Analysis of the Tafel plot in 1 M NaOH solution (Figure 11b) shows a slope of 65 mV·dec-1 for Ru2-GC, 56 mV·dec-1 for Pt/C and 80 mV·dec-1 for Rub which as in the case of acidic media, could be associated to a Tafel-Heyrovsky mechanism where the rds is the H2 formation and desorption.
Long-term stability current-controlled bulk electrolysis experiment at j = -10 mA·cm-2 was also performed in the case of Ru2-GC (1 M NaOH), which shows as well good stability, with η10 increasing in only 25 mV over the 12h electrolysis (Figure 10b). The notorious long-term stability of Ru2-GC in basic media was further evidenced by comparison with that of Pt/C under the same conditions, where η10 increased in more than 250 mV over the 12h electrolytic test.
Again, a Faradaic efficiency of 95% was determined by quantifying the amount of H2
generated during electrolysis (Figure 12a), confirming the production of H2 as the sole reaction taking place. In this case the RuNPs are also still visible on TEM images, indicating the high stability of our nanocatalysts (Figure 12b).
Figure 12. a) H2-monitored (blue) current-controlled bulk electrolysis (black) of Ru2-GC at j = -10 mA·cm
-2 in 1 M NaOH. The production of H2 was monitored in the gas phase by the use of a clark electrode.
Faradaic efficiency (Ɛ) = 97%. b) TEM images of Ru2-GC after 20 min bulk electrolysis at fixed j = -10 mA·cm-2 in 1 M NaOH.
The excellent durability of our catalytic system in acidic and basic conditions indicates both, good mechanical stability of the cathode (no need of polymeric gluing agents between RuNPs and GC) and no aggregation of the RuNPs under turnover conditions.
We believe these findings result from the presence of the PP capping agent that allows maintaining the nanostructured character of the material.
3C.5 Electrocatalytic performance benchmarking
The electrocatalytic performance and short-term stability of Ru1-GC (1 M H2SO4) and Ru2-GC (1 M NaOH) was further compared with that of other electrocatalysts
- 136 -
following the benchmarking methodology reported by Jaramillo et al.15 From the capacitive current in a non-Faradaic zone, which is only associated with double-layer charging, the double-layer capacitance (CDL) was estimated (see Experimental part for calculation details). Then, the electrochemically active surface area (ECSA) of both electrodes was calculated from the CDL (Figure 13).
Figure 13. Left, representative multi CV experiment at different scan rates for CDL determination in, and right, plot of current values at -0.35 V (vs. SCE) for the different scan rates, for CDL determination, in a) 1 M H2SO4 and b) 1 M NaOH.
The roughness factor (RF) was calculated by dividing the estimated ECSA by the geometric area of the electrode (S = 0.07 cm2). The ECSA value allows calculating the specific current density (jS) of the electrode (current density per “real” electroactive area of the system) at a given overpotential. The obtained values of η10 at time = 0 and time = 2h and jS at η = 100 mV (js(η=100)) are reported in Tables A3-A4 and plotted in Figure 14, together with those reported for selected HER catalysts benchmarked with the same methodology in acidic 1 M H2SO4 (η10 < 100 mV) and basic 1 M NaOH (η10 <
150 mV) solutions.
- 137 -
3C
Figure 14. Graphical representation comparison of HEC by Jaramillo’s methodology in 1 M H2SO4 (left) and 1 M NaOH (right).
Both Ru1-GC (acidic conditions) and Ru2-GC (basic conditions) show the lowest η10 (20 and 25 mV, respectively) among the reported systems (see Tables A3-A4). Thus, Ru1-GC and Ru2-Ru1-GC outperform Pt in both electrolytes, which shows η10 of 50 (Ru1-GC, 1M H2SO4) and 30 mV (Ru2-GC, 1 M NaOH) and an increase to 60 mV in both media after 2h of electrolysis. The specific current density values observed at η = 100 mV (0.55 mA·cm-2 for Ru1-GC in acidic media and 0.19 mA·cm-2 for Ru2-GC in basic media) are between 2 and 137 times higher than those reported for all the benchmarked catalysts except Pt which, despite of the same order, shows superior values (see Tables A3-A4).
Further information about the intrinsic electrocatalytic activity of our Ru nanomaterials was obtained by calculating TOF values. This was made on the basis of estimated numbers of active sites determined through the underpotential deposition (UPD) of copper.16,4 The method consists on applying a reductive potential to the WE in an electrochemical set-up with a 5 mM CuSO4 solution, to electroreduce Cu2+ in the form of Cu0 only on the Ru0-active sites. The subsequent polarization curve towards oxidative potentials in a Cu-free H2SO4 solution, displays an oxidative wave devoted to the re-oxidation of Cu0 to Cu2+, with the area below the curve proportional to the number of electrons used for the oxidation and thus proportional to the deposited Cu and Ru-active species. Figure 15 shows the resulting curves for a) Ru1-GC, b) Ru2-GC, c) Pt/C and d) Rub, in acidic solution. As can be seen, Ru2-GC (Figure 15b) shows almost no Cu oxidation after the UPD process, confirming the passivation of the NPs’
surface and thus the decrease on the number of Ru0 surface species.
- 138 -
Figure 15. Copper UPD in 1 M H2SO4 solution before (black line) and after (colorful line) of a) Ru1-GC, b) Ru2-GC, c) Pt/C and d) Ru-black.
The calculated TOF values for Ru1-GC in 1 M H2SO4 at 25, 50 and 100 mV (vs. RHE) are 0.55, 3.06 and 17.38 s-1, respectively, which are of the same order than those of Pt/C (1.65, 5.60 and 23.36 s-1) under the same reaction conditions (Table A1 and Figure 16), and significantly higher than those of Rub. Tables A1-A2 allow to compare these TOF values with those reported for other relevant electrocatalysts for a wide set of transition metals, which highlights the fast kinetics of Ru1-GC, which outperforms the other systems.
Figure 16. TOF vs. E (V) graph of Ru1-GC (red), Pt/C (grey) and Rub (orange) systems in 1 M H2SO4. Data obtained by dividing current intensity i = [mA] by the charge under the Cu-UPD wave in each case.
- 139 -