Solid State Phenomena Vols. 172-174

Abstract: The EKINOX numerical code, formerly developed to simulate high temperature oxidation of Ni alloys, has been recently adapted to solve out the issue of high temperature oxidation of Zirconium alloys. This numerical code is a one dimensional model that simulates the growth of an oxide layer using a specific algorithm for moving boundaries problem. In order to simulate the oxygen diffusion inside Zr alloys, an adaptation of the EKINOX code was necessary. It consisted in adding, first, a non-null oxygen equilibrium concentration in the substrate and second, a new interface in order to simulate the β/α(O) phase transformation due to oxygen diffusion. In this study, EKINOX has also been coupled with the thermodynamic database for zirconium alloys ZIRCOBASE (thermocalc formalism) in order to obtain accurate concentrations values in each phases (considering local equilibrium at each interface). The present paper illustrates the simulation ability of the code by comparing experimental and calculated oxygen diffusion profiles corresponding to different cases, from isothermal oxidations at high temperature (900 < T < 1250°C) to the study of dissolution kinetics of a pre-transient oxide layer under a neutral environment. The influence of pre-hydriding from a few hundreds up to thousands weight-ppm is also derived from the calculations.

Abstract: In order to build the phase diagram of Cu-Ag nanoalloys, we study a 405-atom nanoparticle by means of Monte Carlo simulations with relaxations using N-body interatomic potentials. We focus on a range of nominal concentrations for which the cluster core remains Cu-pure and the (001) facets of the outer shell exhibit two original phenomena. Within the (N,m_Ag-m_Cu,P,T) ensemble, a structural and chemical bistability is observed, which affects all the (001) facets together. For a nanoparticle assembly, this will result in a bimodal distribution of clusters, some of them having their (001) facets Cu-rich with the usual square shape, the other ones having their (001) facets Ag-rich with a diamond shape. This bistability is replaced in the (N_Ag,N_Cu,P,T) ensemble by a continuous evolution of both the structure and the concentration of the (001) facets from Cu-rich square-shaped to Ag-rich diamond-shaped facets as the number of Ag atoms increases. For a nanoparticle assembly, this will result in an unimodal distribution of the cluster population concerning the properties of the (001) facets. This comparison between pseudo grand canonical and isothermal-isobaric results shows that the distribution of a population of bimetallic nanoparticles depends strongly on the conditions under it is elaborated.

Abstract: We compare three models of 2D precipitation kinetics that give access to different time-space scales. Kinetic Monte Carlo simulations (KMC), cluster dynamics (CD) and nucleation-growth-coalescence model (NGCM), based on a same atomic model, lead to an excellent agreement as long as the interfacial free energy is evaluated accurately and the interaction between diffusion fields is taken into account in the CD. The NGCM model noticeably improves the previous approaches of the same kind by using a constrained-equilibrium hypothesis to describe the solid solution. Moreover, in the coalescence regime, we show that CD leads to cluster distributions that are wider and more symmetric than the LSW distribution due to the probabilistic feature of the growth law of a cluster, that makes it differ from the purely deterministic NGCM approach.

Abstract: We address the question of the evolution of a nanostructured system in a metastable state to equilibrium. To this purpose, we use the case study of the transition of an Au_corePd_shell nanoalloy cluster containing up to about 600 atoms toward the equilibrium Au segregated configuration. We start from a molecular dynamics approach with an embedded atom potential. The way the transition develops at low temperatures is found to be very sensitive to the cluster morphology and the way energy is exchanged with the environment. The transition of icosahedral inverse core-shell Au-Pd clusters is predicted to nucleate locally at the surface contrary to clusters with other morphologies, and starting at lower temperatures compared to them.

Abstract: The size distribution and the total density of clusters of a one-dimensional pure deposit can be expressed analytically from the Ising model. For a codeposit, the alloying effect and the presence of broken bonds at the cluster edges lead to inhomogeneities of the chemical composition of the clusters. We investigate the influence of codeposition on the size distribution of clusters in the case of an alloy that forms an ideal solution. We obtain the exact solution for the size distribution of clusters while the complete characterization of the system results from coupled analytical formulae in the grand-canonical ensemble. The results of this analytical model are successfully compared with those obtained by Monte Carlo simulations.

Abstract: Transmission electron microscopy and mechanical testing were employed to investigate the evolution of microstructure and functional superelastic properties of 0.1mm diameter as-drawn Ni-Ti wires subjected to a non-conventional heat treatment by controlled electric pulse current. This method enables a finer control of the recovery and recrystallisation processes taking place during the heat treatment and accordingly a better control on the final microstructure. The best functional properties were obtained for heat-treated Ni-Ti wires having a nanograined microstructure (20-50 nm) partially recovered through polygonization and partially recrystallized. Such microstructure is highly resistant against dislocation slip upon cycling, while microstructures annealed for longer time and showing mostly recrystallized grains were prone to dislocation slip, particularly as the grain size exceeds 100 nm. The density of dislocation defects increased significantly with increasing grain size of the microstructure. The activity of three <100>/{011} slip systems was identified in the largest grains of 500-1200 nm. An additional mode of plastic deformation, {114} compound austenite twinning, was observed in the largest grains of fully recrystallized microstructures. It is proposed that dislocation slip (and possibly deformation twinning) occurring in superelastic cycling is coupled with the stress-induced martensitic transformation.

Abstract: Reactive high energy ball-milling has known a growing interest from both fundamental and applied point of view. We focus here on the specific system Fe-Y₂O₃ metal-oxide nanocomposite because of its potential application to the synthesis of oxide dispersion strengthened steels, which are promising materials for nuclear applications. YFe₃ and Fe₂O₃ were ball-milled during different milling durations in the stoichiometric proportions defined by the chemical reaction 2 YFe₃ + Fe₂O₃ → 8 Fe + Y₂O₃. The obtained milled powder was characterized by XRD, SEM, TEM and their thermal behaviour was investigated by DTA. Through those characterizations, a Mechanically induced Self-propagating Reaction (MSR) was observed and several steps in the ball-milling process were identified: mixing of reactants, chemical reaction propagation, amorphization and refinement of the microstructure. The role of the milling intensity was also examined.

Abstract: Amorphous silicon dioxide layers were implanted with 100 keV Ar ions to a relatively high fluence in a tentative to generate cavities in the oxide. Different oxide layers were used, obtained either by thermally growth or by chemical vapor deposition (CVD) on Si substrate. In all SiO₂ layers, cavities are not formed in the as-implanted state. However, in the transmission electron microscope, under electron beam, the combined effect of irradiation induced defects and implanted rare gas leads to the formation of cavity bands giving the unique opportunity to observed in-situ cavity growth. The cavity morphology and their distribution are found to be dependent on the silicon dioxide growth process. For thermally grown SiO₂ layer, a homogeneous cavity band is formed, centered at the mean ion path, with an average cavity size of 20 nm. For CVD SiO₂ layer, slightly smaller cavities are formed in two distinct bands. The formation of cavities is discussed in light of gas and defects interaction and field-induced migration whereas the cavity distribution is discussed in terms of self-organization.

Abstract: A Fe₅₀Pd₅₀ alloy was severely deformed by High Pressure Torsion (HPT). For a processing temperature ranging from 20°C to 300°C, the Severe Plastic Deformation (SPD) induces a significant grain size reduction (in a range of 50 to 150 nm) but also a strong disordering of the long range ordered L1₀ phase. However, Transmission Electron Microscopy (TEM) data clearly show that few ordered nanocrystals remain in the deformed state. The deformed material was annealed to achieve a nanoscaled long range ordered structure. The transformation proceeds via the nucleation and growth of ordered domains along grain boundaries. Aging at lower temperature (400°C) gives rise to a smallest domain size and thus the highest coercivity.

Abstract: Composite materials and micro- and macrostructure designs have been the focus of numerous scientific studies over the past few years according to their crashworthiness [1-3]. Crashworthiness is concerned with the absorption of energy through controlled failure mechanisms and modes that enable a defined load profile during energy absorption [4]. Cellular materials, such as metal foams, are materials which display a unique combination of physical and mechanical properties, e.g. for crash box applications. The defining characteristic of metal foams is a very high porosity, typically in the range of 70 to 90 vol. %. In principle, cellular metals can be manufactured from gas, liquid or solid phases and currently the most advanced methods involve melt-metallurgical processes [5]. Several groups have produced foam structures by using hollow spheres to form the cells of the material [5, 6]. These materials exhibited plateau stresses of 5 MPa and 23 MPa respectively, with volume specific energy absorptions SEA of 2 MJ/m³ and 10 MJ/m³ respectively, up to 50 % strain [6, 7]. By combining ceramics with ductile metals, failure-tolerant metal matrix composites (MMCs) can be created. With regard to application of the MMCs as wear resistant materials in metal forming tools a prolongation of the life time and the resultant reduced equipment downtimes have been achieved by active steel infiltrating of porous zirconia structures with the aid of Ti as activator [8]. A very promising approach concerning zirconia/steel - composite materials with superior mechanical properties has been demonstrated by Guo et al. using a low-alloyed TRIP steel in combination with an Y-PSZ – ceramic [9, 10]. In a previous study honeycomb structures were formed from composites of high-alloyed austenitic stainless TRIP-steel AISI 304 with Mg-PSZ with different mixing proportions due to ceramic extrusion at room temperature and sintering at 1350 °C for 2 h in an 99.9 % Argon atmosphere [11]. One of the most promising manufacturing route to produce open cell composite foams is based on the patent of Schwartzwalder [12] by the replication method using polyurethane sponge as a template. The polymer foam is impregnated in a powder slurry (this first coating contributes as an adhesive porous layer for further coating processes), the ceramic slurry is squeezed out of the functional pores and cold spray coatings are applied in order to eliminate defects out of the squeezing process and reach the critical wall thickness for acceptable mechanical properties. In [13] the authors reported about foams with 90 Vol% high alloyed TRIP-steel and 10 Vol% Mg-PSZ. Up to 50 % compressive strain a remarkable enhancement of the SEA was observed in comparison to comparable structures with TRIP-steel only.