With a growing population, the need for energy is increasing at the same time. In order to fulfill the demand, appropriate tools are required. MYRRHA (Multi-purpose Hybrid Research reactor for High-tech Applications), as a new flexible fast spectrum research reactor with advantage of both decreasing the toxicity of the waste and the volume of the waste by a factor of 100[1], is developing to resolve the key issues for our future world.
As showed in Figure.1[2], the MYRRHA, which conceived as an accelerator driven system(ADS), is consisted of a proton accelerator, a spallation source and a sub-critical core.
Figure. The scheme of ADS system[2]
The spallation target is a neutron source providing primary neutrons to be multiplied by the surrounding sub-critical core. The spallation reaction of heavy-metal target nuclei provides the primary neutrons, and bombarded by high-energy protons generated by the accelerator. The high power (density) in the target resulting from the heat deposition by the proton beam and the space limitations for optimal neutronic performance, asks for forced convection heat removal by liquid metal.[3] The cooling systems for MYRRHA are designed to evacuate core power of 50-100MWth plus additional heat produced in the spallation target, the decay heat in the in-vessel storage zones, the polonium decay heat and heat produced by all pumps. The inlet and outlet temperatures are 270°C and 400°C[4] , which means that the coolant liquid metal should have low melting point. Lead-bismuth eutectic has been proposed as both spallation target and coolant. It have several advantages:[4][5]chemical inertness with air and water, low vapor pressure over the relevant temperature range high boiling point, high atomic number and favorable neutronics (high scattering and small absorption cross-sections).
2.2 Materials used in lead-bismuth eutectic (LBE)
In Lead-bismuth cooled fast reactors, the structure materials are in contact with both high temperature and high velocity lead-bismuth. This means that the materials contact with LBE should have good corrosion, mechanical and irradiation properties to ensure the safety of the device.
2.2.1 Compatibility of structural materials with LBE
When using LBE as nuclear coolant, it is well known that the materials are very easily corroded. The chemical potential for dissolution of all solid surfaces in contact with liquids is the main driving force for liquid metal corrosion[6].
Two types of corrosion are presented in LBE: uniform and local. [7] Uniform corrosion is characterized by the uniform damage at the surface where the materials contact with the LBE. Several local corrosions, including penetration along the grain boundaries, the specific crystallographic direction, vacancies and pores and previous formed defects, are listed in Figure 2.
Figure . Diagram of corrosion damage of metals in liquid metal media: (a) uniform corrosion; (b) penetration along the grain boundaries; (c) along the specific crystallographic direction; (d) along vacancies and pores; (e) and (f) along previous formed defects [7]. LM represents the liquid metal
Several factors will affect the structural corrosion in LBE. The process involves four following process:[7]
Dissolution of the solid materials into the liquid metal.
Thermal and concentration gradient assisted mass transfer in liquid and solid.
Redistribution of interstitial impurities between the solid and liquid metals.
Diffusion penetration of liquid metals into solid metals with formation of solid solutions or new phases.
Since the liquid metal flows on the surface of the solid, the velocity of the LBE will also influence the corrosion rate. As shown in Figure 3, the effect is summarized:[8]
At low velocity, the corrosion is mass transfer domain, which means that the dissolution rate is larger than mass transfer rate. Above a critical value, corrosion is dissolution domain and independent of velocity. At very high velocities, high shear stress can strip the protective film at interface, erosion-corrosion occurs.
Figure Velocity effects on the corrosion rate [8]
Temperature has important effect on the corrosion rate, because the corrosion product diffusion coefficients in solid and liquid alloys and the liquid metal viscosity are functions of temperature. When the temperature increases, higher diffusion coefficients rates, higher solubilities, higher diffusion coefficients and smaller viscosities will all lead to higher corrosion rate. And sustained corrosion process can only occur in a non-isothermal system such as nuclear coolant system.[9]
Oxygen concentration in LBE is a key parameter for corrosion of structural material. As shown in Figure 4[d], for steel at low concentration of oxygen, corrosion of dissolution occurs whereas for high concentration, oxidation takes place. To control the oxygen activity in LBE systems, the corrosion depth can be minimized.
Figure
2.2.2 Effect of LBE on Mechanical Properties of structural materials
When material is used in the LFR, two influences of the liquid metal, Liquid metal embrittlement (LME) and Environment-assisted cracking, should be taken into account.
Liquid metal embrittlement (LME) is a phenomenon of practical importance, where certain ductile metals experience drastic loss in tensile ductility or undergo brittle fracture when tested in the presence of specific liquid metals.[10]
LME crack occurs by nucleation of a crack at the wetted surface of a solid and then propagation in to the bulk until the final failure. The prerequistites for LME are:[a]
1. Intimate or direct contact at atomic scale between the solid and liquid metal phases
2. Applied stress sufficient to produce plastic deformation
3. The existence of stress concentrators or pre-existing obstacles to the dislocation motion
Lots of experiment about the compatibility of T91 steel with LBE, which has been chosen as the ADS structure material. The requirement to prevent LME effect is summarized[a]:
Excellent surface finish to be free of surface defects
The presence of a protective oxide film on to the steel surface before and during contact with LBE and self-healing in service condition
Prevent metallic embrittling impurities which affect the mechanical properties significantly.
2.2.3 Irradiation effects on compatibility of structural materials with LBE
Another important issue is the material behavior during irradiation. In Jung's [b] work , he demonstrated three effects of irradiation on structural materials: structural changes from displacement and rearrangement of atoms, kinetic effects from enhanced redistribution of atoms by mobile defects with promotion of segregation and phase changes, and chemical changes from the production of new atomic species by nuclear reactions(transmutation).
The products of atomic displacement are vacancies and self-interstitials and clusters. Theses defects would lead to growth of cavities or dislocation loops, climb of dislocations, and produce microvoids, stacking fault tetrahedral acting as barriers to dislocation movement, so the materials would be embrittlement.
The increased concentration of vacancies and self-interstitials cause radiation enhanced diffusion (RED).[c] The flux of defects induces segregation and even precipitation if solubility limits are exceed. The segregation and precipitation at grain boundary would cause transgranular brittle fracture. And the radiation induced segregation also interaction with the environment by promoting corrosion, stress corrosion cracking and LME.[b]
By transmutation, the chemical composition and properties of the structural material may change. The ductility and fracture toughness are reduced by radiation hardening. But when temperature increases, only the density and dimension would change significantly. [c]
2.3 MAX phase ceramics
MAX phase ceramics is a series ternary carbides or nitrides with the general formula Mn+1AXn(n=1,2,3), where M is the early transition metal, A is a group of elements, X is C or N.(showed in Fig 4). It has combination of properties of metals and ceramics, such as good electrical and thermal conductivity, high strength and modulus, high thermal shock resistance, high-temperature oxidation resistance, low density and can be easily machined.
File:MAX phases periodic table.png
Figure
2.3.1 Structure of MAX phase
MAX phase has hexagonal unit cells with the space group P63/mmc and has a common structure with Mn+1Xn layers interleaved with pure A-group element, as shown in Figure 5[11]. Table 1 lists the known MAX phases so far.
Table . List of known MAX phases.[f]
211 phases
211phases
Sc2AlC
Sc2GaC
Hf2TlC
Hf2PbC
Sc2InC
Sc2TlC
Ta2AlC
Ta2GaC
Ti2AlC
Ti2AlN
Ti2SiC
Ti2PC
312 phases
Ti2SC
Ti2GaC
Ti3SiC2
Ti3AlC2
Ti2GaN
Ti2GeC
Ti3GeC2
Ti3SnC2
Ti2AsC
Ti2CdC
V3SiC2
(V0.5Cr0.5)3AlC2
Ti2InC
Ti2InN
Nb3SiC2
Ta3AlC2
Ti2SnC
Ti2TlC
413 phases
Ti2PbC
V2AlC
Ti4AlN3
Ti4SiC3
V2SiC
V2PC
Ti4GeC3
V4AlC3
V2GaC
V2GaN
Nb4AlC3
Ta4AlC3
V2GeC
V2AsC
Ti4GaC3
Cr2AlC
Cr2GaN
514phase
Cr2GaC
Cr2GeC
Ti5SiC4
Zr2AlC
Zr2AlN
Zr2SC
Zr2InC
615phase
Zr2InN
Zr2SnC
Ta6AlC5
Sr2TlC
Zr2TlN
Zr2PbC
Nb2AlC
Nb2PC
Nb2SC0.4
716 phase
Nb2SCx
Nb2GaC
Ti7SnC6
Nb2InC
Nb2SnC
Nb2AsC
Mo2GaC
Hf2AlC
Hf2AlN
Hf2SC
Hf2InC
Hf2SnC
Hf2SnN
Figure
MAX phase has structural similarities with MX phases, so they have many common properties, such as metal-like conductors and stiff. Furthermore, the difference between the MAX phase and MX phase is that in the former, only basal plane dislocations are numerous, multiply, and are mobile at temperature as 77K and higher.[12] Thus less slip systems active in MAX phase are needed than in polycrystalline ductility.
2.3.2 Synthesis of MAX phase
Various synthesis routes of MAX phase have been reported in literatures. Both bulk form and powder form can be synthesis successfully.
Mechanical alloying is a conventional and convenient powder metallurgy process with low cost. C.Yang[13] use Ti ,Al and C as elemental powder to synthesis high purity Ti3AlC2 by this route combining with SPS sintering, and prove that the small amount of excess Al powder will increase the purity of the final product.
Molten salt synthesis is one of the methods of preparing ceramic powders, which use molten salt as the medium. W.-B Tian[14] et al prepared Cr2AlC using chloride as the fluxes, dwell at 1000°C. They prove that proper amount of excessive Al and temperature is important to obtain Cr2AlC. Also The results indicated that the amount of Cr2AlC increases as powder/flux ratio decreases from 2:1 to 1:1 while the further decrement of powder/flux ratio to 1:2 has little effect on its amount.
Several in-situ sintering is used to fully synthesize dense MAX phase ceramics. The most common ones are Hot Pressing (HP) and Spark Plasma Sintering (SPS). A lot experiments[15,16,17,18] prove that high purity MAX phases including Ti2AlN,Ti2AlC,Ti3AlC2 and Nb2AlC can be obtain by HP at different temperature. Ti3AlC2 can be obtained also by SPS process, soaking temperature determines which phases TiC, Ti2AlC or Ti3AlC2 are the main phase obtained in the process.[19]
2.3.3 Mechanical properties of MAX phase
As shown in Table 1, most of MAX phases are elastically quite stiff, If combine it with their density, their specific stiffness values can be high. In general, when A site is lighter elements, like Al, the MAX phase will become stiffer.
Table [11]
Furthermore, Radovic et al.[20] reported on E and shear moduli of several MAX phase in 300-1573K, the values are determined at very slow strains using resonant ultrasound spectroscopy(RUS). They showed that several Al-containing MAX phases and Ti3SiC2 have another useful property: their elastic properties are not a strong function of temperature.
The most unusual but interesting aspect of MAX phases is their nonlinear elastic behavior, which shown in Figure 6[11]. Figure 6b shows that 90% dense solid dissipates more energy per unit volume than its fully dense counterpart. It is prove that the annihilation of incipient kink bands (IKBs) is the explanation of the nonlinear behavior. Incipient kink bands is a kink bands (KB) that does not dissociate into mobile dislocation walls(MDWs).The schematic is show in Figure 8.[11] Frank and Stroh[e] assumed that once a KB nucleated, it would immediately extend to a free surface, and result in two parallel non dislocation walls(shown in Figure 8). The new dislocation wall ultimately forms kink boundaries.
http://www.annualreviews.org/na101/home/literatum/publisher/ar/journals/content/matsci/2011/matsci.2011.41.issue-1/annurev-matsci-062910-100448/production/images/medium/mr410195.f3.gif
Figure [11]
Figure (a) Schematic representation of a thin elliptic cylinder with axes 2α and 2β such that α > β. The sides comprise two dislocation walls of opposite sign and uniform spacing D. (b) Formation of an incipient kink band (IKB) in hard grains (red in panels d, e, and f ) adjacent to soft grains (blue in panels d, e, and f ). The lines in the grains denote basal planes. (c) Schematic of the stress-strain hysteresis loop due to formation and growth of IKBs during loading and their annihilation during unloading. (d) IKBs in hard grains. At this stage, the IKBs are fully reversible. (e) Multiple mobile dislocation walls in a large grain. Dotted lines denote walls that have separated from the source and are moving away from it. This happens only at higher temperatures and/or stresses. (f) Same as panel d, but now the mobile dislocation walls are all subsumed into the kink boundaries. At this stage, the effective grain size is smaller than the original grain size.[11]
Because of the layer structure like graphite, MAX phases also have plastically anisotropic because of the lack of slip system. When polycrystalline samples are loaded, they rapidly develop large internal stresses and uneven states of stress[21] Another consequence is Kink Bands(KB) formation, which plays an important role under mechanical deformation(shown in Figure 7).
http://www.annualreviews.org/na101/home/literatum/publisher/ar/journals/content/matsci/2011/matsci.2011.41.issue-1/annurev-matsci-062910-100448/production/images/medium/mr410195.f7.gif
Figure [22,23]
The MAX phases are relatively soft and can tolerance damage. The coarse-grained microstructure has better damage tolerance than the fine-grained microstructure. The Vickers hardness value of MAX phase is 2-8GPa, which is softer than most of structural ceramics, but harder than most metals. When the load decreases, hardness increases. Their response to nanoindentation is anisotropic. When the basal planes are indented on edge(As shown in Figure 10), they delaminate, and cracks parallel to the basal plane form; but no pileups are observed. To form pileup, both delaminations and kinking are requied.[11]
Figure [11]
2.3.4 Corrosion resistance of MAX phase in LBE
In present study, MAX phases have been proved has good compatibility with liquid metal. In L.A. Barnes et al[24]'s work, for both hot leg and cold leg samples of Ti2AlC, the reaction zone between them and the lead is extremely small; for Ti3AlC2 samples, the lead can even be separated cleanly from the ceramics. Later, Rivai and Takahashi[25] compared the compatibility of steels, refractory metals and Ti3SiC2 with LBE. The result showed that after immersion in LBE at 700°C, no penetration of LBE in to Ti3SiC2 and no trace of corrosion were observed. Heinzel[26] investigated influence of the oxygen content in LBE to the compatibility of Ti3SiC2, and found that under the condition of 10-6 wt% to 10-8wt% oxygen content, 550°C to 750°C, up to 4000h, thin TiOx layer formed at the surface but no dissolution attack was observed.
2.4 Metal matrix composites containing ceramic particles
Metal matrix composites (MMC) is a kind of composite material that disperse reinforce materials like ceramic or organic compound in to metal matrix. In this section the interfacial phenomena of MMC and some MAX phases contained MMC are presented.
2.4.1 Interfacial chemistry
Interfacial has several different definitions. Basically, it is the immediate transition plane from one micro component to another. But in practical system, it is more complex because of non-uniformity of substructure of phases, reaction products, impurity segregation and interaction products. Geometric relationships between the adjacent crystalline phases, chemical interations and electrostatic interactions have to be considered in an interfacial couple of to MMCs with ceramic reinforcing phases.[27]
Thermodynamic can help to indicate whether the MMC system is stable although in practical, it might have some difficulty due to the heterogeneous nature of both reinforce particles and matrix.[27]
The basic ideal to using thermodynamic is the wetting, as shown in Fig 8. The fundamental quantity deciding the mechanical strength of the interface is the ideal work of separation Wsep. It can be calculated using Dupré equation(Equation 1). The large the Wsep, the more energy need to separate the interface.
Wsep=γlg+γsg-γ12 (Equation 1)
The Young equation (Equation2) defines the contact angle Θ
cos Θ=(γsg-γls)/ σ1g (Equation 2)
File:Contact angle.svg
Figure Contact angle of a liquid droplet wetted to a rigid solid surface.
However since the Equations strictly refers to a liquid-solid contact, direct calculation of Θ is problematic, so Wsep is not easy to obtained. The more straight forward thing to do is to calculate Wsep by comparing the total energy of two systems.[28]
Stoneham[29] created a simple model that includes the electrostatic energy. If ions are treated as point charges, Equation 3 is a basis of a simple model of adhesive energy between an ionic crystal and a metal surface.
E(z)=-q2/4(z-zo) Equation 3
where q is the charge, z is the distance z from the charge to the surface, z0 is the position of the image plane.
2.4.2 Property predication using mixture rules
2.4.3 MMC containing MAX phase particles
Combining metal with MAX phase particles to get good properties of both has attracted a lot of interest. Current works are focused on Cu, Ag and Mg based MMC with MAX phases.
Huang[30] prepared Ti3C2-Cu(Al) cermets by pressureless sintering or in-situ hot-extruding a mixure of Ti3AlC2 and Cu powders. No trace of Ti3AlC2 phase can be detected. The intercalation of Al induced to the decomposition of Ti3AlC2, which is transformed to TiCx. Strong bond interface between Ti3C2 layer an Cu(Al) layer results in high fracture toughness of this MMC. Microstructure evolution of this MMC was studied by Zhang[31] and his group. High strength interface layer that bonds Ti3AlC2 and Cu grains was observed at 850°C, at 950°C mild reaction between the two phase was detected. When further increase the Temperature to 1050°C,Ti3AlC2 decomposed into Ti3C2 but strong bonding between it and Cu matrix also generated.
The tribological property of MMCs consisting of Ta2AlC or Cr2AlC and 20 vol% Ag was investigated from ambient to 550°C against Ni-base superalloy and Al2O3[32]. The composites have better wear resistance at low temperature than pure MAX phases. When the temperature increases, the performance is slightly worse. Most importantly, the tribological performance of the MAX/Ag composites appeared to improve both sliding distance and thermal cycling. In practical, they are potential materials for high tribological applications especially under the condition that coating is not possible.
Some researchers[33,34] reported the fabrication of Ti2AlC/Mg composites by pressureless melt infiltration at 750°C for 1h. The stabilized Mg grain size was 35±15nm without changing after annealing. Some Mg dissolved in MAX phase, formed (Ti1-xMgx)2AlC, also some Ti was found in the Mg matrix. The strength of the composite is significantly greater than the pure Mg.
