Poster contributions

abstract: 3.5
Plastic deformation of MgO single crystals studied by DD simulations
PH CARREZ, P CORDIER, Lab de Structure et Proprietes de l'Etat Solide, CNRS-UMR 8008, Universite de Lille 1, France. B DEVINCRE, L KUBIN, Laboratoire d'Etude des Microstructures, CNRS-UMR 104, ONERA Chatillon, France.
Considerable heat and mass transport take place in the Earth's mantle. This process is responsible for plate tectonics and related phenomena such as volcanism or earthquakes. The study of the plastic properties of rocks and minerals is of primary importance to model the rheology of the mantle. Magnesiowustite (Mg,Fe)O is one of the main constituents of the lower mantle (between 670 and 2900 km depth). Unfortunately, deformation experiments on MgO under the extreme P, T conditions relevant of the Earth's lower mantle are very delicate. Numerical simulations provide us today with an alternative to address the behaviour of materials under extreme conditions. In this study, the deformation processes of MgO are simulated at the mesoscopic scale with the help of a 3D Dislocation Dynamics (DD) simulation (see e.g., Devincre B. et al. (2001) Mat. Sci. and Eng. A, 309-310).

The objective of this work is to understand the strong hardening properties of this material through mass simulations. We report here on two preliminary steps, junction formation and dislocation kinetics. MgO is an ionic crystal with the NaCl structure. In agreement with experiments made under normal conditions or at high pressure, the simulation account for two possible slip systems, $1/2 <110> \{1\overline{1}0\}$ and $1/2 <110> \{001\}$. The $\{110\}$ planes can cross at 90° or 60°. Intersections of dislocations gliding in these slip systems potentially leads to the formations of two types of junctions. One results from the interaction between dislocations with perpendicular Burgers vectors (similar to the Hirth locks in fcc crystals), the other results from interactions between dislocations with Burgers vectors at 60°, leading to the formation of a sessile junction (a Lomer lock, following the same analogy as above). The systematic modelling of dislocations intersections shows that only Lomer junction formation is energetically favourable. This is consistent with early literature results on slip line observations. As dislocation motion in MgO is controlled by impurities, an Arhenius equation form is fitted on experimental data taken from literature. This leads to the formulation of a phenomenological law that is valid in a relatively large range of temperature (0-1200K).


abstract: 3.7
MICROSTRUCTURE AND DEFORMATION CHARACTERISTICS OF TITANIUM REINFORCED WITH YTTRIA NANOPARTICLES
VANESSA DE CASTRO, TERESA LEGUEY, ANGEL MUÑOZ, RAMIRO PAREJA, Depto. de Fisica Universidad Carlos III de Madrid, SPAIN.
In situ Ti/Y2O3 nanocomposites have been developed and characterized, and the strengthening effects of the fine dispersion of Y2O3 particles investigated. Samples, containing $0.3\_1.0 wt\%$ of Y2O3 was prepared by the non-consumable arc melting technique. TEM analyses performed on as-cast samples reveal the presence an homogeneous dispersion of Y2O3 particles with sizes $< 100$ nm. Spherical particles are responsible of the pinning of a-type dislocations, in the major slip systems. The mechanical properties have been determined by microhardness measurements and tensile tests at temperatures between 293 and 773 K. The fracture characteristics have also been investigated. Strengthening of the material is noticeable at the investigated temperatures without a significant loss of ductility. The hardness of the composites increases by a factor of 2. Comparison between microhardness and ultimate tensile stress values follows Tabor´s correlation. The validity of the relationships obtained between microhardness and tensile properties for nanocomposites is discussed.


abstract: 3.8
Structure mechanics relationship in neutron-irradiated heat affected zones of austenitic stainless steels welds
RALUCA STOENESCU, DIDIER GAVILLET, Paul Scherrer Institut, Villigen PSI, Switzerland. NADINE BALUC, Centre de Recherches en Physique des Plasmas, Ecole Politechnique Federale de Lausanne, Villigen PSI, Switzerland.
Irradiation assisted stress corrosion cracking is known to appear in the internal welded components of boiling water reactors such as core shrouds. A core shroud is a welded austenitic stainless steel cylinder, located inside the reactor vessel that directs cooling water around the nuclear fuel and provides structural support for the reactor core and fuel assemblies. The welding process itself introduces features, such as slag and other inclusions, dendritic structure, residual stresses, secondary phases, defects, and phase transformations, which degrade the corrosion properties of the welded joints, as compared to the base material. Although the behaviour of austenitic stainless steels as a base material has been thoroughly investigated, studies on weld metal and heat affected zones have been scarce. The goal of this study is to better understand the effects of the welding cycles and neutron-irradiation on the microstructure and mechanical properties of heat affected zones in two types of austenitic stainless steels: AISI 304 and AISI 347. Test materials were welded using the same procedures as the ones used for the real boiling water reactors, and neutron irradiated up 0.3 dpa. The mechanical properties and microstructure of the materials were characterised prior to and upon irradiation by means of tensile testing using miniaturised specimens, and by transmission electron microscopy observations. It has been found that the microstructure of the heat affected zones is strongly influenced by the thermal cycles and residual stresses induced by the welding process. It contains recrystallised zones, relatively free of dislocations, surrounded by a matrix full of dislocations. The dislocation density decreases as the distance from the fusion line increases. The changes in the microstructure significantly influence the mechanical properties: the yield strength in the heat affected zone is around 370 MPa, while for the base material it is only around 230 MPa. Neutron-irradiation to 0.3 dpa clearly yields the formation of small defect clusters that engender hardening and loss of ductility. Detailed results of irradiation and welding effects have been evaluated and correlated.

abstract: 3.13
ASSESSMENT OF IRRADIATION-HARDENING ON ZIRCALOY AND MARTENSITIC STEELS WITH PUNCH TESTS
EMILIANO NEHUEN CAMPITELLI PHILIPPE SPATIG, Fusion Technology-Materials, CRPP-EPFL, Association EURATOM-Confederation Suisse 5232 Villigen PSI Switzerland.
Zirconium alloys are extensively used in fission reactors as cladding tube material and other in-core structures. Tempered martensitic steels are the most promising material candidates for fusion reactor structural applications. Both undergo severe degradation of their mechanical properties under neutron irradiation due to the production and accumulation of defect clusters in the microstructure. Precise knowledge of these changes is mandatory in order to understand their behavior under normal operating conditions or hypothetical accidents and is the purpose of the present work. Having to deal with different kinds of radioactive materials, special effort has been taken to use small specimens making their machining and handling as simple as possible. Thus, small ball punch tests were chosen. This test consists of deforming by bending a small disk of 3 mm in diameter by pushing on it a 1mm diameter ball. The output of the test is the force vs. deflection of the ball. The small ball punch test has been extensively used to extract the mechanical properties of various nuclear materials, however, a clear understanding of the combined effect of the specimen geometry and the material parameters on the force deflection curves is still lacking. In order to extract the latter, a finite element model of the experiment was implemented on ABAQUS. The model was improved and validated by comparing the experimental and the calculated force deflection curves for the unirradiated conditions. For this condition also tensile test were performed in order to assign the proper constitutive parameters to the specimen.In the case of Zirconium alloys it was also necessary to include the effects of plastic anisotropy

We present punch test results obtained on unirradiated 9% Cr tempered martensitic steel and on Zircaloy cladding tube material. Results are also presented for proton irradiated martensitic steel and BWR in service irradiated Zircaloy cladding tubes. It is shown that the irradiation hardening (yield stress increase) and the post-yield behaviour can be satisfactorily estimated by consistently choosing the constitutive parameters that the model requires in order to reproduce the experimental force-deflection curves for the irradiated condition.


abstract: 3.14
First principle calculation of ideal shear strength in two important mantle silicates
D. FERRE, J. DURINCK, PH. CARREZ, P. CORDIER, Lab. Structure et Proprietes de l'Etat Solide CNRS UMR8008 Universite Lille1, France; A. LEGRIS, Lab. Metallurgie Physique & Genie des Materiaux CNRS UMR8517 Universite Lille1, France.

Mantel convection implies very slow deformation of minerals inside the Earth at pressure and temperature generally out of range of laboratory conditions. If recent developments of high-pressure deformation experiments raise the question of the influence of pressure on plastic deformation of minerals and challenge our understanding of their deformation mechanisms, there also have been huge advances in the use of atomistic computer simulation to study mineral properties and behaviour under extreme conditions.

Our approch is based on the Generalised Stacking Fault concept. It allows us to calculate the energy barrier associated with a rigid shear on a given slip system. The aim is to cast some light on the influence of crystal chemistry of a silicate on plastic shear, especially with a view to understand the origin of the strong plastic anisotropy of low-symmetry minerals.

From the profiles of the energy barriers, an upper bound for the yield strength can be established by considering the ideal shear strength of the perfect crystal. It can be calculated from the restoring force in the generalized stacking fault picture. The results then can be linked to the classic models of the Peierls-Nabarro stress for the dissociated dislocation motion.

In this paper, we present a first-principle study of ideal shear strength in two important minerals of the Earth's mantle, forsterite (upper mantle) and perovskite (lower mantle). The calculations have been performed using the ab-initio calculation package VASP that gives access to the total energy of a periodic system without any experimental input but the atomic numbers of atoms.

When incorporating in the calculations ionic relaxations perpendicular to the fault plane to assess quantitatively the stacking fault energy, our work is also aimed at setting the basis for further modelling at the atomic scale of dislocation core structure.


abstract: 3.16
Plastic Deformation of a monoclinic Al$_{13}$Fe$_{4}$ Embedded in a Al-Al$_{13}$Fe$_{4}$ Composite
SHINYA MIYAZAKI, ATSUSHI. KAWACHI, SHINJI. KUMAI, AKIKAZU. SATO, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama, Japan.

Specimens used in this study were prepared by spray-forming coupled with hot-extrusion made at 773 K. The composition of the alloy was Al-8.3 wt % Fe. The microstructure composed of fine grains and small precipitates of Al$_{3}$Fe was further altered by annealing a 973 K for various period to introduce Al$_{3}$Fe particles of small size (a few ?m) and large size (several tens of ?m) embedded in Al matrix. Thin foil specimens were prepared for transmission electron microscopy by electrolytic jet polishing and the subsequent polishing made to perforation.

The Al-Al$_{3}$Fe (Al$_{13}$Fe$_{4}$) composite foils were deformed by compression at room temperature using a hand made apparatus set on a weighing platform. The small particles of a monoclinic structure were not deformable at all but Al matrix deformed to flow into the hole in a foil specimen. The Al matrix was recrystallized into small grains by the heavy deformation to surround the non-deformed Al$_{3}$Fe particles.

The large particles distributed around the hole in a foil specimen started to deform by application of a stress exceeding 1 GPa. The start of deformation was easily recognized from crack initiation by an optical microscope. The deformed foil specimen was examined by a high voltage electron microscope. The cracks were mainly formed along the (100) planes on which twinning was frequently observed. The residual markings of shear deformation were observed edge-on when a specimen was viewed along the [00] direction and the shear planes were analyzed to lie on the most densely packed (10) planes and on the (50) column planes of a pentagon structure. The unit cell consisting of (13+4)x6=102 atoms contains four pentagon arrangements of Fe atoms when viewed along the [010] direction.

In the thick area near a crack, a group of dislocations were observed in the form of concentric circular loops generated at a dislocation source. Each of the dislocations was found to be composed of a number of dislocations, relevant to the atomic configuration composing the monoclinic crystal structure. Thus, the hardly deformable crystal is shown to deform by dislocation motion under special stress condition.


abstract: 3.18
Dislocation nucleation and movement in helium implanted SiGe/Si(001) Heterostructures studied by in-situ TEM
N. HUEGING, M. LUYSBERG, K. URBAN, Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons c/o Institute for Solid State Research, Research Centre Jülich, D-52425 Jülich, Germany; D. BUCA, B. HOLLäNDER, S. MANTL, Institute of Thin Films and Interfaces, Research Centre Jülich, D-52425 Jülich, Germany; M. MORSCHBACHER, P.F. FICHTNER, Universidad Federal do Rio Grande do Sul, Porto Alegre, Brazil; L. LOO, M. CAYMAX, IMEC, Leuven, Belgium.

The utilization of a combined He ion implantation and annealing procedure has been proven to successfully relax lattice mismatched SiGe/Si(001) heterostructures with a Ge content of up to 30%. Following state of the art relaxation models, the implantation and subsequent annealing induces the formation of dislocation loops nucleated at overpressurized He cavities, which finally figure out a misfit dislocation network. To get a more detailed understanding of the structural process involved, in-situ annealing TEM analyses were performed focusing on SiGe layers grown by CVD on Si (001) substrates. For these purposes layer structures with 19 to 27 at% Ge were implanted with helium doses in the range of $7 10^{15}$ to $1.5 10^{16}$ cm-2 and choosing an adapted ion energy which results in an implantation depth of twice the layer thickness. Accompanying elastic recoil detection analyses (ERDA) and Rutherford backscattering (RBS) channelling experiments were performed using samples ex-situ annealed at 200°C to 950°C to determine the He content and the relaxation degree of the heterostructures. At temperatures of about 380°C He platelets are observed to form even if ERDA results indicate that in this stage only about 30% of the initially implanted He dose remains within the sample. At 650°C the He content falls below the detection limit of the ERDA experiment. Although the He content seems to be approximately zero at these elevated temperatures, still He filled and overpressurized platelets are detected. Therefore, even at higher temperatures, nucleation centres for the dislocation formation are present. At temperatures above 600°C the SiGe layer is observed to relax plastically also reaching a relaxation maximum of 70% at 950°C. Present in-situ TEM analyses reveal, that the nucleation of 60° type dislocations loops close to the edge of the He platelets starts off already at temperatures above 400°C, but these dislocations attached at the platelets seem to be immobile at this stage. The enlargement and movement of the loops starts at temperatures of about 650°C. It is demonstrated, that the dislocation velocity of the loop segments is influenced by both, other dislocations close to the heterostructure interface as well as by the elastic strain fields surrounding other helium platelets located in the implantation layer.


abstract: 3.19
SIZE EFFECT IN METAL MATRIX COMPOSITES
SEBASTIEN GROH, Brown University, Division of engineering - Providence RI - USA; BENOIT DEVINCRE, LADISLAS KUBIN, LEM - CNRS/ONERA - Chatillon - France; ARJEN ROOS, FREDERIC FEYEL, JEAN-LOUIS CHABOCHE, LCME - ONERA - Chatillon - France.
The global mechanical behavior of a metal matrix composite reinforced by unidirectional fibers is affected by microstructural properties such as the volume fraction of fibers and the fiber spacing. In addition, the macroscopic mechanical behavior is also affected by the loading direction. When the loading is applied in a longitudinal direction, parallel to the axis of the fibers, the macroscopic mechanical behavior is essentially controlled by the elastic properties of the fibers. On the other hand, when the loading is applied in a transverse direction, perpendicular to the fiber axis, the macroscopic mechanical behavior is mainly governed by the elasto-plastic properties of the matrix.

A discrete-continnum model, resulting from the coupling of a three dimensional discrete dislocation dynamics simulation and a finite element code, is used to compute the macroscopic mechanical behavior of an $Al/Al_{2}O_{3}$ metal matrix composite loaded in the longitudinal or transverse directions for various fiber volume fractions. The model material is a representative cell in which a periodic boundary conditions are applied. The fibers are positioned on a hexagonal lattice and the volume fractions of fibers is chosen between $5 \%$ and $45 \%$. For these volume fractions, the fiber spacing decreases from $0.82 \hbox{ } \mu m$ to $0.27 \hbox{ } \mu m$. The initial dislocation density is chosen to be $10^{14} \; m^{-2}$.

In such materials, the internal stresses have various origins. To isolate these sources of stress, we compared our results to the macroscopic behavior of the mono-crystalline matrix under the same loading conditions and those obtained with a crystal plasticity model with no length scale. The results obtained show the presence of a size effect on the yield stress which can be predicted using Orowan's Law. This size effect is more important for the transverse loading than for longitudinal loading. For transverse loading, dislocations accumulate at each interface perpendicular to the loading direction which leads to polarize the stress profile near the interface. Finally, an effect of triaxiality is observed with the evolution of dislocation density per glide systems in agreement with experiment.

abstract: 3.23
Stability and slip motion of dislocations in a model metallic glass
TAKAAKI YOSHIHARA, YASUSHI KAMIMURA, KEIICHI EDAGAWA, Institute of Indstrial Science, University of Tokyo, Meguro, Tokyo, Japan. SHIN TAKEUCHI, Dept. of Mat. Sci. and Tech., Science Univ. of Tokyo, Noda, Chiba, Japan.

In general, metallic glasses deform plastically with sharp shear-slip markings on the maximum shear stress plane at low temperatures and the ratio Hv/E (Hv: Vickers hardness, E: Young?s modulus) is commonly about 0.06, irrespective of the alloy system. These facts imply the existence of a common slip mechanism in the deformation of the metallic glasses. So far, several people have proposed slip mechanisms by dislocation glide. However, dynamical properties of dislocations and microscopic mechanism of the dislocation glide in metallic glasses are still not very clear, mainly because powerful experimental tools such as transmission electron microscopy, are not available for metallic glasses. In the present study, the stability and slip behavior of dislocations in a model metallic glass has been investigated by computer simulation to reveal detailed microscopic mechanism of the deformation of metallic glasses.

A model metallic glass of Ni33Y67 with N=9700 atoms was constructed by a molecular dynamics simulation using interatomic potentials derived from a hybridized nearly-free-electron tight-binding-bond theory [1]. For the simulation, the initial state was a high-density liquid at 1923K. After equilibration, the temperature was lowered in 8300 steps to 273K by scaling the velocities of atoms with the total volume kept constant. This corresponds to a quench rate of about 1014K/s. After quenching, the system was equilibrated for 4000 steps. We confirm that the model constructed reproduces satisfactorily the pair correlation functions derived experimentally.

We introduced edge and screw dislocations with various magnitudes of the displacement vector into the model and subsequently relaxed it. We found a threshold value (=2.5A) of the initial displacement; for the initial displacement below the threshold value the final displacement is almost zero while for those above the threshold value the final displacement is almost equal to the initial one. This fact indicates that the minimum Burgurs vector of the dislocation allowed in the model metallic glass is 2.5A. The slip behavior of such a stable dislocation is now being investigated and the results will be presented at the conference.
1. Ch. Hausletner and J. Hafner, Phys. Rev. B 45 (1992) 128.


abstract: 3.24
Creep behaviour of nickel-based Superalloy UDIMET 720
L. GUETAZ, S. DUBIEZ, R. COUTURIER, S. TERZI, CEA-Grenoble, DRT-DTEN, 38054 Grenoble cedex 9, France.
UDIMET 720 is a high strength nickel-based Superalloy which is being considered for applications in high temperature turbine disks for nuclear gas cooled reactor. The intermetallic precipitates gamma prime constitute a large volume fraction (50% pct) of the alloy microstructure. Following standard heat treatment, the microstructure contains three variants of the gamma prime precipitates: the primary gamma prime (with an average diameter of 450 nm), the secondary (40 nm) and the tertiary(10 nm) gamma prime . The 650°C and 700°C creep behaviour of the alloy was investigated under stresses ranging from 600 to 900 MPa. Under higher stresses ($>$800 MPa at 650°C) the creep curves show the classical three creep stages whereas under lower stresses early onset of tertiary creep appears after a short primary creep stage. Creep deformation mechanisms were analysed by transmission electron microscopy. Under the higher stresses, dislocations overcome the gamma prime by the Orowan mechanism whereas under the lower stresses they shear the gamma prime, leaving superlattice stacking faults. The stress transition between these two mechanisms is the Orowan stress, which depends on the gamma prime distribution. Such a transition implies that the shear mechanism is thermally activated. Tensile tests conducted under different strain rates confirm that point. Early onset of tertiary creep on the creep curves is therefore related to the shearing mechanism. Precipitate shearing is often considered as a softening mechanism involved many dislocations gliding along the same shear plane. In the crept alloy no slip localisation was observed. The complex stacking fault left by the first shearing dislocation inside the gamma prime probably do not make the slip on the same plane easier. Then the gamma prime shearing mechanism would not directly explain the creep softening. The reason for observed creep softening could be an increase of mobile dislocation density during creep test. This could be ascribed to the low dislocation velocity during shearing. Finally the creep behaviour was modelled by using a physically based model.


abstract: 3.25
Dislocation structure in severely deformed metals studied by X-ray peak profile analysis
J. GUBICZA, Department of Solid State Physics, Eotvos Lorand University, Budapest, Hungary. L. BALOGH, Department of General Physics, Eotvos Lorand University, Budapest, Hungary. R. J. HELLMIG, Y. ESTRIN, Institute for Materials Engineering and Technology, Clausthal University of Technology, Clausthal, Germany. T. UNGáR, Department of General Physics, Eotvos Lorand University, Budapest, Hungary.

Severe plastic deformation (SPD) is an effective tool for producing bulk nanostructured metals. One of the most common SPD methods is equal channel angular pressing (ECAP), a technique that results in a homogeneous sub-micron grain structure (down to the 100 nm range) of the workpiece [1]. Copper specimens have been deformed by ECAP in 1, 2, 4 and 8 passes. The microstructure has been studied by high resolution X-ray diffractometry using the Multiple Whole Profile (MWP) fitting method [2,3]. The crystallite size distribution and the density of dislocations are determined and discussed. The formation of the submicron grain structure is studied as a function of the number of ECAP passes. It is found that the crystallite size is reduced to a few tens of nanometers even after the first ECAP pass and is not further refined by increasing the number of passes. At the same time, the dislocation density increases gradually up to 4 ECAP passes. The strength is found to be controlled rather by the density of dislocations than by the crystallite size. The thermal stability of the microstructure has also been examined by differential scanning calorimetry (DSC). It is found that during the heat-treatment a bi-modal microstructure is formed, being manifested in a special shape of the peak profiles. This type of the evolution of the microstructure is studied as a function of the number of ECAP passes.

References:
1. S. C. Baik, R. J. Hellmig, Y. Estrin, H. S. Kim, Z. Metallkd. 94 (2003) 754-760.
2. T. Ungár, J. Gubicza, G. Ribárik, A. Borbély, J. Appl. Cryst. 34, (2001) 298-310.
3. G. Ribárik, T.Ungár, J. Gubicza, J. Appl. Cryst. 34 (2001) 669-676.


abstract: 3.26
KEY ROLE AND THE UNIVERSALITY OF DEFORMATION MECHANISMS IN PHASE TRANSITIONS IN SOLIDS, LIQUIDS, BIOLOGICAL TISSUES (TUMOR GROWTH, AGING, ADAPTATION TO STRESS AND MEDICAL TREATMENT ARE INCLUDED)
VALERY P. KISEL , NELLY S. KISSEL, Institute of Solid State Physics, 142 432 Chernogolovka, Moscow district, Russia.
Recent investigations irrefutably show that real crystals, glasses, melts, liquids, gases al-ways contain nuclei and nanoclusters of various phases. The interface stresses due to structural and mechanical mismatch between phases play the key role in phase transiti-ons. The first important goal of this work (the request for the invention) is the universa-lity of the deformation and relaxation mechanisms (DRM) during phase transitions in solids, glasses, liquids, melts, gases and biological tissues [1]. This is confirmed by the correlation of transition parameters for various materials: shear moduli, viscosity, sur-face tension, activation energies of deformation and heat of phase transitions, hysteretic character of their variation, the influence of phase prehistory, the similar reactions to physical and chemical effects, the similarity of kinetic curves for crystallization from the melt or glass state, redox reactions, diffusion, electrical conductivity, electrochemical deposition, adsorption-desorption, martensitic and structural transformations, etc. [1]. Mechanical treatments of phase systems induce some of them to grow at the expense of the others up to chemical compounds forming (mechanical alloying, acoustochemistry). Of specific note is the fact that DRM unravel all the features of tumor growth and meta-static processes, adaptation mechanisms to different types of stress and medical treat-ment for biological systems, etc. Second important finding based on literature data shows the same DRM nature of the effects of ultralow doses (ULD) of physical and chemical impacts (chemical agents, the irradiation of particles, light and electromagnetic fields, etc.) on solids, liquids and biological tissues [1]. These effects are due to mecha-nical hardening and softening on the scales of observation from the atomic (molecular) to microscopic cell structures, macroscopic organisms and populations. It is worth stre-ssing that the dependences of hardening-softening on pulse amplitude and duration are the same for micro- and macrodeformation of all the materials. The stress rate and the dwell time between the pulses (frequency), temperature, impurity concentration, irradi-ation dose of particles, electromagnetic fields, currents, etc. dependences of softening have the same V-shaped form for single and nanocrystals, liquids and biological tissues and organisms [1] (various types of adaptation to stress [2], apoptosis and proliferation of cells [3], aging, etc.).

References:
1. Kisel, V.P. In: Untraditional natural resources, innovation technologies and products. Collected scientific works. Issue 10. RANS ed., Moscow, 2003, pp 183-196 (in Russian)
2. Garkavi, L.Kh., Kvakina, E.B., Kuz'menko, T.S. Antistress reactions and activation therapy. Moscow, RANS, IMEDIS, 1998, 617 p. (in Russian).
3.Piruzian, L.A., Malenkov, A.G., Radkevich, L.A. Dokl. Akad. Nauk, 2004, vol.395, No 2, pp. 261-265.



abstract: 3.27
Deformation twinning in single crystals of Ti3Al with an off-stoichiometric composition (Ti-36.5 at.% Al)
KYOSUKE KISHIDA, National Institute for Materials Science, Tsukuba, Ibaraki, Japan.
HARUYUKI INUI, Kyoto University, Kyoto, Japan.
The characteristics of deformation twinning in single crystals of a D019 compound Ti3Al with an off-stoichiometric composition (Ti-36.5at.% Al) have been studied in detail by trace analysis and transmission electron microscopy. Deformation twinning is operative only at high temperatures above 1000°C when the compression axis is parallel to and close to the hexagonal c-axis. All observed deformation twins are assigned to be of the type II and the twinning elements are determined as: K1: {-2 12 -10 3}, K2: {2 0 -2 -1}, Eta1: $<$5 -1 -4 -6$>$, Eta2: $<$-1 3 -2 2$>$, s = 0.346. Since the twin is of the type-II, the indices for K2 and Eta1 are rational while those for K 1 and Eta2 are basically irrational and thus are only approximate. The twinning process is analyzed in the three steps; uniform shear, shuf fles of atoms to the nearest twin sites and interchange shuffles of misplaced atoms. The distances of shuffles in the second step are as small as 10-20 % of the nearest interatomic distance, facilitating the occurrence of twinning. On the other hand, interchange shuffles in the third step require the atom movements by the nearest interatomic distance for 25 % of the constituent atoms on the basis of the Ti3Al stoichiometry and this process is discussed to be the most important step for the occurrence of deformation twinning. The critical resolved shear stress for the {-2 12 -10 3} twin is estimated to be about 50MPa at 1000°C and it gradually decreases with increasing temperature. The reasons why twinning of this type occurs only under a limited condition (i.e., off-stoichiometric compositions, high temperatures and compression axis being close to the c-axis) are discussed in the light of the atom movements by the deformation twinning.



abstract: 3.28
Temperature dependence of solid solution and particle strengthening calculated by dislocation dynamics)
VOLKER MOHLES, Institut für Metallkunde und Metallphysik, RWTH-Aachen, Aachen, Germany.
When a dislocation overcomes glide obstacles by thermal activation (no climb, no cross-slip), these obstacles (single foreign atoms or very small particles) are not overcome singularly (one by one), but several obstacles along the dislocation line are overcome at the same time. This has already been shown by Labusch and Schwarz (Proc. of ICSMA 9, 1992, 47) by computer simulations. Furthermore it has been demonstrated (Mohles and Roennpagel, Comp. Mat. Sci. 7, 1996, 98) that overcoming a row of obstacles arranged along the line is equivalent to overcoming a wall-shaped obstacle. In the present contribution the free activation enthalpy is calculated (analytically and by dislocation dynamics simulations) as a function of the width and the height of an obstacle wall and the external applied stress. Applying this activation enthalpy in an Arrhenius law leads to a quantitative prediction of the flow stress reduction caused by thermal activation. Most importantly, the flow stress derived from the obstacle wall does not vanish at high temperatures. Kocks (1985) had already pointed out that this behaviour is essential for a viable theory of solid solution strengthening. However the author claimed that the model of an obstacle wall (or: potential through) required a pinning effect along the dislocation line, like the Suzuki effect. The present simulations prove that such a pinning effect is not required, but that the effective wall-shape of the obstacles is an intrinsic effect of the limited flexibility of the dislocation. This holds for all volume fractions of foreign atoms or particles which are relevant for a strengthening effect.