A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
An internal standard (ISTD) is a known compound added at a fixed concentration to all samples and standards before analysis, so that the analyte signal is always expressed as a ratio (analyte response / ISTD response) rather than as an absolute value. This ratio corrects for variations in injection volume, detector drift, extraction recovery, and matrix effects — any factor that affects analyte and ISTD equally cancels out. The ideal internal standard is chemically similar to the analyte (so it experiences the same losses and matrix effects), chromatographically resolved from it, absent from the original sample, and stable throughout the procedure. The response factor, defined as the ratio of analyte sensitivity to ISTD sensitivity, must remain constant across the calibration range for quantification to be valid.
Prepare a calibration curve for a GC or HPLC analysis both with and without an internal standard, intentionally varying injection volumes slightly. Compare the %RSD of the two approaches to see how internal standardization dramatically improves precision when injection reproducibility is imperfect.
From your work with calibration curves, you know that quantification depends on a stable relationship between instrument response (peak area, absorbance, etc.) and analyte concentration. In an ideal world, you inject exactly the same volume every time, the detector responds identically from run to run, and every sample behaves like a pure standard solution. In reality, none of these are true — injection volumes vary by a few percent, detectors drift, and real sample matrices suppress or enhance signals unpredictably. The internal standard method solves this by converting absolute measurements into ratios, and ratios are inherently self-correcting.
Here is the logic: you add the same known amount of internal standard to every calibration standard and every sample before any preparation steps. If an injection is 5% low, both the analyte peak and the ISTD peak are 5% low — but their ratio is unchanged. If matrix effects suppress ionization by 20%, both signals drop by roughly 20% — but the ratio is again unchanged. The calibration curve plots the response ratio (analyte area / ISTD area) versus analyte concentration, and samples are quantified from that curve. Because the same ISTD concentration is present everywhere, it cancels out any proportional error that affects both compounds equally.
Choosing the right internal standard is the most important decision. The ideal ISTD is chemically and physically similar to the analyte — it should extract with similar recovery, elute at a similar (but resolved) retention time, ionize with similar efficiency in MS, and be absent from any real sample. In GC-MS and LC-MS, isotope-labeled analogs (deuterated or ¹³C-labeled versions of the analyte) are the gold standard because they are nearly identical in every way except mass, making them the perfect surrogate. When isotope-labeled standards are unavailable or too expensive, a structural analog — a closely related compound with similar functional groups and polarity — is the next best choice. The key test is whether the response factor (RF = analyte sensitivity / ISTD sensitivity) remains constant across the calibration range. If RF drifts with concentration, the ISTD is not behaving like the analyte, and the correction will be unreliable.
A practical subtlety: the ISTD must be added early enough in the workflow to correct for all relevant sources of variability. If you add it after extraction, it corrects for injection and detection variability but not for extraction losses. If you add it before extraction, it corrects for everything — provided the ISTD and analyte have the same recovery. This is why isotope-labeled standards are so valuable: they undergo identical extraction, chromatographic, and ionization behavior, correcting for the entire analytical chain from sample prep to final signal.