Electro-Optical Simulation Of In Ultra-Thin
Photonic Crystal Amorphous Silicon Solar Cells
Abdelhak Merabti
Department:
Exact Sciences. �cole Normale Sup�rieure
B�char, Algeria, BP 417, Bechar, Algeria
E-mail:
merabti73@yahoo.com
Abdelkader Bensliman
�Department: Exact Sciences. �cole Normale Sup�rieure
B�char, Algeria, BP 417, Bechar, Algeria
E-mail:
[email protected]
Mahmoud Habab
Department:
Exact Sciences. �cole Normale Sup�rieure
B�char, Algeria, BP 417, Bechar, Algeria
Abstract
Hydrogenated amorphous Si (a-Si:H) is an important
solar cell material. The critical problem in the a-Si:H-based photovoltaic cell
is increasing the conversion efficiency. To overcome the difficulty,� higher conversion efficiency demands a longer
optical path� to increase optical
absorption. Thus, a light trapping�
structure is needed to obtain more efficient absorption. In this
context, we propose a complete solar cell structure for which a 1D grating is
etched into the ultrathin active absorbing layer of a one-dimensional "CP
1D" photonic crystal a-Si: H characterized by the optimal parameters:
period a = 480 nm, a filling factor ff = 50% and a depth d = 150 nm. This was
selected by varying the CP1D parameters to maximize the absorption integrated
into the active layer. CP1D is suggested as an intermediate layer in the solar
cell concentration system. This study allowed us to model the optical and
electrical behavior of a CP1D solar cell. After optimization of the geometrical
parameters (period and fill factor ... etc.), we concluded that the CP1D led to
greater optical gains than for their unstructured equivalent. The simulation
clearly illustrates that the electric field strongly affects the
electro-optical characteristics of the devices studied, and that it is clear
that 1D PC solar cells as active layer have exhibited a high electric field
distribution. We have focused on the net on the effect of the active layer and
its beneficial role in the sense of expressing the photovoltaic performance of
the devices.
Keywords: Photonic
crystal; finite element method (FEM); Absorption; dimensional; electro-
optical.�������������������������
I. Introduction
With the development of
thin-film solar cells, an important step has been taken in reducing costs by
reducing the thickness of the active layer. However, the low absorption of
light, particularly for wavelengths near the gap of the absorbing material, was
quickly identified as the main limitation of these cells. Special attention has
therefore been given to these optical losses, which are even more critical when
considering ultrathin layers whose thicknesses are of the order of a
micrometer.
To increase the light
harvesting efficiency, classical technologies combine the integration of an
anti-reflection film (ARF),a textured top surface, and a reflector on the
backside [1-4]. Traditional light trapping schemes, used in photovoltaic cells,
are based on geometrical optics. Furthermore, the light trapping approaches
based on wave optics are capable of surpassing geometrical optics approaches in
some cases [5].
In the past years, many wave
optics light-trapping techniques have been explored such as plasmonics based
designs [6,7], scattering into guided modes by metal nanoparticles [8], grating
couplers [9], and pop-tonic crystals (PCs) (in 1D [10], 2D [11] et 3D [12]).
In this study, we will first
describe the best structure of an active layer of a one-dimensional photonic
crystal (CP1D) [13]. In a second step, the study of a one-dimensional photonic
crystal (CP 1D) TE polarization allows better understanding how the electric
field and the electrical potential evolve, through the CP 1D. Finally, we study
the influence of temperature on the electrical properties of a one-dimensional
photonic crystal.
II. Calculation Method
The
Finite element method (FEM) is the method of choice for analysis, complex
geometries and fast simulations of light interaction with photonic crystal
[14].
All
light information is contained in electromagnetic fields. The finite element
method (FEM) performs rigorous simulations of Maxwell's equations. This way of
solving the Maxwell equations makes it possible to calculate the reflectance R,
the transmittance T, and therefore the absorption A = 1-RT of a plane wave
incident on our structures. Due to the modal properties of the PC 1D developed
above. After, we will do most calculations at normal incidence.
In
our study, the finite element method solves the following partial differential
equation that derives from Maxwell's equations:
Where
���
���
�
In
practice, a calculation frequency, a numerical convergence criterion and a
maximum number of iterations are specified. In order to have an optimal
accuracy during a frequency sweep, a high main frequency is generally chosen..
III. Proposed Structure (Cp Solar Cell)
III.1 CP solar cell structure
We introduce the stack of layers constituting our
solar cells by justifying the choice of retained materials and the thickness of
the layers.
The structure consists,
from the front face to the rear face of the cell, of a layer of zinc oxide (ZnO),
of the active layer (a-Si: H), of a layer of the most currently used is indium
oxide doped with tin (In2O3-SnO2, ITO). Because it combines good transparency
in the visible range and good electrical conductivity (TCO), a layer aluminum
(Al), and finally a glass optical supports. A grating is then formed in the
zinc oxide (ZnO) layer and in the active layer to form the
"photonized" solar cell shown in Figure IV.1.
TM D (240nm) TE ITO (80nm) Al (60nm) glass d(150nm) ZnO(90nm)
Figure 1: structure of the solar cell studied with its
main geometrical parameters
III.2 Optical indices of different
materials
We will present here the optical indices that we
will use for the numerical study of the cell. The refractive index (n) and the
extinction coefficient (k) w ere measured by [15-17] for ITO, Al, ZnO and a-Si:
H (Figure IV.2).
|
a |
b |
||
|
b |
e |
||
|
Figure . 2 Optical indices of materials used for simulations numerically a) Al, b) ITO c) ZnO and e) a-Si:H |
|||
III.3 Parameters of the one-dimensional
Photonic Crystal solar cells
The objective of the
one-dimensional Photonic Crystal solar cells structuration is to be able to
work with an ultrafine active layer by taking advantage of the resonances of
the photonic crystal to generate additional absorption over the entire useful
spectrum.
The ZnO layer has a
thickness of 90nm, which ensures efficient lateral transport of charges over
several micrometers to the metal contacts, while limiting the parasitic
absorption of light in this layer. The thickness of the active layer is only
250nm.
To limit the roughness
on the surface of the ITO and to create a barrier against the diffusion of Al
in the silicon, we chose a thickness of 80 nm for the backside ITO.
Finally, the Al layer
is 60 nm thick. This choice is motivated by the need to have a great reflection
of the incident light and a small volume [4].
From an optical
point of view, the ZnO layer is an optical spacer. The active layer is the
absorbent layer and finally, the ITO layer contributes to the reduction of
reflection in the rear face thanks to its refractive index. Note that opaque
electrodes used in devices play a role of mirror.
We injected a plane
wave at normal incidence and we obtained the absorption by performing an energy
balance between the transmitted light power and the power reflected by the
cell. We calculated the absorption spectra of the nanostructured cell for the
optimal parameters of the proposed structure. By examining the profile of these
spectra in TE polarization and TM polarization, we distinguish:
|
|
|
Figure 3: Absorption spectra for the
two polarizations (TE and TM) corresponding to an unstructured layer |
The spectra
corresponding to the structured solar cell differ greatly from the
(unstructured) reference spectrum, and the spectrum pattern depends on the
polarization of the light. However, we see in Figure IV.3 that the
superposition of curves in TE and TM polarizations is not perfect beyond 580nm.
The absorption
increases very rapidly, reaching a first maximum for λ = 420 nm TE
polarization and λ = 520 nm for TM polarization. It then decreases
slightly, returning to a maximum at 730 nm for TE. We can explain this behavior
by the phenomena of interference related to the waves reflected at the
different interfaces of the cell
We can obtain a maximum
of light intensity for a wavelength that gives rise to constructive
interference in the active layer, while destructive interference can degrade
the absorption for a layer of different thickness.
There is a significant
decrease in absorption above 560 nm. We can find the increase in absorption at
low wavelengths thanks to the network-induced antireflection effect and the
creation of absorption peaks at long wavelengths through the coupling of
incident light with CP slow Bloch modes
Thus, the presence of
an Al layer on the rear face in combination with the optical spacer makes it
possible to modulate the coupling force and to create constructive interference
(reinforced coupling) or destructive interference (attenuated coupling) in the
active layer.
We note that the
characteristics of the spectra are little modified whatever the thickness of
the layer. Note, however, that the absorption is slightly degraded from 450 nm
to 550 nm. We can assume that in this wavelength range, the electric field is
unfavorably redistributed in the active layer with respect to the case where
the entire layer is structured.
V. Electrical modeling of one-dimensional
Photonic Crystal sollar cells
The understanding of
electrical phenomena in is essential for the development of a one-dimensional
Photonic Crystal solar cell.
The object of this part
is to introduce the properties of the electric field and current density in a
CP 1D TE polarization with cutting line a) width / 2 and b) width / 3.
|
a |
b |
||
|
Figure 3: cutting line a) width / 2
and b) width / 3 |
|||
V.1 Electric field
Figures IV-4 and IV-5
show electric field growth versus position. Indeed, as seen in the previous
part, the electric field is directly proportional to the applied voltage. It is
then expected to grow the electric field with increasing position. In addition,
we can note that the electric field calculated for the two cutting lines is
zero in the glass layer. The electric field increases rapidly as one moves away
from the glass layer to reach the maximum value of 130 V / m at the layer a-Si:
H. It is important to specify that the electric field is not calculated in the
metals, since none. We will focus only on the value of the electric field at
the active layer and its evolution in the ZnO layer.
Simulation thus gives
important results such as the importance of the thickness of the layer. It also
shows the dependence of the electric field on the shape of the layer.
Figure IV -6 also shows the electric field
distribution for TE polarization.
|
|
|
|
Figure
4: Electric field in a CP 1D polarization TE with cutting line (width / 2) |
|
|
|
|
Figure
5: Electric field in a CP 1D TE polarization with cutting line (width / 3) |
|
a |
b d |
|||
|
|
|
|||
|
c |
d |
|||
|
|
|
|||
|
Figure 6: Electrical field distribution for CP 1D TE
polarization cell: a) 350nm, b)400nm, c)550nm and d)700nm |
||||
The maximum and
minimum surface area of the electric field specified in Table IV.1
Table 1 The maximum and the minimum surface of the
electric field.
|
TE polarization |
||
|
Wave length |
Electric field
surface Max (V/m) |
Electric field
surface Min (V/m) |
|
|
225.173 |
0 |
|
|
180.077 |
0 |
|
|
162.23 |
0 |
|
|
341.195 |
0 |
V.2�
Current density
Figures IV-7 and IV-8 show the evolution of the
current density with the position. The density profile of the current obtained
in the cutting line then has a characteristic Gaussian shape. The peak
positions of the current densities are generally well reproduced. On the other
hand, significant differences in the current density amplitude exist. A
significant increase in current density in the active layer results in a lower
rate of carrier recombination.
Figure IV -9 also shows the current density
distribution for TE polarization.
|
|
|
|
Figure
7 Current density of CP 1D polarization TE with cutting line (width /2) |
|
|
|
|
Figure
8: Current density of� CP 1D
polarization TE with cutting line (width / 3) |
|
a d c |
b |
||||
|
|
|
||||
|
c a |
d |
||||
|
|
|
Figure 9: Current density
distribution for CP 1D TE polarization cell: a) 350nm, b) 400nm, c) 550nm and
d) 700nm
The maximum and minimum area of the current density specified
in Table IV. 2
Table 2 The maximum and the minimum
surface of the current density
|
�TE polarization |
||
|
Wave length |
current
density surface Max (A/m2) |
current
density surface Min (A/m2) |
|
|
1.408E7 |
0 |
|
|
7.3081E7 |
0 |
|
|
5.43.38E7 |
0 |
|
|
5.1201E7 |
0 |
Conclusion
This study
allowed us to model the optical and electrical behavior of a CP1D solar cell.
After optimization of
the geometrical parameters (period and filling factor ... etc.), we concluded
that the CP1D led to greater optical gains than for their unstructured
equivalent.
The simulation clearly
illustrates that the electric field strongly affects the electro-optical
characteristics of the devices studied, and that it is clear that the 1D PC
solar cells as active layer have exhibited a high electric field distribution.
In this part, we have
focused on the net on the effect of the active layer and its advantageous role
in the sense of expressing the photovoltaic performances of the devices.
The points mentioned
above tend to show that our approach makes it possible to significantly
increase the integrated absorption of the cells and to improve their
efficiency. The simulation results also highlight the feasibility of the
proposed structures.
In perspective, we
would like to check other two-dimensional models (2D), three-dimensional (3D)
and other types of organic or inorganic solar cells to have a very concrete and
quantitative study
ACKNOWLEDGMENT
This work is supported by �cole normale sup�rieure
B�char
R�F�RENCES
[1]� H. Ganesha Shanbhogue, C. L. Nagendra, M. N.
Annapurna, S. Ajith Kumar,and G. K. M. Thutupalli, �Multilayer antireflection
coatings for the visible and near-infrared regions,� APPLIED OPTICS 36(25),
6339�6351 (1997).
[2]� Martin F. Schubert, Frank W. Mont, Sameer
Chhajed, David J. Poxson, Jong Kyu Kim, and E. Fred Schubert, �Design of
multilayer antireflection coatings made from co-sputtered and low-refractive
index materials by genetic algorithm,� OPTICS EXPRESS 16 (08), 5290�5298
(2008).
[3]�� Lesley Chan, Dongseok Kang, Sung-Min Lee,
Weigu Li, Hajirah Hunter, and Jongseung Yoon, �Broadband antireflection and
absorption enhancement of ultrathin silicon solar microcells enabled with
density-graded surface nanostructures,� Applied Physics Letters 104, 223905
(2014); 10.1063/1.4881260
[4]� Huihui Yue, Rui Jia, Chen Chen, Wuchang Ding,
Yanlong Meng, Deqi Wu, Dawei Wu, Wei Chen, Xinyu Liu, Zhi Jin, Wenwu Wang, and
Tianchun Ye, �Antireflection properties and solar cell application of silicon
nanostructures,� J. Vac. Sci. Technol, 08 (01), 46�52 (2011).
[5]� Jiun-Yeu Chen, �Design of photonic crystal
enhanced� light-trapping structures for
photovoltaic cells,� Journal of Engineering Technology and Education, 23 (03),
031208 (2011).
[6]�� Harry�
a.� Atwater and� Albert Polman �Plasmonics for improved
photovoltaic� devices,� Nat. Mater.� 9 (3), 205�213 (2010).
[7]� Yuanyuan Li, Jian Pan, Peng Zhan, Shining
Zhu, Naiben Ming, Zhenlin Wang, Wenda Han, Xunya Jiang ,and Jian Zi, �Surface
plasmon coupling enhanced dielectric environment sensitivity in a
quasi-three-dimensional metallic nanohole array,� Opt. Express 18(04),
3546�3550 (2010).
[8]� Peter N. Saeta, Vivian E. Ferry, Domenico
Pacifici, Jeremy N. Munday, and Harry A. Atwater, �How much can guided modes
enhance absorption in thin solar cells?� Opt. Express 17(23), 20975�20990
(2009).
[9]�� S. H. Zaidi, J. Gee, and D. S. Ruby,
�Diffraction grating structures in solar cells,� in Twenty-Eighth IEEE
Photovolt. Spec. Conf. (2000), pp. 395�398.
[10]� J. G. Mutitu, S. Shi, C. Chen, T. Creazzo, A.
Barnett, C. Honsberg, and D. W. Prather, �Thin film solar cell design based on
photonic crystal and diffractive grating structures,� Opt. Express 16(19),
15238�15248 (2008).
[11] O. Painter, R.K.
Lee, A. Scherer, A. Yariv, J.D. O�Brien, P.D. Dapkus, I. Kim, �Two-Dimensional
Photonic Band-Gap Defect Mode Laser,�Science,284, 1819-1821, (1999).
[12] Leo� T. Varghese,�
Yi� Xuan, Ben Niu, Li Fan, Peter
Bermel, and Mi nghao Qi, �Enhanced photon management of thin-film silicon solar
cells using inverse opal photonic crystals with 3D photonic bandgaps,�Adv
Optical Mater. 2013;1:692 � 8.
[13]� A.Merabti, , A.Hasni, and M.Elmir, �Numerical
approach of the influence of geometric properties on the absorbing in photonic
crystal,� JOURNAL OF NANO- AND ELECTRONIC PHYSICS, vol. 8 No 8, 03046(5pp)
(2016).
[14]� S. Guenneau, A. Nicolet, F. Zolla, and S.
Lasquellec, �Modeling of Photonic Crystal Optical Fibers With Finite Elements,�
IEEE Trans. Magn. , vol. 38, No. 2,� Mar.
2002.
[15]
refractiveindex.info, �Optical constants of In2O3-SnO2 (Indium tin oxide, ITO�
(2016). [En ligne]. Available:
http://refractiveindex.info/?shelf=other&book=In2O3-SnO2&page=Moerland
[16]
refractiveindex.info, �Optical constants of Al (Aluminium)� (2016). [En ligne].
Available: http://refractiveindex.info/?shelf=main&book=Al&page=Rakic
[17]
refractiveindex.info, �Optical constants of ZnO (Zinc oxide)� (2016). [En
ligne]. Available:
http://refractiveindex.info/?shelf=main&book=ZnO&page=Bond-o
Copyrights
Copyright for this article is retained by the author(s), with
first publication rights granted to the journal.
This is an open-access article distributed under the terms and
conditions of the Creative Commons Attribution��
license (http://creativecommons.org/licenses/by/4.0/)