Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “MD”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 685 records · Page 38

Temperature dependence of creep compliance of highly cross-linked epoxy: A molecular simulation study

We have used molecular dynamics (MD) simulations to study the effect of temperature on the creep compliance of neat cross-linked epoxy. Experimental studies of mechanical behavior of cross-linked epoxy in literature commonly report creep compliance values, whereas molecular simulations of these systems have primarily focused on the Young’s modulus. In this work, in order to obtain a more direct comparison between experiments and simulations, atomistically detailed models of the cross-linked epoxy are used to study their creep compliance as a function of temperature using MD simulations. The creep tests are performed by applying a constant tensile stress and monitoring the resulting strain in the system. Our results show that simulated values of creep compliance increase with an increase in both time and temperature. We believe that such calculations of the creep compliance, along with the use of time temperature superposition, hold great promise in connecting the molecular insight obtained from molecular simulation at small length- and time-scales with the experimental behavior of such materials. To the best of our knowledge, this work is the first reported effort that investigates the creep compliance behavior of cross-linked epoxy using MD simulations.

36 MATERIALS SCIENCE↗

Molecular dynamics study of Si(100)-oxidation: SiO and Si emissions from Si/SiO{sub 2} interfaces and their incorporation into SiO{sub 2}

Dynamics of Si(100)-oxidation processes at the Si/SiO{sub 2} interface and in the SiO{sub 2} region are investigated focusing on SiO and Si emissions from the interface and the following incorporation into the SiO{sub 2} and/or substrate. Classical molecular dynamics (MD) simulations with variable charge interatomic potentials are performed to clarify these atomic processes. By incorporating oxygen atoms, two-folded Si atoms are formed after structural relaxation at the interface and are emitted as SiO molecules into SiO{sub 2}. The energy barrier of the SiO emission is estimated to be 1.20 eV on the basis of the enthalpy change in an MD simulation. The emitted SiO molecule is incorporated into the SiO{sub 2} network through a Si-O rebonding process with generating an oxygen vacancy. The energy barrier of the SiO incorporation is estimated to be 0.79–0.81 eV. The elementary process of oxygen vacancy diffusion leading to the complete SiO incorporation is also simulated, and the energy barriers are found to be relatively small, 0.71–0.79 eV. The energy changes of Si emissions into the substrate and SiO{sub 2} are estimated to be 2.97–7.81 eV, which are larger than the energy barrier of the SiO emission. This result suggests that, at the ideally flat Si/SiO{sub 2} interface, the SiO emission into the SiO{sub 2} region occurs prior to the Si emission, which is consistent with previous theoretical and experimental studies. The above mentioned typical atomic processes are successfully extracted from some (or one) of MD simulations among many trials in which a statistical procedure is partly employed. Our results give a unified understanding of Si oxidation processes from an atomistic point of view.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Local atomic order, electronic structure and electron transport properties of Cu-Zr metallic glasses

We studied atomic and electronic structures of binary Cu-Zr metallic glasses (MGs) using combined experimental and computational methods including X-ray absorption fine structure spectroscopy, electrical resistivity, thermoelectric power (TEP) measurements, molecular dynamics (MD) simulations, and ab-initio calculations. The results of MD simulations and extended X-ray absorption fine structure analysis indicate that atomic order of Cu-Zr MGs and can be described in terms of interpenetrating icosahedral-like clusters involving five-fold symmetry. MD configurations were used as an input for calculations of theoretical electronic density of states (DOS) functions which exhibits good agreement with the experimental X-ray absorption near-edge spectra. We found no indication of minimum of DOS at Fermi energy predicted by Mott's nearly free electron (NFE) model for glass-forming alloys. The theoretical DOS was subsequently used to test Mott's model describing the temperature variation of electrical resistivity and thermoelectric power of transition metal-based MGs. We demonstrate that the measured temperature variations of electrical resistivity and TEP remain in a contradiction with this model. On the other hand, the experimental temperature dependence of electrical resistivity can be explained by incipient localization of conduction electrons. It is shown that weak localization model works up to relatively high temperatures when localization is destroyed by phonons. Our results indicate that electron transport properties of Cu-Zr MGs are dominated by localization effects rather than by electronic structure. We suggest that NFE model fails to explain a relatively high glass-forming ability of binary Cu-Zr alloys.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Rare event molecular dynamics simulations of plasma induced surface ablation

The interaction of thermal Ar plasma particles with Si and W surfaces is modeled using classical molecular dynamics (MD) simulations. At plasma energies above the threshold for ablation, the ablation yield can be calculated directly from MD. For plasma energies below threshold, the ablation yield becomes exponentially low, and direct MD simulations are inefficient. Instead, we propose an integration method where the yield is calculated as a function of the Ar incident kinetic energy. Subsequent integration with a Boltzmann distribution at the temperature of interest gives the thermal ablation yield. At low plasma temperatures, the ablation yield follows an Arrhenius form in which the activation energy is shown to be the threshold energy for ablation. Interestingly, equilibrium material properties, including the surface and bulk cohesive energy, are not good predictors of the threshold energy for ablation. The surface vacancy formation energy is better, but is still not a quantitative predictor. An analysis of the trajectories near threshold shows that ablation occurs by different mechanisms on different material surfaces, and both the mechanism and the binding of surface atoms determine the threshold energy.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Diffusional motion of redox centers in carbonate electrolytes

Ferrocene (Fc) and N-(ferrocenylmethyl)-N,N-dimethyl-N-ethylammonium bistrifluoromethyl-sulfonimide (Fc1N112-TFSI) were dissolved in carbonate solvents and self-diffusion coefficients (D) of solutes and solvents were measured by {sup 1}H and {sup 19}F pulsed field gradient nuclear magnetic resonance (NMR) spectroscopy. The organic solvents were propylene carbonate (PC), ethyl methyl carbonate (EMC), and a ternary mixture that also includes ethylene carbonate (EC). Results from NMR studies over the temperature range of 0–50 °C and for various concentrations (0.25–1.7 M) of Fc1N112-TFSI are compared to values of D simulated with classical molecular dynamics (MD). The measured self-diffusion coefficients gradually decreased as the Fc1N112-TFSI concentration increased in all solvents. Since TFSI{sup −} has fluoromethyl groups (CF{sub 3}), D{sub TFSI} could be measured separately and the values found are larger than those for D{sub Fc1N112} in all samples measured. The EC, PC, and EMC have the same D in the neat solvent mixture and when Fc is dissolved in EC/PC/EMC at a concentration of 0.2 M, probably due to the interactions between common carbonyl structures within EC, PC, and EMC. A difference in D (D{sub PC} < D{sub EC} < D{sub EMC}), and both a higher E{sub a} for translational motion and higher effective viscosity for PC in the mixture containing Fc1N112-TFSI reflect the interaction between PC and Fc1N112{sup +}, which is a relatively stronger interaction than that between Fc1N112{sup +} and other solvent species. In the EC/PC/EMC solution that is saturated with Fc1N112-TFSI, we find that D{sub PC} = D{sub EC} = D{sub EMC} and Fc1N112{sup +} and all components of the EC/PC/EMC solution have the same E{sub a} for translational motion, while the ratio D{sub EC/PC/EMC}/D{sub Fc1N112} is approximately 3. These results reflect the lack of available free volume for independent diffusion in the saturated solution. The Fc1N112{sup +} transference numbers lie around 0.4 and increase slightly as the temperature is increased in the PC and EMC solvents. The trends observed for D from simulations are in good agreement with experimental results and provide molecular level understanding of the solvation structure of Fc1N112-TFSI dissolved in EC/PC/EMC.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A molecular dynamics study of nuclear quantum effect on the diffusion of hydrogen in condensed phase

In this paper, the quantum effect of hydrogen molecule on its diffusivity is analyzed using Molecular Dynamics (MD) method. The path integral centroid MD (CMD) method is applied for the reproduction method of time evolution of the molecules. The diffusion coefficient of liquid hydrogen is calculated using the Green-Kubo method. The simulation is performed at wide temperature region and the temperature dependence of the quantum effect of hydrogen molecule is addressed. The calculation results are compared with those of classical MD results. As a result, it is confirmed that the diffusivity of hydrogen molecule is changed depending on temperature by the quantum effect. It is clarified that this result can be explained that the dominant factor by quantum effect on the diffusivity of hydrogen changes from the swollening the potential to the shallowing the potential well around 30 K. Moreover, it is found that this tendency is related to the temperature dependency of the ratio of the quantum kinetic energy and classical kinetic energy.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Does fluoride disrupt hydrogen bond network in cationic lipid bilayer? Time-dependent fluorescence shift of Laurdan and molecular dynamics simulations

Time-dependent fluorescence shift (TDFS) of Laurdan embedded in phospholipid bilayers reports on hydration and mobility of the phospholipid acylgroups. Exchange of H{sub 2}O with D{sub 2}O prolongs the lifetime of lipid-water and lipid-water-lipid interactions, which is reflected in a significantly slower TDFS kinetics. Combining TDFS measurements in H{sub 2}O and D{sub 2}O hydrated bilayers with atomistic molecular dynamics (MD) simulations provides a unique tool for characterization of the hydrogen bonding at the acylgroup level of lipid bilayers. In this work, we use this approach to study the influence of fluoride anions on the properties of cationic bilayers composed of trimethylammonium-propane (DOTAP). The results obtained for DOTAP are confronted with those for neutral phosphatidylcholine (DOPC) bilayers. Both in DOTAP and DOPC H{sub 2}O/D{sub 2}O exchange prolongs hydrogen-bonding lifetime and does not disturb bilayer structure. These results are confirmed by MD simulations. TDFS experiments show, however, that for DOTAP this effect is cancelled in the presence of fluoride ions. We interpret these results as evidence that strongly hydrated fluoride is able to steal water molecules that bridge lipid carbonyls. Consequently, when attracted to DOTAP bilayer, fluoride disrupts the local hydrogen-bonding network, and the differences in TDFS kinetics between H{sub 2}O and D{sub 2}O hydrated bilayers are no longer observed. A distinct behavior of fluoride is also evidenced by MD simulations, which show different lipid-ion binding for Cl{sup −} and F{sup −}.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hybrid molecular-continuum simulations using smoothed dissipative particle dynamics

We present a new multiscale simulation methodology for coupling a region with atomistic detail simulated via molecular dynamics (MD) to a numerical solution of the fluctuating Navier-Stokes equations obtained from smoothed dissipative particle dynamics (SDPD). In this approach, chemical potential gradients emerge due to differences in resolution within the total system and are reduced by introducing a pairwise thermodynamic force inside the buffer region between the two domains where particles change from MD to SDPD types. When combined with a multi-resolution SDPD approach, such as the one proposed by Kulkarni et al. [J. Chem. Phys. 138, 234105 (2013)], this method makes it possible to systematically couple atomistic models to arbitrarily coarse continuum domains modeled as SDPD fluids with varying resolution. We test this technique by showing that it correctly reproduces thermodynamic properties across the entire simulation domain for a simple Lennard-Jones fluid. Furthermore, we demonstrate that this approach is also suitable for non-equilibrium problems by applying it to simulations of the start up of shear flow. The robustness of the method is illustrated with two different flow scenarios in which shear forces act in directions parallel and perpendicular to the interface separating the continuum and atomistic domains. In both cases, we obtain the correct transient velocity profile. We also perform a triple-scale shear flow simulation where we include two SDPD regions with different resolutions in addition to a MD domain, illustrating the feasibility of a three-scale coupling.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Low temperature magneto-dielectric measurements on BiFeO{sub 3} lightly substituted by cobalt

Dielectric and magnetodielectric measurements are done on BiFe{sub 1−x}Co{sub x}O{sub 3}: x = 0, 0.01, and 0.02 in the temperature range 70–300 K and up to magnetic field 1.3 T. The dielectric data are well described by Haverliak–Negami expression plus an additional term for the Maxwell Wagner (MW) type relaxations, whose contribution is dominant near room temperature. The parameters obtained from the fitting of data using the above mentioned expression, suggest slowing down of relaxation and approach towards ideal Debye type relaxations, as the temperature is lowered. The dielectric relaxations obey polaronic variable range hopping model with distinct activation energies (E{sub a}) in the extrinsic (6.67T{sup 3/4 }meV) and intrinsic (2.88T{sup 3/4 }meV) regions for the parent sample (x = 0), and thus a distinct transition from extrinsic to intrinsic behavior is seen at 215 K while lowering the temperature. This distinct transition is missing for Co substituted samples probably due to the extrinsic region values of E{sub a} (3.42T{sup 3/4 }meV and 2.42T{sup 3/4 }meV for x = 0.01 and 0.02, respectively) comparable to that of the intrinsic region (see x = 0). The magnetodielectric measurement shows positive magnetodielectricity (MD) in the intrinsic region (T < 215 K for x = 0) and negative MD in the extrinsic region (T > 215 K for x = 0). The extrinsic region is found to be dominated by MW and magnetoresistance effects, whereas MD in intrinsic regions is due to the spin reorientation transitions. The Co substitution is found to increase the extrinsic and non-Debye contributions to dielectricity, which becomes so large that no spin reorientation transitions are seen in x = 0.02 sample. The pyroelectric active region in x = 0 is found to be dominated by the diffusive behavior having contribution of the form ω{sup −0.5}.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

On the importance of collective excitations for thermal transport in graphene

We use equilibrium molecular dynamics (MD) simulations to study heat transport in bulk single-layer graphene. Through a modal analysis of the MD trajectories employing a time-domain formulation, we find that collective excitations involving flexural acoustic (ZA) phonons, which have been neglected in the previous MD studies, actually dominate the heat flow, generating as much as 78% of the flux. These collective excitations are, however, much less significant if the atomic displacements are constrained in the lattice plane. Although relaxation is slow, we find graphene to be a regular (non-anomalous) heat conductor for sample sizes of order 40 μm and more.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Efficient preconditioning of the electronic structure problem in large scale ab initio molecular dynamics simulations

We present an improved preconditioning scheme for electronic structure calculations based on the orbital transformation method. First, a preconditioner is developed which includes information from the full Kohn-Sham matrix but avoids computationally demanding diagonalisation steps in its construction. This reduces the computational cost of its construction, eliminating a bottleneck in large scale simulations, while maintaining rapid convergence. In addition, a modified form of Hotelling’s iterative inversion is introduced to replace the exact inversion of the preconditioner matrix. This method is highly effective during molecular dynamics (MD), as the solution obtained in earlier MD steps is a suitable initial guess. Filtering small elements during sparse matrix multiplication leads to linear scaling inversion, while retaining robustness, already for relatively small systems. For system sizes ranging from a few hundred to a few thousand atoms, which are typical for many practical applications, the improvements to the algorithm lead to a 2-5 fold speedup per MD step.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Phonon-phonon interactions: First principles theory

We present the details of a method to perform molecular-dynamics (MD) simulations without thermostat and with very small temperature fluctuations ±ΔT starting with MD step 1. It involves preparing the supercell at the time t = 0 in physically correct microstates using the eigenvectors of the dynamical matrix. Each initial microstate corresponds to a different distribution of kinetic and potential energies for each vibrational mode (the total energy of each microstate is the same). Averaging the MD runs over many initial microstates further reduces ΔT. The electronic states are obtained using first-principles theory (density-functional theory in periodic supercells). Three applications are discussed: the lifetime and decay of vibrational excitations, the isotope dependence of thermal conductivities, and the flow of heat at an interface.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Shock melting method to determine melting curve by molecular dynamics: Cu, Pd, and Al

A melting simulation method, the shock melting (SM) method, is proposed and proved to be able to determine the melting curves of materials accurately and efficiently. The SM method, which is based on the multi-scale shock technique, determines melting curves by preheating and/or prepressurizing materials before shock. This strategy was extensively verified using both classical and ab initio molecular dynamics (MD). First, the SM method yielded the same satisfactory melting curve of Cu with only 360 atoms using classical MD, compared to the results from the Z-method and the two-phase coexistence method. Then, it also produced a satisfactory melting curve of Pd with only 756 atoms. Finally, the SM method combined with ab initio MD cheaply achieved a good melting curve of Al with only 180 atoms, which agrees well with the experimental data and the calculated results from other methods. It turned out that the SM method is an alternative efficient method for calculating the melting curves of materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Coarse-grained molecular dynamics study of membrane fusion: Curvature effects on free energy barriers along the stalk mechanism

The effects of membrane curvature on the free energy barrier for membrane fusion have been investigated using coarse-grained molecular dynamics (CG-MD) simulations, assuming that fusion takes place through a stalk intermediate. Free energy barriers were estimated for stalk formation as well as for fusion pore formation using the guiding potential method. Specifically, the three different geometries of two apposed membranes were considered: vesicle–vesicle, vesicle–planar, and planar–planar membranes. The free energy barriers for the resulting fusion were found to depend importantly on the fusing membrane geometries; the lowest barrier was obtained for vesicular membranes. Further, lipid sorting was observed in fusion of the mixed membranes of dimyristoyl phosphatidylcholine and dioleoyl phosphatidylethanolamine (DOPE). Specifically, DOPE molecules were found to assemble around the stalk to support the highly negative curved membrane surface. A consistent result for lipid sorting was observed when a simple continuum model (CM) was used, where the Helfrich energy and mixing entropy of the lipids were taken into account. However, the CM predicts a much higher free energy barrier than found using CG-MD. This discrepancy originates from the conformational changes of lipids, which were not considered in the CM. The results of the CG-MD simulations reveal that a large conformational change in the lipid takes place around the stalk region, which results in a reduction of free energy barriers along the stalk mechanism of membrane fusion.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Molecular dynamics simulations of Si etching in Cl- and Br-based plasmas: Cl{sup +} and Br{sup +} ion incidence in the presence of Cl and Br neutrals

Classical molecular dynamics (MD) simulations have been performed for Cl{sup +} and Br{sup +} ions incident on Si(100) surfaces with Cl and Br neutrals, respectively, to gain a better understanding of the ion-enhanced surface reaction kinetics during Si etching in Cl- and Br-based plasmas. The ions were incident normally on surfaces with translational energies in the range E{sub i} = 20–500 eV, and low-energy neutrals of E{sub n} = 0.01 eV were also incident normally thereon with the neutral-to-ion flux ratio in the range Γ{sub n}{sup 0}/Γ{sub i}{sup 0} = 0–100, where an improved Stillinger--Weber potential form was employed for the interatomic potential concerned. The etch yields and thresholds presently simulated were in agreement with the experimental results previously reported for Si etching in Cl{sub 2} and Br{sub 2} plasmas as well as in Cl{sup +}, Cl{sub 2}{sup +}, and Br{sup +} beams, and the product stoichiometry simulated was consistent with that observed during Ar{sup +} beam incidence on Si in Cl{sub 2}. Moreover, the surface coverage of halogen atoms, halogenated layer thickness, surface stoichiometry, and depth profile of surface products simulated for Γ{sub n}{sup 0}/Γ{sub i}{sup 0} = 100 were in excellent agreement with the observations depending on E{sub i} reported for Si etching in Cl{sub 2} plasmas. The MD also indicated that the yield, coverage, and surface layer thickness are smaller in Si/Br than in Si/Cl system, while the percentage of higher halogenated species in product and surface stoichiometries is larger in Si/Br. The MD further indicated that in both systems, the translational energy distributions of products and halogen adsorbates desorbed from surfaces are approximated by two Maxwellians of temperature T{sub 1} ≈ 2500 K and T{sub 2} ≈ 7000–40 000 K. These energy distributions are discussed in terms of the desorption or evaporation from hot spots formed through chemically enhanced physical sputtering and physically enhanced chemical sputtering, which have so far been speculated to both occur in the ion-enhanced surface reaction kinetics of plasma etching.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A polarizable QM/MM approach to the molecular dynamics of amide groups solvated in water

The infrared (IR) spectra of polypeptides are dominated by the so-called amide bands. Because they originate from the strongly polar and polarizable amide groups (AGs) making up the backbone, their spectral positions sensitively depend on the local electric fields. Aiming at accurate computations of these IR spectra by molecular dynamics (MD) simulations, which derive atomic forces from a hybrid quantum and molecular mechanics (QM/MM) Hamiltonian, here we consider the effects of solvation in bulk liquid water on the amide bands of the AG model compound N-methyl-acetamide (NMA). As QM approach to NMA we choose grid-based density functional theory (DFT). For the surrounding MM water, we develop, largely based on computations, a polarizable molecular mechanics (PMM) model potential called GP6P, which features six Gaussian electrostatic sources (one induced dipole, five static partial charge distributions) and, therefore, avoids spurious distortions of the DFT electron density in hybrid DFT/PMM simulations. Bulk liquid GP6P is shown to have favorable properties at the thermodynamic conditions of the parameterization and beyond. Lennard-Jones (LJ) parameters of the DFT fragment NMA are optimized by comparing radial distribution functions in the surrounding GP6P liquid with reference data obtained from a “first-principles” DFT-MD simulation. Finally, IR spectra of NMA in GP6P water are calculated from extended DFT/PMM-MD trajectories, in which the NMA is treated by three different DFT functionals (BP, BLYP, B3LYP). Method-specific frequency scaling factors are derived from DFT-MD simulations of isolated NMA. The DFT/PMM-MD simulations with GP6P and with the optimized LJ parameters then excellently predict the effects of aqueous solvation and deuteration observed in the IR spectra of NMA. As a result, the methods required to accurately compute such spectra by DFT/PMM-MD also for larger peptides in aqueous solution are now at hand.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Near room temperature magnetodielectric consequence in (Li, Ti) doped NiO ceramic

In the quest for high-k dielectrics with decent magnetodielectric (MD) response, ball-milled processed (Li, Ti) doped NiO ceramics have been evaluated by various experimental techniques. Magnetic properties in these ceramics manifest with a pronounced anomaly appearing at ∼260 K, suggesting ferrimagnetic phase (related to cluster magnetism) and co-existence with a glassy-like antiferromagnetic phase at ∼7 K. Room temperature neutron diffraction pattern shows the existence of short-range magnetic correlations. In the magnetically ordered state below 250 K, the magnetic structure is found to be phase coexistence of G-type antiferromagnet and ferrimagnet. Impedance spectroscopy measurements over a wide temperature range can be perfectly described with appropriate microstructural model (internal barrier layer capacitor), based on domain and domain boundary relaxations, justifying the enhancement of the dielectric response. The low-temperature (T < 100 K) dielectric relaxation is polaronic in nature, associated with the charge ordering of a mixed valence states of Ti ions (co-existence of Ti{sup 3+} and Ti{sup 4+}). Finally, our investigations in external magnetic fields up to 15 T reveal the occurrence of negative MD effect near room temperature. This intriguing intrinsic feature has been understood by the mechanism of charge-hopping-mediated MD effects.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A unified relation for the solid-liquid interface free energy of pure FCC, BCC, and HCP metals

We study correlations between the solid-liquid interface (SLI) free energy and bulk material properties (melting temperature, latent heat, and liquid structure) through the determination of SLI free energies for bcc and hcp metals from molecular dynamics (MD) simulation. Values obtained for the bcc metals in this study were compared to values predicted by the Turnbull, Laird, and Ewing relations on the basis of previously published MD simulation data. We found that of these three empirical relations, the Ewing relation better describes the MD simulation data. Moreover, whereas the original Ewing relation contains two constants for a particular crystal structure, we found that the first coefficient in the Ewing relation does not depend on crystal structure, taking a common value for all three phases, at least for the class of the systems described by embedded-atom method potentials (which are considered to provide a reasonable approximation for metals).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗