Keywords

Introduction

The primary life-limiting factor for light water reactors (LWR) is the environmental degradation of reactor components due to radiation damage, corrosion, and high temperature phase transformation. Key components, such as reactor pressure vessel heads and steam generators, are difficult and costly to replace, which makes lifetime extension an important factor in reactor plant license renewal and safety evaluation. Ni-based alloys are widely used in nuclear plants due to their excellent corrosion resistance, fracture toughness, and phase stability during ageing.

Primary components that are now being made with Ni-based (e.g. Alloys 690 and 625) alloys include: steam generator tubing, reactor pressure vessel heads, and control rod drive mechanism penetrations, which are designed for relatively high temperatures (~300 °C) and necessary long lifetimes. The Ni/Cr ratio of these commercial alloys is possibly located in the ordering phase transformation field of the Ni–Cr phase diagram (from 0.65 to 0.7 Ni%) [1]. There is little experimental data in this region to confirm the extent of the ordered phase. Ordering occurs when the attractive force between different atoms is larger than that of same atoms. Thus during the ordering transformation of Ni–Cr system, atoms rearrange from random solid solution into regular patterns [2] (layers of Ni and Cr in MoPt2-type structure shown in Fig. 1) below a critical temperature, Tc = 590 °C [1]. If the alloy is in the ordered phase region or in a two-phase region (bcc Cr + MoPt2), the ordered phase will nucleate and grow [3,4,5,6,7]. Based on the result from previous research on binary Ni–Cr alloys, the ordered phase can cause an increase in tensile strength, a decrease in ductility, and dimensional changes or internal stresses. Previous research results also show a hardening effect from ordering and a linear relationship between Vickers microhardness and yield strength [8]. It is known that the fracture toughness decreases as yield strength increases, and fracture toughness is a key factor in stress corrosion crack (SCC).

Fig. 1
figure 1

Sketch of MoPt2 structure (solid line) with a FCC cell (broken line)

Full size image

The main objectives of this project are to define the relationship between the disorder-order transformation and stoichiometries in the phase diagram. To accomplish the goals, four Ni–Cr binary model alloys with different stoichiometries were isothermally aged and characterized to determine the range of thermal embrittlement susceptibility.

Experimental Methods

The binary model alloys were made in stoichiometry ratios of Ni/Cr (at.%) of 1.8, 2.0, 2.2, and 2.4 to explore the unknown region of the phase diagram of Ni–Cr under 500 °C. Samples were fabricated by small batch (~300 g) arc-melting, then hot-rolled, homogenized in a furnace for 24 h at 1093 °C and water-quenched to room temperature. Compositional analysis was performed by a third-party test laboratory based on method CAP-017 N (ICP-AES) and ASTM 1019-11 (Comb./IGF) (Table 1).

Table 1 Composition of binary model Ni–Cr alloys (at.%)
Full size table

Alloy button plates were cut into 10 × 10 × 5 mm3 specimens by electrical discharge machining to minimize heat generation during the cutting process. All alloys (four stoichiometries) were heat-treated at three temperatures. 373, 418, and 475 °C, for up to 10,000 h. The heat treatment test matrix is shown in Table 2.

Table 2 Heat treatment test matrix of binary model Ni–Cr alloys
Full size table

For micro-hardness tests, the surfaces of samples were ground using SiC papers from 240 to 800 grit followed by 0.05 µm colloidal SiO2 polishing. Micro-hardness measurements were performed using a Micromaterials Vantage nanoindenter with a Berkovich diamond tip and a load of 500 mN. The load was chosen to balance degradation of indenter tip, sensitivity for vibration from environment during testing, resistance on effect of grain orientation, effect of grain boundary on indentation, and the accuracy level of result. A larger load causes faster tip damage, which changes the dynamic area function that used to calculate micro-hardness. A smaller load requires lower environmental vibration. Additionally, grain orientation and presence of grain boundaries will have a greater effect for a low indentation load than for a high load when measuring mirco-hardness. Considering these requirements, microhardness measurements using a grid of (4 × 5) were performed on the sample surface with an distance of 600 µm between indentations.

Transmission electron microscopy (TEM) specimens were prepared by jet electro-polishing in 20% HClO4–80% CH3OH at −35 °C using a Struers Tenupol 5 with a Julabo FP50 closed-cycle refrigeration system. Microstructural characterization of the aged Ni–Cr samples was performed using the FEI Tecnai F30 and FEI Talos F200 field emission gun analytical transmission electron microscopes.

Atom probe tomography (APT) specimens were prepared using an FEI Nova 200 Nanolab FIB. APT specimens were fabricated by utilizing concentric annular milling with a series of decreasing ion currents at 30 kV applied voltage and a 5 kV final milling step. Data were collected using a Cameca LEAP 4000× HR operated in voltage mode using a 20% pulse fraction, 200 kHz pulse rate, and a target detection rate of 0.5% at 50 K. Reconstructions were performed using the IVAS 3.6.12 software. Reconstruction parameters were adjusted to ensure uniform axial and radial density, which resulted in image compression factor values in the range of 1.2–1.7 and k-factor values between 3.3 and 3.5. The evaporation field was taken as that of Ni, 35 V/nm.

Results and Discussion

Transmission Electron Microscopy

The Ni–Cr alloys were characterized by a fully recrystallized equiaxed grain structure; the grain size ranged from 70 to 360 µm. Selected area electron diffraction (SAED) was used to assess the formation of the ordered Pt2Mo-type Ni2Cr precipitates during ageing. The strategy to characterize Pt2Mo-type ordered structure in Ni–Cr alloys via TEM has been well documented in previous research [3, 5, 8, 9]. The additional reflections in the electron diffraction patterns were analyzed to confirm the formation of Ni2Cr precipitates. Figure 2 shows the evolution of secondary reflection in Ni/Cr = 2.0 sample. Note that no Ni2Cr precipitate reflections were detected in the specimens aged at 373 °C or 418 °C for 3000 h, but the presence of Ni2Cr was confirmed after 3000 h at 475 °C.

Fig. 2
figure 2

TEM SAED patterns obtained from Ni/Cr = 2.0 specimens aged for 3000 h at: a 373 °C ([011] zone axis); b 418 °C ([112] zone axis); and c 475 °C ([114] zone axis, ordered precipitate reflection can be observed at 1/3 and 2/3 {311}matrix and {220}matrix)

Full size image

The nano-scale Ni2Cr precipitates formed during ageing at 475 °C are shown in the dark-field TEM image in Fig. 3a. These ultra-fine intragranular Ni2Cr precipitates were less than 5 nm in size. The TEM observations are summarized in Table 3.

Fig. 3
figure 3

a Dark-field TEM image of the Ni2Cr precipitates, and b corresponding [100] SAED pattern showing the ordered precipitate reflections at equivalent 1/3 and 2/3 {220}matrix for Ni/Cr = 2.0 aged at 475 °C for 5000 h

Full size image
Table 3 TEM characterization results for Ni/Cr = 2.0 samples
Full size table

Atom Probe Tomography

Atom probe tomography was used to identify the extent of the ordering transformation. The characterization of Pt2Mo-type precipitates is challenging because of similar matrix and precipitate compositions and evaporation artifacts that lead to significant density variations in the reconstructed data that remain to be explained. These density variations are visible in Fig. 4 (left). Nonetheless, the ordered phase can still be detected by APT using appropriately oriented specimens so that the ordered lattice planes are visible in the datasets [10, 11]. Such an ordered precipitate is highlighted in Fig. 4. The composition of the ordered precipitates was measured from several cube regions encompassing the ordered precipitates, as the one in Fig. 4 but smaller to ensure that the region does not exceed the precipitate. The measured precipitate composition shown in Table 4 agrees with the expected stoichiometry .

Fig. 4
figure 4

Slices from APT datasets showing Cr atoms in pink (left) and Ni (green) and Cr atoms in the ordered precipitates (right). The thickness of the left slice is 2 nm, and that of the right is 5.5 nm

Full size image
Table 4 The composition of Ni2Cr precipitates after ageing at 475 °C for 10,000 h
Full size table

Micro-hardness

The micro-hardness of model alloys as a function of ageing time at three temperatures, 373, 418, and 475 °C, is shown in Fig. 5. At 475 °C (Fig. 5a), the stoichiometric samples have a higher ordering rate than off-stoichiometric samples. For off-stoichiometric samples, the behavior is similar to the low temperature exposure, which means stoichiometric samples experience a faster ordering rate than off-stoichiometric samples. Additionally, behaviors of micro-hardness start getting stable after 1000 h, which indicates ordering transformation is approaching equilibrium. Considering the change of micro-hardness comes from ordering phase, lower amount of ordered phase has been formed in off-stoichiometric samples than in-stoichiometric samples. Based on the relationship between micro-harness and ageing time, comparison of ordered phase fraction between all samples is 2.0 > 1.8 ≈ 2.2 > 2.4.

Fig. 5
figure 5

Micro-hardness versus Ageing time at 373, 418, and 475 °C. °C for Ni/Cr ratios or 1.8, 2, 2.2 and 2.4

Full size image

At 418 °C (Fig. 5b), the ordering behavior is similar for all compositions. Different degrees of hardening can be observed at all compositions. Due to the higher ageing temperature and higher diffusivity of chromium at this temperatures, the rate of hardening (or ordering ) is faster than that at 373 °C. The microhardness results for the samples aged at 373 °C are presented in Fig. 5c, which shows a unclear relationship between micro-hardness and ageing time for both stoichiometric samples (Ni/Cr = 2.0) and off-stoichiometric samples due to the small change on micro-hardness and large uncertainty. All samples show a similar tendency of embrittlement.

The trend of change in micro-hardness (Fig. 5) shows that the effect of hardening will reach saturation after 1000 h of heat treatment at 475 °C, whereas the time to saturation for 418 and 373 °C will be longer. By comparing the change in micro-hardness after ageing for 5000 h, which is shown in Fig. 6, the relationship between embrittlement and stoichiometry can be observed. The larger deviation from stoichiometry leads to less embrittlement.

Fig. 6
figure 6

Change in micro-hardness as a function of Ni/Cr stoichiometry after 5000 h ageing treatment at 373, 418, and 475 °C

Full size image

Conclusions

In this paper, the role of stoichiometry on the Ni–Cr ordering transformation was studied by micro-hardness and the extent of precipitation was confirmed by TEM (imaging and SAED) and APT. The study shows that the Ni2Cr precipitates have an obvious hardening effect on the Ni–Cr alloys. Fine intragranular precipitates (<5 nm) were observed via TEM after disorder-order transformation. The superlattice structure caused by increased local atomic density was detected by APT, whereas the difference in composition between precipitates and matrix phase is insignificant and cannot be used to distinguish them.

The micro-hardness data shows that stoichiometry alloys (Ni/Cr = 2.0) order more rapidly than off-stoichiometry alloys. Larger deviations from stoichiometry leads to slower ordering rates. Off-stoichiometry Ni–Cr samples are not immune to disorder-order transformation, which suggests that off-stoichiometry commercial alloys may exhibit a hardening effect from the Pt2Mo-type ordered phase. In the model alloys , hardening due to Ni2Cr achieved equilibrium after 1000 h ageing at 475 °C. The saturation limit for intermediate (418 °C) and low (373 °C) temperatures is still unclear because of sluggish kinetics.