What Is Atomic Force Microscopy? Operation, Uses, and Limitations

It is a scanning probe technique that moves a sharp probe tip across a specimen and measures forces between the tip and sample. The instrument can map nanometer-scale surface topography, roughness, adhesion, stiffness, friction, and conductivity without relying on visible light.

This guide explains what atomic force microscopy is, how its probe, cantilever, and feedback system operate, and which techniques help materials scientists, engineers, and biologists characterize surfaces across diverse samples.

Surface Imaging Beyond Visible Light

A probe records interactions too small for unaided vision, including features separated by only a few nanometers. Software converts the probe’s movement into a three-dimensional surface map, similar to tracing a miniature landscape with a needle. Your image can reveal ridges, pits, terraces, and particles that appear flat in an optical micrograph.

How AFM Differs from Other Imaging Methods

A scanning probe microscope uses a physical probe to examine a surface point by point. Optical microscopy depends on visible-light illumination, while electron microscopy uses a focused electron beam. The probe-based approach can examine samples directly in air, liquid, vacuum, or a controlled atmosphere without requiring a conductive coating.

Resolution can reach the nanometer or atomic scale, but your result depends on the probe, instrument, environment, sample stability, and feedback performance. That local detail supports work in materials science, nanotechnology, electronics, coatings, polymers, minerals, and biological research.

You gain more from a scan when the measurement matches your question. A height image is suitable for particle dimensions, while a force curve can show how the material deforms or resists separation. A single scan may combine both, provided the probe and operating mode suit the surface.

Information Available from Surface Scans

• ▸Surface topography: You can map particles, grooves, terraces, layers, and defects at nanometer-scale resolution.
• ▸Roughness measurements: Instrument software can calculate height variation across a selected surface region.
• ▸Mechanical properties: Force curves can reveal adhesion, stiffness, friction, deformation, or elasticity.
• ▸Electrical response: Specialized configurations can track local conductance or resistance near the surface.
• ▸Environmental behavior: You can scan dry specimens in air or examine hydrated materials in liquid.

These outputs answer different questions and should not be treated as interchangeable. A height map primarily shows shape, whereas a force curve records mechanical response. Choose the data type that directly supports the property you need to measure.

Once the right data type is chosen, the instrument’s probe, feedback loop, and operating parameters determine how reliably that measurement emerges.

How the Probe, Cantilever, and Feedback System Work

The measurement begins with a sharp probe tip whose apex may measure only a few nanometers across. The tip is attached to a narrow, flexible beam called a cantilever. Interatomic forces bend that beam as it moves over local features, allowing the instrument to detect changes smaller than the tip’s nominal shape.

A laser deflection system directs a laser beam onto the reflective back of the cantilever. Surface-induced movement shifts the reflected beam, and a position-sensitive photodetector converts that shift into an electrical signal. The controller interprets the signal as vertical motion of the probe.

Because every motion value is tied to a known position on the specimen, software combines thousands or millions of measurements. Your resulting map records surface geometry, a force-related signal, or both, depending on the instrument configuration and selected mode.

Probe-Sample Interactions

Mechanical, electrostatic, chemical, and adhesive forces can act between your probe and specimen simultaneously. Short-range chemical forces, electrostatic attraction, and Van der Waals forces influence the signal, while elastic response and local deformation can bend the cantilever.

This mixed interaction creates the central interpretation problem. Your tip may detect a feature while also compressing, pulling, contaminating, or otherwise changing it. Tip geometry, applied force, scan speed, environment, calibration, and sample rigidity therefore belong in your interpretation.

You cannot safely treat every height change as an undisturbed surface feature. A wide groove may reflect convolution between the tip and specimen, while a raised particle may displace the cantilever in a way that changes its apparent width. Careful acquisition helps separate sample structure from probe response.

Feedback and Vertical Movement

Feedback maintains a selected relationship between the tip and surface while the scanner moves in two lateral directions. The controller continually adjusts vertical position, reducing the chance that a steep rise will push the probe away or a deep drop will strike it.

Constant-force operation favors direct topographic tracing, especially on a stable surface. Constant-height operation retains a more fixed tip position and can collect force-related data more directly, but it increases collision risk across rough topography.

Your feedback setting shapes both the image and the analysis available from it. A setting that produces a clear height trace may distort a soft specimen, while a lighter setting may improve delicate-sample protection but reduce signal strength. Test the load on a small area before committing to a full scan.

“Inspecting a labeled scan without its scale bar or scan conditions resembles judging a map without knowing its size. Record the pixel size, scan area, speed, mode, probe, loading conditions, and calibration with your data.

Topography, Roughness, and Material Properties

A height image records probe displacement rather than reflected color or electron intensity. As feedback follows the surface, the instrument stores vertical movement, and software converts those values into a raster image. You can then examine height changes, terraces, particles, grooves, pits, and step edges.

Roughness values summarize vertical variation over a measured area. Common parameters include average roughness, root-mean-square roughness, peak-to-valley height, and the proportion of an area occupied by defects. Your numbers remain dependent on scan size, flattening, plane correction, tip shape, and the selected calculation method.

Beyond height, an instrument can record force-related signals while preserving the same spatial map. Those measurements may support analysis of adhesion, elasticity, friction, conductivity, or molecular organization. One scan can relate surface shape to a local property, although it does not produce a chemical composition map without another technique.

Keep the scan size, processing, probe, and operating conditions consistent when you compare specimens. Otherwise, an apparent change in roughness or contrast may come from the measurement setup. Repeated scans also help you determine whether a feature is reproducible rather than an isolated artifact.

Because setup effects can mimic genuine surface features, choosing a mode that preserves reproducibility is essential before comparing properties.

Operating Modes Matched to Samples and Goals

The operating mode determines whether your tip presses against the surface, taps it briefly, or remains above it. Each choice balances mechanical sensitivity against interference, signal strength, scan speed, and protection from sample damage.

Contact Mode

During scanning, this mode keeps the sharp probe tip pressed continuously against the specimen. Strong probe-sample interaction can produce a strong topographic signal and direct force information, but that same load creates mechanical stress. Soft polymers, biological specimens, and some coatings may stretch, smear, or wear under the tip.

Tapping Mode

An oscillating cantilever scans the surface while its tip makes brief, repeated contact. This intermittent interaction reduces continuous wear and makes the mode a common choice for many moderately soft polymers and coatings. Oscillation amplitude, resonance, feedback, and scan speed still affect image quality.

Non-Contact Mode

Weak interactions are sensed from above the specimen without the tip touching its surface. Delicate surfaces may experience less mechanical disturbance, and force-gradient measurements can reveal surface organization or electrostatic behavior. The weaker signal makes vibration isolation, tuning, and sensitive detection more demanding.

Select the mode from your measurement goal rather than image dimensions alone. Start with surface hardness, required resolution, contamination level, probe shape, operating environment, and whether you need height, stiffness, adhesion, or friction data.

A hard mineral and a biological cell cannot safely share the same loading plan, even when both fields of view have the same lateral size. A low-load approach may protect the cell but sacrifice signal, while a strong signal may flatten the membrane. You should balance resolution against sample integrity before collecting the final dataset.

Applications Across Materials, Devices, and Biology

One reason the technique became valuable is its ability to examine surfaces in several environments. Air and vacuum support many solid-state studies, while liquid operation allows observation of hydrated films, electrochemical surfaces, proteins, and biological structures under controlled conditions. Early work at IBM helped establish scanning probe microscopy as a major nanoscale measurement field.

Materials and Thin Films

Nanomaterials and thin films provide clear, measurable examples of the technique’s value. A scan can measure a nanoparticle’s height and diameter, compare coating uniformity, or identify terraces in a crystalline layer. After a wear experiment, the same instrument can chart grooves, exposed substrate, and debris.

Layered materials also expose a practical limitation because a tip sees the upper surface before deeper structure. Your height measurement may record a single terrace rather than the full film thickness. Cross-sectional preparation or a complementary method may be necessary when internal layers matter.

Semiconductors and Electronic Devices

Local surface control matters in semiconductor and electronic-device development because small defects can affect performance. Engineers examine lithographic features, conductive traces, interfaces, and contamination before those details become larger manufacturing problems. You can relate a visible defect to its height, roughness, or electrical response.

Your scan settings must also account for the device under test. A conductive coating, fragile interconnect, or movable microstructure can respond differently to probe pressure and scan speed. You should locate the feature first, then select a scan area and loading plan that preserve the device.

Biological Samples

Biological-cell membranes, immobilized proteins, lipid structures, and polymeric biomaterials can all be examined with controlled probe loading. Preparation may nevertheless change the features you intend to measure. Buffer conditions, adhesion, scan rate, probe force, and dehydration can alter biological structure.

Your preparation should match the measurement environment rather than rely on a generic dry-surface workflow. A liquid scan can preserve a hydrated sample, but it may require a suitable substrate and probe. A crisp image still demands careful interpretation when immobilization or dehydration contributes to the observed structure.

Because successful imaging can still distort or misread delicate structures, its apparent quality must be checked against resolution limits and artifacts.

• ▸Particle measurement: You can measure a deposited particle’s height and examine the surrounding coating.
• ▸Coating comparison: A force map can help show that one coating is softer than its substrate.
• ▸Electrical mapping: A current map can identify conductive pathways across an otherwise smooth surface.
• ▸Biological comparison: You can compare membrane texture before and after a controlled exposure.
• ▸Wear assessment: A scan can chart grooves, debris, exposed substrate, and changes in surface roughness.

Resolution Limits and Scan Artifacts

A sharp tip can improve spatial detail, but no probe behaves as a perfect mathematical point. Tip-sample convolution can make a narrow groove appear wider and a tall object appear rounded. Atomic-scale claims require a stable instrument, low noise, a suitable cantilever, careful calibration, and unusually rigid sample support.

Your stated resolution must include the conditions that produced it. A nanometer-scale map obtained in air on a hard crystal does not establish the same capability for a soft polymer in liquid. Report the scan environment, probe geometry, load, feedback settings, and processing method with any dimensional claim.

Common Artifacts and Their Causes

• ▸Feature broadening: Tip diameter makes a narrow object appear wider than its actual dimensions.
• ▸Rounded features: Tip shape can compress a tall particle and soften its sharp profile.
• ▸Scan lines: A feedback delay may cause the tip to miss raised or lowered structures.
• ▸Drift rings: Thermal change or slow movement can create repeated circular patterns.
• ▸Contamination loops: Debris on the tip creates repeated features as the probe scans.
• ▸Electrical noise: Interference can alter force signals, phase measurements, or current maps.

Moving too rapidly reduces the controller’s response and produces poorly tracked features. Environmental vibration, electrical noise, contamination, poor calibration, and an unstable mount can also create lines, rings, drift, or repeated patterns that resemble genuine structure. You should inspect such patterns before assigning them physical meaning.

Change the suspected source rather than simply applying stronger image processing. A fresh probe can test contamination, slower scanning can test feedback delay, and repeated scans can test drift. Your correction should address the cause rather than remove evidence that resembles a feature.

Speed, Area, and Sample Protection

AFM scans are slower than many optical or electron-beam acquisition systems, and a rough surface can damage the probe. The instrument measures a local region rather than covering a large specimen instantly. Tip wear changes the contact geometry, so a long scan may require frequent probe checks or a fresh tip.

Resolution also involves a tradeoff between lateral sampling and measurement time. A smaller scan may resolve finer lateral steps, but the shorter duration can make drift more influential. A larger field reveals broader context but may hide small defects unless you select a suitable pixel count and scan rate.

AFM provides a major advantage when you need nanoscale surface contours, local roughness, or force-based properties at one location. SEM offers faster, broader surface coverage, TEM reveals internal structure, and optical microscopy supports rapid inspection when its resolution is sufficient.

You can also combine methods rather than forcing one instrument to answer every question. SEM can locate a defect across a wider area, after which AFM can measure its height and local response. This sequence reduces the time spent searching with a slow local scanner and preserves a clear link between context and measurement.

Bottom Line

A moving probe’s displacement and force response become nanometer-scale surface data through atomic force microscopy. Its strongest capability is the combination of topography and local property measurement, but that value depends on tip condition, sample support, mode selection, calibration, and careful interpretation.

Choose AFM when the local surface and its mechanical or electrical response matter more than speed or wide-area coverage. Start with the property you need, match the mode and probe to the material, and preserve the scan conditions with your results. You can then distinguish genuine specimen behavior from probe or environmental artifacts.

FAQ