Page 120 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 120
Chapter 4
while this number is negative for the CIGS on SLG.
Furthermore, it is observed that the thermal performance of PI samples is worse due to
the relative large decline of the fill factor as a function of temperature (see Figure 4.3).
This was not caused by a high dependence of the fill factor with temperature, but on
a low fill factor at room temperature partly based on the design of the samples used in
these measurements. Therefore, this number as well as the temperature dependency
of the efficiency are more strongly negative than expected for optimised samples.
Additionally, Figure 4.4 and Figure 4.5 show that it should be taken into account that
the normalised temperature dependency is greatly influenced by their values at room
temperature.
This chapter shows that the temperature dependency of CIGS solar cells is comparable
to crystalline silicon solar cells, but it should be noted that the variation within and
between the batches of CIGS samples is large. This is even the case for the samples
deposited on PI, which have been produced on a semi-industrial pilot line. The variation
is also larger than described in reference [4]. This can imply that large differences
between small scale produced CIGS samples exist, but also that differences in cells are
levelled out between modules, which contain a large number of cells. In this chapter,
the impact of the temperature on the various electrical parameters is studied and
described in more detail.
4.4.1 Temperature dependence of the open circuit voltage
In this study, all samples show a steady decrease of open circuit voltage with increas-
o
ing temperature. The temperature coefficient is -2.1±0.2 mV/ C for all samples, regard-
less of the type, substrate and open circuit voltage at room temperature. This is in
o
agreement with the values found in reference [1], which is -2.4±0.1 mV/ C.
When the temperature dependencies of the V of these 42 samples are plotted
oc
versus the V at room temperature (Figure 4.4 and Figure 4.5), it can be seen that
oc
the absolute temperature dependency is slightly decreasing with an increasing V at
oc
room temperature. When the V is considered, a solar cell with a high V is therefore
oc
oc
not only preferable at room temperature, but even stronger at higher temperatures.
This automatically implies that the normalised temperature coefficient is even more
strongly dependent on the V at room temperature.
oc
4.4.2 Temperature dependency of the short circuit current
It was noted that the short circuit current response of the PI simples is different from
the one deposited on SLG (Figure 4.3, Table 4.2 and Table 4.3): most cells on PI fol-
low the classic theory for temperature dependence since they show a small rise on
118
while this number is negative for the CIGS on SLG.
Furthermore, it is observed that the thermal performance of PI samples is worse due to
the relative large decline of the fill factor as a function of temperature (see Figure 4.3).
This was not caused by a high dependence of the fill factor with temperature, but on
a low fill factor at room temperature partly based on the design of the samples used in
these measurements. Therefore, this number as well as the temperature dependency
of the efficiency are more strongly negative than expected for optimised samples.
Additionally, Figure 4.4 and Figure 4.5 show that it should be taken into account that
the normalised temperature dependency is greatly influenced by their values at room
temperature.
This chapter shows that the temperature dependency of CIGS solar cells is comparable
to crystalline silicon solar cells, but it should be noted that the variation within and
between the batches of CIGS samples is large. This is even the case for the samples
deposited on PI, which have been produced on a semi-industrial pilot line. The variation
is also larger than described in reference [4]. This can imply that large differences
between small scale produced CIGS samples exist, but also that differences in cells are
levelled out between modules, which contain a large number of cells. In this chapter,
the impact of the temperature on the various electrical parameters is studied and
described in more detail.
4.4.1 Temperature dependence of the open circuit voltage
In this study, all samples show a steady decrease of open circuit voltage with increas-
o
ing temperature. The temperature coefficient is -2.1±0.2 mV/ C for all samples, regard-
less of the type, substrate and open circuit voltage at room temperature. This is in
o
agreement with the values found in reference [1], which is -2.4±0.1 mV/ C.
When the temperature dependencies of the V of these 42 samples are plotted
oc
versus the V at room temperature (Figure 4.4 and Figure 4.5), it can be seen that
oc
the absolute temperature dependency is slightly decreasing with an increasing V at
oc
room temperature. When the V is considered, a solar cell with a high V is therefore
oc
oc
not only preferable at room temperature, but even stronger at higher temperatures.
This automatically implies that the normalised temperature coefficient is even more
strongly dependent on the V at room temperature.
oc
4.4.2 Temperature dependency of the short circuit current
It was noted that the short circuit current response of the PI simples is different from
the one deposited on SLG (Figure 4.3, Table 4.2 and Table 4.3): most cells on PI fol-
low the classic theory for temperature dependence since they show a small rise on
118
