Page 77 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 77
Stability of Cu(In,Ga)Se 2 Solar Cells
1. Difference in relative humidity.
2. Difference is O /H O ratio.
2
2
3. Encapsulant material interactions, like the formation of acetic acid from EVA.
It is expected that this is not the problem in this case, since PVB and EVA
encapsulated samples degraded similarly.
Sundaramoorthy et al. [105] tested and compared cells with a grid with different back
sheets (TPT (WVTR = 5.5 g/m day), TEFZEL® (WVTR = 4 g/m day), TPAT (WVTR = 1x10 -3
2
2
2
-5
2
g/m day) and glass (reported WVTR = 1x10 g/m day) plus an edge sealant (WVTR <
2
1x10 g/m day). The impact of 50 hours damp heat exposure was studied by electri-
-2
cal measurements via specially designed connection paths. It was observed that the
solar cells with TPT and TEFZEL® back sheets lost most of their initial efficiency due to
damp heat exposure, which was mostly caused by decreasing short circuit current
and fill factor, while the V was reasonably stable. The samples encapsulated in TPAT
oc
were initially stable, but several solar cells became shunted after 20 hours of exposure
to damp heat. The samples with glass back sheets retain about 90% of their efficiency
after 50 hours of damp heat exposure. These results show that barriers with a lower
WVTR limits the degradation of the CIGS solar cells.
The impact of barrier materials was also demonstrated by Olsen et al. [106]. The au-
thors described that unencapsulated cells with a smooth CIGS absorber, an ITO top
contact and a NiAl grid, retain most of their efficiency during 600 to 1000 hours of
damp heat exposure. Once they were encapsulated with a multistack of AlO and
3
2
polymers, significant degradation only occurred after 2500 to 3000 hours of damp
heat exposure. They also showed that this coating protected CIGSSe mini-modules
made from rough CIGSSe, with a ZnO:Al TCO and monolithical interconnection. The
unencapsulated mini-modules degraded very fast (<100 hours) under damp heat ex-
posure, while the encapsulated samples survived longer. The main electrical parame-
ter that decreased due to the exposure of the unencapsulated mini-modules was the
fill factor, which occurred within ten hours. It was concluded that water ingression first
influenced the zinc oxide, after which subsequently the junction was affected, leading
to changes in current transport mechanisms.
Yanagisawa et al. [107] encapsulated CIGS solar cells in EVA and EVA/glass and ex-
posed them as well as an unencapsulated solar cell simultaneously to a damp heat
test and illumination. It was assumed that the degradation of the EVA/glass solar cell
was only caused by temperature and light, while the unencapsulated reference cell
was also affected by humidity. It was shown that the unencapsulated cell degraded
quickly in the beginning, while the sample with EVA started to degrade later.
The degradation of V was proposed to result from an increase in carrier recombina-
oc
75
1. Difference in relative humidity.
2. Difference is O /H O ratio.
2
2
3. Encapsulant material interactions, like the formation of acetic acid from EVA.
It is expected that this is not the problem in this case, since PVB and EVA
encapsulated samples degraded similarly.
Sundaramoorthy et al. [105] tested and compared cells with a grid with different back
sheets (TPT (WVTR = 5.5 g/m day), TEFZEL® (WVTR = 4 g/m day), TPAT (WVTR = 1x10 -3
2
2
2
-5
2
g/m day) and glass (reported WVTR = 1x10 g/m day) plus an edge sealant (WVTR <
2
1x10 g/m day). The impact of 50 hours damp heat exposure was studied by electri-
-2
cal measurements via specially designed connection paths. It was observed that the
solar cells with TPT and TEFZEL® back sheets lost most of their initial efficiency due to
damp heat exposure, which was mostly caused by decreasing short circuit current
and fill factor, while the V was reasonably stable. The samples encapsulated in TPAT
oc
were initially stable, but several solar cells became shunted after 20 hours of exposure
to damp heat. The samples with glass back sheets retain about 90% of their efficiency
after 50 hours of damp heat exposure. These results show that barriers with a lower
WVTR limits the degradation of the CIGS solar cells.
The impact of barrier materials was also demonstrated by Olsen et al. [106]. The au-
thors described that unencapsulated cells with a smooth CIGS absorber, an ITO top
contact and a NiAl grid, retain most of their efficiency during 600 to 1000 hours of
damp heat exposure. Once they were encapsulated with a multistack of AlO and
3
2
polymers, significant degradation only occurred after 2500 to 3000 hours of damp
heat exposure. They also showed that this coating protected CIGSSe mini-modules
made from rough CIGSSe, with a ZnO:Al TCO and monolithical interconnection. The
unencapsulated mini-modules degraded very fast (<100 hours) under damp heat ex-
posure, while the encapsulated samples survived longer. The main electrical parame-
ter that decreased due to the exposure of the unencapsulated mini-modules was the
fill factor, which occurred within ten hours. It was concluded that water ingression first
influenced the zinc oxide, after which subsequently the junction was affected, leading
to changes in current transport mechanisms.
Yanagisawa et al. [107] encapsulated CIGS solar cells in EVA and EVA/glass and ex-
posed them as well as an unencapsulated solar cell simultaneously to a damp heat
test and illumination. It was assumed that the degradation of the EVA/glass solar cell
was only caused by temperature and light, while the unencapsulated reference cell
was also affected by humidity. It was shown that the unencapsulated cell degraded
quickly in the beginning, while the sample with EVA started to degrade later.
The degradation of V was proposed to result from an increase in carrier recombina-
oc
75
