Appl. Phys.

A (2017) 123:758
DOI 10.1007/s00339-017-1371-x

CZTSSe solar cell efficiency improvement using a new band-gap

grading model in absorber layer
Shahram Mohammadnejad1 · Ali Baghban Parashkouh1

Received: 9 May 2017 / Accepted: 31 October 2017 / Published online: 10 November 2017
© Springer-Verlag GmbH Germany, part of Springer Nature 2017

Abstract Earth abundant copper–zinc–tin–chalcogenide conventional energy resources [1]. Solar radiation has both
is an attractive class of materials for the fabrication of high high power and redundancy as well as good accessibility
efficiency, low cost, and sustainable thin-film solar cells. A compared to the other accessible energy resources. By the
CZTSSe solar cell was modeled and the effects of absorber appropriate usage of photovoltaic effect, it is possible to con-
layer band-gap grading were investigated. Regarding the tun- vert this radiation into consumable electricity.
able bandgap of kesterite absorber layers which is between Solar cells are divided into three generations based on
0.95 eV for CZTSe and 1.5 eV for CZTS, several grad- their applied materials. Among thin-film technologies, cad-
ing models were simulated in SCAPS. First, using a set of mium telluride (CdTe) and ­Cu2InGaSe4 (CIGSe)-based solar
comprehensive absorption data for CZTS and CZTSe, the cells have reached to efficiencies around 22.1 and 22.6%,
basic bandgap was selected to be 1.15 eV due to experi- respectively [2, 3]. However, the existence of rare (In, Te
mental aspects, and then, five grading models namely back/ and Ga) and heavy (Cd) elements in these compositions limit
front linear, back/front exponential, and inside graded were their vast usage. Recently, it has been shown that ­Cu2ZnSnS4
explored. The investigation of simulation results showed (CZTS) has a significant photovoltaic performance which
that the recombination rate improvement in back and front attracted the attention of research laboratories.
regions along with current density enhancement is achiev- CZTS is a kesterite-based semiconductor which could be
able by these graded band-gap profiles. Finally, the modi- fabricated by solution processes, and thus, it is possible to
fied inside graded model was reached to a power conversion coat them on large substrates with an optimized material
efficiency of 15.6% which leaded to a considerable output utilization [4]. This kesterite absorber layer has an energy
performance. gap around 1.4–1.5 eV and being tunable is its most pivotal
property. As a result, this layer absorbs a higher amount of
solar radiation with an absorption coefficient more than ­104
1 Introduction ­cm−1 [5].
Adding Se to CZTS shifts its bandgap toward 1 eV. In
According to the growing shortage of fossil fuels and other words, replacing S with Se in absorber layer decreases
global warning, photovoltaic technology and materials have the bandgap. Hence, current density improves significantly
attracted a great attention within past decades. Furthermore, which enhances C ­ u2ZnSn(S,Se)4 (CZTSSe) solar cell effi-
the cost of solar power is being more competitive with the ciency in comparison with CZTS. In addition, there are two
models for grains distribution, small grains stand in bottom
and larger ones place in the top regions of absorber layer
* Ali Baghban Parashkouh [6]. The highest recorded efficiency of CZTSSe solar cell is
alibaghban1369@yahoo.com
about 12.6% which puts minds on producing low-cost elec-
Shahram Mohammadnejad tricity [7]. The different procedures of adding Se to CZTS
Shahramm@iust.ac.ir
and the adjustment of its bandgap are the main issues in the
1
Department of Electrical Engineering, Iran University absorber layer grading. Using high content of Se in CZTSSe
of Science and Technology, Narmak, Tehran, Iran solar cells gives rise to a better power conversion efficiency

13
Vol.:(0123456789)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

758 Page 2 of 9 S. Mohammadnejad, A. Baghban Parashkouh

(PCE) because of easier n- and p-type doping control and front sides of the absorber. For problem simplification, this
the carrier generation enhancement [8]. group assumed that S grading near front side keeps Ec flat
Kesterite-based polycrystalline solar cells have naturally a but spreads Ev [15]. Hence, there are two main reasons for ­Ev
high series resistance and a low shunt resistance. The series down shifting. First, Cu-depleted phase [because of Cu(In,
resistance in CZTSSe solar cells is 3–10 times greater than Ga)3Se5 usage] and second, by Ga/In or Se/S ratio which is
other high efficiency technologies. Mo(S,Se)x layer between intentional. These two methods are described below.
Mo and CZTSSe which forms a Schotcky type contact and
Zn(S,Se) secondary phases are two reasons that justify this 2.1 Intentional and unintentional grading
high resistance. If these secondary phases precipitate in
space charge region (SCR), shunt resistance will be affected From a practical view, grading could be divided in two
too. In addition, these solar cells’ performance is influenced groups: intentional and unintentional. As an instance in
by temperature variations and electron–hole life-time which CIGS solar cells, Cu-poor surfaces play an important role to
depends on defect types. Furthermore, CZTS solar cells have reach higher efficiencies. Unintentional groups are process-
many similar electrical and optical properties with CdTe related and happen by Cu depletion. This depletion grades
and CIGSe technologies which help researchers to design valence band in absorber/CdS interface and thus decreases
more reliable cells. Thus, CZTS solar cells could readily be recombination in this region. On the contrary, intentional
fabricated by thin-film technologies developed for CIGSe grading is caused by controlled variation in absorber’s com-
[6, 9, 10]. position (here, Ga/In and Se/S ratio) and could be tuned
during different steps of the coating process. It is shown that
graded structures help researchers to use both high and low
2 Grading models bandgaps to enhance Jsc and Voc at the same time.
Actually, separating current generation and carrier
CZTS is a quaternary semiconductor and the crystal struc- recombination mechanisms is the main reason of band-gap
ture of CZTSSe is similar to it, but changes in its bandgap engineering in this absorber layers. Totally, ­Voc and ­Jsc are,
appear due to the structural, compositional, electrical, and respectively, depended on the bandgap of SCR and the band-
optical properties [11]. This layers are designed to achieve gap minimum. Current–voltage thermal measurements have
an optimal bandgap which enhances the photon absorption shown that dominant recombination mechanism appears in
and carrier diffusion [12]. Actually, the main goal of compo- the SCR which makes Voc more dependent to bandgap [16].
sitional grading is to obtain a suitable model for bandgap to
get a higher Jsc while maintaining Voc. Yet, there is a 17.4% 2.2 Normal and reverse grading
PCE difference between Shockley–Queisser’s (SQ) limit in a
single junction solar cell and the practical value of CZTSSe Bandgap profiling includes several models like normal,
in the laboratories [7, 13]. reverse, and some mixed forms of these two. Normal and
Bandgap grading in this kind of solar cells is performed reverse grading and their differences in front and back sides
in the range of ­Cu2ZnSnSe4 (Eg = 0.95 eV) to ­Cu2ZnSnS4 of absorber layer are shown in Fig. 1. By getting closer
(Eg = 1.5 eV) and the tuning factor is Se/(Se + S) ratio. As to back contact in the normal form, absorber’s bandgap
a result of this grading, the absorption spectrum of CZTSSe increases and its electron affinity decreases almost linearly.
solar cells transfers linearly toward sections with lower ener- The resulted band-gap gradient makes a semi-electrical
gies [14]. Therefore, the Fermi level splits under illumina- field in solar cell. This extra field enhances electrons dif-
tion and it helps the solar cell to absorb a wider range of fusion length and helps carrier collection including minor-
photons. Similar improvements could be seen in the Jsc and ity carriers. In addition, Voc is increased by the reduction
Voc of CIGS solar cells depending on their single or multi- in recombination rate which is caused by lowered bandgap
ple band-gap profiles. Thus, grading experiences from both near back contact. In addition, the linearly increased band-
kinds of kesterites and chalcopyrites are explored in this gap decreases the position-dependent absorption coefficient
paper. and thus ­Jsc.
It is very important to know how Ec and Ev are affected On the other hand, bandgap decreases linearly near back
by the graded structure. Although, scientists mostly consider contact in reverse grading. Therefore, the recombination
their role in Ec, both bands might be affected using some decreases in this region and due to the higher bandgap in
specific materials. In a Cu(In1−x, ­Gax)(Se1−y, ­Sy)2 absorber SCR the Voc is improved. The reverse electrical field pre-
layer designed by Decock et al., bandgap is graded near the vents Jsc improvement by limiting carrier collection. In other
SCR and back contact region, respectively, with Se and Ga. words, there will not be an impressive progress in current
Adding Ga in the vicinity of the back contact increases band- density [17]. Different models like double and multi-grad-
gap by an offset on Ec, but both bands are affected using S in ing have special characteristics among absorber thickness.

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

CZTSSe solar cell efficiency improvement using a new band-gap grading model in absorber layer Page 3 of 9 758

Fig. 1  a Normal and b reverse grading models. ΔEg is Ec difference

between absorber’s beginning and end

Fig. 2  Band alignment between CZTSe, CZTS, and CdS showing

According to the simulation results, both normal and double offsets between their conduction and valence bands
grading have positive effects on the output performance, but
double grading has a better effect [18].

2.3 Alloy formation

Except zinc-blend compounds and their derivatives, super

lattice orientation plays an essential role in the offset for-
mation between semiconductors during bandgap tuning surface depletions are important factors for reaching high Voc
[12]. Among group IV elements, d level is considered as in buffer/absorber interface. This hole depletion is caused by
valence. Studies have shown that when CZTS and CZTSe valence band offset which increases barrier height near inter-
are placed near each other, Ec in CZTS side lies higher, and face and decreases recombination rate [20, 21]. This problem
its Ev lies lower than the same bands in CZTSe because is controllable with a suitable form of grading.
of p level difference in S and Se (p level in S is lower than Electron–hole generation from incident light and the con-
Se). Thus, electrons and holes will be localized in CZTSe duction band gradient are shown in Fig. 3. The gradient cre-
side. Considering p–d hybridization, a 0.15 eV offset could ated in conduction band makes an extra-electrical field which
be seen between CZTS and CZTSe valence bands. On the could be calculated according to Eq. 1:
other hand, there is a 0.35 eV offset between their conduc- dΔEg
tion bands [12]. Therefore, as the Se amount increases in 𝜀A = , (1)
the composition, conduction band minimum (CBM) down- dx
shifting is expected to be more than valance band maximum where ΔEg is the bandgap along x-direction. This extra-
(VBM) upshifting. This can be seen in Fig. 2 which illus- electrical field acts against built-in field and reduces the
trates the band alignment between CZTSe, CZTS, and CdS. total field in SCR. Therefore, it will have a negative effect
Experiences and existing samples have shown that the on solar cell performance. The increased bandgap in front
PCE of CZTSSe solar cells is better than CZTS, although sides decreases absorption which should be compensated in
the bandgap of CZTS absorber is closer to the optimum the inner parts where bandgap is smaller. On the other hand,
value. According to exact studies on CIGS Solar cells and such extra-electrical forces could be beneficial at absorber/
the existing similarities between CIGS and CZTSSe, it could Mo interface (back side) which has a high recombination
be found that n-type doping of CZTSe is relatively easier speed and rate.
than CZTS. The phase inverted behavior of Se compounds CBM lowering pushes minority electrons near back contact
on the p-type absorber helps to achieve a better carrier sepa- and this makes an acceptable conductivity for majority holes.
ration, and thus, the PCE is improved [12, 19]. Again, optical absorption in this region is lowered by bandgap
Holes heavy density is a limiting factor near surface. Actu- increasing, but it has a small effect on cell’s total performance
ally, enough holes near surface lowers Voc. In a study on Cu- (here short-circuit current) because of small carrier generation
poor surfaces done by Sites and Gloeckler, it was found that rate near back contact [18].

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

758 Page 4 of 9 S. Mohammadnejad, A. Baghban Parashkouh

Fig. 3  a Carrier generation with an incoming photon. b Ec gradient and resulting field in back region, similar field forms near SCR but in oppo-
site direction

3 The suggested solar cell structure and L A,B is the characteristic length [22]. Linear grading is
and absorption analysis the most common and simplest grading law. In addition,
the parabolic dependence could be shown by the following
Solar cell capacitance simulator (SCAPS) is a numerical equation:
solar cell simulation tool which needs an initial guess to Eg (y) = (1 − y)EgA + yEgB − b ⋅ y(1 − y), (6)
extract device characteristics. Defining layers with graded
parameters in the form of ­A1−yBy is an important ability where EgA and EgB are bandgaps of pure materials A and B,
of SCAPS. The compound materials’ properties varying respectively. In the case of CZTSSe alloys, pure A and pure
between A and B along layer thickness are defined as a func- B are respectively referred to CZTS and CZTSe. The bow-
tion of y [22]: ing factor of CZTSSe alloy is composition-independent and
has a value around 0.1. This small bowing factor shows an
P = f [y(x)]. (2) almost linear behavior. The small dimensions of Se and S
Changes in other material properties besides band-gap and their slight chemical differences are the reasons for such
grading is another significant issue which is observed in small bowing factor [24].
SCAPS [23]. Several grading laws such as uniform, linear, In a normal situation where all layers are uniform,
logarithmic, parabolic, exponential, beta function, etc are there are two driving forces for current. First, electro-
included in SCAPS. Equations of linear, parabolic, and static potential gradient which forms drift current and the
exponential laws are presented below: other one is carrier density gradient (∇p, ∇n)forming dif-
fusion current. In the case of a graded layer, electron–hole
PA (yB − y) + PB (y − yA )
Linear → P(y) = (3) continuity and Poisson’s equations should be corrected
yB − yA
with mobility gradient (∇𝜇n , ∇𝜇p ) and dielectric constant
gradient (∇𝜀) respectively. On the other hand, other driv-
PA (yB − y) + PB (y − yA ) (y − y)(y − yA )
Parabolic → P(y) = −b B ing forces like bandgap gradient (∇Eg ), electron affinity
yB − yA (yB − yA )2 gradient (∇𝜒), and effective density of states gradient on
(4) conductive and valence band (∇(log Nc ), ∇(log Nv )) should
yB −y y−yA
Sinh( ) Sinh( )
(5)
LA LB
Exponential → P(y) = P0 + (PA − P0 ) yB −yA + (PB − P0 ) yB −yA ,
Sinh( L ) Sinh( L )
A B

where b is bowing factor, yA and yB are composition val- be considered in the simulations to obtain more accurate
ues in A and B sides, respectively, P0 is background value, outputs [25].

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

CZTSSe solar cell efficiency improvement using a new band-gap grading model in absorber layer Page 5 of 9 758

Most CZTSSe solar cells have similar structure with higher quality and stability. Hence, simulations are based
common layers in different parts. Several back and front on a Mo/CZTSSe/n-CdS/ZnO/ZnO:Al structure, as shown
contacts were investigated in laboratories and simulation in Fig. 4.
studies, and some improvements in output performance The properties of each layer should be entered in
and material aspects have been reached. Kesterite-based SCAPS definition panel to start simulations. Absorption
polycrystalline solar cells have naturally a high series data of each layer are very important in this section. For
resistance and a low shunt resistance. Hence, a 4.25 and a some layers like CdS SCAPS, absorption information is
370 Ω cm2 were included in the simulations, respectively, used, but comprehensive data are entered for CZTS and
as series and shunt resistances [26]. The structure used CZTSe that are the main part of solar cell absorption. A
in this paper is similar to the accepted forms of kesterite summary of material properties is presented in Table 1
thin-film solar cells. Mo is the most suitable contact in [23, 26–28].
the back side of cell and CZTSSe absorber layer is grown The layer optical absorption at different wavelengths
above it. After CZTSSe, a 50 nm n-type CdS buffer layer is an essential parameter to obtain acceptable numerical
is deposited which is 50 times thinner than absorber. results. Thus, absorption coefficient is one of the crucial
Window layer and front contact are n-ZnO and ZnO:Al, issues in grading. The following algorithm explains how
respectively. Although using ZnO:Al in this structure the absorption coefficient is calculated for an ­A1−yBy com-
somewhat attenuates PCE, it forms a better contact with pound with absorption data of A at y = 0 (𝛼A (𝜆)) and B at
y = 1 (𝛼B (𝜆)) [25]:

a. Cut-off wavelength calculation for pure A and B (𝜆gA

and 𝜆gB)
b. Characteristic wavelengths, 𝜆0A and 𝜆0B, should be cal-
culated. Usually, 𝛼A (𝜆) has a maximum near UV that
should be chosen as characteristic wavelength. Other-
wise, an arbitrary value is possible.
c. Calculating the cut-off wavelength of the composition
with y, using Eq. 6 and the following equation:
1240 (nm eV)
𝜆g ≅
Eg (eV)
. (7)

d. The peak wavelength (𝜆0) should be calculated by an

interpolation between 𝜆0A and 𝜆0B (linear for instance).
e. Initial and secondary guesses for𝛼(𝜆) would be calcu-
Fig. 4  CZTSSe solar cell structure used in simulations with electri- lated by substituting 𝜆A and 𝜆B into 𝛼B and 𝛼B:
cal connections

Table 1  Materials properties entered in SCAPS definition panel

Material Properties CZTS CZTSe n-CdS n-ZnO ZnO:Al

Thickness (um) 0.5–2.5 0.5–2.5 0.05 0.2 0.8

Bandgap (eV) 1.5 0.95 2.4 3.3 3.3
Electron affinity (eV) 4.5 4.35 4.2 4.4 4.4
Dielectric permittivity (relative) 10 13.6 10 9 9
CB density of states (1/cm3) 2.200E+18 2.200E+18 2.200E+18 2.200E+18 2.200E+18
VB density of states (1/cm3) 1.800E+19 1.800E+19 1.800E+19 1.800E+19 1.800E+19
Electron Thermal Velocity (cm/s) 1.000E+7 1.000E+7 1.000E+7 1.000E+7 1.000E+7
Hole thermal velocity (cm/s) 1.000E+7 1.000E+7 1.000E+7 1.000E+7 1.000E+7
Electron mobility ­(cm2/Vs) 1.000E+2 1.000E+2 1.000E+2 1.000E+2 1.000E+2
Hole mobility ­(cm2/Vs) 2.500E+1 2.500E+1 2.500E+1 2.500E+1 2.500E+1
Donor density (1/cm3) 1.000E+0 1.000E+0 1.000E+17 1.000E+16 1.000E+18
Acceptor density (1/cm3) 1.000E+18 5.000E+16 0 1.000E+5 1.000E+1
Absorption coefficient Values of [29] Values of [29] SCAPS value Values of [30] Values of [30]

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

758 Page 6 of 9 S. Mohammadnejad, A. Baghban Parashkouh

limits are 𝜆ZnO and𝜆CIGS which correspond to their bandgaps

𝜆gA (𝜆 − 𝜆0 ) + 𝜆0A 𝜆gB (𝜆 − 𝜆0 ) + 𝜆0B [33].
𝜆A = ;𝜆B = . (8) Buffer layer forms a p–n junction with absorber and trans-
𝜆g − 𝜆0 𝜆g − 𝜆0
fers a considerable portion of incident light into it. This layer
f. With a weighted logarithmic average between this two plays an important role in solar cell structure such as junc-
guesses, we have: tion protection against chemical reactions, solar cells band,
and electrical tuning, and forming a high depletion region
log 𝛼 = (1 − y) log 𝛼A (𝜆A ) + y ⋅ log 𝛼B (𝜆B ). (9) with absorber. It should have the ability to transfer photo-
Figure 5 shows the absorption curves for CZTS, CZTSe, generated carriers to the outer circuit. Thus, it needs absorp-
and other applied layers in the simulated solar cell struc- tion losses, surface recombination, and electrical resistance
ture. Detailed absorption coefficients for each wavelength to be as low as possible. Therefore, buffer layer bandgap
are entered in software to obtain more accurate answers. It should be higher than absorber layer to achieve these prop-
is clearly obvious that CZTSe has a higher absorption rate erties [34, 35].
at most wavelengths which helps designers to reach better
output performances [31, 32].
Mo has good conformity with the requirements of a good 4 Results and discussion
contact in thin-film solar cells. It is inert during coating pro-
cesses, which forms big grains and a good ohmic contact. Two bandgaps were tested at the first step of simulations to
Thus, there are a few samples with other back contacts in the determine the absorber basic bandgap, 1.15 and 1.3 eV. The
existing CZTSSe and CIGS solar cells. By decreasing the former energy gap was reported in the champion solar cell
absorber thickness, back metal plays a more important role with 12.6% PCE (1.13 eV) [7] and latter was used in some
as light reflector. In thinner absorbers, light arrives with a simulation works [28, 36]. Uniform, linear, and exponential
higher intensity to absorber/BC interface. A part of this light bandgaps were explored in normal and reverse forms among
is absorbed at metal contacts; thus, the metallization type a 2.5𝜇m thickness of absorber. In the case of uniform grad-
in back surface has a direct effect on the cell performance. ing 1.3 eV gap has a slightly better PCE than 1.15 eV. Grad-
All absorbed photons in the back side of solar cell are ing models are 1.15–1.3 and 1.3–1.4 eV, and according to
considered as lost photons. Thus, lost current density in back simulation results shown in Table 2, it is obvious that former
region could be calculated by the following equation: grading mode has always a 0.7–1.0% higher PCE. Based on
the simulation results and experimental aspects of 1.15 eV
bandgap, it is used as the basic bandgap to find a suitable
𝜆CIGS

∫𝜆ZnO (10)
Jloss = q F1.5 (𝜆)ABC d𝜆,
grading model across absorbers thickness.
Several grading models were explored after applying
where ABC is the back contact optical absorption, F1.5 (𝜆) is 1.15 eV as CZTSSe basic bandgap. Achieving a structure to
the incident spectrum flux density at AM1.5G, and integral reduce the recombination rate in front and back regions and

Fig. 5  Absorption curves of solar cell layers. CZTS, CZTSe, CdS, ZnO, and ZnO:Al

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

CZTSSe solar cell efficiency improvement using a new band-gap grading model in absorber layer Page 7 of 9 758

Table 2  First step simulation results to compare 1.15 and 1.3 eV should be optimized to reach the mentioned goals. The sug-
bandgaps gested band-gap structures are back/front linear, back/front
Bandgap Eg (eV) Voc (V) Jsc (mA/cm2) FF (%) PCE (%) exponential, and inside graded in two different uniform and
profile graded parts thicknesses. Figure 6 shows these grading mod-
els on conduction band. To obtain suitable thickness for dif-
Uniform 1.15 0.70 28.58 63.61 12.91
ferent parts, 0.5 μm and 50 nm uniform parts are simulated
1.3 0.87 22.66 65.97 13
on both sides of absorber which has two graded parts inside.
Normal linear 1.15–1.30 0.63 43.07 56.77 15.35
Simulation results for models shown in Fig. 6 are listed in
1.30–1.40 0.62 39.75 58.33 14.36
Table 3. According to results, 0.5 μm uniform structure in
Reverse linear 1.15–1.30 0.79 27.58 67.89 14.94
back side of solar cell is appropriate for recombination low-
1.30–1.40 0.73 28.62 66.66 14.04
ering and PCE improvement, but the same uniform part in
Normal expo- 1.15–1.30 0.62 44.47 55.61 15.20
nential front side attenuates output performance, and it is better to
1.30–1.40 0.59 41.33 55.57 13.52
decrease its thickness to 50 nm. A shorter 1.3 eV part in
Reverse expo- 1.15–1.30 0.79 23.03 68.24 12.46
nential this region provides more chance for Jsc and other outputs
1.30–1.40 0.72 24.68 67.12 11.90
improvement. PCEs in inside graded model are 13.01, 15.04,
and 15.60%. The detailed outputs of the last and best sam-
ple are Voc = 0.65 V, Jsc=38.81 mA/cm2 and FF = 60.48%.
The 0.5 μm and 50 nm uniform parts reduce recombina-
tion rate near back contact and SCR, respectively, and the
other 1.7 μm graded part provides good absorption ability.
The energy parameters grading and J–V curve of the inside
graded model in its best form (with 15.60% power conver-
sion efficiency) are illustrated in Figs. 7 and 8 respectively.

5 Conclusion

Recombination rate controlling in the back and front regions

of CZTSSe absorber layer is an essential and challenging
Fig. 6  Grading models on conduction band of the absorber: (a) back topic for achieving better outputs. In addition, enhancing
linear, (b) front linear, (c) back exponential, (d) front exponential, and Jsc along absorbers thickness should be done using a mixed
(e) inside graded. Dashed lines are graded parts form of uniform and graded bandgaps. Applying a compre-
hensive absorption information of CZTS and CZTSe thin
also improve current density with absorption enhancement films at different wavelengths, we tried to optimize Voc and
was considered in this simulation step. In addition, absorber Jsc with recombination decreasing and absorption increas-
thickness and its division between uniform and graded parts ing at the same time. Inside graded band profile resulted an

Table 3  Results of grading Band-gap model Thickness (μm) (Tuni, Tgrd) Voc (V) Jsc (mA/cm2) FF (%) PCE (%)
models on absorber layer
Back linear Tuni = 2, Tgrd = 0.5 0.58 44.46 53.76 13.86
Tuni = 2.45, Tgrd = 0.05 0.52 40.97 51.97 11.06
Front linear Tuni = 2, Tgrd = 0.5 0.75 19.44 65.77 9.64
Tuni = 2.45, Tgrd = 0.05 0.72 31.75 58.57 13.51
Back exponential Tuni = 2, Tgrd = 0.5 0.58 43.57 54.01 13.56
Tuni = 2.45, Tgrd = 0.05 0.54 40.18 53.79 11.69
Front exponential Tuni = 2, Tgrd = 0.5 0.74 18.12 64.60 8.7
Tuni = 2.45, Tgrd = 0.05 0.72 31.89 58.60 13.58
Inside graded Tuni = 0.75, 0.75 0.68 35.78 62.18 15.04
Tgrd = 0.50, 0.50
Tuni = 0.75, 0.05 0.67 38.81 60.48 15.60
Tgrd = 0.85, 0.85
Tuni = 0.05, 0.75 0.66 31.34 63.48 13.10
Tgrd = 0.85, 0.85

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

758 Page 8 of 9 S. Mohammadnejad, A. Baghban Parashkouh

Fig. 7  Grading of energy parameters in 15.6% inside graded model

future prospects. J. Mater. Sci: Mater. Electron. 28, 2290–2306

(2017)
2. F. Solar, First solar achieves yet another cell conversion efficiency
world record., 201θ (2016)
3. P. Jackson, R. Wuerz, D. Hariskos, E. Lotter, W. Witte, M. Pow-
alla, Effects of heavy alkali elements in Cu (In, Ga) Se2 solar cells
with efficiencies up to 22.6%. Phys. Status Solidi (RRL)-Rapid
Res. Lett. 10, 583–586 (2016)
4. G. Wang, W. Zhao, Y. Cui, Q. Tian, S. Gao, L. Huang et al., Fab-
rication of a Cu2ZnSn (S, Se) 4 photovoltaic device by a low-
toxicity ethanol solution process. ACS Appl. Mater. Interfaces,
5, 10042–10047 (2013)
5. P. Darvishzadeh, H. Sohrabpoor, N.E. Gorji, Numerical device
simulation of carbon nanotube contacted CZTS solar cells. Opt.
Quantum Electron. 48, 480 (2016)
6. K. Woo, Y. Kim, W. Yang, K. Kim, I. Kim, Y. Oh et al., Band-
gap-graded Cu2ZnSn (­ S1-x, Sex) 4 solar cells fabricated by an
ethanol-based, particulate precursor ink route. Sci. Rep. 3, 3069
(2013)
7. W. Wang, M.T. Winkler, O. Gunawan, T. Gokmen, T.K. Todorov,
Fig. 8  J–V curve of inside graded model with 15.60% PCE. Its open- Y. Zhu et al., Device characteristics of CZTSSe thin-film solar
circuit voltage and short-circuit current are listed in Table 3 cells with 12.6% efficiency. Adv. Energy Mater., 4, (2014)
8. S.M. Yu, K.-S. Lim, A. Khalkar, J.-B. Yoo, Selenization of Cu2Z-
nSnS4 thin film using a Se metal-organic source for solar cell
applications. Appl. Phys. A 122, 767 (2016)
efficiency around 15.60% among tested grading models and 9. D.B. Mitzi, O. Gunawan, T.K. Todorov, K. Wang, S. Guha, The
showed the ability of grading in performance improvement. path towards a high-performance solution-processed kesterite
solar cell. Sol. Energy Mater. Sol. Cells 95, 1421–1436 (2011)
10. B. Shin, O. Gunawan, Y. Zhu, N.A. Bojarczuk, S.J. Chey, S. Guha,
Thin film solar cell with 8.4% power conversion efficiency using
an earth-abundant Cu2ZnSnS4 absorber. Prog. Photovoltaics Res.
References Appl. 21, 72–76 (2013)
11. S. Chen, X. Gong, A. Walsh, S.-H. Wei, Crystal and electronic
1. Z. Shi, D. Attygalle, A.H. Jayatissa, Kesterite-based next genera- band structure of Cu2ZnSnX4 (X = S and Se) photovoltaic
tion high performance thin film solar cell: current progress and

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

CZTSSe solar cell efficiency improvement using a new band-gap grading model in absorber layer Page 9 of 9 758

absorbers: first-principles insights. Appl. Phys. Lett. 94, 41903 25. M. Burgelman, J. Marlein, Analysis of graded band gap solar cells
(2009) with SCAPS, in Proceedings of the 23rd European Photovoltaic
12. S. Chen, A. Walsh, J.-H. Yang, X.-G. Gong, L. Sun, P.-X. Yang Solar Energy Conference, Valencia, 2008, pp. 2151–2155
et al., Compositional dependence of structural and electronic 26. M. Patel, A. Ray, Enhancement of output performance of Cu 2
properties of Cu 2 ZnSn (S, Se) 4 alloys for thin film solar cells. ZnSnS 4 thin film solar cells—a numerical simulation approach
Phys. Rev. B 83, 125201 (2011) and comparison to experiments. Physica B 407, 4391–4397 (2012)
13. S. Siebentritt, Why are kesterite solar cells not 20% efficient? Thin 27. O. Simya, A. Mahaboobbatcha, K. Balachander, Compositional
Solid Films 535, 1–4 (2013) grading of CZTSSe alloy using exponential and uniform grad-
14. H. Wei, Z. Ye, M. Li, Y. Su, Z. Yang, Y. Zhang, Tunable band ing laws in SCAPS-ID simulation. Superlattices Microstruct. 92,
gap Cu 2 ZnSnS 4x Se 4 (1 − x) nanocrystals: experimental and 285–293 (2016)
first-principles calculations. Cryst. Eng. Comm. 13, 2222–2226 28. O. Simya, A. Mahaboobbatcha, K. Balachander, A comparative
(2011) study on the performance of Kesterite based thin film solar cells
15. K. Decock, J. Lauwaert, M. Burgelman, Characterization of using SCAPS simulation program. Superlattices Microstruct. 82,
graded CIGS solar cells. Energy Procedia 2, 49–54 (2010) 248–261 (2015)
16. T. Dullweber, U. Rau, H. Schock, A new approach to high-effi- 29. K. Ito, Copper zinc tin sulfide-based thin film solar cells. Wiley
ciency solar cells by band gap grading in Cu (In, Ga) Se 2 chalco- (2014)
pyrite semiconductors. Sol. Energy Mater. Sol. Cells 67, 145–150 30. H. ElAnzeery, O. El Daif, M. Buffiere, S. Oueslati, K. Ben Mes-
(2001) saoud, D. Agten et al., Refractive index extraction and thickness
17. N.E. Gorji, M.D. Perez, U. Reggiani, L. Sandrolini, A new optimization of Cu2ZnSnSe4 thin film solar cells. Phys. Status
approach to valence and conduction band grading in CIGS thin Solidi (a), 212, 1984–1990 (2015)
film solar cells. Int. J. Eng. Technol. 4, 573 (2012) 31. M. León, S. Levcenko, R. Serna, G. Gurieva, A. Nateprov, J.
18. O. Lundberg, M. Edoff, L. Stolt, The effect of Ga-grading in CIGS Merino et al., Optical constants of Cu2ZnGeS4 bulk crystals
thin film solar cells. Thin Solid Films 480, 520–525 (2005) (2010)
19. T.K. Todorov, K.B. Reuter, D.B. Mitzi, High-efficiency solar cell 32. S. Choi, H. Zhao, C. Persson, C. Perkins, A. Donohue, B. To et al.,
with earth-abundant liquid-processed absorber. Adv. Mater. 22 Dielectric function spectra and critical-point energies of Cu2Zn-
(2010) SnSe4 from 0.5 to 9.0 eV. J. Appl. Phys. 111, 033506 (2012)
20. M. Gloeckler, J. Sites, Efficiency limitations for wide-band-gap 33. K. Orgassa, H.W. Schock, J. Werner, Alternative back contact
chalcopyrite solar cells. Thin Solid Films 480, 241–245 (2005) materials for thin film Cu (In, Ga) Se 2 solar cells. Thin Solid
21. M. Turcu, O. Pakma, U. Rau, Interdependence of absorber com- Films 431, 387–391 (2003)
position and recombination mechanism in Cu (In, Ga)(Se, S) 2 34. M.A. Contreras, M.J. Romero, B. To, F. Hasoon, R. Noufi, S.
heterojunction solar cells. Appl. Phys. Lett. 80, 2598–2600 (2002) Ward et al., Optimization of CBD CdS process in high-efficiency
22. A. Niemegeers, M. Burgelman, K. Decock, J. Verschraegen, “Ste- Cu (In, Ga) Se 2-based solar cells. Thin Solid Films 403, 204–211
faan Degrave SCAPS manual” ed: Version, 2014 (2002)
23. N. Khoshsirat, N.A. Md Yunus, M.N. Hamidon, S. Shafie, N. 35. S. Siebentritt, Alternative buffers for chalcopyrite solar cells. Solar
Amin, Analysis of absorber and buffer layer band gap grading on Energy. 77, 767–775 (2004)
CIGS thin film solar cell performance using SCAPS. Pertanika J. 36. M. Buffiere, G. Brammertz, S. Oueslati, H. El Anzeery, J. Bekaert,
Sci. Technol. 23, 241–250 (2015) K.B. Messaoud et al., Spectral current–voltage analysis of kester-
24. S.H. Wei, A. Zunger, Band offsets and optical bowings of chalco- ite solar cells. J. Phys. D: Appl. Phys. 47, 175101 (2014)
pyrites and Zn-based II-VI alloys. J. Appl. Phys. 78, 3846–3856
(1995)

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

onlineservice@springernature.com