ABSTRACTIn recent years, zinc oxide has been investigated as a front electrode material in hydrogenated amorphous silicon/hydrogenated microcrystalline silicon (a‐Si:H/µc‐Si:H) tandem solar cells. Such as for other transparent conducting oxide materials and applications, a proper balancing of transparency and conductivity is necessary. The latter is directly related to the density and the mobility of charge carriers. A high density of charge carriers increases conductivity but leads to a higher absorption of light in the near‐infrared part of the spectrum due to increased free‐carrier absorption. Hence, the only way to achieve high conductivity while keeping the transparency as high as possible relies on an increase of carrier mobility. The carrier density and the mobility of sputtered Al‐doped zinc oxide (ZnO:Al) can be tailored by a sequence of different annealing steps. In this work, we implemented such annealed ZnO:Al films as a front electrode in a‐Si:H/µc‐Si:H tandem solar cells and compared the results with those of reference cells grown on as‐deposited ZnO:Al. We observed an improvement of short‐circuit current density as well as open‐circuit voltage and fill factor. The gain in current density could be attributed to a reduction of both sub‐band‐gap absorption and free‐carrier absorption in the ZnO:Al. The higher open‐circuit voltage and fill factor are indicators of a better device quality of the silicon for cells grown on annealed ZnO:Al. Altogether, the annealing led to an improved initial conversion efficiency of 12.1%, which was a gain of +0.7% in absolute terms. Copyright © 2013 John Wiley & Sons, Ltd.

The interface between solid-phase crystallized phosphorous-doped polycrystalline silicon (poly-Si(n(+))) and aluminum-doped zinc oxide (ZnO:Al) was investigated using spatially resolved photoelectron emission microscopy. We find the accumulation of aluminum in the proximity of the interface. Based on a detailed photoemission line analysis, we also suggest the formation of an interface species. Silicon suboxide and/or dehydrated hemimorphite have been identified as likely candidates. For each scenario a detailed chemical reaction pathway is suggested. The chemical instability of the poly-Si(n(+))/ZnO:Al interface is explained by the fact that SiO2 is more stable than ZnO and/or that H2 is released from the initially deposited a-Si:H during the crystallization process. As a result, Zn (a deep acceptor in silicon) is "liberated" close to the silicon/zinc oxide interface presenting the inherent risk of forming deep defects in the silicon absorber. These could act as recombination centers and thus limit the performance of silicon/zinc oxide based solar cells. Based on this insight some recommendations with respect to solar cell design, material selection, and process parameters are given for further knowledge-based thin-film silicon device optimization.

Zinc oxide is widely used as transparent contact in thin film solar cells. We investigate the damp heat stability of aluminum doped ZnO (ZnO:Al) films sputter deposited at different conditions. Increase in resistivity upon damp heat exposure was observed for as-deposited ZnO:Al films and the water penetration was directly linked to this degradation. Deuterium was used as isotopic marker to identify the amount of water taken up by the films. Finally, we applied a special annealing step to prepare highly stable ZnO:Al films with charge carrier mobility of 70cm2/Vs after 1000h of damp heat treatment. A grain boundary reconstruction model is proposed to explain the high stability of ZnO:Al films after annealing.

Hard X-ray photoelectron spectroscopy (HAXPES) is used to identify chemical interactions (such as elemental redistribution) at the buried silicon/aluminum-doped zinc oxide thin-film solar cell interface. Expanding our study of the interfacial oxidation of silicon upon its solid-phase crystallization (SPC), in which we found zinc oxide to be the source of oxygen, in this investigation we address chemical interaction processes involving zinc and aluminum. In particular, we observe an increase of zinc- and aluminum-related HAXPES signals after SPC of the deposited amorphous silicon thin films. Quantitative analysis suggests an elemental redistribution in the proximity of the silicon/aluminum-doped zinc oxide interface - more pronounced for aluminum than for zinc - as explanation. Based on these insights the complex chemical interface structure is discussed. (C) 2013 Elsevier B.V. All rights reserved.

TCO films are crucial components of almost all thin-film solar cells and a-Si:H/c-Si heterojunction solar cells. As they are used as front contacts, the requirements for electrical conductivity and optical trnamission are generally very high. Further restrictions are imposed onto the deposition process by the cell manufacturing process, in which e. g. the maximum substrate temperature can be limited. In this paper the optimization of ZnO:Al layer deposited by magnetron sputtering to different solar cells is discussed. For a-Si: H/c-Si heterojunction solar cells the advantages and limitations of different variations of magnetron sputtering of ZnO: Al are discussed and compared to standard ITO deposition. For a-Si:H/mu c-Si:H the usage of post-deposition treatments to improve the optical and electrical performance is briefly discussed.

The electronic structure of the interface between the boron-doped oxygenated amorphous silicon “window layer” (a-SiOx:H(B)) and aluminum-doped zinc oxide (ZnO:Al) was investigated using hard x-ray photoelectron spectroscopy and compared to that of the boron-doped microcrystalline silicon (μc-Si:H(B))/ZnO:Al interface. The corresponding valence band offsets have been determined to be (−2.87 ± 0.27) eV and (−3.37 ± 0.27) eV, respectively. A lower tunnel junction barrier height at the μc-Si:H(B)/ZnO:Al interface compared to that at the a-SiOx:H(B)/ZnO:Al interface is found and linked to the higher device performances in cells where a μc-Si:H(B) buffer between the a-Si:H p-i-n absorber stack and the ZnO:Al contact is employed.

For thin film silicon solar cells with high conversion efficiency and low production costs transparent conducting oxide (TCO) substrates with a high performance are required. TCO layers must have excellent conductivity and high transparency over a wide range in the VIS/NIR wavelength spectrum. Moreover, an advanced light trapping scheme has to be implemented, generally obtained by TCO surface texture. For low production costs it is favorable to keep the TCO and silicon layers as thin as possible. We present methods that accomplish all of these requirements. Tandem solar based on amorphous and microcrystalline silicon deposited on DC magnetron-sputtered zinc oxide (ZnO:Al) coated substrates, which have been capped with silicon and subsequently annealed at 500°C, result in an efficiency improved by 0.4% reaching 12.4%. This is due to the increased carrier mobility, reaching up to 70cm2/Vs, and improved transparency of the ZnO:Al. The high TCO conductivity allows for reducing the layer thickness significantly. In order to demonstrate the potential of combining this improved TCO with alternative light scattering concepts, we show results of a tandem cell on substrates comprising ZnO:Al sputtered on sol–gel coated light-scattering layers with dielectric particles having excellent light trapping properties. Combining both methods will allow for thin film silicon solar cells with high efficiency and potentially low production costs.

Optical methods are powerful and non-destructive means to characterize highly doped transparent conducting oxide thin films. In order to describe the optical properties of high-mobility ZnO films we present a dielectric function composed of different analytic expressions to describe the different contributions to the dielectric function of the films. This allows for the correct description of measured optical spectra and reduces the complex functions to a set of fitting parameters. In a second step we compare the obtained parameters to theoretical models. The basic theories are nicely reproduced and the basic link between optical and electrical properties can be understood. The findings can help on the route to a complete presiction of optical properties from the basic material properties or vice versa.

To improve electrical properties a high temperature annealing treatment was applied to several transparent conductive oxides (TCO), namely tin doped indium oxide (ITO), Ga- or Al- doped ZnO (ZnO:Al/Ga), ion beam assisted deposited (IBAD) ZnO:Ga and Ga doped zinc magnesium oxide (ZnMgO:Ga). All these films were grown by magnetron sputtering. During the annealing process all TCO films were capped with 50 nm of amorphous silicon in order to protect the films from environmental impact. Increase in mobility up to 72 cm2/Vs and low resistivity of 1.6 × 10−4 Ωcm was achieved for ZnO:Al after annealing at 650°C for 24 h. Independent of the deposition conditions and doping or alloying material almost all ZnO based films show a consistent improvement in mobility. Also for ITO films a decrease in resistivity with partially improved mobility was found after annealing. However, not all ITO films show consistent improvement, but carrier density above 1021 cm−3 while ZnO films show no clear trend for carrier density but a remarkable increase in mobility. Thus we propose the healing of defects and the activation of donors to be most significant effects for ZnO and ITO films, respectively.

The chemical and electronic properties of a-Si:H(B)/ZnO:Al and µc-Si:H(B)/ZnO:Al thin-film solar cell structures are studied by hard X-ray photoelectron spectroscopy (HAXPES). Using a combination of different X-ray excitation energies and deliberate sample design, we were able to select the probed volume, i.e., the silicon capping layer only or the silicon and zinc oxide layer (including the buried interface). For the a-Si:H(B) material, we find a higher deposition rate and a smaller value for the modified Auger parameter than for µc-Si:H(B). In addition, we find indications of a pronounced band bending limited to the very surface of the a-Si:H(B) and the µc-Si:H(B) layers, which is more distinct in the latter case.

Thermal post deposition treatments are applied to DC-sputtered aluminum-doped zinc oxide (ZnO:Al) films and lead to a significant improvement of the electrical properties. Protective layers of amorphous silicon are used to protect the films from degradation during the high temperature treatment. Annealing for 6 hours at 500°C leads to a carrier mobility of 48cm2/Vs at a carrier concentration of 5.5·1020cm−3. Furthermore, improvements in the optical as well as in the electrical properties are possible at the same time compared to the as-deposited film. This is achieved by carrying out two thermal treatments to the ZnO:Al film, one prior to the capping with the protective layer and one afterwards. A series of samples with different carrier concentrations allows us to draw conclusions on the specific electrical transport properties.

The chemical structure of the interface between silicon thin films and the transparent conductive oxide ZnO:Al has been investigated by hard x-ray photoelectron spectroscopy. By varying the excitation energy between 2010 and 8040 eV, we were able to probe the Si/ZnO interface buried below 12 nm Si. This allowed for the identification of changes induced by solid phase crystallization (SPC). Based on in-situ SPC annealing experiments, we find clear indications that the formation of Si–O bonds takes place at the expense of Zn–O bonds. Hence, the ZnO:Al acts as the oxygen source for the interfacial Si oxidation.

A postdeposition thermal treatment has been applied to sputtered Al-doped zinc oxide films and shown to strongly decrease the resistivity of the films. While high temperature annealing usually leads to deterioration of electrical transport properties, a silicon capping layer successfully prevented the degradation of carrier concentration during the annealing step. The effect of annealing time and temperature has been studied in detail. A mobility increase from values of around 40 cm2/Vs up to 67 cm2/Vs, resulting in a resistivity of 1.4×10−4 Ω cm has been obtained for annealing at temperatures of 650 °C. The high mobility increase is most likely obtained by reduced grain boundary scattering. Changes in carrier concentration in the films caused by the thermal treatment are the result of two competing processes. For short annealing procedures we observed an increase in carrier concentration that we attribute to hydrogen diffusing into the zinc oxide film from a silicon nitride barrier layer between the zinc oxide and the glass substrate and the silicon capping layer on top of the zinc oxide. Both are hydrogen-rich if deposited by plasma-enhanced chemical vapor deposition. For longer annealing times a decrease in carrier concentration can occur if a thin capping layer is used. This can be explained by the deteriorating effect of oxygen during thermal treatments which is well known from annealing of uncapped zinc oxide films. The reduction in carrier concentration can be prevented by the use of capping layers with thicknesses of 40 nm or more.

The chemical interface structure between phosphorus-doped hydrogenated amorphous silicon and aluminum-doped zinc oxide thin films is investigated with soft x-ray emission spectroscopy (XES) before and after solid-phase crystallization (SPC) at 600 °C. In addition to the expected SPC-induced phase transition from amorphous to polycrystalline silicon, our XES data indicates a pronounced chemical interaction at the buried Si/ZnO interface. In particular, we find an SPC-enhanced formation of Si–O bonds and the accumulation of Zn in close proximity to the interface. For an assumed closed and homogeneous SiO2 interlayer, an effective thickness of (5±2) nm after SPC could be estimated.

We have grown epitaxial grain boundaries of CuGaSe2 by metal organic vapour phase epitaxy onto a GaAs substrate containing a Sigma 3 grain boundary. SEM micrographs show a dense grain boundary. TEM micrographs prove that the grain boundary in the film is the direct continuation of the grain boundary in the substrate. HRTEM shows that the grain boundary in the film is a twin as well and thus a Sigma 3 boundary. Thus, by using a Sigma 3 grain boundary in the cubic GaAs substrate as a template a Sigma 3 grain boundary is obtained in the tetragonal CuGaSe2 film. Kelvin Probe Force Microscopy gives no indication of a space charge around this grain boundary, while in Hall measurements a small barrier of a few 10 meV is evident. This is an experimental indication for the existence of neutral grain boundaries as predicted theoretically. (C) 2007 Elsevier B.V. All rights reserved.

Single grain boundaries in CuGaSe2 have been grown epitaxially. Hall measurements indicate a barrier of 30-40 meV to majority carrier transport. Nevertheless, local surface potential measurements show the absence of space charge around the grain boundary; i.e., it is neutral. Theoretical calculations [Persson and Zunger, Phys. Rev. Lett. 91, 266401 (2003)] have predicted a neutral barrier for the present Sigma3 grain boundary. Thus, we have experimentally shown the existence of a neutral grain-boundary barrier, however, smaller than theoretically predicted.