% 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{Singh:854287,
author = {Singh, Aryak},
othercontributors = {Rau, Uwe and Knoch, Joachim},
title = {{L}aser processing for interdigitated back-contacted
silicon heterojunction solar cells},
school = {Rheinisch-Westfälische Technische Hochschule Aachen},
type = {Dissertation},
address = {Aachen},
publisher = {RWTH Aachen University},
reportid = {RWTH-2022-09441},
pages = {1 Online-Ressource : Illustrationen, Diagramme},
year = {2021},
note = {Veröffentlicht auf dem Publikationsserver der RWTH Aachen
University 2022; Dissertation, Rheinisch-Westfälische
Technische Hochschule Aachen, 2021},
abstract = {The current state-of-the-art industrial solar modules are
based on either Aluminum Back-Surface Field (Al-BSF) or
Passivated Emitter and Rear Contact (PERC) solar cell
technologies. These inherently suffer from relatively low
efficiency values due to the presence of direct
metal-silicon contact in their design- which induces a high
density of electronically active defects in the band-gap of
silicon. In recent years, this issue has been tackled by the
application of passivated contacts - contacts that feature
with an ultra-thin a-Si:H (i) or SiO2 film between the metal
and silicon to electronically separate them. This film,
besides allowing the transport of charge carriers across the
contact interface, also acts as a passivation layer by
passivating the dangling bonds on the surface of the silicon
wafer. Devices that use a-Si:H (i) film as the passivated
contact are referred to as silicon heterojunction solar
cells (SHJ). Silicon Heterojunction solar cells, in
double-side contacted architecture, in addition to promising
higher efficiencies than the current industrial solar cells,
have a simpler processing sequence than the state-of-the-art
industrial PERC technology, and modules based on these solar
cells, have already entered production in recent times.
Although multiple groups have reported high efficiency
results with SHJ solar cells, nonetheless, the double-side
configuration limits the fullest potential of this
technology due to parasitic absorption and reflective
losses- that occur in the doped layers and metal fingers- on
the front side of the device. These losses can effectively
be mitigated by placing both the contacts on the back-side,
in the form of interdigitated fingers. Using such an
architecture, in 2017, Kaneka report the world record
efficiency of $>26\%$ for a single junction c-Si solar cell.
However, given its complex fabrication sequence, no
large-scale industrial process currently exists for the
Interdigitated Back-contacted Silicon Heterojunction solar
cells (IBC-SHJ) cell. The aim of this work was therefore, to
develop a simple fabrication process for the IBC-SHJ solar
cells based on industrially compatible methods for
large-scale production.In the existing literature, most of
the IBC-SHJ solar cells have been fabricated using
photolithography - a high-cost, low throughput method which
is unsuitable for the photovoltaic (PV) industry. Laser
micromachining is a scalable technology that finds
application in a wide range of industrial processes. Due to
its precise nature of material removal and cost
effectiveness, laser ablation can be an attractive tool for
industrial mass production of IBC-SHJ solar cells. In this
thesis, ablation by nanosecond (λ=355nm) and femtosecond
(λ=515nm, 1030nm) laser pulses has been investigated for
patterning ultra-thin (<15nm) n-type and p-type doped
amorphous silicon films into thin interdigitated fingers.
These flingers form the emitter and base contacts at the
rear of the device. At the outset, to study the impact of
laser ablation on the passivation quality of the samples, a
fast, spatially-resolved and calibration-free
characterization method was first required. The dynamic
photoluminescence lifetime imaging method, recently reported
in literature, was adopted for this application. This method
however, suffers from a high degree of image blurring caused
by the migration of luminescent infrared photons. These
photons are generated inside the sample by the radiative
recombination of excess charge carriers, and migrate tens of
cm away before being extracted from the sample and finally
detected. This is especially disadvantageous for analyzing
laser-damaged areas which, despite having less emittance of
their own, can easily increase the out-coupling of these
infrared photons and erroneously show high-lifetime values
in the damaged regions. In the first part of this work
therefore, the dynamic lifetime imaging was improved by
diminishing the blurring effect. To do this, the nature and
spatial extent of infrared blurring was simulated. It was
theoretically estimated that the blurring in the dynamic
lifetime images could be drastically reduced using an
appropriate spectral filter. Finally, it was experimentally
demonstrated that through the application of such a filter,
the migration length of detectable luminesce photons inside
the wafer can be successfully reduced from 80 to below 2mm.
This improved imaging technique was then used throughout
this thesis to study the efficacy of different patterning
approaches. In SHJ solar cells, a <5nm thick a-Si:H (i) film
passivates the surface states by means of hydrogen present
in its matrix. Since hydrogen can easily out-diffuse from
the film at temperatures >250°C, this puts a thermal
constraint on the ablation of doped amorphous silicon films,
that are present on top of the passivating a-Si:H (i) film.
To achieve a damage-free laser ablation of the doped a-Si:H
films (back-surface field and emitter regions), three
different processing sequences, with progressively
increasing complexity, were explored in the second part of
this thesis. i) The first sequence was done directly on a
stack of a-Si:H (i+n) layer with the aim to selectively
remove the top a-Si:H (n) film without hampering the
underlying a-Si:H (i) layer. ii) In the second sequence, a
thick SiO2 hard-mask layer was deposited on the a-Si:H (i+n)
stack. This was first ablated and then used as an etch
resist, to wet-chemically etch away the underlying damaged
silicon. iii) In the third sequence, an additional a-Si:H
(i) hard-mask layer was deposited on top of the SiO2
hard-mask, in order to contain the damage induced by laser
pulses within the a-Si:H (i) mask. The subsequently
patterning of the underlying films was achieved by wet
chemical processes. For each processing sequence, structural
properties of the crater formed on the precursor sample by
single-shot laser pulses were observed and characterized for
their threshold fluence of ablation using light, confocal
and atomic force microscopes. The phase changes induced on
the substrate- in and around the crater- were identified and
explained using Raman spectroscopy. Finally, the
photoluminescence dynamic lifetime imaging, based on the
filter method described in the previous section, was
utilized to assess the impact of laser processing on the
minority charge carrier lifetime of the sample. The
processing sequence perused with SiO2+a-Si:H(i) hard-mask
structure, successfully allowed for a damage-free patterning
of with single and partially overlapping Gaussian pulses.
For this sequence, laser parameters were determined for both
the femtosecond and the nanosecond pulses, in which ablation
was confined only in the top a-Si:H(i)-mask. At the end of
this investigation, the femtosecond and the cost-effective
nanosecond laser were both chosen to pattern the emitter and
back-surface field regions of IBC-SHJ cells. In the last
part of this work, to pattern the metal fingers on top of
the emitter and back-surface filed regions and complete the
device fabrication, another laser-based patterning method
– laser lithography - was developed and applied in this
work. This was done to have an end-to-end mask-less
patterning sequence, that give the advantage of flexibly
designing the device geometry for investigations at research
scale. Subsequently, by using laser ablation and laser
lithography, to respectively pattern the doped a-Si:H films
and metal fingers, a proof-of-concept IBC-SHJ solar cell
with an $>17\%$ efficiency, was demonstrated with
sub-optimal geometry and thin film deposition conditions. At
the end, the two bottlenecks for the device result – i) a
high density of pin-holes in the hard-mask layers and ii)
under-etching below the SiO2 hard-mask during wet-chemical
etching step- were identified and suggestions to mitigate
these issues, and additionally, to further optimize the
device were provided in the outlook. Thus, with this working
device, a foundation stone was laid for fabricating the
IBC-SHJ devices, using cost-effective tools and with simpler
than contemporary laser-based processing approaches in
literature.},
cin = {615610},
ddc = {621.3},
cid = {$I:(DE-82)615610_20140620$},
typ = {PUB:(DE-HGF)11},
doi = {10.18154/RWTH-2022-09441},
url = {https://publications.rwth-aachen.de/record/854287},
}
guest ::