An interdigitated back-contacted (IBC) configuration is proposed for submicron copper indium gallium (di)selenide (CIGS). In a modelling platform, the structure was opto-electrically optimized for maximum efficiency. The results are compared with a reference front/back-contacted (FBC) solar cell with similar absorber thickness and exhibiting 11.9% efficiency. The electrical passivation at the front side is accomplished by an Al₂O₃ layer, which is endowed with negative fixed charges. The results indicate that with an optimal geometry and engineered bandgap grading, the efficiency of the new IBC structure can reach 17%. Additionally, with a reasonably low defect density in the absorber layer, efficiencies as high as 19.7% and open-circuit voltage comparable with that of the record solar cell are possible with the IBC structure.

@en

The optical losses associated with sub-micron absorbers in CIGS solar cells can be reduced by light management techniques. 3-D optical modelling was used to optimize light in-coupling and internal rear reflectance in a 750-nm thick CIGS reference solar cell. At the front side, an effective medium approximation (EMA) approach for describing optical properties of a MgF₂-based anti-reflection coating (ARC) was applied. Taking reflectance as the cost function and sequential nonlinear programming as the optimization algorithm, an optimal porous-on-compact double-layer ARC was determined. This led to a wideband light in-coupling with a 6.8% improvement in the photo-current density (J_ph) with respect to the reference solar cell without ARC. Considering the variation of the sunlight direction due to day and seasonal changes, different light incidence angles were investigated. The results indicate that in this case, our designed double-layer ARC outperforms the standard compact MgF₂ single-layer ARC. By using the EMA approach, the amount of computational memory can be reduced by a factor of 30, shortening the simulation time from four days to one hour. At the rear side of the cell, a point-contacted MgF₂/Al₂O₃ reflector, in combination with our proposed front ARC, enhances the J_ph by 11.3% considering the same reference solar cell. Compared to a much thicker cell (1600-nm thick absorber) with no light management applied, our front-and-rear optical approaches more-than-compensate optical losses resulting from using thinner absorbers. This design is suitable for industrial uptake and practical to realize. Additionally, the approach of using EMA for double-layer ARC optimization is innovative with respect to other ARC approaches applicable to not only chalcopyrite photovoltaic technologies.

@en

An optical investigation of ultra-thin CIGS solar cells and guidelines for elimination of optical losses is presented. Then, a novel back contacted structure for CIGS solar cells is suggested and optimized for best implied photocurrent density.@en

Multi-junction solar cells are considered for various applications, as they tackle various loss mechanisms for single junction solar cells. These losses include thermalization and non-absorption below the band gap. In this work, a tandem configuration comprising copper-indium-gallium-di-selenide (CIGS) and hydrogenated amorphous silicon (a-Si:H) absorber layers is studied. Two main challenges are addressed in this work. Firstly, the natural roughness of CIGS is unfavorable for monolithically growing a high quality a-Si:H top cell. Some sharp textures in the CIGS induce shunts in the a-Si:H top junction, limiting the electrical performance of such a configuration. To smoothen this interface, the possibility of mechanically polishing the intermediate i-ZnO layer has been explored. The second challenge that is addressed, is the significant current mismatch in these tandem architectures. To enhance absorption in the current-limiting top cell, the ZnO:Al front electrode was textured by means of wet-etching the entire tandem stack. We demonstrated that one can manipulate the morphology of the random textures by varying the growth conditions of the ZnO:Al, leading to better light management in these devices.

@en

A 3-D optical modelling was calibrated to calculate the light absorption and the total reflection of fabricated CIGS solar cells. Absorption losses at molybdenum (Mo) / CIGS interface were explained in terms of plasmonic waves. To quench these losses, we assumed the insertion of a lossless dielectric spacer between Mo and CIGS, whose optical properties were varied. We show that such a spacer with low refractive index and proper thickness can significantly reduce absorption in Mo in the long wavelength regime and improve the device’s rear reflectance, thus leading to enhanced light absorption in the CIGS layer. Therefore, we optimized a realistic two-layer MgF2 / Al2O3 dielectric spacer to exploit (i) the passivation properties of ultra-thin Al2O3 on the CIGS side for potential high open-circuit voltage and (ii) the low refractive index of MgF2 on the Mo side to reduce its optical losses. Combining our realistic spacer with optically-optimized point contacts increases the implied photocurrent density of a 750 nm-thick CIGS layer by 10% for the wavelengths between 700 and 1150 nm with respect to the reference cell. The elimination of plasmonic resonances in the new structure leads to a higher electric field magnitude at the bottom of CIGS layer and justifies the improved optical performance.@en

Two types of Cu(In,Ga)Se₂ (CIGS) solar cells, both designed for implementation in CIGS modules, were subjected to temperatures between 25C and 105C. Simultaneous exposure to AM1.5 illumination allowed the measurement of their electrical parameters at these temperatures. These two types of solar cells, produced with different deposition routes on soda lime glass (SLG) and polyimide (PI) substrates, showed large variations in the temperature dependency of their electrical parameters. It was shown that the temperature dependency of the open circuit voltage (V_oc) was dependent on its room temperature value: a high V_oc at 25 °C led to a slower loss of V_oc when the temperature was increased. For the V_oc, the normalised temperature dependency varied between -0.28%/°C and -0.47%/°C, which is in agreement with the literature. The temperature dependency of the short circuit current density (J_sc) showed more surprising results: while the PI samples had the expected positive temperature dependency (0.03 to 0.32%/°C), the SLG samples showed a small negative impact of increasing temperature (-0.01 to -0.05%/°C). A correlation between the temperature dependencies of the J_sc and the ideality factor n was observed. Therefore, this difference in the temperature dependence of the J_sc could be caused by increased recombination for the SLG samples. Furthermore, the temperature coefficients of the fill factor (negative), efficiency (negative), and the series (slightly negative) and shunt (negative) resistances were calculated.

@en

Degradation rates and mechanisms for molybdenum back contacts and ZnO:Al front contacts exposed to damp heat were obtained from literature and experiments. It was found that molybdenum back contacts with a higher density and covered by a MoSe₂ film are more stable than their low density, bare counterparts. For ZnO:Al front contacts, various material changes, including thickness, deposition temperatures and doping concentration increase as well as post-deposition treatments can lead to a more stable material. Moreover, the degradation rate of encapsulated and non-encapsulated CIGS solar cells and modules exposed to damp heat were determined It was observed that in general, exposure to damp heat led to a reduction in efficiency, mostly caused by a reduction in Voc and FF. However, it was found that very large differences exist between samples.

@en

The degradation behavior of Mo/MoSe₂ layers have been investigated using damp heat exposure. The two studied molybdenum based films with different densities and microstructures were obtained by lifting off Cu(In,Ga)Se₂ layers from a bilayer molybdenum stack on soda lime glass. Hereby, a glass/Mo/MoSe₂ was obtained, which resembles the back contact as present in Cu(In,Ga)Se₂ solar cells. The samples were degraded for 150 h under standard damp heat conditions and analyzed before, during and after degradation. It was observed that the degradation resulted in the formation of needles and molybdenum oxide layers near the glass/Mo and the Mo/Cu(In,Ga)Se₂ interfaces. X-ray Photoelectron Spectroscopy measurements have shown that the sodium was also present at the surface of the degraded material and it is proposed that the degraded material consists mostly of MoO₃ with intercalated Na⁺. This element has likely migrated from the soda lime glass. This intercalation process could have led to the formation of Na_xMoO₃ ‘molybdenum bronze’ following this redox reaction: xNa⁺ + MoO₃ + xe⁻ ↔ Na_xMoO₃ It is proposed that the formed oxide layer contains Na_xMoO₃ with different Na⁺ contents and different grades of conductivity. This intercalation process can also explain the high mobility of Na⁺ via the grain boundaries in molybdenum. It was also observed that the molybdenum film with a top layer deposited at a high pressure is more susceptible for damp heat degradation.

@en

Copper-indium-gallium-di-selenide (CIGS) is the present record holder in lab-scale thin-film photovoltaics (TF-PV). One of the problems of this PV technology is the scarcity of indium. Multi-junction solar cells allow better spectral utilization of the light spectrum, while the required current generation per layer is much lower, allowing much thinner absorber layers of CIGS. In this contribution we demonstrate working fabricated devices of CIGS bottom cells that are monolithically integrated with a hydrogenated amorphous silicon (a-Si:H) top cell. The proposed structures are a unique fusion of two distinct fabrication methods, being co-evaporation and plasma enhanced chemical vapor deposition (PE-CVD). In addition, devices without any ZnO have been processed. In those cells a nc-SiO_x:H n-layer acted as an electron recipient and lateral insulator for the CIGS p-layer, and a highly p- and n-doped nc-SiO_x:H layer served as the tunnel recombination junction. The top TCO on the a-Si:H cell was varied with ZnO:Al (AZO) and In₂O₃/Sn₂O₃ (ITO). Efficiencies of the not yet optimized devices have reached 7.9% active area efficiency (with V_oc=1.23V, FF=64%, J_sc= 9.95 mA/cm²).

@en

A 'hybrid' degradation setup, which allows the use of humidity, temperature and illumination as loads in order to accelerate degradation of solar cells and modules, has been designed and constructed. In this setup, the current voltage output of photovoltaic samples is automatically logged and the electrical parameters are calculated. This allows the study of the impact of illumination and damp heat induced degradation by in-situ monitoring. Additionally, this setup also allows the determination of temperature dependency of solar cells by a simple procedure.

@en

Non-encapsulated CIGS solar cells with different contents of sodium (Na) and potassium (K) were simultaneously exposed to damp heat and illumination. The solar cells with higher alkali (Na, K) content exhibited higher initial conversion efficiencies, but degraded severely within 100 hours, while samples with a lower alkali content retained their efficiency longer. The degradation was likely caused by the migration of alkali atoms, leading to shunting of the solar cells and the formation of alkali rich spots.

@en

CIGS solar cells were exposed to liquid water purged with the atmospheric gases carbon dioxide (CO2), oxygen (O2), nitrogen (N2) and air in order to investigate their chemical degradation behavior. The samples were analyzed by electrical, compositional and optical measurements before, during and after exposure in order to follow the degradation behavior of these solar cells in time. The solar cells showed a rapid decrease in conversion efficiency when exposed to water purged with a combination of CO2 and N2 as well as to water purged with air, while their efficiency was slowly reduced in unpurged water and water purged with N2 or O2. Cross-section SEM showed that the exposure of samples to H2O with large concentrations of CO2 led to the dissolution of the ZnO:Al layer, likely starting from the grain boundaries. This resulted in an increased series resistance, which is likely related to an increase in resistivity of the ZnO:Al layer. It also led to a very rapid decrease of the short-circuit current of these samples. Therefore, the conversion efficiency was rapidly lost.@en