EM sample preparation for more reliable SEM, TEM and cryo-EM results
Why EM sample preparation determines image quality
EM sample preparation covers the physical, chemical and cryogenic steps used to make a specimen suitable for electron microscopy while preserving the feature that needs to be measured. The best route depends first on the imaging mode. SEM usually requires a stable surface. TEM requires material thin enough for electrons to pass through. Cryo-EM aims to keep hydrated biological structures vitrified instead of dried or embedded.
Preparation is not a neutral handling step. Fixation, dehydration, staining, coating, ion milling and freezing can improve contrast or stability, but they can also introduce shrinkage, extraction, charging, contamination or beam-damage artifacts. For that reason, the preparation workflow should be selected around the measurement question, not copied from a generic protocol.
For readers comparing instruments, consumables and workflow options, the broader sample preparation category is a useful place to organize related methods across microscopy and laboratory analysis.
Match the preparation route to the microscope and the question
The first decision is what the experiment needs to show: surface morphology, internal ultrastructure, crystallographic or compositional information, or near-native hydrated structure. SEM, TEM and cryo-EM can sometimes be used on related specimens, but their preparation logic is different.
For SEM, the specimen surface must remain stable under vacuum and the electron beam. Dry conductive solids may need only cutting, mounting and cleaning. Hydrated biological specimens commonly require fixation, dehydration, drying and conductive coating. University microscopy facilities and SEM method reviews consistently describe solvent dehydration followed by critical point drying as a common way to reduce collapse in delicate biological materials. Sputter or evaporative coating is also often used to reduce charging on non-conductive surfaces. (pmc.ncbi.nlm.nih.gov)
For TEM, thickness is the central constraint. Electrons must transmit through the specimen, so biological samples are commonly embedded and sectioned, while many materials samples are mechanically thinned, ion milled or prepared as site-specific lamellae. University core protocols commonly describe aldehyde fixation, osmium post-fixation, dehydration, resin infiltration, embedding and ultrathin sectioning for biological TEM, with ultrathin sections often in the 70–90 nm range. (cores.arizona.edu)
For cryo-EM, the goal is different again. The sample is usually kept hydrated, applied to a grid, thinned to a very thin liquid film and vitrified rapidly so water does not crystallize. Cryo-EM preparation reviews define successful vitrification as freezing the aqueous layer fast enough to avoid crystalline ice. They also stress that grid preparation remains a major experimental bottleneck because ice thickness, particle distribution and air-water interface effects can vary from grid to grid. (pmc.ncbi.nlm.nih.gov)
Common EM sample preparation workflows
Biological SEM workflow
A conventional biological SEM workflow often starts with chemical fixation to stabilize morphology, followed by buffer rinses, graded dehydration, drying, mounting and coating. Critical point drying is widely used because direct air drying can create surface-tension forces that collapse soft structures. Some workflows use alternatives such as HMDS or freeze drying, but the choice should be validated against the structure being measured rather than treated as interchangeable.
Biological TEM workflow
Biological TEM preparation usually trades native hydration for stability and contrast. Chemical fixation crosslinks cellular components, osmium tetroxide helps preserve and contrast membranes, dehydration replaces water with organic solvent, and resin embedding supports ultrathin sectioning. Heavy-metal staining can improve contrast. At the same time, overstaining, extraction during dehydration and uneven infiltration can change how membranes, vesicles or dense regions appear. This is why many EM facilities treat pilot processing and section screening as part of the workflow, not as optional troubleshooting.
Materials TEM and FIB lamella workflow
For metals, ceramics, semiconductors and multilayer devices, the preparation goal is often to expose a precise region while keeping the specimen thin enough for TEM or STEM. Mechanical cutting, grinding, dimpling and argon ion milling remain important for many materials. Focused ion beam preparation adds site specificity: a small lamella can be lifted out from a known feature, interface or defect. ISO lists ISO 17297:2025 as a focused ion beam vocabulary standard for TEM specimen preparation, and ISO/FDIS 16887 as an under-development guideline for TEM lift-out using focused ion beam systems. This reflects the importance of standardized language and repeatable FIB workflows in analytical electron microscopy. (iso.org)
Cryo-EM grid preparation
Cryo-EM grid preparation usually involves grid surface treatment, sample application, blotting or dispensing, and plunge freezing into a cryogen such as liquid ethane. The workflow may look simple, but it is highly sensitive to concentration, buffer composition, grid support, blot time, temperature, humidity and transfer handling. Review articles emphasize that semi-automated plungers improve control but do not make the process fully reproducible; screening multiple grid conditions remains normal practice. (pmc.ncbi.nlm.nih.gov)
Preparation choices and typical risks
| Goal | Typical preparation route | Main risk to control |
|---|---|---|
| SEM surface morphology of biological tissue | Fixation, dehydration, critical point drying, mounting and coating | Shrinkage, collapse, residue and charging |
| TEM ultrastructure of cells or tissue | Fixation, osmium post-fixation, dehydration, resin embedding, ultrathin sectioning and staining | Extraction, contrast artifacts, compression and sectioning defects |
| TEM or STEM analysis of a specific device interface | FIB lift-out, protective deposition, thinning and low-energy cleaning where appropriate | Ion damage, redeposition, curtaining and thickness variation |
| Cryo-EM of purified macromolecules | Grid treatment, sample application, blotting or dispensing and vitrification | Ice thickness variation, preferred orientation, aggregation and crystalline ice |
| Particle or nanoparticle imaging | Dispersion, deposition on support film, drying or cryo preparation depending on the question | Agglomeration, drying pattern artifacts and contamination |
Nanomaterial work needs particular caution because preparation can change the object being measured. NIST sample preparation resources for engineered nanomaterials note that particles may agglomerate or undergo transformations during exposure to media and preparation steps. In practice, dispersion conditions, substrate choice, drying method and time between preparation and imaging should be documented as carefully as microscope settings. (nist.gov)
Artifacts are not failures, but they must be recognized
Most EM preparation artifacts are understandable consequences of making a real specimen compatible with vacuum, electrons or cryogenic handling. The issue is not that artifacts exist. The risk is interpreting them as biological or materials features without appropriate controls.
- Shrinkage and collapse: common in soft hydrated specimens when water is removed. Critical point drying and cryo methods are used to reduce this risk, not eliminate it in every case.
- Charging: appears when insulating specimens accumulate charge under the beam. Conductive coating, carbon paint, lower accelerating voltage or variable-pressure SEM may help, depending on the sample and analytical need.
- Extraction: lipids, soluble components or loosely bound material can be lost during fixation, solvent exchange or staining.
- Sectioning damage: chatter, compression, knife marks and folds can mimic internal structure in TEM sections.
- Ion-beam damage: FIB preparation can amorphize, implant ions, heat or redeposit material, especially near final lamella surfaces.
- Ice problems in cryo-EM: crystalline ice, thick ice, thin broken film and uneven particle distribution can all prevent useful data collection.
A strong preparation plan includes at least one validation step. That may mean a lower-magnification survey before high-resolution imaging, a comparison between chemically prepared and cryo-prepared specimens, repeat preparation on independent samples, or a check that the same feature appears across multiple regions and preparation batches. See also: analytical methods.
How to document a reproducible EM preparation method
Reproducibility in EM sample preparation depends on more than the instrument model. A usable method record should allow another trained operator to understand what was done, where variation may enter and which steps are critical for interpretation.
- Define the specimen state before preparation. Record sample age, storage, buffer or medium, moisture condition, orientation and any pre-cleaning.
- Record chemical details. Include fixative composition, buffer, pH where relevant, osmium or stain conditions, solvent series, resin type and cure conditions.
- Record physical processing. Note section thickness target, grid or stub type, coating material and nominal thickness, drying method, polishing steps, ion species and milling energies where applicable.
- Record timing. Delays between fixation, dehydration, drying, coating, freezing and imaging can change sensitive specimens.
- Record screening outcomes. Document rejected grids, cracked sections, charging regions or damaged lamellae, because failed preparations often reveal the practical limits of the method.
This level of detail is especially important when comparing datasets over time. If contrast changes after a stain batch changes, or if a lamella shows more damage after a higher final milling energy, the preparation record becomes the first place to investigate.
Practical selection framework
A useful way to choose an EM preparation route is to ask four questions before writing a protocol. What feature must be preserved: surface texture, membrane organization, crystal structure, elemental distribution or hydrated conformation? Which preparation step is most likely to alter that feature? What control can reveal the alteration? What level of throughput is acceptable?
For example, a researcher imaging bacterial surface morphology by SEM may prioritize fixation quality, dehydration cleanliness and drying method. A semiconductor failure analyst may prioritize site-specific FIB lift-out, lamella thickness and low-damage final polishing. A structural biology team preparing cryo-EM grids may spend most of its optimization effort on buffer, concentration, grid support and freezing conditions rather than microscope time. The correct workflow is the one that protects the measurement objective with the fewest unnecessary transformations.
Cost and equipment access also matter. Critical point dryers, ultramicrotomes, sputter coaters, plunge freezers, cryo-transfer systems and FIB-SEM platforms are specialized tools. If a lab lacks one of them, the alternative should be chosen with a clear statement of the compromise. An easier preparation method may be acceptable for screening but insufficient for quantitative morphology, high-resolution structure or failure analysis.
Frequently asked questions
Is EM sample preparation the same for SEM and TEM?
No. SEM usually images surface-related signals, so preparation focuses on stable, clean and often conductive surfaces. TEM requires electrons to pass through the specimen, so the sample must be thin enough for transmission or sectioned into an ultrathin slice.
Why are biological EM samples often dehydrated?
Conventional high-vacuum SEM and resin-embedded TEM are generally incompatible with bulk water. Dehydration replaces water before drying or embedding, but it can introduce shrinkage or extraction artifacts. Cryo methods are used when hydrated-state preservation is more important than conventional staining or room-temperature handling.
When is FIB preparation useful?
FIB preparation is useful when the region of interest is small, buried or site specific, such as a semiconductor defect, coating interface, grain boundary or multilayer cross-section. It provides precise targeting, but ion damage and thickness control must be managed.
Why do cryo-EM grids often require many screening attempts?
Small changes in concentration, grid chemistry, blotting, humidity, ice thickness and particle behavior can affect whether particles are visible, well distributed and structurally preserved. Screening is part of cryo-EM preparation because successful vitrification does not guarantee ideal imaging conditions.
What is the most important rule for EM preparation?
Start with the measurement question. A protocol that gives attractive images may still be wrong if it changes the structure or composition being interpreted. Good EM preparation balances preservation, contrast, conductivity, thickness and reproducibility.
