The Optimization of Textured a-Si:H Solar Cells with Three Dimensional Simulation

碩士 === 國立臺灣大學 === 光電工程學研究所 === 101 === Increasing the absorber thickness is efficient to improve the energy conversion efficiency of photovoltaic devices. However, it will limit the carrier transport in the depletion region for a-Si:H thin film solar cells due to its short diffusion length. To solve...

Full description

Bibliographic Details
Main Authors: Chun-Yao Lee, 李君垚
Other Authors: Yuh-Renn Wu
Format: Others
Language:en_US
Published: 2013
Online Access:http://ndltd.ncl.edu.tw/handle/62416360808652173617
Description
Summary:碩士 === 國立臺灣大學 === 光電工程學研究所 === 101 === Increasing the absorber thickness is efficient to improve the energy conversion efficiency of photovoltaic devices. However, it will limit the carrier transport in the depletion region for a-Si:H thin film solar cells due to its short diffusion length. To solve this dilemma, textured surface technologies such as Lambertian limit of randomly textured surface, nano light trapping structures, and photonic structures have gotten more and more attention. This thesis studied the a-Si:H solar cell with randomly rough surface for the high-energy conversion efficiency. The 3D numerical modeling is needed to model the characteristics of randomly rough texture. Therefore, we used the 3D finite-difference time-domain method and 3D Poisson and drift-diffusion solver to discuss and analyze the optical and electrical performance of the a-Si:H solar cell and figure out the optimized configuration for the balance between optical and electrical properties. In addition, we assembled 9 pyramids with different geometry scales determined by measurement results to model the randomly textured surface. And this model with pyramids has been proven that it is efficient enough to represent the real device under small efficiency error of 0.10% in this study. The characteristic of rough surface will affect the electrical and optical performance dramatically. The surface with high aspect ratio will increase the light absorption but limit the carrier transport due to the increasing of the defect density. So this thesis studied the influence of different roughness scales such as different average height of roughness, and the average deviation of roughness, which are most important parameters for surface manufacturing technologies. For the high-efficiency performance, the proposed value of average roughness and root mean square roughness are 30.60 nm and 38.50 nm which have maximum energy conversion efficiency. Moreover, the surface texture with smaller variation has lager short circuit current density due to the stronger light absorption. On the other hand, because of the better fill factor and open circuit voltage, the proposed average thickness of i-type a-Si:H layer is 150 nm, which has the maximum efficiency of 10.48%. To improve the electrical performance of solar cells, we investigated the reason for current losses. For the thinner absorber, the hole blocking layer was implemented due to the stronger back surface recombination. The step doping method was implemented into the n-type a-Si:H layer in this thesis to form the back field to block hole carriers. The step doping concentration we proposed is 1.0E19/5.0E20 cm^(-3), which could reduce the back surface recombination by 53.69%. With our fully 3D numerical modeling without many approximations, we can be more accurately calculating the real device performance. Based on these improvement on both optical and electrical properties, the higher stabilized efficiency of the a-Si:H thin film solar cell could be achieved.