Materials Today: Proceedings 56 (2022) 3594–3603

Contents lists available at ScienceDirect

Materials Today: Proceedings

journal homepage: www.elsevier.com/locate/matpr

A review of study on modeling and simulation of additive manufacturing

processes
Phool Singh ⇑, Ranganath M.Singari, R.S. Mishra
Department of Mechanical Engineering, Delhi Technological University, Delhi, India

a r t i c l e i n f o a b s t r a c t

Article history: Nowadays, additive manufacturing (AM) has gained increasing importance, owing to developing pro-
Available online 14 December 2021 cesses and tooling that introduced its benefits to different materials’ applications. Many AM uses its
intrinsic abilities to utilise existing stock feed materials to create parts with features spanning multiple
Keywords: length scales. It was not possible before via another manufacturing process. These processes’ excitement
Modeling and simulation has been tempered by recognising difficulties introduced by AM, including the inducement of mechanical
Additive manufacturing and chemical Properties, undesirable microstructure, and the challenge of maintaining dimensional sta-
Direct energy deposition
bility. Computational modelling is becoming a useful tool in addressing many of the difficulties inherent
FDM
SLM
in the process. This review’s main purpose is to provide some information and discuss ongoing working
EBM progress in the modeling and simulation system as applied to AM and additive manufacturing of metal
3 D printing and polymeric nanocomposite material.
Copyright Ó 2022 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the First International Con-
ference on Design and Materials (ICDM)-2021

1. Introduction mization prediction of the design. In relationship with the multi-

scale, multi-physical process, the process structure begins by mod-
Additive manufacture is one of the striking and silent similari- eling a micro scale, setting up a mechanistic thermal source model,
ties between industrial facilities and processes. Most of the part using meso-scale models of the individual powder particle evolu-
processes are generally noisy. Molding, shaping, stamping, pro- tion, and finally by simulating the process of manufacturing a com-
cessing can be pretty loud. AM frequently generates a sound, plicated product. A high efficiency, self-governance clustering
whether it is metal or polymer, deposition or fusion. It is one facet analyses the mechanical model it is built to capture different mate-
of human comfort in this form of production, but something else rial reactions to the connect with structure and features. The model
needs to be seen here: energy savings. The power required to make incorporates voids, phase structure, inclusions and structures of
a loud noise. The noise is a secondary effect. Therefore, energy not grain, which differentiate AM metals’ distinctive characteristics
spent on noise is saved. This vision calls for the involvement of AM, [1] in the simulation and modeling of the three-dimensional tem-
and AM has a role to play. 3D printing is one way to recycle plastic perature field, known as a birth and death element in multiple lay-
and reintroduce it to the economy in a different form. As a result of ers a powder bed. Many researchers have been challenged to
the disruption caused by AM, more sustainable production may simulate the transient temperature field in additive layer produc-
occur sooner in the product lifetime. A sustainable polymer can tion (ALM) methods. The transient temperature history in the parts
be used to create new parts or products that are made possible processed with ALM that use a motion laser thermal source [2] is
by additive manufacturing and existing parts or products that are crucial for establishing the thermal stress distribution and residual
moved to this workflow (AMT July 2020). stress states. [2].The conventional manufacturing process fre-
The extensive modeling of the relations between the properties quently needs a vast amount of machining and cannot meet the
of Additive Manufacturing (AM) materials and process structure constantly increasing requirements of a sustainable, cost-
includes employing data mining techniques to close the cycle opti- effective, and environmentally friendly in modern industry. Addi-
tive Manufacturing (AM), also referred to as 3D printing, is a direct
digital manufacturing process. 3D digital data can be produced
⇑ Corresponding author. layer by layer with minimum machine and equipment parts or
E-mail address: [email protected] (P. Singh).

https://doi.org/10.1016/j.matpr.2021.12.057
2214-7853/Copyright Ó 2022 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the First International Conference on Design and Materials (ICDM)-2021
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

jobs. Wire arc additive manufacturing (WAAM) is a new technique tions and actual work is shown. A model material may be created
that considerably reduces material usage and production time by to work on the model utilizing primary material as an example as a
providing high deposition rates of near-net-shaped components. computer simulation framework. The findings may be compared
Residual strains and distortions of the work piece deposited are with experimental results after simulations to determine whether
one of the critical problems of this method. Finite element (FE) the model is correct. If the simulation results do not match the test
models are commonly used in understanding and optimization; . data and additional simulations are made, the model should be
however, conventional transient models do not simulate a funda- updated and corrected. The simulation prediction may explain
mental WAAM process [3]. The computer-based modeling benefits molecular characteristics of the whole material system if simula-
from the experimental techniques of polymer nanocomposites tion and experimental results are well agreed on. The results of
(PNCs) that have advanced characteristics owing to their particular the simulation may also obtain data on how experimental methods
hierarchical microstructures The advanced nanocomposite materi- might be developed. The most popular of the many computer sim-
als including a nano-filler such as carbohydrate (CNTs), graphene, ulation techniques for predicting the physical characteristics of
methyl tri methoxysthene, or lignin integrated into the bulk matrix PNCs are molecular dynamics simulation. Simulations of molecular
of polymers, such as epoxy, polystyrene (PS), polypropylene (PP) dynamics may operate as a link between microscopic dynamics
cotton and other materials, are advanced polymer nanocomposites and macroscopic experimental interest (e.g., There are a number
(PNCs) [4]. of factors to consider (glass transition temperature, viscosity, and
Nanocomposite polymers are a novel class of nanoparticles modulus, diffusion coefficients, equations of state, Through the
composed of composites distributed in diverse polymers matrices evaluation of the system’s micro-aspects (such as atom movement,
[i]. In the early years of the composite industry, the development atom interactions, and molecular geometries).
of these materials emerge. The self-assembly in polymer systems
of nano-scale particles like nanospheres, nanodes and nanotubes
is one of the most prominent and promising possibilities to pro- 2. Simulation methods
duce new, high-performance materials. The dispersion across a
polymer or block copolymers of these building blocks can substan- Based on the fundamental theories applied in the property and
tially increase those nanocomposites’ mechanical, optical or elec- structure calculation on PNC, computer simulation can be classi-
trical characteristics. However, a deep knowledge of the fied into the following five major categories:
structural formation and kinetic mechanisms, and structure–prop- 1. Quantum mechanical simulation.
erty connections, is necessary to properly apply them in technolog- 2. Monte Carlo (MC) simulation.
ical applications and assure efficient scale-up. Tailored computer 3. Molecular dynamics (MD) simulation.
simulations will provide a unique technique for investigating 4. Mesoscopic simulation.
structural development and formation and for the determination 5. Simulation on a macroscopic scale, often known as contin-
of structural and property correlations between nanocomposites. uum simulation.
We first describe in this study multi-scale modelling and simula- Fig. 2 shows as compares the time and length scales of the sim-
tion approaches that may be applied to nanocomposites of poly- ulation mentioned above approaches. This graph indicates various
mers. These approaches are then used to many elements, models [11,12] have been developed to forecast the mechanical
including nano compound structures, dynamics, and characteris- characteristics of the AM components, overall predictions of the
tics, consisting of different polymer systems (for example, homo- material response have been challenged, particularly by highly var-
geneous polymers, polymer mixes, and copolymer block ied and anisotropic characteristics. In addition, several variables
mixtures) and nanoparticles with different geometries and pur- influence the construction of a house. The attempts to calibrate
poses. We also focus on the modifications of nanocomposites in and validate even modified versions produced to account for the
multiscale simulation techniques in various approaches, including types of changes in AM materials have revealed that they are
shear, electrical and magnetic fields, and show how external fields somewhat useless about predictive capability in many circum-
may be utilized to tailors the structures and affect the evolution stances that do not fit within the calibration domain.
dynamics of nanocomposites [5]. The ultimate objective of AM modelling is to forecast quantita-
Modeling of AM processes offers essential insights into compet- tively how the end products deploy functionally or load-bearing
ing for physical phenomena that contribute to final material char- capacity, e.g. the characteristics and the life cycle of the input pro-
acteristics and product quality. And application the design space cesses under certain circumstances of load, work. In addition, in
towards usable parts and materials. It is extremely difficult to mea- both structures and properties, we may control the properties
sure the dimensions needed to describe AM processes and predict and design products. Or, to put it another way, the predictive mod-
the item’s properties. The models must cover the length scales els enable the best structural design and property combinations to
resolving the sizes of powder particles and the chamber dimension make optimal use of the basic materials. Defined surface rough-
and the movement to the target of many hundred or thousand ness, microstructures, and even empty spaces in different places
heating sources. The modeling task is thus a multi-scale, multi- can be used to optimize cell adhesion, stiffness, and effective den-
physics endeavor that entailed the complex interaction of several sity. This is because replacement tissues like bones are resistant to
algorithms. Discussing models that are necessary to cover the varying charges at numerous sites and consequently have varied
breadth of AM processes and forecast the material characteristics characteristics. To achieve this, we have presented a data-based
as constructed and residual stresses in the final construction Pro- multi-physic modelling framework for the link between process
vided instances of verification and validation. The aim is to provide structure property performance data mining approaches to com-
an overview of existing methods for modeling how they may con- plete the design-to-predict cycle and process and mechanical
tribute to additives manufacturing. models.
Since then, computer power has grown significantly, and com- For analysing and optimising the manufacturing process at var-
puter simulation has become an essential tool for academics in ious scales, the A.M. Bottom-up approaches are used to understand
several areas in study. The results of computer simulations will the AM process and predict the quality of AM-fabricated materials
be very compatible with experimental work by creating exact force given a set of processing parameters, followed by a top-down strat-
field parameters and the ideal design of simulation methodologies egy to design new material structures and select process parame-
and techniques Fig. 1. The connection between computer simula- ters. Creating and integrating multi-physics models at various
3595
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Fig. 1. The relationship between experiment and computer simulation.

Fig. 2. Comparison of different computer simulation methods in terms of time and duration.

scales. New heat source models derived from micro-scale electron changes in scale, are regularly but do not always vary in their the-
material interaction simulations (for EBSM), meso-scale models oretical properties. [7,8] the hierarchy includes running distinct
simulating individual molten pool tracks at the powder scale, models with some form of parametric connection, whereas [9,10]
and a homogeneous macro-scale model simulating the evolution hybrid involves running models concurrently over several loca-
of additively manufactured products are all steps in this frame- tions in order to mimic them. Although specific methods, for exam-
work’s creation. ple molecular dynamics (MD) and Monte-carlo (MC) are known for
a long time and are now frequently utilised, other techniques like
2.1. Using multi-scale modelling and simulation techniques quantum mechanical (QM), mesoscale modelled, and other
advanced ways of acceleration for simulation are not so common.
Multi-scale modelling refers to the use of two or more distinct Therefore, a basic technical overview for non-specialists is
length and time scale modelling methods that, according to included here [5]

3596
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

A. the quantum mechanical simulation., this study is based on MD simulation requires an input force field file that provides the
functional theories from the first principle density, the ab initio parameters for describing particles, for example, atom mass, bond
molecular dynamics simulation method. Electron correlations are angle, bond length, etc. MD can represent the microstructure and
estimated using electron density and density functional theory, interaction between particles with accuracy, even with approxima-
with corrections for local orbital distortion. Known for its reliabil- tions in the computation. The sizes of the MD system are generally
ity and accuracy, the quantum mechanical simulation does not between 10 Å and 100 nm. The system has hundreds to millions of
incorporate any empirical or changeable factors [13]. Only around atoms, and the simulation takes 100 ns to microseconds. The MD
100 atoms may be simulated; therefore, the simulation duration is simulations may also be divided into molecular dynamics simula-
usually measured in picoseconds or fewer to compensate. tions at the atomic and molecular scales, unified atomic simula-
B. The simulation of Monte Carlo (MC) is an application to tions, and coarse simulations based on the kind of concentrated
dense molecular systems by the class of probabilistic mathematical particles (atomists, molecules, or beads).
models. In the mid 1940 s it was created to investigate, based on a C-1) Atomic and molecular MD focuses on atomic/molecular
random walk, the spread of neutrons into fissionable material [14]. interactions with atomic/molecular dynamics. The advantage is
If only particulates with explicitly defined co orders are collected that it is accurate while limiting the size and length of the simula-
and their energies computed as sums of pair-wise interactions in tion system. The atom/molecular MD are generally not greater
both molecular Dynamics (MD) and MC systems, when it comes than 10 nm in one direction, and with the use of a considerable
to physical size, MCS might be as huge as a medical doctor. MD number of the computer resources the simulation length can
is the opposite force between particles in MC does not have to be exceed microseconds. Regarding MD simulations, PNCs rely on
calculated. On PNCs many MC simulations were used. Material Studio, LAMMPS (for coarse-grained models), GROMACS
(1) The random walking of classical Monte Carlo [15]; (again for coarse-grained models), CGM, and YASP in most cases.
(2) Monte Carlo metropolis [16] that picks configuration based C-2) Molecular dynamics simulation based on united-atom
to weight or a key sampling approach in its selection of force fields: When atoms are combined into one particle, forces
configurations; between particles are calculated without regard to atomic proper-
(3) Monte Carlo coarse-grained [19], used in the Monte Carlo ties and interactions. It is about larger, but less exact, scaled sys-
simulation to characterize particulate matter in the coarse-grain tems than nuclear/molecular MD.
pattern; C-3) Molecular simulation of coarse-grained (CG) dynamics:
(4) Kinetic Monte Carlo [17], which mimics time process devel- tiny clusters of atoms are treated as individual particles or beads,
opment and hence enables time to match MC paths; allowing extended time scale to be handled. It is usually used for
(5) Monte-Carlo offshoot [18], allowing MC to deal with contin- modeling nanostructures materials. The atoms, molecules, and
ual freedoms, and contrary, with the Monte-Carlo Lattice, which beads are inked on. Mapping techniques are intended to convert
prescribes the number for which trial compliance is conducted, the atomic system into coarse-grain systems and then transform
MC can now cope with discrete degrees of freedom, such as coarse-grain systems after long enough simulations to have
Density-functional spins, for example (See Figs. 3 and 3.1). enough precision in the ground grains MD results compared to
C. An example of a molecular dynamics simulation To deter- the atomic MD equivalent [20]. For particle-based MD techniques,
mine microstructures and interactions between the particles, MD the maximum system size and simulation duration are generally
is based on a description of the physical movements of the particle defined by the balance between coarse-graining and the number
for a brief period (for example, in picoseconds to microseconds). of details preserved. This is true for atomic MD, unified atom
Particle dynamics have been computed by the law of Newton. MD, or coarse-grain MD simulations, as well.

Fig. 3. A data-driven multi-scale multi-physics modeling frameworks [1].

3597
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Fig. 3.1. (Section of above figure which is taking the part of Property. Graph between stress (MPa) v/s %age of Elongation.

D. The mesoscopic method is a hybrid approach that combines (1) In addition to thermal conductivity, (2) there is electrical con-
microscopic and macroscopic aspects. The experiment’s time-scale ductivity, (3) bulk modules and complex modulus, (4) transitional
and length are similar in the microsecond range and micrometer- temperature of the glass, (5) coefficient of thermal expansion, and
millimeter range to microseconds .Its fast, but it’s not as precise (6) diffusion coefficient. The Green-Kubo relations [29, 30, and 31]
as experimental or quantum calculations A typical mesoscopic of the equilibrium simulations of the atomic systems can be used,
simulation is Brownian dynamics [21] or Dissipative Particle among them, thermal conductivity and electrical conductivity.
Dynamic Dynamics (DPD) [22]. Brown’s dynamics may be used
to describe the random and continuous movement of the particle. 3.1. Conductivity of heat (thermal)k
Using projection operators, a single molecule or particle of a single
set of components may be applied.. Projection operators, like Lan- It is the linear coefficient relationship. Thermal conductivity is
gevin dynamics, have a means of reducing description on motion defined by Fourier’s law as the linear coefficient linking the macro-
equations in simulation techniques [23]. As the name suggests, scopic heat current (J) to the temperature gradient (T):g the macro-
DPD is a rough-grain design, in which each atom or molecule is scopic heat current ðJÞ to the temperature gradient (rT) (Fourier’s
not explicitly represented but is ground-grained into some of these law):
particles or grains may be able to travel on Nanometers to millime-
J ¼ krT ð1Þ
ters in size and nanoseconds to seconds in length are all covered in
this book. In addition, it can mimic hydrodynamics, including heat Microscopically, it can be calculated using Green-Kubo relations
changes, in an inexpensive method. [28]:
E. The continuum methods utilizes continuity equations, such Z 1
1
as the macroscopic fluid mechanics Navel Stokes (NS) equations k¼ < J ðt ÞxJð0Þ > dt ð2Þ
VKBT 0
[24]. Models are constructed using representative volume ele-
ments (RVEs), which have a sufficient size to disregard the under- Where J is heat current, x represents cross product, <,> denotes the
lying structures of the object. You may set a timer from a few ensemble average in the thermal equilibrium V system volume, T is
minutes to a few days, and you can If you’re interested in using the temperature, and kB is the Boltzmann constant.
the continuous approach, it’s possible that it’s accurate and can With the use of equilibrium molecular dynamics simulations,
run widespread systems for a lengthy period of time; nevertheless, Shenogin et al. [31] determined that the thermal conductivity of
model equations can be problematic As an example, the finite ele- CNT composites is proportional to that of functionalized tube car-
ment approach is a continuum methodology [25]. Another bon atoms.
approach called multi-scaled simulation may be applied with Conductivity of electricity
PNC, according to the simulation methods. Microorganism (e.g. According to a material’s electrical conductivity (r), it is able to
unified atomic or atomic/molecules MD or coarse-grained MD) is transfer charges. A macroscopic calculation utilizing Ohm’s law is
one of the five levels of simulation used in the multi-scale model. possible: J = rE It’s important to remember that J stands for current
It forecasts the characteristics and dynamics of material from sev- density and E stands for
eral levels based on information. Due to the hierarchical architec- Using the Green-Kubo relation method, it may be computed
ture of the Nano-sized fillers contained in the polymer matrix, microscopically [28] as
researchers have created multi-scaled modular modelling methods Z 1
to deal with PNC [26,27]. re¼Vk1 T el el
dt < jx ð0Þjx ðt Þ > ð3Þ
B 0

el el
Where, jx x is the collective current along the x direction, and jx
3. Physical properties calculated using by computer simulation el
at the time t.jx
method
Defined as
Xn
The computer simulation approach is used to determine several el
jx ¼ q
i¼0 i
v xi ð4Þ
of PNC’s dynamic and static properties. Included are the following:
3598
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Fig. 4. At different CNT volume fractions, NCT Co. Ltd. and Y. Ono et al. compared the electrical conductivity values of PNCs estimated from models and from experimental
measurements Ref. [32] was used to create the figure.

Where, N is the total particle number; qi is electric charge on parti- properties. Nanocomposites have less fragile glass because of the
cle i; and v xi is x-direction velocity of particle i. distributed, appealing nanoparticles. Under equilibrium and non-
When Fang et al. [32] simulated the three-dimensional CNT net- equilibrium conditions, the nanometers impact the PNC’s structure
works in the constructed PNCs; they used an efficient RVE model. and dynamics [52]. Additionally, the behavior of the nanocompos-
As illustrated in Fig. 4, they compared the generated electrical con- ite is dependent on nanoparticle size and shape as well. [53]. CG
ductivity to experimental data between the simulation and the Monte Carlo and MD [55] conducted two simulations that exam-
experimental results, a satisfactory agreement was obtained. It ined vacuum development before crazy creation and the early
shows in the graph between simulation of volume fraction and crazy formation.
electrical conductivity true with experimental value. Disadvantages: Computationally inefficient When dealing with
a large number of variables bound by different constraints, this
method requires a lot of time and computations. Poor parameters
4. Simulation results and application and constraints will result in poor outputs.

4.1. A method of quantum simulation

4.3. MD method application
is used. For all-atom the force field parameters for PNC simula-
The nanocomposites have been shown to have improved char-
tions have been determined using ab initio MDs. for analysis and
acteristics when compared to polymer matrices, after the mixing
the calculation of structural and electronic proprieties of interface
of nanosized fillers with polymer matrices. MD simulations were
interaction between the filler nano and polymer matrices with
used by researchers to describe the Structure of PNCs, prediction
defects [36 37-39]. Apart from the initial simulation, ionic poly-
of PNC properties, and exploration of the connections between
merization mechanisms in the manufacture of graphene polymer
chemical composition, properties, and structure of nanofillers, as
nanocomposites were analyzed [40]; the structure, dynamics,
well as mechanical characteristics of PNCs. in order to understand
and salvation of an unordered metal–organic polymer used to be
better the enhancing mechanical composition and the description
explored [42]; and interaction ions of an electrolyte of PNC inves-
of PNCs with optimization properties for various applications.
tigated [43], etc. Initial simulations conduct. As a result of quantum
B. Properties of Dynamic mechanical calculation
mechanical simulation, the simulation does not depend on the spe-
B.1) Atomic MD results
cifics of the applicable experimental field. The simulation is gener-
For example, the Payne effect, loss modulus, and storage modu-
ally based on extremely tiny system volumes due to the enormous
lus and the influence of interactions between nanoparticles and the
use of computer resources in calculations, either the small inter-
surrounding PNC polymer matrix were investigated using MD sim-
face between nanofillers and the polymer or the interface defects
ulations [55,56]. On the other hand, simulations using MD software
or the electron polymers surface (See Fig. 5.1 and 5.2).
were used in the study.
B.2) MD atomic unit Graphene can increase its toughness
4.2. Monte Carlo simulation throughout an extensive temperature range through to the matrix
material. In the rubbery state, In the glassy condition, this impact is
The Monte Carlo simulation method relates structure to the more powerful. As well as, the more influential the graphene-filled
application [46]; NC prediction; thermal conductivity [45]. Monte PNC [57] is placed in polymer matrixes, the stronger the link
Carlo simulations may be used to understand the reinforcement between polymer and graphene. It has been shown that after
mechanism [44] and the molecular optical property of the charge transfer, around the graphene and CNT contact, a new poly-
polymer-based nanocomposites and PNC structure [48] and to pro- mer matrix phase forms, which is stronger and denser than the
vide information between the molecular level (the information in original polymer matrix. In Nano’s electronic application, this can
the polymer chain) and meso (the phase structure) [48,49] among assist explain the mechanical deterioration of hybrid
various applications. The application of efficient Monte Carlo sim- nanocomposites.
ulations, PNC structure, and interaction between J. de Pablo and his B.3) Coarse-grained MD results
co-workers [50,51] examined the nanoparticles and polymer Polyethylene matrixes can be improved by uniformly dispersing
matrix seen in the particles, influencing PNC’s local mechanical (nanoparticles) contribute to the increased strength of the PNCs’
3599
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Fig. 5. CG scheme from the atomic to the mesoscale levels. CG scheme. On the left side of the CG structure the all-atom representation will be presented on the right side. A
comparison of the bond distance and bond angle distributions is shown in the inset for units generated from the simulation of all atoms (line) and coarse grain CG (symbols)
[5].

Fig. 5.1. Section of Fig. 5 is shown in this figure.

elastic module yield during shock wave loading; on the other hand, to the polymer matrix, depending on the volume % of NPs and their
when NPCs are inappropriate by the polymer matrix and the PNC interaction with the polymer matrix. They exhibit greater tough-
can be increased in the microscope and applied in energy absorp- ness efficiency on polymers than CNT and C60, with the same
tion, which leads to the formation of layered PNC structure [59]. An attractive contact between polymers and nanoparticles, and at
increase in PNC strength may be achieved by adding nanoparticles the same loading [57].

3600
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Fig 5.2. (Section of Fig. 5) PEO–PEE diblock copolymer (left panel) to coarse grain model mapping (right panel). There are two main types of polymer: EO and EE. The EO
polymer is represented by the coarse grain unit COE, while the EE polymer is represented by the CH2–CH (CH2–CH3). Because of its symmetry, the COE is preferred to the
ethylether monomer unit –CH2–CH2–O. Interfacial unit CEI is represented by CET and OA, while tertiary butyl and CH2–OH are represented by CEI and CH2–OH, respectively,
by the end groups.

In order to explore this suggestion, different research were con-

ducted in computational, theoretical, and simulation. Thus, for
example, the Swiss and coworkers of microscopic PRISM [60] have
found that four general categories of polymers of nanoparticles are
organized for hard-sphere fillets:

Contact aggregations.
Tight particle bridges on a segmental level.
Thermodynamically stable steric stabilization in ‘‘bond polymer
layers.”
Telebridging.

In addition, three-phase dispersion diagrams may produced for

hard fillers: A macroscopic phase split at a low filler polymer, an
enthalpically stabilised miscible fluid with a moderate interface,
and a microscopic phase split caused by local polymer filler bridg-
ing are all examples of macroscopic phase splitting.
(Fig. 6). Liu et al. [61] demonstrated this theoretical prediction
by means of MD simulations Nanoparticle grafted dispersions of
homopolymers, as well as the self-assembly and organisation of
Fig. 6. The polymer–filler interaction changes the tendency for dispersion state of
nanoparticles in copolymers, were used to develop the medium the filler [4].
field models Ganesan et al. [62] evaluated the development of
equilibrium dispersion and polymer nanocompose structures.
With the Ue Signifying The right elastic stretch tensor.
Study of additional deformations using the right Cauchy-Green
5. Modeling of material tensor C.

C ¼ F t :F ¼ F tp :C e :F p ð6Þ
It is based on a single nanoporous crystal filled with polymer.
Deformation gradient multiplicative breakdown of elastic and The elastic right Cauchy Green tensor C e ¼ F te :F e and plastic left
plastic components as a basis for Kinematics green tensor Bp ¼ F p :F tp the relevant stress measures are the piola-
^  Fp Kirchhoff stress P
F ¼ FeA ð5Þ
P ¼ F e :Se :F t
p ð7Þ
The elastic Green–Lagrange strain tensors defined as
Ee ¼ 1=2½F te :F e  I whose conjugate stress measures the elasticity And second Piola –Kirchhoff stress Se
rotated Cauchy stress T e ¼ Rte :T:Re The polar decomposition theo-
e s:F e
S ¼ F t t
ð8Þ
rem must be applied in order to define elastic rotation. F e ¼ Re :U e
3601
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

Mechanical issues are controlled by liner momentum equilib- [12] P.K. Valavala, G.M. Odegard, Modeling techniques for determination of
rium. The Kirchhoff stress s is represented here as ‘‘”. mechanical properties of polymer nanocomposites, Rev. Adv. Mater. Sci. 9
(1) (2005) 34–44.
[13] N. Metropolis, S. Ulam, The monte carlo method, J. Am. Stat. Assoc. 44 (247)
O ¼ Div P þ bo ð9Þ (1949) 335–341.
[14] Y. Termonia, Polymer chain properties and thermodynamic stability in
Here bo is the body force. oriented-platelet nanocomposites, Phys. Rev. E 88 (1) (2013) 012603.
[15] S. Nam, H.W. Cho, S. Lim, D. Kim, H. Kim, B.J. Sung, Enhancement of electrical
and thermomechanical properties of silver nanowire composites by the
6. Conclusions and future work introduction of nonconductive nanoparticles: experiment and simulation,
ACS Nano 7 (1) (2013) 851–856.
[16] L. Rosenthal, H. Greve, V. Zaporojtchenko, T. Strunskus, F. Faupel, M. Bonitz,
This study reviews; Formation of magnetic nanocolumns during vapor phase deposition of a
metal-polymer nanocomposite: Experiments and kinetic Monte Carlo
simulations, J. Appl. Phys. 114 (4) (2013) 044305, https://doi.org/10.1063/
 Different physical properties computed with simulation meth-
1.4816252.
ods, notably distinct MD approaches. A polymer matrix and [17] H.M. Duong, N. Yamamoto, K. Bui, D.V. Papavassiliou, S. Maruyama, B.L.
nanofiller interactions influence these properties. Wardle, Morphology effects on nonisotropic thermal conduction of aligned
 Understanding the links between nanostructures and PNC prop- single-walled and multi-walled carbon nanotubes in polymer
nanocomposites, J. Phys. Chem. C 114 (19) (2010) 8851–8860.
erties (especially thermo mechanical conductivity) can lead to [18] M. Goikoetxea, Y. Reyes, C.M. de las Heras Alarcón, R.J. Minari, I. Beristain, M.
more efficient PNC use. MD for better PNC design. Paulis, M.J. Barandiaran, J.L. Keddie, J.M. Asua, Transformation of waterborne
 Despite the rapid advancement of PNC experimentation and hybrid polymer particles into films: Morphology development and modeling,
Polymer 53 (5) (2012) 1098–1108.
simulation methods, more work is needed. [19] A.S. Pavlov, P.G. Khalatur, Fully atomistic molecular dynamics simulation of
 It’s difficult to experiment with PNCs with better thermal con- nanosilica-filled crosslinked polybutadiene, Chem. Phys. Lett. 653 (2016) 90–
ductivity because thermal conductivity depends on polymer 95.
[20] S. Chandrasekhar, Stochastic problems in physics and astronomy, Rev. Mod.
matrices, surface chemicals, and nano-filler size. Phys. 15 (1) (1943) 1–89.
 Innovative MD methods will be required in the future to help [21] P.J. Hoogerbrugge, J.M.V.A. Koelman, Simulating microscopic hydrodynamic
build unique PNC structures with better thermal and electrical phenomena with dissipative particle dynamics, EPL (Europhys. Lett.) 19 (3)
(1992) 155–160.
conductivity.
[22] M.P. Allen, D.J. Tildesley, Computer simulation of liquids, Oxford University
 We believe the process structure-ownership models will also Press, 2017.
serve as a roadmap for developing AM monitoring and [23] L.C. Squire, An Introduction to Fluid Dynamics. GK Batchelor. Cambridge
University Press. 1967. 616 pp. Figures. 15s. Aeronaut. J. 71 683 (1967) 801–
diagnostics.
802.
[24] J.YH. Chia, Finite element modelling clay nanocomposites and interface effects
on mechanical properties, Springer, Dordrecht, 2009.
CRediT authorship contribution statement [25] R. Toth, F. Santese, S.P. Pereira, D.R. Nieto, S. Pricl, M. Fermeglia, P. Posocco, Size
and shape matter! A multiscale molecular simulation approach to polymer
Phool Singh: Conceptualization, Writing – original draft. Ran- nanocomposites, J. Mater. Chem. 22 (12) (2012) 5398, https://doi.org/10.1039/
c2jm15763b.
ganath M.Singari: Supervision. R.S. Mishra: Supervision. [26] J. Choi, S. Yang, S. Yu, H. Shin, M. Cho, Method of scale bridging for
thermoelasticity of cross-linked epoxy/SiC nanocomposites at a wide range
of temperatures, Polymer 53 (22) (2012) 5178–5189.
Declaration of Competing Interest [27] D. Frenkel, B. Smit, Understanding molecular simulation: From algorithms to
applications, Comput. Sci. Ser. 1 (2002) 1–638.
The authors declare that they have no known competing finan- [28] M.S. Green, Markoff random processes and the statistical mechanics of time-
dependent phenomena. II. Irreversible processes in fluids, J. Chem. Phys. 22 (3)
cial interests or personal relationships that could have appeared (1954) 398–413.
to influence the work reported in this paper. [31] W. Fang, H.W. Jang, S.N. Leung, Evaluation and modelling of electrically
conductive polymer nanocomposites with carbon nanotube networks,
Compos. B Eng. 83 (2015) 184–193.
References [32] Andrey G. Kalinichev, et al., Molecular models and simulations of layered
materials. No. SAND–2008-7390J. Sandia National Laboratories (United
States). Funding organisation: US Department of Energy (United States), 2008.
[1] W. Yan, S. Lin, O.L. Kafka, C. Yu, Z. Liu, Y. Lian, S. Wolff, J. Cao, G.J. Wagner, W.K.
[36] V. Bala, S.K. Tripathi, R. Kumar, Optical properties of Ga and in doped CdS
Liu, Modeling process-structure-property relationships for additive
nanocomposites: an experimental and first principles study, Mater. Lett. 149
manufacturing, Front. Mech. Eng. 13 (4) (2018) 482–492.
(2015) 18–21.
[2] I.A. Roberts, C.J. Wang, R. Esterlein, M. Stanford, D.J. Mynors, A three-
[37] J. Mazher, A.A. Desta, S. Khan, PAn-Graphene-Nanoribbon Composite Materials
dimensional finite element analysis of the temperature field during laser
for Organic Photovoltaics: A DFT Study of Their Electronic and Charge
melting of metal powders in additive layer manufacturing, Int. J. Mach. Tools
Transport Properties, Sol. Cell Nanotechnol. (2013) 357–407.
Manuf. 49 (12-13) (2009) 916–923.
[38] K.e. Wu, M. Bell, A. Oakes, T. Ratcliff, N.A. Lanzillo, C. Breneman, B.C.
[3] J. Ding, P. Colegrove, J. Mehnen, S. Williams, F. Wang, P.S. Almeida, A
Benicewicz, L.S. Schadler, The effects of nanoparticles and organic additives
computationally efficient finite element model of wire and arc additive
with controlled dispersion on dielectric properties of polymers: Charge
manufacture, Int. J. Adv. Manuf. Technol. 70 (1-4) (2014) 227–236.
trapping and impact excitation, J. Appl. Phys. 120 (5) (2016) 055102, https://
[4] J. Zhao, L. Wu, C. Zhan, Q. Shao, Z. Guo, L. Zhang, Overview of polymer
doi.org/10.1063/1.4959771.
nanocomposites: computer simulation understanding of physical properties,
[39] F. Sheng et al., Simulation on field enhanced electron transfer between the
Polymer 133 (2017) 272–287.
interfaces of ZnO–Ag nanocomposite, J. Phys. Chem. C 117 (36) (2013) 18627–
[5] L.-T. Yan, X.-M. Xie, Computational modeling and simulation of nanoparticle
18633.
self-assembly in polymeric systems: Structures, properties and external field
[40] K. Mylvaganam, L. Zhang, Fabrication of graphene polymer
effects, Prog. Polym. Sci. 38 (2) (2013) 369–405.
nanocompositesthrough ionic polymerization, J. Phys. Chem. A 120 (39)
[7] Julia A. Elliott, Novel approaches to multiscale modelling in materials science,
(2016) 7689–7693.
Int. Mater. Rev. 56.4 (2011) 207–225.
[42] O. Borodin, G.D. Smith, R. Bandyopadhyaya, P. Redfern, L.A. Curtiss, Molecular
[8] Ronald E. Miller, Ellad B. Tadmor, Hybrid continuum mechanics and atomistic
dynamics study of nanocomposite polymer electrolyte based on poly(ethylene
methods for simulating materials deformation and failure, MRS Bull. 32 (11)
oxide)/LiBF 4, Modelling Simul. Mater. Sci. Eng. 12 (3) (2004) S73–S89.
(2007 Nov) 920–926.
[43] J.E. Mark, R. Abou-Hussein, T.Z. Sen, A. Kloczkowski, Some simulations on filler
[9] N. Bernstein, J.R. Kermode, G. Csányi, Hybrid atomistic simulation methods for
reinforcement in elastomers, Polymer 46 (21) (2005) 8894–8904.
materials systems, Rep. Prog. Phys. 72 (2) (2009) 026501, https://doi.org/
[44] Khoa Bui, Dimitrios V. Papavassiliou, Heat Transfer in Nanocomposites with
10.1088/0034-4885/72/2/026501.
Monte-Carlo Simulations, Defect and Diffusion Forum. Vol. 312. Trans Tech
[10] S. Leuders, M. Vollmer, F. Brenne, T. Tröster, T. Niendorf, Fatigue Strength
Publications Ltd, 2011.
Prediction for Titanium Alloy TiAl6V4 Manufactured by Selective Laser
[45] G. Jiang, M.J.A. Hore, S. Gam, R.J. Composto, Gold nanorods dispersed in
Melting, Metall. Mat. Trans. A 46 (9) (2015) 3816–3823.
homopolymer films: optical properties controlled by self-assembly and
[11] R. Hedayati, H. Hosseini-Toudeshky, M. Sadighi, M. Mohammadi-Aghdam, A.A.
percolation of nanorods, ACS Nano 6 (2) (2012) 1578–1588.
Zadpoor, Computational prediction of the fatigue behavior of additively
manufactured porous metallic biomaterials, Int. J. Fatigue 84 (2016) 67–79.

3602
P. Singh, R. M.Singari and R.S. Mishra Materials Today: Proceedings 56 (2022) 3594–3603

[46] M. Vacatello, Predicting the molecular arrangements in polymer-based [61] V. Ganesan, C.J. Ellison, V. Pryamitsyn, Mean-field models of structure and
nanocomposites, Macromol. Theory Simul. 12 (1) (2003) 86–91. dispersion of polymer-nanoparticle mixtures, Soft Matter 6 (17) (2010) 4010–
[48] N. Yan, Y. Zhu, W. Jiang, Self-assembly of AB diblock copolymer confined in a 4025.
soft nano-droplet: a combination study by Monte Carlo simulation and [62] S.O. Nielsen, C.F. Lopez, G. Srinivas, M.L. Klein, Coarse grain models and the
experiment, J. Phys. Chem. B 120 (46) (2016) 12023–12029. computer simulation of soft materials, J. Phys.: Condens. Matter 16 (15) (2004)
[49] G.J. Papakonstantopoulos, K. Yoshimoto, M. Doxastakis, P.F. Nealey, J.J. de R481–R512.
Pablo, Local mechanical properties of polymeric nanocomposites, Phys. Rev. E
72 (3) (2005), https://doi.org/10.1103/PhysRevE.72.031801.
[50] G.J. Papakonstantopoulos, M. Doxastakis, P.F. Nealey, J.-L. Barrat, J.J. de Pablo,
Further reading
Calculation of local mechanical properties of filled polymers, Phys. Rev. E 75
(3) (2007), https://doi.org/10.1103/PhysRevE.75.031803. [6] Q.H. Zeng, A.B. Yu, G.Q. Lu, Multiscale modeling and simulation of polymer
[51] G.N. Toepperwein, N.C. Karayiannis, R.A. Riggleman, M. Kröger, J.J. de Pablo, nanocomposites, Prog. Polym. Sci. 33 (2) (2008) 191–269.
Influence of nanorod inclusions on structure and primitive path network of [29] R. Kubo, Statistical-mechanical theory of irreversible processes. I. General
polymer nanocomposites at equilibrium and under deformation, theory and simple applications to magnetic and conduction problems, J. Phys.
Macromolecules 44 (4) (2011) 1034–1045. Soc. Jpn. 12 (6) (1957) 570–586.
[52] G.N. Toepperwein, R.A. Riggleman, J.J. de Pablo, Dynamics and deformation [30] S. Shenogin, A. Bodapati, L. Xue, R. Ozisik, P. Keblinski, Effect of chemical
response of rod-containing nanocomposites, Macromolecules 45 (1) (2012) functionalization on thermal transport of carbon nanotube composites, Appl.
543–554. Phys. Lett. 85 (12) (2004) 2229–2231.
[53] G.N. Toepperwein, J.J. de Pablo, Cavitation and crazing in rod-containing [33] J. Wu, Density functional theory for chemical engineering: From capillarity to
nanocomposites, Macromolecules 44 (13) (2011) 5498–5509. soft materials, AIChE J. 52 (3) (2006) 1169–1193.
[55] J. Hager, R. Hentschke, N.W. Hojdis, H.A. Karimi-Varzaneh, Computer [34] Z. Gu, P.B. Balbuena, G. Sandi, Molecular dynamics characterization of
simulation of particle–particle interaction in a model polymer polymer confinement in nanocomposite catalytic membranes, J. New
nanocomposite, Macromolecules 48 (24) (2015) 9039–9049. Mater. Electrochem. Syst. 7 (2004).
[56] J. Liu, J. Shen, Z. Zheng, Y. Wu, L. Zhang, Revealing the toughening mechanism [35] J. Moon, S. Yang, M. Cho, Interfacial strengthening between graphene and
of graphene–polymer nanocomposite through molecular dynamics polymer through Stone-Thrower-Wales defects: Ab initio and molecular
simulation, Nanotechnology 26 (29) (2015) 291003, https://doi.org/10.1088/ dynamics simulations, Carbon 118 (2017) 66–77.
0957-4484/26/29/291003. [41] C.C. Jolley, J. Lucon, M. Uchida, C. Reichhardt, M.J. Vaughn, B.J. Lafrance, T.
[57] Y. Zhang, X. Zhuang, J. Muthu, T. Mabrouki, M. Fontaine, Y. Gong, T. Rabczuk, Douglas, Structure, dynamics, and solvation in a disordered metal–organic
Load transfer of graphene/carbon nanotube/polyethylene hybrid coordination polymer: a multiscale study, J. Coord. Chem. 64 (24) (2011)
nanocomposite by molecular dynamics simulation, Compos. B Eng. 63 4301–4317.
(2014) 27–33. [47] Y. Zhu, W. Jiang, Self-assembly of diblock copolymer mixtures in confined
[59] J.B. Hooper, K.S. Schweizer, Theory of phase separation in polymer states: a Monte Carlo study, Macromolecules 40 (8) (2007) 2872–2881.
nanocomposites, Macromolecules 39 (15) (2006) 5133–5142. [54] R. Hentschke, J. Hager, N.W. Hojdis, Molecular modeling approach to the
[60] J. Liu, Y. Gao, D. Cao, L. Zhang, Z. Guo, Nanoparticle dispersion and aggregation prediction of mechanical properties of silica-reinforced rubbers, J. Appl.
in polymer nanocomposites: insights from molecular dynamics simulation, Polym. Sci. 131 (18) (2014) n/a–n/a.
Langmuir 27 (12) (2011) 7926–7933. [58] Y. Fu, J. Michopoulos, J.-H. Song, Dynamics response of polyethylene polymer
nanocomposites to shock wave loading, J. Polym. Sci., Part B: Polym. Phys. 53
(18) (2015) 1292–1302.

3603