AFM sample preparation for reliable nanoscale imaging

Why preparation controls AFM data quality

AFM sample preparation is more than a housekeeping step. It determines whether an image represents the surface itself or a mix of real topography, loose debris, dried salts, adhesive residue, and tip-sample artifacts. A useful AFM specimen is clean, stable, firmly mounted, compatible with the imaging environment, and flat enough for the scanner and probe to maintain feedback. IUPAC describes atomic force microscopy as high-resolution scanning probe microscopy that maps surfaces with resolution on the order of fractions of a nanometer, which is why small residues and weakly attached particles can become visible measurement errors. For related laboratory workflows, see our sample preparation resources. (goldbook.iupac.org)

The practical goal is simple: preserve the feature you need to measure while removing anything that could move, smear, swell, charge, dissolve, or contaminate the probe. The details vary by sample type, but the same logic applies to hard coatings, wafers, powders, polymers, proteins, DNA, and cells.

The preparation goal before the protocol

Before selecting a solvent, substrate, or adhesive, define the measurement question. A roughness measurement on a ceramic thin film, a height measurement on nanoparticles, a phase image of a polymer blend, and a force map of living cells all require different trade-offs. Surface roughness work usually puts flatness, cleanliness, and stable mounting first. Biological imaging may instead require hydration, gentle immobilization, and preservation of native structure.

The 2024 edition of ISO 19606 addresses AFM-based surface roughness measurement for fine ceramic films and specifies a method for evaluating probe-tip adequacy in a defined roughness range. While it is not a universal preparation protocol, it illustrates an important metrology point: AFM results depend on the interaction among the sample surface, probe, scanner, vibration environment, and measurement method, not on the specimen alone. (iso.org)

Cleanliness

Because AFM is a surface-sensitive method, anything left on the surface can be imaged as if it were part of the sample. Finger oils, polishing residue, salts from buffer, airborne dust, dried droplets, and adhesive outgassing can all distort height and phase information. Bruker educational material for materials AFM notes that dirt or dust on a surface will be measured as part of the sample, and it recommends clean handling environments when appropriate. (bruker.com)

Flatness

Flatness does not mean featureless. It means the height variation must stay within the scan range, feedback capability, and measurement objective. A very rough coupon may still be suitable for micron-scale morphology, but it may be unsuitable for sub-nanometer roughness. For cross-sections, coatings, and wafers, polishing and inspection are often preparation steps, not cosmetic steps.

Immobilization

Loose particles, soft films, proteins, and cells must be immobilized strongly enough to withstand scanning forces. Stronger attachment, however, is not always better. Coatings such as poly-L-lysine can improve attachment to mica or glass, but they also add background, change surface chemistry, and may affect delicate specimens. The right choice is the least disruptive method that keeps the target structure stable during imaging.

Preparation choices by sample type

Sample type	Common preparation choice	Main risk to control
Hard films, wafers, ceramics, metals	Cut, polish if needed, clean with compatible solvents, dry with filtered gas, mount rigidly	Scratches, polishing residue, oils, tilt, excessive roughness
Nanoparticles or powders	Dilute suspension, deposit on clean mica, silicon, glass, or graphite, rinse if compatible, dry gently	Aggregation, coffee-ring deposits, loose particles, salt crystals
Thin polymer films	Spin coating, casting, microtoming, or gentle deposition on a flat substrate	Solvent swelling, dewetting, thermal history, tip-induced deformation
Proteins, DNA, biomolecules	Freshly cleaved mica or treated mica, controlled buffer, gentle rinse, air or liquid imaging depending on objective	Poor adhesion, denaturation, dried salts, background from surface treatments
Cells and soft biological samples	Glass or mica with suitable coating, liquid imaging when native hydration matters, fixation only when compatible with the question	Dehydration, over-fixation, weak attachment, altered mechanics

For biological AFM, instrument handbooks and protocols commonly emphasize that mica and glass are common substrates, poly-L-lysine is often used to attach cells or proteins, and fixation must be selected carefully because dehydration and residual fixative can change surface structure or interfere with scanning. (nanophys.kth.se)

A practical AFM sample preparation workflow

1. Define the imaging environment

Decide whether the experiment will be performed in air, liquid, controlled humidity, inert gas, or another specialized environment. Dry imaging often simplifies mounting and reduces fluid-related drift, but it may be unsuitable for hydrated biological structures or samples that change during drying. Liquid imaging can help preserve some native conditions, but it increases the importance of clean buffers, stable immobilization, and control experiments.

2. Select a substrate that is flatter than the feature of interest

Mica is widely used because fresh cleavage exposes a clean, very flat surface suitable for many biomolecules and nanoparticles. Silicon, glass, and graphite are also common, depending on conductivity, optical compatibility, chemistry, and roughness requirements. The substrate should contribute less background than the feature being measured. If a 2 nm particle is deposited on a 5 nm rough substrate, height statistics will be difficult to interpret.

3. Clean without creating a new artifact

Choose cleaning steps that match both the specimen and the measurement. Solvent rinsing, ultrapure water, filtered nitrogen drying, plasma cleaning, UV-ozone treatment, and fresh cleavage can all be useful, but none is universally safe. Plasma may modify polymers. Water may swell hydrophilic films. Acetone may attack adhesives or residues. For surface analysis more broadly, ASTM guidance on specimen preparation stresses minimizing preparation effects that could influence surface-sensitive measurements. (store.astm.org)

4. Deposit or attach the sample under controlled conditions

Use low-concentration suspensions for particles and biomolecules unless aggregation is the measurement target. Deposit small volumes, avoid touching the surface with pipette tips, and keep incubation time consistent between replicates. Some protein protocols use freshly cleaved mica, short incubations, rinsing to remove salts, and drying before tapping-mode imaging. These details should be adapted rather than copied directly, because protein concentration, buffer composition, and surface charge strongly affect adsorption. (stressmarq.com)

5. Mount the specimen so it cannot drift or rock

Mechanical instability often looks like poor AFM performance. Mount the sample on a puck or holder with minimal tilt and no loose edges. Avoid excess tape, wet glue, soft adhesive layers, and overhanging fragments near the scan area. If a conductive or electrical AFM mode is planned, confirm that the mounting path is electrically appropriate and that insulation or oxide layers are intentional rather than accidental. See also: analytical methods.

6. Inspect before scanning

Use an optical microscope when possible. Look for dust, scratches, droplets, salt crystals, cracks, aggregates, edge beads, bubbles, and uneven coating. AFM is slow compared with optical inspection, so a quick pre-check can prevent wasted scan time on a surface that is already visibly compromised.

Common artifacts caused by weak preparation

Observed problem	Likely preparation cause	Corrective action
Streaks in the scan direction	Loose debris, particles not immobilized, sticky residue, or contaminated tip	Improve cleaning, reduce concentration, rinse gently, use a fresh substrate, check tip condition
Large isolated spikes	Dust, salt crystals, aggregates, or dried droplets	Filter solutions, use cleaner handling, rinse compatible samples, dry from one edge
Features appear too wide	Tip convolution, high aspect-ratio structures, blunt or contaminated probe	Use an appropriately sharper probe, compare known standards, report lateral dimensions cautiously
Unstable feedback	Excessive roughness, poor mounting, soft surface, liquid instability, or loose sample	Reduce scan size, improve mounting, select a gentler mode, or prepare a flatter area
Background masks target features	Rough substrate, coating layer, polymer residue, or drying stain	Image a blank substrate, change substrate, dilute sample, adjust deposition method

Treat preparation artifacts as competing explanations, not as afterthoughts. If a feature appears only in one area, follows the scan direction, changes after repeated scanning, or also appears on a blank control, it may not represent the intended specimen.

Documentation that makes AFM results reproducible

AFM images are easier to interpret when preparation metadata are recorded with the same care as scan parameters. At minimum, document the substrate type, cleaning method, deposition volume, sample concentration, incubation time, rinsing method, drying method, imaging medium, mounting adhesive, probe model or nominal properties, and storage time before measurement.

For quantitative work, include blank-substrate scans and replicate areas. Roughness values are especially sensitive to scan size, filtering, line correction, probe condition, and surface defects included in the analysis. If the result will be compared across instruments or laboratories, record the analysis settings rather than only the final number.

Frequently asked questions

Should AFM samples always be dry?

No. Dry samples are common for hard materials, many thin films, and deposited particles, but drying can deform soft or hydrated specimens. Cells, biomolecules, hydrogels, and some polymers may require liquid imaging or controlled humidity to answer the intended question.

Is mica always the right AFM substrate?

No. Mica is useful because it can provide a clean, flat surface after cleavage, but it is not automatically compatible with every sample. Silicon, glass, graphite, metal-coated substrates, or functionalized surfaces may be better when conductivity, optical transparency, chemistry, adhesion, or solvent resistance matters.

How flat does a sample need to be?

It must be flat enough for the scan size, Z range, feedback settings, probe geometry, and measurement target. A surface suitable for imaging micron-scale texture may be unsuitable for sub-nanometer roughness. When in doubt, start with a smaller scan area and verify that the same feature type appears across multiple locations.

Can adhesive tape or glue be used for mounting?

Yes, but use them cautiously. The mounting layer should not creep, outgas, shed particles, contaminate the scan area, or introduce excessive tilt. For electrical modes, it must also support the intended conductive pathway or insulation strategy.

How can I tell whether a feature is real or an artifact?

Scan more than one area, change scan direction, compare trace and retrace, image a blank substrate, reduce scan force where appropriate, and inspect whether features move or smear under repeated scanning. Real features should be physically plausible and reproducible under reasonable imaging changes.

Key takeaway

Effective AFM sample preparation is a controlled compromise between cleanliness, flatness, immobilization, and preservation of the original structure. Start with the measurement question, choose the least disruptive substrate and cleaning method, mount the specimen rigidly, and verify the surface with controls. The outcome is not just a better-looking image; it is AFM data that can be interpreted with fewer hidden assumptions.