Transition metals in multicrystalline silicon solar cells: Understanding the nature, origins, and impacts of metal contamination to minimize its influence on solar cell performance.

摘要

A comprehensive investigation of transition metal impurities in multicrystalline silicon (mc-Si) for solar cell applications is herein detailed. A suite of synchrotron-based analytical X-ray microprobe techniques was employed to characterize the chemical and elemental natures, spatial and size distributions, and recombination activities of metal-rich particles in a range of me-Si materials. Two types of metal-rich particle are typically detected in me-Si: metal silicide nanoprecipitates, and metal-rich inclusions up to tens of microns in size, frequently in an oxidized chemical state. Examination of numerous mc-Si materials yields insights into universal physical principles governing the behavior of metals in me-Si, e.g., the strong interactions between metals and certain types of structural defects, and the dependence of the spatial distribution of a given metal on its diffusivity and solubility. Differences in distribution between mc-Si materials can be explained largely by differences in crystal growth conditions, feedstock purity, and structural defect type and density.; Based on these observations, studies were conducted to determine the effects of several high-temperature processing steps on metal impurities. The final metal distribution (and hence, device impact) is found to be a strong function of processing temperature and cooling rate: fast cools favor homogeneous metal distributions that strongly decrease conversion efficiencies, whereas slow cools allow metal impurities to diffuse to and amass at the most energetically favorable sites, improving material quality elsewhere. Based on these observations, methods are proposed to engineer metallic impurities into less recombination-active states by manipulating their spatial distributions and chemical states, thus reducing their detrimental effects on solar cell performance.