A scanning electron microscope (SEM) builds images by sweeping a focused beam of electrons across the surface of a sample and collecting the signals that bounce or get knocked off at each point, pixel by pixel. The result is a highly detailed picture of surface topography and composition, with resolution that can reach the single-nanometer range in modern instruments. Unlike an optical microscope, which relies on light passing through or reflecting off a sample, an SEM uses electrons whose wavelengths are far shorter, allowing it to resolve features hundreds of times smaller than what visible light can show. That combination of high magnification, sharp depth of field, and surface sensitivity is why SEMs have become standard tools in fields from materials science and semiconductor manufacturing to biology and forensics.
How the Beam and Signals Work
At the top of the instrument sits an electron source, often called the “gun.” Three main types exist. Thermionic guns heat a tungsten filament or a lanthanum hexaboride crystal until electrons boil off; they are affordable and robust but produce a relatively fat beam. Field-emission guns pull electrons from an extremely sharp tip using a strong electric field, yielding a much brighter, finer probe. A middle-ground option known as the Schottky emitter combines heat with an electric field and is generally considered superior to both conventional field emission and pure thermionic sources for low-voltage work.1Journal of Microscopy. Thermal field emission for low voltage scanning electron microscopy The gun type matters because it determines how small the spot on the sample can be, which in turn sets the resolution ceiling.
Electromagnetic lenses below the gun focus the beam down to a tiny probe, and scan coils raster it across the sample surface in a grid pattern. When that probe hits the sample, two signals matter most. Secondary electrons (SE) are low-energy electrons knocked loose from atoms near the very surface. Because they carry information about surface shape, they are the main source of topographic contrast, giving SEM images that familiar three-dimensional look. Backscattered electrons (BSE) are higher-energy electrons from the beam that ricochet back out of the sample after interacting with atomic nuclei deeper inside. Heavier elements scatter more electrons, so BSE signals reveal differences in chemical composition as brightness differences in the image.
Detectors and Why They Matter
The most widely used detector in SEMs is the Everhart-Thornley detector (ETD), introduced in 1960. It sits off to one side of the sample and uses a small voltage bias to attract secondary electrons toward a scintillator-photomultiplier assembly. The catch is that the ETD also picks up backscattered electrons, and for samples made of heavy elements the BSE signal can actually dominate, drowning out the pure surface topography information the operator wants.2Microscopy and Microanalysis. Simultaneous Scanning Electron Microscope Imaging of Topographical and Chemical Contrast Using In-Lens, In-Column, and Everhart–Thornley Detector Systems The off-axis position of the ETD also creates an asymmetric intensity profile on topographic features, producing a shadow-like contrast that can look misleading if you are not aware of it.3Japanese Journal of Applied Physics. A Simulation of the Topographic Contrast in the SEM
Modern instruments get around these limitations by placing additional detectors inside the column, directly above the sample. In-lens and in-column detectors preferentially collect the secondary electrons generated right at the beam’s landing point, filtering out the scattered and re-emitted electrons that blur spatial resolution. Running multiple detector types simultaneously lets the operator separate topographic contrast from compositional contrast in a single imaging session. For nanoparticle work on complex substrates, contrast inversions can occur as you change beam energy or working distance, meaning a particle that looks brighter than its substrate at one setting can flip to darker at another.4PubMed Central. Contrast of Backscattered Electron SEM Images of Nanoparticles on Substrates with Complex Structure Knowing which detector you are using and at what conditions is essential for interpreting what you see.
Preparing a Sample for the Vacuum
A conventional SEM operates under high vacuum, typically on the order of a hundred-thousandth of an atmosphere or lower. That vacuum is necessary so the electron beam can travel without scattering off gas molecules. But it creates a problem: anything wet, oily, or outgassing cannot go in as-is. Biological specimens in particular must be fixed, dehydrated, and dried before imaging.
The gold-standard drying technique for biological tissue is critical point drying (CPD). The sample is soaked in liquid carbon dioxide inside a pressurized chamber, then the temperature and pressure are raised past the critical point where liquid and gas become indistinguishable. The fluid can then be vented without ever forming a liquid-gas boundary, which is what causes surface-tension forces that collapse delicate structures. CPD works well but requires specialized equipment and careful handling.
A cheaper alternative is chemical drying with hexamethyldisilazane (HMDS). You soak the dehydrated sample in HMDS, then let it evaporate in air. For many animal tissues, HMDS produces results indistinguishable from CPD.5PubMed. Comparison of hexamethyldisilazane (HMDS), Peldri II, and critical-point drying methods for scanning electron microscopy of biological specimens The same study found that plant tissues with large fluid-filled vacuoles still showed shrinkage artifacts regardless of method, though CPD generally outperformed HMDS for those samples. More recent work on extremely delicate structures like glandular trichomes found HMDS preservation comparable to CPD, while simple air drying caused severe distortion.6PubMed Central. Replacing critical point drying with a low-cost chemical drying provides comparable surface image quality of glandular trichomes from leaves of Millingtonia hortensis L. f. in scanning electron micrograph For cell biologists studying actin networks, HMDS actually preserved the fine cortical mesh better than standard CPD, which introduced larger artifactual holes in the network.7PLoS ONE. Scanning electron microscopy preparation of the cellular actin cortex: A quantitative comparison between critical point drying and hexamethyldisilazane drying The takeaway is that no single drying method is universally best; the choice depends on the type of tissue and the features you need to preserve.
Non-conductive samples face a second preparation hurdle: charging. When electrons from the beam build up on an insulating surface, they create a local electric field that deflects incoming electrons and produces bright streaks or dark patches in the image. The classic fix is to sputter-coat the sample with a thin layer of metal, often gold, platinum, or a gold-palladium alloy, just a few nanometers thick. The coating drains charge away and also boosts secondary electron yield, improving signal. For delicate or beam-sensitive specimens, carbon coating is gentler, though it gives less secondary electron signal enhancement.
Environmental and Low-Vacuum SEM
Environmental SEM (ESEM) was developed specifically to image samples that would be destroyed or fundamentally altered by high vacuum or metal coating. An ESEM allows a small amount of gas in the specimen chamber, up to about 10 Torr. When that gas is water vapor, hydrated samples can be maintained in something close to their native state.8PubMed. The use of environmental scanning electron microscopy for imaging wet and insulating materials The gas molecules serve double duty: electrons coming off the sample collide with gas molecules and create positive ions, which drift back toward the surface and neutralize charge buildup, eliminating the need for a conductive coating on insulators.
ESEM has found wide use for imaging cultured cells, colloidal particles suspended in liquid, polymers, and geological specimens that would crack or dehydrate under conventional vacuum. It has also been used to study the behavior of water droplets on surfaces, since you can directly observe liquid water in the chamber. There are limits, though. Some specimens that contain a lot of free water lose significant moisture during the initial pump-down before the chamber reaches a stable low-vacuum state, so truly aqueous samples remain challenging.9Microscopy. Observation of wet specimens sensitive to evaporation using scanning electron microscopy Resolution in ESEM is also somewhat lower than in a high-vacuum SEM because beam electrons scatter off the chamber gas.
Low-Voltage Imaging and Surface Sensitivity
Conventional SEM imaging often uses beam energies between 10 and 30 keV. Dropping below 5 keV, into the low-voltage regime, changes the physics in useful ways. The electrons penetrate less deeply, so the signal comes from a shallower volume closer to the true surface. That makes low-voltage SEM more sensitive to thin surface layers, coatings, and contamination. It also reduces beam damage on delicate samples and generates less internal charging on insulators.10Materials Characterization. High-Resolution and Low-Voltage FE-SEM Imaging and Microanalysis in Materials Characterization Field-emission guns are essentially required for good low-voltage performance because the beam must remain tightly focused even at low energy, and thermionic sources do not provide enough brightness under those conditions.
Low-voltage SEM has become especially important in the semiconductor industry, where engineers need to inspect features a few nanometers wide without damaging the device being measured. Running at 1 keV or lower can image the outermost few nanometers of a surface, which is helpful for checking thin-film uniformity or spotting surface contamination that would be invisible at higher voltages.
FIB-SEM and Three-Dimensional Reconstruction
Some SEMs are paired with a focused ion beam (FIB), typically using gallium ions, in a single instrument called a dual-beam or FIB-SEM system. The ion beam can mill away material with nanometer precision while the electron beam images the freshly exposed surface. By alternating between slicing and imaging, the instrument collects a stack of sequential cross-section images that can be assembled into a full three-dimensional reconstruction of a material’s internal structure.11Nanomanufacturing and Metrology. Enhancing 3D Reconstruction Accuracy of FIB Tomography Data Using Multi-voltage Images and Multimodal Machine Learning
This technique, often called FIB tomography, has been applied to everything from fuel-cell electrodes to battery cathodes to biological tissue.12ECS Transactions. Three Dimensional Reconstruction of Solid Oxide Fuel Cell Electrodes Using Focused Ion Beam – Scanning Electron Microscopy When the sample is frozen beforehand, the approach is called cryo-FIB-SEM and allows cross-sectioning of hydrated biological specimens that cannot survive room-temperature milling.13PubMed Central. An introduction to cryo-FIB-SEM cross-sectioning of frozen, hydrated Life Science samples Cryo-SEM on its own, without the FIB, can also reveal internal microstructure by fracturing frozen specimens and imaging the exposed surfaces. One application used cryo-SEM to visualize how water penetration changes the microstructure of skin layers, providing insight into how large drug molecules might cross the skin barrier.14Journal of Pharmaceutical Sciences. Hydration Effects on Skin Microstructure as Probed by High-Resolution Cryo-Scanning Electron Microscopy and Mechanistic Implications to Enhanced Transcutaneous Delivery of Biomacromolecules
Analytical Techniques Built into the SEM
The SEM’s electron beam does more than generate images. When it knocks inner-shell electrons out of atoms in the sample, the resulting X-rays have energies characteristic of each element. An energy-dispersive X-ray spectroscopy (EDS) detector mounted inside the chamber captures those X-rays and identifies the elements present, producing either point analyses, line scans, or full elemental maps overlaid on the SEM image. EDS is standard on most modern SEMs and is the first-line tool for determining what a sample is made of at the micro- and nanoscale.
Electron backscatter diffraction (EBSD) adds crystallographic information. By tilting the sample steeply toward a phosphor screen, the diffracted backscattered electrons form patterns that reveal the crystal orientation at each point. Scanning across the surface yields maps of grain size, grain boundary character, phase distribution, and local texture.15Acta Materialia. Characterization of the microstructure and texture of nanostructured electrodeposited NiCo using electron backscatter diffraction (EBSD) Metallurgists use EBSD routinely to understand how processing conditions affect the internal grain structure of metals and alloys.
Cathodoluminescence (CL) exploits the fact that some materials emit visible or near-visible light when struck by the electron beam. The emitted photon wavelengths depend on the material’s bandgap, impurities, and crystallographic defects, making CL a powerful probe of optical properties in semiconductors, minerals, and certain ceramics. With hyperspectral imaging, each pixel in the SEM image gets a full emission spectrum, mapping spatial variations in luminescence at a resolution of tens of nanometers.16Semiconductor Science and Technology. Structural and luminescence imaging and characterisation of semiconductors in the scanning electron microscope Researchers have even identified different types of nanoparticles by their CL color, using nanodiamonds with nitrogen-vacancy centers for red emission, cerium-doped garnet nanophosphors for green, and nanodiamonds with band-A defects for blue, enabling multicolor correlative imaging under electron-beam excitation.17Scientific Reports. Correlative light and electron microscopy using cathodoluminescence from nanoparticles with distinguishable colours
Image Artifacts and How to Spot Them
SEM images look convincingly photographic, which can create a false sense of objectivity. Several artifacts routinely creep in. Charging artifacts, mentioned earlier, appear when insulating regions accumulate electrons and produce bright flaring or dark voids. Contamination artifacts appear as dark rectangles or streaks in areas that have already been scanned, because the beam cracks hydrocarbon residues in the chamber into a carbon film that changes local secondary electron yield.
Scan line shift artifacts are less obvious. If the beam position drifts between scan lines, vertical features in the image become jagged or distorted. These positional errors produce localized distortions that are especially problematic when the SEM image is being used for precise dimensional measurements or digital image correlation strain mapping.18Ultramicroscopy. Correction of scan line shift artifacts in scanning electron microscopy: An extended digital image correlation framework Vibration, electromagnetic interference, and even thermal drift in the stage can all contribute. Operators learn to recognize these distortions, but automated measurement workflows need explicit correction algorithms to deal with them.
Edge effects are another common trap. At steep topographic edges, secondary electron emission spikes because electrons generated just inside the edge can escape from multiple surfaces. The result is a bright halo around any sharp feature that makes it look slightly larger than it really is. For dimensional metrology at the nanoscale, this “blooming” effect has to be carefully calibrated out.
Where SEM Fits Compared to Other Microscopy
A common question is when to use an SEM rather than a transmission electron microscope (TEM) or an atomic force microscope (AFM). TEM passes electrons through a very thin sample and can achieve atomic-level resolution, but it requires elaborate sample thinning and provides a two-dimensional projection, not a surface view. AFM drags or taps a tiny physical probe across the surface and can measure height with sub-nanometer precision, but it is slow and limited to small scan areas. SEM sits between them in resolution and offers the widest field of view with full three-dimensional surface rendering.
A direct comparison of these methods for measuring nanoparticle dimensions found that SEM works well for particles above about 50 nanometers in diameter and is especially suited to metallic particles, while AFM and TEM give more accurate results with smaller particles. Dynamic light scattering, a solution-based sizing technique, was found to be unreliable for polydisperse or mixed-size samples.19PubMed. A direct comparison of experimental methods to measure dimensions of synthetic nanoparticles In practice, many labs use SEM as their first-look instrument and turn to TEM or AFM for the cases where SEM’s resolution is not enough or where the sample requires a different geometry.
In-Situ Testing Inside the SEM
One of the more dramatic uses of an SEM is watching a material deform or fracture in real time under load. Miniaturized tensile and compression stages have been developed to fit inside the SEM chamber, allowing researchers to apply force to a small specimen while continuously imaging its surface at high magnification. These in-situ setups can capture crack initiation, slip-band formation, and other microstructural changes that are impossible to observe after the fact on a broken sample.20Measurement. Development of in-situ SEM testing apparatus for observing behavior of material at high magnification during tensile test
One recent design uses a pneumatic bellows actuator to load the specimen and a displacement sensor for direct monitoring, keeping the device compact enough to fit inside a standard SEM chamber. This allowed researchers to track how microstructural defects evolved in metals under increasing tensile stress.21Metals. Design and Application of a Miniature Pneumatic Bellows Loading Device for In-Situ Tensile Testing inside the Scanning Electron Microscope In-situ SEM experiments are increasingly being combined with EBSD mapping so that crystal orientation changes during deformation can be tracked grain by grain.
Automation and Benchtop Instruments
For most of its history, the SEM was a large, expensive instrument housed in a dedicated laboratory with vibration isolation, climate control, and a trained operator. That is still the case for high-end research instruments, but the past couple of decades have seen a proliferation of compact benchtop SEMs that cost a fraction of the price and require far less infrastructure. These smaller machines sacrifice some resolution and analytical flexibility, but they bring SEM capability to quality-control labs, teaching settings, and field-adjacent facilities that would never justify a full research instrument.
Automation has also accelerated. Software can now drive the SEM stage, acquire images, and run EDS analysis on thousands of particles without human intervention. One recent framework uses automated stage movement on a benchtop SEM to count and classify elongated mineral fibers in air filter samples, a task that was historically done by hand under a microscope.22Atmospheric Environment. A novel approach for quantifying elongated airborne mineral particles (EMPs) using an automated scanning electron microscope (SEM) Another group developed an end-to-end Python-based framework called AutoEMX that combines automated SEM-EDS measurements with machine-learning analysis to identify and quantify the compositions of individual phases within powder mixtures, demonstrated on a desktop-class instrument.23Nature Communications. Accurate SEM-EDS quantification, automation, and machine learning enable high-throughput compositional characterization of powders This kind of high-throughput, software-driven operation is turning the SEM from a scientist’s exploration tool into a routine analytical workhorse.
A Brief History of the Instrument
The core idea of scanning a focused electron beam across a surface and collecting the resulting signals dates to the 1930s, when Max Knoll in Germany demonstrated the first crude scanning images. Manfred von Ardenne built an improved scanning transmission electron microscope around 1938, and around the same time Vladimir Zworykin’s group in the United States developed their own scanning instrument. None of these early systems produced images good enough for practical use. The real transformation came after World War II, when Charles Oatley at Cambridge University in England revived the concept through a series of doctoral projects. His students, including Dennis McMullan and Kenneth Smith, gradually solved the engineering problems of beam brightness, signal detection, and scan stability. The first industrial application of a Cambridge-built SEM was at a pulp and paper research institute in Canada, followed soon after by semiconductor inspection work at Westinghouse in the United States. Commercial SEMs became available in the mid-1960s from manufacturers in England and Japan, and the technology spread rapidly from there.24Surface and Interface Analysis. The early history and future of the SEM
Since those early decades, the instrument has gained field-emission guns, digital image acquisition, environmental chambers, integrated ion beams, and software-driven automation. Yet the basic operating principle, raster a fine electron probe and build a picture from the response, is the same one Knoll sketched out nearly a century ago. What keeps changing is how small the probe can be, how many types of signal can be collected at once, and how much of the workflow a computer can handle without a human at the controls.