(Color online) Structure of SHJ Voc -sample on an n-type c-Si wafer. Such Voc-samples were used for all investigations in this publication.
(Color online) Dependence of the Voc , ext and Voc , impl measurements on the doping concentration of the a-Si:H(p) layer in an a-Si:H(i+p) layer stack. Lines are guides to the eye.
External/internal Voc-ratio ζ (Voc , ext /Voc , impl ) in dependence of the doping concentration of the a-Si:H(p) layer. The Voc-ratio ζ between the increase of the doping concentration and the decrease of the interface passivation is best in the region around 2000 ppm where the ratio first reaches a saturation level. Since the Voc , ext cannot be higher than the Voc , impl the ratio of 1 is the theoretical limit. The line is a guide to the eye.
(Color online) Schematic energy band diagram of an a-Si:H(p) emitter on a c-Si(n) wafer structure under illumination. Illustrated are the electron affinities χa-Si , χc-Si , the band off-sets ΔEC , ΔEV , the built-in voltage Vbuilt-in-Voc , ext , and the splitting of the quasi Fermi level for the doping concentration representing the regions I and II of Figs. 2 and 3.
Gradients of the valence- and conduction band edge for an optimum doping concentration (comparable to a doping concentration around 2000 ppm of Sect. III, solid black lines) and for a too low doping concentration (comparable to a doping concentration around 600 ppm of Sec. III, broken gray lines) without an external voltage applied and no illumination.
(Color online) Gradients of the valence, conduction band edge, and Fermi energy for an optimum doping concentration (see in Fig. 5) and for a too low doping concentration (see in Fig. 5) under illumination and working condition around the maximum power point of the solar cell (V = 590 mV).
(Color online) AFORS-HET simulations of the influence of the doping concentration of the a-Si:H(p)-layer (varied from Ndop = 1.5·1019 to 2.5·1019 cm−3) on Voc , ext and Vbuilt-in , for a defect density present in the a-Si:H(p) of Ndef = 2·1019 cm−3. The Fermi level splitting within the c-Si(n) at the c-Si(n) back side (at 300 μm) as well as the difference Vbuilt-in-Voc , ext are shown, too.
(Color online) Voc , ext and the implied voltage are plotted against each other for different illumination intensities and doping concentrations. The angle dissector line represents the ratio of 1 between the Voc , ext and Voc , impl. The light intensity values shown at the upper x-axis are only accurate for the 2150 ppm sample. For all other samples, the light intensity values represent only estimates.
(Color online) Voc , ext and implied voltage determined using AFORS-HET simulation software for different illumination densities. The total mid gap defect density Ndef in the a-Si:H(p) layer and the doping concentration Ndot have been varied. An n-type crystalline wafer with a ideal Al-rear side metallization (Sp = 1·10−3 cm/s and Sp = 1·106 cm/s), an 5 nm a-Si:H(p) layer, and 80 nm of TCO on top was simulated. No interface defect density has been taken into account.
Parameter set for front and back side contact model applied for the AFORS-HET simulations of Secs. IV B and V B.
Parameter set for the midgap defect density, Urbach-tail defects, a-Si:H(p)-layer, and c-Si(n) wafer applied for the AFORS-HET simulations of Secs. IV B and V B.
Article metrics loading...
Full text loading...