Final Report Summary - GETMAT (GEn IV and Transmutation MATerials)
Executive Summary:
The objective of the GETMAT project was the investigation, both in the theoretical and experimental domains, of selected material properties that are cross-cutting among the various Generation IV and Transmutation reactor designs. Since the design requirements of the innovative reactors are beyond those for the LWR currently in operation, the selection of the properties to be investigated has been performed by identifying relevant conditions of key components as cores and primary systems. In fact, high temperature, high irradiation doses, corrosive environments as well as high thermal and mechanical loads are features that pose challenges on all materials foreseen for these components. Taking into account this framework and the goals defined by the Generation IV and the waste transmutation approaches as e.g. sustainability in terms of optimisation of use of resources and waste minimisation, high safety standards, economic viability and efficiency, it turned out that innovative materials might be a better choice with respect to conventional nuclear grade steels for the design of the core components and the primary system. Therefore, ODS alloys and 9-12 Cr Ferritic/Martensitic (F/M) steels have been selected as reference for the GETMAT project. In particular, even if ODS alloys are known since several decades, data for nuclear applications are quite scarce. The scientific results produced within the GETMAT project on 9Cr, 12Cr and 14Cr ODS alloys fabricated via the powder metallurgy route as well as the results obtained on the 9-12Cr F/M steels through an important PIE program, contribute considerably to a better understanding of the behaviour of these classes of materials for nuclear application. Moreover, the results have been found to be highly relevant for the identification of further R&D activities defined within the EERA Joint Program for Nuclear Materials (JPNM) and the related projects MATTER and MatISSE supported by the EURATOM FP7. In concordance with the experimental investigation on F/M steels and Ferritic / F/M ODS alloys the theoretical program was devoted to the study of Fe and Fe-Cr alloys. Emphasis has been put on first principle modelling studies and experiments to identify and validate physical models and mechanisms that explain irradiation hardening and embrittlement of these alloys. The results obtained are very promising and supportive in explaining phenomena occurring on Fe-Cr steels. Moreover they have been the basis for further R&D activities included in the JPNM and the MatISSE project.
The GETMAT results contribute also to the development of other technical areas as welding and joining of ODS alloys and the development of corrosion protection barriers for structural materials to be used in HLM-cooled systems.
Finally, within the GETMAT project high priority has been given to Education and Training. The scientific program of GETMAT has allowed the involvement of about 20 PhD students with experimental and theoretical work being part for their respective dissertations. Moreover, educational activities have been further supported through the organisation of two summer-schools and two dedicated workshops. The schools were organised in cooperation with other EU projects like PERFECT60 and MATTER and the workshops had a wide international participation, beyond the European boundaries. These educational activities have been very successful in terms of attraction of both well re-known lecturers in the field of materials science and of a quite significant number of young students.
Project Context and Objectives:
- Project context
The key elements at the origin of the GETMAT project have been identified through a thorough analysis performed on innovative reactors that are of relevance within the European Framework. In particular, reactor systems developed for advanced nuclear fuel cycles as well as advanced LWR were essential. Advanced nuclear fuel cycles include technological approaches that allow an optimised use of resources and a safe and optimised radioactive waste management. Nuclear reactors that are able to respond to these objectives are those with a fast neutron spectrum. Two categories of fast neutron reactors have been gaining attention in Europe, i.e. the sub-critical accelerator driven system (ADS) and the fast reactors (FR). Moreover, Supercritical Light Water Reactors (SCWR), considered as technological advancement of current LWR technology, were included in the analysis. A number of EURATOM supported projects (e.g. EUROTRANS, ELSY/LEADER, ESFR, GoFastR, HPLWR) that were addressing the design of these reactor systems and related materials as well as technological issues have been included and evaluated. The analysis was conducted in order to identify key reactor components for which advancement in the area of nuclear materials resulted to be crucial for the envisaged innovative design solutions. It turned out that the cores and primary systems are the most challenging components for what concerns materials requirements, thus materials selection and qualification. In general, all innovative reactor concepts foresee operational temperatures which are higher with respect to the temperature ranges of LWR (the current technology). Therefore, it can be assumed that all components of these reactors would be exposed to more severe conditions as compared to the established technology. However, the highest temperatures as well as dose rates can be expected at core components and primary systems levels. Therefore, the focus of the GETMAT project has been on these components. The analysis has also shown that candidate materials for cores and primary systems of almost all reactor concepts considered were austenitic and 9-12Cr Ferritic/Martensitic (F/M) steels. However, limits of these two classes of steels could have been identified. These limits impose design constraints that might not be in line with the original objectives declared for an enhanced sustainability and for innovations. Exemplary of these limits and their consequences especially for the fast neutron systems is the selection of the fuel clad materials. The austenitic steel option imposes an upper fuel burn-up limit due to irradiation swelling and embrittlement phenomena that occur on this class of steel for high irradiation doses (as discussed in past FR development programs). The limitation of the fuel burn-up might considerably impact strategies related to sustainability as e.g. the optimisation of use of resources. Limits can be as well identified for the 9-12Cr F/M steels as clad material. Indeed, the temperature window of these steels in an irradiation environment might be limited in the lower range to temperatures above 350°C and in the upper range to temperature below 550°C. The lower limit is due to irradiation hardening and embrittlement phenomena that can occur below 350°C (as discussed e.g. in the SPIRE project) and the upper limit is due to poor thermal creep and stress-to-rupture resistance of these steels above 550°C (as discussed e.g. in the EUROTRANS project). These temperature limits might impact the realisation of innovative design solutions as well as efficiency and economic viability of the reactor systems.
Therefore, within GETMAT the possibility to use oxide dispersed strengthened (ODS) alloys with a ferritic or F/M steel matrix has been proposed as an alternative option to the conventional steels. The focus on ODS alloy was based on data reported in the scientific literature, where it is discussed that these alloys show improved irradiation swelling and improved thermal-creep resistance in the high temperature range. The envisaged applications of ODS alloys were the fuel cladding for all considered reactor systems, i.e. ADS, FR and SCWR, and some structural components of the core as e.g. core support plate. Within this context properties of the ODS alloys that are of relevance for the given applications (and for all reactor systems) have been investigated. The preliminary assessment of the ODS alloys as materials for nuclear application, has been rounded up with an investigation of possible techniques able to join/weld these materials. Moreover, specific tasks related to environmental effects (corrosion and potential mechanical properties degradation) and options for mitigation of these effects through e.g. the development of corrosion protection barrier have been addressed as well. The environmental effects have been analysed for molten lead, helium with controlled impurities and supercritical water.
The development, optimisation and qualification of ODS alloys for nuclear application is considered as a long-term activity due to the fact that knowledge and relevant data on these materials for the nuclear applications are quite scarce. Therefore, selected needs for short- and medium-term applications were taken into account as well. The short- and medium-term plans foresees the availability in Europe of prototype and demonstration FR and ADS systems. For these systems conventional steels as 9-12Cr F/M and the austenitic steels AISI 316L, 15Cr-15Ni/Ti are candidate materials for both core components and primary systems. The GETMAT contribution to the design and construction of the prototype and demonstration systems occurred through the behaviour assessment of these candidate materials in different irradiation fields and in presence of relevant environments (heavy liquid metals (HLM) Pb and Pb-Bi eutectic, sodium and gas atmosphere). The activities were defined such as to take advantage, through post irradiation examinations (PIE), of relevant available and completed irradiation campaigns that were performed in past European projects and national programs. Moreover, it was also possible to include in the project a new irradiation campaign in a fast neutron spectrum and HLM environment and the related PIE.
A further topic considered as cross-cutting for all innovative reactor systems is the physical understanding of materials behaviour in terms of microstructure and mechanical properties evolutions in irradiation, temperature and stress fields. The physical understanding of material behaviour foresees the prediction of basic mechanisms, from atomic to the microscopic level, which determine materials response to applied thermal and mechanical loads, while being exposed to neutron irradiation. This research area includes the development of both physical models and computational tools for the different scales (from atomic to macroscopic level). Moreover, the theoretical activities are supported by dedicated model experiments. Since the multiscale modelling and experimental validation of nuclear material is a quite wide area, modelling and experimental activities included in the GETMAT project have been defined on the basis of the research direction in terms of composition and properties identified for the candidate alloys and steels. Indeed, the focus of the modelling activities was on Fe and Fe-Cr alloys (representative of F/M steels) and key aspects included in the investigation were thermo-chemical data (phase diagram of Fe-Cr) and irradiation hardening and embrittlement of these alloys. The theoretical activities were accompanied by dedicated irradiation experiments (in both ion and neutron fields) to assess assumption and models.
The overall challenges and research area of the GETMAT project can be summarised as follows:
• ODS alloys procurement and their characterisation
• Assessment of joining and welding techniques for ODS alloys
• Development and definition of corrosion protection barriers
• Improvement and extension of conventional steels (9-12 Cr F/M and austenitic steels) qualification
• Improved physical understanding of Fe and Fe-Cr alloys in irradiation and temperature fields and validation experiments.
- Project objectives
To address these challenges the GETMAT project has been structured into four work-packages and for each work package specific objectives have been identified. These objectives are hereafter summarised.
WP1 - Metallurgical and mechanical behaviour
The main objective of this work package is to provide reference material and evaluate the basic physical and mechanical properties for the concepts under investigation, that is, in Gen-IV and Transmutation relevant environments.
WP2 - Materials' Compatibility with coolants
This work package is divided into three main tematic areas: compatibility with different coolants of ODS steels and reference materials, corrosion barriers development and mechanical tests of reference materials when in contact with the coolants themselves. Different coolants have been considered: Pb, LBE, impure He (VHTGFR reference atmosphere) and superctitical water.
WP3 - Irradiation behaviour of structure materials
The objective of WP3 was to enlarge the database concerning the effect of irradiation on the relevant properties (i.e. needed for the design, such as dimensional and mechanical properties) of materials selected for core/target components of future GENIV reactors and ADS. The materials of interest were primarily FM steels, in particular 9Cr steels such as T91 (9Cr1MoVNb) and EM10 (9Cr1Mo) which are reference materials for different nuclear applications, however data on Fe-Cr based ODS alloys and austenitic steels were obtained as well. These data were generated as part of Post Irradiation Experiments (PIE) performed after irradiation campaigns conducted in various radiation environments, in some cases with the investigated materials irradiated in contact with aggressive coolants. This work package was organised in six different tasks, each task being devoted to the PIE of a particular irradiation experiment (with the noteworthy exception of task 3.5 which included both the design and implementation of a new irradiation campaign called LEXURII, followed by the corresponding PIE). The main results obtained within WP3 are presented hereafter and were discussed according to the following topics:
- Behaviour of 9Cr martensitic steels, in terms of dimensional stability and evolution of mechanical properties, after high dose irradiation in a fission environment: this issue is of particular interest for the core components of future GENIV reactors such as the hexagonal duct (wrapper tube) of the fuel assembly. As part of WP3, data up to a maximum dose of 155 dpa were obtained (Boitix9 irradiation in Phénix reactor).
- Effect of irradiation in a spallation environment on the mechanical properties of martensitic steels and ODS alloys: core and target components of ADS will be irradiated with high energy protons and spallation neutrons. In such environment, the combined effects of displacement damage and transmutation-induced impurities, in particular large quantities of helium, are of special concern. Via mechanical tests and microstructural investigation performed after irradiation in the SINQ target, the behaviours of FM steels and ODS alloys were assessed following irradiation in a spallation environment to a maximum dose of about 20 dpa and an accumulated helium content of about 1800 appm.
- Combined effect of irradiation and contact with the coolant: possible synergistic effect of irradiation and embrittlement due to the coolant is a major issue, in particular for systems cooled by heavy liquid metals, and published results were so far extremely scarce. This topic was addressed via the PIE of IBIS (HFR reactor), ASTIR (BR2 reactor) and LEXURII (BOR60) which included tests performed in the liquid metal (liquid Pb-Bi or Pb) on specimens irradiated in contact with the coolant. Furthermore, the PIE of the MEGAPIE target provided unique data regarding the combined effect of irradiation in a spallation environment, in contact with flowing liquid Pb-Bi and with cyclic thermal/mechanical loading on the properties of T91 steel.
WP4 - Multiscale modelling and model experiments on irradiated Fe-Cr alloys
The objective of the WP was to develop physical models to describe the behaviour of iron-chromium alloys when subjected to thermal ageing and irradiation. These alloys were chosen as model materials for high-chromium ferritic/martensitic steels. The processes of interest were nanostructural and microchemical changes that occur under irradiation or ageing and the corresponding changes in mechanical properties, mainly hardening, which in turn causes embrittlement. The approach used was computer-based multiscale modelling (from ab initio to dislocation dynamics), integrated by the performance of specific modelling-oriented experiments. The latter provided detailed data on which models can be contrasted. They also suggested, in combination with fundamental atomistic calculations, physical mechanisms that the nanostructure and microchemical models, as well as mechanical models, needed to incorporate. The link with other work-packages is ensured first of all by the comprehension of the mechanisms that govern the changes induced by irradiation in iron-alloys with varying content of chromium, which allows experimental results on steels performed in the rest of the project, especially WP3, to be better rationalised. Moreoever, the characterisation of technological properties, such as fracture toughness, in model alloys was included in some tasks.
The ultimate objective of the WP was to understand which nanostructural features are the main cause of radiation hardening and embrittlement and to develop a physical model capable of rationalising the experimental data available on this issue. This goal was indeed reached.
Project Results:
WP1 - Metallurgical and mechanical behaviour
Fabrication of ODS alloys is performed following classical pulvi-metallurgical (PM) processes that involve different steps as mechanical alloying (MA), consolidation by HIP and/or hot extrusion, and final thermo-mechanical treatments to achieve the requirements for mechanical performance. This fabrication procedure is expensive and the final quality depends strongly on a large amount of parameters (mechanical alloying time and velocity, consolidation temperature, and final heat treatments) that gives some uncertainties corresponding to the repeatability of heats and increase the fabrication costs. Within the WP1, ODS alloys were produced by PM processes but other techniques that are bases on more conventional steel production, as melting and casting, were explored. Unfortunately, these melting techniques do no allow the production of quality ODS alloys mainly due to the agglomeration of the oxide particles.
Three ODS alloys with different Cr contents (14Cr, 12Cr and 9Cr) where produced by MA and hot extrusion, whereas the trials to produce a P91-ODS alloy consolidated by HIP do not reach the expected results, due to the porosity of the heats that leads to the cracking of the slabs during hot-rolling.
The microstructure of the hot-extruded 14Cr ODS bar is constituted by small elongated grains (cigar shape) along the extrusion direction with a preferential crystallographic orientation of the grains along <110> in ED. This anisotropic microstructure is typical of ODS alloys and affects the mechanical properties as it will be explained later on this report. Nano (Y-Ti-O) phases are uniformly distributed all over the grains and Cr-rich phase decorating some grain boundaries were also observed. The 12Cr ODS plates show a quite different microstructure, with a bimodal grain size distribution: small grains size < 1 micron heavily deformed and some elongated grains size up to 200 microns. The biggest grains have not got a preferential orientation, while the smallest grains seem to have a preferential orientation along <110> in the RD. Both the 12Cr ODS and the P91-ODS show elongated pores along the extrusion direction.
Tensile properties have been determined up to 850ºC by different labs. In general the results form different labs are consistent and the GETMAT 14Cr-ODS and 12Cr-ODS alloys show lower strength than the well-known 14YWT and 12YWT fabricated by ORNL. Interesting to note that all the GETMAT ODS alloys show a peak in the elongation at 600ºC, typical for similar ODS alloys. This enhanced ductility at 600ºC can be attributed to a change on the deformation mechanism from intraganular to intergranular mechanisms, as is reported my Praud et al, and it is also related to a inverse strain rate sensitivity showed by the 14Cr ODS alloy.
The effect of the microstural anisotropy on the tensiel strength is no so evident for temperatures up to 600ºC, for higher temperatures, the transverse orientation is weaker than the longotudinal one. The effect on ductility is larger and more pronounced for the total elongation and reduction in area values, where the loss of reduction in area for the T orientation in respect of the L is around 70% at room temperature and around 50% at 700ºC.
Impact properties of all the ODS alloys has been also obtained by testing small KSLT specimens. In general the DBTT and absorbed energy are strongly dependent to the microstructural features, showing againg the effect of the microstructual anisotropy on the mechanical properties. The observed difference in the values of absorbed energy is a consequence of the relative orientation of the notch regarding to the elongated microstructure. When weak interfaces are present parallel to the longitudinal direction of the impact test bar, the interaction between the weak interfaces and the stress field that is generated by the localized plastic constrain at the notch and/or the crack tip can cause splitting. The two basic geometries are termed ‘‘crack divider’’ and ‘‘crack arrester”, In the crack arrester geometry, the delamination is thought to relax the triaxial stress conditions and to blunt the crack tip. To fracture a material, crack re-initiation is necessary and occurs under conditions of nearly uniaxial tension, which is an unfavourable cleavage. Hence, high absorbed energy is obtained through the delamination in LS and TS samples by the well-known delamination toughening mechanisms. The P-u curves obtained from instrumented Charpy tests are very useful to investigate the effect of splitting on the DBTT. Meanwhile the P-u curves of LT and TL samples present similar behaviour regardless of the test temperature; P-u curves for LS and TS samples present a peculiar behavior. For LT and TL samples, at which ductile fracture surfaces were observed, P gradually decreases after a maximum value Pm is reached as expected. On the other hand, for LS and TS samples, at which extensive delamination was observed, P-u curves exhibit load drops that can be associated to crack arrest, and loads plateaus that can be due to re-initiation of new cracks. In other words, the crack is blunted by delamination. As a result, the absorbed energy becomes larger by the delamination effectct. Moreover, because of the blunted crack by delamination, the sample behaves as an unnotched one during the test.
This anisotropy on the mechanical properties is also observed on the fracture toughness tests, where the TL specimens show very low fracture toughness values at 550ºC and 650ºC.
The issue of the low impact and fracture toughness resistance of the ODS alloys do not restrict their use at high temperature, because these properties are not used on the design of high temperature components, but from a safety point of view, it has to be resolved to assure a safety operation at lower temperature close to room temperature.
High temperature behaviour of the 14Cr and 12Cr ODS alloys have also charactericed by uniaxial creep and low cycle under fully reversed conditions (R = -1) in strain control with a strain rate of 0.2%/s in air and in inert atmosphere. Small punch creep tests were also performed.
The creep curves for the GETMAT 14Cr ODS steel are characterized by very slow secondary creep and sudden rupture at very low deformation with little evidence of tertiary creep. Similar observations have also been reported for other ferritic ODS steels. For a given stress level, the rupture times obtained from the creep tests were found to exhibit considerable scatter. When a test was repeated with the same or nearly the same load, a rupture time differing from the original test by more than one order of magnitude could be recorded. The large scatter in the rupture times is not so much caused by the stochastic nature of a brittle final fracture event, as the documented absence of tertiary creep might suggest, but rather by scatter already in the steady state creep rates.
Fatigue tests have been carried out at 650 ºC and 750 ºC. The alloy is cyclically very stable with practically no hardening/softening. Results from the tests at both temperatures can be described by a common power law. An increase in the test temperature has little influence on the fatigue ductility exponent. For a given total strain level, the fatigue life of the alloy is reduced with increasing temperature. Creep fatigue tests (with 10 min holdtime) show smaller number of cycles to crack initiation, there is nearly no difference between results of tests carried out under inert and air environment.
Irradiation creep and microstructural changes of the 12Cr-ODS and 14Cr-ODS steels have been studied by homogeneously implantation with helium under uniaxial tensile stresses from 40 to 300 MPa. The maximum dose was about 1.2 dpa (5000appm-He) with displacement damage rates of 1 x 10-5 dpa/s at a temperature of 300 ºC. Irradiation creep compliances were measured to be 4.0 x 10-6 dpa-1 MPa-1 and 10 x 10-6 dpa-1 MPa-1 for 12 and 14Cr ODS, respectively. No remarkable effects of Cr content, grain size and dispersoid size on irradiation creep properties are observed. The apparently slightly higher compliance value of the 14Cr ODS cannot be unambiguously related to any material property or microstructur al evolution.
Microstructural evolution was studied in detail by TEM observations. Dislocation loops and helium bubbles are distributed homogenously in the matrix. In the case of high density fine dispersoids, most bubbles are attached to ODS particles. This may suppress loop formation as well as growth of bubbles, thereby increasing the resistance of ODS ferritic steels against helium embrittlement.
Different welding techniques have been explored within GETMAT project: electro magentique pulse technique, diffusion bonding, explosive bounding and friction stir welding. Electron beam and TIG welding techniques have also used.
Diffusion bonding has been performed with PM2000. Hardness measurements of heat treated samples of PM2000 show, that heat treatment above a temperature of 1200 ºC causes a degradation of mechanical properties. TEM investigations as well as light microscopy reveal, that neither a constant particle size distribution nor the changes in the grain sizes of small grains can be identified as a clear reason for the temperature induced loss in hardness. The diffusion bonding model of Hill and Wallach was adapted for ODS materials by using the creep equation for ODS materials introduced by Rösler and Arzt. The influence of pressure, temperature and surface roughness on the bonding time were investigated. For a diffusion temperature of 1100 ºC, a pressure of 50 MPa and a surface roughness of b = 2.5 lm and h = 1.2 lm a similar UTS and Charpy impact energies compared to the bulk material were achieved.
Electro magnetic pulse technique has been used to weld T91 cladding tubes. The sample consists of the cladding tube and an end cap of a specific design, which is entered in the tube. In particular, the angle of collision between the moving component (the tube) and the static central end cap is considered carefully. This angle has a strong influence (supplier's information) on the interaction between the components during the deformation but it also determines the flight distance of the tube prior to collision. This distance is critical in allowing an optimization of the acceleration achieved by the tube as a result of the electromagnetic pulse. The objective of defining a parameter window for the joining of advanced cladding materials (T91 and ODS steel) for Gen IV reactors has been achieved. Future tests will be performed to close in on the optimum conditions. Due attention can be given to a more precise angle of collision and surface roughness, as well as the investigation of the effects of the process on the material's microstructure and properties.
Linear Friction Welding (LFW) is a solid state joining process in which a stationary material part is forced against a part that is linear oscillating to generate friction heat. Important parameters are oscillation frequency and amplitude, friction and forge pressure and burn-off. The joining of 12 Cr ODS steel by LFW has been successfully demonstrated. A clear thermo-mechanically affected zone (TMAZ) with deformed and more refined microstructure containing uniform dispersion of particles was observed in both longitudinal and transverse directions. The micro hardness changes from 370 Hv in the parent material to 330 HV in the TMAZ are observed. The minimum hardness values, 170-190 Hv, occur very close to the weld interface (probably due to dynamic recrystallization).
Explosive Welding (EW) is a solid state process where welding is accomplished by accelerating one of the materials parts at extremely high velocity through the use of chemical explosives. Final explosive welds of 14Cr ODS (produced by CEA) are performed to investigate the welding window for these type of materials. A wide weldability window (between 2000 and 4000 m/s, collision point (welding) velocities) exists for welding 14Cr ODS steel. Due to extremely high strength and low ductility, explosive welding should be done with pre-heating to 200 oC and increased impact velocity (height of explosive layer) compare with conventional steels.
TIG (Tungsten-Inert-Gas) and EB (Electron Beam) welds of P91, P91-ODS, P92 and PM2000 were performed. Dissimilar welds of material with different chemical composition (P91/PM2000) show that a connection is possible, but the Charpy results are more brittle than each of the base material.were be performed and tested by Charpy and creep tests. 3 tests of EB welding of ODS P91 with ODS P91 failed. The tests varied the welding requirements. The results of the P91-PM2000 EB weld show the feasibility joining these two materials, but the test results let advise against this kind of connection. Tests of welding ODS P91 with P91 show the same porosity as the similar welds. Welding P91 using EB is a reasonable joining technology.
WP2 - Materials’ compatibility with coolant
This work package is divided into three main tematic areas: compatibility with different coolants of ODS steels and reference materials, corrosion barriers development and mechanical tests of reference materials when in contact with the coolants themselves. Different coolants have been considered: Pb, LBE, impure He (VHTGFR reference atmosphere) and superctitical water.
The present chapter is thus divided into three paragraphs.
WP 2.1 Compatibility of materials with the coolant and corrosion experiments.
GenIV fast breeder reactors might be cooled by means of liquid metals, gas, supercritical water or molten salts. GETMAT FPVII program focussed mainly on HLM (Heavy Liquid Metals) and on gas and SCW. In the frame of this WP2 area, many institutes performed compatibility experiments of the GETMAT ODS steels both in flowing and stagnant environments. Not only the durations of the tests were different, but the experimental conditions too:
• ENEA: flowing tests in the LECOR loop facility, T= 500°C, flow speed 1 m/s, Oxygen 10-6wt% of 15-15Ti ss up to 8000 hours, ODS steels, up to 5000 hours
• CIEMAT: stagnant tests of 9Cr and 12 Cr ODS in Pb at 600°C and 700°C, at 10-5wt% and 10-7wt%, up to 2000 hours
• CNR: stagnant tests in Pb at 650°C and 700°C of ODS steels and wetting tests
• KIT: corrosion tests of welded materials at 10-6wt% and 10-8wt%.
As above said, the materials to be tested were the GETMAT reference ones, with different compositions.
Where the first two steels are KOBELCO Japan commercial ODS steels, the third the CEA 14Cr ODS manufactured for GETMAT, the last the 15-15Ti DIN commercially made by Sandvik.
Considering the wide amplitude of the experiments and the different approaches in performing them, the outcomes and experimental data were surprisingly similar. A good example of compatibility with Pb and data comparison might be depicted in the micrographies.
Considering the above depicted examples on the 12 Cr ODS materials, a first qualitative observation is possible: the oxidation mechanism of the material appears very similar between the flowing and the stagnant experiments. Many randomly distributed different experimental results were function of the changeability of the original steels’ properties and characteristics.
A synthetic analysis for all the ODS materials might also be drawn.
All the samples tested by difeerent laboratories, do not show any mass loss due to massive corrosion phenomena. The oxidation behavior the three steels are indeed very similar, however some peculiarities of each sample can be highlighted.
The following scheme summarizes the three different behaviors:
• ODS9Cr steel shows an internal oxidation associated to the growth of a thin Fe-Cr-O layer on the surface.
• ODS12Cr steel shows a Fe-Cr-O layer growth on the steel surface and some spots of Cr-O oxide. Any internal oxidation has been detected.
• ODS14Cr shows a double oxide layer; a outer Fe-Cr-O layer and an inner Cr-O layer.
The thickness of the Fe-Cr-O layer ranges, averagely, between 2 μm for ODS9Cr and 10 μm for ODS14Cr showing a slight dependence on the Cr-content in the steel matrix, as well as the Fe-Cr-O composition. The same behavior is shown by continuity of the layer along the interface, in fact ODS14Cr presents a more compact and continuous layer than the other two steels.
The internal oxidation observed in the ODS9Cr steel can be associated to its fine grained structure. This kind of structure can promote the internal oxidation as already reported in literature.
A wide range of tests have been carried on welded materials too. A summary output of the work states that if the chemical composition and the grain size in the weld region (due to a heat treatment) are similar the welded area behaves like the bulk material: in 9Cr F/M steels welded by EB or TIG and one of the friction stir welds.
The smaller the grain size, the lager the oxide layer will develop in 10-6wt% oxygen containing Pb (friction stirr welds 20 and 21 and friction stirr with heat treatment).
Precipitations of elements at the grain boundaries and/or voids can lead to a fast growing oxide or a stronger dissolution attack, like observed at for the P91/PM2000 EB welded specimens.
Changes in the grain orientation and dislocation also may lead to a different behavior of the welding region, like observed at the specimens welded by explosive welding.
Independent on the welding method an oxygen concentration of 10-8wt% is a critical condition for protective oxide formation on welds and bulk material because at least partly dissolution attach must be expected.
HLM are not the only environment consedered as candidate coolant for GenIV reactors. As already mentioned, impure He has been taken into account either.
Three institutes were involved in the performance of He corrosion of 14Cr ODS only, at different temperatures, up to 1000 hours of experiment. The chosen atmosphere, depicted in table 2, is indeed not referring to the GFR GenIV reactor “Allegro”, but to the Very High Temperature GFR one, already very well known and studied in formerly exixting CEA facilities.
ENEA and CIEMAT manufactured dedicate dfacilities to perform the tests, whilst CEA adapted the Coralline loop. The apporaches were quite different: ENEA and CIEMAT obtained the reference atmosphere by means of commercially available tanks, CEA mixed the impurities in situ with a dedicated ingas complex system.
The exposures for all experiments were 250, 500 and 1000 hours, at 750°C, 850°C, 900°C, with a flow rate of 100 l/h ca.
The results gained in the three experiments, validate that the 14Cr ODS did oxidize in the test conditions as expected when the helium composition was selected. It was observed for other high chromium alloys that a sufficient CO partial pressure supports the growth of a dense chromia scale. Based on the general understanding of the surface reactivity in impure helium, it is expected that the surface chromium oxide protects the alloy against reactions with carbon bearing species, carbon monoxide and methane in this case, and thus prevents carburization.
Compatibility of ODS alloys under gas-cooled reactor conditions is virtually unknown. In addition, papers on the high temperature corrosion behavior of Fe-Cr ODS steels are scarce. These alloys were developed as structural materials for nuclear fusion systems but compatibility with environment has rarely been investigated. Few studies referred to other nuclear applications or to the fundamental understanding of their oxidation mechanism.
As far as the oxidation mechanism is concerned, the influence of rare earth elements (RE's) such as yttrium, neodymium and praseodymium on the oxidation behavior of a Fe-Cr alloy has been studies elsewhere A small addition of these REs reduces the oxidation rate. Analysis of the surface oxide morphology revealed that the morphology of oxides formed on Fe-Cr specimens with and without RE's specimens is different. In fact, a fine-grained oxide morphology was observed for alloys with RE's addition which may improve the scale adhesion. Also, it was found that REs increase oxidation selectivity for chromium. The critical Cr content for chromium oxide formation was reduced when REs were added. Basic comparison of the mass gains and scale adhesion for Fe-14Cr steels with and without Y is consistent with a smaller oxidation rate and higher quality scale for the ODS alloys.
Based on these preliminary results, 14Cr ODS steel exhibits a promising corrosion behavior in oxidizing impure helium at 800, 850, 900°C. This comment is important if one thinks that the alloy has yet not been optimized regarding its oxidation properties. However, significant work would still be needed in order to validate the environment compatibility of this material for use in a GFR. An experimental qualification program would at least comprise: parametric studies to establish oxidation kinetics as a function of the environment parameters based on medium to long term oxidation tests, understanding of the oxidation mechanisms to get inputs for lifetime modeling including material aging and long term effects, experimental validation on long duration oxidation tests, investigation of the corrosion behavior in off-normal conditions (temperature outburst, out-of-specification coolant chemistry) and possibly an optimization of the alloy as regards its corrosion performances.
SCW was the third reference environment considered in this WP, work perfomed by VTT only. The tests were done in SCW at two temperatures, 550°C and 650°C. The dissolved oxygen content in the feed water was between 100 and 150 ppb and conductivity was <0.1 μS/cm. The specimens were exposed to SCW at the pressure of 25MPa in supercritical autoclave connected to a recirculation water loop. The environmental control and monitoring in VTT SCW testing system include temperature, pressure, inlet and outlet water conductivity, inlet water dissolved oxygen content, and flow rate.
Based on the weight change testsand data analyses, Al-alloyed high Cr ODS steels PM2000 and MA956 have superior corrosion resistance when compared to 9-14% Cr ODS steels. The oxidation performance of the high Cr Al-alloyed ODS steels result from the formation of a protective Al–Cr–Fe-rich surface oxide layer. On ODS steels with lower alloying element concentrations, i.e. with no Al, the film is less protective with an iron rich outer and iron and chromium rich inner layers. Based on this work, in SCW conditions at 550°C and above, the Cr-content for thin walled ODS components must be higher than that of the 14%.
The results indicate that the ODS steels are less susceptible to SCC than austenitic stainless steels or Ni-base alloys. However, ODS steels have been tested very little in SCW and more data is required in order to make any final conclusions.
WP 2.2 : Advanced corrosion barriers development.
Surface modified layers based on FeCrAl coatings have shown their general capability to protect steels against the influence of liquid lead alloys by the formation of slowly growing stable oxide scales. This Wp2 task focussed on a commercial method, HVOF, and on the GESA method.
In a sense, the goal of this task was twofold: a critical analysis of two different deposition methods, an in deep study of the best compounds to be deposited. This latter was the most important output.
Because the best suitable FeCrAl alloy composition to ensure this protection over a wide temperature range was not addressed so far in a dedicated experimental work. The surface alloying process (GESA) is not the most appropriate to produce samples having a controlled change in composition. Therefore, model alloys were produced by Arc- Melting to evaluate the Al and Cr content required for protective oxide scale formation.
The phenomena occurring during the Fe-Cr-Al alloys exposure to HLMs, containing very small amounts of dissolved oxygen, is scarcely documented .No systematic investigation of the corrosion behaviour of the Fe-Cr-Al system, when exposed to HLM was performed, in order to find the minimum Al content for the formation of an alumina scale. The current paper presents the results of a systematic study concerning the corrosion behaviour of Fe-Cr-Al-base model alloys during their exposure to oxygen-containing (10-6 wt. %) lead in the temperature range 400 - 600 °C. In this vein a map of the alumina stability domain for this temperature range has been drawn in Fe-Cr-Al ternary diagrams and the type of alumina polymorph, formed at different temperatures, was determined.
Exposures were performed up to 600°C. Based on the state of the art oxide maps concerning oxidation behaviour of Fe-Cr-Al-base alloys at 800 and 1000°C in oxygen atmosphere, 10 compositions, belonging to this alloy system, were designed in order to tap the borders of the alumina stability domain, during their exposure to oxygen (10-6 wt%) containing lead, at 400, 500 and 600°C.
Eight alloys, Fe6Cr6Al, Fe8Cr6Al, Fe10Cr5Al, Fe14Cr4Al, Fe16Cr4Al, Fe6Cr8Al, Fe10Cr7Al and Fe12Cr5Al, were found to be protected against corrosion in oxygen containing lead, either by a duplex layer (Fe3O4 + (Fe1-x-yCrxAly)3O4) or by (Fe1-x-yCrxAly)3O4, depending on the temperature at which they were exposed.
Two alloys namely Fe12Cr7Al and Fe16Cr6Al were found to form transient aluminas Al2O3 (at 400 and 500°C) and Al2O3 (at 600°C), as protective oxide scale against corrosion in oxygen containing lead. An oxide map illustrating the stability domain of alumina, grown on Fe-Cr-Al- alloys when exposed to molten, oxygen containing lead, was drawn. The alumina stability domain border shifts with lower temperatures to higher chromium and aluminium concentrations. When CCr ε (12-25 wt%), the minimum concentration of Al, required to form alumina scale on Fe-Cr-Al alloys exposed at 400-600°C in molten Pb, is:
C Al =1.523978 – 0.80805 (CCr) + 0.01561 (CCr)2 [wt%].
For the temperature range and exposure times used during the current evaluation, the growth rate of the alumina scale was low. No area with detached scale was observed and no trace of Al2O3 was detected. Lower aluminium concentrations leads to the formation of Fe-base spinel-type oxides protrusions on extended area. However, this can also happen at high aluminium concentration (7.6 wt%) if the chromium content is lower than 10.4 wt%. The map includes also additional points, extracted from literature and corresponding to alumina forming alloys, when exposed to HLMs, which fit very well with our findings. The influence of the model alloys grain size, of the addition of reactive elements and of longer exposure time on the alumina scale polymorphs type, growth, stability and properties is currently under evaluation.
WP 2.3 Mechanincal testing in environment
Mechanical tests on the GETMAT materials were carried on in Pb, He and SCW too. The performance of mechanical tests in different environments leads to very different approaches and obstacles.
Creep tests in contact with pb were carried on by CIEMAT on 9Cr ODS.
The creep-to-rupture behaviour of the 9 Cr ODS was performed in air and in liquid lead under non-controlled atmosphere at 600oC and also at controlled condition at 650oC, as well as their fractures was examined after the testing. The results show an important effect of liquid lead in the creep behaviour, principally lowering the rupture time of the specimens in liquid Pb under non-controlled conditions. On the contrary, creep test performed at KIT under oxygen controlled conditions show similar tendency in air and in Pb like 12-14CrODS.
A possible reason could be the formation of a thick, porous and brittle oxide layer that breaks during the tests and allow an intimate contact between the liquid and solid metal, leading to a liquid metal embrittlement mechanisms. A deep characterization of the oxide layer formed on the specimens tested under oxygen saturated conditions is on-going to confirm this hypothesis.
Pb has been choosen by ENEA and KIT with the goal to perform fretting experiments, over the 14Cr ODS GETMAT material at ENEA, on GESA and 15-15Ti at KIT. The experimental campaigns differed essentialy in the approach to the experimental device, being carried on both at 500°C, in Oxygen saturated Pb, on teh same material with comparable loads and fretting frequancies. Figures 12, 13, 14 and 15 depict some of the interesting outcomes of the two tests. While the FRETME machine at KIT was able to perform several tests, comparing the reference matarial with others, the ENEA GIORDI device could perform fretting experiments without any debris re deposition phenomena. For all tests, the main goal is to simulate the interaction between the spacer grid and the fuel cladding of the GenIV LFR.
As a common summary of conclusions, it may be stated what follows:
• Favorable load-amplitude combinations can minimize the fretting severity. The fretting penetration and the wear rate decrease with an increasing applied load (after the turning point) and with a decreasing sliding amplitude. As dis¬played by the published fretting maps there are load-amplitude combinations (e.g. load > 75 N and amplitude < 15 µm) that favors the evolution of the fretting contact towards the stick regime. These load-amplitude combinations are expected to lower the specific wear coefficient (and consequently the wear and pene¬tration rate) by up to one order of magnitude compared to e.g. the combination 50 N-15 µm, which in Fig. 36 corresponds to the lowest measured penetration rates. Such conditions can be established with an adequate assembly strategy.
• Suitable material improvement and corrosion counter-measures im¬prove the wear resistance and mitigate the corrosion enhancement of fretting.
• A reduction of the penetration rate up to about 50 % already after 150 h was achieved by using pre-oxidized components. Likewise, for the specific case of austenitic steels, the use of Ni-en¬riched Pb reduces the fretting corrosion severity.
• Additionally, surface treatment and Al-alloying (e.g. GESA-T91) im¬proves remarkably the wear and the corrosion resistance prop¬erty also at high temperature.
• Fretting is a self-mitigating process. Fretting becomes less and less severe with the time. After 930 h, the penetration rate is up to 70% lower than during the run-in period to which the data of Fig. 13 refer.
• The beginning of the fretting process is delayed with respect to the re¬actor start-up. In e.g. the reactor core, the fretting process starts when, due to the irradiation damage and thermal-mechanical loads, the components (e.g. springs, supports or spacer grids) aimed to fasten the fuel cladding tubes (or other parts) lose their mechanical and dimensional properties. Therefore, the fretting process affect the e.g. fuel cladding tube for less than 25000 h.
• Possible design solution like the installation of spacer grids and sup¬ports in the non-active area of the fuel (lower temperature and irradiation) can delay the beginning of the fretting process further.
It is possible to claim with a certain confidence that, the listed aspects will allow the achievement of a penetration rate complying with the operating requirement (0.002 µm/h), which is about one order of magnitude lower than the one measured experimentally (in the best cases from 0.039 to 0.063 µm/h for the three materials).
The small punch test campaign is the last to be quoted in Pb, peformed by RSE.
The characterization of the steels included a preliminary exposure to liquid lead into a controlled-flow rig, followed by a set of classical (tensile, Charpy and creep) and innovative (small punch) mechanical tests.
Difficulties in the production of ODS alloys led to delays in materials procurement, however in the first period of the project RSE activity wasn’t touched, as it concerned the development of the geometries of the samples, the specimens’ and fixtures’ machining and the characterization of P91 (historically one of the first steels having medium content of chromium and often considered as the reference steel of ferritic-martensitic type) by small punch technique. At the delivery time of first ODS material RSE then started a mechanical characterization of virgin material by means of small punch technique with some additional conventional tests if spare material was available. The preliminary characterization of ODS virgin material was later performed for the other two GETMAT distributed ODS steels.
This experimental activity gave good results in terms of test repeatability, and an analysis of material anisotropy was performed, reporting similar trends shown by a few conventional tests performed by other project Partners.
After exposure at ENEA rig LECOR of RSE specimens at 500°C with flowing liquid lead in controlled low content oxygen condition, the final characterisation of exposed materials has been completed allowing the following main considerations:
• No effects of long time exposure on tensile properties, neither when determined by small punch tensile for both 12 and 14 Cr ODS, nor when determined by uniaxial test as observed for 12 Cr ODS.
• Limited effect of exposure on creep rupture time reduction: 14 Cr ODS estimated by small punch creep show some tests with reduced rupture time but data can be considered into conventional creep scatter-band of reference virgin material data, 12 Cr ODS estimated by uniaxial creep tests shows a trend of slightly more pronounced reduction for rupture times but data are always into the scatter of reference virgin material. The observed creep resistance reduction can be related to the presence of penetrating oxidation on specimen surface enhancing localized stress concentration.
• Significant effect of exposure on material embrittlement observed for both ODS steels: a relevant increase of transition temperature estimated by small punch tensile for 14 Cr ODS and a relevant reduction of impact energy for 12 Cr ODS, the latest confirmed by a less relevant (than 14 Cr) but anyway evident increase of small punch tensile transition temperature.
LCF tests were conducted at RT and at 750 °C on 14 Cr-ODS steel specimens in simulated Gas cooled Fast Reactor (GFR) environment, already quoted. Based on the results of such tests, the following main observations can be drawn:
• the cyclic stress-strain curves were obtained for 14 Cr-ODS steel, both at RT and at 750 °C; best-fit relationships were derived from such curves, in order to represent the elasto-plastic behavior of the material under cyclic conditions;
• the general behavior of the 14 Cr-ODS steel during HTLCF tests conducted in simulated GFR environment does not show significant differences as compared appears to other rather similar materials;
• the increase in test temperature from RT to 750 °C appeared to produce a fatigue life reduction by a factor as large as 25.
• the HTLCF data for 14 Cr- ODS steel tested in simulated GFR environment appear to lay near the lower bound of the typical scatterband observed for 304 and 316 SS, when tested in air at the same temperature;
• this could be an indication of an actual lower fatigue strength of 14 Cr-ODS steel as compared to 304 and 316 SS in HTLCF regime, or of a detrimental effect of simulated GFR test environment as compared to air or both. It is however worth noting that the number of tests which were conducted is by far too low for above observations to be regarded as conclusive and/or soundly based. Therefore, a significant extension of the test program, also including for creep-fatigue interaction effects, is strongly recommended.
SCW work, performed again only by VTT was essentially part of the SCW corrosion tests.
The results indicate that the ODS steels are less susceptible to SCC than austenitic stainless steels or Ni-base alloys. However, ODS steels have been tested very little in SCW and more data is required in order to make any final conclusions.
The goal to wrap up such a complex Work Package and to derive conclusions is a hard task. For the corrosion experiments it might be stated that:
• ODS materials showed several similarities, even in very different experiments.
• The presence of a dual layered oxide scale over 9Cr and 12Cr Ods materials is normally detected both in stagnant and flowing conditions. The oxide scale appearance is porous, and the mechanical adhesion is quite poor and random.
• 12 Cr ODS showed unexpected results, in some stagnant tests, where a single layer was detected only on some samples.
• 15-15Ti steel showed good protection due to in situ oxidations, in flowing Bb.
• On weldings: If the chemical composition and the grain size in the weld region (due to a heat treatment) are similar the welded area behaves like the bulk material → in 9Cr F/M steels welded by EB or TIG and one of the friction stirr welds.
• The smaller the grain size, the lager the oxide layer will develop in 10-6wt% oxygen containing Pb (friction stirr welds 20 and 21 and friction stirr with heat treatment).
• Precipitations of elements at the grain boundaries and/or voids can lead to a fast growing oxide or a stronger dissolution attack, like observed at for the P91/PM2000 EB welded specimens.
• Changes in the grain orientation and dislocation also may lead to a different behavior of the welding region, like observed at the specimens welded by explosive welding
• Independent on the welding method an oxygen concentration of 10-8wt% is a critical condition for protective oxide formation on welds and bulk material because at least partly dissolution attach must be expected
• In impure He, ODS materials exhibited an “as expected” oxidation behavior, leading to very small differences between the three temperatures. The manufacturing and surface treatment of the samples had a key role.
General conclusions on the other two subtasks are impossible, due to the very different nature of the experiments.
WP3 - Irradiation behaviour of structure materials
• Effect of high dose irradiation in a fast spectrum on mechanical properties and dimensional stability of a 9Cr martensitic steel.
9Cr martensitic steels are known for the excellent radiation resistance (i.e. negligible void swelling, little hardening and moderate degradation of impact properties) when irradiated in a fast spectrum at temperatures above approximately 380°-400°C up to doses of at least 100 dpa. However significantly higher values for the end-of-life neutron exposure of in-core components such as the hexagonal duct, up to about 150 or higher, are envisaged in the case of future GEnIV reactors. Moreover, it has been argued that the excellent swelling resistance of FM steels is mainly due to a long incubation period (transient regime) and that FM steels should eventually swell at a steady state swelling rate which could be as high as 0.2%/dpa. Therefore, a comprehensive PIE programme was conducted using an hexagonal duct made of 9Cr-1Mo steel (EM10 grade) irradiated in Phénix reactor up to a maximum dose of 155 dpa (Boitix9 experiment). Tensile and impact specimens were machined at different locations along the duct, corresponding to irradiation temperatures ranging from about 385°C to 530°C and doses in the range 43-155 dpa. Mechanical tests confirmed the very good radiation resistance of EM10 steel for these specific irradiation conditions.Irradiation induced hardening for all tested specimens, including after irradiation to 155 dpa, was found to be relatively small. Likewise, a moderate degradation of impact properties after irradiation was observed.
As in other metallic alloys, the irradiation temperature has a drastic impact on the evolution of the mechanical properties. For instance at 325°C, in addition to formation of α’ precipitates, irradiation induced the formation of a high density of small dislocation loops, possibly decorated with solute atoms, which are the main source of the high hardening observed after irradiation (see WP4, T4.2). In the temperature range of interest for the Boitix 9 wrapper, the dislocation microstructure was found by TEM investigations to be very different and consistent with the measured hardening: at 398°C (112 dpa), a low density of large dislocation loops was observed, while at 439°C (155 dpa), hardly any loop could be detected and the dislocation density was found to be similar to that measured prior irradiation.
In addition to the effect of irradiation on mechanical properties, dimensional stability is also of prime importance. Macroscopic density measurement showed that void swelling was very small. This conclusion was confirmed by TEM investigations of the void microstructures. This low void volume fraction and the fact that voids were found to be heterogeneously distributed within the matrix are indications that EM10 steel was still in the incubation stage of void swelling even after irradiation to 155 dpa, which demonstrates the very high swelling resistance of this type of steel.
• Effect of irradiation in a spallation environment on the mechanical properties of martensitic steels and ODS alloys
FM steels and ODS alloys are candidate materials for the structures of spallation targets and core components of ADS. In order to evaluate the evolution of mechanical properties in the relevant radiation environment, tensile and impact specimens of various FM steels and ODS alloys were inserted in the solid spallation target SINQ (PSI) and irradiated with high energy protons and spallation neutrons at temperatures in the range [150-350°C].
For all investigated FM steels, the same tensile behavior was observed.
The specimens irradiated at the highest doses and helium contents displayed very high hardening and a fully brittle, predominantly intergranular fracture mode, as clearly demonstrated by the tensile curve and SEM fracture surface observations for the specimen irradiated to 19.8 dpa and 1750 appm He. This behavior, which is not observed for FM steels irradiated in fission conditions at the same temperature and to similar damage dose, is the result of i) matrix hardening due to the formation of small dislocation loops and a very high density of tiny helium bubbles (~ 3 1016 cm-3 according to TEM measurements) and ii) strong decrease of grain boundary cohesion due to the presence of helium (as shown by ab initio calculations). Likewise, measurements of impact properties after irradiation of FM steels in the SINQ target show a degree of embrittlement significantly greater than that expected after irradiation in fission conditions.
By contrast, ODS alloys displayed a better mechanical behavior after irradiation in spallation conditions. For example, 14Cr ODS (MA957 grade) irradiated in SINQ retained significant ductility when tensile tested at room temperature, whereas, as mentioned above, FM steels subjected to identical irradiation conditions exhibited a brittle intergranular fracture mode. The high density of nanoclusters in ODS alloys are believed to act as effective sinks for point defects as well as trapping sites for helium, thereby decreasing the irradiation induced hardening and protecting grain boundaries from helium embrittlement effect.
• Combined effects of Heavy Liquid Metals (HLM) and irradiation
As several advanced nuclear systems will use a HLM (liquid Lead or Lead Bismuth Eutectic) as coolant, the issue of possible combined effects due to irradiation and HLM is of great importance and has little been addressed in the past. This issue was therefore the focus of several tasks of WP3, in particular via the PIE of ASTIR, LEXURII and MEGAPIE. As part of ASTIR experiment, the behavior of T91 steel in lead Bismuth Eutectic (LBE) was investigated. Liquid Metal Embrittlement (LME) was observed in the case of T91 in the as-received condition. For instance, when tensile tested in LBE, the total elongation was reduced as compared to a test in an inert atmosphere (i.e. fracture occured earlier), while the shapes of the tensile curves in both environments up to the rupture strain were very similar. However, the necessary condition for LME occurrence is an intimate contact between the steel and the liquid metal. This was achieved in the as-received condition for instance by exposing the samples before tensile testing to LBE with low oxygen concentration, which resulted in a partial dissolution of the protective oxide layer typically found on the steel surface.
The effects of test temperature and strain rate on LME were studied in the case of unirradiated specimens and it was found that the reduction of the total elongation has a non-monotonic dependence on both parameters: LME occurs only in a particular range of temperature and strain rate. For a strain rate of about 5 10-5 s-1, the minimum of the ductility trough occurs at about 350°C.
In order to investigate the effect of irradiation on the LME sensitivity of T91, tensile tests were carried out both in inert atmosphere (Ar) and in LBE on T91 specimens irradiated to 2.6 dpa as part of ASTIR irradiation experiments. The tests were performed at 300°C, which was the average irradiation temperature.
A strong LME effect was observed for irradiated specimens tested in LBE. When tested in LBE, multiple cracks initiation occurred at the surface followed by brittle propagation. The average reduction of total elongation due to LME (about 63% reduction compared to tests in Ar atmosphere) was proportionally larger than the average reduction measured in the case of unirradiated samples. Furthermore, ASTIR specimens had not been irradiated in contact with LBE. While for unirradiated specimens, pre-exposure to the liquid metal was found to be a necessary condition of LME occurrence, this was not the case for irradiated specimens. Thus irradiation was found to enhance the LME sensitivity of T91. This conclusion was also supported by the results of PIE performed on T91 specimens irradiated in contact with LBE to about 6 dpa at 350°C (LEXUR II experiment). Also in this case, a drastic reduction of total elongation was observed for irradiated specimens tested in LBE at 350°C compared to values measured in inert atmosphere.
By contrast, for all investigated testing conditions, including tests on specimens irradiated to 2.6 and 6 dpa (as part of ASTIR and LEXURII experiments), austenitic stainless steels (316L and 15-15 Ti grades) did not show any evidence of LME due to LBE.
In addition to tests performed on irradiated specimens, WP3 included as well the PIE of the MEGAPIE target, which was a unique opportunity to investigate the behavior of a real component whose structural materials had been irradiated in contact with flowing LBE and subjected to cyclic thermal/mechanical loadings. It must be emphasized that target dismantling using a large dedicated bandsaw, LME out-melting with a specially designed oven, raw cutting of the PIE structural material samples, cleaning of the raw cut samples (in order to remove alpha contamination), fine-cutting via EDM of a large number of samples which were then packed and sent to partner laboratories, represented a huge amount of work successfully carried out by PSI in its hot laboratories.
Prior to target dismantling, gamma-mapping of target tip was performed and the results were used for evaluating the proton fluence distribution and irradiation doses on the various target components. In addition, gamma analysis of LBE samples taken from different positions in the target showed good agreement between experimentally determined and calculated activities, which indicates that predictions by nuclear codes can be reliably used for safety assessments.
As target dismantling and specimen cutting took much longer than originally planned, only a limited PIE programme could be carried out in the timeframe of the GETMAT project. However, interesting results were obtained. Tensile tests (in air) were performed on specimens cut from the hemispherical beam entrance window made of T91, which was the most irradiated component (the maximum dose was calculated to be about 6 dpa). As expected, significant hardening was measured, however the total elongation of the sample with highest dose was found to be still above 10%, indicating a fully ductile behaviour. These results were in line with available data for T91 irradiated in spallation conditions in the same temperature range (close to 300°C).
No severe LBE corrosion nor erosion phenomena were observed on the T91 and 316L components. These findings are consistent with the results of the investigations performed on coupons irradiated in contact with stagnant LBE (LEXURII PIE) which did not reveal any corrosion damage.
However, SEM and EPMA investigations performed by PSI and CEA have revealed the presence of a deep crack in the Lower liquid Metal Container made of T91 steel, located close to the Electron Beam (EB) weld line (Fig. WP3.6). Such a deep crack was not present before irradiation, but was perhaps nucleated during welding. Cyclic mechanical loading (due to proton beam trips) may have induced rapid crack propagation, since it has been shown that fatigue crack propagation is accelerated in LBE as compared to crack growth rates in inert environment.
WP4 - Multiscale modelling and model experiments on irradiated Fe-Cr alloys
The objective of the WP was to develop physical models to describe the behaviour of iron-chromium alloys when subjected to thermal ageing and irradiation. These alloys were chosen as model materials for high-chromium ferritic/martensitic steels. The processes of interest were nanostructural and microchemical changes that occur under irradiation or ageing and the corresponding changes in mechanical properties, mainly hardening, which in turn causes embrittlement. The approach used was computer-based multiscale modelling (from ab initio to dislocation dynamics), integrated by the performance of specific modelling-oriented experiments. The latter provided detailed data on which models can be contrasted and suggested, in combination with fundamental atomistic calculations, physical mechanisms that the nanostructure and microchemical models, as well as mechanical models, needed to incorporate. The link with other work-packages is ensured not only by the comprehension of the mechanisms governing the changes induced by irradiation in iron-alloys with varying content of chromium, which allows experimental results on steels performed in the rest of the project, especially WP3, to be better rationalised, but also by the inclusion, in some tasks, of the characterisation of technological properties, such as fracture toughness, in model alloys was included in some tasks.
The ultimate objective of the WP was to understand which nanostructural features are the main cause of radiation-hardening and to develop a physical model capable of rationalising the experimental data available on this issue. This goal was indeed reached.
The work-package was organised in three main tasks, the highlights of which are given in what follows.
T4.1 – Modelling of fundamental properties of Fe and Fe-Cr alloys
The main objective here was the development and application of advanced cohesive models for Fe-Cr alloys, capable of describing their thermodynamic properties and the properties of point-defects (vacancies, self-interstitials), as well as extended-defects (dislocations, grain boundaries), possibly including magnetic properties. The development of these cohesive models, then applied for atomistic calculations, required extensive density functional theory (DFT), or first principle (ab initio), calculations.
Highlights of task 4.1
• First principles calculations have provided new insights into several areas of interest for the basic properties of ferritic alloys; specifically, elements commonly found in nuclear steels, not only Cr but also other solutes in Fe-Cr, have been studied. The main findings of these calculations can be so summarized:
• The magnetic coupling has been extensively studied in Fe-Cr, revealing how magnetism influences some of the properties of the alloy, especially the mixing enthalpy, determining the change of its sign from negative to positive in the low Cr region, thereby leading to a modified phase diagram as compared to the standard one. The strong vibrational entropy of Fe-Cr has been verified: it plays an instrumental role for the shape of the Fe-Cr phase diagram, by lowering the temperature at which the misicibility gap closes. Non-configurational entropy should therefore be included in models that describe thermodynamics from an atomistic standpoint.
• For the grain boundary and surface segregation of Cr in Fe, new insights have been gained, namely, Cr appears to have a tendency to segregate at grain boundaries, while it would tend to stay away from free surfaces, especially voids.
• The study of the interaction with point defects with all possible solute atoms in iron revealed that the elements that are found experimentally to form aggregates and to segregate at dislocations or grain boundaries (namely Cr, but also Ni, Mn, Si, P, and Cu) are found to migrate together with point defects, both vacancies and, in a few cases, interstitials. This proves to be a very effective physical mechanism to explain the experimental evidence.
• The development and validation of the interatomic potentials has rendered a new, improved interatomic model for Fe-Cr in very good agreement with experiments and first principles calculations (see above). This potential has been validated and put to use in dislocation studies, defect diffusion, solute-defect interactions, etc. In particular, Metropolis Monte Carlo studies revealed how Cr is expected to re-distribute at steady-state under thermal ageing and, as a tendency, under irradiation. Fig. WP4.1 illustrates graphically the results of these studies.
T4.2 – Modelling of radiation effects in Fe and Fe-Cr alloys
The main objective here was the development of microchemical and nanostructure evolution models, as well as (radiation) hardening models, for Fe-Cr alloys subjected to thermal ageing and irradiation, using as input the results of applying ab initio and/or empirical cohesive models from T4.1. The former category includes several types of kinetic Monte Carlo (KMC) models, each of them with its advantages and limitations, as well as cluster dynamics (or rate theory) model. The latter consistes basically of the coupling of molecular dynamics (MD) with dislocation dynamics (DD): it is the first time that this coupling is successfully applied to rationalise quantitatively the results of mechanical tests, starting from fully physical considerations.
Highlights of task 4.2
• Atomistic KMC (AKMC) techniques have reached maturity as methodology for the simulation of thermal ageing processes (phase separation, segregation, …) driven by diffusion of solute atoms, although some specific difficulties still exist with the diffusion of interstitial species like carbon. The importance of including (semi-empirically) magnetic and vibrational effects emerges from both DFT calculations and experimental studies, and should be considered as one of the main improvements to be introduced in the technique in the long run.
• For AKMC simulations of irradiation processes, important progresses have been made to include multiple vacancies and single-interstitials: the effect of Cr on point defect diffusion properties is explicitly accounted for and simulations of resistivity recovery experiments including effect of Cr and C have been successfully performed. Also the calculations of the Lij phenomenological coefficients (Onsager matrix), for use in radiation-induced segregation models, has been performed and processes such as phase separation and segregation can be now successfully simulated starting from fully physical considerations. However, the modelling of SIA clusters in an AKMC framework remains challenging.
• Object KMC (OKMC) and Cluster Dynamics (or Rate Theory) models have been developed and a direct comparison with experiments has been made to validate them. The key difficulty is the determination of parameters for concentrated alloys: a new method, cellular OKMC, has been developed, with promising results. As alternative, a "grey-alloy" model in which the presence of Cr and C atoms on the nanostructural evolution under irradiation is taken into account by modifying defect stability and mobility parameters, without explicitly introducing solute atoms, has been also developed. The latter allowed good agreement with experimental trends to be obtained, but cannot model precipitation. The former, on the other hand, allows precipitation to be described and in the long run should be able to treat also the other defects, in a similar way to the "grey-alloy" model, providing also more local information, but for the moment these features are not included yet in the code. The rate theory models that have been developed clearly required specific fitting to experimental data.
• Extensive MD simulations have been performed in order to assess the friction stress due to solution strengthening in Fe-Cr alloys of different concentrations, as well as to evaluate the critical stress for unpinning of both edge and screw dislocations at defects, including Cr-decorated loops and Cr-rich precipitates, which are assumed to be the main cause of irradiation hardening in Fe-Cr alloys. The studies have been performed for different defect and box sizes, different temperatures and also different strain rates. They have clearly shown that dislocation loops decorated by Cr atoms (and/or other solute atoms), as a consequence of segregation, are stronger obstacles than undecorated loops and are, therefore, candidates to be the main origin of radiation hardening in these (and other) Fe alloys. A method to extract from MD data the parameters for DD simulations has been developed: this is based on the deduction of the 'friction stress inside the defect', which depends only on the type of defect, as well as the activation energy of unpinning.
• The MD results of strength of irradiation defects (loops, Cr decorated loops, Cr precipitates, …) have been incorporated into a DD model, allowing simulations of radiation strengthening in Fe-Cr alloys as a function of Cr content: it is the first time that a fully physical model manages to account in a quantitative way for the radiation-hardening measured experimentally. Key is the recognition that screw dislocations that interact with loops having collinear Burgers vector may form helical turns, thereby producing hardening effects 10 time larger than the Orowan mechanism. The main sources of hardening are the Cr-NiSiP clusters, if assumed to be small loops with solute atoms segregated on them.
T4.3 – Modelling-oriented experiments in Fe and FeCr alloys
The main objective here was to performance modelling-oriented experiments, in which selected reference materials (Fe-Cr model alloys) were characterised after neutron and ion irradiation, from the nanostructural, microchemical and mechanical point of view, using a combination of experimental techniques.
Highlights of task 4.3
Two main sets of experimental results have been obtained within this task:
1. Characterization of neutron irradiated Fe-Cr alloys
2. Characterization of ion irradiated Fe-Cr alloys
In both cases the goal was to combine different experimental characterization techniques in order to get a complete picture of the nanostructural and microchemical changes occurred in the alloys as functions of Cr content and irradiation dose, as well as, partly, temperature. The materials considered were alloys containing nominally 2.5 5, 9 and 12 %Cr.
Neutron irradiated Fe-Cr alloys (0.06 0.6 and 1.5 dpa, at 300°C)
- Positron Annihilation Spectroscopy (PAS)
Mainly single vacancies are present in FeCr alloys compared to pure Fe: very little vacancy clustering occurs when Cr is present (data for 0.06 dpa).
Higher amount of vacancies is observed at higher dose, but nonetheless the presence of Cr reduces the amount of vacancies drastically.
- Small Angle Neutron Scattering (SANS)
The irradiation-induced clusters detected are Cr-rich alpha’-phase particles for Fe-12%Cr, while in the other alloys they must contain impurity elements, such as C, Ni, P or Si.
There is a pronounced fluence dependence of the concentration and size of alpha’-phase particles below 0.6 dpa and a saturation-like behaviour beyond.
For lower Cr levels, pronounced fluence dependence below 0.6 dpa was also found.
- Atom Probe Tomography (APT)
alpha' clusters were only present in the supersaturated alloys (Fe-9%Cr and Fe-12%Cr)
CrSiPNi enriched clusters were observed in all alloys with the same number density: they are suspected to be small loops on which solute atoms segregated.
Segregation of Si, P and Cr atoms on dislocation lines and grain boundaries was observed.
Zones free of clusters were found around the GBs, suggesting that solute segregation should be the mechanism leading to the formation of the CrSiPNi clusters: they would not form where other sinks attract the solute atoms.
- Transmission Electron Microscopy (TEM)
Dislocation loops are inhomogeneously distributed for 5, 9 and 12 Cr with dislocation loops preferentially located close to grain boundaries and dislocation lines; they are homogeneously distributed for 2.5%Cr: there is an effect of Cr content on the distribution of dislocation loops. This effect might be indirect: it is known that 2.5 %Cr alloy is ferritic and has C in the matrix, while the other alloys are ferritic/martensitic and C has segregated at grain boundaries.
Ion irradiated Fe-Cr alloys (to 1 and 5 dpa, at 100, 300 and 420 °C)
- PAS (Coincidence Doppler Broadening and Positron Annihilation Lifetime Spectroscopy):
Vacancy-type defects were produced by the ion irradiation. The amount and size decrease when %Cr increases and also, when the irradiation T increases. Larger voids and more observed at the lower dose.
- X-ray Magnetic Circular Dichroism (XMCD):
Ion irradiation produces a redistribution of Cr atoms, deduced from the magnetic moment, which can be an indication of Cr clustering also observed by APT.
- APT:
NiSiPCr rich clusters where observed after ion irradiation (Cr is the main element)
No alpha'precipitates are observed in this case, differently from neutron irradiation.
- Extended X-Ray Absorption Fine Structure and X-Ray Diffraction (EXAFS):
This technique indicates the presence of vacancy and interstitial type defects.
- TEM:
Homogeneous distributions of interstitial dislocation loops are found at all conditions.
Indications of denuded zones close to grain boundaries in Fe12Cr.
At 1 dpa: there is no significant effect of Cr content (from 9 to 12) on size or density of defects. Differences are expected in the proportion of ½<111> loops to <100> loops.
At 5 dpa: higher density in 9Cr, while larger size in 12Cr.
Potential Impact:
- Potential impact, exploitation of results and dissemination
It was foreseen that the potential impact of the GETMAT is to advance in the area of materials technology for transmutation and Generation IV reactors. Priority R&D topics were identified and accordingly success indicators defined. Moreover, it was stated that GETMAT would also impact the share of activities among different EU countries and laboratories accounting for their specific area of excellence and by optimising the utilisation of relevant installations.
The potential impact of GETMAT results, as declared at the beginning of the project (and possibly also beyond) is hereafter discussed by evaluating the success indicators as defined in the description of work.
For what concerns the exploitation of results, the project evolution and final outcome can be considered as an essential step towards the emplacement of the Joint Program for Nuclear Materials (JPNM), within the framework of the European Energy Research Alliance. For instance, GETMAT results have been essential in identifying research themes for the JPNM which were translated in Sub-programmes and Pilot Projects.
Moreover, GETMAT has invited a number of external partners to participate to the user group. The user group was composed by steel maker industries as OCAS, and CENTROINOX, the nuclear industry as AREVA and ANSALDO, by research associations as IFE (Norway) and Non-European R&D labs as LANL.
In the section “exploitation of results” a detailed analysis on how the user group has appreciated the GETMAT results will be described.
Finally, GETMAT has done an important work on dissemination of results as well as on education and training. Both aspects were part of the success indicators and will be described in this section. Moreover, in the section 4.3 “Use and dissemination of foreground” further details will be given.
List of success indicators (KPIs)
In order to assess the potential impact of the GETMAT project an analysis of the “Success Indicators” defined at the beginning of the project is hereafter done. The “Success Indicators”, have been identified with the objectives to measure through specific milestones and deliverables the achievement of the declared objectives. The impact of these achievements is then discussed.
Due to technical problems that have been encountered during the project life and that have been solved by e.g. re-addressing activities, it was necessary to adapt the “Success Indicators” accordingly to the updated activities. The identified “Success Indicators” can also be understood as Key Performance Indicators (KPIs) since the “KPIs are commonly used to evaluate the success of e.g. an organization / program / project or the success of a particular activity”.
The list of KPIs (or Success Indicators) for the GETMAT Project, as reported in the description of work, can be classified within two categories. The first category is related to the scientific advancement, while the second category is related to the project output in terms of dissemination and education and training activities.
Hereafter the list of KPIs is given:
GETMAT KPIs (or “Success Indicators”) category I: scientific advancement
1. Distribution to the laboratories of the ODS alloys manufactured in Europe and Japan. Since the fabrication of the ODS alloys is considered a challenging task, due to the amount of material to be produced, the achievement of this milestone will be a successful accomplishment and will give the opportunity to the laboratories to perform the planned experiment within a not too tight schedule.
2. Demonstration of the high temperature performance of the three ODS alloys and their weldability. The accomplishment of this demonstration will be an important task for the definition of the future R&D needs within the area of ODS alloys development for nuclear applications. In addition the compatibility tests of the ODS alloys with different coolants, will complete this framework.
3. The full manufacturing process of “smart coatings” for fuel claddings with optimised chemical composition and standardised qualification process will be available and can be therefore considered in the future consideration concerning materials selection related to the issue of alternative corrosion protection method.
4. The results of the PIE of the samples irradiated in the BOITIX9 and SUPERNOVA experiments (Phénix), MEGAPIE/SINQ, ASTIR (BR2), IBIS/SUMO (HFR) and BOR60 will be collected in several deliverables. A comprehensive representation of these data and the ability to transmit the performance assessment of irradiated materials in presence of coolant to the designers will be an important success indicator.
5. The successful completion of the model parameterisation phase and the timely start of a second phase of model development based on the use of available experimental results will be also used to measure the performance of the project on the modelling side.
GETMAT KPIs (or success indicators) category II: dissemination and education & training
1. Organisation of Workshop / Training Courses / Schools
2. Number of PhDs and Post-Doc students: The reference in the description of work was is 5% of the total budget of the project. This corresponds to a number of PhD/Post-Doc students in the order of 6-10
3. Number and type of publications
Discussion and assessment of the KPIs (or Success Indicators) and their potential impact
GETMAT KPIs (or “Success Indicators”) category I: scientific advancement
1. Distribution to the laboratories of the ODS alloys manufactured in Europe and Japan.
Achievement: Within the GETMAT project ~ 65 kg of ODS were produced and distributed to the GETMAT partners as agreed in the description of work:
o ~ 30 kg 14 Cr ODS
o ~ 20 kg 12 Cr ODS
o ~15 kg 9 Cr ODS
- The distribution of ODS to the laboratories has been achieved with some delay
o ODS manufacturing at large scale in Europe had technical problems at KIT and SCK-CEN
o GETMAT management achieved to get ODS from Japan
Potential impact:
• It has been seen that the large scale production of ODS steel needs further development in Europe.
• All GETMAT partners were working on the same batch of materials, therefore results were comparable. Moreover, the ODS has been distributed such as to optimise the use of facilities and installations in EU.
2. Demonstration of the high temperature performance of the three ODS alloys and their weldability and compatibility tests of the ODS alloys with different coolants
Achievement:
- Tests at high temperature have been performed up to 750°C
o first mapping of ODS performance
o consistency of the mechanical test results coming from different partners, that is an indication of the expertise of the participants
- Investigated welding techniques
o Diffusion Bonding, Friction Stir Welding and Explosive Welding can be applied on ODS upon
- Compatibility with coolants checked
o dependencies from chemistry and temperature identified
Potential Impact:
• Results obtained have made evident the necessity to deepen the microstructural behaviour of ODS when subject to high stresses and to correlate fracture mechanism with microstructure. This will be done within JPNM and related projects.
• It has been suggested that the GETMAT-ODS might be suitable for clad material of advanced nuclear reactors. However, the performances of GETMAT-ODS were not so satisfactory to recommend this material for structural components of reactor systems.
3. The full manufacturing process of “smart coatings” for fuel claddings with optimised chemical composition and standardised qualification process
Achievement
- The thickness of the coating prior to GESA treatment must be < 30µm.
- The surface alloyed layer (after “GESA” treatment) has following required composition: >8wt% Al, between 9 and 14 wt% Cr and the balance is Fe.
- The GESA treatment parameters are fixed regarding electron energy (120keV), pulse duration (40 to 60 µs) and power density (about 2MW/cm²). More than 2 pulses are needed (one with full duration, the other with half duration) to homogenize the final surface
- Short term oxidation experiments in gas Atmosphere is suitable to assess the capability to form thin Al-rich oxide scales in liquid Pb alloys. However the required Al content in gas is at least 2wt% high compared to Pb.
Remaining open issues:
- Deposition process for the coating is still not hundred per cent satisfying. Selected deposition process results in too high oxygen content.
- Can rare earth elements further reduce the required Al content required for thin slowly growing Al-rich oxide formation?
- Process stability and repeatability using the needed GESA IV is still not optimal.
- Qualification process is not working to that extend as expected – an elemental analysis based process (e.g. large area EDX) should be considered and investigated
Potential Impact
• Corrosion protection barrier can be further developed for LFR systems
• Among the two combinations of techniques investigated, i.e. HVOF+Laser treatment and LPPS+GESA treatment, this second one seems to be more promising. However LPPS might be replaced with VPS.
• A qualification process of corrosion protection barrier for nuclear application needs to be defined.
• Relevant activities on corrosion protection barrier for Generation IV and transmutation systems have been included in JPNM and related projects.
4. The results of the PIE of the samples irradiated in the BOITIX9 and SUPERNOVA experiments (Phénix), MEGAPIE/SINQ, ASTIR (BR2), IBIS/SUMO (HFR) and BOR60.
Achievements
- PIE results have been generated on conventional steels as 9-12 F/M and austenitic steels in different environments.
Potential Impact
• The long-term use of 9-12 F/M steel in high temperature and neutron irradiation fields (i.e. for Generation IV and transmutation reactors) is restricted to a temperature range of 350-550°C due to irradiation embrittlement (in the low temperature range) and poor thermal-creep and stress-to-rupture resistance in the high temperature range
• Liquid metal (especially Pb-Bi eutectic and Pb) assisted degradation of 9-12 F/M steel limit the use of this materials for LFR and ADS. Similar degradations have not been observed on austenitic steels.
• In proton-neutron mixed spectrum the high He production rate per dpa might contribute to the hardening and embrittlement of 9-12 F/M steel. These data are relevant for the neutron spallation target development, which is an essential component of ADS systems
5. The successful completion of the model parameterisation phase and the timely start of a second phase of model development based on the use of available experimental results
Achievements
- Fe-Cr phase diagram assessment and role of vibrational entropy
- Definition of an improved interatomic potential for Fe-Cr which has proven to be in very good agreement with experiments.
- Theoretical explanation and experimental verification of irradiation hardening and embrittlement in Fe-Cr alloys
- Identification of differences in the field of ion vs. neutron irradiations
Potential Impacts
• The solubility of Cr in Fe and formation of phases (as given in the phase diagram in the low Cr concentration range) needs to be revised.
• Explanation of irradiation hardening and embrittlement of Fe-Cr alloys can be extended to F/M steels. However, further steps in the theoretical approaches are still needed
• Modelling activities relevant for Generation IV and transmutation systems have been included in JPNM and related projects.
• Even if ion irradiations are relatively easy experiments to be performed. It has been demonstrated that results from ion irradiation have to be handled carefully when used as surrogate of neutron irradiation.
Category II of GETMAT KPIs: dissemination
1. Organisation of Workshop / Training Courses / Schools
Within the GETMAT project it has been agreed to organize two Schools and two workshops. The scientific output of these training activities is here measured in terms of number of attendees at the Schools/Workshop, number of lectures and poster presented.
It should be here also noticed that all GETMAT training activities, but the final international workshop have been organized in cooperation with other EC funded projects and / or international organisations as e.g. IAEA, USA laboratories etc. This approach of organizing training sessions within GETMAT has allowed the improvement of those components of the project that are directly related to the European Research Area and Networking of Excellence.
The training activities organised are:
• First International School on MATerials for nuclear REactors (MATRE-1). 18-23 October 2009 - Auberge de la Ferme - Rochehaut-sur-Semois, Belgium. This school has been organised together with the EC supported PERFORM60 project
• First International Workshop on Dispersion Strengthened Steels for Advanced Nuclear Applications (DIANA-I). 4-8 April, 2011 – Centre CNRS Paul Langevin, Aussois, France. This Workshop has been co-sponsored by: IAEA, LANL (USA), LLNL (USA), ORNL (USA) , OCAS/Arcelor-Mittal (Belgium), SCKCEN (Belgium) CEA (France), KIT (Germany) PSI (Switzerland) and JRC-IE (EU). Selected paper of the DIANA – I workshop have been published on Journal of Nuclear Materials.
• Joint GETMAT/MATTER Summer school - MUNECO: Materials UNder Extreme Conditions. This school has been organised together with the EC supported MATTER project
• Final International GETMAT Workshop
Statistics of the Training Activities:
MATRE – 1 No. of lectures: 18 No. of posters: 24 No. of attendees: 62
DIANA – I No. of lectures: 36 No. of posters: 16 No. of attendees: 84
MUNECO No. of lectures: 25 No. of posters: 32 No. of attendees: 54
Final Int. Workshop No. of lectures: 30 No. of posters: 20 No. of attendees: 75
2. Number of PhDs and Post-Doc students: 30
3. Number and type of publications:
• Conference / Workshop contributions: 4
• Journal Papers: 68
• Others (Newsletter etc.): 4
Exploitation of results: the GETMAT user Group
All participants to the GETMAT user Group were asked to fill in a questionnaire. The questions and the answers are hereafter summarised. It should be noted that not all members of the user group have answered.
• Question 1: Could you exploit the results produced within GETMAT within your institution. Do you have / will have possibilities to disseminate the knowledge acquired within GETMAT?
- Answer from steel industry: The GETMAT project definitively helped to guide and enrich the research done by the steel industry. The knowledge obtained within GETMAT is disseminated inside our organisation and towards strategic partners.
- Answer from nuclear industry: Unfortunately there was no possibility to access any document produced in GETMAT
- Answer from research organisations (EU and non-EU): The results produced within GETMAT were useful in the development of instruments for use in liquid metals and supercritical water. There are various possibilities to disseminate the knowledge acquired within GETMAT (User-Group lecture, Conferences, ..).
• Question 2: Could you identify limits and potentialities of the investigated GETMAT materials, i.e. 9Cr – F/M steels and ODS steels?
- Answer from steel industry: Yes, both for 9Cr-F/M as well as ODS steels the knowledge has has been develop but the industry became also aware of important limitations, mainly for ODS steels.
- Answer from nuclear industry: 9Cr – F/M is still used in both ELFR and ALFRED reactor cooled by Lead.
o The operating temperature has been selected above the Ductile to Brittle Transition threshold under irradiation.
o The lead velocity has been limited lower 2 m/s to limit erosion issue
o The oxygen controlled environment has been adopted to limit corrosion
However the R&D should not be limited to search material suitable for fluent molten Lead but also the influence of the Lead chemistry shall be exploited. We believe that appropriate structural material in conjunction with Lead chemistry (i.e. corrosion inhibitors) have more potentialities to solve the material issues.
- Answer from research organisations (EU and non-EU): ODS steels clearly have potential, but there is still a long way to go (improvement in fabrication routes, avoiding strong anisotropies, and the behaviour under irradiation is not sufficiently explored).
• Question 3: Did you consider the interaction with the GETMAT community through the technical meetings as effective?
- Answer from steel industry: Yes, and there was also the possibility to extent the research network through the GETMAT project.
- Answer from nuclear industry: Designers should be maintained informed about the GETMAT results not only with technical meeting but with appropriate documentation for implementing them in the design.
- Answer from research organisations (EU and non-EU): Yes, the technical meetings were very well organized (and intense) and offered sufficient possibilities for discussions (during the meeting, as well as outside of the meeting).
• Question 4: Do you have any suggestion on how User Groups can be implemented in EC supported R&D future projects on materials for nuclear applications?
- Answer from steel industry: Interaction is mainly related to the technical meetings. It could be beneficial to include maybe one or two additional workshops on sidely related themes like economic impact etc.
- Answer from nuclear industry: Improve the two way communication with dedicated workshops and provide summary and full detailed results.
- Answer from research organisations (EU and non-EU): (1) The case of our institute (IFE) is special because Norway cannot directly receive EC support (as Norway is not in EURATOM). However, we could (technically) contribute through instrumentation and materials testing under irradiation and we have unique capabilities for doing this. If such possibilities are recognized from the start in future projects, subcontracting could be considered. (2) I enjoyed my participation in this user group but it was difficult because of my location out of the US. If the user group has members outside of Europe, it may help participation if some meetings were held as side meetings in conjunction with other related international meetings as long as other members of the user group are also attending this meeting. Otherwise, I found this a very useful user group.
As can be seen from the collected answers, the relation with the User Group occurred only through the technical meetings. This might be the reason why the GETMAT results have been considered more of impact for the steel maker industry and the other research laboratories and less for the nuclear industry.
Moreover, as further suggestion to organise dedicated meetings with the user group to address topics not strictly related to the project might be taken into account for the future. Finally, all members of the user group could exploit the GETMAT results for further develop in their own field.
List of Websites:
The project website has been developed.
Internet address: http://nuklear-server.nuklear.kit.edu/getmat/index.php
The website is continuously up-dated both at the public part and at the member area part. Within the public part all presentations made during the training activities are accessible.
Moreover, the communication between beneficiaries has been always very constructive and good. This communication normally occurred through e-mail and phone calls. Moreover, technical meetings at WP and task levels have been organised on regular basis to discuss technical details of problems and outcomes.
The GETMAT project has also a good communication with other projects as e.g. the MATTER, SEARCH, MAXIMA and with the EERA Joint Program Nuclear Materials. Moreover, during the reporting period GETMAT has contributed to the identification of R&D gaps in the ODS area relevant for the JPNM and the proposal preparation of the MatISSE project.
Through the GETMAT User Group, parties (industries, non-EU research associations) that are external to the project are informed about relevant advancement of the project.
Finally, the GETMAT project has given an essential contribution to the Strategic Research and Innovation Agenda of the SNETP, to the EERA Joint Program on Nuclear Materials and to the EC road-mapping exercise on materials for the SET-Plan.
The objective of the GETMAT project was the investigation, both in the theoretical and experimental domains, of selected material properties that are cross-cutting among the various Generation IV and Transmutation reactor designs. Since the design requirements of the innovative reactors are beyond those for the LWR currently in operation, the selection of the properties to be investigated has been performed by identifying relevant conditions of key components as cores and primary systems. In fact, high temperature, high irradiation doses, corrosive environments as well as high thermal and mechanical loads are features that pose challenges on all materials foreseen for these components. Taking into account this framework and the goals defined by the Generation IV and the waste transmutation approaches as e.g. sustainability in terms of optimisation of use of resources and waste minimisation, high safety standards, economic viability and efficiency, it turned out that innovative materials might be a better choice with respect to conventional nuclear grade steels for the design of the core components and the primary system. Therefore, ODS alloys and 9-12 Cr Ferritic/Martensitic (F/M) steels have been selected as reference for the GETMAT project. In particular, even if ODS alloys are known since several decades, data for nuclear applications are quite scarce. The scientific results produced within the GETMAT project on 9Cr, 12Cr and 14Cr ODS alloys fabricated via the powder metallurgy route as well as the results obtained on the 9-12Cr F/M steels through an important PIE program, contribute considerably to a better understanding of the behaviour of these classes of materials for nuclear application. Moreover, the results have been found to be highly relevant for the identification of further R&D activities defined within the EERA Joint Program for Nuclear Materials (JPNM) and the related projects MATTER and MatISSE supported by the EURATOM FP7. In concordance with the experimental investigation on F/M steels and Ferritic / F/M ODS alloys the theoretical program was devoted to the study of Fe and Fe-Cr alloys. Emphasis has been put on first principle modelling studies and experiments to identify and validate physical models and mechanisms that explain irradiation hardening and embrittlement of these alloys. The results obtained are very promising and supportive in explaining phenomena occurring on Fe-Cr steels. Moreover they have been the basis for further R&D activities included in the JPNM and the MatISSE project.
The GETMAT results contribute also to the development of other technical areas as welding and joining of ODS alloys and the development of corrosion protection barriers for structural materials to be used in HLM-cooled systems.
Finally, within the GETMAT project high priority has been given to Education and Training. The scientific program of GETMAT has allowed the involvement of about 20 PhD students with experimental and theoretical work being part for their respective dissertations. Moreover, educational activities have been further supported through the organisation of two summer-schools and two dedicated workshops. The schools were organised in cooperation with other EU projects like PERFECT60 and MATTER and the workshops had a wide international participation, beyond the European boundaries. These educational activities have been very successful in terms of attraction of both well re-known lecturers in the field of materials science and of a quite significant number of young students.
Project Context and Objectives:
- Project context
The key elements at the origin of the GETMAT project have been identified through a thorough analysis performed on innovative reactors that are of relevance within the European Framework. In particular, reactor systems developed for advanced nuclear fuel cycles as well as advanced LWR were essential. Advanced nuclear fuel cycles include technological approaches that allow an optimised use of resources and a safe and optimised radioactive waste management. Nuclear reactors that are able to respond to these objectives are those with a fast neutron spectrum. Two categories of fast neutron reactors have been gaining attention in Europe, i.e. the sub-critical accelerator driven system (ADS) and the fast reactors (FR). Moreover, Supercritical Light Water Reactors (SCWR), considered as technological advancement of current LWR technology, were included in the analysis. A number of EURATOM supported projects (e.g. EUROTRANS, ELSY/LEADER, ESFR, GoFastR, HPLWR) that were addressing the design of these reactor systems and related materials as well as technological issues have been included and evaluated. The analysis was conducted in order to identify key reactor components for which advancement in the area of nuclear materials resulted to be crucial for the envisaged innovative design solutions. It turned out that the cores and primary systems are the most challenging components for what concerns materials requirements, thus materials selection and qualification. In general, all innovative reactor concepts foresee operational temperatures which are higher with respect to the temperature ranges of LWR (the current technology). Therefore, it can be assumed that all components of these reactors would be exposed to more severe conditions as compared to the established technology. However, the highest temperatures as well as dose rates can be expected at core components and primary systems levels. Therefore, the focus of the GETMAT project has been on these components. The analysis has also shown that candidate materials for cores and primary systems of almost all reactor concepts considered were austenitic and 9-12Cr Ferritic/Martensitic (F/M) steels. However, limits of these two classes of steels could have been identified. These limits impose design constraints that might not be in line with the original objectives declared for an enhanced sustainability and for innovations. Exemplary of these limits and their consequences especially for the fast neutron systems is the selection of the fuel clad materials. The austenitic steel option imposes an upper fuel burn-up limit due to irradiation swelling and embrittlement phenomena that occur on this class of steel for high irradiation doses (as discussed in past FR development programs). The limitation of the fuel burn-up might considerably impact strategies related to sustainability as e.g. the optimisation of use of resources. Limits can be as well identified for the 9-12Cr F/M steels as clad material. Indeed, the temperature window of these steels in an irradiation environment might be limited in the lower range to temperatures above 350°C and in the upper range to temperature below 550°C. The lower limit is due to irradiation hardening and embrittlement phenomena that can occur below 350°C (as discussed e.g. in the SPIRE project) and the upper limit is due to poor thermal creep and stress-to-rupture resistance of these steels above 550°C (as discussed e.g. in the EUROTRANS project). These temperature limits might impact the realisation of innovative design solutions as well as efficiency and economic viability of the reactor systems.
Therefore, within GETMAT the possibility to use oxide dispersed strengthened (ODS) alloys with a ferritic or F/M steel matrix has been proposed as an alternative option to the conventional steels. The focus on ODS alloy was based on data reported in the scientific literature, where it is discussed that these alloys show improved irradiation swelling and improved thermal-creep resistance in the high temperature range. The envisaged applications of ODS alloys were the fuel cladding for all considered reactor systems, i.e. ADS, FR and SCWR, and some structural components of the core as e.g. core support plate. Within this context properties of the ODS alloys that are of relevance for the given applications (and for all reactor systems) have been investigated. The preliminary assessment of the ODS alloys as materials for nuclear application, has been rounded up with an investigation of possible techniques able to join/weld these materials. Moreover, specific tasks related to environmental effects (corrosion and potential mechanical properties degradation) and options for mitigation of these effects through e.g. the development of corrosion protection barrier have been addressed as well. The environmental effects have been analysed for molten lead, helium with controlled impurities and supercritical water.
The development, optimisation and qualification of ODS alloys for nuclear application is considered as a long-term activity due to the fact that knowledge and relevant data on these materials for the nuclear applications are quite scarce. Therefore, selected needs for short- and medium-term applications were taken into account as well. The short- and medium-term plans foresees the availability in Europe of prototype and demonstration FR and ADS systems. For these systems conventional steels as 9-12Cr F/M and the austenitic steels AISI 316L, 15Cr-15Ni/Ti are candidate materials for both core components and primary systems. The GETMAT contribution to the design and construction of the prototype and demonstration systems occurred through the behaviour assessment of these candidate materials in different irradiation fields and in presence of relevant environments (heavy liquid metals (HLM) Pb and Pb-Bi eutectic, sodium and gas atmosphere). The activities were defined such as to take advantage, through post irradiation examinations (PIE), of relevant available and completed irradiation campaigns that were performed in past European projects and national programs. Moreover, it was also possible to include in the project a new irradiation campaign in a fast neutron spectrum and HLM environment and the related PIE.
A further topic considered as cross-cutting for all innovative reactor systems is the physical understanding of materials behaviour in terms of microstructure and mechanical properties evolutions in irradiation, temperature and stress fields. The physical understanding of material behaviour foresees the prediction of basic mechanisms, from atomic to the microscopic level, which determine materials response to applied thermal and mechanical loads, while being exposed to neutron irradiation. This research area includes the development of both physical models and computational tools for the different scales (from atomic to macroscopic level). Moreover, the theoretical activities are supported by dedicated model experiments. Since the multiscale modelling and experimental validation of nuclear material is a quite wide area, modelling and experimental activities included in the GETMAT project have been defined on the basis of the research direction in terms of composition and properties identified for the candidate alloys and steels. Indeed, the focus of the modelling activities was on Fe and Fe-Cr alloys (representative of F/M steels) and key aspects included in the investigation were thermo-chemical data (phase diagram of Fe-Cr) and irradiation hardening and embrittlement of these alloys. The theoretical activities were accompanied by dedicated irradiation experiments (in both ion and neutron fields) to assess assumption and models.
The overall challenges and research area of the GETMAT project can be summarised as follows:
• ODS alloys procurement and their characterisation
• Assessment of joining and welding techniques for ODS alloys
• Development and definition of corrosion protection barriers
• Improvement and extension of conventional steels (9-12 Cr F/M and austenitic steels) qualification
• Improved physical understanding of Fe and Fe-Cr alloys in irradiation and temperature fields and validation experiments.
- Project objectives
To address these challenges the GETMAT project has been structured into four work-packages and for each work package specific objectives have been identified. These objectives are hereafter summarised.
WP1 - Metallurgical and mechanical behaviour
The main objective of this work package is to provide reference material and evaluate the basic physical and mechanical properties for the concepts under investigation, that is, in Gen-IV and Transmutation relevant environments.
WP2 - Materials' Compatibility with coolants
This work package is divided into three main tematic areas: compatibility with different coolants of ODS steels and reference materials, corrosion barriers development and mechanical tests of reference materials when in contact with the coolants themselves. Different coolants have been considered: Pb, LBE, impure He (VHTGFR reference atmosphere) and superctitical water.
WP3 - Irradiation behaviour of structure materials
The objective of WP3 was to enlarge the database concerning the effect of irradiation on the relevant properties (i.e. needed for the design, such as dimensional and mechanical properties) of materials selected for core/target components of future GENIV reactors and ADS. The materials of interest were primarily FM steels, in particular 9Cr steels such as T91 (9Cr1MoVNb) and EM10 (9Cr1Mo) which are reference materials for different nuclear applications, however data on Fe-Cr based ODS alloys and austenitic steels were obtained as well. These data were generated as part of Post Irradiation Experiments (PIE) performed after irradiation campaigns conducted in various radiation environments, in some cases with the investigated materials irradiated in contact with aggressive coolants. This work package was organised in six different tasks, each task being devoted to the PIE of a particular irradiation experiment (with the noteworthy exception of task 3.5 which included both the design and implementation of a new irradiation campaign called LEXURII, followed by the corresponding PIE). The main results obtained within WP3 are presented hereafter and were discussed according to the following topics:
- Behaviour of 9Cr martensitic steels, in terms of dimensional stability and evolution of mechanical properties, after high dose irradiation in a fission environment: this issue is of particular interest for the core components of future GENIV reactors such as the hexagonal duct (wrapper tube) of the fuel assembly. As part of WP3, data up to a maximum dose of 155 dpa were obtained (Boitix9 irradiation in Phénix reactor).
- Effect of irradiation in a spallation environment on the mechanical properties of martensitic steels and ODS alloys: core and target components of ADS will be irradiated with high energy protons and spallation neutrons. In such environment, the combined effects of displacement damage and transmutation-induced impurities, in particular large quantities of helium, are of special concern. Via mechanical tests and microstructural investigation performed after irradiation in the SINQ target, the behaviours of FM steels and ODS alloys were assessed following irradiation in a spallation environment to a maximum dose of about 20 dpa and an accumulated helium content of about 1800 appm.
- Combined effect of irradiation and contact with the coolant: possible synergistic effect of irradiation and embrittlement due to the coolant is a major issue, in particular for systems cooled by heavy liquid metals, and published results were so far extremely scarce. This topic was addressed via the PIE of IBIS (HFR reactor), ASTIR (BR2 reactor) and LEXURII (BOR60) which included tests performed in the liquid metal (liquid Pb-Bi or Pb) on specimens irradiated in contact with the coolant. Furthermore, the PIE of the MEGAPIE target provided unique data regarding the combined effect of irradiation in a spallation environment, in contact with flowing liquid Pb-Bi and with cyclic thermal/mechanical loading on the properties of T91 steel.
WP4 - Multiscale modelling and model experiments on irradiated Fe-Cr alloys
The objective of the WP was to develop physical models to describe the behaviour of iron-chromium alloys when subjected to thermal ageing and irradiation. These alloys were chosen as model materials for high-chromium ferritic/martensitic steels. The processes of interest were nanostructural and microchemical changes that occur under irradiation or ageing and the corresponding changes in mechanical properties, mainly hardening, which in turn causes embrittlement. The approach used was computer-based multiscale modelling (from ab initio to dislocation dynamics), integrated by the performance of specific modelling-oriented experiments. The latter provided detailed data on which models can be contrasted. They also suggested, in combination with fundamental atomistic calculations, physical mechanisms that the nanostructure and microchemical models, as well as mechanical models, needed to incorporate. The link with other work-packages is ensured first of all by the comprehension of the mechanisms that govern the changes induced by irradiation in iron-alloys with varying content of chromium, which allows experimental results on steels performed in the rest of the project, especially WP3, to be better rationalised. Moreoever, the characterisation of technological properties, such as fracture toughness, in model alloys was included in some tasks.
The ultimate objective of the WP was to understand which nanostructural features are the main cause of radiation hardening and embrittlement and to develop a physical model capable of rationalising the experimental data available on this issue. This goal was indeed reached.
Project Results:
WP1 - Metallurgical and mechanical behaviour
Fabrication of ODS alloys is performed following classical pulvi-metallurgical (PM) processes that involve different steps as mechanical alloying (MA), consolidation by HIP and/or hot extrusion, and final thermo-mechanical treatments to achieve the requirements for mechanical performance. This fabrication procedure is expensive and the final quality depends strongly on a large amount of parameters (mechanical alloying time and velocity, consolidation temperature, and final heat treatments) that gives some uncertainties corresponding to the repeatability of heats and increase the fabrication costs. Within the WP1, ODS alloys were produced by PM processes but other techniques that are bases on more conventional steel production, as melting and casting, were explored. Unfortunately, these melting techniques do no allow the production of quality ODS alloys mainly due to the agglomeration of the oxide particles.
Three ODS alloys with different Cr contents (14Cr, 12Cr and 9Cr) where produced by MA and hot extrusion, whereas the trials to produce a P91-ODS alloy consolidated by HIP do not reach the expected results, due to the porosity of the heats that leads to the cracking of the slabs during hot-rolling.
The microstructure of the hot-extruded 14Cr ODS bar is constituted by small elongated grains (cigar shape) along the extrusion direction with a preferential crystallographic orientation of the grains along <110> in ED. This anisotropic microstructure is typical of ODS alloys and affects the mechanical properties as it will be explained later on this report. Nano (Y-Ti-O) phases are uniformly distributed all over the grains and Cr-rich phase decorating some grain boundaries were also observed. The 12Cr ODS plates show a quite different microstructure, with a bimodal grain size distribution: small grains size < 1 micron heavily deformed and some elongated grains size up to 200 microns. The biggest grains have not got a preferential orientation, while the smallest grains seem to have a preferential orientation along <110> in the RD. Both the 12Cr ODS and the P91-ODS show elongated pores along the extrusion direction.
Tensile properties have been determined up to 850ºC by different labs. In general the results form different labs are consistent and the GETMAT 14Cr-ODS and 12Cr-ODS alloys show lower strength than the well-known 14YWT and 12YWT fabricated by ORNL. Interesting to note that all the GETMAT ODS alloys show a peak in the elongation at 600ºC, typical for similar ODS alloys. This enhanced ductility at 600ºC can be attributed to a change on the deformation mechanism from intraganular to intergranular mechanisms, as is reported my Praud et al, and it is also related to a inverse strain rate sensitivity showed by the 14Cr ODS alloy.
The effect of the microstural anisotropy on the tensiel strength is no so evident for temperatures up to 600ºC, for higher temperatures, the transverse orientation is weaker than the longotudinal one. The effect on ductility is larger and more pronounced for the total elongation and reduction in area values, where the loss of reduction in area for the T orientation in respect of the L is around 70% at room temperature and around 50% at 700ºC.
Impact properties of all the ODS alloys has been also obtained by testing small KSLT specimens. In general the DBTT and absorbed energy are strongly dependent to the microstructural features, showing againg the effect of the microstructual anisotropy on the mechanical properties. The observed difference in the values of absorbed energy is a consequence of the relative orientation of the notch regarding to the elongated microstructure. When weak interfaces are present parallel to the longitudinal direction of the impact test bar, the interaction between the weak interfaces and the stress field that is generated by the localized plastic constrain at the notch and/or the crack tip can cause splitting. The two basic geometries are termed ‘‘crack divider’’ and ‘‘crack arrester”, In the crack arrester geometry, the delamination is thought to relax the triaxial stress conditions and to blunt the crack tip. To fracture a material, crack re-initiation is necessary and occurs under conditions of nearly uniaxial tension, which is an unfavourable cleavage. Hence, high absorbed energy is obtained through the delamination in LS and TS samples by the well-known delamination toughening mechanisms. The P-u curves obtained from instrumented Charpy tests are very useful to investigate the effect of splitting on the DBTT. Meanwhile the P-u curves of LT and TL samples present similar behaviour regardless of the test temperature; P-u curves for LS and TS samples present a peculiar behavior. For LT and TL samples, at which ductile fracture surfaces were observed, P gradually decreases after a maximum value Pm is reached as expected. On the other hand, for LS and TS samples, at which extensive delamination was observed, P-u curves exhibit load drops that can be associated to crack arrest, and loads plateaus that can be due to re-initiation of new cracks. In other words, the crack is blunted by delamination. As a result, the absorbed energy becomes larger by the delamination effectct. Moreover, because of the blunted crack by delamination, the sample behaves as an unnotched one during the test.
This anisotropy on the mechanical properties is also observed on the fracture toughness tests, where the TL specimens show very low fracture toughness values at 550ºC and 650ºC.
The issue of the low impact and fracture toughness resistance of the ODS alloys do not restrict their use at high temperature, because these properties are not used on the design of high temperature components, but from a safety point of view, it has to be resolved to assure a safety operation at lower temperature close to room temperature.
High temperature behaviour of the 14Cr and 12Cr ODS alloys have also charactericed by uniaxial creep and low cycle under fully reversed conditions (R = -1) in strain control with a strain rate of 0.2%/s in air and in inert atmosphere. Small punch creep tests were also performed.
The creep curves for the GETMAT 14Cr ODS steel are characterized by very slow secondary creep and sudden rupture at very low deformation with little evidence of tertiary creep. Similar observations have also been reported for other ferritic ODS steels. For a given stress level, the rupture times obtained from the creep tests were found to exhibit considerable scatter. When a test was repeated with the same or nearly the same load, a rupture time differing from the original test by more than one order of magnitude could be recorded. The large scatter in the rupture times is not so much caused by the stochastic nature of a brittle final fracture event, as the documented absence of tertiary creep might suggest, but rather by scatter already in the steady state creep rates.
Fatigue tests have been carried out at 650 ºC and 750 ºC. The alloy is cyclically very stable with practically no hardening/softening. Results from the tests at both temperatures can be described by a common power law. An increase in the test temperature has little influence on the fatigue ductility exponent. For a given total strain level, the fatigue life of the alloy is reduced with increasing temperature. Creep fatigue tests (with 10 min holdtime) show smaller number of cycles to crack initiation, there is nearly no difference between results of tests carried out under inert and air environment.
Irradiation creep and microstructural changes of the 12Cr-ODS and 14Cr-ODS steels have been studied by homogeneously implantation with helium under uniaxial tensile stresses from 40 to 300 MPa. The maximum dose was about 1.2 dpa (5000appm-He) with displacement damage rates of 1 x 10-5 dpa/s at a temperature of 300 ºC. Irradiation creep compliances were measured to be 4.0 x 10-6 dpa-1 MPa-1 and 10 x 10-6 dpa-1 MPa-1 for 12 and 14Cr ODS, respectively. No remarkable effects of Cr content, grain size and dispersoid size on irradiation creep properties are observed. The apparently slightly higher compliance value of the 14Cr ODS cannot be unambiguously related to any material property or microstructur al evolution.
Microstructural evolution was studied in detail by TEM observations. Dislocation loops and helium bubbles are distributed homogenously in the matrix. In the case of high density fine dispersoids, most bubbles are attached to ODS particles. This may suppress loop formation as well as growth of bubbles, thereby increasing the resistance of ODS ferritic steels against helium embrittlement.
Different welding techniques have been explored within GETMAT project: electro magentique pulse technique, diffusion bonding, explosive bounding and friction stir welding. Electron beam and TIG welding techniques have also used.
Diffusion bonding has been performed with PM2000. Hardness measurements of heat treated samples of PM2000 show, that heat treatment above a temperature of 1200 ºC causes a degradation of mechanical properties. TEM investigations as well as light microscopy reveal, that neither a constant particle size distribution nor the changes in the grain sizes of small grains can be identified as a clear reason for the temperature induced loss in hardness. The diffusion bonding model of Hill and Wallach was adapted for ODS materials by using the creep equation for ODS materials introduced by Rösler and Arzt. The influence of pressure, temperature and surface roughness on the bonding time were investigated. For a diffusion temperature of 1100 ºC, a pressure of 50 MPa and a surface roughness of b = 2.5 lm and h = 1.2 lm a similar UTS and Charpy impact energies compared to the bulk material were achieved.
Electro magnetic pulse technique has been used to weld T91 cladding tubes. The sample consists of the cladding tube and an end cap of a specific design, which is entered in the tube. In particular, the angle of collision between the moving component (the tube) and the static central end cap is considered carefully. This angle has a strong influence (supplier's information) on the interaction between the components during the deformation but it also determines the flight distance of the tube prior to collision. This distance is critical in allowing an optimization of the acceleration achieved by the tube as a result of the electromagnetic pulse. The objective of defining a parameter window for the joining of advanced cladding materials (T91 and ODS steel) for Gen IV reactors has been achieved. Future tests will be performed to close in on the optimum conditions. Due attention can be given to a more precise angle of collision and surface roughness, as well as the investigation of the effects of the process on the material's microstructure and properties.
Linear Friction Welding (LFW) is a solid state joining process in which a stationary material part is forced against a part that is linear oscillating to generate friction heat. Important parameters are oscillation frequency and amplitude, friction and forge pressure and burn-off. The joining of 12 Cr ODS steel by LFW has been successfully demonstrated. A clear thermo-mechanically affected zone (TMAZ) with deformed and more refined microstructure containing uniform dispersion of particles was observed in both longitudinal and transverse directions. The micro hardness changes from 370 Hv in the parent material to 330 HV in the TMAZ are observed. The minimum hardness values, 170-190 Hv, occur very close to the weld interface (probably due to dynamic recrystallization).
Explosive Welding (EW) is a solid state process where welding is accomplished by accelerating one of the materials parts at extremely high velocity through the use of chemical explosives. Final explosive welds of 14Cr ODS (produced by CEA) are performed to investigate the welding window for these type of materials. A wide weldability window (between 2000 and 4000 m/s, collision point (welding) velocities) exists for welding 14Cr ODS steel. Due to extremely high strength and low ductility, explosive welding should be done with pre-heating to 200 oC and increased impact velocity (height of explosive layer) compare with conventional steels.
TIG (Tungsten-Inert-Gas) and EB (Electron Beam) welds of P91, P91-ODS, P92 and PM2000 were performed. Dissimilar welds of material with different chemical composition (P91/PM2000) show that a connection is possible, but the Charpy results are more brittle than each of the base material.were be performed and tested by Charpy and creep tests. 3 tests of EB welding of ODS P91 with ODS P91 failed. The tests varied the welding requirements. The results of the P91-PM2000 EB weld show the feasibility joining these two materials, but the test results let advise against this kind of connection. Tests of welding ODS P91 with P91 show the same porosity as the similar welds. Welding P91 using EB is a reasonable joining technology.
WP2 - Materials’ compatibility with coolant
This work package is divided into three main tematic areas: compatibility with different coolants of ODS steels and reference materials, corrosion barriers development and mechanical tests of reference materials when in contact with the coolants themselves. Different coolants have been considered: Pb, LBE, impure He (VHTGFR reference atmosphere) and superctitical water.
The present chapter is thus divided into three paragraphs.
WP 2.1 Compatibility of materials with the coolant and corrosion experiments.
GenIV fast breeder reactors might be cooled by means of liquid metals, gas, supercritical water or molten salts. GETMAT FPVII program focussed mainly on HLM (Heavy Liquid Metals) and on gas and SCW. In the frame of this WP2 area, many institutes performed compatibility experiments of the GETMAT ODS steels both in flowing and stagnant environments. Not only the durations of the tests were different, but the experimental conditions too:
• ENEA: flowing tests in the LECOR loop facility, T= 500°C, flow speed 1 m/s, Oxygen 10-6wt% of 15-15Ti ss up to 8000 hours, ODS steels, up to 5000 hours
• CIEMAT: stagnant tests of 9Cr and 12 Cr ODS in Pb at 600°C and 700°C, at 10-5wt% and 10-7wt%, up to 2000 hours
• CNR: stagnant tests in Pb at 650°C and 700°C of ODS steels and wetting tests
• KIT: corrosion tests of welded materials at 10-6wt% and 10-8wt%.
As above said, the materials to be tested were the GETMAT reference ones, with different compositions.
Where the first two steels are KOBELCO Japan commercial ODS steels, the third the CEA 14Cr ODS manufactured for GETMAT, the last the 15-15Ti DIN commercially made by Sandvik.
Considering the wide amplitude of the experiments and the different approaches in performing them, the outcomes and experimental data were surprisingly similar. A good example of compatibility with Pb and data comparison might be depicted in the micrographies.
Considering the above depicted examples on the 12 Cr ODS materials, a first qualitative observation is possible: the oxidation mechanism of the material appears very similar between the flowing and the stagnant experiments. Many randomly distributed different experimental results were function of the changeability of the original steels’ properties and characteristics.
A synthetic analysis for all the ODS materials might also be drawn.
All the samples tested by difeerent laboratories, do not show any mass loss due to massive corrosion phenomena. The oxidation behavior the three steels are indeed very similar, however some peculiarities of each sample can be highlighted.
The following scheme summarizes the three different behaviors:
• ODS9Cr steel shows an internal oxidation associated to the growth of a thin Fe-Cr-O layer on the surface.
• ODS12Cr steel shows a Fe-Cr-O layer growth on the steel surface and some spots of Cr-O oxide. Any internal oxidation has been detected.
• ODS14Cr shows a double oxide layer; a outer Fe-Cr-O layer and an inner Cr-O layer.
The thickness of the Fe-Cr-O layer ranges, averagely, between 2 μm for ODS9Cr and 10 μm for ODS14Cr showing a slight dependence on the Cr-content in the steel matrix, as well as the Fe-Cr-O composition. The same behavior is shown by continuity of the layer along the interface, in fact ODS14Cr presents a more compact and continuous layer than the other two steels.
The internal oxidation observed in the ODS9Cr steel can be associated to its fine grained structure. This kind of structure can promote the internal oxidation as already reported in literature.
A wide range of tests have been carried on welded materials too. A summary output of the work states that if the chemical composition and the grain size in the weld region (due to a heat treatment) are similar the welded area behaves like the bulk material: in 9Cr F/M steels welded by EB or TIG and one of the friction stir welds.
The smaller the grain size, the lager the oxide layer will develop in 10-6wt% oxygen containing Pb (friction stirr welds 20 and 21 and friction stirr with heat treatment).
Precipitations of elements at the grain boundaries and/or voids can lead to a fast growing oxide or a stronger dissolution attack, like observed at for the P91/PM2000 EB welded specimens.
Changes in the grain orientation and dislocation also may lead to a different behavior of the welding region, like observed at the specimens welded by explosive welding.
Independent on the welding method an oxygen concentration of 10-8wt% is a critical condition for protective oxide formation on welds and bulk material because at least partly dissolution attach must be expected.
HLM are not the only environment consedered as candidate coolant for GenIV reactors. As already mentioned, impure He has been taken into account either.
Three institutes were involved in the performance of He corrosion of 14Cr ODS only, at different temperatures, up to 1000 hours of experiment. The chosen atmosphere, depicted in table 2, is indeed not referring to the GFR GenIV reactor “Allegro”, but to the Very High Temperature GFR one, already very well known and studied in formerly exixting CEA facilities.
ENEA and CIEMAT manufactured dedicate dfacilities to perform the tests, whilst CEA adapted the Coralline loop. The apporaches were quite different: ENEA and CIEMAT obtained the reference atmosphere by means of commercially available tanks, CEA mixed the impurities in situ with a dedicated ingas complex system.
The exposures for all experiments were 250, 500 and 1000 hours, at 750°C, 850°C, 900°C, with a flow rate of 100 l/h ca.
The results gained in the three experiments, validate that the 14Cr ODS did oxidize in the test conditions as expected when the helium composition was selected. It was observed for other high chromium alloys that a sufficient CO partial pressure supports the growth of a dense chromia scale. Based on the general understanding of the surface reactivity in impure helium, it is expected that the surface chromium oxide protects the alloy against reactions with carbon bearing species, carbon monoxide and methane in this case, and thus prevents carburization.
Compatibility of ODS alloys under gas-cooled reactor conditions is virtually unknown. In addition, papers on the high temperature corrosion behavior of Fe-Cr ODS steels are scarce. These alloys were developed as structural materials for nuclear fusion systems but compatibility with environment has rarely been investigated. Few studies referred to other nuclear applications or to the fundamental understanding of their oxidation mechanism.
As far as the oxidation mechanism is concerned, the influence of rare earth elements (RE's) such as yttrium, neodymium and praseodymium on the oxidation behavior of a Fe-Cr alloy has been studies elsewhere A small addition of these REs reduces the oxidation rate. Analysis of the surface oxide morphology revealed that the morphology of oxides formed on Fe-Cr specimens with and without RE's specimens is different. In fact, a fine-grained oxide morphology was observed for alloys with RE's addition which may improve the scale adhesion. Also, it was found that REs increase oxidation selectivity for chromium. The critical Cr content for chromium oxide formation was reduced when REs were added. Basic comparison of the mass gains and scale adhesion for Fe-14Cr steels with and without Y is consistent with a smaller oxidation rate and higher quality scale for the ODS alloys.
Based on these preliminary results, 14Cr ODS steel exhibits a promising corrosion behavior in oxidizing impure helium at 800, 850, 900°C. This comment is important if one thinks that the alloy has yet not been optimized regarding its oxidation properties. However, significant work would still be needed in order to validate the environment compatibility of this material for use in a GFR. An experimental qualification program would at least comprise: parametric studies to establish oxidation kinetics as a function of the environment parameters based on medium to long term oxidation tests, understanding of the oxidation mechanisms to get inputs for lifetime modeling including material aging and long term effects, experimental validation on long duration oxidation tests, investigation of the corrosion behavior in off-normal conditions (temperature outburst, out-of-specification coolant chemistry) and possibly an optimization of the alloy as regards its corrosion performances.
SCW was the third reference environment considered in this WP, work perfomed by VTT only. The tests were done in SCW at two temperatures, 550°C and 650°C. The dissolved oxygen content in the feed water was between 100 and 150 ppb and conductivity was <0.1 μS/cm. The specimens were exposed to SCW at the pressure of 25MPa in supercritical autoclave connected to a recirculation water loop. The environmental control and monitoring in VTT SCW testing system include temperature, pressure, inlet and outlet water conductivity, inlet water dissolved oxygen content, and flow rate.
Based on the weight change testsand data analyses, Al-alloyed high Cr ODS steels PM2000 and MA956 have superior corrosion resistance when compared to 9-14% Cr ODS steels. The oxidation performance of the high Cr Al-alloyed ODS steels result from the formation of a protective Al–Cr–Fe-rich surface oxide layer. On ODS steels with lower alloying element concentrations, i.e. with no Al, the film is less protective with an iron rich outer and iron and chromium rich inner layers. Based on this work, in SCW conditions at 550°C and above, the Cr-content for thin walled ODS components must be higher than that of the 14%.
The results indicate that the ODS steels are less susceptible to SCC than austenitic stainless steels or Ni-base alloys. However, ODS steels have been tested very little in SCW and more data is required in order to make any final conclusions.
WP 2.2 : Advanced corrosion barriers development.
Surface modified layers based on FeCrAl coatings have shown their general capability to protect steels against the influence of liquid lead alloys by the formation of slowly growing stable oxide scales. This Wp2 task focussed on a commercial method, HVOF, and on the GESA method.
In a sense, the goal of this task was twofold: a critical analysis of two different deposition methods, an in deep study of the best compounds to be deposited. This latter was the most important output.
Because the best suitable FeCrAl alloy composition to ensure this protection over a wide temperature range was not addressed so far in a dedicated experimental work. The surface alloying process (GESA) is not the most appropriate to produce samples having a controlled change in composition. Therefore, model alloys were produced by Arc- Melting to evaluate the Al and Cr content required for protective oxide scale formation.
The phenomena occurring during the Fe-Cr-Al alloys exposure to HLMs, containing very small amounts of dissolved oxygen, is scarcely documented .No systematic investigation of the corrosion behaviour of the Fe-Cr-Al system, when exposed to HLM was performed, in order to find the minimum Al content for the formation of an alumina scale. The current paper presents the results of a systematic study concerning the corrosion behaviour of Fe-Cr-Al-base model alloys during their exposure to oxygen-containing (10-6 wt. %) lead in the temperature range 400 - 600 °C. In this vein a map of the alumina stability domain for this temperature range has been drawn in Fe-Cr-Al ternary diagrams and the type of alumina polymorph, formed at different temperatures, was determined.
Exposures were performed up to 600°C. Based on the state of the art oxide maps concerning oxidation behaviour of Fe-Cr-Al-base alloys at 800 and 1000°C in oxygen atmosphere, 10 compositions, belonging to this alloy system, were designed in order to tap the borders of the alumina stability domain, during their exposure to oxygen (10-6 wt%) containing lead, at 400, 500 and 600°C.
Eight alloys, Fe6Cr6Al, Fe8Cr6Al, Fe10Cr5Al, Fe14Cr4Al, Fe16Cr4Al, Fe6Cr8Al, Fe10Cr7Al and Fe12Cr5Al, were found to be protected against corrosion in oxygen containing lead, either by a duplex layer (Fe3O4 + (Fe1-x-yCrxAly)3O4) or by (Fe1-x-yCrxAly)3O4, depending on the temperature at which they were exposed.
Two alloys namely Fe12Cr7Al and Fe16Cr6Al were found to form transient aluminas Al2O3 (at 400 and 500°C) and Al2O3 (at 600°C), as protective oxide scale against corrosion in oxygen containing lead. An oxide map illustrating the stability domain of alumina, grown on Fe-Cr-Al- alloys when exposed to molten, oxygen containing lead, was drawn. The alumina stability domain border shifts with lower temperatures to higher chromium and aluminium concentrations. When CCr ε (12-25 wt%), the minimum concentration of Al, required to form alumina scale on Fe-Cr-Al alloys exposed at 400-600°C in molten Pb, is:
C Al =1.523978 – 0.80805 (CCr) + 0.01561 (CCr)2 [wt%].
For the temperature range and exposure times used during the current evaluation, the growth rate of the alumina scale was low. No area with detached scale was observed and no trace of Al2O3 was detected. Lower aluminium concentrations leads to the formation of Fe-base spinel-type oxides protrusions on extended area. However, this can also happen at high aluminium concentration (7.6 wt%) if the chromium content is lower than 10.4 wt%. The map includes also additional points, extracted from literature and corresponding to alumina forming alloys, when exposed to HLMs, which fit very well with our findings. The influence of the model alloys grain size, of the addition of reactive elements and of longer exposure time on the alumina scale polymorphs type, growth, stability and properties is currently under evaluation.
WP 2.3 Mechanincal testing in environment
Mechanical tests on the GETMAT materials were carried on in Pb, He and SCW too. The performance of mechanical tests in different environments leads to very different approaches and obstacles.
Creep tests in contact with pb were carried on by CIEMAT on 9Cr ODS.
The creep-to-rupture behaviour of the 9 Cr ODS was performed in air and in liquid lead under non-controlled atmosphere at 600oC and also at controlled condition at 650oC, as well as their fractures was examined after the testing. The results show an important effect of liquid lead in the creep behaviour, principally lowering the rupture time of the specimens in liquid Pb under non-controlled conditions. On the contrary, creep test performed at KIT under oxygen controlled conditions show similar tendency in air and in Pb like 12-14CrODS.
A possible reason could be the formation of a thick, porous and brittle oxide layer that breaks during the tests and allow an intimate contact between the liquid and solid metal, leading to a liquid metal embrittlement mechanisms. A deep characterization of the oxide layer formed on the specimens tested under oxygen saturated conditions is on-going to confirm this hypothesis.
Pb has been choosen by ENEA and KIT with the goal to perform fretting experiments, over the 14Cr ODS GETMAT material at ENEA, on GESA and 15-15Ti at KIT. The experimental campaigns differed essentialy in the approach to the experimental device, being carried on both at 500°C, in Oxygen saturated Pb, on teh same material with comparable loads and fretting frequancies. Figures 12, 13, 14 and 15 depict some of the interesting outcomes of the two tests. While the FRETME machine at KIT was able to perform several tests, comparing the reference matarial with others, the ENEA GIORDI device could perform fretting experiments without any debris re deposition phenomena. For all tests, the main goal is to simulate the interaction between the spacer grid and the fuel cladding of the GenIV LFR.
As a common summary of conclusions, it may be stated what follows:
• Favorable load-amplitude combinations can minimize the fretting severity. The fretting penetration and the wear rate decrease with an increasing applied load (after the turning point) and with a decreasing sliding amplitude. As dis¬played by the published fretting maps there are load-amplitude combinations (e.g. load > 75 N and amplitude < 15 µm) that favors the evolution of the fretting contact towards the stick regime. These load-amplitude combinations are expected to lower the specific wear coefficient (and consequently the wear and pene¬tration rate) by up to one order of magnitude compared to e.g. the combination 50 N-15 µm, which in Fig. 36 corresponds to the lowest measured penetration rates. Such conditions can be established with an adequate assembly strategy.
• Suitable material improvement and corrosion counter-measures im¬prove the wear resistance and mitigate the corrosion enhancement of fretting.
• A reduction of the penetration rate up to about 50 % already after 150 h was achieved by using pre-oxidized components. Likewise, for the specific case of austenitic steels, the use of Ni-en¬riched Pb reduces the fretting corrosion severity.
• Additionally, surface treatment and Al-alloying (e.g. GESA-T91) im¬proves remarkably the wear and the corrosion resistance prop¬erty also at high temperature.
• Fretting is a self-mitigating process. Fretting becomes less and less severe with the time. After 930 h, the penetration rate is up to 70% lower than during the run-in period to which the data of Fig. 13 refer.
• The beginning of the fretting process is delayed with respect to the re¬actor start-up. In e.g. the reactor core, the fretting process starts when, due to the irradiation damage and thermal-mechanical loads, the components (e.g. springs, supports or spacer grids) aimed to fasten the fuel cladding tubes (or other parts) lose their mechanical and dimensional properties. Therefore, the fretting process affect the e.g. fuel cladding tube for less than 25000 h.
• Possible design solution like the installation of spacer grids and sup¬ports in the non-active area of the fuel (lower temperature and irradiation) can delay the beginning of the fretting process further.
It is possible to claim with a certain confidence that, the listed aspects will allow the achievement of a penetration rate complying with the operating requirement (0.002 µm/h), which is about one order of magnitude lower than the one measured experimentally (in the best cases from 0.039 to 0.063 µm/h for the three materials).
The small punch test campaign is the last to be quoted in Pb, peformed by RSE.
The characterization of the steels included a preliminary exposure to liquid lead into a controlled-flow rig, followed by a set of classical (tensile, Charpy and creep) and innovative (small punch) mechanical tests.
Difficulties in the production of ODS alloys led to delays in materials procurement, however in the first period of the project RSE activity wasn’t touched, as it concerned the development of the geometries of the samples, the specimens’ and fixtures’ machining and the characterization of P91 (historically one of the first steels having medium content of chromium and often considered as the reference steel of ferritic-martensitic type) by small punch technique. At the delivery time of first ODS material RSE then started a mechanical characterization of virgin material by means of small punch technique with some additional conventional tests if spare material was available. The preliminary characterization of ODS virgin material was later performed for the other two GETMAT distributed ODS steels.
This experimental activity gave good results in terms of test repeatability, and an analysis of material anisotropy was performed, reporting similar trends shown by a few conventional tests performed by other project Partners.
After exposure at ENEA rig LECOR of RSE specimens at 500°C with flowing liquid lead in controlled low content oxygen condition, the final characterisation of exposed materials has been completed allowing the following main considerations:
• No effects of long time exposure on tensile properties, neither when determined by small punch tensile for both 12 and 14 Cr ODS, nor when determined by uniaxial test as observed for 12 Cr ODS.
• Limited effect of exposure on creep rupture time reduction: 14 Cr ODS estimated by small punch creep show some tests with reduced rupture time but data can be considered into conventional creep scatter-band of reference virgin material data, 12 Cr ODS estimated by uniaxial creep tests shows a trend of slightly more pronounced reduction for rupture times but data are always into the scatter of reference virgin material. The observed creep resistance reduction can be related to the presence of penetrating oxidation on specimen surface enhancing localized stress concentration.
• Significant effect of exposure on material embrittlement observed for both ODS steels: a relevant increase of transition temperature estimated by small punch tensile for 14 Cr ODS and a relevant reduction of impact energy for 12 Cr ODS, the latest confirmed by a less relevant (than 14 Cr) but anyway evident increase of small punch tensile transition temperature.
LCF tests were conducted at RT and at 750 °C on 14 Cr-ODS steel specimens in simulated Gas cooled Fast Reactor (GFR) environment, already quoted. Based on the results of such tests, the following main observations can be drawn:
• the cyclic stress-strain curves were obtained for 14 Cr-ODS steel, both at RT and at 750 °C; best-fit relationships were derived from such curves, in order to represent the elasto-plastic behavior of the material under cyclic conditions;
• the general behavior of the 14 Cr-ODS steel during HTLCF tests conducted in simulated GFR environment does not show significant differences as compared appears to other rather similar materials;
• the increase in test temperature from RT to 750 °C appeared to produce a fatigue life reduction by a factor as large as 25.
• the HTLCF data for 14 Cr- ODS steel tested in simulated GFR environment appear to lay near the lower bound of the typical scatterband observed for 304 and 316 SS, when tested in air at the same temperature;
• this could be an indication of an actual lower fatigue strength of 14 Cr-ODS steel as compared to 304 and 316 SS in HTLCF regime, or of a detrimental effect of simulated GFR test environment as compared to air or both. It is however worth noting that the number of tests which were conducted is by far too low for above observations to be regarded as conclusive and/or soundly based. Therefore, a significant extension of the test program, also including for creep-fatigue interaction effects, is strongly recommended.
SCW work, performed again only by VTT was essentially part of the SCW corrosion tests.
The results indicate that the ODS steels are less susceptible to SCC than austenitic stainless steels or Ni-base alloys. However, ODS steels have been tested very little in SCW and more data is required in order to make any final conclusions.
The goal to wrap up such a complex Work Package and to derive conclusions is a hard task. For the corrosion experiments it might be stated that:
• ODS materials showed several similarities, even in very different experiments.
• The presence of a dual layered oxide scale over 9Cr and 12Cr Ods materials is normally detected both in stagnant and flowing conditions. The oxide scale appearance is porous, and the mechanical adhesion is quite poor and random.
• 12 Cr ODS showed unexpected results, in some stagnant tests, where a single layer was detected only on some samples.
• 15-15Ti steel showed good protection due to in situ oxidations, in flowing Bb.
• On weldings: If the chemical composition and the grain size in the weld region (due to a heat treatment) are similar the welded area behaves like the bulk material → in 9Cr F/M steels welded by EB or TIG and one of the friction stirr welds.
• The smaller the grain size, the lager the oxide layer will develop in 10-6wt% oxygen containing Pb (friction stirr welds 20 and 21 and friction stirr with heat treatment).
• Precipitations of elements at the grain boundaries and/or voids can lead to a fast growing oxide or a stronger dissolution attack, like observed at for the P91/PM2000 EB welded specimens.
• Changes in the grain orientation and dislocation also may lead to a different behavior of the welding region, like observed at the specimens welded by explosive welding
• Independent on the welding method an oxygen concentration of 10-8wt% is a critical condition for protective oxide formation on welds and bulk material because at least partly dissolution attach must be expected
• In impure He, ODS materials exhibited an “as expected” oxidation behavior, leading to very small differences between the three temperatures. The manufacturing and surface treatment of the samples had a key role.
General conclusions on the other two subtasks are impossible, due to the very different nature of the experiments.
WP3 - Irradiation behaviour of structure materials
• Effect of high dose irradiation in a fast spectrum on mechanical properties and dimensional stability of a 9Cr martensitic steel.
9Cr martensitic steels are known for the excellent radiation resistance (i.e. negligible void swelling, little hardening and moderate degradation of impact properties) when irradiated in a fast spectrum at temperatures above approximately 380°-400°C up to doses of at least 100 dpa. However significantly higher values for the end-of-life neutron exposure of in-core components such as the hexagonal duct, up to about 150 or higher, are envisaged in the case of future GEnIV reactors. Moreover, it has been argued that the excellent swelling resistance of FM steels is mainly due to a long incubation period (transient regime) and that FM steels should eventually swell at a steady state swelling rate which could be as high as 0.2%/dpa. Therefore, a comprehensive PIE programme was conducted using an hexagonal duct made of 9Cr-1Mo steel (EM10 grade) irradiated in Phénix reactor up to a maximum dose of 155 dpa (Boitix9 experiment). Tensile and impact specimens were machined at different locations along the duct, corresponding to irradiation temperatures ranging from about 385°C to 530°C and doses in the range 43-155 dpa. Mechanical tests confirmed the very good radiation resistance of EM10 steel for these specific irradiation conditions.Irradiation induced hardening for all tested specimens, including after irradiation to 155 dpa, was found to be relatively small. Likewise, a moderate degradation of impact properties after irradiation was observed.
As in other metallic alloys, the irradiation temperature has a drastic impact on the evolution of the mechanical properties. For instance at 325°C, in addition to formation of α’ precipitates, irradiation induced the formation of a high density of small dislocation loops, possibly decorated with solute atoms, which are the main source of the high hardening observed after irradiation (see WP4, T4.2). In the temperature range of interest for the Boitix 9 wrapper, the dislocation microstructure was found by TEM investigations to be very different and consistent with the measured hardening: at 398°C (112 dpa), a low density of large dislocation loops was observed, while at 439°C (155 dpa), hardly any loop could be detected and the dislocation density was found to be similar to that measured prior irradiation.
In addition to the effect of irradiation on mechanical properties, dimensional stability is also of prime importance. Macroscopic density measurement showed that void swelling was very small. This conclusion was confirmed by TEM investigations of the void microstructures. This low void volume fraction and the fact that voids were found to be heterogeneously distributed within the matrix are indications that EM10 steel was still in the incubation stage of void swelling even after irradiation to 155 dpa, which demonstrates the very high swelling resistance of this type of steel.
• Effect of irradiation in a spallation environment on the mechanical properties of martensitic steels and ODS alloys
FM steels and ODS alloys are candidate materials for the structures of spallation targets and core components of ADS. In order to evaluate the evolution of mechanical properties in the relevant radiation environment, tensile and impact specimens of various FM steels and ODS alloys were inserted in the solid spallation target SINQ (PSI) and irradiated with high energy protons and spallation neutrons at temperatures in the range [150-350°C].
For all investigated FM steels, the same tensile behavior was observed.
The specimens irradiated at the highest doses and helium contents displayed very high hardening and a fully brittle, predominantly intergranular fracture mode, as clearly demonstrated by the tensile curve and SEM fracture surface observations for the specimen irradiated to 19.8 dpa and 1750 appm He. This behavior, which is not observed for FM steels irradiated in fission conditions at the same temperature and to similar damage dose, is the result of i) matrix hardening due to the formation of small dislocation loops and a very high density of tiny helium bubbles (~ 3 1016 cm-3 according to TEM measurements) and ii) strong decrease of grain boundary cohesion due to the presence of helium (as shown by ab initio calculations). Likewise, measurements of impact properties after irradiation of FM steels in the SINQ target show a degree of embrittlement significantly greater than that expected after irradiation in fission conditions.
By contrast, ODS alloys displayed a better mechanical behavior after irradiation in spallation conditions. For example, 14Cr ODS (MA957 grade) irradiated in SINQ retained significant ductility when tensile tested at room temperature, whereas, as mentioned above, FM steels subjected to identical irradiation conditions exhibited a brittle intergranular fracture mode. The high density of nanoclusters in ODS alloys are believed to act as effective sinks for point defects as well as trapping sites for helium, thereby decreasing the irradiation induced hardening and protecting grain boundaries from helium embrittlement effect.
• Combined effects of Heavy Liquid Metals (HLM) and irradiation
As several advanced nuclear systems will use a HLM (liquid Lead or Lead Bismuth Eutectic) as coolant, the issue of possible combined effects due to irradiation and HLM is of great importance and has little been addressed in the past. This issue was therefore the focus of several tasks of WP3, in particular via the PIE of ASTIR, LEXURII and MEGAPIE. As part of ASTIR experiment, the behavior of T91 steel in lead Bismuth Eutectic (LBE) was investigated. Liquid Metal Embrittlement (LME) was observed in the case of T91 in the as-received condition. For instance, when tensile tested in LBE, the total elongation was reduced as compared to a test in an inert atmosphere (i.e. fracture occured earlier), while the shapes of the tensile curves in both environments up to the rupture strain were very similar. However, the necessary condition for LME occurrence is an intimate contact between the steel and the liquid metal. This was achieved in the as-received condition for instance by exposing the samples before tensile testing to LBE with low oxygen concentration, which resulted in a partial dissolution of the protective oxide layer typically found on the steel surface.
The effects of test temperature and strain rate on LME were studied in the case of unirradiated specimens and it was found that the reduction of the total elongation has a non-monotonic dependence on both parameters: LME occurs only in a particular range of temperature and strain rate. For a strain rate of about 5 10-5 s-1, the minimum of the ductility trough occurs at about 350°C.
In order to investigate the effect of irradiation on the LME sensitivity of T91, tensile tests were carried out both in inert atmosphere (Ar) and in LBE on T91 specimens irradiated to 2.6 dpa as part of ASTIR irradiation experiments. The tests were performed at 300°C, which was the average irradiation temperature.
A strong LME effect was observed for irradiated specimens tested in LBE. When tested in LBE, multiple cracks initiation occurred at the surface followed by brittle propagation. The average reduction of total elongation due to LME (about 63% reduction compared to tests in Ar atmosphere) was proportionally larger than the average reduction measured in the case of unirradiated samples. Furthermore, ASTIR specimens had not been irradiated in contact with LBE. While for unirradiated specimens, pre-exposure to the liquid metal was found to be a necessary condition of LME occurrence, this was not the case for irradiated specimens. Thus irradiation was found to enhance the LME sensitivity of T91. This conclusion was also supported by the results of PIE performed on T91 specimens irradiated in contact with LBE to about 6 dpa at 350°C (LEXUR II experiment). Also in this case, a drastic reduction of total elongation was observed for irradiated specimens tested in LBE at 350°C compared to values measured in inert atmosphere.
By contrast, for all investigated testing conditions, including tests on specimens irradiated to 2.6 and 6 dpa (as part of ASTIR and LEXURII experiments), austenitic stainless steels (316L and 15-15 Ti grades) did not show any evidence of LME due to LBE.
In addition to tests performed on irradiated specimens, WP3 included as well the PIE of the MEGAPIE target, which was a unique opportunity to investigate the behavior of a real component whose structural materials had been irradiated in contact with flowing LBE and subjected to cyclic thermal/mechanical loadings. It must be emphasized that target dismantling using a large dedicated bandsaw, LME out-melting with a specially designed oven, raw cutting of the PIE structural material samples, cleaning of the raw cut samples (in order to remove alpha contamination), fine-cutting via EDM of a large number of samples which were then packed and sent to partner laboratories, represented a huge amount of work successfully carried out by PSI in its hot laboratories.
Prior to target dismantling, gamma-mapping of target tip was performed and the results were used for evaluating the proton fluence distribution and irradiation doses on the various target components. In addition, gamma analysis of LBE samples taken from different positions in the target showed good agreement between experimentally determined and calculated activities, which indicates that predictions by nuclear codes can be reliably used for safety assessments.
As target dismantling and specimen cutting took much longer than originally planned, only a limited PIE programme could be carried out in the timeframe of the GETMAT project. However, interesting results were obtained. Tensile tests (in air) were performed on specimens cut from the hemispherical beam entrance window made of T91, which was the most irradiated component (the maximum dose was calculated to be about 6 dpa). As expected, significant hardening was measured, however the total elongation of the sample with highest dose was found to be still above 10%, indicating a fully ductile behaviour. These results were in line with available data for T91 irradiated in spallation conditions in the same temperature range (close to 300°C).
No severe LBE corrosion nor erosion phenomena were observed on the T91 and 316L components. These findings are consistent with the results of the investigations performed on coupons irradiated in contact with stagnant LBE (LEXURII PIE) which did not reveal any corrosion damage.
However, SEM and EPMA investigations performed by PSI and CEA have revealed the presence of a deep crack in the Lower liquid Metal Container made of T91 steel, located close to the Electron Beam (EB) weld line (Fig. WP3.6). Such a deep crack was not present before irradiation, but was perhaps nucleated during welding. Cyclic mechanical loading (due to proton beam trips) may have induced rapid crack propagation, since it has been shown that fatigue crack propagation is accelerated in LBE as compared to crack growth rates in inert environment.
WP4 - Multiscale modelling and model experiments on irradiated Fe-Cr alloys
The objective of the WP was to develop physical models to describe the behaviour of iron-chromium alloys when subjected to thermal ageing and irradiation. These alloys were chosen as model materials for high-chromium ferritic/martensitic steels. The processes of interest were nanostructural and microchemical changes that occur under irradiation or ageing and the corresponding changes in mechanical properties, mainly hardening, which in turn causes embrittlement. The approach used was computer-based multiscale modelling (from ab initio to dislocation dynamics), integrated by the performance of specific modelling-oriented experiments. The latter provided detailed data on which models can be contrasted and suggested, in combination with fundamental atomistic calculations, physical mechanisms that the nanostructure and microchemical models, as well as mechanical models, needed to incorporate. The link with other work-packages is ensured not only by the comprehension of the mechanisms governing the changes induced by irradiation in iron-alloys with varying content of chromium, which allows experimental results on steels performed in the rest of the project, especially WP3, to be better rationalised, but also by the inclusion, in some tasks, of the characterisation of technological properties, such as fracture toughness, in model alloys was included in some tasks.
The ultimate objective of the WP was to understand which nanostructural features are the main cause of radiation-hardening and to develop a physical model capable of rationalising the experimental data available on this issue. This goal was indeed reached.
The work-package was organised in three main tasks, the highlights of which are given in what follows.
T4.1 – Modelling of fundamental properties of Fe and Fe-Cr alloys
The main objective here was the development and application of advanced cohesive models for Fe-Cr alloys, capable of describing their thermodynamic properties and the properties of point-defects (vacancies, self-interstitials), as well as extended-defects (dislocations, grain boundaries), possibly including magnetic properties. The development of these cohesive models, then applied for atomistic calculations, required extensive density functional theory (DFT), or first principle (ab initio), calculations.
Highlights of task 4.1
• First principles calculations have provided new insights into several areas of interest for the basic properties of ferritic alloys; specifically, elements commonly found in nuclear steels, not only Cr but also other solutes in Fe-Cr, have been studied. The main findings of these calculations can be so summarized:
• The magnetic coupling has been extensively studied in Fe-Cr, revealing how magnetism influences some of the properties of the alloy, especially the mixing enthalpy, determining the change of its sign from negative to positive in the low Cr region, thereby leading to a modified phase diagram as compared to the standard one. The strong vibrational entropy of Fe-Cr has been verified: it plays an instrumental role for the shape of the Fe-Cr phase diagram, by lowering the temperature at which the misicibility gap closes. Non-configurational entropy should therefore be included in models that describe thermodynamics from an atomistic standpoint.
• For the grain boundary and surface segregation of Cr in Fe, new insights have been gained, namely, Cr appears to have a tendency to segregate at grain boundaries, while it would tend to stay away from free surfaces, especially voids.
• The study of the interaction with point defects with all possible solute atoms in iron revealed that the elements that are found experimentally to form aggregates and to segregate at dislocations or grain boundaries (namely Cr, but also Ni, Mn, Si, P, and Cu) are found to migrate together with point defects, both vacancies and, in a few cases, interstitials. This proves to be a very effective physical mechanism to explain the experimental evidence.
• The development and validation of the interatomic potentials has rendered a new, improved interatomic model for Fe-Cr in very good agreement with experiments and first principles calculations (see above). This potential has been validated and put to use in dislocation studies, defect diffusion, solute-defect interactions, etc. In particular, Metropolis Monte Carlo studies revealed how Cr is expected to re-distribute at steady-state under thermal ageing and, as a tendency, under irradiation. Fig. WP4.1 illustrates graphically the results of these studies.
T4.2 – Modelling of radiation effects in Fe and Fe-Cr alloys
The main objective here was the development of microchemical and nanostructure evolution models, as well as (radiation) hardening models, for Fe-Cr alloys subjected to thermal ageing and irradiation, using as input the results of applying ab initio and/or empirical cohesive models from T4.1. The former category includes several types of kinetic Monte Carlo (KMC) models, each of them with its advantages and limitations, as well as cluster dynamics (or rate theory) model. The latter consistes basically of the coupling of molecular dynamics (MD) with dislocation dynamics (DD): it is the first time that this coupling is successfully applied to rationalise quantitatively the results of mechanical tests, starting from fully physical considerations.
Highlights of task 4.2
• Atomistic KMC (AKMC) techniques have reached maturity as methodology for the simulation of thermal ageing processes (phase separation, segregation, …) driven by diffusion of solute atoms, although some specific difficulties still exist with the diffusion of interstitial species like carbon. The importance of including (semi-empirically) magnetic and vibrational effects emerges from both DFT calculations and experimental studies, and should be considered as one of the main improvements to be introduced in the technique in the long run.
• For AKMC simulations of irradiation processes, important progresses have been made to include multiple vacancies and single-interstitials: the effect of Cr on point defect diffusion properties is explicitly accounted for and simulations of resistivity recovery experiments including effect of Cr and C have been successfully performed. Also the calculations of the Lij phenomenological coefficients (Onsager matrix), for use in radiation-induced segregation models, has been performed and processes such as phase separation and segregation can be now successfully simulated starting from fully physical considerations. However, the modelling of SIA clusters in an AKMC framework remains challenging.
• Object KMC (OKMC) and Cluster Dynamics (or Rate Theory) models have been developed and a direct comparison with experiments has been made to validate them. The key difficulty is the determination of parameters for concentrated alloys: a new method, cellular OKMC, has been developed, with promising results. As alternative, a "grey-alloy" model in which the presence of Cr and C atoms on the nanostructural evolution under irradiation is taken into account by modifying defect stability and mobility parameters, without explicitly introducing solute atoms, has been also developed. The latter allowed good agreement with experimental trends to be obtained, but cannot model precipitation. The former, on the other hand, allows precipitation to be described and in the long run should be able to treat also the other defects, in a similar way to the "grey-alloy" model, providing also more local information, but for the moment these features are not included yet in the code. The rate theory models that have been developed clearly required specific fitting to experimental data.
• Extensive MD simulations have been performed in order to assess the friction stress due to solution strengthening in Fe-Cr alloys of different concentrations, as well as to evaluate the critical stress for unpinning of both edge and screw dislocations at defects, including Cr-decorated loops and Cr-rich precipitates, which are assumed to be the main cause of irradiation hardening in Fe-Cr alloys. The studies have been performed for different defect and box sizes, different temperatures and also different strain rates. They have clearly shown that dislocation loops decorated by Cr atoms (and/or other solute atoms), as a consequence of segregation, are stronger obstacles than undecorated loops and are, therefore, candidates to be the main origin of radiation hardening in these (and other) Fe alloys. A method to extract from MD data the parameters for DD simulations has been developed: this is based on the deduction of the 'friction stress inside the defect', which depends only on the type of defect, as well as the activation energy of unpinning.
• The MD results of strength of irradiation defects (loops, Cr decorated loops, Cr precipitates, …) have been incorporated into a DD model, allowing simulations of radiation strengthening in Fe-Cr alloys as a function of Cr content: it is the first time that a fully physical model manages to account in a quantitative way for the radiation-hardening measured experimentally. Key is the recognition that screw dislocations that interact with loops having collinear Burgers vector may form helical turns, thereby producing hardening effects 10 time larger than the Orowan mechanism. The main sources of hardening are the Cr-NiSiP clusters, if assumed to be small loops with solute atoms segregated on them.
T4.3 – Modelling-oriented experiments in Fe and FeCr alloys
The main objective here was to performance modelling-oriented experiments, in which selected reference materials (Fe-Cr model alloys) were characterised after neutron and ion irradiation, from the nanostructural, microchemical and mechanical point of view, using a combination of experimental techniques.
Highlights of task 4.3
Two main sets of experimental results have been obtained within this task:
1. Characterization of neutron irradiated Fe-Cr alloys
2. Characterization of ion irradiated Fe-Cr alloys
In both cases the goal was to combine different experimental characterization techniques in order to get a complete picture of the nanostructural and microchemical changes occurred in the alloys as functions of Cr content and irradiation dose, as well as, partly, temperature. The materials considered were alloys containing nominally 2.5 5, 9 and 12 %Cr.
Neutron irradiated Fe-Cr alloys (0.06 0.6 and 1.5 dpa, at 300°C)
- Positron Annihilation Spectroscopy (PAS)
Mainly single vacancies are present in FeCr alloys compared to pure Fe: very little vacancy clustering occurs when Cr is present (data for 0.06 dpa).
Higher amount of vacancies is observed at higher dose, but nonetheless the presence of Cr reduces the amount of vacancies drastically.
- Small Angle Neutron Scattering (SANS)
The irradiation-induced clusters detected are Cr-rich alpha’-phase particles for Fe-12%Cr, while in the other alloys they must contain impurity elements, such as C, Ni, P or Si.
There is a pronounced fluence dependence of the concentration and size of alpha’-phase particles below 0.6 dpa and a saturation-like behaviour beyond.
For lower Cr levels, pronounced fluence dependence below 0.6 dpa was also found.
- Atom Probe Tomography (APT)
alpha' clusters were only present in the supersaturated alloys (Fe-9%Cr and Fe-12%Cr)
CrSiPNi enriched clusters were observed in all alloys with the same number density: they are suspected to be small loops on which solute atoms segregated.
Segregation of Si, P and Cr atoms on dislocation lines and grain boundaries was observed.
Zones free of clusters were found around the GBs, suggesting that solute segregation should be the mechanism leading to the formation of the CrSiPNi clusters: they would not form where other sinks attract the solute atoms.
- Transmission Electron Microscopy (TEM)
Dislocation loops are inhomogeneously distributed for 5, 9 and 12 Cr with dislocation loops preferentially located close to grain boundaries and dislocation lines; they are homogeneously distributed for 2.5%Cr: there is an effect of Cr content on the distribution of dislocation loops. This effect might be indirect: it is known that 2.5 %Cr alloy is ferritic and has C in the matrix, while the other alloys are ferritic/martensitic and C has segregated at grain boundaries.
Ion irradiated Fe-Cr alloys (to 1 and 5 dpa, at 100, 300 and 420 °C)
- PAS (Coincidence Doppler Broadening and Positron Annihilation Lifetime Spectroscopy):
Vacancy-type defects were produced by the ion irradiation. The amount and size decrease when %Cr increases and also, when the irradiation T increases. Larger voids and more observed at the lower dose.
- X-ray Magnetic Circular Dichroism (XMCD):
Ion irradiation produces a redistribution of Cr atoms, deduced from the magnetic moment, which can be an indication of Cr clustering also observed by APT.
- APT:
NiSiPCr rich clusters where observed after ion irradiation (Cr is the main element)
No alpha'precipitates are observed in this case, differently from neutron irradiation.
- Extended X-Ray Absorption Fine Structure and X-Ray Diffraction (EXAFS):
This technique indicates the presence of vacancy and interstitial type defects.
- TEM:
Homogeneous distributions of interstitial dislocation loops are found at all conditions.
Indications of denuded zones close to grain boundaries in Fe12Cr.
At 1 dpa: there is no significant effect of Cr content (from 9 to 12) on size or density of defects. Differences are expected in the proportion of ½<111> loops to <100> loops.
At 5 dpa: higher density in 9Cr, while larger size in 12Cr.
Potential Impact:
- Potential impact, exploitation of results and dissemination
It was foreseen that the potential impact of the GETMAT is to advance in the area of materials technology for transmutation and Generation IV reactors. Priority R&D topics were identified and accordingly success indicators defined. Moreover, it was stated that GETMAT would also impact the share of activities among different EU countries and laboratories accounting for their specific area of excellence and by optimising the utilisation of relevant installations.
The potential impact of GETMAT results, as declared at the beginning of the project (and possibly also beyond) is hereafter discussed by evaluating the success indicators as defined in the description of work.
For what concerns the exploitation of results, the project evolution and final outcome can be considered as an essential step towards the emplacement of the Joint Program for Nuclear Materials (JPNM), within the framework of the European Energy Research Alliance. For instance, GETMAT results have been essential in identifying research themes for the JPNM which were translated in Sub-programmes and Pilot Projects.
Moreover, GETMAT has invited a number of external partners to participate to the user group. The user group was composed by steel maker industries as OCAS, and CENTROINOX, the nuclear industry as AREVA and ANSALDO, by research associations as IFE (Norway) and Non-European R&D labs as LANL.
In the section “exploitation of results” a detailed analysis on how the user group has appreciated the GETMAT results will be described.
Finally, GETMAT has done an important work on dissemination of results as well as on education and training. Both aspects were part of the success indicators and will be described in this section. Moreover, in the section 4.3 “Use and dissemination of foreground” further details will be given.
List of success indicators (KPIs)
In order to assess the potential impact of the GETMAT project an analysis of the “Success Indicators” defined at the beginning of the project is hereafter done. The “Success Indicators”, have been identified with the objectives to measure through specific milestones and deliverables the achievement of the declared objectives. The impact of these achievements is then discussed.
Due to technical problems that have been encountered during the project life and that have been solved by e.g. re-addressing activities, it was necessary to adapt the “Success Indicators” accordingly to the updated activities. The identified “Success Indicators” can also be understood as Key Performance Indicators (KPIs) since the “KPIs are commonly used to evaluate the success of e.g. an organization / program / project or the success of a particular activity”.
The list of KPIs (or Success Indicators) for the GETMAT Project, as reported in the description of work, can be classified within two categories. The first category is related to the scientific advancement, while the second category is related to the project output in terms of dissemination and education and training activities.
Hereafter the list of KPIs is given:
GETMAT KPIs (or “Success Indicators”) category I: scientific advancement
1. Distribution to the laboratories of the ODS alloys manufactured in Europe and Japan. Since the fabrication of the ODS alloys is considered a challenging task, due to the amount of material to be produced, the achievement of this milestone will be a successful accomplishment and will give the opportunity to the laboratories to perform the planned experiment within a not too tight schedule.
2. Demonstration of the high temperature performance of the three ODS alloys and their weldability. The accomplishment of this demonstration will be an important task for the definition of the future R&D needs within the area of ODS alloys development for nuclear applications. In addition the compatibility tests of the ODS alloys with different coolants, will complete this framework.
3. The full manufacturing process of “smart coatings” for fuel claddings with optimised chemical composition and standardised qualification process will be available and can be therefore considered in the future consideration concerning materials selection related to the issue of alternative corrosion protection method.
4. The results of the PIE of the samples irradiated in the BOITIX9 and SUPERNOVA experiments (Phénix), MEGAPIE/SINQ, ASTIR (BR2), IBIS/SUMO (HFR) and BOR60 will be collected in several deliverables. A comprehensive representation of these data and the ability to transmit the performance assessment of irradiated materials in presence of coolant to the designers will be an important success indicator.
5. The successful completion of the model parameterisation phase and the timely start of a second phase of model development based on the use of available experimental results will be also used to measure the performance of the project on the modelling side.
GETMAT KPIs (or success indicators) category II: dissemination and education & training
1. Organisation of Workshop / Training Courses / Schools
2. Number of PhDs and Post-Doc students: The reference in the description of work was is 5% of the total budget of the project. This corresponds to a number of PhD/Post-Doc students in the order of 6-10
3. Number and type of publications
Discussion and assessment of the KPIs (or Success Indicators) and their potential impact
GETMAT KPIs (or “Success Indicators”) category I: scientific advancement
1. Distribution to the laboratories of the ODS alloys manufactured in Europe and Japan.
Achievement: Within the GETMAT project ~ 65 kg of ODS were produced and distributed to the GETMAT partners as agreed in the description of work:
o ~ 30 kg 14 Cr ODS
o ~ 20 kg 12 Cr ODS
o ~15 kg 9 Cr ODS
- The distribution of ODS to the laboratories has been achieved with some delay
o ODS manufacturing at large scale in Europe had technical problems at KIT and SCK-CEN
o GETMAT management achieved to get ODS from Japan
Potential impact:
• It has been seen that the large scale production of ODS steel needs further development in Europe.
• All GETMAT partners were working on the same batch of materials, therefore results were comparable. Moreover, the ODS has been distributed such as to optimise the use of facilities and installations in EU.
2. Demonstration of the high temperature performance of the three ODS alloys and their weldability and compatibility tests of the ODS alloys with different coolants
Achievement:
- Tests at high temperature have been performed up to 750°C
o first mapping of ODS performance
o consistency of the mechanical test results coming from different partners, that is an indication of the expertise of the participants
- Investigated welding techniques
o Diffusion Bonding, Friction Stir Welding and Explosive Welding can be applied on ODS upon
- Compatibility with coolants checked
o dependencies from chemistry and temperature identified
Potential Impact:
• Results obtained have made evident the necessity to deepen the microstructural behaviour of ODS when subject to high stresses and to correlate fracture mechanism with microstructure. This will be done within JPNM and related projects.
• It has been suggested that the GETMAT-ODS might be suitable for clad material of advanced nuclear reactors. However, the performances of GETMAT-ODS were not so satisfactory to recommend this material for structural components of reactor systems.
3. The full manufacturing process of “smart coatings” for fuel claddings with optimised chemical composition and standardised qualification process
Achievement
- The thickness of the coating prior to GESA treatment must be < 30µm.
- The surface alloyed layer (after “GESA” treatment) has following required composition: >8wt% Al, between 9 and 14 wt% Cr and the balance is Fe.
- The GESA treatment parameters are fixed regarding electron energy (120keV), pulse duration (40 to 60 µs) and power density (about 2MW/cm²). More than 2 pulses are needed (one with full duration, the other with half duration) to homogenize the final surface
- Short term oxidation experiments in gas Atmosphere is suitable to assess the capability to form thin Al-rich oxide scales in liquid Pb alloys. However the required Al content in gas is at least 2wt% high compared to Pb.
Remaining open issues:
- Deposition process for the coating is still not hundred per cent satisfying. Selected deposition process results in too high oxygen content.
- Can rare earth elements further reduce the required Al content required for thin slowly growing Al-rich oxide formation?
- Process stability and repeatability using the needed GESA IV is still not optimal.
- Qualification process is not working to that extend as expected – an elemental analysis based process (e.g. large area EDX) should be considered and investigated
Potential Impact
• Corrosion protection barrier can be further developed for LFR systems
• Among the two combinations of techniques investigated, i.e. HVOF+Laser treatment and LPPS+GESA treatment, this second one seems to be more promising. However LPPS might be replaced with VPS.
• A qualification process of corrosion protection barrier for nuclear application needs to be defined.
• Relevant activities on corrosion protection barrier for Generation IV and transmutation systems have been included in JPNM and related projects.
4. The results of the PIE of the samples irradiated in the BOITIX9 and SUPERNOVA experiments (Phénix), MEGAPIE/SINQ, ASTIR (BR2), IBIS/SUMO (HFR) and BOR60.
Achievements
- PIE results have been generated on conventional steels as 9-12 F/M and austenitic steels in different environments.
Potential Impact
• The long-term use of 9-12 F/M steel in high temperature and neutron irradiation fields (i.e. for Generation IV and transmutation reactors) is restricted to a temperature range of 350-550°C due to irradiation embrittlement (in the low temperature range) and poor thermal-creep and stress-to-rupture resistance in the high temperature range
• Liquid metal (especially Pb-Bi eutectic and Pb) assisted degradation of 9-12 F/M steel limit the use of this materials for LFR and ADS. Similar degradations have not been observed on austenitic steels.
• In proton-neutron mixed spectrum the high He production rate per dpa might contribute to the hardening and embrittlement of 9-12 F/M steel. These data are relevant for the neutron spallation target development, which is an essential component of ADS systems
5. The successful completion of the model parameterisation phase and the timely start of a second phase of model development based on the use of available experimental results
Achievements
- Fe-Cr phase diagram assessment and role of vibrational entropy
- Definition of an improved interatomic potential for Fe-Cr which has proven to be in very good agreement with experiments.
- Theoretical explanation and experimental verification of irradiation hardening and embrittlement in Fe-Cr alloys
- Identification of differences in the field of ion vs. neutron irradiations
Potential Impacts
• The solubility of Cr in Fe and formation of phases (as given in the phase diagram in the low Cr concentration range) needs to be revised.
• Explanation of irradiation hardening and embrittlement of Fe-Cr alloys can be extended to F/M steels. However, further steps in the theoretical approaches are still needed
• Modelling activities relevant for Generation IV and transmutation systems have been included in JPNM and related projects.
• Even if ion irradiations are relatively easy experiments to be performed. It has been demonstrated that results from ion irradiation have to be handled carefully when used as surrogate of neutron irradiation.
Category II of GETMAT KPIs: dissemination
1. Organisation of Workshop / Training Courses / Schools
Within the GETMAT project it has been agreed to organize two Schools and two workshops. The scientific output of these training activities is here measured in terms of number of attendees at the Schools/Workshop, number of lectures and poster presented.
It should be here also noticed that all GETMAT training activities, but the final international workshop have been organized in cooperation with other EC funded projects and / or international organisations as e.g. IAEA, USA laboratories etc. This approach of organizing training sessions within GETMAT has allowed the improvement of those components of the project that are directly related to the European Research Area and Networking of Excellence.
The training activities organised are:
• First International School on MATerials for nuclear REactors (MATRE-1). 18-23 October 2009 - Auberge de la Ferme - Rochehaut-sur-Semois, Belgium. This school has been organised together with the EC supported PERFORM60 project
• First International Workshop on Dispersion Strengthened Steels for Advanced Nuclear Applications (DIANA-I). 4-8 April, 2011 – Centre CNRS Paul Langevin, Aussois, France. This Workshop has been co-sponsored by: IAEA, LANL (USA), LLNL (USA), ORNL (USA) , OCAS/Arcelor-Mittal (Belgium), SCKCEN (Belgium) CEA (France), KIT (Germany) PSI (Switzerland) and JRC-IE (EU). Selected paper of the DIANA – I workshop have been published on Journal of Nuclear Materials.
• Joint GETMAT/MATTER Summer school - MUNECO: Materials UNder Extreme Conditions. This school has been organised together with the EC supported MATTER project
• Final International GETMAT Workshop
Statistics of the Training Activities:
MATRE – 1 No. of lectures: 18 No. of posters: 24 No. of attendees: 62
DIANA – I No. of lectures: 36 No. of posters: 16 No. of attendees: 84
MUNECO No. of lectures: 25 No. of posters: 32 No. of attendees: 54
Final Int. Workshop No. of lectures: 30 No. of posters: 20 No. of attendees: 75
2. Number of PhDs and Post-Doc students: 30
3. Number and type of publications:
• Conference / Workshop contributions: 4
• Journal Papers: 68
• Others (Newsletter etc.): 4
Exploitation of results: the GETMAT user Group
All participants to the GETMAT user Group were asked to fill in a questionnaire. The questions and the answers are hereafter summarised. It should be noted that not all members of the user group have answered.
• Question 1: Could you exploit the results produced within GETMAT within your institution. Do you have / will have possibilities to disseminate the knowledge acquired within GETMAT?
- Answer from steel industry: The GETMAT project definitively helped to guide and enrich the research done by the steel industry. The knowledge obtained within GETMAT is disseminated inside our organisation and towards strategic partners.
- Answer from nuclear industry: Unfortunately there was no possibility to access any document produced in GETMAT
- Answer from research organisations (EU and non-EU): The results produced within GETMAT were useful in the development of instruments for use in liquid metals and supercritical water. There are various possibilities to disseminate the knowledge acquired within GETMAT (User-Group lecture, Conferences, ..).
• Question 2: Could you identify limits and potentialities of the investigated GETMAT materials, i.e. 9Cr – F/M steels and ODS steels?
- Answer from steel industry: Yes, both for 9Cr-F/M as well as ODS steels the knowledge has has been develop but the industry became also aware of important limitations, mainly for ODS steels.
- Answer from nuclear industry: 9Cr – F/M is still used in both ELFR and ALFRED reactor cooled by Lead.
o The operating temperature has been selected above the Ductile to Brittle Transition threshold under irradiation.
o The lead velocity has been limited lower 2 m/s to limit erosion issue
o The oxygen controlled environment has been adopted to limit corrosion
However the R&D should not be limited to search material suitable for fluent molten Lead but also the influence of the Lead chemistry shall be exploited. We believe that appropriate structural material in conjunction with Lead chemistry (i.e. corrosion inhibitors) have more potentialities to solve the material issues.
- Answer from research organisations (EU and non-EU): ODS steels clearly have potential, but there is still a long way to go (improvement in fabrication routes, avoiding strong anisotropies, and the behaviour under irradiation is not sufficiently explored).
• Question 3: Did you consider the interaction with the GETMAT community through the technical meetings as effective?
- Answer from steel industry: Yes, and there was also the possibility to extent the research network through the GETMAT project.
- Answer from nuclear industry: Designers should be maintained informed about the GETMAT results not only with technical meeting but with appropriate documentation for implementing them in the design.
- Answer from research organisations (EU and non-EU): Yes, the technical meetings were very well organized (and intense) and offered sufficient possibilities for discussions (during the meeting, as well as outside of the meeting).
• Question 4: Do you have any suggestion on how User Groups can be implemented in EC supported R&D future projects on materials for nuclear applications?
- Answer from steel industry: Interaction is mainly related to the technical meetings. It could be beneficial to include maybe one or two additional workshops on sidely related themes like economic impact etc.
- Answer from nuclear industry: Improve the two way communication with dedicated workshops and provide summary and full detailed results.
- Answer from research organisations (EU and non-EU): (1) The case of our institute (IFE) is special because Norway cannot directly receive EC support (as Norway is not in EURATOM). However, we could (technically) contribute through instrumentation and materials testing under irradiation and we have unique capabilities for doing this. If such possibilities are recognized from the start in future projects, subcontracting could be considered. (2) I enjoyed my participation in this user group but it was difficult because of my location out of the US. If the user group has members outside of Europe, it may help participation if some meetings were held as side meetings in conjunction with other related international meetings as long as other members of the user group are also attending this meeting. Otherwise, I found this a very useful user group.
As can be seen from the collected answers, the relation with the User Group occurred only through the technical meetings. This might be the reason why the GETMAT results have been considered more of impact for the steel maker industry and the other research laboratories and less for the nuclear industry.
Moreover, as further suggestion to organise dedicated meetings with the user group to address topics not strictly related to the project might be taken into account for the future. Finally, all members of the user group could exploit the GETMAT results for further develop in their own field.
List of Websites:
The project website has been developed.
Internet address: http://nuklear-server.nuklear.kit.edu/getmat/index.php
The website is continuously up-dated both at the public part and at the member area part. Within the public part all presentations made during the training activities are accessible.
Moreover, the communication between beneficiaries has been always very constructive and good. This communication normally occurred through e-mail and phone calls. Moreover, technical meetings at WP and task levels have been organised on regular basis to discuss technical details of problems and outcomes.
The GETMAT project has also a good communication with other projects as e.g. the MATTER, SEARCH, MAXIMA and with the EERA Joint Program Nuclear Materials. Moreover, during the reporting period GETMAT has contributed to the identification of R&D gaps in the ODS area relevant for the JPNM and the proposal preparation of the MatISSE project.
Through the GETMAT User Group, parties (industries, non-EU research associations) that are external to the project are informed about relevant advancement of the project.
Finally, the GETMAT project has given an essential contribution to the Strategic Research and Innovation Agenda of the SNETP, to the EERA Joint Program on Nuclear Materials and to the EC road-mapping exercise on materials for the SET-Plan.
