The “New Look” on quantitative chromatograms
a real step forward in chromatogram understanding, evaluation, comparison, storage, transmission - at least. This new look shows “Line Chromatograms”

How to get and make them is discussed “here”

Line chromatograms are quantitative and standardised. They are based on the EXPORT standard data file concept.

In the past there was for more than 50 years no improvement seen for better readability, easier (and quantitative) understanding and comparison of chromatograms in gas and column liquid
chromatography. The problems of often too long paper stripes either in evaluation, copying, transmission or just for storage and documentation purposes was finally solved by changing the data into number lines and rows. The latter caused a big loss in chromatogram development and data analysis. But it was only helpful for regulation and standard paperwork.

Gas chromatography analysis covers a very wide range of substance concentrations due to detectors like the flame ionization system covering at least 6 orders of signal magnitude. On paper we can show or see only two orders of signal magnitude when using analog graphics and standard size documents. Thus main peaks are either out scaled or traces are not visible.

This wide range of analytical quantities from the 99 % weight-, mole-, volume- level down to the ppt level of concentration covering 10 orders of magnitudes and its correct evaluation is easily solved by two steps:

The Y-axis in line chromatograms represents concentration data (in a practical unit). The data can be in linear format, based on 100 % for full scale or in any smaller value for a full scale deflection , showing just the highest concentration value of the sample at a selected upper limit. Very often the logarithmic representation is preferable, as this shows the whole analytical concentration range. It allows immediately to see ppb values as well as the main substance concentration. For digital precision we still have the Standard EXPORT file.

The X-axis in line chromatograms represents qualitative chromatography data. This may be the non adjusted (raw) retention time “tms” which equals the sum of the dead time “tm” (better to say “residence time in the mobile phase”) and the net retention time “ts”, better to say “residence time in the stationary phase”. The time data can be the originally measured ones or those corrected by a factor especially for exact line chromatogram comparison. How this looks is shown in the web site part “LIN-LIN”. In case of GC data the X-axis should better be based on retention index values, which are applicable for isothermal as well as pressure or temperature programmed runs. Most informative is the linearized retention index scale as this allows the best comparison of differing chromatograms as seen in this web site under
LOG-LOG.
Equally informative for GC line chromatograms is the “carbon number scale” either as found or linearized. For isocratic HPLC the k-scale is a good choice for the X-axis. It will be shown in this site, that the dead time determination for correct k-values is easily and precisely available in GC and HPLC. The X-axis can be given “as is”, or linearized or logarithmic.

Line chromatograms can easily represent capillary chromatograms of up to 200 peaks.
This quantity is represented in linear or in logarithmic values.

See figure 1 below.

Lineare GC line chromatogram with the solvent peak signal reduced in order to better see the substances with lower concentration.

A practical example showing parts of the high flexibility of “line chromatogram” presentation is a gas chromatogram with a 98% main compound and many substances at low concentration down to 0.01% see figure 2 below.

As the substances analyzed have not yet been qualitatively identified, the peak areas could not get a final quantity value like gram per liter, mole-%, weight-%. Too often chromatographers accept non quantity values like AREA-%. For the necessary transformation into quantitatively correct values like weight-% we need substance specific calibrations. As long as this is not yet done, the Y-axis is named “area-%”. If the substances got no retention index qualification the X-axis must be given in seconds.
As the substance concentrations range over 5 orders of magnitude, only a logarithmic scaling is acceptable. Classically such a chromatogram cannot be represented on standard recorders or printers, as those tools can only show 3 orders of scale magnitudes but already not good enough. Simply the printer lines are too thick.
In figure 2 it is shown which immediate information is available if the analyst wants a quick answer about substances existing above a certain concentration limit - here: above 1 % or if there are ppm-traces.

Y-axis: here logarithmic representation of weight-%

X-axis: here linearized representation of the adjusted retention time ts in seconds

Figure 3 below is an example how to accurately compare chromatograms - here one paraffine sample and the n-alkane test mix for precise chain length identification. The X-axis is given in linearized
C-number values.
Thus equal (saturated normal chain hydrycarbons, n-alkanes) have exactly the same chromatogram position. As the concentration of linear chain saturated hydrocarbons compared to non-n-alkanes (napthenes, aromatic hydrocarbons and others) differs quite strongly and as the concentration range from the short chained alkanes to the high boiling end is also quite different the Y-scale represents the weight-% data in logarithmical scale. This way nearly each concentration level is visible by one look. The analog data representation is available through digital analytical values which are stored in the Standard GC EXPORT files named “sample name.EXP”  - see in the figure “data file C.0000054.EXP”. It contains non adjusted retention times (in seconds), peak width in half height (in seconds), integrated peak area (in seconds times volt or seconds times amperes depending on the detector electronics), and already calculated or manually entered retention index data. See for more in this web site under EXPORT.

Y scale, the QUANTITY scale: LOG scale of concentration ( area % )

X scale, QUALITY scale: here the linearized retention index divided by 100
 = C number. Example C number = 32, n-alkan = C₃₂H₆₆

What may be and what surely is “overlooked” ?

This new chromatogram display concept allows to see nearly everything by “one look” and it can be used for elegant data comparison. It is not necessary to transmit graphics in order to quantitatively communicate with other chromatographers or the research colleagues. Loading the EXPORT data file into the  LIN-LIN, LIN-LOG, LOG-LIN, LOG-LOG line chromatogram software the data display in figure 2 or 3 above. The graphics is available in seconds and the print out time on standard pages depends only on the printer speed. Thus one transmits some few kilobyte ASCII EXPORT attachments by e-mail and not multi megabytes of graphics in binary format or as large PDF file. Thinking about virus, trojaner and phishing problems ASCII eMail attachment transmission is a safer way to communicate with sensitive data.

What we NOT see in line chromatograms is tailing, baseline humps, spikes, peak overlapping and false peak position. If the  separation power is much too small using line chromatograms is of no help but then a qualitative and quantitative data evaluation and transmission is anyway only wasting time.

Many chromatograms - in GC as well as in HPLC - show overloaded solvent peaks with often much too long tailing. If following peaks sit on the slowly ending solvent peak tail the quantitative data not only for these peaks are falsified but the total of data are wrong. There is however no need to have long tailed solvent peaks, as a little packed pre column with a quick backflush keeps the position information of the solvent peak and avoids the falsification of later eluting peak areas. Under those conditions a programmed integration starts quantitation past the solvent signal but still can take and store retention time data precisely.

EXPORT software allows for nearly all necessary corrections but cannot improve false chromatography.

What we also not see in the line chromatogram are peak width values. Although most of the chromatographers see peak width effects only analog but accept that they miss them digitally because their integration program does not take the data or cannot take them correctly. Unfortunately some chromatographers accept to work without any graphics. This is like high speed driving in fog. Peak width data are of great importance for the error analysis. Over broad peak width values are alarm clocks. This becomes obvious when the complete EXPORT data set - data which contain the correct “peak width in half height” values in seconds - are used to graphically show the peak width situation along the retention time scale.

Line chromatograms: How to get them ?

You need an EXPORT reduced raw data ASCII file. This text contains repetitions but the EXPORT concept is too important practically, see for instance the pages “chromatogram combination” for highest precision and accuracy using two or three fully differing analytical systems for one important analysis. Click “here”.

The EXPORT file - if produced according to the advice given here in this site - contains very precise retention time values for each integrated peak.
Very precise does not yet mean “accurate”, but at least time data to the hundreds even thousands of seconds. This precision is by now means practically overdone and easily to reach by qualified chromatogram raw data systems and integration software. Each software company producing chromatography integration programs can add the little piece of EXPORT add-on code, which stores after integration into a beforehand selected folder or disk or memory stick - the data time, width, height, area and one or two text lines containing all relevant details about “when, what, where, how etc. etc”.

Even if only every second a peak height raw data value is taken - but based on hardware, which works on quartz clock driven A-to-D-conversion - it is no problem for a qualified integration program to find the highest peak height value. Now the software calls (at least) three peak height values left from the highest value and three height values right of it. In a fast sub program the software takes these seven height values into a fifth degree polynomial interpolation program and CALCULATES now by interpolation the accurate position of the peak height maximum. This value is used as the corrected peak height and the corresponding calculated time value as the best possible retention time for this peak. This procedure smoothes the signal noise without falsifying the retention time values. Other smoothing routines falsify the peak height value, especially in very fast (micro gas and micro column) chromatography. In addition those smoothing routines offer no precise AND correct retention time data.

Now the integration software integrates the peak area.

The integration software knows all technical details of the A/D converter, the signal amplifier, a possible auto programmable detector/amplifier sensitivity switch and further details of the electronics. Therefore it is possible to give time values correctly in second, peak height values in absolute international standard units like volt or ampere, of course as milli-, micro- nano-units. Thus also the integral can be processed in the mentioned physical units. The reason to have the EXPORT data in international standard units is not only important for quantitative data comparisons - in substance specific ones as well as weight-, volume or other standardized units , but it also allows for the absolute quantitation without the need of critical calibration concentrations. This is a valuable procedure for absolute error checks.

Unfortunately so many chromatographers have forgotten that the peak height represents either a substance flow in grams per second or a concentration in grams per milliliter mobile phase. Therefore the integral “height times time” are absolute values representing grams of substance for each peak. As the chromatographers majority accepts relative chromatography data, the instrument industry and the software programmers take this much easier path to produce “industry standard software”. This reduces precision and accuracy. It makes the detection and reduction of systematic errors a real problem. If the chromatographer would see the separated peaks as grams on a quantity axis instead of 8 to 12 characters digital data in (many) screenful portions and if he would get some ideas about the type of substances - let say as gas chromatography retention index data on a quality axis - then he may immediately get a feel if the analytical results are what he can stand for. He normally produces information for third parties and has to carry responsibility.

Now the chromatographer has the possibility to let calculate the dead time automatically in GC or HPLC. If the instrument runs isocratically, than in GC the time scale can be changed into the best possible qualitative scale “retention index” or in HPLC he can take the k-scale instead of retention time or the widely useless volume scale.

If the runs were programmed and the programs have been kept constant, in GC still we can get correct retention index data. They are precisely and correct calculated by the polynomial interpolation mathematics of fifths degree if there are at least seven index values of n-alkane (hydrocarbons) available enclosing the to be interpolated retention positions. These values must be written into the export file or taken from a qualitative calibration run. The calculation can be done automatically with the proper software (click “here”).

NOTE: what is widely unknown: The HPLC time- or volume- scale can also be treated like the retention index scale in GC. This offers many advantages above a simple time or volume scale.