Losses in CuInSe2-based thin film monolithic tandem solar cells
University of Delaware, Ньюарк, Delaware, United States
DOI: 10.1109/PVSC.2005.1488111 Conference: Photovoltaic Specialists Conference, 2005. Conference Record of the Thirty-first IEEE
Two critical aspects for the development of tandem solar cells using CuInSe2-alloy materials are addressed. First a structure for monolithic tandem solar cells using wide and narrow bandgap Cu(InGa)Se2-alloys in the top and bottom cells is demonstrated. This structure uses an ITO layer as both back contact to the wide bandgap Cu(InGa)Se2 and interconnect to the bottom cell. Three different options for the emitter in the bottom cell are compared. Second, progress on the wide bandgap cell using the Cu(InGa)(SeS)2 system is presented, and different loss mechanisms in Cu(InGa)Se2 and CuInS2 based devices are identified. Efficiency with Cu(InGa)Se2 is reduced by lower optical absorption and poor current collection while with CuInS2, it is reduced by interface recombination.
Available from: Juan Lopez-Garcia
- "For this reason, the replacement of the commonly used opaque Mo back contact by transparent and conductive oxides (TCO) is compulsory. For co-evaporated CIGS and CGS-based thin films solar cells, the use of tin-doped indium oxide (ITO) and aluminium-doped zinc oxide (AZO) as transparent and conductive back contact has been reported      . Theoretical calculations have shown that, under ideal conditions, efficiencies up to 42% could be achieved with a two series junction cell with optimum band gaps of 1.0 and 1.9 eV . "
[Show abstract] [Hide abstract]
ABSTRACT: The development of wide-bandgap materials deposited on transparent conducting oxides for high efficiency and low cost semitransparent photovoltaic devices or tandem cells is becoming important in the last few years. CuIn1−xAlxSe2 (CIAS) thin films with bandgap above 1.95eV for 0.7≤x≤0.9 were deposited onto bare and two different (tin-doped indium oxide, ITO and aluminium-doped zinc oxide, AZO) coated glass substrates by a two stage process consisting of the selenization of metallic precursor layers. Polycrystalline CIAS thin films orientated preferentially along the (112) plane with chalcopyrite structure were obtained. The bandgap energy increased no linearly with the Al addition. X-ray diffractograms showed the coexistence of several CIAS phases. Optical, structural and composition analysis revealed that the ITO and AZO substrates promote not only the incorporation of Se, but also a more homogenous distribution in depth regarding the CIAS samples on bare glass, and acted like a barrier to the oxidation from the glass substrate.
Available from: scholarcommons.usf.edu
[Show abstract] [Hide abstract]
ABSTRACT: Tandem solar cells fabricated from thin films provide promise of improved efficiency while keeping the processing costs low. CdSe as top cell are investigated in this work. CIGS has been a standardized process with lab efficiencies reaching 18% . This dissertation focuses on the development of conductive window layer for the development of a high performance, high bandgap solar cell. ZnSe, Cu2-xSe, and ZnSexTe1-x are investigated as viable window layers of the top cell. ZnSe in undoped form forms a good junction with CdSe films, but the Voc from these devices could never exceed the 360mV mark, while the current densities approached 17.5mA/cm2 .To improve Voc's, the high contact energy at the ZnSe/Cu interface has to be overcome by replacing Cu with a metal having higher work function or doping the window layer to form a tunneling contact with Copper. Deposition of ZnSe from binary sources in presence of nitrogen plasma resulted in films with proper stoichiometry. However, doping could not be accomplished. ZnTe is easily dopable, and was the next alternative. ZnTe doping in presence of Nitrogen plasma resulted in Zn rich films. Hence doping of the ternary compound ZnSexTe1-x was considered. This work focuses on studying the effects of compositional variation on the conductivity of the ZnSexTe1-x films. ZnSexTe1-x films were doped using Nitrogen. Films were deposited by co-evaporation from ZnTe, ZnSe and Se sources. Te/Se ratio was varied by varying the ZnTe thickness and Se Thickness. Films with Zn/Group VI ratio close to 1 were measured for conductivity using IV measurements. Highest conductivity of 2* 10 -8 O-cm was obtained at ZnSe, ZnTe, and Se thicknesses of 2000A, 1500A, and 500A respectively. The actual carrier concentration could be concealed by the current limiting Cu contacts. All films with Zn/Group VI ratio close to 1 showed slight conductivity in the 10-10 O-cm range. Layered ZnSexTe1-x Films doped with Nitrogen had targeted Zn/Group VI ratio of 1, but with a higher Te content. The films were also slightly conductive, in the 10-10 O-cm range. The mechanism limiting the doping in all the films seems to be the same.
[Show abstract] [Hide abstract]
ABSTRACT: Cu(InGa)(SeS)<sub>2</sub> absorber films with a band gap E<sub>g </sub> ges 1.5 eV have been prepared using thermal co-evaporation with different sequences of Cu rich and Cu poor deposition. The composition through the film is, in some cases, non-uniform due to changes in relative S and Se incorporation. The compositional depth profile was characterized by grazing incidence X-ray diffraction and Auger electron spectroscopy with Ar sputtering. The composition ratio [S]/[S+Se] in the films depends on the process and the composition ratio [Cu]/[In+Ga] during deposition. Solar cells with a soda-lime glass/Mo/CIGSS/CdS/ZnO/ITO/(Ni-Al grid) structure were fabricated and a higher FF > 77 % than previous wide bandgap Cu(InGa)(SeS)<sub>2</sub> devices has been demonstrated with these processes. The best Cu(InGa)(SeS)<sub>2</sub> cell with E<sub>g</sub>=1.5 eV is prepared by the two-layer process, and shows an efficiency of 11.2 %, with V<sub>oc </sub>=0.760 V, J<sub>sc</sub>=19.2 mA/cm<sup>2</sup>, and FF=77.3 %. Quantum efficiency of the CIGSS cells shows relatively poor response for wavelength > 600 nm compared to that with CuInS<sub>2</sub>
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.