Atoms measure roughly 0.1 nanometers across, a scale so small that scientists spent more than six decades developing instruments capable of resolving them with any clarity. The journey from the first blurry glimpse of individual atoms in the mid-1950s to the sharp, directly interpretable images produced by modern electron microscopes is a story of incremental breakthroughs, each one limited by the physics of the tools available. That long arc of progress now shapes how researchers design batteries, semiconductors, and drug molecules, turning what was once a theoretical abstraction into an engineering resource.
The First Glimpse: Field-Ion Microscopy in the 1950s
The race to see single atoms began not with electrons but with ions. In 1956, physicists Erwin W. Muller and Kanwar Bahadur published a landmark paper describing how gases ionize at a metal tip under intense electric fields. Their field-ion microscope (FIM) used helium to achieve resolution on the order of a few angstroms, enough to distinguish the positions of individual atoms on a tungsten surface. It was the first time anyone had produced a real-space image of atoms rather than inferring their arrangement from indirect data, and it immediately demonstrated that the “granularity” of matter could, in principle, be made visible.
Yet the FIM had severe constraints. It worked only on sharp metal tips that could withstand enormous electric fields, which meant the technique applied to a narrow set of materials and geometries. The images showed atomic arrangements on the tip’s outermost surface, but they could not reveal the internal structure of bulk crystals or any information about soft matter and biological samples. The requirement for cryogenic temperatures and ultra-high vacuum further limited its practicality. Scientists needed a way to push atomic-scale imaging beyond needle-shaped metal specimens and into more representative materials, and that meant turning to electron beams.
Electron Microscopes Close the Gap
The next major step came in 1970, when Albert Crewe and his collaborators achieved something the FIM could not: they used an electron microscope to image individual atoms directly. A modern review of scanning transmission electron microscopy (STEM) identifies Crewe’s 1970 experiment as the first time isolated atoms were captured by an electron beam. The team relied on specially prepared samples containing isolated atom pairs and used visibility-factor comparisons to confirm that the bright spots in their images corresponded to single heavy atoms rather than clusters or artifacts. This was a proof of concept rather than a routine characterization method, but it showed that electrons, with their short de Broglie wavelength and tunable energy, could do what ions alone could not.
By the end of that decade, high-voltage transmission electron microscopy (TEM) pushed resolution further. A 1979 paper published in Nature described a 600 kV instrument that achieved directly interpretable resolution approaching 0.2 nm. That figure mattered because it brought TEM into the range where atomic columns in thin crystalline samples could be distinguished without heavy computational reconstruction or speculative modeling. With careful lens design and sample preparation, researchers could now map lattice fringes, dislocations, and interfaces at nearly atomic spacing, even if individual light atoms were still difficult to pick out. The gap between “detecting that atoms exist” and “seeing where they sit inside a material” was finally closing, though the technique still demanded ultra-thin specimens, precise alignment, and carefully controlled beam conditions to avoid destroying the sample.
Scanning Probes Rewrite the Rules
While electron microscopists refined their beams, a parallel revolution was taking shape in Zurich. Gerd Binnig and colleagues built a scanning tunneling microscope (STM) in 1981 and quickly demonstrated its power. Their 1982 paper reported atomic-scale topographic images of surfaces, including monoatomic steps and atomic reconstructions on silicon. Unlike TEM, the STM did not fire particles through a sample. Instead, it brought an atomically sharp metallic tip within a nanometer of a conducting surface and measured the quantum-mechanical tunneling current between tip and sample as the tip rastered across. The result was a three-dimensional map of surface atoms with sub-angstrom vertical sensitivity, turning abstract quantum tunneling into a practical metrology tool.
The STM earned Binnig and Heinrich Rohrer the 1986 Nobel Prize in Physics, and it also spawned a family of related instruments. A 1987 study extended the scanning-probe concept to the atomic force microscope, which replaced tunneling current with forces between the tip and sample, allowing operation outside vacuum and on insulating surfaces. This dramatically widened the range of materials that could be studied at atomic or near-atomic resolution, from polymers to biological macromolecules adsorbed on substrates. These probe-based methods gave researchers a direct, tactile sense of surface geometry that complemented the transmission images produced by electron microscopes. Each technique had trade-offs (STM required conducting samples, AFM tips could wear down or contaminate, and neither could peer deep inside a bulk crystal the way TEM could), but together, they covered far more ground than any single instrument.
What “Atomic Resolution” Actually Means
One persistent source of confusion across this 60-year timeline is what scientists mean when they say they have achieved “atomic resolution.” A review article in Microscopy Today addressed this directly, drawing a careful line between average atomic positions measured by diffraction and the imaging of specific atoms through microscopy. X-ray and electron diffraction had long provided precise lattice parameters and symmetry information, essentially telling researchers where atoms sit on average across billions of unit cells. These reciprocal-space techniques are exceptionally powerful for determining crystal structures and phase compositions, but they inherently average over large volumes and cannot reveal exactly where any one atom resides.
That distinction matters for anyone working in materials science or semiconductor fabrication. A diffraction pattern will confirm that a silicon crystal has the expected diamond-cubic structure and may reveal subtle distortions, but only a STEM or STM image can show whether a single arsenic atom has replaced a silicon atom at a specific lattice site or whether a vacancy cluster has formed at a grain boundary. In other words, diffraction excels at describing periodic order, while real-space microscopy exposes disorder, defects, and local chemistry. As devices shrink toward the atomic scale, the ability to see and identify individual atoms, rather than relying on statistical averages, becomes the difference between understanding why a transistor fails and merely knowing that it did. The practical payoff of six decades of microscope development is not just prettier pictures; it is the capacity to diagnose and engineer materials one atom at a time.
From Laboratory Curiosity to Engineering Tool
The progression from Muller’s helium-ion images to modern aberration-corrected STEM did not follow a smooth curve. Each technique hit physical limits that took years or decades to overcome: lens aberrations in electron microscopes, mechanical drift and vibration in scanning probes, and radiation damage in delicate samples. Incremental improvements in vacuum systems, vibration isolation, detector sensitivity, and computational image processing all contributed to making atomic-resolution imaging more robust and more widely accessible. What began as specialized experiments on tungsten tips or carefully chosen heavy atoms has evolved into routine workflows where researchers can, for example, count the number of dopant atoms in a transistor channel or map the oxygen sublattice in a battery cathode.
That transformation has turned atomic imaging from a laboratory curiosity into a core engineering tool. In semiconductor manufacturing, STEM combined with spectroscopic techniques allows process engineers to verify gate oxide thicknesses and interface roughness at the level of individual atomic layers. In catalysis research, STM and AFM reveal how reactant molecules adsorb and rearrange on active surfaces, guiding the design of more efficient catalysts. And in structural biology, variants of electron microscopy now approach resolutions where side chains and bound ions become visible, tightening the feedback loop between structure and function. The original quest to “see atoms” has thus broadened into a more practical mandate: to control matter at the smallest scales where structure, chemistry, and performance intersect, using images that are no longer symbolic stand-ins but direct, atom-by-atom maps of the materials that underpin modern technology.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
