Figure 3. Initial cyclic voltammetry for Mn3O4 films with 0, 30 and 50 % Au. The results were obtained for Rotating Disk
Electrodes which were performed in 1M KOH using a scan rate 20 mV/s and a rotation speed of 1600 RPM. Inset: At each
concentration the activities are compared at an overpotential of 400 mV. The error bars are based on three independent
measurements and the Au concentration is evaluated with XPS.
From the electrochemical results in Figure 3 it is clear that a significant increase in current density is obtained
with increased concentration of Au in the MnOx films. Most striking is a 5-fold improvement in activity at 400
mV overpotential for the MnOx with 50 % Au. When comparing the overpotential at the benchmark value of 10
mA/cm2,42,43 the presence of Au results in a decrease in the overpotential of 65 mV.
The gold could improve the conductivity of the MnOx film, however, the similar Tafel slopes of pure and mixed
films indicate that differences in uncompensated Ohmic losses are not a dominant effect. Changes in the
surface morphology could lead to a higher electrochemically accessible surface area. The pseudo capacitance
measured electrochemically is considered proportional to the active area.43 The change in capacitance between
pure MnOx and Au(50%)-MnOx is below a factor of two as shown in figure S2.
With the XRD and XPS data a simple model can be made to estimate how many Mn-Au interacting sites that
are present for the two concentrations. Here we will use domain size (2 or 3 nm) as particle size for spherical
gold particles distributed on a flat Mn3O4 surface. If the circumferences of such particles are assumed to be
proportional to the number of Mn-Au interacting sites, then the Au(50%)-MnOx should have less than 1.3 times