% 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{Das:974664,
author = {Das, Basita},
othercontributors = {Rau, Uwe and Buonassisi, Tonio},
title = {{D}efect tolerant device geometries for lead-halide
perovskite solar cells},
school = {Rheinisch-Westfälische Technische Hochschule Aachen},
type = {Dissertation},
address = {Aachen},
publisher = {RWTH Aachen University},
reportid = {RWTH-2023-11522},
pages = {1 Online-Ressource : Illustrationen},
year = {2022},
note = {Veröffentlicht auf dem Publikationsserver der RWTH Aachen
University 2024; Dissertation, Rheinisch-Westfälische
Technische Hochschule Aachen, 2022},
abstract = {Motivation, Goal and Task of the Dissertation: Traditional
photovoltaic device optimization efforts rely on reducing
defect density by passivation of surfaces as well as
improved material processing or usage of defect tolerant
materials. However, recombination activity of a defect is
not only a function of defect kinetics but also depends on
the electrostatics and the design of the layer stack of a
photovoltaic device. In this thesis we aim to propose,
develop and prove an alternative approach to solar cell
device optimization, by combining our knowledge of defects
in a material with the impact of device geometry on
recombination via those defects. However, to develop such
guiding principles, we must first understand the
recombination kinetics of defect mediated recombination.
Hence, the first task we undertake is to develop an
analytical model that estimates the electron and hole
capture coefficients of defects. Defect capture coefficients
are difficult to be determined experimentally and in the
absence of better information, they are often heuristically
assumed in device optimization studies. The information on
defect capture coefficients is critical to explore
strategies for modifying device geometry that will deliver a
better performing PV device. So, in our second task, we use
the model of defect capture coefficient in combination of
the photovoltaic device simulator to study the impact of
device geometry on the recombination activity through the
defects inside the device. Device geometry affects the
electrostatics of a device which controls the electron and
hole concentration inside a device. From our efforts to
device optimization in the first and second task it is
apparent that besides the knowledge of defect kinetics, one
also needs the knowledge of material properties of the
different layers as well as their interfaces to pinpoint the
root cause of underperformance to successfully improve a
real device. The electron and hole concentration along with
the defect capture coefficients determine the amount of
defect mediated recombination inside a device. So, in our
third task we take a step further and perform root cause
analysis of underperformance as well as parameter estimation
in perovskite solar cells to make informed decision in our
device optimization efforts.Major Scientific Contributions:
In the course of my doctoral studies, I have achieved all
the three goals outlined above. In my first task I developed
a microscopic model to estimate defect capture coefficients
within the limits of harmonic oscillator approximation. The
model developed in this thesis is a step beyond the state of
the art in the sense that it predicts asymmetric capture
coefficients, which is agreement to the reported values of
capture coefficients in literature for well studies
materials like GaAs. For my second task I performed
extensive device simulation to study the impact of device
geometry both by changing the properties of the charge
transport layers as well as the absorber layer of in a
perovskite solar cell. In this task, I also managed to
achieve agreement with experimental findings in
methylammonium lead halide perovskite solar cells with high
open-circuit voltage. Through this task, I was able to
propose, develop and prove the effectiveness of the new
concept of "Defect tolerant solar cell geometries" and
provide definite guiding principles for future device
optimization efforts. In my third task, I was able to
implement and apply Bayesian inference for root cause
analysis and parameter estimation of perovskite solar cells.
Usage of Bayesian inference techniques on perovskite solar
cells have not been reported before in literature. This task
introduces the perovskite solar cell community to Bayesian
inference as well as machine learning methods for faster and
non-invasive determination of material properties as well as
root cause analysis for underperformance. The three tasks
undertaken in the thesis involves three different kinds of
theory and simulation approaches all equally complex. The
first task involved pure analytical modeling, whereas the
second task is based on device modeling by solution of
coupled differential equation. The third task involved
implementation of a system with Bayesian inference
algorithms in combination of with deep neural networks.
Major Publications: In this thesis I have made three first
author publication covering the studies of the first and
second task. A fourth first author journal article
introducing the open source code developed as a part of the
third task is submitted and a fifth first author journal
article is being written discussing parameter estimation in
perovskite solar cells using Bayesian inference.Published:1.
Das, B., Aguilera, I., Rau, U. Kirchartz, T. What is a deep
defect? Combining Shockley-Read-Hall statistics with
multiphonon recombination theory. Phys. Rev. Mater. 4,
024602 (2020).
https://doi.org/10.1103/PhysRevMaterials.4.0246022.Das, B.,
Liu, Z., Aguilera, I., Rau, U. Kirchartz, T. Defect tolerant
device geometries for lead-halide perovskites. Mater. Adv.
2, 3655–3670 (2021).
https://doi.org/10.1039/D0MA00902D3.Das, B., Aguilera, I.,
Rau, U. Kirchartz, T. Effect of Doping, Photodoping, and
Bandgap Variation on the Performance of Perovskite Solar
Cells. Adv. Opt. Mater. 2101947 (2022).
https://doi.org/10.1002/adom.202101947Unpublished:4. Das,
B., Rau, U., Kirchartz, T. Buonassisi, T. BayesMC: Python
package for doing parameter estimation using Bayesian
inference with Markov Chain Monte Carlo.5. Das, B., Rau, U.,
Buonassisi, T. Kirchartz, T. , Parameter estimation for
perovskite solar cells using Bayesian inference.},
cin = {615610},
ddc = {621.3},
cid = {$I:(DE-82)615610_20140620$},
typ = {PUB:(DE-HGF)11},
doi = {10.18154/RWTH-2023-11522},
url = {https://publications.rwth-aachen.de/record/974664},
}
guest ::