Water electrolysis driven by renewable energy is a promising method to produce green hydrogen that can work as a clean energy carrier for a sustainable energy society. Various water electrolysis systems have been developed and commercialized, however, operations at high current density produce large quantity of bubbles leading to additional overpotentials. Engineering on electrodes, such as surface modifications and 3D-printed structure, is one of the strategies to minimize the bubble influence. Previous studies revealed hydrophilic and periodic structure is preferred for fast bubble departure to reduce overpotential.[1,2] Substrates with three-dimensional structures, including foams and felts, are often employed to deposit sufficient amount of electrocatalysts to minimize overpotentials at a given geometric current density. However, requirements for the substrate thickness have not yet been thoroughly discussed. In this study, we used Ni mesh substrate as a simple model structure and stacked them to vary the thickness to minimize the overpotential during oxygen evolution reaction (OER). Optimization strategy is discussed based on bubble observations, kinetics of electrocatalysts, and electrolyte conductivity.The steady state chronopotentiometries at various current densities were conducted in 1 mol kg−1 KOH at 353 K using pristine and NiFeOx coated Ni mesh (NiFeOx/NM) with various stacking number of meshes. NiFeOx was selected as a model OER electrocatalyst because of its high activity in alkaline media.[3] Tafel slope values were nearly identical among various stacking numbers in the low current density region (<100 mA cm− 2 geo) on both pristine and NiFeOx/NM. However, the value became larger in the high current density region (>500 mA cm− 2 geo) using highly stacked mesh electrodes. The measured potentials at 10 and 800 mA cm− 2 geo as a function of the stacking number were summarized in Figure 1. A monotonic decay of potential was observed at 10 mA cm− 2 geo. Tafel equation predicts surface area enlargement leads to a logarithm potential decay following their Tafel slope values (shown as dashed lines in Figure 1). Their agreements suggest the electrode surface is used uniformly through the thickness direction at 10 mA cm− 2 geo. However, volcano trends were observed at 800 mA cm− 2 geo suggesting the additional overpotential at high current density using thick electrodes. In addition, the optimum stacking number changes by the presence of the catalyst and the aperture of meshes. NiFeOx/NM showed a smaller thickness as an optimum compared to the pristine (Figure 1). The optimum thickness of the stacked electrode with 200 mesh (aperture width of 77 mm) was smaller than that using 20 mesh (aperture width of 1020 mm), which may be due to the bubble accumulation within the stacked meshes.The bubble observation using a high-speed camera was conducted to compare 20 and 200 mesh with various stacking number at 800 mA cm− 2 geo in 1 mol kg−1 KOH at 298 K. Although the average bubble size from 200 mesh was smaller than that from 20 mesh, the size was comparable to the aperture width of 200 mesh. The bubbles filled in the aperture of 200 mesh may reduce the available active sites and the ion conduction path, which can be the reason for the smaller optimum thickness using the dense mesh.To clarify the contribution from the electrocatalysts, particularly Tafel slope values, 1D numerical simulation at 800 mA cm− 2 geo without bubbles was conducted through the stacked direction assuming a porous electrode. The electrode potential no longer decreases above the thickness of 0.5 mm with Tafel slope values of 40 mV dec−1 while continuous decay is predicted with the value of 120 mV dec−1. At 800 mA cm− 2 geo, the electrolyte ohmic voltage loss builds up to 10 – 100 mV within the porous electrode, which results in the loss of local overpotential. The electrode with lower Tafel slope value easily suffers from the additional ohmic voltage loss, which localizes the active region to the near-surface and limits the benefit from the enlarged surface area through the stacking.This study highlights the necessity of substrate structure optimization based on the nature of the deposited electrocatalyst, the electrolyte conductivity, and the dynamics of the bubbles.Reference[1] T. Kou et al., Adv. Energy Mater. 2020, 10, 2002955.[2] J. Das et al, Adv. Funct. Mater. 2024, 34, 2311648.[3] C. C. L. McCrory et al., J. Am. Chem. Soc. 2015, 137, 4347–4357. Figure 1
Read full abstract
Dec 4, 2024
Hongyuan Huang + 6
Open Access
