Correct calibration in planar chromatography is highly critical. If it is based on classical modes, it is nearly outside the legal limits of mathematical statistics, as we then correlate signal pairs measured in analytical systems which too often are incomparable and falsified by a strong systematic error. This may sound over critical, but the key word in the head line is “correct.” The way out is shown in this paper. It mainly means to reduce and avoid all systematic errors, the main source for a too high analytical uncertainty of quantitative results in planar liquid chromatography (PLC).The main steps to reduce the systematically PLC errors are the following:
Select the best possible area on a plate. It is in the center, not at borders.
Repair the unwanted start chromatography after sampling by a complete sample solvent removal and by focusing the already by part pre separated non-homogenous sample area.
Avoid mobile phase interactions which falsifies RF values. This means: suppress any mobile phase chamber effect simply by notusing any chamber.
Reduce drastically the signal worsening by the “second chromatography” effects during the final mobile phase removal from the plate.
Reduce as strong as possible the plate structure effects at signal integration. Multi integration at mini stepwise position movement of the integration track reduces the structure error by the number of steps.
Apply statistically tested outlier removals.
Apply mathematical statistics to qualify and quantify the data correlation of signal over substance mass in the whole range of the calibration line.
Use the statistically qualified polynomial interpolation mathematics.
This way of calibration pays back in a repeatability standard deviation of about “s = ±0.2 to ±0.05% in top PLC quantification, which can reach the best values of high-performance liquid chromatography (HPLC) quantification. The statement “PLC is a semi quantitative technique thus HPLC application is a must in many analytically important areas” is then disqualified as “nonsense.”
R. Kaiser, G. Gottschalk, Elementare Tests zur Beurteilung von Meßdaten, BI — Hochschul-Taschenbücher Band 774, Bibliographisches Institut Mannheim/Wien/Zürich, 1972, 18–21.