All solution processable graded CIGS solar cells fabricated using electrophoretic deposition

Abstract

Graded band gap chalcopyrite solar cells are fabricated based on an all solution processable synthesis method with the aid of electrophoretic deposition and a superstrate structure. The solvothermally synthesized nanoparticles crystallize in the chalcopyrite–wurtzite mixed phase without any trace of impurity phases, as confirmed by XRD and Raman spectroscopy. Crack-free electrophoretic deposited films are formed. However, the porosity of the films is raised due to the Ga content. The J–V characteristics of the CuIn_1−xGa_xS₂ samples are compared with those of back extreme grading (BEG), back moderate grading (BMG), and double moderate grading (DMG) solar cells. The highest efficiency (J_SC = 16.50 mA cm⁻², V_OC = 640 mV, fill factor = 0.61, η = 6.44%) is achieved in DMG cells, mainly due to reducing the minority carrier recombination at the absorber/back contact interface by back grading and increasing the photoresponse or open circuit voltage by front grading in the space charge region. The medium double grading in this structure avoids notch formation in the band structure. Dark diode analysis demonstrates that the lowest series resistance, shunt conductance and reverse saturation current are obtained, as well as a diode ideality factor near 2, in DMG solar cells.

---

1. Introduction

Lack of fossil fuel resources, air pollution, global warming, and population growth lead to the necessity of finding new energy resources. Renewable energies, especially solar energy, have the potential to fulfill the 21^st century’s energy desires. Among all photovoltaic technologies, chalcopyrite solar cells with the advantages of having a direct and tunable band gap (1.04–2.5 eV), high absorption coefficient (>10⁴ cm⁻¹), appreciable record efficiency (19.9%), and respectable efficiency to cost ratio, introduce themselves as a promising technology to fulfill humankind’s future energy desires.^1–3 However, the difficulties in fabricating large area modules, rarity of raw materials, need of costly high vacuum apparatus, and requirement for toxic post selenization treatments, limit the large scale production of these cells.^4,5

The solution processable deposition of Cu(In_1−xGa_x)(S_ySe_1−y)₂ (CIGS) compounds for the fabrication of solar cells attracts a great deal of attention due to its cost effectiveness, high production rate, ability for roll-to-roll production, compositional uniformity over large areas, and low material waste.^6,7 A variety of solution processed methods have been utilized for the deposition of CIGS including spray pyrolysis,⁸ electrodeposition,⁹ spin coating,¹⁰ doctor blading,¹¹ inkjet printing,¹² and electrophoretic deposition (EPD).¹³ EPD is a low cost, simple, fast, and low waste method with the potential for the automation and the deposition of large area films.¹⁴ Although the ability of doctor blading for mass production is higher than EPD, this method always suffers from organic residues used to prepare the paste. EPD has been used for the deposition of quantum dots (QDs) in the fabrication processes for quantum dot sensitized solar cells (QDSCs). CdSe QDs were deposited using EPD to reach an efficiency of 1.48% in QDSCs.¹⁵ Kamat et al. used EPD to deposit CIGS quantum dots for the fabrication of a QDSC with an efficiency of 3.91%.¹⁶ However, the EPD of chalcopyrite compounds for the deposition of a thick absorber layer in solution processable solar cells has rarely been reported. A Cu₂ZnSnS₄ semiconductor was deposited via EPD to achieve a 500 nm thick film for application as an absorber layer.¹⁷

Since the band gap of chalcopyrite compounds can be tuned by changing the Ga/(In + Ga) and S/(Se + S) ratio in the range of 1.04–1.68 eV and 1.04–1.53 eV,¹⁸ respectively, different semiconductors with a slight lattice mismatch and difference in composition can be simply deposited. This composition dependent band gap allows for band gap grading strategies (including front, back, and double gradings) in order to fabricate a device with a tandem structure and surpass the Shockley–Queisser thermodynamic limit for a single planar junction solar cell or to reduce the recombination at the back contact by repelling minority carriers from the interface.¹⁹ In most cases, improved efficiencies are reported.^20–23 In this article, all solution processable band gap graded CuIn_1−xGa_xS₂ inverted type (superstrate) solar cells were fabricated and the effect of the tandem structure on the cell efficiency of these devices is investigated. Fabrication of these kinds of cells may pave the way to achieve flexible low cost solar cells.

2. Experimental

2.1 Materials

Copper chloride (CuCl₂, ≥98%, Merck), indium chloride (InCl₃, 99.99%, Alfa Aesar), sulfur powder (S, 99%, Merck), gallium nitrate (Ga(NO₃)₃, 98%, Merck), absolute ethanol (EtOH, Merck, 99%), 1-propanol (1-PrOH, 99.9%, Merck), deionized water (DIW, 18.2 MΩ), hydrochloric acid (HCl, 32%, Merck), tetrapropyl ortotitanate (TTiP, 98.5%, Merck), cadmium sulfate (3CdSO₄·8H₂O, 99.5%, Merck), thiourea (NH₂–CS–NH₂, 99%, Merck), ammonia (NH₄OH, 30%, Merck), triethylenetetramine (TETA, 95%, Merck), and a fluorine doped tin oxide transparent conductive substrate (FTO, 15 Ω sq⁻¹, Dyesol) were purchased and used as received.

2.2 TiO₂ blocking layer deposition

The sol preparation method for depositing the blocking layer was reported elsewhere.²⁴ Typically, TTiP was dissolved in 1PrOH for 10 min. In another beaker, a solution of 1PrOH, DIW, and HCl was prepared and added drop-wise to the former solution under vigorous stirring and stirred for 1 h. The blocking layer was deposited by dip coating of the prepared sol on pre-cleaned FTO substrates and dried at 100 °C for 10 min. The dip coating and drying processes were repeated twice to reach the thickness of about 100 nm. Then, the film was calcined at 450 °C for 1 h.

2.3 CdS window layer deposition

The CdS n-type layer was deposited through chemical bath deposition as reported elsewhere.²⁵ For this purpose, 0.015 M CdSO₄ in 2 ml DIW, 1.5 M thiourea in 1 ml DIW, and 2.6 ml NH₄OH in 15 ml DIW solutions were separately prepared and added together before deposition. The substrates were placed in the bath at 80 °C for 12 min. The pale orange CdS deposited substrates were washed and dried.

2.4 CIGS electrophoretic deposition and device fabrication

CuIn_1−xGa_xS₂ nanoparticles were synthesized via a modified solvothermal method according to a previous report.²⁶ Briefly, stoichiometric amounts of CuCl₂ (0.2 mmol), InCl₃ (0–0.2 mmol), S (0.4 mmol), and Ga(NO₃)₃ (0–0.2 mmol) were placed into a homemade autoclave in triethylenetetramine as a solvent. The synthesis parameters of 220 °C for 1 h under an internal pressure of 400 kPa and continuous stirring were applied to the samples during the synthesis. Then, the autoclave was allowed to cool naturally to ambient temperature. The precipitates were collected by centrifugation and washing several times with ethanol to remove the organic residues and were then dried at 60 °C for 20 min.

Electrophoretic deposition of the nanoparticles was conducted according to a previous report with slight modification.²⁷ For the electrophoretic deposition, the CuIn_1−xGa_xS₂ nanoparticles were dispersed in EtOH (0.09% wt) by 30 min ultrasonication. A very small amount of TETA (4% vol.) was added to the colloid to increase the dispersibility and deposition current density of the nanoparticles. A DC current with constant voltage and FTO as the anode and TiO₂/CdS coated FTO as the cathode were utilized. The electrophoretic deposition parameters were set to 60 V for 60 min with 0.3 cm as the distance between the anode and cathode. For the deposition of the back extreme grading cell (BEG), the colloids of CuInS₂, CuIn_0.75Ga_0.25S₂, CuIn_0.5Ga_0.5S₂, CuIn_0.25Ga_0.75S₂, and CuGaS₂ were sequentially deposited by applying 60 V for 12 min for each colloid. The colloids of CuInS₂, CuIn_0.75Ga_0.25S₂, and CuIn_0.5Ga_0.5S₂ were also successively applied by 60 V for 20 min for each colloid, for the deposition of the back moderate grading cell (BMG). Finally, the colloids of CuIn_0.75Ga_0.25S₂, CuInS₂, CuIn_0.75Ga_0.25S₂, and CuIn_0.5Ga_0.5S₂ were consecutively deposited by applying 60 V for 15 min for each colloid for the fabrication of the double moderate grading cell (DMG). After deposition, the films were washed by dipping in EtOH and dried naturally. The films were annealed at 250 °C for 30 min in air and at 400 °C for 1 h under an argon/H₂S atmosphere. Finally, the Au layer with a thickness of about 500 nm was deposited on the film using magnetron sputtering through a shadow mask. The schematic structure and FESEM cross section of the CuIn_0.5Ga_0.5S₂ device are shown in Fig. 1.

Fig. 1 (a) Schematic representation of the tandem CIGS graded superstrate structure and (b) FESEM cross-sectional view of the electrophoretic deposited layers before Au deposition.

2.5 Characterization of the thin films and solar cells

The phase structure of the films was studied using X-ray diffraction (XRD), on a Philips X-pert pro PW1730, and using Cu-Kα radiation. Field emission scanning electron microscopy (FESEM), on a Hitachi S4160, was used to investigate the morphology of the films. Raman spectra were recorded using a BRUKER (SETERRA, spectral resolution < 3 cm⁻¹) MicroRaman spectrometer (785 nm argon laser, 25 mW) equipped with a confocal microscope. Photovoltaic measurements were performed using an AM 1.5 solar simulator (Sharif Solar). The power of the simulated light was calibrated to 1000 W m⁻² with a Si photodiode. J–V curves were obtained by applying an external bias to the cell and measuring the generated photocurrent with a PGSTAT digital source meter. All tests were conducted at room temperature. Two cells with the same conditions for each parameter were fabricated for a repeatability test. Diffuse reflectance spectroscopy (DRS), on an Avalight DH-S Avaspec 2048 TEC, was performed to investigate the absorption coefficients and band gaps of the CIGS samples. The spectra were calibrated using the Kubelka–Munk function: F(R) = (1 − R²)/2R, where F(R) and R are the equivalent absorption coefficient and the reflectance, respectively.

3. Results and discussion

Fig. 2(a) shows the XRD patterns of the CuIn_1−xGa_xS₂ nanoparticles synthesized using a solvothermal method. All samples crystallize in both the chalcopyrite and wurtzite crystal structures (JCPDS card no. 27-0159, 47-1372, and 25-0279).^28,29 Although the chalcopyrite phase is thermodynamically more stable, the wurtzite metastable phase can be formed in CIGS nanoparticles with the assistance of the coordinating ligands, sulfur source type, coordinating solvent, sequence of metal reactions, reaction temperature, and concentration.³⁰ The smaller size of the gallium ion compared to indium in the lattice causes a shift toward a higher diffraction angle with increasing gallium content. It is believed that the formation of the intermediate Cu₂S phase is necessary for wurtzite formation.^31,32 TETA can strongly bind to the In³⁺ cations and act as a ligand, manipulating the kinetics of wurtzite growth. The flexibility of the stoichiometry caused by the random distribution of ions in the wurtzite phase, provides the ability to tune the Fermi energy over a wide range, which is remarkably advantageous for photovoltaic applications.

Fig. 2 (a) XRD patterns and (b) Raman spectra of the CuIn_1−xGa_xS₂ nanoparticles.

The Raman spectra of the CuIn_1−xGa_xS₂ nanoparticles are demonstrated in Fig. 2(b). The chalcopyrite semiconductors in the space group of D¹²_2d contain two formulae per primitive unit cell. The A₁ TO modes are the strongest Raman modes at around 293 cm⁻¹ for CuInS₂ and 302 cm⁻¹ for CuGaS₂, which match well with the Raman spectrum of chalcopyrite compounds.³³ The shoulder at around 302 cm⁻¹ in the CuInS₂ sample shows Cu–Au ordering (A*). The wurtzite CIGS phase shows four bands at 256, 288, 302 and 344 cm⁻¹.³⁴ These bands could be observed in the samples. However, they cannot be thoroughly separated from the chalcopyrite Raman modes. The Cu_xS compounds vibrate at 474 cm⁻¹, and could not be detected in these samples, revealing the nonexistence of impurity phases.³⁵

The FESEM images of CuIn_1−xGa_xS₂ films are represented in Fig. 3. The particle size does not dramatically change with Ga content. However, larger agglomerates could be observed in Ga doped samples. Although relatively homogenous crack-free films are formed in all samples, the porosity of the films is affected by the Ga substitution. Large agglomerates in Ga doped samples cause large porosity in those films. The EDX analyses, and R₁ = Cu/In + Ga, R₂ = Ga/In + Ga, and R₃ = S/Cu + In + Ga ratios of the CuIn_1−xGa_xS₂ samples are tabulated in Table 1. The EDX analyses are well matched with the stoichiometry of the chalcopyrite compounds. The R₁, R₂, and R₃ ratios show the occurrence of a slight off-stoichiometry in the samples, which is associated with the chemical synthesis route.

Fig. 3 FESEM images of the CuIn_1−xGa_xS₂ films with x equal to (a) 0, (b) 0.25, (c) 0.5, (d) 0.75 and (e) 1, and (f) graded CIS to CGS.

Table 1 EDX analyses, and R₁ = Cu/In + Ga, R₂ = Ga/In + Ga, and R₃ = S/Cu + In + Ga ratios of the CuIn_1−xGa_xS₂ samples

Sample	Cu (at%)	In (at%)	Ga (at%)	S (at%)	R1	R2	R3
x = 0	29.20	23.28	0.00	47.53	1.25	0.00	0.91
x = 0.25	25.27	19.20	7.30	48.21	0.95	0.28	0.93
x = 0.5	25.62	13.58	15.53	45.26	0.88	0.53	0.83
x = 0.75	31.20	7.92	13.58	47.28	1.45	0.63	0.90
x = 1	26.33	0.00	29.72	43.94	0.89	1.00	0.78

---

Fig. 4 reveals the EDX line scans of BEG, BMG, and DMG CIGS thin film cross sections. The In/In + Ga content is gradually increased with distance from the back contact in BEG cells. The Cu and S content does not dramatically change in the different layers of the graded cell. In the case of BMG cells, the Ga content increases with distance from the back contact but with a lower slope. In the DMG sample, the layer deposition starts with CuIn_0.5Ga_0.5S₂ through CuInS₂ and ends with CuIn_0.75Ga_0.25S₂.

Fig. 4 EDX line scans of graded CuIn_xGa_1−xS₂: (a) BEG, (b) BMG, and (c) DMG CIGS thin film cross section.

Fig. 5(a) displays the DRS spectra of the CuIn_xGa_1−xS₂ solvothermally synthesized samples. The band gaps of the CuIn_xGa_1−xS₂ nanoparticles increase with increasing Ga content as expected. The band gap values of about 1.43, 1.56, 1.77, 1.96 and 2.29 eV could be calculated for x = 0, 0.25, 0.5, 0.75, and 1, respectively. The slightly lower band gaps of the CIGS nanoparticles in this research compared to other reported values may be attributed to the formation of In_Cu, Ga_Cu, Cu_In, and Cu_Ga antisite defects.³⁶ These defects cause a change of up to 0.29 eV in the band edge positions. The In_Cu²⁺ + 2V_Cu⁻ defect clusters are also very common in CIGS compounds, which can affect the band gap. However, the formation energies of V_S, V_Cu, V_In, and Cu_i vacancies and interstitial defects are higher and less likely to occur. The band gap of CuIn_1−xGa_xS₂ compounds can be well approximated by:

E_g(x) = xE_g(1) + (1 − x)E_g(0) − b(1 − x)x (1)

where b is the bowing parameter and is the consequence of bond alternation in the lattice. The minimization of the free energy causes Cu and In atoms to avoid each other as nearest neighbors in the cation sublattice, resulting in short range ordering. b values of nearly 0 to 0.25 have been previously reported for CIGS.² The b value of the nanoparticles is about 0.4, which is higher than the expected value and may be due to the wet chemical synthesis method. Fig. 5(b) shows that the color of the colloidal CIGS nanoparticles changes from dark brown to light brown with increasing Ga content.

Fig. 5 (a) DRS spectra and (b) colloidal dispersion of the CuIn_1−xGa_xS₂ nanoparticles.

The J–V characteristic curves of the CuIn_1−xGa_xS₂, BEG, BMG and DMG solar cells are exhibited in Fig. 6. The values of the short circuit current density (J_SC), open voltage (V_OC), fill factor (FF), and efficiency (η) of the cells are also listed in Table 2. The highest efficiency in the CuIn_1−xGa_xS₂ samples is achieved when x = 0.25. The band gap of CuIn_1−xGa_xS₂ semiconductors increases with Ga content mainly as a result of conduction band (CB) elevation. Shifting of the CB in the space charge region (SCR) causes an increase in V_OC. However, the J_SC is slightly decreased due to the lower absorption spectral region. Moreover, Ga widens the existence of the α-phase (the main chalcopyrite compound with an exact stoichiometry in the In₂S₃–Cu₂S phase diagram) and consequently the tolerance to deviation from the Cu/In + Ga ratio. In addition, the c/a lattice constant ratio equals 2 when x ∼ 0.25, which reduces the defect concentration in this sample. The efficiency gradually drops in compounds with higher Ga content. Not only is there a slight increase in V_OC and a dramatic decrease in J_SC, but there is also a reduction in crystallinity and a more porous film is formed in compounds with higher Ga content, which causes this drop in efficiency. Besides, the lattice mismatch of the CGS film with the back contact is higher than that of the CIS film. Thus, it induces more recombination centers at the back contact. Also, since Ga evaporates more easily than In, it can lead to more 2V_Cu⁻–Ga_Cu⁺ defect clusters during the post treatment procedure, which enormously causes interface recombination centers. The FF of 61% is among the highest reported FFs for solution processed superstrate solar cells.^37–39

Fig. 6 J–V characteristic curves of (a) the CuIn_1−xGa_xS₂ solar cells and (b) BEG, BMG and DMG solar cells under dark and illuminated conditions.

Table 2 J–V characteristic parameters of CuIn_1−xGa_xS₂, BEG, BMG and DMG solar cells

Sample	JSC (mA cm^−2)	VOC (mV)	FF	Eff. (%)
x = 0	14.50	580	0.60	5.05
x = 0.25	13.80	630	0.61	5.30
x = 0.5	10.60	670	0.59	4.19
x = 0.75	8.20	720	0.58	3.42
x = 1	6.50	755	0.56	2.75
BEG	10.30	560	0.45	2.60
BMG	15.80	585	0.60	5.55
DMG	16.50	640	0.61	6.44

---

Schematic representations of the band alignment in BEG, BMG and DMG solar cells are shown in Fig. 7. Of the graded cells, DMG cells show the highest efficiency compared to BMG, BEG and ungraded cells. On the other hand, BEG cells show the lowest efficiency among all cells, while their modified version, BMG cells, exhibit a high enough efficiency of 5.54%. Back grading can potentially increase the performance of the device by repelling CB minority carriers and reducing recombination at the back contact/absorber interface.⁴⁰ Nevertheless, in the case of too strong grading, this results in a notch behind the SCR and increases the probability of charge recombination at the quasi neutral region.⁴¹ Therefore, the efficiency in BEG cells is unexpectedly lower. Moreover, the FF of this device is reduced due to extreme grading. Front grading cannot help to increase the device efficiency unless the grading is applied within the SCR. The concept of double grading (with front grading at the SCR and moderate back and front gradients) is found to be the most efficient grading configuration in CIGS solar cells.^21,23 Since V_OC mainly depends on the band gap in the SCR and J_SC correlates to the overall minimum of the band gap, J_SC can be improved due to a relatively higher spectral response and V_OC can be optimized due to lower minority carrier recombination in DMG cells.

Fig. 7 Schematic representation of the band alignment in BEG, BMG and DMG solar cells.

The diode curves in the dark for the BEG, BMG and DMG cells are demonstrated in Fig. 6(b). The diode curves can be analyzed using:

where j_d, is the diode current density, j₀ is the reverse saturation current density, e is the elementary electron charge, k is the Boltzmann constant, n is the diode ideality factor, R_S is the series resistance, G_Sh is the shunt conductance, j₀₀ is the pre-exponential factor, and E_a is the activation energy of the saturation current.⁴² A plot of dj/dV versus V gives G_Sh from the constant dj/dV when V ≤ 0. A plot of dV/dj versus 1/(j − G_ShV) also estimates R_S and n from the y intercept and slope in the high bias regime, respectively. A plot of ln(j − G_ShV) versus V − jR_S gives j₀ and n from the y intercept and slope, respectively. The values of R_S, G_Sh, n, and j₀ are calculated and show that the DMG device shows the lowest R_S and G_Sh (25 Ω cm² and 2.2 mS cm⁻², respectively) values among all graded cells. The lower j₀ value (1.5 × 10⁻³ mA cm⁻²) of DMG cells indicates less recombination in the SCR and the quasi neutral region (QNR) as a result of the double grading structure. An n equal to 1 corresponds to an ideal diode in which recombination occurs only in the QNR (with no recombination in the SCR). An n value close to 2 describes recombination in the SCR. An n value higher than 2 shows additional recombination mechanisms such as interface recombination and tunneling enhanced recombination.⁴² The n value of about 1.8 shows an appropriate diode ideality factor for this device.

Crossovers, kinks, or roll-overs could not be observed in the dark and illuminated curves of BMG and DMG solar cells, shown in Fig. 6(b). However, the curves obtained in the dark and under illumination of the BEG cells illustrate crossover. The crossover is due to the large electron barrier in the conduction band. Upon photodoping of the buffer layer, the potential drops over the buffer and thus the electron barrier is reduced. The high density of acceptors can also form the electron barrier. The large negative charge in the acceptor states leads to a large potential drop over the buffer layer. The absorber acceptors are filled under illumination and hence the acceptor charge and electron barrier are reduced.

4. Conclusions

All solution processable graded band gap chalcopyrite solar cells are fabricated for the first time using electrophoretic deposition and employing the superstrate structure of Au/CIGS(normal or graded)/CdS/TiO₂/FTO. XRD and Raman spectroscopy confirm that the solvothermally synthesized nanoparticles crystallize in the chalcopyrite–wurtzite mixed phase without any trace of impurity phases. The crack-free films are formed using electrophoretic deposition, possessing different Ga content and using the three different grading strategies of BEG, BMG and DMG. The J–V characteristics of the CuIn_1−xGa_xS₂ samples are compared with those of BEG, BMG and DMG solar cells. The highest efficiency of 6.44% is achieved in DMG cells with J_SC, V_OC and FF of about 16.50 mA cm⁻², 640 mV and 0.61, respectively. The main reasons for obtaining the highest improved efficiency in DMG cells are reducing the minority carrier recombination at the absorber/back contact interface by back grading, increasing the photoresponse or V_OC by front grading in the SCR, and avoiding notch formation in the band structure by using a medium grading strategy. Lowest R_S, G_Sh, J₀, and n near 2 are obtained in DMG solar cell, which exhibit low recombination, appropriate interface quality, and high carrier life time in this cell.

Acknowledgements

We wish to thank The Iran National Science Foundation for supporting this research. The Iran Nanotechnology Initiative Council is gratefully acknowledged for partially supporting this research.

Notes and references

1. D. Aldakov, A. Lefrancois and P. Reiss, Ternary and Quaternary Metal Chalcogenide Nanocrystals: Synthesis, Properties and Applications, J. Mater. Chem. C, 2013, 1, 3756–3776.
2. B. J. Stanbery, Copper Indium Selenides and Related Materials for Photovoltaic Devices, CRC Crit. Rev. Solid State Sci., 2002, 27, 73–117.
3. J. Kolny-Olesiak and H. Weller, Synthesis and Application of Colloidal CuInS₂ Semiconductor Nanocrystals, ACS Appl. Mater. Interfaces, 2013, 5, 12221–12237.
4. S. E. Habas, H. A. S. Platt, M. F. A. M. van Hest and D. S. Ginley, Low-Cost Inorganic Solar Cells: From Ink to Printed Device, Chem. Rev., 2010, 110, 6571–6594.
5. A. C. Arias, J. D. MacKenzie, I. McCulloch, J. Rivnay and A. Salleo, Materials and Applications for Large Area Electronics: Solution-Based Approaches, Chem. Rev., 2010, 110, 3–24.
6. M. Graetzel, R. A. J. Janssen, D. B. Mitzi and E. H. Sargent, Materials Interface Engineering for Solution-Processed Photovoltaics, Nature, 2012, 488, 304–312.
7. W. Liu, D. B. Mitzi, M. Yuan, A. J. Kellock, S. J. Chey and O. Gunawan, 12% Efficiency CuIn(Se,S)₂ Photovoltaic Device Prepared Using a Hydrazine Solution Process, Chem. Mater., 2010, 22, 1010–1014.
8. M. Kaelin, D. Rudmann and A. N. Tiwari, Low Cost Processing of CIGS Thin Film Solar Cells. Sol, Energy, 2004, 77, 749–756.
9. J. S. Wellings, A. P. Samantilleke, S. N. Heavens, P. Warren and I. M. Dharmadasa, Electrodeposition of CuInSe₂ from Ethylene Glycol at 150 °C, Sol. Energy Mater. Sol. Cells, 2009, 93, 1518–1523.
10. D. B. Mitzi, M. Yuan, W. Liu, A. J. Kellock, S. J. Chey, L. Gignac and A. G. Schrott, Hydrazine-Based Deposition Route for Device-Quality CIGS Films, Thin Solid Films, 2009, 517, 2158–2162.
11. M. Kaelin, D. Rudmann, F. Kurdesau, H. Zogg, T. Meyer and A. N. Tiwari, Low-Cost CIGS Solar Cells by Paste Coating and Selenization, Thin Solid Films, 2005, 480, 486–490.
12. W. Wang, Y.-W. Su and C. Chang, Inkjet Printed Chalcopyrite CuIn_xGa_1−xSe₂ Thin Film Solar Cells, Sol. Energy Mater. Sol. Cells, 2011, 95, 2616–2620.
13. W. Guo and B. Liu, Liquid-Phase Pulsed Laser Ablation and Electrophoretic Deposition for Chalcopyrite Thin-Film Solar Cell Application, ACS Appl. Mater. Interfaces, 2012, 4, 7036–7042.
14. L. Besra and M. Liu, A review on Fundamentals and Applications of Electrophoretic Deposition (EPD), Prog. Mater. Sci., 2007, 52, 1–61.
15. M. R. Golobostanfard, H. Abdizadeh and S. Mohajerzadeh, Incorporation of Carbon Nanotubes in a Hierarchical Porous Photoanode of Tandem Quantum Dot Sensitized Solar Cells, Nanotechnology, 2014, 25, 345402.
16. P. K. Santra, P. V. Nair, K. G. Thomas and P. V. Kamat, CuInS₂-Sensitized Quantum Dot Solar Cell. Electrophoretic Deposition, Excited-State Dynamics, and Photovoltaic Performance, J. Phys. Chem. Lett., 2013, 4, 722–729.
17. K. Kornhuber, J. Kavalakkatt, X. Lin, A. Ennaoui and M. Ch, Lux-Steiner, In Situ Monitoring of Electrophoretic Deposition of Cu₂ZnSnS₄ Nanocrystals, RSC Adv., 2013, 3, 5845–5850.
18. I. Aguilera, J. Vidal, P. Wahnon, L. Reining and S. Botti, First-Principles Study of the Band Structure and Optical Absorption of CuGaS₂, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 84, 085145.
19. M. Gloeckler and J. R. Sites, Band-gap Grading in Cu(In,Ga)Se₂ Solar Cells, J. Phys. Chem. Solids, 2005, 66, 1891–1894.
20. D. B. Mitzi, M. Yuan, W. Liu, A. J. Kellock, S. J. Chey, V. Deline and A. G. Schrott, A High-Efficiency Solution-Deposited Thin-Film Photovoltaic Device, Adv. Mater., 2008, 20, 3657–3662.
21. T. Dullweber, O. Lundberg, J. Malmstrom, M. Bodegard, L. Stolt, U. Rau, H. W. Schock and J. H. Werner, Back Surface Band Gap Gradings in CuZnGaSe Solar Cells, Thin Solid Films, 2001, 387, 11–13.
22. S. H. Sohn, N. S. Han, Y. J. Park, S. M. Park, H. S. An, D.-W. Kim, B. K. Min and J. K. Song, Band Gap Grading and Photovoltaic Performance of Solution-Processed Cu(In,Ga)S₂ Thin-Film Solar Cells, Phys. Chem. Chem. Phys., 2014, 16, 27112–27118.
23. S. Schleussner, U. Zimmermann, T. Watjen, K. Leifer and M. Edoff, Effect of Gallium Grading in Cu(In,Ga)Se₂ Solar-Cell Absorbers Produced by Multi-Stage Coevaporation, Sol. Energy Mater. Sol. Cells, 2011, 95, 721–726.
24. M. R. Golobostanfard and H. Abdizadeh, Comparing Incorporation of Carbon Nanotubes in Hierarchical Porous Photoanodes of Quantum Dot and Dye Sensitized Solar Cells, Ceram. Int., 2015, 41, 497–504.
25. M. A. Contreras, M. J. Romero, B. To, F. Hasoon, R. Noufi, S. Ward and K. Ramanathan, Optimization of CBD CdS Process in High-Efficiency Cu(In,Ga)Se₂-Based Solar Cells, Thin Solid Films, 2002, 403, 204–211.
26. A. Khanaki, H. Abdizadeh and M. R. Golobostanfard, Effects of Process Parameters on The Synthesis and Characterization of CuIn_1−xGa_xSe₂ Nanopowders Produced by New Modified Solvothermal Method, Mater. Sci. Semicond. Process., 2013, 16, 1397–1404.
27. A. Khanaki, H. Abdizadeh and M. R. Golobostanfard, Electrophoretic Deposition of CuIn_1−xGa_xSe₂ Thin Films Using Solvothermal Synthesized Nanoparticles for Solar Cell Application, J. Phys. Chem. C, 2015, 119, 23250–23258.
28. Y. Qi, Q. Liu, K. Tang, Z. Liang, Z. Ren and X. Liu, Synthesis and Characterization of Nanostructured Wurtzite CuInS₂: A New Cation Disordered Polymorph of CuInS₂, J. Phys. Chem. C, 2009, 113, 3939–3944.
29. B. Koo, R. N. Patel and B. A. Korgel, Wurtzite-Chalcopyrite Polytypism in CuInS₂ Nanodisks, Chem. Mater., 2009, 21, 1962–1966.
30. Y.-H. A. Wang, X. Zhang, N. Bao, B. Lin and A. Gupta, Synthesis of Shape-Controlled Monodisperse Wurtzite CuIn_xGa_1−xS₂ Semiconductor Nanocrystals with Tunable Band Gap, J. Am. Chem. Soc., 2011, 133, 11072–11075.
31. A. D. P. Leach, L. G. Mast, E. A. Hernandez-Pagan and J. E. Macdonald, Phase Dependent Visible to Near-Infrared Photoluminescence of CuInS₂ Nanocrystals, J. Mater. Chem. C, 2015, 3, 3258–3265.
32. M. E. Norako and R. L. Brutchey, Synthesis of Metastable Wurtzite CuInSe₂ Nanocrystals, Chem. Mater., 2010, 22, 1613–1615.
33. H. Matsushita, S. Endo and T. Irie, Raman-Scattering Properties of I-III-VI₂ Group Chalcopyrite Semiconductors, Jpn. J. Appl. Phys., 1992, 31, 18.
34. M. Gusain, P. Kumar and R. Nagarajan, Wurtzite CuInS₂: Solution Based One Pot Direct Synthesis and Its Doping Studies with Nonmagnetic Ga³⁺ and Magnetic Fe³⁺ Ions, RSC Adv., 2013, 3, 18863–18871.
35. D. Papadimitriou, N. Esser and C. Xue, Structural Properties of Chalcopyrite Thin Films Studied by Raman Spectroscopy, Phys. Status Solidi B, 2005, 242, 2633–2643.
36. S. B. Zhang, S.-H. Wei, A. Zunger and H. Katayama-Yoshida, Defect Physics of The CuInSe₂ Chalcopyrite Semiconductor, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 57, 9642–9656.
37. H. S. An, Y. Cho, S. J. Park, H. S. Jeon, Y. J. Hwang, D.-W. Kim and B. K. Min, Cocktails of Paste Coatings for Performance Enhancement of CuInGaS₂ Thin-Film Solar Cells, ACS Appl. Mater. Interfaces, 2014, 6, 888–893.
38. D. Lee and K. Yong, Superstrate CuInS₂ Photovoltaics with Enhanced Performance Using a CdS/ZnO Nanorod Array, ACS Appl. Mater. Interfaces, 2012, 4, 6758–6765.
39. T. Nakada, T. Kume, T. Mise and A. Kunioka, Superstrate-Type Cu(In,Ga)Se₂ Thin Film Solar Cells with ZnO Buffer Layers, Jpn. J. Appl. Phys., 1998, 37, 499–501.
40. T. Dullweber, U. Rau, M. A. Contreras, R. Noufi and H.-W. Schock, Photogeneration and Carrier Recombination in Graded Gap Cu(In,Ga)Se₂ Solar Cells, IEEE Electron Device Lett., 2000, 47, 2249–2254.
41. A. Chirila, S. Buecheler, F. Pianezzi, P. Bloesch, C. Gretener, A. R. Uhl, C. Fella, L. Kranz, J. Perrenoud, S. Seyrling, R. Verma, S. Nishiwaki, Y. E. Romanyuk, G. Bilger and A. N. Tiwari, Highly Efficient Cu(In,Ga)Se₂ Solar Cells Grown on Flexible Polymer Films, Nat. Mater., 2011, 10, 857–861.
42. Y. S. Lim, H.-S. Kwon, J. Jeong, J. Y. Kim, H. Kim, M. J. Ko, U. Jeong and D.-K. Lee, Colloidal Solution-Processed CuInSe₂ Solar Cells with Significantly Improved Efficiency up to 9% by Morphological Improvement, ACS Appl. Mater. Interfaces, 2014, 6, 259–267.

---