Effect of Heat Treatment Technique on the Low Temperature Impact Toughness of Steel EQ70 for Offshore Structure

Introduction

Offshore structures immersed in the sea suffer from corrosion of seawater and atmosphere as well as pounding of the wave, higher properties, such as strength, weldability [1], corrosion resistance [2] and low-temperature toughness [1],are required for the bad service conditions. And high strength and toughness is always obtained by micro-alloying technology coupled with thermal mechanical control processing (TMCP) and/or appropriate heat treatment technique [3].

Steel plate EQ70 for offshore structure is recently developed as a kind of high strength low alloy steel, whose yield and tensile strength should be higher than 690 MPa and 770 MPa, with the transverse and longitudinal impact energy higher than 46 J and 69 J, respectively. It achieves high strength and toughness through adjusting contents of carbon and alloying elements coupled with appropriate heat treatment technique [3]. The roles of micro-alloying elements are to improve the strength and toughness of steels by grain refinement and precipitation strengthening [4]. However, grain refinement is in conflict with precipitation strengthening. To make the alloying elements come into full play, soaking temperature should be chosen carefully. Too high quenching temperature may result in the deterioration of the properties since the austenite grains are coarsened and deformation as well as cracks may occur in the steel [5, 6], while too low may lead to the decreases in strength and hardness as a small part of ferrite is remained. Appropriate soaking temperature must be the one at which austenite grains are small and distribute evenly and contents of the alloying elements solution in matrix are so enough to precipitate at lower temperature. Thus, the study on the quenching temperature on the low temperature toughness is essential. Gou’s [7] study of effects of quenching and tempering temperatures and quenchants on mechanical properties of offshore steel EQ70 with 2.10 wt. % Ni shows that the best quenching and tempering parameter is that quenching at 880 °C for 90 min and cooling with supersaturated nitrate solution, then tempering at 630 °C for 90 min. However, study of the effect of heat treatment techniques on the low temperature impact toughness of steel EQ70 is few.

While impact property is not only determined by heat treatment, chemical composition, the parameter that has influence on the microstructural parameters containing dislocation density, grain size as well as volume fractions and sizes of the second phase particles is also crucial [8], and the impact energy varies with the prior austenite grain size, marten site packet size, block width and lath width [9]. Nickel adds a solid solution hardening effect, increases quench hardenability, greatly enhances austenitic stability and influences the stacking fault energy of ferrite, the plastic deformation is accommodated at low temperatures [10, 11]. Gulyaye [12] indicated that Ni greatly reduces the temperature of transition to the brittle state. Norstrom et al. found that the addition of nickel decreases the Charpy V impact transition temperature of about 20 °C per percent nickel [13].

The existing researches either focus on the effect of heat treatment or the Ni content on the impact toughness, study on the combined contribution of them is few. Therefore, three kinds of heat treatment techniques were designed to study the effect of heat treatment on the impact toughness of steel EQ70 in this paper, the steels with different Ni contents were used to study the effect of Ni content on the low temperature impact toughness after different heat treatment techniques as well. The results can be a significant reference for actual heat treatment technique in steel plant.

Experimental procedures

The chemical compositions of as-received steel EQ70 were shown in Table 1. Besides Ni, the contents of Cr and Cu are different in steels A and B, respectively. But Cr and Cu in the experimental steel are adding to improve the corrosion resistance, which have little influence on the toughness. So the difference in Cr and Cu contents can be ignored as this study concerns on the effect on toughness. The equilibrium transformation temperatures Ae1 and Ae3 were calculated based on Thermo-Calc thermodynamic software. The Ae3 temperatures for steel EQ70 with 2.10 wt. % Ni (steel A) and 1.47 wt. % Ni (steel B) were 784 °C and 792 °C, respectively. The Ae1 temperatures for steel B were 642 °C, a small part of austenite is still retained at room temperature for steel A, but the mass fraction of the austenite drastically reduced at 653 °C.

Based on the thermodynamic calculation results and the orthogonal experiment done by Gou [7], three kinds of heat treatment modes were designed to study heat treatment technique on the impact toughness of steel EQ70. The regular quenching and tempering (Q&T) process was quenching at 880 °C for 90 min and cooling with 10 wt. % saline (in order to be closer to the actual production), then tempered at 630 °C for 90 min and cooled in the air. The circle quenching and tempering (CQ&T) process was one more quenching at 810 °C for 90 min than the Q&T process. The intercritical quenching and tempering (IQ&T) process was quenching at 720 °C for 90 min, then tempered at 630 °C for 90 min and cooled in the air. The schematic diagram of heat treatment processes mentioned above is shown in Figure 1.

Figure 1:

The schematic diagram of heat treatment processes.

After heat treatment, the steels were cut into standard Charpy V-notch specimen in transverse direction. The impact tests were carried out for three times at –20, –40, –60, –80 and –100 °C, respectively, then the average absorbed energy was obtained. The specimen whose impact energy was closest to the average was chosen for observation of the fracture morphology and microstructure. The fracture morphology was examined by using a scanning electron microscope (SEM, JSM-6480LV), and microstructures etched by 4 vol. % nitric acid were observed on a 9XB-PC type optical microscope.

Results and discussion

Impact toughness of the test steels after different heat treatments

The average impact energies of steels A and B after different heat treatments are shown in Table 2, and the relation between the energy absorbed in fracturing Charpy V-notch specimens and test temperature for steels A and B is shown in Figure 2. It is easy to find that, the impact toughness of steel A is higher than that of steel B at the same test temperature and heat treatment process. For steel B, the energy absorbed is, in descending order, CQ&T, Q&T and IQ&T, while for steel A, that is CQ&T, IQ&T and Q&T. As shown in Table 2, the impact energy of steel A is higher than 120 J at –40 °C after any heat treatment process and higher than 130 J after CQ&T, IQ&T process at –60 °C, which means that toughness grade can be increased from E to F after CQ&T, IQ&T process for steel A. But for steel B the impact energy cannot meet the requirement after IQ&T process, but the toughness grade can also be updated from E to F after CQ&T process.

Table 2:

The average impact energies of steels A and B after different heat treatment processes.

Test temperature/°C Heat treatment process		Average impact energy/J
		−20	−40	−60	−80	−100
Steel A	IQ&T	207.9	180.8	130.9	82.8	25.4
	CQ&T	208.3	207.2	191.1	153.7	84.6
	Q&T	167.9	123.8	59.8	29.8	26.9
Steel B	IQ&T	11.3	5.8	–	–	–
	CQ&T	132.0	95.1	76.3	39.6	19.5
	Q&T	141.1	68.2	42.2	25.1	17.6

Figure 2:

The low temperature impact toughness of steels A and B after different heat treatment processes.

The effect of heat treatment process on the low temperature impact toughness of steel A is more obvious than that of steel B. The difference in impact energy is less than 30 J between steel B after CQ&T and Q&T process at all test temperatures. However, for steel A the difference is larger than 130 J between specimens after CQ&T and Q&T process at –60 °C. With the decrease of test temperature, the difference becomes enlarged firstly then decreased as the temperature is lower than –80 °C.

Microstructure of the test steels

The microstructures of steels A and B after different heat treatment processes are shown in Figure 3. The microstructures are all composed of tempered lath martensite and retained austenite (as shown by the graph’s arrows in Figure 3) except steel B after IQ &T process, which is composed of the carbon-rich M-A constituent (as shown in Figure 3(d)). But the morphologies of martensite are not the same, the spacings of the martensitic laths and the contents of retained austenite are different. As shown in Figure 3(b) and (e), austenite grain sizes and spacing of the martensitic laths of the specimens with CQ&T process are smaller than those of the specimens with Q&T (Figure 3(c) and (f)) or IQ&T (Figure 3(a) and (d)) processes for both steels A and B and more austenite is retained.

Figure 3:

The microstructures of steels A and B after different heat treatment processes: (a) A: IQ&T, (b) A: CQ&T, (c) A: Q&T, (d) B: IQ&T, (e) B: CQ&T, (f) B: Q&T.

The impact toughness is determined by the comprehensive effects of austenite grain size, content and distribution of carbide and volume fraction of the retained austenite. The finer austenite grain results in smaller length and width of the martensite lath [14], so that the path for the propagation of crack becomes longer and the propagation resistance becomes larger. In addition, the increase of grain boundary area decreases the content of the tramp elements along the grain boundary. In one word, the impact toughness can be improved by grain refinement. Carbides can inhibit motion of the dislocation and result in the strengthening effect. However, it also wrecks continuity of the matrix, which deteriorates the toughness. Thus less content but more dispersive distribution of carbides can improve the toughness. Retained austenite can blunt the crack tip and suppress the propagation of cracks, delay the formation of voids and/or cracks and remit stress concentration caused by dislocation pileup for the strain energy has been absorbed during the transformation of metastable retained austenite to martensite induced by plastic deformation.

The comprehensive effects of various factors mentioned above make the impact toughness of specimens after CQ&T process better than others. And the worsening toughness of steel B after IQ&T is due to the continuity wrecking of the matrix caused by the brittle carbon-rich M-A constituents, which result in concentrated stress and cracks, while the better toughness of steel A is due to less carbides precipitating since Ni greatly enhances austenitic stability and more alloying elements are solution in austenite.

Fracture morphology of the test steels

Macroscopic morphologies and microscopic morphologies in radial and fibrous region of specimens test at –40 °C are shown in Figures 4–6. As shown in Figure 4, the lower parts of each individual figure are V-notched position of the impact specimens. The fractures expand from bottom to top during the impact process due to the V-notch. All the fracture morphologies contain fibrous, radial and shear lips region except for steel B after IQ&T process, which does not contain the shear lips region. For steel B after IQ&T process, the energy absorbed is very low, the fracture morphology only contains fibrous and radial region and the radial region takes up dominant percentage, totally more than 90 %. It is obvious that the shear lips region of steel A after any heat treatment process is larger than that of steel B.

Figure 4:

The macroscopic morphologies of the impact specimens test at –40 °C after different heat treatment processes: (a) A: IQ&T, (b) A: CQ&T, (c) A: Q&T, (d) B: IQ&T, (e) B: CQ&T, (f) B: Q&T.

Figure 5:

The morphologies of the impact specimens in radial region test at –40 °C after different heat treatment processes: (a) A: IQ&T, (b) A: CQ&T, (c) A: Q&T, (d) B: IQ&T, (e) B: CQ&T, (f) B: Q&T.

Figure 6:

The morphologies of the impact specimens in fibrous region test at –40 °C after different heat treatment processes: (a) A: IQ&T, (b) A: CQ&T, (c) A: Q&T, (d) B: IQ&T, (e) B: CQ&T, (f) B: Q&T.

The rupture of steel A after CQ&T process is ductile and cleavage fracture in radial region as shown in Figure 5(b), while it is mix of transgranular and cleavage fracture for steel B after IQ&T process as shown in Figure 5(d). The others are all cleavage fracture in the radial region. In addition, the cracks were also identified in the radial region. M/A grains and the precipitates, such as NbC, TiN, may be the nuclei of the crack [15]. The factors influencing the brittle fractures are sizes of the microcracks and the resistances they experience. The former is dependent on the dimensions of M/A grain and/or the precipitates, while the latter is influenced by microstructures the crack will cross. In addition, cracks are more obvious in the samples with Q&T process as shown in Figure 5(c) and (f), which is due to the bigger austenite grains and larger deformation for the higher austeniting temperature. As more alloying elements are solution in austenite and less carbides precipitate, the nucleus of the cracks decrease, more austenite is retained to restrain the crack propagation, the cracks’ number of steel A is less than that of steel B.

In the fibrous region as shown in Figure 6, the dominant rupture is ductile fracture, both voids and dimples are clearly observed. As the toughness of steel B after IQ&T process is worse than others, the fibrous region is limited as shown in Figure 6(d). The micromechanism operated during ductile fracture involves crack initiation and propagation [16]. The hard carbides and/or carbonitrides as well as the inclusions are incorporated in the matrix, which are easy to be the nuclei of voids [17]. The size, density, distribution of the second phase particle not only affects the nucleation but also the propagation of the voids. Ibrahim’s study shows that these parameters will affect propagation process to a much higher degree than voids initiation process [8]. Therefore, the dimples of steel A are deeper than that of steel B after the same heat treatment process.

Conclusions

1. The energy absorbed varies after different heat treatment processes, and the effects are different as the Ni content changes. For steel B, the energy absorbed is, in descending order, CQ&T, Q&T and IQ&T, while for steel A, that is CQ&T, IQ&T and Q&T. The impact grade can be updated by increasing Ni content and/or property heat treatment.

2. The microstructures are all tempered lath martensite and retained austenite. The spacings of the martensitic lath of specimen with CQ&T are smaller than others. More carbides precipitate in steel B than those in steel A after IQ&T process.

3. More austenite is retained and the impact toughness is improved with the increasing Ni content. The effect of heat treatment process on the low temperature impact toughness is more obvious for steel A.