Solar Energy Materials & Solar Cells 133 (2015) 8–14

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

journal homepage: www.elsevier.com/locate/solmat

Theoretical analysis on effect of band offsets in perovskite solar cells

Takashi Minemoto n, Masashi Murata
Department of Electrical and Electronic Engineering, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan

art ic l e i nf o a b s t r a c t

Article history: The effect of band offsets in CH3NH3PbI3-xClx perovskite-based solar cells with planar junction
Received 18 June 2014 configuration was analyzed using one-dimensional device simulator. As widely known in thin-film
Received in revised form compound solar cells, the band offset between buffer/absorber layers is a decisive factor for carrier
30 September 2014
recombination at the interface, determining open-circuit voltage (Voc). In this study, the impact of two
Accepted 25 October 2014
Available online 20 November 2014
kinds of band offsets, i.e., the conduction band offset of buffer (or blocking layer)/absorber layers and the
valence band offset of absorber/hole transport material (HTM) were examined. When the conduction
Keywords: band of the buffer was lower than that of the absorber, the interface recombination became prominent
Solar cells and Voc decreased. In contrast, when the conduction band of the buffer was higher than that of the
Perovskite
absorber by more than 0.3 eV, the collection of photo-generated carriers, i.e. electron in this case, was
Device simulation
impeded by the spike formed by the conduction band offset. Thus, the optimum position of the
Band offset
Buffer layer conduction band of the buffer was 0.0
0.3 eV higher than that of the absorber. Also, the optimum
Hole transport material position of the valence band of the HTM was derived to be 0.0
0.2 eV lower than that of the absorber.
These findings will be useful for new material choice and optimization of buffers and HTMs.
& 2014 Elsevier B.V. All rights reserved.

1. Introduction properties with inorganic materials; especially low binding energy

of exciton, and thus the exciton is Wannier-type [22]. Also, the
Thin film solar cells based on lead methylammonium tri-iodide device structures of the perovskite solar cells with planar junction
perovskite as an absorber have attracted considerable attention as and inorganic thin film solar cells, such as CIGS, are similar.
low-cost and high efficiency organic-inorganic hybrid solar cells The typical structure for the perovskite solar cells is transparent
[1–9]. Initially, high efficiencies of over 10% were reported using conductive oxide (TCO)/blocking layer (or buffer)/absorber/hole
perovskite absorbers with mesoporous scaffold layers [1–5]. After transport material (HTM)/metal back contact [7,8], and that of
the promising results, a similar range of high efficiencies were CIGS solar cells is TCO/buffer/absorber/metal back contact [23–25].
reported by planar junction configuration, i.e., without the meso- From the two facts, solar cell device simulator widely used in CIGS
porous structure [6–9]. Especially, high efficiencies of over 15% solar cells can be applied to the perovskite solar cells. In our
were reported using CH3NH3PbI3-xClx and CH3NH3PbI3 perovskite previous study [26], we analyzed the effect of absorber and
absorbers [7,8]. High light absorption coefficient [10] and long interface qualities and the optimum thickness of the absorber of
carrier diffusion length [2] should be the origin of the flexibility of the perovskite solar cells using the Solar Cell Capacitance Simu-
the device architecture; namely, the mesoporous structure is not a lator (SCAPS) developed by University of Gent [27].
requirement to achieve this level of high efficiency. The CIGS solar cells are heterojunction solar cells, similar to the
To achieve higher efficiency, the understanding of device perovskite solar cells. In the CIGS solar cells, theoretical analysis
operation mechanism is important. To this end, device simulation using device simulation was intensively performed to understand
is useful, and conventional solar cells based on inorganic semi- device operation mechanism and to design optimum layer config-
conductors such as silicon, CdTe, Cu(In,Ga)Se2 (CIGS), and Cu2SnZn uration to increase efficiency. The decrease in carrier recombina-
(S,Se)4 were widely analyzed by device simulators [11–21]. How- tion at the buffer/absorber interface is a requirement to obtain
ever, the detailed theoretical analysis of the perovskite solar cells high open-circuit voltage (Voc) and thus high efficiency. The
using device simulation is few. The perovskite materials are interface recombination depends not only on the defect density
organic-inorganic hybrid materials and have similar material at the interface, but also more importantly on the conduction band
offset (CBO) between the buffer/absorber layers, which is both
n
theoretically supported [16,18,28] and experimentally demon-
Corresponding author. Tel./fax: þ 81 77 561 3065.
E-mail addresses: minemoto@se.ritsumei.ac.jp (T. Minemoto),
strated [29–31]. Also, the effect of the CBO is experimentally
ro0009he@ed.ritsumei.ac.jp (M. Murata). confirmed in different absorber materials such as CuInS2 [32]

http://dx.doi.org/10.1016/j.solmat.2014.10.036
0927-0248/& 2014 Elsevier B.V. All rights reserved.
 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14 9

and SnS [33,34]. In the perovskite solar cells, the absorber is reflectance at the surface and interfaces of each layer is not
reported to be intrinsic [35] and should be fully depleted. There- considered in this simulation.
fore, not only the CBO of the buffer/absorber layers but also the Fig. 1 shows the current density-voltage (J-V) characteristic of
valence band offset (VBO) of the absorber/HTM layers must be the perovskite solar cell calculated under the parameter set in
important. Here, we call a blocking layer as buffer from the Table 1. In the figure, quantum efficiency (QE) curve is also shown
analogy of the CIGS solar cells. Also, there is an experimental in inset. The low QE in short wavelength region (o 355 nm) is due
report on the variation of device behaviors with different HTMs to the absorption by SnO2:F. Here, the CBO of the buffer/absorber
[36], which should be partly ascribed to the effect of the band and the VBO of the absorber/HTM were set to be zero. The similar
offset. In this study, to understand the operation mechanism and ranges of short-circuit current density (Jsc)
22 mA/cm2 and Voc
optimum design of the device, we performed theoretical analysis
1.0 V with the experimental values [6–9] were successfully
of the effect of the CBO and VBO on the solar cell parameters of the obtained, demonstrating that the device simulation can be also
perovskite solar cells with planar junction configuration by SCAPS. used in the perovskite solar cell and the input parameter set was
not far from the real device. For further understanding of the
device operation mechanism, we examine the effect of the band
offsets in the perovskite solar cell.
2. Device simulation parameters IDL1 and IDL2 were inserted between the buffer/absorber inter-
face and the absorber/HTM interface, respectively, to take into
We used SCAPS ver. 3.2.01 for simulation platform. The structure account interface carrier recombination. Fig. 2 depicts the definition
of the perovskite solar cell in the simulation is TCO/buffer/interface of IDL1 and IDL2 in the case of different band offsets. The band
defect layer 1 (IDL1)/absorber/interface defect layer 2 (IDL2)/HTM. offsets were varied by varying the electron affinity of the buffer and
Table 1 summarizes input parameters for each layer. The parameters HTM. The signs (positive and negative) of the CBO and VBO were
of TCO, buffer, absorber, and HTM are based on SnO2:F, TiO2, defined from a barrier height for photo-generated carriers. In the
CH3NH3PbI3-xClx, and 2,2',7,70 -tetrakis(N,N-p-dimethoxy-phenyla- case of the CBO of the buffer/absorber interface, electrons generated
mino)-9,90 - spirobifluorene (Spiro-OMeTAD), respectively. Detailed at the absorber flow to surface side for collection. If the electron
definition of IDL1 and IDL2 are explained later. Here, NA and ND affinity of the buffer (χbuffer) is larger than that of the absorber
denote acceptor and donor densities, εr is relative permittivity, χ is (χabsorber), i.e. the conduction band of the buffer is lower than that of
electron affinity, Eg is band gap energy, μn and μp are mobilities of the absorber as shown in Fig. 2(a); energy cliff is formed at the
electron and hole, and Nt is defect density. Exact values of the physics interface and no barrier for the electron is formed. On the other
parameters are difficult to obtain, especially for new materials, and hand, if the conduction band of the buffer is higher than that of the
we collected and assumed them with our best knowledge in this absorber as shown in Fig. 2. (b), energy spike is formed at the
stage. The thicknesses of TCO, buffer, absorber and HTM were taken interface, which can act as a barrier for electrons. Thus, we defined
from the literature reporting an efficiency of 15.4% [7]. The conduc- negative and positive signs of CBO for former and later cases,
tion types of TCO, buffer, absorber and HTM are n þ , n, i (or n-), p þ , respectively. In the similar manner, negative and positive signs for
respectively, indicating that the perovskite solar cells is n-i-p junction VBO were defined as shown in Fig. 2 (c) and (d), respectively. IDL1
configuration typically used in amorphous and micro-crystalline was set to consider the carrier recombination between electrons at
silicon solar cells. The perovskite absorber was completely depleted
under this parameter set as later shown in the next section, and the 25
absorber depletion was also indicated in the literature [35]. One of Eff =17.9%
the most important parameters to determine the absolute value of
20
Current density (mA/cm )
2

the efficiency is the defect density of the absorber. In this study, we

assumed Nt ¼ 2.5  1013 cm  3 to obtain carrier diffusion lengths of
electron and hole (Ln and Lp) to be 1 μm, which is a similar value to 15
the experiment [2]. Here, Ln and Lp are identical because all the
parameters for the carriers are set to be identical, which is consistent 10
with ambipolar characteristics of the carriers [5]. Other input para-
meters not included in Table 1 were set to be identical: effective
density of states of conduction band and valence band were set to be 5
2.2  1018 and 1.8  1019 cm  3, respectively. Thermal velocity of
electron and hole was 107 cm/s. Defect energy level was the center 0
of band gap and distributed in Gaussian with characteristic energy of 0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.1 eV. The defect type was neutral. Capture cross section of electron Voltage (V)
and hole was 2  10–14 cm2. Pre-factor Aα was 105 to obtain absorp- Fig. 1. Simulated J-V characteristic of perovskite solar cell calculated with the
tion coefficient (α) curve calculated by α ¼ Aα(hv-Eg)1/2. The optical parameter set in Table 1. Inset shows the QE curve of the solar cell.

Table 1
Input parameters of device simulation.

Parameter TCO (SnO2:F) Buffer (TiO2) Interface defect layer, IDL1 Absorber (CH3NH3PbI3-xClx) Interface defect layer, IDL2 HTM (Spiro-OMeTAD)

Thickness (nm) 500 50 10 330 10 350

NA (cm  3)      2  1018 [2]
ND (cm  3) 2  1019 1016 1013 1013 1013 
εr 9.0 9.0 6.5 6.5 [7] 6.5 3.0 [37]
χ (eV) 4.00 3.90 (variable) 3.90 (variable) 3.90 [22] 3.90 2.45 [38] (variable)
Eg (eV) 3.50 3.20 1.55 (variable) 1.55 [3] 1.55 (variable) 3.00 [38]
μn / μp (cm2/Vs) 20/10 20/10 2.0/2.0 2.0/2.0 [1] 2.0/2.0 2  10  4/2  10  4 [38]
Nt (cm  3) 1015 1015 1017 2.5  1013 1017 1015
10 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14

Vacuum level Vacuum level

~
~χ ~
~ ~
~ ~
~χ ~
~χ ~
buffer χIDL1 χabsorber buffer IDL1 χabsorber~
EC CBO (+) EC
CBO (-)

Eg_absorber Eg_IDL1 Eg_absorber

Eg_IDL1
Eg_buffer

EV EV
Eg_buffer χbuffer > χabsorber χbuffer <
= χabsorber
χIDL1 = χbuffer χIDL1 = χabsorber
χIDL1 + Eg_IDL1 = χabsorber + Eg_absorber Eg_IDL1=Eg_absorber

Vacuum level Vacuum level

~
~ ~
~ ~
~ χHTM ~
~ ~
~ ~
~
EC χHTM

χabsorber χIDL2 χabsorber χIDL2 EC

Eg_HTM

Eg_HTM
Eg_IDL2
Eg_absorber Eg_absorber Eg_IDL2
EV
VBO (-)
VBO (+)
EV
χabsorber + Eg_absorber => χHTM + Eg_HTM χabsorber + Eg_absorber < χHTM + Eg_HTM
χIDL2 = χabsorber
χIDL2 = χabsorber , Eg_IDL2= Eg_absorber
χIDL2 + Eg_IDL2 = χHTM + Eg_HTM
Fig. 2. Band alignments of buffer/IDL1/absorber layers with (a) negative and (b) positive CBOs and those of absorber/IDL2/HTM layers with (c) negative and (d) positive
VBOs. Here IDL1 and IDL2 are inserted to take into account the carrier recombination at the interface.

the conduction band of the buffer and holes at the valence band of concentration and HTM are both significantly small. In the
the absorber. Thus, in the case of the negative CBO, the electron following section, we discuss the effect of the CBO of the buffer/
affinity of IDL1 (χIDL1) is set to be identical to χbuffer, and the band absorber interface and the VBO of the absorber/HTM interface on
gap of IDL1 (Eg_IDL1) is adjusted to make the valence band of IDL1 the solar cell parameters together with the defect densities at IDL1
being identical to that of the absorber as shown in Fig. 2(a). In the and IDL2.
case of the positive CBO, χIDL1 and Eg_IDL1 are set to be identical to
χabsorber and Eg_absorber, respectively, as shown in Fig. 2(b). In the
same manner, the band gap of IDL2 was set as shown in Fig. 2 3. Results and discussion
(c) and (d). This methodology to take into account the interface
recombination at heterojunction is used in previous reports on CIGS 3.1. Impact of CBO of buffer/absorber layers
solar cells [18,19], which well explained experimental phenomena
[29–31]. Fig. 3 shows the J-V curves of the perovskite solar cells with
On the other hand, we do not discuss on the effect of the VBO of different CBO values of the buffer/absorber interface. Here, the
the buffer/absorber and the CBO of the absorber/HTM interfaces. VBO of the absorber/HTM interface was set to be zero and the
The carrier recombination associated with the band offsets should defect density of IDL1 was set to be 1016 cm  3. When the CBO is
be negligible because hole concentration at the buffer and electron negative, Voc monotonically decreases with decreasing the CBO
 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14 11

from 0.0 to -0.4 eV, while Jsc is almost unchanged. Fill factor (FF) cell parameters of the perovskite solar cells with different inter-
also decreases; however that is not significant compared to Voc. face defect densities as a function of the CBO of the buffer/
Fig. 4 displays the energy band diagrams of the perovskite solar absorber interface. Here, the VBO of the absorber/HTM interface
cells with (a) negative (  0.2 eV) and (b) positive (0.2 eV) values of
the CBO of the buffer/absorber interface. When the CBO is
negative, a cliff is formed at the buffer/absorber interface as shown
in the inset (close-up of the interface) in Fig. 4(a). The cliff does not
impede photo-generated electron flow toward a front electrode,
and Jsc is almost constant. However, the activation energy for
carrier recombination (Ea) becomes lower than Eg_absorber and Ea is
represented by Eg_absorber – |CBO|. When Ea is lower than Eg_absorber,
the main recombination mechanism of the device is the interface
recombination [39–41]. Thus, Ea directly correlates with Voc, and
the negative CBO reduces Voc. When the CBO is positive, the J-V
curves for CBO ¼0.0, 0.1, 0.2, and 0.3 eV are excellent and almost
overlapped as shown in Fig. 3. However, further increase in CBO
values induces double-diode like curvature, resulting in low FF but
similar Voc. When the CBO is positive, a spike is formed at the
buffer/absorber interface as shown in Fig. 4(b), which can act as a
barrier for photo-generated electron flow toward the front elec-
trode, while Ea is equal to Eg_absorber. The spike impedes the
electron flow at CBO ¼0.4 eV at forward bias state, resulting in
double-diode behavior and the poor FF. In the extreme case of
CBO ¼ 0.5 eV, the spike even impedes the electron flow at short-
circuit state and decreases both Jsc and FF. In contrast, Voc is almost
constant because the spike impedes both forward and photo
currents and also Ea is equal to Eg_absorber. Fig. 5 exhibits the solar

CBO=0.0, 0.1, 0.2, 0.3 (eV)

25
Current density (mA/cm2)

20
0.4
15
-0.1
-0.2
10
0.5
-0.3
5 -0.4

0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Voltage (V)
Fig. 3. J-V curves of perovskite solar cells with different CBO values of buffer/ Fig. 5. Solar cell parameters of perovskite solar cells with different interface defect
absorber interface. Here, the VBO of the absorber/HTM interface was set to be zero densities as a function of CBO of buffer/absorber interface. Here, the VBO of the
and the interface defect density of IDL1 was set to be 1016 cm  3. absorber/HTM interface was set to be zero.

4 4
electron
3 EC 3 electron EC

2 2

1 HTM 1 HTM
Energy (eV)

Energy (eV)

0 0
Ea=Eg-|CBO| EV Ea=Eg EV
-1 absorber -1 absorber
TCO TCO
-2 -2

-3 buffer -3 buffer
-4 -4
0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Depth from surface (μm) Depth from surface (μm)
Fig. 4. Energy band diagrams of perovskite solar cells with (a) negative (-0.2 eV) and (b) positive (0.2 eV) values of CBO of buffer/absorber interface. Insets show close-up
images at the interface.
12 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14

25
VBO = -0.1, 0.0, 0.1, 0.2 (eV)
Recombination current (mA/cm )
2 at open-circuit condition
25

Current density (mA/cm 2)

20
20 0.27
IDL1 15 0.25
15 -0.2
10
0.30
10 -0.3
IDL2 5
-0.4
5
0
0
absorber 0.0 0.2 0.4 0.6 0.8 1.0 1.2
-0.6 -0.4 -0.2 0.0 0.2 0.4 Voltage (V)
CBO (eV) Fig. 7. J-V curves of perovskite solar cells with different VBO values of absorber/
HTM interface. Here, the CBO of the buffer/absorber interface was set to be zero,
and the interface defect density of IDL2 was set to be 1016 cm  3.
at open-circuit condition
Recombination current (mA/cm )
2

25
IDL1
20
4
IDL2 EC
15 3

10 2 hole
HTM

Energy (eV)
5 1
absorber Ea=Eg-|VBO|
0 0
EV
-0.6 -0.4 -0.2 0.0 0.2 0.4
-1
VBO (eV) absorber
TCO
Fig. 6. Recombination current at buffer/absorber interface, absorber, and absorber/ -2
HTM interface of the perovskite solar cells at open-circuit condition as a function of
(a) CBO of buffer/absorber interface and (b) VBO of absorber/HTM interface. -3 buffer
-4
0.0 0.2 0.4 0.6 0.8 1.0 1.2
is set to be zero. Note that the conditions for high defect densities
and high CBO values could not be calculated for convergence Depth from surface (μm)
failure of the program. With increasing the defect density, the
absolute values of the solar cell parameters decrease almost
monotonically while the trends on the CBO are similar. The upper 4
limit of the CBO value is not clear for the conditions of high defect EC
3
densities and high CBO values in this study; however, exact
measurement of the CBO should not be easy in this energy hole
2
resolution and the trend of the solar cell parameters on the CBO
HTM
Energy (eV)

should be important. Especially, the negative CBO increases the 1

interface recombination as shown in the recombination current at Ea=Eg
IDL1, absorber and IDL2 with different CBOs at open-circuit state 0
EV
(Fig. 6(a)), and thus the conduction band of the buffer is at least -1
equal or higher than that of the absorber. Consequently, the absorber
TCO
optimum range for the CBO of the buffer/absorber interface was -2
derived to be 0.0
0.3 (or 0.2 in the case of high defect density at
the interface) eV. -3
buffer
-4
3.2. Impact of VBO of absorber/HTM layers 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Depth from surface (μm)
Fig. 7 shows the J-V curves of the perovskite solar cells with
different VBO values of the absorber/HTM interface. Here, the CBO Fig. 8. Energy band diagrams of perovskite solar cells with (a) negative (-0.2 eV)
and (b) positive (0.2 eV) values of VBO of absorber/HTM interface. Insets show
of the buffer/absorber interface was set to be zero and the defect
close-up images at the interface.
density of IDL2 was set to be 1016 cm  3. Similar to the case of the
CBO of the buffer/absorber interface, when the VBO is negative, Voc
monotonically decreases, while Jsc is almost unchanged. FF also toward a back electrode and Jsc is almost constant. However,
decreases; however that is not significant compared to Voc. Fig. 8 similar to the case of the negative CBO, the activation energy for
displays the energy band diagrams of the perovskite solar cells carrier recombination becomes lower than Eg_absorber and Ea is
with (a) negative ( 0.2 eV) and (b) positive (0.2 eV) values of the represented by Eg_absorber – |VBO|. Thus, the decrease in Ea by the
VBO of the absorber/HTM interface. When the VBO is negative, a negative VBO increases the interface recombination as shown in
cliff is formed at the absorber/HTM interface as shown in the inset Fig. 6(b) and reduces Voc. When the VBO is positive and up to
of Fig. 8(a). The cliff does not impede photo-generated hole flow 0.2 eV, the J-V characteristics show excellent curvature. However,
 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14 13

4 0.3
VBO=0.20, 0.00 eV EC VBO=0.00eV
3 0.0
2 0.27 EV

Energy (eV)
Energy (eV)
1 0.30 -0.3
0.20
0 -0.6
EV 0.27
-1 -0.9
-2 0.27 0.20, 0.00
0.30 -1.2 0.30
-3
-4 -1.5
0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.85 0.90 0.95
Depth from surface ( μm) Depth from surface ( μm)
Fig. 9. Energy band diagrams of perovskite solar cells with high VBO values of absorber/HTM interface: (a) total device and (b) close-up at absorber/HTM interface region
indicated by dashed circle in (a).

further increase in VBO decreases FF. As shown in Fig. 8 (b), the

spike is formed at the absorber/HTM interface similar to the case
of the positive CBO of the buffer/absorber interface. However, the
curvature is not like double-diode which is observed in the case of
the positive CBO. Fig. 9 shows the energy band diagrams of the
perovskite solar cells with high VBO values of the absorber/HTM
interface: (a) total device and (b) close-up at the absorber/HTM
interface region indicated by dashed circle in (a). The increase in
VBO results in the decrease in the electric field across the absorber,
indicating that the complete depletion of the absorber is not given
by the high VBO values. The origin of the electric field is positively-
charged donors in TCO and negatively-charged acceptors in HTM
those are created by the recombination between electrons from
TCO and holes from HTM. The diffusion of holes from HTM to the
absorber side upon junction formation is influenced by hole
concentration and potential. The hole concentrations in the case
of low and high VBOs are, of course, same; however, the high VBO
acts as a barrier for the hole diffusion from HTM to the absorber
side, resulting in the incomplete depletion of the absorber and
thus the poor FF. Fig. 10 exhibits the solar cell parameters of the
perovskite solar cells with different interface defect densities as a
function of the VBO of the absorber/HTM interface. Here, the CBO
of the buffer/absorber interface was set to be zero. With increasing
the defect density the absolute values of the solar cell parameters
decrease but the degree is small compared to the case of the CBO
of the buffer/absorber interface. This is because excess carrier
densities, which is a dominant parameter for carrier recombina-
tion, at the back side (absorber/HTM interface) is significantly
smaller than that of the front side (buffer/absorber interface). Even
if the impact of the defect density of IDL2 is not significant, the
adjustment of the VBO is essential to obtain a high Voc, in other
words to keep Ea ¼Eg_absorber. Consequently, the optimum VBO
range was derived to be 0.0
0.2 eV.
We believe this study helps the understanding of the operation
mechanism of the perovskite solar cells and can be useful for
designing the new material for buffer, absorber and HTM, such as
lead-free new perovskite absorbers and suitable buffer and HTM
for them.

4. Conclusions
Fig. 10. Solar cell parameters of perovskite solar cells with different interface defect
The effects of the band offsets of the perovskite solar cells with densities as a function of VBO of absorber/HTM interface. Here, the CBO of the
buffer/absorber interface is set to be zero.
planar junction configuration were analyzed using one-dimensional
device simulator SCAPS ver. 3.2.01 widely used in CIGS solar cells.
When the conduction band of the buffer is lower than that of the is too high, the spike formed by the CBO acts as a barrier for photo-
absorber, interface recombination becomes prominent because of generated carrier flow. In the similar manner, the higher position
the reduction of the activation energy for carrier recombination, of the valence band of the HTM than that of the absorber leads a
resulting in low Voc. In contrast, if the conduction band of the buffer prominent increase in the interface recombination. On the other
14 T. Minemoto, M. Murata / Solar Energy Materials & Solar Cells 133 (2015) 8–14

hand, too low position of the valence band of the HTM results in [18] T. Minemoto, T. Matsui, H. Takakura, Y. Hamakawa, T. Negami, Y. Hashimoto,
incomplete depletion of the absorber. Consequently, the optimum T. Uenoyama, M. Kitagawa, Theoretical analysis of the effect of conduction
band offset of window/CIS layers on performance of CIS solar cells using
positions of the conduction band and the valance band of the buffer
device simulation, Sol. Energ. Mater. Sol. Cell 67 (2001) 83–88.
and HTM are derived to be 0.0
0.3 eV higher and 0.0
0.2 eV [19] T. Minemoto, J. Julayhi, Buffer-less Cu(In,Ga)Se2 solar cells by band offset
lower than corresponding bands of the absorber, respectively. These control using novel transparent electrode, Curr. Appl. Phys. 13 (2013) 103–106.
results should be one of the important guidelines for designing new [20] M. Murata, D. Hironiwa, N. Ashida, J. Chantana, K. Aoyagi, N. Kataoka,
T. Minemoto, Optimum bandgap profile analysis of Cu(In,Ga)Se2 solar cells with
material combination of buffer, absorber and HTM. various defect densities by SCAPS, Jpn. J. Appl. Phys. 53 (2014) (04ER14-1-4).
[21] D. Hironiwa, M. Murata, N. Ashida, T. Zeguo, T. Minemoto, Simulation of
optimum band-gap profile of Cu2ZnSn(S,Se)4 solar cells with different optical
Acknowledgments and defect properties, Jpn. J. Appl. Phys. 53 (2014) (071201-1-9).
[22] M. Hirasawa, T. Ishihara, T. Goto, K. Uchida, N. Miura, Magnetoabsorption of
the lowest exciton in perovskite-type compound (CH3NH3)PbI3, Phys. B. 201
The author would like to thank Professor Marc Burgelman, (1994) 427–430.
Department of Electronics and Information Systems, University [23] I. Repins, M.A. Contreras, B. Egaas, C. DeHart, J. Scharf, C.L. Perkins, B. To,
of Gent for the development of the SCAPS software package R. Noufi, 19.9% efficient ZnO/CdS/CuInGaSe2 solar cells with 81.2% fill factor,
Prog. Photovolt 16 (2008) 235–239.
and allowing its use. Also, the authors would like to thank [24] P. Jackson, D. Hariskos, R. Wuerz, W. Wischmann, M. Powalla, Compositional
Dr J. Chantana and Dr. Z. Tang of Ritsumeikan University for useful investigation of potassium doped Cu(In,Ga)Se2 solar cells with efficiencies up
discussion. to 20.8%, Phys. Status Solidi RRL 8 (2014) 219–222.
[25] A. Chirilă, S. Buecheler, F. Pianezzi, P. Bloesch, C. Gretener, A.R. Uhl, C. Fella,
L. Kranz, J. Perrenoud, S. Seyrling, R. Verma, S. Nishiwaki, Y.E. Romanyuk,
References G. Bilger, A.N. Tiwari, Highly efficient Cu(In,Ga)Se2 solar cells grown on flexible
polymer films, Nat. Mater. 10 (2011) 857–861.
[1] M.M. Lee, J. Teuscher, T. Miyasaka, T.N. Murakami, H.J. Snaith, Efficient hybrid [26] T. Minemoto, M. Murata, Device modeling of perovskite solar cells based on
solar cells based on meso-superstructured organometal halide perovskites, structural similarity with thin film inorganic semiconductor solar cells, J. Appl.
Science 338 (2012) 643–647. Phys. 116 (2014) (054505-1-6).
[2] S.D. Stranks, G.E. Eperson, G. Grancini, C. Menelaou, M.J.P. Alcocer, T. Leijtens, [27] M. Burgelman, P. Nollet, S. Degrave, Modeling polycrystalline semiconductor
L.M. Herz, A. Petrozza, H.J. Snaith, Electron-hole diffusion lengths exceeding 1 solar cells, Thin Solid Films 361-362 (2000) 527–532.
μm in organometal trihalide perovskite absorber, Science 342 (2013) 341–344. [28] M. Gloeckler, J. Sites, Efficiency limitations for wide-band-gap chalcopyrite
[3] J.H. Noh, S.H. Im, J.H. Heo, T.N. Mandal, S.I. Seok, Chemical management for solar cells, Thin Solid Films 480-481 (2005) 241–245.
colorful, efficient, and stable inorganic-organic hybrid nanostructured solar [29] T. Minemoto, Y. Hashimoto, T. Satoh, T. Negami, H. Takakura, Y. Hamakawa,
cells, Nano Lett. 13 (2013) 1764–1769. Cu(In,Ga)Se2 solar cells with controlled conduction band offset of window/Cu
[4] J. Burschka, N. Pellet, S.-J. Moon, R. H-Baker, P. Gao, M.K. Nazeeruddin, (In,Ga)Se2 layers, J. Appl. Phys. 89 (2001) 8327–8330.
M. Gratzel, Sequential deposition as a route to high-performance perovskite- [30] T. Minemoto, Y. Hashimoto, T. Satoh, W.S. Kolahi, T. Negami, H. Takakura,,
sensitized solar cells, Nature 499 (2013) 316–319. Y. Hamakawa, Control in conduction band offset in wide-gap Cu(In,Ga)Se2
[5] J.M. Ball, M.M. Lee, A. Hey, H.J. Snaith, Low-temperature processed meso- solar cells, Sol. Energ. Mater. Sol. Cell 75 (2003) 121–126.
superstructured to thin-film perovskite solar cells, Energ. Environ, Sci 6 (2013) [31] T. Torndahl, C. Platzer-Bjorkman, J. Kessler, M. Edoff, Atomic layer deposition
1739–1743. of Zn1-xMgxO buffer layers for Cu(In,Ga)Se2 solar cells, Prog. Photovolt 15
[6] J. You, Z. Hong, Y. Yang, Q. Chen, M. Cai, T.-B. Song, C.-C. Chen, S. Lu, Y. Liu, (2007) 225–235.
H. Zhou, Y. Yang, Low-temperature solution-processed perovskite solar cells [32] R. Hamazaki, Y. Oda, S. Fukamizu, A. Yamamoto, T. Minemoto, H. Takakura,
with high efficiency and flexibility, ACS. Nano 8 (2014) 1674–1680. Optimization of compositional ratio of Zn(O,S) window layer in CuInS2 solar
[7] M. Liu, M.B. Johnston, H.J. Snaith, Efficient planar heterojunction perovskite cells, Jpn. J. Appl. Phys. 51 (2012) (10NC10-1-3).
solar cells by vapour deposition, Nature 501 (2013) 395–398. [33] T. Ikuno, R. Suzuki, K. Kitazumi, N. Takahashi, N. Kato, K. Higuchi, SnS thin film
[8] D. Liu, T.L. Kelly, Perovskite solar cells with a planar heterojunction structure solar cells with Zn1-xMgxO buffer layers, Appl. Phys. Lett. 102 (2013) (193901-
prepared using room –temperature solution processing techniques, Nat. 1-4).
Photon. 8 (2013) 133–138. [34] P. Sinsermsuksakul, K. Hartman, S.B. Kim, J. Heo, L. Sun, H.H. Park,
[9] G.E. Eperon, V.M. Burlakov, P. Docampo, A. Goriely, H.J. Snaith, Morphological R. Chakraborty, T. Buonassisi, R.G. Gordon, Enhancing the efficiency of SnS
control for high performance, solution-processed planar heterojunction per- solar cells via band-offset engineering with a zinc oxysulfide buffer layer,
ovskite solar cells, Adv. Funct. Mater. 24 (2014) 151–157. Appl. Phys. Lett. 102 (2013) (053901-1-5).
[10] G.C. Xing, N. Mathews, S.Y. Sun, S.S. Lim, Y.M. Lam, M. Gratzel, S. Mhaisalkar, [35] E. Edri, S. Kirmayer, S. Mukhopadhyay, K. Gartsman, G. Hodes, D. Cahen,
T.C. Sum, Long-range balanced electron- and hole- transport lengths in
Elucidating the charge carrier separation and working mechanism of
organic-inorganic CH3NH3PbI3, Science 342 (2013) 344–347.
CH3NH3PbI3-xClx perovskite solar cells, Nat. Commun. Published online,
[11] P. Basore, Numerical modeling of textured silicon solar cells using PC-1D, IEEE
10.1038/ncomms4461.
T. Electr. Devic. 37 (2) (1990) 337–343.
[36] J.H. Heo, S.H. Im, J.H. Noh, T.N. Mandal, C.-S. Lim, J.A. Chang, Y.H. Lee, H.-J. Kim,
[12] P.P. Altermatt, Models for numerical simulations of crystalline silicon solar
A. Sarkar, M.d.K. Nazeeruddin, M. Gratzel, S.I. Seok, Efficient inorganic-organic
cells – a review, J. Comput, Electron 10 (2011) 314–330.
hybrid heterojunction solar cells containing perovskite compound and poly-
[13] H.-C. Yuan, V.E. Yost, M.R. Page, P. Stradins, D.L. Meier, H.M. Branz, Efficient
black silicon solar cell with a density-graded nanoporous surface: optical meric hole conductors, Nat. Photon. 7 (2013) 486–491.
properties, performance limitation, and design rules, Appl. Phys. Lett. 95 [37] H.J. Snaith, M. Gratzel, Electron and hole transport through mesoporous TiO2
(2009) (123501-1-3). infiltrated with Spiro-MeOTAD, Adv. Mater. 19 (2007) 3643–3647.
[14] A. Niemegeers, M. Burgelman, Effects of the Au/CdTe back contact on IV and [38] D. Poplavskyy, J. Nelson, Nondispersive hole transport in amorphous films of
CV characteristics of Au/CdTe/CdS/TCO solar cells, J. Appl. Phys. 81 (1997) methoxy-spirofluorene-arylamine organic compound, J. Appl. Phys. 93 (2003)
2881–2886. 341–346.
[15] P. Nollet, M. Kontges, M. Burgelman, S. Degrave, R. R.- Koch, Indications for [39] U. Rau, H.W. Schock, Electronic properties of Cu(In,Ga)Se2 heterojunction solar
presence and importance of interface states in CdTe/CdS solar cells, Thin Solid cells - recent achievements, current understanding, and future challenges,
Films 431-432 (2003) 414–420. Appl. Phys. A 69 (1999) 131–147.
[16] R. Klenk, Characterization and modeling of chalcopyrite solar cells, Thin Solid [40] M. Turcu, U. Rau, Fermi level pinning at CdS/Cu(In,Ga)(Se,S)2 interfaces: effect
Films 387 (2001) 135–140. of chalcopyrite alloy composition, J. Phys. Chem. Solids 64 (2003) 1591–1595.
[17] T. Dullweber, O. Lundberg, J. Malmstrom, M. Bodegard, L. Stolt, U. Rau, [41] K. Tanaka, T. Minemoto, H. Takakura, Analysis of heterointerface recombina-
H.W. Schock, J.H. Werner, Back surface band gap gradings in Cu(In,Ga)Se2 tion by Zn1-xMgxO for window layer of Cu(In,Ga)Se2 solar cells, Sol. Energ. 83
solar cells, Thin Solid Films 387 (2001) 11–13. (2009) 477–479.