70 Benchmarking the stability of OER catalysts
chronoamperometry tests can be seen for one sample. The current density decreases
slowly over time which could be due to several reasons: the oxygen gas
formed decreases the number of available sites, active sites are corroded away
or change in lm resistivity. The oxygen gas is a potential problem if the lm
is porous since the small pores could be lled with oxygen that is trapped. Dissolution
of active sites is possible but intuitively new sites would be created
after dissolution of the primary ones. Regarding change in resistivity it has
been suggested that oxygen vacancies play an important role in conductivity
for MnO2 158, and prolonged oxidation of the lm could lead to oxygen incorporation
and higher resistivity. It should be noted that the EIS method
is not suitable for measuring local changes in lm resistivity and the Ohmic
loss correction does not rule out such eects. Instead in-situ characterization
of the resistivity could be carried out with interdigitated array electrodes as
demonstrated by Burke et al. 112. However, detailed investigations of the deactivation
mechanisms have not been pursued in this project, where the focus
is on the mass losses.
Figure 4.5: a) Mass changes based on the frequency change during the two hour
tests. The frequency changes have been converted to mass changes by using the Sauerbrey
equation 202. b) Current density in two hours tests for MnOx lms at both 1.8
and 1.9 VRHE. These tests were carried out in 1 M KOH.
In gure 4.5a the change of mass can be seen for two hour tests at 1.8 and 1.9
VRHE. At both potentials the electrode is unstable and negative mass changes
are evident. Comparing the magnitude strongly indicates that the dissolution
rate is potential dependent as expected, with the faster mass loss occuring at 1.9
VRHE. It is striking that the mass losses are linear and constant which allows for
extrapolation to determine the lifetime of the lms. In practise local thickness
dierences can have an impact on the actual lifetime, but as a rst approach the
initial dissolution rate is useful. As mentioned in the protocol description, the
nal step was a potentiometry test. This was carried out at 20 mA/cm2 and
the results can be seen in gure 4.6. This test shows the same story as in gure
4.5. After an initial rapid deactivation there is a slower deactivation process