Page 69 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 69
Stability of Cu(In,Ga)Se 2 Solar Cells
In the literature, the conditions leading to changes in these electrical parameters as
well as their influence on the composition of the solar cells and modules are described.
2.4.2 Degradation of CIGS cells and mini-modules
In order to learn more about the impact of several degradation conditions on
mini-modules, Feist et al. [21] exposed encapsulated mini-modules to dry heat and
o
damp heat tests (here 85C and 100% RH) and placed them in various water baths
o
(non-purged/purged with O /purged with argon and at room temperature/85 C). Af-
2
ter maximal 1008 hours, the following global trend in the loss of efficiency due to the
treatments was observed:
o
o
Water bath/O /85 C > 85 C/100% RH > water bath/Ar/85 C > control sample (not ex-
o
2
o
posed) > water bath/RT and 85 C/dry > Water bath/RT/Ar and Water bath/RT/O 2
A similar trend was reported in Feist et al. [29], although the control sample was the
most stable in this case. This trend indicates that moist conditions at elevated tem-
peratures are the most detrimental for CIGS mini-modules, which is not surprising
based on the information in the previous chapters. As stated above, the effects of
the damp heat (85°C/100% RH) tests were severe. For this mini-module, the IV curve
quickly changed from an exponential diode-characteristic to a linear ohmic be-
haviour characteristic. O purged water at 85 C had a similar effect and this mini-mod -
o
2
ule already lost all efficiency after 336 hours. The treatments with water at 85 C, water
o
purged with O and damp heat tests led to the formation of voids in the ZnO:Al layer.
2
The voids were concentrated on the ZnO:Al/CdS and CdS/CIGSSe interfaces and were
also present in the CdS and ZnO:Al layer.
For samples exposed to dry heat conditions, it was observed that the efficiency was
not changing, although a roll-over in the current voltage curve measured under for-
ward bias was observed. This might indicate the formation of an additional blocking
barrier, perhaps due to formation of a resistive contact.
In order to learn more about the degradation mechanisms, an as-deposited mini-mod -
o
ule and a mini-module exposed to O purged water at 85C were cleaved at the Mo/
2
CIGSSe interface and studied by TOF-SIMS and SEM-EDX. In as-received devices, unre -
acted Cu-Ga particulates were found at the molybdenum back contact. Corresponding
with these particles were CIS-rich defects in the absorber layer. After exposure to O
2
o
purged water bath at 85 C, these CIS-regions were enriched with sodium and surround -
ed by an area enhanced in oxidation. This may be detrimental to device performance.
It The observation of sodium indicated that sodium may play a role in device degrada -
tion. It was suggested that sodium, in combination with sites of unreacted Cu-Ga and/
or CIS areas, can facilitate oxygen transport, leading to the formation of metallic oxides.
67
In the literature, the conditions leading to changes in these electrical parameters as
well as their influence on the composition of the solar cells and modules are described.
2.4.2 Degradation of CIGS cells and mini-modules
In order to learn more about the impact of several degradation conditions on
mini-modules, Feist et al. [21] exposed encapsulated mini-modules to dry heat and
o
damp heat tests (here 85C and 100% RH) and placed them in various water baths
o
(non-purged/purged with O /purged with argon and at room temperature/85 C). Af-
2
ter maximal 1008 hours, the following global trend in the loss of efficiency due to the
treatments was observed:
o
o
Water bath/O /85 C > 85 C/100% RH > water bath/Ar/85 C > control sample (not ex-
o
2
o
posed) > water bath/RT and 85 C/dry > Water bath/RT/Ar and Water bath/RT/O 2
A similar trend was reported in Feist et al. [29], although the control sample was the
most stable in this case. This trend indicates that moist conditions at elevated tem-
peratures are the most detrimental for CIGS mini-modules, which is not surprising
based on the information in the previous chapters. As stated above, the effects of
the damp heat (85°C/100% RH) tests were severe. For this mini-module, the IV curve
quickly changed from an exponential diode-characteristic to a linear ohmic be-
haviour characteristic. O purged water at 85 C had a similar effect and this mini-mod -
o
2
ule already lost all efficiency after 336 hours. The treatments with water at 85 C, water
o
purged with O and damp heat tests led to the formation of voids in the ZnO:Al layer.
2
The voids were concentrated on the ZnO:Al/CdS and CdS/CIGSSe interfaces and were
also present in the CdS and ZnO:Al layer.
For samples exposed to dry heat conditions, it was observed that the efficiency was
not changing, although a roll-over in the current voltage curve measured under for-
ward bias was observed. This might indicate the formation of an additional blocking
barrier, perhaps due to formation of a resistive contact.
In order to learn more about the degradation mechanisms, an as-deposited mini-mod -
o
ule and a mini-module exposed to O purged water at 85C were cleaved at the Mo/
2
CIGSSe interface and studied by TOF-SIMS and SEM-EDX. In as-received devices, unre -
acted Cu-Ga particulates were found at the molybdenum back contact. Corresponding
with these particles were CIS-rich defects in the absorber layer. After exposure to O
2
o
purged water bath at 85 C, these CIS-regions were enriched with sodium and surround -
ed by an area enhanced in oxidation. This may be detrimental to device performance.
It The observation of sodium indicated that sodium may play a role in device degrada -
tion. It was suggested that sodium, in combination with sites of unreacted Cu-Ga and/
or CIS areas, can facilitate oxygen transport, leading to the formation of metallic oxides.
67
