Effect of composition gradient in Cu(In,Al)Se2 solar cells
Takao Hayashi1, Takashi Minemoto1, Guillaume Zoppi2, Ian Forbes2, Kiyoteru Tanaka1, Satoshi
Yamada1, Tsutomu Araki1, Hideyuki Takakura1
1
College of Science and Engineering, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu,
Shiga 525-8577, Japan
Corresponding author: Takashi Minemoto
Tel/Fax +81-77-561-3065, E-mail: minemoto@se.ritsumei.ac.jp
2
Northumbria Photovoltaics Applications Centre, Northumbria University, Ellison Building,
Newcastle upon Tyne, NE1 8ST, United Kingdom
Abstract
Cu(In,Al)Se2 (CIAS) thin-films were prepared by a three-stage evaporation process. In
this experiment, the composition ratio of Cu/(In+Al) at the end of the second stage (Cu/III2nd)
was changed from 1.1~1.7. The CIAS films showed an Al distribution with a V-shape profile.
The valley depth of the V-shape from the surface increased with increasing the Cu/III2nd ratio.
The valleys of the V-shape for the films with the Cu/III2nd ratio of 1.1~1.7 were located at
approximately 0.3~1.0 m from the film surface, respectively. The rms surface roughness
1
increased from 40 nm for Cu/III2nd=1.1 to 90nm at Cu/III2nd=1.3 and then saturated for greater
Cu/III2nd ratios. Solar cells with the Al/ITO/ZnO/CdS/CIAS/Mo/soda-lime glass structure were
fabricated. The fill factor was seen to decrease while the product of short-circuit current and
open-circuit voltage remained constant. The reverse saturation current increased when the
Cu/III2nd ratio is greater than 1.3 which is a behavior as the surface roughness. Cu/III2nd ratios
greater than 1.3 lead to the distant position of V-shape from the surface and the increase in
surface roughness.
Keywords: Cu(In,Al)Se2 , band gap profile, three-stage process.
2
1. Introduction
CuInSe2 (CIS) and related chalcopyrite-based solar cells have emerged as leading
candidates for low-cost and high-efficiency thin-film solar cells. The energy band gap (Eg) of
the chalcopyrite absorber can be controlled to match the solar spectrum for higher efficiency by
alloying the group III or VI elements (with Ga or Al and S, respectively). The Eg of
Cu(In,Ga)Se2 (CIGS) can be varied from 1.0 eV (CIS) to 1.7 eV (CuGaSe2) by Ga addition. An
increase in Eg of the absorber layer results in a tradeoff of higher open-circuit voltage (Voc) and
lower short-circuit current (Jsc) of the solar cell. The Eg of CIS-based absorber layer can be
adjusted to the theoretically estimated optimum value of 1.4 eV. However, maximum
efficiencies for CIGS solar cells were achieved for Eg = 1.1~1.2 eV by a three-stage evaporation
process [1]. One of the reasons for the performance degradation of the CIGS solar cells with Eg
> 1.3 eV is that a higher-Ga addition induces a degradation of electronic properties of the CIGS
layer leading to losses in Voc and fill factor (FF) [2].
Cu(In,Al)Se2 (CIAS) has been considered as an alternative to the CIGS because it
requires smaller relative alloy concentration than Ga alloy to obtain a comparable Eg. The Eg of
CIAS can be controlled from 1.0 eV (CIS) to 2.7 eV (CuAlSe2) by Al addition [2]. Some group
reported the preparation of CIAS thin-films by multisource elemental co-evaporation and
achieved the solar cell efficiencies of 16.9% [3]. In their report, the crystal grains composing the
3
CIAS film had a columnar shape which is favorable for photo-current collection. However the
grain sizes were relatively small [4] and a large grain size is one of the key parameters to
achieve higher efficiency. The large grain sizes are obtained by a three-stage evaporation
process [5]. However, the three-stage evaporation process results III group element (Ga in CIGS
and Al in CIAS) distribution in depth direction with a V-shape profile, resulting in the Eg
variation in depth direction. The influence of the Eg variation with the V-shape profile in CIAS
needs to be revealed.
In this paper, we report on the effect of composition gradient in CIAS absorber
deposited by the three-stage evaporation process and its effect on solar cell performances. In
this experiment, the composition gradient was changed by changing the compositional
Cu/(In+Al) ratio of the end of the second stage (Cu/III2nd). The film properties and solar cell
parameters were characterized for each experimental condition.
2. Experimental
The CIAS thin-films were deposited by molecular beam epitaxy on a Mo/soda-lime
glass (SLG) substrate by the three-stage evaporation process. The substrate size was 2x2 cm2.
Figure 1 shows the temperature profile of the three-stage evaporation process. In the first stage,
In, Al and Se were deposited on the substrate heated to 350 oC to form a (In,Al)2Se3 layer. In the
4
second stage, Cu and Se were deposited at 580 oC until the film composition becomes Cu-rich
(Cu/III > 1) to form a Cu2-xSe phase to promote the crystal growth. The stoichiometry of the
film composition during the second stage was detected by the substrate temperature fall owing
to the higher emissivity of Cu2-xSe phase than CIS [6]. The substrate temperature was measured
by a pyrometer [7] which was placed perpendicular to the substrate surface. In this experiment,
Cu/III2nd ratio was adjusted to 1.1, 1.2, 1.3, 1.5 or 1.7. In the third stage, In, Al and Se were
deposited at the same temperature as the second stage to obtain the suitable Cu/III ratio of 0.9.
The thickness of all the films is set to ~2 m by adjusting the deposition time of each stage. The
total deposition time for all the films is approximately 100 min.
The elemental composition of CIAS films was estimated by energy dispersive
spectroscopy (EDX). The surface morphology and the grain size of the films were observed by
scanning electron microscopy (SEM). The surface roughness was measured by atomic force
microscopy (AFM). The depth profile of the elements Cu, In, Al and Mo were measured by
secondary ion mass spectroscopy (SIMS).
Solar cells were fabricated with the structure Al-grid/ITO/ZnO/CdS/CIAS/Mo/SLG.
The Mo, ZnO and ITO films were deposited by rf-magnetron sputtering with thicknesses of 0.8,
0.1 and 0.1 m, respectively. The CdS film with a thickness of 60 nm was deposited by a
chemical bath process [8]. An Al-grid with a thickness of 0.2 m was deposited by vacuum
5
evaporation. The active cell area was 0.12 cm2 which was defined by mechanical scribing.
The CIAS solar cells were characterized by current-voltage (J-V) measurement under
100 mW/cm2 AM1.5 illumination at 25 °C and external quantum efficiency (EQE) measurement.
In addition, the Eg was calculated from the EQE. Carrier density was measured by
capacitance-voltage (C-V) method.
3. Results and discussion
3.1 Film characterization
The Cu/III, Al/III and Se/(Cu+In+Al) ratios were approximately 0.90, 0.15, and 1.0,
respectively. Figure 2 shows the SIMS depth profiles of Cu, In, Al and Mo for a Cu/III2nd ratio
of (a) 1.1 and (b) 1.7. The CIAS films showed an Al distribution with a V-shape profile. The
valleys of the Al distribution with a V-shape profile for the film with the Cu/III2nd ratio of 1.1,
1.5 and 1.7 were located approximately 0.3, 0.9 and 1.0 m from the film surface, respectively.
This indicates that the valley position of the V-shape was controllable by the Cu/III2nd ratio. The
CIAS film with the Cu/III2nd ratio of 1.7 has a smoother Al profile than that of 1.1. Similar
relationship between the group III element profile and the Cu/III2nd ratio is also reported in
CIGS [9].
Figure 3 shows cross-sectional (top) and surface (middle) SEM images and surface
6
AFM images (bottom) of CIAS films with the Cu/III2nd ratio of (a) 1.1 and (b) 1.7. The
cross-sectional SEM images showed grains size of approximately 2 m for both films. In this
experiment, the large grain size and the columnar shaped crystal advantageous to collection of
photo-generated carriers were obtained. Compared with the CIAS film with the Cu/III2nd ratio of
1.1, the surface roughness of the CIAS film with the Cu/III2nd ratio of 1.7 was larger as shown in
the surface SEM and AFM images. Figure 4 shows the rms roughness variations of the CIAS
films as a function of the Cu/III2nd ratio. The surface roughness increased for Cu/III2nd ratios up
to 1.3 and saturated for ratio values greater than 1.3. This indicates that the surface roughness is
influenced by the Cu/III2nd ratio.
3.2 Device characterization
Figure 5 and Table 1 show the device parameters of CIAS solar cells as a function of
the Cu/III2nd ratio, as well as Eg and the reverse saturation current (J0). The Eg values were
determined from the absorption edge of the EQE (not shown in this paper). The change of Jsc
and Voc is due to the tradeoff by the Eg variation for the different CIAS films, which can be
understood from the fact that the JscxVoc product is almost constant. The FF decreased with
increasing Cu/III2nd ratio. The efficiency of the CIAS solar cell mainly depends on the FF. The
carrier density of the CIAS films was approximately 1016 cm-3 for all Cu/III2nd ratios. The
7
depletion width is calculated to be approximately 0.3 m. If the valley of the V-shape is located
deeper than the edge of depletion region, the valley acts as a barrier for photo-generated
electrons to contribute photo-current. This case corresponds to the CIAS films with the Cu/III2nd
ratio of 1.1 because the depletion width estimated by C-V measurement is approximately 0.3
m and the valley depth from the surface estimated by the SIMS measurement as shown in
Figure 2 (a) is approximately 0.3 m. With increasing Cu/III2nd ratio, the valley depth increases
so that the valley acts as barrier for photo-generated electrons. This indicates that the
recombination of the photo-generated electrons increases [10]. On the other hand, the J0
increased Cu/III2nd greater than 1.3. This is the similar behavior as rms roughness as shown in
Figure 4, which increases the pn junction area and should affect the coverage of CdS/ZnO/ITO
multilayers. Thus, this indicates that the increase in J0 would be induced by the increase of the
surface area of CIAS films. Film deposited with Cu/III2nd ratio greater than 1.3 leads to the
distant position of the V-shape from the surface and the increase in the surface roughness and
the J0. From the above results, the Cu/III2nd ratio needs to be less than 1.3 for optimum
performance.
4. Conclusions
The CIAS thin-film was prepared by the three-stage evaporation process. The Cu/III2nd
8
ratio was changed to 1.1, 1.2, 1.3, 1.5 or 1.7. The CIAS films showed an Al distribution with a
V-shape profile. The valley position of the V-shape was controllable by Cu/III2nd ratio. The
valleys of the Al distribution with a V-shape profile for the film with the Cu/III2nd ratio of
1.1~1.7 were located approximately 0.3~1.0 m from the film surface. The rms surface
roughness measured by AFM increased from 40 nm for Cu/III2nd=1.1 to 90nm at Cu/III2nd=1.3
and then saturated for ratio values greater Cu/III2nd ratios. The FF decreased with increasing
Cu/III2nd ratio with the efficiency of the CIAS solar cell mainly depending on the FF. This case
corresponds to the CIAS films with the Cu/III2nd ratio of 1.1 because the depletion width is
approximately 0.3 m and the valley depth from the surface is approximately 0.3 m. With
increasing the Cu/III2nd ratio, the valley depth increases so that the valley acts as barrier for
photo-generated electrons. This indicates that the recombination of the photo-generated
electrons increases. The J0 increased the Cu/III2nd ratio of greater than 1.3. This follows a similar
behavior as the rms surface roughness, which increase pn junction area and should affect the
coverage of CdS/ZnO/ITO multilayers. The Cu/III2nd ratio greater than 1.3 leads to the distant
position of V-shape from the surface and the increase in the surface roughness. From the above
result, Cu/III2nd ratio needs to be less than 1.3 for optimum performance.
Acknowledgements
9
This study was supported by Industrial Technology Research Grant Program in 2005 from New
Energy and Industrial Technology Development Organization (NEDO) of Japan.
References
[1] M. A. Contreras, K. Ramanathan, J. AbuShama, F. Hasoon, D. L. Young, B. Egaas, and R.
Noufi, Prog. Photovolt. Res. Appl. 13 (2005) 209-216.
[2] W. N. Shafarman, R. Klenk, and B. E. McCandless, J. Appl. Phys. 79(9) (1996) 7324-7328.
[3] S. Marsillac, P. D. Paulson, M. W. Haimbodi, R. W. Birkmire, and W. N. Shafarman, J. Appl.
Phys. 81(7) (2002) 1350-1352.
[4] W.N. Shafarman, S. Marsillac, P.D. Paulson, M.W. Haimbodi, T. Minemoto, and R.W.
Birkmire, Proceedings of the 29th IEEE Photovoltaic Specialists Conference, Louisiana, 2002 p.
519-522.
[5] K. Ramanathan, G. Teeter, J.C. Keane, R. Noufi, Thin Solid Films 480/481 (2005) 499-502.
[6] N. Kohara, T. Negami, M. Nishitani, and T. Wada, Jpn. J. Appl. Phys 34(9A) (1995)
L1141-L1144.
[7] K. Sakurai, R. Hunger, R. Scheer, C. A. Kaufmann, A. Yamada, T. Baba, Y. Kimura, K.
Matsubara,
P.
Fons,
H.
Nakanish
and
Appl. 12(2/3) (2004) 219-234.
10
S.
Niki,
Prog.
Photovolt.
Res.
[8] Y. Hashimoto, N. Kohara, T. Negami, M. Nishitani, T. Wada, Sol. Energy Mater. Sol. Cells
50(1/4) (1998) 71-77.
[9] T. Negami, T. Satoh, Y. Hashimoto, S. Shimakawa, S. Hayashi, M. Muro, H. Inoue, M.
Kitagawa, Thin Solid Films 403/404 (2002) 197-203.
[10] T. Dullweber, G. Hanna,W. Shams-Kolahi, A. Schwartzlander, M.A. Contreras, R. Nou,
H.W. Schock, Thin Solid Films 361/362 (2000) 478-481.
11
Table 1: Device parameters of CIAS solar cells.
Jsc
(mA/cm2)
Voc
(V)
JscxVoc
(mW/cm2)
FF
Cu/III2nd ratio
Eg
(eV)
Effi.
(%)
J0
(A/cm2)
1.1
1.16
31.5
0.593
18.7
0.674
12.6
3.4x10-6
1.2
1.10
33.7
0.543
18.3
0.680
12.4
6.5x10-6
1.3
1.10
33.6
0.549
18.4
0.663
12.2
2.7x10-5
1.5
1.12
34.7
0.544
18.9
0.648
12.2
4.1x10-5
1.7
1.16
30.8
0.591
18.2
0.638
11.6
5.6x10-5
12
Figure captions
Figure 1: Temperature profiles of the three-stage evaporation process.
Figure 2: SIMS depth profiles with Cu/III2nd ratio of (a) 1.1 and (b) 1.7.
Figure 3: Cross-sectional (top) and surface (middle) SEM images and surface AFM images
(bottom) of CIAS films for Cu/III2nd ratio of (a) 1.1 and (b) 1.7.
Figure 4: Rms surface roughness of CIAS films as a function of Cu/III2nd ratio.
Figure 5: Device parameters of CIAS solar cells as a function of Cu/III2nd ratio.
13
Substrate
temperature
stoichiometry
cooling
580 oC
350 oC
second stage
Cu, Se
first stage
In, Al, Se
R.T.
Time (a.u.)
Figure 1
Hayashi et al.
14
third stage
In, Al, Se Se
CIAS Mo
(a)
Count (cps)
106
In
Al
105
104
103
Cu
Mo
102
0
500
1000
1500
2000
Depth (nm)
CIAS Mo
(b)
Count (cps)
106
In
Al
105
104
103
Cu
102
Mo
0
500
1000
1500
Depth (nm)
Figure 2
Hayashi et al.
15
2000
2500
CIAS
CIAS
2.0 m
Mo
SLG
2.0 m
Mo
SLG
10 m
10 m
X=2.0m/div
Z=0.5m/div
X=2.0m/div
Z=0.5m/div
(a)
(b)
Figure 3
Hayashi et al.
16
rms roughness (nm)
100
80
60
40
20
0
1.0
1.1
1.2
1.3
1.4
1.5
Cu/III2nd ratio
Figure 4
Hayashi et al.
17
1.6
1.7
1.8
2
J0 (A/cm )
FF
2
JscxVoc (mW/cm )
Voc (V)
2
Jsc (mA/cm )
36
34
32
30
28
0.65
0.62
0.59
0.56
0.53
0.50
20.0
19.5
19.0
18.5
18.0
17.5
17.0
0.70
0.68
0.66
0.64
0.62
0.60
10-4
10-5
Effi. (%)
10-6
13.0
12.5
Cu/III (2nd end)
12.0
11.5
11.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
Cu/III2nd ratio
Figure 5
Hayashi et al.
18