5.3 Experimental validation of the concept 89
Ti rich surface could block the catalytic sites and over time make the surface
completely inactive for oxygen evolution. These results are therefore consistent
with the notion that undercoordinated sites on the MnO2 surface are selectively
terminated by Ti, causing a stabilizing eect without signicantly blocking the
oxygen evolution reaction. Further optimization of this concept could lead to
even better performance. It is also possible that other materials could provide
improved protection of the undercoordinated sites. In this study only TiO2 was
investigated experimentally, due to the earlier DFT study. However, since the
MnO2 lm is not crystalline and the Ti modication still leads to a stabilized
surface, it is likely that the most important criteria for successful termination is
using a material with a very low surface formation energy which is also stable
at highly oxidising conditions.
5.3.1 Stabilization in alkaline electrolyte
The Ti-MnO2 lms were also tested for stability in alkaline solution. As shown
in gure 5.3, the mass losses for the pure MnO2 lms were very similar in acid
and alkaline environment. It is therefore expected that the stabilization obtained
by Ti modication works in alkaline as well. Tests conrming this notion were
therefore carried out with 1 M KOH as electrolyte. The test protocol was the
same as described previously, with cyclic voltammetry, stabilization at 1.4 VRHE
and then chronoamperometry at 1.8 and 1.9 VRHE. In gure 5.12a and b the
results are shown for activity and stability, respectively. The deactivation at 1.8
VRHE is 47 % and at 1.9 VRHE it is 30 %. These percentages are higher compared
to the results for acidic electrolyte, where the corresponding deactivation was
19 and 10 % at those two potentials. It is possible that the Ti content in the
samples used for alkaline measurements is slightly higher due to small changes
in deposition rates over time. However, additional calibration of these rates
indicated no such changes. The stability data in gure 5.12 shows that the Ti-
MnO2 exhibits signicantly lower mass losses in alkaline environment compared
to MnO2. At 1.8 VRHE the mass losses are 61 % lower and at 1.9 VRHE the
decrease is 65 %.
The measurements in alkaline therefore conrm that it is possible to obtain a
more stable surface by mixing Ti into the MnO2 lm. In this case the deactivation
turned out to be more signicant but it was matched with a large decrease
in mass loss. The results show that it is possible to obtain better balance between
activity and stability and the results in alkaline adds to the generality of
the concept.
At this point the results are suggesting that the concept of terminating undercoordinated
sites can work. However, it is expected that through optimization of
the deposition process, the deactivation could be avoided while achieving even
lower mass losses. Judiciously chosen annealing treatments could induce mobility
for the Ti atoms in the lm. Furthermore, by varying the concentration
it is possible that the number of undercoordinated sites at the surface can be