Exact solutions for the static bending of nonlocal higher-order

shear beams under various boundary conditions

Noël Challamel 1,* , Abdelhakim Kaci 2, 3 and Abdelouahed Tounsi 3

1
Université de Bretagne Sud, IRDL – UBS – Lorient, IRDL (CNRS UMR 6027), Centre de Recherche, Rue de Saint Maudé –
BP 92116, 56321 Lorient cedex- France, Email : [email protected]
2
Université Dr. Tahar Moulay, Faculté de Technologie, Département de Génie Civil et Hydraulique, BP 138 Cité En-Nasr
20000 Saida, Algérie, Email : [email protected]
3
Material and Hydrology Laboratory, University of Sidi Bel Abbes, Faculty of Technology, Civil Engineering Department,
Algeria, Email: tou [email protected]

Abstract
This paper presents some exact solutions, for the static bending behavior of higher-order shear elastic nanobeams
using the nonlocal differential constitutive relation of Eringen, and under various boundary conditions. The
nonlocal higher-order shear beam referred as a Bickford-Reddy’s beam model, assumes a cubic interpolation field
for the displacement, associated with a parabolic shear strain measure. The governing equations and boundary
conditions are derived using the principle of virtual displacements. The nonlocality is applied to the generalized
higher-order shear constitutive law, which is formulated in term of shear force, bending moment and higher-order
moment. Subsequently, it is shown that the governing equations can be reduced to a single linear sixth-order
differential equation for the transverse deflection, which can be solved using exact methods, for general boundary
conditions. The present approach generalizes the results derived by Reddy (2007) for the static bending of
nonlocal higher-order shear beams under uniform distributed loading. Furthermore, we conduct an in-depth
investigation to quantify the effects of the nonlocal parameter and the length-to-thickness ratio of the nanobeam
on its bending response, for various boundary conditions, including the simply-supported, clamped and free
boundary conditions at each end. Comparative studies demonstrate that our results not only cover the case of
nonlocal higher-order shear beam theories, but can degenerate asymptotically into nonlocal Euler–Bernoulli beam
theory, and nonlocal Timoshenko beam theory, which are better documented for the aforementioned boundary
conditions.

Keywords: Nonlocal theory, Stress gradient model, Length scale effects, Nanobeam.

1. Introduction
Nonlocal elasticity is a continuum theory which intrinsically contains scale effects that can reproduce the
microstructure behaviour at a subscale. Among nonlocal elastic models, the differential nonlocal elastic model of
Eringen (1983) contains one length scale, which may be fitted from atomistic wave dispersion behaviour (see also
Eringen, 1983). The application of this differential nonlocal model (Eringen’s differential model) at the beam
scale has been proposed 20 years later by two independent contributions of Peddieson et al. (2003) and Sudak
(2003). Peddieson et al. (2003) and Sudak (2003) both considered a nonlocal bending moment – curvature elastic
law in the Eringen’s form. They first derived new solutions for the buckling or the static bending of nonlocal
Euler-Bernoulli beams. Peddieson et al. (2003) obtained a fourth-order differential equation of the deflection for
the static bending of the nonlocal Euler-Bernoulli beam, which also includes the second-order derivative of the
distributed lateral loading. Exact solutions are provided by Peddieson et al. (2003) for sinusoidal or uniform
distributed loading. Peddieson et al. (2003) also showed the softening effect induced by the small-scale effects for
such kind of nonlocal theory. Peddieson et al. (2003) also pointed out the absence of scale effects for the nonlocal
cantilever loaded by a concentrated force, which can be seen as a paradoxical result. It has been shown by
Challamel and Wang (2008) (see also Zhang et al., 2010) that this paradox of absence of scale effects, can be
solved using a two-length scale nonlocal beam model (which can be viewed as a two-phase nonlocal Euler-
Bernoulli beam model or a nonlocal strain gradient Euler-Bernoulli beam model – see also Lim et al., 2015).

*
Corresponding author.
E-mail addresses: [email protected]
1
Following the results derived in statics for nonlocal Euler-Bernoulli beams (Peddieson et al., 2003; Sudak, 2003),
researchers later studied the dynamics of nonlocal Euler-Bernoulli beams (Wang, 2005; Lu et al., 2006; Xu,
2006).
The first nonlocal beam solutions derived by Peddieson et al. (2003) or Sudak (2003) are based on nonlocal
Euler-Bernoulli beam theories, which neglects the shear effect. The shear effect can be rigorously introduced
through the first-order-shear beam theory also called Bresse-Timoshenko beam theory (Bresse, 1859;
Timoshenko, 1920; Timoshenko, 1921; Timoshenko, 1922; see also Elishakoff, 2019 or Challamel and
Elishakoff, 2019). The Bresse-Timoshenko beam theory can be seen as a generalization of Euler-Bernoulli beam
theory which accounts for both shear and additional rotary effects. The generalization of the nonlocal Euler-
Bernoulli beam theory to nonlocal Bresse-Timoshenko beam theory is due to Wang (2005), Wang et al. (2006),
Wang and Liew (2007), Wang et al. (2007), Reddy (2007), Wang et al. (2008) and Reddy and Pang (2008).
Whereas Wang (2005), Wang et al. (2006), Wang and Liew (2007), Wang et al. (2007) or Wang et al. (2008) only
considered a nonlocal bending elasticity (with local shear elasticity), Reddy (2007) or Reddy and Pang (2008)
affected both the bending and the shear parts of the elastic law in the formulation of the nonlocal Bresse-
Timoshenko beam model. Reddy (2007) first obtained an analytical solution for the simply supported nonlocal
Bresse-Timoshenko beam, using Navier’s series method. Reddy and Pang (2008) derived exact solutions for the
bending of nonlocal Bresse-Timoshenko beams under various boundary conditions (static behaviour and free
vibration analysis). Simsek M. and Yurtcu (2013) generalized the Navier-type solution of Reddy (2007) to simply
supported nonlocal functionally graded Bresse-Timoshenko beams. Polizzotto (2016) remarked that the statics of
the nonlocal Bresse-Timoshenko beam is governed by a fourth-order differential equation, of the same order than
the one of the local Bresse-Timoshenko beam models.
More advanced shear beam models are the higher-order shear beam models introduced in the 80’s by Bickford
(1982) and Reddy (1984). These models typically contain more refined beam kinematics with cubic displacement
field along the cross section. A classification of Euler-Bernoulli, Bresse-Timoshenko and Bickford-Reddy beam
models is proposed by Polizzotto (2015) through a sequence of increasing order of the kinematic field. A closely
related model also based on a cubic displacement field is the higher-order shear beam model of Shi and Voyiadjis
(2011) (see also Wang and Shi, 2012). Even if the two models, the higher-order Bickford-Reddy beam model and
the higher-order Shi-Voyiadjis beam model statically differ (from the stress point of view), Challamel (2013)
showed that both higher-order shear beam models are governed by the same sixth-order differential equation for
the deflection, which make these two models kinematically equivalent. Challamel (2011), Challamel (2013) or
Challamel et al. (2013) showed that the higher-order shear beam models can be classified as strain gradient
Bresse-Timoshenko beam models. Exact solutions for the static bending of Reddy-Bickford beams are available
in the book of Wang et al. (2000) (see also more recently Ruocco and Reddy, 2023). Reddy (2007) first coupled
nonlocal elasticity to higher-order shear beam theories in his seminal paper. Reddy (2007) derived exact solutions
expressed in series form (Navier-type solution) for the deflection of a nonlocal simply supported Bickford-Reddy
beam under various distributed loading including the uniform loading. Reddy (2007) also presented exact
buckling and vibration solution for the simply supported nonlocal Bickford-Reddy beam. Aydogdu (2009) also
studied the bending, the buckling and the vibration of simply supported nonlocal Bickford-Reddy beam, following
the same methodology and also included the case of alternative higher-order shear kinematics with exponential
variation (instead of the cubic one assumed in the Bickford-Reddy model). Thai (2012) derived exact Navier-type
solution for the simply supported nonlocal Bickford-Reddy and nonlocal Shi-Voyiadjis beam model. Thai (2012)
numerically noticed that both nonlocal higher-order shear beam models give the same results (static deflection,
buckling load or vibration eigenfrequencies), as also theoretically analysed by Challamel (2013) from the
differential equations of the two models in statics and in dynamics (buckling and vibration problem). As shown by
Challamel (2013) (see also Elishakoff et al., 2012), the nonlocal higher-order shear beam mode is governed by a
sixth-order differential equation in space, of the same order than the local higher-order shear beam model.
Challamel et al. (2013-a) studied the buckling of nonlocal higher-order shear beams for various boundary
conditions. To the authors’ knowledge, the available analytical results presented for the static deflection of
nonlocal higher-order shear beams are restricted to simply supported boundary conditions (Reddy, 2007;
Aydogdu, 2009; Thai, 2012). Results derived for other boundary conditions use numerical methods (for instance
the p-Ritz method in the paper of Lin et al., 2020). Pei and Li (2021) studied the static response of nonlocal
2
higher-order shear elastic beams with various boundary conditions. However, the solutions of Pei and Li (2021)
neglected the higher-order shear terms, so that the solutions derived by Pei and Li (2021) is mathematically
equivalent to a nonlocal Timoshenko-type beam. Vaccaro et al. (2022) studied the static response of a stress-
driven nonlocal higher-order shear beam of Reddy’s type and observed the so-called stiffening effect of the small
length scale terms
The present paper aims to present exact solutions for the static bending of nonlocal higher-order shear beams,
not only for simply supported boundary conditions, but also for various standard boundary conditions, including
clamped, hinged or free boundary conditions. In contrast to the study by Pei and Li (2021), the contribution of
higher-order shear moments and forces has not been neglected. The linear sixth-order differential equation for the
deflection is integrated using the sixth variationally-based boundary conditions valid for the higher-order shear
beam model, incorporating the contributions of higher-order moments and shear forces. This approach makes it
possible to capture more subtle and precise effects, providing a more faithful model of mechanical behaviour at
the nanometric scale. By developing a sixth-order equation, we have broadened the spectrum of problems that can
be analysed and solved. This broadens the range of boundary conditions and practical applications, making the
method more versatile and applicable to various nanotechnology and micro-engineering scenarios. By proposing a
theory that, while retaining the necessary complexity, remains analytically soluble for isotropic beams. This
balance between precision and analytical feasibility represents a significant advance in the field.

2. Theoretical Formulations
Consider a nanobeam length L and rectangular cross-section 𝑏 × ℎ, with b being the width and h being the
height. The 𝑥, 𝑦 𝑎𝑛𝑑 𝑧 coordinates are taken along the length, width, and height of the nanobeam, respectively.
The nanobeam is subjected to transverse load of intensity 𝑞(𝑥) per unit length of the nanobeam.

2.1. Assumptions made in theoretical formulation

(1) The in-plane displacement 𝑢 in 𝑥 direction consists of two parts:
a) A displacement component analogous to displacement in elementary beam theory of bending;
b) Displacement component due to shear deformation which is assumed to be hyperbolic in nature with
respect to thickness coordinate.
(2) The transverse displacement w in z direction is assumed to be a function of x coordinate.
(3) One dimensional constitutive law is used.
(4) The nanobeam is subjected to lateral load only.

2.2. The displacement field

Based on the before mentioned assumptions, the displacement field of the present unified refined beam theory
is given as below:
𝑢(𝑥, 𝑧) = 𝑢 (𝑥) − 𝑧𝑤′ + 𝑓(𝑧) 𝜙 + 𝑤 ′ , (1a)
𝑤(𝑥, 𝑧) = 𝑤(𝑥), (1b)
Where𝑢 and 𝑤are the axial displacement along the midline of the beam, 𝜙 rotation of the 𝑦𝑧 planes due to
bending. In order to describe several HSBTs in a unified way, we introduce a parameter𝜁, which takes on different
values in the various HSBTs, depending on the respective kinematic function. If, for instance, we assume a cubic
form function for𝑓(𝑧), associated with a parabolic shear stress distribution, the unified kinematic function can be
presented as
𝑓(𝑧) = 𝜁 𝑧 1 − (2)
ℎ
Where 𝜁 = 1 for the model of Bickford (1982) – Reddy (1984), and 𝜁 = 5⁄4 for the model of Shi and Voyiadjis
(2011) (see more recently Shi (2007) or Wang and Shi (2012)).
With the displacement fields above, normal strain and transverse shear strain for beam are given by:
𝜀 = = −𝑧𝑤 ″ + 𝑓(𝑧) 𝜙 ′ + 𝑤 ″ = 𝑓(𝑧)𝜙 ′ + [𝑓(𝑧) − 𝑧]𝑤 ″ , (3a)
𝛾 = + = 𝑓 ′ (𝑧) 𝜙 + 𝑤 ′ (3b)
By rearranging Eq. (3), the strains can be written as

3
 𝜀 = 𝑧𝜙 ′ + [𝑓(𝑧) − 𝑧] 𝜙 ′ + 𝑤 ″ , (4a)
′ (𝑧) ′
𝛾 =𝑓 𝜙+𝑤 (4b)
Considering first a local orthotropic elastic constitutive law in the following form:
𝜎 =𝐶 𝜀 (5a)
𝜏 =𝐶 𝛾 (5b)
where 𝐶 is the elasticity modulus in the axial direction (denoted by 𝐸 for isotropic beams) whereas 𝐶 is the
transverse shear modulus of the orthotropic beam (denoted by 𝐺 for isotropic beams).

2.3. Equations of motion

The field equations for the static bending of a nonlocal higher-order shear beam can be introduced from the
principle of virtual work:
𝛿𝑈 − 𝛿𝑉 = 0 (6)
where 𝛿𝑈 is the variation of strain energy and 𝛿𝑉is the variation of the potential energy.
In general, the strain energy of the beam of length 𝐿 with an area of cross section denoted 𝐴 can be defined from
the volumetric integral:
𝛿𝑈 = ∫ ∫ (𝜎 𝛿𝜀 + 𝜏 𝛿𝛾 )𝑑𝑥𝑑𝐴
= ∫ 𝑀𝛿𝜙 ′ + 𝑃 𝛿𝑤 ″ + 𝛿𝜙 ′ + 𝑄 𝛿𝑤 ′ + 𝛿𝜙 𝑑𝑥 (7)
where M represents is the bending moment, while Q and P are referred to as higher-order shear forces and higher-
order moments. It is important to note that their contribution to the virtual work is not ignored.
𝑀 = ∫ 𝜎 𝑧𝑑𝐴, 𝑄 = ∫ 𝜏 𝑓 ( ) 𝑑𝑧 and 𝑃 = ∫ 𝜎 [𝑓(𝑧) − 𝑧]𝑑𝑧 (8)
The variation of the potential energy can be expressed as
𝛿𝑉 = ∫ 𝑞𝛿𝑤𝑑𝑥 (9)
Where 𝑞 is the transverse load.
which can be expressed for the higher-order shear beam model, as:
∫ 𝑀𝛿𝜙 ′ + 𝑃 𝛿𝑤 ″ + 𝛿𝜙 ′ + 𝑄 𝛿𝑤 ′ + 𝛿𝜙 𝑑𝑥 − ∫ 𝑞𝛿𝑤𝑑𝑥 = 0 (10)
Integrating each term of the principle of virtual work by part gives the coupled system of equilibrium
equations:
𝑃″ − 𝑄 ′ − 𝑞 = 0
(11)
−𝑀′ − 𝑃′ + 𝑄 = 0
with the variationally based boundary conditions:
[(𝑀 + 𝑃)𝛿𝜙] = 0; [(𝑄 − 𝑃 )𝛿𝑤] = 0 and 𝑃𝛿𝑤 ′ =0 (12)
Eq. (11) can be equivalently rewritten:
𝑃″ − 𝑄 ′ = 𝑞
(13)
𝑀″ = −𝑞
Substituting Eq. (4) and (5) into Eq. (8) and integrating through the thickness of the beam, the stress resultants
are given, in case of local elasticity, as
𝑀 𝐸𝐼 0 −𝑐 𝜅𝐺𝐴 ϕ′
𝑄 = 0 𝜅𝐺𝐴 0 (w′ + ϕ) (14)
𝑃 −𝑐 𝜅𝐴 0 𝑏 𝜅𝐺𝐴 (w" + ϕ′)
where the following notation is introduced for the stiffness parameters:
∫ Ez dA = EI (15a)
∫ G f ′ (z) dA = κGA, (15b)
∫ 𝐸[𝑓(𝑧) − 𝑧] 𝑑𝐴 = 𝑏 𝜅𝐺𝐴, (15c)
∫ 𝐸𝑧[𝑓(𝑧) − 𝑧]𝑑𝐴 = −𝑐 𝜅𝐺𝐴, (15d)

4
 2.4. Stiffness parameters
We will utilize the unified kinematics function in Eq. (2) to obtain dimensionless values for the stiffness
parameters, and we will see that the length scale c0 will vanish in the case of Shi. The dimensionless values of the
stiffness parameters play a key role in describing the various HBTs in a unified framework.
When introducing the unified kinematic function (Eq. (2))
f(z) = ζ z 1 − (16)

into Eq. (15) and integrating over the depth of the beam, we obtain
,
=𝜁 , (17a)
,
=𝜁 − 𝜁 + 1, (17b)
,
=1−𝜁 , (17c)
By introducing the specific values of 𝜁 (5/4 and 1 in the Shi and Reddy model, respectively)
However, prior to doing so, it is convenient to introduce abbreviations by letting the previously defined stiffness
parameters in Eq. (16) be labelled 𝑆 , 𝑆 and𝑆 :
κ ,
S = =ζ − ζ + 1, (18a)
,
𝑆 = =1−𝜁 , (18b)
,
𝑆 = =𝜁 , (18c)
For the Bickford-Reddy’s model, we have 𝑆 = 1⁄21, 𝑆 = 1⁄5 and 𝑆 = 8⁄15, whereas for the Shi-Voyiadjis
model, we have 𝑆 = 1⁄84, 𝑆 = 0 and 𝑆 = 5⁄6 (see Challamel et al., 2013-a; Challamel et al., 2013-b). These
two higher-order shear beam models are associated with the same differential equations for the displacement or
the rotation fields. They are kinematically equivalent for the higher-order shear beam problem (Challamel, 2013)
or for the higher-order shear plate problem (Challamel et al., 2013-b). The same equivalent has been analysed by
di Sciuva (2019) for both the Bickford-Reddy’s higher-order shear plate model, and the Shi-Voyiadjis’s higher-
order shear plate model. The local elastic constitutive law can be written in a single form for the considered cubic
higher-order shear beam models:
M EI 0 −S EI ϕ′
Q = 0 S GA 0 (w′ + ϕ) (19)
P −S EI 0 S EI (w" + ϕ′)

2.5. Nonlocal HSDT model and constitutive relations

Response of materials at the nanoscale is different from those of their bulk counterparts. In the theory of
nonlocal elasticity (Eringen, 1983), the stress at a reference point x is considered to be a functional of the strain
field at every point in the body. For example, in the nonlocal elasticity, the uniaxial constitutive law is expressed
as (Eringen, 1983)
𝜎 −𝜇 = 𝐸𝜀 (20a)

𝜏 −𝜇 = 𝐺𝛾 (20b)
where E is the elasticity modulus and G is the shear modulus. 𝜇 = (𝑒 𝑎) is a nonlocal parameter revealing the
nanoscale effect on the response of nanobeams. 𝑒 is a constant appropriate to each material and a is an internal
characteristic length. Arash and Wang (2012) showed that the value of the nonlocal parameter depends on the
boundary conditions, chirality, mode shapes, number of walls, and the nature of motion. In the investigation of the
nonlocal parameter effect, it is crucial to determine the magnitude of the parameter e0 since it has a significant
influence on the effect of small length scale. So far, no experiments have been conducted to predict the magnitude
of e for carbon nanotubes. In the open literature (Arash and Ansari, 2010; Wang, 2005; Wang and Wang, 2007),
it is suggested that the value of nonlocal parameter can be determined by using a comparison of dispersion curves
from the nonlocal continuum mechanics and molecular dynamics simulation. It should be noted that according to
5
the previous discussions about the values of the nonlocal parameter in detail. (𝑒 𝑎) is usually considered as the
single scale coefficient which is smaller than 2.0 nm for nanostructures (Eringen (1983)).
We assume that the nonlocality of Eringen’s type acts on each part of the constitutive law:
𝑀 − 𝜇𝑀″ 𝐸𝐼 0 −𝑆 𝐸𝐼 𝜙′
𝑄 − 𝜇𝑄″ = 0 𝑆 𝐺𝐴 0 𝑤′ + 𝜙 (21)
𝑃 − 𝜇𝑃 ″ −𝑆 𝐸𝐼 0 𝑆 𝐸𝐼 ″
𝑤 +𝜙 ′

2.6. Equations of nonlocal beam theories

2.6.1. Stress resultants in terms of generalized displacements
Using Eq. (13) and Eq. (21) gives the static variables in term of kinematic ones:
𝑀 = 𝐸𝐼𝜙 ′ − 𝑆 𝐸𝐼 𝑤 ″ + 𝜙 ′ − 𝜇𝑞(𝑥)
(22)
𝑄 − 𝑃′ = 𝑆 𝐺𝐴 𝑤 ′ + 𝜙 + 𝑆 𝐸𝐼𝜙 ″ − 𝑆 𝐸𝐼 𝑤 ‴ + 𝜙 ″ − 𝜇𝑞 ′ (𝑥)
2.6.2. Equations of motion of HSBT
Injecting both expressions of Eq. (22) in Eq. (13) gives the coupled system of differential equations in term of
kinematic variables:
𝐸𝐼𝜙 ‴ − 𝑆 𝐸𝐼 𝑤 ( ) + 𝜙 ‴ = −𝑞 + 𝜇𝑞 ″
(23)
𝑆 𝐺𝐴 𝑤 ″ + 𝜙 ′ + 𝑆 𝐸𝐼𝜙 ‴ − 𝑆 𝐸𝐼 𝑤 ( ) + 𝜙 ‴ = −𝑞 + 𝜇𝑞 ″
It is possible to extract the third derivative of the rotation from the first equation of Eq. (23):
𝐸𝐼(1 − 𝑆 )𝜙 ‴ = 𝑆 𝐸𝐼𝑤 ( ) − 𝑞 − 𝜇𝑞 ″ (24)
Using Eq. (24) and deriving the second equation of Eq. (23) leads to a single sixth-order differential equation
expressed in term of displacement:
( )
− (𝑆 − 𝑆 )𝑤 ( ) + 𝐸𝐼𝑤 ( )
= 1− (1 + 𝑆 − 2𝑆 )𝜕 (1 − 𝜇𝜕 )𝑞 (25)
Note that for Euler-Bernoulli: 𝑆 = 𝑆 = 0 and 𝑆 → ∞
𝐸𝐼𝑤 ( ) = (1 − 𝜇𝜕 )𝑞 (26)
Eq. (26) has been obtained by Peddieson et al. (2003).
and for Bresse-Timoshenko: 𝑆 = 𝑆 = 0 and 𝑆 = 𝜅 = 5⁄6
𝐸𝐼𝑤 ( )
= 1− 𝜕 (1 − 𝜇𝜕 )𝑞 (27)
Eq. (27) has been obtained by Polizzotto (2016).
The nonlocal Higher-order Bickford-Reddy beam model (𝑆 = , 𝑆 = and 𝑆 = 𝜅 = ) or the nonlocal
Higher-order Shi-Voyiadjis beam model (𝑆 = 1⁄84, 𝑆 = 0 𝑎𝑛𝑑 𝑆 = 5⁄6) are both governed by the sixth-order
differential equation:
( )
− 𝑤 ( ) + 𝐸𝐼𝑤 ( )
= 1− 𝜕 (1 − 𝑙 𝜕 )𝑞 (28)
coupled with the 6 variationally-based boundary conditions, three boundary conditions at each end. For simply
supported boundary conditions, we have:
𝑀 + 𝑃 = 0; 𝑤 = 0 and 𝑃 = 0, or equivalently 𝑀 = 0;𝑤 = 0 and 𝑃 = 0 (29)
For clamped boundary conditions, we have:
𝜙 = 0; 𝑤 = 0 and 𝑤 ′ = 0 (30)
For free end boundary conditions, we have:
𝑀 + 𝑃 = 0; 𝑄 − 𝑃 = 0 and 𝑃 = 0or equivalently𝑀 = 0;𝑄 − 𝑃′ = 0and𝑃 = 0 (31)
As a consequence, we need to express the static boundary conditions in terms of kinematic variables.
From the nonlocal differential law Eq. (21), it is possible to show that the static variables are related between each
other by the same differential operators:
(𝑆 𝑀 + 𝑃) − 𝜇(𝑆 𝑀 + 𝑃)″ = 𝐸𝐼(𝑆 − 𝑆 ) 𝑤 ″ + 𝜙 ′ ′
⇒ = (32)
𝑄 ′ − 𝜇𝑄‴ = 𝐸𝐼(𝑆 − 𝑆 ) 𝑤 ″ + 𝜙 ′
We note that in the particular case of Shi-Voyiadjis higher-order shear beam model, we have the following
differential constraint:
′
𝑆 =0⇒ = (33)
6
For the general higher-order shear beam model, and considering Eq. (32) with the equilibrium equations Eq. (13)
leads to the coupled system of differential equations in the shear force:
𝑄 ‴ − 𝑄 ′ = 𝑞(1 − 𝑆 )
(34)
𝑄 ′ − 𝜇𝑄‴ = 𝑆 𝐺𝐴 𝑤 ″ + 𝜙 ′
We then identify:
″ ′ ( )
𝑄′ = (35)

Using Eq. (32), we finally obtain the higher-order moment:

″ ′ ( ) ′ ″ ′

𝑃= (36)

The natural boundary condition can then be expressed in term of kinematic boundary conditions:
𝑃=0 ⇒
𝐸𝐼 𝑆 − 𝑆 𝜇 𝑤 ″ + 𝜙 ′ − 𝑆 𝐸𝐼 1 − 𝜇 𝜙′ + 1 − 𝑆 𝜇 𝑞𝜇 = 0 (37)
For the Shi-Voyiadjis model, this natural boundary condition simplifies:
𝑃 = 0 ⇒ 𝜅𝐺𝐴𝑏 𝑤 ″ + 𝜙 ′ + 𝑞𝜇 = 0 (38)
We then summarize the formulation of boundary conditions for the Shi-Voyiadjis higher-order nonlocal shear
beam model. For simply supported boundary conditions, the boundary conditions can be explicitly given by:
𝑀 = 0; 𝑤 = 0 and 𝑃 = 0 which is equivalent to 𝐸𝐼𝜙 ′ − 𝑞𝜇 = 0; 𝑤 = 0 and 𝜅𝐺𝐴𝑏 𝑤 ″ + 𝜙 ′ + 𝑞𝜇 = 0 (39)
For free end boundary conditions, we have:
𝑀 = 0; 𝑄 − 𝑃 = 0 and 𝑃 = 0 which is equivalent to 𝐸𝐼𝜙 ′ − 𝑞𝜇 = 0; 𝜅𝐺𝐴(𝑤 + 𝜙) − 𝑏 𝜅𝐺𝐴(𝑤 +
𝜙 ) = 0 and 𝜅𝐺𝐴𝑏 𝑤 ″ + 𝜙 ′ + 𝑞𝜇 = 0 (40)
The formulation of boundary conditions for the Bickford-Reddy higher-order shear beam model can be also
deduced from the natural and essential variationally-based boundary conditions.
Eq. (25) can be simplified at the following sixth-order ordinary differential equation for transverse deflection:
𝑤( ) − 𝛽 𝑤( )
= (1 + 𝑆 − 2𝑆 )𝜕 − 1 (1 − 𝜇𝜕 )𝑞 (41)
where
β = (42)
3. General solution
By using the method of varying arbitrary constants (Demidovitch, 1972), the general solution of a sixth-order
ordinary differential equation eq. (41) is in the form:
q0 x4 x3 x2
EIw(x) = C1 cosh(βx) + C2 sinh(βx) + 24
+ C3
6
+ C4
2
+ C5 x
μ ( )
+C − − − μq + C x − C + (43)
β
And by integrating eq. (23), we obtain
Q − P′ (x) = S GA ϕ + w ′ − EI S ϕ + w ′ − S ϕ = −q x − C (44)

M(x) = EIϕ′ − S EI ϕ′ + w ″ − μq = −q −C x−C (45)

EI(1 − S )ϕ(x) = μq x − q −C − C x − C + S EIw ′ (46)

3.1. Simply supported nanobeam under a uniform load

For a nanobeam simply supported at x=0 and x=L with L the length of the nano-beam, the following boundary
conditions should be satisfied
𝑀 = 0; 𝑤 = 0 and 𝑃=0 (47)
Enforcement of the above boundary conditions on Eqs. (37), (43) and (45) yields

7
 1 1 𝛽𝐿
𝐶 = −(1 − 𝑆 ) 𝛺𝑞 𝐿 𝜇̄ 𝐿 − , 𝐶 = (1 − 𝑆 ) 𝛺𝑞 𝐿 𝜇̄ 𝐿 − 𝑡𝑎𝑛ℎ
𝛽 𝛽 2
𝐶 =− , 𝐶 = 0, 𝐶 = (12𝜇̄ + 1) , 𝐶 = (1 − 𝑆 ) [𝛺(𝑆 − 𝑆 ) − 𝜇̄ ]𝛺𝑞 𝐿
𝜇̄ = , 𝛺= , 𝛽 = = (48)
Then the deflection and bending moments are given by
𝐸𝐼𝑤(𝑥) = −2 + + (1 − 𝑆 ) 𝛺𝑞 𝐿 − 𝜇̄ 𝐿
̄ ( )
−𝑡𝑎𝑛ℎ 𝑠𝑖𝑛ℎ(𝛽𝑥) + 𝑐𝑜𝑠ℎ(𝛽𝑥) + − (49)

𝑀(𝑥) = − − (50)

EIϕ(x) = − 4 −6 + 1 + S (1 − S )Ωq L β − μ̄ L
β
β
−tanh 𝑐𝑜𝑠ℎ(𝛽𝑥) + 𝑠𝑖𝑛ℎ(𝛽𝑥) + (S (1 − S )Ω − μ̄ ) 1−2 (51)
For nonlocal Euler–Bernoulli beam theory, we have 𝐒𝟏 = 𝟏; 𝐒𝟐 = 𝟏, 𝐒𝟑 = 𝟎, 𝛃 → ∞. Then the deflection and
bending moments are given by
μ̄
EIw(x) = −2 + + − (52)

M(x) = − − (53)
which coincides with the solution of Reddy and Pang 2008
For nonlocal Timoshenko beam theory, we have 𝑆 = 0, 𝑆 = 0, 𝑆 = 𝜅 and 𝛽 → ∞. Then the deflection and
bending moments are given by
EIw(x) = −2 + + (μ̄ + Ω) − (54)

M(x) = − − (55)
μ̄
EIϕ(x) = − 4 −6 +1 − 1−2 (56)
which coincides with the solution of Reddy and Pang 2008
On the other hand, the Navier solution for simply supported nonlocal HO Reddy-Bickford beam is employed to
obtain the analytical solution. For this purpose, the displacement functions are expressed as product of
undetermined coefficients and known trigonometric functions so as to satisfy the eq. (41) and the above-
mentioned boundary conditions at 𝑥 = 0, 𝐿. The following expansions of the displacement field are assumed:
𝑤(𝑥) = ∑ 𝑊 𝑠𝑖𝑛( 𝜂𝑥) (57)
where 𝑊 is the unknown Fourier coefficients to be determined for each n value and 𝜂 = 𝑚𝜋⁄𝐿.
So that the boundary condition in Eq. (47) is identically satisfied. In addition, the distributed transverse load is
also expanded in simple Fourier series as
𝑞(𝑥) = ∑∞ 𝑄 𝑠𝑖𝑛( 𝜂 𝑥) (58)
where 𝑄 has different values for different kind of loading. For example, for uniform load with magnitude of 𝑞 .
𝑄 = , 𝑚 = 1, 3, 5, … (59)
substituting Eqs. (57) and (58) into Eq. (41), solution form is as follow
w(x, y) = ∑∞ λ [S GA + EI(1 + S − 2S )𝜂 ] sin 𝜂 x (60)
where
𝜆 = 1 + 𝜇𝜂 (61)
The comparison and discussion of the numerical results with our solution for a simply supported is presented in
Table 1.

3.2. Clamped-clamped nanobeam under a uniformly distributed load

Next, we consider a clamped-clamped nanobeam with the following boundary conditions should be satisfied
8
 𝜙 = 0; 𝑤 = 0 and 𝑤′ = 0 (62)
Enforcement of the above BCs on Eqs. (43) and (46) yields
C = 0, C =− , C = (1 + 12μ̄ ),
Ω
C = −μ̄ Ωq L (1 + S − 2S ) + (1 + S − 2S ) − − ,
β β β
Ω Ω
C = (1 − S ) coth , C = (1 − S ) . (63)
β β

μ̄ = , Ω= , β = =
Then the deflection and bending moments are given by
( )
EIw(x) = −2 + + −
( )
+ coth cosh(βx) − sinh(βx) (64)
β

M(x) = − 1−6 +6 − μ̄ q L (65)

( )
EIϕ(x) = − 2 −3 + − 2 −1
( ) 𝛽𝐿
+ coth 2
𝑠𝑖𝑛ℎ(𝛽𝑥) − 𝑐𝑜𝑠ℎ(𝛽𝑥) (66)
For Euler–Bernoulli beam theory, we have 𝑆 = 1; 𝑆 = 1, 𝑆 = 0. Then the deflection and bending moments
are given by
𝐸𝐼𝑤(𝑥) = −2 + (67)

M(x) = − 1−6 +6 − μ̄ q L (68)

which coincides with the solution of Reddy and Pang 2008
For Timoshenko beam theory, we have 𝑆 = 0; 𝑆 = 0; 𝑆 = 𝜅 and β → ∞. Then the deflection and bending
moments are given by
𝐸𝐼𝑤(𝑥) = −2 + + − (69)

M(x) = − − (70)

EIϕ(x) = − 2 −3 + (71)
which coincides with the solution of Reddy and Pang 2008. It is also observed that the transverse deflection of
a C-C beam (clamped at both ends) is completely insensitive to the size effect.

3.3. Clamped-free nanobeam under a uniformly distributed load

In this case, the boundary conditions are
x = 0, 𝜙 = 0; 𝑤 = 0 and 𝑤′ = 0 (72a)
x = L, 𝑀 = 0; 𝑄 − 𝑃 = 0 and 𝑃 = 0 (72b)
Enforcement of the above BCs on Eqs (43) - (46) and (37) yields
C = , C = −q L, C = 0,
μ̄ β (β )
𝐶 =− (1 − 𝑆 ) Ωq L (β )
+ 𝛽 μ̄ L + −1 ,
( ) ̄ ( )
𝐶 = ( )
, 𝐶 = −(1 − 𝑆 )
μ
μ̄ = , Ω= , β = = Ω
(73)
Then the deflection and bending moments are given by
q 0 L4 x 4 x 3 x 2 Ωq0 L4 x x 2
EIw(x) = −4 +6 + (1 − S2 )2 2 −
24 L L L 2 L L
( ) ( ) Ω β (β )
+ ( )
[cosh(βx) + 𝛽𝐿sinh 𝛽(𝐿 − 𝑥)] − (β )
9
 (1−𝑆2 )2 𝛺2
+ 𝑐𝑜𝑠ℎ(𝛽𝐿)
(1 − cosh(βx)) − 𝜇̄ q 𝐿 (74)

M(x) = − 1− (75)

EIϕ(x) = − +3 +3 + 𝜇̄ q 𝐿 + S (1 − S )Ωq L −
( ) ( )Ω
+S ( )
[sinh(βx) − 𝛽𝐿cosh 𝛽(𝐿 − 𝑥)] − S
(β )
β𝜇̄ q 𝐿 sinh(βx) (76)
For the particular case of Euler–Bernoulli and Timoshenko beam theory, we can verify this solution without their
obtained by Reddy and Pang (2008).
For Euler–Bernoulli beam theory, you have 𝑆 = 1, 𝑆 = 1, 𝑆 = 0. Then the deflection and bending moments
are given by
μ̄
EIw(x) = −4 +6 − (77)

M(x) = − 1− (78)
which coincides with the solution of Reddy and Pang 2008

For Timoshenko beam theory, you have 𝑆 = 0; 𝑆 = 0; 𝑆 = 𝜅. Then the deflection and bending moments are
given by
𝐸𝐼𝑤(𝑥) = −4 +6 − 𝜇̄ +𝛺 − 2𝛺 (79)

M(x) = − 1− (80)

EI(1 − S )ϕ(x) = − −3 +3 + μ̄ q L (81)

which coincides with the solution of Reddy and Pang 2008. Moreover, the bending moment is independent of size
effects, while the transverse deflection decreases with the inclusion of the nonlocal parameter.

3.4. Clamped- Simply Supported nanobeam under a uniformly distributed load

In this case, the boundary conditions are
𝜙 = 0; 𝑤 = 0 and 𝑤 ′ = 0 (62)
𝑀 = 0; 𝑤 = 0 and 𝑃=0 (47)
Enforcement of the above boundary conditions on Eqs. (37), (43), (45) and (46) yields
𝐶 =− 𝛼, 𝐶 = 𝛼−𝑞 , 𝐶 = 0, 𝐶 = −(1 − 𝑆 ) 𝜇̅ 𝛺 𝑞 𝐿
( ) ( ) ( ) ̄
𝐶 = 𝛼𝑡𝑎𝑛ℎ(𝛽𝐿), 𝐶 = 𝛼, 𝛼= ( )
(48)
Then the deflection and bending moments are given by
( )
E𝐼𝑤(𝑥) = 𝛺𝑞 𝐿 𝛼[𝑡𝑎𝑛ℎ(𝛽𝐿) 𝑐𝑜𝑠ℎ(𝛽𝑥) + 𝑠𝑖𝑛ℎ(𝛽𝑥)] + −6

− 𝛼 + (12𝜇̅ +12(1 − 𝑆 ) 𝛺 − 3𝛼) − 6𝛼(1 − 𝑆 ) 𝛺

+(1 − 𝑆 ) 𝛺 [α − 2 − 4(𝑆 − 𝑆 )𝛺] (49)

𝑀(𝑥) = − −2 + 𝛼 − 𝛼 +2 (50)
( )
𝐸𝐼𝜙(𝑥) = 𝛺𝑞 𝐿 𝛼[𝑡𝑎𝑛ℎ(𝛽𝐿) sinℎ(𝛽𝑥) + cosℎ(𝛽𝑥)] + −4 + 3𝛼 +

(12 + 24𝜇̅ − 6𝛼) − 𝑆 (1 − 𝑆 ) 24𝛺 − 6𝛼𝛺 (51)

For the particular case of Euler–Bernoulli and Timoshenko beam theory, we can verify this solution without their
obtained by Reddy and Pang (2008).
For Euler–Bernoulli beam theory, you have S = 1, S = 1, S = 0. Then the deflection and bending moments are
given by
EIw(x) = 2 − (5 + 12𝜇̅ ) + (3 + 12𝜇̅ ) (52)

10
 M(x) = − 3 − 15 + 12 − μq L 1 − (53)
For Timoshenko beam theory, you have 𝑆 = 0; 𝑆 = 0; 𝑆 = 𝜅. Then the deflection and bending moments are
given by
𝟒
EIw(x) = −6 − 𝛼 + (12μ + 12Ω − 3α) − 6αΩ (54)

M(x) = − −2 + α − α +2 (55)

𝐸𝐼𝜙(𝑥) = −4 + 3𝛼 + (12 + 24𝜇̅ − 6𝛼) (56)

which coincides with the solution of Reddy and Pang 2008. It should be noted that in the case of EBT, there is
a missing term in the expressions of w(x) and M(x) in the article by Reddy and Pang (2008).

4. Numerical results and discussions

In this section, we evaluate the accuracy of the proposed exact solution for predicting the bending response of
higher-order shear nanobeams (HSBT) using Eringen's nonlocal differential constitutive relations in order to
discuss the influence of small-scale parameters and various boundary conditions on the bending behaviour of the
nanobeam.
The general approach outlined in the previous sections for analyzing the bending of homogeneous nanobeams
is applied here to higher-order shear nanobeams (HSBT) and compared with results obtained using different beam
theories, including Euler–Bernoulli Theory (EBT), Timoshenko Beam Theory (TBT), Refined Beam Theory
(RBT), and the Navier solution.
As a check on the numerical accuracy of the theory and formulation, results of the maximum deflection of the
nanobeam under the uniform distributed load q0 for simply supported (S-S), clamped-clamped (C-C), clamped-
free (C-F) and clamped-simply supported (C-S) boundary conditions are compared with those obtained by Reddy
(2007), Thai (2012) and Benguediab et al. (2013) with Navier solutions, Reddy and Pang (2008) in exact form for
a wide range of nonlocal parameter and thickness ratio.
For convenience, the following nondimensionalization is used:
100𝑤𝐸𝐼
𝑤=
𝑞 𝐿
The numerical results for a simply supported beam subjected to uniform load are presented in Tables 1 to 4 for
various values of thickness ratio 𝐿⁄ℎ and nonlocal parameter 𝜇 . The nonlocal parameters, defined as 𝜇 = (𝑒0 𝑎)2 ,
is considered for values of 0, 1, 2, 3, and 4 𝑛𝑚 . As shown in Table 1, there is a clear agreement between the
present results for the simply supported nonlocal Bickford-Reddy beam and those published by Reddy (2007) and
Benguediab et al. (2013) using Navier solutions. Notably, the results for 𝜇 = 0 correspond to local beam theory.
The effect of the length scale is more pronounced for smaller values of L and decreases as L increases.
Additionally, increasing the nonlocal parameter decreases the stiffness of the beams, and bending results increase.
Table 1: Comparison of dimensionless maximum center deflection under uniform load for simply supported
nonlocal Bickford-Reddy’s beam.
𝐿 ⁄ℎ 𝜇 Navier Solution Navier Solution Present HSBT
(Reddy, 2007) (Benguediab, 2013) (Bickford-Reddy’s beam)
5 0 1.4320 1.4319 1.4320
1 1.5673 1.5673 1.5673
2 1.7027 1.7026 1.7027
3 1.8381 1.8380 1.8381
4 1.9735 1.9734 1.9735
10 0 1.3346 1.3346 1.3346
1 1.4622 1.4622 1.4622
2 1.5898 1.5898 1.5898
3 1.7174 1.7173 1.7174
4 1.8450 1.8449 1.8450
20 0 1.3102 1.3102 1.3102
1 1.4359 1.4359 1.4359
2 1.5615 1.5615 1.5615
11
 3 1.6871 1.6871 1.6871
4 1.8128 1.8128 1.8128
100 0 1.3024 1.3024 1.3024
1 1.4274 1.4274 1.4274
2 1.5525 1.5525 1.5525
3 1.6775 1.6775 1.6775
4 1.8025 1.8025 1.8025
In order to verify the numerical accuracy of the theory and formulation, the results for the dimensionless
maximum centre deflection under uniform loading for homogeneous isotropic nanobeams were obtained using the
present Shi and Voyiadjis's beam model and compared with those obtained by Thai (2012) using Navier solutions,
as shown in Table 2. The agreement between the present results and the published ones can be clearly observed.
Thai (2012) numerically demonstrated that the numerical results for the nonlocal Bickford-Reddy beam and the
nonlocal Shi and Voyiadjis beam coincide.
Table 2: Comparison of dimensionless maximum center deflection under uniform load for simply supported
nonlocal Shi and Voyiadjis’s beam.
𝐿 ⁄ℎ 𝜇 RBT (Thai, 2012) Present HSBT
(Shi and Voyiadjis’s beam)
5 0 1.4320 1.4320
1 1.5673 1.5673
2 1.7027 1.7027
3 1.8381 1.8381
4 1.9735 1.9735
10 0 1.3346 1.3346
1 1.4622 1.4622
2 1.5898 1.5898
3 1.7174 1.7174
4 1.8450 1.8450
20 0 1.3102 1.3102
1 1.4359 1.4359
2 1.5615 1.5615
3 1.6872 1.6871
4 1.8128 1.8128
100 0 1.3024 1.3024
1 1.4274 1.4274
2 1.5525 1.5525
3 1.6775 1.6775
4 1.8025 1.8025
Tables 3 and 4 show the results for different values of the aspect ratio (L/h = 10, 20, 100) and the non-local
parameter (𝜇 = (𝑒 𝑎) = 0, 1, 2, 3, 4 𝑛𝑚2 ) in the case of Euler-Bernoulli Beam Theory (EBT) and Timoshenko
Beam Theory (TBT). As can be clearly seen in Table 3, the non-dimensional deflections are constant for different
values of the slenderness ratio (ℎ⁄𝐿) in the case of Euler-Bernoulli Beam Theory (EBT) for different 𝜇. On the
other hand, it should be noted that EBT gives accurate results when the slenderness ratio (ℎ⁄𝐿) is greater than 20,
while Timoshenko Beam Theory (TBT) has to be used when the slenderness ratio is less than 20. Table 3 also
shows that an increase in the small-scale parameter leads to an increase in the nondimensional maximum
deflection for simply supported boundary conditions.
The main point to highlight is that the Timoshenko Beam Theory (TBT), the RBT and the present theory give
solutions that are almost identical for all values of the thickness ratio L/h and the non-local parameter μ, whereas
the Euler-Bernoulli Beam Theory (EBT) underestimates the deflections. The discrepancy between EBT and shear
deformation theories (i.e. TBT, RBT and the present HSBT theory) is negligible for slender nanobeams, but
significant for deep nanobeams. This is because EBT neglects the effects of shear deformation and rotational
inertia.
In general, the effects of transverse shear deformation and the non-local parameter μ increase deflections. In
other words, the inclusion of shear deformation and nonlocal effects increases deflection magnitudes. The
difference between the EBT and TBT solutions becomes significant as the aspect ratio decreases.
12
 Table 5 shows the nondimensional deflection of a clamped-clamped beam subjected to a uniform load. The
results obtained are compared with those reported by Reddy and Pang (2008) in exact form based on the nonlocal
Timoshenko beam and Euler-Bernoulli beam. It can be seen that the results of the present theory are nearly
identical to those predicted by the Timoshenko Beam Theory (TBT) for all values of the nonlocal parameter and
the length-to-depth ratio, even for short beams where the effects of transverse shear deformation and rotational
inertia are significant. In contrast, the Euler-Bernoulli Beam Theory (EBT) underestimates the deflections. It
should be noted that the TBT requires a shear correction factor to satisfy the free transverse shear stress conditions
on the top and bottom surfaces of the beam, whereas the present theory satisfies these conditions without the use
of shear correction factors. However, this table illustrates that the length scale effect becomes more pronounced at
lower values of L and decreases as L increases. It is also observed that the transverse deflection of a clamped-
clamped beam is completely insensitive to the size effect.
Table 3: Comparison of dimensionless maximum center deflection under uniform load for simply supported
nonlocal Euler-Bernoulli beam.
𝐿 ⁄ℎ 𝜇 EBT EBT with Navier Present
(Reddy and Pang, 2008) Solution (Reddy, 2007) EBT
100 0 1.3021 1.3130 1.3021
1 1.4271 1.4487 1.4271
2 1.5521 1.5844 1.5521
3 1.6771 1.7201 1.6771
4 1.8021 1.8558 1.8021
20 0 1.3021 1.3130 1.3021
1 1.4271 1.4487 1.4271
2 1.5521 1.5844 1.5521
3 1.6771 1.7201 1.6771
4 1.8021 1.8558 1.8021
10 0 1.3021 1.3130 1.3021
1 1.4271 1.4487 1.4271
2 1.5521 1.5844 1.5521
3 1.6771 1.7201 1.6771
4 1.8021 1.8558 1.8021
Table 4: Comparison of dimensionless maximum center deflection under uniform load for simply supported
nonlocal Timoshenko beam.
𝐿 ⁄ℎ 𝜇 TBT TBT with Navier Present
(Reddy and Pang, 2008) Solution (Reddy, 2007) TBT
100 0 1.3024 1.3134 1.3024
1 1.4274 1.4492 1.4274
2 1.5525 1.5849 1.5525
3 1.6775 1.7207 1.6775
4 1.8025 1.8565 1.8025
20 0 1.3102 1.3218 1.3102
1 1.4359 1.4600 1.4359
2 1.5615 1.5981 1.5615
3 1.6871 1.7362 1.6871
4 1.8128 1.8743 1.8128
10 0 1.3346 1.3483 1.3346
1 1.4622 1.4937 1.4622
2 1.5898 1.6391 1.5898
3 1.7173 1.7845 1.7173
4 1.8449 1.9299 1.8449
Table 6 presents the nondimensional deflection of a clamped-free beam subjected to a uniform load. The
results are compared with those reported by Reddy and Pang (2008) in exact form based on the nonlocal
Timoshenko beam and Euler-Bernoulli beam. It is evident that the present theory gives results almost identical to
those of the Timoshenko Beam Theory (TBT) for all values of the nonlocal parameter and the length-to-depth
ratio, even for short beams where the effects of transverse shear deformation and rotary inertia are pronounced. In
contrast, the Euler-Bernoulli Beam Theory (EBT) underestimates the deflections. Notably, the TBT requires a
13
shear correction factor to satisfies the free transverse shear stress conditions on the beam's surfaces, whereas the
present theory meets these conditions without such factors. This table shows that the dimensionless deflection
increases with L and decreases with μ.
Table 7 presents the nondimensional deflection of a clamped- Simply Supported beam subjected to a uniform
load. The results are compared with those reported by Reddy and Pang (2008) in exact form based on the nonlocal
Timoshenko beam and the Euler-Bernoulli beam. It is evident that the present theory yields result nearly identical
to those of the Timoshenko Beam Theory (TBT) for all values of the nonlocal parameter and length-to-depth
ratio, even for short beams where the effects of transverse shear deformation and rotary inertia are pronounced. In
contrast, the Euler-Bernoulli Beam Theory (EBT) underestimates deflections. Notably, the TBT requires a shear
correction factor to satisfy the free transverse shear stress conditions on the beam's surfaces, whereas the present
theory meets these conditions without such factors. This table demonstrates that the dimensionless deflection
increases with μ and decreases with L.
Table 5: Comparison of dimensionless maximum deflection under uniform load for clamped-clamped nanobeams
L ⁄h μ EBT TBT Present
(Reddy and Pang, 2008) (Reddy and Pang, 2008) HSBT
5 0 0.2604 0.3904 0.3846
1 0.2604 0.3904 0.3846
2 0.2604 0.3904 0.3846
3 0.2604 0.3904 0.3846
4 0.2604 0.3904 0.3846
10 0 0.2604 0.2929 0.2922
1 0.2604 0.2929 0.2922
2 0.2604 0.2929 0.2922
3 0.2604 0.2929 0.2922
4 0.2604 0.2929 0.2922
20 0 0.2604 0.2685 0.2684
1 0.2604 0.2685 0.2684
2 0.2604 0.2685 0.2684
3 0.2604 0.2685 0.2684
4 0.2604 0.2685 0.2684
100 0 0.2604 0.2607 0.2607
1 0.2604 0.2607 0.2607
2 0.2604 0.2607 0.2607
3 0.2604 0.2607 0.2607
4 0.2604 0.2607 0.2607
Table 6: Comparison of dimensionless maximum deflection under uniform load for Clamped-free nanobeams
L ⁄h Μ EBT TBT Present
(Reddy and Pang, 2008) (Reddy and Pang, 2008) Present
HSBT
5 0 12.5000 13.0200 13.0085
1 12.0000 12.5200 12.4982
2 11.5000 12.0200 11.9878
3 11.0000 11.5200 11.4774
4 10.5000 11.0200 10.9670
10 0 12.5000 12.6300 12.6286
1 12.0000 12.1300 12.1260
2 11.5000 11.6300 11.6234
3 11.0000 11.1300 11.1208
4 10.5000 10.6300 10.6182
20 0 12.5000 12.5325 12.5323
1 12.0000 12.0325 12.0317
2 11.5000 11.5325 11.5310
3 11.0000 11.0325 11.0304
4 10.5000 10.5325 10.5297
100 0 12.5000 12.5013 12.5013

14
 1 12.0000 12.0013 12.0013
2 11.5000 11.5013 11.5012
3 11.0000 11.0013 11.0012
4 10.5000 10.5013 10.5012

Table 7: Comparison of dimensionless maximum deflection under uniform load for Clamped- Simply Supported
nanobeams
L ⁄h Μ EBT TBT Present
(Reddy and Pang, 2008) (Reddy and Pang, 2008) HSBT
5 0 0.5416 0.6910 0.6866
1 0.5770 0.7285 0.6866
2 0.6125 0.7661 0.7700
3 0.6482 0.8039 0.8121
4 0.6840 0.8418 0.8544
10 0 0.5416 0.5790 0.5784
1 0.5770 0.6148 0.6153
2 0.6125 0.6509 0.6525
3 0.6482 0.6871 0.6898
4 0.6840 0.7235 0.7273
20 0 0.5416 0.5509 0.5509
1 0.5770 0.5864 0.5866
2 0.6125 0.6221 0.6226
3 0.6482 0.6579 0.6587
4 0.6840 0.6939 0.6949
100 0 0.5416 0.5420 0.5420
1 0.5770 0.5774 0.5774
2 0.6125 0.6129 0.6129
3 0.6482 0.6486 0.6486
4 0.6840 0.6844 0.6844
Figure 1 displays the ratio of the maximum deflection under uniformly distributed loads for various boundary
conditions and values of the nonlocal parameter 𝑒 predicted by the Timoshenko beam theory (TBT), Euler beam
theory (EBT), and present High shear beam theory (HSBT) to those predicted by the local EBT, respectively.
Remarkably, our results align significantly with those presented by Reddy and Pang (2008).
Figure 2 illustrates the variation of the static response of nanobeams with the aspect ratio, with nonlocal results
provided for e0a = 1 nm. The aspect ratio ranges from L/h = 10 to L/h = 50. As shown in the figure, deflections
predicted by the nonlocal theory surpass those of the local (classical) results, highlighting the softening effect of
the small-scale parameter on the nanobeam. Interestingly, deflection remains independent of the aspect ratio in the
case of local EBT, but with the consideration of the nonlocal parameter, all responses of EBT become dependent
on the aspect ratio. This dependency arises due to shear deformation effects. The difference between the solution
of EBT and our present approach becomes more significant as the aspect ratio decreases.
Finally, Figure 3 demonstrates the influence of the nonlocal parameter on the static responses of nanobeams, as
obtained using our nonlocal HBT approach. These results show that responses vary nonlinearly with the nonlocal
parameter. Importantly, nanobeams with lower aspect ratios (L/h = 10) are strongly affected by the nonlocal
parameter compared to those with relatively higher aspect ratios. This observation suggests that modeling based
on local (classical) beam theories may not be suitable for nano-sized structures, and nonlocal beam models may
offer a more accurate approximation. It is also observed that the transverse deflection of a C-C beam (clamped at
both ends) is completely insensitive to the size effect (see Eq. (64))

15
 (a) Simply supported (b) Clamped
1,50 1,20
ETB by Reddy and Pang (2008) ETB by Reddy and Pang (2008)
ETB by Reddy and Pang (2008) ETB by Reddy and Pang (2008)
1,40 Present HSBT Present HSBT
1,15
e0=1.00 e0=1.00
1,30 e0=0.67 e0=0.67
e0=0.33 1,10 e0=0.33
Local

Local
EBT

EBT
e0=0.00 e0=0.00
1,20
w /w

w /w
1,05
1,10

1,00
1,00

0,90 0,95
0 10 20 30 40 50 60 0 10 20 30 40 50 60
L/a L/a

(c) Clamped-Free (d) Clamped-Simple supported

1,05 1,40
ETB by Reddy and Pang (2008)
ETB by Reddy and Pang (2008)
1,00 1,30 Present HSBT
e0=1.00
e0=0.67
0,95 1,20 e0=0.33
Local

Local

ETB by Reddy and Pang (2008)

EBT
EBT

e0=0.00
ETB by Reddy and Pang (2008)
w /w

w /w

0,90
Present HSBT 1,10
e0=1.00
e0=0.67
0,85 e0=0.33 1,00
e0=0.00

0,80 0,90
0 10 20 30 40 50 60 0 10 20 30 40 50 60
L/a L/a

Figure 1: Plots of the ratios of the maximum deflection predicted by HSBT, TBT to that predicted by EBT under
uniformly applied load for various boundary conditions and values of the nonlocal parameter e0.

16
 (a) Simply supported (b) Clamped
1,50 0,30
Local EBT Local EBT
Nonocal EBT Nonlocal EBT
1,45 Local Present HSBT 0,29 Local Present HSBT
D im e n s io n le s s d e fle c tio n ,w

D im e n s io n le s s d e fle c tio n ,w
Nonocal Present HSBT Nonlocal Present HSBT

1,40 0,28

1,35 0,27

1,30 0,26

1,25 0,25
10 20 30 40 50 10 20 30 40 50
Aspect, L/h Aspect, L/h

(c) Clamped-Free (d) Clamped-Simply supported

12,8
0,62 Local EBT
12,7
Nonocal EBT
0,61
12,6 Local Present HSBT
D im e n s io n le s s d e fle c tio n ,w
D im e n s io n le s s d e fle c tio n ,w

0,60 Nonocal Present HSBT

12,5

12,4 0,59

12,3 0,58

12,2 0,57

12,1
0,56
Local EBT
12,0 Nonocal EBT 0,55
11,9 Local Present HSBT
Nonlocal Present HSBT 0,54
11,8
10 20 30 40 50 10 20 30 40 50
Aspect, L/h Aspect, L/h

Figure 2: Effect of the aspect ratio on dimensionless deflection for uniform load of homogeneous nanobeams
for 𝑒 𝑎 = 1𝑛𝑚

17
 (a) Simply supported (b) Clamped
1,9 0,31
L/h=10 L/h=10
L/h=20 L/h=20
1,8
L/h=40 0,30 L/h=40
L/h=60 L/h=60
D im e n s io n le s s d e fle c t io n

D im e n s io n le s s d e fle c tio n
1,7

0,29
1,6

1,5 0,28

1,4
0,27

1,3
0,26
0,0 0,5 1,0 1,5 2,0 0,0 0,5 1,0 1,5 2,0
e0a e0a

(c) Clamped-Free (d) Clamped - Simply supported

13,0 0,75
L/h=10
0,73
L/h=20
12,5 0,71 L/h=40
D im e n s io n le s s d e fle c tio n

L/h=60
D im e n s io n le s s d e fle c t io n

0,69
12,0
0,67

0,65
11,5
0,63

0,61
11,0
0,59
L/h=10
10,5 L/h=20 0,57
L/h=40 0,55
L/h=60
10,0 0,53
0,0 0,5 1,0 1,5 2,0 0,0 0,5 1,0 1,5 2,0
e0a e0a

Figure 3: Effect of the nonlocal parameter on dimensionless deflection for uniform load.

5. Conclusions
This study investigated the bending behavior of nanobeams based on the nonlocal third-order shear
deformation theory. Governing equations and boundary conditions were derived using the principle of virtual
displacements.
We developed a mathematical solution for various boundary conditions and established exact formulas for
deflections. The influence of the nonlocal parameter and aspect ratio on the static responses of nanobeams was
explored. Numerical results underscored the significant role played by nonlocal effects in the static responses of
nanobeams. The new nonlocal beam model predicted larger deflections compared to the classical (local) beam
model. Therefore, it is imperative to consider small-scale effects (or nonlocal effects) when analyzing the

18
mechanical behavior of nanostructures. Additionally, the judicious selection of the nonlocal parameter's value is
crucial to ensure the validity of nonlocal beam models.
The non-local size effects on the behaviour of nano-beams reveal contrasting mechanisms depending on the
boundary conditions, as shown by our results and previous studies. In the case of SS, the softening effect of the
small-scale terms is well demonstrated, in agreement with the results obtained by Reddy (2007) for non-local
higher order shear beams. This phenomenon reduces the apparent stiffness of the beam, leading to an increase in
displacements. Similarly, in the CC case, a softening effect is observed, in agreement with the results obtained by
Reddy and Pang (2008) for non-local Euler-Bernoulli (EB) and Timoshenko (TB) beams. On the other hand, a
stiffening effect is observed in the CF case, a paradoxical phenomenon already reported by Peddieson et al. (2003)
and Reddy and Pang (2008). Finally, in the CS case, a stiffening effect is also observed, in agreement with the
results of Reddy and Pang (2008) for non-local Euler-Bernoulli (EB) and Timoshenko (TB) beams. These results
highlight the importance of non-local effects in the mechanical behaviour of nanobeams, and the need for accurate
modelling to predict their load response.
One possible explanation lies in the nature of the boundary condition at the free end. Unlike the SS and CC
cases where the whole beam is stressed, here the free end allows a different redistribution of stresses and non-
local effects. In some cases, taking into account long-range interactions can lead to an increase in apparent
stiffness, in particular due to a modified coupling between internal forces and displacements. This stiffening can
be linked to bending redistribution effects and to the fact that non-locality tends to limit strong curvature
variations in a non-embedded zone.
This paradox merits further analysis, as it shows that the effect of small-scale terms is not always a simple
reduction in stiffness, but can vary depending on the boundary conditions and the type of model used.
In summary, this study advances our understanding of nanobeam bending and emphasizes the critical
importance of nonlocal effects in their mechanical behavior. These findings hold substantial implications for the
design and analysis of nanostructures across various fields of science and technology.

References
Aydogdu, M. (2009). A general nonlocal beam theory: its application to nanobeam bending, buckling and
vibration. Physica E: Low-dimensional Systems and Nanostructures, 41(9), 1651-1655.
https://doi.org/10.1016/j.physe.2009.05.014
Bickford, W. B. (1982). A consistent higher order beam theory. Dev Theor Appl Mech; 11:137–50.
Bresse, J. A. C. (1859). Cours de mécanique appliquée (Vol. 1). Gauthier-Villars.
Challamel, N., & Wang, C. M. (2008). The small length scale effect for a non-local cantilever beam: a paradox
solved. Nanotechnology, 19(34), 345703. DOI 10.1088/0957-4484/19/34/345703
Challamel, N. (2011). Higher-order shear beam theories and enriched continuum. Mechanics Research
Communications, 38(5), 388-392. doi: 10.1016/j.mechrescom.2011.05.004
Challamel, N. (2013). Variational formulation of gradient or/and nonlocal higher-order shear elasticity beams. Composite
Structures, 105, 351-368. http://dx.doi.org/10.1016/j.compstruct.2013.05.026
Challamel, N., Mechab, I., Elmeiche, N., Ahmed Houari, M. S., Ameur, M., & Atmane, H. A. (2013-a). Buckling of
generic higher-order shear beam/columns with elastic connections: local and nonlocal formulation. Journal of Engineering
Mechanics, 139(8), 1091-1109. https://doi.org/10.1061/(ASCE)EM.1943-7889.0000542.
Challamel, N., Kolvik, G., & Hellesland, J. (2013-b). Plate buckling analysis using a general higher-order shear
deformation theory. International Journal of Structural Stability and Dynamics, 13(05), 1350028. DOI:
10.1142/S0219455413500284
Challamel, N., & Elishakoff, I. (2019). A brief history of first-order shear-deformable beam and plate models. Mechanics
Research Communications, 102, 103389. https://doi.org/10.1016/j.mechrescom.2019.06.005
Di Sciuva, M. (2019). On the equivalence of displacement-based third-order shear deformation plate theories. Journal of
Engineering Mechanics, 145(7), 04019044.https://doi.org/10.1061/(ASCE)EM.1943-7889.0001616.
Demidovitch B., “Recueil d’exercices de problemes d’analyse mathematique,” Ellipses Marketing, 11-th edition, 1998,
pp.432–457.
Elishakoff, I. (2019). JP Den Hartog about SP Timoshenko: fifty years later. Mathematics and Mechanics of Solids, 24(5),
1340-1348. https://doi.org/10.1177/1081286518792959
Elishakoff I, Pentaras D, Dujat K, Versaci C, Muscolino G, Storch J, et al. Carbon nanotubes and nanosensors: vibrations,
buckling and ballistic impact. Wiley–ISTE; 2012.
Eringen, A. C. (1983). On differential equations of nonlocal elasticity and solutions of screw dislocation and surface
waves. Journal of applied physics, 54(9), 4703-4710. https://doi.org/10.1063/1.332803

19
 Lim, C. W., Zhang, G., & Reddy, J. (2015). A higher-order nonlocal elasticity and strain gradient theory and its
applications in wave propagation. Journal of the Mechanics and Physics of Solids, 78, 298-313.
https://doi.org/10.1016/j.jmps.2015.02.001
Lin, F., Tong, L. H., Shen, H. S., Lim, C. W., & Xiang, Y. (2020). Assessment of first and third order shear deformation
beam theories for the buckling and vibration analysis of nanobeams incorporating surface stress effects. International Journal
of Mechanical Sciences, 186, 105873. https://doi.org/10.1016/j.ijmecsci.2020.105873
Lu, P., Lee, H. P., Lu, C., & Zhang, P. Q. (2006). Dynamic properties of flexural beams using a nonlocal elasticity
model. Journal of applied physics, 99(7). DOI: 10.1063/1.2189213
Peddieson, J., Buchanan, G. R., & McNitt, R. P. (2003). Application of nonlocal continuum models to
nanotechnology. International journal of engineering science, 41(3-5), 305-312. https://doi.org/10.1016/S0020-
7225(02)00210-0
Pei, Y. L., & Li, L. X. (2021). An uncoupled theory of FG nanobeams with the small size effects and its exact
solutions. Archive of Applied Mechanics, 91, 1713-1728. https://doi.org/10.1007/s00419-020-01849-2
Polizzotto, C. (2015). A unifying variational framework for stress gradient and strain gradient elasticity theories. European
Journal of Mechanics-A/Solids, 49, 430-440. https://doi.org/10.1016/j.euromechsol.2014.08.013
Polizzotto, C. (2016). Variational formulations and extra boundary conditions within stress gradient elasticity theory with
extensions to beam and plate models. International Journal of Solids and Structures, 80, 405-419.
https://doi.org/10.1016/j.ijsolstr.2015.09.015
Reddy, J. (2007). Nonlocal theories for bending, buckling and vibration of beams. International journal of engineering
science, 45(2-8), 288-307. https://doi.org/10.1016/j.ijengsci.2007.04.004
Reddy, J. N. (1984). A simple higher-order theory for laminated composite plates. https://doi.org/10.1115/1.3167719
Reddy, J. N., & Pang, S. D. (2008). Nonlocal continuum theories of beams for the analysis of carbon nanotubes. Journal
of applied physics, 103(2). https://doi.org/10.1063/1.2833431
Ruocco, E., & Reddy, J. N. (2023). Analytical solutions of Reddy, Timoshenko and Bernoulli beam models: A
comparative analysis. European Journal of Mechanics-A/Solids, 99, 104953.
https://doi.org/10.1016/j.euromechsol.2023.104953
Shi, G., & Voyiadjis, G. Z. (2011). A sixth-order theory of shear deformable beams with variational consistent boundary
conditions. https://doi.org/10.1115/1.4002594
Şimşek, M., & Yurtcu, H. H. (2013). Analytical solutions for bending and buckling of functionally graded nanobeams
based on the nonlocal Timoshenko beam theory. Composite Structures, 97, 378-386.
https://doi.org/10.1016/j.compstruct.2012.10.038
Sudak, L. J. (2003). Column buckling of multiwalled carbon nanotubes using nonlocal continuum mechanics. Journal of
applied physics, 94(11), 7281-7287. https://doi.org/10.1063/1.1625437
Thai, H. T. (2012). A nonlocal beam theory for bending, buckling, and vibration of nanobeams. International Journal of
Engineering Science, 52, 56-64. https://doi.org/10.1016/j.ijengsci.2011.11.011
Timoshenko, S. P. (1920). On the differential equation for the flexural vibrations of prismatical rods. Glas. Hrvat.
Prirodosl. Drus., Zagreb, 32(2), 55-57.
Timoshenko, S. P. (1921). On the correction for shear of the differential equation for transverse vibrations of prismatic
bars. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 41(245), 744-746.
https://doi.org/10.1080/14786442108636264
Timoshenko, S. P. (1922). On the transverse vibrations of bars of uniform cross-section. The London, Edinburgh, and
Dublin Philosophical Magazine and Journal of Science, 43(253), 125-131. https://doi.org/10.1080/14786442208633855
Vaccaro, M. S., Barretta, R., Marotti de Sciarra, F., & Reddy, J. N. (2022). Nonlocal integral elasticity for third-order
small-scale beams. Acta Mechanica, 233(6), 2393-2403. https://doi.org/10.1007/s00707-022-03210-w.
Wang, C. M., Kitipornchai, S., Lim, C. W., & Eisenberger, M. (2008). Beam bending solutions based on nonlocal
Timoshenko beam theory. Journal of Engineering Mechanics, 134(6), 475-481. https://doi.org/10.1061/(ASCE)0733-
9399(2008)134:6(475)
Wang, C. M., Reddy, J. N., & Lee, K. H. (Eds.). (2000). Shear deformable beams and plates: Relationships with classical
solutions. Elsevier.
Wang, C. M., Zhang, Y. Y., Ramesh, S. S., & Kitipornchai, S. (2006). Buckling analysis of micro-and nano-rods/tubes
based on nonlocal Timoshenko beam theory. Journal of Physics D: Applied Physics, 39(17), 3904. DOI 10.1088/0022-
3727/39/17/029
Wang, Q. (2005). Wave propagation in carbon nanotubes via nonlocal continuum mechanics. Journal of Applied
physics, 98(12). https://doi.org/10.1063/1.2141648
Wang, Q. (2005). Wave propagation in carbon nanotubes via nonlocal continuum mechanics. Journal of Applied
physics, 98(12). https://doi.org/10.1063/1.2141648

20
 Wang, Q., & Liew, K. M. (2007). Application of nonlocal continuum mechanics to static analysis of micro-and nano-
structures. Physics Letters A, 363(3), 236-242. https://doi.org/10.1016/j.physleta.2006.10.093
Wang, Q., & Wang, C. M. (2007). The constitutive relation and small scale parameter of nonlocal continuum mechanics
formodelling carbon nanotubes. Nanotechnology, 18(7), 075702. DOI 10.1088/0957-4484/18/7/075702
Wang, Q., Varadan, V. K., & Quek, S. T. (2006). Small scale effect on elastic buckling of carbon nanotubes with nonlocal
continuum models. Physics letters A, 357(2), 130-135. https://doi.org/10.1016/j.physleta.2006.04.026
Wang, X. D., & Shi, G. (2012). Boundary layer solutions induced by displacement boundary conditions of shear
deformable beams and accuracy study of several higher-order beam theories. Journal of Engineering Mechanics, 138(11),
1388-1399. https://doi.org/10.1061/(ASCE)EM.1943-7889.0000440
Wang CM, Reddy JN, Lee KH (2000). Shear deformable beams and plates: relationships with classical solutions. Oxford.
UK: Elsevier
Xu, M. (2006). Free transverse vibrations of nano-to-micron scale beams. Proceedings of the Royal Society A:
Mathematical, Physical and Engineering Sciences, 462(2074), 2977-2995. https://doi.org/10.1098/rspa.2006.1712
Zhang, Y. Y., Wang, C. M., & Challamel, N. (2010). Bending, buckling, and vibration of micro/nanobeams by hybrid
nonlocal beam model. Journal of Engineering Mechanics, 136(5), 562-574. https://doi.org/10.1061/(ASCE)EM.1943-
7889.0000107

21