[24] What do Chromatograms tell us? Base line is Rising with a Near Constant Slope..

Chromatograms are like fingerprints.  If you can “read” chromatograms well, you often can find a plausible cause. In this series, we will show a series of GC-chromatograms that are obtained from users and discuss some potential causes for the phenomena. Then we can move into some solutions for improvement.

In chromatography, we always like to get a horizontal base line.  Yes, that would be ideal, but life is not ideal and so is chromatography.(you may quote me on that).  I always get a bit suspicious if I see those beautiful chromatograms that are shown in catalogues, analyzers, new columns..  In real life, chromatograms seldom look like that.  There is always an overloaded, or a tailing/fronting peak, a spike, a ghost peak, a baseline dip or elevation..

Of course we like to have ideal chromatograms, but we also have to be practical about that.

Fig.1 Using alumina column: The base line increases linear with temperature program

Fig. 1 shows an analysis on an alumina column using FID detection. Here light hydrocarbons are separated using PLOT alumina column. There is a strong slope of the base line observed in the beginning which improves after the column is used for 2 days.  What is happening?

This alumina phase provides best selectivity for C1-C6 and is by far better then any siloxane phase.  Several alumina phases are developed the past decades: most recentl an Alumina PLOT column was developed that can be used up to 250C (AluminaBOND / MAPD ) and combined the selectivity of alumina with highest response factors for all hydrocarbons, including the polar ones (methyl acetylene and propadiene).

The practical issue and what sometimes causes a challenge, is that adsorbents have HUGE retention.  That’ means that higher boiling materials are trapped and will slowly “elute” when the column is heated.   The big question is: how do these materials get into the column?

Every capillary column that is made by Restek, has been conditioned for at least 12 hours at Tmax.  This is to ensure the column is “clean”.

Concerning alumina, this material cannot “bleed”. This means that any increase in base line has to come from a hydrocarbon, siloxane or even polar type material that is introduced into the alumina column. Because of the retention these materials will be trapped and elute very, very slow, causing an increase in baseline as shown in fig. 1.

If an alumina PLOT shows high background, there must be cause for this.

Potential causes are :

• Bleed from septa: make sure good septa are used;
• To high inlet temperature:  normally for C1-C6 you do not need high inlet temperature;
• Liner bleed: use  a good liner, consider a “precision” liner.
• Contaminated split-lines; clean these on regular basis;
• Gas lines can be very dirty as metal tubing contains drawing fats. Use clean GC-grade copper tubing and also consider to use carrier as purification. In this case, charcoal filtration may be considered, as they adsorb all hydrocarbons.  Note that molecular sieve traps are not good enough as this material will NOT adsorb the branched hydrocarbons!!

When a TCD  is used, when using a temperature programmed analysis, we often see such a base line. This is caused by small changes in flow between analytical and reference channel. Also sometimes the TCD is not isolated well enough, and the changing oventemperature impacts also the detector temperature.

All points listed are also valid for liquid coated columns like the Rxi-series. Especially thicker films will retain material more and need longer condition times or higher conditioning temperatures to elute possible contaminants.

To check cleanliness of your system, you can keep the column at ambient for 5 minutes and do run to Tmax, without injection. You may already see some small "ghost/system" peaks.  Then do the same, but keep teh column at ambient for 30 minutes. If you see that the peaks are 6x higher you know youare trapping material and you need to check systematically where these peaks originate from.