Chromatography Forum

Since some time I have to deal with 2 old 5890's equipped with a Gerstel CIS2 inlet. Here we are separating a hydrocarbon mixtures from C6 up to appr C30. As I have no experience with this type of inlet I'm trying to understand what happens. I made a mixture of C10/11 and C20 in equal concentrations in a light petroleumether to check column resolution and discrimination. One sysem gives a high concentration of c10/c11, and the other one high c20. A third sysytem , a 2 channel 6890 with an agilent PTV and a split/splitless gives good results on both channels....Can anyone give some help how to tune these Gerstel inlets....I tried some things the with split ratio, but there is also an outlet called "abströhmung" ( I measure 11ml He/min here) but i really dont know what the use is of this flow or what I can do with it.
Some more information: start temp inlet 60C, heating rate 2C /sec, column flow at start temp ( 60C) 15ml/min ( 0.53column) split flow 70ml abstrohmung 11ml/min. I already renewed the septumless sampling head.
Thanks for ur help!

although no one was able to give an answer thank u guys for reading my problem. after some testing I found out that these differences were caused by needle discrimation....the big differences in boiling point in combination with a low boiling solvent and too high inlet temperature were the root cause for this problem. Although the problem hasn't been solved completely yet , I think i will be able to find a way
If anyone has struggled with it before and got something usefull......please don't hesitate .....

Any kind of inlet, except on-column, is sensitive to temperature, gas flow, liner geometry, liner packing, injection speed, pre- and post-injection needle dwell time, split ratio et etc etc. Indeed, the inlet is the weakest link by far in the whole GC chain.

Many (most ?) of the problems with inlets stem from flow disruption when a volatile solvent is injected into an inlet hot enough to evaporate heavy analytes - the usual 250 to 350 C split-splitless injection - or from analyte boiling point discrimination when the inlet is cooler - as with a PTV.

You are probably right about the needle discrimination - during injection at 60 C which is just above/below the BP of your solvent (depending with petroleum ether fraction you are using) you get an inconsistent evaporation / boiling of the sample in or on the needle, which gives eratic transfer of the analytes to the inlet.

With PAHs and pyrethroids in toluene I found that repeatability improved as the temperature of the Gerstel CIS3 PTV inlet during injection was lowered in 20 C steps from 120 C to 60 C - the same as you have found. A faster injection speed also helped - probably by squirting the sample out of and off the needle before it had time to evaporate. When I get a chance I want to follow up on this in more detail.

An alternative to lowering the inlet temperature would be to use a less volatile solvent - I start to think that using hexane, dichloromethane etc even with heavy analytes is GC folklore - does anyone know of any good literature on the influence of solvent BP on discrimination and repeatability for a given set of inlet conditions ??

All the best for 2007

Peter

Splitter discrimination is always present. Peter's comments are excellent as always. DIRECT INJECTION IS MUCH PREFERRED.

Better results require use of a solvent plug behind sample plug in needle,but be careful not to use more than injector volume can tolerate.

If that happens use a slow injection speed to minimize solvent expansion.

Position of needle tip and column tip in splitter inlet is critical to achieve uniform sampling during a split injection.

Older methods also used a larger bore ID column which accepted a larger volume sample plug due to the faster carrier flow.

On column injection might be a solution.


best wishes,

Rod

Peter, you may well have seen this but there are some interesting observations here by Koni Grob.

http://www.restek.com/advantage/adv_koni.asp

Kind regards,

Ralph

These are excellent articles by Dr. Grob. Around the same time I was working on GC methods which had to be performed by capillary columns but even a rather good 1% RSD of sample size was high enough to be unacceptable.

I used special inlet liners which allowed me to have a large inlet volume and which allowed me to connect to capillary columns directly without splitting the sample or using a splitless injector which had its own inherent problems.

BY using direct injection and a small solvent plug behind the sample plug in the syringe coupled with the advantage of a slow injection speed permitted by my Varian autosampler I attained RSD values of less than 0.25% (often much less) for the samples of interest, and some of these were quite difficult.

This also allowed large volume injections to be performed which reduced my LOD by a factor of 5 to 10.

The reduced discrimination of this technique was extremely valuable for the methods I was required to develop. I always expected discrimination in split injections after that work.

best wishes,

Rod

Thanks Ralph for the links - Grob is always one for challenging the status quo !.

Rod, solvent flush is my preferred technique - but not all autoinjectors can do it - Agilents do not, the Gerstel MPS2 does not. It has a few minor drawbacks - total injection volume is bigger so vapour clouds are bigger and solvent purity is more important, but I really cannot understand why major manufacturers do not offer it as an option on hardware that runs under computer control and so should be able to do anything.

Direct injection also has a lot to recommend it - mainly that there are fewer places for sample to go when flows in the inlet fluctuate during injection - but once again there are limitations on the hardware - Gerstel do not make a liner with a press-fit taper at the bottom. I suppose that its main drawback is that it is splitless - which means either diluting the samples more or injecting tiny volumes. Dilution per se should not be a problem - I would guess that a high proportion of analyses have a step in them that actually involves making a solution more concentrated. If analytes elute close to the solvent there has to be a way of sharpening the back edge of the solvent peak (opening the split in classical split/splitless injection) but for heavier analytes, and especially with MS SIM detection, a tail on the solvent peak is completely irrelevant.

I cannot escape the feeling that modern capillary inlets and injection techniques are unsatisfactory compromises dictated by the tradition of getting samples onto a high resolution capillary column via the top three inches of a quarter inch packed column !.

Peter

Peter,

Your comments are valid and I can certainly concur.

I found that examining the size of inlet liners one can sometimes substitute a liner of one brand of GC for a liner of a different GC and get the benefit of the press-fit connection.

I try to avoid being critical of manufacturers of GC hardware but it would appear to me that some of the best selling brands are interested in only addressing the center 80% of the market usage and don't wish to address the needs of the researcher who might need a different and more specialized injection technique.

GC autosamplers should be able to address injection speed and time delay, air gap and solvent wash plug needs. They should be able to wash a syringe from an external and less limited supply of solvent than a simple sample vial.

The option of drying the needle as well is a feature that should be universal.

I have my favorites and will gladly share my preferences privately if anyone wishes to discuss it in a non-public venue.

best wishes,

Rod