% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@PHDTHESIS{Turan:283046,
author = {Turan, Bugra},
title = {{L}aser processing for the integrated series connection of
thin-film silicon solar cells},
volume = {306},
school = {RWTH Aachen},
type = {Dr.},
address = {Jülich},
publisher = {Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag},
reportid = {FZJ-2016-01728},
isbn = {978-3-95806-119-4},
series = {Schriften des Forschungszentrums Jülich Reihe Energie $\&$
Umwelt / Energy $\&$ Environment},
pages = {xii, 188 S.},
year = {2016},
note = {RWTH Aachen, Diss., 2016},
abstract = {The integrated series connection of solar cells is an
essential aspect for thin-film photovoltaic technology. With
a series connection a high output voltage of the module is
achieved while the output current is kept low. Thus, Ohmic
losses in the contact materials are kept low as well. In
thin-film silicon solar technology the steps to create the
interconnection are commonly done by laser ablation
integrated in-between the depositions of the solar cell
layer materials. In three steps laser scribing is used to
selectively remove layers locally in the form of lines
across the module substrate. In a first step the
front-contact is removed for electrical insulation and cell
stripe definition. Afterwards, the absorber is removed
locally exposing the front-contact beneath. Finally, the
interconnection is formed when the back-contact is removed
locally as well. The area that is needed for the
interconnection of two neighboring cells is no longer active
for current generation. Depending on the technology $5-10\%$
of active area is lost. The reduction of this area holds an
attractive potential for an increase of the module
efficiency. The topic of this thesis is the investigation of
the lower geometrical limits for the dead area reduction for
substrate side laser processing of thin-film silicon solar
cells. It is well-known that the interconnection and the
laser processes can have an impact on the performance of the
solar module. Therefore, the characterization of the impact
on the performance is of special importance when laser
processes are used that are capable of generating a reduced
interconnection width. P1: for the front-contact insulation
process it was found out that the scribe quality strongly
depends on the used laser wavelength. Ablation mechanisms
that are driven by material phase changes (scribing with
532nm or 1064nm) can lead to smoother scribe edges compared
to mechanisms dominated by stress-induced removal (355nm)
where non-uniform rip-off at the edges occurs. However, in
certain processing regimes, strong ablation debris
redeposition in direct vicinity of the P1 scribe is observed
when small beam spot radii (<10µm) are used. Such
redeposition has a severe impact on the solar cell
performance in this region. With proper wet-chemical
cleaning the amount of redeposited debris on the
front-contact and the negative impact on the solar module
can be minimized. Parasitic shunting of two neighboring cell
stripes by deposition of absorber material into the P1
scribe increases when the scribe width is reduced.
Measurements show that the overall magnitude of the shunt is
in a value range that impact on the solar module is
negligible for commonly used cell topologies. P2: the width
reduction approach was extended for the absorber removal
process (P2). To ensure the selectivity of silicon removal
without damaging of the front-contact beneath, only 532nm
was used for scribing. For this wavelength ablation is
strongly assisted by mechanical stresses generated by
hydrogen diffusion from the absorber layer and/or thermal
expansion of the absorber layer. Mechanical constraints
limiting the lower scribe width are found that depend on the
absorber thickness and the laser beam spot size. Such
behavior can be explained directly from linear elastic
fracture mechanics where removal of the layer is determined
by the relation between delamination at the interface and
fracture of the absorber along the circumference of the
spot. It can be concluded that for substrate side
laser-induced ablation thin scribe lines are only possible
for thin layers. The parasitic series resistance formed by
P2 also increases as the scribe width is decreased. However,
for processing of amorphous silicon absorbers, with a beam
radius 10µm, the minimal achievable resistance value is
strongly increased. In fact, much more than what would be
expected just by the geometrical contact area reduction.
This is most likely owed to changes of the specific contact
resistance due to increased debris redeposition within the
P2 scribe prior to back-contact deposition. In contrast,
such effects are not observed for processing of tandem
absorber where debris redeposition is less pronounced. Here,
low series resistances, with only minor impact on the module
performance, are achieved for all investigated beam spot
sizes. P3: the back-contact insulation process (P3) is
similar to P2 since the back-contact is removed indirectly
by removal of the absorber beneath. Shunting between front-
and back-contact can occur at the direct P3 scribe edges.
These shunts are possibly formed due to heat generated by
sub-threshold energy intake of excess energy from the
shoulders of Gaussian distribution of the laser. The
mechanical constraints on the minimal achievable scribe
widths are even stronger than what was observed for the
optimization of the P2 process. This is owed to the
additional overall thickness of the layer-stack due to the
back-contact. Furthermore, for tandem solar cell processing
the scribe edges are strongly distorted by delaminated
material while clean edges are obtained for a-Si:H solar
cells. The parasitic shunting by P3 scribing increases by
many orders of magnitude when a processing beam radius of
10µm is used. However, just like it was observed from P2,
an overall weaker deterioration is obtained for scribing of
tandem solar cells than for amorphous silicon cells. It is
possible that material modifications are more localized in
the a-Si:H top-cell. Together with the higher thickness of
the tandem cells (300nm vs. 1.4µm) the impact on the whole
device is not as pronounced.},
cin = {IEK-5},
cid = {I:(DE-Juel1)IEK-5-20101013},
pnm = {121 - Solar cells of the next generation (POF3-121)},
pid = {G:(DE-HGF)POF3-121},
typ = {PUB:(DE-HGF)11 / PUB:(DE-HGF)3},
url = {https://juser.fz-juelich.de/record/283046},
}