Accepted Manuscript

Title: Study of CZTS and CZTSSe solar cells for buffer layers
selection

Authors: Abdelbaki Cherouana, Rebiha Labbani

PII: S0169-4332(17)31329-6
DOI: http://dx.doi.org/doi:10.1016/j.apsusc.2017.05.027
Reference: APSUSC 35962

To appear in: APSUSC

Received date: 9-11-2016

Revised date: 29-4-2017
Accepted date: 3-5-2017

Please cite this article as: Abdelbaki Cherouana, Rebiha Labbani, Study of
CZTS and CZTSSe solar cells for buffer layers selection, Applied Surface
Sciencehttp://dx.doi.org/10.1016/j.apsusc.2017.05.027

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 Study of CZTS and CZTSSe Solar Cells for Buffer Layers Selection

Abdelbaki CHEROUANA, Rebiha LABBANI*

Laboratoire de Physique Mathématique et Subatomique, Département de physique, Université les

frères Mentouri-Constantine1, Route de Ain El bey, 25000 Constantine, Algérie

*Rebiha labbani: labbanire@gmail.com

Highlights

 Simulation of non toxic thin film solar cells is proposed.

 CdS/CZTS, CdS/CZTSSe, ZnS/CZTS and ZnS/CZTSSe were simulated using SCAPS program.
 Efficiency of the cell was affected by the buffer characteristics.
 Optimization of absorber layer parameters improved the performances of the cell.

Abstract

The output characteristics, of a thin film solar cell, are affected by several parameters related to the
hetero-structure that is taken into account. In this paper, nontoxic thin film solar cells were studied.
CdS/CZTS, CdS/CZTSSe, ZnS/CZTS and ZnS/CZTSSe hetero-junctions were numerically simulated
using the Solar Cell Capacitance Simulator (SCAPS). According to the simulation results, we found
that CdS buffer was suitable for the CZTSSe absorber and an efficiency of 9.02 % was obtained. The
ZnS buffer was selected for CZTS absorber and an efficiency of 8.02 % has been reached. By
optimizing the absorber film parameters (thickness and carrier density), we obtained conversion
efficiencies as high as 9.47 % and 10.00 % for CdS/CZTSSe and ZnS/CZTS based solar cells
respectively.

Keywords: CZTS; ; ; ; , CZTSSe, Solar cells, Simulation, SCAPS.

1. Introduction
The manufacturing of thin film photovoltaic devices is an area which has greatly increased in the
last decade [1]. Several materials (such as CdTe, CIS, CIGS) have been employed by researchers to
obtain low cost and high efficiency solar cells [2]. However, some of these materials are expensive or
toxic [3]. Consequently, the researchers were motivated to introduce new environmentally and friendly
materials [1]. Quaternary semiconductors with kestrite mineral structure are promising candidates for
achieving the requirement of low cost and eco-friendly thin film solar cells [4]. The Copper Zinc Tin
Chalcogenide is a quaternary semiconductor of group I-II-IV-VI. It was first elaborated in 1966 and
afterwards, its photovoltaic effect was confirmed in 1988 [5, 6]. Actually, this material is attracting
serious attention for photovoltaic applications.
 On the other hand, due to their excellent properties such as band gap and absorption
coefficient, CZTS and CZTSSe materials have been investigated experimentally and theoretically [4].
Indeed, for a band gap of 1.34 eV, the Shockley–Queisser limit (SQL) is reported to be 33.7 % for a
single junction solar cell [7]. Moreover, the theoretical maximum efficiency is found to be 32.4 %
(corresponding to Jsc=29.6 mA cm–2, Voc=1.21 V, FF=89.9%) for an optimal band gap of 1.48 eV [8,
9]. For the pure sulfide CZTS and mixed sulfo-selenide Cu2ZnSn(S,Se)4 (CZTSSe), the highest power
conversion efficiencies are reported to be 9.2% [10] and 12.6% [11] respectively. In the photovoltaic
device, a buffer layer is usually deposited between the absorber layer and the transparent conducting
oxide (TCO). This is used to provide a suitable band alignment for efficient separation of photo-
generated carriers [12]. It is also employed to avoid the interfacial strain and defects which are
associated with the lattice mismatch between the absorber and the TCO layer [13]. To achieve better
characteristics of solar cells, the selection of buffer materials is crucial especially for thin film
structures [14].

However, the realization of this step requires long times and a lot of money. For this reason,
one should perform simulations prior to experimental operations. Indeed, the use of Computer-based
simulations tools plays a critical role in the design, development and optimization of electronic and
physical devices. By these methods, we can reduce both time and costs. Among the several codes
used in the photovoltaic field, PC1D, AMPS, COMSOL, SCAPS are particularly used [4, 13]. By
these programs, we can predict the effect of various parameters and structures on the functionality of
the device. Quaternary semiconductors are complex in nature and involve hetero-junctions. Hence, the
need for numerical modeling in one dimension is relevant [4, 14].

In this study, numerical simulation of solar cells was performed. We focused on optimization
of CZTS and CZTSSe based solar cells with the suitable buffer layer. We employed the Solar Cell
Capacitance Simulator (SCAPS) [15]. By this program, we were able to study various hetero-
structures namely: CdS/CZTS, CdS/CZTSSe, ZnS/CZTS and ZnS/CZTSSe. Furthermore, we
demonstrated the absorber parameters (thickness and carrier density) impact on the performance of the
thin film device (i.e. solar cell).

2. Device structure
The physical device used in this study is represented in Figure 1. The considered structures of
solar cells (using SCAPS software) were composed of five layers. The superficial (i.e. first) layer was
n type ZnO (Zinc Oxide). It was considered as a window. The second layer was inserted in order to
select the suitable buffer for each absorber (which was the third layer). In this work, the buffer layer
material was supposed to be CdS or ZnS. For the absorber layer, the simulation was run either for
CZTS or CZTSSe (see section 3). For these structures, we included Molybdenum (Mo) as back
contact between the soda lime glass substrate and the absorber layer. The solar cells parameters were
selected from literatures (see table1). These data were estimated theoretically or provided by
experiments. For our simulation, the spectrum was set to the global Am 1.5 standard and the operation
temperature was maintained at 300K.

Table 1:

Figure 1:

3. Results and discussion

3.1. CZTS as an absorber layer

The software SCAPS has been used to predict CZTS based solar cells behaviors with respect to
the most common buffers layers (ZnS or CdS) which are largely used [4]. The parameters used in
simulation are summarized in table 1. In Figure2, we display the output characteristics obtained by
simulation for CdS/CZTS and ZnS/CZTS solar cells. From this figure, it is clear that ZnS/CZTS solar
cell exhibited better results. Indeed, this is revealed by the high obtained value of efficiency (Eta =
8.02 %) as compared to that result for CdS/CZTS hetero-junction (Eta = 7.72 %) (see Table 2). This is
probably due to the higher band gap of ZnS (3.3 eV) compared to the CdS (2.4 eV) [7, 13]. Our
assumption seems to be in agreement with literature in which it is reported that is possible to expand
the photo sensitivity range of the solar cell [16, 17].

Figure 2:

Table 2:

3.2. CZTSSe as an absorber layer

For a solar cell with CZTSSe absorber, the obtained results are displayed in Figure 3 and
summarized in Table 3. The solar cell with CdS as buffer layer has better output performance than
ZnS. Consequently, we have selected CdS layer as buffer for CZTSSe absorber. Our choice is in
agreement with the work of Wang et all [11], where it has been reported that high value of efficiency
(Eta =12.6%) has been reached for this structure.

Figure 3:

Table 3:
3.3. Effect of absorber parameters on ZnS/CZTS based solar cell

 Effect of CZTS thickness:

The absorber layer thickness is a crucial parameter on the performance of thin film solar cells. In this
simulation, the CZTS thickness was varied from 0.2 to 4 µm and the other parameters were kept
unchanged. The simulation results are represented in Figure 4. It has been found that the solar cell
parameters were decreasing with the decrease of the absorber layer thickness. The values of Vco and
FF were almost unchanged in the range {1,4 µm }. Whereas, Isc and Eta increased slowly over 1.5 µm
and 2 µm respectively. In this section, the optimal value of absorber layer was found to be 2.7 µm.

Figure 4:

 Effect of CZTS carriers concentration:

To study the absorber layer carrier concentration effect on the solar cell output parameters, the CZTS
thickness was fixed to the optimized value (which has been obtained above). Figure 5 exhibits the
electrical behaviors as a function of carriers concentration. It is clear that FF, Vco and Eta parameters
increased by the increase in the carrier density. The Isc current decreased with the increasing in the
carrier density but the values were almost unchanged up to 7×1016 cm-3 (which is the optimal value we
have chosen).

Figure 5:

3.4. Effect of absorber parameters on CdS/CZTSSe based solar cell

 Effect of CZTSSe thickness:

For CZTSSe based solar cell, the absorber layer thickness was varied from 1 µm to 5 µm. Figure 6
shows the effect of CZTSSe thickness on the output performances of the solar cell. We observe that,
almost all electric parameters were growing with thickness. The saturating value of fill factor (FF) was
above the thickness of 1.5 µm. The short circuit current density (Isc) and the conversion efficiency
(Eta) increased in the range {1, 2.5 µm } then they became almost unchanged. This is probably caused
by the recombination in the device. In this section, the absorber thickness value of 2.6 µm was
sufficient to give optimal output of the solar cell.
 Figure 6:

 Effect of CZTSSe carriers concentration:

In this simulation, we have studied the effect of CZTSSe carriers concentration (noted Nacztsse) on the
performance of CdS/CZTSSe based solar cell. The open circuit voltage (Voc) and the fill factor (FF)
were decreasing with thickness decreasing (Figure 7). FF patterns exhibited a saturate value above
3×1014 cm-3. The short circuit current was decreasing for high values of Nacztsse and the conversion
efficiency increased rapidly with carrier concentration until 7×1014 cm-3. From these different output
behaviors, we selected Nacztsse equal to 5×1014 cm-3 as optimal value for CZTSSe based solar cell.

Figure 7:

3.5. Study of thin film solar cells with optimal values

Using the optimized parameters obtained in previous sections, we get better performances in
the solar cells outputs. The obtained results from ZnS/CZTS based solar cells were displayed in Figure
8 and summarized in Table 4. We notice an improvement in conversion efficiency from 8.02 % to
10.00 %. For CdS/CZTSSe based solar cells, the results are represented in Fig.9 and illustrated in
Table 5. For this case, we achieve high conversion efficiency up to 9.47 %.

Table 4:

Table 5:

Figure 8:

Figure 9:

In table 6, the optimal results obtained in this work are summarized. Experimental and theoretical
values (provided from literature) are also reported for comparison. According to this table, it is clear
that our results are comparable with the experimental values of literature (see raw 3-7). From that, we
may conclude that Solar Cell Capacitance Simulator is a program able to predict solar cells behavior
similar to real cases. We note that many challenges are present while preparing these hetero-junctions
experimentally. For instance, it is very difficult to grow a layer in such a way that the secondary
materials are easily removable [23]. Moreover, many physico-chemical processes may occur and
change electrical properties of CZTS solar cells [24].
 Table 6:

4. Conclusion
In this paper, a comparative study was performed by simulation using the SCAPS software. Our
goal was to select the better buffer for CZTS and CZTSSe absorber layers and to predict a more
appropriate hetero-junction. In the first time, our simulation demonstrated that ZnS and CdS buffers
were suitable for CZTS and CZTSSe absorber layers respectively. Afterwards, we studied the effect of
the thickness and the carrier concentration of the absorber layer for each selected hetero-junction. By
this investigation, we were able to optimize these two parameters so that we achieved the best voltage
in open circuit (Vco), density current in short circuit (Isc), fill factor (FF) and efficiency (Eta). The
optimized values for the absorber thickness were 2.7 µm and 2.6 µm for CZTS and CZTSSe absorbers
layers respectively. Concerning the carriers concentration, the best values were 7×1017 cm-3 and 5×1014
cm-3 for the CZTS and CZTSSe respectively. By using these optimized values, we were able to
improve the efficiency for each hetero-junction. Indeed, for CdS/CZTSSe based solar cell, the
efficiency increased from 9.02 % to 9.47 %. For the case of ZnS/CZTS, the efficiency was improved
from 8.02 % to 10.00 %. The obtained results were satisfactory and may be used experimentally to
prepare real ZnS/CZTS and CdS/CZTSSe based solar cells.

Acknowledgments

Authors acknowledge Dr. Marc Bargeman, from University of Gent, for providing SCAPS simulation
software.

References

[1] Gunaicha, P. P., et al. "Structural, energetic and elastic properties of Cu 2 ZnSn (SxSe 1− x) 4 (x=
1, 0.75, 0.5, 0.25, 0) alloys from first-principles computations." Solar Energy 102 (2014): 276-281.

[2] O. Simya, A. Mahaboobbatcha, K. Balachander, A comparative study on the performance of

kesterite based thin _lm solar cells using scaps simulation program, Superlattices and Microstructures
82 (2015) 248-261.

[3] Altamura, Giovanni. Development of CZTSSe thin films based solar cells. Diss. Université Joseph-
Fourier-Grenoble I, 2014. URL https://tel.archives-ouvertes.fr/tel-01060095/document.
[4] Patel, Malkeshkumar, and Abhijit Ray. "Enhancement of output performance of Cu 2 ZnSnS 4 thin
film solar cells—a numerical simulation approach and comparison to experiments." Physica B:
Condensed Matter 407.21 (2012): 4391-4397.

[5] Nitsche, R., D. F. Sargent, and P. Wild. "Crystal growth of quaternary 122464 chalcogenides by
iodine vapor transport." Journal of Crystal Growth 1.1 (1967): 52-53.

[6] Ito, Kentaro, and Tatsuo Nakazawa. "Electrical and optical properties of stannite-type quaternary
semiconductor thin films." Japanese Journal of Applied Physics 27.11R (1988): 2094.

[7] Huang, Tang Jiao, et al. "CZTS‐based materials and interfaces and their effects on the performance
of thin film solar cells." physica status solidi (RRL)-Rapid Research Letters 8.09 (2014): 735-762.

[8] Tao, Jiahua, et al. "7.1% efficient co-electroplated Cu 2 ZnSnS 4 thin film solar cells with sputtered
CdS buffer layers." Green Chemistry 18.2 (2016): 550-557.

[9] Ki, Wooseok, and Hugh W. Hillhouse. "Earth Abundant Element Photovoltaics Directly from
Soluble Precursors with High Yield Using a Non‐Toxic Solvent." Advanced Energy Materials 1.5
(2011): 732-735.

[10] Sun, Kaiwen, et al. "Over 9% efficient kesterite Cu2ZnSnS4 solar cell fabricated by using Zn1–
xCdxS buffer layer." Advanced Energy Materials 6.12 (2016).

[11] Wang, Wei, et al. "Device characteristics of CZTSSe thin‐film solar cells with 12.6%
efficiency." Advanced Energy Materials 4.7 (2014).

[12] Kodigala, Subba Ramaiah. Thin Film Solar Cells From Earth Abundant Materials: Growth and
Characterization of Cu2 (ZnSn)(SSe) 4 Thin Films and Their Solar Cells. Newnes, 2013.

[13] Nguyen, Mai, et al. "ZnS buffer layer for Cu 2 ZnSn (SSe) 4 monograin layer solar cell." Solar
Energy 111 (2015): 344-349.

[14] Haque, F., et al. "Prospects of Zinc Sulphide as an alternative buffer layer for CZTS solar cells
from numerical analysis." Electrical and Computer Engineering (ICECE), 2014 International
Conference on. IEEE, 2014.

[15] M. Burgelman, K. Decock, A. Niemegeers, J. Verschraegen, S. Degrave, Scaps manual.

[16] Opanasyuk, A. S., et al. "Properties of the window layers for the CZTSe and CZTS based solar
cells." Journal of Nano-and Electronic Physics 4.1 (2012): 1024-1.

[17] K. Ito, Copper zinc tin sulfide-based thin-film solar cells, edited by Kentaro Ito, Wiley, ISBN
978-1-118-43787-2 (2015).
[18] Benmir, Abdelkader, and Mohamed Salah Aida."Simulation of a thin film solar cell based on
copper zinc tin sulfo-selenide Cu2ZnSn(S,Se)4."Superlattices and Microstructures 91(2016): 70-77.

[19] Cazzaniga, Andrea, et al. "Ultra-thin Cu 2 ZnSnS 4 solar cell by pulsed laser deposition." Solar
Energy Materials and Solar Cells 166 (2017): 91-99.

[20] Park, Ju Young, et al. "Fabrication of earth abundant Cu 2 ZnSnSSe 4 (CZTSSe) thin film solar
cells with cadmium free zinc sulfide (ZnS) buffer layers." Journal of Alloys and Compounds 695
(2017): 2652-2660.

[21] Correa, John M., et al. "Fabrication of solar cells based on Cu2ZnSnS4 prepared from Cu2SnS3
synthesized using a novel chemical procedure." EPJ Photovoltaics 7 (2016): 70305.

[22] Siebentritt, Susanne. "Why are kesterite solar cells not 20% efficient?" Thin Solid Films 535
(2013): 1-4.

[23] Dale, Phillip J., et al. "A review of the challenges facing kesterite based thin film solar
cells." Photovoltaic Specialists Conference (PVSC), 2009 34th IEEE. IEEE, 2009.

[24] Ren, Y. (2017). Annealing of Cu2ZnSn (S, Se) 4 Thin Films: A Study of Secondary Compounds
and Their Effects on Solar Cells (Doctoral dissertation, Acta Universitatis Upsaliensis).

Figure Caption
Figure 1

Figure 2
Figure 3

Figure 4
Figure 5

Figure 6
Figure 7

Figure 8
Figure 9
 Parameters ZnO CdS ZnS CZTS CZTSSe References
thickness (µm) 0.200 0.05 0.050 2.0 2.0 [4, 13, 18]
Band gap (eV) 3.300 2.40 3.300 1.5 1.3 [4]
electron affinity (eV) 4.400 4.2 4.400 4.4 4.2 [18]
dielectric permittivity (relative) 9.00 10.0 10.0 10.0 13.6 [4, 13, 18]
(Conduction Band) effective 2.2E+18 2.200E+18 2.200E+18 2.200E+18 2.2E+18 [4,13]
density of states (1/cm3)
(Valance Band) effective 1.8E+19 1.80E+19 1.80E+19 1.800E+19 1.8E+19 [4,13]
density of states (1/cm3)
electron thermal velocity (cm/s) 1.000E+7 1.000E+7 1.000E+7 1.000E+7 1.0E+7 [4]
hole thermal velocity (cm/s) 191.000E+7 1.0E+7 1.0E+7 1.0E+7 1.0E+7 [4]
electron mobility (cm²/Vs) 1.000E+2 1.000E+2 1.000E+2 1.000E+2 1.0E+2 [4]
hole mobility (cm²/Vs) 2.500E+1 2.500E+1 2.500E+1 2.500E+1 2.5E+1 [13,18]
shallow uniform donor 1.000E+18 1.10E+18 9.000E+17 1.000E+1 1.0E+1 [4]
density ND (1/cm3)
shallow uniform acceptor 1.000E+1 1.00E+0 1.000E+0 1.000E+16 1.0E+15 [18]
density NA (1/cm3)
absorption coefficient cm-1 provided by provided by provided by 5E4 2E4 [4,17]
SCAPS SCAPS SCAPS

Table 1: The solar cells parameters used in the simulation.

Vco (Volt) Isc (mA/cm2) Eta % FF %

ZnS/CZTS 0.55 23.69 8.02 61.21
CdS/CZTS 0.55 22.11 7.72 63.21

Table 2: ZnS/CZTS and CdS/CZTS output characteristics: voltage in open circuit (Vco),
density of short circuit current (Isc), efficiency (Eta) and fill factor (FF).

Vco (Volt) Isc (mA/cm2) Eta % FF %

CdS/CZTSSe 0.63 24.64 9.02 57.62
ZnS/CZTSSe 0.64 24.11 6.30 40.83

Table 3: ZnS/CZTSSe and CdS/CZTSSe output characteristics: voltage in open circuit (Vco),
density of short circuit current (Isc), efficiency (Eta) and fill factor (FF).

Vco (Volt) Isc (mA/cm2) Eta % FF %

ZnS/CZTS before 0.55 23.69 8.02 61.21
optimizing
ZnS/CZTS after 0.64 23.96 10.00 65.20
optimizing

Table 4: CZTS cell output before and after optimizing.

 Vco (Volt) Isc (mA/cm2) Eta % FF %

CdS/CZTSSe before 0.63 24.64 9.02 57.62

optimizing
CdS/CZTSSe after 0.61 26.95 9.47 57.20
optimizing

Table 5: CZTSSe cell output after and before optimizing.

device Eta (%) Isc (mA/cm2) Vco(volt) FF (%) references

CdS/CZTSSe 9.47 26.95 0.61 57.20 This work
ZnS/CZTS 10.00 23.96 0.64 65.20 This work
CdS/CZTSSe 12.6 35.2 0.513 69.8 [11]
CdS/CZTS 5.2, 7.1 17.6, 21.8 0.616, 47.9, [19], [8]
0.614 55.5
ZnS/CZTSSe 3.8 23.5 0.309 54 [20]
ZnS/CZTS 5.47 15.9 0.641 53.7 [21]
Zn0.35Cd0.65CZTS 9.2 19.5 0.747 63.2 [10]
CZTS SQ limit (Eg=1.48 32.4 29.6 1.21 89.9 [8, 9]
ev)
CZTS SQ limit (Eg=1.5 ev) 31.5 29 1.21 91 [22]
CZTS SQ limit With optical 27 25 1.21 91 [7, 22]
losses(Eg=1.5 ev)
CZTSSe SQ limit (Eg=1.15 32.8 42 0.887 89 [7, 22]
ev)
CZTSSe SQ limit With 28 36 0.887 89 [7, 22]
optical losses (Eg=1.15ev)

Table 6: Values of conversion efficiencies (Eta), currents in short circuit (Isc), voltages in
open circuit (Vco) and Fill Factors (FF).