Comparative studies have been conducted for the developed at TsNIITMASh JSC: 15Kh2NMFA steel of class 1 (reference steel), metallurgically improved 15Kh2NMFA steel, and high-nickel steel of different laboratory heats. Specific to these steels is a very low content of impurities, wt. %: P (< 0.005), S (< 0.003) and Cu (0.01–0.003) with the total content of carbide-forming elements (Cr+Mo+V+Nb) in a range of 2.0 to 3.0 wt. %.

Table 1 presents the characteristics of the investigated steels. The heat treatment mode parameters for the high-nickel steel differs in a lower temperature of tempering than for 15Kh2NMFA-type steels.

To estimate the resistance of the investigated steels to grain boundary embrittlement as a result of the reversible temper embrittlement (Balandin et al. 1984), and to estimate the stability of their structural parameters and mechanical characteristics, an embrittlement heat treatment (EHT) mode has been developed. An embrittlement heat treatment is expected to cause the intensive accumulation of grain boundary phosphorus segregation in the investigated steels (Table 2).

A box furnace of the SNOL-2.2,5.1,8/10-I7 type was used for the EHT. The steels were heated at the maximum furnace power up to the first exposure step (600 °С) and cooled to room temperature after the last exposure step (460 °С) at a rate ~20 °C/h.

To estimate the stability of the phase structural state and characteristics of the high-nickel steel, high-nickel steel heats were also examined after a long-term isothermal exposure (IE) in a range of 450 to 550 °C for hundreds and thousands of hours.

To generalize the investigation results and identify the trends for changes in the structural parameters and the mechanical characteristics as a result of a long-term thermal exposure, the investigated states parameters were reduced to the Hollomon-Jaffe parameter (Holloman and Jaffe 1945):

H_P = [(T + 273)·(20 + log(t))]/100 (1)

where T is the exposure temperature, °C, and t is the time of the thermal exposure, h.

The EHT mode was also reduced to parameter H_P using expression (1) by way of recounting each step in terms of the equivalent exposure time with an average temperature of 500 °C.

To detect the changes in the strength characteristics of steels as a result of a long-term thermal exposure, uniaxial static tension tests and microhardness measurements were conducted. The static tension tests were conducted on cylindrical specimens with the test part diameter of 3.0 mm. The tension test results were processed and the yield strength (σ_0.2) was determined in accordance with GOST 1497-84. Microhardness was measured using metallographic specimens in accordance with GOST R ISO 6507-1-2007.

To identify the embrittlement effects as a result of a EHT, impact bending tests were conducted with determining critical temperature of brittleness Т_К, that characterizes the ductile-brittle transition temperature for RPV steels. The impact bending tests were conducted on Charpy-type test specimens. The impact bending test results were processed and Т_К was determined as defined in Appendix E to GOST R 50.05.12-2018.

The qualitative analysis of sizes and number densities for the main strengthening carbide phases in the investigated steels was undertaken by transmission electron microscopy (TEM) using a Titan 80–300 high-resolution microscope (FEI, the USA), as well as using scanning electron microscopy (SEM) using a Merlin high-resolution field emission scanning electron microscope (Zeiss, Germany).

Specimens for TEM and SEM investigations were prepared by method of two-side electrochemical polishing. Thinning was done with the use of Struers Tenupol 5 device within 10% HClO₄ solution in methanol at the temperature below minus 50 °C and voltage about 20 V.

Fractographic investigation of the tested specimen fracture surface makes it possible to obtain direct information about the failure mechanisms, specifically the degree of the grain boundary embrittlement in steel, based on the change in the portion of brittle intergranular fracture (correlates with the grain boundary segregation level) (Naudin et al. 1999; Kameda and Nishiyama 2011; Shtrombakh et al. 2014). For example, for typical RPV steels the portion of brittle intergranular fracture of ≤ 5% shows that the level of the phosphorus grain boundary segregation (C_P) does not exceed 10 to 15% of the phosphorus monolayer coverage and does not have a regular effect on the shift in the critical temperature of brittleness (Naudin et al. 1999; Kameda and Nishiyama 2011; Fedotova et al. 2019).

The fractographic investigation of the Charpy specimens fracture surface after the impact bending tests was carried out using a Supra 40-VP field emission raster electron microscope (Zeiss, Germany). The secondary electron images were acquired with accelerating voltages of 10 to 20 kV in a magnification interval of ×50 to ×3000. The portion of brittle intergranular fracture was determined using the Glagolev method (Saltykov 1970).

The chemical composition of the grain boundary segregation was investigated by the Auger electron spectroscopy (AES) method on the fractures of cylindrical specimens with circular notche (diameter 3.2, height 18 mm) using an Auger Nanoprobe 700 scanning electron microscope (PHI-Ulvac, USA and Japan). The specimen fracture was obtained using a high-vacuum microscope device which was cooled together with the specimen in liquid nitrogen at a pressure ≤7·10^-10 torr. The level of the phosphorus grain boundary segregation as the percentage of the monolayer coverage (C_P) was determined on the intergranular fracture facets.

To determine the phase composition of specimens and evaluate their structural state, diffraction spectra were recorded using a D8 DISCOVER with DAVINCI diffractometer (Bruker, Germany) at the MEPhI’s Laboratory of X-Ray Texture Analysis. Co K_α radiation was used as it has less absorption coefficient in steel in comparison with widely used Cu K_α. BrukerAXS DIFFRAC.EVA v.4.2 software and international database ICDD PDF-2 were used for phase identification. Processing of experimental specter data was done by usage of Rietveld method in the software DIFFRACT.TOPAS v.5.0. The average size of the coherent scattering regions (D) and microdeformation (ε_micro) was estimated based on analyzing the intrinsic broadening of X-ray lines using expressions (Ungar 2004; Ungar et al. 2005):

β cosθ = (λ/D) + 4<ε_micro>sin θ, (2)

where β is the integral width of the diffraction maximum, λ is wavelength of the radiation used, D is the average size of the coherent scattering regions, θ is the Bragg angle of the recorded reflection, and ε_micro is microdeformation.

Fig. 1 presents values of yield strength, σ_0.2, and microhardness, HV_0.1, for the tested steels in the initial state and after a long-term thermal exposure with different parameters, as well as their comparison with the test results for the reference 15Kh2NMFA class 1 steel.

Figure 1. 

Yield strength, σ_0.2, (solid symbols) and microhardness, HV_0.1, (open symbols) compared for the tested steels in the initial state and after a long-term thermal exposure with different parameters.

It can be seen from the data presented in Fig. 1 that both for the improved 15Kh2NMFA steel and the reference 15Kh2NMFA class 1 steel there is a typical (Shtrombakh et al. 2015) for the RPV steels constancy of the σ_0.2 and HV_0.1 values even after thermal exposures with relatively high values of the parameter H_P = 17.8–18.3.

It can be also seen that there is a trend for high-nickel steels towards a minor decrease in the σ_0.2 and HV_0.1 values (by 5 to 10%) for different heats after relatively high values of the thermal exposure parameter (H_P > 18.2). For some high-nickel steel heats, no HV_0.1 reduction has been however observed. Still, on the whole, the σ_0.2 values for the high-nickel steel remain at rather a high level (> 700 MPa), exceeding greatly the typical values for the 15Kh2MFA- and 15Kh2NMFA-type steels currently used as the VVER RPV materials.

The impact bending tests were conducted for different high-nickel steel heats after a segregation provoking embrittling heat treatment (EHT), since after EHT the greatest increase of the Т_K shift are expected due to passage through the entire temperature interval with the maximal effects of the reversible temper brittleness. Table 3 presents the Т_K values for the tested steels in the initial state and after the EHT.

It can be seen from the table that there is a typical for the RPV steels (Shtrombakh et al. 2015) shift in Т_K for the 15Kh2NMFA class 1 steel after the EHT caused by the accumulation of the grain boundary segregation of phosphorus. For the improved 15Kh2NMFA steel with a higher content of Ni, as compared with the 15Kh2NMFA class 1 steel, the post-EHT Т_K value is at the initial state level. This indicates to a higher resistance of the improved 15Kh2NMFA steel against temper embrittlement, despite its increased content of Ni that usually intensify the phosphorus segregation processes and the grain boundary embrittlement respectively (Shtrombakh et al. 2015). It can be also seen that there is untypical for the RPV steels Т_K change towards lower temperatures after EHT. To clarify the reasons for the identified features of changes in the mechanical characteristics of the studied steels after EHT, a complex of structural studies was carried out.

Table 3 presents the fractographic and AES investigation results for different steels after the EHT. For the high-nickel steel, however, no results have been provided by the AES investigation due to the absence of brittle intergranular fracture regions in the AES specimens, which indicates, in turn, to a high cohesive strength of the grain boundaries.

For the improved 15Kh2NMFA steel with a higher content of Ni, as compared with the 15Kh2NMFA class 1 steel, just a slight difference has been found in the portion of brittle intergranular fracture, ν, in the initial state which agrees with the absence of a Т_K change. The absence of Т_K changes and a small portion of brittle intergranular fracture, ν, for the investigated steel was caused by a very low level of the grain boundary phosphorus segregation, C_P, not only in the initial state but also after the EHT which is not enough for the Т_K shift into higher temperatures region (Naudin et al. 1999; Kameda and Nishiyama 2011; Fedotova et al. 2019). The low value of C_P in the improved 15Kh2NMFA steel in the initial state can be explained by a four times smaller grain size and, accordingly, by extended length of grain boundaries which is expected to the distribution of the existing phosphorus through a large surface area.

For the high-nickel steel, no change in the portion of brittle intergranular fracture in the Charpy specimens was found after the EHT which agrees with the absence of a Т_K shift into the higher temperature region (see Table 3). The earlier detected (Kuleshova et al. 2022b) increase in the structural dispersion degree for the high-nickel steels with extended length of high-angle boundaries, as well as the improved purity of steel in terms of the impurity content also defines the low level of the grain boundary phosphorus segregation, C_P, and, accordingly, the absence of brittle intergranular fracture in the initial state and after the EHT.

Fig. 2 presents the characteristics (average size d and number density N) of small (MeC or Me₂C type (Kuleshova et al. 2022b) and large (Me₃C, Me₇C₃, Me₂C, NbC and other type (Kuleshova et al. 2022b) precipitate of the main strengthening carbide phases for the improved 15Kh2NMFA steel, as well as for the high-nickel steel heats with the largest effects of the HV_0.1 and σ_0.2 reduction in various states. Table 4 presents the differences in the contributions of precipitation hardening (Δσ_d), estimated taking into account the Orowan-Ashby equation (Kuleshova et al. 2022b), between the initial state and the state after a long-term thermal exposure for the tested steels.

Figure 2. 

Average size, d, and number density, N, of small nanosized (d_s, N_s) and coarse (d_c, N_c) precipitates of the carbide phases in tested steels in various states (EHT – embrittling heat treatment, IE – isothermal exposure; shown in brackets is the value of parameter H_P corresponding to thermal exposure).

It can be seen from the data in Fig. 2 and in Table 4 that both for 15Kh2NMFA-type steels and different high-nickel steel heats, there are no differences in the characteristics of the carbide phase precipitates (size and number density) in terms of precipitation hardening after a long-term thermal exposures (EHT and IE). Specifically, this is evidenced by the absence of a correlation between the differences in the contribution of precipitation hardening and by the change in the yield strength for the tested steels (see Table 4).

It should be noted that, along with carbide phases, the high-nickel steel structure includes a small fraction (1 to 2%) of residual austenite (Kuleshova et al. 2022b), transformation (dissolution) of which can affect, accordingly, the characteristics of the structure and the mechanical performance of steel (Akhavan et al. 2011). It has been found as a result of the X-ray phase investigation that the volume fraction of residual austenite in a high-nickel steel remains at the initial level (1 to 2%) after a long-term thermal exposure, which is the evidence of its stability in the thermal exposure conditions under consideration.

For the tested steels, Table 5 generalizes the integral characteristics of the structure (average size of coherent scattering regions, D, and microdeformation values, <ε_micro>), obtained as the result of analyzing the X-ray line broadening in the initial state and after a long-term thermal exposure.

It can be seen from the data presented in Table 5 that, for the high-nickel steel, a long-term thermal exposure results in an increase observed in the size of the coherent scattering regions and a decrease in the microdeformations, which is not typical so much for the improved 15Kh2NMFA steel. The key factors that affect the physical broadening of X-ray lines, which evidences the change in the size of the coherent scattering regions and value of microdeformations, are density of the substructural boundaries and characteristics of the dislocation structure (Ungar 2004; Ungar et al. 2005). A high-temperature tempering process is known to involve processes of the equilibrium substructure formation in steel, which are accompanied by the dislocation structure and structural boundary rearrangement (Markov 2012). For example, these processes for the 15Kh2NMFA steel are over at high-temperature tempering for 50 hours under 650 °C (Markov 2012), which corresponds to a thermal exposure of H_P ≈ 20 (see equation (1)). The high-temperature tempering conditions for the improved 15Kh2NMFA steel correspond to H_P ≈ 21, and parameter H_P for the high-nickel steel is smaller and is close to H_P ≈ 19 due to a lower tempering temperature. A smaller value of parameter H_P at the high-nickel steel high-temperature tempering stage indicates to a smaller intensity of the thermal effect, which could lead to the incomplete equilibrium substructure formation in the initial state and the respective yield strength reduction in the event of a further long-term thermal exposure.

At the same time, the observed reduction of Т_K for the high-nickel steel in accordance with the Ioffe scheme (Elmanov et al. 2021) can be explained by the revealed trend for a decrease in the yield strength of the high-nickel steel as a result of a long-term thermal exposure (see Fig. 1). However, no direct proportionality between the yield strength change and the shift in Т_K has been found in the study. Thus, for example, the greatest shift in Т_K is not matched by the greatest change in the yield strength. Therefore, a reduction of both σ_0.2 and Т_K for different heats of high-nickel steel may indicate that the effects in question are expectable, and the absence of proportionality between the changes in σ_0.2 and Т_K is the evidence of a major role played in these effects by the scatter in properties for the material.

It should be noted that despite the trend identified for high-nickel steel towards a minor decrease in the yield strength (by 5 to 10%) and the respective decrease in Т_K in the course of a long-term thermal exposure, high-nickel steel is characterized by much higher yield strength values as compared with traditional VVER RPV steels (see Fig. 1). The yield strength of high-nickel steel is most likely to be preserved at rather a high level (KP65+) after a long-term thermal exposure, which is achieved through the stability of the main strengthening carbide phases (see Fig. 2). And the provided estimations are relatively conservative since the expected conditions of service for the tested steels are matched by a much smaller thermal exposure (H_P < 18) and, accordingly, by a smaller decrease of σ_0.2 in the course of service. Besides, the identified decrease in the Т_K value for high-nickel steel as a result of a long-term thermal exposure just contributes to the increased brittle fracture resistance of steel in the process of service as part of the RPV thanks to the potential compensation of the irradiation embrittlement that leads to the increase of Т_K typical of traditional RPV steels of the 15Kh2MFA and 15Kh2NMFA types. Therefore, the identified peculiarities of the improved 15Kh2NMFA and high-nickel steel behavior during a long-term thermal exposure create opportunities for these steel to be used as the RPV materials in advanced VVER reactors of different designs.

However, in the process of the reactor operation, apart from thermal effects, the RPV material is also subjected to combined effects of neutron irradiation and mechanical loads, which may cause the steel structure and properties to degrade. In this connection, confirming the possibility of using the improved 15Kh2NMFA steel and high-nickel steel as the advanced VVER RPV material requires undertaking complex structural studies and mechanical tests, including for commercial metals, which will be done in future.