Requirements for Beneficial Electrochemical Restructuring: A Model Study on a Cobalt Oxide in Selected Electrolytes

Abstract

The requirements for beneficial materials restructuring into a higher performance oxygen evolution reaction (OER) electrocatalyst are still a largely open question. Here erythrite (Co3(AsO4)2·8H2O) is used as a Co‐based OER electrocatalyst to evaluate its catalytic properties during in situ restructuring into an amorphous Co‐based catalyst in four different electrolytes at pH 7. Using diffraction, microscopy, and spectroscopy, a strong effect in the restructuring behavior is observed depending of the anions in the electrolyte. Only carbonate electrolyte can activate the catalyst material, which is related to its slow restructuring process. While the catalyst turnover frequency (TOF) undesirably reduces by a factor of 28, the number of redox active sites continuously increases to a factor of 56, which results in an overall twofold increase in current of the restructured catalyst after 800 cycles. The activation is attributed to an adequate local order, a high Co oxidation state close to 3+, and a high number of redox‐active Co ions. These three requirements for beneficial restructuring provide new insights into the rational design of high‐performance OER catalysts by electrochemical restructuring.

Stabilization of Co oxide during oxygen evolution in alkaline media by the introduction of Mn oxide

“…Co oxidation states near 3+ were previously found beneficial for catalytic activity in layered Co oxides. [50] In the case of Mn-K edge in the (Co0.7Mn0.3)Ox films, the averaged bulk Mn oxidation state was 3.7+ and did not change after cycling. However, the averaged Mn oxidation state of (Co0.7Mn0.3)Ox is 0.2 oxidation states higher than the previously studied MnOx films (black line in Figure 5d).…”

“…[66] Such effect may prevent the high coverage (or even saturation) of the catalyst with unreacted intermediates or oxygen. Co remained in an oxidation state close to 3+ being optimal for the OER, [50,95] while Mn in (Co0.7Mn0.3)Ox was in oxidation state 3.7+ independent of potential cycling. The latter is different for MnOx, where we previously argued that Mn oxidation is the main irreversible cause of activity loss.…”

“…The increase in the oxidation state can be attributed to the oxidation of Co 2+ sites to Co 3+ sites, likely at potential values around 1.42 V vs. RHE. [50,72] A redox peak was barely observed in the CV experiments of CoOx (inset in Figure 2a) at around 1.5 V vs. RHE, which can be assigned to the oxidation of a small number of Co 2+ sites. Such redox peak is not observed in (Co0.7Mn0.3)Ox studied in the same range of potential.…”

“…The unlikely relevancy of Co 2+ sites in OER suggests that the conversion of Co 2+ into Co 3+ sites should benefit the catalytic activity, [50,72] yet the catalytic activity did not increase but rather decreased according to RRDE measurements. Another possibility, that should not be discarded, is that the catalytically less relevant Co 2+ ions are lost from the near surface region since it is well soluble in aqueous solutions.…”

“…Similar scan rates and number of cycles are typical conditions for film activation during OER. [48,[50][51][52][53][54] Meanwhile, the Pt ring was set at a constant potential of 0.4 V vs. RHE for oxygen reduction. [55] With this setup, the oxygen produced at the disk during continuous potential cycling is consequently reduced at the ring.…”