Preparation of CuIn1−xGaxSe2 thin films by sputtering and selenization process

CuIn_1−xGa_xSe₂ polycrystalline thin films were prepared by a two-step method. The metal precursors were deposited either sequentially or simultaneously using Cu–Ga (23 at%) alloy and In targets by DC magnetron sputtering. The Cu–In–Ga alloy precursor was deposited on glass or on Mo/glass substrates at either room temperature or 150°C. These metallic precursors were then selenized with Se pellets in a vacuum furnace. The CuIn_1−xGa_xSe₂ films had a smooth surface morphology and a single chalcopyrite phase.     

CuIn1−xGaxSe2-based photovoltaic cells from electrodeposited and chemical bath deposited precursors

We have fabricated 13.7%- and 7.3%-efficient CuIn_1−xGa_xSe₂ (CIGS)-based devices from electrodeposited and chemical bath deposited precursors. As-deposited precursors are Cu-rich films and polycrystalline (grain size is very small) in nature. Only preliminary data is presented on chemical bath deposited precursors. Additional In, Ga, and Se were added to the precursor films by physical evaporation to adjust the final composition to CuIn_1−xGa_xSe₂. Addition of In and Ga and also selenization at high temperature are very crucial to obtain high efficiency devices. Three devices with Ga/(In+Ga) ratios of 0.16, 0.26, and 0.39 were fabricated from electrodeposited precursors. The device fabricated from the chemical bath deposited precursor had a Ga/(In+Ga) ratio of 0.19. The films/devices have been characterized by inductive-coupled plasma spectrometry, Auger electron spectroscopy, X-ray diffraction, electron-probe microanalysis, current-voltage characteristics, capacitance–voltage, and spectral response. The compositional uniformity of the electrodeposited precursor films both in the vertical and horizontal directions were studied. The electrodeposited device parameters are compared with those of a 17.7% physical vapor deposited device.     

Quantification of losses in thin-film polycrystalline solar cells

Comparison of measured solar-cell parameters with calculations for ideal cells is a powerful tool to assist fundamental understanding and to focus on the most effective fabrication procedures. The emphasis here will be on quantitative separation of individual loss mechanisms in polycrystalline thin-film cells based on CdTe, CuInSe₂ (CIS), and related alloys such as CuIn_1−xGa_xSe₂ (CIGS). Several techniques to facilitate separation of losses are described.     

Structural changes of CIGS during deposition investigated by spectroscopic light scattering: A study on Ga concentration and Se pressure

We have studied the three-stage deposition process of CuIn_1−xGa_xSe₂ (CIGS) thin films using spectroscopic light scattering (SLS), under varied deposition conditions. The structural changes of CIGS films by (1) Ga composition, (2) Se supply and (3) low deposition temperature, were observed in situ by SLS. The largest changes in SLS profiles by the Ga composition was observed between x=0.3 and 0.5. The SLS profiles changed significantly during stage 1 by varying the Se pressure, while the temperature profiles did not.     

Losses due to polycrystallinity in thin-film solar cells

The highest efficiency thin-film polycrystalline CuIn_1−xGa_xSe₂ (CIGS) and CdTe solar cells are compared directly with crystalline Si and GaAs cells with similar respective bandgaps. The excess efficiency losses (6.3% for CIGS and 9.9% for CdTe) are quantitatively separated into eight categories. In each case, the impact of polycrystallinity is evaluated, and the differential losses are identified as being directly attributable to polycrystallinity or due to other causes. Approximately, two-thirds of the excess loss for polycrystalline cells is clearly due to polycrystallinity. Thus, strategies such as grain passivation to reduce the impact of polycrystallinity should be examined, but at the same time conventional approaches for incremental improvement should receive attention.     

Electrodeposition of CuInxGa1−xSe2 thin films

CuIn_xGa_1−xSe₂ (CIGS) polycrystalline thin films with various Ga to In ratios were grown using a new two-step electrodeposition process. This process involves the electrodeposition of a Cu–Ga precursor film onto a molybdenum substrate, followed by the electrodeposition of a Cu–In–Se thin film. The resulting CuGa/CuInSe bilayer is then annealed at 600°C for 60 min in flowing Argon to form a CIGS thin film. The individual precursor films and subsequent CIGS films were characterized by X-ray diffraction, scanning electron microscopy, energy dispersive spectroscopy and Auger electron spectroscopy. The as-deposited precursor films were found to be crystalline with a crystal structure matching that of CuGa₂. The annealed bi-layers were found to have the same basic chalcopyrite structure of CuInSe₂, but with peak shifts due to the Ga incorporation. Energy dispersive spectroscopy results show that the observed shifts correlate to the composition of the films.     

Cu(In,Ga)Se2 based photovoltaic structure by electrodeposition and processing

CuIn_1−xGa_xSe₂ (CIGS) thin films were formed from an electrodeposited CuInSe₂ (CIS) precursor by thermal processing in vacuum in which the film stoichiometry was adjusted by adding In, Ga and Se. The structure, composition, morphology and opto-electronic properties of the as-deposited and selenized CIS precursors were characterized by various techniques. A 9.8% CIGS based thin film solar cell was developed using the electrodeposited and processed film. The cell structure consisted of Mo/CIGS/CdS/ZnO/MgF₂. The cell parameters such as J_sc, V_oc, FF and η were determined from I–V characterization of the cell.     

CuInxGa1−xSe2 thin films prepared by electron beam evaporation

CuIn_xGa_1−xSe₂ bulk compound of three different compositions x=0.75, 0.80 and 0.85 have been prepared using individual elements of copper, indium, gallium and selenium. Thin films of CuIn_xGa_1−xSe₂ have been deposited using the prepared bulk by electron beam evaporation method. The structural studies carried on the deposited films revealed that films annealed at 400 °C are crystalline in nature exhibiting chalcopyrite phase. The position of the (1 1 2) peak in the X-ray diffractogram corresponding to the chalcopyrite phase has been found to be dependent on the percentage of gallium in the films. The composition of the prepared bulk and thin films has been identified using energy dispersive X-ray analysis. The photoluminescence spectra of the CuIn_xGa_1−xSe₂ films exhibited sharp luminescence peaks corresponding to the band gap of the material.     

Photoconductivity in CuInSe2 films

CulnSe₂ films with different Cu/In ratios (0.4–1.2) were deposited on glass substrates by three source evaporation techniques. The films were polycrystalline in nature with partially depleted grains. Photoconductivity in the films was measured in the temperature range 170–370 K. The data at low temperatures (T < 300 K) were analyzed by using the grain boundary trapping model with monovalent trapping states. The intercrystalline barrier height in dark varied within 0.13 and 0.08 eV with the variation of Cu/In within 0.4 and 1.18. The variation of the barrier height with the intensity of the incident photons in the range 10–75 mW / cm² was analyzed to understand the carrier detrapping effect in the films under various illumination levels.     

Enhanced electrical properties of pulsed laser-deposited CuIn0.7Ga0.3Se2 thin films via processing control

Polycrystalline CuIn_0.7Ga_0.3Se₂ thin films were prepared on soda-lime glass substrates using pulsed laser deposition (PLD) with various process parameters such as laser energy, repetition rate and substrate temperature. It was confirmed that there existed a limited laser energy, i.e. less than 300 mJ, to get phase pure CIGS thin films at room temperature. Particularly, even at room temperature, distinct crystalline CIGS phase was observed in the films. Crystallinity of the films improved with increasing substrate temperature as evidenced by the decrease of FWHM from 0.65° to 0.54°. Slightly Cu-rich surface with Cu_2−xSe phase was confirmed to exist by Raman spectra, depending on substrate temperature. Improved electrical properties, i.e., carrier concentration of ∼10¹⁸ cm⁻³ and resistivity of 10⁻¹ Ω cm at higher substrate temperature for the optimal CIGS films are assumed to be induced by the potential contributions from highly crystallized thin films, existence of Cu_2−xSe phase and diffusion of Na from substrates to films.     

CuIn1−xGaxSe2-based photovoltaic cells from electrodeposited precursor films

We have developed an electrodeposition bath based on a buffer solution so that the stability of the electrodeposition process is enhanced and no metal oxides or hydroxides precipitate out of solution. The buffer-solution-based bath also deposits more gallium in the precursor films. As-deposited precursors are stoichiometric or slightly Cu-rich CuIn_1−xGa_xSe₂. Only a minimal amount of indium was added to the electrodeposited precursor films by physical vapor deposition to obtain a 9.4%-efficient device.     

Present status and future prospects of CIGSS thin film solar cells

Efficiencies of CuIn_1−xGa_xSe_2−yS_y (CIGSS) modules are comparable to those of lower end crystalline-Si modules. CIGSS layers are prepared by reactive co-evaporation, selenization/sulfurization of metallic or compound precursors, reactive co-sputtering and non-vacuum techniques. CuIn_1−xGa_xS₂ (CIGS2) layers are prepared by sulfurization of Cu-rich metallic precursors and etching of excess Cu_2−xS. Usually heterojunction partner CdS and transparent-conducting bilayer ZnO/ZnO:Al layers are deposited by chemical bath deposition (CBD) or RF magnetron sputtering. CIGSS solar cell efficiencies have been improved by optimizing Cu, Ga and S proportions and providing a minute amount of Na. This paper reviews preparation and efficiency improvement techniques for CIGSS solar cells.     

Impact of substrate roughness on CuInxGa1−xSe2 device properties

In comparison to the traditional use of glass substrates, CuIn_xGa_1−xSe₂ (CIGS) deposited onto metal substrates offers improved device cooling under concentration, economical large-scale roll-to-roll processing, and applicability towards lightweight as well as flexible products. However, unlike glass, metal foils tend to exhibit rough surfaces. This study quantifies the effect of substrate roughness on CIGS device performance. Several substrate types with differing average roughnesses were examined. The frequencies of the surface features contributing to roughness are also examined via several different analytical techniques. Devices were prepared and characterized on steel foils, Mo foil, and glass.     

Synthesis and characterization of one-step electrodeposited CuIn(1−x)GaxSe2/Mo/glass films at atmospheric conditions

CuIn_(1−x)Ga_xSe₂ (CIGS) thin films were prepared by one-step electro-deposition technique from a salt bath coupled with thiocyanate complex electrolytes followed by annealing in argon atmosphere at 300 °C. The influence of deposition reduction potentials as well as the salt concentrations on the structure, morphology, composition and the optical properties were performed. A reproducible Cu-In-Ga-Se precursor layer deposition with consistent composition control was demonstrated. The as-deposited films exhibit an amorphous behavior; however the films displayed good crystallization after annealing. The films show very uniform and dense grain formation with platelet-like nanostructures. The optical properties of the films are modified due to annealing. The electrical conductivity measurements demonstrate that the transport mechanism is influenced by three different temperature regions: the ionization, extrinsic and intrinsic regions, respectively, as found in other semiconductors. However, the annealed films display downturn in conductivity at low temperature indicating that there may be trapping at localized sites or scattering of the free carriers, which may be attributed to the over growth and defect sites. The electro-deposition technique demonstrates promise of growing high-quality CIGS thin films.     

The performance of CuIn1−xGaxSe2-based photovoltaic cells prepared from low-cost precursor films

In this paper we report the 15.4%- and 13.4%-efficient CuIn_1−xGa_xSe₂ (CIGS)-based devices from electrodeposited (ED) and electroless deposited (EL) precursors. The efficiency of the device prepared from electroless precursor film has been improved from 12.4% to 13.4%. The dependence of quantum efficiencies on reverse-bias voltage has been measured for a 15.4%-efficient ED device, 18.8%-efficient physical-vapor-deposited device, and 14.2%-efficient Cd-free device. The purpose of this work is to explore and improve the current collection mechanism.     

A comparative study of CuInSe2 and CuIn3Se5 films using transmission electron microscopy, optical absorption and Rutherford backscattering spectrometry

CuInSe₂ and CuIn₃Se₅ films were grown by stepwise flash evaporation onto glass and Si substrates held at different temperatures. Transmission electron microscopy (TEM) studies revealed that the films grown above 370 K were polycrystalline, with CuInSe₂ films exhibiting larger average grain size than CuIn₃Se₅. Optical absorption studies yielded band gaps of 0.97±0.02 and 1.26±0.02 eV for CuInSe₂ and CuIn₃Se₅, respectively. Rutherford backscattering spectrometry (RBS) study of the films on Si showed that CuInSe₂/Si structures included an inhomogeneous interface region consisting of Cu and Si, whereas CuIn₃Se₅/Si structures presented sharp interface.     

New trends to grow the n-CdZn (S1−xSex)2/p-CuIn(S1−xSex)2 heterojunction thin films for solar cell applications

The n-CdZn(S_1−xSe_x) and p-CuIn(S_1−xSe_x)₂ thin films have been grown by the solution growth technique (SGT) on glass substrates. Also the heterojunction (p–n) based on n-CdZn (S_1−xSe_x)₂ and p-CuIn (S_1−xSe_x)₂ thin films fabricated by same technique. The n-CdZn(S_1−xSe_x)₂ thin film has been used as a window material which reduced the lattice mismatch problem at the junction with CuIn (S_1−xSe_x)₂ thin film as an absorber layer for stable solar cell preparation. Elemental analysis of the n-CdZn (S_1−xSe_x)₂ and p-CuIn(S_1−xSe_x)₂ thin films was confirmed by energy-dispersive analysis of X-ray (EDAX). The structural and optical properties were changed with respect to composition ‘x’ values. The best results of these parameters were obtained at x=0.5 composition. The uniform morphology of each film as well as the continuous smooth thickness deposition onto the glass substrates was confirmed by SEM study. The optical band gaps were determined from transmittance spectra in the range of 350–1000 nm. These values are 1.22 and 2.39 eV for CuIn(S_0.5Se_0.5)₂ and CdZn(S_0.5Se_0.5)₂ thin films, respectively. J–V characteristic was measured for the n-CdZn(S_1−xSe_x)₂/p-CuIn(S_1−xSe_x)₂ heterojunction thin films under light illumination. The device parameters V_oc=474.4 mV, J_sc=13.21 mA/cm², FF=47.8% and η=3.5% under an illumination of 85 mW/cm² on a cell active area of 1 cm² have been calculated for solar cell fabrication. The J–V characteristic of the device under dark condition was also studied and the ideality factor was calculated which is equal to 1.9 for n-CdZn(S_0.5Se_0.5)₂/p-CuIn(S_0.5Se_0.5)₂ heterojunction thin films.     

Annealing effects on Zn1−xMgxO/CIGS interfaces characterized by ultraviolet light excited time-resolved photoluminescence

Annealed Zn_1−xMg_xO/Cu(In,Ga)Se₂ (CIGS) interfaces have been characterized by ultraviolet light excited time-resolved photoluminescence (TRPL). The TRPL lifetime of the Zn_1−xMg_xO/CIGS film increased on increasing the annealing temperature to 250 °C, whereas the TRPL lifetime of the CdS/CIGS film had little change by annealing at temperatures lower than 200 °C. This is attributed to the recovery of physical damages by annealing, induced by sputtering of the Zn_1−xMg_xO film. The TRPL lifetime abruptly decreased with annealing at 300 °C. The diffusion of excess Zn from the Zn_1−xMg_xO film into the CIGS interface is clearly observed in secondary ion mass spectroscopy (SIMS) depth profiles. These results indicate that excess Zn at the vicinity of the CIGS surface acts as non-radiative centers at the interface. The TRPL lifetime of the Zn_1−xMg_xO/CIGS film annealed at 250 °C reached values to be comparable to that of the as-deposited CdS/CIGS film. Performance of the Zn_1−xMg_xO/CIGS cells varied with the annealing temperature in the same manner as the TRPL lifetime. The highest efficiency of the Zn_1−xMg_xO/CIGS solar cells was achieved for annealing at 250 °C. The results of the TRPL lifetime on annealing show that the cell efficiency is strongly influenced by the Zn_1−xMg_xO/CIGS interface states related to the damages and diffusion of Zn.     

Properties of CuIn1−xGaxSe2

Polycrystalline bulk samples of CuIn_1−xGa_xSe₂ weregrown with nominal x = 0.15, 0.25 and 0.5. Mobility, conductivity and band gap were measured at room and low temperatures. Mobilities for x = 0.21 were several hundred cm² V⁻¹s⁻¹ at room temperature and for x = 0.15 were 10³ cm² V⁻¹ s⁻¹, all n type. The band gaps were estimated from the spectra of photoelectrochemical cells at room temperature (with 8.5 K photoluminescence estimates shown in brackets), as 1.10 eV (1.14) for x = 0.21, and 1.07 eV (1.093) for x = 0.15. Crystal mechanical properties as regards cracks were not as good as for CuInSe₂, using similar growth techniques.     

Control of conduction band offset in wide-gap Cu(In,Ga)Se2 solar cells

The effects of conduction band offset of window/Cu(In,Ga)Se₂ (CIGS) layers in wide-gap CIGS based solar cells are investigated. In order to control the conduction band offset, a Zn_1−xMg_xO film was utilized as the window layer. We fabricated CIGS solar cells consisting of an ITO/Zn_1−xMg_xO/CdS/CIGS/Mo/glass structure with various CIGS band gaps (E_g≈0.97–1.43 eV). The solar cells with CIGS band gaps wider than 1.15 eV showed higher open circuit voltages and fill factors than those of conventional ZnO/CdS/CIGS solar cells. The improvement is attributed to the reduction of the CdS/CIGS interface recombination, and it is also supported by the theoretical analysis using device simulation.