Page 123 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 123
Temperature dependency of the electrical parameters of CIGS solar cells
of values. For these samples, the absolute temperature dependency varies from
o
-0.034 until -0.071 absolute %/C. When this number is divided by the efficiency at
room temperature, normalised temperature dependencies of -0.36 to -0.70 relative
%/ C are obtained.
o
Figure 4.4 and Figure 4.5 both show the relative and the normalised temperature
coefficients of the efficiency as a function of their efficiency at room temperature. Solar
cells with a high efficiency have a slightly higher absolute temperature dependence,
but because of their high efficiency, the normalised temperature dependency is lower.
Extrapolation of these data points would imply that the temperature dependency of high
efficiency CIGS solar cells and modules is therefore less severe than low efficiency CIGS
samples, but since the various electrical parameters all react different on temperature
increase, part of this difference can be explained by the different compositions.
4.5 Conclusions
The temperature dependency of 42 CIGS solar cells on a soda lime glass and polyimide
substrate was determined. It was observed that the open circuit voltage declinesas a
function of temperature with an absolute value similar for all cells (-2.1±0.2 mV/°C),
regardless of substrate and V at room temperature. The short circuit current density
oc
showed the expected increase with temperature for PI samples, while it decreased
for most SLG samples. We propose that this decrease is largely caused by enhanced
recombination. Additionally, a small decrease in the series resistance was observed
for most cells, while the shunt resistance showed a large decrease with increasing
o
o
temperature. The fill factor (-0.12±0.05%/C) and efficiency (-0.05±0.01%/C) also
showed a decrease with temperature. It is proposed that although the absolute
temperature dependency of the efficiency is higher for high efficiency CIGS solar cells
and modules, the normalised temperature dependency is lower.
4.6 References
[1] S. Liu, E. Simburger, J. Matsumoto, A. Garcia, J. (2008) 2781-2785.
Ross, J. Nocerino, Progress in Photovoltaics: Re - [5] M. Nikolaeva-Dimitrova, R. Kenny, E. Dunlop,
search and Applications 13 (2005) 149-156 Proc. 21 EUPVSEC (2006) 2565-2569
st
[2] D. Mildrexler, M. Zhao, S. Running, EOS, 87 (43) [6] H. Müllejans, A. Burgers, R. Kenny, E. Dunlop, Proc.
(2006) 461-467 19th EUPVSEC (2004) 2455–2458.
[3] T. McMahon, Progress in Photovoltaics: Research [7] A. Virtuani, D. Pavanello, G. Friesen, Proc. 25 EU-
rd
and Applications 12 (2004) 235–248 PVSEC (2010) 4248-4252
[4] H. Mohring, D. Stellbogen, Proc. 23 EUPVSEC [8] J. Wysocki, P. Rappaport, Journal of Applied Phys -
rd
121
of values. For these samples, the absolute temperature dependency varies from
o
-0.034 until -0.071 absolute %/C. When this number is divided by the efficiency at
room temperature, normalised temperature dependencies of -0.36 to -0.70 relative
%/ C are obtained.
o
Figure 4.4 and Figure 4.5 both show the relative and the normalised temperature
coefficients of the efficiency as a function of their efficiency at room temperature. Solar
cells with a high efficiency have a slightly higher absolute temperature dependence,
but because of their high efficiency, the normalised temperature dependency is lower.
Extrapolation of these data points would imply that the temperature dependency of high
efficiency CIGS solar cells and modules is therefore less severe than low efficiency CIGS
samples, but since the various electrical parameters all react different on temperature
increase, part of this difference can be explained by the different compositions.
4.5 Conclusions
The temperature dependency of 42 CIGS solar cells on a soda lime glass and polyimide
substrate was determined. It was observed that the open circuit voltage declinesas a
function of temperature with an absolute value similar for all cells (-2.1±0.2 mV/°C),
regardless of substrate and V at room temperature. The short circuit current density
oc
showed the expected increase with temperature for PI samples, while it decreased
for most SLG samples. We propose that this decrease is largely caused by enhanced
recombination. Additionally, a small decrease in the series resistance was observed
for most cells, while the shunt resistance showed a large decrease with increasing
o
o
temperature. The fill factor (-0.12±0.05%/C) and efficiency (-0.05±0.01%/C) also
showed a decrease with temperature. It is proposed that although the absolute
temperature dependency of the efficiency is higher for high efficiency CIGS solar cells
and modules, the normalised temperature dependency is lower.
4.6 References
[1] S. Liu, E. Simburger, J. Matsumoto, A. Garcia, J. (2008) 2781-2785.
Ross, J. Nocerino, Progress in Photovoltaics: Re - [5] M. Nikolaeva-Dimitrova, R. Kenny, E. Dunlop,
search and Applications 13 (2005) 149-156 Proc. 21 EUPVSEC (2006) 2565-2569
st
[2] D. Mildrexler, M. Zhao, S. Running, EOS, 87 (43) [6] H. Müllejans, A. Burgers, R. Kenny, E. Dunlop, Proc.
(2006) 461-467 19th EUPVSEC (2004) 2455–2458.
[3] T. McMahon, Progress in Photovoltaics: Research [7] A. Virtuani, D. Pavanello, G. Friesen, Proc. 25 EU-
rd
and Applications 12 (2004) 235–248 PVSEC (2010) 4248-4252
[4] H. Mohring, D. Stellbogen, Proc. 23 EUPVSEC [8] J. Wysocki, P. Rappaport, Journal of Applied Phys -
rd
121
