untetheredatom.com › APT › FAQ
For when the reconstruction looks like abstract art and nobody will say why.
Reconstruction artifacts, mass-spectrum headaches, and instrument choices, each answer linked to the interactive part that works through it.
It's a direct consequence of how the reconstruction turns detector hits into a 3-D map: depth comes from evaporation sequence, which is precise, while lateral position comes from projecting a trajectory back through the tip's evolving shape, which is where aberrations creep in. This page walks through why the two directions behave so differently. See it worked out on Reconstruction & Resolution →
Local magnification: a precipitate with a different evaporation field than the matrix locally changes the tip's curvature during field evaporation, which distorts the trajectories of ions from and around it. This page's trajectory-aberration module shows the effect directly. See it worked out on Reconstruction & Resolution →
If a precipitate or second phase has a different evaporation field from the matrix, local magnification distorts trajectories around it and can make solute appear enriched or depleted where no enrichment exists. The trajectory-aberration sandbox on APT-1 lets you set the field ratio, precipitate size, and tip radius and watch the apparent composition profile deviate from the true one. See it worked out on Reconstruction & Resolution →
Lateral blur from trajectory aberrations and the ion detection point-spread function spread every interface in the reconstruction: an atomically sharp boundary in the specimen becomes a profile 1 to 3 nm wide in the atom map, depending on evaporation-field mismatch, tip radius, and crystallography. The APT-1 lateral-resolution sandbox shows the blur, and the APT-2 proxigram builder compares proxigram width to the true width. See it worked out on Reconstruction & Resolution →
Overlapping peaks (from isotopes or multiply-charged species landing at the same mass-to-charge) need to be deconvolved using known isotopic ratios and charge states before you trust a composition number from that range. This page works through untangling a real overlapping spectrum. See it worked out on Data & Analysis →
Counting statistics set a hard floor: in a small analysis volume, the uncertainty on a measured composition can be larger than the difference you're trying to detect. This page's counting-statistics section puts real numbers on what "a few million atoms" actually buys you. See it worked out on Data & Analysis →
Proximity histograms (proxigrams) and iso-concentration surfaces are the standard tools: they let you plot composition as a function of distance from a defined interface, even when that interface is curved or irregular. This page builds both from reconstructed data. See it worked out on Data & Analysis →
The isoconcentration surface that defines where the interface is moves when you change the threshold, and the proxigram is measured from that surface, so a different threshold shifts both the interface position and the composition profile it reports. The APT-2 page's threshold slider shows the sensitivity directly: the interfacial excess is more robust than the peak concentration to this choice. See it worked out on Data & Analysis →
Yes, in general. The Gibbsian excess integrates the solute surplus across the interface, so it is less sensitive to the exact threshold, delocalization, and bin width than the peak concentration is. The APT-2 proxigram builder reports both; sweeping the threshold shows the excess staying nearly constant while the peak height shifts. See it worked out on Data & Analysis →
Thermal tails on mass peaks are characteristic of laser pulsing: the laser heats the tip surface, and atoms that evaporate slightly after the pulse peak arrive later, stretching the peak to higher mass-to-charge. Higher laser energy broadens the tail and can shift the ranging window needed for accurate composition. The APT-2 mass-spectrum explorer has a laser-pulsing mode that shows the tail directly. See it worked out on Data & Analysis →
Voltage pulsing works well for conductive metals and alloys; laser pulsing extends APT to semiconductors and poor conductors by thermally assisting evaporation, at the cost of some thermal-tail mass resolution. This page compares the two directly. See it worked out on Instrument, Prep & Applications →
FIB lift-out followed by annular milling is the standard route: you cut out a wedge containing the feature, mount it on a post, and mill it down to a sharp needle with the feature positioned near the apex. This page covers the lift-out and sharpening steps. See it worked out on Instrument, Prep & Applications →
Yes, correlative APT + TEM on a single site-specific lift-out is a standard approach: TEM confirms you've captured the right feature (and its crystallography) before you run the needle through the atom probe. This page covers how the correlative workflow fits together. See it worked out on Instrument, Prep & Applications →
Hydrogen in an atom probe mass spectrum can come from at least four sources: residual gas in the analysis chamber, hydrogen absorbed during specimen preparation (especially electropolishing or FIB), hydrogen from the specimen itself, and molecular ions (hydrides, water fragments) that complicate the picture further. Distinguishing specimen hydrogen from background is one of the hardest problems in APT, and the answer is rarely a single number. See it worked out on Instrument, Prep & Applications →