scholarly journals Enhanced Molecular Dynamics Approach for Thermal Transport Phenomena in Complex Systems

2021 ◽  
Author(s):  
Sylvie Antoun
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Complex Systems ◽  
Force Field ◽  
Potential Field ◽  
Thermal Transport ◽  
Ternary Mixture ◽  
Soret Effect ◽  
Transport Phenomena ◽  
Integration Scheme

This thesis introduces an enhanced Molecular Dynamics (MD) approach, blended with fine-tuned Force Field (FF) models to reflect more realistic experimental conditions and achieve a precise representation of the atomic interactions in complex systems. Firstly, an enhanced MD algorithm consisting of an upgraded non-equilibrium integration scheme, namely eHEX, coupled with an augmented TraPPE-UA force field, was generated and put to use to predict Soret effect in a binary mixture: n-pentane/n-decane. The results were compared to other MD approaches and validated with respect to benchmarked experimental data. The suggested method showed a closer agreement with experimental data than the previous MD findings. The reinforced potential field (TraPPE-UA) was capable of reflecting the real molecular interactions between the hydrocarbons and reproduce the liquid mixture properties at different conditions. Moreover, the extended HEX method succeeded in conserving the system’s overall energy with minor fluctuations and attaining a stationary state, ensuring the precision of the integration scheme and the satisfaction of local equilibrium. Secondly, the performance of the previously proposed approach was further studied to test its performance on a ternary mixture of methane/n-butane/n-dodecane at five different compositions. Thermodiffusion separation ratio of each component was assessed at 333.15 K and 35 MPa, and compared to the experimental data as well as 3 other MD models from the literature. A good qualitative agreement between the experimental data and the MD model observed in this work was observed, displaying the least deviation when compared to the other MD approaches. The method was capable of adequately representing the physics behind the thermodiffusive separation and deepening the microscopic understanding of the segregation process in a ternary mixture undergoing large thermal gradients. Put differently, the approach elucidates the relative contribution of the cross-interactions found between the unlike species in the mixture and their corresponding composition. Next, an enhanced MD approach was also presented to predict the dynamics and thermophysical properties of suspended γ-alumina nanoparticles (NPs) in acidic aqueous solutions. The previous MD work have unveiled numerous impediments in terms of reproducing the thermal transport phenomena in nanofluids. A hybrid potential field, comprised of refined orce field models (ClayFF and SPC/E), was implemented to allow a precise integration of the nanoscale phenomena into the dynamics and structure of charged alumina NPs, thereby bridging the challenging gap between the solid-liquid interfacial chemistry and the overall thermodynamic properties. The original CLAYFF was augmented to properly account for the energy and momentum transfer between the water molecules and the positively charged NPs, while keeping the number of parameters small enough to allow modeling of a relatively large nanofluidic system.The results were in good agreement with the experimental data. An increase of the NPs volumetric concentration (φ) lead to the enhancement of thermal conductivity along with an increase of viscosity. The results demonstrate the crucial role played by the repulsive electrostatic forces yielding well-dispersed NP suspensions, specially at low φ. The post analysis of Mouromtseff number demonstrated that at lower φ, the system show a higher propensity for stability and enhancement for φ less than 2%, specially at high temperatures. On the contrary, for volumetric concentrations higher than 2%, the system thermal performance deteriorates which is expected due to the fact that the system exhibit a critical condition of aggregation and clogging. With all of the above findings in mind, the MD framework presented in this thesis represents an improved step towards a precise and computationally balanced MD modelling that bridges the relation between molecular signatures and macroscopic features, capable of overcoming the shortcomings present in mainly two emerging thermal applications: 1) Soret effect in hydrocarbon mixture and 2) thermal transport of alumina-water nanofluids.

scholarly journals Enhanced Molecular Dynamics Approach for Thermal Transport Phenomena in Complex Systems

2021 ◽  
Author(s):  
Sylvie Antoun
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Complex Systems ◽  
Force Field ◽  
Potential Field ◽  
Thermal Transport ◽  
Ternary Mixture ◽  
Soret Effect ◽  
Transport Phenomena ◽  
Integration Scheme

This thesis introduces an enhanced Molecular Dynamics (MD) approach, blended with fine-tuned Force Field (FF) models to reflect more realistic experimental conditions and achieve a precise representation of the atomic interactions in complex systems. Firstly, an enhanced MD algorithm consisting of an upgraded non-equilibrium integration scheme, namely eHEX, coupled with an augmented TraPPE-UA force field, was generated and put to use to predict Soret effect in a binary mixture: n-pentane/n-decane. The results were compared to other MD approaches and validated with respect to benchmarked experimental data. The suggested method showed a closer agreement with experimental data than the previous MD findings. The reinforced potential field (TraPPE-UA) was capable of reflecting the real molecular interactions between the hydrocarbons and reproduce the liquid mixture properties at different conditions. Moreover, the extended HEX method succeeded in conserving the system’s overall energy with minor fluctuations and attaining a stationary state, ensuring the precision of the integration scheme and the satisfaction of local equilibrium. Secondly, the performance of the previously proposed approach was further studied to test its performance on a ternary mixture of methane/n-butane/n-dodecane at five different compositions. Thermodiffusion separation ratio of each component was assessed at 333.15 K and 35 MPa, and compared to the experimental data as well as 3 other MD models from the literature. A good qualitative agreement between the experimental data and the MD model observed in this work was observed, displaying the least deviation when compared to the other MD approaches. The method was capable of adequately representing the physics behind the thermodiffusive separation and deepening the microscopic understanding of the segregation process in a ternary mixture undergoing large thermal gradients. Put differently, the approach elucidates the relative contribution of the cross-interactions found between the unlike species in the mixture and their corresponding composition. Next, an enhanced MD approach was also presented to predict the dynamics and thermophysical properties of suspended γ-alumina nanoparticles (NPs) in acidic aqueous solutions. The previous MD work have unveiled numerous impediments in terms of reproducing the thermal transport phenomena in nanofluids. A hybrid potential field, comprised of refined orce field models (ClayFF and SPC/E), was implemented to allow a precise integration of the nanoscale phenomena into the dynamics and structure of charged alumina NPs, thereby bridging the challenging gap between the solid-liquid interfacial chemistry and the overall thermodynamic properties. The original CLAYFF was augmented to properly account for the energy and momentum transfer between the water molecules and the positively charged NPs, while keeping the number of parameters small enough to allow modeling of a relatively large nanofluidic system.The results were in good agreement with the experimental data. An increase of the NPs volumetric concentration (φ) lead to the enhancement of thermal conductivity along with an increase of viscosity. The results demonstrate the crucial role played by the repulsive electrostatic forces yielding well-dispersed NP suspensions, specially at low φ. The post analysis of Mouromtseff number demonstrated that at lower φ, the system show a higher propensity for stability and enhancement for φ less than 2%, specially at high temperatures. On the contrary, for volumetric concentrations higher than 2%, the system thermal performance deteriorates which is expected due to the fact that the system exhibit a critical condition of aggregation and clogging. With all of the above findings in mind, the MD framework presented in this thesis represents an improved step towards a precise and computationally balanced MD modelling that bridges the relation between molecular signatures and macroscopic features, capable of overcoming the shortcomings present in mainly two emerging thermal applications: 1) Soret effect in hydrocarbon mixture and 2) thermal transport of alumina-water nanofluids.

scholarly journals A hierarchical Bayesian framework for force field selection in molecular dynamics simulations

2016 ◽  
Vol 374 (2060) ◽  
pp. 20150032 ◽  
Author(s):  
S. Wu ◽  
P. Angelikopoulos ◽  
C. Papadimitriou ◽  
R. Moser ◽  
P. Koumoutsakos
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Force Field ◽  
Computational Cost ◽  
Md Simulations ◽  
Bayesian Framework ◽  
Underlying Structure ◽  
Hierarchical Bayesian ◽  
Wide Range ◽  
Selection Of

We present a hierarchical Bayesian framework for the selection of force fields in molecular dynamics (MD) simulations. The framework associates the variability of the optimal parameters of the MD potentials under different environmental conditions with the corresponding variability in experimental data. The high computational cost associated with the hierarchical Bayesian framework is reduced by orders of magnitude through a parallelized Transitional Markov Chain Monte Carlo method combined with the Laplace Asymptotic Approximation. The suitability of the hierarchical approach is demonstrated by performing MD simulations with prescribed parameters to obtain data for transport coefficients under different conditions, which are then used to infer and evaluate the parameters of the MD model. We demonstrate the selection of MD models based on experimental data and verify that the hierarchical model can accurately quantify the uncertainty across experiments; improve the posterior probability density function estimation of the parameters, thus, improve predictions on future experiments; identify the most plausible force field to describe the underlying structure of a given dataset. The framework and associated software are applicable to a wide range of nanoscale simulations associated with experimental data with a hierarchical structure.

scholarly journals Molecular Dynamics Calculation Of Thermodiffusion Coefficients In Binary And Ternary Mixtures

2021 ◽  
Author(s):  
Seyedeh Hoda Mozaffari
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Binary Mixtures ◽  
Soret Effect ◽  
Ternary Mixtures ◽  
Computationally Efficient ◽  
Fluid Mixtures ◽  
Binary And Ternary Mixtures ◽  
Different Types ◽  
Non Equilibrium Molecular Dynamics

Thermodiffusion phenomenon in fluid mixtures has been investigated by several scientists in theoretical as well as experimental fields for decades. Nevertheless, due to shortcomings of both methods, interest in searching for alternative approaches to shed some light on molecular scale of the phenomenon has spurred. The objective of this thesis is to develop an accurate molecular dynamics (MD) algorithm that can predict thermodiffusive separation in binary and ternary fluid mixtures. More importantly, the proposed algorithm should be computationally efficient in order to be suitable for integration into multi-scale computational models to simulate thermodiffusion in a large system such as an oil reservoir. In developing such an effective and efficient computational tool, this thesis introduces a modified heat exchange algorithms, wherein, a new mechanism is introduced to rescale velocities which curbs the energy loss in the system and at the same time minimizes the computational time. The performance of the new algorithm in studying Soret effect for binary and ternary mixtures has been compared with other non-equilibrium molecular dynamics (NEMD) models including regular heat exchange algorithm (HEX) and reverse non-equilibrium molecular dynamics (RNEMD). Different types of binary mixtures were studied including one equimolar mixture of argon (Ar)-krypton (Kr) above its triple point, non-equimolar normal alkane mixtures of hexane (nC6)-decane (nC10) as well as hexane (nC6)-dodecane (nC12) for six compositions, three non-equimolar mixtures of pentane (nC5) decane (nC10) at atmospheric temperature and pressure. Additionally, the new algorithm was validated for different ternary mixtures including ternary normal alkanes methane (nC1)-butane (nC4)- dodecane (nC12) for three compositions, and one composition of different types of alkane mixture of 1,2,3,4-tetrahydronaphthalene (THN)-dodecane (nC12)-sobutylbenzene (IBB). The new algorithm demonstrates a significant improvement in reducing the energy loss by nearly 32%. Additionally, the new algorithm is about 7-9% more computationally efficient than the regular HEX for medium and large systems. In terms of direction of thermodiffusive segregations in binary mixtures, in agreement with the experimental data, the new algorithm shows that the heavier component moves towards the cold region whereas the lighter component accumulates near the hot zone. Additionally, the strength of segregation process diminishes as the concentration of heavy component in the mixture increases. The new algorithm improved the prediction of thermodiffusion factor in binary mixtures by 24% in binary mixtures. With respect to the ternary mixtures, similarly to binary mixtures the heaviest and lightest component in the mixture move towards, cold and hot zones, respectively. While the intermediate component shows the least tendency to segregate. In terms of the strength of Soret effect, the new algorithm is about 17% more accurate than the regular HEX algorithm with respect to experimental data.

scholarly journals Molecular Dynamics Calculation Of Thermodiffusion Coefficients In Binary And Ternary Mixtures

2021 ◽  
Author(s):  
Seyedeh Hoda Mozaffari
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Binary Mixtures ◽  
Soret Effect ◽  
Ternary Mixtures ◽  
Computationally Efficient ◽  
Fluid Mixtures ◽  
Binary And Ternary Mixtures ◽  
Different Types ◽  
Non Equilibrium Molecular Dynamics

Thermodiffusion phenomenon in fluid mixtures has been investigated by several scientists in theoretical as well as experimental fields for decades. Nevertheless, due to shortcomings of both methods, interest in searching for alternative approaches to shed some light on molecular scale of the phenomenon has spurred. The objective of this thesis is to develop an accurate molecular dynamics (MD) algorithm that can predict thermodiffusive separation in binary and ternary fluid mixtures. More importantly, the proposed algorithm should be computationally efficient in order to be suitable for integration into multi-scale computational models to simulate thermodiffusion in a large system such as an oil reservoir. In developing such an effective and efficient computational tool, this thesis introduces a modified heat exchange algorithms, wherein, a new mechanism is introduced to rescale velocities which curbs the energy loss in the system and at the same time minimizes the computational time. The performance of the new algorithm in studying Soret effect for binary and ternary mixtures has been compared with other non-equilibrium molecular dynamics (NEMD) models including regular heat exchange algorithm (HEX) and reverse non-equilibrium molecular dynamics (RNEMD). Different types of binary mixtures were studied including one equimolar mixture of argon (Ar)-krypton (Kr) above its triple point, non-equimolar normal alkane mixtures of hexane (nC6)-decane (nC10) as well as hexane (nC6)-dodecane (nC12) for six compositions, three non-equimolar mixtures of pentane (nC5) decane (nC10) at atmospheric temperature and pressure. Additionally, the new algorithm was validated for different ternary mixtures including ternary normal alkanes methane (nC1)-butane (nC4)- dodecane (nC12) for three compositions, and one composition of different types of alkane mixture of 1,2,3,4-tetrahydronaphthalene (THN)-dodecane (nC12)-sobutylbenzene (IBB). The new algorithm demonstrates a significant improvement in reducing the energy loss by nearly 32%. Additionally, the new algorithm is about 7-9% more computationally efficient than the regular HEX for medium and large systems. In terms of direction of thermodiffusive segregations in binary mixtures, in agreement with the experimental data, the new algorithm shows that the heavier component moves towards the cold region whereas the lighter component accumulates near the hot zone. Additionally, the strength of segregation process diminishes as the concentration of heavy component in the mixture increases. The new algorithm improved the prediction of thermodiffusion factor in binary mixtures by 24% in binary mixtures. With respect to the ternary mixtures, similarly to binary mixtures the heaviest and lightest component in the mixture move towards, cold and hot zones, respectively. While the intermediate component shows the least tendency to segregate. In terms of the strength of Soret effect, the new algorithm is about 17% more accurate than the regular HEX algorithm with respect to experimental data.

Predicting Thermal Transport in Bi2Te3: From Bulk to Nanostructures

2011 ◽  
Vol 1329 ◽  
Author(s):  
Bo Qiu ◽  
Xiulin Ruan
Keyword(s):  
Experimental Data ◽  
Thin Films ◽  
Thermal Conductivity ◽  
Molecular Dynamics ◽  
Thermal Transport ◽  
Morse Potential ◽  
Density Functional ◽  
Energy Surface ◽  
Nanowire Diameter ◽  
Potential Form

ABSTRACTTwo-body interatomic potentials in the Morse potential form have been developed for bismuth telluride, and the potentials are used in molecular dynamics (MD) simulations to predict the thermal conductivity of Bi2Te3 bulk, nanowires and few-quintuple thin films. The density functional theory with local density approximations is first used to calculate the total energies for many artificially distorted Bi2Te3 configurations to produce the energy surface. Then by fitting to this energy surface and other experimental data, the Morse potential form is parameterized. Molecular dynamics simulations are then performed to predict the thermal conductivity of bulk Bi2Te3 at different temperatures, and the results agree with experimental data well. We also predicted the thermal conductivity of Bi2Te3 nanowires with diameter ranging from 3 to 30 nm with both smooth (SMNW) and rough (STNW) surfaces. It is found that when the nanowire diameter decreases to the molecular scale (below 10 nm, or the so called "quantum wire"), the thermal conductivity shows significant reduction as compared to bulk value. We find the dimensional crossover behavior of thermal transport in few quintuple layer (QL) thin films at room temperature, and we attribute it to the interplay between phonon Umklapp scattering and boundary scattering. Also, nanoporous films show significantly reduced thermal conductivity compared to perfect thin films, indicating that they can be very promising thermoelectric materials.

Molecular Dynamics Modeling of Heat Transport in Metals and Semiconductors

2010 ◽  
Author(s):  
Sreekant Narumanchi ◽  
Kwiseon Kim
Keyword(s):  
Experimental Data ◽  
Thermal Conductivity ◽  
Molecular Dynamics ◽  
Thermal Transport ◽  
Hybrid Electric Vehicle ◽  
Heat Dissipation ◽  
Phonon Dispersion ◽  
Interfacial Resistance ◽  
Dynamics Modeling ◽  
Molecular Dynamics Modeling

Interfacial thermal transport is of great importance in a number of practical applications where interfacial resistance between layers is frequently a major bottleneck to effective heat dissipation. For example, efficient heat transfer at silicon/aluminum and silicon/copper interfaces is very critical in power electronics packages used in hybrid electric vehicle applications. It is therefore important to understand the factors that govern and impact thermal transport at semiconductor/metal interfaces. Hence, in this study, we use classical molecular dynamics modeling to understand and study thermal transport in silicon and aluminum, and some preliminary modeling to study thermal transport at the interface between silicon and aluminum. A good match is shown between our modeling results for thermal conductivity in silicon and aluminum and the experimental data. The modeling results from this study also match well with relevant numerical studies in the literature for thermal conductivity. In addition, preliminary modeling results indicate that the interfacial thermal conductance for a perfect silicon/aluminum interface is of the same order as experimental data in the literature as well as diffuse mismatch model results accounting for realistic phonon dispersion curves.

scholarly journals In silico modelling of drug–polymer interactions for pharmaceutical formulations

2010 ◽  
Vol 7 (suppl_4) ◽  
Author(s):  
Samina Ahmad ◽  
Blair F. Johnston ◽  
Simon P. Mackay ◽  
Andreas G. Schatzlein ◽  
Paul Gellert ◽  
...  
Keyword(s):  
Experimental Data ◽  
Molecular Dynamics ◽  
Force Field ◽  
Drug Level ◽  
Pharmaceutical Formulations ◽  
Coarse Grained ◽  
Glycol Chitosan ◽  
Heterogeneous Distribution ◽  
Martini Force Field ◽  
Initial Drug

Selecting polymers for drug encapsulation in pharmaceutical formulations is usually made after extensive trial and error experiments. To speed up excipient choice procedures, we have explored coarse-grained computer simulations (dissipative particle dynamics (DPD) and coarse-grained molecular dynamics using the MARTINI force field) of polymer–drug interactions to study the encapsulation of prednisolone (log p = 1.6), paracetamol (log p = 0.3) and isoniazid (log p = −1.1) in poly( l -lactic acid) (PLA) controlled release microspheres, as well as the encapsulation of propofol (log p = 4.1) in bioavailability enhancing quaternary ammonium palmitoyl glycol chitosan (GCPQ) micelles. Simulations have been compared with experimental data. DPD simulations, in good correlation with experimental data, correctly revealed that hydrophobic drugs (prednisolone and paracetamol) could be encapsulated within PLA microspheres and predicted the experimentally observed paracetamol encapsulation levels (5–8% of the initial drug level) in 50 mg ml −1 PLA microspheres, but only when initial paracetamol levels exceeded 5 mg ml −1 . However, the mesoscale technique was unable to model the hydrophilic drug (isoniazid) encapsulation (4–9% of the initial drug level) which was observed in experiments. Molecular dynamics simulations using the MARTINI force field indicated that the self-assembly of GCPQ is rapid, with propofol residing at the interface between micellar hydrophobic and hydrophilic groups, and that there is a heterogeneous distribution of propofol within the GCPQ micelle population. GCPQ–propofol experiments also revealed a population of relatively empty and drug-filled GCPQ particles.

Sign in / Sign up

Export Citation Format

Share Document