Prediction of the melting point of n-alkanes using the molecular dynamics method

Insight into melting point differences of dinitroimidazoles and dinitropyrazoles from the perspective of intermolecular interactions

Dinitroimidazole (DNI) and dinitropyrazole (DNP), along with their congeners, possess similar molecular structures but exhibit distinct melting points. To analyse and elucidate the fundamental reasons for property differences from the perspective of intermolecular interactions, we proposed a simplified approach named binding energy in clusters (BEC) in computing weak interactions within complex crystal systems. Based on the results of the symmetry-adapted perturbation theory (SAPT) calculations, an approximate estimation of the melting point range can be derived by taking into account the cumulative effect (energy of electrostatic, dispersion and induction terms) and repulsive effect (energy of exchange term) values. We have also proposed a formula for calculating the specific melting point, which indicates that stronger intermolecular interactions have a major impact on the melting point, while the distribution of weak interactions also affects the melting point. This work would provide an effective reference for molecular design and structure-performance analysis.

Computational assessment of hexadecane freezing by equilibrium atomistic molecular dynamics simulations

HYPOTHESIS

Upon cooling, alkanes can form intermediate phases between liquid and crystal. They are called "rotator" or "plastic" phases and have long-range positional order with rotational freedom around the long molecular axis which gives them non-trivial and useful visco-plastic properties. We expect that the formation and structure of rotator phases formed in freezing alkanes can be understood much deeper by tracking the process at molecular level with atomistic molecular dynamics.

SIMULATIONS

We defined an appropriate CHARMM36-based computational protocol for simulating the freezing of hexadecane, which contained a sufficiently long (500 ns) equilibrium sampling of the frozen states. We employed it to simulate successfully the freezing of bulk and interface-contacting hexadecane and to provide a pioneering clarification of the effect of surfactant on the crystallization mechanism and on the type of intermolecular ordering in the crystallites.

FINDINGS

The devised computational protocol was able to reproduce the experimentally observed polycrystalline structure formed upon cooling. However, different crystallization mechanisms were established for the two types of models. Crystallites nucleate at random locations in the bulk and start growing rapidly within tens of nanoseconds. In contrast, the surfactants freeze first during the fast cooling (<1 ns), followed by rapid hexadecane freezing, with nucleation starting along the entire surfactant adsorption layer. Thereby, the hexadecane molecules form rotator phases which transition into a more stable ordered phase. This collective transition is first-time visualized directly. The developed robust computational protocol creates a foundation for future in-depth modelling and analysis of solid-state alkane-containing, incl. lipid, structures.

Predicting the Melting Point of Energetic Molecules Using a Learnable Graph Neural Fingerprint Model

Melting point prediction for organic molecules has drawn widespread attention from both academic and industrial communities. In this work, a learnable graph neural fingerprint (GNF) was employed to develop a melting point prediction model using a dataset of over 90,000 organic molecules. The GNF model exhibited a significant advantage, with a mean absolute error (MAE) of 25.0 K, when compared to other featurization methods. Furthermore, by integrating prior knowledge through a customized descriptor set (i.e., CDS) into GNF, the accuracy of the resulting model, GNF_CDS, improved to 24.7 K, surpassing the performance of previously reported models for a wide range of structurally diverse organic compounds. Moreover, the generalizability of the GNF_CDS model was significantly improved with a decreased MAE of 17 K for an independent dataset containing melt-castable energetic molecules. This work clearly demonstrates that prior knowledge is still beneficial for modeling molecular properties despite the powerful learning capability of graph neural networks, especially in specific fields where chemical data are lacking.

Computational Procedure for Analysis of Crystallites in Polycrystalline Solids of Quasilinear Molecules

In the current work, a comprehensive procedure for structural analysis of quasilinear organic molecules arranged in a polycrystalline sample generated by molecular dynamics is developed. A linear alkane, hexadecane, is used as a test case because of its interesting behavior upon cooling. Instead of a direct transition from isotropic liquid to the solid crystalline phase, this compound forms first a short-lived intermediate state known as a "rotator phase". The rotator phase and the crystalline one are distinguished by a set of structural parameters. We propose a robust methodology to evaluate the type of ordered phase obtained after a liquid-to-solid phase transition in a polycrystalline assembly. The analysis starts with the identification and separation of the individual crystallites. Then, the eigenplane of each of them is fit and the tilt angle of the molecules relative to it is computed. The average area per molecule and the distance to the nearest neighbors are estimated by a 2D Voronoi tessellation. The orientation of the molecules with respect to each other is quantified by visualization of the second molecular principal axis. The suggested procedure may be applied to different quasilinear organic compounds in the solid state and to various data compiled in a trajectory.

Molecular dynamics simulations of phase change materials for thermal energy storage: a review

Phase change materials (PCM) have had a significant role as thermal energy transfer fluids and nanofluids and as media for thermal energy storage. Molecular dynamics (MD) simulations, can play a significant role in addressing several thermo-physical problems of PCMs at the atomic scale by providing profound insights and new information. In this paper, the reviewed research is classified into five groups: pure PCM, mixed PCM, PCM containing nanofillers, nano encapsulated PCM, and PCM in nanoporous media. A summary of the equilibrium and non-equilibrium MD simulations of PCMs and their results is presented as well. The primary results of the simulated systems are demonstrated to be efficient in manufacturing phase change materials with better thermal energy storage features. The goals of these studies are to achieve higher thermal conductivity, higher thermal capacity, and lower density change, determine the melting point, and understand the molecular behaviors of PCM composites. A molecular dynamics-based grouping (PCM simulation table) was presented that is very useful for the future roadmap of PCM simulation. In the end, the PCFF force field is presented in detail and a case problem is studied for more clarity. The results show that simulating the PCMs with a similar strategy could be performed systematically. Results of investigations of thermal conductivity enhancement showed that this characteristic can be increased at the nano-scale by the orientation of PCM molecules.

Phase change materials (PCM) have had a significant role as thermal energy transfer fluids and nanofluids and as media for thermal energy storage.

Fluid-solid phase transition of n-alkane mixtures: Coarse-grained molecular dynamics simulations and diffusion-ordered spectroscopy nuclear magnetic resonance

Wax appearance temperature (WAT), defined as the temperature at which the first solid paraffin crystal appears in a crude oil, is one of the key flow assurance indicators in the oil industry. Although there are several commonly-used experimental techniques to determine WAT, none provides unambiguous molecular-level information to characterize the phase transition between the homogeneous fluid and the underlying solid phase. Molecular Dynamics (MD) simulations employing the statistical associating fluid theory (SAFT) force field are used to interrogate the incipient solidification states of models for long-chain alkanes cooled from a melt to an arrested state. We monitor the phase change of pure long chain n-alkanes: tetracosane (C₂₄H₅₀) and triacontane (C₃₀H₆₂), and an 8-component surrogate n-alkane mixture (C₁₂-C₃₃) built upon the compositional information of a waxy crude. Comparison to Diffusion Ordered Spectroscopy Nuclear Magnetic Resonance (DOSY NMR) results allows the assessment of the limitations of the coarse-grained models proposed. We show that upon approach to freezing, the heavier components restrict their motion first while the lighter ones retain their mobility and help fluidize the mixture. We further demonstrate that upon sub-cooling of long n-alkane fluids and mixtures, a discontinuity arises in the slope of the self-diffusion coefficient with decreasing temperature, which can be employed as a marker for the appearance of an arrested state commensurate with conventional WAT measurements.

Structural changes of layered alkylsiloxanes during the reversible melting-solidification process

Through various in situ analyses, we have revealed the structural changes that occur during the reversible melting-solidification process of layered alkylsiloxanes (CnLSiloxanes) with carbon numbers (n) of 18 and 16. In situ high-resolution solid-state (13)C nuclear magnetic resonance (NMR) analysis at controlled temperatures indicates drastic conformational changes of the long alkyl chains during the melting-solidification process. A (13)C NMR signal at 33 ppm, which shows the highest intensity at room temperature (RT), is assigned to an inner methylene group with an all-trans conformation. As the temperature increases, the 33-ppm signal intensity decreases while the signal intensity at 30.5 ppm simultaneously increases. The 30.5 ppm signal is assigned to an inner methylene group with a trans-gauche conformation. Subsequently, upon cooling, the signal at 33 ppm recovers, even after CnLSiloxanes have melted. In situ X-ray diffraction measurements at controlled temperatures reveal that the ordered arrangement of the long alkyl chains becomes disordered with elevating temperatures and reordered upon cooling to RT. In situ high-resolution solid-state (29)Si NMR analysis shows that the melting-solidification process progresses without any structural change in siloxane sheets of the CnLSiloxanes. Thus, the in situ analyses show that disordering of the long alkyl chains causes the CnLSiloxanes to melt. Because the majority of long alkyl chains are packed again in the ordered arrangement with the all-trans conformation upon cooling, the CnLSiloxanes are reversibly solidified and the CnLSiloxane structure is recovered.

On the rational formulation of alternative fuels: melting point and net heat of combustion predictions for fuel compounds using machine learning methods

We report the development of predictive models for two fuel specifications: melting points (T(m)) and net heat of combustion (Δ(c)H). Compounds inside the scope of these models are those likely to be found in alternative fuels, i.e. hydrocarbons, alcohols and esters. Experimental T(m) and Δ(c)H values for these types of molecules have been gathered to generate a unique database. Various quantitative structure-property relationship (QSPR) approaches have been used to build models, ranging from methods leading to multi-linear models such as genetic function approximation (GFA), or partial least squares (PLS) to those leading to non-linear models such as feed-forward artificial neural networks (FFANN), general regression neural networks (GRNN), support vector machines (SVM), or graph machines. Except for the case of the graph machines method for which the only inputs are SMILES formulae, previously listed approaches working on molecular descriptors and functional group count descriptors were used to develop specific models for T(m) and Δ(c)H. For each property, the predictive models return slightly different responses for each molecular structure. Therefore, models labelled as 'consensus models' were built by averaging values computed with selected individual models. Predicted results were then compared with experimental data and with predictions of models in the literature.

Molecular structural property and potential energy dependence on nonequilibrium-thermodynamic state point of liquid n-hexadecane under shear

Extensive computer experiments have been conducted in order to shed light on the macroscopic shear flow behavior of liquid n-hexadecane fluid under isobaric-isothermal conditions through the nonequilibrium molecular dynamic methodology. With respect to shear rates, the accompanying variations in structural properties of the fluid span the microscopic range of understanding from the intrinsic to extrinsic characteristics. As drawn from the average value of bond length and bond angle, the distribution of dihedral angle, and the radius distribution function of intramolecular and intermolecular van der Waals distances, these intrinsic structures change with hardness, except in the situation of extreme shear rates. The shear-induced variation of thermodynamic state curve along with the shear rate studied is shown to consist of both the quasiequilibrium state plateau and the nonequilibrium-thermodynamic state slope. Significantly, the occurrence of nonequilibrium-thermodynamic state behavior is attributed to variations in molecular potential energies, which include bond stretching, bond bending, bond torsion, and intra- and intermolecular van der Waals interactions. To unfold the physical representation of extrinsic structural deformation, under the aggressive influence of a shear flow field, the molecular dimension and appearance can be directly described via the squared radius of gyration and the sphericity angle, R(g)(2) and ϕ, respectively. In addition, a specific orientational order S(x) defines the alignment of the molecules with the flow direction of the x-axis. As a result, at low shear rates, the overall molecules are slightly stretched and shaped in a manner that is increasingly ellipsoidal. Simultaneously, there is an obvious enhancement in the order. In contrast to high shear rates, the molecules spontaneously shrink themselves with a decreased value of R(g)(2), while their shape and order barely vary with an infinite value of ϕ and S(x). It is important to note that under different temperatures and pressures, these three parameters are integrated within a molecular description in response to thermodynamic state variable of density and rheological material function of shear viscosity.

Linear viscoelasticity and thermorheological simplicity of n-hexadecane fluids under oscillatory shear via non-equilibrium molecular dynamics simulations

A small amplitude oscillatory shear flows with the classic characteristic of a phase shift when using non-equilibrium molecular dynamics simulations for n-hexadecane fluids. In a suitable range of strain amplitude, the fluid possesses significant linear viscoelastic behavior. Non-linear viscoelastic behavior of strain thinning, which means the dynamic modulus monotonously decreased with increasing strain amplitudes, was found at extreme strain amplitudes. Under isobaric conditions, different temperatures strongly affected the range of linear viscoelasticity and the slope of strain thinning. The fluid's phase states, containing solid-, liquid-, and gel-like states, can be distinguished through a criterion of the viscoelastic spectrum. As a result, a particular condition for the viscoelastic behavior of n-hexadecane molecules approaching that of the Rouse chain was obtained. Besides, more importantly, evidence of thermorheologically simple materials was presented in which the relaxation modulus obeys the time-temperature superposition principle. Therefore, using shift factors from the time-temperature superposition principle, the estimated Arrhenius flow activation energy was in good agreement with related experimental values. Furthermore, one relaxation modulus master curve well exhibited both transition and terminal zones. Especially regarding non-equilibrium thermodynamic states, variations in the density, with respect to frequencies, were revealed.

Molecular dynamics simulation of nanoconfinement induced organization of n-decane

Molecular dynamics (MD) simulations are used to study the behavior of n-decane under sub-10 nm confinement between two gold {111} surfaces. This confinement and dielectric medium are characteristic of those used in nanoscale electromachining (nano-EM) processes; thus, it is important that the behavior of the nanoconfined dielectric medium be investigated for better process understanding. Results obtained via MD simulations indicate that, when confined down to a thickness less than 1 nm, the mechanical boundary conditions trigger organization in the n-decane medium, resulting in two distinct molecular layers. The n-decane chains lie flat on the {111} gold surfaces and show preferred orientation in the close-packed 110 crystallographic directions. A 4-fold increase in the maximum local density as compared with the experimental bulk (liquid) density is observed at the interface between the molecular medium and the gold {111} surfaces, regardless of confinement spacing. Radial distribution function curves are used to quantitatively examine organization of the medium into molecular layers. The deliberate introduction of ledges (atomic steps) on the gold surface triggers a preferred alignment of the n-decane chains toward the boundaries of the ledges.

Master curves and radial distribution functions for shear dilatancy of liquid n-hexadecane via nonequilibrium molecular dynamics simulations

Shear dilatancy, a significant nonlinear behavior of nonequilibrium thermodynamics states, has been observed in nonequilibrium molecular dynamics (NEMD) simulations for liquid n-hexadecane fluid under extreme shear conditions. The existence of shear dilatancy is relevant to the relationship between the imposed shear rate gamma and the critical shear rate gamma(c). Consequently, as gamma<gamma(c), the intermolecular equilibrium distance of the fluid remains unchanged, while the nonequilibrium state of the fluid approaches equilibrium. In contrast to gamma>gamma(c), the intermolecular distance is lengthened substantially by strong shear deformation breaking the equilibrium thermodynamic state so that shear dilatancy takes place. Notably, a characteristic shear rate gamma(m), which depends on the root mean square molecular velocity and the average free molecular distance, is found in nonequilibrium thermodynamics state curves. Studies of the variations in the intermolecular radial distribution function (RDF) with respect to the shear rate provide a direct measure of the variation in the degree of intermolecular separation. Additionally, the variations of the RDF curve in the microscopic regime are consistent with those of the nonequilibrium thermodynamic state in the macroscopic world. By inspecting the overall shape of the RDF curve, it can be readily corroborated that the fluid of interest exists in the liquid state. More importantly, both primary characteristic values, the equilibrium thermodynamic state variable and a particular shear rate of gamma(p), are determined cautiously, with gamma(p) depending on the gamma(m) value and the square root of pressure. Thereby, the nonequilibrium thermodynamic state curves can be normalized as temperature-, pressure-, and density-invariant master curves, formulated by applying the Cross constitutive equation. Clearly, gamma(c) occurs at which a reduced shear rate gamma/gamma(p) approaches 0.1. Furthermore, the trends in the rates of shear dilatancy in both the constant-pressure and constant-volume NEMD systems under isothermal conditions conform to the cyclic rule of pressure, as a function of density and shear rate.

Material functions of liquid n-hexadecane under steady shear via nonequilibrium molecular dynamics simulations: temperature, pressure, and density effects

Computer experiments of rheology regarding the effects of temperature (T), pressure (P), and density (rho) on steady shear flow material functions, which include viscosity (eta) and first and second normal stress coefficients (psi(1) and psi(2)) depending on shear rate (gamma), have been conducted via nonequilibrium molecular dynamics simulations for liquid n-hexadecane. Straightforwardly, using both characteristic values of a zero-shear-rate viscosity and critical shear rate, eta-gamma flow curves are well normalized to achieve the temperature-, pressure-, and density-invariant master curves, which can be formulary described by the Carreau-Yasuda rheological constitutive equation. Variations in the rate of shear thinning, obviously exhibiting in eta-gamma, psi(1)-gamma, and -psi(2)-gamma relationships, under different T, P, and rho values, are concretely revealed through the power-law model's exponent. More importantly, at low shear rates, the fluid explicitly possesses Newtonian fluidic characteristics according to both manifestations; first and second normal stress differences decay to near zero, while nonequilibrium states are close to equilibrium ones. Significantly, the tendency to vary of the degree of shear thinning in rheology is qualitatively contrary to that of shear dilatancy in thermodynamics. In addition, a convergent transition point is evidently observed in the -psi(2)/psi(1)-gamma curves undergoing dramatic variations, which should be associated with shear dilatancy, as addressed analytically.

Experimental three-dimensional description of the liquid hexadecane/graphite interface

By using an atomic force microscope based on a quartz tuning fork sensor, a 3-dimensional description of the interface between liquid hexadecane and a highly oriented pyrolytic graphite surface can be achieved at room temperature. The C16H34 monolayer in contact with the substrate surface exhibits a lamellar structure whereas no observation at the liquid/graphite interface by scanning tunnelling microscopy was reported for this alkane. The second layer shows very weak corrugations corresponding to lamella boundaries. Force/distance curves show at least four oscillations separated by 0.4 nm except for the first period with a 0.38 nm distance that corresponds to the layer closer the substrate. Such a description agrees well with molecular dynamics results obtained on alkane/solid interfaces.

Eutectic Formation Analysis of Amino Acid Mixtures Using Molecular Dynamics Simulations

The mechanism of eutectic formation was investigated via computer-aided molecular dynamics techniques based on experimental results. The CBZ group mixtures CBZ-l-Asp/d-AlaNH2 x HCl/methanol, CBZ-l-Asp/l-PheOMe x HCl/methanol, and CBZ-l-Tyr/l-ArgNH2 x 2HCl/methanol formed transparent eutectic melts. The non-CBZ group mixtures l-Asp/d-AlaNH2 x HCl/methanol, l-Asp/l-PheOMe x HCl/methanol, and l-Tyr/l-ArgNH2 x 2HCl/methanol did not form eutectic melts. According to molecular dynamics simulation results, increase in the kinetic energy values of eutectic forming mixtures was much larger than the increase in potential energy over a temperature shift from 298 to 333 K. However, the results for non-eutectic forming mixtures were reversed. The Coulomb interaction energies of eutectic forming mixtures significantly decreased, because eutectic melting can increase the mobility of molecules in the mixtures. The enhancement of molecular mobility was confirmed by increased self-diffusion constant values, and the change of solid-to-liquid phase was detected by radial distribution function results. The periodic boundary conditions for calculation of molecular dynamics were found to be reliable.

Thermodynamic and kinetic aspects of fat crystallization

Naturally occurring fats are multi-component mixtures of triacylglycerols (TAGs), which are triesters of fatty acids with glycerol, and of which there are many chemically distinct compounds. Due to the importance of fats to the food and consumer products industries, fat crystallization has been studied for many years and many intricate features of TAG interactions, complicated by polymorphism, have been identified. The melting and crystallization properties of triacylglycerols are very sensitive to even small differences in fatty acid composition and position within the TAG molecule which cause steric hindrance. Differences of fatty acid chain length within a TAG lead to packing imperfections, and differences in chain lengths between different TAG molecules lead to a loss of intersolubility in the solid phase. The degree of saturation is hugely important as the presence of a double bond in a fatty acid chain causes rigid kinks in the fatty acid chains that produce huge disruption to packing structures with the result that TAGs containing double bonds have much lower melting points than completely saturated TAGs. All of these effects are more pronounced in the most stable polymorphic forms, which require the most efficient molecular packing. The crystallization of fats is complicated not just by polymorphism, but also because it usually occurs from a multi-component melt rather than from a solvent which is more common in other industrial crystallizations. This renders the conventional treatment of crystallization as a result of supersaturation somewhat meaningless. Most studies in the literature consequently quantify crystallization driving forces using the concept of supercooling below a distinct melting point. However whilst this is theoretically valid for a single component system, it can only at best represent a rough approximation for natural fat systems, which display a range of melting points. This paper reviews the latest attempts to describe the sometimes complex phase equilibria of fats using fundamental relationships for chemical potential that have so far been applied to individual species in melts of unary, binary and ternary systems. These can then be used to provide a framework for quantifying the true crystallization driving forces of individual components within a multi-component melt. These are directly related to nucleation and growth rates, and are also important in the prediction of polymorphic occurrence, crystal morphology and surface roughness. The methods currently used to evaluate induction time, nucleation rate and overall crystallization rate data are also briefly described. However, mechanistic explanations for much of the observed crystallization behaviour of TAG mixtures remain unresolved.

Molecular dynamics in n-alkanes: Premelting phenomena and rotator phases

Molecular dynamics simulations of the n-alkanes C18H38, C19H40, and C20H42 are reported for temperatures just below the melting point. Besides thermodynamic and average structural data for the ordered phase, we discuss the molecular motions initiating the rotator phases observed in spontaneous phase transitions in isothermal, isostress simulations. The RI phase of C19H40 is initiated by particular cork-screw-like jumps combining a quarter turn about the long molecular axis and a half-chain-period translation along the axis. This motion occurs between the minimum-energy conformation of the ordered crystal and a secondary minimum. Transient analogs of the RI and RII phases of the odd alkanes are found on melting C18H38 and C20H42. Collective motions within lamellae of molecules are prominent in the dynamics.

Toward a robust and general molecular simulation method for computing solid-liquid coexistence

A rigorous and generally applicable method for computing solid-liquid coexistence is presented. The method overcomes some of the technical difficulties associated with other solid-liquid simulation procedures and can be implemented within either a molecular dynamics or Monte Carlo framework. The method consists of three steps: First, relative Gibbs free energy curves are created for the solid and liquid phases using histogram reweighting. Next, the free energy difference between the solid and liquid phases is evaluated at a single state point by integrating along a pseudosupercritical transformation path that connects the two phases. Using this result, the solid and liquid free energy curves are referenced to a common point, allowing a single coexistence point to be determined. Finally, Gibbs-Duhem integration is used to determine the full coexistence curve. To evaluate its utility, this method is applied to the Lennard-Jones and NaCl systems. Results for solid-liquid coexistence agree with previous calculations for these systems. In addition, it is shown that the NaCl model does not correctly describe solid-liquid coexistence at high pressures. An analysis of the accuracy of the method indicates that the results are most sensitive to the transformation free energy calculation.