reagent blank
A reagent blank is a sample that has everything except the thing you are measuring — all the same chemicals, solvents, and steps, but no real sample. It tells you how much signal comes from the supporting cast alone. Think of weighing an empty bowl before you add flour: that empty weight is your blank, and you subtract it so you measure only the flour.
Concretely, you take the reagents and solvents your method uses, carry them through the entire procedure exactly as you treat a real sample, and measure them. Any signal they produce comes from impurities in the reagents, contamination from glassware, or the instrument's own baseline — not from the analyte.
It matters because reagents are never perfectly pure, and even tiny traces add a false floor to every measurement. Subtracting the reagent blank strips that floor away, so the number you report reflects the sample alone. Skipping it is one of the most common causes of results that read too high, especially in trace analysis.
Measuring phosphate in lake water, the analyst runs pure water plus all the color-developing reagents but no sample. It reads 0.02 absorbance — a value subtracted from every real sample.
Whatever the reagents alone produce is subtracted, leaving the sample's true signal.
A reagent blank includes all reagents and procedure steps; a calibration blank usually contains only the solvent or matrix used to make standards. The two overlap but are not identical, and careful methods may use both.