Deposit a dilute dispersion onto a clean, electron-transparent TEM support and let the ethanol evaporate. A practical sequence is to redisperse gently, dilute enough to reduce overlap, place a small drop on a holey or lacey support grid, wick away excess liquid and dry thoroughly. Choose support chemistry compatible with wetting by ethanol.
Look for flakes spanning support holes, where no support film obscures the light-element lattice. Adjacent supported regions help locate and stabilize the flakes. Dilution reduces accidental multilayer stacks, while a clean grid and clean solvent reduce hydrocarbon background. Gentle agitation may be useful; prolonged aggressive sonication can itself change the defects one intends to measure.
Drying prevents a volatile liquid from entering the microscope vacuum. Additional cleaning must be chosen cautiously because heating or energetic plasma treatment can alter oxygen groups and defects in reduced graphene oxide. Record a low-dose survey before atomic-resolution imaging, and distinguish pre-existing defects from changes induced during acquisition.
Use low-voltage, preferably aberration-corrected HRTEM, with about 60–80 kV as a defensible operating range. An approximately 80 kV starting point balances useful lattice resolution against carbon displacement damage; use the lower end if the instrument still resolves the lattice and defect stability is the priority.
A higher accelerating voltage reduces the electron wavelength, but can remove carbon atoms by knock-on displacement. Measurements on pristine single-layer graphene show a strong increase in displacement probability above the vicinity of 80 kV, so simply selecting a conventional 200–300 kV setting is a poor way to preserve the original lattice.[ref]
Accurate Measurement of Electron Beam Induced Displacement Cross Sections for Single-Layer Graphene by Jannik C. Meyer and others. Beam-energy dependence of displacement cross sections, Physical Review Letters 108, 196102
reduced graphene oxide is already defective and chemically modified. Its vulnerable sites need not share the pristine-graphene threshold, and ionization, contamination and chemical etching remain possible at low voltage. Minimize dose, spread the beam while aligning and compare successive short exposures. The voltage choice suppresses one damage mechanism; it does not guarantee damage-free imaging.
Atomic-resolution contrast is primarily electron phase contrast. elastic scattering by the specimen potential changes the phase of the transmitted electron wave. interference between transmitted and diffracted components, modified by the objective lens, converts that phase change into intensity variations. Appropriate defocus, sufficiently small optical aberrations, good spatial optical coherence and transfer of the graphene spatial frequencies are needed. At exact focus an ideal weak phase object need not give a strong visible lattice.
Mass–thickness or amplitude contrast can also distinguish thicker overlap regions, folds, contamination and support film, especially when scattered electrons are excluded by an aperture. It is comparatively weak for a clean monolayer of low- carbon. diffraction contrast becomes important in crystalline, multilayer or differently oriented regions when Bragg-scattered beams are selected or excluded. Relative rotation of layers can additionally produce a moiré pattern.
A dark region alone is not a reliable layer count: electron phase contrast can change sign with defocus, and contamination also changes intensity. Use fold-edge fringes, diffraction information and calibrated image simulations or thickness-sensitive measurements together. Defect identification should likewise compare a structural model with the actual imaging conditions, rather than reading every dark lattice site as a missing atom.
Aberration-corrected scanning transmission electron microscopy with annular dark-field imaging and spatially resolved EELS can locate the Si atoms and test their chemical environment. silicon scatters much more strongly to high angles than carbon, so an isolated Si impurity can stand out in a thin graphene lattice. Neighboring carbon positions reveal whether a site is, for example, three- or four-coordinated in projection.
Acquire a spectrum image at the same time. A silicon core-loss edge identifies the element, while its energy-loss near-edge structure contains information about the unoccupied states and bonding. This combination has experimentally distinguished individual Si coordination environments in graphene.[ref]
Direct Determination of the Chemical Bonding of Individual Impurities in Graphene by Wu Zhou and others. Figures 1–2, atomic-resolution imaging and Si-edge spectra
Advantages are atom-scale spatial localization, simultaneous structural and chemical evidence, and stronger chemical identification than brightness alone. Limitations include low counts at one atom, background subtraction, finite energy resolution, drift and the high dose needed for a useful spectrum. The beam can move or damage an impurity, and a two-dimensional projection does not by itself determine its out-of-plane geometry. Compare spectra and images with plausible structural calculations rather than assigning coordination solely from one bright pixel.
Articles by others on the same topic
There are currently no matching articles.