See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/324963563

Light-absorption enhancement design of ultrathin perovskite solar cells with

conformal structure

Article  in  Journal of Physics D Applied Physics · May 2018

DOI: 10.1088/1361-6463/aac2b1

CITATIONS READS

0 375

8 authors, including:

Xinyu Tan
China Three Gorges University
24 PUBLICATIONS   205 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Long nocoding RNA View project

U1765105 View project

All content following this page was uploaded by Xinyu Tan on 27 June 2018.

The user has requested enhancement of the downloaded file.

Journal of Physics D: Applied Physics

PAPER

Light-absorption enhancement design of ultrathin perovskite solar cells

with conformal structure
To cite this article: Xinyu Tan et al 2018 J. Phys. D: Appl. Phys. 51 245101

View the article online for updates and enhancements.

This content was downloaded from IP address 111.63.11.12 on 22/05/2018 at 05:04

 Journal of Physics D: Applied Physics

J. Phys. D: Appl. Phys. 51 (2018) 245101 (8pp) https://doi.org/10.1088/1361-6463/aac2b1

Light-absorption enhancement design of

ultrathin perovskite solar cells
with conformal structure
Xinyu Tan1 , Lei Sun1, Can Deng1, Yiteng Tu2, Guangming Shen3,
Fengxue Tan3, Li Guan3 and Wensheng Yan4
1
  Key Laboratory of Inorganic Nonmetallic Crystalline and Energy Conversion Materials (CTGU),
College of Materials and Chemical Engineering, China Three Gorges University, 8 University Avenue,
Yichang 443002, People’s Republic of China
2
  College of Electrical Engineering and New Energy, China Three Gorges University, Hubei Provincial
Collaborative Innovation Center for New Energy Microgrid, 8 University Avenue, Yichang 443002,
People’s Republic of China
3
  Department of Physics Science and Technology, Hebei University, 180 Wusi E Rd, Baoding 071000,
People’s Republic of China
4
  Karlsruhe Institute of Technology, Institute of Microstructure Technology,
Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany

E-mail: lguanl@163.com (L Guan) and wensheng.yan@kit.edu (W Yan)

Received 9 February 2018, revised 16 April 2018

Accepted for publication 4 May 2018
Published 21 May 2018

Abstract
We report a structural design of ultrathin perovskite solar cells based on a conformal structure
at the rear surface for potential applications in both single-junction and tandem cells. The light
transmittances of the front and the rear surfaces are calculated in the wavelength range of
300–800 nm via the finite difference time domain numerical simulation method. Compared
with the reference cell, significant photocurrent density enhancement of 27.5% and 29.7%
are achieved when the ratios of height to width of the fluorine doped tin oxide nanoblock
are 2 and 3, respectively. For the case with a hole transport material layer, the enhancements
of photocurrent density enhancements are 19.2% and 29.0%, respectively. When back Au
is removed, the photocurrent density also has notable enhancements of 23.3% and 45.9%,
respectively. The achieved results are beneficial for the development of efficient ultrathin
single-junction and tandem perovskite solar cells.

Keywords: solar cells, perovskite, light absorption, finite difference time domain, thin films,
nanostructures

(Some figures may appear in colour only in the online journal)

1. Introduction with a rapid efficiency progressing from 3.8% in 2009 to

22.7% [2, 3]. Key challenges like instabilities against the
It is expected that photovoltaics (PV) can play a major role humidity and heat have also made important progress [4,
as part of the future renewables-based energy mix. As a con- 5]. It has been widely accepted that the perovskite solar
ventional PV device, crystalline silicon (c–Si) solar cells have cells are the most promising candidate to achieve high
dominated the global PV market with a share of  >90%5 [1]. power conversion efficiency (PCE) at low cost.
However, in the past several years, a kind of new-type solar At present, high-PCE perovskite solar cells are typically
cells, perovskite solar cells, has attracted lots of attention based on lead-halide-perovskite absorbers, where a toxic ele-
5
See www.itrpv.net for the international technology roadmap for photovol- ment of Pb is involved. In addition, the conventional thickness
taic results. of perovskite materials like CH3NH3PbI3 or CH(NH2)2PbI3

1361-6463/18/245101+8$33.00 1 © 2018 IOP Publishing Ltd  Printed in the UK

J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Figure 1.  Configuration of the proposed ultrathin perovskite solar cell architectures. (a) Schematic of the typical planar perovskite solar
cell; (b) 3D schematic of the conformal structured perovskite solar cell with back Au; (c) a cross-sectional view of the perovskite solar
cell, where one periodic simulation region is indicated using the red rectangle. In (c), h and w represent the height and width of the FTO
nanoblock, respectively. The ht and hp are defined as the thicknesses of the TiO2 and perovskite absorber layer, respectively. P represents
periodical length. The arrow represents the incident light.

for solar cells is 300 nm and even above [6, 7]. Recently, some deposition principle. (In the experiment, there is always a
researchers have turned to focus on the perovskite solar cells difference to some extent, particularly, the perovskite thick-
with ultrathin absorber layers less than 300 nm and even less ness at side walls. But, the difference between the experi-
than 100 nm [8–12]. The investigations on ultrathin perovskite ment and ideal case is not considered in the present work.)
solar cells are driven by various motivations. For example, Also, it implies that using same deposition time or con-
ultrathin perovskite solar cells are flexible and light-weight suming equal resource, the conformal structure can produce
[8, 9], which can extend the application to a wider range. Also, larger amount of the active layer. For this case, the light
adopting ultrathin perovskite absorber can reduce the amount absorption enhancement is not only from conformal struc-
of toxic lead element. For instance, the amount of toxic ele- ture but also from the increase of the active amount. The
ment of the 100 nm-thick CH3NH3PbI3 absorber is one third second case is that we set an equivalent volume of perovs-
as low as a 300 nm-thick perovskite absorber. From the kite between planar and conformal cells, which can provide
point of view of carrier recombination, an additional benefit a direct comparison regarding the role of conformal struc-
adopting ultrathin CH3NH3PbI3 layer can reduce the carrier ture in enhancing light absorption. Our results show that a
recombination loss, which is due to the fact that the diffusion significant absorption enhancement is achieved by using
length of electrons and holes of the CH3NH3PbI3 absorber is conformal structure solar cell architecture.
approximately 100 nm [14]. Besides the above advantages,
the investigations on ultrathin perovskite solar cells also are
significant with potential applications in other aspects. For 2.  Modelling and simulation of ultrathin perovskite
example, when the perovskite absorber becomes ultrathin, solar cells
the perovskite layer will become more transparent (or even
semi-transparent), which is promising for the applications in It should be pointed out that although the majority of research
tandem solar cells like perovskite/c–Si tandem cells (or CIGS) includes a hole transport material (HTM) layer, literatures
[15] and all-perovskite tandem cells [16], because it allows where perovskite can also be used as HTMs have been
more light to be absorbed by the bottom sub-cell. reported and perovskite solar cells without HTM or ETM can
The above-mentioned promising applications of ultrathin be fabricated [12, 17–19]. Therefore, in this work, we investi-
perovskite solar cells are at the cost of sacrificing the PCE gate two cases: (i) the light-trapping design without the con-
because of its ultrathin perovskite absorption layer. Hence, sideration of the HTM layer and (ii) the light-trapping design
it is necessary and significant to design high light absorption with the HTM (Spiro-OMeTAD) layer of 50 nm. In addition,
enhancement for ultrathin perovskite solar cells via novel cell we investigate the solar cell architectures with and without
architectures. In this work, we present design investigations full back Au in terms of back electrodes. The solar cells with
on light absorption enhancement for ultrathin perovskite solar full back Au is for single-junction photovoltaic application
cell with conformal architecture for potential applications in while the solar cells without full back Au can be for tandem
single-junction and tandem solar cells. The photocurrent den- photovoltaic application.
sity enhancement is systematically calculated by sweeping the Figures 1(a)–(c) show the models of ultrathin perovs-
geometry parameters. The present research results can benefit kite solar cells with planar structure and conformal struc-
from the development of efficient ultrathin perovskite solar ture. Figure 1 only shows the solar cell architecture without
cells. HTM layer. A 100 nm-thick ultrathin CH3NH3PbI3 absorber
In this work, we have investigated and compared two is chosen to investigate in this work. The reason choosing
cases. The first case is that the planar and conformal per- this thickness is relevant to both (i) trade off consideration
ovskite cells have equal perovskite thickness, which between the absorber thickness and the PCE and (ii) diffusion
assumes that conformal perovskite meets perfect conformal length of electrons and holes of the CH3NH3PbI3 absorber
2
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Table 1.  Simulation parameters of the perovskite solar cell.

Parameter
category Parameter name Parameter setting
Structure Glass 400 nm
FTO 400 nm
TiO2 50 nm
Perovskite 100 nm
Spiro-OMeTAD 50 nm
Au back contact 300 nm
FDTD simulation Mesh accuracy 2
region settings Mesh type Auto non-uniform
Mesh refinement Conformal variant 1
PML type Uniaxial anisotropic PML
Time step dt 0.99 (unitless)
stability factor
Time step dt 0.0087 fs
Minimum mesh 0.25 nm
Figure 2.  Wave length-dependent refractive index (n) and
step
extinction coefficient (k) for perovskite (CH3NH3PbI3).
Simulation time 1000 fs
of about 100 nm [13]. The conformal configuration is con- Plane wave source Source shape Plane wave
structed following a conformal deposition principle. The solar settings Amplitude 1 (Normalized)
cell architecture can be realized by using physical deposition Phase 0°
technique such as physical vapor deposition (PVD) [20] and Polarization 0°
angle
atomic layer deposition (ALD) [21], etc, with a conformal
Plane wave type Bloch/periodic
capacity in terms of fabrication. Time domain Broad band
Figure 1(a) is a planar perovskite solar cell, which is used pulse type
as the reference cell. Figures 1(b) and (c) are the three-dimen- Time domain 666.21 THz
sional (3D) view and cross-sectional view of the proposed frequency
conformal structured perovskite solar cells, respectively. In Pulse length 1.99 fs
the present work, the thicknesses of fluorine doped tin oxide Offset 5.66 fs
(FTO), TiO2, perovskite (CH3NH3PbI3), and Au back contact Band width 666.21 THz
are 400, 50, 100, and 300 nm, respectively. Note that the back
surface of Au electrode shown in the figures  1(b) and (c) is
flat (or even). If at a real conformal deposition, the back sur- In order to implement the present simulations, the refractive
face of Au electrode could be either uneven or approximately indices (n, k) of the materials are required. Optical constants of
flat depending on the specific Au thin film thickness. But, glass, FTO, TiO2, CH3NH3PbI3, and Spiro-OMeTAD are cited
this difference of back surface does not affect the accuracy of from the reported measurements [27, 28] whereas the optical
the present optical calculation results because the present Au constants of Au are taken from Palik’s handbook [29]. The
thickness is already optically thick enough. refractive index (n) and extinction coefficient (k) for perovs-
As shown in figure 1(c), the red region shows a simulation kite (CH3NH3PbI3) are plotted in figure 2. In terms of setting
region, where the incident light is from the bottom of glass the boundary conditions, we use periodic boundary conditions
substrate to up as indicated by an arrow. The height and width on the x-axis and y-axis and perfectly matched layers on the
of the FTO nanoblock are defined using the symbols h and w, z-axis. The detailed FDTD simulation parameter settings are
respectively. Because the perovskite solar cell with conformal listed in table 1.
structure is periodic, the simulation can be conducted by sim- Based on the FDTD method, the light absorption Pabs can
ulating one-periodic structure, where P is defined as periodic be calculated from the divergence of the Poynting vector by
length. Many different numerical methods have been used to using the following equation:
calculate absorptivity of solar cells such as transfer matrix Pabs = 0.5ωεim |E(ω)|2 ,
(1)
method (TMM) [22], finite-difference time-domain method
(FDTD) [23, 24], rigorous coupled-wave analysis (RCWA) where ω is the angular frequency, εim is the imaginary part of
[25] and finite element method (FEM) [26]. Compared with the permittivity and E is the electric field.
other numerical methods, the FDTD can be easy to calculate For the present simulations, the normalized light absorp-
the distribution of electromagnetic field of arbitrary material tivity in the perovskite absorber was determined by
and structure. So the FDTD method is employed to simu-
late the optical behaviors of perovskite nanostructures in this Pabs (λ)
A(λ) =
(2)
paper. Here, the Lumerical FDTD solutions are employed6. Pin (λ)
where Pin(λ) is the power of incident light within the solar cell
6
See www.lumerical.com/ for FDTD solutions. at a wavelength λ.

3
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Figure 3.  The calculated photocurrent density of ultrathin perovskite solar cells with conformal structure as function of the width w and
coverage percentage h of the FTO nanoblock. (a) The case of h/w  =  2; (b) the case of h/w  =  3.

Figure 4.  The calculated normalized light absorption of the conformal structured perovskite solar cells, where (a) is for h/w  =  2 and
h/w  =  3 with Au, and (b) is for h/w  =  3 with and without Au, respectively. The black line represents the normalized light absorption of the
reference cell.

The photocurrent density value is calculated by integrating Table 2.  Calculated photocurrent density of the solar cell with back
light absorption A(λ) over the wavelength as follows: Au and without Au but without HTL.

λ Jsc (mA
ˆ
(3) JSC = e A(λ)IAM1.5 (λ)dλ, Cell architectures cm−2) Enhancement
hc
Planar without Au 13.74
where e is the charge of an electron. h is Plank’s constant. c is
Planar with Au 17.3 Reference
the speed of light in the free space and A(λ) is the light absorp-
Conformal (h/w  =  2) without Au 17.69
tion as function of wavelength. IAM1.5 is air mass 1.5 global
Conformal (h/w  =  2) with Au 22.05 27.5%
solar spectrum.
Conformal (h/w  =  3) without Au 20.49
Conformal (h/w  =  3) with Au 22.43 29.7%
3.  Results and discussion

From equation (1), the light absorption of the conformal struc- shown in equation  (3), the calculated photocurrent density
tured perovskite solar cells can be calculated as function of as function of these variables can be plotted. For the present
wavelength in the range of 300–800 nm, where the height h photocurrent density calculations, we assume the external
and width w of the FTO nanoblock and the period P are varied quantum efficiency is 100%, which is a common practice in
to find the optimal parameters. By integrating the calculated the light-trapping design investigations. In this case, the calcu-
light absorption over the wavelength from 300 to 800 nm as lated photocurrent density is represented using the symbol Jsc.

4
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Figure 5.  The calculated photocurrent density of ultrathin perovskite solar cells with conformal structure which is equivalent to the volume
of 100 nm thickness perovskite as function of the width w and coverage percentage. (a) The case of h/w  =  2; (b) the case of h/w  =  3.

In this work, we investigate the two cases of the FTO with back Au and without back Au are listed in table 2. From
nanoblock with h/w  =  2 and 3, respectively. The w values table 2, for the case of h/w  =  2 with Au, the Jsc of 22.05 mA
are varied from 30 nm to 90 nm with intervals of 10 nm. The cm−2 is achieved, where an improvement of 4.36 mA cm−2 is
coverage percentage C of the perovskite shell is defined as obtained compared with the case of h/w  =  2 without Au. Also,
(w  +  2ht  + 2hp)2/P2, where the ht and hp are 50 and 100 nm, there is an improvement of 1.94 mA cm−2 compared with the
respectively. The range of C is set from 25% to 95% with case of h/w  =  3 without Au for the case of h/w  =  3 with Au.
intervals of 10%. The calculated Jsc as function of w and These comparison shows that the light absorption enhance-
C is shown in figures  3(a) and (b), which corresponds to ment in the longer wavelengths make a main contribution to
the cases of h/w  =  2 and 3, respectively. It can be found in the obtained optimal photocur­rent density enhancement.
figure 3 that the optimal parameters are obviously different in In the present simulations as shown in figure 4, we set the
the two cases. For the case of h/w  =  2, the optimal geometry light source outside the substrate glass and the glass thickness
param­eter w is 50 nm and C is 0.85, which leads to the Jsc is set as 400 nm. By comparison of planar perovskite cell with
of 22.05 mA cm−2. In contrast, for the case of h/w  =  3, the different light source positons, it is found that the setting dif-
optimal w is 80 nm and C is 0.95, which causes the Jsc of 22.43 ference can affect the peaks, particularly, the absorption curve
mA cm−2, where an improvement of 0.38 mA cm−2 is obtained can become gentle above 600 nm (not shown here). However,
compared with the case of h/w  =  2. For enhancement com- it should be pointed out that our comparison between the
parison analysis, the photocurrent density of the reference cell planar and conformal perovskite cells has the same simulation
is calculated with the result of 17.30 mA cm−2. It means that setting. Therefore, the absorption enhancement conclusion is
the relative enhancement in photocurrent density are 27.5% essentially constant.
and 29.5% for h/w  =  2 and 3, respectively. To prove the unique structure can make the perovskite
To identify specific reason of the optimal photocurrent absorb more light, the calculated photocurrent density of
density enhancement, the light absorption as a function of ultrathin perovskite solar cells with conformal structure which
wavelength in the range of 300–800 nm is shown in figure 4, is equivalent to the volume of 100 nm thickness perovskite are
where (a) is for h/w  =2 and h/w  =3 and (b) is for h/w  =3 discussed in figure  5. For the case of h/w  =  2, the optimal
with and without Au, respectively. For comparison, the light geometry parameter w is 50 nm and C is 0.5, which leads to
absorption of the reference cell is also shown in figure 4. It the Jsc of 19.45 mA cm−2. In contrast, for the case of h/w  =  3,
can be seen in figure 4 that the light-absorption enhancement the optimal w is 30 nm and C is 0.3, which causes the Jsc of
is mainly from the longer wavelengths. Specifically, there is 20.12 mA cm−2, where an improvement of 0.67 mA cm−2 is
nearly no difference of the light absorption in the short wave- obtained compared with the case of h/w  =  2. Compared with
length range of 300 to about 410 nm. From 410 to 550 nm, the reference cell, the relative enhancement in photocurrent
the nor­malized light absorption of the perovskite cell with density are 12.4% and 16.4% for h/w  =  2 and 3, respectively.
conformal cell architecture is obviously higher than that of So it demonstrates that the unique structure show excellent
the reference cell. However, the most significant enhance- light-trapping properties.
ment takes place in the longer wavelengths in the range of In addition to the above discussion, the light absorption
550–800 nm. From figure  4(b), the structure with Au show enhancement can be further understood from the point of view
better absorption in the range of 550–800 nm than the struc- of the light absorption density. It is expected that the design
ture without Au. The details of calculated photocurrent den- of the conformal structured perovskite solar cell can affect not
sity and photocurrent density enhancement of the solar cell only the light absorption intensity but also the distribution of

5
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Figure 6.  Two-dimensional light absorption density distribution profiles of the planar perovskite solar cell at three wavelengths of 450 nm
(a), 580 nm (b), and 710 nm (c), and light absorption density distribution profiles of the perovskite cell with conformal structure at the same
three wavelengths of 450 nm (d), 580 nm (e), and 710 nm (f), respectively.

the light absorption density. To verify this opinion, we calcu- Table 3.  Calculated photocurrent density and photocurrent density
late the light absorption density distribution for the reference enhancement of the solar cell with both of back Au and HTL.
cell and the conformal structured cell at h/w  =  2. As a demon- Conformal Conformal
stration, the two dimensional absorption distribution profiles Cell architectures Planar (h/w  =  2) (h/w  =  3)
at three wavelengths of 450 nm, 580 nm, and 710 nm are shown
Jsc (mA cm−2) 17.72 21.12 22.86
in figure 6. By comparison, it is found that the light absorption
Enhancement 19.2% 29.0%
distribution present a clear and remarkable difference.

6
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

Table 4.  Calculated photocurrent density and photocurrent density it can be summarized that the present design of conformal
enhancement of the solar cell without back Au but with HTL. structured perovskite solar cells is promising for the potential
Conformal Conformal applications in both single-junction and tandem solar cells.
Cell architectures Planar (h/w  =  2) (h/w  =  3) It should be pointed out that under the optimal parameters,
significant photocurrent enhancements are obtained for the
Jsc (mA cm−2) 13.97 17.23 20.38
two cases of conformal ultrathin perovskite solar cells with
Enhancement 23.3% 45.9%
and without equivalent volume of 100 nm-thick perovskite.
When the coverage percentage is higher than 85%, the back
The above investigation results are for the light absorption Au structure tends to a mirror. It is expected that the con-
enhancement of the ultrathin perovskite solar cell architecture formal perovskite rather than plasmonic plays a more impor-
with back Au while without a HTM layer. In previous simu- tant role in this case.
lation reports, a two-dimensional array consisting conformal
hemispherical structure was investigated for 100 nm-thick 4. Conclusion
ultrathin perovskite solar cells with back Au but without a
HTM layer [13], where the optimal integrated absorption of In summary, the investigations on ultrathin-film perovskite
18.4% was achieved in the wavelength range of 300–800 nm. solar cells are necessary due to a variety of motivations. The
In contrast, our 100 nm-thick ultrathin perovskite solar cells addressing of the light absorption in the ultrathin perovskite
with conformal structure has demonstrated highest enhance- solar cells can help to achieve higher photocurrent den-
ment of 29.7% in photocurrent density in the wavelength of sity for the improved PCE. We proposed and investigated
300–800 nm. Our design shows a broadband enhancement. In ultrathin perovskite solar cell architecture with a conformal
addition, there is nearly no absorption decrease in the short structure. Light absorption investigations are conducted
wavelengths. In terms of light-trapping physics, our solar for a series of cases including without HTM layer and
cell architecture with conformal structure is similar to [13] with HTM layer as well as without back Au. The optimal
because of photonic grating of conformal structure and sur- results show significant photocurrent density enhancement
face plasmonic mode of conformal Au. But, our design is with 27.5% and 29.7% at the cases of h/w  =  2 and h/w  =  3,
more powerful in light absorption enhancement. respectively for the conformal perovskite cells without HTM
Besides, we also investigate the light absorption enhance- layer. For the case with HTM layer, the photocurrent den-
ment of the solar cells with the addition of the HTM layer. sity enhancements are 19.2% and 29.0%, respectively. When
For our case, a variety of HTM layer thicknesses are inves- Au is removed, the photocurrent density enhancements
tigated and optimal thickness of 50 nm is found. The calcu- become 23.3% and 45.9%, respectively. The simulation
lated optimal photocurrent density and photocurrent density results clearly show significant absorption enhancement by
enhancement of the solar cell with back Au and with a HTM using conformal structure solar cell architecture. In terms
layer of 50 nm are listed in table  3. By comparison of the of light-trapping physics, it is because of photonic grating
conformal structured solar cells between with and without a of conformal structure and surface plasmonic mode of con-
HTM layer, it is found that the addition of 50 nm-thick HTM formal Au. It is expected that the present results can be used
layer reduces the photocurrent density of 0.97mA cm−2 when to help to develop efficient ultrathin perovskite solar cells
the h/w is 2. However, for the case of h/w  =  3, the photo- for single-junction and tandem solar cells. Because the con-
current density has an increase of 0.43mA cm−2. This shows formal deposition is usually physical fabrication, the cost is
that higher ratio of h to w is beneficial to the enhanced light obviously higher than the typical solution-based preparation
absorption, which is attributed to the photonic grating of the method. However, the unique benefit of adopting conformal
conformal structure. structure can produce higher light absorption with poten-
For potential applications for tandem solar cells, we inves- tially unique application areas.
tigate light absorption of the ultrathin perovskite solar cells
without Au but with HTM layer. The calculated results are
listed in table 4. It is found in table 4 that the photocurrent Acknowledgment
density values of the planar and conformal solar cells are
reduced compared with the solar cells with back Au. This Wensheng Yan gratefully thanks the Humboldt Foundation of
decrease is normal because back Au can reflect light to solar Germany for the award of the Humboldt Research Fellowship
cell and thus improve total light propagation path length. It for Experienced Researchers and the DFG for financial support.
can be seen in table  4 that the photocurrent density of the
planar cell is 13.97mA cm−2. However, the photocurrent Funding information
density enhancement values are up to 23.3% and 45.9% for
the conformal structured perovskite solar cells at the cases of National Science Foundation of China (NSFC) (U1765105,
h/w  =  2 and 3, respectively. The simulation results show that 61604087); The Hebei Provincial Young Top-notch
the present design of conformal structure is also very effective Talent Support Program (BJRC2016); The Research
to significantly enhance light absorption for perovskite solar Innovation Program for Graduates of Hebei University
cells without back Au electrode. Based on the above results, (Grant No. X201731); Alexander von Humboldt–Stiftung

7
J. Phys. D: Appl. Phys. 51 (2018) 245101 X Tan et al

(AUS-1141939-HFST-E); German Research Foundation [12] Xie Z J, Sun S R, Wang W, Qin L X, Yan Y, Hou R X and
(DFG) under YA156/1-1 ‘High-performance all-inorganic Qin G G 2018 Opt. Commun. 410 117–22
[13] Chen M, Zhang Y, Cui Y, Zhang F, Qin W, Zhu F and
perovskite/crystalline silicon tandem solar cells’.
Hao Y Y 2017 IEEE J. Photovolt. 7 1324–8
[14] Xing G C, Mathews N, Sun S Y, Lim S, Lam Y M,
ORCID iDs Grätzel M, Mhaisalkar S and Sum T 2013 Science
342 344–7
Xinyu Tan https://orcid.org/0000-0001-6849-0606 [15] Bailie C D et al 2015 Energy Environ. Sci. 8 956–63
[16] Eperon G E et al 2016 Science 354 861–5
Li Guan https://orcid.org/0000-0001-9099-7163 [17] Laban W A and Etgar L 2013 Energy Environ. Sci.
Wensheng Yan https://orcid.org/0000-0002-6364-3471 6 3249–53
[18] Shi J J et al 2014 ACS Appl. Mater. Interfaces 6 9711–8
References [19] Shi J J et al 2014 Appl. Phys. Lett. 104 063901
[20] Liu M, Johnston M B and Snaith H J 2013 Nature
501 395–8
[1] Tan X Y, Yan W S, Tu Y T and Deng C 2017 Opt. Express [21] Sutherland B R, Hoogland S, Adachi M M, Wong C T O and
25 14725–31 Sargent E H 2014 ACS Nano 8 10947–52
[2] Kojima A, Teshima K, Shirai Y and Miyasaka T 2009 J. Am. [22] Arabpour Roghabadi F, Oniy Aghmiuni K and Ahmadi V 2016
Chem. Soc. 131 6050–1 Org. Electron. 34 164–71
[3] www.nrel.gov/pv/assets/images/efficiency-chart.png [23] Tiguntseva E et al 2017 Nanoscale 9 12486–93
[4] Szuromi P 2016 Science 354 192 [24] Yin J, Qu H, Cao J, Tai H, Li J and Zheng N 2016 RSC Adv.
[5] Yang Y and You J B 2017 Nature 544 155–6 6 24596–602
[6] Du Q G, Shen G S and John S 2016 AIP Adv. 6 065002 [25] Fang X, Zhao C Y and Bao H 2014 J. Quant. Spectrosc.
[7] Carretero-Palacios S, Calvo M E and Míguez H 2015 J. Phys. Radiat. Transfer 133 579–88
Chem. C 119 18635–40 [26] Tok R U, Ow-Yang C and Sendur K 2011 Opt. Express
[8] Kaltenbrunner M et al 2015 Nat. Mater. 14 1032–9 19 22731–42
[9] Ong M Z, Chen Z F, Zhang T K, Xiao Y B, Zeng X L, [27] Xing G, Mathews N, Lim S S, Yantara N, Liu X, Sabba D,
Chen J L, Yan K Y and Xu J B 2016 Nanoscale 8 6290–9 Grätzel M, Mhaisalkar S and Sum T C 2014 Nat. Mater.
[10] Gaspera E D, Peng Y, Hou Q, Spiccia L, Bach L, Jasieniak J J 13 476–80
and Cheng Y B 2015 Nano Energy 13 249–57 [28] Ball J M et al 2015 Energy Environ. Sci. 8 602–9
[11] Cai B Y, Peng Y, Cheng Y B and Gu M 2015 Opt. Express [29] Palik E D 1998 Handbook of Optical Constants of Solids
23 A1700–6 (New York: Elsevier)

View publication stats