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

net/publication/348276376

Enhancing the photovoltaic performance of Cd-free Cu2ZnSnS4

heterojunction solar cells using SnS HTL and TiO2 ETL

Article  in  Solar Energy · January 2021

DOI: 10.1016/j.solener.2020.12.020

CITATIONS READS

4 339

1 author:

Atowar Rahman
University of Rajshahi
19 PUBLICATIONS   138 CITATIONS   

SEE PROFILE

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

Fabrication of SiO2 at Low Temperature View project

All content following this page was uploaded by Atowar Rahman on 11 January 2021.

The user has requested enhancement of the downloaded file.

 Solar Energy 215 (2021) 64–76

Contents lists available at ScienceDirect

Solar Energy
journal homepage: www.elsevier.com/locate/solener

Enhancing the photovoltaic performance of Cd-free Cu2ZnSnS4

heterojunction solar cells using SnS HTL and TiO2 ETL
M. Atowar Rahman
Department of Electrical and Electronic Engineering, University of Rajshahi, Rajshahi 6205, Bangladesh

A R T I C L E I N F O A B S T R A C T

Keywords: Cu2ZnSnS4 chalcogenide semiconductor is a very promising absorber material for solar cells and no longer
Cu2ZnSnS4 fulfilled its realistic goals due to the possible challenge of defect-free fabrication, non-optimized buffer layer
SnS HTL alignment and device configuration. In this article, we proposed a new planner structure defined as (SnS/CZTS/
TiO2 ETL
TiO2/ITO) using Ni and Al as a back and front contact materials, respectively, and simulated its photovoltaic (PV)
Cd-free CZTS solar cells
performance using SCAPS-1D. In the simulation, defect densities were added to each layer and the interface
between the buffer layer and the absorber layer. This work demonstrated the promising role of SnS HTL and TiO2
ETL in reducing the Voc deficit of CZTS-based solar cells by forming more favourable band alignment with the
absorber layer, reducing non-radiative recombination at the interfaces. In response to changes in the material
properties of different layers (such as thickness, carrier concentration, defect density), working temperature, and
back contact metal work function, the PV performance of the new architecture for CZTS solar cells was inves­
tigated and optimized. The study of PV performance optimization revealed that a relatively thicker absorber
layer with low carrier concentration showed better PV performances, while ETL and HTL with a thinner and
higher carrier concentration are required to enhance PV performances. Power conversion efficiency of >30%, Voc
of >1.09 V, Jsc of >32 mA/cm2 and FF of >87% was predicted for CZTS solar cells with the new architecture. The
findings of this study suggest that SnS and TiO2 are expected to become HTL and ETL, respectively, for fabri­
cating low cost, high efficiency, and Cd-free CZTS-based heterojunction solar cell.

1. Introduction interference and absorption (Bhattacharya and John, 2019). Over the
past few decades, researchers have focused on finding the alternative
Solar power is one of the most important renewable energy sources low-cost materials and device fabrication technologies owing to various
to satisfy present and future global energy demands in tera-watt (TW) limitations of this c-Si solar cell, for example, weight, rigidity, cost of Si
scale deployment. Toward this path, solar cells are outstanding among wafer, and expensive processing technologies. Thin film solar cells
other potential methods for directly converting sunlight into usable (TFSCs) are getting progressively prominent due to their high-power
energy without creating any destructive results. Because of expanding conversion efficiencies (PCEs) and direct and tunable band gap en­
energy utilization, photovoltaic solar power generation has developed ergies, and in view of the fact that less material is required to construct
exponentially in recent decades. Many researchers in this field have them compared to Si-based solar cells. Single junction cuprous sulfide/
concentrated on finding low-cost, earth-abundant, non-toxic, and highly cadmium sulfide (Cu2S/CdS) solar cell is the first reported TFSCs with
efficient photovoltaic devices. The most noteworthy conversion effi­ the PCE of about 9.1%, but the long-term performances were deterio­
ciency of up to 24.5% have been achieved in most widely used crys­ rated due to the diffusion of copper into the CdS matrix and doping of
talline Si (c-Si) based solar cells (Zhao et al., 1999). The best Si solar cell, the CdS layer, for which further research activities on Cu2S/CdS devices
developed by Kaneka Corporation, is capable of 26.7% conversion ef­ were declined (Wu et al., 2008; Partain et al., 1983). Due to low-cost,
ficiency (Yoshikawa et al., 2017; Green et al., 2018). The theoretical material availability, non-toxicity, low processing temperature, and
maximum PCE of 31% for c-Si solar cells was achieved by using opti­ cost-effective processing technologies of amorphous silicon (a-Si), a-Si-
mized photonic crystal structure and patterning with inverted micro- based solar cells among the TFSCs are preferable to the researcher. The
pyramids with lattice spacing equivalent to the wavelength of near- PCE of a-Si based TFSCs can reach up to 13.6% (Sai et al., 2015). Single-
infrared light, allowing photon capture based on strong wavelength crystal homojunction CdTe solar cell with about 2.0% PCE was first

E-mail address: [email protected].

https://doi.org/10.1016/j.solener.2020.12.020
Received 10 September 2020; Received in revised form 6 December 2020; Accepted 8 December 2020
0038-092X/© 2020 International Solar Energy Society. Published by Elsevier Ltd. All rights reserved.
M. Atowar Rahman Solar Energy 215 (2021) 64–76

demonstrated by Rappaport (1959) and cutting edge highest efficient fluctuations within the material as well as undesirable energy states
CdTe heterojunction TFSCs has been developed by close space subli­ within the bandgap, which are Shockley-Read-Hall (SRH) recombina­
mation method with record PCE of 22.1% in 2018 (Green et al., 2018). tion sites. In standard CZTS/CdS device, there are two possible reasons
However, one of the constituent elements of CdTe-based TFSCs is Cd, which can provide an explanation for the higher defect density at the
which is recognized as a carcinogenic element, and environmentally interface. Firstly, the minority carrier recombination and dislocation/
hazardous materials, and Te is toxic as well as rare earth material defects at the CdS/CZTS interface may increase due to the large lattice
(Candelise et al., 2012). In 1974, S. Wagner et al. were reported the misfit between CZTS and CdS (about 20.5%, see in Table 5). Secondly,
CuInSe2 (CIS)/CdS heterojunction photovoltaic devices for the first time “cliff-like” band alignment at the CdS/CZTS interface induces (Yan et al.,
(Wagner et al., 1974) and began to get a lot of considerations when the 2017) the band bending at the interface, results in equivalent band gap
first high-efficiency (9.4%) CIS TFSCs were reported in 1981 (Mickelsen reduction which will enhance the minority carrier recombination at the
and Chen, 1981). In TFSCs family, CIGS-based solar cells offer higher interface (Yin et al., 2015). As a matter of fact, the Voc of the CdS/CZTS
efficiency at the laboratory level. CIGS has some advantages over other device get reduced significantly (Li et al., 2015; Minemoto et al., 2001b).
TFSCs, as its tunable band gap ranging from 1.0 eV for pure CIS to 1.7 eV Understanding the carrier transport mechanism and the factors that
for pure CGS, high optical absorption coefficient (105 cm− 1) (Jackson affect their performance is very important for performance optimization
et al., 2011), and a technology that is mature for more than 20 years. The process in solar cells. The performance of solar cells also relies upon
champion CIGS solar cell with PCE of 23.35% was fabricated by Naka­ their design, material properties, and fabrication technology. Many re­
mura et al. (2019). The majority of the constituent elements of this solar searchers in this field represent solar cells with improved PV perfor­
cell are abundant within the earth crust, but Indium (In) and Gallium mance over periods of time, in spite of the fact that the overall process is
(Ga) are the rare-earth metal. Moreover, the compounds of Se such as not only quite complex but also expensive and time-consuming (Kabir
H2Se are highly poisonous. Therefore, a negative impact on the sensible et al., 2012). One among of the simplest methods for simulating solar
potential of the CIGS solar cells have been observed in the last decade cells is the numerical approach, which helps the researchers to find out a
due to resource limitations as well as ecological issues. better design and optimized performance. Testing the validity of the
From a technical point of view, TFSCs experience the low efficiency proposed physical structures and fitting the modeling output to experi­
like a-Si, lack of raw material like Te in CdTe, and In and Ga in CIGS mental results are successfully done by numerical modeling and as a
technology, or materials toxicity like Cd in CdTe technology. In this result, simulation has become necessary tools for planning and
context, the p-type chalcogenide semiconductor Cu2ZnSnS4 called CZTS designing high performance solar cells (Kabir et al., 2012).
seems to be a very attractive and high-potential absorbing materials, The configuration of immature CZTS-based solar cells technology
where In and Ga in chalcopyrite structure of CIGS are replaced by less follows the well-known CIGS or CIS technology. Simulation software
expensive, non-toxic and earth-abundant elements such as Zn and Sn, such as SCAPS, wxAMPS, PC1D, etc. can be used to analyze cells with
respectively (Haddout et al., 2019; Kowsar et al., 2019; Jiang, 2013; different buffer layers (Liu et al., 2012), in which various cell structures
Song et al., 2014), and shows high-efficiency potential for the near with different buffer layers can be simulated and optimized. To find the
future (Ki and Hillhouse, 2011)). Compare to CIGS system, CZTS based possible alternative of CdS buffer in both CZTS and CIGS, some nontoxic
solar cells exhibit an improved photovoltaic behavior at high working buffer layers such as ZnSe, SnO2, ZnO, ZnS, ZnMgO, In2S3 etc. were
temperature (Abderrezek et al., 2018). In addition, quaternary kesterite simulated (Hariskos et al., 2005). In this regard, we propose to use an n-
CZTS TFSCs have been very fascinated material from last 10 years, as type TiO2 layer as an alternative to the CdS buffer layer for CZTS thin-
they exhibit high absorption coefficients (>104 cm− 1) (Mitzi et al., film solar cells. It can be a promising electron transport layer (ETL)
2011) and a tunable band gap (1.4 eV–1.6 eV) (Xia et al., 2013; for the CZTS based solar cells due to its suitable band alignment with
Wangperawong et al., 2011; Zhou et al., 2011; Araki et al., 2009; CZTS.
Katagiri et al., 2001; Dhakal et al., 2011) that can enhance the PCE as Different HTLs such as TiN, TiB2, and Ag have been researched
well. (Scragg et al., 2013; Liu et al., 2014; Cui et al., 2014) and received a lot
The heat treatment of a CZTS-based heterojunction solar cell allowed of attention due to their ability to collect holes more efficiently and
the PCE of more than 10.0% (Yan et al., 2017). Green et al. (2018) re­ reducing electron-hole recombination. However, high series resistance
ported the PCE of 11% and Voc of 730.6 mV from CZTS-based solar cell. was observed due to the TiN/CZTS interface that reduced the fill factor
A record PCE of around 12.6% was achieved for a CZTS-based solar cell (FF) and cell performance. In addition, when TiB2 was used as the HTL of
demonstrated by Wang et al., employing a pure hydrazine solution the CZTS solar cell, a decrease in open-circuit voltage was also observed.
approach (Wang et al., 2013). A maximum efficiency of 17.59% and Voc Similarly, inserting a silver layer can improve cell efficiency, but it de­
of 940 mV were observed for a silver-mixed CZTS cell with a CdS/ grades the crystallinity of CZTS films. Substrates coated with Mo are
ACZTS/CZTS structure, as indicated by optoelectronic simulations (Saha widely used for the deposition of CZTS absorber layers. However, the use
and Alam, 2018). Y. H. Khattk et al. demonstrated the theoretical of Mo-coated substrates is restricted due to the instability of the CZTS/
maximum PCE of 23.72%, for CZTS-based solar cells (Khattak et al., Mo interface, the formation of the MoS2 layer, the secondary phase, and
2018). This highly promised CZTS material currently has not achieved the voids at the back-contact interface, which may adversely affect cell
the efficiency comparable to solar cells based on CIGS and CdTe. As the performance. The effects of self-constituent layers (such as CuS, SnS, and
Shockley-Queisser theory shows, the maximum PCE of 32.2% can be ZnS) between CZTS and Mo back contact on the phase purity, crystal­
conceived using single junction CZTS-based solar cells (Shockley and linity, interface quality, and element segregation of CZTS films are
Queisser, 1961). Further research into CZTS solar cells is, accordingly, studied by Chen et al. (2016). The formation of the CuS phase introduces
expected to improve their PV performances. CZTS based TFSCs did no severe interface defects, such as voids. However, the ZnS layer on the
longer meet their feasible objectives due to possible challenge of defect- back contact creates an additional p-n junction, which reduces cell
less absorber layer fabrication, non-optimized buffer layer matching and performance by increasing series resistance. In this context, a layer of p-
non-optimized device configuration. In particular, the larger potential type SnS thin films is proposed as a back-surface passivation layer to
fluctuations and higher defects density at the interface, by comparing enhance the PV performance of the device. The added SnS layer reacts
with CIGS, is identified as the massive Voc deficit in CZTS devices. Since with the secondary phases (such as ZnS and CuS) to form a pure CZTS
Cu and Zn are neighbouring elements in the periodic table, the changes phase, thereby greatly suppressing the formation of these secondary
in bonding are anticipated during the formation of CZTS thin films, this phases in the CZTS absorber layer (Karade et al., 2019). In this work, the
enables Cu and Zn cation intermixing, referred to as antisite point de­ SCAPS-1D software is used to simulate the specific (SnS/CZTS/TiO2/
fects CuZn and ZnCu, as affirmed by neutron diffraction (Schorr et al., ITO) structure of the solar cell to study and optimize the various PV
2007). Antisite point defects induce the electrostatic potential performance parameters of the device. In this study, Nickel (Ni) (1 1 1)

65
M. Atowar Rahman Solar Energy 215 (2021) 64–76

Load 3
Spike offset = + 0.3 eV; key for high VOC

2
EC
Ni
1
SnS

Energy (eV)

TiO2
SnS

ITO
CZTS Light
CZTS 0
Al
EV
TiO2
-1
ITO
Glass Substrate
-2

Light 0.0 0.1 0.2 0.3 5.0 5.1 5.2

Position (µm)
(a) (b)
Fig. 1. (a) Schematic diagram and (b) energy band diagram of the Al/ITO/TiO2/CZTS/SnS/Ni heterojunction solar cell.

with the metal work function of 5.35 eV (Baker et al., 1971; Michaelson,
Table 1
1977) is used as back contact materials, as it is known to form a nearly
Material parameters, used for simulating (Al/ITO/TiO2/CZTS/SnS/Ni) solar cell
ohmic contact with very low contact resistance with SnS (Hajzus et al.,
(Lin et al., 2016b; Mouchou et al., 2020; Benami, 2019).
2018), whereas, Aluminium (Al) is used as front contact materials. The
PV performance parameters are studied by tuning the back- contact Parameters ITO TiO2 ETL Absorber SnS HTL
(CZTS)
metal work function, working temperatures and the thickness, carrier
concentration, defect density of different layers to obtain an optimum Thickness (nm) 50 20–60 1000–5000 200 nm
Bandgap Eg (eV) 3.5 3.26 ( 1.52 (Dhakal 1.6(Yue
cell performance.
Bouachiba et al., 2014) et al.,
et al., 2014) 2009)
2. Materials and methods Electron Affinity, χ 4.6 4.20 4.5 4.1
(eV)
Relative permittivity, 8.9 10 10 13
2.1. Solar cell structure
εr
17 18
Effective conduction 2.2 × 2.0 × 10 2.2 × 10 1.18 ×
TFSCs may be described as containing an absorber layer covered by band density NC 1018 1018
two buffer layers, one above it and the other below it. The absorber layer (cm− 3)
is a semiconductor that absorbs photons, thereby generating electrons in Effective valance 1.8 × 6.0 × 1017 1.8 × 1019 4.46 ×
band density NV 1019 1018
its conduction band and holes in its valence band. One buffer layer is
(cm− 3)
designed to easily transport the photo-generated electrons (PGEs) to an Electron Thermal 1.0 × 1.0 × 107 1.0 × 107 1.0 × 107
adjacent electrode while blocking the photo-generated holes (PGHs) Velocity (cms− 1) 107
from entering that electrode. The other buffer layer, on the opposite side Hole Thermal 1.0 × 1.0 × 107 1.0 × 107 1.0 × 107
of the absorber layer, is used to transmit PGHs, while blocking the Velocity (cms− 1) 107
Electron mobility, μn 10 100 100 15
passage of PGEs into the other electrode. Both buffer layers are semi­ (cm2V− 1s− 1)
conductor materials with wide band gaps, so that very few, if any, Hole mobility, μp 10 25 20 100
incoming incident photons are absorbed within the buffer layers. Fig. 1 (cm2V− 1s− 1)
15 19
(a) shows the schematic of the proposed solar cells, which reads as (Al/ Donor concentration, 1.0 × 10 -10 0.0 0.0
ND (cm− 3) 1021
ITO/TiO2/CZTS/SnS/Ni). In this cell, Al, ITO, TiO2, CZTS, SnS, and Ni
Acceptor 0.0 0.0 1014-1018 1015-
are used as front contact materials, windows layer, electron transport concentration, NA 1019
layer (ETL), absorber layer, hole transport layer (HTL), and back contact (cm− 3)
materials, respectively. CZTS, TiO2 and SnS, which are the absorber, ETL Defect density, Nt 1.0 × 1.0 × 1016 1014-1018 1.0 ×
and HTL, respectively, formed the heterojunction and constitute the key (cm− 3) 1016 1016
− 16 − 16
Capture cross section 1.0 × 1.0 × 10 1.0 × 10 1.0 ×
parts of the device. The simulated energy band diagram of the proposed of electrons (cm− 2) 10− 16 10− 16
heterojunction solar cells is visualized in Fig. 1(b). It is observed from Capture cross section 1.0 × 1.0 × 10− 16
1.0 × 10− 16
1.0 ×
Fig. 1(b) that the conduction band of the CZTS absorber layer is lower of holes (cm− 2) 10− 16 10− 16
than that of TiO2 ETL and the conduction band offset (CBO) between
them is +0.3 eV. The conduction band energy of the effective ETL must
interface. Second, a spike CBO between them removes electrons from
be close to that of the absorber layer. Thus, the photo-electrons have an
the CZTS/TiO2 interface and reduces the non-radiative electrons-holes
easy transit through the absorber/ETL interface and into a collection
recombination at the interface that would reduce Jsc sharply. T. Mine­
electrode whose Fermi energy is not too far from either of these con­
moto et al. reported that spike CBO in between +0.3 to +0.5 eV at the
duction band minimum energies. In our proposed device, the conduction
absorber/ETL interface are beneficial (Minemoto et al., 2001a); beyond
band minimum (CBM) of the n-type TiO2 ETL is slightly higher than that
this level, higher barriers can reduce current flow and hence efficiency
of the p-type CZTS absorber (spike band offset). At first glance, one may
of the device. Because the band gap of TiO2 ETL is 3.26 eV, which is
wonder why it is necessary to form a small barrier for electrons at the
much larger than that of the CZTS absorber, the valence band energy of
absorber/ETL interface. The reasons for this are two-fold. The first is
the absorber is much lower in energy, and thus the PGHs cannot enter
that the ETL CBM below the absorber (cliff band offset) can reduce Voc
the TiO2 buffer layer. Thus, n-type TiO2 establishes an appropriate
by increasing electron-hole non-radiative recombination at the
junction with CZTS to transport PGEs through the CZTS/TiO2 interface,

66
M. Atowar Rahman Solar Energy 215 (2021) 64–76

Table 2 Table 4
Parameters used for defects at interfaces of solar cells. PV performance parameters for different cell configurations.
Parameters SnS/CZTS CZTS/TiO2 Cell Configuration VOC JSC (mA/ FF (%) PEC Device
(Volt) cm2) (%) ID
Defect type Neutral Neutral
Capture cross section of electrons (cm2) 1.0 × 10− 20 1.0 × 10− 20 CZTS/CdS/ITO 0.68 30.13 71.45 14.67 Ref.Cell
Capture cross section of holes (cm2) 1.0 × 10− 20 1.0 × 10− 20 SnS/CZTS/CdS/ 1.03 30.25 76.10 23.77 Cell-1
Energetic distribution Single Single ITO
Total defect density (cm− 2) 1 × 1014–1 × 1018 1 × 1014–1 × 1018 SnS/CZTS/TiO2/ 1.099 31.89 87.85 30.79 Cell-2
ITO
CZTS/TiO2/ITO 0.92 30.99 77.38 22.15 Cell-3

Table 3
Front and back contact parameters used in simulation. recombination (non-radiative recombination). Table 1 lists the baseline
Parameters Back contact Front contact parameters for different layers of the device and their values used to
electrical properties electrical properties perform the calculations. We have introduced one type of single-level
Surface recombination velocity of 1.0 × 105 1.0 × 107 defect in each semiconductor layer. Considering the recombination at
electrons (cms− 1) two interfaces (CZTS/TiO2 and SnS/CZTS), reasonable neutral interface
Surface recombination velocity of 1.0 × 107 1.0 × 105 defects are also used for recombination. Table 2 summarizes the inter­
holes (cms− 1)
facial defect parameters used in the simulation. The reflectivity of the
Work function (eV) 5.35; Ni (1 1 1) 4.06; Al
Working Temperature (K) 280–380
front and back surface is set to 0.1 and 0.9, respectively (Zhao et al.,
2012). Parameters for the front contact, back contact, and working
temperatures are attached in Table 3. For the simulation, the experi­
where holes would be blocked at TiO2. An effective hole-transporting mental absorption coefficient data of the SnS, CZTS,
buffer layer has opposite properties as that of ETL. Its valence band TiO2, and ITO were used from the literature (Yue et al., 2009;
maximum (VBM) energy should closely match that of the absorber layer. Sharmin et al., 2020; Bouachiba et al., 2014; Khusayfan and El-Nahass,
In our device, the VBM energy of SnS HTL is slightly above (spike) to that 2013). All SCAPS-1D simulations are performed under the AM 1.5
of the CZTS absorber so that PGHs from CZTS could be efficiently spectrum, and the shunt resistance and series resistance are fixed at
transported through SnS HTL to the back electrode. The conduction 1000 Ω.cm− 2 and 2.5 Ω.cm− 2, respectively.
band energy of SnS HTL is suitably larger (CBO = +0.4 eV) than that of
the CZTS absorber layer so that PGEs are blocked by this barrier at SnS/ 3. Results and discussions
absorber interface. Therefore, SnS forms a suitable junction with CZTS
absorber to transport holes through the SnS HTL to the back electrode, 3.1. Enhancement of VOC of CZTS solar cells
whereas it blocks electrons to reach the back electrode.
CZTS-based solar cells with four different cell configuration were
simulated: (i) a reference solar cell with the standard CZTS/CdS struc­
2.2. Numerical modeling and material parameters
ture, hereafter we called it Ref.Cell; (ii) a solar cell with an SnS layer, so-
called passivation layer, in-between the back contact and the CZTS
This work is accomplished by Solar Cell Capacitance Simulator
absorber, the (SnS/CZTS/CdS/ITO) structure, hereafter called it Cell-1;
(SCAPS-1D), a one-dimensional software based on solving three basic
(iii) a Cell-2 (SnS/CZTS/TiO2/ITO) structure, where CdS layer in Cell-1 is
semiconductor device equations; Poisson’s equation, the continuity
replaced by a TiO2 ETL layer, and (iv) a solar cell with the device
equations for free holes and electrons, under steady-state conditions
structure of CZTS/TiO2/ITO, hereafter the cell will be identified as Cell-
(Burgelman et al., 2000). We have used the SCAPS-1D simulator to
3. Fig. 2(a) shows representative of four J-V curves for the aforemen­
explore the proposed (Al/ITO/TiO2/CZTS/SnS/Ni) solar cells with
tioned four device structures. The PV performance parameters of the
optimized material parameters for better performance. The software
four solar cell devices are listed in Table 4. Despite good FF (>70%) and
allows for the inclusion of deep bulk level and interface defect

Fig. 2. (a) J-V characteristics of Ref.Cell, Cell-1, Cell-2 and Cell-3 cell with the device structure of (CZTS/CdS/ITO), (SnS/CSTS/CdS/ITO), (SnS/CZTS/TiO2/ITO) and
(CZTS/TiO2/ITO) respectively, (b) External quantum efficiency (EQE) spectra of CZTS-based solar cells with different configuration as indicated on the graph.

67
M. Atowar Rahman Solar Energy 215 (2021) 64–76

Table 5 from 71.5% to 76% in FF, and a small increase from 30.13 mA/cm2 to
Lattice mismatch* of different HTLs, ETLs with CZTS layer. 30.25 mA/cm2 in Jsc. Therefore, the JV analysis shows that the insertion
Layer Lattice Parameters Lattice mismatch of the SnS as HTL in the Ref.Cell structure is beneficial and produces
devices with a better electrical performance in terms of Voc, FF, and Jsc.
a (Å) b (Å) c (Å)
After the introduction of SnS HTL as a passivation layer, due to the lower
CZTS 5.435 5.435 10.843 – interface recombination, the increase in the Voc value in the device is
SnS (HTL) 4.28 11.41 3.96 5.09%
ZnS (HTL) 3.82 3.82 6.257 14.06%
expected and well known (Glunz et al., 1999; Taguchi et al., 2005). In
TiB2 (HTL) 3.035 3.035 3.24 50.0% particular, this increase in Voc may be attributed due to the suitable band
CdS (ETL) – – 6.677 20.5% offset between SnS and CZTS absorber which introduces a built-in po­
TiO2 (ETL) 3.66 3.66 9.76 10.5% tential at the interface and hence reduces minority carrier recombina­
*δ = 2|as − ae|/(as + ae). tion at the SnS/CZTS interface (Xiao et al., 2017; Zhang et al., 2018;
Wolfe et al., 1989). Moreover, the SnS layer not only has all the above-
moderate Jsc values (>30 mA/cm2), we note that the efficiency of the mentioned advantages over other HLTs but also has a lower lattice
Ref.Cell is being limited mostly by low Voc values (<680 mV). We mismatch (5.09%, see Table 5) with CZTS film, resulting in a further
attribute the low Voc values of the Ref.Cell due to the high minority reduction in the carrier recombination at the SnS/CZTS interface. The
carrier surface recombination at the CZTS absorber/back electrode small Jsc gain of the Cell-1 might be due to optical (e.g. from an increased
interface, as the Voc of the solar cell inversely proportional to the loga­ reflection of the back contact) or electronic effects caused by the
rithm of minority carrier surface recombination rate at the interface. passivation layer of SnS.
The minimization of dark current/minority carrier recombination The mismatch of the energy levels at different interfaces was found to
current (Io) is crucial for the optimization Voc in general and hence the be the main factor that leads to non-radiative recombination for energy
performance of solar cells. As the absorber of the Ref.Cell is heavily losses and decreased output voltage (Chueh et al., 2015; Lin et al.,
affected by back contact recombination, making them ideal to evaluate 2016a). Therefore, the energy level matching at the front contact side of
the effects of the passivation layer (SnS HTL) in-between back contact the solar cell is also extremely critical and interactive to reach the
and CZTS absorber. The differences in the PV performance parameters highest the Voc and PCE. In this work, further improvement of PV per­
between the Ref.Cell and passivated cell (Cell-1) are striking (see Fig. 2 formances is achieved by replacing the CdS ETL layer of Cell-1 with the
(a)), the Ref.Cell achieves, on average, an efficiency value of 14.67% TiO2 ETL, as the CdS layer more conveniently forms a cliff type CBO of −
whereas Cell-1 exhibits a value of 23.77%. This large increase in effi­ 0.5 eV with CZTS absorber (Yan et al., 2017; Kumar and Thakur, 2018),
ciency is due to an increase from 680 mV to 1.03 V in Voc, an increase which decreases the activation energy (Ea) for recombination at the

VOC (Volt) JSC (mA/cm2)

1018 1.143 1018 32.80
Carrier Concentration (cm-3)

Carrier Concentration (cm-3)

1.119 32.32

1017 1.097 1017 31.84

1.074 31.35
16
10 1.051 10 16 30.87

1.028 30.39

10 15 1.004 10 15 29.91

0.9815 29.42

1014 0.9585 1014 28.94

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)
FF (%) PCE (%)
1018 87.86 1018 30.80
Carrier Concentration (cm-3)

87.24
Carrier Concentration (cm-3)

30.13

1017 86.62 10 17
29.46

86.00 28.79
16
10 85.38 10 16 28.12

84.76 27.45

10 15 84.14 1015 26.78

83.52 26.11

1014 82.90 1014 25.44

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

Fig. 3. Variations of PV performance parameters depending on the carrier concentration and thickness of the CZTS absorber layer.

68
M. Atowar Rahman Solar Energy 215 (2021) 64–76

1021

Total Recombination Rate (cm-3.s-1)

NA = 1×1014 cm-3

Total Recombination Rate (cm-3.s-1)

NA = 1×1014 cm-3 NA = 1×1015 cm-3
NA = 1×10 15 -3
cm NA = 1×1016 cm-3
1022
19 NA = 1×1016 cm-3 NA = 5×1016 cm-3
10
NA = 5×10 16 -3
cm NA = 1×1017 cm-3
NA = 1×10 17 -3
cm NA = 5×1017 cm-3

NA = 5×1017 cm-3 1021 NA = 1×1018 cm-3

1017 18 -3
NA = 1×10 cm

1015 1020

SnS CZTS

1013 1019
0.14 0.16 0.18 0.20 0.22 0.24 0.26 5.0 5.1 5.2
Position (µm) Position (µm)
(a) (b)
Fig. 4. Total recombination rate (a) at the SnS/CZTS interface and (b) CZTS/TiO2 interface.

CZTS/CdS interface; therefore, it reduces Voc and hence the performance 3.2. Effect of CZTS layer thickness and carrier concentration on the cell
of the cell. Moreover, at forward bias, this cliff CBO at the CZTS/CdS performance
interface of Cell-1 acts as the barrier against injected electrons from ITO
toward the CZTS absorber, results in increasing recombination between The absorber layer thickness and carrier concentration are the key
majority carriers via defects at the CZTS/CdS interface (Minemoto et al., choices for structuring the TFSCs, as it is generally influencing the
2001b). Therefore, as the cliff conduction band offset and the defect photo-generated excitons and charge carrier extraction. Therefore, the
density at the interface increase, the overall recombination increases, optimization of the carrier concentration and thickness of the absorber
while Voc decreases. As it is seen from the Fig. 1(b) that in the simulated layer is crucial to achieve optimum cell performance. According to the
energy band diagram of our proposed Cell-2, where the CdS layer in Cell- literature, the thin CZTS absorber layer of CZTS based solar cells showed
1 is replaced by the TiO2 layer, the TiO2 layer forms a spike type CBO of improved photoconversion efficiency (Shin et al., 2011; Wang et al.,
+0.3 eV at the CZTS/TiO2 interface. The solar cell with TiO2 as a ETL 2010; Wanda et al., 2016). On the contrary, H. Heriche et al. achieved
exhibited enhanced PV performances of the Voc ≈ 1.1 V, Jsc ≈ 31.9 mA/ high conversion efficiency in the CZTS based solar cell with thin buffer
cm2, FF ≈ 88%, and PCE ≈ 30.8% as shown in Table 4. Concerning the layers and a thick absorber layer (Heriche et al., 2014). Qualitative in­
Voc and Jsc values, the 70 mV difference in Voc and 1.65 mA/cm2 dif­ formation has been obtained by investigating the effects of the absorber
ference in Jsc between Cell-1 and Cell-2 can undoubtedly be attributed to layer thickness and carrier concentration on the PV performance pa­
effects caused by the replacement of the CdS layer with the TiO2 layer. rameters of our proposed solar cells (Cell-2), keeping the properties of
Furthermore, compared with the CZTS/CdS interface, the lattice other layers are constant. This is shown in Fig. 3. The carrier concen­
mismatch at the CZTS/TiO2 interface is lower (see Table 5), which re­ tration and thickness of the CZTS absorber layer were varied from 1014
sults in reduced interface defects and reduced carrier recombination, to 1018 cm− 3 and 1.0 to 5.0 µm, respectively. It is observed from the
thereby further improving the PV performance of the device. Therefore, figures that at the carrier concentration > 1017 cm− 3, the Voc was nearly
the replacement of CdS with TiO2 ETL in the structure significantly independent of the thickness of the CZTS absorber layer. It is also seen
improves the PV performances of the solar cells. As a result, this theo­ from the figures that the Jsc was increased with the CZTS thickness at
retical analysis opening a new possibility of using SnS and TiO2 layer as carrier concentration of < 1016 cm− 3, which was resulted due to the
an effective HTL and ETL, respectively, in Cd-free CZTS solar cells. It is increase in absorption of higher wavelength photon in the absorber
also interesting to observe the impact of the SnS layer on the cell per­ layer. In particular, The Jsc was increasing up to the thickness of 3.5 µm
formance by comparing the J-V curve of Cell-2 with that of Cell-3. It is and after that, saturated due to the saturation of the absorption of light.
obvious that both Jsc and Voc have been enhanced by introducing a At the carrier concentration of below 1016 cm− 3, the PCE was poor and
passivation layer (SnS HTL) between back contact and the CZTS the PCE increased with CZTS thickness, as Jsc increased and saturated at
absorber. Fig. 2(b) delineates the EQE spectra of all the solar cells for the 3.5 µm. The FF was constant and not dependent on the thickness of the
wavelength range between 250 nm and 950 nm. By comparing the EQE absorber for the concentration in the range between 5 × 1016 to 5 × 1017
spectra of Ref.Cell and Cell-1, it is possible to observe the influence of the cm− 3. We remarked that all PV performance parameters of the device
SnS passivation layer on the EQE of the proposed cell (Cell-2) and the such as Voc, FF, and PCE were increasing up to the carrier concentration
impact of the TiO2 buffer layer on the EQE of the device can be observed of 5 × 1016 cm− 3 for the absorber, corresponding to the zone where the
by comparing the EQE spectra Cell-1 and Cell-2. In particular, The EQE is device presents a relatively poor efficiency. This is mainly due to the
observed to improve at longer wavelength region when the SnS layer is high recombination of the photo-generated carriers in the bulk region of
introduced. This is due to the back surface field which effectively created the absorber as shown in Fig. 4; photo-generated carriers are recom­
by the SnS layer, thereby efficiently collecting photogenerated carriers. bined before reaching both interfaces.
Moreover, the SnS layer on the back of the CZTS layer can enhance the We also observed that the PV performance parameters of the cell
back reflection of near-infrared light, resulting in a slight increase in were deteriorated at CZTS carrier concentration > 1017 cm− 3. Particu­
near-infrared light absorption in the CZTS layer (Ren et al., 2017). larly, Jsc and FF were reduced at the higher carrier concentration of
Furthermore, an enhanced EQE of the device (Cell-2) at the wavelength CZTS absorber, which might be due to the increase of Auger recombi­
below 520 nm is observed when compared with the Ref.Cell. This is due nation. Moreover, Jsc varies from 32.8 mA/cm2 (the highest value ob­
to the replacement of the CdS layer with the TiO2 buffer layer, as CdS tained during this simulation) for 5 × 1016 cm− 3 to 29.5 mA/cm2 for 1.0
with Eg = 2.42 eV absorb light below 520 nm. × 1018 cm− 3 i.e. a loss of about 3.3 mA/cm2, corresponding to sup­
pressed hole transportation at the CZTS absorber layer due to the

69
M. Atowar Rahman Solar Energy 215 (2021) 64–76

Table 6 interface. This suggests that at higher hole concentration, the enhanced
Comparative analysis of simulated results of CZTS-based solar cells. non-radiative recombination within SCR of the CZTS/TiO2 interface
Solar Voc Jsc FF PCE CZTS References might be another reason for the observed Jsc reduction. Therefore, we
cells (Volt) (mA/ (%) (%) thickness anticipate that in CZTS-based solar cells, the trade-off between the
cm2) thickness of the absorber and the carrier concentration is very important
Ni/SnS/ 1.099 31.89 87.85 30.79 4.0 µm This work to improve its PV performances. From this study, a maximum PCE of ≈
CZTS/ 31% was found at the carrier concentration of 5 × 1016 cm− 3 with a
TiO2/ CZTS absorber layer thickness of > 4.0 µm. The reported results ob­
ITO/Al
Mo/ 1.002 19.31 69.35 13.41 2.68 µm (Patel and Ray,
tained by the numerical simulation are presented in the Table 6. Our
CZTS/ 2012) findings are considered to be in line with the results presented
CdS/ previously.
ZnO/Al
CZTS/ 0.64 23.96 65.2 10.0 2.70 µm (Cherouana
ZnS/ and Labbani,
ZnO 2017)
3.3. Effect of TiO2 ETL thickness and carrier concentration on the cell
Mo/ 0.87 23.04 83.0 16.38 3.00 µm (Zhao et al., performance
CZTS/ 2012)
CdS The effect of TiO2 ETL thickness and carrier concentration on cell
Mo/ 0.64 44.87 82.54 23.72 4.00 µm (Khattak et al.,
performance was investigated, as it is important in terms of stability and
CZTS/ 2018)
CdS/ performance of the device, and presented in Fig. 5. The thickness of ETL
ZnO was considered as thinner than the other layers, whereas, the ETL with
relatively high carrier concentration was adopted for this study. The
carrier concentration and thickness of the TiO2 ETL was varied from
increase of impurity scattering at higher carrier concentration of 1015 to 1019 cm− 3 and 20 to 60 nm, respectively. It is observed from the
absorber, which in turn increased the resistivity of the absorber layer figures that Voc, FF, and PCE showed similar behavior with the thickness
and hence reduced Jsc and FF. It is also interesting to note from Fig. 4(b) and carrier concentration of TiO2 ETL. All of these PV performance
that at higher hole concentration (>1017 cm− 3), carrier recombination parameters of the device decrease with increasing the thickness, but
within the bulk region of CZTS reduced significantly but it increased increases with the carrier concentration of TiO2. This improvement of
sharply within the space charge region (SCR) of the CZTS/TiO2 PV performance with carrier concentration were saturated at the carrier

VOC (Volt) JSC (mA/cm2)

1019 1.081 1019 31.95
Carrier Concentration (cm-3)

Carrier Concentration (cm-3)

1.077 31.92
18
10 1.074 1018 31.88

1.070 31.85
17 17
10 1.066 10 31.82

1.063 31.79

10 16 1.059 10 16 31.76

1.055 31.72

1015 1.052 31.69

1015
20 25 30 35 40 45 50 55 60 20 25 30 35 40 45 50 55 60
Thickness (nm) Thickness (nm)

FF (%) PCE (%)

1019 87.70 1019 30.26
Carrier Concentration (cm-3)
Carrier Concentration (cm-3)

85.36 29.37
18 18
10 83.03 10 28.48

80.69 27.60

1017 78.35 1017 26.71

76.01 25.82

10 16 73.67 10 16 24.94

71.34 24.05

1015 69.00 1015 23.16

20 25 30 35 40 45 50 55 60 20 25 30 35 40 45 50 55 60
Thickness (nm) Thickness (nm)

Fig. 5. Variations of PV performance parameters depending on the carrier concentration and thickness of the TiO2 ETL.

70
M. Atowar Rahman Solar Energy 215 (2021) 64–76

VOC (volt) JSC (mA/cm2)

1E19 1.082 1019 31.96

Carrier Concentration (cm-3)

Carrier Concentration (cm-3)
1.081 31.95
18
1E18 1.081 10 31.95

1.081 31.95

1E17 1.080 1017 31.94

1.080 31.94

1E16 1.080 1016 31.94

1.079 31.93

1.079 1015 31.93

1E15
100 150 200 250 300 350 400 450 500 100 150 200 250 300 350 400 450 500
Thickness (nm) Thickness (nm)

FF (%) PCE (%)

1019 87.66 1019 30.29

Carrier Concentration (cm-3)

Carrier Concentration (cm-3)

87.33 30.18
18
10 18
87.00 10 30.07

86.66 29.96

1017 86.33 1017 29.85

86.00 29.74

1016 85.67 1016 29.63

85.33 29.53

85.00 29.42
1015 1015
100 150 200 250 300 350 400 450 500 100 150 200 250 300 350 400 450 500
Thickness (nm) Thickness (nm)

Fig. 6. Variations of PV performance parameters depending on the carrier concentration and thickness of the SnS HTL.

concentration of > 1017 cm− 3. The improved PCE of the device is of 30 nm which will work better for this cell.
attributed to the increased Jsc and FF for a better collection of the
photogenerated electrons with the thickness of the TiO2 layer varied 3.4. Effect of SnS HTL thickness and carrier concentration on the cell
from 20 nm to 35 nm. When the TiO2 layer was thicker than 35 nm, the performance
PV performance of the device was deteriorated due to increased re­
sistivity of TiO2 ETL, as revealed by decreased Jsc and FF for the devices ETL and HTL play a vital role in the collection of photogenerated
with TiO2 thickness > 35 nm. Moreover, higher absorption of photons in charge carriers in the absorber layer. Due to low binding energy and
thick TiO2 ETL suppresses the light absorption in the CZTS absorber high diffusion lengths of the photo generated excitons in the absorber,
layer, which also may reduce the performance of the device as well. It is they reach to the interfaces and are dissociated by the SCR field of the
important to note that the difference of Voc of the device, when the TiO2 interfaces. Consequently, electrons are transported to ETL and holes to
ETL thickness was varied from 20 nm to 60 nm, was very small, which the HTL layer, simultaneously, whereas holes and electrons trans­
indicates that a variation of the TiO2 ELT thickness produces negligible portation are blocked by the ETL and HTL, respectively. The effect of the
leakage of electrons and holes at the interface. The evolution of Jsc as a carrier concentration and thickness of SnS HTL was evaluated on the
function of carrier concentration presents three zones, such as shown in performance of the device. As illustrated in Fig. 6, Jsc and Voc are slightly
Fig. 5: the first zone is from 1015 to 5 × 1018 cm− 3 and for the thickness increasing with a corresponding increase in the SnS HTL doping con­
between 20 and 30 nm, where Jsc is nearly constant irrespective of centration and almost independent of thickness. The increase of Jsc and
thickness and carrier concentration; the second zone corresponds to > 5 Voc with an increase of carrier concentration signifies the reduction of
× 1018 cm− 3 and for the thickness between 20 and 30 nm, where Jsc photo-generated charge carrier recombination. It is also evidenced from
decreases with the carrier concentration; and the third one corresponds Fig. 6 that the value of FF and PCE is high for the carrier concentration
to > 5 × 1017 cm− 3 and for the thickness > 30 nm; where Jsc decreases > 1018 cm− 3 but again it is independent of the thickness of SnS HTL. A
with the thickness and carrier concentration TiO2 ETL. In the second strong electric field at the HTL/CZTS interface produced by high carrier
zone, the electron transport is limited by the electron–electron scat­ concentration blocks the flow of minority electrons towards the inter­
tering with increasing the carrier concentration, whereas, the behavior face, thus reducing the interface recombination.
of Jsc in the third zone is attributed to the enhanced scattering and de­ Moreover, the Fermi level of HTL shifts towards the valence band due
fects in ETL with increasing the TiO2 carrier concentration and thick­ to high carrier concentration, thus an efficient collection of holes at the
ness. In the study, maximum PCE was observed at a very thin layer of back-contact is achieved by forming a nearly ohmic contact with the
TiO2 and carrier concentration of 1018 cm− 3. However, it is too hard to back-metal electrode (Ni) (Rolland et al., 2017). The optimum PV per­
deposit a very thin ETL layer but it is practicable to achieve a thickness formances of the device were obtained at SnS carrier concentration and

71
M. Atowar Rahman Solar Energy 215 (2021) 64–76

VOC (Volt) JSC (mA/cm2)

1018 1.102 1018 31.91

1.090 31.59
Defect Density (cm-3)

Defect Density (cm-3)

17
10 1.078 10 17
31.28

1.065 30.96

1016 1.053 1016 30.64

1.041 30.32

10 15 1.029 10 15 30.01

1.017 29.69

1014 1.004 1014 29.37

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

FF (%) PCE (%)

1018 88.38 1018 31.08

87.42 30.17

Defect Density (cm-3)

Defect Density (cm-3)

1017 86.47 1017 29.27

85.51 28.36
16 16
10 84.55 10 27.45

83.59 26.54

1015 82.63 1015 25.64

81.68 24.73

1014 80.72 1014 23.82

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

Fig. 7. Variation of PV performance parameters with defect density and thickness of the absorber layer in CZTS solar cells.

thickness of 1018 cm− 3

and 200 nm, respectively. the CZTS absorber layer thickness when the defects density is below 5 ×
1016 cm− 3. The PV performance parameters are started deteriorating
when the defects density exceeds this value. Particularly, when thick­
3.5. Effect of CZTS absorber layer defects density on cell performance
ness > 4.0 µm and defect density < 1016 cm− 3, the highest Voc of 1.10 V
is obtained, whereas Voc drops rapidly to 1.0 V when the defect density
The optoelectrical properties of materials are profoundly reliant on
is > 5 × 1016 cm− 3. In contrast, Jsc shows a maximum value of > 31.9
their thickness, preparation methods, and techniques by which the
mA/cm2 at thicknesses and defect densities > 4.5 µm and < 1017 cm− 3,
properties are measured and it is very difficult to define material
respectively. Decreasing the thickness from 4.5 µm to 1.0 µm causes the
properties accurately based on the variation of thickness (Ferdaous
Jsc to drop by 2.5 mA/cm2, and a decrease in Jsc can also be observed at
et al., 2019; Shahahmadi et al., 2015). Moreover, the optoelectrical
defects > 1017 cm− 3. FF > 88% is observed when the defect density is <
properties of materials can change due to defects present in the material
5 × 1016 cm− 3, and is slightly affected by the thickness. Conclusively,
system. Therefore, it is important to study the influence of the defect
when the thickness is > 4.5 µm and the defect density is < 5 × 1016
density in absorber layer solar cells in various absorber layer thickness,
cm− 3, the highest observed PCE is > 31%. All the above findings suggest
as defects at the absorber layer of solar cells are unfavorable for its
that the defect density of the absorber layer has an adverse effect on the
performance. CZTS is a defect prone system and consists of various types
performance of CZTS solar cells. We found that increasing the thickness
of defects like acceptor and donor vacancies, antisites, and interstitials
of the absorber layer can mitigate this problem.
defects. Characteristic data for all kinds of possible defects in the CZTS
system is rare. There is a provision to introduce different donor and
acceptor defects in SCAPS, however, in this simulation, we introduced 3.6. Effect of interfacial defect density on cell performance
only single acceptor like defects state in the CZTS absorber layer due to
the fact that the formation energy acceptor defects are lower than donor Due to the difference in a lattice (or thermal) mismatch and various
ones (Chen et al., 2010). Fig. 7 presents the effect of defects density and crystallographic structures of the absorber and buffer layers, atomically
thickness of CZTS absorber on the solar cell performance parameters. sudden interfaces with a network of dislocations is formed between two
The effect of defect density from 1014 to 1018 cm− 3 and thickness from layers, which act as electronic defects and promote the recombination
1.0 to 5.0 µm are analyzed to assess the performance of the device. We process. In addition, during the fabrication process, the inter-diffusion of
see that the solar cell performance does not change and is dominated by metal cations between adjacent layers induces the structural defects at

72
M. Atowar Rahman Solar Energy 215 (2021) 64–76

VOC (Volt) JSC (mA/cm2)

1018 1.099 1018 31.90

1.093 31.59
Defect Density (cm-2)

Defect Density (cm-2)

17
10 1.088 1017 31.28

1.082 30.97

1016 1.076 10 16
30.66

1.071 30.36

10 15 1.065 10 15 30.05

1.059 29.74

1014 1.053 1014 29.43

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

FF (%) PCE (%)

1018 87.85 1018 30.80

87.76 30.33

Defect Density (cm-2)

Defect Density (cm-2)

17
1017 87.68 10 29.86

87.59 29.38

1016 87.50 1016 28.91

87.41 28.44

1015 87.32 10 15 27.97

87.24 27.49

1014 87.15 1014 27.02

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

Fig. 8. Variations of PV performance parameters with the defect density of CZTS/TiO2 interface and thickness of the absorber layer in CZTS solar cells.

the interfaces (Kumar and Rao, 2014). The impact of CZTS/TiO2 inter­ 3.7. Effect of back metal work function and working temperature on cell
face defect density on the cell performance was investigated and performance
depicted in Fig. 8. The defect density of the SnS/CZTS interface was
varied from 1014 to 1018 cm− 2 to evaluate the cell performance. The The back-contact material has a great influence on the performance
trend of overall performances of the device is very similar to the pre­ of the solar cell. In order to implement ohmic contact with the absorber
vious results. It is revealed from figures that the performance of the material or the HTL, a relatively high metal work function is required.
device is not dependent on the defect density of the SnS/CZTS interface Until now, molybdenum has been used as back contact materials for
when the thickness of the absorber layer is > 3.5 µm. Fig. 9 shows the CZTS solar cells, like CIGS solar cells. It is assumed that it can provide
impact of defect density of the SnS/CZTS interface on the device per­ better cell performance. However, molybdenum is not the best choice
formance. The cell performance was studied by varying SnS/CZTS because of its non-ohmic contact characteristics with CZTS absorber.
interface defect density from 1014 to 1018 cm− 2. As it can be seen that the Moreover, there is no data on the thermal expansion coefficient of CZTS
interfacial defects have a detrimental effect on cell performances. An absorber, which is crucial for finding suitable materials for back metal
interesting point to be noted here that, compared with previous results, contacts. To investigate systemically the influence of the electron work
this device is very susceptible to SnS/CZTS interface defect density in function of the metal on the rear contact, its value was changed in the
degrading the cell performance. Generally, the previous threshold of range from 4.4 to 5.4 eV in the simulation, and results are illustrated in
1016 cm− 3 for CZTS defect density is reduced to 1015 cm− 2. The Fig. 10(a). As shown in the figure, by increasing the back-contact metal
maximum Voc value (1.1 V) is obtained in a very small area of Fig. 9 with work function from 4.4 eV to 5.4 eV, an amazing improvement in solar
an absorber thickness is > 3 µm and an SnS/CZTS interface defect cell parameters was found. All PV performance parameters show similar
density is < 1015 cm− 2 and it decreases outside this area. FF and Jsc trends, initially increasing with the increase of the metal work function,
showed similar character to Voc, but with higher thickness of the and above the critical metal work function (5.3 eV), these parameters
absorber, they are almost independent of the defect density of the indicate saturation behavior. This is attributed to decreases in the bar­
interface. Finally, the highest PCE > 31.5% is observed at thickness is > rier height for majority charge carriers at the back-contact interface with
4.5 µm and interface defect density < 1015 cm− 2 since PCE is a combi­ increasing metal work function, thereby improving the performance of
nation of the three PV parameters (Voc, Jsc, and FF). the device. These results suggest that back contact metal work function
above 5.3 eV is required for better performance of CZTS absorber based

73
M. Atowar Rahman Solar Energy 215 (2021) 64–76

VOC (Volt) JSC (mA/cm2)

1018 1.111 1018 32.34

1.104 31.97
Defect Density (cm-2)

Defect Density (cm-2)

17
10 1.097 1017 31.60

1.090 31.23

1016 1.082 10 16
30.86

1.075 30.49

10 15 1.068 10 15 30.12

1.061 29.75

1014 1.054 1014 29.38

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

FF (%) PCE (%)

1018 87.88 1018 31.58

87.79 31.01
Defect Density (cm-2)

Defect Density (cm-2)

1017 87.71 1017 30.44

87.62 29.86
16 16
10 87.54 10 29.29

87.45 28.72

1015 87.37 1015 28.15

87.28 27.57

1014 87.19 1014 27.00

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Thickness (µm) Thickness (µm)

Fig. 9. Variations of PV performance parameters with the defect density of SnS/CZTS interface and thickness of the absorber layer in CZTS solar cells.

solar cell, which is well agreed with the results reported in Ref. (Gupta 4. Conclusions
and Dixit, 2018). Metals, such as, Au, Ir, Ni, and W are therefore
compatible as back metal contact in order to get a high efficiency mark The PV performance of the CZTS-based solar cell with the novel
in this system. It is very important to study the temperature (T) behavior heterostructure (Ni/SnS/CZTS/TiO2/ITO/Al) was numerically analyzed
of solar cells, because in applications, they are usually exposed to a and optimized using the SCAPS-1D simulator by incorporating interfa­
temperature range of 288 K to 323 K (Sze and Ng, 1981), and even cial defects and single acceptor/donar like defects in each layer of the
higher temperatures in the desert during the summer period (Kahoul device. This study revealed that the incorporation of SnS and TiO2 to the
et al., 2014), in space and concentrator systems (Landis et al., 2004). In heterostructure as HTL and ETL, respectively, significantly reduced the
this section, PV performances of our proposed solar cells have been large Voc deficit and improved the electrical performance of the CZTS
achieved in various temperatures ranged between 280 and 380 K and based solar cells. Different parameters that influenced the PV perfor­
the results are presented in Fig. 10(b). In Fig. 10(b) there was a signif­ mance of the cell were explained. The thickness, carrier concentration
icant decrease in Voc, FF, and PCE, accompanied by a slight increase in and defect density of the CZTS-absorber, interfacial defects and back
Jsc. When the temperature increases, the band gap narrows (Morales- contact metal work function appeared to be a key factor for optimizing
Acevedo, 2013). Then, the recombination process of the electron-hole the PV performance of the cells. A maximum PCE of ≈ 31% was pre­
pair between the conduction band and the valence band is accelerated dicted for the CZTS solar cells at the CZTS-absorber layer carrier con­
and causes an increase in the dark current in the cell. Therefore, the centration of 5 × 1016 cm− 3 with a thickness of 4.0 µm and defect
observed decrease in Voc at high temperature is due to the increase in density < 1016 cm− 3. The interfacial defects have a detrimental effect on
reverse saturation current density. At the same time, the very small in­ the performance of CZTS solar cells, but can be overcome by increasing
crease in Jsc is due to the decrease of the band gap with increasing thickness of absorber layer. The work function of the back-metal contact
temperature. Consequently, the higher working temperature reduces the has a significant impact on the cell performance, and metals with high
PCE of the solar cell. Also, the increase of the working temperature of the work functions (such as Au, Ir, Ni, and W) could be used as back contact
cells affects the material conductivity by increasing the scattering of material to obtain the highest PCE from CZTS solar cell. We believe that
charge carriers with thermally activated phonons, which leads to a the results of this work will provide important enlightenment for the
decrease in the overall performance of the solar cells. feasible fabrication of cheap and efficient Cd-free CZTS solar cells.

74
M. Atowar Rahman Solar Energy 215 (2021) 64–76

32

PEC (%)

PEC (%)
28
30
21
28
14
26
7
86.4 87.6
FF (%)

FF (%)
79.2 86.4
72.0 85.2
64.8
32.021
JSC (mA/cm2)

VOC (Volt) JSC (mA/cm )

31.896

2
31.878 31.980

31.860 31.939

31.842 31.898
1.159
1.05
VOC (Volt)

1.098
0.84 1.037
0.63 0.976
0.42
4.6 4.8 5.0 5.2 5.4 280 300 320 340 360 380
Work Function (eV) Working Temperature (K)
(a) (b)
Fig. 10. Impact of (a) back contact metal work function and (b) working temperature in PV performance parameters for CZTS solar cells.

Declaration of Competing Interest Cui, H., Liu, X., Liu, F., Hao, X., Song, N., Yan, C., 2014. Boosting Cu2ZnSnS4 solar cells
efficiency by a thin ag intermediate layer between absorber and back contact. Appl.
Phys. Lett. 104, 041115.
The authors declare that they have no known competing financial Dhakal, R., Thapaliya, P., Li, Y., Jiang, M., Yan, X., 2011. TiO2 coated ZnO nanorod to
interests or personal relationships that could have appeared to influence enhance efficiency of hybrid bulk heterojunction solar cells, pp. 000708–000711.
the work reported in this paper. Dhakal, T.P., Peng, C., Tobias, R.R., Dasharathy, R., Westgate, C.R., 2014.
Characterization of a CZTS thin film solar cell grown by sputtering method. Sol.
Energy 100, 23–30.
Acknowledgements Ferdaous, M., Shahahmadi, S., Chelvanathan, P., Akhtaruzzaman, M., Alharbi, F.,
Sopian, K., Tiong, S., Amin, N., 2019. Elucidating the role of interfacial MoS2 layer in
Cu2ZnSnS4 thin film solar cells by numerical analysis. Sol. Energy 178, 162–172.
The authors gratefully acknowledge Dr. Marc Burgelman, University Glunz, S.W., Biro, D., Rein, S., Warta, W., 1999. Field-effect passivation of the SiO2-Si
of Gent, Belgium, for providing SCAPS 1-D simulation software. interface. J. Appl. Phys. 86, 683–691.
Green, M.A., Hishikawa, Y., Dunlop, E.D., Levi, D.H., Hohl-Ebinger, J., Ho-Baillie, A.W.
Y., 2018. Solar cell efficiency tables (version 52). Prog. Photovoltaics Res. Appl. 26,
References 427–436.
Gupta, G.K., Dixit, A., 2018. Theoretical studies of single and tandem Cu2ZnSn(S/Se)4
Abderrezek, M., Fathi, M., Djahli, F., 2018. Comparative study of temperature effect on junction solar cells for enhanced efficiency. Opt. Mater. 82, 11–20.
thin film solar cells. J. Nano- and Electronic Phys. 10, 02027–1–02027–6. Haddout, A., Raidou, A., Fahoume, M., 2019. A review on the numerical modeling of
Araki, H., Kubo, Y., Jimbo, K., Maw, W.S., Katagiri, H., Yamazaki, M., Oishi, K., CdS/CZTS-based solar cells. Appl. Phys. A 125, 124.
Takeuchi, A., 2009. Preparation of Cu2ZnSnS4 thin films by sulfurization of co- Hajzus, J.R., Biacchi, A.J., Le, S.T., Richter, C.A., Walker, A.R.H., Porter, L.M., 2018.
electroplated Cu-Zn-Sn precursors. Phys. Status Solidi (C) 6, 1266–1268. Contacts to solution-synthesized SnS nanoribbons: dependence of barrier height on
Baker, B., Johnson, B., Maire, G., 1971. Photoelectric work function measurements on metal work function. Nanoscale 10, 319–327.
nickel crystals and films. Surf. Sci. 24, 572–586. Hariskos, D., Spiering, S., Powalla, M., 2005. Buffer layers in Cu(In, Ga)Se2 solar cells
Benami, A., 2019. Effect of CZTS parameters on photovoltaic solar cell from numerical and modules. Thin Solid Films 480–481, 99–109.
simulation. Journal of Energy and Power Engineering 13, 32–36. Heriche, H., Rouabah, Z., Selmani, L., 2014. Thickness optimization of various layers of
Bhattacharya, S., John, S., 2019. Beyond 30% conversion efficiency in silicon solar cells: CZTS solar cell. J. New Technol. Mater. 4, 27–30.
A numerical demonstration. Sci. Rep. 9, 12482. Jackson, P., Hariskos, D., Lotter, E., Paetel, S., Wuerz, R., Menner, R., Wischmann, W.,
Bouachiba, Y., Bouabellou, A., Hanini, F., Kermiche, F., Taabouche, A., Powalla, M., 2011. New world record efficiency for Cu(In, Ga)Se2 thin-film solar
Boukheddaden, K., 2014. Structural and optical properties of TiO2 thin films grown cells beyond 20%. Prog. Photovolt. Res. Appl. 19, 894–897.
by sol-gel dip coating process. Mater. Sci.-Poland 32, 1–6. Jiang, M., Y, X., 2013. Cu2ZnSnS4 thin film solar cells: Present status and future
Burgelman, M., Nollet, P., Degrave, S., 2000. Modelling polycrystalline semiconductor prospects. In: Solar Cells - Research and Application Perspectives. InTech.
solar cells. Thin Solid Films 361–362, 527–532. Kabir, M.I., Shahahmadi, S.A., Lim, V., Zaidi, S., Sopian, K., Amin, N., 2012. Amorphous
Candelise, C., Winskel, M., Gross, R., 2012. Implications for CdTe and CIGS technologies silicon single-junction thin-film solar cell exceeding 10% efficiency by design
production costs of indium and tellurium scarcity. Prog. Photovolt. Res. Appl. 20, optimization. Int. J. Photoenergy 2012, 1–7.
816–831. Kahoul, N., Houabes, M., Sadok, M., 2014. Assessing the early degradation of
Chen, H.J., Fu, S.W., Wu, S.H., Wu, H.T., Shih, C.F., 2016. Comparative study of self- photovoltaic modules performance in the Saharan region. Energy Convers. Manage.
constituent buffer layers (CuS, SnS, ZnS) for synthesis Cu2ZnSnS4 thin films. Mater. 82, 320–326.
Lett. 169, 126–130. Karade, V., Lokhande, A., Babar, P., Gang, M.G., Suryawanshi, M., Patil, P., Kim, J.H.,
Chen, S., Yang, J.H., Gong, X.G., Walsh, A., Wei, S.H., 2010. Intrinsic point defects and 2019. Insights into kesterite’s back contact interface: A status review. Sol. Energy
complexes in the quaternary kesterite semiconductor Cu2ZnSnS4. Phys. Rev. B 81. Mater. Sol. Cells 200, 109911.
Cherouana, A., Labbani, R., 2017. Study of CZTS and CZTSSe solar cells for buffer layers Katagiri, H., Saitoh, K., Washio, T., Shinohara, H., Kurumadani, T., Miyajima, S., 2001.
selection. Appl. Surf. Sci. 424, 251–255. Development of thin film solar cell based on Cu2ZnSnS4 thin films. Sol. Energy
Chueh, C.C., Li, C.Z., Jen, A.K.Y., 2015. Recent progress and perspective in solution- Mater. Sol. Cells 65, 141–148.
processed interfacial materials for efficient and stable polymer and organometal Khattak, Y.H., Baig, F., Ullah, S., Marí, B., Beg, S., Ullah, H., 2018. Enhancement of the
perovskite solar cells. Energy Environ. Sci. 8, 1160–1189. conversion efficiency of thin film kesterite solar cell. J. Renew. Sustain. Energy 10,
033501.

75
M. Atowar Rahman Solar Energy 215 (2021) 64–76

Khusayfan, N.M., El-Nahass, M.M., 2013. Study of structure and electrooptical Schorr, S., Hoebler, H.J., Tovar, M., 2007. A neutron diffraction study of the stannite-
characteristics of indium tin oxide thin films. Adv. Condensed Matter Phys. 2013, kesterite solid solution series. Eur. J. Mineral. 19, 65–73.
1–8. Scragg, J.J., Kubart, T., Wätjen, J.T., Ericson, T., Linnarsson, M.K., Platzer-Björkman, C.,
Ki, W., Hillhouse, H.W., 2011. Earth-abundant element photovoltaics directly from 2013. Effects of back contact instability on Cu2ZnSnS4 devices and processes. Chem.
soluble precursors with high yield using a non-toxic solvent. Adv. Energy Mater. 1, Mater. 25, 3162–3171.
732–735. Shahahmadi, S.A., Yeganeh, B., Huda, N., Asim, N., Hafidz, M., Alam, M., AlOthman, Z.
Kowsar, A., Farhad, S.F.U., Rahaman, M., Islam, M.S., Imam, A.Y., Debnath, S.C., A., Sopian, K., Amin, N., 2015. Properties of a-SiGe thin films on glass by co-
Sultana, M., Hoque, M.A., Sharmin, A., Mahmood, Z.H., 2019. Progress in major sputtering for photovoltaic absorber application. J. Nanosci. Nanotechnol. 15, 1–6.
thin-film solar cells: Growth technologies, layer materials and efficiencies. Int. J. Sharmin, A., Bashar, M.S., Sultana, M., Mamun, S.M.M.A., 2020. Sputtered single-phase
Renew. Energy Res. 9, 2. kesterite Cu2ZnSnS4 (CZTS) thin film for photovoltaic applications: Post annealing
Kumar, A., Thakur, A.D., 2018. The simulation of CZTS solar cell for performance parameter optimization and property analysis. AIP Adv. 10, 015230.
improvement, AIP Conf. Proc. 1953. p. 050009. Shin, B., Gunawan, O., Zhu, Y., Bojarczuk, N.A., Chey, S.J., Guha, S., 2011. Thin film
Kumar, S.G., Rao, K.S.R.K., 2014. Physics and chemistry of CdTe/CdS thin film solar cell with 8.4% power conversion efficiency using an earth-abundant
heterojunction photovoltaic devices: fundamental and critical aspects. Energy Cu2ZnSnS4 absorber. Prog. Photovoltaics Res. Appl. 21, 72–76.
Environ. Sci. 7, 45–102. Shockley, W., Queisser, H.J., 1961. Detailed balance limit of efficiency of p-n junction
Landis, G., Rafaelle, R., Merritt, D., 2004. High temperature solar cell development. In: solar cells. J. Appl. Phys. 32, 510–519.
2nd International Energy Conversion Engineering Conference, 19th European Song, X., Ji, X., Li, M., Lin, W., Luo, X., Zhang, H., 2014. A review on development
Photovoltaic Science and Engineering Conference, Paris, France. prospect of CZTS based thin film solar cells. Int. J. Photoenergy 2014, 1–11.
Li, W., Chen, J., Yan, C., Hao, X., 2015. The effect of ZnS segregation on Zn-rich CZTS Sze, S.M., Ng, K.K., 1981. Physics of Semiconductor Devices, p. 264 (Chapter-14).
thin film solar cells. J. Alloy. Compd. 632, 178–184. Prentice-Hall series in solid state physical electronic, John Wiley and Sons, New
Lin, Q., Armin, A., Burn, P.L., Meredith, P., 2016a. Organohalide perovskites for solar York.
energy conversion. Acc. Chem. Res. 49, 545–553. Taguchi, M., Terakawa, A., Maruyama, E., Tanaka, M., 2005. Obtaining a higher Voc in
Lin, S., Li, X., Pan, H., Chen, H., Li, X., Li, Y., Zhou, J., 2016b. Numerical analysis of HIT cells. Prog. Photovoltaics Res. Appl. 13, 481–488.
InxGa1-xN/SnS and AlxGa1-xN/SnS heterojunctionsolar cells. Energy Convers. Wagner, S., Shay, J.L., Migliorato, P., Kasper, H.M., 1974. CuInSe2/CdS heterojunction
Manage. 119, 361–367. photovoltaic detectors. Appl. Phys. Lett. 25, 434–435.
Liu, F., Sun, K., Li, W., Yan, C., Cui, H., Jiang, L., Hao, X., Green, M.A., 2014. Enhancing Wanda, M.D., Ouédraogo, S., Tchoffo, F., Zougmoré, F., Ndjaka, J.M.B., 2016. Numerical
the Cu2ZnSnS4 solar cell efficiency by back contact modification: Inserting a thin investigations and analysis of Cu2ZnSnS4 based solar cells by SCAPS-1D. Int. J.
TiB2 intermediate layer at Cu2ZnSnS4/Mo interface. Appl. Phys. Lett. 104, 051105. Photoenergy 2016, 1–9.
Liu, Y., Sun, Y., Rockett, A., 2012. A new simulation software of solar cells-wxAMPS. Sol. Wang, K., Gunawan, O., Todorov, T., Shin, B., Chey, S.J., Bojarczuk, N.A., Mitzi, D.,
Energy Mater. Sol. Cells 98, 124–128. Guha, S., 2010. Thermally evaporated Cu2ZnSnS4 solar cells. Appl. Phys. Lett. 97,
Michaelson, H.B., 1977. The work function of the elements and its periodicity. J. Appl. 143508.
Phys. 48, 4729–4733. Wang, W., Winkler, M.T., Gunawan, O., Gokmen, T., Todorov, T.K., Zhu, Y., Mitzi, D.B.,
Mickelsen, R.A., Chen, W.S., 1981. Development of a 9.4% efficient thinfilm CuInSe2/ 2013. Device characteristics of CZTSSe thinfilm solar cells with 12.6% efficiency.
CdS solar cell. In: in: 15th photovoltaic specialists conference. Institute of Electrical Adv. Energy Mater. 4, 1301465.
and Electronics Engineers Inc, Kissimmee, FL, New York, pp. 800–804. Wangperawong, A., King, J., Herron, S., Tran, B., Pangan-Okimoto, K., Bent, S., 2011.
Minemoto, T., Hashimoto, Y., Satoh, T., Negami, T., Takakura, H., Hamakawa, Y., 2001a. Aqueous bath process for deposition of Cu2ZnSnS4 photovoltaic absorbers. Thin
Cu(In, Ga)Se2 solar cells with controlled conduction band offset of window/Cu(In, Solid Films 519, 2488–2492.
Ga)Se2 layers. J. Appl. Phys. 89, 8327–8330. Wolfe, C.M., Holonyak, N., Stillman, G.E., 1989. Calculation of the contact potential
Minemoto, T., Matsui, T., Takakura, H., Hamakawa, Y., Negami, T., Hashimoto, Y., (built-in voltage). Prentice-Hall series in solid state physical electronic, illustrated
Uenoyama, T., Kitagawa, M., 2001b. Theoretical analysis of the effect of conduction ed.,. Prentice Hall, Englewood Cliffs, N.J.
band offset of window/CIS layers on performance of CIS solar cells using device Wu, Y., Wadia, C., Ma, W., Sadtler, B., Alivisatos, A.P., 2008. Synthesis and photovoltaic
simulation. Sol. Energy Mater. Sol. Cells 67, 83–88. application of copper(I) sulfide nanocrystals. Nano Lett. 8, 2551–2555.
Mitzi, D.B., Gunawan, O., Todorov, T.K., Wang, K., Guha, S., 2011. The path towards a Xia, D., Zheng, Y., Lei, P., Zhao, X., 2013. Characterization of Cu2ZnSnS4 thin films
high-performance solution-processed kesterite solar cell. Sol. Energy Mater. Sol. prepared by solution-based deposition techniques. Physics Procedia 48, 228–234.
Cells 95, 1421–1436. Xiao, D., Li, X., Wang, D., Li, Q., Shen, K., Wang, D., 2017. CdTe thin film solar cell with
Morales-Acevedo, A. (Ed.), 2013. Solar Cells - Research and Application Perspectives. NiO as a back contact buffer layer. Sol. Energy Mater. Sol. Cells 169, 61–67.
InTech. Yan, C., Sun, K., Huang, J., Johnston, S., Liu, F., Veettil, B.P., Sun, K., Pu, A., Zhou, F.,
Mouchou, R., Jen, T., Laseinde, O., Ukoba, K., 2020. Numerical simulation and Stride, J.A., Green, M.A., Hao, X., 2017. Beyond 11% efficient sulfide kesterite
optimization of p-NiO/n-TiO2 solar cell system using SCAPS. Materials Today: Cu2ZnxCd1-xSnS4 solar cell: Effects of cadmium alloying. ACS Energy Lett. 2,
Proceedings. URL: https://doi.org/10.1016%2Fj.matpr.2020.04.880, doi:10.1016/j. 930–936.
matpr.2020.04.880. Yin, L., Cheng, G., Feng, Y., Li, Z., Yang, C., Xiao, X., 2015. Limitation factors for the
Nakamura, M., Yamaguchi, K., Kimoto, Y., Yasaki, Y., Kato, T., Sugimoto, H., 2019. Cd- performance of kesterite Cu2ZnSnS4 thin film solar cells studied by defect
free Cu(In, Ga)(Se, S)2 thin-film solar cell with record efficiency of 23.35%. IEEE J. characterization. RSCAdvances 5, 40369–40374.
Photovolt. 9, 1863–1867. Yoshikawa, K., Kawasaki, H., Yoshida, W., Irie, T., Konishi, K., Nakano, K., Uto, T.,
Partain, L.D., McLeod, P.S., Duisman, J.A., Peterson, T.M., Sawyer, D.E., Dean, C.S., Adachi, D., Kanematsu, M., Uzu, H., Yamamoto, K., 2017. Silicon heterojunction
1983. Degradation of a CuxS/CdS solar cell in hot, moist air and recovery in solar cell with interdigitated back contacts for a photoconversion efficiency over
hydrogen and air. J. Appl. Phys. 54, 6708–6720. 26%. Nat. Energy 2, 17032.
Patel, M., Ray, A., 2012. Enhancement of output performance of Cu2ZnSnS4 thin film Yue, G.H., Wang, L.S., Wang, X., Chen, Y.Z., Peng, D.L., 2009. Characterization and
solar cells—a numerical simulation approach and comparison to experiments. optical properties of the single crystalline SnS nanowire arrays. Nanoscale Res. Lett.
Physica B 407, 4391–4397. 4, 359–363.
Rappaport, P., 1959. The photovoltaic effect and its utilization. Sol. Energy 3, 8–18. Zhang, L., Jiang, C., Wu, C., Ju, H., Jiang, G., Liu, W., Zhu, C., Chen, T., 2018. V2O5 as
Ren, Y., Richter, M., Keller, J., Redinger, A., Unold, T., Donzel-Gargand, O., Scragg, J.J. hole transporting material for efficient all inorganic Sb2S3 solar cells. ACS Appl.
S., Björkman, C.P., 2017. Investigation of the SnS/Cu2ZnSnS4 interfaces in kesterite Mater. Interfaces 10, 27098–27105.
thin-film solar cells. ACS Energy Lett. 2, 976–981. Zhao, J., Wang, A., Green, M.A., 1999. 24.5% efficiency silicon PERT cells on MCZ
Rolland, A., Pedesseau, L., Kepenekian, M., Katan, C., Huang, Y., Wang, S., Cornet, C., substrates and 24.7% efficiency PERL cells on FZ substrates. Prog. Photovoltaics Res.
Durand, O., Even, J., 2017. Computational analysis of hybrid perovskite on silicon 2- Appl. 7, 471–474.
T tandem solar cells based on a si tunnel junction. Opt. Quant. Electron. 50, 21. Zhao, W., Zhou, W., Miao, X., 2012. Numerical simulation of CZTS thin film solar cell. In:
Saha, U., Alam, M.K., 2018. Boosting the efficiency of single junction kesterite solar cell 2012 7th IEEE International Conference on Nano/Micro Engineered and Molecular
using ag mixed Cu2ZnSnS4 active layer. RSC Adv. 8, 4905–4913. Systems (NEMS), Kyoto, Japan, IEEE. pp. 502–505.
Sai, H., Matsui, T., Koida, T., Matsubara, K., Kondo, M., Sugiyama, S., Katayama, H., Zhou, Y.L., Zhou, W.H., Du, Y.F., Li, M., Wu, S.X., 2011. Sphere-like kesterite Cu2ZnSnS4
Takeuchi, Y., Yoshida, I., 2015. Triple-junction thinfilm silicon solar cell fabricated nanoparticles synthesized by a facile solvothermal method. Mater. Lett. 65,
on periodically textured substrate with a stabilized efficiency of 13.6%. Appl. Phys. 1535–1537.
Lett. 106, 213902.

76

View publication stats