Dispersion HIE-FDTD method for simulating
graphene-based absorber
ISSN 1751-8725
Received on 3rd November 2015
Revised on 26th May 2016
Accepted on 3rd July 2016
doi: 10.1049/iet-map.2015.0707
www.ietdl.org
Ning Xu, Juan Chen, Jianguo Wang
✉
, Xijia Qin, Jiepei Shi
Department of Electronic and Information Engineering, Xi’an Jiaotong University, Xi’an, People’s Republic of China
✉ E-mail: wanguiuc@163.com
Abstract: A dispersion hybrid implicit–explicit finite-difference time-domain (HIE-FDTD) method is proposed for simulating
the graphene-bas ed absorber at terahertz frequencies. The surface conductivity of the graphene derived from the Kubo
formula is incorporated into the conventional HIE-FDTD algorithm by using the auxiliary differential equation
technique. The time step size of this method is not related with the fine cells inside the graphene layer. The simulation
result of the i nfinite graphene sheet shows that the dispe rsion HIE-F DTD method has high computational accuracy and
the computational time is greatly reduced compared wi th the dispersion FDTD method. By using the dispersion HIE-
FDTD method, it is validated that the graphene sheet can be used as a broadband absorber.
1 Introduction
Graphene [1] is a two-dimensional (2D) atomic layer of carbon atoms
in the hexagonal pattern. Compared with noble metals widely
employed in modern electronics, graphene presents extraordinary
physical properties such as tunability, extreme confinement, and low
losses [2–7]. One of the most advantageous features is that the
conductivity of the graphene is tunable due to its specific
hybridisation. Furthermore, graphene is linear dispersive, which
leads to high electron mobility. With the unique characteristics,
graphene has attracted much interest recently. Hence, there is a
growing need for studying and developing the application of
graphene in the area of microwave and terahertz (THz) spectra.
Among the microwave and THz devices, absorber is one of the
most important applications. Conventional absorbers are generally
realised by using metals and other composite structures, rarely
designed by using graphene. Considering the distinctive properties
of graphene, the graphene-based absorber promise enhanced
performances. As a result, accurate modelling and simulation of
the graphene-based absorber will be highly desirable.
It is known that some analytical and numerical methods have been
developed to simulate THz graphene-based devices [8–19]. Among
them, the dispersion finite-difference time-domain (FDTD) method
is relatively common [8, 9]. However, considering that graphene is
a one-atom thick layer, the dispersion FDTD method needs an
extremely fine spatial discretisation inside the graphene layer to
obtain accurate simulation results. As the Courant–Friedrich–Levy
(CFL) stability condition [9] is mainly limited by the smallest cell
size, which results in a long computational time, the dispersion
FDTD method is computationally expensive when it is used to
simulate graphene devices.
To solve this problem, some improved dispersion FDTD methods
have been proposed [15, 16]. In [15], the graphene sheet is
modelled as a polarisation current source by using the Dirac delta
function. With the help of equivalent circuit representation, the
intraband electron transition of the graphene was considered in [16].
The graphene layer does not need to be discretised across its
thickness by using these methods, so the fine spatial discretisation
inside the graphene layer can be avoided. However in these
methods, the infinite graphene sheets are required, so they cannot be
used to simulate practical graphene devices, such as the
graphene-based absorber.
To deal with the problem of the CFL stability condition for the
dispersion FDTD method, a new hybrid implicit–explicit
(HIE)-FDTD method has been proposed [20–24]. The time step
size of this method is only related with two space discretisations
instead of the smallest one, so it is extremely efficient for
simulating the graphene-based devices which have a thin layer
only in one direction.
Considering that the graphene is dispersive, here we develop a
novel HIE-FDTD method, namely dispersion HIE-FDTD method.
The auxiliary differential equation (ADE) technique, which is
often used to represent the surface conductivity of graphene, is
incorporated into the HIE-FDTD method. As the time step size in
the proposed method is not confined by the smallest space cell, it
has higher computational efficiency than the conventional
dispersion FDTD method.
This paper is organised as follows. In Section 2, the 3D
computation formulas for the dispersion HIE-FDTD method are
given, meanwhile, the update equations are presented. In Section
3, in order to verify the accuracy of the dispersion HIE-FDTD
method, an infinite graphene sheet is simulated both by the
dispersion HIE-FDTD method and the dispersion FDTD method.
The results show that the dispersion HIE-FDTD method has high
accuracy, and it is extremely more ef ficient than the dispersion
FDTD method due to its large time step size. In Section 4, this
method is used to simulate the graphene-based absorber. Finally,
some conclusions are drawn in Section 5.
2 Formulations
In the method, graphene can be described as an infinitesimally thin
conducting surface, which is located in the y–z plane with one
biasing electrostatic field in the x-axis. The surface conductivity of
the graphene involves the intraband conductivity and the interband
conductivity. However, in microwave and THz frequency range,
the conductivity is dominated by the intraband conductivity which
can be obtained from the Kubo formula [25–27]
s =−
je
2
K
B
T
p
h
−
2
v
− j2G
×
m
c
K
B
T
+ 2ln e
−(
m
c
/K
B
T)
+ 1
(1)
here, e is the electron charge; K
B
is the Boltzmann constant; T is the
operating temperature in Kelvin; h
−
is the reduced Planck’s constant;
ω is the angular frequency; G = 1/ 2
t
()is the scattering rate and
t
is
the phenomenological scattering time; μ
c
is the chemical potential
IET Microwaves, Antennas & Propagation
Research Article
IET Microw. Antennas Propag., 2017, Vol. 11, Iss. 1, pp. 92–97
92
&
The Institution of Engineering and Technology 2016