X feg

All samples were examined in an FEI Titan³ Themis G2 operated at an accelerating voltage of 300 kV, equipped with a field emission gun (X‐FEG) operating at an extraction voltage of 4.5 kV and a monochromator. HR images and diffraction patterns were acquired using a Gatan OneView CMOS camera and an FEI BF‐STEM detector was used to acquire SMF images. The pixel size at each magnification and diffraction pattern had previously been calibrated using a standard of gold nanoparticles on graphite.
The electron flux in CTEM was controlled by adjusting the monochromator focusing lens and the C2 condenser lens and was set to 0.08 e⁻/(Ų s) when searching for areas of interest at low magnification and acquiring selected area electron diffraction (SAED) patterns. In STEM, the total electron fluence per image was controlled by altering the magnification, reducing the probe current to 5 pA using the monochromator focusing lens and setting the dwell time per pixel to 10 µs. The electron flux and measured probe current in CTEM and STEM were based on a flu‐cam current reading which had been previously calibrated using a Faraday cup, and therefore reported uncertainties in the electron flux and probe current measurements are readout error from the flu‐cam. In TEM, this corresponds to ±0.01 e⁻/(Ų s) and in STEM ±1 pA.

Conventional TEM analyses were performed with a JEOL JEM 1400Plus microscope operating at 120 kV, with a LaB₆ source and a GATAN Orius SC600 CCD camera. High resolution HAADF STEM images were acquired by using the FEI Titan 50–300 PICO microscope installed at the Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons (ER-C), Jülich (Germany), equipped with a Schottky type high-brightness electron gun (FEI X-FEG), here operated at 80 kV, a monochromator unit, a Cs probe corrector (CEOS DCOR), a Cs-Cc achro-aplanat image corrector (CEOS CCOR+), and a post-column energy filter system (Gatan Quantum 966 ERS) as well as a 16 megapixel CCD system (Gatan UltraScan 4000 UHS)