How determinant is the iron precursor ligand in Fe-N-C single-site formation and activity for oxygen reduction reaction?

Metal-Nitrogen-Carbon (M-N-C) materials have emerged as one of the best non-PGM alternatives to Pt/C catalysts for the electrochemical reduction of O2 in fuel cells. In this work, we explore how decisive is the iron precursor ligand in the Fe-Nx site formation when a commercial carbon black (Vulcan XC72) is pyrolyzed in the presence of an iron complex. Iron complexes coordinating different nitrogen ligands such as phthalocyanine, 1,10-Phenanthroline, 2,2’-Bipyridyl, imidazole, EDTA, etc., were tested. Different Fe-N-C catalysts were synthetized by using twelve different complexes to achieve a better understanding on the role of ligands on site formation and activity, i.e. the site density (SD) and the turn-over frequency (TOF), respectively. It was found that Fe-Nx active site can be effectively formed only when ligands containing aromatic nitrogen rings are used and that the ligand decomposition leads to a substantial change in the carbon morphology developing an increased micropore volume, while mesopore surface area is reduced. The activity was checked in both acidic and alkaline electrolyte by RRDE technique showing that the catalytic activity deeply depends on the type of iron complex, even if such difference is much less important when KOH was used as electrolyte. Site density was determined according to nitrite stripping method, and it was found that the activity (E1/2) in sulphuric acid electrolyte well correlate with the SD, which reaches its maximum when Fe(phen)3Cl2 was used as precursor (SD = 4.24 × 1018 sites g−1; TOF = 6.66 e− sites−1 s−1). Indeed, Fe(phen)3Cl2 involves on one hand a higher mesopore occlusion (from 93 to 34 m2 g−1), but on the other hand the etching due to the small molecules developed by the pyrolysis of the ligand leads to higher micropore surface (from 132 to 246 m2 g−1), which is an important aspect for the formation of Fe-Nx active sites. This clearly places a limit on the ability to increase catalytic activity simply by playing on the initial content of complex, since while we try to increase the SD by increasing the complex loading at the same time the textural properties worsen. Stability test and post-stress test site density analysis on the best catalysts show a higher stability in alkaline electrolyte. In acid electrolyte, the kinetic current decreases by 60% after 7000 cycles and this was proven to depends by the decreasing of the SD, attesting the labile nature of Fe-Nx sites when used in demanding conditions.