The effects of device geometry and TCO/buffer layers on damp heat accelerated lifetime testing of Cu(In,Ga)Se2 solar cells

Abstract

In Cu(In,Ga)Se2 solar cells encapsulated with polyethylene terephthalate (PET) or glass top sheets, the effects of damp heat (D-H) accelerated lifetime testing (ALT) depend on water vapor transmission rate (WVTR) of both transparent conducting oxide (TCO) and the intrinsic zinc oxide (i-ZnO) buffer, as well as device geometry. PET top sheets have a WVTR of ∼10 g/m2·day, and glass has a WVTR of 0. Previously, coupons encapsulated with PET degraded to 50% of initial efficiency after 1000 h D-H ALT. We show that PET encapsulated coupons degrade at the same rate as glass encapsulated coupons after 2000 h D-H ALT to 92% of initial efficiency. The only change from previous work is that, here, i-ZnO covers the entire coupon surface, not the just active area. The WVTR of the i-ZnO/TCO stack is 2 × 10−3 g·H2 O/m2·day. A set of unencapsulated devices went through D-H ALT, one where scribing was used to define the active area of the device and another without scribing; both were protected only by 50-nm i-ZnO. The bare-unscribed device performed as well as the previous glass and PET encapsulated coupons after 1500 h D-H ALT; the bare-scribed device degraded to 78% of initial efficiency, indicating that TCO integrity is a critical ALT parameter.