SEM Sample Preparation: Fixation, Drying and Coating Step by Step

October 31, 2026 · 7 min read

SEM sample preparation turns a wet, fragile or non-conductive specimen into one that survives the vacuum and images cleanly under the electron beam. This step-by-step guide covers fixation, dehydration, critical point drying, mounting and sputter coating, with notes on bacteria, powders and the artifacts each step can cause.

Why SEM sample preparation matters

A scanning electron microscope images a surface with a focused electron beam in a vacuum chamber. That environment is hostile to most specimens: water evaporates, soft tissue collapses, loose particles can detach and non-conductive surfaces charge up. SEM sample preparation exists to deliver a specimen that is dry, stable, firmly attached and electrically conductive while still showing the surface as it was in life or after processing.

If you want a refresher on electron sources, detectors and vacuum modes first, see our overview of how a scanning electron microscope works. This guide concentrates on the bench work that comes before you load the stub.

Two broad workflows exist. Biological samples (cells, bacteria, tissues, plants) need chemical fixation and careful drying. Materials samples (metals, ceramics, polymers, powders) usually skip fixation but still require cleaning, mounting and often a conductive coating.

SEM sample preparation steps at a glance

Step	Purpose	Typical approach
1. Primary fixation	Crosslink proteins, stabilise structure	Glutaraldehyde (often 2–2.5%) in cacodylate or similar buffer
2. Post-fixation (optional)	Fix lipids and membranes, add conductivity	Osmium tetroxide, often 1%
3. Dehydration	Replace water gradually	Graded ethanol or acetone series
4. Drying	Remove solvent without surface-tension damage	Critical point drying with CO₂ or chemical drying with HMDS
5. Mounting	Attach and ground the sample	Aluminium stub, carbon adhesive disc, silver or carbon paint
6. Coating	Prevent charging, increase signal	Sputtered gold, Au/Pd or platinum; evaporated carbon for EDS

Fixation and dehydration for biological samples

Primary fixation with aldehydes

Aldehydes such as glutaraldehyde and formaldehyde crosslink proteins and lock the ultrastructure in place, as the University of Gothenburg core facility explains in its SEM preparation notes. Many facility protocols use 2–2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer. The McGill Facility for Electron Microscopy Research, for example, fixes cells and bacteria in 2.5% glutaraldehyde in 0.1 M cacodylate for 30–60 minutes, noting that fixation at 4 °C may help some specimens.

Always handle fixatives in a fume hood with gloves. Glutaraldehyde, cacodylate (an arsenic compound) and osmium tetroxide are toxic and need proper waste disposal under your institution's rules.

Secondary fixation with osmium

Osmium tetroxide fixes membrane lipids so they are not extracted by solvents during dehydration. The deposited osmium also improves conductivity and reduces charging. For cells and bacteria the step is often optional: McGill suggests either 1% osmium for 30–60 minutes or extending primary fixation to 24–72 hours at 4 °C, which also reduces cracking during drying.

Graded ethanol dehydration

Water must be replaced with a solvent in small steps to avoid shrinkage and distortion. A typical series:

Rinse out the fixative with buffer (for example 3 × 5 minutes).
Pass through 30%, 50%, 70%, 80% and 90% ethanol, about 10 minutes each.
Finish with 100% ethanol, repeated three times.
If you must pause, store samples overnight in 70% ethanol at 4 °C rather than in 100%.
Move straight to drying after the final change; do not leave samples in absolute ethanol for long, since this can cause shrinkage.

Critical point drying vs HMDS: removing the solvent

Letting ethanol or acetone simply evaporate is the classic beginner's mistake. These liquids have a high surface tension, and as the meniscus passes through the specimen it pulls delicate structures together, causing collapse and what the Gothenburg facility describes as micro-ripping of the surface.

Critical point drying (CPD)

In a critical point dryer, the ethanol is exchanged for liquid CO₂ under pressure. The chamber is then heated past the critical point of CO₂ (about 31 °C and 7.4 MPa), where liquid and gas become indistinguishable. Venting the gas slowly leaves a dry specimen without any liquid–gas interface ever crossing it. CPD is the standard for tissues and fragile three-dimensional structures. Make sure exchanges with CO₂ are complete, because residual ethanol can cause local collapse.

Chemical drying with HMDS

Hexamethyldisilazane (HMDS) has a low surface tension and evaporates with relatively little distortion. Gothenburg uses it for cell preparations, and McGill describes a graded ethanol–HMDS series as an alternative to CPD. It needs no special equipment, but HMDS is hazardous and must be used in a fume hood. Results on soft, highly hydrated tissues are usually less reliable than with CPD.

SEM sample preparation for bacteria and cultured cells

Bacteria and cells must be attached to a substrate before fixation and dehydration, or they will wash away. A common approach, used in the McGill protocol, is to coat 12 mm glass coverslips with 0.01% poly-L-lysine, let the cell or bacterial suspension adhere for 15–30 minutes, and then continue with fixation, ethanol dehydration and CPD or HMDS drying.

Wash cells in filtered buffer before fixation to remove medium, salts and debris that would crystallise on the surface.
Never let the coverslip dry out between steps; leave a thin film of liquid when changing solutions.
Keep coverslips in labelled holders so you know which side carries the cells.
For biofilms, gentle rinsing is essential, since vigorous pipetting strips the extracellular matrix you want to image.

McGill finishes its bacterial samples with about 5 nm of sputtered platinum before imaging, which illustrates how thin a coating can be for fine surface detail.

SEM sample preparation for powders and nanoparticles

Powders need no fixation, but they need even dispersion and a firm grip on the stub. Particles that come loose in the vacuum contaminate the chamber and can damage the column. A Nanoscience Instruments guide describes several methods:

Flick method (roughly 500 µm to 1 µm): load powder on a stiff-handled swab, strike the handle so particles scatter onto carbon tape, then blow off excess with compressed gas.
Dish method: spread a thin layer in a dish and press a taped stub onto it to pick up a near-monolayer.
Drop casting: disperse particles in ethanol or isopropanol and let a drop dry on the stub. Below about 200 nm, use a silicon wafer instead of carbon tape, because fine particles can sink into the adhesive.
Vacuum dispersion: dedicated devices disperse powder under controlled vacuum for reproducible, de-agglomerated samples.

Key tip: after mounting any powder, hold the stub upside down and blow it with clean compressed gas before loading. Anything that falls off on the bench would have fallen off in the chamber.

Mounting and sputter coating: gold, platinum or carbon?

Mount the dried specimen on a stub with a carbon adhesive disc; a line of silver or carbon paint from the sample to the stub improves grounding. Non-conductive materials such as polymers accumulate charge, which shows up as bright areas and drift. Thermo Fisher lists three remedies: grounding with conductive tape, coating with a conductive layer, or imaging at a lower vacuum level.

The coating should be just thick enough to stop charging. The Gothenburg facility recommends about 10 nm of gold, warning that thicker films hide surface detail. Choose the material by the analysis you plan, following JEOL's guidance on conductive coatings:

Coating	Best for	Limitations
Gold (sputtered)	Routine imaging, polymers, up to roughly 50,000×	Grain becomes visible at higher magnification; interferes with EDS
Platinum	Higher magnification (about 50,000–200,000×)	Still adds X-ray peaks to spectra
Iridium	Ultrahigh-resolution imaging; EDS on heat-sensitive samples	Needs a high-vacuum sputter coater
Carbon (evaporated)	EDS and quantitative X-ray work; single low-energy peak	Evaporation heat can damage sensitive samples; less ideal for fine surface imaging

If you only need imaging of a polymer or a dried biological sample, gold or gold–palladium is usually fine. If you need elemental maps, plan for carbon coating, or image uncoated in low-vacuum mode.

Common SEM preparation artifacts and how to avoid them

Collapse and wrinkling: caused by air drying from solvent. Use CPD or HMDS.
Shrinkage: too rapid a dehydration series or long storage in 100% ethanol. Use smaller steps and dry promptly.
Cracking: stresses during drying, especially in large or poorly fixed samples. Fix thoroughly; extended aldehyde fixation can help.
Salt and medium crystals: residue from culture medium or phosphate buffer. Rinse well before fixation.
Charging: bright streaks and image drift. Improve grounding, coat more evenly, or lower the accelerating voltage.
Coating grain and masking: too thick a film or gold at very high magnification. Use thinner films or platinum/iridium.
Contamination: fingerprints and dust. Handle stubs only with tweezers and gloves, and store finished samples in a desiccator.

Summary

Good SEM sample preparation follows a logical chain: stabilise the structure, replace water gently, dry without surface tension, attach firmly and make the surface conductive with the thinnest suitable coating. Biological samples rely on fixation, a graded ethanol series and critical point drying, while powders depend on dispersion and secure mounting. Treat published protocols as starting points, record every parameter you change, and adapt times and coating thickness to your sample and instrument.

Frequently asked questions

What are the main steps of SEM sample preparation?

For biological samples: fixation, optional osmium post-fixation, graded ethanol dehydration, critical point or HMDS drying, mounting on a stub and conductive coating. Materials samples usually need only cleaning, mounting and coating.

Why are SEM samples coated with gold?

Non-conductive samples accumulate electrons and charge, which distorts the image. A thin sputtered gold layer, typically around 10 nm, conducts the charge away and increases the secondary electron signal.

How do you prepare bacteria for SEM?

Attach bacteria to a coated substrate such as poly-L-lysine coverslips, fix in glutaraldehyde, dehydrate in a graded ethanol series, dry by critical point drying or HMDS and sputter coat with a thin metal layer.

Is critical point drying necessary for SEM?

For soft, hydrated biological samples it is the standard way to avoid collapse from surface tension. HMDS chemical drying is a common alternative for cells, and dry materials samples do not need it.

Which coating should I use for EDS analysis?

Carbon is usually preferred because its single low-energy X-ray peak does not interfere with other elements. Gold and other metals add their own peaks and absorb low-energy X-rays.

How do you prepare powder samples for SEM?

Disperse the powder onto carbon tape by the flick or dish method, or drop cast a suspension in ethanol or isopropanol. Blow off loose particles with compressed gas, and use a silicon wafer for particles below about 200 nm.