Search

The enthalpy of mixing in the liquid phase is a thermodynamic property reflecting interactions between elements that is key to predict phase transformations. Widely used models exist to predict it, but they have never been systematically evaluated. To address this, we collect a large amount of enthalpy of mixing data in binary liquids from a review of about 1000 thermodynamic evaluations. This allows us to clarify the prediction accuracy of Miedema's model which is state-of-the-art. We show that more accurate predictions can be obtained from a machine learning model based on LightGBM, and we provide them in 2415 binary systems. The data we collect also allows us to evaluate another empirical model to predict the excess heat capacity that we apply to 2211 binary liquids. We then extend the data collection to ternary metallic liquids and find that, when mixing is exothermic, extrapolations from the binary systems by Muggianu's model systematically lead to slight overestimations of roughly 10% close to the equimolar composition. Therefore, our LightGBM model can provide reasonable estimates for ternary alloys and, by extension, for multicomponent alloys. Our findings extracted from rich datasets can be used to feed thermodynamic, empirical and machine learning models for material development., Comment: Manuscript: 27 pages, 5 figures ; Supplementary notes: 3 pages, 5 figures

An assessment of the Sn unary system is presented. First, the literature on phase equilibria, the thermodynamic properties, the volume and related properties, and shock compression of tin is thoroughly reviewed. Second, the Sn system is investigated by means of synchrotron X-ray diffraction in a diamond-anvil cell up to pressures and temperatures of 57 GPa and 730 K. New information is obtained on the thermal stability and thermal expansion coefficient of the {\gamma} (I4/mmm) and {\gamma}" (Im-3 m) phases. Third, density functional theory calculations are conducted on the six allotropic phases of tin observed in experiments using both a local density approximation (LDA) and a generalized gradient approximation (GGA) functional. This combined experimental and theoretical investigation provides further insights on the pronounced metastable nature of Sn in the 30 - 70 GPa range. Last, a Gibbs energy modeling is conducted using the recently proposed Joubert-Lu-Grover model which is compatible with the CALPHAD method. Special emphasis is placed on discussing extrapolations to high pressures and temperatures of the volume and of the thermodynamic properties. While the description of the heat capacity is approximate at moderate pressure, all available data are closely reproduced up to 2500 K, which is 5 times higher than the atmospheric pressure melting point of tin, and 150 GPa, which is almost 3 times the standard bulk modulus of \b{eta}-Sn., Comment: Accepted Manuscript: 58 pages (double-spaced), 13 figures. Supplementary notes: 12 pages, 7 figures

High-pressure Gibbs energy approaches are promising for establishing multi-component thermodynamic databases. However, so far, they have often been considered unsuccessful, because of their tendency to lead to unphysical extrapolations of the thermodynamic properties at elevated temperatures and pressures. Beyond this symptom, the root causes of the problem are rarely investigated. In this work, it is identified that these shortcomings are caused by (i) an inconsistent treatment of the SGTE method of extrapolation, and (ii) an insufficient knowledge of the models to describe the volume that are the key to extend CALPHAD databases toward high pressures. Because the first step toward solving the problem is to focus on the later issue, several models to describe the volume that are built upon the Grover empirical law are investigated, namely, the Lu-Grover, Joubert-Lu-Grover, Jacobs-Grover, and the Anand-Saxena-Grover approaches. In addition, an original description built upon the concept of thermal pressure, and a new scheme based on the equivalence between temperature and pressure when computing the volume are discussed. For each of these models, limitations and possibilities are identified from a practical standpoint., Comment: Manuscript: 42 double-spaced pages, 8 figures. Supplementary Notes: 13 pages, 6 figures

Machine learning (ML) methods are becoming integral to scientific inquiry in numerous disciplines, such as material sciences. In this manuscript, we demonstrate how ML can be used to predict several properties in solid-state chemistry, in particular the heat of formation of a given complex crystallographic phase (here the $\sigma-$phase, $tP30$, $D8_{b}$). Based on an independent and unprecedented large first principles dataset containing about 10,000 $\sigma-$compounds with $n=14$ different elements, we used a supervised learning approach, to predict all the $\sim$500,000 possible configurations within a mean absolute error of 23 meV/at ($\sim$2 kJ.mol$^{-1}$) on the heat of formation and $\sim$0.06 Ang. on the tetragonal cell parameters. We showed that neural network regression algorithms provide a significant improvement in accuracy of the predicted output compared to traditional regression techniques. Adding descriptors having physical nature (atomic radius, number of valence electrons) improves the learning precision. Based on our analysis, the training database composed of the only binary-compositions plays a major role in predicting the higher degree system configurations. Our result opens a broad avenue to efficient high-throughput investigations of the combinatorial binary calculation for multicomponent prediction of a complex phase.

Boron and Silicon site occupancies of the T2 phase in the Cr-Si-B system were investigated experimentally and by first-principles electronic-structure calculations within the scope of the Density Functional Theory (DFT). A sample with nominal composition Cr$_{0.625}$B$_{0.175}$Si$_{0.2}$ was arc-melted under argon, encapsulated in a quartz-tube and heat-treated at 1400{\deg}C for 96 hours. It was then analyzed using Scanning Electron Microscopy (SEM) and X-Ray Diffractometry (XRD) with synchrotron radiation. An excellent agreement was obtained between experiments and theoretical calculations, revealing that Si occupies preferably the $4a$ sublattice of the structure due to the presence of weak B bonds, making the site preferences a key factor for its stabilization. The results of this work provide important information to support a better description of this phase in alloys with Si and B, since T2 phases are known to occur in many important Transition Metal-Si-B ternary systems, such as Nb/Mo/W/Ta/V-Si-B.

High-throughput calculations are a very promising tool for screening a large number of compounds in order to discover new useful materials. Ternary intermetallic are thus investigated in the present work to find new compounds potentially interesting for thermoelectric applications. The screening of the stable non-metallic compounds required for such applications is obtained by calculating their electronic structure by DFT methods. In a first part, the study of the density of states at the Fermi level of well-known chemical elements and binary compounds allows to empirically optimize the selection criteria between metals and non-metals. In a second part, the TiNiSi structure-type is used as a case-study through the investigation of 570 possible compositions. This screening method leads to the selection of 12 possible semiconductors. For these selected compounds, their Seebeck coefficient and their lattice thermal conductivity are calculated in order to identify the most interesting one. TiNiSi, TaNiP or HfCoP could thus be compounds worth an experimental investigation.

Aiming at increasing the yield strength of transformation and twinning induced plasticity (TRIP and TWIP) titanium alloys, a dual-phase $\alpha$/$\beta$ alloy is designed and studied. The composition Ti 7Cr 1.5Sn (wt.%) is proposed, based on an approach coupling Calphad calculations and classical Bo-Md design tool used in Ti-alloys. Its microstructure is made of 20% of $\alpha$ precipitates in a $\beta$ matrix, the matrix having optimal Bo and Md parameters for deformation twinning and martensitic transformation. The alloy indeed displays a yield strength of 760 MPa, about 200 MPa above that of a Ti 8.5Cr 1.5Sn (wt.%) single beta phase TRIP/TWIP alloy, combined with good ductility and work-hardening. In situ synchrotron X ray diffraction and post-mortem electron back-scattered analyses are performed to characterize the deformation mechanisms. They evidence that the TRIP and TWIP mechanisms are successfully obtained in the material, validating the design strategy. The interaction of the precipitates with the {332}<113> $\beta$ twins is analyzed, evidencing that the precipitates are sheared when hit by a twin, and therefore do not hinder the propagation of the twins. The detailed nature of the interaction is discussed, as well as the impact of the precipitates on the mechanical properties., Comment: 29 pages; 6 figures

Within 4 different crystal structures, 2280 ternary intermetallic configurations have been investigated via high-throughput density functional theory calculations in order to discover new semiconducting materials. The screening is restricted to intermetallics with the equimolar composition TMX, where T is a transition metal from the Ti, V, Cr columns, Sr, Ba, Y and La, M an element from the first line of transition metals and X a sp elements (Al, P, Si, Sn and Sb), i.e. to a list of 24 possible elements. Since the calculations are done combinatorically, every possible ternary composition is considered, even those not reported in the literature. All these TMX configurations are investigated in the 4 most reported structure-types: TiNiSi, MgAgAs, BeZrSi and ZrNiAl. With an excellent agreement between calculations and literature for the reported stable phases, we identify 472 possible stable compounds among which 21 are predicted as non-metallic. Among these 21 compositions, 4 could be considered as new semiconductors.