1 Introduction

The development of advanced materials for high-technology applications is an ever-active domain with consistent growth and breakthroughs. Modern technologies, including clean energy sector, require materials which have specific properties for their intended applications and are being constantly looked for (Ref 1,2,3,4,5,6). Oxide Dispersion-Strengthened (ODS) steels are one such class of materials with enhanced creep strength at elevated temperatures up to 973 K, which are being developed as a core structural material for next-generation Fast Breeder Reactors (FBRs) (Ref 1, 7). In particular, ODS alloys are being developed as clad tube, which houses the nuclear fuel pellets. Primary requirement for these ODS alloys is the superior creep strength at operating temperatures which could be as high as 923–973 K and additionally, these alloys also need to have superior void swelling resistance and sustained high temperature mechanical strength (Ref 1, 7, 8). While Austenitic steels are currently being used as nuclear structural materials in critical components, ferritic steels are increasingly being considered for next-generation sodium-cooled FBRs due to their open body-centered cubic (bcc) structure, which is known to provide superior void swelling resistance compared to their austenitic counterparts. However, high temperature creep strength is an issue with ferritic steels, leading to their restricted usage at higher temperatures. This issue can be resolved by addition of fine ceramic dispersions, as in the case of ODS alloys, wherein the ceramic dispersions block the mobile dislocations to enhance their high temperature creep strength. The addition of ceramic dispersions, however, cannot be done through conventional melting route owing to the density differences between the alloy and dispersoid and have to be carried out through powder metallurgical/ball milling route. Although there are persistent explorations on alternate methods of ODS alloy synthesis (Ref 9,10,11,12,13), the powder metallurgy synthesis is still extensively used and well-established. The most common dispersoid used in ODS steels is Y2O3, and it has been studied widely. However, the limitation with Y2O3 is its coarsening at elevated temperatures during consolidation of ball milled powders which will affect the mechanical properties of ODS steel and it is a serious limitation in its applications (Ref 14). This problem could be overcome by adding Ti to Y2O3 to facilitate the formation of complex oxides in the form of Y2Ti2O7 and YTiO5, which are finer and stable and subject of wide research (Ref 15, 16). Another approach to overcome this issue is to look for the alternate oxides which have better dispersoid refinement properties and lower coarsening kinetics. Owing to its huge potential, this approach is capable of producing viable outcomes and possibly solve the issue we face with dispersoid coarsening.

Alternate dispersoids are being investigated in the recent past for their strengthening effects in alloys, specifically at high temperatures. In addition, since the irradiation response is a vital factor in nuclear structural materials, alternative dispersoids with varying chemistry are of significant interest. This is because, the evolution of the nano-sized oxides under irradiation depends on the chemical composition of the oxides (Ref 17), and hence, there could be novel dispersoids with better irradiation response than the widely used Y2O3/Y2O3-Ti complexes. Few of the reports on alternate dispersoids specific to ODS alloys include using ZrO2, HfO2, CeO2, La2O3 and even the ex situ synthesized Y2Ti2O7 and Y4Zr3O12 as a dispersion strengthener (Ref 18,19,20,21,22,23,24). In one of the reports, Hoffmann et al. (Ref 22) investigated the mechanical properties of various steels strengthened by dispersoids, namely ZrO2, CeO2, HfO2 and La2O3 and reported superior mechanical properties with ZrO2 and La2O3 and proved the merit of exploring alternate dispersoids. This exploration is also fueled with the geopolitical scenarios with non-uniform distribution of rare earths in various parts of world as reported by Pasebani et al. (Ref 20, 21), who explored La2O3 as an alternate dispersoid owing to its abundance in USA and reported promising results with it.

Lately, there has been a growing interest in utilizing the advantages of ZrO2 as a dispersion strengthener due to its smaller unit cell structure and relatively lower absorption cross section for fast neutrons. (Ref 22, 25,26,27,28,29). The ZrO2 has a monoclinic structure which is stable up to 1443 K with no intermediate phase transformations and also has a smaller unit cell structure with 12 atoms in its unit cell (mP12). The smaller unit cell structure is predicted to be beneficial in retaining its crystal structure during the mechanical milling process. In addition to the direct addition of ZrO2 in to the steel matrix as reported in literatures, the Zr is also reported as a potential microalloying element to produce ODS steels with favorable mechanical performance, irradiation response (Ref 30,31,32) and to refine the dispersoid sizes and optimize nanoprecipitate dispersion characteristics (Ref 33,34,35). A good combination of tensile strength and ductility is reported by Zhou et al. where Zr is used as a microalloying element and ZrO2 and complex Y-Zr-O oxide formation is reported (Ref 36). There are promising results on irradiation response of Zr added ODS steel with formation of Y-Zr-O precipitates, which are superior to Y-Al-O and even Y-Ti-O precipitates (Ref 32, 37). Promising irradiation response with less susceptibility to radiation hardening is also reported in an austenitic ODS steel strengthened by Y2O3-ZrO2 (Ref 38), further emphasizing the importance of exploring alternate dispersoids such as ZrO2. With indications of being a potential strengthener, the ZrO2 needs to be explored more in order to exploit its full potential.

Conventionally the ODS alloys are prepared through powder metallurgy route followed by consolidation through hot isostatic pressing (HIP) or spark plasma sintering (SPS). Generally, this process is followed with a heat treatment sequence for achieving the desired microstructure of the material. It is known that these processing steps play a huge role in final microstructure developed, and hence, it becomes essential to understand the process of microstructure evolution. We have previously reported (Ref 26) the microstructure development in the ball milled powder, and, in this article, we report the microstructure and mechanical property analysis of a SPS’d ZrO2 dispersion-strengthened 9Cr steel alloy. The choice of 9Cr variant steel as base matrix is driven by the fact that it is at a higher technology readiness level for fuel cladding tubes among all the candidate steels, and hence, it is one of the closest materials to actual industrial implementation phase (Ref 39). The objective of the current study is to demonstrate the feasibility of using ZrO2 as a dispersion strengthener in 9Cr steel system and to evaluate its high temperature microstructure and mechanical properties. In order to accomplish this, an extensive study of the material is performed by utilizing electron microscopy and allied techniques to evaluate its microstructure and the results of the analysis are presented in this report.

2 Experimental

The synthesis of 9Cr ODS steel powder with 0.35 wt.% ZrO2 is carried out in a planetary ball mill in an Ar atmosphere. Detailed chemistry of the initial powder is given in Table 1. The particulars of the synthesis, its microstructure and microchemical characterization and the optimization of milling conditions are reported previously (Ref 26, 28, 29). The optimized milling duration was reported to be 100 h, and the synthesized powders at the optimal milling conditions had a grain size range of 1–20 μm (Ref 26). The process of consolidating the milled powders was carried out using Dr. Sinter make Spark Plasma Sintering (SPS) instrument at a temperature of 1323 K and a pressure of 50 MPa, applied for 10 min. The applied heating rate was 50 °C/min, and the samples were 20 mm in diameter and 5 mm height. The choice of Temperature and Pressure was based on the literature reports from similar systems (Ref 40,41,42,43,44,45,46,47) by considering a high relative density and reduced porosity. The density of the consolidated specimens is calculated by Archimedes principle. Five specimens were cut from various regions of the consolidated material and used for the measurements. The conventional Normalizing and Tempering (N&T) treatment is carried out for about 1 h each at temperatures of 1323 and 1023 K in a muffle furnace, to homogenize the microstructure (Ref 39). Further, the long-term annealing of the N&T sample at 973 K is carried out for 400 and 1000 h, to understand the microstructural stability. The heating rate was maintained at ~ 20 °C/min to reach the desired temperature. Surface microstructure of the specimens was analyzed through a FEI make Dual Beam Helios NanoLab-600i FEG–Scanning Electron Microscope (SEM) attached with an EDAX make Apollo X Silicon Drift Detector for Energy-Dispersive Spectroscopy (EDS) microanalysis. The grain structure was examined through an EDAX make Electron Backscattering Diffraction (EBSD) detector, attached with the SEM. A 200 kV Philips make CM-200 Transmission Electron microscope (TEM) attached with TVIPS make 2 × 2 K CCD camera was used to observe the microstructure. The mechanical hardness of the material was analyzed through Vickers hardness measurements with a 50 g load. A total of 20 measurements were performed for each sample, before reporting the average hardness value.

Table 1 Chemical composition of the 9 Cr steel powder (Wt.%)
Full size table

3 Results and Discussion

3.1 Density Measurement of Consolidated 9Cr ZrO2 ODS Steel

The density of the consolidated specimen was calculated by Archimedes principle, as explained previously. The objective here was to calculate an overall average density of the sample and look for any possible inhomogeneity. The measured values of the specimens taken from various regions of the consolidated sample are tabulated in Table 2.

Table 2 Measured densities of 9 Cr ZrO2 ODS
Full size table

It can be observed from the above table that relative densities vary between ~ 92 and 96% with an average of ~ 94%. This indicates a uniform distribution of porosity and less probability of local inhomogeneity within the sample during consolidation. It also indicates that the sintering parameters are optimum to achieve the higher efficiency of thermal energy, which aided in faster diffusion and compaction. For further analysis, sample region with relatively high density was used.

3.2 Microstructural Analysis of as-SPS’d 9Cr ZrO2 ODS Steel

Typical microstructure of the as-SPS steel is shown in Fig. 1(a). The specimen showed a fairly homogeneous distribution of grains and pores, and the grain features were clearly smaller. The overall distribution of pores is about ~ 50–80 nm size and exhibits a tendency to network along grain boundaries. Figure 1(b) shows the EBSD Inverse Pole Figure (IPF) of the same sample at a relatively lower magnification superimposed with grain boundary, after the standard post processing to assign average orientation to non-indexed points based on its surrounding pixels. It reveals equiaxed grain morphology with nearly similar grain shape/size. Grain boundaries with a misorientation 5–15° and > 15° were identified as high angle and low angle boundaries, respectively. The average grain size was measured to be 0.8 ± 0.3 μm. The colors denote the type of crystallographic plane aligned with the sample surface, and the myriad of colors denotes an essential random type of texture in the specimen. It could be argued here that grain growth kinetics is restricted in the process due to minimum holding time at high temperature during consolidation through SPS and subsequent pinning effect of ZrO2 dispersoids.

Fig. 1
figure 1

Microstructure of as-SPS 9Cr ZrO2 ODS steel (a) Secondary Electron micrograph (b) Inverse Pole Figure map showing the equiaxed grain structure

Full size image

Figure 2 shows the TEM microstructure of the as-SPS’d 9Cr ZrO2 ODS steel. The diffraction contrast from Fig. 2(a) reveals a microstructure, in which some fine faceted ferrite grains with high dislocations density are seen. This is an indicative of retention of strain even after consolidation. In addition to it, few larger precipitates are also seen from this figure. TEM investigation at higher magnification revealed the presence of uniform distribution of nano-sized dispersoids inside the grains as shown in Fig. 2(b) with inset showing the selected area electron diffraction (SAED) patterns with unique reflections from ZrO2 highlighted with an arrow mark. It is observed from further analysis that these nano-sized dispersoids in the matrix after consolidation through SPS retain their fine size ranging between 2 and 12 nm. The size distribution here is in close agreement with the reported values of < 5 nm, for the same system in as milled condition (Ref 26). The results are also in line with the reported values in the literature for similar ODS systems consolidated through SPS. Meza et al. (Ref 48) have reported a size distribution of 2—22 nm for a 14Cr-Y2O3 ODS system which reduced to 2—12 nm with Boron addition and the average size decreased from 8.11 to 4.26 nm. In another study, Li et al. (Ref 49) showed an average dispersoid size of 7 nm with a size distribution of 2–22 nm in a 14Cr-ODS system, just after SPS, with majority of dispersoids possessing a size less than 10 nm. The superiority of SPS in terms of its short residence time compared to other consolidation techniques is evident here, as we could retain the fine structure of the dispersoids without any detrimental increase in size for the nano-dispersoids. Further, owing to this strained inhomogeneous microstructure as observed through TEM, it is concluded that this issue needs to be addressed before carrying out further investigation. Accordingly, the samples were normalized at 1323 K for 1 h followed by a tempering treatment at 1023 K for 1 h to homogenize the microstructure. In addition to that, to assess the long-term effects on high temperature exposure, the samples were treated at 973 K for 400 and 1000 h durations and subsequently characterized. The results of the same are discussed in subsequent sections.

Fig. 2
figure 2

Transmission Electron Micrographs of as SPS 9Cr ZrO2 ODS steel showing (a) microstructure with grains and precipitate and (b) Grain dispersed with finer dispersoids with inset showing the SAED pattern

Full size image

3.3 Effect of High-Temperature Treatment on 9Cr ZrO2 ODS Steel—Normalizing & Tempering and Aging Studies

3.3.1 Microstructural Analysis by Scanning Electron Microscopy and Electron Backscatter Diffraction

Figure 3(a) shows the EBSD crystal orientation map of N&T 9Cr ZrO2 ODS steel. A similar microstructure in comparison with as SPS condition is observed here with an average grain size of ~ 1 ± 0.4 μm, also exhibiting a random orientation. Further, the long-term thermal treatment of the N&T sample at 973 K is carried out for 400 and 1000 h, to understand the microstructural stability as discussed previously. The temperature chosen here is typical at the service condition of reactor operation, and hence, this study is expected to provide vital information about the material microstructure. Figure 3(b) and (c) shows the corresponding EBSD crystal orientation map for the 400 and 1000 h aged samples, respectively. The grain size distribution is quantified, and analysis is carried out and shown in Fig. 3(d). All three samples (N&T, N&T + 400 h aged and N&T + 1000 h aged) show a uniform size distribution with little shift in peak grain size toward higher values. The average value of ~ 1.0 ± 0.4 μm, 1.1 ± 0.5 μm and 1.3 ± 0.6 μm was observed for N&T, 400 and 1000 h samples, respectively. The grain size distribution of as SPS sample is also been included in the graph and shown a value of ~ 0.8 ± 0.3 μm. The finer grain size distribution is a characteristic of SPS’d steels, and similar values are reported for other 9Cr ODS system in the literature (Ref 50). In addition, the retention of fine grain size after aging is also in line with previously reported studies (Ref 51, 52), where a grain size of 1.9, 2.2 and 2.5 μm is reported after SPS, aging at 973 K for 5000 and 10000 h, respectively. The finer grain sizes even after prolonged annealing are indicating the superior stability of the microstructure in this time–temperature domain which could be attributed due to the presence of hasten pinning effect by uniformly distributed fine dispersoids even at 973 K.

Fig. 3
figure 3

EBSD Inverse Pole Figure map superimposed with grain boundary (> 10°) of 9 Cr ZrO2 ODS under different thermal exposure sequences. (a) Normalized and Tempered, (b) Normalized and Tempered and 400 h aged and (c) Normalized and Tempered and 1000 h aged specimen, and (d) comparative grain size distribution map

Full size image

The superior stability of the microstructure with no substantial grain growth and saturated strain during this prolonged annealing is believed to be arising also from the stability of the dispersoids at this time–temperature domain. The dispersoids are known to effectively pin the grain boundaries thus inhibiting their growth is believed to be the case here, which is also in line with what is reported in the literatures (Ref 30). It was reported previously (Ref 28) by us that the dispersoids were quite stable at 1223 K retaining their nanocrystallinity with a size range of 2–14 nm, even in a model alloy and similar stability of the dispersoids are expected here.

3.3.2 Microstructural Analysis by Transmission Electron Microscopy

To shed more light on the microstructural stability displayed by the sample, as seen via EBSD analysis, further investigation through TEM is carried out and discussion is made herewith.

Figure 4(a) shows the TEM Bright Field (BF) micrograph of N&T sample exhibiting a finer microstructure. Owing to a small quantity of carbon (~ 0.01 wt.%), the martensitic structure is not obviously seen, but only finer martensitic lath features are revealed, which are marked in the figure. Figure 4(b) shows the BF micrograph at a higher magnification, showing uniformly distributed nano-dispersoids. Further, the microstructure of long-term aged samples is also investigated and shown in the figure. Figure 4(c) and (d) displays the BF micrographs that correspond to N&T + 400 h aged sample at low and high magnifications, respectively. A matrix with equiaxed grains, in line with the EBSD analysis, is observed here also. The nano-features, as seen in from a grain interior in Fig. 4(d), show the retention of fine nature of these features. The results are incessant with 1000 h aged samples also, and shown in Fig. 4(e) and (f), wherein the inset shows the SAED pattern, confirming the matrix and dispersed phase. Further, the quantification of nano-dispersoids in the materials is carried out and the results are summarized in Fig. 4(g), which shows a dispersoid size distribution histogram including the size distribution for as-SPS sample. It was observed that the peak of the size distribution chart is shifting slightly toward a higher size from N&T to 1000 h. with a slight difference in the maximum value which varies from 13 nm in-as SPS to 22 nm in 1000 h aged sample. It is quite evident from the figure that the dispersoids retain their fine structure throughout the annealing duration with an average size varying from ~ 4 nm in the as SPS sample to ~ 7 nm after 1000 h aging, exhibiting extraordinary stability by the dispersoids. Additionally, the normal distribution to the histogram is also plotted and shown in the graph, which reveals that, the distribution of sizes is relatively narrower for the as-SPS sample in comparison with 400 and 1000 h aged samples. It is to be mentioned here that in as milled condition, the dispersoids possess a size distribution which varies from 2 to 12 nm, as reported by us previously (Ref 26) and current results show the dispersoid sizes not varying drastically from this size distribution indicating its superior stability. The value of dispersoid sizes also compares well with the reported values in the literature in a similar system. Zheng et al. (Ref 51) reported a dispersoid size of 6.9 nm after consolidation, which increased only slightly to 8.1 nm after 10000 h aging at 973 K. Here as well, the maximum size of dispersoids was limited to ~ 20 nm, in close agreement to present work. In an another study (Ref 50), Y2O3 size was reported to be 9.2 nm, immediately after sintering, which increased to 12 nm after annealing at 1073 K, once again in the same size range as reported here.

Fig. 4
figure 4

(a) & (b) TEM Bright Field micrographs of Normalized and Tempered ZrO2 ODS steel. (c)&(d) TEM Bright Field micorgraphs of Normalized and Tempered and 400 h aged ZrO2 ODS steel. (e)&(f) TEM Bright Field micorgraphs of Normalized and Tempered and1000h aged ZrO2 ODS steel. (g) Dispersoids size distribution histogram for as SPS, Normalized and Tempered and 400 and 1000 h aged samples

Full size image

3.4 Evaluation of Mechanical Property of 9Cr ZrO2 ODS Steel

3.4.1 Hardness Measurements

The microstructural analysis as discussed earlier indicated at a stable microstructure with ultrafine grains and high-density of nano-scale oxides. Attempts are made here to correlate the observed behavior to mechanical properties through Vickers Hardness measurements. Table 3 shows the variation of hardness for 9Cr ZrO2 ODS steel under different conditions. A hardness value of ~ 479 HV0.05 is observed in as SPS sample, which decreased to ~ 402 HV0.05 after N&T. Further, the hardness shows a little dip in going from N&T to 400 h aged, but shows saturation from 400 to 1000 h aging. The steep decrease in hardness from as SPS to normalizing and tempering is understood in terms of annihilation of dislocations in the material during the thermal exposure, mainly taking place during the initial hours. This is expected, as the ball milling is known to induce high amount of dislocations (Ref 29) and there will be insufficient time during consolidation through SPS for relaxation of these excess strain and consequently, strain relieving through dislocation annihilation takes place during the Normalizing and Tempering treatment and corresponding decrease in hardness is expected. The hardness values are reported in the literature for various ODS systems in summarized in Table 4 along with the values from this work. The observed spread of hardness values in the table is understood as due to the varying chemistry, process conditions and thermomechanical treatment sequences in these materials. Further, the saturation of hardness from 400 to 1000 h is also consistent with the comparable grain sizes and dispersoid sizes at these conditions, essentially indicating at a stable microstructure and it is in line with reported data on similar system. Zheng et al. reported a hardness value of 382 HV for a 9Cr ODS system, after 10000 h aging at 973 K (Ref 51). The retention of a high hardness value even after a long-term thermal exposure at 973 K is promising and expected to be beneficial from its practical application point of view.

Table 3 Vickers hardness values of the various samples with 50 g load
Full size table
Table 4 The reported values for Hardness and Yield Strength in various ODS steel systems
Full size table

3.4.2 Estimation of Yield Strength

To understand the strengthening behavior in the 9Cr ZrO2 ODS steel due to the contribution from dispersoids and fine grains, the microstructural details discussed so far are utilized and arrived at an estimation of yield stress. The yield stress at room temperature may be estimated from the equation; (Ref 53),

$$\sigma_{y } = \sigma_{m } + \sigma_{{{\text{d}} }} + \sigma_{i }$$
(1)

where, σmis the matrix yield stress, σd is the direct strengthening due to the dispersoids or nano-clusters, and σi is the indirect strengthening due to the Hall–Petch effect. The direct strengthening due to dispersoids may be written as (Ref 53),

$$\sigma_{d } = \frac{MGb}{{{\text{d}}_{p } }} \left[ {\frac{6f}{\pi }} \right]^{{^{1/2} }}$$
(2)

where, M, G, b are Taylor factor, shear modulus and the Burgers vector, respectively. The strengthening due to dispersoids can be estimated with an input from the volume fraction f of the dispersoids and its diameter, dp. The indirect strengthening due to grain size is also known as Hall–Petch strengthening and can be calculated from,

$$\sigma_{i } = k \times {\text{d}}^{ - 1/2}$$
(3)

where k is a Hall–Petch constant and d is the grain size.

Using the above formulas, the analysis of yield stress is carried out for N&T, N&T + 400 h and N&T + 1000 h samples. The grain size and average dispersoid diameter are obtained from the EBSD and TEM investigations, as discussed earlier. The other material-related parameters for the calculation of yield stress are taken from literature and listed in Table 5.

Table 5 The material constants used in the estimation of yield stress
Full size table

The various contributions for strengthening in 9Cr ZrO2 ODS are evaluated using the above parameters and summarized in Fig. 5.

Fig. 5
figure 5

Comparison of measured and calculated yield stress along with various contributions to strengthening

Full size image

It can be observed from the above graph that the major contribution for yield strength of 9Cr ZrO2 ODS steels comes mainly from dispersoids. The finer grain size also contributes substantially to the yield strength thanks to SPS. The calculated yield stress mentioned in the above graph is obtained by using the numerical relation that yield stress = 3*Vickers Hardness (Ref 54). A good correlation between measured and calculated values is obtained. The superior strength of the alloy at room temperature and a large contribution of strength from dispersoids indicate at a superior strength at elevated temperatures because dispersoids in ODS steels can suppress the dislocation movement, as well as inhibit the grain boundary migration through pinning. Thus, it is expected that material shows superior properties even at high temperatures. In order to assess the material in comparison with existing ODS systems, the literature reported Yield Strength values are compared and shown in Table 4. The observed variations in Yield Strength values in various systems are due to the varying chemistry, processing condition and thermomechanical treatment sequence, similar to the hardness values. However, when compared with material systems with similar processing and heat treatment sequences, the Yield Strength values were in fairly good agreement (Ref 55, 56).

4 Conclusions

  • The consolidation of 9Cr ZrO2 ODS powder through SPS is carried out, and a specimen of ~ 94% density is obtained.

  • Microstructure of as-SPS sample characterized through SEM/EBSD and TEM revealed a fine-grained morphology with the presence of dislocations indicating strained microstructure.

  • The studies on N&T and high-temperature long-term aged specimens indicated an excellent stability of the microstructure with retention of fine-grained structure and dispersoids.

  • The mechanical properties are measured through Vickers Hardness and correlated with Yield stress. The Yield stress is also calculated through predictive tools, and a good correlation between the two is observed. Major contributor to strength was proved to be dispersoids, with a substantial contribution from fine grains.

  • The stable microstructure with fine grains, stable dispersoids and high strength, indicates at promising high temperature properties of the material, thus validating the superiority of 9Cr ZrO2 ODS.