Analysis of sub-stoichiometric hydrogenated silicon oxide films for surface passivation of crystalline silicon solar cells

Abstract

Thermal stability of passivating layers in amorphous/crystalline silicon (a-Si/c-Si) heterojunction solar cells is crucial for industrial processing and long-term device stability. Hydrogenated amorphous silicon (a-Si:H) yields outstanding surface passivation as atomic hydrogen saturates silicon dangling bonds at the a-Si/c-Si interface. Yet, a-Si surface passivation typically starts to degrade already at annealing temperatures in the range of 200 to 250 °C depending on annealing time, and optical absorption in front layers of a-Si reduces the short circuit current density. We show that oxygen incorporation into a-Si:H films enhances the thermal stability of the passivation and reduces parasitic absorption. We further show that for good passivation of the a-Si/c-Si interface, a compact material structure of the a-Si:O:H films is required where atomic hydrogen is the dominating type of diffusing hydrogen species. For plasma deposited a-Si:O:H films, oxygen incorporation of up to 10 at. % leads to an increase of the optical band gap while the hydrogen concentration is almost constant at approximately 10 at. %. For oxygen concentrations below 3%, the films yield surface recombination velocities as low as 10 cm/s on p-type wafers, and the temperature stability improves by about 50 K compared to pure a-Si:H. For films with relatively low oxygen content, hydrogen effusion spectra and Fourier transform infrared spectroscopy (FTIR) indicate a compact microstructure where only atomic H diffuses. For oxygen concentrations above 3%, the passivation quality reduces and H effusion and FTIR suggest the formation of an open, void-rich material where molecular H2 diffuses. In this case, annealing above 400 °C results in improved interface passivation, presumably due to a densification of the material. Likely, this densification results in an increased density of atomic H, which saturates Si dangling bonds near the c-Si interface.

Amorphous silicon oxide window layers for high-efficiency silicon heterojunction solar cells

“…31,32 Here, for an increasing CO 2 /SiH 4 ratio, a clear decrease of the high-temperature peak signal and a simultaneous increase of the low-temperature peak signal is observed, pointing to increased porosity. Similar results have been reported by Einsele et al 20 With TDS and FTIR measurements, they showed that by adding oxygen to the layer its microstructure is changed significantly. For higher oxygen contents, they observed an increase in the stretching mode at 2100 cm…”

“…Widebandgap doped amorphous silicon oxide layers with nanocrystallites (nc-SiO x :H) are used in thin-film silicon tandem solar cells as window layers and intermediate reflectors 16,17 showing the potential for improved light management. In SHJ devices, intrinsic amorphous silicon oxide (a-SiO x :H) as well as doped nc-SiO x :H layers were tested as passivation and charge-carrier collection layers, respectively, and promising results were reported on p-type [18][19][20] and n-type crystalline silicon wafers. However, a direct comparison of such layers to state-of-the-art a-Si:H films has been lacking so far, making an assessment of their true potential difficult.…”

Nano-composite microstructure model for the classification of hydrogenated nanocrystalline silicon oxide thin films

“…It should be mentioned, that the solubility of H 2 , which is the most favorable configuration of not bonded H that cannot be detected by FTIR, is very low in c-Si. In addition, it is known from a-SiO X :H films that the presence of H 2 is negligible as compared to bonded H, except for a-SiO X :H films with an increased amount of voids that can host H 2 molecules [16]. Additionally, it should be noted that no Si-C related peaks were found in the FTIR spectra of the deposited films, even though CO 2 was used as the oxygen source.…”

Back-Contacted Silicon Heterojunction Solar Cells: Optical-Loss Analysis and Mitigation

“…To further decrease the short-wavelength losses, it is thus required to seek for alternatives to the standard a-Si:H layer. Possible approaches include wide-bandgap materials such as a-SiO x :H [9], [58], indirect bandap materials such as μc-Si:H [59] and perhaps the use of black silicon surfaces passivated by atomiclayer deposited aluminum oxide [60], [61]. Recently, Wan et al [62] also demonstrated PECVD a-SiN x :H layers featuring parasitic absorption less than 0.1 mA/cm 2 and excellent surface passivation.…”