Introduction For the realization of efficient hydrogen generation on alkali water electrolysis, there is an urgent need to develop high catalytic active, stable, cost-effective oxygen evolution reaction (OER) catalysts. Up to now, many efforts have been devoted to developing OER oxide catalysts by cation doping, thus, there is little room for the future exploration and alternative developing strategies are required. Anion doping is one of promising catalysts design strategies. Recently, to realize a more flexible control of anion composition, we have developed a novel electrochemical anion-doping technique using anion-conductive solid electrolyte.[1] We have successfully demonstrated electrochemical F-doping to a perovskite oxide La0.5Sr0.5CoO3-δ (LSC55) and found that the surface morphology can be controlled by F-doping current value. By high-rate F-doping, the mixture of amorphous and polycrystalline can be created on the surface of LSC55 particles. On perovskite OER catalysts, improvements of OER activity of perovskite oxides are previously reported by F-doping and surface amorphization.[2][3] Therefore, our electrochemically F-doped samples may have good OER activity. In this study, by preparing two kinds of F-doped samples and evaluating their OER activities, we aimed to elucidate the influence of the electrochemical F-doping on the OER activity. Experiment LSC55 was synthesized by Pechini method. For the F-doping to LSC55, an electrochemical cell, PbF2-Pb|BaF2|BaF2-LSC55, was assembled. Here, PbF2-Pb and BaF2 were the F-source and the solid electrolyte respectively. By controlling the F-doping current density at 60 or 3000 μA/g, two kinds of F-doped samples with different surface morphology were prepared. Each F-doped sample is expressed as LSCF-low and LSCF-high in this abstract. By controlling the amount of charge, 20mol% of F was supplied to each sample. The crystal structure, distribution of F and surface morphology of the prepared samples were evaluated by x-ray diffractometry (XRD), time-of-flight secondary ion mass spectrometry (TOF-SIMS), annular bright field scanning transmission electron microscopy (ABF-STEM).For the OER activity measurements, catalyst inks were prepared by mixing the prepared sample, K+-exchanged nafion and tetrahydrofuran. 6.4 μL of the ink was dropped on a glassy carbon rotating disk electrode. 0.1 M KOH aq., Pt and HgO-Hg were used as an electrolyte, a counter electrode and a reference electrode respectively. OER activities were measured by cyclic voltammetry (CV) between 0.3 and 0.9 V for 100 cycles. Results and discussion The XRD measurements of LSCF-low and LSCF-high revealed that they have the same perovskite structure as the pristine sample. To evaluate the distribution of F in the particle, we performed TOF-SIMS analyses. While a negligible F signal was detected on the pristine sample, clear F signals were detected on the two F-doped samples, which supports the success in the electrochemical F doping. Furthermore, surface F-concentration of LSCF-high was much higher than that of LSCF-low, which indicates the creation of an F-rich surface phase on the LSCF-high. The ABF-STEM observation on the pristine particle revealed that crystalline structure was highly maintained around the surface. On the surface of the LSCF-low particle, while the crystal structure was relatively maintained, the about 3 nm thick amorphous layer was created. On the LSCF-high particle, the about 20 nm thick modified surface layer composed of amorphous and nano-size crystalline domains was observed. These results support that we achieved to prepare F-doped samples with different surface morphologies by the electrochemical F-doping.OER activities of each sample were evaluated by CV. LSCF-high showed the highest initial OER activity and the following order is the pristine and LSCF-low. The maximum current density was about 33% larger than that of the pristine. On the other hand, the cycle performance showed the opposite trend. LSCF-low has the best stability and the degradation rate of LSCF-high was most large. These results clearly support the distinction of OER activity due to the difference of F-doping condition. In the presentation, we will discuss the cause of the different OER activities of electrochemically F-doped samples. Reference [1] T. Katsumata, et al., Adv. Func. Mater., 33, 50, 2307116 (2023).[2] K. Iwase, et al., Chem. Mater., 35, 7, 2773 (2023)[3] Y. Wang, et al., ACS. Appl. Mater. Interfaces, 13, 49, 58566 (2021). Acknowledgement This work was supported by JST, PRESTO (JPMJPR20T6), the establishment of university fellowships towards the creation of science technology innovation (JPMJFS2102) and JSPS KAKENHI (JP22H02174).
Read full abstract