Introduction

Ternary chalcopyrite compounds CuInX2 (X = S, Se, Te) are being considered as important semiconductor materials for thin film solar cell applications because it has a direct band gap (1.08 eV) which can be tailored by the addition of gallium, high optical-absorption coefficient, good stability, largest efficiency, cost effective and chemically and environmentally stable (Arya et al. 1991; Chaure et al. 2005). Furthermore, these compounds can be easily deposited either n- or p-type thin film form by controlling the elemental composition (Chaure et al. 2004; Chraibi et al. 2004). A CIGS solar cell achieved the highest efficiency of over 20% on laboratory scale with high-vacuum techniques (Repins et al. 2008) which indicate that CIS-based solar cells are able to compete with Si-based solar cell. Various other deposition techniques apart from electrodeposition are used by researcher such as sputtering (Yamaguchi et al. 1992), co-sputtering of the elements (Arya et al. 1991), Rapid thermal processing (Karg et al. 1993), Pulse laser deposition (Wang et al. 2010), Spray pyrolysis (Bates et al. 1983) and Molecular beam epitaxy (Niki et al. 2002), etc.

In this paper we report the cathodic electrodeposition of CuInSe2 on a Fluorine doped tin oxide (FTO) substrates. Advantages of electrodeposition are low equipment cost, high deposition rate, negligible waste of chemicals and possibility to deposit over large area. Moreover, an isothermal process mainly controlled by concentration, temperature, pH of the electrolyte and applied potential. A Pourbaix diagram has been used as a guide to fix the preliminary electrodeposition conditions, pH and temperature of the electrolyte, concentration of ionic species etc. Thin film deposition under various conditions have been performed to understand the deposition mechanism and to optimize the deposition condition. The elemental composition and structural properties of the CIS thin films were studied by using Energy dispersive X-rays analysis and the X-ray diffraction technique.

Experimental details

CuInSe2 (CIS) thin films were deposited on FTO-coated glass substrates of sheet resistance 15–20 Ω/cm2. A conventional three-electrode electrochemical cell was employed for the deposition of thin films consisting FTO as a working electrode, graphite as a counter electrode and Ag/AgCl as a reference electrode (Pandey et al. 1996). The ionic source for Cu, In and Se was CuSO4 ·5H2O, In2(SO4)3·5H2O and H2SeO3. The details are reported elsewhere (Chaure et al. 2004). Lithium sulfate was used as a supporting electrolyte, to bring deposition potential of all species closer for a better co-deposition of Cu, In and Se. Triethanolamine (TEA) was used as a complexing agent which forms complex with copper that reduces the rate of copper reduction and shifts copper deposition potential towards cathodic direction. Triethanolamine also improves the adhesion of film to the substrate. Similar results are reported by Whang et al. (2009) and Yu et al. (2010). Indeed, we found that the films deposited in presence of TEA were highly uniform, very well adhesive to the substrate and free of loosely bound particles on the surface of sample. The above precursors were dissolved in double-distilled water with appropriate molar concentration. The pH of bath was adjusted ~2 using H2SO4 or NaOH. Cyclic voltammetry (CV) and potentiostatic deposition of CIS layers were carried out using μ3AUT 70762 AUTOLAB potentiostat of working temperature ~40°C without stirring. The information observed in CV was used to optimize the film deposition potential. The films were deposited at various growth potentials ranging from −0.6 to −0.9 V with respect to Ag/AgCl. The layers were annealed ~400°C for 10 min to improve homogeneity and crystalline nature. Subsequently, the films were characterized using range of characterization techniques.

Results and discussion

Cyclic voltammetry was recorded to understand the co-deposition procedure of Cu, In and Se. Figure 1 shows the CV recorded for CIS bath on FTO working electrode. The features related to the deposition of elemental, binary and ternary phases of CIS were observed. Initially, at lower cathodic potential up to −0.4 V, the features noticed were related to the reduction of Cu and Se by charge-transfer reaction;

Fig. 1
figure 1

Cyclic voltammogram recorded on FTO in the presence of Cu, In and Se in electrolytic bath

Full size image
$$ {\text{Cu}}^{2 + } + 2{\text{e}}^{ - } \to {\text{Cu}}({\text{s}}) \, ; \, E_{0} = \, 0.34{\text{ V }}({\text{NHE}}) $$
(1)
$$ {\text{HSeO}}_{2}^{ + } + 4{\text{e}}^{ - } + 3{\text{H}}^{ + } \to {\text{Se}}({\text{s}}) + 3{\text{H}}_{2} {\text{O}}; \, E_{0} = 0.74\,{\text{V }}({\text{NHE}}) $$
(2)

The deposition of indium was observed beyond the potential −0.6 V with respect to Ag/AgCl reference electrode, by following charge-transfer reaction;

$$ {\text{In}}^{3 + } + 3{\text{e}}^{ - } \to {\text{In}}({\text{s}}); \, E_{0} = - 0.34\,{\text{V }}({\text{NHE}}) $$
(3)

A plateau observed in between −0.7 and −0.9 V favors the deposition of CIS alloys. Sharp linear rise in cathodic current observed beyond −0.9 V could be due to the metallic deposition of indium along with hydrogen evolution. CIS can be formed by the chemical reaction:

$$ {\text{Cu}}^{2 + } + {\text{In}}^{3 + } + 2{\text{SeO}}_{2}^{ - } + 12{\text{H}}^{ + } + 13{\text{e}}^{ - } \to {\text{CuInSe}}_{2} + 6{\text{H}}_{2} {\text{O}} $$
(4)

During anodic curve, the dissolution (stripping) peaks for In, Cu and Se were observed and marked as a, b and c, respectively.

Theoretical thickness of the films deposited at various potentials was calculated using Faraday’s law:

$$ T = \frac{1}{nFA}\left[ {\frac{itM}{\rho }} \right] $$
(5)

where ‘n’ is the number of electron transferred, ‘F’ is the Faraday’s constant, ‘A’ is the area of electrode, ‘i’ is the current, ‘t’ is the deposition time, ‘M’ is the formula weight, and ‘ρ’ is the density of intrinsic CIS (Friedfeld et al. 1999).

Thickness measured for the samples deposited at potential −0.8 and −0.9 V was 590 and 650 nm (±20 nm), respectively. Note that the deposition time was kept constant (10 min) for the deposition of above samples.

Structural characterization was carried out with the help of X-ray diffraction (D8 Bruker). The X-ray diffractogram of the CIS thin film deposited at different potential is shown in Fig. 2. The presence of sharp peaks in the X-ray diffractogram shows that CIS is polycrystalline in nature. The CIS is found to exhibit chalcopyrite tetragonal structure with a preferred orientation along certain crystallographic planes which have been identified as (112), (204/220), (116/312). The secondary binary phases Cu x Se and In x Se were observed for the film deposited for −0.6, −0.7 and −0.9 V, whereas stoichiometric film along with high intensity CIS peaks were observed for the layer deposited at −0.8 V. Note that no secondary phases were observed for film deposited at −0.8 V. The substrates (FTO) peaks were also observed in XRD spectra which are marked as solid circles. Lattice parameter for tetragonal structure can be obtained by the formula,

Fig. 2
figure 2

XRD spectra of as-deposited film at potentials. a −0.6 V, b −0.7 V, c −0.8 V and d −0.9 V with respect to Ag/AgCl reference electrode

Full size image
$$ \frac{1}{{d^{2} }} = \left[ {\frac{{h^{2} + l^{2} }}{{a^{2} }} + \frac{{k^{2} }}{{c^{2} }}} \right]. $$
(6)

The lattice parameter was calculated to be 5.783 and 11.73 Å which gives tetragonal distortion i.e., c/a ratio is approximately equal to 2.

Post deposition heat treatment on CIS films was carried out in the furnace at 400°C (± 5°C) for 10 min. After heat treatment the peak intensity improves which indicates the improvement in crystallinity of CIS thin films (Kemell Marianna et al. 2000). The X-ray patterns obtained for as-deposited and annealed CIS thin film are shown in (Fig. 3).

Fig. 3
figure 3

XRD spectra of CIS deposited film at potential −0.8 V; a as-deposited and b annealed at 400°C for 10 min

Full size image

The grain size of as-deposited and annealed sample was calculated using the Scherrer equation,

$$ t = \frac{0.9\lambda }{\beta \cos \theta } $$
(7)

where λ is the wavelength, β is the FWHM and θ is the diffraction angle (Cullity and Stock 1978). Grain size of the as-deposited and annealed sample of CIS was calculated approximately 13 and 24 nm, respectively (Panchal et al. 2009).

Compositional and morphological analysis was carried out using SEM and EDAX (JEOL JSM-6360A). Figure 4a–d shows the SEM images of as-deposited CIS, deposited at potential −0.6, −0.7, −0.8 and −0.9 V, respectively. All the films show uniform and compact deposition. CIS layer deposited at −0.8 V was compact, spherical and uniform, whereas the mixed morphology consisting uniform grains along with irregular shaped clusters were observed for other samples. After annealing CIS films become compact and power free with grayish metallic finishing. Heat treated sample (Fig. 5) can be clearly seen the grain enhancement along with noticeable improvement in crystallinity.

Fig. 4
figure 4

SEM images of as-deposited films at potentials. a −0.6 V, b −0.7 V, c −0.8 V and d −0.9 V with respect to Ag/AgCl reference electrode

Full size image
Fig. 5
figure 5

SEM images of CIS deposited film at potential −0.8 V; a as-deposited and b annealed at 400°C for 10 min

Full size image

The elemental composition was detected by EDAX at accelerated voltage 20 kV and probe current 1nA. Table 1 shows the chemical composition of as deposited CIS films at potentials −0.6, −0.7, −0.8 and −0.9 V.

Table 1 Atomic percentage composition of CIS thin films by EDAX
Full size table

CIS layer deposited at −0.8 V was nearly stoichiometric ~ 24:25:51 for Cu:In:Se. The growth potential plays an important role to control the stoichiometry of layer. The Cu-rich and In-rich layers are p-type and n-type, respectively (Chaure et al. 2005). The optical band gap was estimated from Fig. 6 by plotting the graph (αhν)2 as a function of (hν). The electronic transition between the valance and conduction band in case of CIS was direct transition. The transition of electron can be allowed as permitted by the transition probability or forbidden where no such probability exists. The transition probability is given by the Eq. 8. The absorption coefficient was calculated of the order of 104 per cm. Similar value is reported by Friedfeld et al. (1999).

$$ \left( {\alpha h\upsilon } \right)^{2} = \alpha_{0} \left( {h\nu - E_{\text{g}} } \right) $$
(8)

where Eg denotes the band gap, hν the energy of the incident photon and α0 is a constant. Value of Eg was obtained by extrapolating the linear portion of the graph to intercept the photon energy axis. After annealing, the band gap comes to very closer to that of ideal value which is 1.04 eV.

Fig. 6
figure 6

Plot of (αhν) 2 versus (hν) of CIS deposited film at potential −0.8 V; a as-deposited and b annealed at 400°C for 10 min

Full size image

Electrical measurements

The electrical studies were performed on FTO/CIS/Al structure (Fig. 7).

Fig. 7
figure 7

I–V characteristics of CIS films deposited at potential −0.8 V; a as-deposited and b annealed at 400°C for 10 min

Full size image

An electrical measurement shows Schottky behavior of diode. IV curve clearly shows that the improvement in conductivity after heat treatment. The improvement in current indicates the decrease in grain boundaries and defects.

Conclusion

CIS films were potentiostatically electrodeposited from aqueous bath. A plateau observed between −0.7 and −0.9 V was found to be suitable for the deposition of CIS films. CIS films deposited at reported potentials were polycrystalline with tetragonal chalcopyrite structure. The optical band gap was estimated to be 1.04 eV. SEM study showed that the CIS films are uniform, densely compact, and free from pores. From EDAX, it has been observed that the growth potential plays an important role to control the stoichiometry of film. Stoichiometry of the thin film deposited at −0.8 V is very close to the desired composition of film. The current density has been increased for heat treated sample could be due to change in conductivity of material as well as the defects present in as-deposited sample may be reduced. Further study on CIS is under progress and the development of superstrate and substrate solar cell devices will be the topic of future work.