The Microstructure and Tensile Properties of New High-Manganese Low-Activation Austenitic Steel

Abstract

Using X-ray diffraction, scanning and transmission electron microscopy, the microstructure of a new low-activation chromium-manganese austenitic steel with a high content of manganese and strong carbide-forming elements is studied. Its structure, dislocation character and particle composition are detailed. The processes taking place in the steel under cold-rolling deformation are described. It is shown that the mechanical properties of the new high-manganese steel revealed by testing at 20 and 650 °C are comparable with those of well-known analogs or exceed them. Relying on the structural studies, this is attributed to the dispersion and substructural strengthening. Better plastic properties of the steel are associated with the twinning-induced plasticity effect. It is shown that the steel fracture after tension at the test temperatures is mainly ductile dimple transcrystalline with the elements of ductile intercrystalline fracture (at 20 °C), while at 650 °C the signs of the latter disappear. The low-activation chromium-manganese austenitic steels characterized by increased austenite stability are thought to be promising structural materials for nuclear power engineering.

1. Introduction

Austenitic steels are among the most common structural materials for nuclear power engineering. Chromium-nickel austenitic steels are currently used as fuel element cladding and other structural parts of nuclear reactors. They have a number of attractive properties—increased heat resistance, resistance to radiation and helium embrittlement (up to radiation doses of about 100 dpa), high corrosion resistance, lack of viscous brittle transition at low temperatures, etc. [1,2,3,4,5]. However, there are some disadvantages such as the appreciable radiation swelling (more than 6%) and the tendency to the helium embrittlement at radiation doses above 100 dpa. Another problem with these steels is their high activation after prolonged operation under irradiation conditions in a nuclear reactor. A high level of induced radioactivity will persist for more than 1000 years. This is determined by the presence of such highly activated elements as Ni, Nb, Mo, and Co in the steels composition.

Generation IV nuclear reactors, which are designed for a wider temperature range and longer fuel campaigns (compared to Gen. III reactors) and higher radiation doses (more than 110 dpa) need new structural materials with higher resistance to heat, radiation and radiation swelling. The concept of low-activation structural materials [5,6,7], i.e., materials for which the induced radioactivity drops over 50–100 years to a level that allows their safe processing, significantly limits the possibility of using alloying elements in steels. It implies that Fe, Mn, V, W, Ti, Ta, Zr, and C are acceptable low-activation elements, while the content of Ni, Cu, Nb, Mo, Co, Al, N, and O should be reduced to the minimum possible values.

Promising low-activation structural materials for nuclear reactors are 9–12% Cr ferritic-martensitic steels of the EUROFER, F82H, CLAM, RUSFER types [1,2,3,8,9] and chromium-manganese (Ni free) austenitic steels of the Fe-12Cr-(20-25)Mn system [5,6,7,10,11,12]. Significant progress has been made to date in the development and comprehensive study of the microstructure and mechanical properties of ferritic-martensitic steels [1,2,3]. It has been shown that these steels have significantly lower values of radiation swelling (including high radiation doses) compared to the chromium-nickel austenitic steels and a number of other attractive properties [1,2,3,9,10]. The development of low-activation chromium-manganese austenitic steels was started at the end of the 20th century and continued in the early 2000s [5,6,7,10,11,12,13,14]. In [5,6,7,10,11,12,13], an advantage of these steels over the chromium-nickel steels was shown in terms of their short-term and long-term mechanical properties. Later, however, the interest decreased and the studies dealing with the topic were not so numerous [15,16,17]. It should also be noted that an important merit of the austenitic steels as fcc materials is the absence of a tendency to low temperature embrittlement in contrast to bcc materials.

The experiments on irradiation of low-activation chromium-manganese austenite [13] revealed a tendency towards pore formation due to the radiation-stimulated γ-α transformation, which is largely determined by the austenite stability. In these steels, the latter can be enhanced by increasing the content of Mn and C and by limiting the content of Cr and other ferrite-stabilizing elements. It should be underlined that when designing the compositions of such steels, it is necessary to avoid the formation of the σ-phase (Fe-Cr), which can precipitate during heat treatment and during long-term holding in certain temperature ranges and contribute to the loss of steel ductility [5,12,13].

It is known [4,18,19,20,21] that the resistance of the austenitic steels to radiation swelling can be improved by forming a high density of defects, twin boundaries and disperse particles. A large number of the latter can be achieved by aging or thermomechanical treatments in the case where sufficient amounts of carbon and carbide formers are available in the solid solution.

The aim of this work is to develop a new low-activation austenitic steel with a high content of manganese and carbide-forming elements and to study the microstructure, mechanical properties, and fracture after its solution treatment and cold rolling.

2. Materials and Methods

A 2.5-kg steel ingot of the following composition: Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C, wt. %, was melted in a vacuum induction furnace. The target manganese content of the steel was increased compared to known analogues [5,6,7,10,11,12,13], and the carbon content was near the upper limit for known compositions to improve the austenite stability. The content of the elements with a high tendency to carbide formation (Ti, V, Zr, Ta and W) was increased. To meet the low-activation requirement, the content of such high-activation elements as Ni, Cu, Nb, Mo, Co, Al was minimized.

The elemental composition was studied by a Shimadzu LAB CENTER XRF-1800 X-ray fluorescence spectrometer (Shimadzu, Kyoto, Japan), in an ISKROLINE 300 K optical emission spectrometer (ISKROLINE, St. Petersburg, Russia), and in an Octane Elect Super (EDAX, Mahwah, New Jersey, United States) X-ray microanalysis attachment to an Apreo 2 S Scanning Electron Microscope (SEM, Thermo Fisher Scientific, Waltham, MA, USA). After melting, the ingot was homogenized at 1100 °C for 2 h followed by hot forging and hot rolling at 1100 °C. Then it was solution treated (ST) at 1100 °C for 1 h followed by air cooling. Multi-pass cold rolling (CR) was performed on a 2-roll mill at T = 22 °C to the strain degree of ε = 20%. The microstructure and the elemental and phase compositions of the steel were studied after its ST and CR treatments.

The phase content of the steel was determined by X-ray diffraction (XRD) performed in a DRON-7 X-ray diffractometer (Bourevestnik JSC, St. Petersburg, Russia) using Co Kα radiation. The Bragg-Brentano symmetric geometry was used in an angular range of 2θ = 25–115° with a scanning step of 0.05°. The exposure time at each point was 10 s. The X-ray diffractograms were processed by the method of full-profile analysis (Rietveld method).

The microstructural characterization was performed using an Apreo 2 S SEM (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a Field Emission Gun (FEG) and a Tescan MIRA 3 LMU FEG SEM (TESCAN ORSAY HOLDING, Brno, Czech Republic). A Pegasus Integrated EDS-EBSD system with the Octane Elect Super and Velocity Super detectors (EDAX) were used with the Apreo 2 S SEM. An Oxford Instruments Nordlys F EBSD detector (Oxford Instruments, High Wycombe, UK) and an energy dispersive X-ray (EDX) microanalysis system with an Ultim MAX 40 detector was used for the EBSD and EDX investigations with the Tescan MIRA 3 LMU SEM.

The samples for the SEM examination were prepared by mechanical polishing followed by ion milling as a final polishing step. The latter was performed in a Technoorg Linda SEMPrep 2 system (Technoorg Linda Co. Ltd, Budapest, Hungary). The EBSD data were obtained with step sizes of 2 µm for the area of 1 mm², 1 µm for the area of 3600 µm² and 0.1 µm for the area of 30 × 20 µm². The microstructure analysis was carried out by the Oxford Instruments AZtec software (version 3.1, Oxford Instruments, High Wycombe, UK) using the results obtained on the Tescan MIRA 3 and by the EDAX OIM Analysis software (version 8.6) and EDAX Apex software (version 2.1) using the results obtained on the Apreo 2 S.

The boundaries with the misorientation angles of 2° < θ ≤ 10° and θ > 10° were taken to be low- (LABs) and high-angle boundaries (HABs), respectively. The boundaries with a misorientation of ˂2° were neglected. The average grain size was taken as the equivalent circle diameter of a region with the same orientation and HABs.

The studies by Transmission Electron Microscopy (TEM) were conducted using a JEOL JEM-2100 electron microscope (JEOL Ltd., Akishimam Tokio, Japan) at an accelerating voltage of 200 kV in the TEM and Scanning Transmission Electron Microscopy (STEM) modes. Thin foils were prepared by electropolishing in an electrolyte containing 450 mL of orthophosphoric acid and 50 g of chromic anhydride. The dislocation density was calculated by the linear intercept method from bright-field TEM images. The average transverse dimensions of the twins and the particle sizes (diameter) were also estimated. The EBSD and TEM images after CR were taken in the longitudinal section.

The mechanical tension tests were carried at a strain rate of ≈2 × 10⁻³ s⁻¹ at room temperature (20 °C) and at 650 °C (close to the maximum nuclear reactor operating temperature) using dog-bone samples with the gage length of 13 mm and the gage section of 2 × 1 mm². The samples were cut from the section parallel to the rolling plane and subjected to tension in the rolling direction. The high temperature tensile tests were carried out in a vacuum of ≈7 × 10⁻³ Pa. The fracture surfaces of the samples after the tensile tests were studied by the Apreo 2 S SEM.

3. Results

3.1. The Elemental and Phase Composition of Steel

The elemental composition of the new steel determined by the above methods is presented in

Table 1. Elemental composition of new high-manganese austenitic steel (wt. %, base Fe).

Cr	Mn	Ni	Cu	W	V	Ti	Ta	Zr	Nb	Mo	Co	Si	S	P	C	B
11.9	28.7	0.06	0.08	0.9	0.4	0.11	0.2	0.03	0.01	<0.01	0.02	0.5	0.013	<0.01	0.26	0.004

. The data demonstrate that the content of high-activation elements (Ni, Cu, Nb, Mo, Co) in the steel is minimal and there are no such elements as Al, N, O. Thus, the resulting composition satisfies the low-activation requirements.

The XRD results (Figure 1) showed that the new steel in both ST and CR states represents a single-phase austenite. This confirms the austenite stability to γ-α and γ-ε transformations under the studied deformation conditions. No carbide and any other phases were detected by this method, which is most likely due to their low volume fractions. The lattice parameters, coherent-scattering regions and microdistortions (Δd/d) for the ST steel are 0.3623 ± 0.00005 nm, 43 ± 25 nm and 3.2 × 10⁻³, respectively. After cold rolling, these parameters change insignificantly: 0.36221 ± 0.00003 nm, 50 ± 35 nm and 4.0 × 10⁻³, respectively. A higher value of the lattice parameter of the new steel, in comparison with the known austenitic steels, is probably due to a high content of the alloying elements in the austenite solid solution, primarily Mn and C. It was shown [22] that the austenite lattice parameter increases as a function of the manganese content. An XRD pattern taken after cold rolling demonstrtaes that the (220)γ line intensity increases while that of the (111)γ line decreases, which indicates the formation of a rolling texture.

3.2. Microstructural Investigation

Figure 2 shows the EBSD orientation map of the new steel in an ST state and its statistical analysis. In this state, the austenite grains are observed ranging in size from a few to 136 µm with an average grain size of 25.6 µm (Figure 2b). The orientation map (Figure 2a) shows the elongated grains corresponding to the annealing twins. This is confirmed by the presence of a peak near 60° in the distribution of boundaries over the misorientation angles (Figure 2c), corresponding to the twin boundaries (special boundaries ∑3). This distribution also shows a peak of lower intensity near 39°, corresponding to the secondary twins (special boundaries ∑11). It should be noted that in this state, an increased density of LABs (dislocation substructures) is observed only in individual grains (Figure 2a). The presence of relatively small (a few micrometers) grains in the corresponding grain size distribution (Figure 2b) may be due to the presence of thin annealing twins and relatively coarse carbide particles (discussed in detail below).

The SEM method reveals a significant amount of dispersed particles (Figure 3) on the polished surface of the steel in the state after ST, which are located both along the boundaries and in the body of the grains. The particle size (diameter) varies from 0.05 to 5.2 µm, with an average value of 1.1 µm. The fraction of particles (estimated as the ratio of the area occupied by the particles to the area of the entire surface under study), calculated from SEM images, is 0.012. An examination of the elemental composition of some particles (points 1 and 2 in Figure 3b) showed that they are enriched in Ta, Ti, W, Zr, and V. For comparison, in the matrix (point 3 in Figure 3b) the content of these elements is much lower and close to the composition presented in Table 1.

The SEM EBSD studies of the CR steel microstructure showed that the austenite grains after this treatment are elongated in the rolling direction. In many grains, a significant number of LABs are formed (Figure 4a). An increased density of dislocation substructures can also be seen in the geometrically necessary dislocation maps (GND, Figure 4b).

On the magnified orientation and GND maps (Figure 4c), the deformation twins and dispersed carbide phase particles are visible, near which the GND density increases, compared with the neighboring areas of the material. From the analysis of grain size distribution (Figure 4d), it follows that as a result of cold rolling, new fine grains are formed, and the average grain size decreases to 15 µm. On the histogram of the distribution of grains over misorientations, the fraction of LABs significantly increases (compared to the ST state, Figure 2b). This is due to the intensive formation of dislocation substructures during deformation. Despite the decrease in the twin boundary fraction relative to the ST state, the distribution exhibits a peak near 60° associated with the twin boundaries. In this case, under the deformation conditions, the annealing twins are partially retained and the deformation twins are formed. However, a significant part of the deformation twins is not indexed by the EBSD method at the used scanning step due to their small (tens to hundreds of nm) thickness.

The TEM studies of the ST steel revealed flat dislocation pileups, dispersed particles (Figure 5a) and microtwins (Figure 5b). An estimation of the average dislocation density from the TEM images yielded the values of 3.3 × 10¹⁴ m⁻², while in some areas it can reach 2 × 10¹⁵ m⁻². The dispersed particles observed by TEM have the sizes from tens of nanometers to several microns, while the nanoscale particles pin the dislocation substructure of the steel. The presence of an increased dislocation density and the formation of microtwins in the steel structure in the initial (undeformed) state may be due to quenching stresses—local internal stresses that occur during rapid cooling from the solution treatment temperature near the dispersed particles. According to the estimates carried out in the JMatPro program [23], the stacking fault energy (SFE) of the new steel is 25 mJ/m². Such SFE values provide low twin shear stresses, which contributes to the formation of twins already in the quenched state.

In the CR structural state, a high density of microtwins is shown by TEM studies (Figure 6). The selected area electron diffraction (SAED) pattern (Figure 6b) shows the twin reflections and rod-like reflections characteristic of thin twin plates. The dark-field (DF) images in the twin (Figure 6c) and matrix reflections (Figure 6d) confirm the presence of twins, and the average width of the twin plates is about 50 nm. In addition to microtwins, there are dispersed particles and dislocation substructure clearly visible on the BF and DF images. The dislocation density as a result of CR increases (relative to the ST state) and in some areas can reach 10¹⁶ m⁻², in which case no individual dislocations are any longer resolved.

A more detailed study of the dispersed particles in the ST and CR states was provided by the STEM EDX methods. It was shown that the particles lying along the grain boundaries have preferentially elongated shapes (Figure 7a), while the particles in the grain body have rounded shapes (Figure 7a,d). The latter are enriched in Ti, Ta, W and Si (Figure 7c). The elongated particles (Figure 7a) are enriched in Cr. The SAED pattern (Figure 7b) shows a series of reflections belonging to the M₂₃C₆ carbide (M = Fe, Cr). Similar particles lying along the grain boundaries were previously shown in [7,24]. A large number of such particles located at the grain boundaries can lead to a “wavy” shape and faceting of the boundaries [7]. The rounded particles (Figure 7a,d) and the particles in Figure 3b are the complex-alloyed fcc carbides of the MC-type (M = Ta, Ti, W, V, Si and Zr). This is confirmed by the SAED patterns (Figure 7e) and EDX maps (Figure 7f).

There are also individual coarse (micron-sized) particles (Figure 8). The elemental analysis along the line passing through one of these particles, the SAED patterns and the elemental map confirm its belonging to the M₂₃C₆ carbide. Figure 8 shows that under the CR conditions, a local stress state is formed in the matrix near an M₂₃C₆ coarse non-deformable particle. The azimuthal significant “twisting” of the matrix reflections (Figure 8b), the formation of numerous thin extinction contours in the matrix, and the increased dislocations density near the particle (Figure 8a) provide an evidence of this fact.

3.3. Mechanical Properties

The stress-strain curves from the Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C steel obtained during tensile tests at 20 and 650 °C are shown in Figure 9. The average values of yield strength, tensile strength and elongation to failure of the new steel compared to the known data for the reactor chromium-nickel steel (Fe–16Cr–19Ni–2Mo–2Mn–Nb–Ti–B) [25] and other low-activation chromium-manganese steels [7,12] are presented in

Table 2. Mechanical Properties of Steels.

Steel/Treatment	Tensile Tests Temperature, °C	Yield Strength, MPa	Tensile Strength, MPa	Elongation to Failure, %
Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C, ST	20	331	662	75.5
	650	95	353	42
Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C, CR	20	750	894	35.7
	650	389	495	14.3
Fe-16Cr-19Ni-2Mo-2Mn-Nb-Ti-B, ST [23]	20	201	539	47.4
	650	95	360	31
Fe–20Mn–12Cr–W–0.25C ST [12]	20	249	687	50.2
	600	115	318	38.9
Fe–20Mn–12Cr–Ti–W–V–P–B, ST [7]	20	304	915	54.9
	600	100	400	39
Fe–20Mn–12Cr–Ti–W–V–P–B, CR [7]	20	915	1114	11.3
	600	500	620	8

. From the analysis of these data, it follows that for the new steel in the ST state, the yield strength values are higher than 330 MPa and elongation to failure are higher than 75%, which exceeds the corresponding values for other steels. However, the tensile strength values for the Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C steel in the ST state are close to those obtained in [12] for the Fe–20Mn–12Cr–W–0.25C steel and somewhat lower than those for the Fe–20Mn–12Cr–Ti–W–V–P–B steel [7].

At the elevated tensile temperature (650 °C), the yield strength and tensile strength values of the Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C steel in the ST state are comparable with those of the analogues, while elongation to failure is higher than in the analogues (Table 2).

The flow curve at 650 °C (Figure 9) exhibits the characteristic features of a serrated flow, which is due to the dynamic strain aging effects. At the same time, carbon and the elements with a high tendency to carbide formation (Ti, V, Ta) present in the steel composition can form atmospheres and segregations on the dislocation substructures. The separation of dislocations from the atmospheres and segregations under the conditions of elevated deformation temperatures can cause the formation of characteristic teeth on the flow curves [26].

Cold rolling of the steel leads to an increase in its strength properties at 20 and 650 °C (Figure 9, Table 2) compared to the ST state. In this case, the plasticity of the material decreases. The most significant reduction in the elongation to failure is observed at 650 °C.

It should be noted that in the CR state, the flow curve at 650 °C does not have any serrated flow pecularities, which may be due to the features of the structural state formed under the cold rolling conditions. An increased density of dislocations and deformation microtwins can suppress the dynamic strain aging, therefore no serrated flow effects are visible on the curves.

3.4. Fractography

Figure 10 shows the fractograms of the tensile samples fracture surfaces of the ST and CR steel samples tested at 20 and 650 °C. As can be seen from this figure, both the treatment mode and the test temperature have a noticeable effect on the fracture type.

At room temperature, the test samples fracture mainly by the mechanism of ductile dimple transcrystalline fracture (Figure 10a,c). In addition, there are the areas of ductile intercrystalline fracture. The differences between the treatments are in the number of such areas. In the case of the ST samples, there are numerous areas of intercrystalline fracture in the zone of radial crack propagation (Figure 10a). The dimensions of steel structural elements outlined by the cracks in Figure 10a are comparable to the dimensions of the austenite grains obtained by the EBSD studies, while the fracture surface of the CR samples contains only single canyons formed along the boundaries of the steel structural elements (Figure 10c).

It is commonly believed [27] that intercrystalline fracture contributes to a decrease in the material ductility. However, as shown above (Section 3.3), the initial steel (ST) ductility is ≈2 times higher compared to the state after cold rolling (CR). The presence of numerous slip lines found on the walls of the canyons of intercrystalline fracture of the samples after ST (Figure 10a), in our opinion, indicates the plastic strain localization followed by the formation of necks between the individual austenite grains. This is one of the factors responsible for the higher plasticity values compared to the CR state. Possible reasons for this behavior are the segregation of impurity atoms and/or the precipitation of second phase particles at the austenite grains boundaries in the ST samples.

As the test temperature is increased to 650 °C, the signs of intercrystalline fracture of the steel after both treatment modes disappear (Figure 10b,d). Numerous ductile fracture dimples are observed on the sample fracture surfaces (Figure 10b,d and Figure 11c). Most of the dimples are submicron in size. In addition to them, there are much larger dimples up to several tens of micrometers in diameter. It should be noted that at room test temperature such bimodality in the dimple size distribution is less pronounced.

On the fractograms from all of the studied samples at the bottom of the dimples of ductile fracture, numerous particles of various carbide phases differing in shape and size were found (Figure 11a,c). The sizes of the smallest particles are ≈100 nm, the largest ones can reach several (3–5) micrometers. The diameter of the dimples is largely determined by the size of these particles and the distance between them.

Carbide precipitates are stress concentrators, and they contribute to the formation of discontinuities (micropores) at the particle/matrix interface during tensile tests. Due to the increased hardness and brittleness, some particles (for example, the Ta-based particle marked 3 in Figure 11a) fracture under the applied stresses.

Using the EDX analysis, it was established (Figure 11b,d) that 4 types of particles are present in the steel fractures—particles enriched in Ti, Ta, Zr and Cr, respectively. Apparently, the first three particle types are MC-carbides based on the above-listed elements (regions 1, 3, 4 in Figure 11b,d). The fourth type corresponds to the Cr-based carbides (region 2 in Figure 11b). These results are consistent with the data of a SEM EDX analysis of the steel sample polished surfaces and the TEM data indicating the formation of the MC and M₂₃C₆ carbides.

4. Discussion

In this paper, a new low-activation chromium-manganese austenitic steel with an increased manganese content compared to known analogues [5,6,7,10,11,12,13] is investigated. The carbon content in it is 0.26%, which is close to the maximum values for similar steels. The content of strong carbide-forming elements (Ti, V, Ta, W, and Zr) is higher than in the analogues. These differences in the elemental composition predetermine the structure and phase state of the steel and its mechanical behavior (including high-temperature performance).

Primarily, an increased Mn content and a sufficiently high C content increase the austenite stability. According to the Scheffler’s diagram [11,16] in its modified form for high manganese (Ni free) steels, the Ni equivalent = 22.2, and the Cr equivalent = 12.7. This determines the position of the studied composition in the single-phase austenitic region sufficiently far from the boundaries of both martensitic and ferritic regions. The increased austenite stability is confirmed by the obtained experimental results. Neither after the ST nor CR treatments there are any phases except for the austenite and an insignificant amount of the MC and M₂₃C₆ carbides.

As noted in [13], the tendency of steels to radiation swelling under irradiation conditions largely depends on the austenite stability, therefore, a more stable chromium-manganese austenite may have lower values of radiation swelling compared to the known analogues; however, additional studies are needed to confirm this assumption. The calculations of nuclear-physical characteristics [28] showed that a stable chromium-manganese austenite may have a lower tendency to helium embrittlement compared to the stable chromium-nickel austenite in the Fe–16Cr–19Ni–2Mo–2Mn–Nb–Ti–B type steel used as an advanced structural material for the fuel cladding shells in the reactors in Russia. It should also be noted that the increased stability of austenite makes it possible to consider the steel studied in this work as a promising structural material for nuclear fusion reactors.

Low-activation austenitic steels are generally studied after their ST and (or) CR treatments [6,7,10,11,12,13]. In contrast to the low-activation ferritic-martensitic steels [1,2,3,8,9,29,30,31], no heat or special thermomechanical treatments are applied to precipitate the disperse particles. In the present work an increased content of strong carbide-forming elements with a carbon content of 0.26% resulted in the formation of complexly alloyed dispersed particles of the MC type (M = Ti, Ta, V, W, Zr and Si) with a volume fraction of ≈1.2% with a mean size of 1.1 µm and M₂₃C₆ particles in the steel. Note that the volume fraction could be further increased by special treatments (aging or thermomechanical treatments), favoring a precipitation of new nanosized particles. It was shown earlier [18,19,29,30,31], that nanoscale particles pin the dislocation substructure of steels and contribute to an increase in their strength properties by increasing the efficiency of dispersion and substructural strengthening. A higher volume fraction of nanoscale particles can have a positive effect on the steel resistance to radiation swelling under irradiation conditions [18,19]. The effects of special treatments and irradiation will be studied in the near future.

The strength properties of the steel (Table 2) are comparable with those of the known analogues or exceed them [7,12] with the exception of Fe-20Mn–12Cr–Ti–W–V–P–B complexly alloyed steel subjected to CR [7]. At the same time, the plastic properties of the new steel are 1.5–3 times higher than those of the known analogues (Table 2). Based on the structural studies, the advanced strength and plastic properties of the new steel have been attributed to a higher efficiency of dispersion and substructural strengthening and a high (29%) content of manganese. The latter provides a sufficiently low stacking fault energy (25 mJ/m²) and contributes to the twinning-induced plasticity effect under tension at 20 °C.

The fracture mechanism of the steel is structure sensitive and is controlled by the treatment mode (ST or CR), on the one hand, and by the tensile test temperature (20 and 650 °C), on the other hand. At room temperature tests, it is the ductile intercrystalline fracture (along with the transgranular ductile dimple fracture) which is likely to favor a higher plasticity of the ST steel over that of the CR steel. In the former, relatively soft/plastic austenite grains seem to be more easily deformed than the grain boundaries strengthened by the second-phase particles and impurity atom segregations. During tensile tests at 20 °C, this provides a higher stress concentration near the grain boundaries and an intensive formation of intercrystalline fracture canyons. After CR, due to the formation of a highly defective twin microstructure significantly (more than 2 times) increasing the steel strength properties and decreasing its plasticity compared to the ST treated steel. The number of stress concentrators inside the matrix grains becomes larger and the stress distributions along the grain boundaries and in the grain bulk become comparable.

5. Conclusions

A new low–activation chromium–manganese austenitic Fe–29Mn–12Cr–W–Si–Ta–Ti–V–Zr–0.25C steel with an improved austenite stability and dispersed particles of the complexly alloyed MC carbides (M = Ti, Ta, V, W, Zr and Si) and M₂₃C₆ (M = Fe, Cr, Mn) carbides has been designed. Its microstructure and mechanical properties after solid solution treatment and cold rolling have been investigated. The main results are the following:

1. Due to the low stacking fault energy, the solution-treated steel microstructure is characterized by flat dislocation pileups and microtwins. Cold deformation promotes the intensive mechanical twinning and the formation of dislocation substructures. Near the dispersed particles, the dislocation density increases both as a result of quenching after solution treatment and after cold rolling.

2. The yield strength of the new steel in the solution treated state at 20 °C exceeds that of similar steels. At elevated temperatures, the strength properties in both solution-treated and cold-rolled states are comparable to or exceed the corresponding values for the analogues. This is due to the formation of micro-twins, increased dislocations density and availability dispersed particles. The new steel at all test temperatures is characterized by higher (1.5–3 times) values of elongation to failure. At 20 °C, this is due to the twinning-induced plasticity effect.

3. At room temperature testing, the main fracture mechanism is the mechanism of ductile dimple transcrystalline fracture. There are also elements of ductile intercrystalline fracture. At 650 °C, the signs of intercrystalline fracture disappear. At the bottom of the ductile fracture dimples there are dispersed particles enriched with Ti, Ta, Zr and Cr, which are the crack-initiating stress concentrators.

4. The proposed low-activation chromium-manganese austenitic steel with increased austenite stability and dispersed carbide particles is of interest for nuclear power engineering as a material that may have a reduced tendency to radiation swelling under irradiation conditions.