Editorial

Dear Colleagues, The fourth issue of Journal of Planar Chromatography (JPC) 2022 includes ten original research papers on the topic high-performance thin-layer chromatography (HPTLC). Most of them describe a quantitative evaluation of the sample and all of them are worth reading. All of the papers presented use the method of elution chromatography, based on the work of Mikhail Tswett [1]. He was the first to apply the sample onto the stationary phase and then performed chromatography by passing the mobile phase through the stationary phase to elute the analytes. Elution chromatography is by far the most commonly used chromatographic method and is usually referred to when "chromatography" is mentioned. Besides elution chromatography, only the techniques of frontal and displacement chromatography are known. In the frontal technique, the sample is dissolved in the mobile phase and continuously added to the stationary phase. This is the separation method of solid phase extraction (SPE), but is also known in TLC [2]. In the displacement technique, a sorbed component of the sample is displaced (i.e., released) from the stationary phase by a highly sorbing component from the mobile phase. The technique is mainly used in ion chromatography, but is also applied in TLC [2, 3]. According to the theory of elution chromatography by Martin and Synge [4], the analyte distributes between the stationary and mobile phases, resulting in a binomial distribution. The analyte moves through the stationary phase in the form of a binomial function. The Gaussian function (normal error curve), which is mathematically easier to describe, is a satisfactory approximation to the binomial function. We refer to such a distribution as a peak and recognize it by its characteristic bell shape [5]. A successful separation results in a bell-shaped peak for every analyte of the sample. After HPTLC track scanning, the graphical response, called a densitogram, shows all peaks and gives the information of how many analytes are separated. To quantify the analytes, their peaks are integrated. Integration in this context means that all signals belonging to the analyte will be summed up. By comparing this signal area (or height) with a standard signal of an analyte known in content, the amount of analyte in the sample can be calculated. Figure 1 shows an HPTLC densitogram in which a single organic acid is separated after application on a silica gel HPTLC plate using a mixture of toluene and ethyl acetate as mobile phase. The analyst recognizes a pattern of bell-shaped peaks in the densitogram and assigns it to the various analytes. Deciding where a single peak begins and ends depends on our arbitrary recognition of the bell shape. Peak integration is a mini-max calculation method. The aim is to cover as much peak area as possible with a minimum number of data points. The reverse is true for the baseline. The baseline, by definition, is what remains of the densitogram after all peaks have been virtually removed. The aim is to construct a baseline by using a maximum of data points to sum a minimum of signal values. For integration, we construct the baseline as a flat line starting with the noise before the peak and ending with the noise after the peak. Over the years, more and more HPTLC papers have evaluated peaks that deviate from this accepted chromatographic theory. A typical integration of such a peak is shown in Fig. 2, where the peak area is drawn in blue. With this type of integration, the baseline is no longer part of the densitogram. It is now the line where the detector signal is zero, probably on the assumption that no substance is present when no signal is detected. The peak boundaries were indicated by the bell shape of the peak, but then—in a magical process—the bell-shaped peak turns into a more rectangular zone containing a large fraction of the analyte and carrying a Gaussian-shaped peak at the top. In summary, a chromatographic evaluation method is presented that results in non-Gaussian peaks. * Bernd Spangenberg spangenberg@hs-offenburg.de

A successful separation results in a bell-shaped peak for every analyte of the sample. After HPTLC track scanning, the graphical response, called a densitogram, shows all peaks and gives the information of how many analytes are separated. To quantify the analytes, their peaks are integrated. Integration in this context means that all signals belonging to the analyte will be summed up. By comparing this signal area (or height) with a standard signal of an analyte known in content, the amount of analyte in the sample can be calculated. Figure 1 shows an HPTLC densitogram in which a single organic acid is separated after application on a silica gel HPTLC plate using a mixture of toluene and ethyl acetate as mobile phase.
The analyst recognizes a pattern of bell-shaped peaks in the densitogram and assigns it to the various analytes. Deciding where a single peak begins and ends depends on our arbitrary recognition of the bell shape. Peak integration is a mini-max calculation method. The aim is to cover as much peak area as possible with a minimum number of data points. The reverse is true for the baseline. The baseline, by definition, is what remains of the densitogram after all peaks have been virtually removed. The aim is to construct a baseline by using a maximum of data points to sum a minimum of signal values. For integration, we construct the baseline as a flat line starting with the noise before the peak and ending with the noise after the peak.
Over the years, more and more HPTLC papers have evaluated peaks that deviate from this accepted chromatographic theory. A typical integration of such a peak is shown in Fig. 2, where the peak area is drawn in blue.
With this type of integration, the baseline is no longer part of the densitogram. It is now the line where the detector signal is zero, probably on the assumption that no substance is present when no signal is detected. The peak boundaries were indicated by the bell shape of the peak, but then-in a magical process-the bell-shaped peak turns into a more rectangular zone containing a large fraction of the analyte and carrying a Gaussian-shaped peak at the top. In summary, a chromatographic evaluation method is presented that results in non-Gaussian peaks.
* Bernd Spangenberg spangenberg@hs-offenburg.de 1 Offenburg University of Applied Sciences, Offenburg, Germany Is it possible to have a chromatographic process in which a sharp zone of pure analyte appears at a specific location surrounded by unidentified signals? Is it possible that this sharp zone contains pure analyte and outside sharply separated unidentified compounds that we only know are not analytes, and is it possible that this zone was formed during the chromatographic process? Is it possible to have a peak with a "dirty" environment where there is no coelution within the peak? Or in other words, is the occurrence of a non-Gaussian peak in chromatography even possible? The answer is a clear No. In elution chromatography, the peaks always show a bell-shaped binomial distribution. The example presented in Fig. 2 is elution chromatography, not frontal separation or displacer chromatography. It could be that the analyte has displaced all other compounds at its position and thus contains no impurities. This would then be a combined elution and displacement chromatography. In such a case, a bell-shaped peak of the analyte should sit directly on the green line and all displaced compounds should be visible as additional peak(s) in the flow direction to the right of the analyte's peak. Long peak boundaries perpendicular to the baseline are not to be expected in this way. The peak of the analyte would have a bell shape and not a rectangular structure.
Chromatographic separations always yield bell-shaped peaks that are never bounded by flat lines perpendicular to the baseline. Mistakenly, many chromatographs make an exception for overlapping peaks, where instead of sophisticated mathematical peak reconstruction based on Gaussian-shaped signals, unresolved peaks are simply separated by a vertical line, similar to the integration limits shown in Fig. 2. However, one should always be aware that this is an inadequate compromise, and in Fig. 2, we have no additional peaks to the left and right of the analyte peak, so we cannot assume that a sharp separation of the signals into "part of the analyte peak" and "not part of the analyte peak" would be useful.
To conclude, there are some major objections against an integration process as shown in Fig. 2.
• A theory of separation leading to non-Gaussian distributions of analytes would be new and, to my knowledge, has not yet been published. • The baseline in Fig. 2 is arbitrarily constructed from only two data points rather than hundreds as in Fig. 1, so the integration in Fig. 2 is less reliable compared to Fig. 1.   Fig. 1 Separation of a single organic acid on a silica gel HPTLC plate using a mixture of toluene and ethyl acetate as the mobile phase. The integration limits and the peak maximum are marked in red. The baseline is the entire densitogram without the peak, drawn as blue line • In Fig. 2, the peak shape is selected by setting a new baseline in the sense that a detector signal of zero indicates no substance, which is definitely false. A signal is always zero if the (arbitrarily chosen) reference intensity of the detector without sample is equal to the signal intensity. It should be noted that in chromatography, the relative difference between peak and baseline is used for quantification and not the measured data generated by using a blank reference, as is the case in UV-VIS spectrometry. • The newly defined baseline theory assumes that the peak to be integrated is accompanied by defined peaks on the left and right consisting of "dirt", referring to a procedure often (incorrectly) performed when integrating overlapping peaks. • The implied new theory of baseline construction distinguishes between peaks, baseline and additional signals ("dirt"). Scientific reasons are not given why the commonly used definition of the baseline is rejected. • The user of such an integration procedure should be aware that the peak area can change significantly (by about 25% in Fig. 2) just because two data points are chosen more or less arbitrarily as baseline.
• Last but not least, chromatographer who uses such kind of a non-Gaussian integration makes a logical mistake. They identify a peak by its geometrical structure, setting integration limits from its knowledge of having a Gaussian bell-shaped peak and after that they decide not to have a Gaussian peak.
To all colleagues who use non-Gaussian peaks as the basis of their integration method, I warmly recommend to reconsider their approach.
Bernd Spangenberg (Editor-in-Chief) Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.