4.4 Conclusion 73
MnOx M1:8 VRHE M1:9 VRHE M20 mA/cm2
EQCM -462 131 -1128 229 -735 93
ICP-MS -332108 -1570 447 -793 194
Table 4.1: EQCM and ICP-MS results for MnOx with chronoamperometry and
chronopotentiometry tests. All values are in units of ngoxide/cm��2 and are measured
over two hours.
MnO2 is therefore 36 hours. Such a lifetime is naturally completely useless in an
actual electrolyzer which should be able to run for several thousands of hours.
At 1.8 VRHE the dissolution rate is 462 ng/cm2 for two hours corresponding to
87 hours which is rather long for a catalyst developer, who may want to vary a
large number of parameters for a new and exciting material. The combination
of EQCM and ICP-MS makes it possible to reveal the stability in short and
reproducible tests that can be used to extend the understanding about materials
at an early stage. For catalysts that are very stable these tests can also be used,
but the level of accuracy must be kept in mind. With tests of two hours it is
challenging to keep the drift of the frequency below 1 Hz. This corresponds
to approximately 9 ng/cm2 per hour. For a 40 nm lm such a loss rate gives
a lifetime of around 2000 hours. Using ICP-MS the detection limits can be
extremely low so with this technique several years of stability could be predicted.
As an example the detection limit for Mn is between 1 and 10 ppt. With a mass
loss of 10 ppt or less in a 200 ml solution during 2 hours of electrochemical
testing, the predicted lifetime of 40 nm would be 4100 hours. Such an example
is quite idealised and assume a perfectly homogeneous lm which may not hold
in practise. However, it does give an impression of what can be done with short
term tests. In an actual device the reaction conditions may dier substantially
from idealised model investigations. Therefore, long term testing in relevant
settings is necessary for establishing durability for nalized catalysts.
4.4 Conclusion
In this chapter the focus has been on developing a protocol to elucidate mass
losses for OER catalysts with short term tests. To this end, manganese oxide
thin lms were deposited on EQCM samples and subjected to typical chronoamperometry
and chronopotentiometry conditions. The activity reported here
compares favourably to literature results for other Mn based catalysts. During
two hour tests the catalyst shows some deactivation but more strikingly
a constant loss of mass. These mass losses are measured carefully with both
EQCM and ICP-MS, resulting in a good match between the methods and good
reproducibility. The results show the importance of measuring mass losses instead
of relying on purely electrochemical data. The mass loss rates can be used