Potential-induced degradation of encapsulant-less p-type crystalline Si photovoltaic modules

Currently, photovoltaic (PV) energy is one of the leading renewable energy sources, and its use will continue to increase. ¹⁾ Most crystalline silicon (c-Si) PV modules contain an encapsulant, such as ethylene-vinyl acetate (EVA), for long-term reliability and mechanical strength. However, the encapsulant can also cause several problems. For example, it is difficult to separate the materials used in the c-Si PV modules that are bonded with the encapsulant, leading to difficulty in material recycling. ^2,3) EVA reacts with water from the atmosphere to produce acetic acid, which causes electrode corrosion. ⁴⁾ In addition, the encapsulant can be a migration pathway for Na⁺ ions from the cover glass to the cells. Under the negative-bias application to the cells with respect to a grounded aluminum (Al) frame, Na⁺ ions move to the cells and cause performance degradation of the PV modules, referred to as potential-induced degradation (PID). ^5,6) In particular, in the case of p-type c-Si PV modules, Na-decorated stacking faults penetrating through the p–n junction of solar cells are formed by the negative-bias PID stress, and the fill factor of the current–voltage (I–V) characteristics is degraded. ⁷⁾

To solve these problems, we have been working on developing PV modules that do not contain encapsulants. ⁸⁾ Figure 1 shows the structures of the conventional and our novel encapsulant-less modules. ⁸⁾ To date, there have been several preceding studies of encapsulant-less modules. ^9–22) These modules have glasses on both sides that are fixed on their edges. In contrast, the encapsulant-less modules in our study have an openable front cover glass put on a base made of plastic. Thus, the modules have advantages, such as repairability by replacing individual degraded cells and easy disassembly for recycling. However, the novel concept modules can exhibit problems of long-term reliability that are different to those of the conventional encapsulated modules. In particular, moisture can easily reach the surface of the cells, and related degradation modes should be investigated. In addition, the optimal base material and its structure should be selected.

Zoom In Zoom Out Reset image size

In this study, we investigated the PID resistance of our novel encapsulant-less modules using conventional p-type c-Si PV modules in dry and damp-heat (DH) conditions. We confirmed the high stability of our encapsulant-less modules against PID stress both in dry and DH conditions. We also observed the formation of precipitates in the modules under DH conditions.

Conventional and encapsulant-less PV modules were fabricated using commercially available p-type cells with an Al back surface field on their rear. The cells were cleaved into 20 × 20 mm²-sized pieces using a diamond pen, and tab electrodes were then soldered. The conventional modules were fabricated by laminating cover glass/EVA/cell/EVA/backsheet using a vacuum laminator (LM-50X50-S, NPC). A more detailed procedure for the lamination process was described in our previous articles. ^23,24) The encapsulant-less modules were fabricated by just stacking the cover glass/cell/polycarbonate (PC) base. The components of the encapsulant-less modules were lightly fixed with Kapton tape. The PC bases have a cell-size groove at their center with a depth of 1 or 3 mm for cell placement. Figure 2 shows the top view of the encapsulant-less module. Note that there is no edge sealing structure, and the components in the encapsulant-less modules are exposed to ambient air, such as moisture.

Zoom In Zoom Out Reset image size

PID tests were performed for the conventional and the encapsulant-less modules using the Al-plate method in dry and DH conditions. ²⁵⁾ Figure 3 shows the structure diagram of the PID test performed in this study. Conductive rubber was inserted between the Al plate and the cover glass to improve electrical contact between the Al plate and the cover glass. A negative bias of −1000 V was applied to the cell with respect to a grounded Al plate placed on the cover glass of the modules for up to 480 h. The PID tests in dry and DH conditions were performed at 85 °C, and the relative humidity was set to be 85% for the conventional module and the encapsulant-less module having a PC base with a 3 mm groove. We also performed a DH test without bias application for comparison.

Zoom In Zoom Out Reset image size

The PID of the PV modules was evaluated by current-density–voltage (J–V) measurements under 1-sun light irradiation and in the dark at 25 °C. Shunt resistances (R_sh) of the cells were estimated by a one-diode model fitting to the dark J–V data. As will be mentioned later, we observed the formation of yellow precipitates in the encapsulant-less modules after the DH test. We conducted tests in the DH environment with and without PID stress on the encapsulant-less modules for 480 h. To identify the origin of the precipitates, a pristine PC base and the precipitate were analyzed by scanning electron microscopy coupled with energy-dispersive spectrometry (SEM-EDS) (TM3030Plus, Hitachi) and by attenuated total reflection IR (ATR-IR) spectroscopy (Spectrum 100, PerkinElmer).

Figure 4 shows the J–V curves of the conventional and the encapsulant-less PV modules before and after the PID tests in a dry environment for up to 480 h. The encapsulant-less modules show a slightly lower initial short-circuit current density (J_sc) compared to the conventional module, probably caused by larger optical reflection due to the existence of air between the glass and the cells. The conventional modules show significant degradation by the PID test for 24 h, due to serious shunting. In contrast, the encapsulant-less modules show no significant degradation even after the PID test for 480 h. Note that the PID test for the encapsulant-less module with a 1 mm groove shows slight degradation. Figure 5 shows the R_sh of the PV modules as a function of PID-stress duration. The R_sh of the conventional module drops rapidly within a PID stress for 50 h. The encapsulant-less module with a 1 mm groove shows a reduction in R_sh to ∼100 Ωcm² in the first 24 h PID test, and then the R_sh is saturated. In contrast, the encapsulant-less module with a 3 mm groove shows no reduction in R_sh after the 480 h PID test. These results indicate that the encapsulant-less modules show higher PID resistance compared to the conventional modules. This is because the encapsulant can be the invasion pathway for Na ions, and the absence of the encapsulant leads to the suppression of the Na drift and resulting PID. Note that the encapsulant-less modules with shallower grooves show slight degradation, while the PID is completely suppressed in the modules with deeper grooves. We speculate that the shallow grooves may cause the contact between the tab electrodes of the cells and the cover glass, resulting in the Na invasion to the cells through the contact and the emergence of PID. One can see that R_sh is saturated after its rapid drop. The reason for the saturation is unclear at present. A possible reason is the precipitation of a large amount of Na on the cell electrode and the resulting blockage of further Na invasion. These results indicate that the base of the encapsulant-less module must have a groove with enough depth so that the cells do not have contact with the cover glass.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Figure 6 shows the J–V curves of the conventional and the encapsulant-less modules with a 3 mm groove before and after the PID test in a DH environment for up to 480 h. The conventional modules show serious degradation of the PV performance. The performance degradation of the conventional module is not only due to the creation of shunting paths by the PID stress, but also to the generation of acetic acid through the reaction of EVA with moisture and the resulting corrosion of the Ag electrodes and an increase in series resistance. ²⁶⁾ We also observed the partial delamination of EVA from the cell surface, which may also enhance optical reflection loss and decrease J_sc. In contrast, the electrical characteristics of the encapsulant-less modules are not degraded by the PID stress, even in a DH environment. The absence of EVA may lead to the suppression of moisture-induced degradation as well as PID.

Zoom In Zoom Out Reset image size

Figure 7 shows photographs of the encapsulant-less module before and after the DH test. Yellow precipitates are found in the encapsulant-less modules after DH. Note that the yellow precipitates are formed without the application of bias stress, and water ingress is involved in the formation of the precipitates. Although the precipitates do not affect the electrical properties of the modules, the appearance of the modules becomes worse, and the precipitates should be avoided in the actual application. We also confirmed that the yellow precipitates are formed on the glass/PC and the electrode/PC interfaces. We also separately confirmed that the exposure of a PC base alone to the DH environment for 1 month does not lead to the formation of the precipitates. This indicates that the precipitates may be formed through a reaction not with moisture but with liquid water on the interfaces between PC and another solid. We performed SEM-EDS analysis for a pristine PC base and for the precipitates and found that both materials are mainly composed of C (84%) and O (15%), while Na is not detected. This clearly indicates that the formation of the precipitates is not related to the invasion of Na from the cover glass, and the precipitates originate from the PC base. Figure 8 shows the IR absorption spectra of the pristine PC and the precipitates obtained by ATR-IR analysis. Compared to the spectrum of the pristine PC, the peaks of C=O at 1790 cm⁻¹ and O–C–O at 1165–1232 cm⁻¹ become weaker, while the peaks of O–H at 3300 cm⁻¹ and C–C at 1600 cm⁻¹ are enhanced in the spectra of the precipitates. This spectral change can be explained by considering the hydrolysis of PC and the resulting formation of bisphenol A (BPA). ^27–30)

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

In summary, we demonstrated the superior PID resistance of the encapsulant-less modules, while we found the new degradation mode of the encapsulant-less modules, and the formation of yellow precipitates by the hydrolysis of the PC. For the industrialization of the encapsulant-less PV modules, measures for the suppression of precipitate formation must be considered. A possible way to suppress the precipitate formation is using the base material free of hydrolysis. Another possible way is to prepare a barrier structure along the edges of the encapsulant-less modules. To maintain the advantage of the module with openable glass, the usage of O-ring-like rubber gaskets should be considered in the future. It is also important to develop the commercially available large-sized novel encapsulant-less modules and to investigate their mechanical strength.

We investigated the PID resistance of encapsulant-less p-type c-Si PV modules with a PC base in dry and DH environments. In both PID tests, the encapsulant-less modules exhibited superior PID resistance. This is due to the absence of EVA, which can be a pathway for the invasion of Na from the cover glass and form acetic acid through a reaction with moisture. We found that encapsulant-less modules with shallow grooves, in which the tab electrode is in contact with the cover glass, experience slight PID. We speculate that the contact between the glass and the tab electrode of the cell can be a pathway for Na drift to the cell, which causes PID. We also found that yellow precipitates are formed on the glass/PC and the electrode/PC interfaces under DH environment. The origin of the yellow precipitate is probably BPA precipitated by the hydrolysis of PC. The newly found degradation modes should be prevented by modifying the material and/or structure of the modules.

The authors thank Prof. Atsushi Masuda of Niigata University, Prof. Yasuaki Ishikawa of Aoyama Gakuin University, and Prof. Yasushi Sobajima of Gifu University for their fruitful discussion, and Mr. Kodai Nakamura of JAIST for designing the structure of the encapsulant-less modules. This work was supported by the New Energy and Industrial Technology Development Organization (NEDO).