nickel-based alloys containing chromium carbides: Part I. Study of the Ni-30wt% Cr-xC system over the [0-2.0wt% C] range, Calphad, 32(2008), p. 485. V. Kneˇzevi J. Balun, G. Sauthoff, G. Inden, and A. Schneider, Design of martensitic/ferritic heat-resistant steels for application at 650◦ with supporting thermodynamic modelling, Mater. Sci. Eng. A, 477(2008), p. 334.