MOLECULAR DYNAMICS FOR PASSIVE HEAT ENHANCEMENT

BY

TIMOTHY, PROMISE EMMANUEL EES/19/20/0574

“A BACHELOR OF ENGINEERING (B.Eng.) PROJECT REPORT SUBMITTED

TO THE DEPARTMENT OF MECHANICAL ENGINEERING, FACULTY OF
ENGINEERING, COLLEGE OF ENGINEERING AND ENVIRONMENTAL
STUDIES, IBOGUN CAMPUS, IN PARTIAL FULFILLMENT OF THE
REQUIREMENT FOR THE AWARD OF BACHELOR OF ENGINEERING
DEGREE IN MECHANICAL ENGINEERING.”

“OLABISI ONABANJO UNIVERSITY

P.M.B. 2002, AGO-IWOYE, OGUN STATE
NIGERIA”

AUGUST, 2024
 DECLARATION
I, Timothy Promise Emmanuel, solemnly declare that the project herein presented has been

accomplished through the utilization of various simulations facilitated by assigned software tools,

complemented by rigorous study conducted through extensive perusal of scholarly journals,

citations, dissertation papers, and theses authored by diverse researchers. This project has not been

previously submitted, either in part or in full, for the attainment of any other academic qualification

or degree, nor has it been presented to any other institution for similar purposes. The reference

section included herein is intended solely for the acknowledgment of the sources of material

utilized in this project.

_______________________ _____________________
TIMOTHY Promise Emmanuel Date

ii
 CERTIFICATION
This is to certify that the project “Molecular Dynamics for Passive Heat Enhancement” is accepted
and approved for its contributions to engineering knowledge and literacy presentation and
complies with the rules governing the award of the degree of Bachelor of Engineering from Olabisi
Onabanjo University, Ago Iwoye.

________________ ____________________
Dr. S.O. ADEKOMAYA Date
Project Supervisor

________________ ____________________
Dr. S.A. AASA Date
Project Coordinator

_________________ _____________________
Dr. C.N. NWAOKOCHA Date
Head, Department of Mechanical Engineering

_________________ _____________________
Prof. D.A. FADARE Date
External Examiner

iii
 ABSTRACT
This study investigates passive heat enhancement performance using molecular dynamics

simulation strategy Large scale atomic and molecular massively parallel simulator (LAMMPS).

The nano scale propagation of heat enhance of fluid-solid interactions of copper are modeled in

LAMMPS. Heat flux, temperature range of 0 oC - 65 oC, is applied. Heat flows through the endwall

along the mainstream geometry are simulated for 10000 time steps. The simulation results reveal,

the model response for smooth and roughen surface. The potential energy, deformation rate,

temperature gradient distribution and mass flow average are measured. It is observed that roughen

surface deformation significantly influences how heat flux is distributed across the heat sink even

at nanoscale level. Roughened surface endwall tends to have localised heating, which affect the

overall efficiency of heat dissipation. Besides, roughness copper model endwalls turbulate flows

even at nano scale level, thereby potentially improve the heat transfer rate.

iv
 DEDICATION
This research project is dedicated to Almighty God, whose grace has been ever-present since the

inception of my journey in this esteemed institution. It is through His unwavering support and

guidance that I have found strength and perseverance to navigate the challenges encountered along

the way.

Furthermore, I dedicate this research to myself, TIMOTHY Promise Emmanuel. This dedication

serves as a testament to the countless hours of dedication, determination, and self-discovery poured

into this endeavor. May it stand as a reminder of my resilience, growth, and unwavering

commitment to academic excellence

v
 ACKNOWLEDGEMENT
My utmost gratitude goes to Almighty God for guiding me through this research. Without His

grace, this project would not have become a reality.

My sincere appreciation goes to my parent Mr. and Mrs. Timothy whom I’m greatly indebted to

for bringing me up with love and affection.

I extend my sincere appreciation to my supervisor, Dr. Adekomaya S.O, for his invaluable

assistance and contribution in reviewing this work. I would like to especially thank Dr. Aasa A.S.

for his remarkable and immense contribution in ensuring that this project was successfully

completed, despite his busy schedule.

I would also like to extend my gratitude to the Lecturers in the Department of Mechanical

Engineering for the knowledge imparted on me.

Finally, I would like to express my gratitude to my colleagues; Soretire Olayinka, Olajide Samuel,

Elenbalulu Joshua, Adetola Akorede, Abdullahi Ibraheem, and my friends Akiladejo Blessing,

Dangana Job, Adedeji Taiwo, and Kehinde. I especially appreciate all of you for your efforts and

support during the course of obtaining this degree. I also thank everyone who directly or indirectly

contributed significantly to our academic journey and to the success of this project.

vi
 TABLE OF CONTENTS

DECLARATION ii
CERTIFICATION Error! Bookmark not defined.

ABSTRACT iv

DEDICATION v

ACKNOWLEDGEMENT vi

LIST OF FIGURES vii

CHAPTER ONE 1

1.0 INTRODUCTION 1

1.1 Background of the study 1

1.2 Problem Statement 3

1.3 Aim and Objectives 4

1.3.1 Aim 4

1.3.2 Objectives 4

1.4 Justification of Study 4

1.5 Scope of Study 5

CHAPTER TWO 6

Literature Review 6

2.0 Heat Enhancement 6

2.1 Heat Enhancement Techniques 6

2.1.1 Active Heat Enhancement 7

2.1.2 Passive Heat Enhancement 7

2.1.3 Compound Heat Enhancement 7

2.2 Literature Review 8

Anil Katarkar et al. (2020) 8

Tejas Sonawane et al. (2016) 9

vii
 Fasano et al., Passive heat transfer enhancement by 3D printed Pitot tube based heat sink
(2020) 10

Zhang et al. (2022) 12

2.3 History of Molecular Dynamics 13

2.4 Softwares for Molecular Dynamics 13

CHAPTER THREE 15

3.1 Molecular Dynamic 15

3.2 LAMMPS 17

3.2.1 Conventions of LAMMPS 18

3.2.2 LAMMPS Files 21

3.3 Data Reduction 25

CHAPTER FOUR 28

RESULTS AND DISCUSSIONS 28

4.1 INTRODUCTION 28

4.2 Data representation 28

CHAPTER FIVE 37

CONCLUSION AND RECOMMENDATION 37

5.1 Conclusion 37

5.2 Recommendation 38

REFERENCES 39

viii
 LIST OF FIGURES
Figure 2.1: Scenario showing two major techniques of industrial boiling system. Katarkar et al.,

Review on Passive Heat Enhancement Techniques in Pool Boiling Heat Transfer, pg. 2

(2020) 9

Figure 2.2: Schematic view of twisted tape & wire coil Figure 2 Tejas Sonawane et al. (2016) 10

Figure 2.3: Schematics of the air flow patterns in the Pitot heat sink with (a) open and (b) closed

configurations. (Fasano et al., Passive heat transfer enhancement by 3D printed Pitot tube-

based heat sink, 2020) 11

Figure 2.4: The normalized overlap of phonon bands of different diameter CNTs and metals at

different frequencies. (a) 0–20 THz; (b) 20–40 THz; (c) 40–60 THz. (Figure 2.4 interfacial

thermal resistance between metals and carbon nanotubes (CNTs) using molecular dynamics

simulation, (2022) 12

Figure 3.1: LAMMPS software interface 18

Figure 3.2: typical cubic atom geometry 19

Figure 3.3: showing the exact location of atom P in a 3-dimension plane- axis 19

Figure 3.4: snapshot showing Boundary condition command on LAMMPS 20

Figure 3.5: Algorithm for a typical structure.dat input file 22

Figure 3.6: Algorithm for a typical in.filename input file 22

Figure 3.7: Algorithm for a typical pot.mod input file 23

Figure 3.8: structure.dat file containing the force fields embedded in the atoms 24

Figure 3.9: Log.lammmps output file 25

Figure 4.1: energy visualization showing the various energy levels in a simulation box 29

Figure 4.2: Plot of potential energy against time steps 30

Figure 4.3: Plot of potential energy against time steps for deformed copper atoms 31

ix
Figure 4.4: Plot comparing potential energy of the smooth and deformed (rough) copper atoms

against time steps 31

Figure 4.5: Plot of deformation against time steps 32

Figure 4.6: Plot of potential energy against time steps 33

Figure 4.7: Plot representing the computed temperature across the atoms of the top, bottom

surface and the whole box of atoms against time steps 34

Figure 4.8: Plot of the computed strain in the atom box against timesteps 35

Figure 4.9: Plot representing the Centre of mass against the simulation frames 36

x
 CHAPTER ONE

1.0 INTRODUCTION

1.1 Background of the study

Over the years, there have been various evolving technologies in the heat transfer systems, thanks

to the recent development on various heat transfer mechanisms, there has been significant

achievements when it comes to the aspect of passive heat enhancement. Various methods have

been employed by various researchers from diverse background to achieve advanced results in the

aspect of heat transfer methods considering cost, space requirements, and available materials. One

such method in which this remarkable feat has been able to attain a notable achievement in passive

heat enhancement is molecular dynamics. Molecular dynamics (MD) simulations predict how

every atom in a protein or other molecular system will move over time based on a general model

of the physics governing interatomic interactions karplus and McCammon, (2002) Molecular

dynamics is a computer simulation method for studying the transient physical movements of atoms

and molecules, and is thus engaging in type of N-body (N-Body which is a term used to describe

manipulation of facial atoms of heat transfer materials) simulation by making use of computer

softwares packages which are designed for this purpose. Molecular dynamics software like large-

scale atomic/molecular massively parallel simulator, LAMMPS software, are used in carrying out

study on molecular dynamics for passive heat enhancement calls for the understanding on how

atoms and molecules of various heat transfer mechanism operates . Molecular dynamics

simulations are often used in combination with a wide variety of experimental structural biology

techniques, including X-ray crystallography, cryoelectron microscopy (cryo-EM), nuclear

magnetic resonance (NMR), electron paramagnetic resonance (EPR), and Forster resonance

energy transfer (FRET). Importantly, such simulations can also predict how molecules will

respond—at an atomic level—to perturbations as stated by Scott A. and Ron.O. Dror “molecular

1
dynamics simulations for all” (2016) According to Sumith Yesudasan Daisy, “Molecular

Dynamics Study of Solid-Liquid Heat Transfer and Passive Liquid Flow”, (2016), atoms and

molecules on the surface of materials used for heat transfer are allowed to interact each other using

a potential for a desired period of time, giving a view of the dynamical evolution of the system. In

the most common version, the trajectories of atoms and molecules are determined by numerically

solving Newton's equations of motion for a system of interacting particles, where forces between

the particles and their potential energies are calculated using interatomic potentials or molecular

mechanics force fieldss according to Sumith Yesudasan Daisy, “Molecular Dynamics Study of

Solid-Liquid Heat Transfer and Passive Liquid Flow” (2016). The rate of heat transfer can be

increased passively by increasing the surface area, roughness, and by changing the boundary

conditions. The active method involves addition of nano sized, high thermal conductivity, and

metallic powder to the base fluid, to increase the heat transfer rate. Passive techniques, where

inserts are used in the flow passage to enhance the heat transfer rate, are best suited compared to

active techniques as highlighted by Tejas Sonawane et al., (2016). Passive method generally uses

surface or geometrical modifications to the flow channel by incorporating inserts or additional

devices. For example, inserts extra component, swirl flow devices, treated surface, rough surfaces,

extended surfaces, displaced enhancement devices, coiled tubes, surface tension devices and

additives for fluids heat transfer enhancement techniques have been extensively developed to

improve the thermal performance of heat exchanger systems with a view to reducing the size and

cost of the systems according to S. liu and M. Sakr, “A comprehensive review on passive heat

transfer enhancements in pipe exchanger” (2012). In a research paper by Anil Katarkar et al (2020),

Various methods like incorporating swirl flow, micro and nano surface coatings, modifying the

characteristics of the surfaces like modifying the coating, roughness, wettability and contact angle

(CA) or by enhancing the area of the surface for heat exchanging like enlarged surfaces, coiled

2
tubes, tube inserts, metallic foams, micro fins and porous coatings are been highlighted as methods

which can simply be used to enhance passive heat transfer Other researchers such as Thejaraju

Rajashekariah, (2019), have also conducted studies on passive heat enhancement specifically on

design and analysis of passive heat enhancement techniques. His findings indicate that

implementing various additional 7 parts like fins, strip inserts, swirl generators (in the aspect of

industrial heat exchangers), and coiled wires leads to the attainments of an economical passive

heat transfer system. Therefore, this study intends to apply the knowledge of molecular dynamics

for surface heat enhancement of a system and how this system could give rise to system optimal

thermal performance

1.2 Problem Statement

With notable advancements in passive heat enhancement techniques through the application of

molecular dynamics, there remains a pressing issue in understanding and optimising the

effectiveness of such methods in practical heat transfer applications. Specifically, there is a need

to address the challenge of enhancing heat transfer rates efficiently and economically, while also

considering factors such as material availability, space requirements, and cost- effectiveness.

Furthermore, a significant gap exists in the basic understanding and techniques of molecular

dynamics among researchers and practitioners in the field of passive heat enhancement

 Complexity of atoms and molecules on the surface of heat transfer components and

materials

 Limited understanding on inter-facial dynamics

 Computational cost of substituting other passive heat transfer mechanisms

Bridging this gap is essential for identifying and implementing improvements in passive heat

enhancement methods. Therefore, the primary problem addressed in this project report is twofold:

3
firstly, the need for further investigation and optimization of molecular dynamics- based passive

heat enhancement techniques to improve their practical applicability and effectiveness in real-

world heat transfer systems; and secondly, the necessity of enhancing the understanding and

utilization of basic molecular dynamics principles and techniques to facilitate advancements in

passive heat enhancement technologies

1.3 Aim and Objectives

1.3.1 Aim

My aim is to embark on an innovation to execute highly effective molecular dynamics simulations

for passive heat enhancement

1.3.2 Objectives

The main objective of this research is:

a) To carry out molecular dynamic simulation on copper atoms

b) To compare and analyze various parameters which describes the behavior of heat transfer

surface

c) To suggest techniques suitable for proper passive heat enhancement when factors like

material availability and economic factors are considered

1.4 Justification of Study

Passive heat enhancement offers a significant improvement in thermal management strategies,

particularly in micro/nano-scale systems, where traditional cooling methods may prove

inadequate. By leveraging the inherent properties of materials and interfaces without active energy

input, passive techniques present an attractive and sustainable approach to address the escalating

thermal challenges in modern electronic devices and packaging. Molecular dynamics (MD)

4
simulations emerge as a powerful tool in this context, enabling detailed investigations into the

intricate mechanisms governing heat transfer at the atomic level

1.5 Scope of Study

This research will be showing the simulation analysis and designs of molecular dynamics for

passive heat enhancement using program and design softwares, LAMMPS which is an open-

source software equipped with necessary tools and platforms for this purpose

5
 CHAPTER TWO

Literature Review

2.0 Heat Enhancement

Heat enhancement involves employing various techniques in the design of heat exchangers to

improve the transfer of heat while reducing the input energy and work required. As noted by

Ventsislav Zimparov in "Energy conservation through heat transfer enhancement techniques"

(2001), enhanced heat transfer surfaces have been successfully utilized in the heat transfer industry

to create more compact and efficient units. These techniques aim to conserve the useful part of

energy by achieving higher heat transfer rates compared to standard surfaces and flow

configurations.

2.1 Heat Enhancement Techniques

Heat enhancement techniques refer to various methods used to improve the transfer of heat in heat

exchangers and other heat transfer systems. Heat enhancement techniques comprises a range of

methods employed to enhance heat transfer within heat exchangers and similar heat transfer

systems. The primary goal of these techniques is to increase heat transfer rates while reducing

energy usage and operational expenses alongside. Typically, heat enhancement techniques are

classified into two main categories: active and passive. These techniques aim to enhance heat

transfer rates while minimizing energy consumption and operational costs. However, it's important

to note the existence of a third method, which combines aspects of both active and passive

techniques. This hybrid approach is known as the compound method. By integrating elements from

both active and passive methods, the compound method offers a comprehensive strategy for

optimizing heat transfer efficiency in various applications.

6
2.1.1 Active Heat Enhancement

This method involves the utilization of external power input to improve heat transfer efficiency.

Active methods encompass various techniques, including induced pulsation through cams and

reciprocating plungers, employing magnetic fields to disrupt seeded light particles within a flowing

medium, utilizing mechanical aids, surface and fluid vibration, electrostatic fields, suction or

injection, and jet impingement. These methods necessitate an external activator or power supply

to facilitate the enhancement process. Elshafei et al., "Experimental study of heat transfer in

pulsating turbulent flow in a pipe.” (2008)

2.1.2 Passive Heat Enhancement

Passive heat transfer enhancement methods don't require any outside power source. In convective

heat transfer, one way to boost heat transfer is by increasing the surface area and how long the heat

transfer fluids stay in contact with surfaces. This creates swirling motion in the fluid, disrupting

its layers near the surface. As a result, the effective surface area and the time the fluid spends near

the surface increase. This also improves the heat transfer rate. Common methods for achieving this

include using inserts, extending the surface area, modifying the surface texture, and adding certain

substances. Tejas Sonawane et al., "A Review on Heat Transfer Enhancement by Passive

Methods," (2016)

2.1.3 Compound Heat Enhancement

When multiple techniques, such as passive and active methods, are used together to enhance the

heat transfer of a device, resulting in greater enhancement compared to using each technique

separately, it's termed as compound enhancement technique, as noted by Tejas Sonawane et al. in

"A Review On Heat Transfer Enhancement By Passive Methods" (2016). It's crucial to recognize

that employing the compound method for heat enhancement may lead to increased energy

consumption by the system, which is an important factor to consider in the heat enhancement

7
process. However, despite the higher energy consumption, the compound method ensures effective

results in enhancing heat transfer efficiency

2.2 Review of work done by past researchers

This part discusses literature based on relevant studies pertaining to passive heat enhancement and

molecular dynamics simulation to understand the concept

Anil Katarkar et al. (2020)

In this publication, the researchers studied various thermo features on heat transfer mechanisms

which affect the rate at which heat is been transferred in various heat exchangers, specifically in

boiling applications. This features include thermo-physical features of material & the working

fluid, modification of surface techniques (pin-fins treated, roughness, extended, artificial cavities

pattern, micro-channels), thin-film (nano-scale, micro-scale, macro-scale surfaces, and hybrid

scale), coating method, working fluid (water or refrigerants) and additives for fluids. The current

status of these passive heat enhancement methods for boiling activities for example in pool boiling

heat transfer with various kinds, improvement potentials, future challenges, surface modification

techniques, and their potential industrial and technical applications was also examined in this work.

It was highlighted in this work that majority of this modifications and excessive utilization of

conventional energy resources has given rise to several issues like climatic variations, global

warming, and depletion of energy resources. Hence, the enhancement of energy efficiency and

protection of the natural environment has become an essential factor in the world.

8
 Fig 2.1: Scenario showing two major techniques of industrial boiling system. Katarkar et al.,
Review on Passive Heat Enhancement Techniques in Pool Boiling Heat Transfer, pg. 2 (2020)

They pointed out the fact that by enhancing the performance efficiency of applications of various

heat exchangers, evaporators in the refrigerators, boilers in power generation plants, the impacts

of global warming can be mitigated. This research demonstrated that development of novel and

unusual performances of those materials used for heat transfer purpose in the areas of applications

like optoelectronics, electro-optics, dielectrics, and micro and nano surface coatings methods can

be employed as inventive methods to enhance the performance of heat transfer to hold periodic

dry out noting out that thin films perform a major function in the advancement and investigation

of the materials with unique and novel characteristics. This was further demonstrated by using a

thin layer of nano- micron material on the surface is termed as a thin film, where the material

thickness is in the range of a few nanometers to the number of micrometers. It was proposed that

these thin films have a prominent influence on recent technology and can go on to be manufactured

by adopting physical and chemical depositions

Tejas Sonawane et al. (2016)

Tejas Sonawane et al., (2016) carried out an in-depth study on heat enhancement methods to

improve heat transfer in the industrial heat exchangers and also explored the techniques used in

9
areas like thermal power plant, air conditioning, automobile and aerospace. In this research, they

concluded that by using treated surfaces we enhance heat transfer rate which are applicable for

boiling and condensing. This was demonstrated by the use of twisted tapes in heat exchangers. The

twisted tapes are made of mild steel and have tape width (w) of 10 mm, 15 mm & 20mm. Tape

thickness (d) of 0.8 mm, and tape length (l) of 900 mm. Also a wire coil having pitch of 30 mm is

used to generate co-swirl. All tapes were prepared with different twist ratios, y/w = 3.5, 2.66 and

2.25 respectively where twist ratio is defined as twist length (l) to tape width (w). On the other

hand, it was also stated that to avoid an additional friction in the system that might be caused the

thicker tape

Figure 2.2: Schematic view of twisted tape & wire coil Figure 2 Tejas Sonawane et al. (2016)

Fasano et al., Passive heat transfer enhancement by 3D printed Pitot tube based heat sink

(2020)

The study conducted by fasano et al. (2016) Passive heat transfer enhancement by 3D printed Pitot

tube based heat sink" sheds light on the innovative use of 3D printing technology, specifically

10
additive manufacturing (AM), to fabricate heat transfer devices with enhanced performance. The

authors highlight the potential of AM techniques in revolutionizing the design and manufacturing

of heat transfer devices, particularly in terms of shape complexity and performance optimization.

The research laid great emphasis on the importance of effective heat transfer enhancement in

various technological applications, ranging from turbine blades cooling to thermal management of

electronic devices. It emphasizes the promising nature of additive manufacturing processes, as a

means of manufacturing heat transfer devices with improved performance characteristics. In their

study, Fasano et al. (2016) present an innovative heat sink design incorporating Pitot tubes

fabricated using AM techniques. The design aims to exploit the Pitot tube effect to induce

secondary flows within the heat sink, thereby enhancing heat transfer efficiency. The authors

conducted preliminary tests on the proposed heat sink, demonstrating a significant enhancement

in heat transfer performance compared to conventional heat sinks

Figure 2.3: Schematics of the air flow patterns in the Pitot heat sink with (a) open and (b) closed
configurations. (Fasano et al., Passive heat transfer enhancement by 3D printed Pitot tube-
based heat sink, 2020)

11
The experimental results indicate that the proposed Pitot tube-based heat sink achieved up to 98%

heat transfer augmentation compared to conventional heat sinks. This remarkable improvement

underscores the potential of AM-based techniques in enhancing heat transfer efficiency

Zhang et al. (2022)

The study by Zhang et al. (2022) investigates the interfacial thermal resistance between metals and

carbon nanotubes (CNTs) using molecular dynamics simulation. They analyze the effects of CNT

diameter, metal type, and mean interfacial temperature on thermal resistance. Employing

temperature difference and heat flow methods, they find that thermal resistance decreases

exponentially with increasing CNT diameter and identify thermal rectification due to different

interfacial temperatures. Additionally, they quantitatively analyze the thermal transfer mechanism

at the metal-CNT interface, providing insights into heat transfer enhancement at the nano-scale

interface.

Fig 2.4: The normalized overlap of phonon bands of different diameter CNTs and metals at
different frequencies. (a) 0–20 THz; (b) 20–40 THz; (c) 40–60 THz. (Figure 2.4 interfacial
thermal resistance between metals and carbon nanotubes (CNTs) using molecular dynamics
simulation, (2022)

This research contributes to a deeper theoretical understanding of micro-nano-scale interface heat

transfer, essential for optimizing the design of electronic devices and improving heat transfer

capabilities

12
2.3 History of Molecular Dynamics

Molecular Dynamics (MD) has its roots in the early 1950s, building on Monte Carlo simulations

and statistical mechanics. The methodology gained prominence at Los Alamos National

Laboratory, where Rosenbluth and Metropolis introduced the Metropolis–Hastings algorithm. The

interest in studying the time evolution of N-body systems dates back to the seventeenth century

with Newton's work and continued into the following century, focusing on celestial mechanics and

issues like solar system stability Numerical methods like the Verlet integration algorithm,

developed by Jean Baptiste Joseph Delambre, date back to 1791. Researchers began integrating

equations of motion for manybody systems using analog computers as early as 1941. Some

researchers used labor intensive methods, such as constructing physical models with macroscopic

spheres to model atomic motion. MD's application expanded with the discovery of microscopic

particles and the rise of computing power. Fermi's proposal in 1953, realized on the MANIAC I

computer at Los Alamos National Laboratory, marked a significant step in understanding

irreversibility 14 through the study of many-body systems subjected to various force laws. Alder

and Wainwright used IBM 704 computers to simulate elastic collisions between hard spheres in

1957, while Gibson et al. conducted one of the earliest realistic simulations of matter in 1960.

Rahman's 1964 simulations of liquid argon using a Lennard-Jones potential demonstrated

remarkable agreement with experimental data, cementing the Lennard-Jones potential as a

cornerstone in MD simulations

2.4 Softwares for Molecular Dynamics

Molecular dynamics simulations can be conducted utilizing various software packages tailored to

the specific characteristics of materials and desired results. Some of these software options

encompass; Large-Scale Atomic/Molecular Massively Parallel Simulator (LAMMPS),

GROningen Machine for Chemical Simulations (GROMACS), Assisted Model Building with

13
Energy Refinement (AMBER), and Chemistry at HARvard Molecular Mechanics (CHARM)

which was developed by a research group from Harvard University.

However, it is noteworthy to mention that algorithmic solutions can also be crafted using native

programming languages like C++ to fulfill specific requirements, nevertheless, the proposed

methodology for the passive heat enhancement research employing molecular dynamics involves

leveraging the LAMMPS software. LAMMPS is an open-source platform equipped with

specialized tools dedicated solely to conducting molecular dynamics simulations

14
 CHAPTER THREE

3.1 Molecular Dynamic

Molecular dynamics is a computational method used in scientific experiment to simulate the

behavior and motion of atoms and molecules over time. It involves solving Newton's equations of

motion to track the positions and velocities of individual particles within a system. By integrating

these equations numerically, molecular dynamics simulations can provide insights into the

dynamic properties, structural changes, and interactions of molecules at the atomic level. The

algorithm for molecular dynamics simulations include some simple steps, which are; to calculate

the energy between the atoms, calculate the force between the atoms using newton’s equation of

force,

Generally, newton’s equation for a typical translational motion of a spherically symmetric

molecule is giving as

𝑚𝑑2 𝑟
𝐹= 3.1
𝑑𝑡 2

Where F is the sum of the forces exerted on the molecule by the other molecules in the system, r

is the position vector of the molecule, t is the time, and m is the mass of the molecule.

In molecular dynamics, when calculating the force between two molecules which happens to be

an electromagnetic force, the interaction force between a pair of molecules can be derived from

this potential by the following relation:

𝐹𝑖𝑗 = −∇𝑢𝑖𝑗 3.2

15
Where 𝐹𝑖𝑗 is the force acting on the ith molecule by the jth molecule and 𝑢𝑖𝑗 is the two-body

potential between the ith molecule and the jth molecule.

Summing up all the forces exerted on one molecule by other molecules yields the total force on

this molecule. After the total force on every molecule in the simulation domain has been calculated,

one can integrate equation in time for each molecule using the model.

𝜎 12 𝜎 6
𝑢(𝑟) = 4𝜀 {( 𝑟 ) − (𝑟 ) } 3.3

Where σ [m] and ε [J] are scales for length and energy, respectively

12 6
𝜎 𝜎 𝑞 𝑞 𝑒2
𝐸𝑏 = ∑𝐴𝑖 ∑𝐵𝑗 [4𝜀𝑖𝑗 {( 𝑟 𝑖𝑗 ) − ( 𝑟 𝑖𝑗 ) } + 4𝜋𝜀
𝑖 𝑗
] 3.4
𝑖𝑗 𝑖𝑗 𝑟 0 𝑖𝑗

Where 𝑞𝑖 and 𝑞𝑗 are the partial charges on united pseudo-atoms i and j, 𝑟𝑖𝑗 is the distance between

these atoms.

In order to obtain the Lennard-Jones potential parameters 𝜎𝑖𝑗 and 𝜀𝑖𝑗 for variety of combinations

of molecules, the well-known combination rule as follows is used.

𝜎𝑖𝑖 + 𝜎𝑗𝑗
𝜎𝑖𝑗 = 3.5
2

𝜀𝑖𝑗 = √𝜀𝑖𝑖 𝜀𝑗𝑗 3.6

The total energy derived from the system is hereby given as

1
𝐸𝑡𝑜𝑡𝑎𝑙 = ∑𝑖 𝐹𝑖 (𝜌 ℎ, 𝑖) + 2
∑𝑖 ∑𝑗(≠1) ∅𝑖𝑗 (𝑅𝑖𝑗 ) 3.7

16
Here, (𝜌 ℎ, 𝑖) is the host electron density at atom i due to the remaining atoms of the system, 𝐹𝑖 (ρ)

is the energy (gain) to embed atom i into the background electron density ρ, and ∅𝑖𝑗 (𝑅𝑖𝑗 ) is the

core-core pair repulsion between atoms i and j separated by the distance𝑅𝑖𝑗 .

3.2 LAMMPS

LAMMPS is a classical molecular dynamics (MD) code employed for modeling ensembles of

particles in various states such as liquid, solid, or gaseous. It is versatile, capable of simulating

atomic, polymeric, solid-state (including metals, ceramics, and oxides), granular, coarse-grained,

or macroscopic systems using different interatomic potentials (force fields) and boundary

conditions.

LAMMPS carries out molecular dynamics simulation by executing 6 main steps which include

 Defining the geometry and coordinates of the atoms about to be simulated

 Specifying the boundary conditions

 Defining the atomic interaction potentials

 Scripting the type of simulation to be run

 Running the simulation

 Results visualization, analysis, and comparisons

The range of systems it can model spans from small-scale 2D configurations to large-scale 3D

systems with billions of particles integrates Newton’s equations of motion to simulate interactions

among particles, which can represent atoms, molecules, electrons, coarse-grained clusters, or

larger material clumps. Interaction models within LAMMPS primarily focus on short-ranged

interactions, with some long-range models also included

17
 Fig 3.1: LAMMPS software interface

3.2.1 Conventions of LAMMPS

The conventions of LAMMPS refer to the standard practices and rules followed when using the

LAMMPS software for molecular dynamics simulations. These conventions encompass various

simulation setup procedures

3.2.1.1 Geometry
Geometry plays a crucial role in defining the arrangement of atoms and molecules within a

simulated system. This involves defining the location and specifying the positions of atoms, their

types, and any periodic boundary conditions of the system to be simulated.

For a cubic box, 𝑙𝑥 = 𝑙𝑦 = 𝑙𝑧 = 𝑎 = 𝑒𝑑𝑔𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑏𝑜𝑥 𝑎𝑛𝑑 𝑥𝑦 = 𝑥𝑧 = 𝑦𝑧 = 0

𝑥𝑙𝑜 = 0, 𝑥𝑙𝑖 = 𝑙𝑥 = 𝑎 3.8

𝑦𝑙𝑜 = 0, 𝑦𝑙𝑖 = 𝑙𝑦 = 𝑎 3.9

𝑧𝑙𝑜 = 0, 𝑧𝑙𝑖 = 𝑙𝑧 = 𝑎 3.10

18
 Fig 3.2: typical cubic atom geometry

The position of each atom is defined in the Cartesian coordinate system, denoted by three axes: x,

y, and z. This coordinate system provides a standardized way to represent the spatial arrangement

of atoms in three-dimensional space, allowing users to specify the exact location of atoms and

define the geometry of the simulated system accurately.

Fig 3.3: showing the exact location of atom P in a 3-dimension plane- axis

19
3.2.1.2 Periodic Boundary Conditions
Periodic Boundary Conditions (PBC) in LAMMPS software refer to a simulation technique used

to mimic an infinite system by imposing periodicity along the simulation box edges. This means

that when an atom moves out of the simulation box boundary on one side, it re-enters the box from

the opposite side.

Fig 3.4: snapshot showing Boundary condition command on LAMMPS

3.2.1.3 Thermodynamic ensemble

In a thorough molecular dynamics simulation with LAMMPS software, it is vital to clearly specify

the thermodynamic properties we intend to replicate within the system. There are categories,

known as thermodynamic ensembles, which help us achieve this:

Microcanical ensemble; this method maintains a constant number of particles, volume, and energy

within the system. It's like keeping a closed box with a fixed number of particles and energy levels,

allowing no exchange of heat with the surroundings

Canonical ensemble; in this ensemble, the number of particles, volume, and temperature are kept

constant. It's akin to regulating the system's temperature while allowing for energy exchange with

the environment. In this experiment in which the focus involves studying the behavior of heat

20
transfer and its effect on temperature, maintaining a constant temperature on the surface of heat

exchangers, the suitable thermodynamic ensemble is the canonical ensemble

3.2.2 LAMMPS Files

In LAMMPS software, two distinct file types are employed to direct tasks: input files, which

contain commands and parameters steering the setup and execution of simulations, and output

files, generated by LAMMPS throughout simulations, housing data, results, and specified output

information.

3.2.2.1 Input files

LAMMPS performs calculations by sequentially reading commands from an input script, which is

a text file. It processes one command at a time, and when the input script reaches its end, LAMMPS

concludes its execution. This differs from programs that analyze the entire input before

commencing calculations.

Each command in LAMMPS prompts an immediate action without considering subsequent

commands. These actions include setting internal variables, reading files, or conducting

simulations. These actions can generally be classified into three main categories:

3.2.2.1.1 Structure.dat-data file

This contains the size of the system, the number of atoms, the number of atom types, the mass of

each type, the position of each atom (and potentially their velocity)

21
 Fig 3.5: Algorithm for a typical structure.dat input file

3.2.2.1.2 In.filename-input file

This extension contains the instructions of the task to perform, the parameters and potentially the

force-field, or empirical potential. It can also contain the position of each atom if it is been

calculated analytically like in the case of crystal atoms

Fig 3.6: Algorithm for a typical in.filename input file

22
3.2.2.1.3 pod.mod-potential file
This contains information about the potential to be used

Fig 3.7: Algorithm for a typical pot.mod input file

3.2.2.2 Output files

The output file in LAMMPS is a crucial component of the simulation process. It serves as a

repository for a diverse range of information generated by the software throughout the simulation.

This includes storing not only raw data but also processed results and any specific output

information specified by the user. By providing a comprehensive record of the simulation's

progress and outcomes, the output file enables researchers to analyze and interpret the results

effectively.

Importantly, the output file is generated in four compartments

3.2.2.2.1 Final-Structure.dat - data file

This is identical to the input data file, however, it contains information about the final system

which can also be used as a starting input data file for another simulation

23
 Fig 3.8: structure.dat file containing the force fields embedded in the atoms

3.2.2.2.2 Md.lammpstrj – trajectory file

The trajectory file defines the positions of the atoms every N steps after simulation for visualization

of further analysis

3.2.2.2.3 Log.lammps - log file

This extension stores all the actions performed by LAMMPS which is a useful tool for

troubleshooting should in case a simulation goes wrong

24
 Fig 3.9: Log.lammmps output file

3.3 Data Reduction

Without loss of generality, LAMMPS sets the fundamental quantities mass σ, ϵ, and the Boltzmann

constant𝑘𝑏 = 1.

The masses, distances, energies which is specified are multiples of these fundamental values. The

formulas relating the reduced or unitless quantity to the same quantity with units is also given.

Thus, in this simulation, I will be making use of the mass σ and ϵ values for the selected material

and convert the result from a unit less LJ simulation into physical quantities in which the unit of

𝑚𝜎2
time has to conform to the relation ϵ = since energy is a derived unit
𝜏2

The following formulas are been applied

𝑀
 Mass = mass or m, where M* = 𝑚 3.11

25
 𝑥
 Distance = σ, where ϰ* = 𝜎 3.12

𝜖
 Time = T, where T* = T√𝑚𝜎2 3.13

𝐸
 Energy = ϵ, where E* = 𝜖 3.14

𝑇𝑘𝑏
 Temperature = reduced LJ temperature, where T* = 3.15
𝜖

𝜎 𝜏
 Velocity = , 𝑤ℎ𝑒𝑟𝑒 𝑣 ∗ = 𝑣 𝜎 3.16
𝜏

𝑇𝑘𝐵
 𝑡𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒𝑟 = 𝑟𝑒𝑑𝑢𝑐𝑒𝑑 𝐿𝐽 𝑡𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒 𝑤ℎ𝑒𝑟𝑒 𝑇 ∗ = 3.17
𝜖

𝜎3
 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 = 𝑟𝑒𝑑𝑢𝑐𝑒𝑑 𝐿𝐽 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒, 𝑤ℎ𝑒𝑟𝑒 𝑃 ∗ = 𝑝 𝜖
3.18

1
 𝑐ℎ𝑎𝑟𝑔𝑒 = 𝑟𝑒𝑑𝑢𝑐𝑒𝑑 𝐿𝐽𝑐ℎ𝑎𝑟𝑔𝑒 𝑤ℎ𝑒𝑟𝑒 𝑞 ∗ = 𝑞 3.19
√4𝜋𝜀0 𝜎𝜖

√4𝜋𝜀0 𝜎𝜖𝜎
 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐 𝑓𝑖𝑒𝑙𝑑 = 𝑓𝑜𝑟𝑐𝑒/𝑐ℎ𝑎𝑟𝑔𝑒 𝑤ℎ𝑒𝑟𝑒 𝐸 ∗ = 𝐸 3.20
𝜖

These quantities are related to energy models of an heat exchanger system, it is also important to

consider the conductivity of the selected material in which simulation will be carried on, the

common model used for this purpose is the Green-Kubo method for thermal conductivity. In this

method, thermal conductivity which is denoted as 𝑘𝑢𝑣 at a given correlation at time t is expressed

as
1 𝑡
𝑘𝑢𝑣 (𝑡) = ∫0 𝑑𝑡′𝐶𝑢𝑣 (𝑡′) 3.21
𝐾𝐵 𝑇2𝑉

Where 𝐾𝐵 is Boltzmann’s constant, T is the absolute temperature, and V is the volume of the

simulation cell. The 𝐶𝑢𝑣 (𝑡′) is defined as the ensemble average of the product of two heat currents

separated by t. In the MD simulation, the ensemble average is substituted by a time average.

26
3.3.1 Velocity extraction
The velocity of the system is computed using the embedded function extracting the velocity of a

system. This measures the crystal structure around the atoms at a selected surface. This is achieved

by the embedded formula in lammps

𝑛𝑖 𝑛𝑖𝑗 2
1
𝑄𝑖 = ∑ ‖∑ 𝑅⃗𝑖𝑘 + 𝑅⃗𝑖𝑗 ‖
𝑛𝑖
𝑗=1 𝑘=1

Where the index 𝑗 goes over the 𝑛𝑖 nearest neighbors of atom , and the index 𝑘 goes over

the 𝑛𝑖𝑗 common nearest neighbors between atom i and atom j. 𝑅⃗𝑖𝑘 and 𝑅⃗𝑖𝑗 are the vectors

connecting atom k to atoms i and j

27
 CHAPTER FOUR

RESULTS AND DISCUSSIONS

4.1 INTRODUCTION

The simulation was conducted on a selected variety of copper, an efficient material for designing

heat sinks. The operation involved running simulations on copper using two different scenarios as

criteria. The first simulation was performed using atoms from a smooth surface copper material,

followed by another simulation using atoms from a rough surface copper material. The objective

was to determine the rate of heat dissipation in relation to various thermodynamic properties at

different timesteps. The thermodynamic properties examined included pressure, energy,

equilibration, volume, rate of heat dissipation across the surface, entropy, and the effect of shear

stress on the material's surface. The movement of heat flux across the defined geometry of both

the smooth and rough surfaces was considered in the context of the multiphase flow of air. This

chapter analyzes the performance of a copper heat sink under various thermodynamic conditions

4.2 Data representation

This portion analyze and compare the important parameters of the simulated system after executing

the simulation over 10000 time steps at a very minute range of 0.0001 picoseconds which is very

ideal in order to gain full insights on the activities of the atoms in the defined geometry

4.2.1 Potential energy of the atoms

Potential energy (eV) represents the total energy stored within the atomic structure of copper due

to atomic interactions. In this study, we simulated copper atoms with a smooth surface to evaluate

and enhance their performance in heat sinks for effective heat dissipation. Afterward, we deformed

the smooth surface to create a rough one and measured the potential energy.

28
The extracted data showed that the potential energy of the rough surface varied more, leading to

higher heat transfer efficiency. This indicates that atoms gained energy due to heat flux applied

from the bottom of the atoms within the set boundaries. Using Ovito to visualize the dump files,

we observed how atoms with initially low energy gained energy as heat flux was applied, showing

the direction of heat flow and energy distribution.

The smoother surface exhibited more stable potential energy, implying a more stable atomic

arrangement. In contrast, the rough surface, with its higher and more varied potential energy,

suggested increased atomic interaction and enhanced heat transfer efficiency due to surface

irregularities. This analysis helps understand the trade-offs between stability and heat dissipation

efficiency in heat sink designs.

Figure 4.1: energy visualization showing the various energy levels in a simulation box

29
The graph for the actual simulation with the geometry of atoms with smooth surfaces establish that

the system experienced lower potential energy at the initial stage which increases with respect to

the timestep, this phenomenon explain that with the increase in potential energy, the efficiency of

heat dissipation reduces.

0 2000 4000 6000 8000 10000 12000

-27950

-28000
Potential Energy(eV)

-28050

-28100

-28150

-28200

Time (ps)

Figure 4.2: Plot of potential energy against time steps

On the other hand, after applying deformation to the specified geometry to create a rough surface

for heat dissipation enhancement, it was observed that the potential energy in the enhanced

simulation setup experience declined which leads to high heat dissipation efficiency.

30
 0 10 20 30 40 50
-25720
Potential Energy(eV) -25740
-25760
-25780
-25800
-25820
-25840
-25860
-25880
-25900
-25920
Time (ps)
Figure 4.3: Plot of potential energy against time steps for deformed copper atoms

comparison between the extracted data’s from both the smooth and rough surface setup is

described using the graph below, The rough surface, despite its higher potential energy, tends to

increase surface area which can enhance heat transfer. The rough surface has more variations in

potential energy, which can be attributed to the defects on the surface affecting the energy

distribution.

0 2000 4000 6000 8000 10000 12000

-25500
potential energy for
deformed cu (eV)

-26000
-26500
pot-energy(smooth)
-27000
PotEng (rough)
-27500
-28000
-28500
Time (ps)

Figure 4.4: Plot comparing potential energy of the smooth and deformed (rough) copper atoms
against time steps

31
4.2.2 Deformation Parameter
In the course of running the simulation, the deformation experienced by the atoms box across

various timesteps was extracted. The graph illustrates the relationship between deformation and

timesteps for the simulated copper atoms with deformation measured in Ångströms (Å). At the

beginning of the time period, there is a sharp increase in deformation, peaking around 2.5×10⁻¹⁸

Å, after the initial peak, deformation rapidly declines and stabilizes. The sharp initial peak gives

the insight that the material undergoes significant structural stress relaxation immediately after the

application of the heat flux which represents thermal load and thermal gradients within the defined

copper material, as the simulation approached 1000ps to 2000ps, a rapid decline in deformation

which indicates that the copper atoms had adjust to the thermal and mechanical stresses, gradually

reaching a more stable configuration as it dissipates the initial thermal energy. This stability

implies that the copper atoms is effectively managing the heat dissipation without much

compromise

3.00E-18

2.50E-18
Deformation (Å)

2.00E-18

1.50E-18

1.00E-18

5.00E-19

0.00E+00
0 2000 4000 6000 8000 10000 12000
Time (ps)

Figure 4.5: Plot of deformation against time steps

32
4.2.2 Temperature gradient
Temperature being a very critical and important parameter in the evaluation of heat sinks

performance gave a notable insights in understanding the behavior of copper atoms under different

temperature conditions. The system of copper atoms was simulated such that the temperature of

the box was varied from 00C to 650c (converted to kelvin) which is the working range temperature

of mild electronic heat sinks. The graph indicated that the system experienced a rapid rise in

temperature from less than 150k in which towards this temperature, copper tend to lose its

electrical resistance to 280k within the first 2000 picoseconds. The rapid increase in temperature

demonstrates the coppers high thermal conductivity, it was subsequently followed by a

stabilization state in which the temperature increases but a very lower rate, this describes the

behavior in which heat flux is added to the system and this spanned across the duration of 6000

picoseconds of the total run time where the system tends to experience equilibrium. A near steady

state was reached in which the system appears to stabilize around 300k to 350k shortly before it

approach the 10000th timestep

400
350
300
Temperature

250
200
150
100
50
0
0 2000 4000 6000 8000 10000 12000
Time (s)

Figure 4.6: Plot of potential energy against time steps

33
4.2.4 Temperature distribution profile
The graphs below demonstrates the temperature profiles of copper atoms in the simulation over

the specified time of 10000 picoseconds with separate lines representing the temperature of the

bottom surface, top surface and the whole box of atoms. At the beginning of the simulation, within

the timestep of o to 1000 picoseconds, a steep increase across all profiles was observed which

indicates that the systems quickly absorbs heat thereby increasing the total energy present in the

atoms. However, within the timesteps of 1000ps to 3000ps, the graphs starts to level off and

fluctuate which indicates that the system is reaching thermal equilibrium where the rate of heat

input matches the rate of heat dissipation. As soon as it reaches the timestep interval of 3000ps to

10000ps, the temperature profiles exhibit small fluctuations but generally remained stable with the

top and bottom surfaces have slight differences in temperature, which could be due to heat flux

dynamics

400

350
TEMPERATURE (kelvin)

300

250

200

150

100
c_temp_bottom
50 c_temp_top
c_temp_all
0
0 2000 4000 6000 8000 10000 12000
Time (s)
Figure 4.7: Plot representing the computed temperature across the atoms of the top, bottom
surface and the whole box of atoms against time steps

34
4.2.5 Computed strain in the box of atoms
The extracted strain datas from the output file is graphically represented with a significant initial

spike observed within the timesteps of 0ps to 1000ps with values ranging from positive and

negative which is as a result of tensile and compressive force experienced by the atoms due to

deformation, however, it is observed that within the timestep of 1000ps to 10000ps, the strain

values fluctuate around zero, This indicates that the system reaches a quasi-equilibrium state where

the strain oscillates without a significant trend upwards or downwards. The presence of both

positive and negative strain values shows that the simulated copper atoms experienced both

compressive and tensile stresses

10000

5000

0
Strain

0 2000 4000 6000 8000 10000 12000

-5000

-10000

-15000
Time (ps)

Figure 4.8: Plot of the computed strain in the atom box against timesteps

35
4.2.6 Centre of mass
The graph shows the positions of the first atom in the copper heat sink along the X, Y, and Z

coordinates across 40 simulation frames. Each colored line represents the position in a different

coordinate direction. The positions of the atom fluctuate around zero in all three dimensions,

indicating dynamic movement during the simulation, the fluctuations in position range

approximately between -0.2 and 0.2 units, indicating the extent of atomic movement. The X and

Z positions appear to have more pronounced fluctuations compared to the Y position, which

suggests different degrees of movement or forces acting in these directions. The consistent

movement of atoms indicates that the material is dynamically responding to thermal loads

0.2
First_Atom_Pos_X
0.15
First_Atom_Pos_Y
0.1
ATOMS POSITION

First_Atom_Pos_Z
0.05
0
0 10 20 30 40 50
-0.05
-0.1
-0.15
-0.2
SIMULATION FRAMES

Figure 4.9: Plot representing the Centre of mass against the simulation frames

36
 CHAPTER FIVE

CONCLUSION AND RECOMMENDATION

5.1 Conclusion

This research executed molecular dynamics simulations on copper using LAMMPS to investigate

and analyze various nano parameters to enhance passive heat transfer across the surface. The study

focused on the effects of deformation, surface roughness, and the center of mass dynamics on the

heat dissipation efficiency of copper heat sinks.

 Deformation significantly influences how heat flux is distributed across the heat sink.

Regions experiencing high deformation tend to have localized heating, which can affect

the overall efficiency of heat dissipation

 Smooth surfaces typically experience less deformation under thermal and mechanical

stresses, resulting in more uniform heat dissipation. However, they might not utilize the

maximum potential surface area for heat transfer, potentially limiting their effectiveness in

enhancing heat dissipation.

 Introducing roughness to the surface of a copper heat sink can lead to localized deformation

at the microscopic level. This increased roughness enhances the surface area, potentially

improving the heat transfer rate.

 Smooth surfaces exhibit lower initial strain levels under thermal loads due to uniform heat

distribution. Despite this, they might not enhance heat dissipation as effectively as rough

surfaces because of limited surface area.

 Tracking the center of mass across simulation frames provides insights into the dynamic

stability and structural behavior of heat sinks. Stable center of mass fluctuations indicate

consistent performance, while significant center of mass shifts can reveal areas of high

stress and potential structural weaknesses.

37
 Incorporating deformation, surface roughness, and center of mass dynamics as parameters in

the analysis of copper heat sinks provides a comprehensive understanding of their performance

under thermal loads. This approach helps in optimizing the design and material properties to

enhance heat dissipation while ensuring structural integrity.

5.2 Recommendation

Having gone through the fundamentals of thermal molecular dynamics, the following are my

recommendations.

 The LAMMPS should be used to study thermal flow in Couette fluid.

 The effect of MD in studying pressure, friction factor, temperature, Nusselt number and

performance index for rough geometry extended surfaces.

 Establish molecular influence in surface heat enhancement system as applicable in heat

exchanger devices such as boilers, radiators, heat recovery steam generator (HRSG) which

is used in power plants to recover heat from exhaust gases, furnace heat exchanger and

condensers

38
 REFERENCES

Liu, S., & Sakr, M. (2012). A comprehensive review on passive heat transfer enhancements in pipe

exchangers. Renewable and Sustainable Energy ReviewsKarplus, M., & McCammon, J. A. (2002).

General model of the physics governing interatomic interactions.

Fan, Z., Pereira, L. F. C., Wang, H.-Q., Zheng, J.-C., Donadio, D., & Harju, A. (2015). Force and

heat current formulas for many-body potentials in molecular dynamics simulation with

applications to thermal conductivity calculations. COMP Centre of Excellence, Department of

Applied Physics, Aalto University, Helsinki, Finland.

Costeniuc, M., Ellis, R. S., Touchette, H., & Turkington, B. (2018). The generalized canonical

ensemble and its universal equivalence with the microcanonical ensemble. Department of

Mathematics and Statistics, University of Massachusetts, Amherst, Massachusetts, USA.

Zimparov, V. (2002). Energy conservation through heat transfer enhancement techniques.

International Journal of Energy Research

Moore, S. (2018). Modeling supra-molecular structures with LAMMPS. Sandia National

Laboratories. SAND2018-7268C. National Technology and Engineering Solutions of Sandia,

LLC., a subsidiary of Honeywell International, Inc., for the U.S. Department of Energy's National

Nuclear Security Administration.

Fasano, M., Ventola, L., Calignano, F., Manfredi, D., Ambrosio, E. P., Chiavazzo, E., & Asinari,

P. (2016). Passive heat transfer enhancement by 3D printed Pitot tube based heat sink.

International Communications in Heat and Mass Transfer

39
Alam, T., & Kim, M.-H. (2017). A comprehensive review on single phase heat transfer

enhancement techniques in heat exchanger applications. Renewable and Sustainable Energy

Reviews

Sonawane, T., Patil, P., Chavhan, A., & Dusane, B. M. (2016). A review on heat transfer

enhancement by passive methods. International Research Journal of Engineering and Technology

(IRJET)

Thejaraju, R., & Girish, K. B. (2019). A comprehensive review on design and analysis of passive

enhancement techniques in double pipe heat exchanger. International Journal of Scientific &

Technology Research

Shah, R. K., & Sekulib, D. R. (n.d.). Heat exchangers. In University of Kentucky. Retrieved from

Delphi Harrison Thermal Systems, Lockport, New York.

Maruyama, S. (2002). Molecular dynamics methods in microscale heat transfer. In Heat transfer

& fluid flow in microchannels (pp. 2.13.7-1). Begell House, Inc

Mukherjee, S., Datta, S., & Das, A. K. (2017). Molecular dynamic study of boiling heat transfer

over structured surfaces. Journal of Heat Transfer

Liebenberg, L., & Meyer, J. P. (2007). In-tube passive heat transfer enhancement in the process

industry. International Journal of Refrigeration, 30(6), 1133-1144

40
Poulikakos, D., Arcidiacono, S., & Maruyama, S. (2003). Molecular dynamics simulation in

nanoscale heat transfer: A review. Microscale Thermophysical Engineering

Hospital, A., Goñi, J. R., Orozco, M., & Gelpí, J. L. (2015). Molecular dynamics simulations:

Advances and applications. Advances and Applications in Bioinformatics and Chemistry

Fan, Z., Pereira, L. F. C., Wang, H.-Q., Zheng, J.-C., Donadio, D., & Harju, A. (2015). Force and

heat current formulas for many-body potentials in molecular dynamics simulation with

applications to thermal conductivity calculations. School of Mathematics and Physics, Bohai

University; COMP Centre of Excellence, Department of Applied Physics, Aalto University;

Departamento de Física Teórica e Experimental, Universidade Federal do Rio Grande do Norte;

Key Laboratory of Semiconductors and Applications of Fujian Province, Department of Physics,

Xiamen University; Fujian Provincial Key Laboratory of Mathematical Modeling and High-

Performance Scientific Computation, Department of Physics, Xiamen University; Max Planck

Institut für Polymerforschung; Donostia International Physics Center; IKERBASQUE, Basque

Foundation for Science.

Yesudasan Daisy, S. (2016). Molecular dynamics study of solid-liquid heat transfer and passive

liquid flow (Doctoral dissertation, Syracuse University).

Hollingsworth, S. A., & Dror, R. O. (2018). Molecular dynamics simulation for all. Neuron

Favata, A., Micheletti, A., Ryu, S., & Pugno, N. M. (2020). An analytical benchmark and a

Mathematica program for MD codes: Testing LAMMPS on the 2nd generation Brenner potential.

Journal of Computational Chemistry

41
Sheikholeslami, M., Gorji-Bandpy, M., & Domiri Ganji, D. (2015). Review of heat transfer

enhancement methods: Focus on passive methods using swirl flow devices. Renewable and

Sustainable Energy Reviews

Hetsroni, G., Mosyak, A., Segal, Z., & Ziskind, G. (2002). A uniform temperature heat sink for

cooling of electronic devices. International Journal of Heat and Mass Transfer

42