2.5 Oxygen Evolution 27
O + H2O + 2OH�� + 2e�� ! HOO + H2O + OH�� + 3e�� (2.20)
HOO + H2O + OH�� + 3e�� ! O2 + 2H2O + 4e�� (2.21)
The binding energies of the intermediates, O, OH and OOH are central
to the model put forward by Rossmeisl and co-workers. Using the quantum
mechanical method Density Functional Theory, DFT, it is possible to calculate
these energies on various surfaces such as metals and metal oxides. From the
binding energies each step in the reaction mechanism can be evaluated and for a
reaction step to be thermodynamically allowed the change in free energy should
be zero or negative. When all steps are zero or downhill in energy the reaction
can run at "signicant" rates. The calculations are summarized in free energy
diagrams as shown in gure 2.6, where a rutile (110) RuO2 surface is used as
catalytic surface.
Figure 2.6: Free energy diagram for the oxygen evolution reaction based on the
binding of intermediates to a rutile RuO2 surface, shown for 0, 1.23 and 1.6 V. At the
equilibrium potential, 1.23 V the rst three steps are still not energetically favorable,
which means that an extra overpotential is needed for the reaction to proceed at
signicant rates. Figure taken from 90.
The free energy prole of the reaction taking place at U= 0 V has all steps
going uphill in energy and the nal step is 4.92 V higher than the beginning.
This value ts with the total energy needed to split water, which has so far been
stated as 1.23 V per electron. When the potential is increased to 1.23 V it turns
out that the energy prole is not at, but instead three out of four steps are
still uphill in energy. In fact, to get all steps at or downhill it requires 1.60
V. For the ideal catalyst all steps would require 1.23 V to become at. The