Abstract

This project aims to study in the wave propagation behavior of carbon

nanotubes. Highly superior mechanical and physical properties of CNTs have
led to wide-ranging studies in the field of nanotechnology. Lately in the past
years, may research have reported on this superiority of nonlocal models of
CNTs to their local counterparts for modelling the vibration of carbon
nanotubes. This research is about using such models to study transverse and
torsional wave propagation in carbon nanotubes. The scale effect of wave
propagation of the carbon nanotube is investigated for different types such as
open-ended and capped ends, different lengths, and also CNTs under initial
compression stress.
Introduction
A material that is 10-9 in unit is called a nanoparticle or nanomaterial which is
1/50000th of human hair. Nanomaterials are often employed in the field of
nanotechnology. They are usually engineered, produced as ab by-product or
they occur naturally – organics and non-organic. Based on the confinement of
electrons, they are classified into Quantum Well, Quantum Wires and
Quantum Dots. Carbon has a Unique structure that enables it to combine with
other element resulting in new components. It occurs in four allotropes such as
diamond, graphite, fullerene and amorphous. Nanotube is a structure that is
rolled up to form a tube and is in nanometer scale(10 -9m) and the organic
nanotubes are referred as Carbon Nanotubes (CNT). These CNTs are divided
into Single Walled Nano Tube (SWNT) and Multi Walled Nano Tube (MWNT).
The figure 1 below shows the types of nanotubes based on their walls. The
MWNT is further classified into Russian Doll Structure and Parchment
Structure.
Single-walled carbon nanotubes are seen as graphene sheets that are made to
be rolled into sheets as shown in the figure as zigzag, armchair and chiral
types. SWCNTs are several times stronger that steel but 1/6 th of steel’s weight.
This property makes them valued material for reinforcing to increase the
strength of composite materials.
The chiral angle is calculated with the formula:
θ = cos−1 ¿

Figure 1-Schematic view of different kinds of nanotubes: a zigzag, b armchair and c chiral

The CNT is classified under fullerene. On the basis of chirality, CNTs are
divided into Chiral and Non-Chiral, where the Non-Chiral is divided into Zigzag
and Armchair. They have better thermal, electrical, Physic-Chemical, Field
emission and optical properties. CNTs are the strongest materials discovered
with reference to their tensile strength (up to 63 GPa) and elastic modulus. The
bonding property is found to be as same as graphite. The Carbon Nano Tubes
propagate waves in terms of terahertz (THz). The C-C bond of the carbon
nanotubes can be described via various potential functions like harmonic, cubic
anharmonic, quartic, Morse and other functions. Total potential energy is the
total sum of energies because of bonded interactions as well as non bonded
interactions. It is expressed in the equation:
V = ΣV r + Σ Vθ + Σ V φ + Σ Vω+ ΣVvdw
Where V r - bond stretch interaction
V θ - bond angle bending

Vφ - dihedral torsion angle

V ω - improper torsion

Vvdw – non-bonded van der Waals interaction

These energy terms are explained clearly in the figure (below)

Figure 2- Energy terms of bonded and nonbonded interaction(Gajbhiye SO, 2015)

From the potential functions TB potential is found to be more accurate

and is used widely in molecular dynamics simulations. It is used to express a C-
C bond and also the stretching of a bond and bond angle bending force
constant. The energy between the atoms i and j is determined through the
formula:
 Where rij is the distance between the atoms i and j that is same as the length
of bond at initial equilibrium configuration. And the value of f c(r) is 1, that
reduces the time of computation by limiting the potential range and includes
the immediate neighbor shell of the atoms of carbon. The figure(below) shows
the part of angle bending of a nanotube.

Figure 3-Part of nanotube with angle bending (Gajbhiye SO, 2015)

General background
The Carbon Nanotubes are first discovered in the year 1991 by Japanese
physicist Sumio Iijima, which gave rise to branch of interest in the field of
material sciences which is the nanoscience. So far, millions of dollars have
been spent on these nanomaterials which is totally worth it as they have
diverse and unique properties. The Carbon Nano Tubes are 100 times stronger
than steel on molecular level, but they weigh only 1/6 th of steel with high
aspect ratio. (Azonano, n.d.) The figure below shows the footprints of CNT in
Nature and it has been around for ages.
 Figure 4 - Footprints of CNT in Nature(www.nanowerk.com)

CNTs were in the Damascus steel dating to 17 th century, out of which

legendary swords were made. The carbon nanotubes were first produced by
Laser ablation in the year 1995. It is an expensive technique as it requires high
power input and also expensive lasers. The most capable method of producing
CNTs is thee Chemical Vapor Deposition (CVD). It uses energy of 600-900°C to
atomize gaseous sources of carbon such as Acetylene, Methane and Carbon
monoxide. Then the carbon atoms from the result diffuse with a catalyst-
coated substrate and finally condense to give carbon nanotubes. Crafting
materials with CNTs as additives result in plastic composites with better
mechanical strength and electrical conductivity. In biomedical applications,
CNTs acts as a vehicle for delivering drugs and in nerve cell regeneration. (Ren,
n.d.). They were found to create sufficient heat to kill cancer cells when
directed a near infrared laser at the tumour cells in mice.
Recent studies reveal that, Carbon Nano Tubes also show promise for 5G
antennas and found to be as effective as copper in wireless applications. The
figure below shows metal-free antennas that are made out of strong, thin and
flexible CNT films.
 Figure 5 - CNTs in Antenna(Image:theengineer.co.uk)

Figure 6-Bronze mask before and after coated with Vantablack(Image:news.artnet.com)

Vantablack is known as one of the darkest materials on earth that can absorb
upto 99.96% of light. They were made with bulk carbon nanotube arrays. The
image below shows a bronze mask that is before and after coated with
vantablack paint. It is used in solar panels thus increasing the rate of heat
absorption and also for military purposes, especially in thermal camouflage.
CNTs are also used in tissue engineering that helps in scaffolding bone growth.
The image below shows a flexible hydrogen sensor that is made by scientists
with single walled carbon nanotubes. These sensors can detect leaks from
hydrogen powered automobiles and fuel stations before becoming highly
hazardous. (Berger, 2007)
 Figure 7- Flexible hydrogen sensor made of Single-walled CNTs((Image: Dr. Sun/Argonne)

Literature Review
The mechanical behavior of CNTs has been studied using experimental,
theoretical and computational modelling methods. The study of wave
propagation is made by the determination of the number and speed (phase
and group) of waves. Gajbhiye and Singh (2015) worked on the vibrational
characteristics of open and capped- end single-walled CNTs with the help of
multiscale analysis technique thar incorporates Tersoff-Brenner potential. They
have also demonstrated the vibrational characteristics of single-walled carbon
nanotubes depending their length, diameter and atomic structure. Wang et al.
(2015) predicted wave propagation in nonlinear curve single-walled carbon
nanotube theoretically. They made a computer model with the combined
effects of geometrical nonlinear and imperfection. The model is useful for
characterizing the mechanical properties of CNTs and their applications.
Xu et al. (2015) investigated the dispersion relation in single-walled
carbon nanotubes. They made studies on the effect of two small scale
parameters on the angular frequency and phase velocity in longitudinal,
torsional and radial directions. Hussain et al. (2017) made investigations on the
vibration analysis of single-walled carbon nanotubes on the on the basis of
Donnell thin shell theory. They used MATLAB to generate the frequencies of
SWCNTs. They found the results to compromise the previously known
molecular dynamics simulations.
Tang et al. (2016) performed studies on an embedded viscoelastic single-
walled carbon nanotube on the basis of nonlocal strain gradient theory. They
observed that the blocking diameter is dependent on damping coefficient, the
nonlocal and strain gradient length scale parameters and also the Winkler
modulus of the surrounding elastic medium. Hussain et al. (2019) analysed the
vibration of zigzag and chiral rotating, functionally graded single-walled CNTs
and studied with Love’s shell theory with wave propagation applications. They
used MATLAB for calculating forward frequencies. Then, they compared the
results produced with the previous literatures with simply supported condition
and found a good coincidence.
Ansari and Hemmatnezhad (2011) studied the difficulties of non-linear
vibrations of MWCNT that are fixed to an elastic medium. The layers have been
made to be in intact with each other by van der Waals forces. To get
amplitude–frequency curves for single walled double walled and triple walled
carbon nanotubes variation iteration method has been employed. They
investigated the changes in material constant around the elastic medium that
influences the vibrational characteristics of multi walled carbon nanotubes.
Their results readily agreed with the solutions that are available from the open
literature. The current studies capability has been clarified numerically and
also with computational efficiency. Deng and Yang (2014) made studies on
vibrational characteristics of multiwalled CNTs that are filled with fluid based
on the model of non-local elastic Flugge shell. Having the scale effects in mind
they formulated governing equations for vibration for N-layer CNTs. They
observed that smaller the frequencies of vibration higher the scale coefficient
and the length, wall number effects and inner radius are negligible. They have
concluded that the number of walls in the nanotube influences the results off
cut-off frequencies.
Li et al. (2016) derived the governing equation in a fluid-conveying
single-walled carbon nanotube on the basis of nonlocal strain gradient theory
and the Kelvin-Voigt viscoelastic model. They observed that the damping
coefficient’s effect on the phase velocity of both up and down stream waves to
be negligible at low wave numbers but also remarkable at high numbers.
Janghorban and Zare (2011) made analysis of free vibration on carbon
nanotubes on the basis of Timoshenko beam model with different thicknesses.
They assumed the properties to be in longitudinal direction. They have used
differential quadrature method to obtain the equations of motion. The results
obtained could be used as a benchmark for future works.
Ehteshami and Hajabasi (2012) investigated vibrational characteristics of
multiwalled carbon nanotubes that have modelled on the basis of nonlocal
Euler beams. The nanotubes remain intact with each other by means of van
der Waals force. They have obtained governing equations by means of
analytical approach. They have used various boundary conditions. When the
nonlocal parameters are introduced the natural frequency and the inter-tube
frequencies decrease. Inter tube frequencies showed less reduction when
compared to natural frequencies. At higher mode numbers, the natural
frequencies have a tendency to frequencies of constituent beams for multiple
classical Euler beam systems. Wang et al. (2015) studied terahertz wave
propagation under temperature and magnetic fields in a fluid-conveying single-
walled carbon nanotube. With the help of wave analysis, they obtained the
equation of dispersion characteristics of wave propagation. They confirmed
the numerical results obtained with the validity of the present model.
Zhen et al. (2017) studied transverse wave propagation in fluid-conveying
viscoelastic single-walled carbon nanotubes based on elastic theory considered
with surface effect. They noticed that the wave frequency increases as the fluid
flow velocity increases but the same decreases as the tube diameter and wave
number increases. They also observed that in case of larger wave numbers and
nonlocal parameter, the real part of frequency ratio raises. Li et al. (2019)
made experimental studies and perceived that the Young’s modulus of CNTs
increase sharply with the decrease of tube diameter. They used Hamilton
principle to derive boundary conditions and equations of equilibrium. With the
help of the laws of thermodynamics, they obtained the constitutive relations.
Yu et al. (2017) created a dynamic model of Euler-Bernoulli beams for fluid-
filled single-walled carbon nanotube was created based on high-order
nonlogical strain gradient theory and slip boundary conditions of nano-scale
fluid. They found that the behaviours of wave propagation to be damped at
high fluid velocity as the energy transmission is damped by scale effects.
Ugur Guven (2014) performed analytical studies with presumed stress
initially on single walled CNTs in magnetic field for transverse vibrations. They
modelled the nanotube similar to Timoshenko beam model. In the result they
obtained solutions for combined strain/inertia as well as stress/strain gradient
theories of elasticity. They made numerical analysis for single walled carbon
nanotubes with boundary conditions as simply supported. The results they
obtained on propagation of wave in carbon nanotubes might play a major role
in designing and applications in nano-electric mechanical systems. Ansari et al.
(2012) investigated the vibrational characteristics of single walled and double
walled carbon nanotubes under different boundary conditions for different
lengths. They have used Molecular Dynamics simulation on the basis of
Tersoff-Brenner and Lennard-Jones potential energy functions. They have also
considered effects such as tensile and compressive strains initially. The results
show that natural frequency depends upon boundary conditions particularly
with shorter length tubes. The natural frequency decreases with increase in
tube length. Natural frequencies are also highly sensitive to tensile strains and
compressive strains. Their results give information on the calibration of non-
local models for small-scale parameters of CNTs.
Yang et al. (2018) made a dynamic Timoshenko beam model on the basis
of the new nonlocal strain gradient theory and slip boundary theory and
studied the wave propagation behaviour of fluid-filled carbon nanotubes at
nanoscale. They found that the fluid’s flow velocity inside the carbon
nanotubes increases due to the slip boundary effect, thus resulting in the
promotion of wave propagation in dynamic system. Hu et al. (2102)
investigated vibrations of SWCNT and performed studies using Molecular
Dynamic simulation, non-local rod and beam models. The non-local models are
vital in studying the vibrational characteristics of carbon nanotubes. They have
summarised the results of the both non-local beam models on the basis of MD
simulation.
Zidour et al. (2012) developed a non-elastic Timoshenko beam model for
analysis of free vibration of zigzag single-walled CNT on the basis of nonlocal
theory of thermal elastic mechanics. They also considered the thermal effect.
They have used Eringen’s nonlocal constitutive equations for formulations.
Appropriate Young’s modulus and shear modulus values have been used for
the zig zag single walled carbon nanotubes. They found that the natural
frequency depends on the change of temperature and also the chirality of
zigzag SWCNT. Their results are vital for designing of devices that has CNTs.
Yang et al. (2019) used Timoshenko’s beam model to study the wave
propagation in a fluid conveying carbon nanotube with the help of nonlocal
stress-strain gradient coupled theory and nonlocal fluid theory. They then
derived the equations for the motion of CNTs. They compared the numerical
solution and validated with Monte Carlo simulation and interval analysis
method. Zhang et al. (2014) made the study of vibration analysis of CNT
carrying fluid by means of wave method. The carbon nanotube’s constitutive
relation was found with the help of the theory of nonlocal elasticity. They have
also obtained the governing equations on micro scale. They used wave method
to find different speeds of first 5 orders of natural frequencies of the CNTs
carrying fluid. They have concluded with making a difference in the results
between the approach of wave method and other previous studies.
Pang et al. (2015) made a viscoelastic single walled carbon nanotube
adhered by surface material on the basis of nonlocal elastic theory and Kelvin
model and formulated the general governing equation of transverse wave
motion. They also studied the properties of wave propagation. They finally,
concluded that the small scale and surface effects and the impact of structural
damping on the properties of transverse wave propagation are dependent on
the tube diameter and wave number. Oveissi and Ghassemi (2018) studied and
investigated longitudinal and transverse wave propagation of stationary and
axially moving CNTs with and without fluid flowing inside. They used nano-rod
model for longitudinal vibration and Euler-Bernoulli model for transverse
vibrations to obtain the model of stationary CNT. They reported that the phse
velocity decreases with the increase in wave number on both longitudinal and
transverse cases.
Bisheh and Wu (2019) solved wave propagation problem in smart
laminated CNT that is of reinforced composite cylindrical shells which is was
coupled on top with piezoelectric layers while the bottom in hygrothermal
environment for the first time. They observed that the dispersion solutions of
smart laminated CNT-reinforced composite cells were moderately influenced
by temperature/moisture. Safaei et al. (2019) observed that adding of carbon
nanotubes in small amounts into face sheets of sandwich structurers improve
their thermo-mechanical responses. They concluded with the fact that
distribution and cluster size of carbon nanotubes had their effects on speed
and amplitude of thermoelastic stress propagation of waves in nanocomposite
sandwich plates.
Gul and Aydogdu (2017) studied flexural and axial wave propagation in
double-walled CNTs embedded in an elastic medium and axial wave
propagation in single walled CNTs. They found that the doublet mechanics
gave exact results for both flexural and axial wave propagation in nanotubes.
Huang et al. (2013) analysed the effects of flexural waves in CNTs. They
proposed an exact higher- order model to analyse dynamic behaviour of
nonlocal elastic beams that has circular cross section. They have considered
both shear deformation and rotary inertia without having introduced the
factor of shear correction. With the help of this model they obtained
dispersion of wave and the relation of dispersion for both single and double
walled carbon nanotubes respectively. They have also explained the scale
effects of coefficient, elastic medium that is being surrounded and the force of
van der Waals on phase velocity.
Haynayeb and Khadem(2012) performed studies on double walled nanotubes
with clamped-clamped conditions for forced vibrations. They used Euler
Bernoulli beam models for both inner and outer layers of CNTs. They applied
electrostatic actuation with both AC and DC voltages between the CNTs and
the electrode. They found that changing DC voltage’s vale brought changes in
the behaviour of the double walled carbon nanotubes. And also under primary
and secondary condition, both the layers of double walled carbon nanotubes
vibrate with the similar frequencies. Arda and Aydogdu (2016) investigated the
torsional wave propagation in multiwalled carbon nanotubes. They attained
the governing equation of motion of multiwalled carbon nanotube with the
help of Eringen’s nonlocal elasticity theory. They reported that the group and
escape velocity of waves converged to a limit value in nonlocal elasticity
approach.
Bisheh and Wu (2018) studied and analysed the propagation of waves in
piezoelectric cylindrical composite shell reinforced with CNTs with the help of
Mori-Tanaka micromechanical model and for the first time, considered rotary
inertia and transverse shear effects through the first order shear deformation
shell theory. They also examined the effects of CNTs and shell geometry
parameters on dispersion solutions. Yas and Samadi (2012) performed studies
on free vibrations and analysis of buckling of nanocomposite beams
strengthened by single walled carbon nanotubes that is placed on elastic
medium. They imagined the single walled carbon nanotubes to be in perfectly
aligned. They performed analysis for natural frequencies and obtained results
for different conditions of boundaries. They also studied the effects of fraction
of volume of nanotube, stiffness of foundation, ratio of slenderness, carbon
nanotubes distribution and boundary conditions for natural frequency as well
all critical buckling load. They concluded with results indicating that the
parameters play a vital role on free vibration and characteristics of bucking.
 Bisheh and Wu (2019) investigated analytically the straight carbon
nanotubes that are angled and randomly oriented with the help of first order
shear deformation shell theory along with transverse shear effects and rotary
inertia for the first time. They used Mori-Tanaka method for mechanical
modelling. They observed that the different orientations of the carbon
nanotubes show different behaviour of dispersion. They also reported that the
volume fraction and shell geometry effects of CNTs to be more obvious at high
circumferential wave numbers. Ansari et al (2012) studied characteristics of
vibration of SWCNT on the basis of gradient elasticity theories. They used
Euler-Bernoulli and the Timoshenko theories for theoretical formulations. They
observed that the strain gradient elasticity is more reliable for shorter lengths
of single walled carbon nanotubes. They have also mentioned that Timoshenko
beam model’s results are close to the results of Molecular Dynamics
simulations.
Eltaher et al. (2019) investigated modal participation of SWCNTs with
fixed-fixed boundary conditions with vacancies. The bonding between the
atoms is modelled with the help of 3D beam element and circular cross section
is used. They implemented the model of energy equivalent and observed the
relationship between the energy stored in atomic chemical bonding and
potential energy of the mechanical beam structure. Metin Aydogdu (2013)
performed studies on the axial vibration characteristics of SWCNT on the basis
of non-local elastic theory that is fixed in elastic medium. He has used
Eringen’s nonlocal constitutive equations. Then other parameters such as
elastic medium’s stiffness, different boundary conditions and also other non-
local parameters are discussed. It is observed that the axial frequencies of
vibration are so high that ignored the scale of small length.
Eltaher et al. (2019) again studied the characterization and behaviours of
SWCNT with the approach of equivalent-continuum mechanics. They evaluated
Young’s modulus of SWCNTs with the help of tensile test at any angle of
orientation. They also made use of infinite element simulation to study
buckling and natural frequencies of single walled carbon nanotubes. Wang et
al. (2013) investigated the propagation of wave in SWCNTs carrying fluid and
also set in various physical fields that includes magnetic as well as temperature
fields. The modelled carbon nanotubes similar to Timoshenko beams. They
performed various numerical simulations to verify the developed model and
also to understand the effects of nonlocal parameters such as the velocity of
fluid, density of flow, temperature, change in magnetic field flux and the elastic
medium surrounded. They observed that the nonlocal parameters showed
influence on the nanotube and the fluid inside. And also the physical effects
such as magnetic and temperature fields showed vital role in the single walled
carbon nanotube’s propagation of wave.
Bocko and Lengvarsky (2014) investigated the free bending vibrations of
SWCNTs. The nonlocal theory of beam bending is used for computation of
frequencies. They obtained solution for 4 different boundary conditions
namely: clamped-free, simply-simply supported, clamped-simply supported
and clamped-clamped. The cases that characterises the influence of
parameters are given in terms of graphical outputs. Metin Aydogdu (2015)
studied multiwalled CNTs on the basis of longitudinal wave propagation with
the help of the theory of non-local elasticity. The interpretation suggested a
model of multiwalled rod. They considered the forces of can der Waals to be in
the axial direction. The conclusions they made will be useful for designing
oscillators, motors and also electro-mechanical systems in nanoscale.

Research Motivation
Designing the perfect shape and structure of CNT using simulation
software is often challenging. This is because the design process of CNT in
simulation software are not always straightway and also because the codes
generated for modelling CNTs in the literature are often kept secret from the
viewers (Saini, 2012). Therefore, this project aims on creating a general
method for creating computational CNT and also vibration analysis to be
performed. It is noted that there is a lot of analyses on taking the chirality,
length and diameter of the tube as parameter but a very few on the wave
numbers and diameters as different parameters. Carbon-Carbon bonding is
defined in different sections like circular, rectangular and even irregular ones.

Research Objectives
The objective of this project is to use ANSYS software package along with
Nanotube Modeler and Wing Pro Python, to develop a general method to
model computational carbon nanotubes. The scale effect on the carbon
nanotube’s wave propagation will be investigated for different aspects and
boundary conditions with the help of numerical simulations.
Significance of Current Research
The properties exhibited by the CNTs that make it out stand for the
development of new applications are their nearly perfect structure of
geometry, higher strength, lower density of mass and their nano scaled
structure. They are better conductors of electricity than the copper itself and
better transmitter of heat than the diamonds. In recent years, there have been
many reports from studies reporting the superiority of nonlocal models to
their local counter parts for modelling the vibration of carbon nanotubes.
Therefore, this current project aims at developing a general method to create
the nanotube and perform numerical with different parameters and boundary
conditions. Hopefully, this project would pave wave for many other future
research analyses on the scale effect of CNTs.

Methodology
Many analyses and researches have been done on carbon nanotubes with
varying structure, tube length and tube diameter of the tubes with different
boundary conditions have been done to study the vibrational analysis of CNTs.
The methodology mainly consists of modelling and simulation. The software
used for simulation and analyzation is the Ansys Mechanical APDL 2019 R3.
Nanotube modeller 1.8.0(below) is used to obtaining the geometry of the CNT
as a protein data bank(.pdb) file.
Finite Element Analysis
Carbon nanotubes are made up of carbon atoms that are bonded by means of
covalent bonds thus forming hexagonal lattices. Every atom has been displaced
by means of external forces that are constrained by means of bonds. So, the
total deformation brought to the nanotube is nothing but the outcome of
interaction between the bonds. Those bonds are elements that carry loads that
connect the atoms. Carbon nanotubes are created as space-frame models. The
figure below represents carbon atoms connected via covalent bonds forming
hexagonal lattices.
 Figure 8- Carbon atoms connected with covalent bonds(Image: www.understandingnano.com)

Modelling
The main objective is to model an accurate geometry of carbon anotube
in Ansys software. Therefore the model is created with the help of nanotube
modeller first and then the obtained .pdb files are converted to Ansys pre-
processor script. Then the script is given as input to the Ansys.
Modelling with Nanotube modeller
The version of version of nanotube modeller used is Nanotube modeller
1.8.0. It is a professional software that gives the coordinates of the nanotubes.
The input parameters are shown in the table below.
Table 1-Input parameters

Here the chirality i.e. (n, m), tube length, bond length, number of walls
and diameter of the tube are defined. The figure below gives the general
layout of the Nanotube Modeller.
Figure 9-Layout of the Modelling application

The model from the modeller cannot be imported directly into the
Ansys. So the information of the coordinates are saved as .pdb format as
shown in the figure below which needs to be in format that Ansys can read.
Figure 10-Output from the Nanotube Modeller in .pdb format

Generating Ansys pre-processor script

A Python script is executed (below) that converts .pdb file to pre-
processing script for Ansys. A software called Wing IDE is used that acts as a
compiler for executing the Ansys script for python.
 Figure 11-Wing IDE for convertiing .pdb file to Ansys script

The obtained Ansys pre-processor script is shown in the figure below. K stands
for keypoints or atoms and L stands for bonding lines. Ansys mechanical APDL
gives the geometry of the nanotube once the script is given as input.

Figure 12- Ansys pre-processor script

Preprocessor
Wireframe model
Once the script has been given as input, the wire frame model has been
generated as shown in the below.
Scaling Factor
For beam structure, the finite element analysis is insensitive to the very small
scale of dimensions such as nanometer. So, some adjustments have been
made converting the nano-scaled units to micro-scaled units. The values for
the analysis have been given in after applying the scaling factors. The table
below shows the scaling factor that has been implemented for analysis. (M. A.
Eltaher et al, 2019)
Scaling factor for the finite-element model software
Divisions Units Actual scale Given scale
Frequency Hertz f 10-8 E
Length m L 1010 L
Density Kg/m3 ρ 10-4 ρ
Young’s modulus Pascal E E

Material properties
Then we input the appropriate material properties for the carbon
nanotubes with scaling factor. The carbon nanotubes are considered as
 linear elastic material. The figures below show the Ansys window of
material properties

Figure 13-Material properties window

Beam Section
The beam tool is used to create cross section of the beams. There are
some predefined shapes available such as circular, rectangular, I-Section, etc.
Then the offset of the section for the required cross section is also selected.
 Centroid – Node of the beam will be offset to the centroid which is
default
 Shear Cen - Node of the beam will be offset to the center of the
shear
 Origin - Node of the beam will be offset to the cross-section’s
origin
 Location - Node of the beam will be offset to the specified location
We have assumed the carbon nanotube to have circular cross
section(below). And we enter the appropriate values.
R = Radius of the circular cross section
N = Total no. of divisions along the circumference of the
circular cross section
T = Total no. of divisions along the radius of the circular
cross section
 Figure 14- Beam tool Ansys Window

With the preview button, the Ansys gives the preview of the cross
section with the type and dimensions specified as demonstrated in the figure
below.
Figure 15-Preview of the cross section

We can also get the preview of the mesh as shown in the figure below
Figure 16-Mesh view of the cross section

List section gives the summary of data of the cross section that has been
applied to the Ansys database. It is explained in the figure below.

Figure 17- List section Ansys window

Meshing
There are 4 types of manual size meshing:
  Global meshing
 Areas meshing
 Lines meshing
 Keypoints meching
 Layers meshing
We go for Lines meshing and select ‘All Lines’. There we give the size and
number of divisions. The element size window of Ansys is shown in the figure
below.

Figure 18-Element sizes window

Then we go for Main Menu>Preprocessor>Meshing>Mesh>Lines. Then a

window appears as shown in the figure below.
 Figure 19- Mesh lines window

We can mesh by selecting one line at a time, as box, as polygon, as circle, as

loop or by pick all lines. We select pick all and select okay to have the mesh to
be generated as shown in the figure below.
Figure 20- Carbon nanotube after being meshed

And once the mesh has been generated, we can start applying boundary
condition.

Solution
The type of analysis, boundary conditions and loading conditions are
given. There are 7 types of analysis available(shown in figure below):
 Static – Done for static analysis - For all DOF
 Modal – Done for modal analysis – For structural and fluid DOF
 Harmonic – Done for harmonic analysis – For structural, magnetic,
fluid and electrical DOF
 Transient – Done for transient analysis – For all DOF
 Spectrum – Done for spectrum analysis – For structural DOF only
 Eigen Buckling – Done for buckling analysis – for structural DOF
only
 Substructuring/CMS – Done for substructure analysis – for all DOF
Figure 21- Types of Analysis window

For modal analysis different extraction methods are available:

 Block Lanczos
 PCG Lanczos
 Supernode
 Subspace
 Unsymmetric
 Damped
 QR Damped
We go for Block Lanczos. And the number of mode shapes to extract and
the number of modes to expand are specified (demonstrated in the figure
below).
 Figure 22- Modal analysis window

Then we define the type of load and boundary conditions. There are different
types of loads available:
 Displacement
 Force/Moment
 Pressure
 Temperature
 Inertia
In this project we have use displacement and force/moment. And these loads
can be applied on lines, on areas, on keypoints, on nodes or on node
components. We select key points as box and specify the constraint as shown
in the figure below. The value of displacement is specified and also the degrees
of freedom to be constrained.
Figure 23-Applying structural displacement

Then for Force/Moment we define the type of load and boundary

conditions. The force can be applied on keypoints, on nodes or on node
components. Then we specify the direction, distribution and also the value of
force/moment to be applied (as shown below).

Figure 24- Applying force/moment

Once the loading and boundary conditions are specified, we solve. We solve
for current load step(below).
Figure 25-Solving the current load step

General Post Processing

Once solved the solution part is done. We can get the Results Summary
from the General Postprocessing. We pick a result from read results to get a
plot
And for plotting the result five different plots are available:
 Contour Plot
 Vector Plot
 Plot path item
 Concrete plot
 Thin film
We select Contour plot and under contour plot we go for Nodal solution.
Nodal solution>DOF Solution>Displacement vector sum
Figure 26- Contour nodal solution data

After selecting displacement vector sum we obtain the nodal solution as

shown in the sample figure below.
Figure 27-Sample of nodal solution

Similarly, the analysis is done for the required geometries of carbon

nanotube, material properties, loading and boundary conditions.

Results and discussion

1. Comparison Studies
i) Open-Ended CNTs
The comparison studies have been made with D. Karlicic et al. (2015).
Armchair typed carbon nanotubes with chirality (8,8) for different lengths to
diameter ratios have been obtained from the nanotube modeller and
converted to preprocessor script with the help of Python. Once the script is
entered in the Ansys MAPDL, for preprocessing, the element type is defined as
Beam 188. This type of element doesn’t require and real constants in the
MAPDL. Then the material properties are given, thickness (h = 0.34nm),
Poisson’s ratio (v= 0.3), mass density ( ρ = 2300 kg/m3), Young’s modulus (E =
1.1 TPa). Then, the meshing of the lines is made. Then moving to the solution
part, the type of analysis done is Modal. Then in Analysis options, Block
Lanczos method has been selected for the required number of modes to
extract. The ends of the carbon nanotubes are fixed with zero displacements.
Then solved and the results obtained are compared with (Ansari et. al., 2012).
The table below demonstrates the fundamental frequency of (8,8) armchair
nanotubes.

L/ Euler-Bernoulli beam Timoshenko beam theory Prese Perce

D theory (Ansari et al., 2012) (Ansari et al., 2012) nt nt
resul error
MD Classi Stress MD Classi Stress
ts [%]*
simula cal Strain/in simula cal Strain/in
tion theor ertia tion theor ertia
y gradient y gradient
theory theory
8.3 0.5299 0.538 0.5300 0.5299 0.530 0.5299 0.516 2.54
0 6 4
10. 0.3618 0.364 0.3615 0.3618 0.360 0.3604 0.354 2.13
1 0 6 1
13. 0.1931 0.198 0..1978 0.1931 0.197 0.1971 0.198 2.74
7 2 2 4
17. 0.1103 0.124 0.1234 0.1103 0.124 0.1240 0.123 11.6
3 4 0 1
20. 0.0724 0.085 0.0852 0.0724 0.085 0.0851 0.086 19.3
9 3 1 4
24. 0.0519 0.062 0.0621 0.0519 0.062 0.0620 0.062 19.8
5 1 0 2
28. 0.0425 0.047 0.0472 0.0425 0.047 0.0471 0.047 11.3
1 2 1 3
31. 0.0358 0.037 0.0373 0.0358 0.037 0.0373 0.037 5.02
6 3 3 6
35. 0.0287 0.029 0.0299 0.0287 0.029 0.0299 0.030 4.87
3 9 9 1
39. 0.0259 0.024 0.0244 0.0259 0.024 0.0244 0.024 5.40
1 4 4 5
*
Percent error is calculated with respect to the MD simulation.
The formula used for finding the percent error is:
¿
Percent error = ¿ V true−V observed ∨ V true ×100 ¿

It is observed that the results obtained from Molecular Dynamics Simulation

correlate with the finite element analysis with a maximum difference up to
19.8% for length-diameter ratio 28.1. But percent error is 1.5% for Classical
theory. The figures below show the fundamental frequencies of the CNTs of
different length to diameter ratios.

Figure 28-Mode shapes of the fundamental frequencies of CNTs with different L/D ratios
 From the results obtained, it is observed that the method used for
analysis readily agrees with the MD simulation results of both Euler-Bernoulli
beam theory and Timoshenko beam theory. Therefore this method of
approach is observed to be reliable. It is observed that the frequency
decreases with the increase in L/D aspect ratios.
ii) Capped CNTs
Then the comparison studies have been made with (Gajbhiye SO, 2015).
Caped ended armchair typed carbon nanotubes with chirality (5,5) has been
generated from the nanotube modeller and converted to preprocessor script
with the help of Python. Once the script is entered in the Ansys MAPDL, for
preprocessing, the element type is defined as Beam element. This type of
element doesn’t require and real constants in the MAPDL. Then the material
properties are given, thickness (h = 0.34nm), Poisson’s ratio (v= 0.28), mass
density ( ρ = 1.9 kg/m3), Young’s modulus (E = 21.607 TPa). Then, the meshing
of the lines is made. Then in the solution part, the type of analysis done is
Modal. The ends of the carbon nanotubes are fixed with zero displacements.
Then in Analysis options, Block Lanczos method has been selected for the
required number of modes to extract. Then solved and the results obtained
are compared with (Gajbhiye SO, 2015). The table below demonstrates the
fundamental frequency of (5,5) caped ended armchair nanotubes.
Comparison of the first 7 frequency values f(THz)

Frequency Gajbhiy Obtained Percent

e& result error
Singh
1 0.046 0.044 4.35
2 0.046 0.044 4.35
3 0.261 0.295 13.02
4 0.261 0.295 13.02
5 0.347 0.368 6.05
6 0.597 0.551 7.71
7 0.646 0.552 14.55
The formula used for finding the percent error is:
¿
Percent error = ¿ V true−V observed ∨ V true ×100 ¿

The figure below shows the fundamental frequency of the CNT. From
the table above it is observed that the results produced by Ansys agree with
(Gajbhiye SO, 2015).
The maximum difference of 14.55% has been obtained, this is because
(Gajbhiye SO, 2015) has used elastic beam element with lumped carbon
masses at the ends. And also they have used beam elements with mass
elements while the current model uses only beam elements. The results
produced from the analysis agrees with (M. A. Eltaher et al, 2019). Therefore
this method of approach is observed to be reliable.

Figure 29-Mode shapes of (5,5) capped CNT with length L=69.29 Å, fixed at both ends.

2. Numerical Examples

i) CNT under initial compression stress

A system that has been analyzed in its unsteady state is called transient
analysis. To study CNT under initial compression stress, a carbon nanotube
with chiralities (7,7) of length 40 Å has been modelled with the help of with Nanotube
Modeller and converted to preprocessor script with the help of Python. Once the script is
entered in the Ansys MAPDL, for preprocessing, the element type is defined as Beam 188.
This type of element doesn’t require and real constants in the MAPDL. Then the material
properties are given, thickness (h = 0.34nm), Poisson’s ratio (v= 0.3), mass density ( ρ = 2300
kg/m3), Young’s modulus (E = 1.1 TPa). We use the scaling factor for the FEA. Then the lines
have been meshed. Once it is done, Transient type of analysis is selected. The solution is
done in 3 steps. First the constraining boundary conditions are given. Then, 3 LS files are
written. First with time step 0.001 and write it as Load step 1. For the 2 nd LS file we apply
forces of 100N on both ends and give 0.001 as the time at the end of the load step. Then
write the LS file as 2. Then for the last LS file we delete the forces and also enter the time at
the end of load step as 1. And save the LS as 3. Once solving load steps from 1-3, we obtain
the solution. In time history post processor, we obtain the graph that is shown below.

Figure 30-Value/Frequency chart

ii) Effect of CNT lengths on natural frequencies

To determine the effects of CNT lengths on natural frequencies, three
chiralities of nanotubes were selected namely (5,5), (8,8) & (10,10). And for
each chirality, three lengths were selected – 30Å, 50Å and 70Å. Each model is
generated with Nanotube Modeller and converted to preprocessor script with
the help of Python. Once the script is entered in the Ansys MAPDL, for
preprocessing, the element type is defined as Beam 188. This type of element
doesn’t require and real constants in the MAPDL. Then the material properties
are given, thickness (h = 0.34nm), Poisson’s ratio (v= 0.3), mass density ( ρ =
2300 kg/m3), Young’s modulus (E = 1.1 TPa). We use the scaling factor for the
FEA. Then, the meshing of the lines is made. Then in the solution part, the type
of analysis done is Modal. Then in Analysis options, Block Lanczos method has
been selected for the required number of modes to extract. Then the
frequencies obtained from each model has been plotted into a graph and
compared.
The figures below show the fundamental frequencies for three lengths
30Å, 50Å & 70Å of (5,5) armchair nanotubes. The table below demonstrates
the first 7 fundamental frequencies of 3 different lengths of CNTs.
 Frequency Frequency (THz) Frequency (THz) l
Set (THz) l=30 l=50 = 70
1 0.8811 0.3934 0.2122
2 1.4406 0.8511 0.5224
3 1.9058 0.9163 0.5933
4 2.1542 1.2876 0.9016
5 2.8996 1.5559 0.9171
6 3.147 1.7067 1.1882
7 3.5101 2.2637 1.365

l=30Å l=50 Å l=70 Å

The graphs below show that shorter the length, higher the frequency is
in carbon nanotubes. Therefore, products with less frequency are highly
subjected to buckling. The frequency obtained from the analysis is tabulated
below.
The figures below show the fundamental frequencies for three lengths 30Å,
50Å & 70Å of (8,8) armchair nanotubes. It is observed from the graph below
that shorter the length, higher the frequency is in the models.
Mode Frequency (THz) Frequency (THz)
no. l=30 l=50 Frequency (THz) l=70
1 0.9669 0.4808 0.2795
2 1.3941 0.8214 0.5719
3 1.7888 1.0307 0.6324
4 1.9852 1.2867 0.9014
5 2.1472 1.6258 1.0544
6 2.426 1.6485 1.146
7 2.8114 1.6828 1.5098
l=30Å l=50 Å l=70 Å
The figures below show the fundamental frequencies for three lengths 30Å,
50Å & 70Å of (10,10) armchair nanotubes. From the graph below it is observed
that shorter the length, higher the frequency is in the models.
Model Frequency (THZ) l=30 Frequency (THZ) Frequency (THZ) l =
no l = 50 70
1 0.9868 0.6491 0.31
2 1.3548 0.7874 0.5666
3 1.3824 1.0578 0.673
4 1.9958 1.0894 0.9009
5 2.1062 1.2187 1.0867
6 2.1339 1.2406 1.0994
7 2.7602 1.2491 1.1354

L=30Å l=50 Å l=70 Å

 The figures above demonstrate the fundamental frequency of (10,10)
single walled carbon nanotubes of length 30Å, 50Å & 70Å
Conclusion
A general method to create the nanotube and perform numerical with
different parameters and boundary conditions. The following conclusion and
observations have been made:
• The developed method for creating computational carbon nanotube
readily agrees with the results from other models namely Euler-Bernoulli
and Timoshenko beam models.
• From the open-ended model it is observed that frequency decreases
with an increase in L/D ratio.
• And the results from the Capped CNTs obliges with the previous model
of Gajbhiye and Singh(2015).
• It is observed that for CNTs under compression the frequency increases
once the load is applied.
• And also length plays a vital role, shorter the length higher the
frequency and larger the diameter higher the frequency.
References
Ansari, R., & Hemmatnezhad, M. (2011). Nonlinear vibrations of embedded
multi-walled carbon nanotubes using a variational approach.
Mathematical and Computer Modelling.
Ansari, R., Ajori, S., & Arash, B. (2012). Vibrations of single- and double-walled
carbon nanotubes with layerwise boundary conditions: A molecular
dynamics study. Current Applied Physics.
Ansari, R., Gholami, R., & Rouhi, H. (2012). Vibration analysis of single-walled
carbon nanotubes using different gradient elasticity theories.
Composites Part B: Engineering.
Arda, M. A. (2016). Torsional wave propagation in multiwalled carbon
nanotubes using nonlocal elasticity. Applied Physics A: Materials
Sciences and Processing, 1-10.
Aydogdu, M. (2012). Axial vibration analysis of nanorods (carbon nanotubes)
embedded in an elastic medium using nonlocal elasticity. Mechanics
Research Communications.
Aydogdu, M. (2014). Longitudinal wave propagation in multiwalled carbon
nanotubes. Composite Structures.
Azonano. (n.d.). A History of Carbon Nanotubes. Retrieved from Azonano:
https://www.azonano.com/article.aspx?ArticleID=4843
Bisheh, H. W. (2018). Analysis of wave propagation characteristics in
piezoelectric cylindrical composite shells reinforced with carbon
nanotubes. International Journal of Mechanical Sciences, 200-220.
Bisheh, H. W. (2019). Wave propagation in piezoelectric cylindrical composite
shells reinforced with angles and randomly oriented carbon nanotubes.
Composites Part B: Engineering, 10-30.
Bisheh, H. W. (2019). Wave propagation in smart laminated composite
cylindrical shells reinforced with carbon nanotubes in hygrothermal
environments. Composites Part B: Engineering, 219–241.
Bocko, J., & Lengvarský, P. (2014). Bending Vibrations of Carbon Nanotubes by
Using Nonlocal Theory. Procedia Engineering.
Deng, Q., & Yangb, Z. (2014). Vibration of fluid-filled multi-walled carbon
nanotubes seen via nonlocal elasticity theory. Acta Mechanica Solida
Sinica.
Ehteshami, H., & Hajabasi, M. A. (2011). Analytical approaches for vibration
analysis of multi-walled carbon nanotubes modeled as multiple nonlocal
Euler beams. Physica E: Low-dimensional Systems and Nanostructures.
Gajbhiye SO, S. S. (2015). Vibration characteristics of open-end and capped-
end single walled carbon nanotubes using multi-scale analysis technique
incorporating Tersoff-Brenner potential. Acta Mech.
Gajbhiye, S. O., & Singh, S. P. (2015). Vibration characteristics of open- and
capped-end single-walled carbon nanotubes using multi-scale analysis
technique incorporating Tersoff–Brenner potential. Acta Mech .
Gul, U. A. (2017). Wave Propagation in double walled carbon nanotubes by
doublet mechanics theory. Physica E: Low-Dimensional Systems and
Nanostructure, 345-357.
Guven, U. (2014). Transverse vibrations of single-walled carbon nanotubes
with initial stress under magnetic field. Composite Structures.
Hajnayeb, A., & Khadem, S. E. (2012). Nonlinear vibration and stability analysis
of a double-walled carbon nanotube under electrostatic actuation.
Journal of Sound and Vibration.
Hu, Y.-G., Liew, K. M., & Wang, Q. (2012). Modeling of vibrations of carbon
nanotubes. Procedia Engineering.
Huang, Y., Luo, Q.-Z., & Li, X.-F. (2013). Transverse waves propagating in carbon
nanotubes via a higher-order nonlocal beam model. Composite
Structures.
Hussain, M. N. (2017). Vibrational behavior of single-walled carbon nanotubes
based on cylindrical shell model using wave propagation approach. AIP
Advances.
Hussain, M. N. (2019). Rotating response on the vibrations of functionally
graded zigzag and chiral single walled carbon nanotube. Applied
Mathematical Modelling, 506 -520.
Li, C. G. (2019). Nonlocal second-order strain gradient elasticity model and its
application in wave propagating in carbon nanotubes. Microsystem
 Technologies, 2215-2227.
Li, L. H. (2016). Wave propagation in fluid-conveying viscoelastic carbon
nanotubes based on nonlocal strain gradient theory. Computational
Materials Science, 282-288.
M. A. Eltaher et al. (2019). Modal Participation of fixed-fixed single-walled
carbon nanotube with vacncies. International journal of Advanced
Structural Engineering.
M. A. Eltaher, T. A. Almalki, K. H. Almitani, K. I. E. Ahmed, & A. M. Abdraboh.
(2019). Modal participation of fixed–fixed single-walled carbon nanotube
with vacancies. International Journal of Advanced Structural Engineering
, 151–163.
MaziarJanghorban, & Zare, A. (2011). Free vibration analysis of functionally
graded carbon nanotubes with variable thickness by differential
quadrature method. Physica E: Low-dimensional Systems and
Nanostructures.
Meeks-Matous, J. (2018, February). Who is the world's largest producer of
carbon nanotubes? Retrieved from Quora:
https://www.quora.com/Who-is-the-worlds-largest-producer-of-carbon-
nanotubes
Mohamed A. Eltaher, T. A. (2019). Characterization and behaviors of single
walled carbon nanotube by equivalent-continuum mechanics approach .
Advances in Nano Research, 39-49.
Oveissi, S. G. (2018). Longitudinal and transverse wave propagation analysis of
stationary and axially moving carbon nanotubes conveying nanofluid.
Applied Mathematical Modelling, 460-477.
Pang, M. Z. (2015). Transverse wave propagation in viscoelastic single-walled
carbon nanotubes with small scale and surface effects. Journal of
Applied Physics.
Ren, G. (n.d.). Carbon Nanotube. Retrieved from Britannica:
https://www.britannica.com/science/carbon-nanotube
Safaei, B. M.-D. (2019). Determination of thermoelastic stress wave
propagation in nanocomposite sandwich plates reinforced by clusters of
carbon nanotubes. Journal of Sandwich Structures and Materials.
Saini, D. (2012). Vibration Analysis of Carbon Nanotube through Finite Element
Method. Patiala.
Tang, Y. L. (2016). Viscoelastic wave propagation in viscoelastic single walled
carbon nanotubes based on nonlocal strain gradient theory. Physica E:
Low-Dimensional Systems and Nanostructures, 202-208.
Wang, B. D. (2015). Terahertz wave propagation in a fluid-conveying single-
walled carbon nanotube with initial stress subjected to temperature and
magnetic fields. Acta Mechanica, 3031-3043.
Wang, B. D. (2015). Wave propagation analysis in nonlinear curved single-
walled carbon nanotubes based on nonlocal elasticity theory. Physica E:
Low-Dimensional Systems and Nanostructures, 283-292.
Wang, B., Deng, Z., Ouyang, H., & Zhang, K. (2013). Wave characteristics of
single-walled fluid-conveying carbon nanotubes subjected to multi-
physical fields. Physica E: Low-dimensional Systems and Nanostructures.
Xu, X.-J. D.-C. (2015). Effects of strain and higher order inertia gradients on
wave propagation in single walled carbon nanotubes. Physica E: Low-
Dimensional Systems and Nanostructures, 101-110.
Yang, Y. W. (2018). Wave Propagation in Fluid-Filled Single-Walled Carbon
Nanotube Based on the Nonlocal Strain Gradient Theory. Acta
Mechanica Solida Sinia, 484-492.
Yang, Y. Y. (2019). Study on the small-scale effect on wave propagation
characteristics of fluid-filled carbon nanotubes based on nonlocal fluid
theory. Advances in Mechanical Engineering.
Yas, M. H., & Samadi, N. (2012). Free vibrations and buckling analysis of carbon
nanotube-reinforced composite Timoshenko beams on elastic
foundation. International Journal of Pressure Vessels and Piping.
Yu, Y. Y. (2017). Zhendong yu Chongji/Journal of Vibration and Shock. Wave
propagation of fluid-filled single walled carbon nanotubes based on the
nonlocal-strain gradient theory, 1-8.
Zhang, Z., Liu, Y., & Li, B. (2014). Free vibration analysis of fluid-conveying
carbon nanotube via wave method. Acta Mechanica Solida Sinica.
Zhen, Y.-X. (2017). Wave propagation in fluid-conveying viscoelastic single-
walled carbon nanotubes with surface and nonlocal effects. Physica E:
 Low-Dimensional Systems and Nanostructures, 275-279.
Zidour, M., Halim, K., Abdelwahed, B., Mokhtar, S., Hichem, N., Belhadj, A., . . .
Tounsi, A. (2012). The thermal effect on vibration of zigzag single walled
carbon nanotubes using nonlocal Timoshenko beam theory.
Computational Materials Science.