Hydrogen fuel cells currently utilize expensive platinum group metal nanoparticles as catalysts [1]. To reduce the cost of catalyst materials, large efforts to develop viable platinum group metal-free (PGM-free) catalyst materials, which often consist of single transition metal atoms embedded in nitrogen-doped graphitic carbon structures, are being undertaken [2]. Given the atomic-scale nature of the metal sites in these materials, their performance and stability are highly dependent on the local coordination environment and consequently electronic structure of the sites [3]. A detailed understanding of the local elemental composition and lattice structure of individual sites, and how these properties vary between sites, is therefore needed to further develop improved PGM-free catalyst materials. While many bulk techniques are available that can provide average properties of these sites, such as Mössbauer and X-ray absorption spectroscopies, techniques with high spatial resolution are needed to measure the properties of individual sites. As a result, scanning transmission electron microscopy (STEM) combined with electron energy-loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy (EDS) has proven an invaluable tool for characterization of these materials due to its ability to probe structural and electronic properties at the atomic scale [4]. However, combined with the instability of many metal sites under the electron probe, the quantity and type of information accessible by conventional STEM instrumentation and acquisition methods is limited. To better understand the local coordination environment and electronic structure of individual sites, as well as how these properties vary site to site, new techniques and hardware are therefore needed.Here, we demonstrate advancements in STEM-EELS and -EDS that allow the quality and quantity of spectroscopic information obtained from individual sites to be significantly improved over conventional method. First, we demonstrate direct real-space mapping of metal sites by using a direct electron detector in the EELS spectrometer. This increases the data acquisition rate and signal-to-noise ratio sufficiently to map metal atom positions at the atomic scale. In addition, we show how automated metal atom position identification and subsequent probe positioning allow the rate at which metal sites are probed to be increased by another order of magnitude while simultaneously increasing the quantity of data obtained from each site. Using EELS and EDS signals acquired in parallel, this automated acquisition method enables a wide range of metal site elements to be identified, as well as the local coordination environment around the sites to be determined. Combined with electronic structure measurements of the metal atoms themselves, this technique provides significantly more information about individual sites than conventional methods. The techniques presented here are therefore poised to significantly improve our understanding of the local coordination environment and electronic structure of individual PGM-free metal sites, as well how these properties vary site to site, accelerating development of these low-cost hydrogen fuel cell catalyst materials [5].
Read full abstract