BENZ4
BENZ4
BENZ4
A R T IC LE I N F O ABS TRA CT
Keywords: In this research work, SCAPS-1D (Solar Cell Capacitance Simulator in one Dimension) is used to
BSF simulate the CZTSSe (Cu2ZnSn(S,Se)4) solar cell with Al/ZnO:Al/ZnO(i)/CdS/CZTSSe/Mo
CZTSSe structure. The simulation results have been compared and validated with real experimental re-
SnS sults. After that, an effective receipt is proposed with the aim of improving the efficiency of the
CZTSe
CZTSSe solar cell, in which a BSF layer is inserted using various materials (SnS, CZTSSe and
Solar cell
Thin film
CZTSe). The obtained results show that the efficiencies of CZTSSe solar cells are increased from
12.3% to 15.7%, 15.3% and 15% by the insertion of SnS, CZTSSe and CZTSe materials as BSF
layers, respectively. This enhancement corresponds with a BSF layer thickness of 30 nm and
doping concentration of 1E18 cm−3. Next, an optimization of BSF layers thickness has been
conducted. The optimum value of thickness is considered at 40 nm with an enhancement ratio in
efficiency of 36.70%, 26.21% and 21.53% for SnS, CZTSSe and CZTSe, respectively. Better
performances have been noted for SnS material. The optimized CZTSSe solar cell with SnS as a
BSF layer achieves an efficiency of 16.95% with JSC = 36.34 mA/cm2, VOC = 0.69 V, and
FF = 67% under Standard Test Conditions (AM1.5 G and cell temperature of 25 °C).
1. Introduction
The Quaternary semiconductor Cu2ZnSn(S,Se)4 (CZTSSe)-based Thin-film solar cells have been considered as an environment-
friendly and cost-effective material for photovoltaic applications. It has recently gotten a lot of attention because of its promising
performances and earth-abundant elements [1,2]. CZTSSe is regarded as an interesting alternative to other thin-film photovoltaic
materials; it has an appropriate band gap energy (Eg) of 1.0–1.5 eV and a high absorption coefficient surpassing 104 cm−1 [3]. Over
the past few years, a remarkable progress in the power conversion efficiency (PCE) of CZTSSe solar cell has been accomplished, with
the best PCE record of 12.7% [4]. Nevertheless, this outcome shows low CZTSSe efficiencies and it should be enhanced [5]. Currently,
the dominant parameter responsible for the efficiency limitation of the kesterite CZTSSe solar cells is the low open-circuit voltage
(VOC) [6]. The causes of this problem have been a subject of intense discussions [7]. One of the effective solutions proposed to realize
a high efficiency CZTSSe solar cell with high VOC is by stacking a heavily doped thin layer in the back side of the cell between the
absorber layer and the back contact called a Back Surface Field (BSF) layer. It pushes the minority carrier to the front (depletion area)
and minimizes recombination in the back side of the solar cell [8]. It also improves electrical performances of the solar cells,
specifically the open circuit voltage (VOC). For that reason, the materials selected for the realization of this layer should have a mesh
⁎
Corresponding author.
E-mail addresses: mahfoud_cbi@yahoo.fr, halimbenz40@yahoo.fr (M. Abderrezek).
https://doi.org/10.1016/j.cjph.2019.11.020
Received 4 August 2019; Received in revised form 23 October 2019; Accepted 14 November 2019
Available online 30 November 2019
0577-9073/ © 2019 The Physical Society of the Republic of China (Taiwan). Published by Elsevier B.V. All rights reserved.
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
Fig. 1. Schematic diagram of the studied CZTSSe structure: (a) Experimental design [6], (b) The proposed design.
parameter close to the mesh parameter of the base layer with a high doping concentration of the order of 1018 cm−3 [9].
In this research paper, a comparative study on CZTSSe-based thin film solar cell performances has been carried out with various
back surface field (BSF) layers (SnS, CZTSSe and CZTSe). These materials are prepared in thin film with several features: the pho-
tovoltaic conversion efficiency, the optical gap, the absorption coefficient, low cost and non-toxic nature. The main objective is to
investigate the thickness effect of the BSF layers on the CZTSSe solar cell performances such as VOC and Jsc with the aim of producing
high efficiency CZTSSe cells with the appropriate BSF layer. First and foremost, the one-dimensional simulator (SCAPS-1D) is used to
simulate and numerically analyse the performances of CZTSSe solar cell without BSF and compare the obtained results with the
previous reported experimental results [6] to approve the developed simulation model. Second, the new proposed structure of
CZTSSe solar cell with various BSF layers (SnS, CZTSSe and CZTSe) is investigated in terms of thickness effect. Finally, an optimi-
zation of the proposed BSF layers is carried out to determine the optimum thickness of each BSF material.
The CZTSSe solar cell presented in Fig. 1(a) is formed from the following layers: ZnO:Al/ZnO-i/CdS/ CZTSSe/Mo. This structure is
adopted in simulation to compare it with the cell realized by Yang et al. [6]. It was prepared using different ratios of Se/SeS2. The best
achieved efficiency is 12.30%.
The CZTSSe solar cell structure has been modeled by using SCAPS-1D simulator. The program was developed at the Electronics
and Information Systems Department, Gents University, Belgium [6,8]. The SCAPS-1D simulation program is able to solve the basic
semiconductor equations, which are: the Poisson equation, relating the charge to the electrostatic potential φ and the continuity
equations for electrons and holes [10,11], which are denoted below with constitutive following equations:
∂ ⎛ ∂Ψ ⎞ ρdef
ε0 εr = −q (p − n + ND+ − N A− + )
∂x ⎝ ∂x ⎠ q (1)
∂Jn ∂n
− − Un + G =
∂x ∂t (2)
∂Jp ∂p
− − Up + G =
∂x ∂t (3)
The charge carrier transport is described by the drift and diffusion is expressed by the following equations:
μn n ∂EFn
Jn = −
q ∂x (4)
μp p ∂EFp
Jp = −
q ∂x (5)
where, Ψ is the electrostatic potential, ɛ0, ɛrare the permittivity of vacuum and semiconductor, n and p are the free carrier con-
centration, ND+and N A− are the density of ionized donors and acceptors , ρdef is the charge density of defects, q is the magnitude of the
232
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
Table 1
Parameter values for the different materials layers used for the simulation.
ZnO:Al ZnO-i CdS CZTSSe CZTSe SnS
charge of an electron, G is the generation rate, Jn and Jp are the electron hole current density, respectively, Un and Up are the
recombination rates for the electrons and holes, respectively, EFn, EFp are the quasi-Fermi level for the electrons and holes, respec-
tively [2]. The physical parameters used in this work for the different materials layers were all cited from experimental study of [6]
and literatures [8,11], which are summarized in Table 1.
The numerical simulation of the CZTSSe solar cell using the baseline parameters presented in Table 1 has achieved an efficiency of
12.4% with (VOC = 0.532 V, JSC = 33.75 mA/cm2, FF = 68.9%), Figs. 2 and 3 demonstrate the comparison results between ex-
perimental and simulation curves of J-V and quantum efficiency, respectively. The simulation results and the experimental data
reported by Yang et al. have been matched. Table 2 summarizes the obtained electrical parameters of simulation results in com-
parison with experimental results.
In this part, a highly doped thin layer of 0.03 µm thickness and 1E18 cm−3 doping concentration is added between the absorber
layer (CZTSSe) and back contact (Mo) as it plays the role of BSF layer. Various BSF materials (SnS, CZTSSe, CZTSe) are used with the
structure validated so far (using SCAPS-1D). The schematic diagram of the proposed structure is presented in Fig. 1(b).
Fig. 2 shows the comparison between simulated J-V curves of the proposed structures of the solar cells (with various BSF layers)
and the previous cases (solar cells without BSF layers). It is clear compared with same structure without BSF layer (simulated and
experimental) that the structures with various BSF layers show more efficient performances. Table 2 indicates the functional para-
meters obtained of these solar cells. It exhibits that both parameters VOC and JSC of the structures with BSF layers are increased
leading to boost the efficiency of the solar cell. A better performances of SnS material as a BSF layer are noticed with an efficiency of
15, 7% (VOC = 0.613 V, JSC = 36.25 mA/cm2, FF = 70.72%). The quantum efficiency as a function of wavelength for these solar cells
is exhibited in Fig. 3. It indicates a high performance achieved with CZTSSe solar cell with various BSF considering the same structure
without BSF. The enhancement is observed in the high wavelength region (650–1100 nm).
40
Current density (mA/cm2)
30
Simulation
Experimental
20 BSF(SnS)
BSF(CZTSSe)
BSF(CZTSe)
10
0
0.2 0.4 0.6
Voltage (V)
Fig. 2. J-V characteristics of CZTSSe solar cell without BSF (Experimental and simulation) and with various BSF types.
233
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
100
80
60
QE(%)
40 Simulation
Experimental
BSF(SnS)
20 BSF(CZTSSe)
BSF(CZTSe)
0
400 600 800 1000 1200
Wavelenght (nm)
Fig. 3. Quantum efficiency curves for CZTSSe solar cell without BSF (experimental and simulation) and with various BSF.
Table 2
Functional parameters of the CZTSSe solar cells without and with various BSF.
VOC (Volt) JSC (mA/cm2) FF (%) Ƞ (%)
Solar cells without BSF Experimental data 0.520 34.64 67.2 12.1
Simulation data 0,532 33,75 68,9 12,4
Solar cells with divers BSF layers CZTSSe 0.593 36.06 71.3 15.3
CZTSe 0.587 35.89 71.38 15.0
SnS 0.613 36.25 70.72 15.7
40
C u rre n t d e n s ity (m A /c m 2 )
30
10nm
20nm
30nm
20 40nm
50nm
60nm
10
0
0.0 0.2 0.4 0.6 0.8
Voltage (V)
Fig. 4. Effect of BSF (SnS) thickness on J-V characteristic curves of CZTSSe solar cells.
234
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
100
80
10nm
60 20nm
Q E (% )
30nm
40nm
40 50nm
60nm
20
0
400 600 800 1000 1200
Wavelenght (nm)
Fig. 5. Effect of BSF (SnS) thickness on the Quantum efficiency of CZTSSe solar cells.
40
Current density (mA/cm2)
30
10nm
20nm
30nm
20 40nm
50nm
60nm
10
0
0.0 0.2 0.4 0.6
Voltage (V)
Fig. 6. Effect of BSF (CZTSSe) thickness on J-V characteristic curves of CZTSSe solar cells.
100
80
10nm
60 20nm
QE(%)
30nm
40nm
40 50nm
60nm
20
0
400 600 800 1000 1200
Wavelenght (nm)
Fig. 7. Effect of BSF (CZTSSe) thickness on the Quantum efficiency of CZTSSe solar cells.
Fig. 7 presents the variation of QE in respect to BSF layer(CZTSSe) thickness. The QE increases after 10 nm (0.01 µm) thickness in
the long wavelength (780–1000 nm). No change is made by the rise in BSF (CZTSSe) thickness over 20 nm (0.02 µm) thickness; it
reaches the saturation thickness.
235
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
40
10
0
0.0 0.2 0.4 0.6
Voltage (V)
Fig. 8. Effect of BSF (CZTSe) thickness on J-V characteristic curves of CZTSSe solar cells.
100
80
10nm
60 20nm
QE(%)
30nm
40nm
40 50nm
60nm
20
0
400 600 800 1000 1200
Wavelenght (nm)
Fig. 9. Effect of BSF (CZTSe) thickness on the Quantum efficiency of CZTSSe solar cells.
Fig. 9 presents the variation of the CZTSSe solar cells QE in respect to BSF (CZTSe) thickness, QE is affected only in the long wave
length (780–1000 nm) by the BSF (CZTSSe) thickness rise.
In this section, the effect of various BSF thicknesses on the performances of the studied solar cell (Voc, Jsc and ƞ) is investigated.
Figs. 10–13 present the variation of solar cell performances as function of thickness for various BSF layers. For each BSF layer, a
simulation is carried out in which the thickness is varied to get the performances (η, Voc and Jsc) of the cell corresponding to that
thickness. Then, every parameter is plotted in respect to various BSF layers in the same curve as it is shown in Figs. 10–13. The
0.75
CZTSSe
SnS
0.70 CZTSe
0.65
V o c (V )
0.60
0.55
10 20 30 40 50 60
Thickness (nm)
Fig. 10. Variation of Open circuit voltage (Voc) as a function of various BSF thicknesses.
236
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
36.4
36.0
J s c (m A /c m 2 )
35.6
35.2 CZTSSe
SnS
CZTSe
34.8
10 20 30 40 50 60
Thickness (nm)
Fig. 11. Variation of Short Circuit Current (Jsc) as a function of various BSF thicknesses.
17
CZTSSe
SnS
CZTSe
16
Efficiency (%)
15
14
13
10 20 30 40 50 60
Thickness (nm)
Fig. 12. Variation of electrical Efficiency (ƞ) as a function of various BSF thicknesses.
72
71
70
69 CZTSSe
F F (% )
SnS
68 CZTSe
67
66
65
10 20 30 40 50 60
Thickness (nm)
Fig. 13. Variation of Fill Factor as a function of various BSF thicknesses.
thickness of BSF layers is varied between 10 and 60 nm and the doping concentration is fixed at 1E18 cm−1. A slight increase has
been noticed in Voc of CZTSSe and CZTSe materials between 10 and 40 nm thicknesses as it is presented in Fig. 10. After that, it
stagnates between 40 and 60 nm thickness around the value of 0.6 V. For SnS material, a surge in Voc between 10 and 40 nm is
remarked. A slight increase reaching the maximum value of Voc of 0.723 V at 60 nm thickness followed. The experimental results
affirmed that SnS improves significantly the open circuit voltage VOC. This positive contribution can be attributable to the passivation
effect at the interface between CZTSSe and Mo [13,14].
The Short Circuit Current (Jsc) variation is exhibited in Fig. 11. All the materials have the same pattern. A surge in Jsc between 10
and 20 nm thickness and a slight decrease between 20 and 30 nm has been observed. Then, a quasi-saturated point is reached after
237
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
30 nm thickness. The maximum Jsc values of various BSFs layers are 36.35 mA/cm2 at 40 nm, 36.14 mA/cm2 at 60 nm and
35.89 mA/cm2 at 30 nm for SnS, CZTSSe and CZTSe materials, respectively.
Fig. 12 presents the variation of the electrical efficiency (ƞ) as a function of various BSF layers thicknesses. It has been observed
that CZTSSe and CZTSe materials have the same variation form; an increase until 40 nm thickness followed by a saturation of the
efficiency after 40 nm thickness around the value of 15.5%. Whereas, for the SnS material a sharp rise in efficiency between 10 and
40 nm thickness is remarked. After 40 nm thickness, the efficiency stabilizes around the value of 17.08%.
The Fill Factor (FF) variation is shown in Fig. 13. CZTSSe and CZTSe materials have the same pattern. An increase of FF between
10 and 20 nm thickness and a quasi-saturated behavior after 20 nm thickness. Whereas, in the case of SnS a decrease trend is noticed
after 20 nm thickness. The maximum FF values of various BSFs layers are 71.39% and 71.165% at 20 nm and 71.363% at 60 nm for
SnS, CZTSSe and CZTSe materials, respectively.
Even though the SnS material has better performances as BSF layer. Especially the Voc with an increased trend as it shown in
Fig. 10, the case is different for the FF, a decreased trend is noticed from Fig. 13. This could be explained by the relation between FF
and Voc.
In theory the fill factor (FF) is defined as the ratio between the theoretical Power (Pth=Isc × Voc) and the maximal possible power
(Pmax=Impp× Vmpp) as described by the equation below [12]:
Pmax Vmpp × Jmpp
FF = =
Pth VOC × JSC (6)
According to Eq. (1), FF is dependent on the combined effect of Vmpp, Jmpp, Voc and Jsc. On one hand, the high Voc values reached
by the CZTSSe solar cell with SnS as BSF could be attributed to its high band gap (Eg = 1.3 eV) in comparison with other materials
(CZTSSe with Eg = 1.096 eV and CZTSe with Eg = 0.95 eV). With increasing band gap energy, the Voc is increasing too even though
the Jsc decreasing [15]. On the other hand, the increase in BSF layer may be attributes to the increased resistivity which results in
high series resistance thereby dropping the fill factor [16–18]. It is worth mentioning here that in practice the FF will be reduced by
the presence of parasitic resistive losses, it is essential to reduce Rs to approximately zero to achieve a high fill factors [19]. In our
situation, Voc of the CZTSSe solar cell with SnS material as BSF layer increases more than other materials leading to a large deficit in
FF.
4. Conclusion
In this research work, a comparative study of thickness effect for various back surface field (BSF) layers (SnS, CZTSSe and CZTSe)
has been conducted on the performances of CZTSSe Based Thin Film Solar cell such as the open-circuit voltage (Voc), the short-circuit
current density (Jsc) and the conversion efficiency (ŋ) using SCAPS simulation software. A maximum efficiency of 17.10% at 60 nm
thickness, 15.76% at 60 nm thickness and 15.07% at 40 nm thickness has been achieved using SnS, CZTSSe and CZTSe materials,
respectively, as BSF layers. Moreover, the rise of BSF layers thicknesses indicates an enhanced VOC of all CZTSSe solar cells. The
optimum value of thickness is considered at 40 nm for all materials. The Simulation results exhibit high performances for the SnS
material as a BSF layer in comparison with CZTSSe and CZTSe. The proposed design of solar cell with SnS material as back surface
field (BSF) layer has a better conversion efficiency compared with the other BSF layers (CZTSSe and CZTSe).
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Acknowledgment
The authors wish to thank Dr. M. Burgelman's group of Electronics and Information Systems (ELIS). University of Gent for the
SCAPS-1D program tool.
References
[1] R.B.V. Chalapathy, et al., Performance of CZTSSe thin film solar cells fabricated using a sulfo-selenization process: influence of the Cu composition, Sol. Energy
159 (2018) 260–269, https://doi.org/10.1016/j.solener.2017.10.085.
[2] J. Li, D. Wang, X. Li, Y. Zeng, Y. Zhang, Cation substitution in earth-abundant Kesterite photovoltaic materials, Adv. Sci. (2018) 1–22, https://doi.org/10.1002/
advs.201700744 5,4,1700744.
[3] M. Dey, et al., Modeling of Cu2ZnZnS4 solar cells with bismuth sulphide as a potential buffer layer, Proceedings of the 5th International Conference on
Informatics, Electronics and Vision (ICIEV), Dhaka: Bangladesh, 2016, pp. 1095–1099, , https://doi.org/10.1109/ICIEV.2016.7760168.
[4] A. Cherouana, R. Labbani, Study of CZTS and CZTSSe solar cells for buffer layers selection, Appl. Surf. Sci. 424 (2017) 251–255, https://doi.org/10.1016/j.
apsusc.2017.05.027.
[5] M. Dey, M. Dey, S. Alam, A.K.S. Gupta, N.K. Das, M.A. Matin2, N. Amin., Proceedings of the Design of Ultra-Thin CZTS Solar Cells with Indium Selenide as Buffer
Layer, (ECCE 2017), Cox's Bazar: Bangladesh, 2017, pp. 946–950, , https://doi.org/10.1109/ECACE.2017.7913040.
[6] K.J. Yang, et al., A band-gap-graded CZTSSe solar cell with 12.3% efficiency, J. Mater. Chem. A 4 (2016) 10151–10158.
[7] T. Gershon, et al., Unconventional Kesterites: the quest to reduce band tailing in CZTSSe, Curr. Opin. Green Sustain. Chem. J. 4 (2017) 29–36, https://doi.org/
10.1039/C6TA01558A.
[8] C. Yan, K. Sun, F. Liu, J. Huang, F. Zhou, X. Hao, Boost Voc of pure sulfide Kesterite solar cell via a double CZTS layer stacks, J. Sol. Energy Mater. Sol. Cells 160
(2017) 7–11, https://doi.org/10.1016/j.solmat.2016.09.027.
238
A.E. Benzetta, et al. Chinese Journal of Physics 63 (2020) 231–239
[9] M.K. Omrani, M. Minbashi, N. Memarian, D.H. Kim, Improve the performance of CZTSSe solar cells by applying a SnS BSF layer, Solid State Electron J. 141
(2017) 50–57, https://doi.org/10.1016/j.sse.2017.12.004 v.
[10] A.D. Adewoyin, M.A. Olopade, O.O. Oyebola, M.A. Chendo, Development of CZTGS/CZTS tandem thin film solar cell using SCAPS-1D, Opt. Int. J. Light Electron
Opt. 176 (2018) 132–142, https://doi.org/10.1016/j.ijleo.2018.09.033.
[11] M. Sijanur, R. Robin, M. Mansoor, M. Rasmi, Numerical modelling and analysis of ultra-thin film Cu(in,Ga)Se2 solar cell using SCAPS-1D, Proceedings of the 3rd
International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), Dhaka: Bangladesh, 2016, https://doi.org/10.1109/
CEEICT.2016.7873169 ID: 1477.
[12] A.D. Adewoyin, M.A. Olopade, M. Chendo, Enhancement of the conversion efficiency of Cu2ZnSnS4 thin film solar cell through the optimization of some device
parameters, Optik J. 133 (2017) 122–131, https://doi.org/10.1016/j.ijleo.2017.01.008.
[13] E.H. Ihalane, L. Atourki, H. Kirou, A. Ihlal, K. Bouabid, Numerical study of thin films CIGS bilayer solar cells using SCAPS, Mater. Today Proc. 3 (2016)
2570–2577, https://doi.org/10.1016/j.matpr.2016.04.004.
[14] P.D. Antunez, et al., Back contact engineering for increased performance in Kesterite solar cells, Adv. Energy Mater. 1602585 (2017) 1–5, https://doi.org/10.
1002/aenm.201602585.
[15] S. Rühle, Tabulated values of the Shockley–Queisser limit for single junction solar cells, Sol. Energy 130 (2016) 139–147, https://doi.org/10.1016/j.solener.
2016.02.015.
[16] A.C. Lokhande, et al., Fabrication of pulsed laser deposited Ge doped CZTSSe thin fi lm based solar cells : influence of selenization treatment, Sol. Energy Mater.
Sol. Cells 161 (2016) 355–367.
[17] S. Zhuk, A. Kushwaha, T.K.S. Wong, S. Masudy-panah, Critical review on sputter-deposited Cu2ZnSnS4 (CZTS) based thin film photovoltaic technology focusing
on device architecture and absorber quality on the solar cells performance, Sol. Energy Mater. Sol. Cells 171 (2017) 239–252, https://doi.org/10.1016/j.solmat.
2017.05.064.
[18] H. Cui, H., W. Li, X Liu, et al., Optimization of precursor deposition for evaporated Cu2ZnSnS4 solar cells, Appl. Phys. A 118 (2015) 893–899, https://doi.org/10.
1007/s00339-014-8806-4.
[19] O.K. Simya, A. Mahaboobbatcha, K. Balachander, A comparative study on the performance of Kesterite based thin film solar cells using Scaps simulation
program, Superlattices Microstruct. 82 (2015) 248–261, https://doi.org/10.1016/j.spmi.2015.02.020.
239