Synthesis and characterization of Fe-15wt.% ZrO2

Powder Technology 287 (2016) 190–200

Contents lists available at ScienceDirect

Powder Technology journal homepage: www.elsevier.com/locate/powtec

Synthesis and characterization of Fe-15 wt.% ZrO2 nanocomposite powders by mechanical milling K.G. Raghavendra a, Arup Dasgupta a,⁎, Pragna Bhaskar a, K. Jayasankar b, C.N. Athreya c, Padmalochan Panda d, S. Saroja a, V. Subramanya Sarma c, R. Ramaseshan d a

Microscopy and Thermo-Physical Property Division, Physical Metallurgy Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603102, India Advanced Materials Technology Department, CSIR – Institute of Minerals & Materials Technology, Bhubaneswar 751013, India Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India d Surface and Nano Science Division, Materials Science Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603102, India b c

a r t i c l e

i n f o

Article history: Received 21 July 2015 Received in revised form 1 October 2015 Accepted 3 October 2015 Available online 09 October 2015 Keywords: Oxide dispersion strengthened alloys Nanocomposites Microstructure Nanocrystallinity Mechanical milling

a b s t r a c t Fe-15 wt.% ZrO2 nanocomposite powder was synthesized via mechanical milling with an aim to study the morphology of the powder particles and refinement of the oxide and Fe crystallites during milling. Detailed microstructural and microchemical investigations were carried out in order to optimize the milling condition and to highlight the advantages for the choice of ZrO2. A homogeneous mixture was confirmed by X-ray mapping and ZrO2 dispersoids were observed to retain crystallinity even after 100 h of milling. It was also observed that both Fe and ZrO2 crystallites refine to very fine nanocrystalline (nc) sizes after such milling. The result has 2fold significance: (a) Yttria, which is a standard dispersoid in oxide dispersion strengthened (ODS) steels, usually amorphised under similar conditions, which is detrimental to its structural stability and (b) nanocrystallites of Fe have useful magnetic properties. Modified Williamson–Hall technique (mod. W–H) was employed to measure the size and dislocation density of the matrix ferrite phase. Nanoindentation technique was used to evaluate the nanohardness of the milled powder as a function of milling duration. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Mechanical milling/alloying (MA) is a widely used technique for the synthesis of nanostructured materials. The process involves repeated fracture and cold welding between the powders and refinement of crystallite sizes down to nanometers is possible with this technique [1–3]. This powder metallurgical route serves as an excellent technique to incorporate fine ceramic dispersions into the metallic matrix including ODS alloys [4–8] and nanocomposite materials [9–13], which is otherwise difficult via conventional melting owing to differences in density. The MA process is strongly influenced by process parameters such as ball to powder ratio, rotation speed of the mill and process control agent [14]. MA is known to produce microstructures which are far from equilibrium. This attribute of MA is the cause for the attractive properties observed in the processed materials [15]. ODS alloys are promising candidate structural materials for in-core applications in next generation nuclear reactors, owing to their superior mechanical properties at elevated temperatures along with high resistance to irradiation defects [4,6,16–23]. The stable nanometer sized oxide particles (dispersoids) [24], which are added to the base matrix to get uniform distribution of dispersoids are known to strengthen the ⁎ Corresponding author. E-mail address: [email protected] (A. Dasgupta).

http://dx.doi.org/10.1016/j.powtec.2015.10.003 0032-5910/© 2015 Elsevier B.V. All rights reserved.

material along with other mechanism such as solid solution hardening and dislocation hardening [25]. Yttria-titania complexes are conventionally used as oxide dispersoid in ODS alloys [5,26]. However, concerns remain with regard to coarsening of such dispersoids, although Ti is expected to help in maintaining their fine sizes, predominantly within 10 nm [4,27]. Moreover, the behavior of other oxide dispersoids in ODS alloys is not extensively studied. Attempts have been initiated to explore the performance of oxide dispersoids other than Yttria (Y2O3) in the recent past. Preliminary results of Hoffman et al. [28] and Pasebani et al. [29,30] suggest that use of alternative rare earth oxide such as Magnesia (MgO), Lanthana (La2O3), Ceria (Ce2O3) and Zirconia (ZrO2) as dispersoids in ODS alloys can yield promising results. In the present study ZrO2 has been chosen, considering its high temperature stability [31], low fast neutron absorption cross section [16,32] and fewer number of atoms per unit cell (mP12) compared to Y2O3 (cI80). Yttrium and Zirconium (Zr) are expected to exhibit similar physical and chemical properties due to their positions in the periodic table [33]. The role of Zr in grain size stability at elevated temperatures has attracted attention recently based on both experiments [34–36] and modeling [37,38]. Also, literature shows that Zr addition to Y2O3 based ODS alloys resulted in enhancement of thermal stability [39] and mechanical properties by formation of fine Y–Zr–O precipitates [40]. ODS steels contain a very small volume fraction of the dispersoids which

K.G. Raghavendra et al. / Powder Technology 287 (2016) 190–200

leads to difficulty in reliable characterization of the milled powders, especially through X-Ray Diffraction (XRD) [41,42]. Also high weight fraction of dispersoid has yielded useful insights about the milling mechanism and dispersoid evolution [43,44]. Hence a concentrated Fe-15 wt.% ZrO2 composite was chosen for this study. Such a concentrated mechanically alloyed nanocomposite granular solids has attracted significant attention due to their interesting magnetic properties [45–47]. High magnetization, high Curie temperature, low hysteresis loss and low dielectric loss at high frequencies are some of the important properties exhibited by ceramic – metal composites [48]. The nc nature of Fe has profound effects on the magnetic behavior of the material owing to surface effects. The magnetic moment, for example, has large dependency on the dimensionality of the material, which in turn changes the value of structure sensitive properties, like coercivity [49]. Enhanced coercivity for ball milled and annealed Fe– ZrO2 powder consisting of varying volume fraction (~30–68%) of ZrO2 is reported [47]. Moreover, nature of the magnetic domains in the nc materials differ from their bulk counterparts, leading to an interesting magnetic phenomenon such as Superparamagnetism and Giant magneto resistance [49]. The key factor in these applications is to synthesize nc-Fe and retaining its nanocrystallinity. Superior magnetic properties in the nc materials can be retained or enhanced by the separation of magnetic nano particles by non magnetic material, wherein the two systems are immiscible [46]. ZrO2 in Fe matrix could play a similar role. Further, investigations on Fe–ZrO2 system have also been driven by their application as metal matrix composites owing to their mechanical properties. ZrO2 reinforced Fe, synthesized by mechanical alloying followed by compaction and sintering was shown to exhibit enhanced mechanical [10] properties. Keeping in mind the above mentioned usefulness of a Fe–ZrO2 composite system, the objectives of this study include optimization of milling duration based on the achievement of desired microstructure and microchemical uniformity. The evolution of Fe crystallites and ZrO2 dispersoids, with milling duration, was also investigated systematically. 2. Experimental Fe (sieve size of +325 mesh) and ZrO2 powders were procured from M/s, HIMEDIA, India with a commercial purity grade of 99.5% and 99%, respectively. The powders were blended together and ball milled in the proportion of Fe–15 wt.% ZrO2. High energy ball milling was carried out in a planetary ball mill consisting of austenitic stainless steel (ss) vial of volume 500 ml. Austenitic ss balls of diameter 10 mm were used for the milling. The ball to powder ratio and the rotation speed were maintained at 10:1 and 300 rpm, respectively. Both the ball milling process and removal of powder from the vial were carried out under high purity Argon atmosphere in order to avoid unwanted oxygen pickup. Process control agents were not used in order to avoid C pickup during milling. Milling was carried out for different durations, viz., 0 (blended powder), 10, 30, 60 and 100 h. The XRD analysis using Co-Kα radiation was carried out using a Bruker AXS D8 discover instrument. Co X-rays were preferred over Cu to avoid the loss of information due to excess fluorescence. The X-ray Beam voltage and current were maintained at 35 kV and 25 mA respectively. Instrumental broadening was calculated by collecting the XRD spectrum for a strain free Corundum sample with a large grain size. Mod. W–H technique was employed for the determination of matrix crystallite size, strain and dislocation density. FEI make Helios NanoLab-600i dual beam field emission Scanning Electron Microscope (SEM) was used to study the size and morphology of the milled powders. Energy Dispersive Spectroscopy (EDS) microanalysis was carried out using an Apollo X Silicon Drift Detector attached to the SEM. Transmission Electron Microscope (TEM) specimen from powder particle were prepared by dispersing the powder in epoxy resin and hardener mixture and then ground followed by dimple thinning and ion milling. When the powder particle size was large

191

(N10 μm), Focused Ion Beam (FIB) technique using Ga+ ions, was employed for TEM sample preparation. Pt was used for welding the specimen to Omniprobe needle and subsequently to TEM grids. TEM characterization was carried out using a Philips CM-200 Analytical TEM operated at 200 kV attached with TVIPS make 2 K × 2 K CCD camera. ImageJ software was used for the analysis of TEM results. Nanohardness measurements on the powder particles were carried out in a Nanoindenter (M/s CSM, Switzerland) equipped with a three sided pyramidal diamond (Berkovich) tip with an end radius of ~30 nm. Loading and unloading rates were maintained at 20 mN/min and indentation was carried out for a fixed depth of 200 nm. At least 15 indentations were carried out on each sample and the average value of the hardness is measured after ignoring the erroneous readings. The measurements were carried out on the powder milled for different durations and hardness value as a function of milling time is analyzed. Fig. 1 represents flowchart of the experimental procedures.

3. Results and discussion 3.1. Structure and chemistry of milled powders Fig. 2(a) and (b) shows the secondary electron (SE) micrographs of the Fe and ZrO2 powder particles prior to mixing. Fe powder particles are irregularly shaped with rounded edges and fine cracks on their surface. The particle sizes measured over ~200 particles, as analyzed using ImageJ software, are found to be in the range 50–100 μm. On the other hand, the SE image of ZrO2 particles reveals agglomeration of the fine particles, with the agglomerates measuring about 10–20 μm. The agglomeration of the fine powder particles is a common phenomenon resulting from Van der Waals forces, static electrification and mechanical interlocking [50]. The sizes of grains making up the powder particles were determined by TEM dark field (DF) analysis. (110) reflection of Fe, ð111Þ reflection of ZrO2 were used for the DF imaging. Fig. 2(c and d) show the TEM-DF images of Fe and ZrO2 powders, respectively. Fine bright spots in these images represent the individual crystallites. Grain size distribution analysis has been carried out from the DF images and given as insets in the figure. The Fe grains are nc and distributed over a size range of 5–45 nm. On the other hand, the ZrO2 grains are also nc and spread over a range of 5–90 nm. It can be observed from the peak distribution that majority of the grains have a size of about 5–10 nm. Fig. 3 shows the gradual changes in the morphology of the powder during mechanical milling for various durations. Fig. 3(a) shows the SE micrograph of the 10 h milled powder. It exhibits a broad range of sizes with irregular morphology of the powder particles. Fig. 3(b) shows the SE micrograph of the powder obtained after 30 h milling, revealing a decrease in the average particle size. More rounded particles are observed but a wide range of particle size persists, indicating incomplete homogenization. Fig. 3(c) shows the SE micrograph after 60 h milling. Reduction in the particle size from that of the initial powder is apparent from this figure. It is also evident from this figure that the size range is reduced and more number of particles has rounded edges. Fig. 3(d) shows SE image of the 100 h milled powder. A fairly uniform particle size distribution and rounded edges for almost all particles are observed. Fig. 3(e) shows a plot of maximum and mean particle sizes and the mean aspect ratio of the powder particles as a function of milling time. Particle size of the Fe powder was used for the largest particle size and mean of average of Fe and ZrO2 powder particle sizes were used for 0 h milling time here. It is observed that the maximum and mean particle sizes decrease from ~150 μm and 50 μm at the beginning of milling i.e., at 0 h to ~ 10 μm and 3 μm, respectively after 100 h milling. Relatively small difference between the maximum and mean particle sizes for the powder milled for 100 h is a signature of the desired condition of more uniform particle size distribution. At this stage, a balance is achieved between rate of cold welding and rate of

192


Fig. 1. Flowchart of the experimental procedures.

fracturing due to severe plastic deformation by milling [1–3]. It may also be mentioned here that a uniform distribution in particle size is beneficial to the consolidation process during fabrication of a component [51, 52]. The mean aspect ratio of the powder particles, are found to decrease from about 2.5 to 1.5 till 30 h of milling duration and then saturating at

that value. The lower aspect ratio indicates a better degree of sphericity for the powder particles, which is helpful during consolidation. Irregularly shaped particles leads to porosity and poor density during consolidation [53]. Therefore, from size distribution and aspect ratio considerations, 30 h of milling duration appears to be sufficient.

Fig. 2. Analysis of the initial powder microstructure showing SE images of (a) Fe (b) ZrO2 and TEM DF images of (c) Fe (d) ZrO2.


193

Fig. 3. (a–d) SE micrographs of the milled powders after 10 h, 30 h, 60 h and 100 h durations, respectively; (e) plot of maximum and mean particle sizes as well as mean aspect ratio as a function of milling durations.

194


Fig. 4. (a, c, e) SE micrographs of a representative powder particle milled for 10 h, 30 h and 100 h, respectively. (b, d, f) Fe-Kα (Red) and Zr Lα (Cyan) X-ray maps corresponding to electron images shown in Figs (a, c, e) for 10 h, 30 h and 100 h of milling, respectively. (g) EDX spectrum of the 100 h milled powder along with the elemental composition. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


Microchemical investigations were carried out on the milled powder by energy dispersive X-ray (EDX) analysis. Fig. 4(a) shows the SE image of a typical powder particle milled for 10 h. The corresponding X-ray map for the characteristic emissions of Fe-Kα (red) and Zr-Lα (cyan) are shown in Fig. 4(b). A careful inspection of the regions of interest (ROI 1 and 2) marked in the figures, reveals Zr rich (ROI – 1) and Zr lean (ROI −2) regions, indicating incomplete mixing of the ZrO2 dispersoids in the Fe matrix after 10 h of milling duration. Similarly, Fig. 4(c) and (d) show the typical SE image of a powder particle and its corresponding Fe-Kα (red) and Zr-Lα (cyan) X ray maps for the 30 h milled powder, respectively. Again, a careful inspection of the ROIs (3 and 4) marked in the figures, still reveals Zr rich (ROI – 3) and Zr lean (ROI −4) regions. Considering the microchemical inhomogeneity, the milling duration of 30 h is insufficient as against what was conjectured earlier. On the other hand, Fig. 4(e) and (f) show the typical SE image of a powder particle and its corresponding Fe-Kα (red) and ZrLα (cyan) X ray maps for the 100 h milled powder, respectively. The figures indicate a desired uniform mixing of the two phases, Fe and ZrO2. EDX spectrum of the area mapped [Fig. 4(e)] is shown in Fig. 4(g) and elemental quantification yielded weight percents of 84.6, 9.2 and 6.2 for Fe, Zr and O, respectively. Ideally, with 15 wt.% of ZrO2, these values should have read 85.0, 11.1 and 3.9, respectively. Thus, it is observed that though the Fe concentration is comparable to the ideal value, those for Zr is less and O is higher. This is attributed to O pickup during loading into microscope. These results demonstrate that 100 h of milling duration is indeed required for optimum synthesis of the powder. Fig. 5 shows the XRD patterns of the milled Fe-15 wt.% ZrO2 powder at various milling durations, viz., 0 (only physically mixed in suitable weight ratio), 10, 30, 60 and 100 h, respectively. Various peaks observed in the diffractograms were indexed for bcc-Fe and monoclinic-ZrO 2 with reference to the ICDD Card Nos.: 00-0060696 and 00-036-0420, respectively. The peaks identified are marked in the Fig. 5. A careful inspection of the three strongest peaks of Fe, viz., (110) or a, (200) or b and (211) or c, shows a gradual increase in the FWHM, which is in agreement with the XRD analysis of milled powders reported earlier [1–3]. XRD peak broadening is attributed to decrease in crystallite size as well as increase in strain [54]. Debye Scherrer analysis [55] of the strongest peaks corresponding to Fe and ZrO2, namely (110) and (110), respectively, was carried out to evaluate variation of crystallite size with milling duration. Accordingly, Fig. 6 shows that crystallite size of Fe and ZrO2 decrease from 26 and 36 nm to 10 and 12 nm, respectively, with increase in milling duration from 0 to 100 h. It is observed that the sizes of ZrO2 refine faster in the initial stages of milling than Fe owing to brittle nature of the former. However, beyond 10 h of milling duration,

Fig. 5. X-Ray diffraction patterns for the Fe-15 wt.% ZrO2 powders for different durations (0–100 h).

195

Fig. 6. Devolution of Fe and ZrO2 crystallite sizes as a function of milling duration.

the sizes of the ZrO2 and Fe crystallites are comparable. The Fe crystallites work-harden during milling and then fragments with further progress in milling. It may be mentioned here that, ≤ 10 nm dispersoid size is desirable during fabrication of steel [56,57] for structural applications. This is because the steel will contain b1 vol.% of the dispersoid; and a uniform distribution of ~ 10 nm dispersoids can ensure just enough separation between them so as to block mobile dislocations and thus impart high strength to these materials. Therefore 100 h of milling is suitable from this point of view. Fig. 7(a–d) represents selected area electron diffraction (SAED) patterns of the milled powders after milling for 10, 30, 60 and 100 h, respectively. The patterns resemble several circular rings overlapped with a few diffraction spots. The ring patterns indicate polycrystalline nature of the milled powders. However, the presence of few strongly diffracting crystals is evident from the spots observed in the pattern. SAED patterns were indexed with reference to the ICDD data card Nos. 00-006-0696 and 00-036-0420 corresponding to bcc-Fe and monoclinic-ZrO2 (room temperature phase of ZrO2), respectively. Accordingly, rings were marked as, (i – v) in Fig. 7(a) correspond to (110), (200), (211), (220) and (310) diffraction planes of the bcc-Fe, while rings marked as (1–4) in this figure correspond to (111), (111), (002) or (020) (too close to be distinguished in SAED) and (220) or (122) of monoclinic ZrO2. It may be mentioned here, that almost all monoclinic ZrO2 peaks beyond the radial limits of ‘i’, overlap with those of Fe. It is significant to observe that crystalline ZrO2 rings and spots are observed after 100 h of milling duration, indicating that the ZrO2 dispersoids retain crystallinity even after prolonged duration of milling in contrast to the behavior of Y2O3 [7,58]. This is attributed to the following reasons: (1) large unit cell structure (cI80) of Y2O3 which poses difficulty to retain its crystallinity below a critical size, and (2) smaller structure (mP12) of monoclinic ZrO2. This is in agreement with the reports that a structure with lesser number of atoms in the unit cell is stable under ion irradiation compared to a complex structure [59]. As reported earlier [58], an amorphous dispersoid coarsens rapidly to very large grains during a subsequent high temperature process and is undesirable for dispersion strengthening. Fig. 8 shows bright field (BF) and dark field (DF) TEM micrographs from identical locations for evaluation of the Fe and ZrO2 grains during mechanical milling for different durations. Fig. 8(a) shows the bright filed (BF) micrograph of a typical powder particle from the 10 h milled powder. The DF micrographs corresponding to (110) Fe and ð111Þ ZrO2 crystallites are shown in Fig. 8(b) and (c), respectively. Given the

196


Fig. 7. (a–d) SAED patterns obtained at different milling durations: 10 h, 30 h, 60 h and 100 h, respectively.

polycrystalline nature of the Fe and ZrO2 crystallites (ring patters in Fig. 7), these DF micrographs are representative. Both Fe and ZrO2 grains are observed to be nc bearing similar dimensions. Similarly, Fig. 8(d–f) show the BF, DF images corresponding to Fe and ZrO2 crystallites, respectively of the 100 h milled powder. Analysis of Fig. 8(a–f) show that both Fe and ZrO2 crystallites are more refined and uniformly dispersed, in case of 100 h milling. The statistical distribution of the Fe and ZrO2 crystallites are shown in Fig. 8(g) by means of normalized cumulative frequency plots. As is observed, only 81% of Fe grains and 72% ZrO2 grains are below 10 nm in case of 10 h milling, while as much as 89% of Fe as well as ZrO2 grains are below 10 nm after 100 h of milling. These results are in agreement with the sizes determined from XRD results and shown earlier in Fig. 6. This indicates refinement as well as more uniform distribution of grains after 100 h of milling. This result is most important for the ZrO2 dispersoids, where it exhibits a uniform distribution as well as refinement without amorphisation, essential for superior dispersion strengthening and higher creep strength in ODS ferritic alloys [60]. The nanocrystallization of α-Fe below 10 nm is also critical to its superparamagnetic effect [49], wherein the nano-magnetic domains can remain in single domain states while thermal fluctuations can randomize them in the absence of a magnetic field. The effect has immense application in biomedical field [61].

3.2. Measurement of strain and dislocation density of the ferrite matrix Further investigations were carried out by X-ray analysis in order to evaluate the strain induced during the ball milling process. A detailed analysis of the peak broadening was carried out to study the evolution of strain as a function of milling time. Due to the limitation on the number of peaks with adequate intensity for the ZrO2, this analysis was carried out for the matrix phase alone. During analysis of the peak broadening effect, it was assumed that predominant contribution to strain is from dislocations [62,63] resulting from severe plastic deformation of the grains due to milling [63]. Mod. W–H method [64] was employed for the determination of dislocation density. The mod. W–H equation for estimation of dislocation density [64–66] is given below: 2

ΔK ¼

0:9 πM2 b þ 2 d

!2

ρ1=2 KC 1=2 þ O K 2 C :

ð1Þ

2cosθ Δθ where, K ¼ 2sinθ . Δθ, θ and λ are the notations for λ and ΔK ¼ λ FWHM, diffraction angle and characteristic wavelength of the X-rays, respectively. In the present study λ is 0.1789 nm corresponding to the


197

Fig. 8. (a) TEM BF micrograph of a powder particle milled for 10 h; (b–c) Corresponding DF micrographs from Fe (110) and ZrO2 (111) crystallites, respectively; (d) TEM BF micrograph of a powder particle milled for 100 h; (e–f) corresponding DF micrographs from Fe (110) and ZrO2 (111) crystallites, respectively; (g) Graph showing the statistical distribution of Fe and ZrO2 crystallites during 10 and 100 h of milling durations.

198


Co-Kα radiation. d, ρ and b are the average grain size, dislocation density and magnitude of the Burgers vector, respectively. M is a constant depending on the effective outer cut-off radius of dislocations. C is the contrast factor. O stands for non-interpreted higher-order terms in KC1/2 which may be neglected [62]. The contrast factors of dislocations characterize the “visibility” of dislocations in the diffraction experiments, whose value depends on a number of factors, such as, the elastic constants of the material, the relative orientation of the diffraction vector (g), the Burgers vector (b), the line vector (l) and the normal vector of the slip plane (n) of the dislocations [67]. The first term on the RHS of Eq. (1) represents the peak broadening due to size while the second term, the strain contribution to the broadening. The mod. W–H technique is best suited for treating the ball milled powder since it takes into consideration the contrast effect C of dislocations on the peak broadening [68–70]. The average value of the contrast factor (C) for Fe was considered to be 0.061, 0.285, 0.118 and 0.061 for the reflections {110}, {200},{211} and {220}, respectively [68]. Using these values for ‘C’, the parameter KC1/2 was calculated. A plot of ΔK Vs KC1/2 for various milling durations (0, 10, 30, 60 and 100 h) is shown in Fig. 9. The figure also shows the linear fits for each set of data points, so that its slope and the intercept may be used to determine the strain (ε) and average grain size (d), respectively. The dislocation density in the material is related to the strain in case of BCC materials by the following relationship [71]. pffiffiffi ε2 1=2 ρD ¼ 2 3 Db

with the literature [1–3] and is typical of the mechanical alloying process. The crystallite size is found to be around 7 nm after 100 h of milling which is in good agreement with the Debye Scherer analysis value and also with the TEM results [Fig. 8(g)]. 3.3. Evaluation of nanomechanical properties

ð2Þ

where, Burgers vector ‘b’ of dislocations equals (a√3)/2 [69]. The plot of dislocation density (ρ) and grain size (d) as a function of milling duration are shown in Fig. 10. It is observed from this figure that while dislocation density increases almost monotonically, grain size decreases initially rapidly and gradually at higher milling durations. Dislocation density is calculated as 5 × 1015/m2 at the onset of milling and increased to 2.3 × 1016/m2 after 100 h of milling. The near monotonous increase of dislocation density till 100 h of milling duration indicates that the material has not undergone dynamic recrystallization (which would have resulted in a drop in ρ) as in other SPD processes [72]. A correlation between crystallite size and dislocation density may be drawn, since an increasing dislocation can result in the crystallite size refinement via increasing strain in mechanically milled systems, as has been reported in a Cu–Cr system [73]. Further, the variation of d is in agreement

Fig. 9. Mod. W–H plots for the powder milled for different durations (0–100 h).

Fig. 10. Plot of Fe crystallite size and dislocation density as function of milling time.

As ZrO2 and Fe powder particles get blend into a uniform composite matrix with reduction of their individual crystallite sizes, measurement of their nano-mechanical properties become very interesting. These properties could also influence the milling and compaction nature of these powders [74]. The nanohardness measurements were carried out on the milled powders to evaluate the nano-mechanical property as a function of milling time, the results of which are given below [Fig. 11]. The error bars represents the standard deviation from the measurement. The increasing tendency for the hardness with increase in the milling duration can be seen from the graph. The hardness of Fe particles has increased from 7.2 to 11.2 GPa for un-milled to 100 h of milling,

Fig. 11. Nano hardness of the powder particles as a function of milling duration and inverse square root of crystallite size, d−1/2.


respectively. The variation of hardness as function of d−1/2 (Hall–Petch relationship) [75,76] is also plotted in Fig. 11. Though not exactly linear, an increase in hardness with decrease in grain size (assuming; grain size = crystallite size), establishing a near Hall–Petch behavior. The deviation from linearity is possibly due to a combined effect of crystallite size decrease and dislocation density increase with increase in milling duration, as have been shown in Fig. 10. The observation is similar to the published results on Fe based ODS powder [77,78]. Higher hardness for the composite in the nc regime is an added advantage for the material from its mechanical strength point of view. 4. Conclusion The Fe-15 wt.% ZrO2 nanocomposite has been synthesized by mechanical milling for various milling durations and characterized with SEM, TEM, XRD and Nano indentation. The results are summarized as follows: • the powder particles showed reduction in size, more uniformity in sizes and increased sphericity with increase in milling time. From these point of view 30 h milling duration seemed enough for achieving desired conditions for consolidation. However, microchemical investigations revealed that 100 h of milling duration is required for uniform mixing of the Fe and ZrO2 phases; • XRD and TEM investigation also revealed that the size and size distribution of both Fe and ZrO2 dispersoids continued to improve up to 100 h of milling. Crystallite sizes of both Fe and ZrO2 were reduced to size ~5–10 nm after 100 h together with narrow crystallite size distribution, indicating that this period milling duration is required for superior consolidation; • even after 100 h of milling, the ZrO2 phase was found not to amorphise unlike the other popular oxide dispersoid, yttria. This was understood in terms of smaller number of atoms per unit cell for the former and considered a desired property. Amorphisation can lead to an undesired nucleation and rapid growth of the dispersoids during high temperature consolidation process. This establishes the rationale behind choice of ZrO2 as an alternate dispersoid material for ODS alloys; • nanocrystallization of Fe below 10 nm is suitable for superparamagnetic applications; • matrix strain analysis by mod. W–H method indicated an increase in the dislocation density of Fe with increase in the milling duration from 3 × 1015/m2prior to milling to about 2.3 × 1016/m2 after 100 h of milling, indicating that the Fe crystallites have not undergone dynamic recrystallization during milling; and • nanohardness of the powder was observed to be increasing with increase in milling duration. Hardness value reached to 11.2 GPa after 100 h of milling. Increase in hardness was attributed to the collective consequence of (i) Hall – Petch strengthening (ii) Dispersion strengthening and (iii) Strain hardening.

Acknowledgment The authors hereby gratefully acknowledge Dr. S. Venugopal, Director, Metallurgy and Materials Group and Dr. M. Vijayalakshmi, Associate Director, Physical Metallurgy Group for their enthusiastic support and useful discussion. The authors would also like to acknowledge the experimental support provided by UGC-DAE-CSR node at Kalpakkam. One of the authors Raghavendra K. G. acknowledges DAE, India for the research fellowship and the Homi Bhabha National Institute, Mumbai for facilitating the research. References [1] C. Suryanarayana, Mechanical Alloying and Milling, Taylor & Francis, 2004. [2] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (2001) 1–184.

199

[3] B. Murty, S. Ranganathan, Novel materials synthesis by mechanical alloying/milling, Int. Mater. Rev. 43 (1998) 101–141. [4] S. Ukai, M. Fujiwara, Perspective of ODS alloys application in nuclear environments, J. Nucl. Mater. 307–311 (2002) 749–757 (Part 1). [5] S. Ukai, M. Harada, H. Okada, M. Inoue, S. Nomura, S. Shikakura, K. Asabe, T. Nishida, M. Fujiwara, Alloying design of oxide dispersion strengthened ferritic steel for long life FBRs core materials, J. Nucl. Mater. 204 (1993) 65–73. [6] S. Ukai, T. Nishida, H. Okada, T. Okuda, M. Fujiwara, K. Asabe, Development of oxide dispersion strengthened ferritic steels for FBR core application, (I), J. Nucl. Sci. Technol. 34 (1997) 256–263. [7] L. Zhang, S. Ukai, T. Hoshino, S. Hayashi, X. Qu, Y2O3 evolution and dispersion refinement in co-base ODS alloys, Acta Mater. 57 (2009) 3671–3682. [8] H. Zhang, Y. Huang, H. Ning, C.A. Williams, A.J. London, K. Dawson, Z. Hong, M.J. Gorley, C.R.M. Grovenor, G.J. Tatlock, S.G. Roberts, M.J. Reece, H. Yan, P.S. Grant, Processing and microstructure characterisation of oxide dispersion strengthened Fe– 14Cr–0.4Ti–0.25Y2O3 ferritic steels fabricated by spark plasma sintering, J. Nucl. Mater. 464 (2015) 61–68. [9] C. Suryanarayana, T. Klassen, E. Ivanov, Synthesis of nanocomposites and amorphous alloys by mechanical alloying, J. Mater. Sci. 46 (2011) 6301–6315. [10] P. Jha, P. Gupta, D. Kumar, O. Parkash, Synthesis and characterization of Fe–ZrO2 metal matrix composites, J. Compos. Mater. 48 (2014) 2107–2115. [11] S.E. Hernández-Martinez, J.J. Cruz-Rivera, C.G. Garay-Reyes, C.G. Elias-Alfaro, R. Martínez-Sánchez, J.L. Hernández-Rivera, Application of ball milling in the synthesis of AA 7075 – ZrO2 metal matrix nanocomposite, Powder Technology 284 (2015) 40–46. [12] A. Canakci, T. Varol, H. Cuvalci, F. Erdemir, S. Ozkaya, E.D. Yalcın, Synthesis of novel CuSn b sub N 10b/sub N −graphite nanocomposite powders by mechanical alloying, micro & nano letters, Institution of Engineering and Technology, 2014 109–112. [13] A. Canakci, T. Varol, C. Nazik, Effects of amount of methanol on characteristics of mechanically alloyed Al–Al2O3 composite powders, Mater. Technol. 27 (2012) 320–327. [14] A. Canakci, F. Erdemir, T. Varol, A. Patir, Determining the effect of process parameters on particle size in mechanical milling using the Taguchi method: measurement and analysis, Measurement 46 (2013) 3532–3540. [15] T. Varol, A. Canakci, Synthesis and characterization of nanocrystalline Al 2024–B4C composite powders by mechanical alloying, Philos. Mag. Lett. 93 (2013) 339–345. [16] C.R.F. Azevedo, Selection of fuel cladding material for nuclear fission reactors, Eng. Fail. Anal. 18 (2011) 1943–1962. [17] S. Saroja, A. Dasgupta, R. Divakar, S. Raju, E. Mohandas, M. Vijayalakshmi, K. Bhanu Sankara Rao, B. Raj, Development and characterization of advanced 9Cr ferritic/martensitic steels for fission and fusion reactors, J. Nucl. Mater. 409 (2011) 131–139. [18] G.R. Odette, M.J. Alinger, B.D. Wirth, Recent developments in irradiation-resistant steels, Annu. Rev. Mater. Res. 38 (2008) 471–503. [19] K.L. Murty, I. Charit, Structural materials for Gen-IV nuclear reactors: challenges and opportunities, J. Nucl. Mater. 383 (2008) 189–195. [20] A.K. SURI, Material development for India's nuclear power programme, Sadhana 38 (2013) 859–895. [21] Q.X. Sun, Y. Zhou, Q.F. Fang, R. Gao, T. Zhang, X.P. Wang, Development of 9Cr-ODS ferritic–martensitic steel prepared by chemical reduction and mechanical milling, J. Alloys Compd. 598 (2014) 243–247. [22] J.-H. Ahn, H.-J. Kim, I.-H. Oh, Y.-J. Kim, Preparation of nano-sized ODS alloys by ballmilling using metallic salts, J. Alloys Compd. 483 (2009) 247–251. [23] W. Hoffelner, Damage assessment in structural metallic materials for advanced nuclear plants, J. Mater. Sci. 45 (2010) 2247–2257. [24] L. Zhang, X. Qu, X. He, D. Rafi-ud, M. Qin, H. Zhu, Hot deformation behavior of Cobase ODS alloys, J. Alloys Compd. 512 (2012) 39–46. [25] T. Shanmugasundaram, M. Heilmaier, B.S. Murty, V.S. Sarma, On the Hall–Petch relationship in a nanostructured Al–Cu alloy, Mater. Sci. Eng. A 527 (2010) 7821–7825. [26] R.L. Klueh, J.P. Shingledecker, R.W. Swindeman, D.T. Hoelzer, Oxide dispersionstrengthened steels: a comparison of some commercial and experimental alloys, J. Nucl. Mater. 341 (2005) 103–114. [27] L. Dai, Y. Liu, Z. Ma, Z. Dong, L. Yu, Microstructural evolution of oxide-dispersionstrengthened Fe–Cr model steels during mechanical milling and subsequent hot pressing, J. Mater. Sci. 48 (2013) 1826–1836. [28] J. Hoffmann, M. Rieth, R. Lindau, M. Klimenkov, A. Möslang, H.R.Z. Sandim, Investigation on different oxides as candidates for nano-sized ODS particles in reducedactivation ferritic (RAF) steels, J. Nucl. Mater. 442 (2013) 444–448. [29] S. Pasebani, I. Charit, D.P. Butt, J.I. Cole, A preliminary study on the development of La2O3-bearing nanostructured ferritic steels via high energy ball milling, J. Nucl. Mater. 434 (2013) 282–286. [30] S. Pasebani, I. Charit, Y.Q. Wu, D.P. Butt, J.I. Cole, Mechanical alloying of lanthanabearing nanostructured ferritic steels, Acta Mater. 61 (2013) 5605–5617. [31] T.B. Reed, Free Energy Formation of Binary Compounds, MIT Press, Cambridge, Massachusetts, 1971. [32] D.M.H. Bailly, C. Prinier, The Nuclear Fuel of Pressurized Water Reactors and Fast Neutron Reactors, Lavoisier Publishing, Paris, 1999. [33] W.Z. Xu, L.L. Li, M. Saber, C.C. Koch, Y.T. Zhu, R.O. Scattergood, Nano ZrO2 particles in nanocrystalline Fe–14Cr–1.5Zr alloy powders, J. Nucl. Mater. 452 (2014) 434–439. [34] K.A. Darling, R.N. Chan, P.Z. Wong, J.E. Semones, R.O. Scattergood, C.C. Koch, Grainsize stabilization in nanocrystalline FeZr alloys, Scr. Mater. 59 (2008) 530–533. [35] K.A. Darling, B.K. VanLeeuwen, C.C. Koch, R.O. Scattergood, Thermal stability of nanocrystalline Fe–Zr alloys, Mater. Sci. Eng. A 527 (2010) 3572–3580. [36] M. Saber, H. Kotan, C.C. Koch, R.O. Scattergood, Thermal stability of nanocrystalline Fe–Cr alloys with Zr additions, Mater. Sci. Eng. A 556 (2012) 664–670.

200


[37] M. Saber, H. Kotan, C.C. Koch, R.O. Scattergood, Thermodynamic stabilization of nanocrystalline binary alloys, J. Appl. Phys. 113 (2013) 063515. [38] M. Saber, H. Kotan, C.C. Koch, R.O. Scattergood, A predictive model for thermodynamic stability of grain size in nanocrystalline ternary alloys, J. Appl. Phys. 114 (2013) 103510. [39] H. Kotan, K.A. Darling, R.O. Scattergood, C.C. Koch, Influence of Zr and nano-Y2O3 additions on thermal stability and improved hardness in mechanically alloyed Fe base ferritic alloys, J. Alloys Compd. 615 (2014) 1013–1018. [40] R. Gao, T. Zhang, X.P. Wang, Q.F. Fang, C.S. Liu, Effect of zirconium addition on the microstructure and mechanical properties of ODS ferritic steels containing aluminum, J. Nucl. Mater. 444 (2014) 462–468. [41] J.H. Kim, J.H. Lee, J.Y. Min, S.W. Kim, C.H. Park, J.T. Yeom, T.S. Byun, Cryomilling effect on the mechanical alloying behaviour of ferritic oxide dispersion strengthened powder with Y2O3, J. Alloys Compd. 580 (2013) 125–130. [42] L. Toualbi, M. Ratti, G. André, F. Onimus, Y. de Carlan, Use of neutron and X-ray diffraction to study the precipitation mechanisms of oxides in ODS materials, J. Nucl. Mater. 417 (2011) 225–228. [43] L. Dai, Y. Liu, Z. Dong, Size and structure evolution of yttria in ODS ferritic alloy powder during mechanical milling and subsequent annealing, Powder Technol. 217 (2012) 281–287. [44] S.T.Y. Kimura, S. Suejima, R. Uemor, H. Tamehiro, Ultra grain refining and decomposition of oxide during super-heavy deformation in oxide dispersion ferritic stainless steel powder, ISIJ Int. 39 (1999) 176–182. [45] A.K. Giri, C. de Julián, J.M. González, Coercivity of Fe‐SiO2 nanocomposite materials prepared by ball milling, J. Appl. Phys. 76 (1994) 6573–6575. [46] S.R. Linderoth, M.S. Pedersen, Fe‐Al2O3 nanocomposites prepared by high‐energy ball milling, J. Appl. Phys. 75 (1994) 5867–5869. [47] S.C. Axtell, R. Schalek, Magnetic properties and grain growth stability of nanocomposite Fe–ZrO2 granular solids prepared by mechanical milling, J. Appl. Phys. 79 (1996) 5263–5265. [48] M.N.I. Zhitomirsky, A. Petric, Synthesis and magnetic properties of Ni–zirconia composites, Mater. Manuf. Process. 18 (2003) 719–730. [49] A.P. Guimarães, Principles of Nanomagnetism, Springer-Verlag Berlin Heidelberg, Berlin Heidelberg, 2009. [50] R. Kaiser, The Agglomeration of Zinc Oxide Powders, Massachusetts Institute of Technolgy, Massachusetts, 1961. [51] J. Ma, L.C. Lim, Effect of particle size distribution on sintering of agglomerate-free submicron alumina powder compacts, J. Eur. Ceram. Soc. 22 (2002) 2197–2208. [52] F.-S. Shiau, T.-T. Fang, T.-H. Leu, Effects of milling and particle size distribution on the sintering behavior and the evolution of the microstructure in sintering powder compacts, Mater. Chem. Phys. 57 (1998) 33–40. [53] C.D. Sagel-Ransijn, A.J.A. Winnubst, B. Kerkwijk, A.J. Burggraaf, H. Verweij, Production of defect-poor nanostructured ceramics of Yttria-Zirconia, J. Eur. Ceram. Soc. 17 (1997) 831–841. [54] B.D. Cullity, S.R. Stock, Elements of X-ray Diffraction, Prentice Hall, 2001. [55] P.S.P. Debye, Phys. Z. 17 (1916) 277–283. [56] P. He, On the Structure–property Correlation and the Evolution of Nano features in 12–13.5% Cr Oxide Dispersion Strengthened Ferritic Steels, KIT Scientific Publishing, Karlsruhe, 2013. [57] Z.P.L.J.H. Schneibel, S.H. Shim, Nanoprecipitates in Steels, in: R.R. Judkins (Ed.), Twenty First Annual Conference on Fossil Energy Materials 2007, pp. 212–219. [58] P.K. Parida, A. Dasgupta, K. Jayasankar, M. Kamruddin, S. Saroja, Structural studies of Y2O3 dispersoids during mechanical milling and annealing in a Fe-15 Y2O3 model ODS alloy, J. Nucl. Mater. 441 (2013) 331–336.

[59] L.S. Hung, M. Nastasi, J. Gyulai, J.W. Mayer, Ion‐induced amorphous and crystalline phase formation in Al/Ni, Al/Pd, and Al/Pt thin films, Appl. Phys. Lett. 42 (1983) 672–674. [60] P. Susila, D. Sturm, M. Heilmaier, B.S. Murty, V. Subramanya Sarma, Effect of yttria particle size on the microstructure and compression creep properties of nanostructured oxide dispersion strengthened ferritic (Fe–12Cr–2 W–0.5Y2O3) alloy, Mater. Sci. Eng. A 528 (2011) 4579–4584. [61] T. Neuberger, B. Schöpf, H. Hofmann, M. Hofmann, B. von Rechenberg, Superparamagnetic nanoparticles for biomedical applications: Possibilities and limitations of a new drug delivery system, J. Magn. Magn. Mater. 293 (2005) 483–496. [62] R.K. Khatirkar, B.S. Murty, Structural changes in iron powder during ball milling, Mater. Chem. Phys. 123 (2010) 247–253. [63] I. Groma, X-ray line broadening due to an inhomogeneous dislocation distribution, Phys. Rev. B 57 (1998) 7535–7542. [64] T. Ungár, A. Borbély, The effect of dislocation contrast on x‐ray line broadening: a new approach to line profile analysis, Appl. Phys. Lett. 69 (1996) 3173–3175. [65] T. Ungár, G. Tichy, The effect of dislocation contrast on X-Ray line profiles in untextured polycrystals, Phys. Status Solidi A 171 (1999) 425–434. [66] T. Ungár, S. Ott, P.G. Sanders, A. Borbély, J.R. Weertman, Dislocations, grain size and planar faults in nanostructured copper determined by high resolution X-ray diffraction and a new procedure of peak profile analysis, Acta Mater. 46 (1998) 3693–3699. [67] G. Ribárik, Modeling of Diffraction Patterns Based on Microstructural Properties, Department of Materials Physics, Eötvös Loránd University, Budapest, Hungary, 2008. [68] T. Ungar, Strain Broadening Caused by Dislocations, Advances in X-ray Analysis (AXA) — the proceedings of the Denver X-ray Conferences, JCPDS — International Centre for Diffraction Data, 1996. [69] S. Vives, E. Gaffet, C. Meunier, X-ray diffraction line profile analysis of iron ball milled powders, Mater. Sci. Eng. A 366 (2004) 229–238. [70] A. Pandey, H. Palneedi, K. Jayasankar, P. Parida, M. Debata, B.K. Mishra, S. Saroja, Microstructural characterization of oxide dispersion strengthened ferritic steel powder, J. Nucl. Mater. 437 (2013) 29–36. [71] M. Mhadhbi, M. Khitouni, L. Escoda, J.J. Suñol, M. Dammak, Characterization of mechanically alloyed nanocrystalline Fe(Al): crystallite size and dislocation density, J. Nanomater. (2010) 8. [72] P. Bhaskar, A. Dasgupta, V.S. Sarma, U.K. Mudali, S. Saroja, Mechanical properties and corrosion behaviour of nanocrystalline Ti–5Ta–1.8Nb alloy produced by cryorolling, Mater. Sci. Eng. A 616 (2014) 71–77. [73] Q. Fang, Z. Kang, An investigation on morphology and structure of Cu–Cr alloy powders prepared by mechanical milling and alloying, Powder Technol. 270 (2015) 104–111 (Part A). [74] L.J. Taylor, D.G. Papadopoulos, P.J. Dunn, A.C. Bentham, J.C. Mitchell, M.J. Snowden, Mechanical characterisation of powders using nanoindentation, Powder Technol. 143–144 (2004) 179–185. [75] E.O. Hall, The deformation and ageing of mild steel, Proc. Phys. Soc. Lond. Sect. B 64 (1951) 747–753. [76] N.J. Petch, The cleavage strength of polycrystals, J. Iron Steel Inst. Lond. 173 (1953) 25–28. [77] R. Vijay, M. Nagini, J. Joardar, M. Ramakrishna, A.V. Reddy, G. Sundararajan, Strengthening mechanisms in mechanically milled oxide-dispersed iron powders, Metall. Mater. Trans. A 44 (2013) 1611–1620. [78] P. Susila, D. Sturm, M. Heilmaier, B.S. Murty, V. Subramanya Sarma, Microstructural studies on nanocrystalline oxide dispersion strengthened austenitic (Fe–18Cr–8Ni– 2 W–0.25Y2O3) alloy synthesized by high energy ball milling and vacuum hot pressing, J. Mater. Sci. 45 (2010) 4858–4865.