First-principles Molecular Dynamics Calculations Research Articles

Thermal performance of MgCl2–NaCl–KCl eutectic salt for the next generation concentrated solar power and correlation between structure and thermophysical properties: Insights from atomic and electronic levels

MgCl2–NaCl–KCl (MgNaK) eutectic salt has emerged as a highly promising candidate for concentrated solar power (CSP), with extensive research conducted in recent years. However, the thermal property data for the MgNaK eutectic salt are still very limited. Here, we present a novel and accurate many-body potential trained by machine learning (ML) for MgNaK eutectic salt (with mole fractions of 45.4 %, 33 %, and 21.6 % respectively), developed using energies and forces extracted from first-principles molecular dynamics calculations (FPMD). The performance of the potential is well tested. We conduct a detailed investigation on the temperature-dependent structural and corresponding thermal properties, analyzing them from both the atomic and electronic perspectives. The introduction of Na+ or K+ ions disrupts the net structure formed by corner-joint and edge-joint MgClx units, thereby affecting the transport properties. The calculated density (ρ) and constant pressure specific heat capacity (CP) exhibit agreement with experimental data within a margin of 2 %. We observe and discuss the typical λ of negative linear temperature correlation, which is similar to other molten alkali chloride salts. Finally, based on the simulated and experimental values, we provide reliable recommendations for the λ and viscosities (η) of MgNaK eutectic salt across its entire operating temperature range.

Degassing of CO2 triggers large-scale loss of helium from magma oceans

The moon-forming impact caused widespread melting and vaporization of the proto-Earth, leading to the formation of a protolunar disk1,2. Volatile elements trapped in the inner parts of the disk condensed together with heavier elements into a global magma ocean3,4. By subsequent cooling and degassing, volatiles contributed to the formation of the Earth’s secondary atmosphere5. The thermodynamic conditions at the magma ocean-atmosphere interface profoundly impacted the degree of degassing that occurred. Today, our understanding of these conditions remains unclear. Here, we simulate the degassing of carbon and helium from a magma ocean with a bulk silicate Earth composition6 at presumed typical conditions for the early Earth5,7. We employ first-principles molecular dynamics calculations. We find that volatile loss is determined by pressure, temperature, and melt composition. Carbon and helium devolatilize more easily when both elements are present in the melt, leading to pronounced loss from the magma ocean. Our findings suggest that the early Earth atmosphere was carbon-rich, with a high fraction of helium and other noble gases, and thicker and hotter than previously thought. We evaluate the implications for Earth’s composition and thermal evolution, as well as other planets that may undergo similar processes with different chemistries.

Structure and dynamical properties during solidification of liquid aluminum induced by cooling and compression

Abstract The structural transformation from a liquid to a crystalline solid is an important subject in condensed matter physics and materials science. In the present study, firstprinciples molecular dynamics calculations were performed to investigate the structure and properties of aluminum during the solidification which was induced by cooling and compression. During both the cooling and compression processes, it was found that the icosahedral short-range order was initially enhanced and then began to decay, the face-centered cubic short-range order eventually became dominant before it transformed to a crystalline solid.

Quantitative descriptor of lattice anharmonicity in crystal

<sec>Anharmonic effect is often one of the physical root causes of some special material properties, such as soft mode phase transition, negative thermal expansion, multiferroicity, and ultra-low thermal conductivity. However, the existing methods of quantifying the anharmonicity of material do not give a clear and accurate anharmonicity descriptor. The calculation of the anharmonic effect requires extremely time-consuming molecular dynamics simulation, the calculation process is complex and costly. Therefore, a quantitative descriptor is urgently needed, which can be used to implement quick calculation so as to understand, evaluate, design, and screen functional materials with strong anharmonicity.</sec><sec>In this paper, we propose a method to quantify the anharmonicity of materials by only phonon spectrum and static self-consistent calculation through calculating and analyzing the material composed of germanium and its surrounding elements. In this method, the lattice anharmonicity is decomposed into the anharmonic contribution of independent phonon vibration modes, and the quantitative anharmonicity descriptor <inline-formula><tex-math id="M3">\begin{document}$ {\sigma }_{\boldsymbol{q},j}^{A} $\end{document}</tex-math><alternatives><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M3.jpg"/><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M3.png"/></alternatives></inline-formula> of phonons is proposed. Combining it with the Bose-Einstein distribution, the quantitative descriptor <inline-formula><tex-math id="M4">\begin{document}$ {A}_{{\mathrm{p}}{\mathrm{h}}}\left(T\right) $\end{document}</tex-math><alternatives><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M4.jpg"/><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M4.png"/></alternatives></inline-formula> of temperature-dependent material anharmonicity is proposed. We calculate the bulk moduli and lattice thermal conductivities at 300 K of nine widely representative materials. There is a clear linear trend between them and our proposed quantitative descriptor <inline-formula><tex-math id="M5">\begin{document}$ {A}_{{\mathrm{p}}{\mathrm{h}}}\left(T\right) $\end{document}</tex-math><alternatives><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M5.jpg"/><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="5-20231428_M5.png"/></alternatives></inline-formula>, which verifies the accuracy of our proposed descriptor. The results show that the descriptor has the following functions. i) It can systematically and quantitatively classify materials as the strength of anharmonicity; ii) it intuitively shows the distribution of the anharmonic effect of the material on the phonon spectrum, and realizes the separate analysis of the phonon anharmonicity that affects the specific properties of the material; iii) it is cost-effective in first-principles molecular dynamics calculations and lays a foundation for screening and designing materials based on anharmonicity.</sec><sec>This study provides an example for the high-throughput study of functional materials driven by anharmonic effect in the future, and opens up new possibilities for material design and application. In addition, for strongly anharmonic materials such as CsPbI<sub>3</sub>, the equilibrium position of the atoms is not fixed at high temperatures, resulting in a decrease in the accuracy of quantifying anharmonicity using our proposed descriptor. In order to get rid of this limitation, our future research will focus on the distribution of atomic equilibrium positions in strongly anharmonic materials at high temperatures, so as to propose a more accurate theoretical method to quantify the anharmonicity in strongly anharmonic materials.</sec>

Phase transition process and mechanism of LaOF in air: From experiment to theory

Lanthanum oxyfluoride (LaOF) is widely used in multiple fields, such as optoelectronics, photocatalysis, all solid-state fluoride ion batteries, etc., and the requirements for its phase structure vary in different fields. However, there is currently no clear understanding of the phase transition process of LaOF in an air atmosphere. Therefore, in this article, LaOF with different phase structures was obtained by calcining LaCO3F in air. TG-DSC analysis, XRD, Raman, SEM, TEM, and first-principles calculations based on DFT were used to illustrate the phase transition mechanism and phase stability of LaOF in air. The experimental results indicate that when calcined in air, a rhombohedral phase (γ-LaOF) is obtained at 600 °C, a tetragonal phase (β-LaOF) is obtained at 750 °C, but at 925 °C, β-LaOF undergoes a phase transition again, transforming into γ-LaOF. In addition, the products begin to decompose into La2O3 when the calcination temperature reaches 1300 °C. Raman-assisted XRD data demonstrated the phase transition law of LaOF, namely γ-LaOF → β-LaOF → γ-LaOF → γ-LaOF + β-LaOF → γ-LaOF + La2O3, and γ-LaOF is more stable. The HSE functional calculation, based on first principles, was used to analyze the phase structures of three-dimensional LaOF. The results confirmed that the stability of γ-LaOF is higher than that of β-LaOF. Furthermore, the phonon spectrum calculation results indicate that the stability order is γ-LaOF > β-LaOF > α-LaOF. Additionally, first-principles molecular dynamics calculations were employed to explain the phase transition law of LaOF at different temperatures by examining the excess energy of γ-LaOF and β-LaOF.

Are High-Temperature Molten Salts Reactive with Excess Electrons? Case of ZnCl2.

New and exciting frontiers for the generation of safe and renewable energy have brought attention to molten inorganic salts of fluorides and chlorides. This is because high-temperature molten salts can act both as coolants and liquid fuel in next-generation nuclear reactors. Whereas research from a few decades ago suggests that salts are mostly unreactive to radiation, recent experiments hint at the fact that electrons generated in such extreme environments can react with the melt and form new species including nanoparticles. Our study probes the fate of an excess electron in molten ZnCl2 using first-principles molecular dynamics calculations. We find that on the time scale accessible to our study, an excess electron can be found in one of three states; the lowest-energy state can be characterized as a covalent Zn2Cl5•2- radical ion, the other two states are a solvated Zn•+ species (ZnCl3•2-) and a more delocalized species that still has some ZnCl3•2- character. Since for each of these, the singly occupied molecular orbital (SOMO) where the excess charge resides has a distinct and well-separated energy, the different species can in principle be characterized by their own electronic spectra. The study also sheds light onto what is commonly understood as the spectrum of a transient radical species which can be from the SOMO onto higher energy states or from the melt to pair with the excess electron leaving a hole in the liquid.

First-principles calculations and device characterizations of formamidinium-cesium lead triiodide perovskite crystals stabilized by germanium or copper

To avoid formation of the photo-inactive δ-phase of formamidinium-cesium lead triiodide, copper or germanium was added to the perovskite compounds to stabilize the photoactive α-phase. It was found that the substitution of lead by germanium (Ge) or copper (Cu) provided the stabilization of the α-phase in the present work. The first-principles molecular dynamics calculations indicated that displacements of formamidinium molecules were suppressed by the Ge doping. X-ray diffraction results indicated that the Ge or Cu doping of the perovskite compounds could be effective for suppression the phase transition from α- to δ-phase.

Atomistic simulations of thermal conductivity in novel GeC channel materials from first-principles molecular dynamics calculations

Silicon carbide (SiC) has emerged as a candidate material for next-generation power devices to replace traditional silicon power devices. They feature smaller size, faster switching speed, simpler cooling, and greater reliability than Si-MOSFETs. To date, however, the thermal conductance of GeC-based power MOSFETs is unclear. This work explains the heat transfer of GeC by simulating the thermal conductivity through molecular dynamics (MD) and proposes a potential 4H-GeC power MOSFET with wide bandgap and high thermal conductivity to replace Si-MOSFETs.

Atomic and electronic structures of interfaces between amorphous (Al2O3)1−x(SiO2)x and GaN polar surfaces revealed by first-principles simulated annealing technique

We report first-principles molecular-dynamics calculations with the simulated annealing technique that clarify the atomic and electronic structures of the semiconductor–insulator interfaces consisting of GaN (0001) and (0001¯) faces and the amorphous (Al2O3)1−x(SiO2)x. We confirm that the obtained interfaces are free from dangling bonds, as predicted by our previous calculations, irrespective of the thickness of the amorphous (Al2O3)1−x(SiO2)x layer. This is due to the high atomic density and large mean coordination number near the interfaces caused by atomic diffusion from inside of the insulator to the interfaces. The calculated local density of states of the (Al2O3)1−x(SiO2)x/GaN system quantitatively shows clear band offsets and, more importantly, the absence of deep states in the GaN energy gap. Interestingly, we find that the band alignment causing the offset is not abrupt at the interface but varies gradually near the interface, predicting the existence of transition layers. We determine the thicknesses of the transition layers in the (Al2O3)1−x(SiO2)x/GaN system to be about 10 Å. We argue that those structural characteristics prevent the formation of the dangling-bond origin carrier traps at the interface, and this is a superior feature of the (Al2O3)1−x(SiO2)x as a gate oxide for the GaN-based metal–oxide–semiconductor devices.

Impact of Dispersion Force Schemes on Liquid Systems: Comparing Efficiency and Drawbacks for Well-Targeted Test Cases.

First-principles molecular dynamics (FPMD) calculations were performed on liquid GeSe4 with the aim of inferring the impact of dispersion (van der Waals, vdW) forces on the structural properties. Different expressions for the dispersion forces were employed, allowing us to draw conclusions on their performances in a comparative fashion. These results supersede previous FPMD calculations obtained in smaller systems and shorter time trajectories by providing data of unprecedented accuracy. We obtained a substantial agreement with experiments for the structure factor regardless of the vdW scheme employed. This objective was achieved by using (in addition to FPMD with no dispersion forces) a selection of vdW schemes available within density functional theory. The first two are due to Grimme, D2 and D3, and the third one is devised within the so-called maximally localized Wannier functions approach (MLWF). D3 results feature a sizeable disagreement in real space with D2 and MLWF in terms of the partial and total pair correlation functions as well as the coordination numbers. More strikingly, total and partial structure factors calculated with D3 exhibit an unexpected sharp increase at low k. This peculiarity goes along with large void regions within the network, standing for a phase separation of indecipherable physical meaning. In view of these findings, further evidence of unconventional structural properties found by employing D3 is presented by relying on results obtained for a complex ionic liquid supported on a solid surface. The novelty of our study is multifold: new, reliable FPMD data for a prototypical disordered network system, convincing agreement with experimental data and assessment of the impact of dispersion forces, with emphasis on the intriguing behavior of one specific recipe and the discovery of common structural features shared by drastically dissimilar physical systems when the D3 vdW scheme is employed.

Stability of and conduction in single-walled Si2BN nanotubes

We explore the possibility and potential benefit of rolling a ${\mathrm{Si}}_{2}\mathrm{BN}$ sheet into single-walled nanotubes (NTs). Using density functional theory (DFT), we consider both structural stability and the impact on the nature of chemical bonding and conduction. The structure is similar to carbon NTs and hexagonal boron-nitride (hBN) NTs and we consider both armchair and zigzag ${\mathrm{Si}}_{2}\mathrm{BN}$ configurations with varying diameters. The stability of these ${\mathrm{Si}}_{2}\mathrm{BN}$ NTs is confirmed by first-principles molecular dynamics calculations, by exothermal formation, an absence of imaginary modes in the phonon spectra. Also, we find the nature of conduction varies from semiconducting over semimetallic to metallic, reflecting differences in armchair/zigzag-type structures, curvature effects, and the effect of quantum confinement. We present a detailed characterization of how these properties lead to differences in both the bonding nature and electronic structures.

Degenerate Lattice-Instability-Driven Amorphization under Compression in Metal Halide Perovskite CsPbI3.

Halide perovskites have been intensively investigated for photovoltaic applications because of their good optoelectronic properties and low cost. Various high-pressure experiments have shown that these materials generally undergo reversible phase transitions between different crystalline phases as well as between crystalline and amorphous phases under external pressure. Herein, using first-principles density functional theory (DFT) and ab initio molecular dynamics (AIMD) calculations, we investigate the origin of the pressure-induced amorphization in CsPbI3. We find that the amorphous-like structures obtained from AIMD calculations become more stable than the orthorhombic Pbnm phase above 6.66 GPa, in good agreement with the experimental value (4.44 GPa). We further find that an imaginary flat band appears in the phonon dispersion of the orthorhombic CsPbI3 phase across the Brillouin zone at 10 GPa, leading to degenerate lattice instabilities. These energetically degenerate phonon modes are related to PbI6 octahedral tilting modes and provide random local distortions, leading to amorphization.

Rotation of complex ions with ninefold hydrogen coordination studied by quasielastic neutron scattering and first-principles molecular dynamics calculations

Quasielastic neutron scattering (QENS) and neutron powder diffraction of the complex transition metal hydrides ${\mathrm{Li}}_{5}{\mathrm{MoH}}_{11}$ and ${\mathrm{Li}}_{6}{\mathrm{NbH}}_{11}$ were measured in a temperature range of 10--300 K to study their structures and dynamics, especially the dynamics of the hydrogen atoms. These hydrides contain unusual ninefold H-coordinated complex ions (${\mathrm{MoH}}_{9}^{3\ensuremath{-}}$ or ${\mathrm{NbH}}_{9}^{4\ensuremath{-}}$) and hydride ions $({\mathrm{H}}^{\ensuremath{-}})$. A QENS signal appeared >150 K due to the relaxation of H atoms. The intermediate scattering functions derived from the QENS spectra are well fitted by a stretched exponential function called the Kohlrausch-Williams-Watts functions with a small stretching exponent \ensuremath{\beta} \ensuremath{\approx} 0.3--0.4, suggesting a wide relaxation time distribution. The $Q$ dependence of the elastic incoherent structure factor is reproduced by the rotational diffusion of $M{\mathrm{H}}_{9}$ ($M=\mathrm{Mo}$ or Nb) anions. The results are well supported by a van Hove analysis for the motion of H atoms obtained using first-principles molecular dynamics calculations. We conclude that the wide relaxation time distribution of the $M{\mathrm{H}}_{9}$ rotation is due to the positional disorder of the surrounding Li ions and a unique rotation with $M{\mathrm{H}}_{9}$ anion deformation (pseudorotation).

Effect of carbon atoms on the reliability of potassium-ion electrets used in vibration-powered generators

Potassium-ion electrets, which are key elements in vibration-powered microelectromechanical generators, can store negative charge almost permanently. However, our experiments show that the charge storing capability of potassium ion electrets is severely degraded when a SiC plate instead of the usual SiO2 plate is used in the fabrication process, indicating the negative influence of carbon incorporation. We theoretically studied the atomic and electronic structures of amorphous silica (a-SiO2) with and without potassium and carbon atoms using first-principles molecular-dynamics calculations. Our calculations show that negative charge accumulates in fivefold-coordinated Si atoms without carbon contamination. However, carbon contamination of the potassium-ion electret causes oxygen deficiencies in SiO2, hindering the formation of fivefold-coordinated Si atoms. As a result, carbon contamination causes degradation of the negative charge storing capability of potassium-ion electrets.

Experimental and theoretical studies on atomic structures of the interface states at SiO2/4H-SiC(0001) interface

We investigated the atomic structures of the interface states (gap states) at the SiO2/4H-SiC(0001) interface using hard x-ray photoelectron spectroscopy (HAXPES), operando hard x-ray photoelectron spectroscopy, extended x-ray absorption fine structure, and first principles molecular dynamics (FPMD) calculations. For the interface states, two types were observed: continuous interface states in the whole gap and interface states with discrete energy levels near the conduction band minimum (CBM). From HAXPES, we found that carbon clusters and carbon vacancies were formed at the SiO2/4H-SiC(0001) interface. FPMD calculations on the SiO2/4H-SiC(0001) interface showed that the interface states in the whole gap were attributed to the various atomic geometries of the CßSi3 species and the carbon clusters with various sizes and surrounding atoms. For the interface states with a discrete energy level near the CBM, we could not find their atomic structure using our current calculations. We calculated the carbon vacancies prepared on the side of an SiC substrate at the SiO2/4H-SiC(0001) interface, indicating the formation of a discrete energy level in the midgap. It is likely that carbon vacancies formed at the step of the interface may modulate the midgap energy level to energy below the CBM. Therefore, we propose that the interface states with discrete energy levels near the CBM could be attributed to the carbon vacancies formed on the steps at the SiO2/4H-SiC(0001) interface.

First-Principles Dynamics of Fluorine Adsorption on Clean and Monohydrogenated Si{001}.

The interaction of highly reactive species with solid surfaces can result in modes of adsorption quite distinct from the classic molecular and dissociative events that are usually thought to dominate. For instance, compelling experimental evidence suggests that adsorption of F2 at the Si{001} surface is often initiated by abstraction (and binding at the surface) of just one fluorine atom from the molecule; the second fluorine atom subsequently experiences either a separate atomic adsorption event or ejection from the surface altogether. Molecular dynamics simulations using empirical potentials support this concept but massively overestimate the prevalence of atomic ejection. In this work, we report first-principles molecular dynamics calculations that correctly show atomic ejection to be rare while providing insight into the details of abstractive adsorption. In addition, we also examine the case of F2 adsorption onto a monohydrogenated Si{001} surface, finding evidence for a different type of abstractive adsorption, in which a hydrogen atom may be removed from the surface to form a short-lived HFF intermediate. The latter rapidly decomposes to produce either HF or (via reaction with another surface hydrogen atom) H2.

Probing the Disorder Inside the Cubic Unit Cell of Halide Perovskites from First-Principles.

Strong deviations in the finite temperature atomic structure of halide perovskites from their average geometry can have profound impacts on optoelectronic and other device-relevant properties. Detailed mechanistic understandings of these structural fluctuations and their consequences remain, however, limited by the experimental and theoretical challenges involved in characterizing strongly anharmonic vibrational characteristics and their impact on other properties. We overcome some of these challenges by a theoretical characterization of the vibrational interactions that occur among the atoms in the prototypical cubic CsPbBr3. Our investigation based on first-principles molecular dynamics calculations finds that the motions of neighboring Cs-Br atoms interlock, which appears as the most likely Cs-Br distance being significantly shorter than what is inferred from an ideal cubic structure. This form of dynamic Cs-Br coupling coincides with very shallow dynamic potential wells for Br motions that occur across a locally and dynamically disordered energy landscape. We reveal an interesting dynamic coupling mechanism among the atoms within the nominal unit cell of cubic CsPbBr3 and quantify the important local structural fluctuations on an atomic scale.

Equation of State Determination for Rhenium Using First-Principles Molecular Dynamics Calculations and High-Pressure Experiments

The room-temperature bulk modulus of rhenium (Re) was measured in the pressure range 0 to 115 GPa using a laser-annealing diamond anvil cell and the synchrotron X-ray diffraction method. Thermal properties of Re were investigated up to 4000 K based on first-principles molecular dynamics calculations, and the equation of state for Re was determined using experimental and calculated data. A Vinet equation of state fitted to the 300 K data yielded a bulk modulus of KT0 = 384 GPa and a pressure derivative of K T 0 ′ = 3.26. The contribution of thermal pressure was determined to have the form ΔPth = [αKT(Va) + (∂KT/∂T)Vln(Va/V)]ΔT. When αKT(Va) was assumed to be constant, the fit to the data yielded αKT(Va) = 0.0056 GPa/K. In contrast, the volume dependence of the thermal pressure was very small, and fitting yielded a value of (∂KT/∂T)V = −0.00042.

Genesis of a CO2-rich and H2O-depleted atmosphere from Earth's early global magma ocean.

The magma ocean was a important reservoir for Earth’s primary volatiles. Understanding the volatile fluxes between the early atmosphere and the magma ocean is fundamental for quantifying the volatile budget of our planet. Here we investigate the vaporization of carbon and hydrogen at the boundary between the magma ocean and the thick, hot early atmosphere using first-principles molecular dynamics calculations. We find that carbon is rapidly devolatilized, while hydrogen mostly remains dissolved in the magma during the existence of a thick silicate-bearing atmosphere. In the early stages of the magma ocean, the atmosphere would have contained significantly more carbon than hydrogen, and the high concentrations of carbon dioxide would have prolonged the cooling time of early Earth.

Soft X-ray Emission Studies on Hydrate-Melt Electrolytes.

Highly salt-concentrated aqueous solutions are a new class of electrolytes, which provide a wide potential window exceeding 3 V and, hence, realize possibly inexpensive, safe, and high-energy-density storage devices. Herein, we investigate the evolution of the coordination structure and electronic state depending on the salt concentration through soft X-ray emission spectroscopy and first-principles molecular dynamics calculations. Close to the concentration limit, categorized as a "hydrate melt," a long-range hydrogen-bond network of water molecules disappears with emerging localized electronic states that resemble those in the gas phase. Such localized electronic states are attributed not only to their geometrically isolated nature but also to their dominant electrostatic interaction with Li+ cations. Therefore, the electrical properties of water in the hydrate melt can be more gaslike than liquidlike.