The good news: AFM demands far less preparation than SEM or TEM. There's no staining, no conductive coating, no dehydration required, and it works in air, liquid, or vacuum (Nanosurf, 2024). Still, results vary wildly depending on substrate choice, adhesion chemistry, and how well you control your environment.
This guide walks through the exact steps, materials, and parameters that determine whether your AFM scan succeeds or fails—plus the mistakes that quietly ruin good samples.
TL;DR
- Sample prep involves three phases: substrate selection, activation/adhesion, and rinsing/drying
- Substrate flatness and adhesive choice drive image quality more than any other factor
- Biological and liquid samples need dedicated hydration management
- Contamination and poor fixturing cause most failed scans
- The right approach depends on sample type: rigid, soft, biological, or powder
How to Prepare Samples for AFM: Step-by-Step
Step 1: Select and Prepare the Substrate
Your substrate choice sets the ceiling on your image resolution. The core rule, per NIST's nanoparticle protocol: substrate roughness must be much smaller than the smallest feature you're imaging (NIST, 2009).
Common substrate options:
- Mica — cleaves to an atomically flat, negatively charged surface; ideal for DNA, ferritin, and lipid bilayers
- Silicon — practical flat option for nanoparticles; roughness as low as 0.13 nm RMS at small fields of view
- HOPG — smooth and conductive but hydrophobic, which complicates adsorption of hydrophilic samples such as DNA
- Glass — untreated cover slips are rough; KOH/ethanol treatment can cut RMS roughness ~10× for membrane-protein work
- Gold-on-mica — used for electrochemical studies and calibration; annealed films can show 0.8–1.0 µm terraces
Cleave or clean immediately before use. Even air-cleaved mica can pick up nanometer-scale contamination if left exposed.
Step 2: Activate the Surface for Adhesion
A flat substrate only helps if the sample adheres. Activation creates the charge or chemistry needed for binding:
- Poly-L-lysine — positive charge for negatively charged biomolecules; dilute 0.1% PLL 1:10 and incubate mica ~30 minutes
- APTES/APS silanes — for nucleic acid immobilization on mica; APS is more hydrolytically stable for liquid imaging
- Divalent cations (Mg²⁺, Ni²⁺) — alternative for DNA immobilization in HEPES buffer, without a separate silane step
Match surface charge to your sample, and confirm the adhesive works in your imaging mode—contact, tapping, or liquid. A promoter that's stable in air can behave differently once submerged.
Step 3: Deposit and Adhere the Sample
Concentration is where most first attempts go wrong. Too dense, and particles clump into unreadable masses. Too dilute, and you scan empty substrate for hours.
- Dilution series — start with a range rather than guessing one concentration
- Incubation time — scales with particle size; NIST's gold nanoparticle protocol uses 10 minutes for 60 nm particles and 30 seconds for 10 nm particles
- Rinse and dry — rinse gently with deionized water, then dry with nitrogen to remove unbound material without disturbing what adhered
Practical starting point for DNA: published protocols often use 1–20 ng/µL with a 1–3 minute incubation before rinsing. Treat these as baselines—your optimum will differ by sample class.
Step 4: Fixture and Inspect Before Imaging
Mount the substrate on a magnetic AFM disc with epoxy or non-outgassing tape. Adhesives that outgas in the chamber contaminate high-resolution scans—a common, avoidable failure.
Before loading into the AFM:
- Inspect under an optical microscope to confirm dispersion quality
- Check for clumping or empty regions that would waste scan time
- Set up hydration if working with biological or liquid samples
For temperature-sensitive biological work, an environmental chamber limits buffer evaporation during long scans. Temperature-controlled stages (for example, ranges from −30°C to 250°C on Molecular Imaging systems) support phase-transition studies and help reduce thermal drift on extended runs.
When Should You Follow This Preparation Process?
This workflow suits most rigid and soft nanomaterials, thin films, and biomolecules imaged in near-native conditions. Use it for:
- Nanoparticle sizing and distribution studies
- Polymer morphology characterization
- Thin-film surface analysis
- Biomolecule imaging in liquid or air
Skip this workflow for a few sample types:
- Highly porous samples
- Extremely tall or irregular specimens
- Materials that need cross-sectional analysis
Those cases usually need microtoming or resin embedding instead of standard deposition.
Scale changes the priority. Large-batch prep needs consistent dilution and tight control over exposure time so early and late samples match. Single-particle studies flip the focus: control dispersion density first, not throughput.
What You Need Before Preparing AFM Samples
Having materials on hand before you start prevents mid-protocol delays that can ruin time-sensitive incubation steps.
Equipment and Materials Checklist
- AFM system with cantilevers/probes matched to your imaging mode
- Magnetic specimen discs for mounting
- Substrates: mica, silicon, HOPG, or gold, depending on sample type
- Adhesives: epoxy, UV-cure glue, or non-outgassing double-sided tape
- Deionized water for rinsing
- Nitrogen gas for drying
Stock calibration samples, cantilevers, mica substrates, and gold-on-mica products from Molecular Imaging's web store so these supplies stay on hand between experiments.
Sample and Environmental Readiness
- Confirm your sample is chemically compatible with the intended substrate and adhesive
- Check humidity or temperature sensitivity in advance
- For liquid or biological imaging, verify buffer composition
- Confirm a hydration method (open cell or environmental chamber) is ready
Key Parameters That Affect Results When Preparing AFM Samples
Even a correctly followed protocol can produce poor images if these variables drift outside acceptable ranges.
Substrate Roughness and Flatness
Features smaller than your substrate's roughness cannot be resolved: topography noise swamps the signal. This is why mica dominates biomolecule imaging (it's smooth enough to reveal single strands of DNA) while nanoparticle work can tolerate silicon's slightly higher roughness, since particle features are typically larger.
Adhesion Strength and Chemistry
Weak adhesion means your sample moves during scanning, producing streaks or complete dropout. Adhesion force must exceed the tip-sample interaction force. Studies measuring particle adhesion have observed weakly bound particles swept off the substrate mid-scan, so only strongly adhered material ever gets successfully imaged.
Sample Concentration and Dispersion
Overly concentrated samples clump; overly dilute samples waste scan time. A published TiO2 nanoparticle study tested dilution ratios of 1:10, 1:40, and 1:50 from a stock suspension rather than relying on one fixed number. This dilution-series approach is the safest default for any new sample class.
Hydration and Environmental Conditions
Evaporation doesn't just dry out your sample. It concentrates the buffer salts left behind, which can distort or destroy biomolecular structure.
One lipid-bilayer study found a 20 µL buffer drop evaporated in about 15 minutes under an open setup. The same work held stable imaging for 3–4 hours with an integrated bath holding up to 420 µL (Nature Scientific Reports, 2018).
Closed liquid cells and environmental chambers solve this by maintaining consistent humidity and volume over long scans, which matters for anything beyond a quick snapshot image.
Common Mistakes and Troubleshooting in AFM Sample Preparation
Two mistakes account for a disproportionate share of failed sessions:
- Skipping optical inspection before loading the sample. A 30-second microscope check catches poorly dispersed regions that would otherwise waste scan time
- Using excess adhesive or outgassing packaging. Even small amounts of the wrong tape or glue can contaminate high-resolution scans
When images still fail, match the symptom to a likely cause:
| Problem | Likely Cause | What to Check |
|---|---|---|
| Streaky or noisy images | Poor sample adhesion or tip contamination | Re-check surface activation; clean or swap the probe |
| Buffer evaporation | Open liquid cell left exposed too long | Switch to a closed cell or chamber; top up with deionized water, not buffer |
| Inconsistent results across replicates | No standardized protocol | Log environmental conditions, incubation times, and rinse steps precisely |
That last point matters more than most labs admit. Two researchers running "the same" protocol can get different images simply because one rinsed for 10 seconds longer. Write down every step, every time.
Frequently Asked Questions
What are the steps of sample preparation?
Sample preparation follows four steps: select and prepare your substrate, activate its surface for adhesion, deposit and adhere the sample, then fixture and inspect it before loading into the AFM.
Why is sample preparation important for AFM imaging?
AFM relies on direct physical interaction between the probe and your sample. Improper preparation causes streaking, poor resolution, or complete imaging failure since there's no way to compensate electronically for a poorly adhered surface.
What substrate is best for AFM sample preparation?
Mica is the standard for atomically flat biomolecule imaging. Silicon, glass, or HOPG suit other material types depending on the flatness and conductivity your sample requires.
How do you prepare biological samples for AFM without damaging them?
Use gentle deposition, an appropriate buffer for your target molecule, and continuous hydration through a liquid cell or environmental chamber. This preserves native structure instead of forcing a dried, potentially distorted conformation.
What causes poor image quality despite correct sample preparation?
Contamination on the tip, probe wear, and improperly calibrated scan parameters are the usual culprits when your prep steps were done correctly but images still look off.
How long does AFM sample preparation typically take?
Rigid samples like nanoparticles on silicon can take minutes. Biological or resin-embedded samples often require hours between incubation, buffer setup, and hydration configuration.
