Performance Degradation, Polymer Encapsulant Degradation, and Estimating Lifetimes for Photovoltaic Modules from Accelerated Testing

Abstract

The purposes of this paper are to provide background information about encapsulants for photovoltaic (PV) modules and to evaluate estimates of the durability of different materials used in PV modules. We summarize field-deployed module degradation in performance, the established degradation mechanisms of ethylene vinyl acetate (EVA) copolymer, and potential degradation mechanisms in other module components and at materials interfaces. Most of the emphasis is on estimates for the commercially used EVA formulations A9918 and 15295. These formulations have degraded in field-deployed modules to produce acetic acid and a yellow to brown color from polyenes and the module efficiencies have been reduced by 10% to 70% in 4 to 12 years. Yet, projections were made by several different research groups in the 1980s that the EVA lifetime could range from 20 to 100 years, the production of acetic acid would only be 0.006% in 20 years, and that silver metallization would survive attack from acetic acid for 300 years at 55°C. Those authors did not use appropriate variables or specimen configurations that simulate reality, and based their projections using the generalization that chemical reaction rates double for every 10°C increase in T. We review the inherent errors in their assumptions about the Arrhenius relation. We also show how degradation in efficiency (from current-voltage data) comparable to field experience is obtained using minimodules with a construction that simulates reality and by using appropriate variables in accelerated testing, i.e., UV, T, and RH. A test protocol is outlined that simulates reality and permits making suitable service lifetime projections.