Photovoltaic (PV) modules are susceptible to reliability concerns under reverse bias conditions, particularly when subjected to partial shading. When a solar cell is shaded, it generates less current than its unshaded counterparts, causing excess current to flow in reverse through the affected cell. This can result in a substantial reverse voltage, potentially leading to junction breakdown. Under sustained reverse bias stress, localized regions of the cell can become highly conductive, drawing in more current and dissipating it as heat. This heat accumulation can lead to the formation of localized "hotspots, " a phenomenon observed across all PV technologies. However, the severity of reverse bias damage varies depending on the material system and device architecture. A key reliability concern associated with reverse bias damage is the presence of positive feedback effects, which can trigger thermal runaway. A positive feedback loop where the hotspot becomes more conductive at higher temperatures, leads to an unstable system where the temperature rises quickly. If the temperature surpasses a critical threshold, the cell undergoes irreversible damage. Thin-film solar technologies are particularly susceptible to reverse bias-induced thermal runaway due to their structural and thermal properties. The encapsulation materials commonly used in PV modules have poor thermal conductivity, limiting their ability to dissipate heat efficiently. Additionally, the inherently thin absorber layers of thin-film modules provide little thermal mass, making them more prone to rapid temperature increases compared to conventional wafer-based PV technologies. Among thin-film solar cells, copper indium gallium selenide (Cu (In, Ga) Se₂ or CIGS) is a particularly promising candidate due to its high absorption coefficient of approximately 10^5\, cm^-1 in the visible spectrum (400-700\, nm). This property enables ultra-thin absorber layers (1-2\, ), facilitating lightweight and flexible applications such as vehicle- and building-integrated photovoltaics (VIPV and BIPV) and unmanned aerial vehicles (UAVs). However, like other thin-film technologies, CIGS is prone to reverse bias damage due to its low thermal mass and the limited heat dissipation capabilities of standard encapsulation materials. A well-documented reverse bias degradation mechanism in CIGS solar cells is the formation of so-called "wormlike defects". In this thesis, I propose a three-phase model describing the progression of reverse bias damage in CIGS solar cells: (i) the nucleation phase, (ii) the growth phase, and (iii) the wandering phase. During the nucleation phase, thermal activation of the junction breakdown current leads to an initial instability, initiating a thermal runaway. This phase concludes when thermal decomposition occurs, segregating the CIGS material and forming a conductive shunt-like defect, marking the transition to irreversible damage. In the growth phase, the defect expands from the nanometer to the micrometer scale. Finally, in the wandering phase, the defect propagates through the cell, forming elongated wormlike structures. To investigate the nucleation phase, I developed a novel characterization method called Laser-Induced Hot-Spot Lock-In Thermography (HS-LIT) to visualize and quantify the interplay between thermal heat and electrical power. Initial measurements revealed that laser-induced hotspots caused localized power redistribution, leading to temperature increases. To enhance measurement precision, I introduced laser modulation, enabling the direct quantification of the loop gain driving the thermal runaway. A thermal runaway occurs when the loop gain exceeds 1. Typically, a loop gain above 1 is unmeasurable because it signifies an unstable system. However, in a modulated experiment, the system can temporarily exhibit loop gains above 1 while remaining stable over an entire modulation cycle. This allows for the non destructive quantification of a loop gain above 1, demonstrating that the system would be unstable under DC conditions. Experimental results confirmed that thermal runaway is more likely at higher voltages. For instance, a commercial CIGS solar cell on a glass substrate exhibited loop gains of 1. 05, 1. 10, and 2. 03 at reverse biases of 2, 2. 5, and 3\, V, respectively, demonstrating a superlinear scaling trend. Additionally, HS-LIT measurements revealed that hotspots near the P1 scribe line exhibited higher loop gains compared to those near P3. Further experiments on flexible CIGS modules with steel substrates revealed significantly lower loop gains compared to cells on glass, attributed to the high thermal conductivity of the steel substrate, which effectively dissipated heat from the local hotspot. To analyze the growth phase, I implemented a coupled electro-thermal finite element model (FEM) with high spatial and temporal resolution to simulate defect expansion. The model assumes that after the initial thermal runaway, a small ohmic defect is formed. This defect subsequently exhibits its own positive feedback loop, as power dissipation within the defect heats the surrounding material. If the temperature surpasses a critical threshold, the material undergoes decomposition, forming a conductive phase and thereby enlarging the shunt-like defect. The model successfully replicated experimentally observed defect growth, predicting that an initial defect with a radius of 10\, nm could expand to a 5\, defect within just 1\, ms. For this result we assume a defect with a resistivity of R₃₄₅=3. 14 10^-7\, ². Simulations demonstrated that lower-resistivity defects grow more rapidly, whereas higher-resistivity defects tend to stabilize or cease expanding. Furthermore, the study examined how different device stack layers influence defect evolution, confirming that compared to glass, steel foil substrates suppress thermal runaway due to enhanced heat dissipation. Finally, the wandering phase was explored using the same simulation model employed for the growth phase. These simulations examined how defect location influences propagation dynamics. Results showed that defects originating near P1 interconnections exhibited little wandering, tending to expand along the P1 line. In contrast, defects forming near the P3 scribe line propagated rapidly toward the P1 line. A 10\, nm defect near P3 expanded over a length of 13. 5\, within 18. 5\, , driven by applied power. These results align well with observations in the literature. This work provides a comprehensive framework for understanding reverse bias damage in CIGS solar cells, integrating experimental and modeling approaches to characterize and mitigate thermal runaway. The findings highlight the crucial influence of defect location and substrate selection on the reliability of CIGS thin-film photovoltaics.
Nofal Suheir (Wed,) studied this question.