Page 15 - Mirjam-Theelen-Degradation-of-CIGS-solar-cells
P. 15
Introduction
The functional layers used in this thesis are summarised:
Substrate: Soda lime glass (SLG): A substrate is used to offer support to the thin solar
cells. SLG is mostly used as substrate since it is relative cheap and contains
sodium as well as potassium. Flexible substrates, such as polyimide or metal
foils, can also be used. In these cases, sodium and in some cases potassium are
mostly added from an alternative source. In this thesis, SLG has been used as a
substrate, unless mentioned otherwise.
Barrier (SiN, Al O , SiO ): optional layer to prevent migration of species from the
x
3
x
2
substrate. This is required for e.g. stainless steel substrates to prevent diffusion
of iron, but it is also used to prevent sodium and potassium migration from soda
lime glass, so these elements can be added at another point of the production
process, in desired quantities. SiN deposited by Plasma Enhanced Chemical
x
Vapour Deposition (PECVD) has been used as barrier within this thesis, this
layer is not depicted in Figure 1.6.
Back contact: Molybdenum: The metal molybdenum is used as conductive layer
beneath the absorber, in order to transport the generated holes. Molybdenum
layers show good conductivity, acceptable adhesion, are reasonably smooth
and allow beneficial sodium diffusion, which is necessary for the formation of
good CIGS absorbers [8,9]. Furthermore, they can withstand the temperatures
required for CIGS production processes and do not react with the metals
in the CIGS absorber (copper, indium and gallium). The main drawback of
molybdenum is the relatively low reflectance compared to other metals.
Molybdenum films are deposited by direct current (DC) sputtering. The
sputtering conditions, including sputter pressure and power, which have been
varied in this thesis, have a large impact on the CIGS absorber composition.
Back contact: Molybdenum diselenide (MoSe ): Molybdenum reacts with selenium
2
during the CIGS deposition process, which results in the formation of beneficial
molybdenum diselenide (MoSe ) layers [9], which function as an ohmic contact
2
between molybdenum and CIGS.
Absorber: Cu(In,Ga)Se: In this layer, light is absorbed and electron-hole pairs are
2
generated. This layer consists of copper indium gallium diselenide (or CuIn
1-
Ga Se , Cu(In,Ga)Se or CIGS) and is a p-type polycrystalline semiconductor
x x 2 2
with a direct bandgap and a high optical absorption coefficient. These features
allow the use of relatively thin (2 µm) absorber layers [10].
Cu(In,Ga)Se is a solid solution of CuInSe and CuGaSe and can have all
2
2
2
compositions for Ga/(In+Ga) from 0 to 1. With the variation of the gallium and
indium ratio, the lattice parameters as well as the bandgap of the material can
be changed: Pure CuInSe has a bandgap of 1.04 eV, while pure CuGaSe has a
2
2
13
The functional layers used in this thesis are summarised:
Substrate: Soda lime glass (SLG): A substrate is used to offer support to the thin solar
cells. SLG is mostly used as substrate since it is relative cheap and contains
sodium as well as potassium. Flexible substrates, such as polyimide or metal
foils, can also be used. In these cases, sodium and in some cases potassium are
mostly added from an alternative source. In this thesis, SLG has been used as a
substrate, unless mentioned otherwise.
Barrier (SiN, Al O , SiO ): optional layer to prevent migration of species from the
x
3
x
2
substrate. This is required for e.g. stainless steel substrates to prevent diffusion
of iron, but it is also used to prevent sodium and potassium migration from soda
lime glass, so these elements can be added at another point of the production
process, in desired quantities. SiN deposited by Plasma Enhanced Chemical
x
Vapour Deposition (PECVD) has been used as barrier within this thesis, this
layer is not depicted in Figure 1.6.
Back contact: Molybdenum: The metal molybdenum is used as conductive layer
beneath the absorber, in order to transport the generated holes. Molybdenum
layers show good conductivity, acceptable adhesion, are reasonably smooth
and allow beneficial sodium diffusion, which is necessary for the formation of
good CIGS absorbers [8,9]. Furthermore, they can withstand the temperatures
required for CIGS production processes and do not react with the metals
in the CIGS absorber (copper, indium and gallium). The main drawback of
molybdenum is the relatively low reflectance compared to other metals.
Molybdenum films are deposited by direct current (DC) sputtering. The
sputtering conditions, including sputter pressure and power, which have been
varied in this thesis, have a large impact on the CIGS absorber composition.
Back contact: Molybdenum diselenide (MoSe ): Molybdenum reacts with selenium
2
during the CIGS deposition process, which results in the formation of beneficial
molybdenum diselenide (MoSe ) layers [9], which function as an ohmic contact
2
between molybdenum and CIGS.
Absorber: Cu(In,Ga)Se: In this layer, light is absorbed and electron-hole pairs are
2
generated. This layer consists of copper indium gallium diselenide (or CuIn
1-
Ga Se , Cu(In,Ga)Se or CIGS) and is a p-type polycrystalline semiconductor
x x 2 2
with a direct bandgap and a high optical absorption coefficient. These features
allow the use of relatively thin (2 µm) absorber layers [10].
Cu(In,Ga)Se is a solid solution of CuInSe and CuGaSe and can have all
2
2
2
compositions for Ga/(In+Ga) from 0 to 1. With the variation of the gallium and
indium ratio, the lattice parameters as well as the bandgap of the material can
be changed: Pure CuInSe has a bandgap of 1.04 eV, while pure CuGaSe has a
2
2
13
