FTIR spectroscopy sample preparation for ATR, KBr pellets, liquids, and films

What sample preparation needs to achieve

FTIR spectroscopy sample preparation is the process of presenting a solid, liquid, film, powder, coating, or gas to the infrared beam so the spectrum represents the material of interest, not moisture, excessive thickness, poor contact, scattering, or the preparation matrix. For quick qualitative identification, ATR is often the most convenient route because many samples can be measured with little preparation. For bulk composition, method transfer, archived reference spectra, or quantitative work, transmission, KBr pellets, liquid cells, gas cells, diffuse reflectance, or specular reflectance may be more suitable.

The practical choice starts with the analytical question: surface or bulk, qualitative or quantitative, destructive or non-destructive, trace or major component, and wet or dry sample. ASTM E168-16(2023) treats preparation, operation, and calculation as linked parts of infrared quantitative analysis, which is a useful reminder that sample handling cannot be separated from interpretation. (store.astm.org)

For readers building a broader workflow, this article fits with other sample preparation topics because the same principle applies across analytical methods: the sample has to match the measurement mode.

Choose the sampling route before touching the sample

A common mistake is preparing every material for the accessory already installed on the spectrometer. A better approach is to select the route from the sample form and the question being asked. Thermo Fisher sample-handling guidance separates solids, films, liquids, viscous materials, surface analysis, and gases into different preparation situations, while Shimadzu notes that powders can be measured by more than one method, including KBr pellets, Nujol mulls, diffuse reflection, and ATR. (documents.thermofisher.com)

Analytical situation	Commonly practical route	Main preparation control
Clean, hard, flat solid surface	ATR	Consistent contact pressure and a clean crystal
Dry powder for routine identification	ATR or diffuse reflectance	Particle size, packing, and moisture control
Dry powder where bulk transmission data are needed	KBr pellet	Dilution, grinding, pellet clarity, and dry KBr
Free-flowing nonaqueous liquid	ATR or liquid transmission cell	Path length, volatility, and window compatibility
Aqueous liquid	ATR or a validated liquid-cell method	Short path length, background, and water absorption
Thin free-standing film	Transmission	Uniform thickness and no wrinkles in the beam
Coating on reflective metal	Specular or reflection-absorption method	Surface smoothness, angle, and substrate background
Gas or vapor	Gas cell	Cell path length, pressure, purge, and background gas

Preparation guidance by sample type

Powders and crystalline solids

Powders look simple, but they are easy to prepare inconsistently. A powder can often be placed directly on an ATR crystal, although loose or coarse particles may give weak contact and a noisy baseline. Grinding can improve reproducibility, but it should be controlled to avoid heating, hydration change, polymorphic change, or contamination from the mortar.

If the purpose is surface analysis, ATR may be useful because it emphasizes the near-surface region. If the purpose is bulk characterization, a KBr pellet or a well-controlled diffuse reflectance preparation may represent the whole powder more reliably.

Liquids, viscous samples, and pastes

Liquids can be measured by ATR, as a thin film between IR-transparent windows, or in a fixed-path cell. The main preparation variable is optical path length. Too much liquid path length can saturate strong bands; too little can push weak bands into the noise.

Viscous liquids and greases usually need a uniform, bubble-free layer. Volatile samples require fast handling or a sealed cell. For aqueous samples, the analyst must account for strong water absorption and choose windows, seals, and cleaning steps that are compatible with the solvent.

Films, fibers, coatings, and layered materials

Thin films can be excellent transmission samples when they are uniform and thin enough for the beam to pass through. Thick or dark films may need ATR, microtoming, or reflectance instead. Fibers should be aligned or bundled consistently if orientation affects the bands.

Layered packaging, paints, and coatings require a decision about depth. ATR may emphasize the outer layer, transmission may mix layers, and reflectance methods may be preferred when the material is supported by a reflective substrate.

Gases and vapors

Gas-phase FTIR preparation is mainly about controlling the cell rather than grinding or mounting a sample. The cell should be clean, leak-checked, and purged or evacuated as required by the method. Background gas, pressure, temperature, and path length need to be recorded because they affect band intensity and line shape. If the gas is reactive, corrosive, toxic, or condensable, the cell material and safety controls become part of the preparation plan.

ATR sample preparation without hiding its limits

ATR is widely used because it is fast and usually needs little or no sample preparation. Thermo Fisher describes ATR as a versatile infrared sampling technique for rapid analysis with minimal or no preparation, and Bruker explains that ATR interacts with only the first few microns of the sample rather than transmitting through the full thickness. (knowledge1.thermofisher.com)

That convenience does not remove the need for control. The sample must make intimate contact with the crystal; hard grains, warped films, and rough plastics can leave air gaps. Contact pressure should be consistent between samples when spectra will be compared. The crystal must be clean before the background and again before each sample. Abrasive powders can scratch softer crystals, and strongly absorbing materials may still produce distorted or saturated bands.

• Use a fresh background after cleaning the crystal or changing purge conditions.
• Cover the active crystal area completely with liquid, paste, powder, or film.
• Apply consistent pressure for solids and powders when comparing spectra.
• Check that the cleaning solvent does not leave residue or attack the crystal or seals.
• Compare ATR spectra with ATR libraries, or apply a validated correction when comparing them with transmission spectra.

ATR is often the right first measurement for unknowns, but it is not always enough. If a contamination layer sits on a bulk material, ATR may mainly report that layer. If a minor component is buried in a thick solid, transmission, extraction, microtomy, or another sampling strategy may be needed.

Transmission, KBr pellets, and liquid cells

Transmission FTIR requires the infrared beam to pass through the sample. That is why sample thickness, dilution, and matrix choice matter. Bruker describes transmission as a method that often requires the sample to be very thin or diluted; solids are commonly ground and mixed with potassium bromide, then pressed into a pellet, while very thin samples can be placed on a KBr window. (bruker.com)

For the KBr pellet method, Shimadzu gives a practical reference point: approximately 0.1 to 1.0 percent sample mixed into 200 to 250 mg of fine alkali halide powder for a 13 mm pellet, followed by pressing under vacuum. The same guidance warns that inadequate vacuum can lead to fragile pellets that scatter light and recommends drying KBr at about 110 °C for two to three hours before storage in a desiccator. (ssi.shimadzu.com) See also: analytical methods.

1. Dry the KBr or other alkali halide when the method and material require dry handling.
2. Grind the sample and KBr finely and uniformly, avoiding excess pressure or heat.
3. Use a low starting concentration for strongly absorbing samples, then adjust if bands are too weak.
4. Press a clear, mechanically stable pellet and reject cloudy, cracked, or uneven pellets for critical work.
5. Run a suitable background, ideally accounting for the holder, blank KBr, and moisture behavior required by the method.

Concentration and path length should be optimized by the spectrum, not by the recipe alone. Shimadzu notes that about 1 percent concentration is generally suitable for KBr pellets, but the actual concentration should be adjusted by transmittance or absorbance; it gives approximately 10 percent transmittance, or about 1 absorbance unit, as a useful maximum peak-intensity target for the KBr pellet example. The same source advises adjusting film thickness, liquid membrane thickness, or microscope transmission path length in a similar way. (shimadzu.eu)

Mulls and liquid films are useful when pellet pressing is not suitable, but the matrix can create its own bands. For example, Shimadzu lists Nujol absorption near 3000 to 2800 cm-1, 1460 cm-1, 1375 cm-1, and 730 cm-1, so regions of interest should be checked against the mull medium rather than assumed to be sample bands. (ssi.shimadzu.com)

Diffuse reflectance and surface reflection methods

Reflectance methods are not one technique. Bruker separates reflection-absorption, specular reflection, and diffuse reflection, each with different sample requirements. Reflection-absorption suits very thin material on a reflective substrate; specular reflection requires a smooth reflective surface; diffuse reflectance, commonly called DRIFTS, measures radiation scattered from particles. (bruker.com)

For DRIFTS, preparation quality is central. Bruker states that particle shape, size, compactness, and refractive index affect the amount of diffusely reflected light, and that samples should be ground, well mixed, and diluted with an IR-transparent solid such as KBr. Shimadzu similarly describes background measurement on the diluent, followed by measurement of sample diluted in KBr or another diluent. (bruker.com)

Raw diffuse reflectance spectra should not be treated as simple transmission spectra. Shimadzu explains that diffuse reflected light undergoes repeated transmission and reflection, which emphasizes low absorption bands compared with transmission, and that Kubelka-Munk transformation is used when comparison or quantitative handling is required. Bruker also notes that DRIFTS spectra differ from transmission spectra and may need processing before comparison. (ssi.shimadzu.com)

Quality checks before interpreting peaks

The final preparation step is a sanity check. If the spectrum has saturated peaks, extreme noise, a tilted baseline, unexpected water bands, matrix bands, or poor repeatability, check the preparation route before assigning functional groups. Many apparent chemistry problems are sample-handling problems.

Observation	Likely preparation issue	Corrective action
Flat-topped or clipped bands	Sample too thick, too concentrated, or path length too long	Dilute, reduce thickness, or shorten path length
Weak spectrum with high noise	Too little sample interaction or poor ATR contact	Improve contact, increase scans, or choose another mode
Sloping or scattering baseline	Coarse powder, cloudy pellet, rough surface, or particle-size variation	Improve grinding, mixing, packing, or surface preparation
Unexpected broad water features	Moist KBr, wet sample, or changing purge/background	Dry compatible materials and rerun background
Peaks from oil or mull medium	Preparation matrix contributes bands	Run blank matrix and choose another medium if needed

A complete record should include the sampling mode, accessory, crystal or window material, background method, sample mass or dilution ratio where relevant, drying conditions, cell path length, pressure or contact setting, number of scans, spectral resolution, and any correction or transformation. These details make spectra reproducible and help another analyst judge whether differences come from chemistry or preparation.

Frequently asked questions

Is ATR always the best FTIR sample preparation method?

No. ATR is often the fastest first option, especially for routine identification, but it emphasizes the near-surface region and depends on contact with the crystal. Transmission, KBr pellets, liquid cells, gas cells, or reflectance methods may be better when the question involves bulk composition, defined path length, trace components, thin films, or reflective substrates.

How much sample should be used in a KBr pellet?

As a starting point, Shimadzu describes approximately 0.1 to 1.0 percent sample mixed into 200 to 250 mg of fine KBr for a 13 mm pellet. The final amount should be adjusted according to band intensity because different materials absorb infrared radiation differently. (ssi.shimadzu.com)

Why do ATR and transmission spectra look different?

They use different optical geometries. Transmission measures light after it passes through the sample, while ATR measures interaction at the crystal-sample interface. Bruker notes that ATR and transmission can produce slightly different spectra, and Shimadzu cautions that ATR peak intensity and shape can be affected by wavelength dependence and anomalous dispersion, especially for some inorganic or high-refractive-index samples. (bruker.com)

Can wet or aqueous samples be measured by FTIR?

Yes, but they need a suitable method. ATR can be convenient for many liquids, while liquid transmission methods require careful path-length control and compatible window materials. Strong water absorption, bubbles, evaporation, and residue from cleaning are common preparation issues to check before interpreting the spectrum.

What is the safest way to compare FTIR spectra from different laboratories?

Compare spectra generated by the same sampling mode whenever possible, or document and validate the conversion or correction method. At minimum, record whether the spectrum is ATR, transmission, diffuse reflectance, specular reflectance, or gas-cell data, because preparation route can change relative intensities, baselines, and apparent band shapes.