arXiv subjects

N. M. Chtchelkatchev

Publications and source records attributed to N. M. Chtchelkatchev.

At least 19 recordsLinked to original sources

arXiv2026

Pressure-dependent melting and crystallization of B2-NiAl from neural-network molecular dynamics

We investigate pressure-dependent melting of ordered B2-NiAl using neural-network molecular dynamics with a Deep Potential interatomic model. Melting temperatures are determined from two-phase solid-liquid coexistence simulations over a broad pressure range, yielding the melting curve $T_m(P)$. Relative to available experimental and previous molecular-dynamics results, the present calculations predict a stronger increase of the melting temperature with pressure at elevated compression. To assess the thermodynamic consistency of the calculated melting line, we evaluate the enthalpy and volume changes upon melting and compare the Clapeyron slope with the derivative of the fitted $T_m(P)$ curve. The two estimates are in good agreement over most of the investigated pressure range, supporting the internal consistency of the coexistence results. To probe the character of melting, we perform a layer-resolved composition analysis of the coexistence configurations and find that the coexisting liquid remains essentially equiatomic at all studied pressures, with deviations of the aluminum fraction from the stoichiometric value not exceeding $5\times10^{-3}$. This provides direct atomistic evidence that melting of B2-NiAl remains congruent within the present model. Together, these results establish a thermodynamically consistent pressure-dependent melting description of B2-NiAl and clarify the character of its melting under compression.

cond-mat.mtrl-sci↗

arXiv2026

The New High-entropy Compound RhMnFeCoGe4 with Cubic Non-centrosymmetric B20 Structure

A novel high-entropy compound, RhMnFeCoGe$_4$, with a cubic non-centrosymmetric B20 struc- ture, has been synthesized under conditions of high pressure and temperature. The electrical transport and magnetic properties of the obtained compound at both ambient and elevated pres- sures have been investigated. In addition, nuclear magnetic resonance (NMR) spectra were obtained at 4.2 K and ab initio calculations were performed. The new material exhibits ferromag- netic behavior with a critical temperature of $T_C$ = 146 K and a spontaneous moment of 2.5 $μ_B$ per formula unit. The magnetization data obtained at the critical region yielded the critical temperature and exponents, which were found to be $T_C$ = 146(1) K, $β$ = 0.337(1), $γ$ = 1.121(1), and $δ$ = 4.326(1). The magnetic moments of Mn and Co were determined from NMR spectra to be 2.2 $μ_B$ and 0.5 $μ_B$, respectively. Ab initio calculations yielded reasonable values for the lattice parameter and the magnetic moments of all constituents. The density of states and band structure are determined for both paramagnetic and ferromagnetic states. Lattice compression results in the increase in the $T_C$.

cond-mat.mtrl-sci↗

arXiv2026

Adaptive fine-tuning of foundation models for crystal structure prediction: Discovery of high-pressure phases in the CaFeNi system

The prediction of crystal structures is a key challenge in chemistry and materials science, but evolutionary crystal structure prediction (CSP) remains computationally expensive because it relies on repeated \textit{ab initio} relaxations and energy ranking. Machine learning interatomic potentials (MLIPs) can accelerate CSP, yet their use is limited by the need for large training sets and by the difficulty of choosing which candidate structures should be labeled by density functional theory (DFT). Here we introduce a self-consistent, foundation-model-assisted CSP workflow that combines evolutionary search with adaptive data selection and fine-tuning. Starting from a pretrained MLIP, the algorithm rapidly explores configuration space while iteratively selecting compact, representative, and physically relevant subsets of structures for DFT labeling, thereby reducing redundant calculations and improving a system-specific potential. We apply the method to the chemically complex Ca--Fe--Ni ternary system. The workflow reproduces the known low-pressure convex hull and enables efficient high-pressure exploration. It predicts a previously unreported compound, Ca$_6$FeNi, which becomes thermodynamically stable above 100~GPa. These results show that foundation-model-based, data-efficient CSP can greatly reduce computational cost while preserving accuracy and enabling the discovery of new materials in complex multicomponent systems.

cond-mat.mtrl-sci↗

arXiv2026

Machine learning potential as a guide for eutectic in ultra-refractory multicomponent ceramics

The experimental determination of eutectic points is a long-established and widely used technique, but it is generally only practical for systems with relatively low melting points. Many modern, promising materials, however, are ultra-refractory, with melting points exceeding 3000 K. For these systems, conventional melting experiments become prohibitively expensive and technically challenging. Advanced AI modeling can serve as a powerful precursor to guide successful experimentation in such cases. This work proposes a novel criterion for determining the eutectic point concentration in ultra-refractory alloys. The approach is verified using the Ti-B-C system - the most thoroughly studied three-component refractory system to date. The core of the algorithm is a machine-learning interatomic potential, based on a neural network, which achieves accuracy comparable to ab initio methods. Crucially, the algorithm operates effectively in the liquid phase, eliminating the need for information about the solid alloy's crystalline structure to estimate eutectic points.

cond-mat.dis-nn↗

arXiv2025

Accuracy and Limitations of Machine-Learned Interatomic Potentials for Magnetic Systems: A Case Study on Fe-Cr-C

Machine-learned interatomic potentials (MLIPs) have become the gold standard for atomistic simulations, yet their extension to magnetic materials remains challenging because spin fluctuations must be captured either explicitly or implicitly. We address this problem for the technologically vital Fe-Cr-C system by constructing two deep machine learning potentials in DeePMD realization: one trained on non-magnetic DFT data (DP-NM) and one on spin-polarised DFT data (DP-M). Extensive validation against experiments reveals a striking dichotomy. The dynamic, collective properties, viscosity and melting temperatures are reproduced accurately by DP-NM but are incorrectly estimated by DP-M. Static, local properties, density, and lattice parameters are captured excellently by DP-M, especially in Fe-rich alloys, whereas DP-NM fails. This behaviour is explained by general properties of paramagnetic state: at high temperature, local magnetic moments self-average in space and time, so their explicit treatment is unnecessary for transport properties but essential for equilibrium volumes. Exploiting this insight, we show that a transfer-learning protocol, pre-training on non-magnetic DFT and fine-tuning on a small set of spin-polarised data, reduces the computational cost to develop magnetic MLIPs by more than an order of magnitude. Developing general-purpose potentials that capture static and dynamic behaviors throughout the whole composition space requires proper accounting for temperature-induced spin fluctuations in DFT calculations and correctly incorporating spin degrees of freedom into classical force fields.

cond-mat.mtrl-sci↗

arXiv2024

Kob-Andersen model crystal structure: genetic algorithms vs spontaneous crystallization

For the first time, the crystal structure of the Kob-Andersen mixture has been probed by genetic algorithms calculations. The stable structures of the system with different molar fractions of the components have been identified and their stability at finite temperature has been verified. A possibility to obtain these structures by spontaneous crystallization of a liquid has been checked.

cond-mat.soft↗

arXiv2023

Local structure, thermodynamics, and melting curve of boron phosphide at high pressures by deep learning-driven ab initio simulations

Boron phosphide (BP) is a (super)hard semiconductor constituted of light elements, which is promising for high demand applications at extreme conditions. The behavior of BP at high temperatures and pressures is of special interest but is also poorly understood because both experimental and conventional ab initio methods are restricted to studying refractory covalent materials. The use of machine learning interatomic potentials is a revolutionary trend that gives a unique opportunity for high-temperature study of materials with ab initio accuracy. We develop a deep machine learning potential (DP) for accurate atomistic simulations of solid and liquid phases of BP as well as their transformations near the melting line. Our DP provides quantitative agreement with experimental and ab initio molecular dynamics data for structural and dynamic properties. DP-based simulations reveal that at ambient pressure tetrahedrally bonded cubic BP crystal melts into an open structure consisting of two interpenetrating sub-networks of boron and phosphorous with different structures. Structure transformations of BP melts under compressing are reflected by the evolution of low-pressure tetrahedral coordination to high-pressure octahedral coordination. The main contributions to structural changes at low pressures are made by the evolution of medium-range order in B-subnetwork and at high pressures by the change of short-range order in P-sub-network. Such transformations exhibit an anomalous behavior of structural characteristics in the range of 12--15 GPa. Analysis of the results obtained raise open issues in developing machine learning potentials for covalent materials and stimulate further experimental and theoretical studies of melting behavior in BP.

cond-mat.mtrl-sci↗

arXiv2022

Solidification of the Glass-Forming Al86Ni2Co6Gd6 Melt under High Pressure

High pressures allow the synthesis of new metastable compounds that remain intact for a sufficiently long time at normal conditions. Until now, it has not been fully understood how pressure, glass-forming ability and solidification of liquids are interconnected. We have investigated the structure of the glass-forming eutectic alloy Al86Ni2Co6Gd6 obtained by rapid cooling from the melt having a temperature of 1800 K under a pressure of 10 GPa. X-ray diffraction analysis and electron microscopy show that the samples are homogeneous and dense. The structure is finely dispersed. New stable crystalline phases with cubic (cP4/2) and tetragonal (tI26/1) structures are formed in the alloy. The studies have shown that the average microhardness of the samples obtained at 10 GPa is almost 2 times higher than that of the original sample at atmospheric pressure and is about 1700 MPa. To understand the results, we used ab initio molecular dynamics and studied how the melt changes with pressure. It is shown that at a temperature of 1800 K, high pressure increases the concentration of icosahedral clusters in the melt so that at 10 GPa atoms inside the icosahedra form a percolation cluster, while at atmospheric pressure they do not. Thus, the glass-forming ability of a melt increases at high pressure strongly influencing solidification processes.

cond-mat.mtrl-sci↗

arXiv2022

Structure and solidification of the (Fe0.75B0.15Si0.1)100-xTax (x=0-2) melts: experiment and machine learning

Fe-B-Si system is a matrix for synthesis of new functional materials with exceptional magnetic and mechanical properties. Progress in this area is associated with the search for optimal doping conditions. This theoretical and experimental study is aimed to address the influence of Ta alloying on the structure of undercooled (Fe0.75B0.15Si0.1)100-xTax (x=0-2) melts, their undercoolability and the processes of structure formation during solidification. Small concentration of Ta complicates standard ab initio and machine learning investigations. We developed a technique for fast and stable training of machine learning interatomic potential (MLIP) in this case and uncovered the structure of the undercooled melts. Molecular dynamic simulations with MLIP showed that at Ta concentration of 1 at.% there is a sharp change in the chemical short-range ordering in the melt associated with a change in the interaction of Ta atoms. This effect leads to a restructuring of the cluster formation in the system. At the same time, our experimental investigation shows that melts with a Ta content of 1 at.% have the greatest tendency to undercoolability. Alloying with Ta promotes the formation of primary crystals of Fe2B, and at a concentration of more than 1.5 at.% Ta, also of FeTaB. Herewith, near 1 at.% Ta, the crystallization of the melt proceeds nontrivially: with the formation of two intermediate metastable phases Fe3B and Fe2Ta Laves phase. Also, the highest tendency to amorphization under conditions of quick quenching is exhibited by a melt with a Ta concentration of 1 at.%. The results not only provide understanding of optimal alloying of Fe-B-Si materials but also promote a machine learning method for numerical design of metallic alloys with a small dopant concentration.

cond-mat.mtrl-sci↗

arXiv2022

Time-dependent exchange creates the time-frustrated state of matter

Magnetic systems governed by exchange interactions between magnetic moments harbor frustration that leads to ground state degeneracy and results in the new topological state often referred to as a frustrated state of matter (FSM). The frustration in the commonly discussed magnetic systems has a spatial origin. Here we demonstrate that an array of nanomagnets coupled by the real retarded exchange interactions develops a new state of matter, time frustrated matter (TFM). In a spin system with the time-dependent retarded exchange interaction, a single spin-flip influences other spins not instantly but after some delay. This implies that the sign of the exchange interaction changes, leading to either ferro- or antiferromagnetic interaction, depends on time. As a result, the system's temporal evolution is essentially non-Markovian. The emerging competition between different magnetic orders leads to a new kind of time-core frustration. To establish this paradigmatic shift, we focus on the exemplary system, a granular multiferroic, where the exchange transferring medium has a pronounced frequency dispersion and hence develops the TFM.

cond-mat.str-el↗

arXiv2022

The resistance of quantum entanglement to temperature in the Kugel-Khomskii model

The Kugel--Khomskii model with entangled spin and orbital degrees of freedom is a good testing ground for many important features in quantum information processing, such as robust gaps in the entanglement spectra. Here, we demonstrate that the entanglement can be also robust under effect of temperature within a wide range of parameters. It is shown, in particular, that the temperature dependence of entanglement often exhibits a nonmonotonic behavior. Namely, there turn out to be ranges of the model parameters, where entanglement is absent at zero temperature, but then, with an increase in temperature, it appears, passes through a maximum, and again vanishes.

quant-ph↗

arXiv2022

Split of the magnetic and crystallographic states in Fe$_{1-x}$Rh$_{x}$Ge

We report on a comprehensive experimental and theoretical study of Fe$_{1-x}$Rh$_{x}$Ge compounds, within the entire concentration range $x \in \left[0.0 - 1.0\right]$, using X-Ray diffraction, small-angle neutron scattering, magnetometry and theoretical calculations. While FeGe and RhGe are single phase helimagnet and unconventional superconductor, respectively, an internal splitting of the crystallographic and magnetic states is found for intermediate compositions $x \in \left[0.2 - 0.9\right]$. A theoretical analysis of the stability of the two detected phases, together with the experimental data, indicate that this splitting preserves a common space group and occurs within single crystallites. Despite their apparent similarity, these two phases however display different magnetic structures, with distinct ferro- and helimagnetic character.

cond-mat.mtrl-sci↗

arXiv2022

Ground-state structure, orbital ordering and metal-insulator transition in double-perovskite PrBaMn2O6

In recent years, A-site ordered half-doped double-perovskite manganites $\rm RBaMn_2O_6$ (R=rare earth) have attracted much attention due to their remarkable physical properties and a prospect of application as magnetoresistance, multiferroic, and oxygen storage materials. The nature of the ground state in ${\rm RBaMn_2O_6}$ as well as sequence of phase transitions taking place at cooling are not yet well understood due to complexity in both experimental and theoretical studies. Here we address the origin of the ground-state structure in PrBaMn$_2$O$_6$ as well as its electronic and magnetic properties. Utilizing GGA+U approach and specially designed strategy to perform structural optimization, we show that the system has two competing AFM-A and AFM-CE magnetic structures with very close energies. The AFM-A structure is a metal, while AFM-CE is an insulator and the transition to the insulating state is accompanied by the charge Mn$^{3+}$/Mn$^{4+}$, and orbital $3x^2-r^2$/$3y^2-r^2$ orderings. This orbital ordering results in strong cooperative Jahn-Teller (JT) distortions, which lower the crystal symmetry. Our findings give a key to understanding contradictions in available experimental data on ${\rm PrBaMn_2O_6}$ and opens up the prospects to theoretical refinements of ground-state structures in other ${\rm RBaMn_2O_6}$ compounds.

cond-mat.str-el↗

arXiv2021

Deep machine learning potentials for multicomponent metallic melts: development, predictability and compositional transferability

The use of machine learning interatomic potentials (MLIPs) in simulations of materials is a state-of-the-art approach, which allows achieving nearly \textit{ab initio} accuracy with orders of magnitude less computational cost. Multicomponent disordered systems have a highly complicated potential energy surface due to both topological and compositional disorder. That arises issues in MLIPs developing, such as optimal design strategy of potentials and their predictability and transferability. Here we address MLIPs for multicomponent metallic melts taking the ternary Al-Cu-Ni ones as a convenient example. We use many-body deep machine learning potentials as implemented in the DeePMD-kit to build MLIP that allows describing both atomic structure and dynamics of the system in the whole composition range. Doing that we consider different sets of neural networks hyperparameters and learning schemes to create an optimal MLIP, which allows archiving good accuracy in comparison with both \textit{ab initio} and experimental data. We find that developed MLIP demonstrates good compositional transferability, which extends far beyond compositional fluctuations in the training configurations. The results obtained open up prospects for simulating structural and dynamical properties of multicomponent metallic alloys with MLIPs.

cond-mat.mtrl-sci↗

arXiv2021

Investigation of the degree of local structural similarity between the parent-liquid and children-crystal states for a model soft matter system

We investigate the degree of local structural similarity between the parent-liquid and children-crystal states for a model soft-matter system of particles interacting through the harmonic-repulsive pair potential. At different pressures, this simple system crystallizes into several significantly different crystal structures. Therefore, the model is well suited for addressing the question under consideration. In our studies, we carefully analyze the developments of the pair and triple correlation functions for the parent-liquid as the pressure increases. In particular, these considerations allow us to address the similarities in the orientational orderings of the corresponding liquid and solid phases. It is demonstrated that the similarities in the orientational ordering between the two states extend beyond the first and second neighbors. Currently, it is widely accepted that orientational ordering is important for understanding the behaviors of liquids, supercooled liquids, and the development of detailed theories of the crystalization process. Our results suggest that, up to a certain degree, it might be possible to predict the structures of the children-solids from studies of the parent-liquids. Our results raise anew a general question of how much insight into the properties of the liquid-state can be gained from drawing a parallel with the solid-state.

cond-mat.soft↗

arXiv2020

Quantum entanglement, local indicators and effect of external fields in the Kugel-Khomskii model

Using the exact diagonalization technique, we determine the energy spectrum and wave functions for finite chains described by the two-spin (Kugel--Khomskii) model with different types of intersubsystem exchange terms. The found solutions provide a possibility to address the problem of quantum entanglement inherent to this class of models. We put the main emphasis on the calculations of the concurrence treated as an adequate numerical measure of the entanglement. We also analyze the behavior of two-site correlation functions considered as a local indicator of entanglement. We construct the phase diagrams of the models involving the regions of nonzero entanglement. The pronounced effect of external fields, conjugated to both spin variables on the regions with entanglement, could both enhance and weaken the entanglement depending on the parameters of the models.

cond-mat.str-el↗

arXiv2020

The uncertainty of glass transition temperature in molecular dynamics simulations and numerical algorithm for its unique determination

When the cooling rate $v$ is smaller than a certain material-dependent threshold, the glass transition temperature $T_g$ becomes to a certain degree the "material parameter" being nearly independent on the cooling rate. The common method to determine $T_g$ is to extrapolate viscosity $ν$ of the liquid state at temperatures not far above the freezing conditions to lower temperatures where liquid freezes and viscosity is hardly measurable. It is generally accepted that the glass transition occurs when viscosity drops by $13\leq n\leq17$ orders of magnitude. The accuracy of $T_g$ depends on the extrapolation quality. We propose here an algorithm for a unique determining of $T_g$. The idea is to unambiguously extrapolate $ν(T)$ to low temperatures without relying upon a specific model. It can be done using the numerical analytical continuation of $ν(T)$-function from above $T_g$ where it is measurable, to $T\gtrsim T_g$. For numerical analytical continuation, we use the Pade approximant method.

physics.chem-ph↗

arXiv2020

Freezing of two-length-scale systems: complexity, universality and prediction

Two-length-scale pair potentials arise ubiquitously in condensed matter theory as effective interparticle interactions in molecular, metallic and soft matter systems. The existence of two different bond lengths generated by the shape of potential causes complex behavior in even one-component systems: polymorphism in solid and liquid states, water-like anomalies, the formation of quasicrystals and high stability against crystallization. Here we address general properties of freezing in one-component two-length-scale systems and argue that the formation of solid phases during cooling a liquid is essentially determined by the radial distribution function (RDF) of the liquid. We show that different two-length-scale systems having similar RDF freeze into the same solid phases. In most cases, the similarity between RDFs can be expressed by the proximity of two dimensionless effective parameters: the ratio between effective bond lengths, $λ$, and the fraction of short-bonded particles $ϕ$. We validate this idea by studying the formation of different solid phases in different two-length-scale systems. The method proposed allows predicting effectively the formation of solid phases in both numerical simulations and self-assembling experiments in soft matter systems with tunable interactions.

physics.chem-ph↗