Evolutions of Microstructure and Mechanical Properties of the Welding Heat-affected Zone in 800 MPa-grade Complex Phase Steel

Abstract

A thermomechanical simulator Gleeble 3800 was used to simulate the thermal cycles experienced by various heat-affected zones (HAZ) during the welding process. The influence of peak temperature (T_p, 500°C–1320°C) on the hardness, microstructure, precipitates, and properties of complex steel 780FB with microalloyed elements Ti, Nb, and V was systematically studied. The contributions of dislocation strengthening, precipitation strengthening, fine grain strengthening, and phase transformation strengthening increments to strength changes of samples after different thermal cycles were quantified, and the calculated results were found to be consistent with the experimental data. Compared with 780FB, there was little change in microstructure and properties when T_p was 500°C. When T_p was 650°C, the increase in VC density from 43/µm² to 288/µm² caused the enhancement of hardness and strength. The precipitation strengthening increment (49.84 MPa) played a dominant role in strength improvement. As partial bainite in the microstructure of 780FB transformed into ferrite at T_p of 800°C, the weakening of phase transformation strengthening (−57.5 MPa) became the main factor in strength change. The softening and strength reduction further increased when T_p was up to 980°C, as 780FB completely recrystallized and transformed into ferrite and MA islands. The phase transformation strengthening further reduced by 74.75 MPa. When T_p was 1320°C, the VC density decreased from 43/µm² to 13/µm², and the (Ti,Nb)C density decreased from 34/µm² to 14/µm², leading to severe grain growth (2.24 µm to 19.89 µm) and bainite transformation. The decrease in precipitation strengthening (−26.86 MPa) and fine grain strengthening (−87.91 MPa) counteracted with the increase in phase transformation strengthening (51.62 MPa), resulting a slight decrease in hardness and strength.

1. Introduction

The increasing emphasis of energy conservation and emission reduction has driven automotive industry towards lightweight design, which promoted the development of high-strength steel.^1,2) Various high-strength steels such as dual-phase steel,^3,4) transformation-induced plasticity steel,^5,6) martensitic steel,⁷⁾ and complex phase steel⁸⁾ have been developed. Compared with other high-strength steels, complex phase steel has excellent combination between strength and formability making broad application prospects in the manufacturing of automotive chassis structures such as torsion beams and control arms.^9,10) The microstructure of complex steel is ferrite and bainite with a small amount of martensite and retained austenite. Obviously, the bainite can alleviate the performance difference between soft-phase ferrite and hard-phase martensite.^11,12) However, the performance of the structural components not only depends on the properties of the materials but also the properties of the welded joints.

In recent years, the weldability^13,14) of high-strength complex phase steel, as well as the microstructure^15,16) and properties^17,18,19,20) of welded joints, have been widely investigated. M. Zhang et al.¹¹⁾ studied the weldability of 3.5 mm thick Ti microalloyed steel CP800. Results showed that no cold or low-temperature cracks were detected in the weld seam and HAZ after welding. The tensile strength of the joint was greater than 620 MPa, meeting the usage requirements. X. Zhu et al.¹⁵⁾ and P. Svec et al.¹⁶⁾ all found the hardening of the HAZ when studying the 1 mm and 1.5 mm thick CP780 laser welded joints, which is related to the formation of lath martensite and bainite. Besides, P. Svec et al.¹⁶⁾ also found a significant softening in the subcritical zone with the increase of welding heat input, which was formed by the coarsening of grains. L. Hu et al.¹⁸⁾ conducted a study on the mechanical properties of 3 mm thick CP1000 laser welded joints, and the results showed that the toughness of the fusion zone was higher than that of the base material and fusion line. The study by L. Yang et al.²⁰⁾ showed that the hardness of the 2.8 mm thick CP800 welded joints under different heat inputs was higher than that of the base metal, with inverse trend of elongation. When the heat input was 5.25 kJ/cm, the elongation of the welded joint was the best, reaching 76.8% of the base metal. From the above research, it can be concluded that the thin gauge complex steel has good weldability, but the HAZ may undergo softening or hardening after welding, resulting in a decrease in performance compared to the base metal. Although the weldability of thin gauge multiphase steels have been extensive investigated, there is still a lack of systematically analysis on the mechanism of changes in hardness, microstructure, and properties in the HAZ, especially for microalloyed complex phase steel.

In this paper, a thermomechanical simulator simulator Gleeble 3800 was used to conduct heat treatment at different temperatures on the 800 MPa-grade multiphase steel with microalloyed elements Ti, Nb, and V added, to simulate the thermal cycling process experienced by each HAZ during welding. The effects of the peak temperature of the thermal cycling on the hardness, microstructure, precipitates, and properties of the sample were systematically analyzed. Moreover, the increment of dislocation strengthening, precipitation strengthening, fine grain strengthening, and phase transformation strengthening were calculated to determine the dominant factors to the strength changes at each peak temperature, which can provide theoretical support for studying and improving the welding properties of high-strength complex phase steel.

2. Materials and Methods

The 2.6 mm thickness 800 MPa grade multiphase steel 780FB with Ti, Nb, and V added was used in this study, and the specific chemical composition and mechanical properties are shown in Tables 1 and 2, respectively. The chemical composition of Nb, Ti and V used in this study was the middle value of the range listed in Table 1. The phase transformation curve of 780FB was measured using a DIL402C thermal dilatometer, and the Ac1, Ac3 and Ar3 measured using the tangent method were 740.6°C , 901.8°C and 831°C, respectively.

Table 1. The chemical composition of 780FB (mass%).

C	Si	Mn	Cr	Nb	Ti	V	N
0.05	0.59	1.73	0.47	0.036–0.06	0.062–0.090	0.010–0.030	0.0043

Table 2. The mechanical properties of 780FB.

Yiele strength/MPa	Ultimate tensile strength/MPa	Total elongation/%
730	818	18.5

The thermomechanical simulator Gleeble 3800 was used to simulate the thermal cycles experienced by various HAZ during the welding process. The peak temperatures (Tp) of the thermal cycles were chosen between 500°C–1320°C, to study the changes in hardness, microstructure, and mechanical properties at different positions in the welding HAZ. The sub-regions of the HAZ corresponding to each Tp were listed in Table 3. 500°C and 650°C (below Ac1) corresponded to the subcritical zone, 800°C (Ac1–Ac3) corresponded to the intercritical zone, 980°C (above Ac3) corresponded to the fine-grained zone, and 1320°C (above Ac3) corresponded to the coarse-grained zone. In the thermal simulation test, the temperatures of the samples were measured by a type-K thermocouple, which was welded to the center of the samples.

Table 3. The HAZ corresponding to the peak temperature of the thermal cycle curve.

HAZ	CGHAZ	FGHAZ	ICHAZ	SCHAZ
Temperature/°C	1320	980	800	650, 500

Note: CG: coarse-grian; FG: fine-grain; IC: inter-critical; SC: sub-critical.

The samples were designed as shown in Fig. 1 to ensure that the specimen fractured in the corresponding HAZ during the tensile test and that the longitudinal axis of the samples was parallel to the rolling direction. Adopting a two-step heating method, which first heated at a rate of 300°C/s to 50°C below the peak temperature, and then changed to a rate of 50°C/s to the peak temperatures and isothermal for a certain period. When the peak temperature was 800°C–1320°C, the isothermal time and t_8/5 (cooling time from 800°C to 500°C) were 1 second and 15 seconds, respectively. The isothermal time were 8 and 10 seconds when the peak temperatures were 500°C and 650°C, respectively. The cooling curves were calculated based on the Rykalin-2D formula. The choice of the peak temperature, isothermal time, and t_8/5 for each zone was based on the measurement results of mixed gas shielded welding with a heat input of 3.89 kJ/cm. The actual measured thermal cycle curves in thermal simulation process were shown in Fig. 2.

Fig. 1. The schematic diagram of the thermal simulation samples.

Fig. 2. Thermal simulation curves with different peak temperatures in experimental. (Online version in color.)

After thermal simulation, the middle parts of the samples were removed for mechanical grinding and polishing. 4 vol% nital solution was used to etch the samples for microstructure observations via JSM-7001F scanning electron microscope (SEM), and the software Image Pro was used to calculate the fraction of each phase in the microstructure. The full width at half maximum (FWHM) and retained austenite in 780FB and the samples with different peak temperatures were characterized by X-ray diffraction (XRD). For electron backscattered diffraction (EBSD) analysis, the samples were electrolytically polished at 15 V in a solution of 10 vol% perchloric acid in alcohol. EBSD mappings were performed by TESCAN GAIA3 (accelerating voltage: 20 kV, step size: 0.15 μm for low magnification and 0.06 μm for high magnification) to acquire the crystal structure information, and then analyzed by AZtecCrystal. The precipitates were extracted by carbon extraction replica, and characterized by transmission electron microscopy (TEM). The types of precipitates were determined by energy-dispersive X-ray spectroscopy (EDX). The average number of precipitates under 10 fields of view was used as the precipitation density. When calculating the initial precipitation temperatures of the precipitates, the element with a low mass fraction was used as the benchmark, and the other element was calculated using the ideal chemical ratio. The surface oxide skin of the samples was ground off before the tensile test, and the average value of three tests was taken as the results of yield strength, ultimate strength, and uniform elongation.

3. Results

3.1. Microhardness

The microhardness values of 780FB and the thermal-simulated samples with different T_p were shown in Fig. 3. The average microhardness of 780FB was 269 HV, and slightly decreased to 256 HV when T_p was 500°C. As Tp increased to 650°C, the microhardness significantly increased to 284 HV, even higher than the base metal. When Tp was 800°C, the microhardness decreased again to 240 HV. When T_p was 980°C, the microhardness decreased to 224 HV, indicating a significant softening phenomenon. As Tp increased to 1320°C, the microhardness rose again, reaching to 258 HV.

Fig. 3. Microhardness of 780FB and the thermal-simulated samples at different peak temperatures (The temperature of the base metal was 25°C). (Online version in color.)

3.2. Microstructure

The microstructure of 780FB consisted of ferrite (61.3%), bainite (32.5%), and MA (martensite and austenite) islands (6.20%), with the grain sizes of ferrite and bainite around 2.24 μm, as shown in Fig. 4(a). The width lath bainite was about 500 nm, and the MA islands were distributed at the grain boundary, as shown in the magnified image of Fig. 4(b). The MA islands mentioned in this paper might be martensite, austenite, or a mixture of martensite and austenite.

Fig. 4. (a) SEM and (b) TEM microstructures of 780FB.

When T_p was 500°C, the microstructure was ferrite (61.8%), bainite (34.8%), MA islands (2.6%), and cementite, as shown in Fig. 5(a). Compared with 780FB, the grain sizes of ferrite and bainite have slightly increased, the ferrite content remained unchanged, the bainite content slightly increased, and the MA islands content decreased. The edge of the MA islands became blurred, and the blocky structure was not obvious. This is because the carbon atoms in the MA islands (main) and bainite diffused towards the grain boundaries as the Tp reached 500°C, forming the cementite with sizes of 100–300 nm, as indicated by the white arrow in Fig. 5(b). At the same time, the decarbonized MA islands transformed into bainite. TEM results (Fig. 5(b)) showed that 780FB had an obvious recovery, that was, the bainite laths disappeared or merged so that they cannot be distinguished, but the sub-grain (700 nm) with unclear boundary can be seen. This was the main reason for a slight decrease in the microhardness at 500°C.

Fig. 5. (a) SEM and (b) TEM microstructures of the thermal-simulated samples with Tp=500°C.

When T_p was 650°C, the microstructure transformed to ferrite (61.0%), bainite (36.5%), and cementite, as shown in Fig. 6(a). Compared to 500°C, the size of ferrite and bainite slightly increased (2.37 μm), and the MA islands completely disappeared. This was because the higher the peak temperature, the faster the diffusion velocity of carbon atoms, resulting in the complete transformation of MA islands into cementite. However, 780FB had a low carbon content (0.05%) making the total amount of cementite less. The TEM results indicated that (Fig. 6(b)) the width of the sub-grains continued to increase to approximately 800 nm.

Fig. 6. (a) SEM and (b) TEM microstructures of the thermal-simulated samples with Tp=650°C.

From Fig. 7(a), the microstructure remained ferrite (81.44%), bainite (15.46%), and MA islands (3.1%) as T_p increased to 800°C. MA islands were distributed at the grain boundaries, as shown in the magnified image of Fig. 7(b). However, there was a significant difference in the grain sizes of ferrite, which was consistent with the inter-critical zone of the actual welded joints. When the T_p was between Ac1 and Ac3, partial bainite and MA islands in 780FB were transformed into austenite. During the cooling process, most of the austenite transformed into ferrite, and others remained in the microstructure. The original ferrite size increased during the thermal cycling process, while the newly generated proeutectoid ferrite size was smaller, causing a significant difference in ferrite grain size. The main reason for the decrease in microhardness compared with 780FB was the decrease in the contents of bainite and MA island, while the increase in ferrite.

Fig. 7. (a) SEM and (b) TEM microstructures of the thermal-simulated samples with Tp=800°C.

As T_p increased to 980°C, the microstructure transformed into ferrite and MA islands, as shown in Fig. 8(a). The ferrite was equiaxed with a size of 2.52 μm, indicating the complete disappearance of rolling characteristics. The content of MA islands were about 4.5%, distributed at grain boundaries, as shown in Fig. 8(b). After being heated to 980°C (above Ac3), 780FB was completely austenitized and the microstructure completely changed to austenite, and the austenite grains growth was not obvious for the short residence time above Ac3. Under the experimental cooling rate, small-sized austenite grains tended to transform into ferrite²¹⁾ (Fig. 2). Compared with 780FB, the grains grew slightly, but the ferrite content significantly increased, resulting in the dramatical softening of the microstructure.

Fig. 8. (a) SEM and (b) TEM microstructures of the thermal-simulated samples with Tp=980°C.

The microstructure transformed into bainite with a grain size of 20 μm when T_p was 1320°C, as shown in Fig. 9(a). Compared to 980°C, the longer residence time at high temperatures allowed the more obvious growth of the austenite at 1320°C. Because the coarsened austenite grain first transitioned to the lath bainite transformation zone (Fig. 2), the microstructure transformed into bainite directly with a large grain size. The width of the bainite laths was about 1 μm, and cementites with an average aspect ratio of 7 distributed between bainite laths, as shown in Fig. 9(b).

Fig. 9. (a) SEM and (b) TEM microstructures of the thermal-simulated samples with Tp=1320°C.

3.3. Precipitates

The precipitate morphologies of 780FB and the samples after different thermal cycles observed under TEM were shown in Fig. 10. According to the EDX analysis, the types of precipitates in each sample remained unchanged, including circular VC, elliptical or circular (Ti,Nb)C, and a small amount of square TiN. However, as the peak temperature increased, the Ti content in (Ti,Nb)C increased.

Fig. 10. Precipitates morphologies of 780FB and the samples with different thermal cycles.

According to the statistics of the size distributions and densities of precipitates in 780FB and the samples after different thermal cycles, it can be seen that the densities of VC and (Ti,Nb)C changed with the change of T_p, and the TiN size increased with T_p increased. However, the Tp had little effect on the particle sizes and densities of VC and (Ti,Nb)C. The typical precipitate size distributions were drawn in Fig. 11. The average size of VC in 780FB was 5.25 nm with a density of 43/μm² (Fig. 11(a)), the (Ti,Nb)C was 22.41 nm with a density of 34/μm² (Fig. 11(b)), and the TiN is 59.02 nm. When T_p was 500°C, the types and densities of precipitates little changed. When Tp was 650°C, the sizes and densities of (Ti,Nb)C and TiN changed little, while the density of VC increased to 288/μm². As T_p increased to 800°C, the size and density of VC changed little, while the size distribution range of (Ti,Nb)C increased (Fig. 11(d)), with the average particle size of 23.12 nm and density of 43/μm². The average size of TiN was 63.19 nm. When Tp was 980°C, the size of VC changed little with the density of 24/μm², the size distribution range of (Ti,Nb)C continued to increase (Fig. 11(d)) with the density of 58/μm², and the average size of TiN was 75.35 nm. As T_p increased to 1320°C, the sizes of VC and (Ti,Nb)C changed little, while the densities decreased to 13/μm² and 14/μm², respectively. The TiN size increased to 91.32 nm.

Fig. 11. Size distributions of precipitates. (Online version in color.)

3.4. Mechanical Properties

The mechanical properties of 780FB and the samples with different thermal cycles were shown in Fig. 12. Obviously, the changing trend of ultimate strength and yield strength with T_p variation was consistent, but opposite for uniform elongation. Compared with 780FB, the ultimate strength and yield strength were slightly decreased when T_p was 500°C, but increased at 650°C, even exceeding 780FB. The uniform elongation was only 4.36% at 650°C, less than half of 780FB. The ultimate strength and yield strength continued to decrease at 800°C and 980°C and reached the lowest values at 980°C, which were 57% and 83% of 780FB, respectively. When T_p rose to 1320°C, the ultimate strength and yield strength slightly increased again, but still lower than 780FB.

Fig. 12. Mechanical properties of 780FB and the samples with different thermal cycles: (a) yield strength and ultimate strength; (b) uniform elongation. (Online version in color.)

4. Discussions

4.1. The Effects of Peak Temperature on the Precipitates

The potential precipitates in the microstructure included TiN, TiC, NbN, NbC, VC, and VN. Based on the solubility product formulas (1), (2), (3), (4), the maximum initial precipitation temperatures for TiN, TiC, NbN and NbC were calculated as 1593°C, 980°C, 994°C and 864°C. Due to the mass fraction ratio of Ti to N in 780FB being 17.67, which was much higher than the ideal chemical ratio of 3.43 for TiN, and the highest precipitation temperature of TiN, there was no remaining (or very little) N bound to Nb or V. Considering the precipitation in austenite, the initial precipitation temperatures of TiC, NbC, and VC in ferrite were calculated as 831°C (Ar3), 831°C (Ar3), and 722°C by formulas (5), (6), (7).

The solubility product formulas of TiN²²⁾ in austenite was

  

$$log{[(x_{Ti}\cdot x_N)]}_{\gamma }=3.94-15\mathrm{\hspace{0.17em} }190/T,$$(1)

where x_Ti (wt.%) and x_N (wt.%) were the solid solubility of Ti and N atoms, respectively; T (K) was the temperature. The solubility product formulas of NbN²²⁾ in austenite was

  

$$log{[(x_{Nb}\cdot x_N)]}_{\gamma }=2.80-8\mathrm{\hspace{0.17em} }500/T,$$(2)

where x_Nb (wt.%) was the solid solubility of Nb atoms. The solubility product formulas of TiC²³⁾ in austenite was

  

$$log{[(x_{Ti}\cdot x_C)]}_{\gamma }=2.75-7\mathrm{\hspace{0.17em} }000/T,$$(3)

Where x_C (wt.%) was the solid solubility of C atoms. The solubility product formulas of NbC²²⁾ in austenite was

  

$$log{[(x_{Nb}\cdot x_C)]}_{\gamma }=3.42-7\mathrm{\hspace{0.17em} }900/T.$$(4)

The solubility product formulas of TiC²⁴⁾ and NbC²⁴⁾ in ferrite were

  

$$log{[(x_{Ti}\cdot x_c)]}_{\alpha }=4.40-9\mathrm{\hspace{0.17em} }575/T,$$(5)

  

$$log{[(x_{Nb}\cdot x_c)]}_{\alpha }=3.90-9\mathrm{\hspace{0.17em} }930/T.$$(6)

The solubility product formulas of VC²⁵⁾ was

  

$$log(x_V\cdot x_C)=4.55-8\mathrm{\hspace{0.17em} }300/T,$$(7)

where x_V (wt.%) was the solid solubility of V atoms.

When the peak temperatures were 500°C and 650°C, no precipitate dissolved, and no significant growth of the precipitates due to the low temperatures. The possible reason for the drastic increase of VC density (288/μm²) at 650°C compared to 500°C (43/μm²) was that 650°C corresponds to the shortest time for VC precipitation.²⁶⁾

800°C was located in the precipitation temperature range of TiC and NbC hence promoting new (Ti, Nb)C precipitated. Moreover, 800°C was higher than the initial precipitation temperature of VC, causing some VC in 780FB to dissolve, but re-precipitated during the cooling process. Therefore, after undergoing a thermal cycle with a peak temperature of 800°C, the amount of VC did not change much, while the density of (Ti,Nb)C increased from 34/μm² to 43/μm².

Compared to 800°C, more VC was dissolved during the heating stage at 980°C, and there was not enough time for complete reprecipitation during the cooling stage, causing a decrease in VC density from 43/μm² to 24/μm². Moreover, due to the higher peak temperature, enabling new (Ti,Nb)C precipitated and grew more sufficient. (Ti,Nb)C density increased from 34/μm² to 58/μm² and the original (Ti,Nb)C size increased from 22.41 nm to 26.37 nm.

The T_p of 1320°C was higher than the initial precipitation temperatures of VC, NbC, and TiC, all of which dissolved in the austenite, except for TiN grew through the Ostwald mechanism. During the cooling stage, a small amount of VC and (Ti,Nb)C precipitated again. Therefore, after the thermal cycle with T_p of 1320°C, the TiN size increased from 59.02 nm to 91.32 nm, and the density of VC and (Ti,Nb)C decreased from 13/μm² and 14/μm².

4.2. The Effect of Peak Temperature on the Strength

The strength of steel was a comprehensive result of solid solution strengthening, dislocation strengthening, precipitation strengthening, fine grain strengthening, and phase transformation strengthening. After undergoing different thermal cycles, the strengthening forms varied in different degrees except for solid solution strengthening.

The strength increment (Δσ_d, MPa) caused by dislocation density could be calculated by formula (8),^27,28)

  

$$\mathrm{\Delta }{\sigma }_d=2\alpha Gb\mathrm{\Delta }{\rho }^{1/2},$$(8)

where α was a constant, 0.5; G was the shear elastic module, 80650 MPa; b was the magnitude of the Burgess vector, 0.248 nm; Δρ (/m²) was variation of the dislocation density. The XRD patterns of 780FB and the samples with different peak temperatures were shown in Fig. 13(a), and the corresponding dislocation densities were calculated by the W-H method²⁹⁾ based on the FWHM. The calculated results were listed in Table 4, and the corresponding strength variations were plotted in Fig. 14. The calculated results showed that after different thermal cycles, the dislocation density of 780FB decreased, and the corresponding strength increment decreased. The dislocation density decreased most with 3.19×10⁶/mm² at 980°C, the corresponding strength increment decreased by 35.7 MPa. According to section 3.2, the decrease in dislocation density at 500°C and 650°C was due to the recovery of bainite, while at 800°C and 980°C was due to the partial or complete recrystallization and transformation into ferrite. At 1320°C, 780FB was completely recrystallized, but bainite was formed during the cooling process, resulting in a smaller decrease in dislocation density compared to 980°C.

Fig. 13. XRD patterns of 780FB and the samples with different peak temperatures: (a) before tentisile testing; (b) after tensile testing. (Online version in color.)

Table 4. The reduction of dislocation density (/mm²) and the average effective grain size (μm) of different samples.

Samples	780FB	500°C	650°C	800°C	980°C	1320°C
Dislocation density	1.50×107	8.41×106	7.75×106	7.20×106	5.76×106	1.16×107
Effective grain size	2.24	2.37	2.43	2.52	2.94	19.89

Fig. 14. The effects of peak temperature on the variations of strength increment. (Online version in color.)

Combined with formula (9)³⁰⁾ and the sizes and quantities of precipitates (section 3.3), the strength increment (Δσ_p, MPa) caused by precipitates after different thermal cycles could be calculated. The corresponding strength variations were plotted in Fig. 14.

  

$$\mathrm{\Delta }{\sigma }_p=\frac{Gba}{4r}ln\left(\frac{d-2r}{2b}\right)\left[\frac{1}{(d-2r)/2}\right],$$(9)

Where a was the constant, 2; r (nm) was the radius of the precipitates; and d (nm) was the distance between precipitates. Obviously, the factors affecting the precipitation strengthening increment were the particle size and density of the precipitates. From section 3.3, the VC with 5.25 nm significantly increased at 650°C, thus the corresponding precipitation strengthening increment (49.84 MPa) was the largest. Most of the (Ti,Nb)C and VC disappeared due to dissolution at 1320°C, causing a sharp decrease (26.86 MPa) in precipitation strengthening increment.

The average effective grain size of 780FB and the samples with different thermal cycles were listed in Table 4. The strength increment (MPa) caused by changes in grain size could be calculated by formula (10),³¹⁾ and the corresponding strength variations were shown in Fig. 14.

  

$$\mathrm{\Delta }{\sigma }_g=k_y{\overline{d}}^{-1/2},$$(10)

where k_y was the proportional coefficient related to the grain size, 17.4 MPa·mm^1/2 above 3 μm, 10 MPa·mm^1/2 below 3 μm; $\overline{d}$ was the average effective grain size. The calculation results indicated that the grain size had a significant impact on the increment of fine grain strengthening. After the thermal cycle at T_p of 980°C, the grain size only increased by 0.7 μm, but the fine grain strengthening increment decreased by 26.86 MPa. The grain size increased by 17.65 μm at T_p of 1320°C, the fine grain strengthening increment sharply decreased by 87.91 MPa.

Combined with formula (11)^32,33) and the changes of bainite laths (Figs. 4(b)–9(b)), the transformation strengthening increment (Δσ_pt, MPa) could be calculated. The corresponding strength variations were shown in Fig. 14.

  

$$\mathrm{\Delta }{\sigma }_{pt}=115\times C_B/\overline{L},$$(11)

Where C_B was the content of bainite, and $\overline{L}$ (μm) was the width of the bainite lath. Calculated results showed that the phase transition of the samples with different thermal cycles had a significant impact on the changes in the strength increment. Even if the microstructure changes little after heat treatment at 500°C (the width of bainite lath increased from 500 nm to 700 nm), the decrease of strength increment caused by phase transformation was still 17.58 MPa. At T_p of 980°C, the decrease in strength increment caused by phase transformation was maximum with 74.75 MPa because the bainite nearly completely transformed into ferrite, which was the main reason for the decrease in strength. When T_p rose to 1320°C, the microstructure all transformed into bainite making the strength increased by 51.62 MPa.

The difference between the calculated comprehensive effect of each strength increment variation of the samples with different thermal cycles and the experimental results did not exceed 10%, indicating a good agreement between the two results, as shown in Fig. 14. It was thus clear that the calculated results of each strength increment change in this paper are reliable.

4.3. The Effect of Peak Temperature on the Uniform Elongation

As we all know, the retained austenite in the microstructure can effectively coordinate deformation and improve the uniform elongation during the tensile test. The retained austenite contents in 780FB and the samples with different thermal cycles before and after tensile testing were quantitatively calculated by Figs. 13(a) and 13(b), respectively, and the calculated results were listed in Table 5. There was no retained austenite in the samples with Tp of 650°C and 1320°C. The retained austenite contents of the samples with Tp of 500°C and 800°C were no changed after the tensile testing, indicating that the retained austenite had no effect on the uniform elongation. For 780FB and the sample with Tp of 980°C, the retained austenite contents of decreased from 2.6% and 1.9% to 1.8% and 1.7%, respectively. It meant some retained austenite transformed into martensite during the tensile test, which triggered the TRIP (Transformation induced plasticity) effect and contributed to the uniform elongation.

Table 5. Retained austenite contents in 780FB and the samples with different thermal cycles.

Samples	780FB	500°C	650°C	800°C	980°C	1320°C
Before tensile testing	2.6%	1.2%	0	1.5%	1.9%	0
After tensile testing	1.8%	1.2%	–	1.5%	1.7%	–

Figure 15 was the grain boundary misorientation maps of 780FB and the samples with different peak temperatures. When T_p was less than 980°C, as T_p increased, the proportion of high-angle grain boundaries gradually increased. The increase in high-angle grain boundaries increased the effective hindrance of dislocation slip during the tensile test and improved the uniformity of microzone deformation. However, due to the TRIP effect of 780FB, the uniform elongation of the sample with Tp of 500°C slightly increased compared with 780FB. For the sample with Tp of 650°C, a large amount of VC precipitation promoted the plane-slip of dislocations and limited the cross-slip, which caused the uneven distribution of slip and formed the thick slip bands. Thus, the uniform elongation decreased to 4.36%. When Tp was 800°C, the proportion of high-angle grain boundaries increased to 79.4% and the ferrite content increased to 81.44%, resulting in a higher uniform elongation than 780FB. As Tp rose to 980°C, the contents of high-angle grain boundaries and ferrite significantly increased, coupled with the TRIP effect, making the uniform elongation increased to 18.17%. When T_p was 1320°C, the proportion of small angle grain boundaries significantly increased, but the VC content significantly decreased. The comprehensive effects made the uniform elongation lower than the T_p of 980°C, but higher than 780FB.

Fig. 15. Grain boundary misorientation maps of 780FB and thermal-simulated samples at different peak temperatures: (a) 780FB; (b) 500°C; (c) 650°C; (d) 800°C; (e) 980°C; (f) 1320°C. (Online version in color.)

5. Conclusions

In this paper, the influential mechanism of thermal cycles with different peak temperatures on the hardness, microstructure, precipitates, and properties of the 800 MPa-grade complex phase steel 780FB was systematically studied. More importantly, the effects of the peak temperatures on the incremental changes in dislocation strengthening, precipitation strengthening, fine grain strengthening, and phase transformation strengthening were quantitatively calculated, and the contributions of each incremental change to the strength change were clearly explained. The main conclusions were drawn as:

(1) When T_p was 500°C, there was little change in microstructure, strength and uniform elongation compared to 780FB. When T_p was 650°C, a large amount of VC precipitation lead to the dominant effect of precipitation strengthening increment, while also causing the uneven distribution of slip. The strength exceeded 780FB, but the uniform elongation decreased.

(2) When T_p was 800°C and 980°C, partial or all bainite transformed into ferrite, causing the dislocation density decreased and the proportion of high-angle grain boundaries increased. The strength decreased for the reduction of phase transformation strengthening increment and dislocation strengthening increment, while the uniform elongation increased.

(3) When T_p was 1320°C, the microstructure completely transformed into large grain sized bainite and the precipitates significantly decreased, resulting in an increase in phase transformation strengthening increment and a decrease in fine grain strengthening increment and precipitation strengthening increment. With the comprehensive effect, the strength decreased and the uniform elongation increased.

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

• 1)   C.  Song,  H.  Wang,  Z.  Sun,  J.  Xu,  H.  Chen and  W.  Yin: Scr. Mater., 212 (2022), 114583. https://doi.org/10.1016/j.scriptamat.2022.114583
• 2)   D. W.  Fan,  H. S.  Kim and  B. C. D.  Cooman: Steel Res. Int., 80 (2009), 241. https://doi.org/10.2374/SRI08SP131
• 3)   Y.  Qin,  J.  Liu,  K.  Zeng,  B.  Xing and  J.  Jiang: Mater. Trans., 64 (2023), 657. https://doi.org/10.2320/matertrans.MT-M2022155
• 4)   F.  Badkoobeh,  H.  Mostaan,  M.  Rafiei,  H. R.  Bakhsheshi-Rad and  F.  Berto: Metals, 12 (2022), 101. https://doi.org/10.3390/met12010101
• 5)   L.  Qian,  K.  Li,  F.  Huang,  D.  Li,  T.  Wang,  J.  Meng and  F.  Zhang: Scr. Mater., 183 (2020), 96. https://doi.org/10.1016/j.scriptamat.2020.03.029
• 6)   M.  Soleimani,  A.  Kalhor and  H.  Mirzadeh: Mater. Sci. Eng. A, 795 (2020), 140023. https://doi.org/10.1016/j.msea.2020.140023
• 7)   B.  Gao,  L.  Wang,  Y.  Liu,  J.  Liu,  L.  Xiao,  Y.  Sui,  W.  Sun,  X.  Chen and  H.  Zhou: Mater. Sci. Eng. A, 23 (2023), 881. https://doi.org/10.1016/j.msea.2023.145370
• 8)   Y.  Chang,  M.  Lin,  U.  Hangen,  S.  Richter,  C.  Haase and  W.  Bleck: Mater. Des., 203 (2021), 109620. https://doi.org/10.1016/j.matdes.2021.109620
• 9)   Y.  Li,  S.  Kuan,  L.  Jia,  Y.  Wang,  T.  Chen,  W.  Ju,  Y.  Zhang and  Y.  Zheng: Iron Steel Van. Titan., 42 (2021), 184 (in Chinese). https://doi.org/10.7513/j.issn.1004-7638.2021.01.030
• 10)   H.  Sun,  K.  Wei,  X.  Yang,  Z.  Xiao and  Y.  Wu: Int. J. Fatigue, 131 (2019), 105364. https://doi.org/10.1016/j.ijfatigue.2019.105364
• 11)   X.  Li,  A.  Ramazani,  U.  Prahl and  W.  Bleck: Mater. Charact., 142 (2018), 179. https://doi.org/10.1016/j.matchar.2018.05.038
• 12)   B.  Erice,  C. C.  Roth and  D.  Mohr: Mech. Mater., 116 (2018), 11. https://doi.org/10.1016/j.mechmat.2017.07.020
• 13)   M.  Zhang,  Q.  Li,  C.  Huang,  R.  Wu,  R.  Fu,  L.  Li and  P.  Fang: Adv. Mater. Res., 634 (2013), 2899. https://doi.org/10.4028/www.scientific.net/AMR.634-638.2899
• 14)   R.  Koganti and  C.  Orsette: ASME IMECE, 4 (2009), 501. https://doi.org/10.1115/imece2009-13319
• 15)   X.  Zhu,  Z.  Liu,  G.  Yin,  Q.  Zhou,  H.  Qu and  N.  Li: Weapon Mater. Sci. Eng., 45 (2022), 120 (in Chinese). https://doi.org/10.14024/j.cnki.1004-244x.20220801.003
• 16)   P.  Svec,  A.  Schrek and  M.  Domankova: Kov. Mater., 56 (2018), 29. https://doi.org/10.4149/km_2018_1_29
• 17)   H.  Kun,  Z.  Weiming,  Z.  Mei and  L.  Lin: Hot Work. Technol., 40 (2011), 31 (in Chinese). https://doi.org/10.3969/j.issn.1001-3814.2011.07.010
• 18)   L.  Hu,  Q.  Yan,  J.  Huang and  Y.  Wu: Adv. Mater. Res., 97 (2010), 21. https://scientific.net/AMR.97-101.3957
• 19)   B.  Njock,  P.  Kah,  B.  Mvola,  P.  Layus and  X.  Cai: AAIC, 65 (2020), 159. https://doi.org/10.5937/zzk2004159N
• 20)   L.  Yang,  X.  Wang,  P.  Huan,  Z.  Liu,  M.  Zhang,  C.  Chen and  Z.  Hu: Appl. Laser, 38 (2018), 4 (in Chinese). https://doi.org/10.14128/j.cnki.al.20183804.562
• 21)   X.  Wang,  X.  Li and  C.  Shang: The 5th Chinese Youth Symposium on Materials Science and Technology, (2011), 47 (in Chinese). https://www.docin.com/p-1214448459.html
• 22)   K.  Narita: Trans. ISIJ, 15 (1975), 145. https://doi.org/10.2355/isijinternational1966.15.145
• 23)   L.  Yong,  Z.  Liu,  X.  Sun,  J.  Cao,  X.  Zha and  Y.  Zhang: Iron Steel Van. Titan., 26 (2005), 12 (in Chinese). https://doi.org/10.3969/j.issn.1004-7638.2005.02.005
• 24)   K. A.  Tailor: Scr. Mater., 32 (1995), 7. https://doi.org/10.1016/S0956-716X(99)80002-8
• 25)   J. A.  Todd and  P.  Li: Metall. Trans., 17A (1986), 1191. https://doi.org/10.1007/BF02665318
• 26)   Y. J.  Zhang,  G.  Miyamoto,  K.  Shinbo,  T.  Furuhara,  T.  Ohmura,  T.  Suzukic and  K.  Tsuzaki: Acta Mater., 84 (2015), 375. https://doi.org/10.1016/j.actamat.2014.10.049
• 27)   M. S.  Park: J. Mech. Sci. Technol., 28 (2014), 2587. https://doi.org/10.1007/s12206-014-0615-3
• 28)   W.  Sun,  S.  Hu,  G.  Li and  H.  Yu: HSLA Steels, 91 (2016), 733. https://doi.org/10.1007/978-3-319-48767-0_91
• 29)   S.  Takebayashi,  T.  Kunieda,  N.  Yoshinaga,  K.  Ushioda and  S.  Ogata: ISIJ Int., 50 (2010), 875. https://doi.org/10.2355/isijinternational.50.875
• 30)   J.  Cao,  Q.  Yong,  Q.  Liu and  X.  Sun: J. Mater. Sci., 42 (2007), 10080. https://doi.org/10.1007/s10853-007-2000-4
• 31)   K.  Ishida,  T.  Matsuo and  M.  Imafuku: Metal Sci. Forum, 941 (2018), 218. https://doi.org/10.4028/www.scientific.net/MSF.941.218
• 32)   C. H.  Young and  H. K. D. H.  Bhadeshia: Mater. Sci. Technol., 10 (1994), 209. https://doi.org/10.1179/mst.1994.10.3.209
• 33)   X.  Lei,  S.  Dong,  J.  Huang,  J.  Yang,  S.  Chen and  X.  Zhao: Mater. Sci. Eng. A, 718 (2018), 437. https://doi.org/10.1016/j.msea.2018.02.008