The influence of higher processing temperatures on the formation reaction of Cu(In,Ga)(Se,S)2 thin films using a three step reactive annealing process and on the device performance has been investigated. High process temperatures generally lead to the formation of larger grains, decrease the amount of void formation and their distribution at the back Mo/Cu(In,Ga)(Se,S)2 interface, and lead to a much faster formation reaction that shortens the overall reaction process. However, high temperature processing also leads to a decrease in device performance. A loss in open circuit voltage and fill factor could be attributed to enhanced interface recombination processes for the samples fabricated at higher process temperatures, which itself may be caused by a lack of Na and subsequent poor passivation of interface defect states. The lack of Na resulted in a decrease in free charge carrier concentration by two orders of magnitude.

NaF precursor layers used for providing Na to Cu(In,Ga)Se2 (CIGS) grown on Na-free substrates have been studied. The NaF layers were deposited on top of the Mo back contact prior to the CIGS co-evaporation process. The co-evaporation process was interrupted after the preheating steps, and after part of the CIGS layer was grown. Completed samples were also studied. After the preheating, the NaF layers were analyzed with X-ray Photoelectron Spectroscopy and after growing part and all of the CIGS film, the Mo/NaF/CIGS stack was characterized using transmission electron microscopy (TEM) and secondary ion mass spectrometry (SIMS). The NaF layers were found to be stable in thickness and composition during the pre-heating in selenium containing atmosphere before the CIGS process. The TEM analyses on the partly grown samples show a layer at the CIGS/Mo interface, which we interpret as a partly consumed NaF layer. This is corroborated by the SIMS analysis. In finalized samples the results are less clear, but TEM images show an increased porosity at the position of the NaF layer.

In this work, we investigate the effects of Sb, Bi, or Te interlayers at the Mo/Cu-In-Ga interface on the reaction to form Cu(In,Ga)(Se,S)2 in order to control void formation and improve adhesion. Interlayers with 10 nm thickness were evaporated onto the Mo back contact prior to sputtering the metal precursors. CIGSS absorber layers were formed by a three-step H2Se/Ar/H2S reaction and solar cells were fabricated. The influences of each interlayer were characterized in the precursor and reacted films in terms of the density of the void formation, film structure and morphology, adhesion, and device performance.

We have investigated the migration energy of Cd atom in CuInSe2 (CIS) with a Cu vacancy by first-principles calculations. The activation energy of Cd migration in CIS and migration pathways are obtained by means of the combination of linear and quadratic synchronous transit (LST/QST) methods and nudged elastic band (NEB) method. The theoretical migration energy of Cd atom in CIS is 0.99 eV. The migration energy of Cd atom (Cd→VCu) in CIS is comparable to that of Cu migration (Cu→VCu) in CIS (1.06 eV). This result indicates that Cd diffusion in CIS easily occurs like Cu diffusion.

CIGS solar cells were irradiated with 250 keV electrons, which can create only Cu-related defects in the cell, to reveal the radiation defect. The EL image of CIGS solar cells before electron irradiation at 120 K described small grains, thought to be those of the CIGS. After 250 keV electron irradiation of the CIGS cell, the cell was uniformly illuminated compared to before the electron irradiation and the observed grains were unclear. In addition, the EL intensity rose with increasing electron fluence, meaning the change in EL efficiency may be attributable to the decreased likelihood of non-irradiative recombination in intrinsic defects due to electron-induced defects. Since the light soaking effect for CIGS solar cells is reported the same phenomena, the 250 keV electron radiation effects for CIGS solar cells might be equivalent to the effect.

The tolerance of photovoltaic performances of Cu(In,Ga)Se2-based (CIGSe) solar cells prepared from 3-stage grown absorbers to cadmium sulfide (CdS) buffer layer thickness was investigated. We focus on the influence of the maximum Cu content y = [Cu]/([In]+[Ga]) reached during the co-evaporation process on this tolerance. By increasing the duration of the 2nd stage we varied ymax from 0.93±0.11 up to 1.06±0.12. Although final Cu content and CIGSe surface morphology seem to be similar for all absorbers, the photovoltaic performance of cells with higher maximum Cu content are better; moreover they tolerate much thinner CdS buffers (down to 10 nm-thick) without open circuit voltage or fill factor loss. Cells with lower ymax exhibit more erratic performance and J(V,T) measurements show a specific voltage distribution for thin CdS. From these results it appears possible to decrease the CdS buffer layer thickness if it is deposited on adapted absorbers.

Zn(S,O,OH) Chemical Bath Deposited (CBD) remains one of the most studied Cd-free buffer layer for replacing the CBD-CdS buffer layer in a Cu(In,Ga)Se2-based (CIGSe) solar cells and has already demonstrated its potential to lead to high-efficiencies. However, in order to further increase the deposition rate of the Zn(S,O,OH) layer during the CBD, the inclusion of additives can be a reasonable strategy, as long as the efficiencies of solar cells are maintained. The aim of this work is to understand the effect of the introduction of additives such as hydrogen peroxide (H2O2), H2O2+ethanolamine (C2H7NO) and H2O2+tri-sodium citrate (Na3C6H5O7) during CBD on the deposition mechanism, the growth rate and the quality of the buffer layer. It has been shown that the combined use of H2O2 and citrate in the bath formulation allows the deposition of Zn(S,O,OH) via a mix of “ion-by-ion” and “cluster-by-cluster” mechanisms that have good properties as buffer layers leading to high efficiency solar cells.

Pulsed d.c Magnetron Sputtering (PdcMS) has been investigated for the first time to study the deposition of copper indium gallium diselenide (CIGS) thin films for photovoltaic applications. Pulsing the d.c. in the mid frequency region enhances the ion intensity and enables long term arc-free operation for the deposition of high resistivity materials such as CIGS. It has the potential to produce films with good crystallinity, even at low substrate temperatures. However, the technique has not generally been applied to the absorber layers for photovoltaic applications. The growth of stoichiometric p-type CIGS with the desired electro-optical properties has always been a challenge, particularly over large areas, and has involved multiple steps often including a dangerous selenization process to compensate for selenium vacancies. The films deposited by PdcMS had a nearly ideal composition (Cu0.75In0.88Ga0.12Se2) as deposited at substrate temperatures ranging from no intentional heating to 400 °C. The films were found to be very dense and pin-hole free. The stoichiometry was independent of heating during the deposition, but the grain size increased with substrate temperature, reaching about ∼ 150 nm at 400 °C. Hot probe analysis showed that the layers were p-type. The physical, structural and optical properties of these films were analyzed using SEM, EDX, XRD, and UV-VIS-NIR spectroscopy. The material characteristics suggest that these films can be used for solar cell applications. This novel ion enhanced single step low temperature deposition technique may have a critical role in flexible and tandem solar cell applications compared to other conventional techniques which require higher temperatures.

Sodium plays an important role in the development of device quality CIGS (Cu-In-Ga-Se) and CIGSeS (Cu-In-Ga-Se-S) chalcopyrite thin film solar cells. In this study the effect of location of sodium precursor on the device properties of CIGS solar cells was studied. Reduction in the surface roughness and improvement in the crystallinity and morphology of the absorber films was observed with increase in sodium quantity from 0 Å to 40 Å and to 80 Å NaF. It was found that absorber films with 40 Å and 80 Å NaF in the front of the metallic precursors formed better devices compared to those with sodium at the back. Higher open circuit voltages and short circuit current values were achieved for devices made with these absorber films as well.

CIGS solar cells were fabricated on a hybrid back contact comprised of a TCO layer (ZnO:Ga (GZO)) and Mo layers. It was discovered that an additional Mo layer introduced underneath the TCO layer promotes sodium diffusion through the TCO back contact into the upper CIGS absorber layer. Improvement in VOC and JSC values relative to those of sodium-free solar cells was achieved with the Mo/GZO/Mo hybrid back contact as a result of the enhanced sodium diffusion.

The junction formation when Cu(InGa)Se2 is deposited onto ZnO in a superstrate configuration (glass/window/buffer/Cu(InGa)Se2/contact) is investigated by x-ray photoelectron spectroscopy and analysis of device behavior. When Cu(InGa)Se2 is deposited on ZnO, a Ga2O3 layer is formed at the interface. Approaches to avoid the formation of this unfavorable interlayer are investigated. This includes modifications of the process to reduce the thermal load during deposition and improvement of the thermal stability of the ZnO buffer layer. It was demonstrated that both lowering of the substrate deposition temperature and deposition of the ZnO buffer layer at elevated temperature limits the Ga2O3 formation. The presence of Ga2O3 at the junction does affect the device behavior, resulting in a kink in JV curves measured under illumination. This behavior is absent in devices with limited Ga2O3 formation.

High temperature multi-source co-evaporation has been the most successful approach to fabricate record efficiency Cu(InGa)Se2 devices, yet many groups have been unable to replicate this success when transferring these methods to the Cu2ZnSnSe4 system. The difficulties stem from the dramatic differences in the thermochemical properties which result in decomposition and loss of volatile species, such as Zn and SnSe, at temperatures needed for growth. In co-evaporation, decomposition and element loss must be managed throughout the entire growth process, from the back contact interface to the final terminating surface of the film. The beginning and ending phases of deposition encompass different kinetic regimes suggesting a phased approach to growth may be helpful. A series of depositions with different effusion profiles were used to demonstrate the effects of decomposition during different stages of growth. Secondary phase detection can be challenging in CZTSe, but a combination of SEM imaging and thin cross-section depth profile by EDS were found to best identify and locate the secondary phases that occur during different phases of growth for co-evaporated Cu2ZnSnSe4 films.

Deposition with a uniform incident flux followed by shuttered vacuum cool-down yielded films with a ZnSe phase at the absorber/Mo interface and Cu-rich composition at the surface of the exposed film. Devices from these absorber layers never exceeded conversion efficiencies of 1%. Decomposition at the surface could be prevented by continuing effusion of Se and Sn during the cool-down of the substrate. Resulting films demonstrated more faceted grains as well as significantly improved device performance. Secondary phases that traditionally form at the back contact during the beginning of growth were minimized by decreasing the substrate temperature to 300°C during the initial stages of deposition which reduced the ZnSe formed at the Mo interface. The thermochemical origin of the secondary phases will be discussed and the performance of representative devices will be presented.

The kesterite semiconductor Cu2ZnSnS(e)4 is seen as a suitable absorber layer to replace Cu(In,Ga)Se2 in thin film solar cells, if thin film photovoltaics are to be deployed on the terawatt scale. Currently the best devices, and hence the best kesterite absorber layers are grown away from stoichiometry and are zinc rich and copper poor, presumably leading to the formation of ZnS(e). However, it has been shown that secondary phases present in an absorber layer reduce device performance. If growth in Zn rich conditions seems to be mandatory, then any secondary phases formed should be grown on the surface of the absorber layer so that they may be easily removed by etching. Therefore, it is important to know how and why secondary phases form, and if possible, how to segregate them to the surface of the absorber layer.

Here we show that ZnSe is formed at the initial stages of absorber formation from annealing metal stacks in selenium vapor. Further we demonstrate that the way the precursor metals are distributed on the substrate leads to different absorber layer performances in full devices. The importance of selenium vapor pressure is highlighted in respect to the order of selenisation of the metals, Zn before Cu. Additionally, the importance of selenium and tin selenide vapor pressure during annealing is reviewed with regard to avoiding a decomposition of the Cu2ZnSnSe4 to ZnSe and Cu2Se phases. Regardless of the atmosphere above the absorber, the reaction of the absorber with molybdenum appears unavoidable without the use of a passivation strategy. Counter-intuitively, it is demonstrated that for our absorber layers grown under Zn-rich conditions, removal of the ZnSe is harmful for device performance.

The vibrational properties of kesterite Cu2ZnSnS4 (CZTS) single crystals were studied by polarization-dependent Raman scattering measurements. The CZTS crystals grown by chemical vapor transport technique using iodine trichloride as a transport agent consist of several mirror-like planes. The detailed analysis of the experimental spectra obtained from different planes allows determining the symmetry assignment of the observed Raman-active modes. The wavenumber values of Raman-active modes are compared with the results of recent theoretical calculations. The presented data are useful for examination of CZTS absorber films applied for solar cells to clarify the existence of structural or phase inhomogeneities.

This work deals with the influence of sodium on the properties of CZTSSe material and solar cells. For that purpose, two types of substrates are compared, one with low sodium content (borosilicate glass), the other one with higher sodium content (soda-lime glass). In each case the Na-content in the CZTSSe passing from the substrate through the Mo back contact is quantified by secondary ion mass spectroscopy analysis. Photoluminescence spectroscopy indicates that better quality material is achievable when increasing the Na-content in the CZTSSe. The material characterization results are compared to the photovoltaic properties. Index Terms — Cu2ZnSn(S1-xSex)4, CZTSSe, CZTS, CZTSe, Sodium, Kesterite, thin film, solar cell.

Quaternary semiconductors, Cu2ZnSnS4 and Cu2ZnSnSe4 which contain only earth-abundant elements, have been considered as the alternative absorber layers to Cu(In,Ga)Se2 (CIGS) for thin film solar cells although CIGS-based solar cells have achieved efficiencies over 20 %. In this work we report an air-stable route for preparation of Cu2ZnSn(Sx,Se(1-x))4 (CZTSSe) thin film absorbers by a solution process based on the binary and ternary chalcogenide nanoparticle precursors dispersed in organic solvents. The CZTSSe absorber layers were achieved by spin coating of the ink precursors followed by annealing under Ar/Se atmosphere at temperature up to 580°C. We have investigated the influence of the annealing temperature on the reduction or elimination of detrimental secondary phases. X-ray diffraction combined with Raman spectroscopy was utilized to better identify the secondary phases existing in the absorber layers. Solar cells were completed by chemical bath deposited CdS buffer layer followed by sputtered i-ZnO/ZnO: Al bi-layers and evaporated Ni/Al grids.

Large-area Cu2ZnSnS4 (CZTS) thin films were deposited by low-cost spray pyrolysis technique on Mo-coated soda-lime glass (SLG) substrates at varied substrate temperatures of 563-703°K. Deposition conditions were optimized to obtain best quality films and effect of post deposition thermal processing of the as-deposited films under H2S ambient were investigated. Structural, morphological, and compositional characterization of as-deposited and H2S treated CZTS absorber layers were carried out by x-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy (SEM) and energy dispersive x-ray analysis (EDX). Optical and electrical properties were measured by UV-Vis spectroscopy, van der Pauw, and Hall-effect measurements. Films grown at ∼360°C substrate temperature showed superior optoelectronic properties, improved stoichiometry and smoother morphology compared to films grown at much higher or lower temperatures. Film properties were significantly improved after the H2S processing. Our results show that large area high quality CZTS films can be fabricated by low-cost spray pyrolysis technique for high throughput commercial production of CZTS based heterojunction solar cells.

Cu2ZnSnSe4 p-type semiconductors currently investigated for use in thin film solar cells can be synthesized by firstly depositing a metallic precursor and secondly annealing the precursor in selenium vapor. Differently stacked Cu-Sn-Zn metallic precursors were characterized after a soft annealing at 350°C under nitrogen atmosphere. For the stack where the Sn and Zn were in direct contact with sufficient Cu to form a stable alloy, a bi-layered structure consisting of Cu-Sn on the bottom and Cu-Zn on the top was formed. Contrarily, when Zn was not in direct contact with Cu, the metals diffused to form a stable alloy and the system segregates horizontally, forming a mixed columnar structure. These two types of precursors were selenized under exactly the same conditions to form kesterite absorbers for solar cell devices. Using this approach the improvement from 0.44% power conversion efficiency for the bi-layered precursor to 4.5% for the mixed precursor was achieved.

Fundamental aspects of (electro-)luminescence of Cu(In,Ga)Se2 solar cells and modules are investigated by means of spectrally and spatially resolved measurements. The validity of the reciprocity relation between spectrally resolved electroluminescence emission and photovoltaic quantum efficiency is verified for the case of industrially produced ZnO/CdS/Cu(In,Ga)Se2 heterojunction solar cells. Further we find that photo- and electroluminescent emission in these devices obey a superposition principle only in a limited range of the applied electrical or illumination bias. This range depends on the light soaking history of the sample and extends up to an injected current density of approximately 15 mAcm-2 after 3 h of light soaking at a temperature of 400 K. In the state prior to light soaking this range is limited to 4 mAcm-2. At higher bias, a characteristic discrepancy between electroluminescence and electro-modulated photoluminescence appears. We attribute this anomaly to a potential barrier behavior close to the CdS/ Cu(In,Ga)Se2 interface. Metastable defect reactions induced by holes injected into the space charge region partly reduce this barrier. We further find that the luminescence efficiency is enhanced by a factor of 3 by light soaking at 400 K. Spatially resolved electroluminescence measurements conducted during application of voltage or current bias at ambient temperature in the dark are qualitatively compatible with the conclusions drawn from the spectrally resolved measurements.

The purpose of the present paper is to focus on the impact of oxygen gas partial pressure during the sputtering of i-ZnO and ZnMgO on the transient behavior of solar cells parameters when a CBD-ZnS buffer layer is used. Based on electrical characterization of cells, we have observed that the effect of light-soaking is different on J-V characteristics depending on the quantity of oxygen present during the first deposition time of the i-ZnO or ZnMgO layers. In fact, we have noticed that, when cells are prepared with standard i-ZnO, the efficiencies are very low and a pronounced transient behavior is observed. However, when the i-ZnO or ZnMgO is first formed by a few nanometers sputtered layer without any additive oxygen, depending on the thickness of this layer, the transient effects strongly decrease. It is then possible to reach efficiencies quite similar to the CdS reference cells, especially with ZnMgO, without any post-treatments.