Several schools of thought exist as to why this difference may occur. Sheng et al. have suggested that hydrogen has greater bonding energy to surfaces in alkaline solutions.  This aligns closely with the work presented by Herranz et al. The inefficacy of pH in changing binding energy of weak-binding metals offers other problems to ponder .  Others have suggested substrate effects may affect the ability of the hydroxide to bind and subsequently the kinetics of the reactions. It may be possible that the partial surface charges of various thermally oxidized metals influence the orientation of the water molecule and subsequent hydrogen availability to the surface. If this were the case, then the size of the catalyst particles on the substrate would greatly affect the HOR/HER kinetics. [4, 5] The circumference of the particles would relate to the proportion of edge affected by the substrate, making smaller particles ideal to observe this phenomenon.
In this study, we attempt to answer two questions to improve fundamental understanding of the alkaline HER/HOR. First, does water orientation affect the kinetics of these reactions? Second, does particle size have an effect as a result of these substrate effects on water orientation? We hypothesize that the particles of smaller size will have a higher proportion of edge effects, thus they will be affected differently than their larger counterparts on the two substrates. The negative surface charge of chromium oxide should promote water in the H-down orientation, facilitating the reaction. Conversely, the positive surface charge on zirconium oxide would orient the water molecule H-up.  To test this hypothesis, oxide films were thermally synthesized from metal foils. Platinum particles were then electrodeposited on the oxide surfaces. The hydrogen underpotential deposition (H-UPD) and HER/HOR kinetics on the oxide-supported Pt nanoparticles were studied in 0.1 M KOH using cyclic voltammetry.
Chromium and zirconium oxides were synthesized by heating metal foils to 100°C for 60 minutes in a tube furnace. to This ensured formation of a chemically stable oxide to inhibit substrate alteration during electrochemical experimentation. Platinum nanoparticles were prepared by pulse electrodeposition. The pulse time was varied between 10 milliseconds and 1 second to control the nanoparticle size. The H-UPD kinetics of different particles on different substrates were compared by measuring the peak splitting of the platinum 110 peaks, then normalizing the current at these associated peaks by electrochemical surface area (ECSA). Results are plotted in Figure 1.
Figure 1: Peak splitting vs the natural log of normalized peak current
Figure 1 illustrates the insignificant influence of substrate on the reaction rates. Thus, either water orientation is not dominant in dictating the reaction kinetics in alkaline solutions or the nanoparticles were too large to observe substrate effects. Although the deposition pulse time has a large effect on kinetics, it was the opposite of the anticipated trend. Since the current was normalized by platinum loading, the sheer differences in platinum loading cannot cause this difference. We hypothesize that smaller, higher defect particles are less active than their larger and less defected counterparts . Future work will investigate the possibility that other factors, such as binding energy, control the reaction rates more strongly than the surface charge effect on water orientation.
 Durst et al, Energy Environ. Sci. 2014, 7, 2255.
 Sheng et al, Nat. Commun. 2015, 6, 5848.
 Herranz et al, Nano Energy 2016, 1-25.
 Zheng et al, J. Electrochem. Soc. 2016, 163, 6, 499-506.
 J. Durst et al, ECS Trans. 2014, 64, 3, 1069-1080.