ATR-FTIR Spectroscopy

Attenuated Total Reflectance (ATR) is one of the most widely used FTIR sampling techniques — it lets you collect infrared spectra from solids, liquids, pastes, and powders with virtually no sample preparation.

What Is ATR-FTIR?

ATR-FTIR stands for Attenuated Total Reflectance Fourier Transform Infrared spectroscopy. It combines ATR sampling with the speed and sensitivity of an FTIR spectrometer.

In a standard ATR measurement, the sample is placed in direct contact with a crystal that has a high refractive index — typically diamond, zinc selenide (ZnSe), or germanium. An infrared beam enters the crystal at an angle beyond the critical angle for total internal reflection. At each reflection point, a small amount of energy — the evanescent wave— extends a fraction of a wavelength beyond the crystal surface into the sample, where it is partially absorbed. The attenuated beam that exits the crystal carries the sample's infrared absorption signature.

How ATR-FTIR Works

When infrared light travels from a high-refractive-index medium (the ATR crystal) toward a lower-refractive-index medium (the sample) at a steep enough angle, it undergoes total internal reflection rather than passing through. The beam is not actually reflected at a single point on the surface — instead, it penetrates slightly into the sample as an evanescent wave before returning to the crystal.

This evanescent wave decays exponentially with distance from the crystal surface. Its penetration depth is typically about 0.5–5 μm across the mid-IR for diamond or ZnSe (about 2 μm at 1000 cm⁻¹), and roughly 0.2–1.1 μm for germanium (about 0.66 μm at 1000 cm⁻¹), meaning ATR probes only the outermost layer of the sample. The portion of the evanescent wave that is absorbed at specific frequencies produces the ATR spectrum — which looks like a conventional FTIR absorption spectrum, though low-wavenumber bands appear relatively stronger and strong bands can shift a few cm⁻¹ to lower wavenumber compared with transmission spectra.

Good optical contact between the sample and the crystal is essential. For solids, a pressure clamp or anvil pushes the sample firmly against the crystal. Liquids simply need to cover the crystal surface. Powders are pressed to fill air gaps. Poor contact leads to weak, noisy spectra that do not represent the true sample composition.

ATR Crystal Materials

The three most common ATR crystal materials differ in refractive index, spectral range, chemical resistance, and cost. Your choice of crystal affects the depth of penetration, spectral coverage, and which sample types you can safely measure.

Diamond

n = 2.4

Range: 4000–400 cm⁻¹ (reduced signal at ~2300–1900 from diamond’s own absorption; low-wavenumber limit set by the supporting optics)

Durability: Excellent — hardest known material

Approx. cost (2026): High (complete single-reflection accessory about $3,000–$8,000)

Best for: Abrasive solids, corrosive liquids, general-purpose use

ZnSe

n = 2.4

Range: 4000–650 cm⁻¹

Durability: Moderate — scratches easily; attacked by acids and strong bases (usable pH about 5–9)

Approx. cost (2026): Moderate (replacement crystal about $500–$1,500)

Best for: General liquids and soft solids, near-neutral samples (pH about 5–9)

Germanium (Ge)

n = 4.0

Range: 4000–600 cm⁻¹

Durability: Moderate — brittle, susceptible to thermal shock

Approx. cost (2026): Moderate–High (crystal about $1,000–$3,000)

Best for: Carbon-black-filled, high-refractive-index or very strongly absorbing samples (smallest penetration depth)

Diamond vs multi-reflection crystals

Most modern ATR accessories use a single-reflection diamond crystal. Older or specialized accessories may use multi-reflection ZnSe or Ge troughs (6–12 reflections) for enhanced sensitivity with liquids and thin films. More reflections increase absorption signal but require larger sample volumes.

Depth of Penetration

The effective sampling depth in ATR is governed by how far the evanescent wave extends into the sample. The depth of penetration (dp) is defined as the distance at which the evanescent wave amplitude falls to 1/e (~37%) of its value at the crystal surface:

dp = λ / (2π · n1 · √(sin²θ − (n2/n1)²))

λ = wavelength of IR light  |  n1= crystal refractive index  |  n2= sample refractive index  |  θ = angle of incidence

Worked example

Diamond crystal (n1 = 2.4) with a typical organic sample (n2 ≈ 1.5) at θ = 45° and λ = 6 μm (≈ 1667 cm⁻¹):

sin²(45°) = 0.500
(n₂/n₁)² = (1.5/2.4)² = 0.391
√(0.500 − 0.391) = √0.109 = 0.330
dp = 6 / (2π × 2.4 × 0.330) = 6 / 4.98 ≈ 1.2 μm

Factors that affect penetration depth

  • Wavelength —Longer wavelengths (lower wavenumbers) penetrate deeper. This is why ATR spectra show stronger low-wavenumber bands relative to transmission spectra.
  • Angle of incidence —Angles of incidence closer to the critical angle increase penetration; larger angles of incidence decrease it. Most single-reflection accessories use a fixed 45° angle.
  • Refractive indices —A smaller difference between crystal and sample refractive indices increases penetration depth. Germanium (n = 4.0) gives the shallowest depth; ZnSe and diamond (both n ≈ 2.4) give essentially the same, deeper penetration.

Advantages of ATR over Transmission

Minimal sample preparation

Place the sample on the crystal and press — no grinding, diluting, or pellet-pressing required. This saves time and eliminates preparation artefacts.

No KBr pellets needed

Transmission IR requires mixing the sample into a KBr pellet (time-consuming and hygroscopic) or a Nujol mull (whose C-H bands near 2950–2850, 1460, 1377 and (weakly) 720 cm⁻¹ hide those regions of the sample spectrum). ATR bypasses this entirely.

Works with liquids, pastes, and powders directly

Liquids are simply placed on the crystal surface. Pastes and powders are pressed to ensure contact. There is no need for special cells, spacers, or windows.

No scattering from particles in the beam

In transmission, large or irregularly shaped particles scatter light and distort the baseline. ATR avoids this because the beam does not pass through the sample; the evanescent wave interacts only with the contact layer. Coarse or hard powders should still be ground for good crystal contact.

Reproducible results

For liquids and soft solids in full contact with the crystal, the effective pathlength is set by the crystal, angle and wavelength, so there is no variation from pellet thickness or film uniformity. For hard solids and powders, reproducibility depends on consistent clamping pressure.

Quick turnaround

Clean the crystal, place the sample, measure, and clean again — a full ATR analysis can take under a minute. This makes ATR well-suited for quality control, screening, and high-throughput work.

Limitations and Considerations

While ATR is the most convenient FTIR sampling method, it is not without trade-offs. Understanding these limitations helps you decide when a different sampling technique may be more appropriate.

Surface technique only

ATR probes only the outermost few micrometres (about 0.5–5 μm, depending on wavelength and crystal) of the sample. If the surface composition differs from the bulk (e.g., oxidized layers, coatings, or surface contamination), the spectrum may not represent the true bulk material.

Pressure-dependent contact

Solid samples must be pressed firmly against the crystal. Insufficient pressure leads to weak spectra; excessive pressure can damage soft crystals (ZnSe) or deform compliant samples. Reproducibility depends on consistent clamping force.

Crystal contamination and care

The crystal must be cleaned thoroughly between samples. Residual material produces ghost peaks in subsequent spectra. ZnSe is attacked by strong acids and bases. Even diamond crystals can be fouled by strongly adhesive compounds.

ATR correction needed for library matching

Because penetration depth varies with wavelength, ATR spectra have stronger low-wavenumber bands relative to transmission spectra. To search against transmission-based libraries, an ATR correction must be applied first. Most modern FTIR software includes this as a standard function. For guidance on spectral processing, see spectrum interpretation.

Frequently Asked Questions

What is the difference between ATR-FTIR and transmission FTIR?

In transmission FTIR, the infrared beam passes entirely through the sample, requiring it to be thin or transparent to IR light. In ATR-FTIR, the beam bounces inside a crystal and only an evanescent wave (about 0.5–5 μm deep, depending on wavelength and crystal) probes the sample surface. ATR is faster, needs less preparation, and works with thick, opaque, or liquid samples that would block a transmission measurement.

Which ATR crystal should I use?

Diamond is the safest default — it handles nearly any sample and is extremely hard and chemically inert, though it can crack under impact or excessive point pressure. Use ZnSe for routine liquids and soft solids when cost matters, but keep to about pH 5–9 and avoid acids. Choose germanium for carbon-black-filled, high-refractive-index or very strongly absorbing samples, because its higher refractive index produces a shallower penetration depth and avoids total absorption artefacts.

Do I need to apply an ATR correction before searching a library?

Yes, if your reference library contains transmission spectra. ATR spectra have stronger absorption at lower wavenumbers because the evanescent wave penetrates deeper at longer wavelengths. Most FTIR software includes an ATR correction that rescales band intensities for the wavelength-dependent penetration depth; some packages also offer an "extended" or "advanced" ATR correction that additionally corrects the small band shifts caused by refractive-index changes near strong bands.

References

  1. Harrick, N. J. Internal Reflection Spectroscopy. Interscience, 1967.
  2. Griffiths, P. R.; de Haseth, J. A. Fourier Transform Infrared Spectrometry, 2nd ed. Wiley, 2007.