Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy

Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy

Author’s Accepted Manuscript Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy Qin Shen, Xiangyu...

2MB Sizes 0 Downloads 20 Views

Author’s Accepted Manuscript Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy Qin Shen, Xiangyuan Xiong, Tong Li, Hao Chen, Yangming Cheng, Wenqing Liu www.elsevier.com/locate/msea

PII: DOI: Reference:

S0921-5093(18)30400-3 https://doi.org/10.1016/j.msea.2018.03.053 MSA36246

To appear in: Materials Science & Engineering A Received date: 31 January 2018 Revised date: 10 March 2018 Accepted date: 13 March 2018 Cite this article as: Qin Shen, Xiangyuan Xiong, Tong Li, Hao Chen, Yangming Cheng and Wenqing Liu, Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy, Materials Science & Engineering A, https://doi.org/10.1016/j.msea.2018.03.053 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effects of co-addition of Ni and Al on precipitation evolution and mechanical properties of Fe-Cu alloy Qin Shena, Xiangyuan Xiongb, Tong Lia, Hao Chena, Yangming Chenga,Wenqing Liua* a

Institute of Materials, School of Materials Science and Engineering, Shanghai University, Shanghai 200444, China.

b

College of Materials Science and Engineering, Beijing University of Technology, Beijing 100124, China.

ABSTRACT Fe-Cu and Fe-Cu-Ni-Al alloys were aged at 500 °C for different time after solution treatment at 900 °C for 2 h. The influence of co-addition of Ni and Al on the microstructure evolution of Cu-rich phase was studied using atom probe tomography (APT). APT results showed that the addition of Ni and Al effectively increased the number density of Cu-rich particles and reduced their sizes with a narrow size distribution. In the peak hardness condition, the precipitates in the Fe-Cu-Ni-Al alloy exhibited a core-shell structure with the Cu-rich phase in the core and NiAl phase in the outer shell, leading to a dramatic improvement of peak hardness and strength. The NiAl shell of the precipitates impeded the growth and coarsening of the Cu-rich phase by decreasing the interfacial energy and the diffusion rate of Cu, Ni and Al atoms in the shell. After further ageing, the core-shell structure of the precipitates decomposed, forming separate Cu-rich phase and NiAl phase. KEY WORDS Cu-rich phase, Precipitation hardening, Atom probe tomography Fe-Cu-Ni-Al alloy, Ageing *Corresponding author: Wenqing Liu Address: No. 99 Shangda Road, Shanghai 200444, China. Tel.: +86 21 66135027. E-mail addresses: [email protected].

1. Introduction Cu precipitation strengthening in steels has been studied extensively and has 1

become the cornerstone for the development of high-strength low-carbon steels [1-4]. Similar to the Cu precipitation strengthened steels, the Ni and Al containing steels were found to be hardened by ageing at temperatures above 400 °C [5-8]. The strengthening effect was due to the precipitation of NiAl nanoparticles [7]. With a proper addition of Ni and Al to the Cu-containing steels, a higher ultimate tensile strength of ~1.6 GPa was achieved for the precipitation-strengthened steel [9]. The co-precipitation of nanoscale Cu-rich phase and NiAl phase in the CuNiAl-containing steels offers a promising way to increase strength of steels significantly. A great deal of efforts has been devoted to the characterization of precipitates

and

understanding

the

co-precipitation

mechanism

in

the

CuNiAl-containing steels [10-14]. Kolli et al. [15] showed that after ageing at 500 °C for 1024 h, the precipitate of the NU-170 steel had a core-shell structure with Cu atoms in the core and Ni, Al and Mn atoms surrounded in the shell. However, Wang et al. [16] reported that the Cu-rich particle and the NiAl particle were distributed side-by-side to form the composite precipitates, which was different from core-shell structured precipitates. Furthermore, it was found that during the structure evolution, the composition of the Cu-rich phase and the NiAl phase might undergo complex changes with ageing time. Such as Ni and Al, partitioned to the Cu-rich phase cores during the early stage of ageing, and then segregated at the Cu-rich phase/matrix interfaces [17-19]. In these reports, there is less work addressing the effect of the formation of the NiAl phase on the evolution of the Cu nanoparticles. The detailed mechanism for how the co-addition of Ni and Al affects the nanoscale precipitation of Cu-rich phase is not fully understood yet. To clarify this mechanism, it is essential to further explore Cu-rich phase evolution in the alloys with and without Ni and Al addition. To this end, the atom probe tomography (APT) is a suitable and powerful tool to identify the precipitates and determine the precipitate sizes and compositions, especially for samples at the early stage of ageing. In this study, two model alloys, Fe-Cu and Fe-Cu-Ni-Al, have been selected to investigate the effects of co-addition of Ni and Al on the precipitation of nanoscale 2

Cu-rich phase by using the APT. The corresponding mechanical properties of the alloys have been discussed.

2. Materials and methods The compositions of the Fe-Cu and Fe-Cu-Ni-Al alloys in this study are given in Table 1. These alloys were prepared by melting pure metals with purities above 99.99 wt.% in a vacuum induction furnace. And the melted alloys were cast into a water-cooled copper mold, followed by forging at 1100 °C and hot rolling at 1200 °C by 50% in thickness. Then each resulting alloy was subjected to solution treatment at 900 °C for 2 h, followed by water quenching to room temperature and ageing at 500 °C for various selected times. The hardness measurements were conducted on polished surface of the samples using Vickers micro-indentation (500 g load and indentation time 10 s), and at least seven indents were measured to obtain an average value for each sample. Tensile tests were performed using a materials testing system (MTS CMT 5205) machine at a strain rate of 10-3 s-1. The tensile samples with a cross-section of 3.2×1.6 mm and a gauge length of 25 mm, were cut by electro-discharge machine and ground carefully on each side with SiC paper through 2000-grit. Fracture surfaces were examined by scanning electron microscopy (SEM). The yield strength was determined using the 0.2% offset plastic strain method. In order to prepare tips for APT, small rods with a cross-section of 0.5×0.5 mm2 were cut out of the aged bulk alloys. Subsequently, the tip samples were prepared by a two-stage electro-polishing method [20]. The analysis was performed on a local electrode atom probe (LEAP4000X HR) at a temperature of about 50 K with a target evaporation rate of 0.5% and a pulse voltage fraction of 20% in an ultra-high vacuum of ~ 10-9 Pa. The voltage pulse repetition rate was 200 kHz. The Integrated Visualization and Analysis Software (IVAS 3.6.8) was used for 3D reconstructions. The identification and characterization of Cu-rich precipitates were undertaken by the maximum separation method [21, 22] with a maximum separation distance (dmax) of 0.5 nm and a minimum solute atom number (Nmin) of 20. The radius of a precipitate, 3

rp, was equated to that of the volume equivalent sphere, given by Eq. (1) [23]: 3

3𝑛𝛺

𝑟𝑝 = √ , 4𝜋

(1)

where n is the total number of atoms in each precipitate, Ω is the atomic volume (1.178 × 10-2 nm3 for Fe), and

is the estimated detection efficiency of the

microchannel plate detector in the LEAP system, which is 0.37 in this case.

3. Results 3.1 Mechanical properties Fig. 1 displays the hardness as a function of ageing time for the Fe-Cu and Fe-Cu-Ni-Al alloys. The Fe-Cu alloy has a hardness of ~145 Hv in the as-quenched (AQ) state. At the early stage of ageing (0.1 h), the hardness increased slightly by ~8 Hv. With increasing ageing time, the hardness gradually increases and reaches a peak value of ~211 Hv at 0.25 h, which is ~66 Hv higher than that of the AQ sample. This peak hardness appears stable and does not drop until aging for 4 h. With a further increase of the ageing time, the hardness decreases. Owing to the solid solution hardening of Ni and Al atoms, the hardness of the Fe-Cu-Ni-Al alloy in the AQ state is higher than that of the Fe-Cu alloy. After ageing for 0.1 h, the hardness is increased by ~22 Hv, which is more significant than that of Fe-Cu alloy. The position of peak hardness shifts to a longer time at 4 h, with a peak hardness of ~337 Hv. And the peak hardness plateau for the Fe-Cu-Ni-Al alloy is longer than that of Fe-Cu alloy. From the hardness tests, it is evident that the precipitation hardening took place in the Fe-Cu and Fe-Cu-Ni-Al alloy during ageing. More importantly, the addition of Ni and Al atoms significantly affects the age-hardening response. Fig. 2 shows the tensile stress-strain curves of the Fe-Cu and Fe-Cu-Ni-Al alloys in the AQ and peak hardness states. The yield strength, ultimate tensile strength and elongation-to-failure are summarized in Table 2. The fracture surfaces of the tensile test samples are shown in Fig. 3. The Fe-Cu alloy shows a yield strength of 207 MPa and an ultimate tensile strength of 317 MPa in the AQ state. Elongation of the AQ 4

Fe-Cu alloy sample is 25.3%, with a dimple fracture surface (Fig. 3a). After ageing for 0.25 h, the yield strength and the ultimate tensile strength are improved to 400 MPa and 534 MPa, while the elongation is decreased to 6.1%. The sample shows a larger ductile zone area, consisting of high density of dimples combined with a cleavage fracture mode (Fig. 3b). The co-addition of Ni and Al increased considerably the strength of the AQ Fe-Cu-Ni-Al alloy. As seen in Table 2, accompanying the improved strength, the elongation is reduced, but the sample is still ductile, with 10.9% elongation-to-failure as well as a dimple fracture surface (Fig. 3c). When aged for 4 h, the yield strength and ultimate tensile strength are dramatically increased to 1023 MPa and 1053 MPa, while the ductility remains nearly unchanged with an elongation-to-failure of 10.1%, exhibiting a good combination of high strength and high ductility. The failure mode of the ultrahigh strength alloy is dimple fracture, mixed with ductile transgranular failure (Fig. 3d). These dramatic increases in the yield strength and ultimate tensile strength are in agreement with the hardness measurements, demonstrating that the addition of Ni and Al enhances the age-strengthening effect of the alloy significantly. 3.2 The evolution of precipitates in the Fe-Cu and Fe-Cu-Ni-Al alloys To understand the effects of Ni and Al atoms on the microstructural evolution of the Fe-Cu-Ni-Al alloy, the APT work was performed on both the Fe-Cu and the Fe-Cu-Ni-Al alloys. Fig. 4 shows Cu, Ni and Al atom maps of the Fe-Cu and Fe-Cu-Ni-Al alloys aged at 500 °C for 0.1 h, 0.25 h, 4 h and 128 h, respectively. Based on the APT results in Fig. 4, the average sizes (rp) and number densities (Nv) of Cu precipitates in the alloys aged at 500 °C for different times are calculated and shown in Table 3. As seen in Fig. 4a, Cu atoms in the Fe-Cu alloy aged for 0.1 h do not show obvious precipitation. However, from the nearest neighbor distribution (NND) analysis of Cu atoms (Fig. 5a), it can be seen that there is a slight deviation between experimental data and random data curves, suggesting a slight clustering of Cu atoms. On the other hand, a considerable number of Cu clusters are observed in the 5

Fe-Cu-Ni-Al alloy aged for 0.1 h (Fig. 4b), which led to the development of nonrandom distributions of Cu by NND analysis, as shown in Fig. 5b. The Ni and Al atoms of the Fe-Cu-Ni-Al alloy only show slight segregation, as seen in Fig. 5c and 5d. After ageing for 0.25 h (Figs. 4c and d), Cu clusters become obvious in both alloys. Ni and Al atoms in the Fe-Cu-Ni-Al alloy show clustering, indicating the formation of the NiAl phase, which is in agreement with the previous studies [10-16]. As seen in Table 3, the average radius rp of the Cu clusters in the Fe-Cu alloy is 1.2 nm, while in the Fe-Cu-Ni-Al alloy is 1.0 nm, slightly smaller than that in the Fe-Cu alloy. The number density Nv of the Cu clusters is 9.55×1023 m-3 in the Fe-Cu alloy and 14.2×1023 m-3 in the Fe-Cu-Ni-Al alloy. So, it can be concluded that, the addition of Ni and Al atoms increases the number density of Cu-rich precipitate, while reduce the average of Cu-rich precipitate. At the ageing time of 4 h, the average radius rp of the Cu-rich phase in the Fe-Cu alloy increases to 2.3 nm and its number density Nv is 1.7×1023 m-3, decreased significantly compared with the 0.25 h aged sample. While in the Fe-Cu-Ni-Al alloy, the average radius rp of the Cu-rich phase increases to 1.4 nm and its number density Nv is 12.2×1023 m-3, reduced slightly compared with the 0.25 h aged sample. It can be inferred that due to the effect of co-addition of Ni and Al, the Cu cluster in the Fe-Cu-Ni-Al alloy grew at a slower rate than those of the Fe-Cu alloy. With increasing ageing time to 128 h, the precipitates of the Fe-Cu and Fe-Cu-Ni-Al alloys grow, accompanied by changes in morphology from equiaxed to disk and elongated, respectively. In the Fe-Cu alloy the average size rp of Cu precipitates is 14.2 nm and its number density Nv is 0.02×1023 m-3; while in the Fe-Cu-Ni-Al alloy the average size rp of Cu precipitates is 3.1 nm, much smaller than that in the Fe-Cu alloy, and the number density Nv is 0.9×1023 m-3, one order of magnitude higher than that in the Fe-Cu alloy. These results show that the coarsening of Cu precipitates in the Fe-Cu-Ni-Al alloy is much slower than that in the Fe-Cu alloy. The compositions of the Cu-rich phase in the Fe-Cu and Fe-Cu-Ni-Al alloys after 6

ageing for 0.25 h, 4 h and 128 h are given in Table 4. The concentration values in Table 4 were obtained utilizing the proximity histogram based on the isoconcentration surface defined by the threshold concentration of 20 at.% Cu with the grid size 0.2 nm and the delocalization distance 3 nm [24, 25]. The uncertainty for each concentration c is [c(1- c)/N]0.5, where N is the total number of atoms used for calculating the concentration c. It can be seen that these Cu precipitates contain not only Cu atoms but a significant number of Fe atoms as well as Ni and Al atoms for the Fe-Cu-Ni-Al alloy. The solubilities of Fe, Ni and Al atoms in the Cu precipitates, found in the present study, are in good agreement with the observations by Jiao et al. [26]. With increasing ageing time, the Cu concentration in the precipitates increases and their Fe concentration decreases for both the Fe-Cu and the Fe-Cu-Ni-Al alloys, while the Ni and Al concentrations remain almost constant for the Fe-Cu-Ni-Al alloy. However, in the same ageing condition, the Cu concentration of the Cu precipitates in the Fe-Cu-Ni-Al alloy is lower than that in the Fe-Cu alloy. This may be due to the segregation of Ni and Al atoms in the Cu precipitates. After ageing for 128 h, the Cu concentration at the Cu precipitate center in the Fe-Cu alloy is 98.4 ± 3.8 at.%, close to pure Cu metal, but only 76 ± 3.8 at.% in the Fe-Cu-Ni-Al alloy. The 3-nm-thick atom maps sliced through the centers of the precipitates in the Fe-Cu-Ni-Al alloy aged for different times are shown in Fig. 6. The circles around the Cu segregation zone correspond to 20 at.% Cu isoconcentration surfaces, whereas the circles around the Ni and Al segregation zone correspond to 15 at.% (Ni+Al) isoconcentration surfaces. Figs. 6d-f show one-dimensional composition profiles measured with a cylindrical probe through the center of precipitates in Figs. 6a-c. At the ageing time of 0.25 h, Ni, Al and Cu atoms segregate from the matrix (Fig. 6a). Fig. 6d exhibits that Ni segregates at the Cu-rich precipitate/matrix interface, while Al is found enriched in the Cu-rich clusters. With increasing ageing time to 4 h, the precipitates exhibit a core-shell structure, with Cu precipitates in the core and NiAl precipitates in the shell (Fig. 6b), which is similar to the core-shell structure reported by Kolli et al. [15]. Fig. 6e shows the changes in the distribution of Cu, Ni and Al atoms. The Cu-rich core becomes more enriched in Cu atoms, while Ni and Al atoms 7

are further rejected from the core and become enriched around the Cu-rich precipitates. The Cu-rich precipitates are not in the core of NiAl precipitates and are enriched along a preferred direction in the sample aged for 128 h (Fig. 6c). The NiAl precipitates are only observed on one side of a Cu-rich precipitate in Fig. 6f. This result is consistent with the results of Zhang et al. in the study of a Fe-Cu-Ni-Al-Mn alloy aged for 2000 h [19], which showed that Ni, Al and Mn atoms were expelled away from the Cu-rich core to form the NiAl(Mn) phase on one side of the Cu-rich core.

4. Discussion 4.1 Effect of Ni and Al on the precipitation of Cu-rich phases With the addition of Ni and Al, the number density of Cu-rich precipitates increases dramatically, as shown in Table 3. It can be inferred that the co-addition of Ni and Al promotes the precipitation of Cu-rich phases. According to the classical nucleation theory [27], the nucleation of Cu-rich precipitates is described by the equation: −∆𝐺 ∗

𝑑𝑁 𝑑𝑡 𝑛𝑢𝑐𝑙𝑒𝑎𝑡𝑖𝑜𝑛

∝ exp(

𝑘𝐵 𝑇

)

(2)

where kB is Boltzmann’s constant, T is the temperature and ∆𝐺 ∗ is the critical energy for nucleation, which is described by [27], ∆𝐺 ∗ =

16𝜋𝛾3 3(∆𝐺𝑉 +∆𝐺𝜀 )2

(3)

where γ is the interfacial energy between the nanoparticles and the matrix, ΔGV is the chemical driving force for nucleation and ΔGɛ is the elastic strain energy. The beneficial effects of Ni and Al atoms on the nucleation of Cu particles can be discussed in terms of the chemical driving force ΔGV and the interfacial energy γ of nucleation. Firstly, the heat of mixing between Cu-Al is negative (-1 kJ mol-1) [28], therefore at the early ageing time (0.25 h), the Al co-precipitate with Cu to form Cu-Al-Fe precipitates, as shown in Fig. 6d. It is equivalent to increasing the total concentration of the nanoparticle-forming elements. The matrix supersaturation of Al 8

and Cu in the Fe-Cu-Ni-Al alloy is higher as compared with that of Cu in the Fe-Cu alloy, resulting in an increase in chemical driving force ΔGV for the nucleation of the nanoparticles. Second, due to the positive mixing enthalpy (+2 kJ mol-1) between Cu and Ni [28], Ni atoms are favored to be located at the interface between matrix and Cu precipitates, which is consistent with previous APT results [29] and first principle theoretical calculations [30]. The interface segregation of Ni decreases the interfacial energy γ between the matrix and Cu precipitates. As a result, the chemical driving force ΔGV is increased and the interfacial energy γ is decreased, so the critical energy for nucleation of Cu nanoparticles is reduced, leading to an increased nucleation rate of Cu precipitates in the Fe-Cu-Ni-Al alloy. Consequently, the number density of Cu particles in the Fe-Cu-Ni-Al alloy is increased. 4.2 Effect of Ni and Al on the growing and coarsening of Cu-rich phases As shown in Table 3, the Cu-rich phases in the Fe-Cu and Fe-Cu-Ni-Al alloys increase in sizes and decrease in number densities with ageing time prolonging. In addition, at the same ageing time, the size of Cu-rich phase in the Fe-Cu-Ni-Al alloy is always smaller than that in the Fe-Cu alloy, indicating that the co-addition of Ni and Al atoms reduces the growth and coarsening rate of the Cu-rich phase. This result can be understood as follows: Elements Al and Ni have a large negative mixing enthalpy (-22 kJ mol-1). With increasing ageing time, they form a B2-NiAl phase at the interface between Cu-rich phase and matrix. When the ageing time is 4 h, the Cu-rich phase and NiAl phase of the Fe-Cu-Ni-Al alloy exhibit a core-shell structure (Fig. 6b). The shell of the NiAl phase in the Fe-Cu-Ni-Al alloy plays an important role in impeding the growth and coarsening of the Cu-rich phase. The B2 structure NiAl phase has a lattice constant (0.2887 nm) larger than Fe (0.2866 nm) but smaller than bcc Cu (0.2951 nm), it can serve as a buffer layer with a small lattice mismatch to lower interfacial energy and lattice strain energy of matrix/Cu precipitates interface, leading to a slow growth of Cu precipitates. Moreover, since the NiAl phase is an intermetallic phase with high thermal stability [9], the Cu, Ni and Al atoms diffuse more slowly through the B2 9

NiAl phase than through the Fe matrix. As a result, the Cu-rich phase in the Fe-Cu-Ni-Al alloy grows at a slower rate than that in the Fe-Cu alloy. Another reason for the slower growth rate of the Cu-rich phase in the Fe-Cu-Ni-Al alloy should be that the Cu-rich precipitates in the Fe-Cu-Ni-Al alloy are more uniform in size. As seen in Fig. 7, the sizes of the Cu-rich precipitates in the Fe-Cu-Ni-Al alloy are always smaller than those in the Fe-Cu alloy for the same ageing time, and the size distribution of the Cu-rich precipitates in the Fe-Cu-Ni-Al alloy is relatively narrower than that in the Fe-Cu alloy, therefore, the differences in sizes of the precipitates in the Fe-Cu-Ni-Al alloy are smaller. As the driving force of coarsening of precipitates depends on the chemical potential difference ∆μ between the precipitates of different sizes, as given by [31] ∆𝜇 = 2𝛺𝛾(

1 𝑟2



1 𝑟1

),

(4)

where Ω is the molar volume, γ is the interfacial energy, and r1 and r2 are the radius of the precipitates. Therefore, the larger the difference in sizes of the precipitates, the greater the driving force of coarsening is given. The difference in sizes of the Cu-rich precipitates in the Fe-Cu-Ni-Al alloy is smaller than that the Fe-Cu alloy, in favor of a smaller driving force of coarsening and a slower coarsening rate of the Cu-rich phase in the Fe-Cu-Ni-Al alloy. 4.3 Effect of Ni and Al on mechanical properties The ageing response is different for the Fe-Cu and Fe-Cu-Ni-Al alloys due to the co-addition of Ni and Al, resulting in quite different mechanical properties. There are a considerable number of Cu clusters in the 0.1 h aged Fe-Cu-Ni-Al sample (Fig.4b), leading to a higher hardness increment than the Fe-Cu alloy (Fig.1). The peak hardness of the Fe-Cu alloy appeared at 0.25 h, while for the Fe-Cu-Ni-Al alloy it appeared at 4 h. Although the time is different, the peak hardness always corresponds to the high number density of precipitates with small sizes. We estimated the precipitation strengthen improvement is attributed to the dislocation-precipitate interaction. In the peak hardness states, the precipitates are very small, the gliding dislocations would shear the precipitates rather than bypass them. For the particle 10

shearing mechanism, the increase in yield strength mainly results from the total contributions of chemical strengthening ( ∆𝜎chemical ), coherency strengthening (∆𝜎coherency ) and modulus strengthening (∆𝜎modulus ), which are predicted by the following equations [32]: ∆𝜎𝑐ℎ𝑒𝑚𝑖𝑐𝑎𝑙 =

3

2𝑀 1 𝑏𝐿𝑇 2

(𝛾𝑖𝑛𝑡𝑒𝑟𝑓𝑎𝑐𝑖𝑎𝑙 𝑏)2 3

1

𝑅 1

∆𝜎𝑐𝑜ℎ𝑒𝑟𝑒𝑛𝑐𝑦 = 4.1𝑀𝐺𝜀 2 ƒ2 ( )2 𝑏

∆𝜎𝑚𝑜𝑑𝑢𝑙𝑢𝑠 = M

𝐺𝑏 𝐿

[1 − (

𝐸𝑝 2 3/4 ) ] 𝐸𝑚

(5) (6) (7)

where M = 3 is the Taylor factor for bcc matrix, b = 0.25 nm is the Burgers vector of the matrix, L is the mean precipitate spacing in the slip plane, calculated by 1⁄ 2,

0.866/(𝑅𝑁)

R and N are the average precipitate radius and number density, T is

the line tension associated with the dislocation, calculated by Gb2/2, and G = 80 GPa is the shear modulus of the ferrite matrix, γinterfacial = 0.37 J m-2 is the apparent surface 2 ∆𝑎

energy [19], 𝜀 = ( ) is the constrained lattice parameter mismatch, f = 3/4πR3N is 3

𝑎

the volume fraction of precipitates, Ep and Em are the dislocation line energy in the matrix and the precipitates, respectively, and the ratio of Ep/Em depends on the particle radius R [33]. The R and N of the Cu-rich precipitates in the Fe-Cu alloy can be obtained by the Table 3. However, the Cu-rich phase and NiAl phase in the Fe-Cu-Ni-Al alloy exhibit a core-shell structure, we combine them as a whole to calculate the strength. The R and N of the co-precipitates are 1.7±0.5 nm and 11.7×1023 m-3, respectively. Using the experimental data determined by APT, the strengthening contributions from each individual mechanism are calculated as shown in Fig.8. The total precipitation strengthening are determined to be ~209 and ~332 MPa for the Fe-Cu and Fe-Cu-Ni-Al alloys at the peak hardness states, which are close to the experimental values from the tensile results (Table 2). Therefore, this indicates that the strengthening of the peak hardness states is controlled by the dislocation-precipitate cutting mechanism. With increasing ageing time, the Cu-rich phases of the Fe-Cu and Fe-Cu-Ni-Al 11

alloys increase in size and decrease in number density. After ageing for 128 h, the precipitate size of the Fe-Cu alloy grows to 14.2 nm, with the number density decreasing to 0.02×1023 m-3. The Cu-rich precipitates of the Fe-Cu-Ni-Al alloy increase to 3.1 nm in size and the number density decreases to 0.9×1023 m-3. The coarsening of precipitates in both alloys leads to the decrease in hardness (Fig.1). In the coarsening, the Cu precipitates of the Fe-Cu-Ni-Al alloy begin to transform from bcc to 9R and fcc Cu [34-36]. Because of the difference in structure, the NiAl phase will prefer to grow on one side of Cu-rich phase (Fig.6c).

5. Conclusions (1) The co-addition of Ni and Al atoms to the Fe-Cu alloys has caused both Cu-rich phase and NiAl phase precipitation, leading to a strong precipitation strengthening effect on top of the alloy strengthening. (2) The co-addition of Ni and Al atoms increases the number density of the Cu-rich precipitates, which is due to the fact that Ni and Al can increase the nucleation rate of Cu-rich phase. (3) At the peak ageing condition (4 h ageing), the precipitates in the Fe-Cu-Ni-Al alloy exhibit a core-shell structure with the Cu-rich phase in the core and NiAl phase in the outer shell. With increasing ageing time to 128 h, the core-shell structured precipitate decomposed, forming separate NiAl precipitate and Cu-rich precipitate. (4) The B2 structured NiAl shell in the Fe-Cu-Ni-Al alloy impedes the growth and coarsening of the Cu-rich phase by decreasing the interfacial energy and the diffusion rate of Cu, Ni and Al atoms in the shell. Moreover, the Cu-rich phase in the Fe-Cu-Ni-Al alloy has a small difference in sizes as compared with that in the Fe-Cu alloy, leading to a slow coarsening rate.

Acknowledgements This work was supported by National Key Research and Development Program of China [No. 2017YFB0703002] and the State Key Lab of Rolling and Automation of Northeastern University Development Fund [No. 2016002]. 12

References [1] W.H. Zhou, H. Guo, Z.J. Xie, C.J. Shang, R.D.K. Misra, Copper precipitation and its impact on mechanical properties in a low carbon microalloyed steel processed by a three-step heat treatment, Mater. Des. 63 (2014) 42-49. [2] N. Maruyama, M. Sugiyama, T. Hara, H. Tamehiro, Predipitation and phase transformation of cooper particles in low alloy ferritic and martensitic steels, Mater. Trans. JIM. 40 (1999) 268-277. [3] M. Hattestrand, J. O. Nilsson, K. Stiller, P. Liu, M. Andersson, Precipitation hardening in a 12%Cr–9%Ni–4%Mo–2%Cu stainless steel, Acta. Mater. 52 (2004) 1023-1037. [4] Z.W. Zhang, C.T. Liu, X.L. Wang, M.K. Miller, D. Ma, G. Chen, J.R. Williams, B.A. Chin, Effects of proton irradiation on nanocluster precipitation in ferritic steel containing fcc alloying additions, Acta. Mater. 60 (2012) 3034-3046. [5] X.Q. Xu, X.F. Zhang, G.L. Chen, Z.P. Lu, Improvement of high-temperature oxidation resistance and strength in alumina-forming austenitic stainless steels, Mater. Lett. 65 (2011) 3285-3288. [6] R.P. Kolli, Z. Mao, D.N. Seidman, D.T. Keane, Identification of a Ni0.5(Al0.5-xMnx) B2 phase at the heterophase interfaces of Cu-rich precipitates in an α-Fe matrix, Appl. Phys. Lett. 91 (2007) 241903. [7] Z.K. Teng, M.K. Miller, G. Ghosh, C.T. Liu, S. Huang, K.F. Russell, M.E. Fine, P.K. Liaw, Characterization of nanoscale NiAl-type precipitates in a ferritic steel by electron microscopy and atom probe tomography, Scr. Mater. 63 (2010) 61-64. [8] H. Leitner, M. Schober, R. Schnitzer, Splitting phenomenon in the precipitation evolution in an Fe-Ni-Al-Ti-Cr stainless steel, Acta. Mater. 58 (2010)1261-1269. [9] M. Kapoor, D. Isheim, G. Ghosh, S. Vaynman, M.E. Fine, Y.W. Chung, Ageing characteristics and mechanical properties of 1600 MPa body-centered cubic Cu and B2-NiA precipitation-strengthened ferritic steel, Acta. Mater. 73 (2014) 56-74. [10] D. Isheim, R.P. Kolli, M.E. Fine, D.N. Seidman, An atom-probe tomographic study of the temporal evolution of the nanostructure of Fe-Cu based high-strength 13

low-carbon steels, Scr. Mater. 55 (2006) 35-40. [11] 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-2.5Cu-1.5Mn-4.0Ni-1.0Al multicomponent ferritic alloy, Acta. Mater. 61 (2013) 2133-2147. [12] D. Isheim, D.N. Seidman, Nanoscale studies of segregation at coherent heterophase interfaces in α-Fe based systems, Surf. Interface Anal. 36 (2004) 569-574. [13] 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 (2013) 5996-6005. [14] 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, Materials Taday 20 (2017) 142-154. [15] R.P. Kolli, D.N. Seidman, The temporal evolution of the decomposition of a concentrated multicomponent Fe-Cu-based steel, Acta. Mater. 56 (2008) 2073-2088. [16] X.J. Wang, G. Sha, Q. Shen, W.Q. Liu, Age-hardening effect and formation of nanoscale composite precipitates in a NiAlMnCu-containing steel, Mater. Sci. Eng. A 627 (2015) 340-347. [17] D. Isheim, M.S. Gagliano, M.E. Fine, D.N. Seidman, Interfacial segregation at Cu-rich precipitates in a high-strength low-carbon steel studied on a sub-nanometer scale, Acta. Mater. 54 (2006) 841-849. [18] Q. Shen, H. Chen, W.Q. Liu, Effect of Cu on Nanoscale precipitation evolution and mechanical properties of a Fe-NiAl alloy, Microsc. Microanal. 23 (2017) 350-359. [19] Z.W. Zhang, C.T. Liu, M.K. Miller, X.L. Wang, Y.R. Wen, T. Fujita, A. Hirata, M.W. Chen, G. Chen, B.A. Chin, A nanoscale co-precipitation approach for property enhancement of Fe-base alloys, Sci. Rep. 3 (2013) 1327. [20] M.K. Miller. Atom Probe Tomography: Analysis at the Atomic Level. Kluwer Academic/Plenum Publishers, New York , 2000, pp. 160. 14

[21] D. Vaumousse, A. Cerzo, P.J. Warren, A procedure for quantification of precipitate microstructures from three-dimensional atom probe data, Ultramicroscopy 95 (2003) 215-221. [22] M.K. Miller, E.A. Kemik, Atom probe tomography: A technique for nanoscale characterization, Microsc. Microanal. 10 (2004) 336-341. [23] R.P. Kolli, D.N. Seidman, Comparison of compositional and morphological atom-probe tomography analyses for a multicomponent Fe-Cu Steel, Microsc. Microanal. 13 (2007) 272-284. [24]

O.C.

Hellman,

J.

Rusing,

D.N.

Seidman,

Efficient

sampling

for

three-dimensional atom probe microscopy data, Ultramicroscopy 95 (2003), 199-205. [25] O.C. Hellman, J.A. Vandenbrouche, J. Rusing, D. Isheim, D.N. Seidman, Analysis of Three-dimensional Atom-probe data by the proximity histogram, Microsc. Microanal. 6 (2000) 437-444. [26] 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. [27] H.I. Aaronson, F.K. LeGoues, An assessment of studies on homogeneous diffusional nucleation kinetics in binary metallic alloys, Metall. Trans. A 23 (1992) 1915-1945. [28] A. Takeuchi, A. Inoue, Classification of bulk metallic glasses by atomic size difference, heat of mixing and period of constituent elements and its application to characterization of the main alloying element, Mater. Trans. 46 (2005) 2817-2829. [29] M.K. Miller, P. Pareige, M.G. Burke, Understanding pressure vessel steels: an atom probe perspective, Mater. Charact. 44 (2000) 235-254. [30] A. Seko, N. Odagaki, S.R. Nishitani, I. Tanaka, H. Adachi, Free-Energy Calculation of Precipitate Nucleation in an Fe-Cu-Ni Alloy, Mater. Trans. 45 (2004) 1978-1981. [31] Z.C. Liu, Z.X. Yuan, Y.C. Liu, Solid Phase Transformation. Machinery Industry Press, Beijing, 2010, pp. 238. [32] Z.B. Jiao, J.H. Luan, M.K. Miller, C.Y. Yu, C.T. Liu, Effects of Mn partitioning 15

on nanoscale precipitation and mechanical properties of ferritic steels strengthened by NiAl nanoparticles, Acta. Mater. 84 (2015) 283-291. [33] K.C. Russell, L.M. Brown, A dispersion strengthening model based on different elastic moduli applied to the Fe-Cu system, Acta. Metall. 20 (1972) 969-974. [34] P.J. Othen, M.L. Jenkins, G.D.W. Smith, High-resolution electron microscopy studies of the structure of Cu precipitates in α-Fe, Phil. Mag. A 70 (1994) 1-24. [35] G. Han, Z.J. Xie, Z.Y. Li, B. Lei, C.J. Shang, R.D.K. Misra, Evolution of crystal structure of Cu precipitates in a low carbon steel, Mater. Des. 135 (2017) 92-101. [36] R. Monzen, M.L. Jenkins, A.P. Sutton, The bcc-to-9R martensitic transformation of Cu precipitates and the relaxation process of elastic strains in an Fe-Cu alloy, Philos. Mag. A 80 (2000) 711-723.

16

Fig. 1 Microhardness as a function of ageing time at 500 °C for the Fe-Cu and Fe-Cu-Ni-Al alloys Fig. 2 Room-temperature tensile stress-strain curves of the Fe-Cu and Fe-Cu-Ni-Al alloys in the as-quenched (AQ) and peak hardness states Fig. 3 Scan electron micrographs of room-temperature fracture surfaces for (a) Fe-Cu, as-quenched; (b) Fe-Cu, aged for 0.25 h; (c) Fe-Cu-Ni-Al, as-quenched; and (d) Fe-Cu-Ni-Al, aged for 4 h. The scale bars in (a), (c) and (d) are 20 μm, the scale bars in (b) is 50 μm Fig. 4 Three-dimensional atom maps of Fe-Cu (a, c, e, g) and Fe-Cu-Ni-Al (b, d, f, h) alloys after ageing at 500 °C for 0.1 h (a, b), 0.25 h (c, d), 4 h (e, f) and 128 h (g, h) Fig. 5 The nearest-neighbor distribution (NND) analysis of Cu atoms in the Fe-Cu alloy (a) and Cu, Ni and Al atoms in the Fe-Cu-Ni-Al alloy (b, c, d) after ageing for 0.1 h Fig. 6 Cu (green), Ni (red) and Al (blue) atom maps and one-dimensional composition profiles of precipitates in the Fe-Cu-Ni-Al alloy 500 °C aged for (a, d) 0.25 h, (b, e) 4 h, (c, f) 128 h in the 15 × 15 × 3 nm3 selected regions. The 3-nm-thick selected regions were cut through the centers of the precipitates in the Fe-Cu-Ni-Al alloy. The circle around the Cu segregation zone is a 20 at.% Cu isoconcentration surface, the circle around the Ni and Al segregation zone is a 15 at.% (Ni+Al) isoconcentration surface Fig. 7 Size distribution integrated of Cu-rich phases in the Fe-Cu (a, b, c) and Fe-Cu-Ni-Al (d, e, f) alloy samples aged at 500 °C for (a, d) 0.25 h, (b, e) 4 h and (e, f) 128 h Fig.

8

Strengthening

contributions

of

chemical

strengthening,

coherency

strengthening and modulus strengthening of precipitates in the Fe-Cu and Fe-Cu-Ni-Al alloy Table 1 Chemical compositions of the experimental alloys (wt.%) Table 2 The yield strength (YS), ultimate tensile strength (UTS) and elongation-to-failure (EL) of the Fe-Cu and Fe-Cu-Ni-Al alloys in the as-quenched 17

(AQ) and peak hardness states Table 3 Average radius (rp) and number density (Nv) of Cu-rich phase in the Fe-Cu and Fe-Cu-Ni-Al alloys Table 4 Core compositions (at.%) of the Cu-rich phases obtained by proxigrams after ageing at 500 °C for 0.25 h, 4 h and 128 h

Fig. 1 Microhardness as a function of ageing time at 500 °C for the Fe-Cu and Fe-Cu-Ni-Al alloys

18

Fig. 2 Room-temperature tensile stress-strain curves of the Fe-Cu and Fe-Cu-Ni-Al alloys in the as-quenched (AQ) and peak hardness states

Fig. 3 Scan electron micrographs of room-temperature fracture surfaces for (a) Fe-Cu, 19

as-quenched; (b) Fe-Cu, aged for 0.25 h; (c) Fe-Cu-Ni-Al, as-quenched; and (d) Fe-Cu-Ni-Al, aged for 4 h. The scale bars in (a), (c) and (d) are 20 μm, the scale bars in (b) is 50 μm

Fig. 4 Three-dimensional atom maps of Fe-Cu (a, c, e, g) and Fe-Cu-Ni-Al (b, d, f, h) alloys after ageing at 500 °C for 0.1 h (a, b), 0.25 h (c, d), 4 h (e, f) and 128 h (g, h)

20

Fig. 5 The nearest-neighbor distribution (NND) analysis of Cu atoms in the Fe-Cu alloy (a) and Cu, Ni and Al atoms in the Fe-Cu-Ni-Al alloy (b, c, d) after ageing for 0.1 h

21

Fig. 6 Cu (green), Ni (red) and Al (blue) atom maps and one-dimensional composition profiles of precipitates in the Fe-Cu-Ni-Al alloy 500 °C aged for (a, d) 0.25 h, (b, e) 4 h, (c, f) 128 h in the 15 × 15 × 3 nm3 selected regions. The 3-nm-thick selected regions were cut through the centers of the precipitates in the Fe-Cu-Ni-Al alloy. The circle around the Cu segregation zone is a 20 at.% Cu isoconcentration surface, the circle around the Ni and Al segregation zone is a 15 at.% (Ni+Al) isoconcentration surface

22

Fig. 7 Size distribution integrated of Cu-rich phases in the Fe-Cu (a, b, c) and Fe-Cu-Ni-Al (d, e, f) alloy samples aged at 500 °C for (a, d) 0.25 h, (b, e) 4 h and (e, f) 128 h

Fig. 8 Strengthening contributions of chemical strengthening, coherency strengthening and modulus strengthening of precipitates in the Fe-Cu and Fe-Cu-Ni-Al alloy

23

Table 1 Chemical compositions of the experimental alloys (wt.%) Alloy Fe-Cu Fe-Cu-Ni-Al

Cu 1.63 1.62

Ni

Al

3.26

1.10

Fe Bal. Bal.

Table 2 The yield strength (YS), ultimate tensile strength (UTS) and elongation-to-failure (EL) of the Fe-Cu and Fe-Cu-Ni-Al alloys in the as-quenched (AQ) and peak hardness states Model alloys YS(MPa) UTS(MPa) EL(%) Fe-Cu, AQ 207±4.9 317±3.8 25.3±4.1 Fe-Cu, 0.25 h 400±8.5 534±15.6 6.1±0.4 Fe-Cu-Ni-Al, AQ 655±39.0 749±37.9 10.9±0.9 Fe-Cu-Ni-Al, 4 h 1023±13.2 1053±1.0 10.1±1.4

Table 3 Average radius (rp) and number density (Nv) of Cu-rich phase in the Fe-Cu and Fe-Cu-Ni-Al alloys r /nm Fe-Cu Fe-Cu-Ni-Al 1.2±0.5 1.0±0.4 2.3±0.8 1.4±0.5 14.2 3.1±0.8

Ageing Time/h 0.25 4 128

Nv/×1023 m-3 Fe-Cu Fe-Cu-Ni-Al 9.55 14.2 1.7 12.2 0.02 0.9

Table 4 Core compositions (at.%) of the Cu-rich phases obtained by proxigrams after ageing at 500 °C for 0.25 h, 4 h and 128 h at.%

0.25 h

4h

128 h

Fe-Cu

Fe-Cu-Ni-Al

Fe-Cu

Fe-Cu-Ni-Al

Fe-Cu

Fe-Cu-Ni-Al

Cu

82.7±3.6

63.0±3.2

91.2±1.2

70.0±2.1

98.4±3.8

76.0±4.7

Fe

17.0±3.6

20.9±2.7

8.8±1.2

12.9±1.6

1.5±2.6

7.4±1.9

Ni

--

5.8±1.6

--

5.5±1.1

--

5.8±2.7

Al

--

9.9±2.0

--

11.5±1.5

--

10.8±2.1

24