How The Electron Microscope Sees Tiny Things

A standard light microscope cannot resolve anything smaller than about 200 nanometers, roughly the width of a large virus, no matter how well its lenses are ground. That ceiling is not a manufacturing failure; it is physics, fixed by the wavelength of visible light itself. Ernst Ruska and Max Knoll broke through it on March 9, 1931, at the Technische Hochschule zu Berlin, by replacing light with a beam of electrons whose wavelength can be thousands of times shorter, and by the early 1930s their prototype was already outperforming the best optical microscopes of the era.

Why light itself sets the limit

In the 1870s, German physicist Ernst Abbe worked out that any optical instrument's maximum resolution is limited to roughly half the wavelength of the light it uses to form an image. Visible light spans about 400 to 700 nanometers, and that range caps useful magnification at somewhere around 1,000 to 1,500 times before the image simply dissolves into a blur, regardless of how precisely the glass is shaped. This ceiling is called the diffraction limit, and Abbe's formula still governs every optical microscope built today, from a classroom instrument to a research-grade confocal system.

Diffraction limit — the physical boundary on resolution imposed by a wave's own wavelength, below which no amount of lens refinement can separate two closely spaced points. It is not a flaw in engineering; it is a property of waves themselves, whether those waves are visible light, sound, or, as Ruska and Knoll eventually exploited, electrons.

The diffraction limit explains why, for three centuries after Antonie van Leeuwenhoek's pioneering work with hand-ground lenses in the 1670s, biologists could see bacteria but never resolve a virus, a protein, or the internal machinery of a cell. Viruses typically range from about 20 to 300 nanometers, well below what any light-based instrument could separate into distinct shapes. Researchers knew something smaller than bacteria caused diseases such as tobacco mosaic disease and polio, because the infectious agent passed straight through filters fine enough to trap bacteria, but nobody had actually seen one. That gap between inference and observation is exactly what electron optics closed.

The key insight, developed independently by Louis de Broglie in his 1924 doctoral thesis, was that matter itself behaves as a wave under the right conditions, not just light. De Broglie proposed that every moving particle has an associated wavelength inversely proportional to its momentum, a relationship now written as λ = h/p, where h is Planck's constant and p is momentum. For everyday objects, this wavelength is absurdly small and irrelevant, but for a lightweight particle like an electron accelerated to high speed, it becomes a workable optical tool.

How an electron beam replaces a glass lens

Electrons cannot pass through ordinary glass lenses the way photons do; a lens ground from silica does nothing to a stream of charged particles. Instead, electron microscopes bend the beam using precisely wound electromagnetic coils, called magnetic lenses, that exploit the fact that a moving charge feels a force when it crosses a magnetic field. Vary the current through the coil, and the focal length of that magnetic lens changes exactly the way turning a focus ring changes a camera lens's focal length.

Picture steering a stream of charged marbles through a tunnel lined with magnets rather than bouncing a flashlight beam off a set of mirrors. The underlying physics of "focusing" is completely different from optical refraction, but the goal, bringing scattered signal back together into a sharp point, plays the same role. Ruska built the first working two-stage magnetic lens system with Knoll in the spring of 1931, and by September 23, 1933, Ruska had captured images at roughly 12,000x magnification with an improved prototype, already surpassing the practical ceiling of contemporary light microscopes (TU Berlin Institutional History; Nobel Prize Foundation).

Ruska received the 1986 Nobel Prize in Physics for this foundational design work, sharing the prize with Gerd Binnig and Heinrich Rohrer, who were separately honored for inventing the scanning tunneling microscope. Notably, Max Knoll, Ruska's collaborator and the person who supervised his original 1928 diploma thesis on magnetic lens effects, had died in 1969 and could not be included, since Nobel rules bar posthumous awards (SciHi Blog, history of electron microscopy).

An electron accelerated to a typical transmission electron microscope operating voltage of 100 kilovolts has a de Broglie wavelength of roughly 3.7 picometers, about 100,000 times shorter than a wavelength of green light. Pushing the accelerating voltage to 300 kilovolts, common in modern research instruments, shortens that wavelength further, to roughly 2 picometers, smaller than most atoms.

Wavelength versus voltage: doing the math

The de Broglie relationship means an electron microscope's theoretical resolution is tunable simply by adjusting the accelerating voltage of the electron gun. Higher voltage means faster electrons, faster electrons mean shorter wavelengths, and shorter wavelengths push the Abbe diffraction limit down toward the size of individual atoms. This is the entire physical reason electron microscopy works at all: it is the same Abbe formula that limits light microscopes, just applied to a wave with a vastly smaller wavelength.

Why practical resolution never quite reaches the theoretical limit

In practice, magnetic lenses suffer from imperfections called aberrations, chiefly spherical and chromatic aberration, that blur the image well before the electron's own wavelength becomes the limiting factor. Modern instruments correct for these using additional multipole lens elements, a technique refined since the 1990s, pushing achievable resolution down toward roughly 0.05 nanometers in the best aberration-corrected transmission electron microscopes, close enough to see individual columns of atoms in a crystal lattice (University of Utah, Advanced Microscopy Core Facility).

Two very different ways to build an image

Not every electron microscope images the same way, and the distinction matters as much as the underlying physics. A transmission electron microscope (TEM) fires a broad electron beam through a specimen sliced thinner than 100 nanometers, and the pattern of electrons that pass straight through versus those that scatter off atoms forms the image, conceptually similar to how an X-ray image forms from radiation passing through tissue and bone at different rates.

A scanning electron microscope (SEM), commercialized in the 1960s and made practical largely through work by Charles Oatley's group at Cambridge University, works on an entirely different principle. Rather than transmitting electrons through a thin slice, an SEM rasters a tightly focused beam back and forth across a bulk sample's surface, detecting secondary and backscattered electrons that bounce off or get knocked loose from the surface. The detector signal is reassembled point by point into an image, producing the strikingly three-dimensional-looking topography familiar from insect close-ups and dust-particle photographs, rather than an internal cross-section.

Feature	TEM	SEM
What it images	Internal structure of ultra-thin sections	External surface topography
Sample prep	Slices under 100 nm, often stained with heavy metals	Surface coating, often gold or carbon, for conductivity
Typical resolution	Below 0.1 nanometer in aberration-corrected instruments	1 to 20 nanometers
Image formation	Electrons transmitted through the specimen	Electrons scattered or emitted from the surface
Best for	Viruses, cell organelles, crystal lattices, thin films	Insect anatomy, fracture surfaces, powders, dust

TEM and SEM solve different imaging problems with the same core physics, and many research labs run both instruments side by side.

What resolution actually means in practice

Resolution measures the smallest distance between two points that a microscope can still display as two separate features rather than one indistinguishable blur. A microscope with 1-nanometer resolution cannot show two objects sitting 0.5 nanometers apart as anything other than a single smear, even if its magnification setting is cranked far higher. Magnification and resolution are related but distinct: magnifying a blurry image further just makes a bigger blur, which is why marketing claims about magnification numbers alone can be misleading without a stated resolution figure to back them up.

Why electron microscopes need a vacuum, and what that costs

Air molecules would scatter an electron beam long before it ever reached the specimen, the same way thick fog scatters a car's headlights into a diffuse glow instead of a sharp cone. To keep the electron path clear from source to detector, electron microscopes operate inside a vacuum chamber, typically pumped down to pressures many orders of magnitude below normal atmospheric pressure.

That vacuum requirement carries a real cost for biology. Living, hydrated tissue cannot survive the combination of vacuum exposure and intense electron bombardment; water boils off instantly in vacuum, and the specimen collapses or burns under the beam. Traditional sample preparation for TEM and SEM therefore involves chemically fixing tissue, dehydrating it through a graded alcohol series, and often embedding it in resin or coating it with a thin conductive metal layer, a process that can introduce its own artifacts into the final image.

Cryo-electron microscopy, or cryo-EM, solves much of this problem by flash-freezing a hydrated specimen so fast that the water forms a glassy, non-crystalline solid, called vitreous ice, instead of ice crystals that would tear the sample apart. Jacques Dubochet, Joachim Frank, and Richard Henderson shared the 2017 Nobel Prize in Chemistry for developing this technique and the computational methods to reconstruct high-resolution 3D structures from thousands of individual frozen-particle images, a breakthrough that let structural biologists determine atomic-resolution protein shapes without needing to grow the notoriously difficult crystals that X-ray crystallography requires (NobelPrize.org, 2017 Chemistry press release).

The unglamorous work of preparing a sample

None of the optics matter if the specimen itself falls apart under the beam, and getting a sample ready for a transmission electron microscope is closer to a slow, exacting kitchen process than to snapping a photograph. A biological tissue sample first has to be chemically fixed, usually in glutaraldehyde and then osmium tetroxide, to lock its proteins and membranes in place so they stop changing shape the moment they are removed from a living organism. Only after fixation can the sample survive the dehydration, embedding, and vacuum exposure that come next.

Once fixed, the tissue is dehydrated through a graded series of alcohol or acetone baths, each step replacing more of the water with solvent, because water and resin do not mix. The dehydrated sample is then infiltrated with a liquid epoxy resin and cured into a hard block, essentially the same idea as embedding an insect in amber, except the "amber" here is a synthetic polymer chosen for how evenly it polymerizes around delicate structures. That resin block is what actually gets cut.

Why sections have to be thinner than a wavelength of light

A transmission electron microscope only works if electrons can pass through the specimen, so the resin block is sliced on a device called an ultramicrotome into sections roughly 60 to 70 nanometers thick, thinner than the wavelength of visible light itself and about a thousand times thinner than a sheet of household plastic wrap (ScienceDirect, Ultramicrotomy overview). The ultramicrotome advances the block a few nanometers at a time past a diamond or freshly cleaved glass knife edge, and the resulting ribbon of sections is floated onto water and picked up on a small copper support grid barely larger than a hole punch confetti.

Cutting a section this thin is closer to slicing a single layer off an onion with a razor than to using a kitchen knife, and a diamond knife's edge, magnified, is still smooth enough to cut cleanly at the nanometer scale without tearing the tissue. Because carbon, hydrogen, and oxygen (the elements that make up most biological tissue) scatter electrons weakly and produce almost no contrast on their own, the sections are then stained with heavy-metal salts such as uranyl acetate and lead citrate, whose electron-dense uranium and lead atoms scatter the beam strongly and outline cell membranes, organelles, and macromolecular structures that would otherwise be nearly invisible (Frontiers in Veterinary Science, sectioning and staining review). Materials scientists preparing metal or semiconductor samples skip the biological fixation steps entirely but face their own version of the same problem, thinning a solid chip down to an electron-transparent wafer using mechanical polishing followed by a focused ion beam that mills away material atom-layer by atom-layer until light can nearly pass through what is left.

Naming the first virus anyone actually saw

The gap between inferring a virus and seeing one closed on a specific date. In 1939, Gustav Kausche, Edgar Pfankuch, and Helmut Ruska, Ernst Ruska's brother, published the first electron micrographs of tobacco mosaic virus in the journal Naturwissenschaften, under the title "Die Sichtbarmachung von pflanzlichem Virus im Übermikroskop," revealing its now-famous rigid, rod-shaped particles (ResearchGate, Kausche/Pfankuch/Ruska TMV images). Tobacco mosaic virus had already been central to virology for decades, since Martinus Beijerinck had identified it in 1898 as a disease agent that passed through filters fine enough to remove all known bacteria, but until 1939 nobody knew what shape that invisible agent actually was.

That single image mattered beyond botany. Tobacco mosaic virus became the standard test object for calibrating new electron microscopes for the next several decades, the same way a resolution test chart calibrates a camera lens, precisely because its rod shape and known dimensions made it easy to check whether an instrument was performing correctly. Later electron micrographs extended the same approach to poliovirus, bacteriophages, and eventually the coronaviruses responsible for SARS and COVID-19, each new outbreak prompting electron microscopists to image the pathogen within days of receiving a sample, a diagnostic speed that traces its lineage directly back to that 1939 tobacco leaf sample.

Why this changed biology and materials science

Before 1931, virology and much of cell biology were built on inference and silhouette. Electron microscopes made viruses, ribosomes, mitochondria, and the internal scaffolding of cells directly visible for the first time, converting biology from a field that reasoned about invisible structures into one that could photograph them. The polio virus, tobacco mosaic virus, and bacteriophages, all previously known only through their filterability and infectious behavior, became objects researchers could actually see and measure.

Materials science benefited on a comparably large scale. Engineers now routinely examine grain boundaries in metal alloys, dislocations and defects in semiconductor wafers, and the atomic arrangement of new battery cathode materials using the same core instrument Ruska and Knoll built in a Berlin laboratory. That capability underpins quality control across industries from aerospace manufacturing to microchip fabrication, where a single misplaced atomic layer in a transistor can determine whether a chip functions at all.

Space science and forensics use electron microscopes too, though for different reasons: SEM analysis of meteorite fragments and gunshot residue relies on the same surface-imaging principle used to photograph insect exoskeletons, just aimed at very different subjects.

Common misconceptions

People sometimes assume electron microscopes are simply stronger versions of light microscopes, using the same optical idea turned up to a higher setting. In reality, they solve a fundamentally different physical problem. The obstacle a light microscope hits is the diffraction limit imposed by photon wavelength, not weak lenses, and no amount of lens polishing, however skilled, can push a light microscope past that wall. Only switching to a wave with a shorter wavelength, in this case the electron's de Broglie wavelength, gets around it.

Another common misconception is that electron microscope images show true color the way a photograph does. TEM and SEM images are, by their nature, grayscale maps of electron intensity, since electrons carry no color information analogous to the wavelength of visible light. Any color visible in a published electron micrograph, such as the vivid blues and greens seen in popular science coverage of viruses, has been added digitally afterward for contrast or clarity, not captured by the instrument itself.

A third misconception conflates magnification with resolution, treating a bigger displayed image as automatically a more detailed one. As covered above, magnifying past an instrument's resolution limit only enlarges the blur; useful magnification always tops out at whatever the resolution and the desired sharpness allow.

The takeaway

Ruska and Knoll did not simply build a stronger microscope; they swapped an entire category of wave, photons for electrons, to sidestep a diffraction limit that no amount of lens refinement could ever fix. That single substitution, grounded in de Broglie's 1924 wave-particle duality and realized as a working instrument by 1931, opened a scale of biology and materials science that had remained physically invisible through three centuries of optical microscopy. Everything from the structure of a ribosome to the atomic lattice of a battery electrode traces back to that Berlin laboratory bench.

---

Read next:

• The Millikan oil drop experiment explained
• The Rutherford gold foil experiment explained
• How the Geiger counter detects radiation

FAQ

Q: Who invented the electron microscope?

Ernst Ruska and Max Knoll built the first working electron microscope on March 9, 1931, at the Technische Hochschule zu Berlin, achieving two-stage magnetic-lens magnification. Ruska later won the 1986 Nobel Prize in Physics for the achievement; Knoll had died in 1969 and could not be included under Nobel rules against posthumous awards.

Q: Why can electron microscopes see smaller details than light microscopes?

Resolution is fundamentally limited by wavelength under Ernst Abbe's diffraction-limit formula, and electrons accelerated in a microscope's electron gun have de Broglie wavelengths measured in picometers, thousands of times shorter than visible light's 400 to 700 nanometer range. That shorter wavelength pushes the physical resolution limit down toward the size of individual atoms.

Q: What is the difference between a TEM and an SEM?

A transmission electron microscope (TEM) fires electrons through an ultra-thin sample to reveal internal structure, achieving resolutions below 0.1 nanometer in modern aberration-corrected instruments. A scanning electron microscope (SEM) scans a focused beam across a surface and detects scattered or emitted electrons to build a topographic image of external features, typically at 1 to 20 nanometer resolution.

Q: Why do electron microscopes require a vacuum?

Air molecules would scatter the electron beam before it reached the sample, degrading or destroying the image, the same way fog scatters a car's headlights. A vacuum chamber keeps the electron path clear from the electron gun to the detector.

Q: Can electron microscopes image living cells?

Generally no. The vacuum environment and the intense electron beam are incompatible with living, hydrated tissue, so most biological samples are chemically fixed, dehydrated, or embedded before imaging. Cryo-electron microscopy, which won its developers the 2017 Nobel Prize in Chemistry, works around this by flash-freezing hydrated specimens into vitreous ice rather than drying them, preserving near-native structure for imaging.

Q: How much shorter is an electron's wavelength than visible light?

An electron accelerated through a typical 100-kilovolt transmission electron microscope has a de Broglie wavelength of about 3.7 picometers, roughly 100,000 times shorter than the 400 to 700 nanometer range of visible light. That gap is the entire physical basis for electron microscopy's superior resolution.

Q: Why are biological samples stained with heavy metals before electron microscopy?

Biological tissue is made mostly of carbon, hydrogen, and oxygen, elements that scatter electrons too weakly to produce visible contrast on their own. Staining ultrathin sections with heavy-metal salts such as uranyl acetate and lead citrate adds electron-dense atoms that scatter the beam strongly, outlining membranes and organelles that would otherwise blend into the background.

Q: What was the first virus ever imaged with an electron microscope?

Gustav Kausche, Edgar Pfankuch, and Helmut Ruska published the first electron micrographs of tobacco mosaic virus in 1939, revealing its rod-shaped particles for the first time and finally giving a visible form to an infectious agent virologists had only inferred since Martinus Beijerinck's 1898 filtration experiments. Tobacco mosaic virus went on to become a standard calibration specimen for electron microscopes for decades afterward.

Q: How thin are the sections cut for transmission electron microscopy?

Samples are typically sliced on an ultramicrotome into sections roughly 60 to 70 nanometers thick, thinner than a wavelength of visible light and about a thousand times thinner than household plastic wrap. Sections this thin are necessary because a transmission electron microscope forms its image from electrons that pass through the specimen, and anything thicker would scatter too many of them to produce a usable picture.