Photovoltaic Power Generation - Renewable Energy: Technology, Economics and Environment - page 241

of absorbed photons per time unit, the photocurrent of a solar cell increases with a
decreasing energy gap.
However, the energy gap also determines the upper limit of the potential barrier
within the p-n-junction (see diffusion voltage in Fig. 6.5). A small energy gap is
thus always associated with a small open-circuit voltage. Since power is the prod-
uct of current and voltage, very small energy gaps only have small efficiencies.
Large energy gaps create high open-circuit voltage, but only allow absorbing a
very limited portion of the solar spectrum. Photocurrent thus only has small val-
ues, and finally the product of current and voltage is small.
This analysis of extreme cases reveals that there is an optimum energy gap with
regard to the choice of semiconductor material for photovoltaic application.
Fig. 6.8 shows the corresponding calculation of the theoretical solar cell effi-
ciency in relation to the energy gap E g of the semiconductor material for a average
solar spectrum /6-12/. Depending on the respective applied material, simple solar
cells (i.e. no tandem solar cell or other type of combined cell) can achieve maxi-
mum theoretical efficiencies of approximately 30 %.
Due to other effects, the efficiencies of real solar cells are much lower than the
indicated theoretical efficiencies (also refer to /6-28/). This is, among other fac-
tors, mainly attributed to the following mechanisms.
− Part of the incident light is reflected by the finger-type contact system or con-
ducting grid mounted on the front side (see Fig. 6.6). By choosing small grid
contacts with maximum spacing in between reflection losses are kept to a
minimum. Yet, for a low-impedance transition resistance between semiconduc-
tor layer and grid contact maximum contact areas are required. Also the spac-
ing between the grid contacts must not exceed inadmissible limits to minimise
the resistance losses of the charge carriers on their way through the semicon-
ductor.

40

GaAs
CdTe
CuInS
CuGaSe
a-Si

35

InP

Si

CuInSe

2

30 2

2

25

20

15

10

5

0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Fig. 6.8 Theoretical efficiencies of various types of simple solar cells under average condi-
tions (see /6-12/)

Energy gap in eV