More Than You Ever Wanted to Know About Calibration, Part 7 – Internal Standards, Surrogates, and Isotope Dilution

In my first of this series of calibration blog posts I talked about the differences between types of calibrations, and addressed the use of internal standard calibrations. Since then I’ve received a few questions about the technique, especially about the difference between extracted internal standards and surrogates, so let’s take some time to de-mystify them. We’ll start with some definitions from some EPA methods.

Internal Standard

From Chapter 1 of SW-846 – “Internal standards may be spiked into prepared field samples and QC samples (or sample extracts). Their recovery is generally used to demonstrate the lack of severe matrix effect in the instrumental analysis by setting criteria for the internal standard response in comparison to a response in a sample with a known lack of matrix effect (i.e., a standard). Internal standards are also used to account for matrix effects and/or variability in instrument response by normalizing the response of the target analytes and surrogates, thereby decreasing measurement bias to the extent that their behavior mimics that of the target analytes.”

Essentially, internal standards are compounds added to a sample or extract to monitor matrix effects or instrument drift and generate relative response factors that work to correct for those effects. The monitoring of matrix effects is usually done by comparing area counts from the samples to the most recent calibration or calibration check. They can be added pre-sampling, pre-extraction, post-extraction, or any mix of those. Post-extraction spiking allows for correction of matrix effects and drift of the instrument alone, pre-extraction spiking allows for correction of extraction effects, and pre-sampling spiking allows for the correction for losses or interferences during sampling.

Isotope Dilution

From EPA 533 – “An analytical technique for measuring analyte concentration using the ratio of the peak area of the native analyte to that of an isotopically labeled analogue, added to the original sample in a known amount and carried through the entire analytical procedure.”

Isotope dilution is essentially a pre-extraction internal standard, but specifically uses isotopes of the target compounds. This allows for greater accuracy, as the isotopes should have the same matrix effects as the target compounds, allowing for a more accurate response factor with matrix effects or extraction losses.

Surrogates

Again from Chapter 1 of SW-846 – “Surrogates are most commonly used to monitor the performance of organic analyses using methods such as high performance liquid chromatography (HPLC), gas chromatography (GC), and gas chromatography/mass spectrometry (GC/MS).  Surrogate spikes are added to field samples and QC samples for organic analyses at known amounts, and their recoveries are used to assess matrix effects and, to some extent, verify proper processing and instrument performance for each sample.  The analytes used as surrogates mimic the behavior of the target analyte(s) throughout sample preparation and instrument determination.  Surrogates are organic compounds which are similar to the target analytes in chemical composition and behavior in the analytical process, but are not normally found in environmental samples.  Surrogates added to LCS samples and blanks are used to assess recovery in a matrix known to be free from interference.  This information can be used to determine the magnitude of matrix interference effects on environmental sample results.”

Surrogates are generally added pre-extraction, and are used to monitor matrix effects. Where they differ from internal standards is that they are not used to generate response factors, and instead of using area counts for monitoring they are calibrated to generate true % recoveries.

To summarize, internal standards monitor matrix effects, usually through area counts, and are used to generate response factors that can correct for those matrix effects. Surrogates only monitor matrix effects, usually through calibrated recoveries, but don’t correct for those effects. However, this gets a bit muddied when you look at methods that combine pre- and post-extraction internal standards, like EPA 533, EPA 1633, and OTM-45. In these cases, a pre-extraction internal standard is spiked and used to generate response factors for native PFAS compounds, and the post-extraction internal standard is used to generate response factors for the pre-extraction internal standards, allowing a true calibrated recovery to be calculated for them. In this case, the pre-extraction internal standards are pulling double duty as internal standards and surrogates.

If you’re running an EPA or other official method, the choice on whether to use internal standards or surrogates has already been made for you. If you’re making an in-house method though, there are a few key points to consider. If your matrix has no effects on extraction and your extraction has no losses, then pre-extraction internal standards and surrogates don’t add any value, so it’s better to skip them. If the matrix has no effect on the instrument and the instrument isn’t prone to drift, then post-extraction internal standards don’t add value and also can be skipped.

When choosing internal standards and surrogates, the more similar they are to your target compounds the better. Isotopes of your analytes would be preferred, but that may not be possible or cost effective. In addition, for pre-extraction internal standards it’s critical that the compounds chosen have similar extraction behavior, otherwise you’ll introduce inaccuracies from the internal standard over or under correcting the matrix effects. It’s for this reason that isotope dilution is considered the gold-standard pre-extraction internal standard technique. If you can’t get a comprehensive set of well matching internal standards, you may be better off with a short list of surrogates. This will give you an indication of matrix effects and extraction efficiency without introducing error from mismatched internal standard responses.

View all of the posts in the "More Than You Ever Wanted to Know About Calibrations" series.