Infrared Spectrometry (IR/ATR) – Troubleshooting & DIY

Troubleshooting Infrared Spectrometry (IR/ATR)

Developments in ATR have caused it to become the most common form of IR spectroscopy over the past two decades. Its unparalleled convenience has led to widespread use in preliminary and routine analyses. But the deceptive simplicity of ATR can lead to some common, but unexpected issues.

Although ATR is attractive because it requires little to no sample prep, be aware that the prep work has just been shifted to the instrument itself – and is still as important as ever. Since your sample’s spectrum will only be as good as your contact with the crystal – your first step will be proper cleaning.

Prep Work & Preventative Maintenance

The ATR crystal should be cleaned thoroughly at the beginning of every work day, and after any analyses which might leave a residue. These can include foods, oily products, plasticizers, and objects exposed to the outdoors. First, wipe the lens and surrounding area using lint-free optic lens cleaning paper, such as Kimwipes. Next, remove any leftover residue with a fresh wipe soaked in an organic solvent such as methanol or MEK (the higher the volatility, the better).

Important!  Water should never be used! Although some crystals can tolerate exposure, there’s always a better solvent choice. It has high IR absorption, evaporates slowly, and avoiding water altogether will remove the potential for expensive mistakes. Acetone is suitable, but should be avoided – stray droplets will mar the instrument’s plastic surfaces.

After cleaning the surface with a solvent, it is very important to allow for full evaporation before taking a background. If the solvent doesn’t fully evaporate first, the spectrum will be skewed by the solvent peaks. If the ATR window is under an argon or helium gas purge, the close the gas valve and open the chamber to room air. Nitrogen purges can remain in place and will be fully evaporated after ten minutes. Evaporation in standard room air takes approximately 20 minutes. Permanently enclosed chambers require even longer.

Important!  Solvent evaporation depends on its attraction to molecules in the air. Inert purge gases like helium and argon simply don’t have the intermolecular attraction to carry the solvent away at a reasonable pace.

Dry and powdered samples should be cleaned after every use by wiping the area thoroughly with lens cleaning paper. Follow this by using a non-abrasive powder brush to dislodge any stubborn particles which may still be caked onto the sides of the crystal. Once the instrument is clean, double check your method to ensure that it’s set up properly for your source and ATR window. Ensure that you have a reasonable scan resolution and count, and take a new background. Then perform a scan of a standard such as PS, PTFE, or LDPE film. Compare this to a reference spectrum.

Important! If your peaks are shifted from the correct wavenumber, your instrument could need realigned or recalibrated. If your peaks are growing weaker than your reference spectrum, your source could be aging; the ATR window might need polished/replaced; or the clamp could be giving poor contact with the window.

I’ve got this weird band around 3500 cm^-1 that keeps creeping in every few days. I can’t figure it out!

Important! That’s part of the water spectrum – you’re seeing the change in humidity. Controlling the environmental conditions in your lab is an topic unto itself, which we’ve explained in detail here.

If your IR is cooled or purged by gas from a liquid nitrogen tank, condensation is also a possibility.

My spectra used to be smooth but now they’re jagged and I don’t know what happened!

Important! Most likely, the scan resolution was changed somehow. This can happen when a different method gets loaded. Most labs run their instruments at low resolutions for smooth graphs with easily identifiable peaks. Higher resolutions will suffer a signal-to-noise ratio drop. To maintain the same quality in resolution half as wide, you need to double your scan number (and 1/4 width requires 4x the number of scans; and so on). This quickly leads to long sample collection times for instruments which aren’t designed for high-res data collection.

What’s a sine wave doing in my data?

Important! Your sample is probably reflective. If the problem persists through different samples, then a mirror within the instrument may have fallen out of position.

I’m getting results that don’t make any sense!

Important! Spectrum database search software often gives questionable results, especially when analyzing composite / proprietary samples. Create a table of molecular features from the peaks and bands; and supplement your analysis with data from other instruments when needed.

I’ve got a massive backlog of samples to identify or compare. Have you got any tricks that can help me? 

Important!  Try rapid qualitative analysis to quickly identify samples of interest:

First, configure your IR method with a resolution and scan count that produces an acceptable spectrum in less than five seconds. Next, clean the window with solvent, wait 60 seconds, and immediately perform a background. Then run each sample exactly 60 seconds after cleaning the previous residue from the window. The residual solvent peaks will be suppressed by the background. Repeat samples of interest using proper techniques, as these spectra will not be suitable for publication, consumer goods, or physical evidence.


Is there something else that you think should be included on this page?

Let us know with a comment below!