Summary report of XL CALPHAD—Rio de Janeiro, Brasil, 2011

CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Contents lists available at SciVerse ScienceDirect

CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry journal homepage: www.elsevier.com/locate/calphad

Summary report of XL CALPHAD—Rio de Janeiro, Brasil, 2011

article info Keywords: Conference CALPHAD Thermodynamics Modelling First principle calculations

abstract XL CALPHAD—Computer Coupling of Phase Diagrams and Thermochemistry conference was held in Rio de Janeiro, Brazil, May 22–27, 2011. The conference had an audience of 192 persons. There were 88 oral and 100 poster presentations. Presentations were divided according to themes: ab initio; the CALPHAD method, and assessments; applications – energy; applications – steel, superalloys, oxides and experiments and databases. A total of 90 companies were represented among authors and audience. In this summary, a brief description of the highlights of the conference is presented as well as the abstracts for the contributed papers.

1. Introduction In its first edition in the Southern Hemisphere, XL CALPHAD— Computer Coupling of Phase Diagrams and Thermochemistry conference was held in Rio de Janeiro, Brazil, May 22–27, 2011. The conference had an attendance of 192 persons and a total of 88 oral and 100 poster presentations. Presentations were divided according to themes: ab initio; the CALPHAD method, and assessments; applications – energy; applications – steel, superalloys, oxides and experiments and databases. The ab initio sessions provided a broad overview of the relevant topics in the area, with featured papers by J.M. Sanchez, J. Neugebauer, T. Mohri, I. Abrikosov, P. Turchi, C. Colinet. In the CALPHAD method sessions lively discussions were initiated by the overview given by J. Agren in a session chaired by L. Kaufman. The situation concerning lattice stabilities and the current ‘‘inverted pyramid’’ analogy were discussed also in the featured presentations by L. Kaufman, B. Sundman and S. Fries. In the energy applications, featured presentations by P. Rogl, C. Gueneau and B. Sundman lead the sessions with a large number of applications related to nuclear materials, which harmonized with the view given by P. Turchi on the actnides and K. Ishida’s overview on new Co based superalloys. The featured presentations by T. Matsumiya and M. Selleby focused on steels and advanced alloys while R. Schmid-Fetzer discussed databases for magnesium alloys. A significant presence of experimental contributions from well known researchers was also a highlight of this CALPHAD. Finally, B. Sundman, U. Kattner and S. Fries gave overviews of new developments in CALPHAD software, including a significant project on ‘‘open CALPHAD software’’. The schedule was very busy but the attendance was consistently high even with the excellent weather in Rio de Janeiro and an interesting social program. Fig. 1 presents the official XL CALPHAD photograph. The poster sessions were held during four nights after dinner, with a drink service and there was extensive interaction and also lively discussions during these sessions.

http://dx.doi.org/10.1016/j.calphad.2012.07.003

During the CALPHAD banquet, the APDIC Industrial Award was presented to SANDVIK, as well as the CALPHAD best paper and best poster awards, and the APDIC best paper award. L. Kaufman was presented a plaque from the participants of the CALPHAD, in commemoration of his 80th birthday. A significant audience of ‘‘young calphadians’’ took active part in the conference (Fig. 2). Scholarships for students and academics were offered by PETROBRAS, CALPHAD Inc. and STT (Foundation of Computational Thermodynamics, Stockholm, Sweden).

2. Oral presentations 2.1. Ab initio J.M. Sanchez The cluster expansion method: first principles or phenomenology? The cluster expansion (CE) method has been used extensively over the last 25 years to obtain effective cluster interactions (ECIs) that are expected to describe the energy of disordered alloys exhibiting short range order. Furthermore, these ECIs are extracted from the energies of a relatively small set of ordered compounds calculated using first-principles density functional theory (DFT). Thus, from its inception, the objective and/or promise of the CE has been to provide an important ingredient in the development of a first-principles thermodynamic theory of alloys. In this presentation, we will briefly review a rigorous mathematical formulation of the CE [1] that shows that the cluster basis developed by Sanchez, Ducastelle and Gratias [2] is a multidimensional discrete Fourier transform while the general formalism of Sanchez [3] corresponds to a multidimensional discrete wavelet transform. Weaknesses and strengths of the CE method, as well as its potential role in the quest for a firstprinciples thermodynamic theory of alloys, will be explored for several cases ranging from a simple pair potential model to energies obtained using DFT.

112

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Fig. 1. The XL CALPHAD group photo on the St. John Fortress, Rio de Janeiro.

Fig. 2. Young calphadians with L. Kaufman on the occasion of the XL CALPHAD.

References [1] J.M. Sanchez, Phys. Rev. B 81 (2010) 224202. [2] J.M. Sanchez, F. Ducastelle, D. Gratias, Physica 128A (1984) 334. [3] J.M. Sanchez, Phys. Rev. B 48 (1993) 14013. J. Neugebauer, B. Grabowski, A. Glensk, T. Hickel Accuracy and limitations of ab initio approaches in predicting free energies for binaries and unstable phases In order to extend and improve the algorithms and databases needed for the construction of phase diagrams, additional information on, e.g., the volume/pressure dependence of thermodynamic properties, the detailed balance of the various physical excitation mechanisms, and the free energy of unstable phases is crucial. Since an experimental approach to this information is partly not even feasible, the interest in ab initio based methods, in

particular density functional theory (DFT), has been in the last few years re-intensified. Recently we have demonstrated that the presently available DFT exchange–correlation functionals are sufficient to predict the temperature dependence of free energies, heat capacities, thermal expansion coefficients and even defect properties of unary fcc metals with an accuracy that often rivals experimental data [1–3]. While these first results are very promising, many challenging questions remain. For example: how to compute free energies of instable phases that defy well established concepts based, e.g., on the quasiharmonic approximation (due to unphysical imaginary phonon modes) or molecular dynamics? Or: can these methods be extended to chemically more complex alloys and/or to compare several phases on an absolute energy scale? To address these questions, we carefully selected benchmark systems. We have chosen Mg–Si alloys as an example for a binary system and Ca for a system with an unstable phase and a phase transition. Extending our previous approach originally developed for unary systems we were able to compute all relevant free energy contributions such as configurational,

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

electronic, and vibrational contributions. Besides the comparison of the thermal expansion, heat capacity, or the temperature dependent bulk modulus with experimental data, we also compared directly the constant pressure DFT free energies. The results showed over a large temperature range an excellent agreement with CALPHAD data. An extension of the formalism to also include anharmonic contributions also allowed to include unstable phases and to accurately predict the fcc/bcc phase transition of Ca. References [1] B. Grabowski, T. Hickel, J. Neugebauer, Phys. Rev. B 76 (2007) 024309. [2] B. Grabowski, Ph.D. Thesis, University of Paderborn, 2009. [3] B. Grabowski, L. Ismer, T. Hickel, J. Neugebauer, Phys. Rev. B 79 (2009) 134106; see also the Viewpoint in Physics: G. Grimvall, Physics 2 (2009) 28. ˜ o, C.G. Schon H.M. Petrilli, N. Sodre´, P.G. Gonzales-Ormen Ab-initio calculation of the bcc order/disorder relations in the Al–Cr–Fe (aluminum–chromium–iron) system It is widely recognized (e.g., [1]) that chromium is the most important alloying element used in the development of iron aluminides. It is surprising, therefore, that the order/disorder relations in this system are scarcely studied in the literature. Most published works (e.g., [2]) in this system address only the Al-rich corner or the phase relations involving the more complex Al-rich intermetallic phases, leaving the large bcc field empty (although they recognize that B2 order occurs). This lack of experimental and theoretical information is probably caused by extreme experimental difficulties (e.g., the similarity between the atomic numbers of Fe and Cr, which limit the usage of X-ray diffraction data) as well as theoretical difficulties due to the complex magnetic ordering in the Fe and Cr-bearing compounds. This work presents, to the knowledge of the authors, the first ab-initio thermodynamic description of the bcc Al–Cr–Fe system. The calculations are performed in the framework of the so-called cluster expansion method, by combining full potential, linear augmented plane wave (FP-LAPW) electronic structure calculations and the cluster variation method (CVM) in the irregular tetrahedron approximation. References [1] R. Balasubramaniam, On the role of chromium in minimizing room temperature hydrogen embrittlement in iron aluminides, Scr. Mater. 34 (1996) 127–133. [2] M. Palm, The Al–Cr–Fe system—phases and phase equilibria in the Al-rich corner, J. Alloys Compd. 252 (1–2) (1997) 192–200. B. Burton, A. van de Walle First principles phase diagram calculations for the systems: SiC-AlN; SiC-GaN; SiC-InN; and ZrO_X (0oXo1) The ATAT-package [1] was used to perform cluster-expansion method first principles phase diagram (FPPD) calculations for the wurtzite-structure quasibinary systems SiC-AlN, SiC-GaN and SiCInN. In SiC-AlN, planewave pseudopotential formation-energy calculations predict low-energy metastable states with formation energies, Delta-E_f  0.04 eV/mole (mol¼one cation þone anion). The crystal structures of these states are all of the form (SiC)_m(AlN)_n(SiC)_o(AlN)_p...(m,n,o,p integers), where (SiC)_m indicates m SiC-diatomic-layers _9_ to the hexagonal c-axis and similarly for (AlN)_n, (SiC)_o and (AlN)_p.

113

The presence of low-energy layer-structure metastable states helps to explain why one can synthesize (SiC)_{1 X}AlN)_X films, or single crystals, with any value of X, in spite of the apparently strong tendency toward immiscibility. In SiC-GaN, ordered structures are predicted at X¼1/4, 1/2, and 3/4 (Pm, Pmn2_1 and Pm, respectively). In SiC-InN, one Cmc2_1 ordered phase is predicted at X¼1/2. Reference [1] A. van de Walle, G. Ceder, J. Phase Equilib. 23 (2002) 348. G. Hug, P. Jund, R. Viennois, C. Colinet, J.-C. Te´denac Ab initio modeling in the Mg–Si system Mg2Si is a promising material for energy production from thermoelectric effect. In order to control the processing of such materials, an accurate knowledge of the Mg–Si phase diagram and the behavior of the different intrinsic doping defects in required. The Mg–Si system is relatively simple and contains only one defined compound of the cubic anti-fluorite structure (CaF2 prototype) at the Mg2Si stochiometry. However, some discrepancies still remain in the literature between the experimental formation enthalpy of Mg2Si and calculated ones. In particular the existence or not of a high pressure Lave phase in not well established. The formation energy of point defects such as vacancies, interstitials and antisite are also scarcely determined. We present a theoretical study of the relative stability under pressure of the anti-fluorite structure as compared to four other possible structures (Cu2Mg, MgZn2, Ni2In, MgNi2 prototypes). Phonon calculations are performed in selected cases in order to study the temperature relative stability as well. Finally, the formation energy of the different intrinsic defects has been determined. These calculations have been done using several wellestablished ab initio methods including pseudo-potential plane wave, full potential plane waves and full potential local orbitals associated with different exchange and correlation potentials. T. Mohri Progress of theoretical study of alloy phase equilibria based on cluster variation method Cluster variation method (CVM) is one of the most reliable theoretical tools for studying alloy phase equilibria. The conventional CVM, however, does not incorporate local lattice distortion effects into the free energy and, therefore, the calculated phases are still in the excited state. One way to circumvent such inconvenience is to employ continuous displacement cluster variation method (CDCVM). Recently, the author extended CDCVM calculations from two-dimensional square lattice to fcc lattice. In the present talk, review of theoretical calculations of phase equilibria based on CVM is provided.

J. Vˇresˇta´l, J. Pavlu˚ Laves phases in the Hf–V and V–Zr systems: Stability analysis using ab initio data and phase diagrams Systems Hf–V and V–Zr are well known systems where transformation from non-Laves phase to Laves phase exists by increasing temperature [1–4]. The transformation from bodycentered tetragonal or orthorhombic HfV2 or rhombohedral ZrV2 structure to cubic C15 Laves phase at about 100–115 K was described experimentally more in detail in [2–4]. Analysis of this anomaly using ab initio calculated total energies is presented and

114

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

consequences of this analysis for phase diagram of both systems are discussed. Financial support of Czech Science Foundation by Grant no. P108/10/1908 and of Ministry of Education of the Czech Republic no. MSM0021622410 are gratefully acknowledged. References [1] F. Stein, M. Palm, G. Sauthof, Intermetallics 13 (2005) 1056–1074. [2] Y. Zhao, F. Chu, R.B. von Dreele, Q. Zhu, Acta Crystallogr. B 56 (2000) 601. [3] F.R. Drymiotis et al., Phys. Rev. B 72 (2005) 024543. [4] O. Rapp, G. Benediktsson, Phys. Lett. A 74 (197). ˇ J. Pavlu˚, J. Vˇreˇsˇta´l, M. Sob Energetics and magnetism of Mo-based sigma phases The overview of ab initio calculated energies of formation of the sigma phases in Co–Mo, Cr–Mo and Fe–Mo systems will be given. These energies were evaluated with respect to the structures of pure constituents which are stable at standard ambient temperature and pressure (i.e., ferromagnetic hcp Co, antiferromagnetic bcc Cr, ferromagnetic bcc Fe, nonmagnetic bcc Mo) and using various methods [1–5]. These total energy differences are compared with available data. The magnetism of various configurations was also investigated and the stabilizing effect of magnetic ordering in Co–Mo and Fe–Mo was proved. The magnetic moments per atom and their relation to the composition and atomic site will be discussed. Acknowledgements This research was supported by the Grant Agency of the Czech Republic (Project no. P108/10/1908), Ministry of Education of Czech Republic (Project no. MSM0021622410) and Academy of Sciences of the Czech Republic (Project no. AV0Z20410507). The access to the computing facilities of the MetaCenter of the Masaryk University, Brno is acknowledged. References [1] G. Krier, O. Jepsen, A. Burkhardt, O.K. Andersen, Computer ¨ Code TB-LMTO-ASA version 4.6, Max-Planck-Institut fur ¨ Festkorperforschung, Stuttgart, 1994. [2] P. Blaha, K. Schwarz, J. Luitz, Computer Code WIEN97, Vienna University of Technology, 1997 (improved and updated Unix version of the original copyrighted WIEN code, which was published by P. Blaha, K. Schwarz, P. Sorantin, S.B. Trickey, Comput. Phys. Commun. 59 (1990) 399). [3] P. Blaha, K. Schwarz, G.K.H. Madsen, D. Kvasnicka, J. Luitz, WIEN2k, An Augmented Plane Wave Plus Local Orbitals Program for Calculating Crystal Properties, Karlheinz ¨ Wien, Austria, 2001. Schwarz, Technische Universitat, ¨ [4] G. Kresse, J. Furthmuller, Comput. Mater. Sci. 6 (1996) 15. ¨ [5] G. Kresse, J. Furthmuller, Phys. Rev. B 54 (1996) 11169. S.R. Debiaggi, C.D. Toro, G.F. Cabeza, A.F. Guillermet Thermodynamic properties of M–In–Sn (M ¼Cu,Ni) intermetallics: Ab initio database and systematics The phase stability properties of the Cu–In–Sn and Ni–In–Sn intermetallic phases (IPs) are currently a subject of attention in connection with the development of lead-free soldering alloys. In particular, considerable efforts have been devoted to the development

of predictive approaches to the thermodynamics of binary and higher order IPs of actual and potential technological relevance for the leadfree diffusion soldering methods [1,2]. The theme of the present work is the use of ab initio methods for developing a database with theoretical information of various IPs of interest in connection with the thermodynamic analysis of the Cu–In–Sn, Ni–In–Sn and their binary subsystems. Specifically, systematic ab initio density-functional-theory calculations have been performed to establish trends in the 0 K energies of formation (EOF) of several IPs which are stable in the Cu–In, Cu–Sn, Ni–In and Ni–Sn binaries. We used the VASP code [3] with the projector augmented wave potentials and the generalized gradient approximation for the exchange and correlation functions. In this way, various stable, metastable or hypothetical compounds involved in the compound-energy model (CEM) [4] treatment of these binary and higher-order systems are studied. The present contribution includes a discussion of the trends in the EOF vs. composition [5,6]. In addition, new correlations for the EOF are presented which open up the possibility of using these results in CALPHAD optimizations using the CEM. References [1] G. Ghosh, M. Asta, J. Mater. Res. 20 (2005) 3102. [2] G. Ghosh, Metall. Mater. Trans. 40A (2009) 4. ¨ [3] G. Kresse, J. Furthmuller, Comput. Mater. Sci. 6 (1996) 15. [4] M. Hillert, J. Alloys Compd. 320 (2001) 161. [5] S. Ramos de Debiaggi, G.F. Cabeza, C. Deluque Toro, A.M. Monti, S. Sommadossi, A. Ferna´ndez Guillermet, J. Alloys Compd., 509 (2011) 3238–3245. [6] C.E. Deluque Toro, S. Ramos de Debiaggi, A.M. Monti, Phys. B: Condens. Matter, 407 (2012) 3236–3239. A. Talekar, D. Chandra Ternary phase diagrams for plastic crystal energy storage materials—Computational and experimental approach Ternary phase diagrams of PE-AMPL-TRIS, PE-PG-TRIS and PGNPG-TRIS ‘Plastic Crystals’ are important for low and high temperature passive energy storage. Based on our previous work we have found that there are low temperature transitions near 273–300 K, which were not observed in the binaries. In order to predict phase transitions by adding a third component to the binaries, ternary phase diagrams are important. We will present the above ternary phase diagrams using CALPHAD methodology. Low temperature, tetragonal, monoclinic, and orthorhombic phases of these organic compounds transform to orientationally disordered high temperature FCC or BCC phases. The low temperature phases are assumed to be ideal based on the relatively small interaction between them, whereas the high temperature phases are assumed to be subregular. The excess molar Gibbs energies for the binaries involved are fitted using Redlich–Kister polynomial built into the TCC software. We are also performing low and high temperature in-situ X-ray diffraction and differential scanning calorimetry on these ternaries to determine experimental phase diagram of one system for corroborating with the computational work. PE: Pentaerythritol C5H12O4, PG: Pentaglycerine- C5H12O3, NPG: Neopentylglycol-C5H12O2, AMPL: 2-amino-2-methyl-1,3-propanediolC4H11NO2, and TRIS: Tris(hydroxymethyl) aminomethane-C4H11NO3 Program funded by Intel Corporation. I. Abrikosov, B. Alling, M. Ekholm, H. Lind, F. Tasnadi Configurational thermodynamics of multicomponent complexa lloy systems from first-principles theory.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

We review recent advances in first-principles theory of alloy thermodynamics [1] and illustrate its predictive power in studies of decomposition behavior of multicomponent nitride alloys for hard coating applications [2], as well as in studies of influence of magnetic excitations on chemical order–disorder transition in Fe–Ni permalloy [3]. To model the phase diagram of cubic Ti1 xAlxN we develop a unified cluster expansion approach. In our scheme the alloy Hamiltonian is expanded in terms of concentration and volume dependent effective cluster interactions, where the chemical fixed-lattice and local lattice relaxation terms are incorporated. Based on the obtained results, we propose a design route for the next generation of nitride alloys via a concept of multicomponent alloying based on selforganization on the nanoscale via a formation of metastable intermediate products of the spinodal decomposition. We show that in case of TiCrAlN alloys it leads to considerable improvement of thermal stability of thin films for hard coating applications [2]. For random Ni-rich Fe–Ni alloys, we demonstrate from ab initio calculations that deviations of the global magnetic state from ideal ferromagnetic order due to temperature induced magnetization reduction have a crucial effect on the chemical transition temperature. We propose a scheme where the magnetic state is described by partially disordered local magnetic moments, which allows us to reproduce the transition temperature in good agreement with experimental data [3].

115

Very recently, the authors have reported that the variation in stacking fault energy with annealing at 1173 K in Czochralski-grown silicon crystals heavily doped with n- or p-type dopant atoms [1]. In n-type crystals, the stacking fault energy decreased with increasing annealing time. As the concentration of dopant atoms increases, the stacking fault energy decreases. On the other hand, the energy was unchanged during annealing in p-type and non-doped crystals. These results imply that n-type dopant atoms segregate nearby a stacking fault, via their thermal migration. In this research, we investigated that the dopant effect of phosphorus is confirmed by the first principles calculations. The slab model including a stacking fault has been constructed, and been calculated by the VASP (Vienna Abinitio Simulation Package) code with the cut-off energy of 1000 eV and PAW potentials. One P atom is putted in the model and compare the energies and interatomic distances. Fig. 1 shows the site dependency of the energy. The left pannel shows the schematic model of the slab, where the c and h represent the cubic and hexagonal stacking sequences, respectively. Thus the stacking fault are located at the site numbers of 9–10 and 11–12. In this case, because of the small unit size of slab, the replacement of one atom means that the atoms in a whole layer are replaced. The structure energies of P added Si slab show the relative stable when the P atoms are located around the sacking fault. This result strongly suggests that the lowering of the stacking fault energy is due to the P enrichment nearby the stacking fault during the annealing.

References References [1] A.V. Ruban, I.A. Abrikosov, Rep. Prog. Phys. 71 (2008) 046501. [2] B. Alling, A.V. Ruban, A. Karimi, L. Hultman, I.A. Abrikosov, Phys. Rev. B (submitted, available as arXiv:1012.3120v1), H. Lind, R. Forse´n, B. Alling, N. Ghafoor, F. Tasnadi, M.P. Johansson, I.A. Abrikosov, M. Ode´n, in manuscript. [3] M. Ekholm, H. Zapolsky, A.V. Ruban, I. Vernyhora, D. Ledue, I.A. Abrikosov, Phys. Rev. Lett. 105 (2010) 167208. S. R. Nishitani, K. Togase, Y. Ohno, Y. Tokumoto, I. Yonenaga First principles calculations of stacking fault energy of P doped Si crystal

[1] Y. Ohno, T. Taishi, Y. Tokumoto, I. Yonenaga, J. Appl. Phys. 108 (2010) 073514. S.H. Lee, Z.-K. Liu Computational and experimental investigations of defects and stability of (La1–xCax)FeO3  d perovskites (La1  xCax)FeO3  d perovskites have a wide range of applications such as gas separation membranes and cathodes in solid oxide fuel cells. Its performances are dictated by defect structures under service conditions, and in the present work, the defect structures and energetics of (LaxCa1  x)FeO3  d are studied.

Fig. 1. Schematic slab model and the energy dependency on site.

116

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Energetics of various defects are investigated by first-principles calculations based on density functional theory to understand their relative dominance. The interactions of multicomponent and multiple-defects are modeled by the CALPHAD method using the thermochemical data from first-principles calculations and oxygen nonstoichiometry and phase equilibrium data from experimental investigations in the literature and obtained in the present work. Ionic sublattice model is used to model the perovskite phase, and the calculated phase diagrams are in good agreement with experimentally reported phase equilibrium data. Based on the developed models, the oxygen vacancy concentration as a function of partial pressure of oxygen, temperature, and calcium concentration are obtained. T. Markus, J. Emmerlich, D. Music, L. Yang Phase stability and thermomechanical properties of LSCF and BSCF perovskites Active materials applicable as oxygen separation membranes have to exhibit both high ionic and electronic conductivity. Two of the promising materials are La0.6Sr0.4Co0.2Fe0.8O3 d (LSCF) and calculate the thermodynamic stabilities and thermo-chemical properties of LSCF and BSCF such as phase transition temperatures, stabilities, heat capacities, based on ab initio calculations. With the abinitio stability data in hand, thermo chemical software packages will be used to estimate the stability of LSCF and BSCF against chemically aggressive agents. Furthermore, the aim is to calculate elastic constants for the most relevant structural modifications of LSCF and BSCF. To verify our modeling activities, experimental studies using differential scanning calorimetry (DSC) and differential thermal analysis combined with thermo gravimetric measurements (DTA/TG) were conducted. J. Wang, Y. Du, H. Xu, L. Sun, Z.-K. Liu Formation and migration of native defects in complex hydrides Among various solid state storage materials, complex hydrides, which are referred to alkali and alkalineearth metal salts of [NH2]  , [BH4]  and [AlH4]  , have received considerable attention due to their high volumetric and gravimetric density. The major problem associated with the dehydrogenation/hydrogenation of complex hydrides is the slow kinetics. Hydrogen reactions in complex hydrides are generally more complicated than those in conventional metal hydrides since both hydrogen and other constituent elements are involved at various steps of the reactions. Understanding the microscopic mechanism of the hydrogenation and dehydrogenation processes is the key to enhance the kinetics and decrease the decomposition temperature. We have studied defect aspects in Na3AlH6 (the intermediate compound for decomposition of NaAlH4) and LiNH2 (lithium amide) by first-principles calculations based on density functional theory in the generalized gradient approximation. Defect induced local atomic structures, formation energies of defects and migration of dominant defects were investigated. We find that for the above two hydrides energetically favorable defects are in charged states other than in neutral states. For Na3AlH6, we characterized several migration scenarios for all types of H vacancy. Our work demonstrates that the barrier for local diffusion is lower than that for corresponding nonlocal diffusion and H transport in Na3AlH6 is dominated by mobility of positively charged H vacancies. For LiNH2, our investigations show that both Hiþ and H0i correspond to the formation on an NH3 molecular. The energetically favorable  þ H- and Li-related defects are VH  , Hiþ , Hi , VLi , and ILi .

Single positively charged NH2 vacancy (VNH2 ) has the lowest formation energy in N-related defects and the formation energies of N interstitials are substantially higher. The migration barrier for VH  is relatively high (0.88 eV), whereas charged H interstitial has moderate diffusion barriers (0.40 eV for Hi and 0.46 eV for Hiþ ). The Li-related defects are predicted to be diffusion readily  þ with a migration barrier of just 0.20 eV for VLi and 0.38 eV for ILi . The predicted activation energies for self-diffusion show that the formation of H interstitial is the bottleneck for H transport and Lirelated defects are the predominant diffusion species in LiNH2. Acknowledgements The financial support from the National Natural Science Foundation of China (Grant nos. 20833009, 50721003 and 50831007) is greatly acknowledged. References [1] J.C. Wang, Y. Du, et al., Int. J. Hydrogen Energy 35 (2009) 609–613. [2] J.C. Wang, Y. Du, et al., Appl. Phys. Lett. 95 (2009) 111910/1–3. B.-J. Lee, J.-W. Jang, K.-S. Han, C.-H. Han A comparative analysis of the effect of particle pinning and solute segregation on the inhibition of grain boundary movement Microstructural evolution has a decisive effect on properties of polycrystalline materials. Inhibition of grain boundary movement is often an important process to control the microstructure. Particle pinning has long been used as a means to inhibit the grain boundary movement. Even though a similar inhibition is expected from the segregation of solute atoms on grain boundaries, its effect is not quantitatively known. The purpose of the present study is to quantitatively evaluate the possibility of the use of grain boundary segregation as an inhibitor of grain boundary movement, by comparing the pinning pressure from AlN precipitation and Sn segregation in bcc Fe using atomistic simulation techniques (molecular dynamics and Monte Carlo simulation). The way of evaluating grain boundary mobility [1], molecular dynamics simulation to estimate the pinning pressure, interatomic potentials [2] for the atomistic approaches and the way of extrapolating simulation results to real scale will be presented as well as the final results, the relativeability between AlN pinning and Sn segregation for the inhibition of grain boundary movement in bcc Fe. References [1] K.G.F. Janssens, D. Olmsted, E.A., Holm, S.M., Foiles, S.J. Plimpton, P.M. Derlet, Nat. Mater. 5 (2006) 124–127. [2] B.-J. Lee, W.-S. Ko, H.-K. Kim, E.-H. Kim, CALPHAD 34 (2010) 510–522.

Patrice E.A. Turchi, Alexander I. Landa Actinide alloys and their challenges The current interest in a better usage of actinide-based mixtures in nuclear reactors and the applicability of nuclear fuels in hybrid fusion–fission systems raises challenging questions on the increasing role of minor actinides and fission products and gases on fuel stability

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

and performance. Nuclear materials under extreme conditions of temperature and irradiation with a temporal evolution of materials chemistry require a fundamental knowledge that must feed into physics-based fuel performance codes to build in predictability and to tackle the question of fuel behavior under normal and also accidental operating conditions. Hence, the prediction of phase stability trends and phase diagrams of multi-component complex actinide-based alloys is undoubtedly the Holy Grail of alloy physics and materials properties simulations. In this presentation, we will first discuss some of the most promising fuel forms that are currently considered for advanced nuclear reactors and hybrid fusion-fission engines. Then, we will show that despite the urgent need for more experimental work, the CALPHAD methodology, combined with appropriate first-principles electronic structure calculations, is a powerful tool to investigate and predict the thermodynamic properties of actinide based alloys. We will focus on three different themes: the Pu solubility challenge in nuclear molten salts, the Pd attack in nuclear TRISO particles, and the possible improvement in metallic fuels to minimize the so-called site-redistribution. Ultimately, since experiments on actinide alloys are costly and challenging, we will show that despite their own limitations ab initio results may provide useful guidance for well-chosen experiments that can lead to full validation and verification of the thermodynamic driving force that is required for subsequent work on materials performance. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344.

117

[6] G. Kresse, D. Joubert, Phys. Rev. B 59 (1998) 1758. [7] J.P. Perdew, Y. Wang, Phys. Rev. B. 45 (1992) 13244. [8] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C. Fiolhais, Phys. Rev. B 46 (1992) 6671. [9] J. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865. J.-M. Joubert, A. Mascaro, J.-C. Crivello,C. Toffolon-Masclet Ab-initio calculations of Er–H compounds: Application to the modelling of the phase diagram Metal–hydrogen systems are interesting from both the fundamental and application point of views. Up to now, not so many systems have been studied by ab initio calculations and/or modelled by the Calphad technique and it is a growing field of research. In the present work, the reaction of hydrogen with erbium has been studied by first-principles calculations. The enthalpies of formation of the compounds Er, ErH, ErH2 and ErH3 generated by ordered insertion of hydrogen atoms in the tetrahedral and octahedral (or triangular) sites of the fcc and hcp structures have been calculated. The results of the calculations are consistent with the experimental observation of the sequence Er (hcp)-ErH2 (fcc)-ErH3 (hcp), qualitatively, and with the measured enthalpies of formation, quantitatively. These results have been used to describe the stable and metastable end-members generated by the compound energy formalism for the two phases and a description of the phase diagram assessed by the Calphad technique will be proposed.

C. Colinet, J.-C. Tedenac

J.-C. Crivello, M. Palumbo, T. Hammerschmidt, S.G. Fries, J.-M. Joubert

Structural stability of the ternary compounds Mo5SiB2 and Nb5SiB2 Experimental investigations in Mo–Si–B [1,2] and Nb–Si–B [3] systems show the existence of the ternary phases Mo5SiB2 and Nb5SiB2 crystallizing in the B3Cr5-prototype structure. In the present work, we have performed ab-initio calculations of lattice parameters and energies of formation for the Mo5SiB2 and Nb5SiB2 ternary compounds by considering three possible competiting structure types: Mn5Si3-prototype (D88), W5Si3-prototype (D8m or T1) or B3Cr5-prototype (D81 or T2). These calculations have been carried out using the Vienna ab initio simulation package (VASP) [4], making use of potentials constructed by the projector-augmented waves (PAW) technique [5,6]. For the GGA exchange correlation function, the Perdew-Wang (PW91) [7,8] or the Perdew, Burke and Ernzerhof (PBE) [9] parameterizations are employed. We show that the Mo5SiB2 and Nb5SiB2 ternary compounds are effectively stable in the B3Cr5prototype structure. The calculated lattice parameters as well as the internal parameters are in good agreement with the experimental determinations. Both electronic and size effects allow to explain the stability of the Mo5SiB2 and Nb5SiB2 ternary compounds in the B3Cr5prototype structure.

Calculating heats of formation using DFT: Guidelines for the study of TCP phases Calculating heat of formation using density functional theory (DFT) has been in common usage for several decades, in particular in conjunction with the Calphad method. In fact, Calphad thermodynamic modeling requires a complete energetic description of a system, which could be limited by experimental works alone. To compensate for the lack of information, DFT calculations can provide formation enthalpies at 0 K of any ordered compounds, even in metastable states. The aim of this work is to present our experience of first-principles calculations to estimate the heat of formation of compounds formed at 0 K, with a focus on the complex topologically-close packed phases (TCP). Taking the chi and sigma phases as examples, we present calculations of all ordered configurations for several rhenium-based systems. From these results, we discuss the influence on the converged energy (i) of the basic pseudo-potential parameters, (ii) of the exchange– correlation functional LDA and GGA (PW91 or PBE), (iii) of the magnetic contribution, and (iv) of the number of valence electrons considered with different pseudo-potentials. In each case, we pay attention to the validity of ab initio results in comparison with the experimental data available: lattice parameters, internal coordinates, site occupancies and formation enthalpies.

References [1] J.H. Schneibel, C.T. Liu, L. Heatherly, L. Kramer, Scr. Mater. 38 (1998) 1169. [2] C.A. Nunes, R. Sakidja, Z. Dong, J.H. Perepezko, Intermetallics 8 (2000) 327. [3] D.N.P. Junior, C.A. Nunes, G.C. Coelho, F. Ferreira, Intermetallics 11 (2003) 251. ¨ [4] G. Kresse, J. Furthmuller, Comput. Mater. Sci. 6 (1996) 15; ¨ G. Kresse, J. Furthmuller, Phys. Rev. B 54 (1996) 11169. ¨ [5] P.E. Blochl, Phys. Rev. B 50 (1994) 17953.

W. Xiong, P. Korzhavyi, Q. Chen, M. Selleby Bridging ab initio and CALPHAD by a robust magnetic model The ab initio and CALPHAD methods have developed rapidly during the last decades as the main tools for materials design. However, both methods have their own limitations. CALPHAD shows its power above 300 K while ab initio usually does at 0 K. Therefore, extending the CALPHAD method down to temperatures

118

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

below 300 K is an interesting topic. In this work, the state-of-theart magnetic model of the CALPHAD approach is reviewed. It is found that, in order to accurately predict magnetic contributions to the Gibbs energy of alloys, the Inden–Hillert–Jarl (IMJ) model [1,2] needs to be improved. Firstly, the relation between magnetic moment and magnetic entropy is elucidated to assure the correct estimation of the magnetic entropy. Secondly, a new function is adopted for describing the Curie/Ne´el temperature and magnetic moment in order to evaluate the magnetic ordering energy due to the transformation between different magnetic states. The Fe–Cr– Ni system is studied to illustrate the advantages of the present model. The model-predicted magnetic contributions are distinctly different using the IMJ model and the present one. The present work demonstrates a way to improve the thermodynamic descriptions at low temperatures by adopting the modified magnetic model and the new generation of lattice stability [3]. This work was performed within the VINN Excellence Center Hero-m, financed by VINNOVA, the Swedish Government Agency of Innovation Systems, Swedish Industry and KTH.

iron, carbon and cementite [1,2]. The metastable Fe5C2-H¨agg and Fe2C-epsilon carbides are included. A four-sublattice model is also developed to describe thermodynamic properties of body centered tetragonal (bct) martensite as an ordered form of the bcc phase. The proposed model can describe the order–disorder transformation and the spinodal decomposition in Fe–C martensite. Using the proposed model the effect of ordering of the stability of martensite is evaluated and compared with experimental data and previou References [1] Q. Chen, B. Sundman, J. Phase Equilib. 22 (6) (2001) 631– 644. [2] B. Hallstedt, D. Djurovic, J. von Appen, R. Dronskowski, A. ¨ Dick, F. Kormann, T. Hickel, J. Neugebauer, CALPHAD 34 (1) (2010) 129–133. D. Chandra, A. Talekar, J. Lamb, W.-M. Chien, D. Phanon, N. ˇ ´ , K.Yvon, J.-C. Crivello, M. Latroche, M. Gupta Penin, R. Cerny

References [1] G. Inden, in: Proceedings of the CALPHAD, V. Max Planck Institut fuer Eisenforschung, Duesseldorf, 1976, p. 1. [2] M. Hillert, M. Jarl, CALPHAD 2 (1978) 227. [3] Q. Chen, B. Sundman, J. Phase Equilib. 22 (2001) 631. 2.2. The CALPHAD method, and assessments ˚ J. Agren Calphad – what did we hope for, what did we get – where do we go? In the early 1980s the Calphad technique started to spread outside the Calphad community, which was quite small at that time and had a hard core of less than 50 individuals. Today the technique has become a standard scientific and engineering tool among other tools that are used to solve various problems. In the early 1970 many scientists were inspired by the pioneering work of Larry Kaufman and of Mats Hillert which so far had proceeded quite independently. Thanks to the newly launched Calphad conferences and the European efforts within SGTE an intensive international collaboration started. At that time the visions were clear and simple; once general softwares and databases were available almost any thermodynamic calculation would be possible. In this talk the initial visions will be analyzed. Soon unforeseen problems related to the compilation of databases and the deficiencies of the softwares lead to a modification of the visions. On the other hand, it was perhaps unexpected that Calphad could be applied more or less directly in industrial engineering which led to new requirements. The use of Calphad as a cornerstone in other types of simulations became evident. For some years the practical applicability led to commercial values and divergence of the databases. In contrast to the original vision of the single unified database several high-precision special purpose databases became commercially available. When we now meet the future it may come as a surprise that it is the industrial applications that call for a new generation of highprecision databases based on much more fundamental science.

Ternary phase diagram of Li–N–H complex hydrides for hydrogen storage systems Light-weight complex hydrides with high hydrogen capacity are becoming important for vehicular applications. The pioneering work of Chen et al. [1] led to discovery of about 10 wt% hydrogen using amideimide system, but the capacity decreases as function of cycling to  3 wt%H. Several other investigations [2–8] led to a better understanding of the system. We present our work on Li3N (under the USDOE’s Metal Hydride Center of Excellence program), Li amide–imide, in which we detected new phases during hydrogenation of the Li3N compound. The understanding of the phase stability was mainly derived from thermodynamic calculations of the Li–N–H ternary systems at different temperatures and pressures by using the FACT program. These calculations explained the stability of phases obtained experimentally. Hydriding of Li3N (starting compound) yielded a Li coordinated (interstitial type) Li4NH compound formed at low hydrogen pressures (  0.0003 MPa), Li2NH ( 0.0005 MPa), a new phase at  0.02 MPa Li1.5NH1.5, and LiNH2 ( 0.1 MPa). Calculated ternary phase diagrams predicted reaction pathways and also suggested as to why there were losses in hydrogen capacity. An important observation is that the addition of N2 to hydrogen gas improved the reversible hydrogen capacity in this system. We developed Li– N–H complex hydrides that use 80/20H2/N2 mixtures that show significant enhancement of the reversible capacity as compared to cycling without nitrogen additives. We attribute this enhancement to the reaction of nitrogen with liquid lithium during cycling. The Gibbs free energy of formation of Li3N (DGo ¼ 98.7 kJ/mol) is more negative than that of LiH (DGo ¼ 50.3 kJ/mol). Based on the computations, we propose that the mitigation of hydrogen capacity losses is due to the destabilization of the LiH phase that tends to accumulate during cycling. We will present calculated ternary phase diagrams, experimental and calculated thermodynamic, neutron and X-ray diffraction results and structure electronic by DFT calculations. References

R. Naraghi, M. Selleby Reassessment of the thermodynamic description of the Fe–C system A new thermodynamic assessment of the Fe–C system is performed based on the new more physically accurate descriptions of

[1] P. Chen, Z. Xiong, J. Luo, J. Lin, K. Tan, Nature 21 (2002) 302. [2] F.E. Pinkerton, J. Alloys Compd. 400 (2005) 76. [3] T. Ichikawa, N. Hanada, S. Isobe, H.Y. Leng, H. Fujii, J. Phys. Chem. B 108 (2004) 7887. [4] M.P. Balogh, C.Y. Jones, J.F. Herbst, L.G. Hector Jr., M. Kundrat, J. Alloys Compd. 420 (2006) 326.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

[5] E. Weidner, D. Bull, I. Shabalin, S. Keens, M. Telling, K. Ross, Chem. Phys. Lett. 444 (2007) 76. [6] W.I.F. David, M.O. Jones, D.H. Gregory, C.M. Jewell, S.R. Johnson, A. Walton, P.P. Edwards, J. Am. Chem. Soc. 129 (2007) 1594. [7] F. Zhang, Y. Wang, M.Y. Chou, Phys. Rev. B 82 (2010) 094112. [8] J.C. Crivello, M. Gupta, M. Latroche, D. Chandra, Phys. Rev. B 81 (2010) 104113. C. Li Thermodynamic assessment of V2O3–TiO2 System The composite oxides involving the V2O3, TiO2 and other oxides could be used as the oxidation catalyst or the refractory for melting the high reactive alloy, the phase diagram of the V2O3–TiO2 pseudobinary system is the fundamental to these practical applications, however, no diagram fully assessed of this system is published. Therefore, a set of critically evaluated thermodynamic parameters of the V2O3–TiO2 pseudo-binary system is urgent needed. In order to obtain an assessed V2O3–TiO2 diagram which corresponds with the experimental data well, the existing phase diagram and thermodynamic data of V2O3–TiO2 system are assessed, and the phase diagram are calculated and optimized by using the CALPHAD technology, and some compounds with argument are discussed. In this calculation, the liquid phase, V2O3, rutile TiO2, V2TiO5, V2Ti2O7 and V2Ti3O9 are treated with the substitutional solution model. A self consistent set of the optimized thermodynamic parameters has been derived, which is in better agreement with most of the experiments. Compared with experimental data and the results in this work as well as the results reported previously, it is demonstrated that the present thermodynamic assessment is in better agreement with most of the experiments.

119

more than 100 years. Applying in-situ neutron diffraction it was shown that the e phase has the formula Fe5Al8 and its structure is of the body-centred cubic g1 brass type.

H. Wang, N. Warnken, R.C. Reed Phase field simulation on beta/alpha transformation in TiAlV system coupled with CALPHAD assessment Ti alloys are used widely for aerospace applications. However, in practice, there are some aspects of the physical metallurgy of these materials – and particularly the beta-alloys which display the body-centered cubic (bcc) structure – which are not well understood. Our aim here is to assess the available thermodynamic information relevant to Ti–Al–V system, and thus to design a thermodynamic database which could then be used to identify alloy compositions which are prone to the ordering reaction. This then might be used for the purposes of alloy design; since the ordering reaction is known to affect properties, one might then avoid compositions where this effect might occur. Based upon the thermodynamic driving forces calculated by the new thermodynamic database, interdiffusion in the disordered A2 phase of the ternary Ti–Al–V system is assessed and optimised. A set of parameters describing the atomic mobilities of the disordered beta phase are given, which is used in the quantitative models of alpha precipitation from beta matrix. References [1] [2] [3] [4]

I. Steinbach, Modell. Simul. Mater. Sci. Eng. 17 (2009) 1. F. Zhang, et al., J. Mater. Eng. Perform. 14 (2005) 717. N. Dupin, I. Ansara, Z. Metallkd. 90 (1999) 76. Q. Chen, et al., Scr. Mater. 50 (2004) 471.

F. Stein, M. Palm, S. Voß, C. He, O. Dovbenko, O. Prymak Experimental investigations of phases, phase equilibria, and melting behaviour in the systems Fe–Al–Nb and Co–Al–Nb and their terminal binary systems Experimental isothermal sections and partial liquidus surfaces of the ternary systems Fe–Al–Nb and Co–Al–Nb and new results on their binary boundary systems Fe–Al, Co–Al, Fe–Nb, and Co– Nb are presented. Microstructures, phase compositions, crystal structures, and phase transition temperatures were studied by applying differential thermal analysis, electron probe microanalysis, X-ray and high-temperature neutron diffraction as well as various metallographic methods. The ternary sections of both systems are dominated by extended phase fields of a C14-type Laves phase which in case of the Co-containing system coexists with a C15 and a C36 Laves phase. Other remarkable features are extended phase fields of the m phase in both ternary systems and the occurrence of the very slowly forming L21-type Heusler phase Co2AlNb. Two particularly interesting results from the binary systems concern the B2 CoAl phase, which we found to have a significantly higher melting temperature than previously reported in the literature, and the e hightemperature phase of the Fe–Al system, the crystal structure of which was unknown until recently although the existence of this phase is known since

Larry Kaufman Computation materials design The development of high performance corrosion resistant materials (HPCRM) alloys that are damage tolerant and applicable as a coating to a wide variety of substrates under a wide range of conditions is a remarkable achievement that has been accomplished is just over two years by the dedicated efforts of experienced scientists and engineers [1] who made judicious choices in charting the direction of their work. The reduction to practice based on amorphous metal alloys has left some gaps in knowledge which has been filled by using computational thermodynamics based on CALPHAD Methods employed to define the unstable Fe3B phase as the nucleation site and the glass transition and T(lim), the transformation limiting temperature, as illustrated below. The iron based amorphous glasses of interest contain iron, boron, carbon, chromium, manganese, silicon, yttrium and tungsten and are subjected to a wide variety of heating and cooling rates that cover sixteen orders of magnitude in the time variable that can be used to describe the synthesis, fabrication and application of these materials as bulk components or corrosion resistant coatings of useful structures.

120

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Reference [1] L. Kaufman, Computational materials design, J. Phase Equilib. Diffus. 30 (5) (2009) 418–428. E. Povoden-Karadeniz, E. Eidenberger, P. Lang, H. Leitner, E. Kozeschnik Thermodynamics of the Fe–Co–Mo system and modeling of early precipitation in Fe–25 at%Co–9 at%Mo alloy Solution-treated and quenched Fe–25 at%Co–9 at%Mo alloys show a remarkable increase in hardness during ageing at elevated temperatures [1]. The tremendous role of semicoherent m-phase, (Co,Fe)7Mo6 for precipitation hardening has been demonstrated in several experimental studies. Systematic investigations of the microstructure and composition of these precipitates suggest a close relation of m-phase formation to initial decomposition processes [2]. However, up to now the mechanisms, which govern these early decomposition stages in the Fe–Co–Mo system are not well understood, and explained controversially in the literature. Chain-like regions along elastically soft o1004 Matrix-directions have been interpreted either as spinodal decomposition [3], or as metastable coherent, nanoscale bcc-structured nuclei with compositions that differed significantly from the composition of macroscopic precipitates predicted by the phase diagram [4]. We apply classical nucleation theory in combination with a CALPHAD assessment of thermodynamic model parameters of the Fe–Co–Mo system, in order to verify the existence of potential precursor phases of the m-phase in Fe–25 at%Co–9 at%Mo alloy. The theoretical nucleation and growth regime is distinguished from the composition range of spinodal decomposition in the Fe–Co–Mo system by calculating the coherent spinodal, with the Gibbs energies of the phases in the Fe–Co–Mo system being assessed over a wide range of temperatures. For compositions outside the spinodal, the most probable nucleus composition of nano-scaled particles can then be predicted by minimization of the critical energy of the nucleation event. Recently, this strategy has been successfully applied to the simulation of bcc-Cu precipitation in Fe–Cu [5]. Since the molar Gibbs energy is included in the minimum energy equation for the nucleation event, the

thermodynamic assessment of the Fe–Co–Mo system is an essential part for a correct prediction of the nucleus composition. The determined nucleus composition of potential pre-m precipitates in Fe–25 at%Co–9 at%Mo alloy is used as input parameter for the thermo-kinetic simulation of the precipitation sequence during continuous heating of solution-treated and quenched Fe–25 at%Co– 9 at%Mo alloy. We predict nucleation and growth of bcc-type particles before the precipitation of m-phase. The calculated composition of the pre-m phase is in good agreement with 3D atom probe measurements. References ¨ [1] W. Koster, W. Tonn, Arch. Eisenhuttenwes 12 (1932) 51. ¨ [2] E. Eidenberger, M. Schober, T. Schmolzer, E. Stergar, H.Leitner, P. Staron, H. Clemens, Phys. Status Solidi A 207 (2010) 2238. [3] T. Kozakai, H. Aihara, M. Doi, Trans. ISIJ 25 (1985) 159. [4] D. Isheim, O.C. Hellman, D.N. Seidman, F. Danoix, D. Blavette, Scr. Mater. 42 (2000) 645. [5] E. Kozeschnik, Scr. Mater. 59 (2008) 1018.

M. Hosseinifar Experimental investigation and thermodynamic optimization of the Al–Mg–Nd system An addition of rare-earth (RE) metals to Mg–Al cast products results in an improved creep resistance which makes these alloys (known as AE family) suitable for automotive powertrain applications [1]. To make this material economically competitive, instead of adding rather costly individual RE elements, a less expensive mixture of them (mischmetal) is utilized. In order to realistically model the mischmetal addition, the corresponding thermodynamic database must at least include major constituents of mischmetal, namely, cerium, lanthanum and neodymium. The Al–Mg–Ce and Al–Mg–La systems are already optimized [2,3]. To make this database complete Al–Mg–Nd system needs to be assessed as well. An obstacle hindering the assessment of the Mg–Al–Nd system is that even though limited

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

experimental information is available in the literature [4,5], the reliability of these data is questionable. For instance, it is proved beyond doubt that the reported liquidus temperatures are burdened with a systematic error [2,3]. In the present investigation, the 400 1C isotherm of the Al–Mg–Nd system was re-examined using diffusion couple technique. The homogeneity ranges of ternary phases in this system were determined utilizing EPMA and SEM-EDS analysis of three diffusion couples. Similar couples were also prepared at 500 1C. The determination of the solubility limits of ternary phases at two temperatures was instrumental in fixing the temperature dependent variables during optimization process. Thermal analysis experiments were conducted on two Al-rich ternary alloys. Substantially higher liquidus temperatures were recorded in the present work compared to previously reported data [5]. These high temperatures correspond to the primary solidification region of the Al11Nd3 phase. The Al–Mg–Nd system is optimized based on the experimental findings of the present study. References [1] M.O. Pekguleryuz, A.A. Kaya, Adv. Eng. Mater. 5 (12) (2003) 866. [2] J. Grobner, D. Kevorkov, R. Schmid-Fetzer, Intermetallics 10(5) (2002) 415. [3] M. Hosseinifar, D.V. Malakhov, J. Alloys Compd. 505(2) (2010) 459. [4] K.H.O. Odinaev, et al., Izv. Vyssh. Uchebn. Zaved., Tsvetn. Metall. 4 (1988) 94. [5] K.H.O. Odinaev, I.N. Ganiev, A.Z. Ikromov, Russ. Metall. 4 (1996) 153. E. Povoden-Karadeniz, P. Warczok, P. Lang, A. Falahati, M.R. Ahmadi, E. Kozeschnik Thermodynamic modeling of metastable phases in Al–Cu–Fe– Mg–Mn–Si and applications to precipitation kinetics simulations Due to their good hardenability response by nano-scale metastable precipitates, Al-alloys of the Al–Cu–Fe–Mg–Mn–Si system, such as AA2024 (Al–Cu), AA6016 (Al–Mg–Si), AA6082 (Al–Mg–Si– Fe–Mn), are widely used as high-strength materials in many industrial applications. Whereas the stable phases of the Al–Cu–Fe–Mg– Mn–Si system, such as the AlFeSi-phases [1] and the Q-phase [2] have been assessed recently, a systematic CALPHAD assessment of metastable phases is not yet available. In this paper, we present thermodynamic models of metastable Mg–Si co-clusters, GP-zones,

121

b00 -, b0 -, B0 -, Y-, y00 -, and y0 -phases, which are successfully tested in thermo-kinetic precipitation simulations. Particular focus is put on the model development for disordered co-clusters and coherency strain-induced ordered particles (GP-zones) at the very early stages of precipitation (low temperatures from room temperature up to about 180 1C). Specific properties of these early structures, such as their affinity to vacancies that form during quenching after solution treatment, obviously influence the precipitation of hardening phases like b00 at higher temperatures. MgSi co-clusters and GP-zones are almost fully coherent with the fcc Al-Matrix. Hence, in a first approximation these structures are associated with the model description used for the fcc Al-Matrix: Mg–Si co-clusters are simply described as highly metastable Mg–Si solid solution (Mg,Si)(Va), and GP-zones are treated as an fccbased ordered phase by using a splitmodel with 4 substitutional sublattices. For Al–Mg–Si GP-zones, the preferred sublattice occupancy then reads (Al)(Al)(Mg)(Si)(Va), analogous to the L10 structure in the case of chemical ordering. The crystallographic Mg–Si layering proposed by Matsuda et al. [3] can be reproduced. The thermodynamic model parameters of Mg–Si co-clusters are based on new energy data determined by firstprinciples and optimized with experimental differential scanning analysis data. Subsequent thermokinetic test-simulations aim at giving best reproduction of experimental particle sizes and number densities determined after various heat treatments [4]. References [1] J. Lacaze, L. Eleno, B. Sundman, Metall. Mater. Trans. A 41 (2010) 2208. [2] K. Chang, S. Liu, D. Zhao, Y. Du, L. Zhou, L. Chen, Thermochim. Acta 512 (2011) 258. [3] K. Matsuda, H. Gamada, K. Fujii, Y, Uetani, T. Sato, A. Kamio, S. Ikeno, Metall. Mater. Trans. A 29 (1998) 1161. [4] A. Falahati, E. Povoden-Karadeniz, P. Lang, P. Warcok, E. Kozeschnik, Int. J. Mater. Res. 101 (2010) 1089. P. Zhou, S. Cui, D. Liu, L. Zhang, Y. Du, W. Jie Assessment of the atomic mobilities for Fcc_A1 and Bcc_A2 Cu–Fe–Zn alloys Based on critically reviewed experimental diffusivities available in the literature, the assessment of atomic mobilities for fcc Fe–Zn, bcc Cu–Fe, Fe–Zn, Cu–Zn and Cu–Fe–Zn alloys was

Fig. 1. Calculated inter-diffusion coefficients of bcc Fe–Zn (a) and bcc Cu–Zn (b) alloys in comparison with the experimental data. A constant M is added to separate the data.

122

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

performed using the DICTRA software. Comprehensive comparisons show good agreements between the presently calculated diffusion coefficients and the experimental data. The reliability of the atomic mobilities obtained was further verified by comparing various model-predicted diffusion phenomena with the experimental information, such as concentration profiles, interdiffusion flux, and diffusion paths in a series of binary and ternary diffusion couples Fig. 1. Acknowledgement The financial support from National Basic Research Program of China (Grant no. 2011CB610401) is greatly acknowledged.

out to determine lattice parameters, energies, bulk moduli and the vibrational contribution to the free energy in the harmonic and quasi-harmonic approach. One of the results, which we will discuss, is the partitioning of physical effects in the heat capacity. We have tested a combined approach using first principles and fitting methods. We show results of this approach by comparing our calculations with experimental data. Both the Debye and Einstein model plus additional contributions for electronic excitations, quasi and anharmonic vibrations and effects of lattice vacancies were tested by adapting optimization programs from Lukas [4]. We show a numerical scheme to calculate the Debye integral, which we implemented in this program. Acknowledgement

K. Ishida, T. Omori New Co-base superalloys—phase equilibria and applications Phase stability of intermetallic compounds in Co-base alloys are reviewed compared with those in Ni-base alloys focusing on the GCP (geometrically closed-packed) phase. Although only Co3Ti is known to be a stable g0 (L12) compound, there are some metastable g0 phases in the Co–X binary system such as Co3Al, Co3W, Co3Ta, etc. Recently, we discovered Co3(Al,W) ternary compound with L12 structure [1,2], which shows the positive temperature dependence on flow stress [3]. Therefore, the Co–Al– W base alloys have potential for new-type of superalloys. In this paper, the phase equilibria and some properties of Co–Al–W base alloys will be shown. The applications of these new Co-base superalloys will also be presented.

ThyssenKrupp AG, Salzgitter Mannesmann Forschung GmbH, Robert Bosch GmbH, Bayer Materials Science AG, Bayer Technology Services GmbH, Benteler Stahl/Rohr AG, the state of North Rhine-Westphalia and the European Commission in the framework of the European Regional Development Fund (ERDF). References [1] B.J. Lee, A summary of the CALPHAD XXXIX Conference, CALPHAD 35 (1) (2011) C1–C21. [2] A.T. Dinsdale, CALPHAD 15 (4) (1991) 317. [3] M.H.G. Jacobs, R. Schmid-Fetzer, Phys. Chem. Min. 37 (10) (2010) 721. [4] H.L. Lukas, Reference Manual, 1995 /http://www.met.kth. se/ bosse/BOOK/lukamanu.zipS.

References [1] J. Sato, T. Omori, K. Oikawa, I. Ohnuma, R. Kainuma, K. Ishida, Science 312 (2006) 90. [2] K. Ishida, Mater. Res. Soc. Symp. Proc. 1128 (2009) 357. [3] A. Suzuki, T. M. Pollock, Acta Mater. 56 (2008) 1288. M. Palumbo, S. G. Fries, T. Hickel, A. Dal Corso, M.H.G. Jacobs, U.R. Kattner, B. Sundman The challenge of covering thermodynamic properties not only at high temperature but also at low temperature: A progress report As it was pointed out recently [1], a revision of the lattice stabilities presently used in the CALPHAD community [2] is necessary to achieve a description of thermodynamic properties not only in the temperature regime above room temperature, but also in the temperature regime from room temperature down to zero Kelvin. That requires consideration of physically realistic models in Calphad methodology [3]. Our efforts in this direction are underway as part of the Sapiens project, presented in last CALPHAD XXXIX meeting held in Korea. In the present paper we report progress on this subject. To test our formalism we have selected the pure elements Cr and Ni as they are part of the selected elements of the Sapiens project. We started our analysis by collecting experimental data from the literature and results calculated by first principles methods. The reliability of these data was critically evaluated using several test criteria, such as experimental conditions, uncertainties in these data, and measuring methods. Additional, first principles calculations were carried

D.V. Malakhov Non-statistical thermodynamic optimization: an irrelevant topic or a useful approach? Since the CALPHAD technique is related to the least-squares method, the statistical nature of input data is implicitly assumed; it is supposed that an experimentally measured property can always be characterized by its accuracy. Even if no attention is paid to the statistical aspects of an experimental study in a publication, it is believed that the accuracy is calculable. Since such a belief is not helpful to assign statistical weights to experimental points, one asks questions like ‘‘What was the purity of substances used? Was a thermocouple calibrated? Was the equilibrium achieved?’’ instead of guessing, for instance, how many times a measurement was repeated. By finding reasonable answers, one guesstimates the accuracies of corresponding measurements and provides BINGSS/TERGSS, PARROT, etc. with the weights, but they are no longer traditional statistical weights! In contemporary publications, raw experimental data are replaced with figures visualizing them; error bars are not always shown, and even if they are, it is seldom clearly explained what they represent. Consequently, the methodological foundation of the least-squares technique becomes shaky. It is not intended to commence a hopeless struggle against a reproachable way in which experimental knowledge is disseminated today. It is contemplated to augment the CALPHAD method by a new assessment technique based on the ideas of a non-statistical treatment of experimental information formulated by Kantorovich [1] and successfully employed for solving economic problems. Methodologically, this technique better corresponds to a statistical fuzziness of experimental data used in the CALPHAD

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

The key idea is that a measurement can be characterized by its ultimate non-statistical error. By introducing the ‘‘interval calculus’’, one ceases dealing with such statistical entities as number of degrees of freedom, Student’s distribution, ‘‘three sigma’’ rule, and so on, but instead creates a non-statistical corridor of errors associated with experimental data and demands that any function describing them must reside within it. The left figure exemplifies a case when a property was measured in three independent examinations, A, B and C. It is worth accentuating that all error bars are the non-statistical intervals of ‘‘absolute trust’’ assigned to each measurement. At first glance, the results of all three experiments are in a good agreement, but an inspection of the domains of permissible slopes and intercepts shown in the right figure reveals that the outcome of the experiment B does not conform to the observations made in the experiments A and C whose domains Ya and Yc overlap. The example presented above is deliberately made very simple to clarify some distinctive features of the non-statistical optimization. In the presentation, more sophisticated CALPHAD related cases are considered.

123

cooling were performed for all alloys and the same microstructure characterization as for the as-cast alloys was carried out. Natures and fractions of the different phases, as well as the solidus and liquidus temperatures, were compared to results issued from Thermo-Calc calculations. All the studied nickel alloys are hypereutectic with presence of coarse pro-eutectic carbides in addition to the {Ni matrixþchromium carbides} eutectic compound. Concerning the cobalt alloys, for carbon contents lower than 3 wt% their microstructures are hypo-eutectic with presence of dendrites of Co matrix delimiting the {Co matrixþ chromium carbides}eutectic spaces, while for higher carbon contents coarse pro-eutectic carbides appear while dendrites are not present anymore. Lamellar graphite also appears for the carbon contents higher than 3.5 wt%C in the nickel alloys while it appears only for 5 wt%C in the cobalt alloys. In the two systems Cr7C3 carbides are present in all alloys but Cr3C2 carbides also exist in the carbon-richest alloys. Keywords: Nickel alloys, Cobalt alloys, Very high carbon, Chromium carbides, Graphite, Phase fractions, Melting range, Thermodynamic calculations.

Reference References [1] L.V. Kantorovich, On new approaches to computational methods of data treatment, Sib. Math. J. 3 (5) (1962) 701–709. P. Berthod, L. Aranda, O. Hestin, E. Souaillat,M. Ba, A. Dia Experimental and thermodynamic study of ternary Ni–30Cr– xC and Co–30Cr–yC carbon-rich alloys with x and y varying from 2.5 to 5.0 wt% Among the alloys used as coatings deposited by thermal spraying for achieving wear-resistance properties, there are Cobased alloys containing high quantities of tungsten carbides. Other alloys, containing chromium carbides and simply elaborated by foundry, may be also of interest. This study consists in an exploration of the microstructures and of the melting temperature ranges of alloys belonging to the ternary families Ni or Co (bal.)– 30Cr–xC (in wt%), with x increasing from 2.5% to 5%, with comparison with thermodynamic calculations results. Six alloys per family were synthesized by casting. In the as-cast condition they were metallographically characterized using scanning electron microscopy, with measurement of the surface fractions of the present phases (carbides, and eventually graphite). The meltingstart and meltingend temperatures were also measured by DTA/ DSC. Exposures to 1000, 1100 and 1200 1C for 50 h, followed by air

[1] P. Berthod, E. Souaillat, O. Hestin, As-cast microstructures of {M–30Cr–0 to 5%C} ternary alloys. Part 1: Nickel-base alloys, Mater. Sci.: Indian J. 6(4) (2010) 260–266. [2] P. Berthod, O. Hestin, E. Souaillat, As-cast microstructures of {M–30Cr–0 to 5%C} ternary alloys. Part 2: Cobalt-base alloys, Mater. Sci.: Indian J. 7 (1) (2010). D. Jendrzejczyk-Handzlik, W. Gierlotka, K. Fitzner Thermodynamic properties and phase equilibria in gold-antimony-tin system determined from E.M.F., Calorimetric, and DTA/DSC methods Using calorimetric, E.M.F. and DTA/DSC methods thermodynamic properties and phase equilibria of the Au–Sb–Sn system were investigated. Consequently, drop calorimetric measurements were carried out along the following cross-sections: X(Au)/X(Sb)¼2:1 and 1:2 at single temperature at 1073 K—X(Au)/X(Sb)¼1:1 at two temperatures 1073 and 923 K and integral enthalpies of mixing of liquid ternary alloys were determined at those temperatures. Next, E.M.F. measurements were done using solid oxide galvanic cells with zirconia electrolyte. The E.M.F.’s of the cells Au(1 x  y) SbxSny,

124

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

SnO2//ZrO2 þY2O3//Ni, NiO (I) were measured in the temperature range from 973 to 1273 K for two cross-sections X(Au)/X(Sb)¼2:1 and 1:2, and activities of tin were determined for the alloys in this temperature range. Finally, DTA/DSC measurements were done along two cross-sections – X(Au)/X(Sb)¼1:1 where X(Sn) was changed from 0.8 to 0.1 – X(Au)/X(Sn)¼1:1 where X(Sb) was changed from 0.8 to 0.1 and liquidus temperature and invariant reactions were identified. Additionally, an isothermal section of the ternary system was investigated at 473 K. All the data taken together with the information about binary systems contributed to the database used to optimize the whole ternary system. O. Fabrichnaya, H.J. Seifert Experimental study and thermodynamic modelling of phase relations in the ZrO2–Sm2O3–Y2O3 and Sm2O3–Y2O3–Al2O3 systems Phase relations in the ZrO2–Sm2O3–Y2O3–Al2O3 system are important to estimate stability of new materials for thermal barrier coating applications. Thermodynamic databases of the ZrO2–Sm2O3–Y2O3 and Sm2O3–Y2O3–Al2O3 systems were derived based on binary extrapolations into ternary systems and used for selection of compositions for experimental studies. Samples obtained by co-precipitation technique were heat treated at 1250, 1400 and 1600 1C. Phase assemblages formed during heat treatment were identified by XRD. Microstructures were investigated by scanning electron microscopy (SEM/EDX). Selected samples were investigated by differential thermal analysis (DTA). The obtained experimental results were used for more advanced modelling of pyrochlore phase taking into account solubility of Y2O3 in pyrochlore structure, Sm2Zr2O7. Ternary parameters were introduced into fluorite and C phases in the ZrO2–Sm2O3–Y2O3 system to reproduce experimental data. Isothermal sections of the ZrO2–Sm2O3–Y2O3 system at 1250, 1400 and 1600 1C were calculated using the assessed parameters. Phases with perovskite (YAP), monoclinic (YAM) and garnet (YAG) structures in the Sm2O3–Y2O3–Al2O3 system were modelled by compound energy formalism with mutual substitution of Sm and Y in the sublattices. Calculated isothermal sections of the Sm2O3–Y2O3–Al2O3 system based on ideal mixing in YAP, YAM and YAG phases were in reasonable agreement with experimental results. Liquidus surface was predicted. S.G. Fries The role of CALPHAD in the science of materials CALPHAD was born as a method to calculate phase diagrams from modeled Gibbs energies. These models developed so soaked in experimental facts that not only phase diagrams but other quantities, like Cp curves, calculated using CALPHAD, are sometimes mistaken as they would be the genuine experiments. These calculations became so associated to CALPHAD that sometimes the method is mistaken as a computer code. An outsider looking briefly to the word can erroneously infer that CALPHAD is a code that calculates experimental facts. This misunderstanding can lead to erroneous and dangerous conclusions: no experiments are need anymore, no new models are need, all is already in CALPHAD. However, independently of the way the word CALPHAD is understood, there is an irrefutable evidence: CALPHAD is part of the needs in materials development. One would like to have a method that would predict materials behavior like the idealized CALPHAD. Density functional theory can help in that point. However, like experimental values, the observable quantities predicted by the theory are discrete and models are still necessary to connect them. The generalized CALPHAD concept is a dynamical network of consistent information that permeates scales and adapts

itself as a whole by the demands of the application in agreement to new theoretical achievements. This generalized CALPHAD should be constructed (SAPIENS project is a part of it, see Ursula Kattner and Mauro Palumbo’s lectures) and taught in materials science courses to create a connected way of hinking, independently of discipline, scale or technique. In this lecture some examples will illustrate that even if CALPHAD is just a part of alloys design it is a decisive one. Acknowledgement ThyssenKrupp AG, Salzgitter Mannesmann Forschung GmbH, Robert Bosch GmbH, Bayer Materials Science AG, Bayer Technology Services GmbH, Benteler Stahl/Rohr AG, the state of North Rhine-Westphalia and the European Commission in the framework of the European Regional Development Fund (ERDF). Y.H. Jo, I. Jung, H.M. Lee Synthesis of size and composition controlled Sn–xCu nanoparticles: Effect on the phase diagram and application to highly conductive ink Various sizes of Sn–Cu nanoparticles were synthesized using a modified polyol process. Monodispersive Sn–0.7Cu (bulk eutectic) nanoparticles with diameters of 21, 18 and 14 nm were synthesized. The peak melting temperatures of the 21, 18 and 14 nm Sn–0.7Cu nanoparticles were 212.9, 207.9 and 205.2 1C, respectively, as compared with that of the bulk alloy, 230.6 1C. To decrease the melting temperature further, we synthesized 14 nm Sn–xCu (x¼0, 0.7, 2.1, 4.1, 5.3 and 6.6) nanoparticles and the resultant melting temperature decreased to 200.3 1C for the Sn–5.3Cu, which is 4.9 1C lower than that of Sn–0.7Cu nanoparticles. The increase in the solubility limit of Cu in Sn was observed in nano-sized particles, and the composition of Sn–5.3Cu is the eutectic point in the 14 nm nanoscale phase diagram. Finally, to fabricate a low-cost, highly conductive ink for inkjet printing, we synthesized a gram scale of uniformly sized Sn nanoparticles. A 20 wt% of Sn nanoparticles was dispersed in the 50% ethylene glycol: 50% isopropyl alcohol mixed solvents. To improve the electrical property, we applied the surface treatments of hydrogen reduction and plasma ashing. The two treatments had the effect of diminishing the sheet resistance. addition, conductive patterns (1 cm  1 cm) were successfully drawn on the Si wafer using an inkjet printing instrument with conductive Sn ink although the maximum resistivity for an hour of sintering at 250 1C was six times higher than the bulk Sn resistivity.

I.-H. Jung, M.-A. Van Ende, H. Choi, T.-S. Kim Thermodynamic database development of the Nd–Fe–B–Mg system for the Nd–Fe–B magnet scrap recycling process More than 90% of RE metals are currently produced from China. However, due to the increasing usage of RE in high-valuable electronic devices and important components for electronic, communication, automotive, etc., shortage of supply is expected in the near future. Moreover, the Chinese government limits the export of RE metals and may use it as a political weapon. In particular, Nd is one of the most critical RE elements that could face a supply constraint soon. The usage of Nd shall be increased drastically due to the application to the electrical motors in current and future hybrid and electric automotives. In addition, Nd is a quite promising alloying element for Mg alloys for high temperature and sheet applications. Thus, recycling of Nd from Fe–Nd–B magnet scraps is inevitable to keep the balance between supply and demand. Since the recycling process of the

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

125

magnet materials has not been well established yet, undamental research on the thermodynamic behaviour of Fe–Nd–B magnet materials and the chemical reactions between the magnet and solvent medium (Mg) are critical to understand and improve the Nd recycling process. In this presentation, the recent preliminary thermodynamic modeling (CALPHAD) results of the quaternary Fe– Nd–B–Mg system will be presented. Then, the applications of the database for the pyrometallurgical recycling of the Nd–Fe–B magnet using liquid Mg solvent will be presented.

used to describe the Gibbs energies of the ordered/disordered transitions for Fcc_A1/L12 and Bcc_A2/Bcc_B2. A set of selfconsistent thermodynamic parameters of the Mn–Ni–Si system has been obtained. Comprehensive comparisons between calculated and measured phase diagrams show that the measured isothermal sections at 800 and 1000 1C are satisfactorily accounted for by the thermodynamic description Fig. 1.

S. Zhou, R.E. Napolitano

The financial support from the National Natural Science Foundation of China (Grant nos. 50971135, 50721003 and 50831007) is greatfuly acknowledged.

Nonequilibrium phase transformation calculation with energetics The energetics is employed to calculate the nonequilibrium phase transformation. For the extreme condition, i.e., no chemical partitioning, T-zero lines are used to describe diffusionless phase transformation. Between two-terminal cases, i.e., phase equilibrium and zero chemical partitioning, a continuum exists. To describe this continuum, we introduce a parameter, eta¼ ((realMu) (T-zeroMu))/ ((equMu)(T-zeroMu)), named as the degree of chemical partitioning. For a given temperature, the variation of eta from 0 to 1 indicates the increasing minimum diffusional burden that accompanies the phase transition for the composition range between T-zero and Tequ. Thus, level curves with respect to eta in the composition-temperature domain represent possible nonequilibrium phase transformation. The nonequilibrium phase transformations for Al–Sm and Pd–Si systems will be presented.

B. Hu, H. Xu, S. Liu, Y.Du, C. He, C. Sha, W. Zhang, Y. Peng, D. Zhao, Y. Li Experimental investigation and thermodynamic calculation of the Mn–Ni–Si system The Mn–Ni–Si system has been investigated via thermodynamic modeling coupled with key experiments. The isothermal section at 1000 1C was determined by means of XRD and SEM/ EDX. The present experimental data confirmed 7 out of 8 ternary compounds (t1, t3, t4, t5, t6, t7, t11 and t12) reported in the literature. Based on the critically reviewed experimental data available in the literature and the present experimental data, a thermodynamic modeling is performed. One single function was

Acknowledgements

A.U. Khan, P. Broz, H. Niu, X. Chen, P. Rogl The phase diagrams Ta–V–{Si,Ge} Due to low induced activity under neutron irradiation Ta–V are important constituents in CrWVTa martensitic steels developed for fusion reactors. Therefore, reliable experimental phase equilibria data as well as thermodynamic description of phase stability in the Ta–V–Si subsystem are needed for successful materials design. Although phase relations have been reported for an isothermal section of the Ta–V–Si system at 1400 1C [1] contradictions arise with Laves phases Ta(V,M)2 claimed to exist [2] for M¼Si and for the homologue Ge for which no phase diagram is hitherto available. Phase relations in the systems Ta– V–{Si,Ge} have been reinvestigated in this work (based on optical microscopy, X-ray microanalyses and X-ray powder analyses of arc melted and annealed alloy specimens) and are characterized by continuous solid solutions (Ta,V)Si2, (Ta,V)5Si3, extended solids solubilities (Ta,V)6Si5 and Laves phase compounds with two structure types MgZn2 and MgCu2. Based on thermodynamic assessments of binary subsystems [3–5] and ab-initio data for the ground state energies of the end members of the solid solutions a CALPHAD modeling was performed to provide a consistent set of thermodynamic data in the Ta–V–Si system. ¨ AD Financial support by MEB 060915 and CZ 11/2009 of WTZ-O are gratefully acknowledged. Phase equilibria were calculated by means of Thermocalc code. X.-Q. Chen acknowledges the support from the ‘‘Hundred Talents Project’’ of CAS.

2.3. Applications—Energy P. Rogl, H. Noel

Fig. 1. Calculated isothermal section at 800 1C along with the experimental data.

The phase diagram basis U–M–Si for LEU fuels; M ¼Al, Sc to Pt The talk intends to present an overview in the phase relations, compound formation and compound properties in the ternary systems U—metal M—silicon, where M stands for aluminium and for a transition metal of groups III–VIII. The research reported herein is related to low enriched uranium (LEU) proliferation resistant reactor fuel systems. Phase relations in the systems U– M–Si were mainly established for isothermal sections in the range from 800 to 1400 1C based on optical microscopy, X-ray microanalyses and X-ray powder analyses of arc melted and annealed alloy specimens. X-ray single crystal data analyses and neutron powder refinement assisted in elucidating the crystal as well as magnetic structures of ternary compounds. Most of the ternary compounds for M¼Ti to Re are weakly magnetic with rather low saturation moments—The magnetic properties will be discussed as well as systematic trends in the formation and crystal chemistry of the compounds in the ternary systems U–M–Si. Stability

126

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

of competing phases has been evaluated for selected cases by means of ab-initio calculations. J.-Y. Park, H.-G. Kim, B.-K. Choi, S.-Y. Park, Y.-I. Jung, D.-Jun Park Experimental study on the microstructure factors controlling the corrosion behavior of Zr alloys The development of the advanced nuclear fuel claddings with improved corrosion resistance has been widely carried out in many countries operating nuclear power plants. The fact that Nb was selected as a major alloying element in Zr-based alloy is the common characteristics of the newly developed fuel claddings. Although a number of researches were carried out to evaluate the corrosion behavior of Nb-containing Zr alloys, the optimum Nb content for corrosion performance was changed according to chemical compositions and manufacturing processes. It is essentially required to investigate the microstructure factors controlling the corrosion behavior of Zr alloys for developing the advanced nuclear fuel cladding with superior corrosion resistance. In this study, the effects of alloying elements, microstructure and the processing Parameters on the corrosion behavior have been intensively investigated using the various types of Nbcontaining Zr alloys. The microstructure of the alloys was observed by a transmission electron microscopy. The types of precipitates in the alloys were identified by analyzing the selected area diffraction pattern and the chemical composition from the energy dispersive X-ray spectroscopy. Corrosion tests of the alloys were performed in PWR-simulating condition by using static autoclaves. The addition of Sn showed a deleterious effect of the corrosion resistance of Nb-containing Zr alloys as in Zircaloy-type alloys. The corrosion behavior of Zr–xNb alloys was significantly different depending on the Nb content and the annealing temperature during the manufacturing process, which in turn determines the precipitate characteristics. The corrosion resistance of the Nb-containing Zr alloys was apparently degraded with having beta-Zr phase and increasing size of the precipitated intermetallic compounds, which was resulted from increasing of intermediate and final annealing temperature.

investigated using two different numerical tools. The first one is a numerical code called EKINOX formerly developed to treat HT oxidation of Fe and Ni alloys [2]. This code has been recently adapted in order to solve out HT oxidation of Zr alloys [3]. It has also been coupled with the Zircobase thermodynamic database [4] via the TQ interface in order to obtain accurate equilibrium concentrations values in each phase. The second tool is the DICTRA software coupled with the Zircobase and the Zircomob database. Zircomob is the first mobility database dedicated to Zr alloys [5]. Oxygen diffusion profiles have been calculated inside the three phases involved in this dissolution process using both tools and compared. References [1] J.-C. Brachet, et al., J. ASTM International 5, 5, Paper ID JAI101116, 2008, available online at www.astm.org. [2] N. Bertrand, et al., Mater. Sci. Forum 595–598 (2008) 463. [3] C. Corvalan-Moya, et al., J. Nucl. Mater. 400 (2010) 196c. O. Beneˇs, R.J. M. Konings, D. Staicu

 C. Toffolon-Masclet, C. Desgranges, B. Mazeres, D. Monceau

Physico-chemical properties investigation of plutonium Actinides are 5f elements that are unstable, undergoing radioactive decay. As a result of the complexity of their handling and also their availability, it is difficult to experimentally investigate their properties. Special cases are uranium and plutonium which have been studied extensively as they are fissile materials used in nuclear technology, e.g., as a fuel in nuclear reactors. One of the key properties that determine thermodynamic stability of a material is the heat capacity. In case of plutonium element, several experiments have been performed to investigate this quantity; however the data are in significant disagreement as shown in Fig. 1. Moreover for the high temperature e-phase and for the liquid phase only very few points have been measured, with very large scatter. Therefore new measurements are needed to obtain more reliable description. In this study we present new experimental data of the enthalpy increments of the a-, b-, g-, d-, d0 -, e- and liquid plutonium which were measured in the Institute for Transuranium Elements (JRC-ITU) using a drop calorimeter. For the liquid phase determination special measures have been introduced. From the obtained data the heat capacity has been derived and correlated with a direct measurement using differential scanning calorimeter, performed in JRC-ITU as well. Both apparatuses are installed in an alpha tight glove box which made it is feasible to handle this radioactive element. Furthermore the enthalpies of transitions between various phases were identified and correlated with the literature. The set of new data has been implemented in the thermodynamic database of the actinide elements and used in the thermodynamic modelling of the p–T

Simulation of oxide dissolution in Zr alloys: Comparison between numerical EKINOX code and DICTRA calculations High temperature (HT) oxidation (T4 1000 1C) of Zr alloys involves phase transformations controlled by oxygen diffusion in the substrate. Indeed, once the O solubility limit in the b-Zr phase is reached, the phase transformation b-Zr-a-Zr begins, leading to the growth of both an oxygen stabilized a-Zr (O) layer and an outer oxide (ZrO2) layer once the O solubility in a-Zr has been reached. In the case of HT oxidation of a pre-oxidized material, that is a Zr alloy with an external transient ZrO2 layer formed at lower temperatures (i.e.,  350–400 1C), a dissolution of the transient oxide is observed in the first steps of the subsequent HT oxidation [1]. The kinetics of this dissolution stage has been

Fig. 1. A compilation of the high temperature heat capacity data of plutonium element taken from [1].

References [1] J.Y. Park, B.K. Choi, Y.H. Jeong, Y.H. Jung, J. Nucl. Mater. 340 (2005) 237–246. [2] J.Y. Park, B.K. Choi, S.J. Yoo, Y.H. Jeong, J. Nucl. Mater. 359 (2006) 59–68. [3] J.Y. Park, Y.H. Jeong, Y.H. Jung, Met. Mater. Int. 7 (2001) 447–455.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

phase diagram of plutonium which will be shown in this study. In order to describe this system the unknown thermal expansion coefficients and the bulk modulus of plutonium phases had to be optimized, giving the novel data.

127

compound. The calculated phase diagrams and thermodynamic properties are in good agreement with the experimental data. The thermodynamic database includes the commonly used elements in the nuclear materials, for example: Th, Pu, U, Al, C, Co, Cr, Cu, Fe, Mg, Mn, Mo, Nb, Ni, Pb, Sc, Si, Sn, Y, Zn and Zr.

References [1] R.J.M. Konings, O. Beneˇs, The thermodynamic properties of f-elements. Part I. The lanthanide and actinide metals, J. Phys. Chem. Rev. Data 39 (2010) 3. M. Beilmann, O. Beneˇs, R. Konings Thermodynamic assessment of a molten salt reactor fuel The molten salt reactor (MSR) is a nuclear reactor type with great potential in terms of safety and security. For this reason it is considered in the Generation IV International Forum as one of the six next-generation nuclear energy systems. In this reactor type the fissile material is dissolved in a matrix of molten fluoride salts which acts as solvent and coolant in parallel. The whole fluid salt mixture circulates between the reactor core, where it is heated up, and the heat exchangers. Today the research concentrates on concepts with a non-moderated (fast) neutron spectrum which offers new prospects. Depending on the reactor design the MSR can be operated as actinide burner as well as breeder reactor in which 233U is bred from 232Th. In this context CALPHAD is a very strong tool to support further reactor developments and the decision about the final fuel composition. With this method several thermodynamic properties of the salt are accessible. As a possible multi component fluoride salt mixture for the breeder design we assessed the LiF–NaF–CaF2–ThF4–UF4 system using the CALPHAD approach. The excess Gibbs energies of the liquid solutions have been modelled with the modified quasi chemical model and for the solid solutions we applied a polynomial formalism. To model the higher order systems the binary systems were extrapolated according the asymmetric Toop formalism. These calculations are based on the available literature data as well as on own differential scanning calorimetry (DSC) measurements on several subsystems using special tools to work with fluoride salts and with radioactive samples. From the obtained description we were able to derive some important thermodynamic properties such as melting temperature, vapour pressure or boiling points. X.J. Liu, C.P. Wang The thermodynamic database of nuclear material system A good understanding of nuclear materials is important to develop a safe nuclear reactor with high efficiency, however, traditional methods of materials research solely based on experimental investigations are far from being suitable for probing nuclear materials properties because of the stringent experimental conditions. The investigation of phase diagrams of nuclear material systems is essential for the development of new nuclear materials. Our goal is to develop a thermodynamic database and phase diagrams for nuclear materials. In this work, the thermodynamic assessments of phase diagrams in the binary and ternary systems with rare earth elements have been carried out by using CALPHAD method on the basis of the experimental data including thermodynamic properties and phase equilibria. Gibbs free energies of the solution phases were described by subregular solution model with the Redlich–Kister equation, and those of the intermetallic compounds and gas phase were, respectively, described by sublattice model and ideal gas model. A consistent set of thermodynamic parameters have been derived for describing the Gibbs free energies of each solution phase and intermetallic

C. Gueneau, P. Gotcu, O. Beneˇs, B. Sundman, N. Dupin, R.J.M. Konings Thermodynamic modelling of oxide fuels containing minor actinides (U, Pu, Am, Np)O2 7x Uranium dioxide UO2 and mixed oxide (U, Pu)O2 fuels have already been used in thermal and fast reactors. For the next generation of nuclear reactors, the recycling of minor actinides (Np, Am, Cm) in the fuel is considered. The goal is to destroy the minor actinides by neutron fission (transmutation). Fuels with low (few percent) and high content (up to 40%) of minor actinides are studied for the fast reactors. In fast reactors, a steep temperature gradient exists in the fuel pins. In some regions, the fuel can be exposed to temperatures above 2300 K during operation. Thus the use of such fuels requires a good knowledge of the high temperature thermodynamic properties of the U–Pu–Am–Np–O system. For that, a thermodynamic modelling of this complex system has been undertaken using the Compound Energy Formalism to describe the fluorite oxide phase (U, Pu, Am, Np)O2. A three sublattice model with ionic species is used: (U3 þ , U4 þ , U5 þ , Am3 þ , Am4 þ , Np3 þ , Np4 þ )1(O2  , Va)2(O2  , Va)1 where ‘‘Va’’ designate vacancies. The model will be described. Calculations of key properties such as oxygen potentials and vapour pressure data will be presented and compared to available experimental data. L. Zhang, P.J. Masset Investigation in the CaO–SiO2–M2O (M ¼Na, K) system The aim of VIRTUHCON is the theoretical modelling of high temperature conversion processes and metallurgy and gasification. The thermodynamic properties of molten oxides systems (slag) are essential as important feature for the modeling of such processes. Compared to experimental methods thermodynamic equilibrium modeling is able to generate results within less time at lower costs. In order to perform calculations concerning systems which have a complicated structure and strong interactions between the constituents an adequate thermodynamic model and assessed thermodynamic data are needed. The accuracy of the calculation depends on the reliability of the database and adaptability of the model. The goal of the present work is to describe the thermodynamic properties of the ternary CaO–SiO2–M2O where M is an alkali oxides. Some of the binary systems (SiO2–M2O) have been already optimized and the parameters data in the literatures have been used to extrapolate in the ternary system. The missing binary systems have been assessed and the results were used for the optimization of the ternary system. The ionic two-sublattice model was applied to represent the phase relations in the CaO–SiO2–Na2O and CaO–SiO2–K2O. The phase equilibria calculated the new optimized slag solution data show good agreement with the experimental points. M.A. Duchesne, R.W. Hughes, D.Y. Lu, D. McCalden, A. Macchi, E.J. Anthony Application of FactSage for the study of slagging in entrainedflow gasifiers FactSage is a powerful tool for the prediction of thermodynamic properties and equilibrium states. To demonstrate its

128

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

relevance in the field of fossil fuel gasification, particularly slagging behaviour, three applications are discussed. The liquid and solid phases present in fly ash and on the gasifier wall dictate ash sticking propensity. Hence the first application is the calculation of ash equilibrium states and liquidus temperatures for coal and petroleum coke ash. The FactSage calculated results are compared to SEMEDX analysis of quenched fly ash samples and wall slag samples collected from CanmetENERGY’s high pressure 1 ton/d gasifier. Second, since slag-refractory interactions such as dissolution and spalling are major concerns for the design and operation of gasifiers, two types of refractory test rods (SiC and 70 wt%Cr2O3–30 wt%Al2O3) were exposed to ash constituents in CanmetENERGY’s gasifier. Phases predicted via FactSage are compared to compounds detected by SEM-EDX at the slagrefractory interface. Thirdly, slag viscosity affects slag-refractory interaction and slag accumulation along the reactor walls. The latter will affect heat transfer within the reactor and could possibly lead to slag tap plugging. Slag viscosity models which make use of FactSage calculations are quantitatively compared to viscosity easurement results, and qualitatively compared to slag thickness on CanmetENERGY’s gasifer refractory. Many challenges remain with the use of thermodynamic modeling to predict gasifier slagging phenomena, such as determining vanadium equilibrium states, implementation of thermodynamic models within comprehensive reactor models, working with complex multi-component systems and adjusting predictions for nonequilibrium conditions. These are discussed in the context of each application.

function of burn-up and initial composition (carbon content) for representative operating temperature values. For that purpose, we have used the isotopic composition data given in [1] and compared our results to those obtained with the Solgasmix-PV computer program code and also to experimental observations on irradiated fuels from [2]. Special attention is paid to the evolution of the carbon over metal (C/M) content with burn-up. Indeed, as the C/M ratio decreases with burn-up, the initial M2C3 phase present in the fuel might disappear as soon as C/M decreases below one and due to the narrow nonstoichiometry of the MC phase [3,4], a liquid (U,Pu) phase might appear, which is not recommended for safety criteria. Finally, comparison of our results obtained on of un-irradiated (U,Pu)C fuel (isothermal ternary sections, carbon and plutonium activities) with those calculated by the TCC Software using the CALPHAD description of the (U–Pu–C) system developed in the frame of the FuelBase Project [5] shows significant differences. Those last results lead us to conclude that a complete CALPHAD description of the whole (U–Pu–C–FP) system would be helpful if we want to improve the operating conditions of this kind of fuel.

P. Masset, C. Guo

2.4. Applications—steel, superalloys, oxides

Evaluation of the thermodynamic properties of the gas phase involved in gasification processes Gasification processes are operated at elevated temperatures (1000–1500 1C) depending of the technology used in order to produce syngas (mainly hydrogen and carbon monoxide) employed in chemical engineering plants. At high temperatures, part of the mineral matter (SiO2, Na2O, K2O) contained in the ash/ slag arising from the original coal vaporizes. In the reactor volatiles species react with oxygen, water vapour and carbon (CO, CO2) based gaseous species. This work aims at determining the thermodynamic properties of species formed at elevated temperatures and their range of stability as a function of the process conditions (temperature, oxygen, CO/CO2 partial pressures, total pressure) encountered in such high temperature process. This evaluation provides a new description of the gas phase properties which may react with solid phases (ash, structural materials) or liquid phases (slag) and may be used in flowsheets of process description. J.-C. Dumas, J.-P. Piron, C. Martial Evaluation of the chemical composition of irradiated mixed carbide fuel The mixed carbide (U,Pu)C is a reference material for Generation IV (gas-cooled fast reactor or sodium fast reactor) nuclear fuels. The creation of fission products (FP) can modify the physical and chemical properties of the fuel behaviour during irradiation. As they control the carbon activity in the fuel, it is important to assess the influence of burn-up and temperature on the speciation of FP (solid, liquid and gaseous phases). The first part of this study is dedicated to the thermodynamic modelling of the irradiated fuel introduced into the FactSage code. The second part deals with the evaluation of the chemical state of the fission products as a

References [1] R. Agarwal, V. Venugopal, J. Nucl. Mater. 359 (2006) 122 . [2] Hj. Matzke, Science of Advanced LMFBR Fuels, Elsevier Science Publishers B.V., 1986. [3] H. Kleykamp, J. Nucl. Mater. 221 (1985) 131. [4] H. Holleck, Thermodyn. Nucl. Mater. IAEA Vienna, 1975.

T. Matsumiya Applications of CALPHAD in steelmaking and future scope CALPHAD has been used to control the chemical compositions of nonmetallic inclusions such as for sulfide shape control in antihydrogen-induced-cracking steel for line pipe and softening oxide inclusions in austenitic stainless steel for wire, to design the compositions of continuous casting flux which prevents from casting sticking to the mold for titanium bearing steel, to optimize hot metal demanganization process, and so on in steelmaking field. In order to utilize CALPHAD, thermodynamic data are required. When they are not available, ab itnitio based calculations of phase equilibrium and thermodynamic property values are expected to give help. From those viewpoints, we conducted phase diagram calculation of Au–Cu system with the combination of Monte Carlo simulation and interaction energy values obtained by application of cluster expansion method to total energy calculated ab initio and estimated activity coefficients and interaction parameters of solutes in dilute silicon solution by the use of density functional theory. In addition, excess entropy of mixing was included in the estimation of activity coefficient of oxygen in silicon by applying grand canonical Monte Carlo simulation. Since high performance computers will be available in near future, such as 10 PFLOPS computer in Japan in 2012, ab initio estimation of thermodynamic properties and atomistic simulation should be utilized more than before in the combination with CALPHAD method in various ways. A. Schneider Application of the CALPHAD method for ferritic boiler steels An overview of past and current applications of the CALPHAD method on various questions concerning ferritic boiler steels is

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

given. The focus is on the development of high-temperature steels, on welding and on life-time predictions. The different concepts on the improvement of properties such as creep strength, corrosion resistance and weldability are reported. In view of developing or improving steels for the application in conventional power plants various elements have been tested with respect to a controlled precipitation of chromium carbides, carbonitrides and the intermetallic Laves phase. The thermodynamic calculations were mostly used for identifying alloy compositions being necessary for the evolution of the desired microstructures and precipitations. The kinetic simulations are often focussed on diffusioncontrolled transformations during heat-treatment, welding and application in the power plant. A keyissue is the life-time prediction for 100,000 h and more including all aspects such a nucleation, growth and coarsening of precipitates. C. Cicutti, C. Capurro Development of a model to predict inclusions composition in steels deoxidized with aluminum, silicon and manganese The composition of non-metallic inclusions may affect steel castability and also have a strong impact on the properties of final product. In the present work, a model to predict the composition of inclusions in equilibrium with liquid steel was developed. Knowing the total composition of steel and temperature, the amount of dissolved elements in the melt and the composition of inclusions are calculated. Following the procedure proposed by Ericksson [1,2], an in-house code was developed which minimizes the free energy of the system keeping a mass balance of the involved species. In this particular case, the model was applied to study the inclusions formed during deoxidation of steel with Al, Si and Mn. Activities of dissolved elements in liquid steel were calculated using Wagner’s formalism [3]. Activities of oxides in Al2O3–SiO2–MnO system were estimated using the sub-regular model proposed by Zhang [4,5]. Model results were compared with measurements previously reported in the literature [6,7] and also with experimental work carried out in this study. In general, a reasonable agreement between measured and calculated values was obtained. Hence, the model was used to analyze the effect of steel composition on the type of non-metallic inclusions expected. The amount of aluminum and ferroalloys that have to be added to ensure (or to avoid) alumina as deoxidation product was determined. References [1] [2] [3] [4] [5] [6]

G. Eriksson, Acta Chem. Scand.25 (1971) 2651. G. Eriksson, E. Rose´n, Chem. Scr. 4 (1973) 193. G. Sigworth, J. Elliot, Metal. Sci. (1974) 8 (9) 298. X. Zhang, et al., Calphad 21 (3) (1997) 301. X. Zhang, et al., Calphad 21 (3) (1997) 311. M. Garlick, et al., Ironmaking Steelmaking 29 (2) (2002) 140. [7] K. Ogawa, et al, Tetsu Hagane´ 71 (2) (1985) A29. ¨ B. Bottger Phase-field modelling of technical alloy systems Despite of the importance of thermodynamics for material science and processing, material structures and properties or optimised parameters for industrial processes typically cannot be obtained from pure thermodynamic considerations. The reason for this is the predominant role of diffusion and interfacial kinetics for microstructure formation and thus for the materials behaviour and properties during the process.

129

This conclusion does not curtail the needs for improved thermodynamic descriptions, but highlights the importance of a further coupling of these thermodynamic data to simulation tools like MICRESS [1] or DICTRA [2] which include diffusion and phase transformations. In this paper, details are shown how a phasefield model [3] has been coupled to CALPHAD-type thermodynamic databases. The aim is to develop and continuously improve the commercial software tool MICRESS which allows for modelling of microstructure formation in technical alloys during manifold processes like casting, heat treatment, deformation and service. Emphasis is placed on the pragmatic formulation of the phasefield model, which allows for use at different length scales, as well as on a general and effective coupling scheme to thermodynamic data by integration of the Gibbs energy minimising software Thermo-Calc [2], facilitating the use with manifold alloy systems and databases. Furthermore, specific topics related to technical alloy systems like the treatment of stoichiometric phases, composition sets, miscibility gaps etc. are discussed. References [1] /http://www.micress.deS. [2] /http://www.thermocalc.comS. [3] J. Eiken, B. B¨ottger, I. Steinbach, Phys. Rev. E 73 (2006) 066122.

T. Go´mez-Acebo, F. Castro Diffusion in Fe–Ni PM alloys: Microstructure and DICTRA simulations In the Fe–Ni–C alloy system the interdiffusion of Fe and Ni has been studied under several combinations of time and temperature. The interest in studying this system is contemplated within the frame of powder metallurgy, so that Fe, Ni and carbon powder mixtures have been used as raw materials for the preparation of the studied samples containing different amounts of admixed Ni. The studied compositions are Fe–0.8 wt%C–2 to 6 wt%Ni, with varying particle size of the Ni powders (o1.0 and E7 mm) while the mean particle size for the Fe powders are E70 mm. Sintering was carried out at temperatures between After sintering the microstructures of the experimentally obtained samples were characterised by optical and scanning electron microscopy. The experimental results were compared with diffusion simulations using the last available thermodynamic [1] and kinetic [2] descriptions of the Fe–Ni system, using DICTRA. The simulations have confirmed recent experimental observations [3], in which Fe diffusivity is higher than that of Ni. Interpretation of the different microstructures, obtained at different combinations of time and temperature, was allowed by considering effective diffusion distances including the influence of particle sizes. For high Ni particle sizes, there are sintering times and temperatures in which the chemical composition of the Ni-containing regions cannot be homogenised. This has proved very useful for the design of sintering cycles of Fe–Ni components made by powder metallurgy. References [1] G. Cacciamani, A. Dinsdale, M. Palumbo, A. Pasturel, The Fe–Ni system: Thermodynamic modeling assisted by atomistic calculations, Intermetallics 18 (2010) 1148–1162. [2] B. Jonsson, Mobilities in Fe–Ni alloys—assessment of the mobilities of Fe and Ni in fcc Fe–Ni alloys, Scand. J. Metall. 23 (1994) 201–208. [3] M.L. Yunker, J.A. Van Orman, Interdiffusion of solid iron and nickel at high pressure, Earth Planet. Sci. Lett. 254 (2007) 203–213.

130

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

P. Schaffnit, J. Mentz, J. Konrad Simulation of an industrial solidification process by coupling CALPHAD and phase-field The solidification of small portions of steel melts is part of almost all welding operations in steel applications. During the investigated welding operations the prepared edges of the steel material are heated up to high temperatures and pressed together. A characteristic depletion of alloying elements can be observed in the bond line formed by material in a semi-liquid state. This depletion influences the mechanical properties of the steel construction. The mechanisms leading to this element distribution have not been fully described yet. The highly dynamic nature of the process precludes experimental investigations, making computational simulation a unique mean of gaining insight in this industrial process. The microstructure evolution and subsequent element redistribution upon the heating of the material were simulated by combining thermodynamic calculations and phase field modelling with the commercial code MICRESS. Experimental data provided the element distribution and microstructure at room-temperature. The redistribution of elements in presence of a liquid phase was calculated on the basis of CALPHAD data, and the final element distribution was determined assuming that most of the liquid phase has been squeezed out of the bonding zone. The simulated bond line reveals a zone depleted of alloying elements and small regions of resolidified material enriched in alloying elements. The simulation results were compared to microprobe measurements of industrially solidified samples and show a good agreement regarding the element contents and distribution. As a result of these findings the knowledge based optimization of process parameters should lead to improved mechanical properties of the bonding. M. Selleby How applications drive thermodynamics towards fundamental considerations Many of the core binary systems for steels were evaluated during the 1980s. In 1991 SGTE compiled thermodynamic descriptions for pure elements, and those descriptions were used to harmonize the development of databases. Now, 20–30 years later, it is time to renew these descriptions of elements, extend them to 0 K, better describe the magnetic contribution and revisit all important binary and ternary systems. The VINN Excellence Center Hero-m at KTH is working closely together with industry to be able to improve industrial materials like high strength steels, advanced stainless steels, cemented carbides but also new materials like bulk amorphous alloys. In order to fully understand the underlying mechanisms of microstructural development we need to model, e.g., the martensitic transformation, the spinodal decomposition in duplex stainless steels and the formation of bainite. In our attempts to do so, it has proven necessary to revisit and reevaluate some of the current descriptions used in many of the commercial databases. It will be demonstrated how applications to real industrial processes have urged us to reevaluate both thermodynamic models and descriptions. This work was performed within the VINN Excellence Center Hero-m, financed by VINNOVA, the Swedish Government Agency of Innovation Systems, Swedish Industry and KTH (Royal Institute of Technology).

H. Strandlund, S. Norgren Applications of CALPHAD in cemented carbide development Cemented carbides used for metal cutting tools and rock drilling are most often WC-Co based alloys. Within the cemented carbide industry there is a great demand for developing new and better WC-Co based grades. To meet the demands and to guarantee a rapid development computational thermodynamics is an important and useful tool. In this work several examples will be presented showing how the use of thermodynamic calculations support experimental work in exploring new areas of interest. Specifically the use of computational thermodynamics when inhibiting grain growth in submicron tungsten carbide, High Pressure High Temperature (HPHT) processed diamondcemented carbide shear cutters and gradient formation in turning- and rock drill inserts will be discussed. B. Jansson, J. M. Ullbrand, F. Tasna´di, L. Hultman, M. Ode´n A multi-scale modeling approach for transition metal nitride alloy coatings The performance of cemented carbide based cutting tools can be considerably improved by thin film coatings. Several mechanisms for the improvement can be identified such as diffusion barrier, decrease friction and increased hardness for better wear resistance. Chemically unstable thin films of transition metal nitride alloys having a homogeneous composition can be synthesized at relatively low temperatures by, e.g., reactive arc sputtering. The coatings might undergo controlled precipitation hardening through spinodal decomposition by tuned heat treatment or high temperature and mechanical load application. Firstprinciples density-functional theory (DFT) for a relatively small system ( 100 atoms) can be used to solve the quantum mechanical problem combined with thermodynamic (CALPHAD) methods to model the microstructure evolution on the nm to mm scale utilizing continuum models and the phase field approach. This multi scale modeling approach has experienced certain success. However, presently the different methods used for different scales are not interfacing fully and are often treated separately. For example, DFT calculations emphasize the electronic subsystem, while atomic motions are often neglected. Classic CALPHAD models for large time and length scales often lack adequate experimental information, which is compensated by oversimplified interpolations. Examples of required input data for adequate modeling the microstructure evolution by the phase field approach that can be generated from DFT calculations are:

 Thermodynamic properties of homogeneous solid solution phases.

 Anisotropic elastic properties as a function of composition and pressure.

 Lattice parameter as a function of composition and pressure.  Surface tension for diffuse interfaces, e.g., gradient energy.  Activation energy of diffusion. Experimental information regarding the microstructure and heat evolution is also required to tune the phase field models and DFT calculations:

 Differential scanning calorimetry to study kinetics and evolved heat in situ

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

 X-ray diffraction (SAX/WAX) to study the kinetics of the 

microstructure evolution Characterization of the microstructure by TEM and APT (atom probe tomography) after heat treatments.

The ability to model the spinodal decomposition and its effect on mechanical properties can thus be utilized as a tool to design and to tailor coatings to desired properties. Q. Chen Challenges of precipitation kinetic modelling. I. Study on growth rate models The temporal evolution of particle size distribution is determined via a continuity equation by system thermodynamics and kinetics embodied in the nucleation rate and growth rate of different particles. Formulating appropriate models for the nucleation and growth and solving the governing equation in an efficient and robust way constitute the major challenges for the kinetic simulation of precipitation. In this presentation we focused on developing various approximations for the growth of precipitates under the diffusion controlled condition. We have implemented them in TC-PRISMA, a computational tool for kinetic simulation of precipitation in multicomponent and multiphase systems. We then applied these models to study concurrent nucleation, growth, and coarsening processes in different alloy systems, and discussed their advantages and limitations on the basis of the simulation results. I. Ohnuma, S. Shimenouchi, T. Omori, R. Kainuma, K. Ishida Phase equilibria at low temperatures (o600 1C) and thermodynamic evaluation in the Fe-base binary systems. In some Fe-base alloys, peculiar phase equilibria at low temperatures below 600 1C have been proposed by experimental studies and predicted by thermodynamic calculation. For instance in the Fe–Si binary system, the bcc phase exhibits two-fold ordering from the disordered A2 to the intermediately ordered B2 and further to the ordered D03 structures. In addition, miscibility gaps among these 3 phases, B2 þD03 and A2þD03, were confirmed, which is consistent with the conventional BWG calculation. On the other hand in the Fe–Al system, even though the two-fold ordering transition is identical to the Fe–Si system, the miscibility gaps, A2þ B2 and A2þD03, are inconsistent with the BWG prediction. Furthermore, even though phase equilibria at low temperatures in the other Fe-base binary systems, such as the Fe–Ni, Fe–Cr, etc. were also studied, precise equilibrium compositions were scattered and suspicious due to the difficulty in the low temperature equilibration. In this study, extremely deformed powders of the Fe-base alloys were equilibrated at low temperatures below 600 1C and phase equilibria were determined by a FEEPMA (JEOL:JXA-8500F) with high special resolution ( o1 mm). Gas-atomized powders of Fe–X (X¼Ni, Si, Al, Cr, etc.) were prepared and deformed extremely by converge milling method. Each powder sample was encapsulated in an evacuated quarts capsule and equilibrated at temperatures between 400 and 700 1C for various durations up to 3 months. After that, each powder sample was molded in a conductive resin, mechanically polished and finished by a vibratory polisher with 0.1 mm diamond paste. Microstructure of cross section of the powder samples was examined and equilibrium compositions were determined by the FE-EPMA with low accelerating voltage of 6 kV. Making use of the experimental findings, thermodynamic re-evaluations of the Fe–X binary systems were carried out.

131

References [1] Lukas, Fries and Sundman, Computational Thermodynamics, The Calphad Method, Cambridge University Press, 2007. B. Sundman, C. Gueneau, N. Dupin Thermodynamic modelling of defects in U–Pu–O–C The growing interest in using more complex fuels in nuclear reactors poses several problems both in manufacturing, during operation and for extreme conditions. A thorough knowledge of the thermodynamic properties of the system is necessary to understand how to control the different transformations during processing and under varying conditions during energy production. The modelling of the defects in different oxide phases will be described using the compound energy formalism (CEF) and compared with the Wagner– Schottky model and using the Kroger–Fink notation. The effect of the defects on various thermodynamic properties like solubility and heat capacity will be described. The modelling of additions of higher actinides will be discussed.

A.J. Ramirez, J. Unfried S. Design of a ductility-dip cracking resistant Ni–Cr–Fe alloy Engineering of precipitates size, morphology and distribution, grain boundaries morphology and solute partitioning within FCC matrix can be an effective method to improve the resistance of Ni– Cr–Fe alloys to the high temperature fracture phenomena known as ductility-dip cracking (DDC). DDC is a solid state intergranular fracture phenomenon associated with welding and high temperature forming that plagues several FCC materials, including the solid solution strengthened Ni-base alloys, Cu-base alloys, and stainless steels [1]. This phenomenon is characterized by a severe ductility reduction within homologous temperature range between 0.4 and 0.8. Several works have reported that the high temperature grain boundary sliding is the fundamental mechanism that governs DDC for alloy 690 [2]. Calphad methodology was applied in this work to perform both thermodynamic and kinetic modeling of the alloy 690 including few chemical composition modifications in order to optimize the DDC resistance without compromising other properties [3]. Primary carbonitrides precipitation and solidification have been evaluated for as-welded Ni-base alloy 690 modified with Nb, Mo, and Hf additions. Actual as-welded samples were produced and characterized using electron microscopy (SEM, TEM, and EBSD). The DDC resistance was evaluated using a dedicated SEM-in situ thermomechanical setup. The additions of Nb (o2.5 wt%) and Hf (o0.4 wt%) increased the mass fraction and the primary carbonitrides precipitation temperature. Consequently, the material presented a delayed grain boundary migration and the grain boundaries morphology changed from flat to undulated for the aswelded condition. The occurrence of wavy grain boundaries and the Mo atoms distribution (o4.0 wt%) in gamma matrix could be related to the grain boundary sliding blocking and to the dislocation movement restriction at high temperatures, which has been shown to improve DDC resistance in the redesigned 690 alloy. References [1] A.J. Ramirez, C.M. Garzo´n, Hot Cracking Phenomena in Welds II, 200, pp. 427–454. [2] A.J. Ramirez, J.C. Lippold, Hot Cracking Phenomena in Welds, 2005, pp. 19–41.

132

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

[3] J.S. Unfried, E.A. Torres, A.J. Ramirez, In situ observations of ductility-dip cracking mechanism in Ni–Cr–Fe alloys. in: J.C. Lippold, T.H. Boellinghaus, C.E. Cross (Eds.). Hot Cracking Phenomena in Welds III. Berlin: Springer, 2011, p. 295–315. H. Ipser, R. Mishra The ternary Ni–Sb–Sn system: Phase diagram and thermochemistry Ternary Sb–Sn alloys are being considered as high-temperature solders where the liquidus temperature can be adjusted by varying amounts of a third element, among them also Ni. On the other hand, Ni is frequently used as substrate material in electronics. Therefore, the ternary Ni–Sb–Sn phase diagram is of high importance for lead-free electronics. The phase diagram of this ternary system was investigated, with particular emphasis on the Sn-rich corner, by X-ray powder diffraction, electron microprobe analysis and thermal analyses. A number of isothermal sections, different isopleths, as well as the liquidus projection will be presented. Partial pressures of Sb were determined by an isopiestic vapor pressure method along three sections through the ternary system, i.e., xNi/xSn ¼3/1, 3/2, and 3/4. From these measurements it was possible to derive partial Gibbs energies (i.e., thermodynamic activities) of Sb, and from their temperature dependence the other partial thermodynamic properties could be obtained. All these experimental data provide an input for a CALPHAD-type optimization of the ternary Ni–Sb–Sn system, of which a very first version will be shown.

References [1] A.D. Pelton, P. Chartrand, G. Eriksson, Metall. Mater. Trans. 32A (2001) 1409–1416. [2] C. Robelin, P. Chartrand, J. Chem. Thermodyn. 43 (2011) 377–391. [3] E. Renaud, C. Robelin, A.E. Gheribi and P. Chartrand, submitted for publication to J. Chem. Thermodyn. (2011). P. Chartrand, C. Robelin, M.s Heyrman, N. Tripathi The cryolite database for the Hall–Heroult Alumina reduction process A molten Na3AlF6(cryolite)–AlF3–CaF2 solution is used in all alumina reduction smelters around the world for the production of tens of millions of tons of aluminum every year. Additives (LiF, KF, MgF2) and impurities (P, S, C, Fe, Be) play an important role in the process operation and efficiency, and they affect its environmental impacts. For the last ten years, an extensive thermodynamic database coupled with physical property databases (viscosity, density, electrical conductivity) have been developed by the authors with the support of aluminum producers. An overview of the thermodynamic database, and how it is intimately coupled with the physical property models will be shown, with examples of their applications to the Hall–Heroult process. The structure of cryolite melts and how it is taken into account in the modeling will also be discussed.

C. Robelin, P. Chartrand

V. Prostakova, J. Chen, E. Jak, S.A. Decterov

Overview of the molten chlorides, fluorides and chloro-fluorides databases of the FactSage thermochemical software The main salt databases currently available in the FactSage thermochemical software are: LiCl–NaCl–KCl–RbCl–CsCl–MgCl2–CaCl2– SrCl2–BaCl2, LiCl–NaCl–KCl–MgCl2–CaCl2–MnCl2–FeCl2 (–FeCl3)– CoCl2–NiCl2–AlCl3, NaCl–KCl–MgCl2–CaCl2–ZnCl2, and LiCl–LiF– NaCl–NaF–KCl–KF–MgCl2–MgF2–CaCl2–CaF2–SrCl2–SrF2. The liquid was modelled using the Modified quasichemical model in the quadruplet approximation [1] that evaluates coupled 1st- and 2ndnearest-neighbour short-range order. All binary subsystems as well as all higher-order (mostly ternary and ternary reciprocal) subsystems for which experimental data were available have been considered. A complete critical evaluation of all available phase diagram and thermodynamic data (enthalpy of mixing of the liquid, EMF and vapour pressure measurements) has been performed for the condensed phases (liquid, solid solutions, stoichiometric compounds) and relevant gaseous species of the various systems, and optimized model parameters have been found. Emphasis will be put on the modelling of AlCl3-based systems, and on the recent additions of ZnCl2 [2] to the NaCl–KCl–MgCl2–CaCl2 system and of Sr2 þ [3] to the Li þ , Na þ , K þ , Mg2 þ , Ca2 þ //Cl  , F  reciprocal system. In particular, the binary systems ACl–AlCl3 (where A¼Li, Na or K) show strong negative deviations from ideality at the equimolar composition (due to shortrange ordering in the liquid phase) and the binary mixtures exhibit a region of liquid–liquid immiscibility at high AlCl3 content. The existence in such melts of the AlCl4 and Al2Cl7 species was observed by Raman spectroscopy. In order to introduce two different compositions of maximum short-range-ordering near the AAlCl4 and AAl2Cl7 compositions, pure liquid aluminum chloride was modelled as a mixture of AlCl3 and Al2Cl6 (with two paired aluminum cations). The existence of a sharp minimum of solubility of MCl2 in ACl–AlCl3 melts near the equimolar composition (where M is a divalent cation such as Co2þ ) will also be discussed.

Development of a thermodynamic database for Ni-containing oxide systems for simulation of Ni extraction from laterite ores The present study reports the optimization of the CaO–MgO–NiO– SiO2 system. It is part of an on-going research project to develop a self-consistent thermodynamic database for nickel extraction from laterite ores. The database development is done by thermodynamic modeling which is closely related to experimental study of phase equilibria. Initially, parameters of the models are optimized to fit the experimental data collected from the literature. When there is not enough data to constrain the model parameters, or significant discrepancies in the available data are revealed, experimental measurements are initiated for temperatures and compositions which are most useful for thermodynamic modeling. The experimental procedure involves equilibration and ultra rapid quenching followed by electron probe X-ray microanalysis (EPMA) of quenched samples. Since the analysis takes place after equilibration, the changes in composition during equilibration do not affect the accuracy of the results. Tie-lines between equilibrated liquid and solid phases are measured directly, providing essential data for subsequent thermodynamic modeling. The modified quasichemical model [2] is used for the slag (molten oxide) phase. The models based on the compound energy formalism [3] have been developed for the olivine, spinel and pyroxene solid solutions. These physically meaningful models are based on the structure of the corresponding solution and, therefore, have good ability to predict the properties of multicomponent systems. However, accurate data in low-order subsystems are often needed to properly constrain parameters of these models. Such data are often missing in the literature because they are of no direct importance for practical applications. A limited number of experimental measurements specifically designed to constrain the models significantly reduces the amount of work required to obtain an accurate thermodynamic description of a multicomponent system.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

References [1] C.W. Bale, P. Chartrand, S. Decterov, G. Eriksson, K. Hack, R.B. Mahfoud, J. Melancon, A.D. Pelton, S. Petersen, Calphad 26 (2) (2002) 189–228. [2] A.D. Pelton, M. Blander, Metall. Trans. B 17B (1986) 805–815. [3] M. Hillert, B. Jansson, B. Sundman, Z. Metallkd. 79 (2) (1988) 81–87.

133

from ffi4 to 10, than those previously reported, which were nearly all calculated from data on coarsening assuming ideal-solution thermodynamics. In the TIDC theory the width of the interface, d, is allowed to increase with particle size, r. A simple equation relating s to the ratio of the gradient energy and d is used to show that s can remain constant even though d increases with r. The magnitudes of s in the different alloys are discussed in the context of qualitative considerations of bonding in the various alloys. The results on the important Ni–Al alloy system are compared with the predictions of various atomistic simulations of the energy of the Ni3Al/Ni–Al interface.

N. Warnken, R.C. Reed D. Cai, L. Zhang, Y. Du A design driven superalloy development approach The development of Ni-base superalloys over the last few decades has led to alloys with remarkable complexity; many alloy specifications comprise about 10 major alloying elements. It becomes obvious that systematic experimental studies, leading to new alloy compositions are quite complicated, as these would have to include thousands of alloys. As an alternative approach, design rules are applied to identify promising superalloy compositions. These simple rules characterise the most important properties, such as creep resistance, microstructural stability, oxidation resistance, manufacturability, density and cost. In order to determine these, CALPHAD type and other calculations are performed on a very large number – of the order of 1000,000 – of alloy compositions. From these results a few ideal compositions are selected according to design specifications which are suitable for either aeroengine blades or disk alloys or as blade alloys for industrial gas turbines. Future enhancements of the method are discussed. A.J. Ardell Interfacial free energies from data on coarsening plus assessments of Gibbs free energies of mixing in Ni-base cXc0 alloys Data on coarsening of c0 -type precipitates (Ni3X, with the L12 crystal structure) in a variety of binary and ternary Ni-base alloys are re-evaluated in light of recent (TIDC) and classical (LSW) theories of coarsening, with the objective of ascertaining the best values possible of interfacial free energies, s, of the precipitate/matrix interfaces. The re-evaluations include fitting of the particle size distributions, reanalyzing available data on the kinetics of particle growth and kinetics of solute depletion, and using hermodynamic assessments of the binary alloy phase diagrams to calculate curvatures of the Gibbs free energies of mixing. The product of the work is 2 sets of interfacial free energies, one set for the analysis using the recent TIDC theory and the other for the analysis using the classical LSW theory. The TIDC-based analysis yields lower values of s by about a factor of 2/3. All the interfacial energies are considerably larger, by factors ranging

Phase-field simulation of grain growth in anisotropic systems The phase-field method has proven to be a powerful tool for grain growth and related phenomena during the past decades. With the available grain boundary energy and mobility plus thermodynamic properties, the evolution of grain growth can be simulated in 2D and 3D using the phase-field method, and the corresponding macroscopic parameters, i.e., the grain size and the grain orientation distributions, are thus obtained. Considering the great challenge to perform quantitative simulation of grain growth in anisotropic systems, a phase-field code based on the continuum-field approach was developed in the present work. The misorientation between neighboring grains and the grain boundary inclination were also taken into account. With grain orientations randomly distributed, the phase-field simulation of grain growth evolution in an anisotropic system was performed using the present phase-field code, as shown in Fig. 1. Four types of triple junctions between small/large angle misorientation grains can be predicted. The present work also demonstrates that the boundaries between large angle misorientation grains combine with each other or disappear much faster than other kinds of boundaries. In order to quantitatively simulate the austenite grain growth in the Fe–C system, the Mont-Carlo and molecular dynamic simulations were made to compute the grain boundary energy. With the computed boundary energy, the evolution of austenite grain growth in the Fe–C system was simulated using the present phase-field code, and the simulated results were compared with the corresponding experimental data. In addition, the pinning effect was also investigated in the present work. Acknowledgements The financial support from the National Basic Research Program of China (Grant no. 2011CB610401) and the National Natural Science Foundation of China (Grant nos. 50721003 and 50831007) is greatly acknowledged.

Fig. 1. Phase-field simulation of grain growth evolution in an anisotropic system using the present phase-field code. Black lines are the grain boundaries with large angle misorientations, while grey ones the boundaries with small angle misorientations.

134

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

J.M. Ullbrand, B. Jansson, F. Tasna´di, L. Hultman, M. Ode´n The effect of elastic anisotropy on the spinodal decomposition in (Ti,Al)N—A phase field study The phase field approach has been used to solve the Cahn– Hilliard equations for a number of (AB)-model systems, but few results are published on real material systems and compared to experiments. In this study the phase field method is used to simulate spinodal decomposition in the (Ti,Al)N system. (Ti,Al)N thin films are of great industrial interest as hard wear protective coatings, e.g., on inserts for metal cutting. Coatings with compositions inside the spinodal can be synthesized at low temperatures by physical vapor deposition processes. The temperature at the edge can reach up to 1000 1C during cutting and the coating will then decompose. Thus, by proper aging of the coating a considerable hardening effect can be achieved during application. The variation of lattice parameter and elastic properties with composition strongly affect the microstructure evolution during the spinodal decomposition. In order to appropriately address (Ti,Al)N alloys, enthalpy of mixing, the elastic compliance tensor and the lattice parameter, determined by firstprinciples densityfunctional theory (DFT) calculations, are used as input parameters. The elastic problem is solved exactly in every time step and the atomic mobilities are adjusted by comparison with experiments. At compositions above 40% AlN the microstructure exhibits nm-sized domains with preferred growth directions in the elastically compliant directions. At AlN contents below 40% where the elastic anisotropy is less pronounced spherical domains are instead formed. The initiation of the spinodal decomposition is notably slower when the composition is displaced from the center of the miscibility gap towards lower amounts of AlN. In addition, the evolved modulation in elastic properties itself slows down the decomposition. The origins of the observed kinetic differences are discussed and the microstructure evolution is compared to experimental observations by, e.g., scanning transmission electron microscopy.

D. Shishin, C. Bale, E. Jak, S.A. Decterov Thermodynamic database for copper smelting and converting A thermodynamic database for copper smelting and converting developed previously [1,2] has been further improved in the present work. The most important phases are matte (liquid sulfide), slag (liquid oxide), liquid metal, gas and oxide and sulfide solid solutions. Liquid metal and matte are modeled as one extended solution, but the slag is described as a separate solution. The core chemical system is Cu–Ca–Fe–Al–Mg–O–Si–S. The database also includes important minor elements such as Zn and Pb. The strong short range ordering in liquid solutions is taken into account by using the modified quasichemical model [3]. The models relating the structure and thermodynamic properties of solid solutions have been developed within the framework of the compound energy formalism [4]. The goal of the present study was to add oxygen, lead and zinc into the combined metal–matte liquid solution. This solution must describe the experimental data on the solubility of oxygen in both liquid metal and matte. It must also be consistent with the slag to reproduce the important metal–matte–slag equilibria. The developed database in combination with the FactSage thermochemical software can be used to calculate phase and chemical equilibria of interest for the production of copper. An example of its application to the real industrial process will be given.

References [1] S. Decterov, A. Pelton, Metall. Mater. Trans. B 30 (4) (1999) 661–669. [2] P. Waldner, Internal Report, Centre for Research in Computational Thermochemistry, Montreal, 2007. [3] A.D. Pelton, M. Blander, Metall. Mater. Trans. B: Process. Metall. 17 B (1986) 805–815. [4] M. Hillert, J. Alloys Compd. 320 (2) (2001) 161–176. 2.5. Experiments and databases R. Schmid-Fetzer, J. Groebner, A. Kozlov, M. Hampl Progress in magnesium alloy database development New classes of magnesium materials with promising property profiles have been developed in recent years. The transfer of interest from classical alloy systems like AZ, AM and ZK to new innovative classes forming LPSO and related phases has opened a new field of materials with high strength, increased ductility and high fracture toughness. Mg-RE based alloys and Mg–Zn–RE alloys, including Y, play a key role in these new classes of alloys. Progress in development of our thermodynamic Mg alloy database in that direction is reported, with examples of phase formation in as-cast and heat treated alloys. In addition, the erroneous description of Mg-hcp and Zn-hcp, which spread from the COST507 database to other databases, is revealed. Not only the description as two separate phases, hcp_A3 and hcp_Zn, but also the peculiar lattice stabilities is considered unrealistic. A remedy for that erroneous description is presented for Mg–Zn and related multicomponent systems. M. Medraj, M. Mezbahul-Islam, E. Essadiqi The Mg–Cu–Ni–Y quaternary system: Thermodynamic modeling coupled with key experiments Mg based alloys provide a promising candidate for bulk metallic glass which has high mechanical strength and good corrosion resistance. Among these alloys, the Mg–Cu–Y system has the largest supercooled liquid region. Also, the Mg–Ni–Y system has been found to be a potential candidate for the Ni-metal hydride batteries. These batteries are supposed to replace the existing Ni/Cd rechargeable batteries due to environment concerns. Hence, a sound description of the Mg–Cu–Ni–Y quaternary system is essential. A thorough review and critical evaluation of phase equilibria and thermodynamic data of the phases in the Mg–Cu–Y, Mg–Ni–Y, Mg–Cu–Ni and Cu–Ni–Y and their constituent binary systems have been performed over the entire composition range from room temperature to above the liquidus. These systems are being modeled for the first time using the modified quasichemical model for the liquid phase to account for the presence of the short-range ordering properly. The Gibbs energies of the different phases have been modeled, and the optimal model parameters that reproduce all the experimental data simultaneously within experimental error limits have been obtained. Diffusion couples and key alloys are used to verify the calculated ternary systems. The alloys for diffusion couples and key experiments were prepared from pure metals (Mg—99.8 wt%, Ni—99.9 wt%, Cu—99 wt%, Y—99.9 wt%). Some of the samples have also been prepared using arc melting or induction melting furnaces. New ternary phases have been observed and their crystallographic data and homogeneity ranges have been identified using SEM/EPMA and XRD. Also, DSC has been used to

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

determine the melting temperature of the key alloys. The calculated phase diagrams are compared with the current experimental results. M. Jiang, H.X. Li, Y.P. Ren, S.M. Hao Experimental study on the long period ordered (LPO) phase in magnesium alloys Mg-base alloys have been attracted as lightweight and high strength structural materials because of their low density, high specific strength and stiffness. It has been shown that an Mg97Zn1Y2 (at%) alloy, made through rapidly solidified powder metallurgy (RS P/M) processing, reveals excellent mechanical properties including maximum tensile yield strength 610 MPa and elongation 5% at room temperature [1,2]. It has been proved that a long period ordered (LPO) structure is formed in the alloy, with a stacking sequence of ABCBCB0 , where A and B0 layers are significantly enriched by Zn and Y [3,4]. The high-strength of the Mg97Zn1Y2 alloy is due to not only a grain refinement through RS P/M, but also to the LPO phase in aMg grain. Since then more attentions have been paid to the LPO phase in the Mg-based alloys. It has been known that the LPO phase can form in Mg–TM–RE (TM¼Cu, Ni, or Zn; RE ¼rare earth metals) alloys [5,6]. As a strengthening phase in Mg-based alloys, thermodynamic stability, phase composition and volume fraction are all important information for alloy design. However, the phase diagram information for the Mg–TM–RE systems is very limited. In this work, a systematic experimental study has been performed to the Mg–TM (TM¼ Cu, Ni, Zn)–RE (RE ¼Y, Gd) alloys. It has been proved that the LPO phase is stable in all these systems with limited solubility ranges of Cu, Ni, Zn, Y and Gd. The equilibrium compositions of the LPO phase have been determined accurately, which are different from the earlier reports. Ideal TM/RE ratios to get the LPO phase, which are important for the composition design of the LPO Mg-based alloys, have been obtained. Phase relationships concerning the LPO phase, as well as the phase diagrams in the Mg-rich region of the Mg–TM–RE ternary systems have been established. References [1] Y. Kawamura, K. Hayashi, A. Inoue, T. Matsumoto, Mater. Trans. 42 (2001) 1172. [2] A. Inoue, Y. Kawamura, M. Matsushita, K. Hayashi, J. Koike, J. Mater. Res. 16 (2001) 1894. [3] E. Abe, Y. Kawamura, K. Hayashi, A. Inoue, Acta Mater. 50 (2002) 3845.

A. Kroupa, A. Dinsdale, A. Watson, J. Vrestal, A. Zemanova, P. Broz The COST MP0602 thermodynamic database for high-temperature lead-free solders COST Action MP0602 entitled ‘‘HISOLD—Advanced Solder Materials for High Temperature’’; has as its main objective, according to the Memorandum of Understanding, ‘‘yto increase the fundamental knowledge of the crucial properties of alloys that can be used as environmental friendly alternatives to hightemperature solders. The aim is to identify promising materials with a set of suitable properties, such as melting point, wettability and surface tension, which will allow them to be used successfully in a variety of industrial applications. Furthermore, health and economic issues must be taken into account during the

135

evaluation in addition to the physical, chemical and mechanical behaviour.’’ The Action started officially on May 15, 2007, and its official end is May 15, 2011. 25 countries signed the Memorandum of Understanding. Of these, 22 countries involving approx. 60 research teams were actively engaged in research as part of COST MP0602. One of the main outcomes of the Action, closely related to the topic of the CALPHAD conference, is a selfconsistent thermodynamic database developed for high temperature lead-free solders (approx. 260–350 1C) and their interaction with selected substrate materials. This database is based on the existing 11 component SOLDERS database (itself a product of the COST 531 Action) with additional elements and systems, important for HT solders. Currently the database contains 18 elements (Ag, Al, Au, Bi, Co, Cu, Ga, Ge, Mg, Ni, P, Pb, Pd, Sb, Si, Sn, Ti, Zn), but not all binary system assessments are included. The database has been tested extensively using MTDATA, Thermo Calc and Pandat to ensure portability between different platforms. Experimental thermodynamic and phase equilibrium data and thermodynamic assessments generated within the action have been included where appropriate. This work was supported by the Czech Ministry of Education, Youth and Sports projects No. OC08053 and OC09010. L. Zabdyr, G. Garzel EMF Measurements in the liquid Ag–Bi–Cu–Sn lead-free solder alloys Electromotive force measurements were carried out in the liquid quaternary Ag–Bi–Cu–Sn alloys by the use of solid oxide electrolyte galvanic cells with air reference electrode. Experiments were made within temperature interval: 950–1300 K along four composition paths of constant ratios: XAg:XBi:XCu¼ 1, XAg:(XBi:XCu)¼3/2, XBi:(XAg:XCu)¼3/2 and XCu:(XAg:XBi)¼ 3/2 and tin content changing from 0.1 to 0.9 mole fraction, every 0.1. All the results were approximated by straight line equations: EMF vs T, and tin activities were then calculated in two arbitrary temperatures. Results were presented by graphs and listed in tables. Unusual activity curve for XBi:(XAg:XCu)¼3/2 composition path was most probably caused by miscibility gap detected earlier in Bi–Cu–Sn ternary liquid alloys. Results will be used along with planned calorimetric measurement results for optimization of the quaternary system under accord by CALPHAD approach. U.R. Kattner, B. Sundman, M. Palumbo, S.G. Fries Open CALPHAD—Software and databases Calphad type databases and various application software have become a necessary tool for industrial development of new materials and alloys. In particular for applications in phase transformations and thus the simulation of microstructures and materials properties the thermodynamic and kinetic databases have now gained an established position. But this is also a field of intense research as new data and first principle techniques provide new ideas for models as well as thermodynamic data in composition ranges not accessible by experimental techniques. Thus there is a need for constant revision of databases and software to include these ideas and data. Unfortunately the companies providing the software and databases are too small to carry out heavy research projects. Since today all major thermodynamic software and databases are commercial this leaves very little opportunities for researchers to develop and test new ideas. There are some minor open source software packages available for Calphad applications but the Sapiens

136

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

project at ICAMS in Bochum, Germany, has taken the initiative to start developing open software with full modeling facilities and provide this to the scientific community. It is called Open Calphad and it is still in the cradle but will be made available on the web, without cost, either as free or open software, making it possible to modify the source code and to implement it in any other software. This facility has already been utilized for evaluating new heat capacity models (see presentation by Mauro Palumbo). This software, in part utilizing code that was developed by Leo Lukas [2], is written in Fortran 90 with a command driven user interface and a software interface based on the TQ definition [3]. We hope that several programmers will be interested to provide interfaces to other languages as well as a more advanced GUI. Acknowledgements ThyssenKrupp AG, Salzgitter Mannesmann Forschung GmbH, Robert Bosch GmbH, Bayer Materials Science AG, Bayer Technology Services GmbH, Benteler Stahl/Rohr AG, state of North RhineWestphalia and European Commission in the framework of the European Regional Development Fund (ERDF).

B. Hallstedt Thermodynamic evaluation of the Ti–Al–C system The Ti–Al–C system is of interest for several applications. The intermetallic g-TiAl (L10) is a candidate light-weight material for high temperature applications. It normally contains a small amount of carbon, which precipitates as Ti3AlC on short term annealing and as Ti2AlC after longer times. Ti3AlC has the perovskite structure, i.e., L12 with the central interstitial site filled with C. There is a phase with this structure also in, e.g., the Fe–Al– C system (Fe3AlC) where it is called k carbide. Ti2AlC and Ti3AlC2 belong to a group of phases called MAX-phases. The MAX-phases have layered crystal structures, sometimes called nano-laminates, and show properties intermediate between metallic and ceramic materials. This makes them attractive for a number of applications. All three ternary carbides show considerable non-stoichiometry with respect to carbon. Here a thermodynamic evaluation of the Ti–Al–C system is presented based on evaluations of the three binary sub-systems from literature. Experimental data on the carbon solubility in the intermetallic Ti–Al phases and the stoichiometry of the ternary carbides are in some cases strongly contradictory. For Ti2AlC first-principles DFT calculations were used to study the energies of some of the possible defects, including C vacancies. The focus of the presentation will be on basic modelling issues and the selection of experimental data.

enthalpy and entropy to lattice stability. In this presentation, our recent efforts in predicting the enthalpy and entropy contributions to lattice stability of Ti [5] and Zr [6] will be discussed. References [1] Z.K. Liu, First-principles calculations and CALPHAD modeling of thermodynamics, J. Phase Equilib. Diffus. 30 (2009) 517–534. [2] Y. Wang, S. Curtarolo, C. Jiang, R. Arroyave, T. Wang, G. Ceder, L.Q. Chen, Z.K. Liu, Ab initio lattice stability in comparison with CALPHAD lattice stability, Calphad 28 (2004) 79–90. [3] P.E.A. Turchi, I.A. Abrikosov, B. Burton, S.G. Fries, G. Grimvalle, L. Kaufman, P. Korzhavyi, V.R. Manga, M. Ohno, A. Pisch, A. Scott, W.Q. Zhang, Interface between quantum-mechanical-based approaches, experiments, and CALPHAD methodology, Calphad 31 (2007) 4–27. [4] V. Ozolins, First-principles calculations of free energies of unstable phases: the case of fcc W, Phys. Rev. Lett. 102 (2009) 065702. [5] Z.G. Mei, S.L. Shang, Y. Wang, Z.K. Liu, V. Ozolins, Ab initio study of hcp-to-bcc phase transition in Ti, in preparation. [6] V. Ozolins, First-principles calculations of free energies of bcc and hcp Zr, in preparation. M. Hindler, A. Mikula Thermodynamic properties of the Au–Sb–Sn and Au–Sb system Some Au–Sb–Sn alloys might become useful to replace lead containing solders at temperatures between 300 and 450 1C. We determined the thermodynamic properties of such alloys in the liquid state with an emf method, using a liquid electrolyte and with calorimetric measurements. The results of both experiments are in excellent agreement with each other. See Fig. 1. In order to calculate the integral quantities with the Gibbs– Duhem equation the thermodynamic data of the binary systems, in our case the Au–Sb system has to be known. Since we found only one set of data in the literature we had to use them. As it

Z.-K. Liu, Z.-G. Mei, V. Ozolins Lattice stability by ab initio molecular dynamic simulations Thermodynamic modeling based on the CALPHAD technique relies on available thermochemical and phase equilibrium data, and the lattice stability is its foundation. First-principles calculations based on density functional theory are becoming a powerful approach in providing thermochemical data in addition to experimental investigations [1]. However, it remains challenging in predicting the lattice stabilities of unstable phases from firstprinciples calculations [2,3]. The recent breakthrough demonstrated the applicability of ab initio molecular dynamics (AIMD) in predicting the lattice stability of unstable fcc W [4]. Furthermore, it was shown that various lattice stability values proposed in the literature could be correlated through the relative contributions of

Fig. 1

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

turned out with this set of data we could not get reliable results for the integration. So we had to re-measure the properties of the Au–Sb system. The problem for the emf measurements in this system was, as it has already been pointed out by the previous author, the high vapour pressure of SbCl3. We prepared the KCl– LiCl electrolyte, powdered it and mixed it with SbCl3. The mixture was sealed in a tube and heated in a furnace for a longer time at 110 1C. The reason was to form a stable compound of LiCl and SbCl3. This was used as a stable electrolyte for the emf measurements. The results of the Au–Ab measurements fit very well into the results of the ternary system. ¨ D. Kobertz, M. Muller Experimental studies and re-assessment of the quasi-binary systems containing the sulfates of sodium, potassium, and calcium by differential thermal analysis and X-ray diffraction The combustion of lignite containing high amounts of sulfur, calcium, potassium, and sodium induces the formation of sulfatic depositions and corrosion related processes on refractory metal and ceramic material. These depositions diminish the heat conductance and their spalling leads to damages in the power plant. The mechanisms for the formation of these conglomerations are initiated by low-melting or related to adhesive sulfatic phases. Limitation of fossil fuel resources requires an increase of the efficiency of power plants by using combined cycle power systems. The pressurized pulverized coal combustion (PPCC) combined cycle is a coal fired combined cycle concept which is able to achieve efficiencies in excess of 53%. The direct use of the hot flue gas for driving a gas turbine requires a hot gas cleanup to achieve corrosion prevention of the turbine blading. Hot corrosion on turbine blading is initiated by deposition or condensation of corrosive species like alkali sulfates and dependent on concentrations of alkalis in the hot flue gas. Type I hot corrosion is caused by the formation of liquid Na2SO4 above its melting point (884 1C). Type II hot corrosion is caused by the formation of an eutectic melt of NiSO4 and Na2SO4 above 671 1C. NiSO4 itself is formed by the reaction of the oxide scale of Ni base alloys in blading with SO3 in dependence of the SO3 partial pressure in the hot flue gas. The motivation of our studies is to understand the aforementioned alkali sulfate related processes under applied conditions and accrued phases. Multi-component sulfatic systems can only be really understood after comprehending the basic binary sub-systems. This work includes the experimental studies of the three sub-systems Na2SO4–CaSO4, K2SO4–CaSO4, and Na2SO4–K2SO4 since there are variant and even contradictory results and interpretations in the literature (e.g., [1–11]). differential thermal analysis (DTA) with simultaneous thermo gravimetry (TG) as well as X-ray diffraction (XRD) measurements were done both to achieve the transition temperatures and the phase compositions in the sulfatic systems.

3. Posters 3.1. Applications to process and materials design and development A.S. Gandhi, A. Muralidhar, K.C. Hari Kumar Effect of surface energy on the zirconia-yttria phase diagram: A CALPHAD approach Yttria stabilised zirconia ceramics (YSZ) are widely used for their attractive properties such as transformation toughening, high ionic conductivity, low thermal conductivity and high

137

thermal expansion coefficient. The useful phases and microstructures of zirconia are almost always metastable. Zirconia phase stability is influenced by surface/interfacial energies of the phases [1]. Stresses also influence phase stability. In this work we have modified Gibbs energy functions of a few phases of ZrO2–YO1.5 system by adding surface energy contributions [2]. The phases selected are monoclinic (m), tetragonal (t) and cubic (c). Phase boundaries and important T0 lines have been calculated for particle sizes down to 20 nm. The results are correlated with the observed phase stability reported in the literature. For instance, transformation toughening requires presence of a tetragonal phase with Ms temperature just below room temperature. This leads to stress-induced martensitic transformation in the crack wake during fracture with enhanced fracture toughness. The Ms temperature is related to the T0 (t/m). It is known that pure t-zirconia can form at room temperature if the surface area is sufficiently large, e.g., in nanoparticles. The t-YSZ can be stabilised if the surface area is large or stresses are present. Also, aging of non-transformable t0 in thermal barrier coatings and cYSZ in fuel cells are important issues. Implications of our results for these scenarios will be discussed. References [1] M.W. Pitcher, S.V. Ushakov, A. Navrotsky, B.F. Woodfield, G. Li, J. Boerio-Goates, B.M. Tissue, J. Am. Ceram. Soc. 88 (2005) 160. [2] M.J. Mayo, A. Suresh, W.D. Porter, Rev. Adv. Mater. Sci. 5 (2003) 100.

J.-C. Dumas, J.-P. Piron, C. Martial Evaluation of the chemical composition of irradiated mixed carbide fuel The mixed carbide (U,Pu)C is a reference material for Generation IV (gas-cooled fast reactor or sodium fast reactor) nuclear fuels. The creation of fission products (FP) can modify the physical and chemical properties of the fuel behaviour during irradiation. As they control the carbon activity in the fuel, it is important to assess the influence of burn-up and temperature on the speciation of FP (solid, liquid and gaseous phases). The first part of this study is dedicated to the thermodynamic modelling of the irradiated fuel introduced into the FactSage code. The second part deals with the evaluation of the chemical state of the fission products as a function of burn-up and initial composition (carbon content) for representative operating temperature values. For that purpose, we have used the isotopic composition data given in [1] and compared our results to those obtained with the Solgasmix-PV computer program code and also to experimental observations on irradiated fuels from [2]. Special attention is paid to the evolution of the carbon over metal (C/M) content with burn-up. Indeed, as the C/M ratio decreases with burn-up, the initial M2C3 phase present in the fuel might disappear as soon as C/M decreases below one and due to the narrow nonstoichiometry of the MC phase [3,4], a liquid (U,Pu) phase might appear, which is not recommended for safety criteria. Finally, comparison of our results obtained on of un-irradiated (U,Pu)C fuel (isothermal ternary sections, carbon and plutonium activities) with those calculated by the TCC Software using the CALPHAD description of the (U–Pu–C) system developed in the frame of the FuelBase Project [5] shows significant differences. Those last results lead us to conclude that a complete CALPHAD description of the whole (U–Pu–C–FP) system would be helpful if we want to improve the operating conditions of this kind of fuel.

138

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

References [1] R. Agarwal, V. Venugopal, J. Nucl. Mater. 359 (2006) 122. [2] Hj. Matzke, Science of Advanced LMFBR Fuels, Elsevier Science Publishers B.V., 1986. [3] H. Kleykamp, J. Nucl. Mater. 221 (1985) 131. [4] H. Holleck, Thermodyn. Nucl. Mater., IAEA Vienna, 1975. F.D. Murari, A. Costa e Silva, R.R. Avillez Effect of boron on the microstructure and mechanical properties of cold rolled multiphase steels The influence of boron content on the phase transformation characteristics, microstructure and mechanical properties of multiphase steels was investigated by means of computational thermodynamics (Thermo-Calc), dilatometry, quantitative metallography and tensile tests. The steels were prepared as 50 kg ingots in an induction furnace operating under argon gas atmosphere with boron contents in the range of 0 to 47 ppm. The ingots were cut into blocks of 35 mm thick, which were reheated to 1250 1C for 1 h and hot rolled in six passes to 7.0 mm thick. The hot rolled sheets were machined and then cold rolled to a final thickness of 1.2 mm. The continuous annealing cycles were ¨ carried out in a Bahr dilatomer and in a Gleeble machine. It was found that the boron did not influence the amount of austenite formed during the heating and soaking steps in the continuous annealing simulations. However, the boron influenced the austenite transformation during the cooling step, reducing the amount of ferrite and increasing the amount of bainite. Regarding mechanical properties, the boron addition increased the mechanical resistance and reduced the ductility. The steels containing boron levels of up to 27 ppm exhibited the largest effect. The results of amount of austenite achieved by means of the ThermoCalc presented themselves overestimated compared to those obtained by dilatometry and metalography, especially for soaking temperatures below 800 1C. References [1] D.T. Llewellyn, Ironmaking Steelmaking, 20 (5) (1993) 338–343. [2] G.M. Pressouyre, G. Primon, R. Blondeau, Int. Conf. HSLA Steels’ 85 (1985) 335–350. [3] H.K.D.H. Bhadeshia, L.E. Svensson, Model for boron effects in steels welds, in: T. Zacharia (Eds.), Proceedings of the International Conference on Modelling and Control of Joining Processes, Orlando, Florida, American Welding Society, 1993, pp. 153–160.

F. Hodaj, O. Mailliart, V. Chaumat The role of atmosphere on the interfacial interactions between molten oxide glasses and silicon carbide A method for joining silicon carbide between 1300 and 1600 1C under air using calcium aluminosilicate glasses has been developed [1]. This technology avoids protective atmosphere resulting in high production cost and allows repairing more easily damaged joined components. Many studies have been performed on the joining of SiC using glass-ceramics under vacuum or in neutral gases. Only a few studies are available in the literature on interfacial interactions between molten glasses and SiC as well as

on the joining of SiC by glass-ceramics under air. In this work interfacial interactions between calcium aluminosilicate glasses and SiC under air are studied by performing wetting and joining experiments on polycrystalline SiC between 1100 and 1600 1C. Some specific experiments are also performed: (i) on silica or monocrystalline SiC under air and (ii) on polycrystalline SiC under argon or vacuum. The interfacial interactions between glass and SiC strongly depend on the temperature, the furnace atmosphere and the content of silica glass. Moreover, under air, there is an enormous difference between the reactivity in a sessile drop configuration and in a brazing configuration. The purpose of this presentation is to focus on fundamental issues of interfacial interactions on the molten oxide glasses SiC system. The confrontation of results of thermodynamic calculations with experimental observations allowed to elucidate the role of oxygen on these interfacial interactions. References [1] O. Mailliart, V. Chaumat, F. Hodaj, French Patent Application No. 08 55857 filed on September, 1, 2008. ~ H.L.G. Magalhaes, C.A. da Silva, A. Costa e Silva Improved cleanliness of SAE 1045 Al-killed steel grade with applied in the automobilistic industry, using computational thermodynamic Steelmaking has been devoting large amount of resources to improve processes and products in order to face a strong competition based on quality requirements. Inclusion engineering has been included in the steelmaking vocabulary as a means of forecasting the type of inclusion, its morphology, size and distribution in the steel as well as its thermo-mechanical behavior. One of the several inclusion engineering tools is computational thermodynamics. Its application allows the inclusion chemistry to be foreseen as a function of steel composition, pressure and temperature. In this work Thermo-Calcs and CEQCSI software have been used to evaluate the possibility of formation of various types of inclusion. It has been also investigated the concept of ‘‘casting window’’ and the formation of macroinclusions (complex inclusions). The experiment included some statistic tools employed with the main objective of measuring the impact of variables: size, distribution, morphology and area in percentage of the CaS and oxisulfides inclusions where the occurrence of macroinclusion is minimized. The identification of the inclusions was made through SEM-EDS. Box-plot diagrams suggest that the best configuration to achieve a criterion based on reducing size variability of type 3 oxisulfide and calcium sulfide. Main effect diagrams show the trend to higher calcium sulfide content for increasing CaSi addition and sulfur content. The manganese sulfide content increases for increasing CaSi addition. References [1] A. Costa e Silva, Equilibrium calculation at steel shop by computational thermodynamics. Tecnol. Metal. Mater. Sa~ o Paulo 03 (01) (2006) 45–52. [2] A. Kirsch-Racine, A. Bomont-Arzur, Calcium treatment of medium carbon steel grade for machinability enhancement: from the theory to industrial practice. Revue Me´tall.—CIT 12 (2007) 591–597.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

I.G. Kim, J.Y. Lee Computational thermodynamics and spheroidisation of bearing steels Bearings are vital components of almost all machines. Steels supplied to the bearing manufacturer for making raceways are in the form of tubes or forgings, whereas the rolling elements are made by cold forging drawn wire. The aim of spheroidise annealing is to facilitate machining and cold-forming operations by inducing a microstructure which is a mixture of relatively coarse cementite particles and ferrite; the roughness of machined surfaces is also reduced in the process. Reasonably large cementite particles and a small number density is conducive to less wear on the tools used for machining. Lamellar carbides of the type associated with incomplete spheroidisation lead to enhanced tool-degradation. Spheroidisation reduces the hardness of the steel supplied to a bearing manufacturer to about 230 HV. There are many properties required from bearing steels. This study is in its first stage limited to exploring modifications which enhance the rate of spheroidisation, with a focus on the ACM temperature. Beginning with a typical bearing steel as the base, Fe–1C–1.45Cr– 0.25Si–0.35Mn, having about 910 1C of ACM temperature, attempts were made to increase the ACM temperature and the fraction of cementite present at the ACM temperature. Using the ThermocalcTM with the TCFE6 database, we build up a software package to generate phase diagrams affected by all the possible elements from the database systematically. It is found that B, P, S and Cr are effective to increase ACM. We found that the effect of boron seems abnormal, which suggest to refine thermodynamic data of boron. Interestingly, the broadening of eutectoid reaction region was observed for phosphorus and sulfur cases, but a further study should be required by allowing the P and S compound phases. The volume fraction of cementite around the ACM temperature of each composition was made, and it is concluded that P and Cr are effective in providing large volume fraction with high ACM temperature. J.B. Bacalhau, E.N. Souza, A.B. Farina, C.A. Barbosa Thermodynamic simulation of nitrogen and manganese contents in Delta ferrite formation of plastic mold stainless steel Some grades of low carbon and low nickel martensitic stainless steels are applied in components of plastic injection molds like hot runners and refrigeration plates because of their corrosion resistance. The production of these steels has special controls due the low temperature range in forging or rolling. If the deformation temperature is too high, there is the precipitation of delta ferrite in steel microstructure, which damages the material corrosion resistance. In the case of low temperature it is required excessive power of the forging press or rolling mill. The present work discusses the use of thermodynamic simulations to develop a new stainless steel with higher deformation temperature range. The effect of nitrogen and manganese contents in delta ferrite formation has been modeled in the thermocalc software with TCFE6 thermodynamic database. The results obtained by computer modeling it were evaluated in alloys produced in pilot scale in a VIM furnace. Samples of these materials were heated to different temperatures and by metallographic analysis the amount of delta ferrite precipitated was measured. The accuracy of results between modeling and laboratory investigation were extremely high. A good combination of chemical composition, mainly N and Mn contents, enabled a steel development with deformation temperature range of 40–90 1C higher than others commercial such grade steels.

139

References [1] G. Roberts, G. Krauss, R. Kennedy, Tool Steel, fifth ed., ASM International, 1998. [2] M. Hillert, The compound energy formalism, J. Alloys Compd. 320 (2001) 161–176. [4] ASM Handbook, Properties and Selection: Stainless Steels, Tool Materials and Special-purpose Metals, ninth ed., vol. 3. M.F. Campos, S.B. Ribeiro, J.F.C. Lins, A.B. Farina Microstructural evolution in the aging of inconel 718 In the case of superalloys, formation of precipitates in grain boundaries is deleterious for the creep properties. Thus, CALPHAD method is an important aid to predict superalloys compositions able to resist longer periods at high temperatures. It should be reminded that each INCONEL 718 has slightly different chemical composition. It is relevant to note that minor alloying elements may change the amount of volumetric fraction of the phases and also the precipitation sequence. The main hardening mechanism of Inconel 718 is due to the formation of a semicoherent precipitate gamma double prime (Ni3Nb) (boy centered tetragonal) and, in a lesser extent, the coherent precipitate gamma prime Ni3(Al,Ti) (fcc L12 structure) also contributes for the mechanical properties. However, with increasing aging time, coherent gamma double prime (Ni3Nb) tends to transform in Ni3Nb-delta [orthorhombic, DO22 structure]. This delta phase usually precipitates at grain boundaries. In the present study, the precipitation hardening in Inconel 718 alloys was studied in detail, using techniques for microstructural characterization, as SEM-EBSD (Scanning Electron Microscope with Electron Backscatered Diffraction) and X-ray Diffraction (XRD) Synchrotron with Rietveld Analysis. The studied Inconel 718 alloy was submitted to solubilization at 980 1C during 30 min under vacuum, followed by a quenching in water. The samples were subsequently submitted to annealing at the temperature of 680 1C during 1, 2, 4, 8, 10, 20, 50 and 100 h. Carbides as MC were identified in the microstructure. After 10 h of heat treatment, delta phase can be identified along grain boundaries. These results are in agreement with ThermoCalc calculations (database TCNI8), which predicted the presence of delta phase at lower temperatures than that suggested in some previous studies [1]. Acknowledgements FAPERJ CNPq LNLS, Research Project D10B—XPD-9924. References [1] C.T. Sims, N.S. Stoloff, W.C. Hagel, Superalloys II, 2nd ed., Wiley-Interscience, New York, 1987. M.G. Lage, A. Costa e Silva Evaluating segregation in HSLA steels using computational thermodynamics The level of segregation in steel plates used to manufacture large diameter high strength pipe is important to pipe manufacture and performance. In this work, as part of the alloy design process, the level of segregation of different HSLA steel compositions was evaluated using computational thermodynamics. Equilibrium and Scheil–Gulliver model calculations were used to identify

140

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

the most promising compositions. Phase diagrams, phase fraction diagrams and the progress of the composition of the various phases as a function of temperature were evaluated, to quantify the segregation towards the end of solidification. Results indicate that S is the stronger segregating element, as expected, but also that Nb, C and P segregate considerably. P and Mn segregation are well known to be associated with banding in these products and the present results confirm this. The enrichment of the liquid phase in S, P and Nb is related not only to the content of these elements but also to the steel C content. Even using only these two extreme models – Equilibrium and Scheil – the results are useful to support the alloy design, in special concerning the evaluation of segregation of C, Mn, S, P, Nb e Ti, elements that have a significant influence both on mechanical properties and on the resistance to hydrogen embrittlement of the steels used for large diameter piping. In the next step, these results will be compared with the results of the continuous cast and controlled rolled plates. References [1] D. Chakrabarti, et al., Metall. Mater. Trans. A 39A (2008) 1963–1977. [2] J.J.R. Mondrago´n, et al., ISIJ Int. 48 (4) (2008) 454–460. [3] A. Schneider, et al., Int. J. Mater. Res. 6 (2008) 674–679. M. Ba, A. Dia, P. Berthod, L. Aranda, Th. Schweitzer, P. Villeger Experimental and thermodynamic study of ternary Fe–30Cr–xC carbon-rich alloys with x varying from 2.5 to 5.0 wt% White cast irons are not expensive very hard alloys which can be used for wear resistance. Other iron alloys candidates for the same applications, and offering a better resistance against high temperature oxidation or corrosion, as well as a higher refractoriness, are Fe-based chromium—rich alloys also containing a great quantity of carbon. The aim of this work is to explore the microstructures and of the melting temperature ranges of alloys belonging to the ternary family Fe(bal.)–30Cr–xC (in wt.%), with x increasing from 2.5% to 5%, and to compare these metallographic data with thermodynamic calculations results. Six alloys were synthesized by foundry practice and they were metallographically characterized in their as-cast state as well as in their heat-treated states (50 h at 1000, 1100 and 1200 1C). This was done by using scanning electron microscopy, measuring their carbides surface fractions. Additionally DTA/DSC experiments were carried out to assess the melting-start and melting-end temperatures of the alloys. The characterization of both matrix and carbides was completed by performing X-Ray diffraction. Natures and fractions of the different phases, as well as the solidus and liquidus temperatures, were compared to results issued from Thermo-Calc calculations. For a carbon content equal to 2.5 wt% the alloy microstructure is hypo-eutectic with presence of dendrites of Fe matrix delimiting the {Fe matrix þchromium carbides} eutectic spaces, while for carbon contents equal to 3 wt% and higher dendrites are replaced by coarse pro-eutectic carbides. Contrarily to what can be observed for similar alloys based on nickel or cobalt lamellar graphite never appeared in these iron alloys for the same carbon contents. Matrix is Back Centered Cubic, and carbides are exclusively Cr7C3 for all the as-cast alloys. References [1] P. Berthod, E. Souaillat, O. Hestin, As-cast microstructures of {M–30Cr–0 to 5%C} ternary alloys. Part 1: Nickel-base alloys, Mater. Sci.: Indian J. 6 (4) (2010) 260–266.

[2] P. Berthod, O. Hestin, E. Souaillat, As-cast microstructures of {M–30Cr–0 to 5%C} ternary alloys. Part 2: Cobalt-base alloys, Mater. Sci.: Indian J. 7(1) (2010). P. Berthod Thermodynamic calculations used for characterizing the sub-surfaces changes induced by high temperature oxidation for carbides-reinforced alloys In addition to the external oxide scale, oxidation at high temperature (typically 1000 1C) induces subsurface modifications as depletion in Cr or Al, formation of internal oxides and other phenomena depending on the alloy. In the specific case of superalloys or other refractory alloys strengthened by carbides, frequent phenomena are the disappearance of carbides over a depth increasing with time and/or, in some cases, the precipitation of new carbides along an internal alloy band separating the outer carbide-free zone and the not modified bulk. It can be useful to characterize this whole zone affected by oxidation to complete the knowledge of its new chemical composition by determining the new carbon content. Image analysis and thermodynamic calculations allow specifying this characteristic, and furthermore the new level of local refractoriness. Different refractory alloys – cobalt-based, nickel-based or iron-based – containing different types of interdendritic carbides (Cr7C3, TaC, etc.) were elaborated by foundry and subjected to oxidation at different high temperatures (1000–1300 1C). The oxidized samples were metallographically prepared and their sub-surface microstructures examined by scanning electrons microscopy. The depths of the formed carbidefree zones were measured, the new carbides were identified by microprobe/WDS measurements and image analysis was performed to measure their surface fractions. Thermodynamic calculations using Thermo-Calc were performed first to assess the new carbon distribution, and second for the estimation of the new solidus temperatures displayed by the outer zone of alloy. It appears that the C content is now extremely low in the carbidefree zone, leading to a higher local refractoriness which can eventually protect the {partially melted}-bulk alloy from catastrophic oxidation in case of sudden temperature increase. After oxidation at 1000 1C new carbides precipitated in a band between the carbides-free zone and the bulk. References [1] D.J. Young, High Temperature Oxidation and Corrosion of Metals, Elsevier, Amsterdam, 2008. [2] P. Berthod, Experimental and thermodynamic study of nickel-based alloys containing chromium carbides. Part II: Study of the sub-surface characteristics of Ni–30 wt%Cr–xC alloys oxidized, Calphad 32 (3) (2008) 492–499.

~ R.F. Guimaraes, H.C. Miranda, H.F.G. Abreu, F.H.C. Sabo´ia The effect of Cr and Mo in the formation of intermetallic phases Molybdenum (Mo) containing stainless steels are commonly used to prevent naphthenic acid corrosion (NAC) in the refining industry. The development of new alloys with high content of Mo associated with Cr or with Cr and Ni is one of the strategies to reduce the effects of naphthenic acid corrosion. In this work 100 kg ingots of novel alloys with Cr of 9, 15 and 17 wt% and Mo 5, 7 and 9 wt% were cast, forged and hot rolled. Prior to the

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

manufacturing stage a study of phases and intermetallic precipitates was done using Thermocalc. Phase diagrams for different chromium and molybdenum contents were plotted to identify and calculate phases formed at different temperatures. To validate the results obtained in the simulation, microstructural characterization was performed by scanning electron microscope (SEM), electron back scattered diffraction (EBSD) and energydispersive X-ray spectroscopy (EDX). The addition of chromium and molybdenum favors a ferritic microstructure of the alloys studied as well as the formation of intermetallic phase as chi, sigma, mu and carbides M6C and M23C6. The analysis of phases by SEM and EDX confirmed the formation of the ferritic phase and carbides M6C and M23C6. The content of elements present in the carbides formed were influenced by the content of chromium and molybdenum present in each experimental alloy. The intermetallic phases chi, sigma, mu were well identified by EBSD. The results indicated that the use of thermodynamic software is a good tool to foresee possible problems in the development of Fe–Cr–Mo alloys.

141

Fig. 1. Geometry model for multicomponent diffusion simulation between base metal and y foil.

References [1] ASM Handbook, Casting, Nickel and Nickel Alloys, vol. 15, 10th ed., ASM, Materials Park, OH, 1993. [2] ASM Handbook, Welding, Brazing and Soldering, Selection of nickel, nickel–copper, nickel–chromium, and nickel– chromium–iron alloys, vol. 6, 10th ed., ASM, Materials Park, Ohio. R.N.C. Siqueira, R.R. Avillez, E.A. Brocchi Chlorination of vanadium oxide under the presence of graphite Chlorination is an important process for extracting metals of high economic value from minerals and industrial wastes, in particular, if volatile chlorides and or oxychlorides are produced, which in a further step can be separated through selective condensation. Graphite has proven to be an important reducing agent to chlorination, as it reacts with oxygen, promoting the process driving force. The present work focuses on the chlorination of vanadium oxide under the presence of graphite. Thermodynamic data for the vanadium chlorides and oxychlorides are reviewed and compared with the available database. The equilibrium thermodynamic conditions for the vanadium oxide chlorination are calculated for different gas atmospheres with and without graphite as a reducing agent. The atmosphere composition and the reaction temperatures are determined from predominance diagrams. T.Maehsima, K. Tanaka, T. Ohshima, H. Takao Prediction of liquid phase behavior during the rapid transient liquid phase bonding process of steel using cementite filler metals A plane-to-plane bonding by the surface liquid reaction from an inserted foil enables a simplified forming of hollow or complex steel parts. We have proposed a rapid transient-liquid-phase (TLP) bonding process of steel using monophase cementite (y) foils which was intentionally-created by substituting chromium for a part of iron [1]. The y filler metal is suitable as inserted foils because the emerged liquid layer ensures the dissolution of alloying elements from base steel with melting its surface, followed by quick isothermal solidification owing to the rapid

Fig. 2. Changes in maximum liquid thickness with the heating rates.

carbon diffusion. In this study, the effect of heating rates on the eutectic liquid behavior during TLP bonding process was investigated using a multicomponent diffusion simulation. One-dimensional half model as illustrated schematically in Fig. 1 is considered due to the inherent symmetry of the bond. The simulation showed that the maximum thickness of liquid layer increased with an increasing heating rate (Fig. 2). The maximum liquid thickness can be estimated by residual thickness of the y foil just before emergence of the liquid during heating. The result is interpreted as heating rates affect the degree of solid dissolution of y foils resulting in determination of the maximum liquid thickness. References [1] K. Tanaka, H. Takamiya, N. Iwata, K. Nakanishi, Tetsu Hagane 96 (2010) 6. V. Lutsyk, A. Zelenaya ‘‘Wrong and Useful’’ models of SiO2–CaO–Al2O3 T–x–y diagram Existing T–x–y diagram models of CaO–Al2O3–SiO2 system solve the distinct applied tasks, but have the significant disagreement with the topological structure. In particular, the immiscibility surfaces and four liquidus surfaces (3CaO  SiO2, 3CaO  2SiO2, 3CaO  Al2O3, CaO  6Al2O3) are absent in V. Danek’s thermodynamical model [1, p. 147]. In the model of Berman and Brown [2] one liquidus surface (5CaO  3Al2O3) is internal and does not adjacent to the binary system CaO–Al2O3, in distinction to the experimental data. The model of Shobu, calculated by the thermodynamics software CaTCalc [3], does not consider three liquidus surfaces (3CaO  SiO2, 3CaO  2SiO2, 5CaO  3Al2O3). In our work, a kinematical method of diagram surfaces description [4] is used for the simulation of computer model CaO–Al2O3–SiO2. In this case, the surface is given by the

142

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

motion of the forming line along to the directing lines, described by the interpolating polynomials. The coordinates of binary and ternary points as well as the points defining the curvatures of lines on the contour surfaces and the surfaces themselves were used as an initial data for model realization. The obtained model is formed by 15 liquidus surfaces, an immiscibility surface, 80 ruled surfaces, 16 horizontal complexes at the temperature of invariant points and 16 vertical planes of triangulation; and it involves 33 two-phase regions and 46 three-phase ones. Such model makes possible to consider the boundaries of twoand onedimensional concentration fields with the different unique crystallization schemes, and to visualize the phase’s trajectories in any diagram part with the decoding of crystallization stage on the mass balance diagrams. Comparing different phase diagrams models for the same ternary system, it has been confirmed once more, that ‘‘All models are wrong, but some are useful’’ [5]. References [1] V. Danek, Physico-Chemical Analysis of Molten Electrolytes, Elsevier, 2006. [2] R.G. Berman, T.H. Brown, Geochim. Cosmochim. Acta 48 (1984) 661. [3] E.M. Levin, C.R. Robbins, H.F. McMurdie, Phase diagrams for ceramists: Ohio, Am. Ceram. Soc., 1964. [4] K. Shobu, CALPHAD 33 (2009) 279. [5] V.I. Lutsyk, A.M. Zyryanov, A.E. Zelenaya, Russ. J. Inorg. Chem. 53 (2008) 792. [6] M. Gaune-Escard, Molten Salts Ionic Liquids Bull. 98 (2009) 1. W.V. Bielefeldt, A.C.F. Vilela Thermodynamic analysis of the inclusions formation in steels for the automotive industry At present days, there are higher quality requirements for the production of special steels, remarkably those intended for automotive industry. Furthermore, such steels are considered critical in relation to their production by continuous casting (CC) process, since contents of elements such as Al, S, Ca, and O must be controlled. Steels for automotive parts need to have improved machinability and mechanical properties (grain size and fatigue resistance). In this sense, the production of aluminium killed, high

sulphur steels is increasing. Concerning continuous casting process, many factors have influence on steel castability, such as: refractory materials, slag composition, steel covering in the tundish, valve material and design, temperature control during the process, waiting times, among others. The control of inclusions is only one among these factors; however, it is very important for improving castability. Calcium alloys are widely employed to control composition, distribution and morphology of oxide and sulfide inclusions in steels. The computational thermodynamics shows to be an excellent tool for comprehending physicochemical phenomena which occur in steelmaking, and, moreover, it can help engineers from industrial plant taking decision. The general purpose of this work is the thermodynamic study of non-metallic inclusions in the CC tundish for SAE 1141 and SAE 8620 steels. The specific purposes are: (1) obtaining the phases formed and their composition in inclusions by applying computational thermodynamic (FactSage); and (2) establishing conditions of steel composition for the improvement of steels castability. SAE 1141 quality combines both good conformability and machinability in the same steel grade. It is widely employed in forging segments to produce components with both relative complexity and mechanical demand, such as forks and shaft ends for automotive industry. SAE 8620 steel is classified by SAE-AISI as nickel–chromium–molybdenum–steel, which is a low alloy steel, used in components under the cementation process. SAE 8620 steel makes part of a series of steels for cementation, which have sufficient hardenability to be oil-quenched, acquiring, in their core, good values of ductility. They are employed for gear building sets, pieces for light works, small mechanisms, pins, etc, generally, materials which present as the most important characteristic wear resistance. Thermodynamic simulations were performed with the co mmercial software FactSage 5.5 and databases (FToxid and FSstel). Simulations were carried out by using both global chemical composition and average temperature of liquid steel in the CC tundish. The result obtained is the steel composition and non-metallic inclusions (oxides and sulfides) at the temperature of 1520 1C (SAE 1141) and 1540 1C (SAE 8620). Curves obtained in the figures (resulting in castability or liquid window): Calcium-aluminates saturation line (CaO–Al2O3, CaO–2Al2O3), considering a minimum of 30 and 35% of CaO in inclusions, CaS saturation lines: maximum of 5 and 10% of CaS, Lines of starting liquid, 40% and 100% of liquid phase in inclusions Figs. 1 and 2

Fig. 1. Liquid window for SAE 1141 steel and liquid phase formation, Total oxygen ¼ 24.5 ppm; S ¼0.101% and T ¼1520 1C.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

143

Fig. 2. Liquid window for SAE 8620 steel, including formation curves of liquid fraction and CaO–Al2O3, Total oxygen ¼18.0 ppm; S¼ 0.025% and T¼ 1540 1C.

For SAE 1141 steel conditions, 10–45 ppm Al, there is a great difference between formation curves of 35% CaO, and the points of liquid fraction. However, from 45 ppm Al, they coincide. For Al¼50 ppm, there is not 2CaO–Al2O3–SiO2 formation, and the liquid phase starts with lower calcium content. However, for Al¼80 ppm, formation of CaO–Al2O3 occurs, backing to increase the calcium content necessary to form the liquid phase. Both conditions of chemical composition for obtaining window of liquid inclusions and the influence of aluminum content on forming inclusions for SAE 1141 steel were evaluated satisfactorily. For an average Al content of 275 ppm for SAE 8620 steel, inclusions window with a minimum of 35% of CaO and a maximum of 5% of CaS is between 10.9 and 14.2 ppm of calcium. The curve of 35% of CaO from SAE 8620 steel liquid window practically coincides with the initial formation of the liquid phase. One can observe modification of alumina inclusions by calcium, forming varied calcium–aluminates for different contents of Al and Ca under SAE 8620 steel conditions. Curves of liquid windows normally starts with Al contents higher than 100 ppm (indicating Al-killed steels), because it is from this value, the calciumaluminates formation becomes predominant. As a general conclusion, a very good correlation was obtained between the thermodynamic predictions carried out in this study and the results obtained from the castability index of SAE 1141 and SAE 8620 steels produced by the continuous casting in the industrial plant. Thus, this research can contribute for the study of inclusions in other critical steels, as the employment of the methodology proposed here.

References [1] W. V. Bielefeldt, A.C.F. Vilela, Computational thermodynamic study of inclusions formation in the continuous casting of SAE 8620 steel, Steel Res. Int. 81 (2010) 1064– 1069. [2] W.V. Bielefeldt, A.C.F. Vilela, C.A.M. Moraes, P.C. Fernandes, Computational thermodynamics application on the calcium inclusion treatment of the SAE 8620 steel, Steel Res. Int. 78 (2007) 857–862. [3] W.V. Bielefeldt, A.C.F. Vilela, Thermodynamic study of nonmetallic inclusion formation in SAE 1141 steel, in: Proceedings of the First TMS-ABM Congress, 2010, Rio de Janeiro, p. 4474–4483.

3.2. CALPHAD assessments A. Zemanova, A. Kroupa, R. Mishra, H. Ipser, H. Flandorfer Thermodynamic assessment of the Ni–Sn–X (X ¼In, Sb) ternary systems The In–Ni–Sn and Ni–Sn–Sb alloys are potential candidates of the Pb-free solders, where Ni is used as substrate material in electronics. The aim of the present study is to develop the thermodynamic descriptions of the In–Ni–Sn and Ni–Sn–Sb ternary systems using the CALPHAD method. Phase equilibria in the ternary systems have been studied using X-ray powder diffraction, electron microprobe analysis, scanning electron microscopy and thermal analyses. All these experimental results and data from the literature were used for a CALPHADtype optimization of In–Ni–Sn and Ni–Sn–Sb ternary systems. The unary data were taken from new version of SGTE (4.4) database. The thermodynamic data for binary systems used in these assessments are taken from the SOLDERS thermodynamic database. The complete solubility was found between Ni3Sn2 and Ni2In phases in the In–Ni– Sn ternary system and between Ni3Sn2 and NiSb phases in the Ni–Sn– Sb system, which are of the Ni2In and NiAs type, respectively. Therefore, identical thermodynamic model has to be used for these phases, with respect to the existing experimental data and previous description of these phases within the SOLDERS thermodynamic database. Acknowledgements The authors are grateful to the Ministry of Education of the Czech Republic (Projects no. OC08053 and MEB 061005) for financial support. A. Kroupa, A. Zemanova, D. Legut The study of Z-phase in ternary Cr–Nb–N and Cr–V–N systems The presence of Z-phase – the Cr2(Nb,V)2N2 nitride – in modern Cr–Mo steels is of crucial importance and influences significantly the mechanical properties of such materials. Good thermodynamic description of such phases is very important for complete thermodynamic assessment of multicomponent systems and reliable prediction of properties of complex advanced materials. Therefore the experimental and theoretical study of Cr–Nb–N and Cr–V–N systems

144

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Fig. 1. Calculated Ba–Mg phase diagram by the present thermodynamic description and comparison with the experimental data: (a) entire phase diagram; (b) enlarged section.

is important to obtain more detailed information about these systems and, if possible, their proper thermodynamic description. The samples of above mentioned system were annealing for a long time to obtain states close to thermodynamic equilibrium and the experimental results were used in combined CALPHAD and ab-initio approach to describe above mentioned Cr–X–N systems, as they are crucial for Z-phase and plays important role in modeling of phase equilibria in advanced steels. Acknowledgements The authors are grateful to the National Science Foundation of the CR (Project no. 106/10/1908) and to Academy of Sciences of the Czech Republic no. AV0Z20410507. B. Lindahl, M. Selleby

Diffusion multiples for subsystems within the Fe–Al–C–Mn system Due to their outstanding mechanical properties, there is an increasing interest in TWIP (Twinning Induced Plasticity) steels and their low density, high aluminium counterpart L-IP (Lightweight steels with Induced Plasticity). The combination of high strength, high ductility and low density is of special interest in the automotive industry where the demand for inexpensive and environmentally friendly materials has increased drastically. The TWIP steels contain high amounts of manganese and the L-IP steels contain high amounts of aluminium as well. In the present work alloys with up to 40 mass% manganese and 15 mass% aluminium are being studied. The high interest in these steels gives rise to a demand for better knowledge of the thermodynamic properties of the relevant systems. Conventional steels only contain low amounts of Mn and Al therefore the existing databases are not adequate. In phase diagram determination, diffusion multiples are very handy for determining phase relations in systems. In this work re-assessments of some of the subsystems in the quaternary system Al–Fe–Mn–C were done using diffusion multiples and equilibrium annealing guided by simulations of diffusion paths. C. Wu, X. Su, J. Wang, Y. Liu, H. Peng, H. Tu First-principles calculations and thermodynamic assessment of the Te–Zr system

The Te–Zr binary system was thermodynamically assessed by the CALPHAD approach with the help of first-principles calculations. First-principles calculations of the energies of formation of seven Te–Zr compounds were performed using the Vienna abinitio simulation package (VASP). The enthalpies of formation of Te5Zr, Te3Zr, Te2Zr, Te6Zr5, TeZr, Te4Zr5 and TeZr3 are  51.7,  69.7, 88.5,  88.2,  87.6, 83.7 and  49.1 KJ/mol atom, respectively. The calculated enthalpy of formation of Te2Zr and Te4Zr5 agrees well with experimental data. The optimization of the Te–Zr system was performed by PARROT module using available experimental data from literature. The Gibbs free energy of the liquid was described by a sub-regular solution model with the Redlich–Kister equation. And all the Te–Zr binary compounds except Te2Zr are stoichiometric ones. The Te2Zr was described by two-sublattice model (Te)1(Te,Zr)2. The enthalpies of formation obtained by first principle calculation were used as input data in evaluating the Gibbs energy functions of those phases. The thermodynamic parameters were optimized to consistently reproduce the available experimental data with satisfactory agreement. Acknowledgements The financial support from Qinlan project and National Natural Science Foundation of China (nos. 50971110 and 50971111) are greatly acknowledged. X. Ren, C. Li, Z. Du, C. Guo, S. Chen

Thermodynamic modeling of the Ba–Mg binary system On the basis of the thermochemical and phase equilibrium experimental data, the phase diagram of the Ba–Mg binary system has been assessed by the CALPHAD technique. The liquid phase is of unlimited solubility and modeled as a solution phase using the Redlich–Kister equation. The intermetallic compounds, Mg17Ba2, Mg23Ba6 and Mg2Ba, with no solubility ranges are treated as strict stoichiometry compounds by the formula MgmBan. Two terminal phases, hcp-Mg and bcc-Ba, are kept as two pure element phases, since Mg and Ba do not dissolve in one another. After optimization, a set of selfconsistent thermodynamic parameters has been obtained. The calculated values agree well with the available experimental data Fig. 1.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Acknowledgments The authors would like to acknowledge National Natural Science Foundation of China (nos. 50731002 and 50671009) for the financial support. Thanks to Royal Institute of Technology and CompuTherm LLC for supplying Thermo-Calc and Pandat Software packages, respectively.

145

The three allotropic modifications of La, dhcp-La, fcc-La and bccLa, were kept as pure element phases since there is no solubility of Ge in La. A set of self-consistent thermodynamic parameters of the Ge–La binary system was obtained. The calculation results agree well with the available experimental data from literatures Figs. 1 and 2. Acknowledgements

M. Liu, C. Li, Z. Du, C. Guo, C. Niu Thermodynamic modeling of the Ge–La binary system The Ge–La binary system was critically assessed by means of the calculation of phase diagram (CALPHAD) technique. The liquid phase with unlimited solubilities of Ge and La into each other and the terminal solid solution diamond-(Ge) with a small solubility of La were described using the substitutional model, in which the excess Gibbs energy was formulated with the Redlich–Kister equation. The compounds with homogeneity ranges, aGe1.7La, bGe1.7La and GeLa, were modeled using two sublattices as a(Ge, La)1.7La, b(Ge, La)1.7La and (Ge, La)(Ge, La), respectively. The intermediate phases with no solubility ranges, Ge4La5, Ge3La4, Ge3La5 and GeLa3, were treated as stoichiometric compounds.

Fig. 1. The Ge–La binary phase diagram compared with experimental data.

Thanks to the Royal Institute of Technology Sweden for supplying the Thermo-Calc software. The authors would like to acknowledge National Natural Science Foundation of China (nos. 50731002 and 50671009) for the financial support. Thanks also to Dr. Viktor Kusnetsov from Lomonosov Moscow State University for literature collection.

D. Kapush, K. Korniyenko, V. Petyuch, T. Velikanova, V. Shemet, B. Grushko Experimental study of the Al–Ni–Pt alloy system Pt-modified nickel aluminide coatings are used in order to improve the high-temperature oxidation resistance of the components of the gas turbines produced from Ni-based superalloys. This explains an interest in the low-Al part of the Al–Ni–Pt system. On the other hand, the interest in the Al-rich part is because of the formation of complex periodic and quasiperiodic intermetallics attractive for both basic and applied research. In our work we present phase equilibria in the Al–Ni–Pt alloy system investigated at 1100 and 1300 1C as well as in the temperature range of crystallization in the whole compositional range and also at 1000, 900 and 790 1C from 50 to 100 at%Al using SEM/EDX, XRD and DTA. The liquidus and solidus surfaces, reaction scheme as well as isothermal sections at the aforesaid temperatures are constructed. 17 invariant four-phase reactions with participation of liquid were determined. The Al2Pt phase was found to extend up to  Al53Ni23Pt24 composition, separating the Al3Pt2 and Al3Ni2 phase fields. Isostructural AlNi and high-temperature AlPt phases are probably formed continuous region of solid solutions at elevated temperatures. Four ternary phases were revealed: two of them—in the Al-rich region, namely the w phase ( Al73Ni7Pt20, rhombohedral structure, a¼1.20953(25), c¼2.6932(4) nm, the Al28Ir9-type) and the e phase ( Al72Ni12Pt16, the e6-type orthorhombic structure, a¼2.34, b¼ 1.65, c¼1.24 nm). Apart from the ternary structure of the before known AuCu-type also its C-centered orthorhombic superstructure (Pt5Ga3-type structure) was revealed along  35 at%Al. Its lattice parameters for the Al33.5Ni7.2Pt59.2 composition are a¼0.79022(10), b¼0.72583(8), c¼0.39319(5) nm. At 1100 and 1300 1C it coexists in a wide compositional range with the ternary extension of the low-temperature AlPt and AlNi(Pt) b-phase. E.R. Pinatel, M. Palumbo, P. Rizzi, M. Baricco

Fig. 2. Calculated partial molar enthalpies of La in liquid at 1892–1905 K.

Thermodynamic modelling of the LaNi(5  x)AlxHy (0ox oy) The system LaNiH is well known for hydrogen storage applications. A dehydrogenation enthalpy ofabout 30 kJ/mol H2 allows the release of hydrogen close to room temperature and under 1 bar. This systemalone already shows good properties for hydrogen storage applications but Ni can be partially substituted with other elements in order to tune the adsorption–desorption behaviour and to improve cyclability [1]. Al is one of the best substituents because it shows a solubility up to LaNi3.5Al1.5, it is able to significantly reduce the disproportionation/decomposition of the hydride and to lower the plateau pressure [2,3]. For these

146

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

reasons, the thermodynamic behaviour of the Ni–Al substitution in LaNi5 is of particular interest. The aim of this work is to extend the CALPHAD description of the La–Ni–H system [4] in order to model the LaNi(5  x)AlxHy solid solution. The system was investigated experimentally by means of PCI (pressure composition isotherm) for several temperatures (25, 40, 60, 80 and 100 1C) and for different Al contents (x ¼0, 0.2, 0.4, 0.6 and 0.8).Samples were prepared by arc melting and checked by X-ray powder diffraction, SEM (scanning electron microscope) and EDS (energy dispersive X-ray spectroscopy) analysis. According to structural information [2], a 3 sublattices model (La)(Ni,Al)5(H,Va)7 has been chosen to describe the solid solution. First principles calculations using density functional theory (DFT) were performed to estimate the formation energies of the end-members (LaAl5 and LaAl5H7). The calculations employed the generalized gradient approximation of Perdew–Burke–Ernzerhof (PBE) as implemented in the VASP code with PWS pseudopotentials. The free energy of the solid solution as a function of composition and temperature was assessed from data available from experiments, first principles calculations and literature using the CALPHAD approach. The assessed thermodynamic functions for this phase show a good agreement with available experimental data.

References

temperature in air atmosphere. The samples were heated to and held at 1873 K for 24 h in order to achieve the equilibrium state. Thereafter the samples were quenched in air. The chromium distribution and phase compositions were studied using SEM, EDS and XRD techniques. FACTsage software was used for the phase equilibrium calculations. The experimental results obtained from the present work were compared with the calculation results from FACTSage software. It is found that the spinel formation at 1873 K is favored in the slag basicity range of 1.2–1.6. G. Huang, H. Qi, L. Liu, Z. Jin Thermodynamic assessment of Ca–Cu–Mg sytem Bulk metallic glasses (BMGs) research has been a hotspot since the metal-metal system such as Ca-, La-, Mg-, Zr-based alloys were prepared by stabilization of supercooled liquid. And the glass formation ability (GFA) is vital to develop new BMGs. Thus the Scientific efforts have been done to assess GFA. One of the successful approaches is using CALPHAD method to calculate the onset driving force (ODF) for crystalline phases from supercooled liquid. The purpose of this paper is to bulid up the thermodynamic database of Ca–Cu–Mg and predicts the glass formation composition range. All the binary parameters come from our group’s Magnesium alloy database and the ternary parameters derived from experiments of

[1] H.H. Van Mal, K.H.J. Buschow, A.R. Miedema, J. Less-Common Met. 35 (1) (1974) 65. [2] H. Diaz, A. Percheron-Gu{gan, J.C. Achard, C. Chatillon, J.C. Mathieu, Int. J. Hydrogen Energy 4 (5) (1979) 445. [3] Y. Hashimoto, K. Asano, Y. Iijima, Mater. Trans. 43 11 (2002) 2696. [4] M. Palumbo, J. Urgnani, D. Baldissin, L. Battezzati, M. Baricco, Calphad 33 (1) (2009) 162.

¨ G. Jelkina, L. Teng, B. Bjorkman Effect of Basicity on the chromium partition in CaO–MgO– SiO2–Cr2O3 synthetic slag at 1873 K The objective of the present work is to get an understanding of the phase relationships in the CaOMgO–SiO2–Cr2O3 slags with a view to control the precipitation of Cr-spinel in the slag phase. It was reported that Cr in the spinel phase has lower leaching levels than in some silicate phases. The equilibrium phases in CaO–MgO–SiO2–Cr2O3 slag system at 1873 K have been studied. The Cr2O3 and MgO contents in the slag were fixed to 6 and 8 wt%, respectively. The basicity (CaO/ SiO2) of slag was changed in the range of 1.0–2.0. Gas/slag equilibrium technique was adopted to synthesize the slag at high

Fig. 2. Calculated the isothermal section at 623 K.

Fig. 1. Calculated the mixing enthalpy of liquid with experiments.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

147

compounds optimized in every system were calculated using the MIVM. The Fe–Ni and Fe–Co phase diagrams and some correlated thermodynamic properties were calculated. The results are in agreement with the experimental data and then indicate that the model is reliable and convenient. References [1] D.P. Tao, Thermochim. Acta 363 (2000) 105–113. [2] D.P. Tao, B. Yang, D.F. Li, Fluid Phase Equilib. 193 (2002) 167–177. [3] H.W. Yang, D.P. Tao, Metall. Mater. Trans. A 39 (2008) 945–949. [4] H.W. Yang, B.Q. Xu, B. Yang, W. Ma, D.P. Tao, Fluid Phase Equilibr. 314 (2012) 78–81. Fig. 3. Calculated the iso-ODF lines for the whole composition.

Myles [1] and Sommer [2]. Figs. 1 and 2 show that the calculated results are agree with the experiments. Therefor the built up thermodynamic data is reasonable to calculate the ODF. The glass formation of Ca–Cu–Mg had been studied by many authors [3–6]. The iso-ODF of Ca–Cu–Mg is shown in Fig. 3 comparing with the range of complete glass formation [3]. It indicates that the low ODF is the most important factor to form the glass metal. References [1] K.M. Myles, The ternary system copper–magnesium–calcium, J. Less-Common Met. 20 (2) (1970) 149–154. [2] F. Sommer, H.G. Krull, J.J. Lee, Calorimetric studies of liquid Ca–Cu–Mg alloys, J. Less-Common Met. 169 (2) (1991) 361–368. [3] F. Sommer, W. Vogelbein, B. Pre del, Glass formation of ternary Ca–Cu–Mg alloys, J. Non-Cryst. Solids 51 (3) (1982) 333–343. [4] U. Mizutani, et al., Electronic structure and electron transport in Ca–Mg–Cu metallic glasses, Mater. Sci. Eng. 99 (1–2) (1988) 295–299. [5] K.J. Laws, et al., Synthesis of copper-based bulk metallic glasses in the ternary Cu–Mg–Ca system, J. Alloys Compd. 486 (1–2) (2009) L27–L29. [6] K. Laws, et al., Prediction of glass-forming compositions in metallic systems: copper-based bulk metallic glasses in the Cu–Mg–Ca system, Metall. Mater. Trans. A 41(7) (2010) 1699–1705. H. Yang, Y. Wang, D. Tao

Application of the molecular interaction volume model to phase diagram calculations of Fe–Ni and Fe–Co systems We calculate the phase diagrams of Fe–Ni and Fe–Co systems by means of a combination of the molecular interaction volume model (MIVM) calculations and the CALPHAD approach. The molecular interaction volume model (MIVM) was obtained on a physical basis. It is a two-parameter model and is able to predict component activities and enthalpies in liquid alloys and solid solutions at the required temperatures. A significant advantage of the MIVM lies in its ability to calculate thermodynamic properties of multicomponents alloys using only the binary infinite dilute activity coefficients. It, therefore, avoids empirical fitting for the pair potential energy interaction parameters. Using the obtained thermodynamic data of every phase in the systems, the Gibbs energy of solution phases and

K.C. Hari Kumar, S. Ghosh, B. Prabhakara Reddy, K. Nagarajan, P.R. Vasudeva Rao Thermodynamic assessment of LiCl–UCl3 and KCl–UCl3 binary systems by coupling CALPHAD and first principles calculations We report on the thermodynamic assessment of LiCl–UCl3 and KCl–UCl3 carried out by combining CALPHAD and ab initio methods. The liquid and solid solutions were modelled by ionic two-sublattice model. The enthalpy of mixing in these systems was calculated using the surrounded-ion model for asymmetric molten salt systems [1–4] and were found to be within 78%. The enthalpy of formation of K2UCl5 which is a congruently melting compound in KCl–UCl3 system was calculated using a full potential linearized augmented plane waveþlocal orbital method as implemented in Wien2k [5] using generalized gradient approximation within the framework of density functional theory. The spin orbit coupling was included by applying the second variational procedure for the scalar relativistic eigenstates, for which the basis was extended by relativistic 6p1/2 local orbitals centered at U atom for all spin orbit calculations. The number of k-points in the irreducible Brillouin zone and RKmax were taken to be 1248 and 7.00, respectively, where R is the smallest muffin tin radius taken to be 2.5 and Kmax is the largest k-vector in the reciprocal lattice used in the plane wave basis set expansion. The number of k-points and Kmax were sufficient enough to achieve a self consistent total energy value with an energy convergence criteria of 0.01 mRy and also until the difference in the charge between the successive iterations was less than 0.0001 electrons. Since the molecular species could not be directly treated. References [1] M. Gaune-Escard, J.C. Mathieu, P. Desre, Y. Doucet, J. Chim. Phys. 6 (1973) 1003. [2] M. Gaune-Escard, J.C. Mathieu, P. Desre, Y. Doucet, J. Chim. Phys. 7–8 (1973) 1033. [3] G. Hatem, M. Gaune-Escard, J. Chim. Phys. 4 (1975) 463. [4] G. Hatem, M. Gaune-Escard, J. Chim. Phys. 77 (1980) 925. [5] P. Blaha, K. Schwarz, G.K.H. Madsen, D. Kvasnicka, J. Luitz, WIEN2k, An Augmented Plane WaveþLocal Orbitals Program for Calculating Crystal Properties, 2001, ISBN-3-9501031-1-2. L. Liu, L. Zhang, H. Bo, Z. Jin Investigation of the phase diagrams and microstructure evolution of Al–Cu–RE alloys Interest in aluminum alloys is increasing continuously due to their properties such as the low density, high strength, high

148

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

corrosion-resistance as well as their potential applications in the automotive and aerospace industries. Microalloying has been used to improve the high temperature strength of the selected age hardening aluminum alloys. Rare earth elements are benefit to casting process in the conventional aluminum alloys resulted in improving tensile strength, heat resistance and corrosion resistance, etc. for many years. The invesitagiation on the glass forming ability (GFA), especially on the thermodynamic modeling of amorphous forming system, is an important issue on the amorphous alloys research. The key to the thermodynamic modeling of GFA of alloys is the reliable binary and ternary phase diagram and thermodyniac properties. Associated with a project entited ‘‘Investigation of phase diagram and microalloying of Al–Cu based alloy systems’’ supported by National Science Foundation of China (no. 50771106), the Cu–RE (RE¼Y, Nd, Gd and Dy) binary systems and Al–Cu–X (X¼Dy, Er, Gd, Nd, Sc, Ti, Y, Yb, Zr) systems have been investigated by means of experimental determination and CALPHAD method. The experimental result showed that Cu7RE phase forms in the Cu-rich region of the Cu–Y and Cu–Dy binary systems, and in Cu–Y and Cu–Dy binary systems the Cu-rich compounds are Cu6RE phases. It was shown that Cu7Dy phase was only stabilized at high temperature, this phase was decomposed to Fcc(Cu)þCu5Dy_L in low temperature. A ternary compounds(t1-Al8Cu4Er) and six ternary compounds (t1-Al8Cu4Nd, t2-Al9Cu8Nd2, t3-Al6Cu7Nd, t4-Al2.4Cu8.6Nd, t5-Al3CuNd and t6-AlCuNd) were determined in the Al rich part of the isothermal 673 K section of the Al–Cu–Er and Al–Cu–Nd systems. Based on the experimental phase relations and literature results, the thermodynamic parameters of the Cu–Gd, Cu–Dy, and Cu–Er binary systems and the Al–Cu–X (X¼Dy, Er, Gd, Nd, Sb, Sc, Ti, Y, Yb, Zr) ternary systems have been obtained. In order to verify the quality of the thermodynamic parameters, the electric-arc melted cast-state sample of Al–Cu–MM (MM¼Y, Nd, Gd, Dy, Er and Ti) were prepared. The equilibrium solidification and the solidification according to the Scheil model were calculated for the alloys. The calculated results were successfully applied to understand the experimental microstructures of the above mentioned alloys in Al–Cu–MM (MM¼Y, Nd, Gd, Dy, Er and Ti) system. Based on the thermodynamic database, the relationship between the GFA and the driving forces of crystalline phases was investigated. It was revealed that the alloys with good glass forming ability in the Al–Cu–RE system have the low driving forces of the crystalline phases. These minimum driving forces are the criterion for selecting the composition for formation of Al–Cu–MM amorphous alloys. L. Jin, Y.-B. Kang, P. Chartrand, A. Gheribi, D. Kevorkov, M. Medraj Modeling of thermodynamic properties and phase equilibria in Mg–Al–Mischmetal Systems Magnesium–aluminum-based alloys are widely used because of their low density, high strength-to-weight ratio, specific rigidity, satisfactory salt-spray corrosion resistance and good ductility. Adding mischmetal to Mg–Al alloys, which is a mixture of rare earth (RE) elements with typical composition ranges of La (25–34%), Ce (48–55%), Pr (4–7%) and Nd (11–17%) [1], may improve the creep resistance [2] and strength at elevated temperatures due to the precipitation of the intermetallic phases (like Al11RE3 and Al2RE phases), and suppression of probably detrimental Mg17Al12 phase in the interdendritic or grain boundary region. The aim of current study is to optimize or predict the thermodynamic properties and phase diagrams of Al–RE (La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu), Mg–Al–RE (La, Ce, Pr, Nd, Sm, Gd, Tb, Dy, Ho and Er) systems and Al–RE0 –RE00 and to build a complete thermodynamic atabase concerning Mg, Al and RE elements, which will guide the magnesium and aluminum alloys design. Optimized model parameters of the

Gibbs energies for all phases in these systems have been obtained. The optimization procedure is biased by putting a strong emphasis on the observed trends in the thermodynamic properties of Al–RE, Al–Mg–RE, Al–RE0 –RE00 systems, like enthalpy of formation, enthalpy of mixing. The modified quasichemical model [3], which takes short-range ordering into account, is used for the liquid and the compound energy formalism is used for the solid solutions in the binary systems. Equilibrated key alloys experiments are carried out for Mg–Al–(Ce, La, Nd) systems in order to validate the modeling parameters. Furthermore, first-principles calculations using VASP software are performed for the partial enthalpies of mixing in the Al–(Ce, La, Pr, Nd) and Mg–(Ce, La, Pr, Nd) binary systems and the modified Miedema’s model, are used to calculated the enthalpies of formation of the binary ternary intermetallic compounds. References [1] J. Gnobner, R. Schmid-Fetzer, Thermodynamic modeling of the Mg–Ce–Gd–Y system, Scr. Mater. 63 (2010) 674–679. [2] G. Pettersen, et al., Microstructure of a pressure die cast magnesium–4 wt% aluminium alloy modified with rare earth additions, Mater. Sci. Eng. A 207 (1996) 115–120. [3] A.D. Pelton, et al., The modified quasichemical model I-binary solutions, Metall. Mater. Trans. B 31B (2000) 651–659. ¨ L. Eleno, M.G.V. Cuppari, C.G. Schon Experimental investigation and thermodynamic modeling of the Fe–Cr–Mo–C system A new assessment of the Fe–Cr–Mo–C quaternary system was performed, mainly to take into account new experimental information obtained recently by our group. Several alloys, with different compositions, were induction-melted under vacuum and later homogenized at 1170 1C for 72 h. The alloy compositions were determined by setting the Chromium content at 5 wt% and varying the alloys were then encapsulated in quartz and equilibrated at 600, 700 and 900 1C. The thermodynamic optimization consisted in altering the parameters of the carbides from the TCFE database, especially the M23C6, while keeping all other parameters. A recent model for the cementite [1] was also incorporated into the assessment. The results show a tendency to increase the stability of the M23C6 carbide over the M7C3, in agreement with the experimental information. A better overall agreement also resulted from the optimization. References [1] B. Hallstedt, et al., Calphad 34 (2010) 129. M.F. Campos, A.B. Farina Thermodynamic modeling of the binary system Co–Sm In this study, it is described the Thermodynamic modeling of CoSm binary phase diagram using the CALPHAD (calculation of phase diagrams) methodology. A detailed thermodynamic description of binary system Co–Sm was obtained, leading to a thermodynamic database compatible with experimental data from literature. Reliable Curie temperature for the phases Co2Sm, Co3Sm, Co7Sm2 ,Co19Sm5, Co5Sm, Co17Sm2 [1] provided consistent parameters for the calculated phase diagram. All phases were considered as line compounds, except the phases Co5Sm and

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Co17Sm2. The sub-lattice model was employed for the thermodynamic description of all phases. The solubility of the Co17Sm2 phase was successfully modeled, obtaining values coherent with experimental data from literature. The model assumes that Sm replaces pair of Co atoms into Co17Sm2 structure. A model for the description of the solubility of Co5Sm phase has also been proposed. Taking into account experimental data [1], it is assumed that the solubility of Sm into SmCo5 is negligible, and that dumbbell substitution of Sm atoms by pairs of Co atoms account for the solubility of Co into SmCo5. The enthalpy of formation of the phase Co5Sm has been calculated:  8609 J/mol. This value is in agreement with published experimental data [2]. A new model for the liquid phase was adopted in order to improve the description of recent experimental data. References [1] M.F. de Campos, F.J.G. Landgraf, J. Phase Equilib. 21 (5) (2000) 443–446. [2] F. Meyer-Liautaud, C.H. Allibert, R. Castanet, J. Less-Common Met. 127 (1987) 243–250. A. Mascaro, J.-M. Joubert, C. Toffolon-Masclet, C. Raepsaet Experimental study and thermodynamic assessment of the ER–H binary system In order to increase cycle length and fuel burn up of pressurized water reactors (PWR), new fuel claddings are studied. In this framework, a new concept is being developed by the CEA: a cladding constituted by a liner of zirconium–erbium alloy between two liners of industrial zirconium alloys [1,2]. Indeed, erbium, thanks to its high neutron absorption cross section, can be used as a solid burnable poison for controlling the nuclear reaction. This kind of configuration provides a more homogeneous power distribution within the reactor core and liberates some extra space for more fuel. Each cladding is surrounded by a pressurized water environment. Furthermore, some hydrogen released from the dissociation of water molecules diffuses inside the material. Thus, it is of high interest to have a good knowledge of the Er–H–Zr ternary system. In this study, the Er–H system, one of the 3 constitutive binary systems has been studied experimentally and modelled by the Calphad method. The samples have been synthesised by hydrogenation using the Sieverts method. They have been analysed by X-ray diffraction and to quantify and locate the hydrogen, the nuclear microprobe has been used with PIXEþERDA combination (particule induced X-rays emission and elastic recoil detection analysis). Several pressure-composition isotherms have been measured, complementing the data available in the literature. This system has been modelled using the CALPHAD method. Due to the existence of homogeneity domains, the solid phases have been modelled with the sub-lattice model. Formation enthalpies of stable and metastable phases have been calculated ab-initio [3] and incorporated in the model. A set of parameters describing the Gibbs energies of the different phases is given and reproduces well the available experimental data. References [1] J.-C. Brachet and colleagues, French CEA Patent: BD 1725, 2006. [2] J.-C. Brachet and colleagues, Proceedings of the Top Fuel 2009, paper 2189, Paris (France) September 6–10, 2009. [3] J.-M. Joubert, A. Mascaro, J.-C. Crivello, Toffolon-Masclet, Ab-initio Calculations of Er–H Compounds: Application to the Modelling of the Phase Diagram.

149

M. El Guendouzi, R. Azougen, S.M. Aboufaris E., I. Hamdouny Prediction of solubility from the water activities in the mixed aqueous solutions YCl2þ Y0 Cl2þ  H2O with Y0 ; Y Mg2 þ ; Ca2 þ ; or Ba2 þ The thermodynamic computing models used for the prediction of behavior solutions solid–liquid–gas equilibria close to experimental accuracy have wide applicability. They can simulate the complex changes that occur in nature and can replicate conditions that are difficult to duplicate in the laboratory. Such models can be powerful predictive and interpretive tools to study the geochemistry of natural waters and mineral deposits, solve environmental problems, and optimize industrial processes. The specific interaction approach for describing electrolyte solutions to high concentrations introduced by Pitzer, represents significant advance in physical chemistry that has facilitated the construction of accurate thermodynamic models. These theory models of electrolyte solution have been developed by others authors, which reliably predict mineral solubility in the complex brine system. The main objective of this study is the development of a thermodynamic model for solution behavior and solid–liquid equilibria. In this investigation, the mixed aqueous solutions of chloride with alkaline earth metal {yYCl2þ (1y) Y0 Cl2} (aq) with Y; Y0 RMg2 þ ; Ca2þ ; and Ba2 þ have been studied using the hygrometric method at 298.15 K. The water activities are measured at total molalities from 0.20 mol/kg to saturation for different ionic-allow the deduction of osmotic coefficients. From these measurements, the mixing ionic parameters yYY0 and cY0 YCl are determined and used to predict the solute activity coefficients in the mixtures. The values of the pure electrolyte, mixing ionic-interaction parameters, which give the best fit of the activity data in binary, ternary solutions and solubility data. Thermodynamic characteristics, solubility products Kosp and the standard molar Gibbs energy of formation DGofm of the crystallizing solids are determined. These systems magnesium, barium and calcium chloride minerals are the higher solubility minerals of the complex sea-salts and brine. Solid sea salt particles take up water in the atmosphere and form aqueous droplets. The prediction for electrolytic systems is the prerequisite before applying the software to the pilot and industrial processes, such as the wet-process phosphoric acid production at the industrial scale. References [1] K.S. Pitzer, Thermodynamics of electrolytes. I. Theoretical basis and general equations, J. Phys. Chem. 77 (1973) 268–277. [2] C. Harvie, N. Moller, J. Weare, The prediction of mineral solubilities in natural waters: the Na–K–Mg–Ca–H–Cl– SO4–OHHCO3–CO3–CO2–H2O system from zero to high concentration at 25 1C, Geochim. Cosmochim. Acta 48 (1984) 723–751. [3] R. Pabalan, K.S. Pitzer, Thermodynamics of concentrated electrolyte mixtures and the prediction of mineral solubilities to high temperatures for mixtures in the system Na–K–Mg–Cl–SO4–OH–H2O, Geochim. Cosmochim. Acta 51 (1987) 2429–2443. [4] C. Christov, N. Moller, Chemical equilibrium model of solution behaviour and solubility in the H–Na–K–OH–Cl–HSO4–SO4– H2O system to high concentration and temperature, Geochim. Cosmochim. Acta 68 (2004) 1309–1331. [5] M. El Guendouzi, A. Errougui, Solubility in the ternary aqueous systems containing M, Cl  , NO3  -, and SO4.2  with M¼NH4 þ , Li þ , or Mg2 þ at T¼298.15 K, J. Chem. Eng. Data 54 (2009) 376–381. [6] R. Azougen, M. El Guendouzi, A. Rifai, J. Faridi, Water activities, activity coefficients and solubility in the binary

150

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

and ternary aqueous solutions LiCl þ YCl2 þ  H2O with YRMg2 þ ; Ca2 þ ; or Ba2 þ , Calphad 34 (2010) 36–44. [7] M. Azaroual, C. Kervevan, A. Lassin, L. Andre´, M. Amalhay, M. El Guendouzi, Modeling and analysis of the thermodynamic conditions of mineral fouling in the wet-process phosphoric acid production (25oTo100 1C), in: Proceedings of the COVAPHOS III, Marrakech, Morocco March 18–20, 2009. S. Lin, D. Morgan Coupling compound energy formalism CALPHAD modeling with first principle calculation for ionic materials Phases with homogeneity ranges are conventionally well described by the calculation of phase diagram (CALPHAD) approach using the compound energy formalism (CEF). However, in ionic materials, the wide composition ranges are usually coupled with variety of valance states of constituent elements, which results in numerous compositional end-members in the CEF model. Moreover, most of these end-members are not charge neutral, ensuring that no direct measurements of these compounds can be obtained. In practice, determining the large number of model parameters in the CEF from typical experimental data sets requires many, often poorly justified, approximations. First principle calculations offer a way to reduce the approximations used in applying the CEF to complex ionic systems. Here we discuss approaches to treat the problems of modeling complex charged end-members to enable energetics from first principle calculation to be applied in the CALPHAD CEF in ionic systems. Initial results for La1 yMnO37 d perovskite, a base oxide for many important cathode materials in solid oxide fuel cells and other applications, will be presented. M.I. Ivanov, V.V. Berezutsky, M.O. Shevchenko, V.G. Kudin, V.S. Sudavtsova Thermodynamic properties of the liquid alloys of the Al–Ln binary systems The thermochemical properties of alloys of the binary Al–Y(La, Eu, Yb) systems were determined using the calorimetry method. Integral mixing enthalpies of alloys of the Al–Y(La) systems were defined to reach a minimum  41 kJ/mol at xY¼0.4 and xLa¼0.36, and Al–Eu(Yb) ones— 23 kJ/mol at xEu(Yb)¼0.39. The activities of components of the Al–Y(Sc, La, Ce, Nd, Eu, Yb) alloys were calculated using the co-ordinates of the liquidus lines of these systems and the theory of ideal associated solutions (TIAS).

reaches a components at 1800 K show essential positive deviations from the Raoult’s law, which are not enough, however, for a miscibility gap appearance. Integral mixing Gibbs energy is close to  3.5 kJ/ mole at entropy can be considered about zero. Estimation of thermodynamic properties of liquid alloys in other Ti–R systems has shown that the integral mixing enthalpies of the melts are positive with components also show large positive deviations from the Raoult’s law. Regular behavior of thermodynamic properties of liquid Ti–R melts is accounted for by a difference of atomic radii of the components (dimensional factor). V. Iina, T. Pekka Thermodynamic assessments of the [Co,Cr,Fe,Ni]–Pb binary systems Phase diagrams are basic information with high practical use, e.g., in high temperature metal processing. With CALPHADmethod it is possible to assess critically the phase diagrams and build internally consistent databases with thermodynamic data of specific systems, while reducing expensive and time-consuming laboratory work. In this work the Pb-binaries of the system Co– Cr–Fe–Ni–Pb have been assessed with CALPHAD-method using experimental data from the literature and from the isothermal equilibration experiments reported in this study. The phase diagram and the excess Gibbs energy values of the solution phases, molten alloy and the solid solutions were calculated using the Redlich–Kister polynomials. The data was fitted by a least squares method using MTDATA software tool. The isothermal equilibrationexperiments were carried out in evacuated, argon filled and sealed quartz ampoules in a chamber furnace at about 1273–1573 K. A neck was made in the middle of the ampoule according to Fig. 1 in order to separate the metal saturated Pb melt and Pb saturated metal foil after equilibration by turning the ampoule 1801 before quenching the ampoule into ice water. After quenching the ampoules were broke down and the compositions of saturated metals were analyzed with ICP and EPMA. The aim behind this paper is to build a database of lead based alloys as a part of the NPL-MIRO MTOX oxide database. The binaries are thus assessed or extrapolated to ternaries and higher order systems.

M.O. Shevchenko, Yu.V. Fartushna, V.G. Kudin, V.S. Sudavtsova, M.V. Bulanova Thermodynamic properties of liquid Ti–Dy alloys Titanium alloys with rare earths (R) are interesting from both fundamental and practical points of view. Titanium affinity to the rare earth elements is low. As a result, Ti–R systems can be divided into two groups. The systems of the first group (R—from La to Gd and Y) show miscibility gap in a liquid, decreasing along the row. The systems of the second group (R—from Tb to Lu) are of a simple eutectic type. Both titanium and rare earths are chemically aggressive elements, so experimental study of Ti–R based systems is rather difficult. Especially this concerns thermodynamic properties of liquid alloys, which are unknown. Therefore, the goal of this work is to estimate thermodynamic properties of liquid Ti–Dy alloys on the basis of a phase diagram constructed by us, and using a methodology proposed by us. Resulting integral mixing enthalpy of Ti–Dy melts is positive and

Fig. 1. The quartz ampoule and the position of the samples inside the ampoule (e.g., Fe–Pb binary system).

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

151

Fig. 1. The effect of Al2O3 content on low melting area of CaO–SiO2–Al2O3–MnO system based on FactSage software.

F. Wang, H. Wang, C. Li Control on low melting point area in CaO–SiO2–Al2O3–MnO system The composition area of low melting point inclusions in CaO– SiO2–Al2O3–MnO system was calculated and analyzed by using thermodynamic software FactSage. The results show that the composition areas of low melting point inclusions increase at first, then decrease with alumina and calcium oxide contents increasing, respectively, as shown in Fig. 1. However, it always increases with silica and manganese oxide contents rising, respectively. To obtain low melting point inclusions, alumina and silica contents should be approximately controlled to be 20%, 30%, respectively, the CaO content restricted to be 25–30%, and calcium oxide content/silica content should be 0.8–1. Then iso-activity lines were drawn in CaO–SiO2–20%Al2O3–MnO diagram. Based on above data, the inclusions with good plasticity can be obtained in cord wire steel. Acknowledgements Thanks to Thermfact and GTT-Technologies for the support of the FactSage software. The authors would like to acknowledge National Natural Science Foundation of China (no. 50874007). X.Fang, K. Li, D. Zhao, Y. Du Precipitation sequence of Al–Mg–Si–(Cu) alloys and its verification by first-principles calculations The whole precipitation sequences of two cast Al–Mg–Si alloys (containing no Cu or minor Cu) were investigated via X-ray diffraction, transmission electron microscopy and hardness examinations. The precipitation sequence of the Cu-free alloy can be expressed as: super-saturated solid solution (SSSS)-G.P. zonespre-bv-b00 -U2 þ Si þ b-Si þ b, while that of the Cu-containing alloy is identified as: SSSS-G.P. zones-pre-b00 -b00 -Q0 þ b þ SiQþ b þSi. The addition of Cu accelerates the microstructural evolution, induces the formation of Q0 and Q precipitates, and decreases the stability of b00 precipitates. The finite-temperature thermodynamic properties and stability of the metatable phases in

Fig. 1. The isothermal section of the Tb–Fe–Cr ternary system at 600 1C.

the sequence have been computed by means of first-principles calculations, and the calculated results are consistent with the experimental observed sequence. Acknowledgements The financial support from the National Natural Science Foundation of China (Grant nos. 50721003 and 50831007) is greatly acknowledged. Y. Zhong, H.-Y. Zhou, C.-H. Hu, S.-K. Pan

Phase relationship in the Tb–Fe–Cr ternary system at 600 1C The phase relationships in the Tb–Fe–Cr ternary system at 600 1C have been investigated mainly by X-ray diffraction analysis and first-principles density functional theory calculations. The

152

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

result shows that the isothermal section consists of 9 singlephase regions, 16 two-phase regions and 8 three-phase regions. The compound TbFe12–xCrx with the ThMn-type structure is found to have a broad solubility ranging from x ¼1.4 to x¼2.6. The XRD patterns and total-energy calculations suggest that Tb2Fe17 crystallizes with theTh2Zn17-type rhombohedral structure under our experiment conditions. The solid solutions in this ternary system are formed by the substitution of Cr for Fe. The maximum solid solubilities of Cr in (Fe), Tb2Fe17, Tb6Fe23 and TbFe2 are about 17, 12.5, 3 and 7 at% Cr, respectively (Fig. 1). Acknowledgements Y. Zhong and H.Y. Zhou are thankful for the financial support from the National Natural Science Foundation of China under Grant nos. 50631040, 50961005 and 50901023. Y. Peng, Y. Du, H. Xu, P. Wang, C. Sha Experimental investigation and thermodynamic calculation of the Mg–Ni–Zn System The isothermal section of the Mg–Ni–Zn system at 400 1C was determined by using 19 alloys prepared with powder metallurgy method. The samples, which were annealed at 400 1C for 40 days, were examined by means of optical microscopy, X-ray diffraction, and scanning electron microscopy with energy dispersive X-ray spectroscopy. The experimental results confirm the existence of a ternary compound t reported in the literature. The composition range of this compound was determined to span from about 20 to 60 at% Zn. Thermodynamic modeling of the Mg–Ni–Zn system was performed, and a set of self-consistent thermodynamic parameters in the Mg–Ni–Zn system was obtained by using the CALPHAD approach taking into account the present experimental data and the literature data. Good agreement between the calculated and measured phase diagrams indicates that the present thermodynamic description of this system is reasonable. Due to the high volatility of Zn, it is difficult to obtain the experimental information near the Zn-rich region. Future experiment is needed for this region Fig. 1.

Acknowledgements The financial support from the National Natural Science Foundation of China (Grant nos. 50971135, 50721003 and 50831007) is greatly acknowledged. L. Chen, Y. Du, K. Chang, B. Yang, W. Pan, Y. Peng Microstructure and mechanical properties of TiAlN/TiN nanomultilayer coating Multilayer Ti–Al–N coatings are used for various applications where hard, wear and oxidation resistant materials are needed. Here, we prepare TiAlN/TiN nano-multilayer coatings with modulation period of less than 20 nm in order to further improve the mechanical properties of Ti–Al–N coatings. Multilayer structure results in an increase in adhesion with substrates from 72 N for Ti–Al–N coating to 98 N for TiAlN/TiN nano-multilayer. In addition, the interfaces of TiAlN/TiN nano-multilayer coating retard the outward diffusion of metal atoms (Al and Ti) and inward diffusion of O while exposing coating in air atmosphere at elevated temperatures, and thus improve its oxidation resistance. An improved machining performance in both continuous cutting and milling is obtained by TiAlN/TiN nano-multilayer coated inserts, which can be attributed to the combined effects of higher adhesion with substrates and better oxidation resistance. Acknowledgements The financial support from National Natural Science Foundation for Youth of China (Grant no. 51001120) and the postdoctoral foundation of China (Grant no. 20100470060) is greatly acknowledged. Z. Du, C. Guo, M. Li, C. Li A thermodynamic modeling of the Pd  Ti system The phase equilibria and thermodynamic properties of the PdTi system were optimized using the CALPHAD (CALculation of PHAse Diagram) technique. The solution phases, liquid, bcc, fcc, and hcp were modeled with the substitutional-solution model. Both the compounds Pd2Ti and PdTi2 having a tetragonal MoSi2-type structure were treated as one phase with the formula PdTi(Pd, Ti) by a threesublattice model with Pd on the first sublattice, Ti on the second and Pd and Ti on the third one, respectively. The intermetallic compounds Pd3Ti, Pd3Ti2, and PdTi3 were treated as stoichiometric compounds. The intermetallic compound aPdTi, which had a homogeneity range, was treated as the formula (Pd,Ti)(Pd,Ti) by a two-sublattice model. A two-bPdTi in order to cope with the second-order transition between bPdTi with CsCl-type structure (B2) and body-centered cubic solution (A2) in the Pd Ti system. A set of self-consistent thermodynamic parameters was obtained. This work was supported by National Natural Science Foundation of China (NSFC) (Grant nos. 50971027 and 50934011). Z. Du, D. Lu, C. Guo, C. Li

Fig. 1. Calculated isothermal section at 400 1C along with the present experimental data.

Thermodynamic description of the Mg–Sn–Y system The Sn–Y and Mg–Sn–Y systems were assessed by means of the CALPHAD technique on the basis of available experimental information. In the Sn–Y binary system, the solution phases (liquid, bcc, bct and hcp) are described by the substitutional solution model, Sn3Y5 intermetallic compound, which has a homogeneity range, was treated by a three-sublattice model as the formula (Sn,Y)3(Y,Sn)2Y3 in accordance with the site occupancies, and

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

153

the others compounds, Sn4Y5, Sn10Y11, Sn2Y, Sn5Y2 and Sn3Y were treated as stoichiometric compounds. Combined with the thermodynamic parameters of the Mg–Sn and Mg–Y binary systems in literature, the thermodynamic modeling of the Mg–Sn–Y ternary system was performed. The liquid phase is described by the associated solution model with an associate Mg2Sn, while the solid solution phases, bcc, hcp, are modeled by the substitutional solution model. The compound Sn3Y5 in the Sn–Y system was treated as (Sn,Y,Mg)3(Y,Sn,Mg)2Y3 in the Mg–Sn–Y system. The MgSnY ternary compound was treated as stoichiometric compound on the basis of the experimental information. A set of self-consistent thermodynamic parameters of the Mg–Sn–Y system was obtained. This work was supported by National Natural Science Foundation of China (NSFC) (Grant nos. 50731002, 50801004). J. Liu, C.Guo, C. Li, Z. Du Thermodynamic re-assessment of the Ni–Sn system The Ni–Sn system was critically re-assessed by means of the CALPHAD technique. The excess Gibbs energy of the solution phases (liquid, fcc and bct) were modeled with the Redlich–Kister equation. Five intermetallic compounds, Ni3Sn_HT (which is stable at high temperature), Ni3Sn_LT (which is stable at low temperature), Ni3Sn2_HT, Ni3Sn2_LT and Ni3Sn4 in the Ni–Sn system were treated as the formula (Ni)0.3333(Ni,Sn)0.3334(Sn)0.3333, (Sn)0.2(Ni,Sn)0.4(Ni)0.4, and thermodynamic parameters of the Ni– Sn system was obtained. The calculated Ni–Sn phase diagram and enthalpies of mixing of liquid at 1550 K and comparison with the experimental data, as shown in Figs. 1 and 2. This work was supported by National Natural Science Foundation of China (NSFC) (Grant nos. 50801004 and 50931002). Z.-K. Liu, C. Zacherl, S.L. Shang, Y. Wang Predicting diffusion coefficients in magnetic phases by firstprinciples methodologies

Fig. 2. Calculated enthalpies of mixing of liquid at 1550 K in the Ni–Sn system in comparison with the experimental data. The reference states of elements are liquid for Ni and Sn.

Predicting self-diffusion and impurity-diffusion coefficients by computational methods is an attractive alternative to lengthy and costly experimental techniques. Methods for computational calculation of cubic and hexagonal systems have been developed and are based on parameter free first-principles calculations. However, a method for calculation diffusion coefficients of magnetic systems without an empirical model for representing the magnetic transition has not been discovered. As a baseline, diffusion coefficients in antiferromagnetic Cr are predicted based on current first-principles techniques. Cluster expansion, Debye model, and phonon first-principles approaches are used in the present work to examine and represent the properties of magnetic Ni at its ferromagnetic to paramagnetic transition. The work being presented focuses on progress made toward the development of an entirely first-principles prediction of self-diffusion and impurity diffusion in magnetic Ni. 3.3. Database development A. Costa e Silva, M.A. da Cunha, L.N.A. Vieira, F. Rizzo

Fig. 1. Calculated Ni–Sn phase diagram by the present thermodynamic description with experimental data measured by Schmetterer et.al.

The thermodynamics of aluminum nitride precipitation in steels revisited In previous work [1] the possibilities and limitations of the application of calculations in the Al–Fe–N system to describe the precipitation of AlN mostly in austenite (FCC) and in during the solidification were discussed. Assessed values in computational thermodynamic databases were compared to ‘‘classical’’ solubility products and experimental data in the usual ranges of austenite thermomechanical processing. The results were also used to evaluate the possibility of ‘‘rock candy’’ fracture. Presently, the precipitation of AlN in ferrite (BCC) is discussed. This is specially relevant for two reasons: (a) some classes of deep drawing steels have their processing designed to favor the precipitation of this nitride concurrently with the annealing treatment, when the steel is mostly ferritic and (b) some silicon alloyed electric steels make extensive use of the precipitation of this nitride at relatively high temperatures, when these steels can have significant fractions of

154

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

BCC in their microstructure. The precise knowledge of the precipitation–dissolution behavior of AlN in these two classes of steels is of great importance to their correct processing. In this work, the information available on the solubility of AlN in BCC is compiled and compared, considering, as much as possible, the interaction with other relevant solutes, mostly silicon, carbon and manganese. Some discrepancies are identified and some possible reasons for these are indicated. Furthermore, the impact of the use of different sources of data on steel processing development is discussed and the need for further studies highlighted. References [1] A. Costa e Silva, F. Rizzo, J.G. Speer, A study of the thermodynamics of aluminum nitride precipitation in steels and ferrous alloys, Proceedings of the XXXI CALPHAD, Stockholm, Sweden, May 2002. A. Costa e Silva, A.Q. Bracarense, E.C.P. Pessoa, M. Monteiro, R.R. Avillez, F. Rizzo, V.R. Santos Thermodynamics of flux behavior in steel welding with focus on hydrogen in steel In previous work the fundamentals of hydrogen absorption from moisture in liquid steel were review and applied to provide a preliminary interpretation of results of underwater welding with different electrodecoatings [1]. Hydrogen can cause several problems in steel such as embrittlement, cracks (sometimes called ‘‘flakes’’ in the steel industry), pores and bubbles. The extremely high mobility of hydrogen in iron aggravates these problems. A large portion of the knowledge about the thermodynamic of hydrogen in steels and slags is well established. In this work, the basic aspects of the thermodynamics of hydrogen in steel entry in the liquid state are reviewed: the importance of oxygen activity and iron content in slag and how alloying elements partition between slag and melt affect these variables is discussed. The effect of the atmosphere above the liquid steel is also discussed. The application of this knowledge to welding, in special underwater welds is presented: results of hydrogen absorption in welds using different electrode coatings are compared with computational thermodynamic calculations, in which an attempt to calculate the slag-metal partition of the different elements and of oxygen is made. The results are discussed and some guidelines for slag selection, purely from the thermodynamic point of view, are proposed. The lack of reliably assessed data for some of the oxides in the slag requires the use of preliminary assessed coefficients, which might limit the accuracy of the predictions. The most relevant compounds for which data is missing are listed, in order to initiate a project to assess their Gibbs energy in a slag described by the Kapoor–Frohberg–Gaye model. References [1] A. Costa e Silva, A.Q. Bracarense, E.C.P. Pessoa, M. Monteiro, R. Avillez, F. Rizzo, V.R.Santos, Thermodynamics of hydrogen in steel—application to underwater welding, in: Proceedings of the XXXVIII CALPHAD, Prague, Czech Republic, May 2009. ¨ A. Markstrom, Y. Du, S.H. Liu, L.J. Zhang, L. Kjellqvist, ¨ J. Bratberg, A. Engstrom TCAL1: A new thermodynamic database for aluminium alloys Thermodynamic databases developed using the CALPHAD methods have been successfully applied to the modelling and

simulation of Al based alloys for more than fifteen years. Such databases when combined with suitable software can be used for accelerating alloy design as well as improving understanding of existing alloys in terms of their processing and in-service behaviour. Additionally, such databases are also essential to the modelling of microstructural evolution using methods such as phase field codes and steady state approaches. A new thermodynamic database has been developed for Al-base alloys based on the critical evaluation, using the CALPHAD method, of all the constituent binary systems across their full range of composition and 59 ternaries, 15 quaternaries and 1 quinary. This new database contains all the important Al-based alloy phases within a 26-element framework [Al Cu Fe Mg Mn Ni Si Zn B C Cr Ge Sn Sr Ti V Zr Ag Ca H Hf K La Li Na Sc] and in total 345 solution and intermetallic phases are included in this database. Also of note is that the ordered and disordered bcc (A2 and B2) and fcc (A1 and L12/g0 ) phase have been modelled with a two sub-lattice model using a single Gibbs energy curve, and this type of description is of particular importance to be able to predict second order transformations between A2 and B2. Calculations on different kind of commercial alloys show a very good accuracy of the database, both when it comes to prediction of phases that are formed and the temperature for formation. Solidification simulations are shown for various kinds of commercial Al-alloys and are compared with experimental data. C.P. Wang, S.X. Gan, X.J. Liu A thermodynamic database of phase diagram in alloy systems including rare earth elements The rare earth elements have good combination of magnetic, optical, electrical and thermal characteristics, and have been widely used to prepare new materials. As the map for materials design, the phase diagram can give an important direction for development of the rare earth alloy materials. Thus, development of the thermodynamic database of the phase equilibria in the rare earth alloy systems is important and necessary. In this work, the thermodynamic assessments of phase diagrams in the binary and ternary systems with rare earth elements have been carried out by using CALPHAD (calculation of phase diagrams) method on the basis of the experimental data including thermodynamic properties and phase equilibria. Gibbs free energies of the solution phases were described by subregular solution model with the Redlich–Kister equation, and those of the intermetallic compounds and gas phase were, respectively, described by sublattice model and ideal gas model. A consistent set of thermodynamic parameters have been derived for describing the Gibbs free energies of each solution phase and intermetallic compound. The calculated phase diagrams and thermodynamic properties are in good agreement with the experimental data. The thermodynamic database of alloy systems with rare earth elements has been developed, which will provide important information including phase diagrams and various thermodynamic properties for development of alloy materials with rare earth elements. ~ N.M. Guimaraes, D.A.P. Reis, C. Moura Neto, G.C. Coelho, D.S. Almeida Thermodynamic modeling of the Y2O3–Nb2O5 system A great effort has been done by the scientific community to increase efficiency and decrease greenhouse gas emissions of energy generation equipments. This has lead materials scientists and engineers to develop new substrate and coating materials. Oxide coatings are commonly used to protect Ni-based

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

superalloys against oxidation and hot-corrosion. Thermal barrier coatings (TBCs) are used on top of other coating systems in order to lower the metal surface temperature and increase its lifetime. TBCs are usually composed by yttria-stabilized zirconia (YSZ) ceramic materials. Co-doping with rare earth oxides increases the thermal insulation efficiency [1,2]. Hot-corrosion resistance has been improved by Ta2O5 additions [3]. A thermodynamic database specialized in TBC systems is commercially available, namely STBC1 – SGTE thermal barrier coating database [4], containing however information only for the ZrO2–Al2O3–Gd2O3–Y2O3 system. The present work aims at contributing to the development of thermal barrier coating databases through the modeling of the Y2O3–Nb2O5 system. All information available in the literature together with our own experimental results such as phase quantification and transition temperatures have been used as a basis for the thermodynamic modeling. The calculated phase diagram reproduces well the available experimental data. However, this result has still a preliminary character because of the lack of thermochemical information on the system.

155

important density of grain or interphase boundaries. The atomic transport along these short-circuits diffusion is several orders of magnitude faster than in the crystal. At low temperatures, the diffusion processes are often not properly considered. In fact, even when bulk diffusion can be neglected the atomic displacement along grain boundaries is important and the whole diffusion processes becomes relevant. The Co presents a fast diffusion in the bulk of Zr based alloys [1] and recent experiments show his ultra-fast diffusion along alpha-Zr grain boundaries [2]. Particularly for the Co60 radioisopes production for the nuclear medicine the diffusion process of this element plays a key role in order to guarantee the security conditions. In this work the simulation results of Co diffusion in Zr grain boundaries in different Harrison kinetics [3] are showed in the [373–573] K temperature range at different times. The simulations have been performed with DICTRA software. The results are discussed in comparison to previous experimental data. Finally the segregation factor and the validity of the different kinetic regimes are evaluated. References

References [1] J.R. Nicholls, Advances in coating design for high-performance gas turbines, MRS Bull., 2003. [2] O. Fabrichnaya, C.H. Wang, M. Zinkevich, C.G. Levi, F. Aldinger, Phase equilibria and thermodynamic properties of the ZrO2–GdO1.5–YO1.5 system, J. Phase Equilib. Diffus. 26 (6) (2005) 591–604. [3] F.M. Pitek, A Study of the Zirconia–Yttria–Tantala System as a Potential Thermal Barrier Oxide, Dissertation of Doctor Degree, University of California, 2006. [4] Thermo-Calc Database Guide, Foundation of Computational Thermodynamics, Stockholm, Sweden. W. Gong, Y. Wu, M. Gaune-Escard Thermodynamic assessment of CsBr–LnBr3 system (Ln¼La, Ce, Tb) Lanthanide halides and their mixtures with alkali metal halides play a significant role in a number of industrial applications largely based on molten salt technologies, many still under development. Their thermodynamic and transport properties can provide basic information for process development and optimization. However these data are scarce and not easily accessible in literature. Accordingly, intensive efforts are being made at an international level both in R&D aspects and also in database development. Based on the experimental phase diagram and thermodynamic data, thermodynamic assessment on the CsBr–LnBr3 (Ln¼La, Ce, Tb) mixtures were performed using the CALPHAD method. A two sub-lattice ionic solution model, (Cs þ )P:(Br  , LnBr6  3, LnBr3)Q, was used to describe the liquid phase. The phase diagrams, thermodynamic properties and the general features of the CsBr– LnBr3 mixtures (Ln¼La, Ce, Tb) were obtained reliably. And a critical discussion was given. 3.4. Diffusion and phase transformation modeling C. Corvalan, M. Iribarren, J. Sola Calculation of different Co diffusion kinetics along alpha-Zr grain boundaries at power reactor temperatures under normal conditions Zirconium is the base material for nuclear devices. During normal operation conditions this structural material presents an

[1] G.V. Kidson, Philos. Mag. A 44 (1981) 341. [2] C. Corvalan Moya, et al., Defect Diffus. Forum 283–286 (2009) 669. [3] L.G. Harrison, Trans. Faraday Soc. 57 (1961) 1191. K.C. Hari Kumar, V.B. Rajkumar, B.S. Srinivas Prasad Numerical simulation of precipitate evolution in ferritic-martensitic power plant steels Knowledge about the nature of precipitates present and their evolution with respect to time is found to be very critical to increase the operating temperature in power plant steels [1]. Evolution of most important precipitates in P91, RAFM and E911 steels are studied using the equilibrium thermodynamics tool Thermo-Calc and the kinetic simulation tool MatCalc. We have used an ingeniously developed thermodynamic database for steels and a modified version of public domain mobility data base to achieve this. Variation of precipitate mean radii and phase fractions of M23C6, MX, Laves, Z phase, M7C3 precipitates with time were simulated. Effect of certain alloying elements like molybdenum, tantalum, etc., on the precipitation phenomenon in P91, RAFM and E911 steels is studied. Simulations are carried out for different heat treatment schedules and the results are compared with the experimental data from literature. References [1] F. Abe, Strengthening mechanisms in steel for creep and creep rupture, in: F.Abe, T.U. Kern, R. Viswanathan (Eds.), Creep-resistant Steels, Woodhead Publishing Ltd., Cambridge, 2008, pp. 279–304. M. Kajihara, K. Motojima, T. Asano Microstructure evolution by solid-state reactive diffusion in the (Ni Cr)/Sn system Owing to high electrical conductivity, Cu-base alloys are widely used as conductor materials in the electronics industry. If the Cu-base conductor is interconnected with a Sn-base solder, Cu6Sn5 and Cu3Sn are formed at the interconnection between the conductor and the solder during soldering and then gradually

156

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

grow during energization heating at solid-state temperatures. The Cu  Sn compounds are brittle and possess high electrical resistivities. Thus, the growth of the compounds deteriorates the mechanical and electrical properties of the interconnection. To inhibit the formation of the compounds, the Cu-base conductor is usually plated with a Ni layer. The addition of Cr into the Ni layer improves corrosion resistance and may influence the compound formation. To examine this influence, the kinetics of the solidstate reactive diffusion in the (Ni Cr)/Sn system was experimentally observed by a metallographical technique. In the experiment, Sn/(Ni Cr)/Sn diffusion couples with Cr concentrations of 5.0, 11.7, 15.0, 23.2 and 28.2 at% were isothermally annealed at temperatures of 453 473 K for various times up to 1200 h. During annealing, a multiphase layer consisting of Ni3Sn4, CrSn2 and Sn is formed at the initial (Ni Cr)/Sn interface in the diffusion couple. This multiphase layer is merely called the intermetallic layer. Diffusion paths at 473 K for various diffusion couples are shown in Fig. 1. On the other hand, Fig. 2 indicates the result of 473 K for the mean thickness l of the intermetallic layer versus the annealing time t. As can be seen, l is proportional to a power function of t, and the exponent n of the power function

takes values of 0.60 0.75. This means that the interface reaction as well as the interdiffusion contributes to the rate-controlling process for the growth of the intermetallic layer. At the experimental annealing times, the overall growth rate of the intermetallic layer reaches the maximum value at a Cr concentration of 12 at%. M.F. Campos, J.A. Castro, S.A. Romero, M.F. Moreira, F.J.G. Landgraf Phase transformations in the annealing of Sm–Co–Fe–Cu–Zr magnets The processing of Sm2Co17 type magnets – in fact (SmZr)2(CoFeCu)17 – includes a complex heat treatment, with the following sequence (steps 1–4): (1) solubilization at 1200 1C (4 h), (2) quenching, (3) precipitation heat treatment at 800–850 1C (or 820 1C) for several hours (7 h) (4) slow colling, from 820 to 400 1C, at 1 1C/min This process involves several phase transformations, leading to a nanostructure, which is responsible for the excellent magnetic properties of these alloys [1,2]. In the present an isotropic magnet with composition Sm0.104Co0.60Fe0.195Cu0.072Zr0.02 was submitted to the heat treatment described above. The nanoscale microstructure was evaluated with FEG SEM (Field Emission Gun Scanning Electron Microscope) and with Synchrotron X-Ray Diffraction (XRD), used for Rietveld Analysis. At the step 1, the only phase present is disordered 2:17R (rhombohedral, 166 structure), which may present different chemical compositions At the step 3, temperature around 820 1C, the following phase transformation takes place, 2:17R (disordered)-1:3þ1:5þ 2:17R (ordered). There is a formation of nanoscale structure with Sm2(Co,Fe)17 cell phase (rhombohedral 166 structure), Sm(CoCu)5 cell boundary phase (hexagonal 191 structure) and the lamellae phase, (ZrSm)1(CoFeCu)3. The 1:3 phase has been described as Rhombohedral Sm1Zr2Co9, a stable phase in the ternarium Sm–Zr–Co It is presented a diffusion model (based on the finite volume method). This model takes into account copper and samarium diffusion. The model is able to describe the kinetics of the phase transformations, which results in the excellent magnetic properties found in these alloys. Acknowledgements FAPERJ CNPq.

Fig. 1. Diffusion paths at 473 K.

References [1] M.M. Corte-Real, M.F. de Campos,Y. Zhang, G.C. Hadjipanayis, J.F. Liu, Phys. Status Solidi A 193 (2002), 302. [2] S. A. Romero, M. F. de Campos, H. Rechenberg, F.P. Missell, J. Magn. Magn. Mater. 320 (2008) 181 S. Sankaran, N. Maheswari, V.B. Rajkumar, S.G. Chowdhary, K.C. Hari Kumar Kinetic simulation of carbon migration during quenching and partitioning (Q and P) treatment of steels

Fig. 2. The dependence of l on t at 473 K.

Quenching and partitioning is a novel heat treatment process which can be used to generate steels with optimum strength and ductility. The process involves quenching the steel below the martensite-start temperature (MS), followed by a partitioning treatment (reheating slightly below MS) to enrich the remaining austenite with carbon, thereby stabilizing it to room temperature. During this process the competing reactions, mainly carbide precipitation, can be suppressed by appropriate alloying with Si

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

and Al [1–3]. Desirable alloy compositions selected for the present simulation are in the range of C ¼0.2–0.3, Si¼0.4–1.6, Mn¼ 1.2–1.75, Al¼ 1–1.6. Thermodynamic and kinetic calculations were performed using MatCalc and Thermo-Calc software. The partitioning of carbon from martensite (here considered as supersaturated BCC matrix) to austenite (FCC matrix) has been modeled using the indigenously developed thermodynamic database combining it with an open source mobility database. A carbon diffusion profile has been calculated and it will be useful to decide the proper partitioning temperature and time to obtain the desired microstructure. References [1] A.J. Clarke, J.G. Speer, M.K. Miller, R.E. Hackenberg, D.V. Edmonds, D.K.Matlock, F.C. Rizzo, K.D. Clarke, E. De Moor, Acta Mater. 56 (2008) 16–22. [2] J.G. Speer, D.V. Edmonds, F.C. Rizzo, D.K. Matlock, Curr. Opin. Solid State Mater. Sci. 8 (2004) 219–237. 3.5. Experimental measurements A. Antoni-Zdziobek, S. Lay, C. Crozet Experimental determination of phase equilibria in the ironrich side of the Cu–Fe–Ni system between 873 and 1273 K Although the Cu–Fe–Ni system has been the subject of several investigations, published information of the phase diagram of this system is relatively incomplete, particularly in the Fe-rich side of the system. Only few data concerning the crystallographic nature of phases have been reported. A new experimental study has been carried out to contribute to the knowledge of phase equilibria of this system. Two key Cu/Fe/Ni diffusion multiple and 23 decisive alloys have been prepared, directed by a critical evaluation of the data available in the literature and by computer calculations. A combination of EPMA and metallography is used to determine accurately the phase boundaries. XRD is used on equilibrated alloys to identify the crystallographic nature of the phases. It is shown that it is not possible to retain the high-temperature phase g(fcc) to room temperature even by a fast quench and that g(fcc) to a(bcc) partitionless transition takes place. However, heating/cooling experiments which could avoid the possibility of a phase change during quenching are not suitable: it is shown by dilatometry that metastable structures are produced even during very slow cooling process as 0.03 1C/s. Using diffusion multiple and equilibrated alloys, tie-lines of the g1/g2 (where g1 is enriched in Fe and Ni and g2 is enriched in Cu) phase equilibria were obtained at 1073 and 1273 K. Tie-lines of the aFe/g1 and aFe/g2 phase equilibria were obtained at 1073 K. The three phase field aFeþ g1þ g2 was determined at 873 K. The calculated miscibility gap is less extended for high Ni content than indicated by the experimental values. On the contrary, Fe solubility in the g1 phase is higher than the calculated values. It appeared that phase equilibria in the Cu-rich corner were dependant on the method for the equilibrated alloys preparation. Therefore a comparison of the different methods used in the litterature, cast or sintered alloy, is given and our findings are discussed. C.A. Nunes, G.C. Coelho, B.B. Lima, T.F. Villela, N. David, J.M. Fiorani, M. Vilasi Experimental investigation and thermodynamic assessment of the V–Si–B system Based on the technological interest of Me–Si–B for structural, high-temperature applications, this investigation focused initially

157

on the experimental evaluation of the V–Si–B system in the V–VB–VSi2 region. It was possible to establish an isothermal section at 1600 1C and a liquidus projection. The isothermal section data indicated a negligible solubility of B in the V3Si and V5Si3 (T1) and V6Si5 phases as well as of Si in V3B2 and VB phases. Two ternary phases presenting the structures known as T2 (Cr5B3-prototype) and D88 (Mn5Si3-prototype) were observed in both as-cast and heattreated samples. With respect to the liquidus projection data, nine regions of primary solidification were identified (Vss, V3Si, T2, V5Si3, V6Si5, D88, VSi2, VB) and eight invariant four-phase reactions. From these experimental and literature data (T–x data; Cp, enthalpy of formation), the thermodynamic optimization of the Gibbs energy functions of the phases was carried out using the Parrot Module of Thermo-Calc. In general, a good agreement between the experimental and calculated data was achieved.

C. Ve´bert, Y. Hamini, L. He´richer, S. Michon, L. Aranda, P. Berthod Casting and characterization of Ni-, Fe(Ni)- and Fe-based 30%Cr-containing alloys strengthened by TaC; comparison with thermodynamic calculations Tantalum carbides are of a great interest for reinforcing refractory alloys since they are among the most stable ones at high temperatures. It is why they are often used in cast cobaltbased superalloys, alone or in addition to other types of carbides less stable (e.g., chromium carbides), to efficiently strengthen these ones in service for resisting creep deformation under applied mechanical stresses. Such TaC carbides are less common in the microstructures of nickel-based or iron-based alloys, which can be also considered for high temperature applications. TaC carbides are potential strengthening particles in such alloys when they need containing chromium for resisting hot corrosion instead of aluminum. Three nickel-based alloys, three {ironþ nickel}-alloys and three iron-based alloys, all containing 30 wt%Cr as well as carbon and tantalum quantities allowing them to possibly develop interdendritic TaC-carbides are solidification, were elaborated by foundry under inert atmospheres. They were metallographically examined in the as-cast condition and in three heat-treated states (at 1000, 1100 and 1200 1C). Their melting temperature ranges were also specified by DTA/DSC measurements. All these experimental data were thereafter compared to results of thermodynamic calculations performed with Thermo-Calc. If some TaC carbides were obtained in the nickel-based alloys after solidification in the Ni-based alloys (mixed with chromium carbides), these ones tended to disappear during the subsequent exposures to high temperature. In contrast, TaC carbides precipitated in much greater quantities during solidification in the (Fe, Ni)-based and Fe-based alloys and they remained stable during the heat-treatments at high temperatures. These observations, as well as the measured solidus and liquidus temperatures were not always in good agreement with thermodynamic calculations, revealing that the used home-made database needs to be improved. References [1] S. Michon, P. Berthod, L. Aranda, C. Rapin, R. Podor, P. Steinmetz, Calphad 27(3) (2003) 289–294. [2] P. Berthod, L. Aranda, C. Ve´bert, S. Michon, Calphad 28 (2004) 159–166. [3] P. Berthod, Y. Hamini, L. Aranda, L. He´richer, Calphad 31 (2007) 351–360. [4] P. Berthod, Y. Hamini, L. He´richer, L. Aranda, Calphad 31 (2007) 361–369.

158

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

C. Wu, X. Su, J. Wang, Y. Liu, H. Peng, H. Tu Experimental investigation and thermodynamic modeling of the Zn–V–X (X ¼Sn, Ti, Si) ternary systems The Zn–V–Sn and Zn–V–Ti alloys can well suppress high Sireactivity during galvanizing. Investigation of the Zn–V–Si system helps to understand the effect of V on galvanizing Si-containing steel. To build a Zn-based alloys thermodynamic database, the phase equilibria of these ternary systems, including isothermal section 450 and 600 1C, were experimentally investigated by means of X-ray powder diffraction (XRD), differential thermal analysis (DTA) and scanning electron microscopy coupled with energy/wave dispersive spectroscopy (SEM-EDS/WDS). In addition, these systems were thermodynamic assessed in the present work. Two ternary compounds, V13Sn23Zn64 and V11Sn13Zn76, which have narrow composition range, were found in the Zn– Sn–V system. While no ternary compound was found in the Zn–V– Ti and Zn–V–Si systems. The experimental results indicated that neither Si nor Sn can dissolve into VZn3 and V4Zn5. However, TiZn3 and VZn3 form a continuous solid solution. And the solubility of Zn in VSn2, V3Sn, VSi2, V5Si3 and V3Si are about 2.5, 4.8, 1, 2.6 and 2.1 at%, respectively. Based on present results and available information, these systems were thermodynamically assessed. Acknowledgement The financial support from Qinlan project and National Natural Science Foundation of China (nos. 50971110 and 50971111) are greatly acknowledged.

thermo gravimetry (TG) as well as X-ray diffraction (XRD) measurements were done both to achieve the transition temperatures and the phase compositions in the sulfatic systems. References [1] [2] [3] [4]

J.H. Gladstone, Q. J. Chem. Soc. 6 (1854) 106 M. Le Chatelier, Bull. Soc. Chim. 47 (1887) 300 K. Kubierschky, History (1902) 404–413 J.H. van’t Hoff, H. Barschall, Sitzungsber. K. Preuss. Akad. Wiss. 18 (1903) 359–371. ¨ [5] E. Janecke, Z. Phys. Chem. Stoechiom. Verwandschaftsl. 64 (3) (1908) 343–356. [6] R. Nacken, Zentralbl. Mineral., Geol. Palaeontol. (1910) 262–271 ¨ [7] H. Muller, Neues Jahrb. Mineral., Geol. Palaeontol. Beilageband 30 (1910) 1–54. [8] W. Grahmann, Z. Anorg. Chem. 81 (1913) 257–314. [9] W. Eysel, Am. Mineral. 58 (7–8) (1973) 736 [10] J. M. Sangster, A.D. Pelton, in: Critical coupled evaluation of phase diagrams and thermodynamic properties of binary and ternary alkali salt systems, special report to the phase equilibria program, Part B: Evaluations for 54 common-ion binary systems involving (Li, Na, K) and (F, Cl, OH, NO3, CO3, SO4), American Ceramic Society, Westerville, Ohio, 1987, pp. 4–231. [11] N. Mofaddel, R. Bouaziz, M. Mayer, Thermochim. Acta 185 (1) (1991) 141–153

H. Ipser, R. Ganesan ¨ D. Kobertz, M. Muller Experimental studies and re-assessment of the quasi-binary systems containing the sulfates of sodium, potassium, and calcium by differentia thermal analysis and X-ray diffraction The combustion of lignite containing high amounts of sulfur, calcium, potassium, and sodium induces the formation of sulfatic depositions and corrosion related processes on refractory metal and ceramic material. These depositions diminish the heat conductance and their spalling leads to damages in the power plant. The mechanisms for the formation of these conglomerations are initiated by lowmelting or related to adhesive sulfatic phases. Limitation of fossil fuel resources requires an increase of the efficiency of power plants by using combined cycle power systems. The pressurized pulverized coal combustion (PPCC) combined cycle is a coal fired combined cycle concept which is able to achieve efficiencies in excess of 53%. The direct use of the hot flue gas for driving a gas turbine requires a hot gas cleanup to achieve corrosion prevention of the turbine blading. Hot corrosion on turbine blading is initiated by deposition or condensation of corrosive species like alkali sulfates and dependent on concentrations of alkalis in the hot flue gas. Type I hot corrosion is caused by the formation of liquid Na2SO4 above its melting point (884 1C). Type II hot corrosion is caused by the formation of an eutectic melt of NiSO4 and Na2SO4 above 671 1C. NiSO4 itself is formed by the reaction of the oxide scale of Ni base alloys in blading with SO3 in dependence of the SO3 partial pressure in the hot flue gas. The motivation of our studies is to understand the aforementioned alkali sulfate related processes under applied conditions and accrued phases. Multi component sulfatic systems can only be really understood after comprehending the basic binary sub-systems. This work includes the experimental studies of the three sub-systems Na2SO4– CaSO4, K2SO4–CaSO4, and Na2SO4–K2SO4 since there are variant and even contradictory results and interpretations in the literature (e.g., [1–11]). Differential thermal analysis (DTA) with simultaneous

Thermochemical study of rare earth–cadmium systems: The system Gd–Cd The pyrometallurgical process is one of the promising methods for reprocessing of spent nuclear fuel. In this process liquid Cd is used as cathode in the electrorefining step. The solubility and chemical activity of actinides and rare earth fission products in Cd are important parameters in the process. When the solute concentrations exceed their solubility limits in liquid Cd, intermetallic compounds between solutes and Cd are formed. An understanding of the chemical activity of Cd in such intermetallic compounds would help in optimizing the process parameters. Thus the thermochemical study of fission products-Cd systems assumes importance. In the present work, a detailed study of the thermochemical behavior of Gd–Cd alloys was carried out by an isopiestic vapor pressure method. The Gd samples were equilibrated in the temperature range 730– 1122 K, at various known cadmium vapor pressures, viz., 3– 133 mbar. Data points were obtained between approximately 0 and 85 at% Cd. The obtained equilibrium phases were characterized and their cadmium activity was determined as a function of temperature and composition. From the temperature dependence of the activities, the thermochemical behavior of Gd–Cd alloys was derived. The derived data could be used as an input in the optimization of the Gd–Cd phase diagram by a CALPHAD approach. H. Li, Y.P. Ren, N.N. Zhao, M. Jiang, G.W. Qin, S.M. Hao Experimental investigation on the ternary compound and c-Ni solvus in the Ni-rich corner of the Nb–Ni–Ti system The equilibrium phase structures and compositions in the Nirich corner of the Nb–Ni–Ti system have been researched through the equilibrated alloy method by using scanning electron microscopy assisted with energy dispersive spectroscopy of X-ray,

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

electron probe microanalysis, X-ray diffraction and transmission electron microscopy. It has been shown that one ternary compound with the composition about 10Nb–80Ni–10Ti (atomic fraction, %) exists in the Nb–Ni–Ti system, which could be in equilibrium with the Ni-based solid solution. The ternary compound could also be in three-phase equilibrium with g-Ni þ Ni3Nb at 1000 1C. It has the hexagonal structure with the structure based solid solution increases for the third addition of Ti. The maximum solubility of Nb in g-Ni is about 11.7 at%, and the total solubility of Ti and Nb in g-Ni is about 14.2 at% at 1000 1C. Acknowledgements The financial support from the National Natural Science Foundation of China (Grant no. 50971037) is greatly acknowledged. J. Pstrus, T. Gancarz, P. Fima Design of new lead-free solders on the example of Sn–Zn–In Toxic properties of lead are the reasons for worldwide legislation, for example the RoHS EU directive, prohibiting the use of cadmium and lead in consumer products. Efforts were undertaken on finding replacements to conventional tin–lead solder, that would have similar properties. Literature data [1–3] point to eutectic Sn–Zn, as one of the most promising systems for soldering application. Unfortunately it is characterized by relatively poor wettability [4,5] and thus the addition of indium was proposed [6]. Thermodynamic analysis [7] and our own data show that the addition of In to Sn–Zn eutectic should not exceed 4 wt%. DSC measurements were performed on the alloy containing 4 wt% of indium. Wetting test procedure was developed based on initial measurements made on copper with traditional SnPb solder at 2501 with the use of rosin. Currently, wetting measurements Sn–Zn eutectic-based with the use of industrial flux in nitrogen atmosphere. In order to progress with designing process, it is necessary to collect the results of dilatometric tests, electrical resistance, mechanical strength and structural studies. Acknowledgement This work is financed under the project ‘‘POIG.01.01.02-00015/09-00 – ZAMAT – Advanced materials and technologies of their production—a task 4.300 , in the years 2010–2013. References [1] S. Ganesan, M. Pecht, Lead-free Electronics, Wiley-Interscience, 2006. [2] J. Glazer, Int. Mater. Rev. 40 (1995) 65. [3] M. Abtew, G. Selvaduray, Mater. Sci. Eng. R 27 (2000) 95. [4] P. Vianco, R. Rejent, J. Electron. Mater. 28 (1999) 1127. [5] Z. Chen, et al., J. Electron. Mater. 35 (2006) 1734. [6] M. McCormack, et al., Appl. Phys. Lett. 63 (1993) 15. [7] S. Vaynman, M.E. Fine, Scr. Mater. 41 (1999) 1269. K.C. Hari Kumar, S. Balakrishnan, K. Anathasivan, S. Anthonysamy, V. Ganesan, P.R. Vasudeva Rao Solidus and liquidus measurements in the U–Zr binary system Experimental data on the phase transitions in U–Zr system are useful in establishing the isothermal sections of the ternary U– Pu–Zr. The latter is relevant to metallic fast reactor fuels. Measurements on the solidus and liquidus in the U–Zr system are rather very limited [1–4]. Summers-Smith [1] had reported

159

the solidus of U–Zr alloys (10, 30, 50 at% Zr) by metallographic examination while the liquidus (above 40 at% Zr) was measured by optical pyrometry. These values were later adopted by Masalski [5]. Leibowitz et al. [4] reported the solidus and liquidus temperatures for an alloy with a composition of 19.3 at% zirconium. Kanno et al. [6] arrived at the solidus and liquidus temperatures from their (measured) values on the activity of uranium in U–Zr alloys and Ogawa [7] calculated the same using the data reported in ref [5]. Maeda et al. [2] had measured the liquidus (24.4 and 39.3 at% Zr) by Knudsen effusion mass spectrometry which is in commensurate with Summers-Smith [1]. Recently Kurata [8] optimized this phase diagram based on the data available in the literature. The solidus and liquidus was not measured over the entire range of composition by any of these authors and hence, the investigation of the same was undertaken by using the spot technique. For measuring the temperatures of phase transitions involving liquids in radioactive alloys beyond 1273 K, an experimental system based on Spot technique devised by Ackermann and Raugh [9] was set up in a glove box. This technique has been widely used to study the high temperature phase transitions that involve liquid in actinide bearing systems [9,10]. The accuracy in the temperatures of transitions was determined by measuring the melting points of pure Au, Cu, Ni, Pt and Zr. These melting points could be determined within 75 K. References [1] D. Summers-Smith, J. Inst. Met. 83 (1954–1955) 277. [2] A. Maeda, et al., J. Alloys Compd. 179 (1992) L21–L24. [3] R.I. Sheldon, et al., Bull. Alloys Phase Diag. 10 (1989) 15–171. [4] L. Leibowitz, et al., J. Nucl. Mater. 167 (1989) 76–81. [5] T.B. Massalski, Acta Metall. 6 (1958) 243. [6] M. Kanno, M. Yamawaki, T. Koyama, N. Morioka, J. Nucl. Mater. 154 (1988) 154. [7] T. Ogawa, T. Iwai, J. Less-Common. Met. 170 (1991) 101. [8] M. Kurata, IOP Conf. Ser.: Mater. Sci. Eng. 9 (2010) 012022. [9] R.J. Ackermann, E.G. Raugh, High Temp. Sci. 21 (1972) 161. [10] K. Ananthasivan, S. Anthonysamy, C. Sudha, A.L.E. Terrance, J. Nucl. Mater. 300 (2002) 217–229.

K. Shinagawa, T. Omori, K. Oikawa, I. Ohnuma, K. Ishida, R. Kainuma Experimental determination and thermodynamic analysis of phase equilibria in the Ni–W and Ni–Al–W systems Ni-base superalloys strengthened by coherent precipitates of g0 Ni3Al (L12) phase in g(A1) matrix have been widely used as high temperature materials. Tungsten (W) is one of the essential alloying elements for improving high-temperature properties, and its effect on the phase equilibria and the mechanical properties has been reported by some researchers. However, even for the fundamental systems, such as the Ni–W binary and the Ni–Al–W ternary systems, phase diagrams have not been established in detail yet. Moreover, it was found that discrepancies between experimental results and calculation, especially in the Ni–Al–W system, were caused by inaccuracies in the respective sub-systems. In this study, eutectic, liquidus and solidus temperatures and phase equilibria among solid phases in the Ni–W binary system were determined experimentally. Isothermal section diagrams at temperatures between 900 and 1300 1C in the Ni–Al–W ternary system were also established. In the Ni–W system, the Ni4W phase is obtained by alloying method and the temperatures of the peritectoid reactions, (Ni)þNiWZNi4W and (Ni)þ(W)ZNiW are confirmed to be 1031 and 1056 1C, respectively, by DSC measurement. It is interesting to note that partition behavior of W between g

160

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

and g0 phases drastically changes with increasing W content and that the g0 phase region extends to low Al content in the Ni–Al–W system. Thermodynamic evaluation of the Ni–Al–W system was carried out taking into account obtained experimental results. Calculated phase diagrams are in good agreement with experimental data. K. Yaqoob, J.-M. Joubert Experimental determination of isothermal sections of Mo–Ni–Re system at 1200 and at 1600 1C Mo–Ni–Re system is of great interest in relation to the development of some nickel based superalloys. However if concentration of Mo and Re is too large, Frank–Kasper phases may form during exposure to high temperature [1–3]. These phases have adverse affect on the mechanical properties of the material. Therefore in order to avoid precipitation of these brittle intermetallic phases, it is important to determine the presence and homogeneity domains of different phases which may appear in the Mo–Ni–Re system. Comparison of the previous attempts [1–3] made for the experimental determination of Mo–Ni–Re phase diagrams revealed the presence of contradictory results. Therefore in order to develop a better understanding, a complete redetermination of Mo–Ni–Re system was carried out. Samples of different compositions were prepared, annealed, and investigated by means of metallography, X-ray diffraction and electron probe microanalysis. The isothermal section determined at 1200 1C showed a large extension of binary s(Mo–Re) and d(Mo–Ni) phases into the ternary region and presence of a ternary phase was also observed. However, no ternary phase was observed in the isothermal section determined at 1600 1C. For the ternary s phase, a complete structural characterisation including the determination of ternary site occupancies was carried out for different ternary compositions. They will be compared to the calculated site occupancies [4] obtained by ab-initio calculations. References [1] A.A. Kodentsov, S.F. Dunaev, E.M. Slyusarenko, E.M. Sokolovskaya, A.N. Priimak, Vestn. Mosk. Univ., Ser. 2., Khim., 28 (2) (1987) 153–158. [2] E.M. Slyusarenko, A.V. Peristyi, E.Y. Kerimov, M.V. Sofin, D.Y. Skorbov, J. Alloys Compd. 264 (1998) 180–189. [3] Y. Feng, R. Wang, K. Yu, D. Wen, Rare Met. 27 (1) (2008) 83–88. [4] K. Yaqoob, J.-C. Crivello, J.-M. Joubert, Inorg. Chem. 51 (2012) 3071–3078.

to access the multi-component phase equilibrium, and phase transformation for the system. The ICP-MS results showed close agreement with Thermo-Calc simulations for manganese sulfide precipitate, however aluminum nitride showed a different solubilization temperature. TEM results showed a cupper sulfide phase, and ThermoCalc simulation considered cupper in solid solution in ferrite and austenite mainly. L. Zhang, P.J. Masset, L. Liu Thermodynamic analysis of the Al–Cu–Er ternary system For ternary and multicomponent phase diagrams, the experimental data is often scarce and sometimes contradictory. It is obviously an extremely time- and cost-consuming process to establish such phase diagrams solely by experimental work over wide ranges of compositions and temperatures. The CALPHAD method has proved to be a powerful tool to reduce the experimental effort by selecting small number of key experiments that are decisive for the thermodynamic description of the multicomponent system. In this work, experimental work has been done and combined with the CALPHAD method to generate a consistent thermodynamic description of the Al–Cu–Er ternary system, especially for Al-rich alloys. The viability of a procedure for the selection of multicomponent key samples is demonstrated for this multicomponent system. Dedicated thermal analysis with DSC on sealed samples was performed and the microstructure of solidified alloys was analyzed using SEM/EDX. The phase formation in ternary alloys is analyzed using the tool of thermodynamic equilibrium and Scheil calculations for the solidification paths and compared with present experimental data. The significant ternary solid solubilities of intermetallic phases are quantitatively introduced in the ternary Al–Cu–Er system phase diagram and validated by experimental data. And the experimental data of energy transformation (thermo effect) during the phase transformation are also simulated by our thermodynamic calculations. References [1] L.G. Zhang, L.B. Liu, G.X. Huang, H.Y. Qi, B.R. Jia, Z.P. Jin, Thermodynamic assessment of the Al–Cu–Er ternary system, CALPHAD 32(2008) 527–534. [2] L.-G. Zhang, P.J. Masset, F. Cao, F. Meng, L. Liu, and Z.P. Jin, Phase relationships in the Al-rich region of the Al–Cu–Er system, J. Alloys Compd. (509) 2010, 3822–3831. M.M. Asadov

L.N.A. Vieira, M.A. Cunha, F.L. Alcantara, R.A.N.M. Barbosa Microstructural aspects and computational thermodynamics of secondary phase precipitation in Fe–Si 3 wt% steel The magnetic properties of electrical steels are affected by precipitation of secondary phases. The precipitates can work as a key controlled requirement as part of the manufacturing process or as an unwanted harmful residual in the final product. In this work, nitrides and sulfides precipitates in high permeability grain oriented electrical steels (HGO) were studied after soaking at different temperatures for long annealing time. The samples were encapsulated in a vacuum sealed quartz tube in order to avoid changes in the chemical composition during high temperature annealing and then the samples were analyzed by transmission electron microscopy (TEM), scanning electron microscopy (SEM), inductively coupled plasma mass spectrometry (ICP-MS) and also optical microscopy (OM). The Thermo-Calc software with a TCFE6 database was applied

Physicochemical and thermodynamic properties of the GeSe2–A2B6 (A2 ¼Hg; Cd; B6 ¼S, Te) systems With the purpose of definition of important parameters of intermediate phases and limited solid solutions of the threefold mutual Cd(Hg), Ge//S(Se),Te systems we investigated physicochemical and thermodynamic properties of the pseudo-binary GeSe2–CdTe, GeSe2–HgTe, GeSe2–HgS, GeSe2–CdS systems. The phase diagrams of GeSe2–A2B6 (A2 ¼Hg; Cd; B6 ¼ S, Te) systems were plotted by the methods of differential thermal and X-ray diffraction analyses, by the measurement of electromotive force, microhardness and density. It is established, that phase equilibriums in the pseudo-binary GeSe2– CdTe, GeSe2–HgTe, GeSe2–HgS, GeSe2–CdS systems are characterized by formation of the limited solid solutions on basis of basic components and fourfold intermediate phases such as A2GeSe2Te2 and A4GeS4Se2:Cd2GeSe2Te2 (hexagonal system; a¼5.69; ˚ Cd4GeS4Se2, Hg2GeSe2Te2 (tetragonal system; a¼7.50; c¼11.32 A), ˚ Hg2GeS2Se2 (hexagonal system; a¼7.20; c¼ 36.64 A), ˚ c¼36.48 A), ˚ Hg4GeS4Se2 (monoclinic system; a¼12.38; b¼7.14; c¼ 12.40 A).

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

New dependences of the important physicochemical properties of solid solutions on basis of the GeSe2 crosscuts of GeSe2–A2B6 (A2 ¼ Hg; Cd; B6 ¼S, Te) on composition are obtained. Thermodynamic characteristics of Cd2GeSe2Te2 and Hg2GeSe2Te2 phases were determined.

161

[3] I. Ansara, A.T. Dinsdale, M.H. Rand (Eds.), Thermochemical Database for Light Metal Alloys, vol. 2, COST 507, Luxembourg, 1998. [4] J. Miettinen, CALPHAD 27 (2003) 263. [5] H.S. Liu, J. Wang, Z.P. Jin, CALPHAD 28 (2004)363.

Md. Mezbahul-Islam, M. Medraj

P. Fima, J. Pstrus, T. Gancarz

New intermetallic compounds in the Mg–Ni–Y system Although the Mg–Ni–Y system is considered to be one of the promising candidates for Ni-metal hydride batteries and an important metallic glass system, it has not been investigated experimentally for the whole composition range. In this work, this system has been investigated experimentally at 400 1C and new findings have been incorporated in a self-consistent database using the CALPHAD approach. Diffusion couples and key alloys have been used to investigate this system. Phase relations and solubility limits have been determined for the ternary phases using EPMA and XRD ¨ to A6, ¨ have been found techniques. Six new ternary compounds, A1 ¨ has been found to have wide solubility in this system where A6 range. The crystal structures of these compounds are being investigated in this work. The binary compounds Ni3Y and NiY are found to have similar solid solubility limits in the ternary system of approximately 2–3 at% where Mg substitutes Y atoms. Whereas, the MgNi2 compound forms ternary solubility where Y substitutes Ni atoms with maximum solubility of approximately 4 at%. The isothermal section of the Mg–Ni–Y phase diagram at 400 1C has been constructed using the current experimental results and compared with the literature. The constituent binary systems have been modelled using the modified quasichemical model for the liquid phase. The ternary compounds have been included in the modelling and a self-consistent thermodynamic database has been constructed for this system. The ternary intermetallic compounds with homogenity range have been modelled using the sublattice model.

Thermodynamic and thermophysical properties of Sn–Zn–In alloys Worldwide studies on Pb-free replacement for traditional solders lead to two groups of alloys that can be used commercially. One of them is SAC alloys, whereas the others are alloys based on Sn–Zn eutectic. The great advantage of Sn–Zn eutectic alloy is that its melting temperature (471 K) is close to the melting temperature of Sn–Pb eutectic solder (456 K). The aim of this work is to study the effect of indium on thermodynamic and thermophysical properties of three Sn–Zn–In alloys of varying In content. Phase transitions and melting temperature were studied by means of the DSC technique. It was found that the addition of indium lowers melting temperature. The electric resistivity was studied by four points technique at 298 K. It was found that electric resistivity of Sn–Zn–In alloys is lower compared to Sn–Zn eutectic alloy. The electric resistivity studies over 298–423 K temperature range as well as thermal expansion studies by dilatometric technique over 223–423 K are planned. Acknowledgement This work is financed under the project ‘‘POIG.01.01.02-00015/09-00 – ZAMAT – Advanced materials and technologies of their production—a task 4.300 , in the years 2010–2013. R.N.C. Siqueira, R.R. Avillez, A.M.S. Gomes, E.G.M. Silva

P. Broz, G. Vassilev, V. Gandova, J. Bursı´k Study of phase transformations and phase equilibria in the Ni– Sn–Zn system Electronic, automotive, aerospace and other industries cannot be realized without solder materials. In the last decade a big effort has been made for replacements of lead containing solders which pose a serious threat from their toxicity point of view. One of the potential lead-free replacements is materials containing Ni, Sn and Zn. The knowledge on the phase transformations in the Ni–Sn–Zn system is, therefore, very useful. In the present work phase transformation and phase equilibria studies of the Ni–Sn–Zn system were performed by means of thermal analysis and analytical electron microscopy in the region of potential leadfree solder technology. The results together with previously published ones [1,2] will be used for a future CALPHAD phase equilibria type modeling based on a thermodynamic prediction of phase equilibria for binary subsystems [3–5] which is only available at the moment. This work has been supported by the Ministry of Education of the Czech Republic under the project OC09010 (COST MP0602), MSM0021622410 and the University of Plovdiv (NPD) project RS09HF-023. References [1] J. Chang, S-K. Seo, H.M. Lee, J. Electron. Mater. 39 (2010) 2643. [2] V. Gandova, P. Broz, J. Burˇsı´k, G. P. Vassiler, Thermochim. Acta, 524 (2011) 47–55.

Heat capacity at low temperatures and molar entropy of stoichiometric Al2MnO4 The heat capacity at constant pressure of the Al2MnO4 spinel was obtained by relaxation calorimetry on the range between 2 and 300 K. The data indicate the occurrence of a reversible transition around 33 K. The integration of the heat capacity from 0 to 298.15 K resulted in an entropy value equal to 116.4 J/mol K at 298.15 K (25 1C). This value is significantly greater than previous published values that did not consider the entropy associated with the low temperature transition anomaly detected at very low temperatures. The molar entropy associated with the transition was estimated to be 8.170.3 J/mol K. To analyze possible effects due to the crystalline structure, X-ray diffraction experiment was performed with synchrotron radiation over the same temperature range. The thermal dependence of the lattice parameter was computed and the reasons for the heat capacity anomaly are discussed. S. Shimenouchi, I. Ohnuma, T. Omori, R. Kainuma, K. Ishida Experimental investigation of phase equilibria at low temperatures ( o600 1C) in the Fe–Ni binary system. In the Fe–Ni system, interesting phase equilibria at low temperatures below 600 1C have been proposed by experimental studies and predicted by thermodynamic calculations. For instance, the existence of the Nishizawa horn along the Curie temperature of the fcc (Fe, Ni) solution phase and that of the ordered FeNi (L10) phase were suggested by experimental

162

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

investigation on meteorites [1] and predicted by the CALPHAD [2] and ab-initio [3] calculations. Furthermore, it was reported that the solubility of Ni in the a-Fe shows ‘‘retrograde solubility’’, i.e., the Ni solubility increases with decreasing temperature below A3 point of pure Fe and starts decreasing again below 500 1C until the invariant reaction, g-(Fe, Ni)Z a-Feþ g0 -Ni3Fe (L12). The solubility of Ni in the a-Fe at low temperatures below 500 1C was determined by Goldstein et al. [4], Romig et al. [5] and Zhang et al. [6], in which small particles of g-(Fe, Ni) precipitated in a0 -Fe martensite matrix were examined by XRD, EPMA, TEM, etc. However, it seems that extra energies due to the martensite structure might cause the deviation of the solubility limit from the true equilibrium composition, especially at low temperatures. In this study, extremely deformed powders of the Fe–Ni alloys were equilibrated at low temperatures below 600 1C and phase equilibria were determined by a FE-EPMA (JEOL:JXA-8500F) with high special resolution (o1 mm). Gas-atomized powders of Fe–Ni alloys were prepared and deformed extremely by converge milling method. Each powder sample was encapsulated in an evacuated quarts capsule and equilibrated at temperatures between 400 and 700 1C for various durations up to 3 months. After that, each powder sample was molded in a conductive resin, mechanically polished, and finished by a vibratory polisher with 0.1 mm diamond paste. Microstructure of cross section of the powder samples was examined and equilibrium compositions were determined by the FEEPMA with low accelerating voltage of 6 kV. Dual-phase structures consisting of strain free particles of the a-Fe and g-(Fe,Ni) phases, whose grain size reached up to 1 mm, were obtained after heattreatment above at 400 1C for 3 months. Results of experiments suggest that the equilibrium compositions of the g-(Fe,Ni) are in consistent with the previous data in literatures. However, it was confirmed that the Ni solubility in the a-Fe increases monotonically with decreasing temperature, which suggests that the ‘‘retrograde solubility’’ seems to be incorrect in the Fe–Ni system. References [1] C.-W. Yang, D.B. Williams, J.I. Goldstein, J. Phase Equilib. 17 (1996) 522. [2] I. Ohnuma, R. Kainuma, K. Ishida, CALPHAD and Alloy Thermodynamics, P.E.A. Turchi, A. Gonis, R.D. Shull (Eds.), TMS, 2002, p. 61. [3] T. Mohri, Y. Chen, Y. Jufuku, CALPHAD 33 (2009) 244. [4] J.I. Goldstein, R.E. Ogilvie, Trans. Metall. Soc. AIME 223 (1963), 2083. [5] A.D. Romig Jr., J.I. Goldstein, Metall. Trans. A 11A (1980) 1151. [6] J. Zhang, D.B. Williams, J.I. Goldstein, Metall. Trans. A 25A (1994) 1627.

1300 1C for 36 h. A cylindrical hole with a diameter of 1 mm was formed by drilling near the diffusion zone of the Pt/Fe DC, and some Al chips were inserted into the hole. DTs were finally obtained by diffusion-annealing at 1200 1C for 18 h and at 1000 1C for 84 h in evacuated quartz tubes. The obtained DTs were chemically analyzed using an EPMA. According to experimental results on the Pt–Fe–Al system, the solubility of Pt for the b phase (B2) in the Fe–Al system reaches about 50 at% at 1200 1C, which decreases rapidly with decreasing temperature down to 5 at% at 1000 1C. On the other hand, in the Pt–Fe–In system, binary compounds, g0 -Pt3Fe (L12) and g0 Pt3In (L12) make continuous solution, but no b phase (B2) was obtained. Furthermore, by microstructure examinations, martensite phases were identified in the b phase of the Pt–Fe–Al alloys and in the g0 phase of the Pt–Fe–In alloys, and iso-Ms(¼R.T.) concentration lines for the martensitic transitions were successfully determined. T.F. Villela, V.M. Chad, F. Ferreira, G.C. Coelho, C.A. Nunes, N.David, J.-M. Fiorani, M. Vilasi Thermodynamic modeling of the Cr–Si–B system and experimental evaluation of critical points in the Cr-rich region Materials with multiphase microstructures belonging to TM–Si–B systems (TM¼transition metals) are good candidates for high-temperature structural applications. These alloys may be used both as substrate or coating materials. This work presents experimental results and thermodynamic modeling of the Cr–Si–B ternary system. Our own experimental data regarding the liquidus projection and the 1200 1C isothermal section, both in the Cr-rich region, were used as the basis for the thermodynamic modeling. Binary and ternary samples were produced by arc-melting high-purity pieces of Cr, Si and B, followed by heat treatment at 1200 1C for 200 h. The samples were characterized by X-ray diffraction, scanning electron microscopy and electron microanalysis. The polymorphic transformation aCr5Si3– bCr5Si3, which according to Chang [1] occurs at 1505 1C, has not been verified by differential thermal analysis experiments performed up to 1550 1C. No thermodynamic modeling for this ternary system was found in the literature. The binary coefficients of Coughanowr, Ansara and Lukas [2] for the Cr–Si, Campbell and Kattner [3] for the Cr–B and Fries and Lukas [4] for the Si–B were used in the optimization of the ternary system. The model adopted for aCr5Si3 (T1) was the solution with three sublattices (Cr)4(Cr)1(Si,B)3 and for Cr5B3 (T2) was the solution with three sublattices (Cr)5(B,Si)2(B)1. The substitutional and interstitial models were used to describe the solubility of Si and B in the BCC structure, respectively, resulting in a solution with two sublattices (Cr,Si)1(B,Va)3. The experimental data are well reproduced by using the coefficients of the present optimization. References

T. Miyamoto, I. Ohnuma, K. Ishida, R. Kainuma Combinatorial technique for phase equilibria determination of Pt–Fe–X (X ¼Al, In) sytems In general, determinations of phase equilibria consume much time, manpower, money, etc. Therefore, it is important to pursue effective experimental methods to find out not only phase equilibria but also martensitic and magnetic transition boundaries in a simultaneous procedure. Combinatorial technique, which performs on diffusion triples (DT) to determine equilibrium compositions and the transition boundaries, has great potential to realize such as ecoexperiments. In this study, applications of the combinatorial technique for phase equilibria determination in novel shape memory Pt–Fe–X (X¼Al, In) alloys is introduced. A Pt/Fe diffusion couple (DC) prepared by clamping at 1000 1C was first diffusion treated at

[1] Y.A. Chang, Trans. Metall. Soc. AIME 242 (1968) 1509–1515. [2] C.A. Coughnanowr, I. Ansara, H.L. Lukas, COST507 Database for Light Metals Alloys. [3] C.E. Campbell, U.R. Kattner, CALPHAD 26(3) (2002) 477–490. [4] S.G. Fries, H.L. Lukas, COST507 Database for Light Metals Alloys. T. Gancarz, P. Fima, J. Pstrus Thermodynamic and physicochemical properties of Zn–Al–In alloys Zn–Al–In solders containing 6 wt% Al and 1–3 wt% In are developed for ultra high temperature applications. The solder is designed to have a liquidus temperature between 650 and 660 K. The Al content improves electrical resistivity, but the In content is

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

supposed to lower melting temperature and improve the spreadability. The aim of this work is to study the effect of indium on thermodynamic and physicochemical properties of three Zn–Al–In alloys of varying In content. Phase transitions and melting temperature were studied by means of the DSC technique. It was found that the addition of indium to Zn–Al eutectic alloy decreases their melting temperature. The electric resistivity was studied by four points technique at 298 K. It was found that electric resistivity of Zn–Al–In alloys is higher compared to Zn–Al eutectic alloy. The electric resistivity studies over 298 to 423 K temperature range as well as thermal expansion studies by dilatometric technique over 223 to 423 K are planned. Acknowledgement This work in financed under the project ‘‘POIG.01.01.02-00015/09-00 – ZAMAT – Advanced materials and technologies of their production—a task 4.300 , in the years 2010–2013. Y. Yu, C.P. Wang, X.J. Liu, R. Kainuma, K. Ishida Experimental confirmation and thermodynamic analysis of existence of fcc-type miscibility gap at high temperatures in some Cu–Ni base alloy systems Spinodal-hardening is one of the most important strengthening mechanisms in the Cu–Ni alloys, and is largely related to the facecentred cubic (fcc)-type miscibility gap (MG). The fcc-type MG in the Cu–Ni alloy system has been predicted to be at temperatures lower than 354.5 1C, suggesting to use spinodalhardening in the Cu–Ni alloys is limited below 354.5 1C. Thus, to increase the existing temperatures of the fcc-type MG in the Cu–Ni base alloy systems by the addition of a third alloying element, is of both technical importance and academic interest. The present work systemically deals with the experimental confirmation of the fcc-type MG at high temperatures in the Cu–Ni–X (X¼ Mo, V, W, Nb, Ta) alloy systems by equilibrated alloy and diffusion couple methods, and thermodynamically analyzes the formation of the fcc-type MG at high temperatures in the Cu–Ni–X (X¼Mo, V, W, Nb, Ta, Cr, Fe, Co) alloy systems by using the CALPHAD (calculation of phase diagrams) method. As an example, the calculated isothermal section of the Cu–Ni–V system at 1100 1C, is shown in Fig. 1(a), where a fcc-type MG (fcc1þfcc2) exists in the Cu–Ni rich side. Typical fcc and (fcc1þfcc2) microstructures were observed in the Cu47Ni47V6 (at%) (Fig. 1(b)) and Cu46Ni46V8 (at%) (Fig. 1(c)) alloys equilibrated at 1100 1C, respectively.

163

A. Debski, ˛ W. Ga˛sior, Z. Moser, Ł. Major, R. Major Formation enthalpy of the BLi and B3Li intermetallic compounds by solution calorimetric method Solution calorimetric method was used for the determination of formation enthalpy of two intermetallic phases from the boronlithium system. Preparation of the BLi and B3Li phases was conducted by direct reaction of boron and lithium at 773 K. Next, the alloys obtained were short time annealed (about one hour) at 1073 K followed by three days annealing at 873 K. Samples of phases were analyzed by the X-ray diffraction and transmission electron microscopy. The results show that the assumed phases were formed. Attempts of dissolution of the boron in Cu, Al and Se were undertaken however without success, because the long time was necessary for solution of boron in mentioned liquid solvents. The tests of solution of studied intermetallic phases in water were conducted and it was found that they fine dissolved in short time (about dozen minutes) and that the boron accumulated at the bottom of glass vessel. So, the water was chosen as the solvent for BLi and B3Li intermetallic phases. Calibration of the calorimeter was performed by dissolution of NaOH, LiOH, KMnO4 and ice in water and next the calorimeter was tested by measurement of the formation enthalpy of CaLi2 which was earlier measured by the dissolving it in Al and Sn bath (solvent). The results obtained were very similar for three different bathes so, the experiments with the solution of boron–lithium phases in water were undertaken. The heat effects of solution of Li, BLi and B3Li were measured and the formation enthalpy of the phases was calculated. The calorimetric studies of B–Li intermetallic phases have been carried out in the frame of research program advanced materials and their production technologies (ZAMAT) POIG.01.01.02-00-015/ 09-00.

W. Gasior, W. Zakulshi, A. Debski Phase relations in the Ca–Li system The Ca–Li alloys are a candidate for safe and ecological hydrogen storage as an energy source for the application in the automobile industry. Additionally that system plays a very importante role in the aluminum and magnesium industries. The lack of thermodynamic data for the Ca–Li system both for the activities and for he heats of mixing in the liquid state was clearly emphasized in Bale, Pelton [1] assessment, as well as afther 20 years in the paper of Groebner, Schmid-Fetzer et al. [2] where the authors intended to model the thermodynamics and

Fig. 1.

164

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

phase relations in he ternary Ca–Li–Mg system. The lack of the entire thermodynamic description of that system is not an effect of accidental negligence but is related to the extreme experimental difficulties, due to the high chemical reactivity of calcium and lithium with oxygen, hydrogen, nitrogen and the moisture. References [1] C.W. Bale, A.D. Pelton, Bull. Alloy Phase Diag. 8 (2) (1987) 125–127 [2] J.Groebner, et al, Thermochim. Acta 389 (2002) 85–94 3.6. First principle calculations C.G. Schon, R. Arro´yave On the cluster expansion method The cluster expansion method (CEM), provides a simple and efficient way of predicting alloy phase diagrams using ab-initio calculations. It combines the calculation of a set of stoichiometric compound’s electronic structures at the ground state (T¼0 K) using standard methods and a statistical mechanics formalism to extrapolate the thermodynamics of the system at finite temperatures. The method corresponds to a Fourier expansion in the configuration space, possessing thus important mathematical (topological and functional analytic) properties [1]. A key issue in the CEM is the choice of the truncation point, or the basis used, for the expansion (i.e., which compounds should be added). There are currently two approaches in use in the materials science community. One based on an algorithm developed by van de Walle and Ceder [2] requires introducing as many compounds as necessary for the self-consistent description of the energy of all possible configurations in the system (the so-called converged expansion). This approach requires introducing a large number of compounds (typically over 50) and the converged basis is systemdependent. The second approach is based on the choice of a small truncation point (a ‘‘basic cluster’’) which is common to all systems of a given class (e.g., the irregular tetrahedron cluster in bcc lattices, see for example [3]). The first approach is mathematically sound, but requires the use of very specific computational tools. The second one is more flexible, but the expansion is not converged. There are evidences, however, suggesting that the main thermodynamic properties of a bcc system are already reproduced in the irregular tetrahedron approximation [4], hence the convergence of the expansion should only contribute with second or higher-order corrections over these results. We compare here ab-initio calculations for the bcc systems Co–Ga and Ni–Ga using both approaches, allowing the evaluation of their limitations

new lead-free soldering materials. A particular motivation for the current research efforts is the lack of experimental information on the properties which are involved in the usual thermodynamic (CALPHAD-type) assessments, or in studies of the thermophysics of the various types of IPs. A possible way of facing these challenges would be to build a rather general database with consistent information on various kinds of physico-chemical properties. The purpose of the present contribution is to present a progress report on the development of a theoretical database of such characteristics using a densityfunctional-theory ab initio method implemented in the VASP code [1]. We present the calculated molar volume, bulk modulus and its pressure derivative, the energy of formation from the constituent elements and the electronic density of states of a variety of stable and metaestable Me–X IPs with Me¼Cu,Ni and X¼In,Sn. The results are compared with the available experimental data and previous ab initio results [2–5]. We also establish trends as a function of composition, which should be useful both in alloy design work and in the understanding of the phase-stability systematics. References ¨ G. Kresse, J. Furthmuller, Comput. Mater. Sci. 6 (1996) 15. G. Ghosh, M. Asta, J. Mater. Res. 20 (2005) 3102. G. Ghosh, Metall. Mater. Trans. 40A, (2009) 4. S. Ramos de Debiaggi, G.F. Cabeza, C. Deluque Toro, A.M. Monti, S. Sommadossi, A. Ferna´ndez Guillermet, J. Alloys Compd., 509 (2011) 3238–3245. [5] C.E. Deluque Toro, S. Ramos de Debiaggi, A.M. Monti, Phys. B: Condens. Matter, 407 (2012) 3236–3239. [1] [2] [3] [4]

F. Kormann, A. Dick, T. Hickel, J. Neugebauer

Integrating finite temperature magnetism into ab initio free energy calculations The Gibbs free energy of a system is a fundamental quantity for predicting phase diagrams, finite temperature materials parameters, or kinetic barriers. An ab initio derivation of it makes a highly accurate evaluation of all excitation processes mandatory. One of the most challenging – but for many engineering materials crucial – contribution is coming from the magnetic degrees of freedom. We have developed an approach that eliminates shortcomings of conventional approaches and that is firmly based on an exact solution of an effective Heisenberg model employing the quantum Monte Carlo approach (QMC). We demonstrate the high accuracy achievable by the new approach by computing magnetizations, magnetic heat capacities and free energies for the magnetic pure elements Fe, Co, and Ni, and by extending it to magnetic compounds such as cementite (Fe3C). Generally, an excellent agreement with experimental data is found.

References [1] D. DeFontaine. Solid State Phys. 47 (1994) 33–176. [2] A. van de Walle, G. Ceder. J. Phase Equilib. 23 (2002) 348–359. [3] N. Sodre´, et al. CALPHAD 33 (2009) 576–583. ¨ G. Inden. Acta Mater. 4 [4] C.G. Schon, C.E.D. Toro, S.R. Debiaggi, G.F. Cabeza, A.F. Guillermet Ab initio study of thermodynamic, structural and electronic properties of stable and metastable compounds in the Me–X systems (Me ¼Cu, Ni; X ¼In, Sn) The physico-chemical properties of Me–X (Me¼Cu, Ni; X¼In, Sn) based intermetallic phases (IPs) and related compounds have received considerable attention in connection with the search for

H. Kim, J.-Y. Noh Ab initio calculations on the effect of Mn substitution in Fe3AlC The dispersion of k-carbide, (Fe,Mn)3AlC, within austenic matrix is expected to enhance the mechanical properties such as the ductility and the stability of alloys. The stoichiometric (Fe, Mn)3AlC has perovsktietype structure (E21) in which iron (or manganese) atoms are located at the face centers, aluminium atoms are at the corners of the cube, and the carbon atoms take the body-center site of the cube. In this work, we have performed the density functional theory calculations using the projector augmented wave (PAW) potentials and the Perdew–Burke–Ernzerhof generalized gradient approximation (GGA) for the exchangecorrelation functional. We calculated

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

structural, magnetic, elastic, energetic, and electronic properties for both crystalline and supercell structures and investigated the changes in physical properties of k-carbide accompanied by the varying contents of Mn. The crystalline Fe2MnAlC containing around 30% of Mn has the largest formation energy indicating that it is energetically most favorable. To understand the initial Mn alloying process, we substituted one Mn atom for one Fe atom in the 2  2  2 and the 2  2  4 supercell of Fe3AlC. We will present the Mn substitution induced changes in physical properties by analyzing the electronic structures. This work has been supported partly by the POSCO, Inc. and partly by the Strategic Supercomputing Support Program form Korea Institute of Science and Technology Information (No. KSC2008-S02-0010). I. Polishuk Hybridizing SAFT and cubic EOS: What can be achieved? This study deals with creating a concept of the hybrid model gathering the advantages of both cubic EOS and SAFT approaches. The proposed idea is revision of the Chapman’s et al. SAFT by addressing the problem of the numerical pitfalls and the issue of the space available for dispersive interactions with further attaching the SAFT part by the cubic EOS’s cohesive term. It is demonstrated that the resulting model on one hand preserves the characteristic for SAFT accuracy in estimating the liquid compressibility, and on the other one—the characteristic for cubic equations capability of simultaneous modeling of critical and subcritical data. Moreover, on the basis of the comprehensive set of thermodynamic properties of 8 challenging for modeling compounds (including n-hexatriacontane, water and methanol) it has been demonstrated that the proposed EOS has an over-all superiority comparing even to one of the most successful versions of SAFT, namely the SAFT-VR-Mie. These results indicate that tracking the trends established by experimental data and using the century-long experience with developing semi-empirical engineering models might sometimes be preferred over the exclusive relying on advanced molecular theories.

165

Fig. 1. shows two atomistic structure models of Cu3Si based on the Zintl phases [1]; one is the D03 type structure called ‘twisted’, and the other is arranged on the position of Si sites called ‘parallel’. The reported stable intermetallic structure of Cu7Si2 is also calculated. The energy calculations with the outer and inner relaxes of atomistic structures are performed by the VASP (Vienna ab-initio simulation package) code.

Fig. 2. shows that two Cu3Si phases are both more stable than the segregation limit of pure Cu and Si. Although the experimentally observed precipitates show the very close structure with the Cu3Si twisted model, the parallel Cu3Si is more stable than the twisted one. The outer shape of the parallel model, however, is very irregular, and its c-axis is 41.2% larger than the a-axis. Both models of Cu3Si show no band gap, which indicates that the phases are conductor.

˜ o, N. Sodre´, H.M. Petrilli C.G. Schon, P.G. Gonzales-Ormen

˜ o, N. Sodre´, H.M. Petrilli, C.G. Schon P.G. Gonzales-Ormen

Ab-initio calculation of the BCC Fe–M (M ¼Ti, V, Cr, Zr, Nb, Mo): New results on a correlation between compound stability and magnetism The metastable phase diagrams of the bcc-based ordering equilibria in the Fe–M (M¼Ti, V, Cr, Zr, Nb, Mo) systems have been calculated by the cluster expansion method, through the combination of full potential-linear augmented plane wave (FP-LAPW) electronic structure calculations and cluster variation method (CVM) thermodynamic calculations in the irregular tetrahedron approximation. In spite of being neighbors in the periodic table, these six alloying elements result in radically different phase diagram topologies when alloyed with iron. This result contrasts with usual regularities (e.g., the Hume–Rothery rules) typically observed in other intermetallic compounds. The results are discussed considering the electron density of states (DOS) functions of the B2–FeM and D03–Fe3M compounds in the six systems and correlating their Formation Energies and the stability against the disordering reactions (B2A2–FeþA2–M and D03–Fe3M-A2–FeþB2–FeM) with the total magnetic moment. A linear correlation between compound stability and total magnetic moment such that the highest magnetic momenta lead to the lowest compound stabilities.

Ab-initio calculation of the metastable bcc Al–M (M ¼Sc, Ti, V, Cr, Y, Zr, Nb, Mo) phase diagrams Transition metal aluminides, in particular those based on stable ordered intermetallic phases of the Ni–Al, Ti–Al and Fe– Al systems, are currently considered as good candidates the development of technological materials for structural applications involving high temperatures in aggressive environments. Technological materials, however, rarely consist of simple binary or even ternary systems, on the contrary, most of them consists of multicomponent thermodynamic systems with a large number of components. The present work is part of a major effort to build a multicomponent thermodynamic database dedicated to the description of the bcc phase in transition metal aluminides using ab-initio data, combining full potential-linear augmented plane wave (FP-LAPW) electronic structure calculations and the cluster variation method (CVM) in the irregular tetrahedron approximation. Additionally important informations are gained about the thermodynamics of the bcc phase in the high-aluminum limit, hence in the limit of aluminum alloys, for which there is a natural lack of experimental informations.

166

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

R. Taniguchi, S.R. Nishitani, Y. Ohno, I. Yonenaga First principles calculations of the copper silicide precipitates Very recently, the authors have reported the experimental observation of the precipitates of Cu3Si in a heavily copper doped Si. In this research, the energetic assessment of this precipitation behavior has been investigated by the first principles calculations (Figs. 1 and 2). The other phase of Cu7Si2 is also more stable than the segregation limit. The tie line between pure Cu and Cu7Si2 drawn by the dashed line indicates that the twisted model of Cu3Si is metastable to Cu þ Cu7Si2, and the parallel model of Cu3Si is most stable. Because the twisted Cu3Si shows cubic shape and the good lattice coherency with Si, it would appear as the structure of the precipitates. References [1] N.E. Christensen, Phys. Rev. B 32 (1) (1985) 209–228.

V. Kulish, M.-F. Ng, Z. Chen, P. Wu First-principles study on Si clusters supported on carbon nanotubes (CNTs) and graphene Recently silicon has attracted a great attention among the novel anode materials for Li-ion batteries, because it has the highest known theoretical capacity (4200 mA h/g in Li22Si5 phase) and it is the second most abundant material on the Earth. The practical application of Si anodes, however, is hindered by the enormous volume changes ( 4400%) occurring in Si during lithiation, which eventually lead to the cracking of the anode material and large capacity fade. An attractive strategy to overcome the volume expansion problem is to build Si-based nanocomposites or hybrid materials. In both approaches, the choice of the secondary material and the strategy for the materials design play an important role. In the present study, we use density functional theory (DFT) to study hybrid nanostructures, consisting of Si-clusters adsorbed on the carbon nanotubes (CNTs) and graphene. We study the effects of Si cluster size and orientation, as well as the effect of support on the properties of hybrid systems. We first examine the structural stability of Si/CNT and Si/graphene systems. Adsorbed clusters exhibit different energetically favourable structural configurations, while the most stable one is the cluster orientation with three Si atoms, adjacent to CNT/graphene. We further investigate the effect of Si-Si spacing on the stability of Si/CNT complex. From the analysis of electronic structures, it is found that adsorption of Si clusters does not change the electronic properties of the support. Moreover, the density of state (DOS) of Si/CNT system shows more states around the Fermi level than in the pristine CNT. This indicates that Si/CNT should have higher conductivity than its standalone parts (Si and CNT), which is beneficial for the Li-ion battery application.

enthalpy and heat capacity of these grain boundaries are studied in terms of quasiharmonic approach with the vibrational contribution to Helmholtz free energy described by the Debye model. The newly developed approach is used to predict the grain boundary structures with arbitrary orientations. Y. Masaki, M. Sugimoto, Y. Yamamoto, Y. Miyamoto, S.R. Nishitani First principles calculations of 18R Mg alloys Long periodic stacking ordered (LPSO) Mg alloys show excellent mechanical properties, and the the mechanism of such a query microstructure formation is the one of the key issues of the newly developing light-weight Mg materials. The LPSO Mg alloys, its typical composition of Mg97Zn1Y2, show 18R structure with the long period stacking sequence of the mixture of hexagonal and cubic layers and added Zn and Y atoms are enriched in cubic layers. In this research, we have investigated the energetic assessment of these structures by the first principles calculations in order to reveal the mechanism of LPSO formation in Mg alloys. We have performed the atomistic model constructions using MedeA and the structure energy calculations using Vienna Ab Initio Simulation Package (VASP) code. The cut-off energy was 600 eV and the k-point meshes were determined from the shapes of models Figs. 1 and 2. Y. Yamamoto, S.R. Nishitani, T. Kaneko First principles calculations on vibration free energies of SiC polytypes SiC has been recently focused on the next generation materials for the power electronic devices due to its excellent physical and chemical properties. SiC shows many polytypes such as, 3C, 4H and 6H. The phase stabilities of these polytypes are still under discussions. The novel crystal growth method of SiC polytypes, so-called metastable solvent epitaxy (MSE), has recently been developed by the authors, in which a similar mechanism to that of the high-pressure synthesis of diamond is used. The driving force of the epitaxial growth of stable 4H–SiC is the metastability of the polytype of SiC. The stability of the polytypes have been explored by the first principles calculations with the phonon free energy, but the difference is very small. In this research, the authors investigate the accuracy of the calculations. The VASP (Vienna ab-initio simulation package) code has been used. After adjusting

W.Y. Wang, H.Z. Fang, S.L. Shang, Y. Wang, Z.K. Liu, S. Mathaudhu Predicting grain boundary structures and properties from ab initio molecular dynamics and first-principles calculations A solid–liquid interface model is developed to predict grain boundary structures through ab initio molecular dynamic calculations. Using the coincidence site lattice model for Cu, the twinning S3[011]() and S5[011](210) grain boundary structures are created and their related energies from our calculations match well with the previous reported results. Additionally, entropy,

Fig. 1. Shows the energies of the stacking sequence ratios between pure hexagonal(2H) to pure cubic (3C) stacking sequences. The 18R structure locates on the line between 2H and 3C, and either do the other structures, which indicates that the inter-layer interactions are short and good estimations of linear combination between hexagonal and cubic stacking energies.

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

167

Fig. 2. Shows the energy differences between cubic and hexagonal structures of the different sizes with the Zn and Y additives. The energies are almost linearly dependent on the system size, which are expected from the linear dependency among 2H and 3C. The exceptions are observed in 24 atoms models with Y and Yþ Zn, which indicate that these structures are relatively more stable than the simple averages. The other detailed discussions will be given on the poster. We will also report the results of Monte Carlo simulation of atomistic level LPSO formations.

some parameters of the VASP code, the major controlling parameter on the total energy has been appeared to be the cut-off energy. Fig. 1 shows the cutoff energy dependence of the total energy of 3C–SiC; the energies approach to a value of 1000 eV or above of the cut-off energy. Fig. 2 shows the temperature dependency of the phase stabilities of 3C–SiC measured from 4H–SiC. Two lines calculated by the cut-off energies of 600 and 400 eV are shown by the solid and dashed lines, respectively. The dashed line with the cut-off energy of 400 eV shows that 4H–SiC is more stable than 3C–SiC on the entire temperature range. The solid line with the cut-off energy of 600 eV shows

Fig. 1. Cut-off energy dependence of the total energy of 3C–SiC.

(i) 3C-SiC is stable at the lower temperatures, (ii) 4H-SiC is stable at the higher temperatures, and (iii) the transition temperature is located around at 1000 K. These results are consistent with the experimental ones. O. Malyi, Z. Chen, P. Wu Effects of sulfur and nitrogen impurities on the properties of the solid oxide fuel cell anode materials Solid oxide fuel cells (SOFC) are one of the most promising energy conversion devices that produce electricity directly from oxidizing a fuel. However, its high cost is a significant limitation to its residential and commercial applications if pure hydrogen is used as the fuel. As a result, the search of cheaper SOFC materials and cheaper fuels becomes vital. In the past few years, various types of fuels have been considered as alternative fuels to hydrogen, and it has been shown that impurities (carbon, sulfur, nitrogen etc.), contained in them, may not only change SOFC performance significantly but also lead to the degradations of SOFC performance. Therefore for the development of new SOFC anode materials the studies of the effect of impurities on the SOFC materials are needed. Based on an ab initio density functional theory, we evaluate the effects of sulfur and nitrogen impurities on the properties of SOFC anode materials. It is observed that sulfur and nitrogen impurities have tendency to place at the zirconia surface that may have significant effect on the stability of metal/zirconia interfaces, and as a result, on the particle

Fig. 2. Temperature dependency of the phase stabilities of 3C–SiC measured from 4H–SiC. The solid and dashed lines were calculated with the cut-off energies of 600 and 400 eV, respectively.

agglomeration rates. The study of the effect of sulfur impurity on the stability of metal/c-ZrO2 interfaces has proposed an explanation for the experimentally observed degradation of SOFC performance in case of using sulfur-containing fuels (at low concentration of sulfur impurity in a fuel gas) and Ni/YSZ anode materials. Finally, based on the first principle studies of the effects of sulfur and nitrogen impurities on mixed conductor properties

168

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

we suggest criteria for the intelligent design of sulfur/nitrogen tolerant anode materials. 3.7. Modeling and fundamental aspects J. Romanowska Predicting thermodynamic properties of ternary systems by means of finding a function value inside a Gibbs triangle This paper presents a new numerical approach to modeling of ternary systems on the basis of thermodynamic properties of binary systems included in the investigated ternary system. Ternary interaction parameters in the Muggianu extension of the Redlich–Kister formalism are calculated numerically by use of the Excel program and Solver. Unless all binary parameters are known, the idea of calculating parameters can be regarded as solving an equation, where all boundary conditions (binary ij alloys) are known. This approach is contrary to finding a function value outside a certain area, if the function value inside this area is known. The approach proposed in this paper is as follow: if we know all boundary conditions, values on all legs of the Gibbs triangle and a function inside the triangle, we find a function value inside the triangle. Results of calculations were compared with the literature data for Ag–Au–Bi, In–Sn–Zn and Cu–Fe–Sn systems.

J. Vˇresˇta´l, J. Pinkas, J. Sopousek, P. Broz, J. Bursı´k Calculation of nanoalloys phase diagrams Calphad-type calculation is most effective and useful tool for practical applications because thermodynamic data for various multicomponent systems are available. First-principles approach is restricted to the several hundred atoms only and molecular dynamic simulations have a limitation when applied to multi-component systems. Calphad-type calculations make us possible to take surface energy of nanoparticles into account and to calculate phase stability in this situation. Procedure of including surface dependent parameters of pure components and of alloys into Calphad modeling was explained on the model system Ag–Au [1]. In the present contribution, the Cu–Ni and Ag–Cu systems were used as a system with practical importance in electronic industry for calculation of phase diagram of nanoalloys–nanoparticles of different composition. Calculated nanoalloy phase diagram could be verified by calorimetric determination of temperatures of phase transformation of synthesized nanoalloys, characterized by electron microscopy. Acknowledgement Financial support of Ministry of Education, Youth and Sports of Czech Republic under Grant nos. 11046 and MSM0021622410 and Czech Science Foundation of Grant no. 106/09/0700 are gratefully acknowledged.

References [1] Y.-M. Muggianu, M. Gambino, J.-P. Bros, J. Cnim. Phys. 72 (1975) 83–88. [2] O. Redlich, T. Kister, Ind. Eng. Chem. 40 (1948) 345–348. [3] E. Zoro, C.Servant, B. Legendre, CALPHAD 31 (2007) 89–94. [4] Y. Ciu, X.J. Liu, I. Ohunuma, H. Ohtani, K. Ish. 3.8. Nanomaterials A. Zemanova, J. Bursı´k, J. Sopousek, J. Vˇresˇta´l, P. Broz Calculation of Ag–Sn phase diagram including the size effect CALPHAD approach is a very useful technique for calculation of phase diagrams of bulk materials based on thermodynamic database containing data such as chemical potentials of pure substances and excess Gibbs energy of mixtures as a function of composition, temperature and pressure. In order to extend the use of CALPHAD approach to small metallic particles on submicron and nano scale, due to the surface effect, the chemical potentials and the excess Gibbs energy should be expressed with an additional parameter: the particle size. In this study, the nanoalloy Ag–Sn binary phase diagram was calculated by using the CALPHAD method. Silver–tin alloy is one of the promising alternatives for Sn/Pb solders. The calculated nanoalloy phase diagram was verified using experimental data from the literature and also the data from own experimental work. Acknowledgement Financial support of the Czech Science Foundation (grant No. 106/ 09/0700) and of Ministry of Education, Youth and Sports of the Czech Republic (Grant nos. LD11024, 11046 and MSM0021622410) is gratefully acknowledged.

References [1] J. Park, J. Lee, CALPHAD 32 (2008) 135. 4. List of registered participants Abiodun, Jimoh Sakiru Abrikosov, Igor

Hosseinifar, Mehdi

Norgren, Susanne

Hu, Biao

Nunes, Carlos Angelo Ohnuma, Ikuo

Huang, Guoxing ˚ Agren, John Albertsson, Galina Hug, Gilles Alves, Celso Luiz Moraes Antoni-Zdziobek, Annie Araujo, Giselle de Mattos Ardell, Alan J. Avillez, Roberto Ribeiro Bacalhau, Jose´ Britti Baltazar, Anderson W.S. Bamberger, Menahem Barros, Simone Kessler Beilmann, Markus Benes, Ondrej

Ipser, Herbert

Osman, Renata Fortini M. Palumbo, Mauro

Ishida, Kiyohito

Park , Joohyun

Islam, Md. Mezbahul

Park, Jeong-Yong

Jansson, Bo JendrzejczykHandzlik, Dominika Jiang, Min

Pavlu, Jana Peng, Yingbiao

Jin, Liling

Perrot, Pierre

Petrilli, Helena Maria Joubert, Jean-Marc Pimenta, Hamilton Porto Kajihara, Masanori Pinatel,Eugenio Riccardo Kattner, Ursula Polishuk, Ilya Kaufman, Lawrence P. PovodenKaradeniz, Erwin

Rio de Janeiro / CALPHAD: Computer Coupling of Phase Diagrams and Thermochemistry 39 (2012) 111–169

Berthod, Patrice

Kim, Hanchul

Bielefeldt, Wagner Kim, Ingee Viana ¨ Bottger, Bernd Kobertz, Dietmar Broz, Pavel

Konrad, Joachim

Burton, Benjamin

¨ Kormann, Fritz

Cai, Dan

Kozeschnik, Ernst

Campos, Marcos Fla´vio Capurro, Constantino Carboni, Marcelo Carboneri Carvalho, Ricardo Nolasco Chandra, Dhanesh

Kroupa, Ales

Lee, Hyuck Mo

Chen, Qing

Lee, Jee yong

Coelho, Daniel Massari de S. Coelho, Gilberto Carvalho Colinet, Catherine Corvalan, Carolina

Lee, Sung Hoon

Costa e Silva, Andre´ Craidy, Pedro da Silva Crivello, JeanClaude Decterov, Sergei A. Defendi, Guilherme Antonio Du, Yong Du, Zhenmin Duchesne, Marc Dumas, JeanChrisophe Dupin, Nathalie Eckstein, Carlos Bruno El Guendouzi, Mohamed Eleno, Luiz T. F. Fabrichnaya, Olga

Kumar, K.C. Hari Lage, Meire Guimara~ es Lee, Byeong-Joo

Lehmann, Jean

Prostakova, Viktoria Pstrus, Janusz

Farina, Alexandre Bellegard Ferreira, Fla´vio

˜ o, Ramirez London Antonio J. Ramos de Debiaggi, Susana B. Ribeiro, Rafaella Martins Rizzo Assunc- a~ o, Fernando C. Robelin, Christian

Figueiredo, Alfredo L.L. Fima, Przemyslaw Fiorani, Jean Marc Fitzner, Krzysztof Fonseca, Antonio Sergio M. Fridman, Daniel Pallos Fries, Suzana G. Gancarz, Tomasz Gasior, Wladyslaw Gomez-Acebo, Tomas Gueneau, Christine Guimara~ es, Nara Miranda Guo, Cuiping Hallstedt, Bengt Hammerle, Janyne Rodrigues Hauegen, Christien Guisard He, Cuiyun Hodaj, Fiqiri

Rocha, Elias E. de Arau´jo Rogl, Peter Sanchez, Juan M. Santos, Dilson Silva Schmid-Fetzer, Rainer Schneider, Andre

Li, Chonghe

Schon, Claudio Geraldo Selleby, Malin Shimenouchi, Shota Shinagawa, Kazuya

Li, Hongxiao

Shishin, Denis

Lima, Belmira Benedita Lin, Shih-kang Lindahl, Bonnie C.B.

Silva, Raphael Mendes R. Sippola, Hannu Souza, Jean Bruno R.

Lino, Roney Eduardo Liu, Libin Liu, Miao Liu, Xingjun

Stein, Frank Strandlund, Henrik Sundman, Bo Talekar, Anjali

Liu, Zi-Kui Lopes, Aline Aguiar

Taniguchi, Ryo Tedenac, JeanClaude Toffolon-Masclet, Caroline Turchi, Patrice E.A. Ullbrand, Jennifer M.

Leszek, Zabdyr Li, Changrong

Lutsyk, Vasily Maeshima, Takashi Magalha~ es, Humberto L.G.

169

Malakhov, Dmitri V.

Vaajamo, Iina

¨ Markstrom, Andreas

Viana, Cristia dos Santos Vieira, Leandro Nakamura A. Vrestal, Jan Wang, Fuming Wang, Hang Warnken, Nils

Markus, Torsten Masaki, Yoshihiro Matsumiya, Tooru Medraj, Mamoun Medrano, Julio H.P. Meis, Constantin Menichetti, John Mikula, Adolf Miyamoto, Takashi Mohri, Tetsuo

Xiong, Wei

Muggianu, Yves

Yamamoto, Yosuke Yan, Yu Yaqoob, Khurram Zacherl, Chelsey L. Zakulski, Wojciech

Murari, Fabio Dian

Zhang, Ligang

Naraghi, Reza ¨ Neugebauer, Jorg Nicolas, David

Zhong, Yan Zhou, Peng Zhou, Shihuai

Nishitani, Shigeto R.

Noh, Ji-Young

Acknowledgements The XL CALPHAD was organized by ABM. The support of the following companies and organization is gratefully acknowledged: CSN, VALE, V&M do Brazil, CBMM, GERDAU, Villares Metals, PETROBRAS, RHI, SANDVIK, USIMINAS, MRS Logı´stica S.A., Thermo-calc Software AB, FactSage-Thermofact, GTT Technologies, Micress, SGTE and CALPHAD Inc. The support of the following Brazilian agencies is also gratefully acknowledged: CAPES, CNPq and FAPERJ.

A. Costa e Silva n EEIMVR-UFF, Volta Redonda and IBQN, Brazil E-mail address: [email protected] (A. Costa e Silva) F. Rizzo PUC-RIO, Rio de Janeiro, Brazil

Received 24 April 2012

n

Corresponding author.