Spatial discretized poisoning of SOFC using CFD tools
Kevin Futterer
11:15 - 12:15 Wednesday 15 May 2019 SE PH01150 Solid oxide fuel cells (SOFC) nd more and more attention in driving technologies and
stationary systems for energy production. A lot of eort is put into the durability and
reliability of SOFC to reach durability targets of 40000 to 80000 hours.
For an optimal operation of a fuel cell all damaging failure modes have to be known.
In most cases the most damaging modes will looked at and a model will be developed
to calculate the degradation over time. Through this damaging factor the reliability of
a cell in regard to this failure mode can be determined.
Within this master thesis the eorts were concentrated on four failure modes: Carbon
deposition, Nickel Oxidation, Nickel agglomeration on the anode and Chrome deposition
on the cathode. After development of the models, they were implemented into AVL Fire
and a CFD analysis was carried out. In the end all damage models could be calculated
and spatially illustrated in three dimensions in the software.
Furthermore a new ansatz was developed for damage modeling for fuel cells, which de-
scribes the probability for damage dependent on the conditions.
The rst aim of the master thesis was to develop damage models for Carbon deposi-
tion, Nickel Oxidation, Nickel Agglomeration and Chrome deposition.
Carbon deposition:
The calculation of Carbon deposition on the Nickel based anode of a SOFC is based on
the Langmuir-Hinshelwood kinetics. Methane cracking and the Boudouard reaction are
taken into account for carbon deposition. The degradation is dependent on the partial
pressure of all gases present in the anode and the temperature. It was also possible to
make the carbon deposition rate dependent on the current which is drawn at the cell.
Nickel Oxidation:
The Nickel oxidation model was de ned through the partial pressure of Oxygen on the
anode side. If the partial pressure of Oxygen became higher, than the equilibrium pres-
sure for the formation of NiO, oxidation could occur on the triple phase boundary. If
NiO is build, the anode is going to expand, which then leads to a breakthrough of the
electrolyte. Nickel Oxidation should be avoided at all cost.
Nickel Agglomeration:
Nickel agglomeration is a failure mode which cannot be avoided, during long term op-
eration of a solid oxide fuel cell. Over time the Nickel atoms in the anode start to
clump together, which leads to less active surface area for the reactions to occur, which
furthermore decreases the performance, especially at the end of the lifetime. Nickel ag-
glomeration is mainly dependent on the temperature. Since the temperature is higher at
the fuel outlet Nickel agglomeration is of greater importance at this region.
Chrome Deposition:
Chrome deposition on the cathode is also a failure mode which cannot be avoided.
Through the pipes, which transport the air to the cathode, chrome is introduced to
the system. The Chrome is deposited on the triple phase boundary of the cathode and reduces the active surface area and the performance of the cell. The Chrome deposition
rate is strongly dependent on the partial pressure of water on the cathode side, the tem-
perature and the current which is drawn at the cell.
The second aim of the master thesis was to perform a CFD analysis of a solid oxide
fuel cell. Results like, the current distribution, the temperature pro le, the over poten-
tials of the electrodes and the partial pressures were demonstrated over the cell. After
implementation of all four damage models into the source code of AVL Fire, it was possi-
ble to to calulate and to show the damage spatially in three dimension over the electrodes.
The third and last aim, which processed during the master thesis was to design a new
progress method for durability modeling of fuel cells.
A risk factor could be developed, which described the probability between dierent op-
eration conditions, that damage could occur. This new ansatz was improved, not only
for certain operating conditions, but for whole test cycles.
It is now possible to describe the damage threshold and the probability for damage before
the damage even occured at he cell.
|
Institute of Solid State Physics
