1. Introduction
Nanoscale structures/materials have very different characteristics. Nanostructures/nanomaterials have attracted great attention because their extraordinary features, such as high strength, low density, high elasticity modulus, and high hardness [
1,
2,
3,
4], have become the focus of researchers. The aforementioned unique properties of such structures, materials, and rapid developments in nanotechnology has led to use of these structural elements in designing micro- and nanoelectro mechanical systems (MEMS and NEMS) such as resonators, atomic force microscopes, switches, actuators, and sensors.
Some experimental studies have revealed the deformation behaviors of micro-/nanosized structures [
5,
6]. However, experiments are very difficult and quite expensive on these scales because high precision test devices are needed. On the other hand, atomistic modeling such as molecular dynamic simulations is computationally expensive and requires a long period of time. Consequently, this option is limited to structures that have only a few atoms [
7].
To understand and accurately interpret the mechanical properties and behaviors of nanoscale structures, use of models based on continuum mechanics may be a better alternative than experiments and atomistic modelling. Unfortunately, classical continuum theories are not sufficient to predict and estimate size dependency because they lack internal/additional material length scale parameters. In ultrasmall scales (micrometer dimension, nanometer dimension) interactions between atoms and molecules have increasing importance and cannot be neglected. Therefore, the solution to classical continuum theories, which does not take into account size effects, does not give accurate results. In order to obtain more accurate results, higher-order continuum theories such as couple stress theory [
8,
9,
10], modified couple stress theory [
11], strain gradient theory [
12], modified strain gradient theory [
13], and nonlocal elasticity theory [
14] have been developed and contain various length scale parameters. These theories were used by many researchers for various analyses such as buckling [
15], bending [
16,
17], free vibration [
18,
19], forced vibration [
20], and nonlinear vibration [
21].
Rahmanian et al. [
22] presented free vibrations of single-walled carbon nanotubes (SWCNT) on a Winkler elastic foundation via nonlocal elasticity theory. In this study, SWCNT was modeled as both beam and shell structures. Demir and Civalek [
23] reported thermal vibration formulation of a nonlocal Euler–Bernoulli beam embedded in an elastic matrix. Finite element formulation for Eringen’s nonlocal elasticity theory was employed via Hermitian cubic shape functions. Thermal vibrational behaviors of silicon carbide nanowire on an elastic matrix were investigated for simply supported (S-S) and clamped–clamped (C-C) boundary conditions. Finite element formulations of nonlocal elastic Euler–Bernoulli and Timoshenko beam theories were achieved by Pradhan [
24]. Vibration, buckling, and bending analyses of carbon nanotubes with four different boundary conditions were performed by the Galerkin finite element technique. Rajasekaran and Bakhshi Khaniki [
25] reported static deformation, stability, and free vibration responses of small-scale beams. A finite element model of axial, functionally graded, nonuniform small-scale beams was investigated by using nonlocal strain gradient theory. Eltaher et al. [
26] pursued free vibration analysis of functionally graded Euler–Bernoulli nanobeams by using Eringen’s nonlocal elasticity theory. Finite element results were given for a dynamic analysis of the nanobeam. Nejad and Hadi [
27] studied bending analysis of non-homogeneous nanobeams. Eringen’s nonlocal elasticity theory was utilized in Euler–Bernoulli nanobeams made of bi-directional, functionally graded material. Murmu and Pradhan [
28] studied the thermo-mechanical vibration response of embedded carbon nanotubes surrounded by an elastic matrix based on nonlocal elasticity theory. Reddy [
29] developed nonlocal beam models based on four different beam theories. Static bending, free vibration, and buckling analyses of nanobeams are performed in this study. Tornabene et al. [
30] presented a multiscale approach for three-phase carbon nanotube (CNT)/polymer/fiber-laminated nanocomposite structures. Detailed formulations can be found in the literature [
31,
32,
33,
34,
35] about CNT-reinforcement or the finite element method (FEM). More recently, Uzun et al. [
36] investigated the free vibration responses of carbon nanotubes and boron nitride nanotubes based on nonlocal elasticity theory. Nonlocal natural frequencies are obtained for various cross-section geometries.
In the present study, free vibration analysis of three kinds of nanowires resting on a Winkler elastic foundation with various boundary conditions are performed. Simply supported (S-S), clamped simply supported (C-S), clamped–clamped (C-C), and clamped-free (C-F) boundary conditions are selected. Silicon carbide nanowire (SiCNW), silver nanowire (AgNW), and gold nanowire (AuNW) are modeled as nonlocal Euler–Bernoulli beams, and their vibration behaviors are investigated using the finite element method (FEM). A Galerkin weighted residual method is utilized to govern equations and matrices, and the Winkler foundation parameter and small-scale parameter are gained. Effects of boundary conditions, temperature rise, and small-scale and Winkler foundation parameters of frequency values are investigated and compared for three kinds of nanowires.
2. Euler–Bernoulli Nanobeam Resting on a Winkler Elastic Foundation
The nonlocal stress tensor
at point
x is expressed as follows [
14]:
where
is the Kernel function,
is the distance in the Euclidean form,
is a material constant that depends upon the internal characteristic lengths (
) and external characteristic length (
), and
is a material constant that is determined experimentally.
and
represent the fourth-order elasticity and the strain tensors, respectively, and
is the region occupied by the body.
The nonlocal constitutive formulation is [
28]:
x,
y,
z depict length, width, and height of the beam, respectively and
u1,
u2,
u3 are the displacements in the
x,
y,
z directions. The displacements for a Bernoulli–Euler beam can be written as below [
17]:
is the strain tensor, expressed as:
From Equation (5) we find the strains of the Euler–Bernoulli beam as follows:
Stress
for the linear elastic materials is expressed as follows:
where
E is the elastic modulus of the material. If
, the only nonzero component of strain, is written in Equation (7),
is obtained as:
Moment (
M) and the moment of inertia (
I) are given by:
where
A represents the cross-section area.
For the transverse vibration of an Euler–Bernoulli beam (shown in
Figure 1) resting on a Winkler elastic foundation, the equilibrium conditions are:
where
ρ,
q(
x,
t), and
kw are the mass density, distributed load, and Winkler foundation parameter, respectively.
The nonlocal constitutive relations can be simplified in the following form for a one-dimensional case [
14,
29]:
By multiplying
z on both sides of Equation (13) and integrating the cross-sectional area of the beam, we obtain:
Substituting Equations (6) and (9) into (14), we get:
By differentiating Equation (15) twice with respect to variable
x and substituting Equation (12) into Equation (15), we obtain the governing equation for the vibration of an Euler–Bernoulli nanobeam resting on a Winkler elastic foundation, as below:
4. Results and Discussion
In this section, frequency values of nanowires were obtained with various non-dimensional small-scale parameters (
e0a/
L), different non-dimensional Winkler foundation parameters (
KW), different boundary conditions, and different number of elements (
N). The material properties for the three nanowires are listed in
Table 1. The results obtained were shown in tables and graphs. The dimensionless Winkler parameter used for the results is expressed as the formula below:
Table 2 presents the natural frequencies of nanowires with C-C, C-S, S-S, and C-F boundary conditions. Finite element solutions for the three nanowires were compared with each other. It was clearly observed from the table that the highest frequency value occurred for both SiCNW and the C-C boundary condition, while the lowest value was seen in both AuNW and C-F boundary conditions.
Table 3,
Table 4,
Table 5 and
Table 6 show natural frequencies of embedded simply supported nanowires for
KW = 1,
KW = 10,
KW = 100, and
KW = 1000, respectively, with different
e0a/
L values. It was found from these tables that frequency values increased as
KW values increased, but frequencies decreased by increasing
e0a/
L. Moreover, it was evident that small-scale effects became more considerable for higher modes.
In
Table 7, the frequency values of C-C nanowires are given by analytical and finite element solutions for
e0a/
L = 0.2. For the finite element solution, as the element number increased, the results approached the real value.
Effects of both temperature rise and the Winkler parameter on the first three natural frequencies of nanowires are revealed in
Table 8. It is apparent from the table that an increase temperature rise led to a decrease in frequency, contrary to the Winkler parameter. Also, it can be emphasized that the frequencies of AuNW were more affected than the other nanowires because of their related material properties, given in
Table 1. Moreover, it can be observed that the influence of temperature rise was more prominent for lower modes and smaller Winkler parameters.
Effects of nanoscale and foundation parameters on the first five natural frequencies are respectively depicted in
Figure 2 and
Figure 3. It can be concluded from these figures that size dependency was more pronounced for higher modes, while the natural frequencies in lower modes were more affected from foundation parameters. As stated before, it was clear that the natural frequencies decreased and increased by increasing
e0a/
L and
KW, respectively.
Figure 4 displays the variation of fundamental frequencies of the three embedded nanowires with respect to temperature rise for various small-scale parameter values. It can be recognized from the figure that the effect of
e0a/
L was more prominent for SiCNW than the other ones. On the other hand, the influence of temperature rise was more significant for gold and silver nanowires than the silicon carbide nanowire.
5. Conclusions
Three types of nanowires, SiCNW, AgNW, and AuNW, are modeled as nonlocal Euler–Bernoulli nanobeams resting on a Winkler elastic foundation. Frequency values of these nanowires are obtained via a finite element solution, and results are given by tables and graphs. The effect of temperature change on the vibrational responses of simply supported nanowires is also examined as a case study. It can be concluded from the results that among all boundary conditions, C-C has the highest frequency values and C-F has the lowest ones. Also, it can be emphasized that by increasing the Winkler parameter value, frequency values increase, while by increasing non-dimensional small-scale parameter (e0a/L), frequency value decreases. When we compare the frequency values of nanowires, SiCNW has the highest frequency values, while AuNW has the lowest frequency values because of its different material properties. Additionally, it is revealed that the effects of temperature change and small-scale parameters on the frequencies of nanowires are both considerable and negligible depending on the values of the involved material properties.