Theory of Calibration
Calibration Types (Linear)

If only one calibration sample of a single standard substance is analyzed for the calibration, you have to enter only one concentration value in the first Amount column of the amount table in the QNT Editor. (For more information about the editor, refer to Data Representation and Reprocessing  The QNT Editor.) The result is exactly one Calibration Point. Connecting the calibration point with the origin forms the calibration curve. It is described by the function derived from the Lambert Beer law:

The slope of the line corresponds to the proportionality factor c1 (leftmost image); c1 is also called RF value.

If one calibration sample is analyzed several times, several points can be entered in the amount/area diagram. The points of one concentration are called replicates. With an increasing number of replicates, the impact of imprecision decreases after averaging. Although several replicates exist, only one amount/area ratio is determined. This is referred to as multi-point calibration on one calibration level (center image).

The result is more reliable if several concentrations are measured. Of course, several replicates can be used per concentration. As a result, calibration points at different concentrations are received in addition to the replicates of one concentration. This is called a multiple point calibration on several levels (for example, 3-level calibration (see the rightmost image)). The calibration curve does not necessarily have to go through the origin. The linear Calibration Function is therefore corrected by an offset.

 Caution: 

The decision whether it is physically sensible to use a calibration type other than the linear default is the responsibility of the user. (With a calibration type that is physically not sensible, no y values may appear for the actually measured values. For example, if, for the Quadratic calibration type, the calibration function is a bottom-down parabola-shaped curve, the values measured above the summit cannot be evaluated. In this case, Chromeleon reports "n.a.")

 Notes on Standard Addition mode: 

In Standard Addition calibration mode, the calibration curve is forced through the average of all unspiked injections, not through the origin.

If you calibrate using the Standard Addition method and if no calibration points with Amount = 0 are available (Spiked samples only), Chromeleon calibrates with an offset.

If calibration points with Amount = 0 are available (Unspiked samples, also) and if you calibrate using the Linear calibration method, the calibration curve does not go through the origin. Instead, it is forced through the mean of all samples for which Amount = 0; i.e., all unspiked samples for this substance. Please note that the results may be different from those obtained by calibrating with the Linear with Offset calibration method.

Average Calibration Factor (AvCF)

Each calibration level defines a single linear calibration curve described by the corresponding calibration factor (that is, the slope of the calibration curve). The curve always goes through the origin. The average calibration factor is computed by averaging all calibration factors of all calibration levels. (Calculation of the calibration curve is not based on the method of least squares.)