VHF-Deposited a-SiC:H Alloys for High-Bandgap Solar Cells: Combining High Voc and Reasonable Stability
- PDF / 505,062 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 15 Downloads / 182 Views
ABSTRACT The material properties of a-SiC:H alloys deposited by VHFGD are studied, with a special emphasis on the effect of hydrogen dilution of the plasma on layer quality. By incorporating these layers into p-i-n solar cells the authors compare layer properties and cell performance. Special attention is paid to the stability of the solar cells against light soaking. Furthermore, the authors show that the insertion of a buffer layer can, also in the case of entirely a-SiC:H cells, lead to a substantial increase of Voc. A reasonable stability of these cells is maintained by an appropriate doping of the buffer layers.
INTRODUCTION The bandgap of a-Si:H is increased by carbon alloying. A-SiC:H. based materials and cells have thus been investigated intensively in the past. While high-voltage a-SiC:H cells have been demonstrated', those cells are still not applied in today's best stable a-Si:H stacked cell modules 2 , because of their insufficient stability. From work on a-Si:H cells, the stability is known to be determined by both the film quality and the device structure. In the present paper we address first the film quality issue in a study combining the proven method of hydrogen plasma dilution with that of VHF plasma excitation, so as to thereby optimise a-SiC:H film properties. In the second part, we apply these films to high-voltage p-i-n cells, emphasising, in particular, the role of p-i buffer layers, which have already been shown to play a crucial role in a-Si:H solar cell stability.
MATERIAL STUDY Experimental Intrinsic layers of amorphous hydrogenated silicon carbon alloys were deposited from silane (SiH 4 ) and methane (CH 4 ) as feedstock gases, in a single chamber reactor using the very high2 frequency glow discharge technique (VHFGD) at 70MHz. The RF input-power of 30mW/cm (for some films, 60mW/cm 2 ) is coupled capacitively to the plasma. The pressure during deposition was always 0.4mbar. Dow Corning 7059 glass and single-crystal Si-wafers (for infrared measurements) were used as substrates. The film thickness was measured using an cc4 1 Step profiler. As a measure for the bandgap energy, we determined E04 [o(XE 0 4 )=10 cm- ] and Etauc (extrapolation of 1r•E-) from the evaluation of visible/UV transmission and reflection spectra. The Urbach energy was determined by PDS. Absolute values were obtained by calibrating the spectra in the high-energy range to the (absolute) absorption spectra from transmission/reflection measurements. The microstructure factor of the samples as well as the hydrogen content were determined evaluating IR absorption spectra. Conductivity measurements were executed using coplanar aluminium contacts; the samples were heated up to 180'C and2 afterwards slowly cooled down. Photoconductivity was measured under 100mW/cm illumination. Hydrogen dilution (dil.=[H 2 ]/([SiH4 ]+[CH 4 ])) was varied from zero dilution to a dilution of 20, and the methane fraction [CH 4 ]/([SiH 4 ]+[CH 4 ]) was varied from 0 to 90%. Gas fluxes were between 5 and 20sccm for SiH 4 and CH 4 , and up to 100
Data Loading...
