Thermal stability and grain boundary strengthening in

Materials and Design 134 (2017) 426–433

Contents lists available at ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/matdes

Thermal stability and grain boundary strengthening in ultrafine-grained CoCrFeNi high entropy alloy composite Praveen Sathiyamoorthi a,⁎, Joysurya Basu b, Sanjay Kashyap c, K.G. Pradeep d, Ravi Sankar Kottada a a

Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600 036, India Department of Metallurgical Engineering, Indian Institute of Technology BHU, Varanasi 221 005, India Department of Materials Engineering, BML Munjal University, Gurgaon, Haryana 122413, India d Materials Chemistry, RWTH Aachen University, Kopernikusstr.10, 52074 Aachen, Germany b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Excellent resistance to anneal softening at 0.56 to 0.68 Tm • Negligible decrease in hardness even after annealing at 0.56 Tm for 600 h • High growth exponent (~4) in ultrafinegrained CoCrFeNi alloy • Major contribution to strengthening is from grain boundary strengthening.

a r t i c l e

i n f o

Article history: Received 10 May 2017 Received in revised form 23 August 2017 Accepted 26 August 2017 Available online 30 August 2017 Keywords: High entropy alloy Thermal stability Atom probe tomography Grain growth, strengthening

a b s t r a c t Thermal stability of CoCrFeNi high entropy alloy in as-milled and sintered conditions was investigated using Xray diffraction, differential scanning calorimetry, transmission electron microscopy, and atom probe tomography. Composite microstructure consists of FCC and carbide with a fine dispersion of oxide was observed in the sintered condition. Unsolicited contamination of carbon and oxygen in the as-milled powder due to the milling medium had led to the formation of composite microstructure. An exceptional thermal stability was observed upon exposure of sintered compact to higher temperatures (0.56 Tm to 0.68 Tm) for the prolonged duration of 600 h. Sintered compact exposed to 700 °C (0.56 Tm) for 600 h showed negligible change in hardness and grain size. Analysis based on the modified Hall-Petch model for two phase alloy indicates the phase boundaries act as a strong obstacle while the major contribution to strengthening comes from grain boundaries. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction Alloy design, since its discovery [1,2], is based on one principal element with the addition of other alloying elements in minor amounts to get the desired properties. Limitations on alloy design with only one or two principal elements are based on the premise that increasing the number of principal elements would result in the formation of complex compounds, and the associated difficulties in obtaining ⁎ Corresponding author. E-mail address: [email protected] (P. Sathiyamoorthi).

http://dx.doi.org/10.1016/j.matdes.2017.08.053 0264-1275/© 2017 Elsevier Ltd. All rights reserved.

thermodynamic data to predict phase formations [2,3]. On the contrary, Yeh et al. and Cantor et al. had independently shown that alloys with multi principal elements form a simple structure with less number of phases [4,5]. Yeh and his co-workers proposed that the increase in configuration entropy associated with a number of principal elements as the possible reason for the formation of simple structures [6,7], and hence the name High Entropy Alloys (HEAs). The HEAs are reported to have four core effects: high entropy effect which aids in stabilizing the solid solution, sluggish diffusion effect which aids in slowing down the kinetics, lattice distortion effect which influences the properties to a great extent, and cocktail effect helps in achieving a composite effect

P. Sathiyamoorthi et al. / Materials and Design 134 (2017) 426–433

on properties [8–10]. The discovery of HEAs has widened the window of alloy design, and their four core effects have motivated major research groups to investigate various properties of HEAs [11–17] synthesized through various routes such as vacuum arc melting [18], vacuum induction melting [19], mechanical alloying (MA) [20] followed by sintering [21], sputtering [22], plasma nitriding [23], and cladding [24]. Among the processing routes, MA is a solid state processing route originally developed to produce oxide dispersion strengthened (ODS) nickel and iron base superalloys [25]. Versatility and advantages of MA are very well documented in the literature and hence are not elaborated here [26,27]. Similarly, among powder consolidation techniques, spark plasma sintering (SPS) has the ability to retain the nanocrystalline nature while attaining the fully dense samples [28]. Thermal stability of dense consolidated samples is of immense interest if they have the potential to be considered in moderate to higher temperature applications. Thus, the present study is aimed at exploiting the intrinsic cocktail properties and sluggish diffusivity of HEAs together with the nanocrystalline structure obtained via MA and SPS for thermal stability studies. In spite of the initial promise by HEAs for a large number of compositions, most of the studies are limited to few elemental compositions. Among them, HEAs consisting of Co, Cr, Fe, and Ni with the addition of other elements such as Al/Ti/Mn are more prominent than any other combinations [29–33]. [Refer Supplementary information for detailed information]. However, thermal stability studies of these HEAs are limited in the literature [34–37]. For example, Wu et al. studied the thermal stability of a family of FCC phase alloys: CoCrFeNiMn, CoCrFeNi, and subsets of CoCrFeNi synthesized by arc melting followed by cold rolling and annealing [34]. Grain growth [38] and sluggish diffusivity [39] studies are explored in CoCrFeNiMn alloy synthesized by casting route. In one of our earlier studies on CoCrFeNi alloy synthesized by MA followed by SPS, we reported an exceptional resistance to grain growth and attributed it to the composite microstructure obtained due to the contamination of C during ball milling [40]. However, the effect of impurities on thermal stability, grain growth kinetics, and strengthening contributions has not been analyzed in detail and has not been reported so far. Thus, based on this brief background, this study is aimed at answering the following specific questions: a. how do the carbides and oxides influence the thermal stability and grain growth kinetics in HEAs?, b. what is the extent of strengthening by carbides and oxides in these kinds of HEAs?, c. Does the analysis presented here provide cues to design thermally stable microstructures?.

427

660), following the procedures described in Ref [41]. APT data evaluation was performed using IVAS 3.6.10 software provided by Cameca Instruments. Vickers hardness (Wolpert Wilson) with the load of 1 kgf and dwell time of 10 s was used to measure the hardness of the sintered pellets. Hardness values reported here are an average of 10 readings performed on both sides of the pellets. The grain size is measured using ImageJ software (open source software developed by National Institute of Health (NIH) of USA). The longest distance/intercept along the grain is measured from one end of the grain boundary to another end of the grain boundary within a grain. More than 150 grains were measured to get an average value and standard deviation. 3. Results 3.1. Phase evolution

2. Experimental details

Fig. 1 illustrates the evolution of phases in CoCrFeNi from the starting powder to as-milled condition to as-sintered condition. In comparison to starting powder, XRD pattern of as-milled powder shows the absence of starting elemental peaks and formation of major FCC phase with peak shift and significant peak broadening. The absence of elemental diffraction peaks together with peak shift suggest the dissolution of elements and alloy formation, whereas peak broadening suggests the refinement of crystallite size and generation of strains during MA. The crystallite size and the strain estimated from the FCC (111-most intense peak) using line profile analysis were 5 nm and 0.7%, respectively. The XRD pattern of sintered CoCrFeNi (Fig. 1) indicates the evolution of Cr7C3 and Cr2O3 as secondary phases with the retention of FCC as a primary phase, and a significant decrease in peak broadening as compared to the as-milled condition. The decrease in peak broadening suggests the crystallite growth and/or the annihilation of strains. The estimated crystallite size and strain from the FCC (111) peak of sintered CoCrFeNi alloy were 60 nm and 0.06%, respectively. It is important to note that the second phase was interpreted to be a sigma phase [21,42,43] in the very early studies on this material, due to lack of access to advanced characterization techniques such as TEM or APT. However, the reported sigma phase is conclusively identified to be a carbide phase based on the detailed analysis of TEM, in our recent studies [40] and in APT studies, which will be discussed in the later sections. Since carbon/oxygen was not added deliberately, the formation of carbide and oxide suggests contamination of carbon and oxygen in the as-milled powder. The source of contamination could be the milling medium and/or the process controlling agent (PCA) used. It has been

The equimolar CoCrFeNi alloy was synthesized by MA from pure metal powders and subsequently consolidated at 900 °C in an SPS unit. More details on the processing of CoCrFeNi equimolar powders by MA and subsequent SPS were described in a previous study [40]. Thermal analysis and heat treatment of powders were carried out in Differential Scanning Calorimetry (DSC; Netzsch, Exton, PA) using alumina crucibles with the heating rate of 30 °C per min in an argon atmosphere and then furnace cooled to room temperature. Heat treatment of sintered pellets was carried out in a tubular furnace in the air, and the pellets were air cooled to room temperature. Fully dense sintered pellets were annealed with a regular interval of 120 h for 600 h at 700 °C, 800 °C, 900 °C and for 240 h at 1000 °C. PANalytical Xpert Pro XRD with Cu target and X'celerator detector was utilized for investigating the structural evolution at room temperature. XRD pattern was acquired using the scanning parameter with 0.02° step size and 20 s time per step. Transmission electron microscopy (TEM) studies were carried out using FEI-Tecnai G2 F30 TEM. The tip temperature was maintained at ~60 K. Elemental distribution at the atomic scale in the sintered compacts was investigated using a local electrode atom probe (LEAP™ 4000X HR, Cameca Instruments) applying laser pulse at 250 kHz repetition rate. Samples for atom probe tomography (APT) were prepared using a dual beam focused ion beam (FEI Helios Nanolab

Fig. 1. XRD patterns of CoCrFeNi showing phase evolution after MA and SPS illustrating the transformation of BCC after MA to Cr7C3, and formation of Cr2O3 after SPS.

428


widely reported that the alloys synthesized through MA route are contaminated with interstitial impurities due to the milling medium and the PCA used [26,27]. The estimated oxygen and carbon content in the as-milled powder using LECO analyzer was 1.2 wt% and 1.3 wt%, respectively. 3.2. Thermal analysis of MA CoCrFeNi The DSC trace of CoCrFeNi MA powder (Fig. 2a) shows one exothermic peak at 600 °C and two endothermic peaks at 1280 °C and 1455 °C. The endothermic peak at 1455 °C can be attributed to the melting point of the alloy. In one of the studies, Wu et al. also reported that the melting of CoCrFeNi starts at 1422 °C [34]. To determine the origin for other peaks, MA powder was annealed for 15 min at different temperatures followed by furnace cooling and subjected to XRD at room temperature. The XRD patterns of the heat treated powder at different temperatures (Fig. 2b) clearly indicate the evolution of Cr7C3 around 600 °C, complete dissolution of minor BCC above 600 °C, and dissolution of Cr7C3 above 1000 °C. The decrease in BCC peak intensity above 500 °C and the appearance of Cr7C3 peaks above 500 °C (Fig. 2b) suggest that BCC phase transforms to carbide phase. This transformation can be attributed to the exothermic peak observed at 600 °C. No further phase changes were observed in the milled powders when exposed in the temperature range of 700 °C–1000 °C. However, carbide peaks intensity gradually decreased upon exposure above 1000 °C and disappeared above 1200 °C. This observation suggests that the carbide is thermally stable till 1000 °C and above which it dissolves leading to the formation of single FCC phase. Thus, this dissolution of carbide can be attributed to the endothermic peak observed around 1300 °C. The TGA curve shown along with DSC curve in Fig. 2a indicates a gradual increase in weight percentage above 500 °C till 1000 °C, and a subsequent gradual decrease in weight percentage above 1000 °C till the melting point. The discussion on TGA curve will be presented in Section 4.1. 3.3. Thermal stability of as-sintered CoCrFeNi Fig. 3a shows the DSC trace with TGA curve of the sintered pellet. Since Cr7C3 phase has already formed during sintering (Fig. 2a), no exothermic peak was observed at 600 °C. However, two endothermic peaks were observed, similar to DSC trace of MA powders, one at 1315 °C which can be attributed to Cr7C3 dissolution, and other at 1455 °C which can be attributed to the melting of the alloy. Similar to the TGA curve of MA powder, the TGA curve of sintered pellet shows a decrease in weight percentage above 1000 °C till the melting point. More details with regard to TGA behavior will be discussed in Section 4.1. Since the

carbide phase is thermally stable till 1000 °C, the thermal stability of the sintered pellet was investigated in the temperature range of 700 °C to 900 °C. Fig. 3b shows the XRD patterns of sintered pellets exposed to high temperatures for a duration of 600 h (25 days) at 700 °C to 900 °C. At all the temperatures, phases observed were similar to the phases observed in the sintered pellet. This indicates an excellent phase stability of this alloy. Fig. 4 shows the TEM image of the as-sintered condition and at different annealing temperatures for the holding time of 600 h. The microstructure consists of FCC phase (A), carbide phase (B), and oxide precipitate (C). The average grain size (FCC + carbide) calculated from the TEM images using ImageJ analysis were 120 ± 30 nm, 130 ± 40, 150 ± 35, and 260 ± 75 nm for as-sintered, 700 °C - 600 h, 800 °C - 600 h, and 900 °C - 600 h heat treated samples, respectively. The grain growth is almost negligible for samples annealed at 700 °C and 800 °C, and it is increased only by a factor of 2 for the sample annealed at 900 °C. The grain size of FCC phase and carbide phase was almost similar at all conditions, and hence only the average grain size is reported here. The volume fraction of phases estimated using reference intensity ratio method from the XRD pattern of SPS and heat treated conditions are almost similar, and the estimated values are 72%, 19%, and 9% for FCC, carbide, and oxide respectively. Fig. 5 illustrates the hardness and grain size variation with time at different annealing temperatures of 700, 800, and 900 °C, respectively. The change in hardness value is negligible for the sample annealed at 700 °C for 600 h (~580 ± 5 to 565 ± 5 HV1). This indicates an excellent resistance to anneal softening. However, a decrease in hardness was observed for the samples heat treated above 700 °C. Fractional decrease in hardness after 600 h at 700 °C, 800 °C, and 900 °C was 2.6%, 9.5%, and 25%, respectively. Fig. 6 illustrates the elemental distribution of Co, Cr, Fe, Ni, and C atoms in the as-sintered condition acquired using APT. It can be seen clearly that the elements are in-homogeneously distributed and are clearly partitioned into two distinct regions: one rich in Co, Fe, and Ni, and the other with Cr and C. Fig. 7 represents the one-dimensional concentration depth profile taken along the cylinder, positioned across the two regions with 0.5 nm bin width. It is clear from the concentration depth profile that Co, Fe, and Ni are equally distributed with a lesser amount of Cr, and the carbide phase is enriched with Cr with a lesser amount of Fe, Co, and Ni in the decreasing order. The Cr\\C depleted and Cr\\C enriched regions could be attributed to the FCC and the Cr7C3 phase observed in the XRD patterns, respectively. It is also interesting to note that the composition of the carbide phase in the as-sintered condition and after heat treatment at 1000 °C for 240 h was almost similar [Table S2 of Supplementary information]. This clearly suggests that not only the phases are stable even the composition of the phases is stable.

Fig. 2. a) Thermal analysis (DSC and TGA) of CoCrFeNi MA powder, b) XRD patterns indicating the phase changes in CoCrFeNi powder after heat treatment.


429

Fig. 3. a) Thermal analysis (DSC and TGA) of CoCrFeNi SPS, b) XRD patterns illustrating the phase stability of sintered CoCrFeNi after heat treatment.

4. Discussion Having established the formation of FCC phase together with the carbide phase and oxide precipitates, it is necessary to rationalize the stability of carbide and oxide. Similarly, it is also important to examine the possible reasons for an excellent thermal stability observed in this alloy. Thus, the following discussion illustrates the various aspects involved in formation and stability of Cr7C3 and Cr2O3, and exceptional resistance to anneal softening and grain growth. 4.1. Formation and stability of Cr7C3 and Cr2O3 In general, there are three types of stable chromium carbides: Cr3C2, Cr7C3, and Cr23C6. The type of carbide that can be formed in an alloy

depends on the activity of C [44]. In most of the steel literature, it has been reported that a transition carbide forms first, followed by the formation of stable carbide due to difficulties in the kinetics of stable carbide nucleation [45,46]. Consistent with this observation, in a recent study on HEA, the transformation of (Cr,Fe)7C3 carbide to (Cr,Fe)23C6 carbide is reported in Al0.3CoCrFeNiC0.1 HEA when heat treated above 600 °C [47]. However, in the present study, no such carbide transformation occurred on annealing the CoCrFeNi sintered pellets, but only Cr7C3 was observed. In order to understand the carbide formation, few ternary alloy powder combinations containing Co, Cr, Fe, and Ni were prepared and were subsequently annealed in the temperature range of 500–700 °C. It is interesting to note that the transformation of Cr7C3 carbide to Cr23C6 carbide was observed in CrFeNi heat treated powder, but not in CoCrNi heat treated powder (Fig. S1 of Supplementary information).

Fig. 4. TEM images of CoCrFeNi alloy depicting composite microstructure: a) as-sintered, b) 700 °C – 600 h, c) 800 °C – 600 h, and d) 900 °C – 600 h. [A-FCC, B-Carbide, C-oxide.]

430


phenomenon that Cr2O3 oxide is very stable till 1000 °C, and above which it transforms to volatile CrO3 oxide [49]. Hence the decrease in weight loss above 1000 °C can be attributed to the volatilization of Cr2O3. It is also possible that the Cr2O3 can form while cooling and hence the observation of oxide peaks in the XRD patterns of the powders heated above 1000 °C (Fig. 2b). Although the experiments were carried out in the argon atmosphere, oxygen present as contamination in the milled powder might be a possible reason for the oxide formation. 4.2. Kinetics of grain growth A general expression for grain growth kinetics is given by n

n

d −do ¼ kt

Fig. 5. Hardness and grain size variation of CoCrFeNi at different heat treatment conditions.

where d is the average grain size after time t, do is the average initial grain size, k is the rate constant, and n is the grain growth exponent. Rate constant k can be expressed by Arrhenius equation as k ¼ ko exp

Carbon contamination estimated in these alloys were ≈1.1 to 1.3 wt%. Thus, for almost similar carbon wt%, the occurrence of Cr7C3 to Cr23C6 transformation in CrFeNi alloy and absence of such transformation in CoCrFeNi and CoCrNi alloys indicate that the presence of Co seems to stabilize Cr7C3 carbide. It is interesting to note that Fang et al. observed the formation of Cr23C6 type carbide in low Co content-Al0.5CrFeNiCo0.3C0.2 HEA upon sintering of MA powder [48]. In general, the formation of Cr7C3 requires higher C activity than Cr23C6 [44]. Since Co is a graphitizing element, for the similar carbon content, it is possible that the activity of C might be higher in alloys with Co and hence it favors/stabilizes the Cr7C3 carbide rather than Cr23C6. As shown in TGA curves of Figs. 2a and 3a, there is a weight gain in the temperature regime of 400–1000 °C, and subsequently weight loss above 1000 °C. Also, the increase in weight gain observed in TGA of CoCrFeNi MA powder is substantially higher than that of the sintered pellet. Clearly, this weight gain can be attributed to the formation of Cr2O3, as it is also consistent with the observation of Cr2O3 peaks in the XRD patterns of annealed powders (Fig. 2b). Although the oxide diffraction peaks are not clearly visible at 500 °C and 600 °C (Fig. 2b), it is possible that fraction of oxides formed at these temperatures could be very less. The TGA of sintered pellet does not show much of the weight gain when compared to the as-milled powder as the available surface area will be less in sintering compact than the MA powder. However, the decrease in weight loss above 1000 °C is quite similar in both MA powder and sintered pellet. This weight loss is quite intriguing since there is no element of CoCrFeNi that can volatilize at these temperatures. The possible explanation for such a decrease in weight percentage can be explained based on the stability of Cr2O3. It is a well-known

ð1Þ

−Q

RT

ð2Þ

where ko is a constant, Q is the activation energy, R is the gas constant, and T is the absolute temperature. Since the increase in grain growth is not significant even after annealing at 900 °C for 600 h, estimation of grain growth exponent (n) by plotting log d vs log t is not possible with the available experimental data. However, it is possible to make an approximate estimate of n based on assuming Q for grain growth to be equivalent to that of lattice diffusion of rate controlling element in HEA. Although diffusivity data for HEAs are quite limited in the literature, recently, Tsai et al. reported an activation energy for diffusion in CoCrFeMnNi to be in the range 290 to 320 kJ/mol [39]. Similarly, in another study, Liu et al. reported an activation energy of 321.7 kJ/mol for grain growth in CoCrFeMnNi [38]. Thus, assuming the activation energy for CoCrFeNi alloy in the present study to be ≈ 320 kJ/mol, the grain growth exponent (n) estimated for t = 600 h using the Eqs. (1) and (2) is ≈4. The estimated grain growth exponent (n) is almost similar to the grain growth exponent reported in the microcrystalline CoCrFeNi (n ≈ 4) [34], and it is higher than the grain growth exponent reported in microcrystalline CoCrFeMnNi (n ≈ 3) [38]. In general, the growth rate is inversely proportional to the average grain size, and hence the grain growth rate in the nanocrystalline material is anticipated to be higher than that of the microcrystalline counterpart. However, observation of similar/higher grain growth exponent (n ~ 4) in the present CoCrFeNi alloy with the grain size much finer than microcrystalline HEAs indicates a stronger resistance to grain growth. This excellent resistance to grain growth can be attributed to the composite microstructure obtained in the present study. Fig. 8 shows the schematic of composite microstructure [two phase aggregate with a fine dispersion of oxide] similar

Fig. 6. APT shows distribution of individual atoms in as sintered CoCrFeNi compact for an analyzed volume of 107 × 106 × 193 nm3.


431

Fig. 7. Concentration-depth profile (right) of individual elements in the sintered compact of CoCrFeNi taken along the analyzed volume of 10 × 10 × 60 nm3 (left).

to that of the as-sintered microstructure of the present CoCrFeNi alloy [Fig. 4a]. Grain A, consisting of the multi-principal elements, appears to be having reduced diffusivities inherently. Grain B, with volume fraction lower than A, requires long range diffusion for its grain to grow. Though the volume fraction of B is low, the grain size of A and B in the present study is of almost similar size. Interestingly, the grain size of both A and B is similar even after heat treatment. Oxide precipitate, distributed randomly along the grain A and B, can act as a pinning factor to both A and B. A comprehensive report on exceptional resistance to grain growth is reported elsewhere [40]. Although the present three phase microstructure is a result of unsolicited contamination of C and O, it has resulted in exhibiting excellent thermal stability up to 700 °C.

4.3. Grain boundary strengthening Sriharitha et al. [50] estimated the contribution of various strengthening mechanisms including the Hall-Petch contribution to the hardness of AlxCoCrCuFeNi alloys prepared by MA and SPS. Similarly, He et al. [51] estimated the contribution of tensile strength due to different strengthening mechanisms in the cast CoCrFeNi alloy with the small addition of Ti and Al. Thus, in the present CoCrFeNi alloy, an analysis was made to estimate the contribution from various phases to overall hardness of the alloy. Hardness dependence on grain size can be expressed in the form of Hall-Petch equation, which is given by −1 H ¼ Ho þ kh d =2

ð3Þ

where H is the hardness in HV at a grain size d, Ho is the hardness intercept at d−1/2 = 0, and kh is the Hall-Petch slope. From the hardness vs d−1/2 plot (Fig. 9), the values of Ho and kh were estimated as 108 HV and 163 HV μm−1/2, respectively.

Fig. 8. Schematic of composite microstructure observed in the as-sintered condition. [A FCC-HEA phase, B - carbide, C - oxide].

Fig. 9. Hardness as a function of grain size depicting the Hall-Petch equation. [The data points are fitted using linear regression, and the equation of the line is y = 162.87× + 108.28].

432


Wu et al. also observed the similar values for single FCC phase CoCrFeNi alloy [Ho: 118 HV, kh: 165.5 HV μm1/2] [34]. In a similar study, Liu et al. reported the value of Ho = 125 HV and kh = 70 HV μm1/2 for single phase CoCrFeMnNi alloy [38]. In a recent study, a very high friction stress of 218 MPa is reported for CoCrNi medium entropy alloy, and it is attributed to the severe local distortion [52]. The Hall-Petch relation expressed in Eq. (3) is for single phase alloys. Since in the present study, two phase mixture (FCC + Cr7C3) is observed. The modified Hall-Petch relation for two phase alloys proposed by Fan et al. [53] is used to understand the contribution of strengthening from grain boundaries and phase boundaries. The modified Hall-Petch relation is expressed as −1=2 −1=2 H ¼ Ho;α þ kα dα f αc þ Ho;ε þ kε dε f εc −1=2 Fs þ Ho;αε þ kαε dαε

ð4Þ

The overall efficiency of strengthening, kh, to the efficiencies of different boundaries can be expressed as [53] kh ¼ kα f αc þ kε f εc þ kαε Fs

ð5Þ

The overall hardness Ho at d = ∞ is expressed as [45] Ho ¼ Ho;α f αc þ Ho;ε f εc þ Ho;αε Fs

strengthening may be primarily because of precipitation strengthening and grain boundary strengthening. Thus, the total hardness can be expressed as H ¼ Hi þ ΔHg þ ΔHp

ð7Þ

where Hi is the intrinsic hardness, ΔHg and ΔHp are the contribution from grain boundary strengthening and precipitation strengthening, respectively. The Ho of CoFeNi alloy [97.3 HV] is taken as the intrinsic hardness. The hardness contribution from the grain boundary strengthening (H\\Ho) can be estimated as 470 HV. Consequently, the grain boundary strengthening and the intrinsic hardness account for 567.3 HV out of the measured hardness of 580 HV. In a recent study on AlCoCrCuFeNi, it is reported that strengthening contribution from grain boundary strengthening accounts for 85% of the observed flow stress [55].The remaining hardness of 12.7 HV could be attributed to the presence of oxides. It is clear that the major source of strengthening in the present study is grain boundary strengthening and the role of precipitates in strengthening appears to be very low. In one of our earlier studies on oxide dispersed HE alloy, the influence of oxide in strengthening is observed to be negligible [56]. 4.4. Designing thermally stable microstructure

ð6Þ

where H is the overall hardness, Ho,α, Ho,ε, and Ho,αε are the hardness intercept for FCC, Cr7C3, and mixture of FCC and Cr7C3 phases, respectively, kα, kε, and kαε are the Hall-Petch coefficient of FCC phase, Cr7C3 phase, and mixture of FCC and Cr7C3 phases, respectively, fαc and fεc are continuous volume fraction of FCC and carbide phase, respectively, and Fs is degree of separation of an FCC and Cr7C3 mixture. The detailed explanation of the topological parameters of fαc, fεc, and Fs were described in the Ref [53]. Based on the APT analysis, it is quite reasonable to assume that FCC phase in the present CoCrFeNi alloy is primarily composed of Co, Fe, and Ni. Wu et al. reported the value of Ho = 97.3 HV and kh = 131.1 HV μm1/2 for single FCC phase CoFeNi alloy [34]. Hence the value of Ho and kh reported for CoFeNi alloy is taken for FCC phase (Ho,α and kα) in the present study. Since the value for Cr7C3 is not available in the literature, the Ho and kh value for Cr [Ho = 341 HV and kh = 140 HV μm1/2] from the literature is used for the carbide phase (Ho,ε and kε) [54]. Incorporating the topological parameters and Hall-Petch coefficients value in Eq. (5) and Eq. (6), Ho,αε and kαε are estimated as 110 HV and 243. 47 HV μm1/2, respectively. The estimated kαε value is higher than the kα and kε. Since the Hall-Petch coefficient indicates the efficiency of the boundary in preventing the dislocation motion, the higher kαε value indicates that the phase boundaries act as strong obstacles than the grain boundaries in the present alloy. Since the Hall-Petch coefficient only indicates the efficiency of strengthening and in order to estimate the individual contribution of strengthening from grain boundaries and phase boundaries, the HallPetch coefficient is multiplied by the topological parameter [53]. The individual contribution from α-α grain boundaries, ε-ε grain boundaries, and α-ε phase boundaries are given by kαfαc = 90.2 HV μm1/2, kεfεc = 4.0 HV μm1/2, and kαεFs = 68.7 HV μm1/2, respectively. Though the phase boundaries act as a strong obstacle, the major contribution to strengthening comes from α-α grain boundaries. The first, second, and third part of the right-hand side in Eq. (4) represent the individual contribution of FCC (Hα), carbide (Hε), and a mixture of FCC and carbide (Hαε), respectively. Accordingly, Hα, Hε, and Hαε are estimated to be 327 HV, 21 HV, and 230 HV, respectively. Consequently, the overall estimated hardness is 578 HV, which is in good agreement with the measured hardness (580 HV). In a polycrystalline material, the strength of an alloy is a combination of several strengthening mechanisms. In the present study, the

It is of our opinion that implications of this study are quite profound and suggestive for further ideas in the direction of designing thermally stable microstructures. Thus, based on this comprehensive study, the following are the possible variants of the microstructure that can be thought of giving similar or better thermal stability. a. By varying the carbon and oxygen content or both of them independently, one can vary the concentration of carbide phase (B) and oxide phase (C) in an HEA matrix phase, and hence their thermal stability can also be controlled depending on the application requirement. Thus, such alloys can also be anticipated to be having better high-temperature properties due to their thermal stability. b. A similar strategy will be to design phases similar to B or C that can be coherent phases (similar to gamma prime phase in superalloys) within HEA matrix phase, so that their room temperature strength and creep strength can be increased quite significantly. Recently, He et al. [44] utilized this kind of approach to develop precipitation hardened HEAs by adding small additions of Ti and Al to CoCrFeNi alloy that exhibited high tensile strength. c. Another approach that is similar to ODS steels is to have nanocrystalline clusters similar to YTiO3 or Y2TiO5 or Y2Ti2O7 produced within HEA matrix phase so that these materials can be anticipated to have higher creep resistance.

Although the suggested strategies b and c are quite rudimentary and speculative in approach at this stage, they provide ample opportunities in the alloy design. 5. Summary Thermal stability of CoCrFeNi alloy synthesized by MA and SPS was investigated in the present work. Upon exposure to higher temperature, the MA CoCrFeNi powders showed microstructural changes such as crystallite growth, strain recovery, carbide and oxide phase formation, and dissolution of carbide phase. Dissolution of carbide and evaporation of oxide phase was observed above 1000 °C. The APT of sintered compact revealed that FCC phase mainly composed of Co, Fe, and Ni in almost equiatomic proportion and the carbide phase was enriched in Cr with a lesser amount of Fe, Ni, and Co in the decreasing order. The sintered CoCrFeNi pellet exhibited exceptional thermal stability at 700 °C (0.57 Tm) for 600 h. Grain growth kinetics showed better


resistance to grain growth in the present ultrafine-grained alloy than the microcrystalline counterpart. The major contribution to strengthening comes from the grain boundary strengthening while the phase boundaries act as strong obstacles. Based on the present study, an alloy design has been proposed with thermally stable composite microstructure composed of two-phase aggregate with oxide dispersion. Acknowledgment Authors thank Professor B.S. Murty for granting unconditional access to use his laboratory equipment, and also for invaluable interactions during the initial stages of the present work. Authors would like to thank Naval Research Board (NRB), Directorate of Naval R&D, New Delhi, India for the financial support through the project # DNRD/05/4003/ NRB/190. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.matdes.2017.08.053. References [1] A.C. Reardon, Metallurgy for the non-metallurgist, 2nd ed. ASM International, Materials Park, Ohio, 2011. [2] S. Ranganathan, Alloyed pleasures: multimetallic cocktails, Curr. Sci. 85 (10) (2003) 1404–1406. [3] A.L. Greer, Confusion by design, Nature 366 (6453) (1993) 303–304. [4] J.-W. Yeh, High-entropy Multielement AlloysGoogle Patents 2002. [5] B. Cantor, I.T.H. Chang, P. Knight, A.J.B. Vincent, Microstructural development in equiatomic multicomponent alloys, Mater. Sci. Eng. A 375-377 (2004) 213–218. [6] J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T. Shun, C.H. Tsau, S.Y. Chang, Nanostructured high-entropy alloys with multiple principal elements: novel alloy design concepts and outcomes, Adv. Eng. Mater. 6 (5) (2004) 299–303. [7] J.-W. Yeh, S.-J. Lin, T.-S. Chin, J.-Y. Gan, S.-K. Chen, T.-T. Shun, C.-H. Tsau, S.-Y. Chou, Formation of simple crystal structures in Cu-Co-Ni-Cr-Al-Fe-Ti-V alloys with multiprincipal metallic elements, Metall. Mater. Trans. A 35 (8) (2004) 2533–2536. [8] D.B. Miracle, O.N. Senkov, A critical review of high entropy alloys and related concepts, Acta Mater. 122 (2017) 448–511. [9] M.C. Gao, J.-W. Yeh, P.K. Liaw, Y. Zhang, High-entropy Alloys: Fundamentals and Applications, Springer, Switzerland, 2016. [10] B. Murty, J.-W. Yeh, S. Ranganathan, High-entropy Alloys, Butterworth-Heinemann, 2014. [11] M.A. Hemphill, T. Yuan, G.Y. Wang, J.W. Yeh, C.W. Tsai, A. Chuang, P.K. Liaw, Fatigue behavior of Al0.5CoCrCuFeNi high entropy alloys, Acta Mater. 60 (16) (2012) 5723–5734. [12] Y. Zou, S. Maiti, W. Steurer, R. Spolenak, Size-dependent plasticity in an Nb25Mo25Ta25W25 refractory high-entropy alloy, Acta Mater. 65 (2014) 85–97. [13] O.N. Senkov, J.M. Scott, S.V. Senkova, F. Meisenkothen, D.B. Miracle, C.F. Woodward, Microstructure and elevated temperature properties of a refractory TaNbHfZrTi alloy, J. Mater. Sci. 47 (9) (2012) 4062–4074. [14] K.G. Pradeep, C.C. Tasan, M.J. Yao, Y. Deng, H. Springer, D. Raabe, Non-equiatomic high entropy alloys: approach towards rapid alloy screening and propertyoriented design, Mater. Sci. Eng. A 648 (2015) 183–192. [15] Y. Deng, C.C. Tasan, K.G. Pradeep, H. Springer, A. Kostka, D. Raabe, Design of a twinning-induced plasticity high entropy alloy, Acta Mater. 94 (2015) 124–133. [16] Y. Zhang, T.T. Zuo, Z. Tang, M.C. Gao, K.A. Dahmen, P.K. Liaw, Z.P. Lu, Microstructures and properties of high-entropy alloys, Prog. Mater. Sci. 61 (2014) 1–93. [17] D.Y. Li, C.X. Li, T. Feng, Y.D. Zhang, G. Sha, J.J. Lewandowski, P.K. Liaw, Y. Zhang, Highentropy Al0.3CoCrFeNi alloy fibers with high tensile strength and ductility at ambient and cryogenic temperatures, Acta Mater. 123 (2017) 285–294. [18] C.J. Tong, Y.L. Chen, S.K. Chen, J.W. Yeh, T.T. Shun, C.H. Tsau, S.J. Lin, S.Y. Chang, Microstructure characterization of AlxCoCrCuFeNi high-entropy alloy system with multiprincipal elements, Metall. Mater. Trans. A 36a (4) (2005) 881–893. [19] C.P. Lee, Y.Y. Chen, C.Y. Hsu, J.W. Yeh, H.C. Shih, The effect of boron on the corrosion resistance of the high entropy alloys Al[sub 0.5]CoCrCuFeNiB[sub x], J. Electrochem. Soc. 154 (8) (2007) C424. [20] S. Varalakshmi, M. Kamaraj, B.S. Murty, Synthesis and characterization of nanocrystalline AlFeTiCrZnCu high entropy solid solution by mechanical alloying, J. Alloys Compd. 460 (1–2) (2008) 253–257. [21] S. Praveen, B.S. Murty, R.S. Kottada, Alloying behavior in multi-component AlCoCrCuFe and NiCoCrCuFe high entropy alloys, Mater. Sci. Eng. A 534 (2012) 83–89. [22] T.-K. Chen, M.-S. Wong, T.-T. Shun, J.-W. Yeh, Nanostructured nitride films of multielement high-entropy alloys by reactive DC sputtering, Surf. Coat. Technol. 200 (5– 6) (2005) 1361–1365. [23] W.-Y. Tang, M.-H. Chuang, H.-Y. Chen, J.-W. Yeh, Microstructure and mechanical performance of new Al0.5CrFe1.5MnNi0.5 high-entropy alloys improved by plasma nitriding, Surf. Coat. Technol. 204 (20) (2010) 3118–3124. [24] C. Huang, Y.Z. Zhang, R. Vilar, Microstructure characterization of laser clad TiVCrAlSi high entropy alloy coating on Ti-6Al-4V substrate, Adv. Mater. Res. 154-155 (2010) 621–625.

433

[25] P.S. Gilman, J.S. Benjamin, Mechanical alloying, Annu. Rev. Mater. Sci. 13 (1) (1983) 279–300. [26] B.S. Murty, S. Ranganathan, Novel materials synthesis by mechanical alloying/milling, Int. Mater. Rev. 43 (3) (1998) 101–141. [27] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (1–2) (2001) 1–184. [28] H.W. Zhang, R. Gopalan, T. Mukai, K. Hono, Fabrication of bulk nanocrystalline Fe–C alloy by spark plasma sintering of mechanically milled powder, Scr. Mater. 53 (7) (2005) 863–868. [29] S.G. Ma, S.F. Zhang, J.W. Qiao, Z.H. Wang, M.C. Gao, Z.M. Jiao, H.J. Yang, Y. Zhang, Superior high tensile elongation of a single-crystal CoCrFeNiAl0.3 high-entropy alloy by Bridgman solidification, Intermetallics 54 (2014) 104–109. [30] W. Ji, W. Wang, H. Wang, J. Zhang, Y. Wang, F. Zhang, Z. Fu, Alloying behavior and novel properties of CoCrFeNiMn high-entropy alloy fabricated by mechanical alloying and spark plasma sintering, Intermetallics 56 (2015) 24–27. [31] V. Maier-Kiener, B. Schuh, E.P. George, H. Clemens, A. Hohenwarter, Nanoindentation testing as a powerful screening tool for assessing phase stability of nanocrystalline high-entropy alloys, Mater. Des. 115 (2017) 479–485. [32] B. Gwalani, V. Soni, M. Lee, S.A. Mantri, Y. Ren, R. Banerjee, Optimizing the coupled effects of Hall-Petch and precipitation strengthening in a Al0.3CoCrFeNi high entropy alloy, Mater. Des. 121 (2017) 254–260. [33] H.R. Sistla, J.W. Newkirk, F. Frank Liou, Effect of Al/Ni ratio, heat treatment on phase transformations and microstructure of AlxFeCoCrNi2 − x (x = 0.3, 1) high entropy alloys, Mater. Des. 81 (2015) 113–121. [34] Z. Wu, H. Bei, F. Otto, G.M. Pharr, E.P. George, Recovery, recrystallization, grain growth and phase stability of a family of FCC-structured multi-component equiatomic solid solution alloys, Intermetallics 46 (2014) 131–140. [35] B. Schuh, F. Mendez-Martin, B. Völker, E.P. George, H. Clemens, R. Pippan, A. Hohenwarter, Mechanical properties, microstructure and thermal stability of a nanocrystalline CoCrFeMnNi high-entropy alloy after severe plastic deformation, Acta Mater. 96 (2015) 258–268. [36] C. Zhang, F. Zhang, H.Y. Diao, M.C. Gao, Z. Tang, J.D. Poplawsky, P.K. Liaw, Understanding phase stability of Al-Co-Cr-Fe-Ni high entropy alloys, Mater. Des. 109 (2016) 425–433. [37] Y.D. Wu, Y.H. Cai, X.H. Chen, T. Wang, J.J. Si, L. Wang, Y.D. Wang, X.D. Hui, Phase composition and solid solution strengthening effect in TiZrNbMoV high-entropy alloys, Mater. Des. 83 (2015) 651–660. [38] W.H. Liu, Y. Wu, J.Y. He, T.G. Nieh, Z.P. Lu, Grain growth and the Hall–Petch relationship in a high-entropy FeCrNiCoMn alloy, Scr. Mater. 68 (7) (2013) 526–529. [39] K.Y. Tsai, M.H. Tsai, J.W. Yeh, Sluggish diffusion in Co–Cr–Fe–Mn–Ni high-entropy alloys, Acta Mater. 61 (13) (2013) 4887–4897. [40] S. Praveen, J. Basu, S. Kashyap, R.S. Kottada, Exceptional resistance to grain growth in nanocrystalline CoCrFeNi high entropy alloy at high homologous temperatures, J. Alloys Compd. 662 (2016) 361–367. [41] H. Zhang, K.G. Pradeep, S. Mandal, D. Ponge, H. Springer, D. Raabe, Dynamic straininduced transformation: an atomic scale investigation, Scr. Mater. 109 (2015) 23–27. [42] S. Praveen, B. Murty, R.S. Kottada, Effect of molybdenum and niobium on the phase formation and hardness of nanocrystalline CoCrFeNi high entropy alloys, J. Nanosci. Nanotechnol. 14 (10) (2014) 8106–8109. [43] S. Praveen, B.S. Murty, R.S. Kottada, Phase evolution and densification behavior of nanocrystalline multicomponent high entropy alloys during spark plasma sintering, JOM 65 (12) (2013) 1797–1804. [44] G.Y. Lai, High-temperature Corrosion and Materials Applications, ASM International, 2007. [45] A. Inoue, T. Masumoto, Carbide reactions (M3C → M7C3 → M23C6 → M6C) during tempering of rapidly solidified high carbon Cr-W and Cr-Mo steels, Metall. Trans. A. 11 (5) (1980) 739–747. [46] B.A. Marinkovic, R.R.d. Avillez, S.K. Barros, F.C.R. Assunção, Thermodynamic evaluation of carbide precipitates in 2.25Cr3/4 1.0Mo steel for determination of service degradation, Mater. Res. 5 (4) (2002) 491–495. [47] T.-T. Shun, Y.-C. Du, Age hardening of the Al0.3CoCrFeNiC0.1 high entropy alloy, J. Alloys Compd. 478 (1–2) (2009) 269–272. [48] S. Fang, W. Chen, Z. Fu, Microstructure and mechanical properties of twinned Al0.5CrFeNiCo0.3C0.2 high entropy alloy processed by mechanical alloying and spark plasma sintering, Mater. Des. 54 (1980–2015) (2014) 973–979. [49] C.S. Tedmon, The effect of oxide volatilization on the oxidation kinetics of Cr and FeCr alloys, J. Electrochem. Soc. 113 (8) (1966) 766. [50] R. Sriharitha, B.S. Murty, R.S. Kottada, Alloying, thermal stability and strengthening in spark plasma sintered AlxCoCrCuFeNi high entropy alloys, J. Alloys Compd. 583 (2014) 419–426. [51] J.Y. He, H. Wang, H.L. Huang, X.D. Xu, M.W. Chen, Y. Wu, X.J. Liu, T.G. Nieh, K. An, Z.P. Lu, A precipitation-hardened high-entropy alloy with outstanding tensile properties, Acta Mater. 102 (2016) 187–196. [52] S. Yoshida, T. Bhattacharjee, Y. Bai, N. Tsuji, Friction stress and Hall-Petch relationship in CoCrNi equi-atomic medium entropy alloy processed by severe plastic deformation and subsequent annealing, Scr. Mater. 134 (2017) 33–36. [53] Z. Fan, P. Tsakiropoulos, P.A. Smith, A.P. Miodownik, Extension of the Hall-Petch relation to two-ductile-phase alloys, Philos. Mag. A 67 (2) (1993) 515–531. [54] D. Wu, J. Zhang, J.C. Huang, H. Bei, T.G. Nieh, Grain-boundary strengthening in nanocrystalline chromium and the Hall–Petch coefficient of body-centered cubic metals, Scr. Mater. 68 (2) (2013) 118–121. [55] R.S. Ganji, P.S. Karthik, K.B.S. Rao, K.V. Rajulapati, Strengthening mechanisms in equiatomic ultrafine grained AlCoCrCuFeNi high-entropy alloy studied by microand nanoindentation methods, Acta Mater. 125 (2017) 58–68. [56] S. Praveen, A. Anupam, T. Sirasani, B.S. Murty, R.S. Kottada, Characterization of oxide dispersed AlCoCrFe high entropy alloy synthesized by mechanical alloying and spark plasma sintering, Trans. Indian Inst. Metals 66 (4) (2013) 369–373.