Optimization of hydrogenated amorphous silicon for its use in different photovoltaic technologies

Abstract

[eng] The main objective of this work is to study and optimize the amorphous silicon deposited in a PECVD reactor of recent acquisition, making it suitable for thin film and c-Si based solar cells. Optimization of intrinsic and doped a-Si:H was developed based on several optical and electrical properties. Intrinsic a-Si:H as active layer was optimized with series in three parameters: Depletion, hydrogen flow and temperature. The optimum layer was obtained at low depletion values, with a hydrogen flow ratio of 2:1 and at 200ºC. The series developed for p a- Si:H (TMB flow, temperature, pressure and CH4 flow) led to obtaining device quality material (σdark=1.1·10-5 S/cm, Ea=0.43 eV, Eg=2.02 eV). For p μc-Si:H, after four optimization series (temperature, TMB flow, power and SiH4 flow) device quality material was also obtained (σdark=1.32 S/cm, Eg=2.07 eV, Φ=0.596). In the complete device, the front doped layer thickness and the back reflector optimizations led to increments on the short circuit current of 11% and 12% respectively, and improvements in other parameters. A working PIN a-Si:H solar cell fully deposited at the UB with an efficiency of 7.08% was obtained. During the stage in the LPCIM, a study in polymorphous PIN and NIP solar cells was developed aiming to improve Voc. a-Si:H buffer layers before the pm-Si:H intrinsic material produced an increase in the Voc due to the formation of better interface with the existing μc-SiOx doped layer. Conversely, a-Si:H layers deposited over pm-Si:H and before μc-SiOx material caused the failure of the cell, due to a bad growth of μc-SiOx over amorphous material. To study the light induced degradation of a-Si:H and the associated seasonal effect, a light soaking system was designed and constructed. An optimization of the intrinsic layer thickness was developed; finding an optimal thickness of 200 nm. The role of the temperature in seasonal effect was studied in both T-Solar cells and laboratory UB cells by degradations at different temperatures. Samples going from lower to higher temperatures experienced a recovery of their properties. Conversely, samples going from higher to lower temperatures experienced a sudden drop of their properties. The final state of the parameters of the sample does not depend on the previous history of the sample, but only on the final degradation temperature. Then, a-Si:H deposited at the UB was used as part of c-Si solar cells in collaboration with the UPC. First, the role of sputtered alumina combined with a-Si:H in passivation was studied. A very thin layer of a-Si:H was deposited over the crystalline wafer to act as intermediate layer. This a-Si:H layer acts as physical barrier that protects the wafer from the ion bombardment in the sputtering and serves as hydrogen source that contributes to the saturation of surface defects. The carrier lifetimes obtained with the a-Si:H layer are more than one order of magnitude higher than those without it. Finally, i a-Si:H, p a-Si:H and p μc-Si:H deposited at the UB reactor were used to develop a p- type emitter on an n-wafer. A reoptimization of these materials for this purpose was done. The use of optimized UB’s ITO produced an absolute increment of 3% in the efficiency of devices with n-emitter. The thickness of the p-layers for the emitter was adjusted on an a-Si:H solar cell: The Voc is good (0.875 V), but the absorption has still to be reduced. First trials of p-emitters showed values of the lifetimes limited by the passivation of the a-Si:H, which has to be improved. A study on the influence of carbon concentration and deposition temperature in a- Si:H passivation revealed that including some methane and increasing the temperature help passivation.

[spa] En este trabajo se estudia y optimiza el silicio amorfo depositado en un reactor PECVD de reciente adquisición, a fin de ser empleado en células de lámina delgada y como parte de células basadas en silicio cristalino. La optimización del silicio de diferentes tipos (, i a-Si:H, p a-Si:H y p μc-Si:H) basado en varias propiedades ópticas y eléctricas han permitido obtener una célula de silicio amorfo del 7,08% de eficiencia depositada completamente en la UB. El estudio con capas buffer para aumentar el Voc, ha revelado que epositar una capa de a-Si:H intermedia antes de depositar el pm-Si:H produce un aumento del Voc. Se ha estudiado la degradación por iluminación del a-Si:H. Primero se optimizó el espesor (200 nm) y posteriormente se estudió el rol de la temperatura en el LID. Las células que pasaron de temperaturas más bajas a más altas experimentaron una recuperación en sus propiedades, mientras que las que pasaron de temperaturas más altas a más bajas sufrieron una brusca disminución de sus propiedades. El estado final no depende de la historia de la célula, sino solamente de la temperatura final de degradación. En cuanto al uso de a-Si:H combinado con otras tecnologías, se ha observado que depositar una capa muy delgada de silicio amorfo antes de pasivar una oblea de c-Si con alúmina depositada por sputtering mejora el tiempo de vida en más de un orden de magnitud, al proteger la oblea del bombardeo de iones y servir como fuente de hidrógeno para saturación de enlaces. Se ha realizado una reoptimización de ITO, i a-Si:H, p a-Si:H y p μc-Si:H depositados en la UB con el objetivo de desarrollar un emisor tipo p. El espesor de las capas del emisor tipo p se ajustó en función de su funcionamiento en células de a-Si:H. Las primeras pruebas de emisores presentaron valores de tiempo de vida bajos, limitados por la pasivación del silicio amorfo. Se ha observado que incluir metano y aumentar la temperatura en el depósito de a-Si:H es beneficioso para la pasivación, que aún debe ser mejorada.