high-resolution TEM
High-resolution TEM (HRTEM) is the mode that produces those famous images where a crystal appears as a neat grid of dots, each dot sitting where a column of atoms runs through the foil. It feels like a direct photograph of the atoms — but understanding what it really is guards you against being fooled. Instead of using just one beam (as in ordinary diffraction-contrast imaging), HRTEM opens the aperture wide to let many diffracted beams through, and lets them recombine and interfere in the image plane. Where the waves reinforce you get a bright fringe; where they cancel, a dark one. Line the crystal up along a zone axis and these fringes cross to make a two-dimensional lattice image.
So the crucial, honest point: an HRTEM image is an interference pattern, not a literal snapshot of atoms. Whether the atomic columns come out bright on a dark background or dark on a bright one — and even whether the dots sit exactly on the columns — depends on the microscope focus, the specimen thickness, and the lens aberrations. There is a special setting called Scherzer defocus where, for a thin enough specimen, the bright (or dark) dots do correspond faithfully to atomic columns, which is why operators aim for it. But change the focus and the contrast can reverse; make the crystal thicker and the pattern can turn into something that no longer maps simply onto the structure. This is captured formally by the contrast transfer function, the microscope's filter on which spatial details pass through and with what sign.
Because of this, careful HRTEM work compares the recorded image against computer simulations for a range of thicknesses and focuses, rather than trusting the eye. The payoff is enormous: HRTEM reveals how atomic planes bend and terminate at a dislocation core, how two crystals stitch together across an interface, how a thin film sits on its substrate, and where stacking faults or twins interrupt the pattern. The arrival of aberration correctors, which cancel much of the lens error, has made images far more directly interpretable, but the underlying caution never fully goes away: a lattice image is a phase-contrast fringe pattern, and its dots are not automatically atoms.
At Scherzer defocus with a very thin silicon crystal viewed down [110], the HRTEM image shows pairs of dark dots spaced 0.14 nm apart — the famous 'dumbbells' of silicon's atomic columns. Change the focus by a few tens of nanometres and those same columns can turn bright, a vivid reminder that the dots are interference fringes, not atoms photographed directly.
The same atomic columns look bright or dark depending on focus — HRTEM contrast must be interpreted, not read off.
The beginner's error is to treat an HRTEM lattice image as a direct photograph of atoms. It is a phase-contrast interference pattern; the sign and position of the dots depend on defocus, thickness, and aberrations, so image simulation is needed to say which dots are atoms.