Strategies for strengthening-ductility and hierarchical

Materials Science & Engineering A 739 (2019) 225–234

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Strategies for strengthening-ductility and hierarchical co-precipitation in multicomponent nano-precipitated steels by Cu partitioning

T

Jiacheng Yana,b, Hongwei Xua,b, Xiaowei Zuoa,b, , Tao Jiac, Engang Wanga,b ⁎

a

Key Laboratory of Electromagnetic Processing of Materials (Ministry of Education), Northeastern University, Shenyang 110819, China School of Metallurgy, Northeastern University, Shenyang 110819, China c State Key Lab of Rolling and Automation, Northeastern University, Shenyang 110819, China b

ARTICLE INFO

ABSTRACT

Keywords: Ultra-high-strength steel Co-precipitates Ductility Strength Hierarchical structure Intercritical heat treatments

We investigated the dependence of both the strength and ductility in Cu-bearing nano-precipitated steels on compositions, tempering and aging, where Fe-2Cu (wt%, Cu-steel) steel by Cu-rich precipitates (CRPs), Fe-2Cu5Ni-3Mn-1.5Al steel (Cu/Ni steel) by co-precipitates of CRPs+NiAl, and Fe-2Cu-5Ni-3Mn-1.5Al-1Mo (Mo-steel) steel by co-precipitates of CRPs+NiAl+Mo2C precipitates were paid special attentions on. The strength and ductility of Cu-steel were sensitive to tempering temperatures rather than aging time. Tempering and aging in Cu/Ni steels remarkably increased the strength with seriously sacrificing of ductility because larger-sized blocky CRPs segregated along the grain boundaries where Cu partitioning along grain boundaries was partially dominate, resulting in brittle fracture in almost all of aged Cu/Ni steels. Mo-steel, however, showed superior strength and ductility than Cu/Ni steel because Mo addition reduced the possibility of blocky CRPs by decreasing Cu partitioning along grain boundaries, and promoted the formation of hierarchical co-precipitates where largesized coarsening co-precipitates contributed to the ductility and small-sized secondary co-precipitates had significant strengthening, thus resulting in the improvement of strengthening-ductility.

1. Introduction The increase in strength is generally accompanied by the loss in both ductility and toughness in metallic materials and the trade-off is one of critical issues for R&D of ultrahigh-strength Cu-bearing steels [1–3], which have desirable applications in aerospace, power generation, bridges, ship building and automotive industries [4–6]. Nanoscale co-precipitation, which involves different kinds of precipitates such as Cu-rich precipitates (CRPs), NiAl nano-particles and M2C particles, is the most powerful way to improve both strength and ductility with superior combination of mechanical, welding and irradiation properties [4,7–10]. The strengthening from nanoparticles precipitates by impeding the movements of dislocations [7,11,12] is highly dependent upon microstructural characterizations of the precipitates involved the structure, morphology, size, number density, and interparticles spacing [13–16]. Specifically, coherent nanoparticles in the body-centered cubic (bcc) ferritic/martensitic steels have recently been realized to provide extremely high strengthening responses by precipitating out precipitates with sufficiently fine scales (less than 5 nm in diameter) [4]. Although high density nanoscale co-precipitation particles, which are almost coherent with ferritic/martensitic matrix, and Cu-rich

⁎

precipitates (CRPs) for ultrahigh-strength Cu-bearing steels [17] are inevitably beneficial for strengthening by precipitation strengthening, the balance between strength and ductility might be broken and the ductility changes abnormally because of the presence of abundant precipitates. In Fe–2.5Cu alloy (weight percent, similarly hereinafter), the presence of CRPs by aging remarkably increased the ultimate tensile strength (UTS) from 380 MPa to 720 MPa with slightly decreasing the elongation (EL) from 36% to 23% [12]. The co-precipitation of nanoscale CRPs and NiAl particles by aging in the 1.5Cu-3.25Ni-Al steels displayed strength of 1053 MPa and EL-to-failure of 10.1% [18], demonstrating that CRPs and NiAl particles significantly enhanced the age-hardening ability. The co-precipitation of CRPs and Ni(Al,Mn) particles in the 1.5Cu-3.25Ni-Al-Mn steels, however, showed strength of 1300 MPa and ductility of 11% [18], suggesting that Mn promoted to partition the NiAl nanoparticles and increased the particle number density. The steel with the composition of Ni-2Al-3Mn-1.5Cu-1.5Mo1.5W-0.07 Nb exhibited an UTS of 1.9 GPa, an EL-to-failure of 10% and a reduction in area of 40% [7], indicating a good combination of ultrahigh strength and good ductility, where the precipitation of Nb(C,N) and (Nb,Mo)C carbides precipitated during annealing refined the grain size and inhibited the recovery of dislocations [19,20]. Meanwhile,

Correspondence to: Northeastern University, No.3-11, Wenhua Road, Heping District, P.O. Box 314, Shenyang 110819, Liaoning Province, China. E-mail address: [email protected] (X. Zuo).

https://doi.org/10.1016/j.msea.2018.10.036 Received 21 August 2018; Received in revised form 8 October 2018; Accepted 9 October 2018 Available online 11 October 2018 0921-5093/ © 2018 Elsevier B.V. All rights reserved.


J. Yan et al.

M2C carbide precipitates and cementite or M7C3 also played important roles in influencing the second burst of nucleation of CRPs [10,21,22], where carbide precipitates acted as nucleation sites for CRPs and thus increased the density of heterogeneous sites for nucleation of CRPs. Although the M2C carbides precipitated slower than CRPs [21], the precipitation of M2C particles offset the loss of strength during the process of aging. The aged Fe-5Ni-1Al-5Mn steel showed strength of 1308 MPa and uniform EL of less than 1%, where more Mn addition resulted in the transformation from NiAl-type nanoparticles to Ni2AlMn-type Heusler nano-precipitates and grain boundary (GB) embrittlement by Mn enrichment, promoting brittle cleavage and intergranular fractures [23]. Additionally, the aged Fe-0.012C-6.84Mn2.44Ni-1.63Al-1.36Cu steel had strength of 1215 MPa and uniform EL of 0.9%, where Mn diffusion and segregation played influential roles in the low EL [24]. These results explicitly showed the composition-dependent relationship between strength and ductility. A detailed data summary can be found in the Supplementary Table S1. Therefore, it is still one of interests to reveal and balance the strength and ductility in multicomponent nano-precipitated steels. Applying intercritical heat treatments (IHTs) [25] or quenchingpartitioning -tempering (QPT) [24] is one of strategies to improve the trade-off between strength and ductility by the formation of the dual hierarchical structure [4,7,23,25,26]. The strategies were designed as [26]: (1) coarsening of precipitates due to heat treatment at a higher temperature, (2) the recovery of the matrix, (3) formation of reverted austenites, and (4) reduced GB segregation of impurities and Cu. By carefully tuning the size and number density and diffusion pathways of related solute atoms via the optimization of heat treatments, the co-precipitated features may yield desired high strength without significantly degrading both impact toughness and ductility [25,27]. The co-precipitation of nanoparticles in dual phases was thought to be closely related to the cooperative austenite reversion process, which was believed to be governed by Mn diffusion and segregation, leading to a different element enrichment in dual phases [24,28]. However, the interactions and elemental segregations among Mn, Cu, Ni and other additional elements are exceptionally complicated, and it is necessary to carefully explore the mechanism for strengthening-ductility and coprecipitation in multicomponent nano-precipitated steels. In this paper, Fe-2Cu (wt%, Cu-steel) steel strengthened by Cu-rich precipitates (CRPs), Fe-2Cu-5Ni-3Mn-1.5Al steel (Cu/Ni steel) by coprecipitates of CRPs+NiAl, and Fe-2Cu-5Ni-3Mn-1.5Al-1Mo (Mo-steel) steel by co-precipitates of CRPs+NiAl+Mo2C precipitates were subjected to different IHTs. The dependence of both the strength and ductility in the multicomponent nano-precipitated steels on compositions and IHTs (tempering and aging) was emphatically investigated by tensile tests, fracture characterization using scanning electron microscopy (SEM), and the morphology and compositions of precipitates characterized using transmission electron microscopy (TEM) and 3D atom-probe tomography (APT). Cu partitioning was carefully placed emphasis on, which might be one of determinable elements to optimize strengthening-ductility and co-precipitation in our experimental steels.

Table 1 Chemical compositions of the experimental steels. Alloy Cu-steel Cu/Ni-steel Mo-steel

wt% at% wt% at% wt% at%

Cu

Ni

Al

Mn

Mo

C

Fe

2.07 1.82 1.88 1.62 1.90 1.64

– – 5.01 4.68 5.02 4.70

– – 1.98 4.03 2.13 4.34

– – 3.38 3.37 3.37 3.37

– – – – 0.92 0.53

0.01 0.05 0.03 0.14 0.04 0.18

Bal. – Bal. Bal. Bal. –

value was calculated on at least 3 samples. The fractures of the tensile specimens were observed on the Zeiss scanning electron microscope. The foils for TEM images were twin-jet polished in an 8% perchloric acid alcohol solution at temperature of −30 ℃ and the voltage of 35 V, and then finally cleaned by argon ion-milling with a voltage of 3 keV at an incidence angle of 4° with a cryogenic stage cooled by liquid nitrogen. The foils were observed on a FEI-Tecnai G20 TEM operating at 200 kV and equipping with an energy dispersive x-ray spectrometer (EDS) system. APT was performed with a Cameca local electrode atom probe (LEAP 3000X HR). Imago Visualization and Analysis Software was used for three-dimensional (3D) reconstruction, composition analyses and the creation of isoconcentration surfaces. The mean radius and number density of precipitates were statistically determined by APT data [29]. 3. Results 3.1. Strength-ductility in nano-precipitated steels 3.1.1. Cu-steel After Cu-steels were quenched, tempered and aged, the engineering stress-strain curves (Fig. 1) show larger fluctuations from the effect of tempering temperatures (Fig. 1b) rather than the aging time (Fig. 1a). The ultimate tensile strength (UTS) as a function of EL of Cu-steels in different cases (Fig. 1c) shows that IHT-Cu-steels have higher EL (larger than 10%) with lower UTS (lower than 600 MPa), and the EL is higher at higher tempering temperatures, which is verified that the fracture show good ductility because of the presence of abundant dimples (Fig. 1d). The morphology of fracture surfaces (Supplementary Fig. S1) at different heat treatments show that the depth and densities of dimples were sensitive to tempering temperatures rather than aging time, suggesting that the strengthening-ductility of Cu-steel was sensitive to tempering temperatures rather than aging time [30,31]. 3.1.2. Cu/Ni steel Compared to Cu-steels, the strength of the Cu/Ni steels is significantly increased because of the co-precipitates of NiAlMn-type and Cu-rich precipitates by the addition of Ni, Al and Mn [16]. The posttempered Cu/Ni-steels at different temperatures without further aging (Fig. 2a) show that the UTS is higher than 700 MPa, and both the strength and EL change greatly with the different tempering temperatures. Specifically, the strength of post-tempered Cu/Ni-steel at 625 °C is highest up to 1100 MPa with a limited EL, however, the UTS of posttempered Cu/Ni-steel at 700 °C decreases to 800 MPa (by 28%) with good ductility of 10%, suggesting that proper tempering treatments are helpful to increase the ductility of post-tempered Cu/Ni-steels. Interestingly, the Cu/Ni steels treated by tempering and aging keep higher strength rather than good ductility (Fig. 2b). The fracture morphology of post-tempered Cu/Ni steels (Fig. 2c) shows a micro-void coalescence fracture mode with a lot of fine dimples [32], presenting a characteristic mode of a ductile fracture [7,12]; post-aged Cu/Ni steel (Fig. 2d), however, shows the flat facets of fracture with quasi-cleavage facets and intergranular cracks, characterizing a typical brittle cleavage fracture [23]. Although the measured compositions are a little far away the true values, SEM-EDS analysis (Supplementary Fig. S2) shows that the

2. Experimental procedure Three experimental steels were respectively casted in a vacuum induction furnace and the chemical compositions were analyzed by optical spectroscopy (Table 1). The steels were hot forged, hot rolled, solution-treated at 900 °C under Ar atmosphere for 1 h and waterquenched. The solution-treated samples were then sectioned and tempered at temperatures from 625 ℃ to 700 ℃ for 10 min, and then aged at 550 °C for ranging from 0 to 240 min under Ar atmosphere. At least 3 samples were experimentally conducted for each heat-treatment case. Tensile tests of the heat-treated samples were performed on Instron5969 universal testing systems at the rate of 1 mm/min and the strain was measured using digital image correlation techniques. The average 226


J. Yan et al.

Fig. 1. (a) The engineering stress-strain curves of Cu-steels, which were tempered at 650 °C and followed by aging temperature of 550 °C for different time; (b) The engineering stress-strain curves of Cu-steels which were tempered at different tempering temperatures, and aged at the temperature of 550 °C for 2 h; (c) UTS and EL of Cu-steels at different IHTs; (d) the fracture morphology of Cu-steel tempered at 650 °C followed by aging at 550 °C for 4 h.

increasing aging time (Fig. 3c), the UTS and EL of Mo-steels show the nucleation/growth process of co-precipitates. The treatments of higher tempering temperatures show quicker nucleation and growth, and the peak aging time is shorter. The post-tempered samples show better ductility than aged samples. The fracture morphology of Mo-steels (Fig. 3d) tempered at 675 °C for 10 min and aged at 550 °C for 4 h shows coalescence fracture mode with a lot of dimples, indicating a characteristic mode of a ductile fracture. The proper tempering temperature for Mo-steels is 650 °C or 675 °C, with the strength of 1200 MPa and the ductility of 7%.

tendency of the matrix of aged Cu/Ni steels is more serious Cu-depleted than post-tempered ones, suggesting more serious Cu-enrich along grain boundaries in aged Cu/Ni steels than post-tempered ones. TEM morphology on post-tempered and aged Cu/Ni steels (Fig. 2e and f) shows that the matrix retains lath-like morphology with substructures after solution treatments [14]. A small amount of fine precipitates are found along the GB in the post-tempered Cu/Ni steel, but coarsening blocky precipitates apart from fine acicular precipitates are found in aged Cu/Ni steel. TEM-EDS also showed the composition of acicular and blocky precipitates was Cu-rich. Acicular CRPs were found in both the post-tempered (Fig. 2e) and aged Cu/Ni-steel (Fig. 2e), which might be irrelevant to the brittle cleavage fracture. However, blocky CRP (BCRP), which was confirmed by the following APT results (Fig. 5a), might be one of the reasons for the poor ductility of aged Cu/Ni steels.

3.2. Co-precipitation of nano-precipitated steels The elemental distribution maps explicitly show the partitioning of elements with compositions. In aged Cu/Ni-steel (Fig. 4a), elemental and interfacial segregations of all elements are obviously found. The atoms of Cu, Ni, Al and Mn severely segregate at almost same sites, indicating the co-precipitation of CRPs and NiAlMn-type precipitates [4,16,23]. In aged Mo-steel, the atoms of Cu, Ni, Al and Mn segregate at almost same sites and show larger segregation (Fig. 4b), however, Mo atoms are homogeneously distributed without apparent partitions. Numerous Mo2C particles nucleated along the CRPs and NiAl precipitates are observed after aging for 40 min, suggesting the heterogeneous nucleation behind co-precipitation of CRPs and NiAl precipitates, which is consistent with previous work [10,21,33]. Isoconcentration surfaces of co-precipitates (Fig. 5a) show that in Cu/Ni-steel a large-sized precipitate is clearly recognized along the GB,

3.1.3. Mo-steel With the addition of a small amount of Mo into Cu/Ni-steels, the strength and ductility of Mo-steels are improved significantly. In the post-tempered Mo-steels (Fig. 3a), the strength increases first and then decreases with increasing tempering temperatures. The values of the EL in tempered Mo-steels are obviously higher than those in Cu-Ni steels (Fig. 2a) except for the samples tempered at 625 ℃, in which the strength increases greatly by 29% compared to the solution-treated Mosteel and the highest EL is 13.5% in the sample tempered at 675 °C. When the post-tempered Mo-steels were superimposed of aging at 550 °C for 40 min, the strength increases monotonously with increasing tempering temperatures and the EL is higher than 6% (Fig. 3b). With 227


J. Yan et al.

Fig. 2. (a) UTS, yield strength (YS) and EL of post-tempered Cu/Ni-steels as a function of tempering temperatures; (b) UTS and EL of Cu/Ni-steels, which were tempered at different temperatures for 10 min and followed by aging at 550 °C for 40 min, as a function of tempering temperatures; the fracture morphology of (c) post-tempered Cu/Ni-steel at 700 °C for 10 min and (d) Cu/Ni-steel tempered at 700 °C for 10 min and followed by aging at 700 °C for 2 h; TEM images of (e) posttempered Cu/Ni-steel at 700 °C for 10 min and (f) Cu/Ni-steel tempered at 700 °C for 10 min and followed by aging at 700 °C for 40 min.

which is consistent with the morphology of BCRP in Fig. 2f. The onedimensional compositional profile of the selected precipitate (Fig. 5c) shows that Cu concentration of the precipitate is as high as 70 at%, indicating that the precipitate reaches its stable stoichiometric Cu composition. Additionally, a small number of Ni and Mn segregates on both sides of the precipitate, demonstrating as one of the large-sized CRP at GB, which might be responsible for the brittle fracture in Cu/Nisteel because of Cu segregation (Fig. 2d). In Mo-steel, the co-precipitates display two kinds of different sizes, one as small-sized precipitates and another as large precipitates (Fig. 5b). The two co-precipitates are found to be hierarchically surrounded by Mo2C carbides in the Mo-steel (Fig. 5b) after intercritical temperature tempering (675 °C) and aging. The co-precipitates of CRPs and NiAl nanoparticles with Mo2C carbides are thought to nucleate firstly surrounding the carbides [25,34], then large-size co-precipitates might coarsen further in the fresh secondary martensitic zone when tempering at intercritical temperatures, named as coarsening co-precipitates (CCPs). Further aging promoted fine-dispersed secondary co-precipitates (SCPs) to precipitate

hierarchically from the supersaturated matrix [35], forming hierarchical structure, where the CCPs might contribute on the good ductility and fine SCPs contribute to strengthening. The formation of hierarchical structure might be responsible for the improvement of Mosteels compared with Cu/Ni-steels. The one-dimensional composition profile of the selected CRP (Fig. 5d) shows that the co-precipitates are neighbor-like structure and the NiAl precipitates are gradually encircled by CRPs during the coarsening. The size and number density of the co-precipitations are two important parameters which are closely related the strengthening effects. The mean radius and number density of Cu and NiAl-type particles in aged Cu/Ni-steel and Mo-steel (Table 2) were quantitatively analyzed by setting a critical mean radius of 3 nm because two different sized coprecipitates are divided by about 3 nm (Fig. 6b). In Cu/Ni-steel, the densities of fine Cu and NiAl-type particles is 1.68 × 1023/m3 and 1.19 × 1024/m3, respectively. The mean radius of fine Cu and NiAl particles are 1.54 nm and 0.90 nm, respectively. When adding Mo into Cu/Ni-steel, the densities of fine Cu and NiAl-type particles decrease by 228


J. Yan et al.

Fig. 3. (a) UTS, YS and EL of post-tempered Mo-steels as a function of tempering temperatures; (b) UTS and EL of Mo-steels, which were tempered at different temperatures for 10 min and followed by aging at 550 °C for 40 min, as a function of tempering temperatures; (c) UTS and EL of Mo-steels, which were tempered at different temperatures for 10 min and followed by aging at 550 °C, as a function of aging time; (d) The fracture morphology of Mo-steel tempered at 675 °C for 10 min, followed by aging at 700 °C for 4 h.

80% (3.30 ×1022/m3) and 24% (9.03 ×1023/m3), respectively. The average radius of fine Cu and NiAl particles are 1.02 nm and 0.84 nm in the Mo-steel, respectively, and the fine Cu and NiAl-type particles in Mo-steel are less than those in Cu/Ni-steel, which shows remarkable inhibition by Mo addition. Results in Table 1 also show that the size of precipitates of tempered Mo-steel is much different, compared with the untempered Mo-steel [1,16]. The mean radius of small size precipitates are similar in both tempered and untempered Mo-steel, which indicates that these types of precipitations are formed during aging process. However, the big precipitates only exist in tempered Mo-steel, demonstrating that these precipitates are formed in two-step heat treatments. First, tempering step at 675 ℃ may lead to co-nucleation of Cu and NiAl particles, and then these precipitations continue to grow and coarsen during aging process. The varying size and hierarchical arrangement of CRPs indirectly demonstrate that they nucleate at different steps during multistage heat treatments [25]. Thus, histogram of radius distribution is used to show the specific numbers in different radius range (Fig. 6). We only choose the small-size (Rp < 3 nm) precipitates for studying the co-precipitations behavior during aging process. Within CRPs in Cu/Ni-steel (Fig. 7a), Cu atoms segregate towards the cores of CRPs with the concentration of 80 at%. Specifically, the concentrations of Ni and Mn are 23 at% and 13 at%, respectively. The concentration-peak positions of Ni and Mn are away from the cores for 4 nm and 2.5 nm, respectively, indicating obvious interfacial segregations. In Mo-steel, however, the concentration of Cu greatly decreases to 40%, and the segregation of Ni and Mn displays similar enrichment patterns with 30% and 10%. The concentration peaks reduce to half, but the peak concentration positions of Ni and Mn are away from the

cores for 2 nm, showing a slight interfacial segregation. These suggest that CRPs in Mo-steel is less than those in Cu/Ni steel. Within NiAl precipitates in Cu/Ni-steel (Fig. 7b), Cu atoms segregate towards the cores of NiAl precipitates with the concentration of 27 at%. The concentrations of Ni and Mn are 37 at% and 21 at%. The concentration of Cu and Mn reach peaks at the central cores, showing slight interfacial segregations. In Mo-steel, Cu atoms segregate towards the cores of NiAl precipitates with the concentration of 23% which displays similar enrichment patterns. However, the concentration of Ni increases to 50 at% and the concentration of Mn keeps 20 at%. The peak concentration positions of Cu are away from the cores for 4 nm, showing interfacial segregations. Although Cu, Ni and Mn are changeable with increasing aging time, the concentration of Mo keeps invariable no matter in CRPs and NiAl precipitates, further showing that Mo might restrictedly affect the nucleation growth of co-precipitation of CRPs and NiAl precipitates. 4. Discussion 4.1. Cu-partitioning-induced low ductility in the Cu/Ni-steel In post-tempered and aged Cu/Ni steels, we found similar lath-like matrix with substructures after solid-solution treatments (Fig. 2e and 2 f), which was consistent with previous literatures [10,12,14,25]. After tempering and/or aging treatments, reverted austenites were nucleated and grown, which was believed to facilitate for the nano-scale fine precipitates [24,25]. In our case, the tempering time was too short to find much reverted austenites. Only a small amount of precipitates was observed at the GBs in the post-tempered Cu/Ni-steel (Fig. 2e), but 229


J. Yan et al.

Fig. 4. Atom maps of Cu, Mn, Ni, Al, Mo and C of (a) Cu/Ni-steels tempered at 700 ℃ for 10 min and aged at 550 ℃ for 40 min; (b) Mo-steels tempered at 675 ℃ for 10 min and aged at 550 ℃ for 40 min.

through aging treatments, both small-size precipitates and large-size precipitates were found at the GBs of the aged Cu/Ni steel (Fig. 2f). Unfortunately, nearly all of aged Cu/Ni-steels showed low ductility even after intercritical heat treatments (Fig. 2a and b). Previous study [24,28] thought that Mn partitioning was one of the reasons to promote the GB embrittlement. In Fe-7Mn-2.5Ni-1.5Al Cufree steels, the content of Mn increased from 7 at% to 15 at% through the interface, indicating severe Mn interfacial enrichments [28]. In Fe6.84Mn-2.44Ni-1.63Al-1.36Cu Cu-bearing steels [24], where the elemental ratio of Mn/Cu was 5.0. Local one-dimensional concentrations of Mn and Cu were 20 at% and 30 at%, respectively, and the concentration ratio of Mn/Cu was nearly 0.6. Higher Mn concentrations drove the diffusion of Cu through the martensites into the CRPs, thereby contributing to a high coarsening rate [24]. It was concluded that Mn was easier to segregate on GBs with the increasing elemental ratio of Mn/Cu, and the concentration ratio on GBs was closely dependent on elemental ratio of Mn/Cu. As the elemental ratio of Mn/Cu and interfacial concentration ratio of Mn/Cu were higher, Mn partitioning was dominate rather than Cu partitioning and vice versa. In our samples, the elemental ratio of Mn/Cu was nearly 1.8, and the concentrations of Mn and Cu were 13 at% and 70 at%, respectively, and the concentration ratio of Mn/Cu was nearly 0.2, which was distinctly lower than the previous work [24,28]. Because the partitioning was governed by the elemental ratio and interfacial concentration ratio of Mn/Cu, as elemental ratio and interfacial concentration ratio of Mn/Cu

in our case was lower, the partitioning was governed by Cu partitioning rather than Mn partitioning in previous work [24,28]. Both our APT results (Fig. 5c) and TEM results (Fig. 2f) found a huge BCRP (mean radius 12 nm) along the GB, suggesting that the huge-size non-spheroidal CRP located at GB might contribute to the brittle failure of Cu/Ni-steel instead of the Mn partitioning as described in previous work. The presence of two main types large precipitates including carbide precipitates and CRPs was as a result of the short-circuit diffusion of Cu [33], where diffusion of solute elements along a ground boundary was as much as several orders of magnitude greater than in the bulk matrix. The high-diffusivity along a GB reduced nucleation times, increased the growth rate, and caused an earlier onset of coarsening [36]. Therefore, the precipitates nucleated along GBs were larger than those distributed in the martensite. Gao et al. [37] demonstrated that the appearance of CRPs was thought to be as one of the main reasons of embrittlement and the Cu clusters in the GB were favored to be with large size and low density rather than a high number of randomly distributed smaller sized clusters because of the positive binding energy. In addition, Table 1 shows another possibility that the number density of small-sized Cu precipitates was very high about 1.68 × 1023, which enhanced the precipitation strengthening. Hence, it was plausible to attribute the evidently degraded plasticity to the presence of well-dispersed nanoscale CRPs with a high number density (Nv) rather than that of the large CRPs with a relatively low Nv. 230


J. Yan et al.

Fig. 5. (a) Isoconcentration surfaces of Cu/Ni-steel; (b) isoconcentration surfaces of Mo-steel; (c) 1D composition distributions of Cu/Ni-steel; (d) 1D composition distributions of Mo-steel.

Thus, in Cu/Ni-steel, which had high strength but little ductility, large-sized BCRPs on GBs were regarded as one of the main possibilities to promote the GB embrittlement. The formation of these CRPs attributed to the high Cu contents in steel and the short-circuit diffusion of Cu and the Cu clusters’ absorption of positive binding energy.

the ductility decreased with the temperature. Higher tempering temperature and aging time contributed to higher strength increase, but appropriate tempering without superimposing aging displayed higher ductility. Mapping of strength increase plotted as the ductility reduction of Mo-steel (Fig. 8) indicated that the best combination of strength and ductility was superimposed by tempering at 650 °C (the strength increase by 19.4% and the ductility increase by 3%) or 675 °C (the strength increase by 37.6% and ductility decreases by 11%) for 10 min and aging for 120 min. We believed that hierarchical precipitation via Mo addition, which was introduced by multistage heat treatments including quenching, tempering and aging, was still one of important reasons to improve the strengthening-ductility of Mo-steels in terms of reverted austenite and hierarchical precipitates [24,25,38]. The dispersed reverted austenites were generally formed at martensitic lath boundaries which were the preferential sites for

4.2. Improved Cu partitioning and strengthening-ductility in Mo-steel In temper-aged Mo-steel, the strength and ductility (Fig. 3) were improved significantly compared to Cu/Ni steels (Fig. 2) partially because of Mo addition. Tempering-aging treatments was an effective way to improve the strength and ductility of Mo-steel. The largest strength increase was 37.6% when Mo-steel was tempered at 675 ℃ for 10 min and followed by aging for peak time of 40 min, while the minimum ductility reduction was under post-tempered condition. Generally, after tempered at proper temperature (> 650 ℃), the strength increased and

Table 2 Mean radius and number density of CRPs and NiAl particles in aged Cu/Ni-steel (APT results showing in Fig. 4a) and aged Mo-steel (APT results showing in Fig. 4b). Cu particles Rp/nm Cu/Ni-steel (Fig. 4a) Mo-steel (Fig. 4b) Mo-steel (no tempered) [16]

Rp Rp Rp Rp

< > < >

3 nm 3 nm 3 nm 3 nm

1.54 4.29 1.02 5.27 1.06

± ± ± ± ±

NiAl particles Nv/m

0.75 2.51 0.38 1.02 0.34

3

1.68 × 1023 1.3 × 1022 3.30 × 1022 2.47 × 1022 1.66 × 1024

231

Nv/m3

Rp/nm Rp Rp Rp Rp

< > < >

3 nm 3 nm 3 nm 3 nm

0.90 4.20 0.84 6.65 1.10

± ± ± ± ±

0.51 1.20 0.30 2.04 0.67

1.19 × 1024 0.19 × 1024 9.03 × 1023 0.30 × 1023 5.29 × 1024


J. Yan et al.

Fig. 6. Mean radius distribution of CRPs and NiAl particles in tempered and aged Mo-steels and Cu/Ni-steel. (a) Cu/Ni-steel; (b) Mo-steel.

Fig. 8. Strength increase plotted as ductility reduction for Mo-steel after IHTs, which is based on the data summary of strength and elongation in our work (Supplementary Table S2).

steels. Reducing the possibility of forming BCRPs along GB might be another strategy for strengthening-ductility. Previous study [24,28,39] demonstrated that the coarsening precipitates was formed and grew along the austenite GB, which caused this orientation relationship to be destroyed and a dramatic increase of the coherency strain energy at GBs. Furthermore, serious embrittlement of GBs caused that cleavage cracks trend to propagate along the grain boundaries. In our cases, the brittle cleavage fracture might be formed due to the BCRPs on GBs. However, the BCRPs could not be found in Mo-steel, which could account for the ductility improvement. Hierarchical precipitates were found in our work, which might be another reason to improve the strengthening-ductility of Mo-steels. Through tempering treatment, a small amount of Cu-precipitates nuclear in the matrix because the solid solubility of the precipitation increases after the tempering at over 625 °C. After aging treatment, these nucleated Cu-precipitates further Ostwald ripened [40] to CCRPs in the fresh secondary martensitic zone (FSM) [25], which were accounted for the enhancement of ductility. Around the CCRPs, fine dispersed SCRPs (Fig. 5b) were detected, which might strengthen by precipitation strengthening. These two kinds of different precipitates with different size co-existed in matrix to improve the strengthening-ductility. Mo addition obviously slowed down the diffusion of Mn and Cu, delaying the evolution of elemental concentrations, time of peak hardness, and the nucleation sequence of Ni(Al,Mn) precipitates, which

Fig. 7. Proximity histograms of (a) CRPs and (b) Ni(Al, Mn) particles in Cu/Nisteels tempered at 700 ℃ for 10 min and aged at 550 ℃ for 40 min, and Mosteels tempered at 675 ℃ for 10 min and aged at 550 ℃ for 40 min.

heterogeneous nucleation [24,25,38]. Increasing temperature and time allowed the pertinent solute atoms to diffuse more rapidly and promoted the austenite reversion. It was believed that reverted austenites were a comparatively ‘‘soft’’ phase compared to the martensitic matrix and its volume fraction was usually small [38]. Although reverted austenites were not found in our work because of limited tempering time, we believed the presence of reverted austenites, and they should be responsible for the improvement of strengthening-ductility of Mo232


J. Yan et al.

was attributed to the strong repulsive interactions from the positive mixing enthalpy between Mo and Mn/Cu, affecting the atom diffusion, increasing diffusion activation energy of both Cu and Mn, and decreasing the diffusion coefficients [16,41]. This might be one of origins that Mo reduced the Cu partitioning along GBs and promoted to coprecipitate CRPs and NiAl precipitates, thus increasing strengtheningductility.

3843–3851. [7] Z.B. Jiao, J.H. Luan, M.K. Miller, C.T. Liu, Precipitation mechanism and mechanical properties of an ultra-high strength steel hardened by nanoscale NiAl and Cu particles, Acta Mater. 97 (2015) 58–67. [8] Z.B. Jiao, J.H. Luan, Z.W. Zhang, M.K. Miller, W.B. Ma, C.T. Liu, Synergistic effects of Cu and Ni on nanoscale precipitation and mechanical properties of high-strength steels, Acta Mater. 61 (16) (2013) 5996–6005. [9] D. Jain, D. Isheim, D.N. Seidman, carbon redistribution and carbide precipitation in a high-strength low-carbon HSLA-115 steel studied on a Nanoscale by atom probe tomography, Metall. Mater. Trans. A 48A (7) (2017) 3205–3219. [10] D. Jain, D. Isheim, A.H. Hunter, D.N. Seidman, Multicomponent high-strength lowalloy steel precipitation-strengthened by sub-nanometric Cu precipitates and M2C carbides, Metall. Mater. Trans. A 47A (8) (2016) 3860–3872. [11] D. Isheim, M.S. Gagliano, M.E. Fine, D.N. Seidman, Interfacial segregation at Curich precipitates in a high-strength low-carbon steel studied on a sub-nanometer scale, Acta Mater. 54 (3) (2006) 841–849. [12] Y.R. Wen, Y.P. Li, A. Hirata, Y. Zhang, T. Fujita, T. Furuhara, C.T. Liu, A. Chiba, M.W. Chen, Synergistic alloying effect on microstructural evolution and mechanical properties of Cu precipitation-strengthened ferritic alloys, Acta Mater. 61 (20) (2013) 7726–7740. [13] A.J. Ardell, Precipitation hardening, Metall. Trans. A 16 (12) (1985) 2131–2165. [14] Y.R. Wen, A. Hirata, Z.W. Zhang, T. Fujita, C.T. Liu, J.H. Jiang, M.W. Chen, Microstructure characterization of Cu-rich nanoprecipitates in a Fe-2.5 Cu-1.5 Mn4.0 Ni-1.0 Al multicomponent ferritic alloy, Acta Mater. 61 (6) (2013) 2133–2147. [15] Z. Sun, G. Song, J. Ilavsky, P.K. Liaw, Duplex precipitates and their Effects on the Room-temperature fracture behaviour of a NiAl-strengthened ferritic alloy, Mater. Res Lett. 3 (3) (2015) 128–134. [16] J. Yan, D. Huang, X. Zuo, Co-precipitation kinetics, microstructural evolution and interfacial segregation in multicomponent nano-precipitated steels, J. Alloy Compd. Rev. (2018). [17] Z.B. Jiao, J.H. Luan, M.K. Miller, C.Y. Yu, C.T. Liu, Effects of Mn partitioning on nanoscale precipitation and mechanical properties of ferritic steels strengthened by NiAl nanoparticles, Acta Mater. 84 (2015) 283–291. [18] Q. Shen, H. Chen, W. Liu, Effect of Cu on nanoscale precipitation evolution and mechanical properties of a Fe-NiAl alloy, Microsc. Microanal. 23 (2) (2017) 350–359. [19] G. Han, Z.J. Xie, L. Xiong, C.J. Shang, R.D.K. Misra, Evolution of nano-size precipitation and mechanical properties in a high strength-ductility low alloy steel through intercritical treatment, Mat. Sci. Eng. A-Struct. 705 (2017) 89–97. [20] I. Altpeter, G. Dobmann, K.H. Katerbau, M. Schick, P. Binkele, P. Kizler, S. Schmauder, Copper precipitates in 15 NiCuMoNb 5 (WB 36) steel: material properties and microstructure, atomistic simulation, and micromagnetic NDE techniques, Nucl. Eng. Des. 206 (2–3) (2001) 337–350. [21] M.D. Mulholland, D.N. Seidman, Nanoscale co-precipitation and mechanical properties of a high-strength low-carbon steel, Acta Mater. 59 (5) (2011) 1881–1897. [22] X.Y. Cheng, H. Li, X.B. Cheng, Carbides and possible hydrogen irreversible trapping sites in ultrahigh strength round steel, Micron 103 (2017) 22–28. [23] Z.B. Jiao, J.H. Luan, M.K. Miller, C.Y. Yu, Y. Liu, C.T. Liu, Precipitate transformation from NiAl-type to Ni2AlMn-type and its influence on the mechanical properties of high-strength steels, Acta Mater. 110 (2016) 31–43. [24] Y. Li, W. Li, W.Q. Liu, X.D. Wang, X.M. Hua, H.B. Liu, X.J. Jin, The austenite reversion and co-precipitation behavior of an ultra-low carbon medium manganese quenching-partitioning-tempering steel, Acta Mater. 146 (2018) 126–141. [25] Q. Liu, J. Gu, Hierarchical Cu precipitation in lamellated steel after multistage heat treatment, Philos. Mag. 87 (27) (2017) 2361–2374. [26] M. Kapoor, D. Isheim, S. Vaynman, M.E. Fine, Y.W. Chung, Effects of increased alloying element content on NiAl-type precipitate formation, loading rate sensitivity, and ductility of Cu- and NiAl-precipitation-strengthened ferritic steels, Acta Mater. 104 (2016) 166–171. [27] M.E. Fine, S. Vaynman, D. Isheim, Y.W. Chung, S.P. Bhat, C.H. Hahin, A. New, Paradigm for Designing high-fracture-energy steels, Metall. Mater. Trans. A 41A (13) (2010) 3318–3325. [28] Y. Li, W. Li, N. Min, W.Q. Liu, X.J. Jin, Effects of hot/cold deformation on the microstructures and mechanical properties of ultra-low carbon medium manganese quenching-partitioning-tempering steels, Acta Mater. 139 (2017) 96–108. [29] R.P. Kolli, D.N. Seidman, Comparison of compositional and morphological atomprobe tomography analyses for a multicomponent Fe-Cu steel, Microsc. Microanal. 13 (4) (2007) 272–284. [30] W.J. Kim, K.E. Lee, S.H. Choi, Mechanical properties and microstructure of ultra fine-grained copper prepared by a high-speed-ratio differential speed rolling, Mat. Sci. Eng. A-Struct. 506 (1–2) (2009) 71–79. [31] Y. Sun, X.F. Li, X.Y. Yu, D.L. Ge, J. Chen, J.S. Chen, Fracture morphologies of advanced high strength steel During deformation, Acta Metall. Sin.-Engl. Lett. 27 (1) (2014) 101–106. [32] Z.K. Teng, C.T. Liu, M.K. Miller, G. Ghosh, E.A. Kenik, S. Huang, P.K. Liaw, Room temperature ductility of NiAl-strengthened ferritic steels: effects of precipitate microstructure, Mat. Sci. Eng. A-Struct. 541 (2012) 22–27. [33] R.P. Kolli, D.N. Seidman, Co-Precipitated and collocated carbides and Cu-Rich precipitates in a Fe-Cu steel characterized by atom-probe tomography, Microsc. Microanal. 20 (6) (2014) 1727–1739. [34] B.S. Lee, M.C. Kim, J.H. Yoon, J.H. Hong, Characterization of high strength and high toughness Ni-Mo-Cr low alloy steels for nuclear application, Int J. Pres. Ves. Pip. 87 (1) (2010) 74–80. [35] M. Bahmani, R. Elliott, N. Varahram, The relationship between fatigue strength and microstructure in an austempered Cu-Ni-Mn-Mo alloyed ductile iron, J. Mater. Sci. 32 (20) (1997) 5383–5388.

5. Conclusions (1) The strengthening-ductility of Cu-steel by CRPs was sensitive to tempering temperatures rather than aging time. Elevated temperature tempering reduced the strength and increased the plasticity of Cu-steel. IHT-Cu-steels had higher EL (larger than 10%) with lower strength (lower than 600 MPa). (2) The strength of Cu/Ni steel by co-precipitates of CRPs+NiAl was much higher than that of Cu-steel, but the plasticity was much lower. Compare to tempered and aged Cu/Ni steel, the strength of post-tempered Cu/Ni steel decreased slightly with increasing ductility. Almost all aged Cu/Ni steels showed brittle fractures, which was likely to be the presence of large-sized blocky Cu-rich precipitates along GBs due to Cu partitioning. (3) The Cu-bearing Mo-steel with the main strengthening phase of Cu/ NiAl co-precipitation and Mo2C showed better strengthening-ductility. Tempering and aging heat treatment was capable of achieving the optimal combination of strength and ductility. The best heat treatment process was tempering at 650 °C or 675 °C for 10 min and aging at 550 °C for 2 h, and the strength was 1200 MPa and the ductility was 7%. (4) After intercritical tempeing and aging, the atoms of Cu, Ni, Al and Mn showed obvious segregation except Mo atoms. Mo might restrictedly affect the nucleation growth of co-precipitation of CRPs and NiAl precipitates Hierarchical co-precipitation, where largesize coarsening co-precipitates formed in the fresh secondary martensite zone, and fine secondary co-precipitates hierarchically nucleated from the supersaturated matrix, contributed to the improvement of strengthening-ductility in the Mo-steel. Acknowledgements This work was supported by the National Natural Science Foundation of China (Nos. U1860103 and 51474066), and the Fundamental Research Funds for the Central Universities (No. N170904007). The authors are grateful for the APT analysis and fruitful discussions from Prof. Wenqing Liu at Shanghai University, China. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.msea.2018.10.036. References [1] K. Li, J.G. Shan, C.X. Wang, Z.L. Tian, Influence of aging temperature on strength and toughness of laser-welded T-250 maraging steel joint, Mat. Sci. Eng. A-Struct. 669 (2016) 58–65. [2] R.O. Ritchie, The conflicts between strength and toughness, Nat. Mater. 10 (11) (2011) 817–822. [3] A. Cerezo, S. Hirosawa, I. Rozdilsky, G.D.W. Smith, Combined atomic-scale modelling and experimental studies of nucleation in the solid state, Philos. Trans. R. Soc. Lond. Ser. A-Math. Phys. Eng. Sci. 361 (1804) (2003) 463–476. [4] Z.B. Jiao, J.H. Luan, M.K. Miller, Y.W. Chung, C.T. Liu, Co-precipitation of nanoscale particles in steels with ultra-high strength for a new era, Mater. Today 20 (3) (2017) 142–154. [5] J. Zelenty, G.D.W. Smith, K. Wilford, J.M. Hyde, M.P. Moody, Secondary precipitation within the cementite phase of reactor pressure vessel steels, Scr. Mater. 115 (2016) 118–122. [6] N. Ueshima, T. Maeda, K. Oikawa, Effect of Cu addition on precipitation and growth behavior of MnS in silicon steel sheets, Metall. Mater. Trans. A 48A (8) (2017)

233


J. Yan et al. [36] J.J. Hoyt, On the coarsening of precipitates located on grain boundaries and dislocations, Acta Metall. Mater. 39 (9) (1991) 2091–2098. [37] N. Gao, A. Ghoniem, X. Gao, P. Luo, K.F. Wei, Z.G. Wang, Molecular dynamics simulation of Cu atoms interaction with symmetrical grain boundaries of BCC Fe, J. Nucl. Mater. 444 (1–3) (2014) 200–205. [38] Q. Liu, H. Wen, H. Zhang, J. Gu, C. Li, E.J. Lavernia, Effect of multistage heat treatment on microstructure and mechanical properties of high-strength low-alloy steel, Metall. Mater. Trans. A 47A (5) (2016) 1960–1974. [39] Y.F. Feng, R.B. Song, Z.Z. Pei, R.F. Song, G.Y. Dou, Effect of aging isothermal time

on the microstructure and room-temperature impact toughness of Fe-24.8Mn-7.3Al1.2C austenitic steel with kappa-carbides precipitation, Met Mater.-Int 24 (5) (2018) 1012–1023. [40] Q.D. Liu, C.W. Li, J.F. Gu, Interactive formation of Cu-rich precipitate, reverted austenite, and alloyed carbide during partial austenite reversion treatment for highstrength low-alloy steel, J. Mater. Res 32 (12) (2017) 2325–2334. [41] A. Takeuchi, A. Inoue, Calculations of mixing enthalpy and mismatch entropy for ternary amorphous alloys, Mater. Trans. Jim 41 (11) (2000) 1372–1378.

234