Page 49 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 49
Stability of Cu(In,Ga)Se 2 Solar Cells
ambient conditions was also reported. Furthermore, the formation of sulphate was
reported as a result the damp heat induced oxidation of sulphur in Cu(In,Ga)(Se,S)
2
solar cells.
Sodium was found to have a very large impact on CIGS absorber stability: in the pres -
ence of sodium and water, oxidation of CIGS absorbers occurred. Experiments showed
that this oxidation catalysed by water promotes an enhanced removal of selenium
from the absorber layer containing NaSe compounds leading to severe efficiency
X
2
loss, mainly following reduced shunt resistance. Additionally, the presence of a large
sodium content due to porous molybdenum also led to the formation of physical dis-
tortions in the CIGS absorber layer.
2.3.3 Buffer degradation
In CIGS solar cells, a n-type buffer is used to form the pn-junction together with the
p-type CIGS absorber. The most used buffer is CdS, which is deposited via Chemical
Bath Deposition (CBD). However, for both environmental and practical issues and the
relative low bandgap of CdS (2.4-2.5 eV), many alternatives for CdS, like Zn(OOH S )
y z
x
and In S have been studied and implemented [52]. In this chapter, the stability of CdS
x y
as well as alternative buffers are described. It should be noted that the same experi-
mental restrictions as described for CIGS absorber layers also exists for buffer studies,
so no studies about individual buffer layers are included.
2.3.3.1 CdS buffer degradation
Wennerberg et al. [38] studied the degradation of the window layer with model
structures, which consisted of a substrate (glass or silicon)/CdS (no/one/two CBD-
dips)/ZnO/ZnO:Al. ZnO:Al grown on a substrate without CdS buffer were more stable
regarding the sheet resistance under damp heat conditions. Furthermore, the TCO
sheet resistance degradation was enhanced by an increasing CdS buffer thickness.
Therefore, the CdS seems to have a negative influence on the ZnO:Al stability.
CIGS solar cells on glass with CBD CdS buffer layers were studied with respect to
their degradation behaviour under damp heat exposure (100 hours, 1000 hours) by
Schmidt et al. [40]. IV measurements revealed that, besides V , most significantly the
oc
fill factordegrades. The degradation of the ZnO/CdS window layer was considered as
likely explanation. Decreasing ZnO conductivity with increasing duration of damp
heat was argued to lead to the observed losses in fill factor. The authors stated that
different ZnO deposition techniques and various CdS thicknesses lead to more or less
degradation of the ZnO/CdS window layer.
Heske et al. [48] analysed ZnO/CIGSSe and ZnO/CdS/CIGSSe interfaces by XES (soft
47
ambient conditions was also reported. Furthermore, the formation of sulphate was
reported as a result the damp heat induced oxidation of sulphur in Cu(In,Ga)(Se,S)
2
solar cells.
Sodium was found to have a very large impact on CIGS absorber stability: in the pres -
ence of sodium and water, oxidation of CIGS absorbers occurred. Experiments showed
that this oxidation catalysed by water promotes an enhanced removal of selenium
from the absorber layer containing NaSe compounds leading to severe efficiency
X
2
loss, mainly following reduced shunt resistance. Additionally, the presence of a large
sodium content due to porous molybdenum also led to the formation of physical dis-
tortions in the CIGS absorber layer.
2.3.3 Buffer degradation
In CIGS solar cells, a n-type buffer is used to form the pn-junction together with the
p-type CIGS absorber. The most used buffer is CdS, which is deposited via Chemical
Bath Deposition (CBD). However, for both environmental and practical issues and the
relative low bandgap of CdS (2.4-2.5 eV), many alternatives for CdS, like Zn(OOH S )
y z
x
and In S have been studied and implemented [52]. In this chapter, the stability of CdS
x y
as well as alternative buffers are described. It should be noted that the same experi-
mental restrictions as described for CIGS absorber layers also exists for buffer studies,
so no studies about individual buffer layers are included.
2.3.3.1 CdS buffer degradation
Wennerberg et al. [38] studied the degradation of the window layer with model
structures, which consisted of a substrate (glass or silicon)/CdS (no/one/two CBD-
dips)/ZnO/ZnO:Al. ZnO:Al grown on a substrate without CdS buffer were more stable
regarding the sheet resistance under damp heat conditions. Furthermore, the TCO
sheet resistance degradation was enhanced by an increasing CdS buffer thickness.
Therefore, the CdS seems to have a negative influence on the ZnO:Al stability.
CIGS solar cells on glass with CBD CdS buffer layers were studied with respect to
their degradation behaviour under damp heat exposure (100 hours, 1000 hours) by
Schmidt et al. [40]. IV measurements revealed that, besides V , most significantly the
oc
fill factordegrades. The degradation of the ZnO/CdS window layer was considered as
likely explanation. Decreasing ZnO conductivity with increasing duration of damp
heat was argued to lead to the observed losses in fill factor. The authors stated that
different ZnO deposition techniques and various CdS thicknesses lead to more or less
degradation of the ZnO/CdS window layer.
Heske et al. [48] analysed ZnO/CIGSSe and ZnO/CdS/CIGSSe interfaces by XES (soft
47
