In the conventional three-stage co-evaporation process to grow Cu(In,Ga)Se2 (CIGS) film, a large grain is achieved by the co-evaporation of Cu and Se on (In,Ga)2Se3 layer at 550 °C in the second stage and then a p-type is achieved by the co-evaporation of In, Ga, and Se in the third-stage. We reported a new process where a CIGS film with a large gain and p-type is achieved by evaporation of Cu only in the second stage at 400 °C and by the Se annealing in the third stage. In the new process, thermal budget was lowered and the third-stage co-evaporation process was eliminated. It was found that the CIGS gain size increased when the Cu/(In + Ga) ratio was above 0.7 and an addition thin CIGS layer appeared on the CIGS surface. The reaction path with Cu was described in the Cu-In-Se ternary phase diagram. The cell conversion efficiency increased from 9.6 to 15.4% as the Se annealing temperature increased from 400 to 550 °C in the third stage, mainly due to the increase of open-circuit voltage and fill factor. Our process demonstrated a new route to grow a CIGS film with a less thermal budget and simpler process in the co-evaporation process. 相似文献
In this study, Cu(In,Ga)(Se,S)2 (CIGSS) thin films were deposited onto a bi-layer Mo coated soda-lime glass by co-sputtering a chalcopyrite Cu(In,Ga)Se2 (CIGS) quaternary alloy target and an In2S3 binary target. A one-stage annealing process was performed to form CIGSS chalcopyrite phase without post-selenization. Experimental results show that CIGSS films were prepared by the proposed co-sputter process via CIGS (70 W by radio frequency) and In2S3 (30 W by direct current) with a substrate temperature of 373 K, working pressure of 0.67 Pa, and one-stage annealing at 798 K for 30 min. The stoichiometry ratios of the CIGSS film were Cu/(In + Ga) = 0.92, Ga/(In + Ga) = 0.26, and Se/(S) = 0.49 that approached device-quality stoichiometry ratio (Cu/(In + Ga) < 0.95, Ga/(In + Ga) < 0.3, and (Se/S) ≈ 0.5). The resistivity of the sample was 14.8 Ω cm, with a carrier concentration of 3.4 × 1017 cm−3 and mobility of 1.2 cm2 V−1 s−1. The resulting film exhibited p-type conductivity with a double graded band-gap structure. 相似文献
Local current mapping and surface potential distributions on polycrystalline Cu(In,Ga)Se2 (CIGS) films are investigated by conductive atomic force microscopy and Kelvin probe force microscopy. The two kinds of samples fabricated by co-evaporation had extremely different conversion efficiencies of 10% and 0.2% for stoichiometric and Cu- and Se-deficient compositions, respectively. We examined the microscopic reasons for the differences in the local electrical properties. Current mapping and current–voltage behaviors were measured at intragrain regions (IGs) and grain boundaries (GBs). Electronic transport between a Pt scanning probe and the CIGS layer is explained by the Schottky conduction mechanism. The surface potential distribution shows an intriguing relation with topological variation, inferring that a local built-in potential is possibly formed on positively charged GBs. The surface potential is about 100 mV, which shows energy band bending near GBs in the films. Exciton separation near GBs is explained by the bending of the conduction and valence bands, which is sensitive to compositional and structural inhomogeneities. 相似文献
Decreasing the absorber layer thickness of thin‐film solar cells can be an effective solution for cost reduction of photovoltaic electricity generation. Unfortunately, this reduction leads to detrimental effects such as incomplete photon absorption and increased charge carrier recombination at the rear electrode. To tackle these losses in ultra‐thin 0.5 µm Cu(In,Ga)Se2 (CIGS) solar cells, we developed different passivation structures made of MgF2 and Al2O3 at the molybdenum–CIGS interface, leading to localized back contacts. The influence of the distance between those contacts on the cell performance was studied by varying the periodicity of the applied 1D patterns from 6 μm to 30 μm. Thus, an increase in performance was measured for microstructured layers with a periodicity of up to 12 µm. More precisely, a MgF2 layer yielded an increase in power conversion efficiency (PCE) of up to 9%rel compared to an unpassivated cell design, and a passivation layer comprising Al2O3 led to up to a 5%rel increase in PCE. The gains were primarily attributed to an increased reflectivity of the back contact, while the formation of a negative backside field in the case of Al2O3 might have contributed to this increase by preventing electrons from recombining at the backside interface. Our findings indicate a high lateral conductivity for holes inside the multicrystalline CIGS compound over few tens of micrometres, which allows an independent design of future back contacts and light‐trapping schemes.
False‐colour scanning electron microscopy cross‐section picture of a passivated solar cell, with the front contact layers coloured in green, the 0.5 µm CIGS absorber in dark red, the MgF2 passivation layer in blue, and the Mo back contact in grey. 相似文献
Cu(In,Ga)Se2 (CIGS) solar cells were fabricated by varying the film thickness of the cracker-ZnS (c-ZnS) buffer layer from 0 nm to 20 nm, and performance was found to depend on c-ZnS film thickness. The best cell efficiency of approximately 8% was obtained from the CIGS solar cell with an 8 nm thick-c-ZnS buffer layer. To investigate the primary factor to determine the cell performance, we utilized the impedance spectroscopy (IS) reflecting interface qualities, and capacitance-voltage (CV) profiling sensitive to bulk properties. In IS results, an equivalent circuit model including the resistance and capacitance was proposed to interpret cell performance, and carrier lifetime was obtained in connection with recombination probability at p-n junction. In CV profiling, the carrier concentration in the CIGS bulk, the depletion width, and the charge distribution related to the defect states along the depth direction were evaluated. The formation mechanism of c-ZnS buffer layer is suggested by measuring the chemical states, which is closely associated with the IS and CV results. The depletion width substantially increased at c-ZnS film thickness more than 15 nm due to the diffusion of Zn atoms toward CIGS layer, resulting in negative influence on cell performance. From this study, we demonstrated that IS and CV profiling are complementary analysis tools for interpretation of the solar cell operation concerning the interface and bulk properties. 相似文献
Sn-based thin films as new buffer layer for Cd-free Cu(In,Ga)Se2 (CIGS) solar cells were developed. The Sn(O,S)2 films were formed on CIGS substrates by chemical bath deposition from an alkaline ammonia solution by reacting tin(IV) chloride with thiourea. Optimization of the growth process allowed the smooth and conformal coverage of the films on the CIGS substrates with a thickness of 20 nm that was a self-limited thickness in the chemical bath deposition process. XPS analysis revealed that the as-deposited films contained Sn–O, Sn–OH, and Sn–S bondings and the ratio of Sn–S bonding to Sn–O bonding was 0.3. The CIGS solar cell fabricated with a 20-nm thick Sn(O,S)2 buffer layer had the best efficiency of 11.5% without AR coating. The open circuit voltage, short circuit current, and fill factor were 0.55 V, 34.4 mA/cm2, and FF = 0.61, respectively. The open circuit voltage and fill factor were low compared to the conventional CIGS solar cell with a 50-nm thick CdS buffer due to too thin Sn(O,S)2 buffer layer. 相似文献
The present work is an attempt to prepare well defined surfaces of Cu(In,Ga)Se2 (CIGS) thin films in order to answer to basic questions about the relationship between bulk and surface composition. The approach is to use an oxidative etch with an aqueous bromine solution, known to lead to specular surfaces. The CIGS surface is then analyzed by mechanical profilometry, SEM and XPS, allowing for determination of the surface roughness and the nature of surface species. After short time bromine etch, a Se0 film is formed on the CIGS surface which can be completely removed by KCN treatment, leaving a CIGS specular surface. An highlight result is that under specific conditions, the surface composition is close to the stoichiometry of the Cu(In,Ga)3Se5 copper deficient phase. This is the first time that such a study is conducted on technology relevant thin polycrystalline CIGS film. It is expected that the method described will help conducting experiments (e.g. Angle resolved XPS, SIMS, etc.) with an improved resolution. 相似文献
The composition of Cu(In,Ga)Se2 (CIGS) films employed in CIGS solar cells is Cu deficient. There can be point defects, including Cu vacancies, Se vacancies, and metal anti-site defects. The surface composition and defects are not well controlled right after CIGS film fabrication with a three-stage co-evaporation process. This fabrication technique can result in a large variation in cell efficiency. In order to control the CIGS film in a reproducible way, we annealed the CIGS film in air, S, or Se. With this annealing procedure, the Cu content of the CIGS surface was significantly reduced and Ga content was strongly increased. An intrinsic CIGS layer with a lower valence-band maximum and a wider ban gap was formed at the surface. By annealing the CIGS film, the open-circuit voltage and fill factor were significantly improved, which indicates that the surface intrinsic layer acts as a hole-blocking layer so that the surface recombination rate is suppressed. In addition to CIGS film annealing, with subsequent annealing of the completed devices using rapid thermal annealing, the efficiency and reproducibility of CIGS solar cells were markedly improved. 相似文献