Page 40 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 40
Chapter 2
2.3.1.1.1 Impact of aluminium incorporation
Wang et al. [32] looked at the addition of aluminium to molybdenum in order to im-
prove the stability. Mo Al (x=14-39 at%) was deposited by co-sputtering and ex-
x
1-x
posed to a damp heat test for 168 hours. It was shown that these films retained their
electrical and optical properties better than the pure molybdenum films. It was pro-
posed that the formation of amorphous Mo-Al alloy regions reduced the surface area
where oxygen and water molecules can diffuse. Furthermore, the formation of a pas-
sive Al O film instead of MoO on the surface was reported to enhance the stability of
3
3
2
the film. However, the presence of aluminium led to the formation of a thinner MoSe 2
layer and its impact on the cell efficiency was not included.
2.3.1.1.2 CIGS cells from degraded molybdenum
The impact of molybdenum oxidation on the efficiency of the subsequent CIGS solar
cells is not necessarily detrimental as was shown by Salomé et al. [33]. Both as-depos-
ited molybdenum and molybdenum stored in a dry N cabinet without H O but with
2
2
low concentrations of O were used to prepare CIGS cells. Electrical and composition-
2
al characterisation of the cells showed that devices prepared with degraded molyb-
denum actually had a higher efficiency (+0.8 %-points) than devices prepared with
as-deposited molybdenum. This was caused by an increase of the V , while the J had
oc
sc
decreased. Elemental profiling via SIMS showed that the sodium concentration in the
molybdenum near the CIGS interface might be higher for the degraded samples, but
no further conclusions could be drawn.
As-deposited molybdenum was also placed in an oxygen atmosphere (60 minutes
at 200 C) in order to simulate the degradation process. This resulted in a very small
o
increase of the V , but it is not clear whether the same mechanism is involved as it is
oc
for the high N /low O degraded samples.
2
2
Hempel et al. [34] also deposited CIGS solar cells on degraded molybdenum. The mo -
lybdenum was exposed to damp heat or elevated temperatures (up to 300C). Both
o
treatments were reported to result in the formation of MoO, which was reduced to
3
o
MoO after CIGS deposition. The exposure to a damp heat test and 200C and 300 C
o
2
dry heat tests were reported to lead to decreased cell performances. This could be
linked to the observed gallium in-diffusion in the molybdenum, as well as the preven -
tion of the formation of MoSe when molybdenum oxide is present.
2
2.3.1.2 Long term stability of the molybdenum in the CIGS cells
While the previous chapters have shown that molybdenum can oxidise very easily,
molybdenum degradation does not seem to be the key-parameter when the loss of
38
2.3.1.1.1 Impact of aluminium incorporation
Wang et al. [32] looked at the addition of aluminium to molybdenum in order to im-
prove the stability. Mo Al (x=14-39 at%) was deposited by co-sputtering and ex-
x
1-x
posed to a damp heat test for 168 hours. It was shown that these films retained their
electrical and optical properties better than the pure molybdenum films. It was pro-
posed that the formation of amorphous Mo-Al alloy regions reduced the surface area
where oxygen and water molecules can diffuse. Furthermore, the formation of a pas-
sive Al O film instead of MoO on the surface was reported to enhance the stability of
3
3
2
the film. However, the presence of aluminium led to the formation of a thinner MoSe 2
layer and its impact on the cell efficiency was not included.
2.3.1.1.2 CIGS cells from degraded molybdenum
The impact of molybdenum oxidation on the efficiency of the subsequent CIGS solar
cells is not necessarily detrimental as was shown by Salomé et al. [33]. Both as-depos-
ited molybdenum and molybdenum stored in a dry N cabinet without H O but with
2
2
low concentrations of O were used to prepare CIGS cells. Electrical and composition-
2
al characterisation of the cells showed that devices prepared with degraded molyb-
denum actually had a higher efficiency (+0.8 %-points) than devices prepared with
as-deposited molybdenum. This was caused by an increase of the V , while the J had
oc
sc
decreased. Elemental profiling via SIMS showed that the sodium concentration in the
molybdenum near the CIGS interface might be higher for the degraded samples, but
no further conclusions could be drawn.
As-deposited molybdenum was also placed in an oxygen atmosphere (60 minutes
at 200 C) in order to simulate the degradation process. This resulted in a very small
o
increase of the V , but it is not clear whether the same mechanism is involved as it is
oc
for the high N /low O degraded samples.
2
2
Hempel et al. [34] also deposited CIGS solar cells on degraded molybdenum. The mo -
lybdenum was exposed to damp heat or elevated temperatures (up to 300C). Both
o
treatments were reported to result in the formation of MoO, which was reduced to
3
o
MoO after CIGS deposition. The exposure to a damp heat test and 200C and 300 C
o
2
dry heat tests were reported to lead to decreased cell performances. This could be
linked to the observed gallium in-diffusion in the molybdenum, as well as the preven -
tion of the formation of MoSe when molybdenum oxide is present.
2
2.3.1.2 Long term stability of the molybdenum in the CIGS cells
While the previous chapters have shown that molybdenum can oxidise very easily,
molybdenum degradation does not seem to be the key-parameter when the loss of
38
