Dynamic Stabilization of Nickel-Based Oxygen Evolution Anodes by Phosphate Additive Functioning in the Presence of Chloride Ions

Abstract

Listen

Translate article

The technology of seawater splitting driven by renewable energy potentially produces green and inexpensive hydrogen by simplifying the complex process lines and saves scarce sources of safe and fresh water. Chloride ions (Cl−), a dominant impurity in seawater, is reported to corrode the earth-abundant Ni-based oxygen evolution anode materials in both non-extreme and alkaline pH mediums, resulting in the short operation time.[1,2] Thus, engineering of the reaction field mainly consisting of the electrodes and electrolytes is a recent issue to achieve the prolonged operation time. In this study, we used one of the most active electrodes of NiFeOx/Ni foam (NF) and newly introduced the phosphate ions into the potassium borate electrolyte (K-borate) at pH 9.2. Added phosphate dynamically and selectively formed passivation layer with Ni, preventing corrosion from Cl− and leaching of Ni, which led to longer operation time.[3] Firstly, the redox behavior of NiFe hydroxides (NiFeOx)/NF was investigated using the cyclic voltammetry (CV) in 1.0 mol kg−1 K-borate buffer solution with 0.5 mol kg−1 KCl at pH 9.2 (Figure 1). The electrode was corroded with the formation of green solid Ni(OH)x. In contrast, when the phosphate at 1.0 mol kg−1 was added to the system, the corrosion current was not observed. In addition, the chronopotentiometry testing was conducted at the commercially relevant current density of 500 mA cm−2 at 353 K in K-borate and K-borate/phosphate electrolytes, both of which contained KCl. The rapid corrosion within 10 min happened only in K-borate electrolytes. However, the potential was stable at 1.68 V vs. RHE in phosphate-containing electrolytes for 10 h. These results highlight the advantages of mixing the phosphate into the borate electrolytes. A previous study proposed that the adsorbed phosphate works as a charge repulsion layer in alkaline mediums[2] In the following sections, the role of phosphate is clarified by using spectroscopic and electrochemical analysis.CVs were recorded in both K-borate and K-borate/phosphate electrolytes. In the latter electrolyte, the reduction peak of Ni shifted to more positive than that in pure K-borate. This potential shift was also observed in a single K-phosphate electrolyte. Ex-situ X-ray photoelectron spectroscopy (XPS) testing over the sample tested in phosphate-containing electrolyte shows a lower area ratio of Ni3+/Ni2+ than that tested in K-borate. In addition, operando X-ray absorption near edge structure (XANES) spectra depict the lower Ni absorption edge in the presence of phosphate additive than that in the K-borate electrolyte. In contrast, the spectra of Fe K-edge did not shift under various conditions. These electrochemical and spectroscopic analysis insist on the selective interaction between Ni2+ and phosphate additive. The stability testing was conducted at 50 mA cm−2 and 298 K. The potential increased as time passed in 24 h in K-borate electrolyte, but it was not the case for the phosphate-containing electrolyte. The dissolution amount of Ni in K-borate was 3 times higher than that in the presence of phosphate after the testing, suggesting that the stabilization of Ni2+ by phosphate led to a longer operation. The potential profile and dissolution amount of Ni in the presence of Cl− was similar to without Cl− condition in K-borate/phosphate electrolyte, highlighting that the stabilized Ni was highly tolerant toward the Cl−.The charge repulsion effect is discussed by using the redox probes of other halide ions (Br− and I−). The CVs were recorded in K-borate and K-borate/phosphate with 0.5 mol kg−1 KBr and KI. In K-borate electrolyte, the corrosion current was observed, which disappears by phosphate addition. In contrast, the Faradaic efficiency of O2 (FEO2) at 50 mA cm−2 was close to zero in the presence of both Br− and I−. The two oxidation potentials are 1.31 and 1.59 VRHE which are lower than the onset potential of OER, which would trigger the halide ion oxidation. Given that halide ions were found to react on the surface, the charge repulsion layer did not form on the surface at the present pH level. This is contrary to the previous reports in alkaline mediums, highlighting the uniqueness of non-extreme pH conditions where the phosphate works as a stabilizer toward NiFeOx/NF.This study shows that a concerted design of the electrocatalyst and electrolyte can introduce unique functions.Reference[1] H. Komiya, T. Shinagawa, K. Takanabe, ChemSusChem 2022, 15, e202201088.[2] M. Yu, J. Li, F. Liu, J. Liu, W. Xu, H. Hu, X. Chen, W. Wang, F. Cheng, J. Energy Chem. 2022, 72, 361–369.[3] H. Komiya, K. Obata, T. Honma, K. Takanabe, J. Mater. Chem. A 2024, 12, 3513–3522. Figure 1