Splitless Mode Flow

Chromatography Forum: GC Archives: Splitless Mode Flow
Top of pagePrevious messageNext messageBottom of pageLink to this message  By kalaitzoglou on Wednesday, January 9, 2002 - 12:09 pm:

I don't have experience in gas chromatography so your help would be valuable.
I have an HP 5890(II)GC attached to a 5971MSD.I try to determinate PAHs using the splitless mode. My question is how much should be the flow at the inlet vent. The manual values are inlet vent- split vent 50ml/min , purge vent 3ml/min and column flow 1ml/min.
Can I reduce the flow at the inlet vent (from the total flow control) in order to save carrier gas especially when I don't use the instrument?
Someone told me that it is better to double the inlet vent flow (100ml/min) in order to keep the injector port clean. Is this true?

Top of pagePrevious messageNext messageBottom of pageLink to this message  By David McCalley on Thursday, January 10, 2002 - 02:30 am:

The column flow which is optimum will depend on the diameter of your column, although if you are using something like a 0.25mm ID capillary, 1ml/min. is about right. The low flow of 3ml/min (septum purge)is designed to sweep volatiles from the septum out of the injector, and the flow is not too critical. Flows through the split are usually about 50-100ml/min. The purpose of this flow is to clean the injector of residual sample which does not enter during the splitless period (usually about 1 min.) The higher flow may sweep out residual sample and solvent more rapidly. The biggest waste of gas is clearly through the split vent and you can turn it off after a run. The only problem is that less volatile residues and impurities are more likely to collect in the injection port and transferto the column. If you leave your column at low temperatures when not in use, these impurities are likely to be released on your first temperature program run when the instrument is re-used.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Jason Ellis on Thursday, January 10, 2002 - 08:18 am:

Because the flow through the split vent line is adjusted by the total flow controller, you can never shut down the flow out the split vent without also shutting off column flow. When the split valve is shut in this instrument the split flow is diverted across the top of the inlet and out through the same vent (this is why if you measure flow through the vent while turning the valve on and off you will see no change). Thus, you would not want to try to shut down the flow completely through the split vent, even when the instrument is sitting idle.

I usually recommend a minimum flow of about 30-40 mL/min through the split vent. This keeps the flows in the inlet in an efficient zone and also performs an effective "flush out" of the inlet when the split valve opens.

Running higher flows (up to about 100 mL/min) can help to reduce contamination in the inlet at the expense of higher carrier gas use. This certainly won't cause any harm to your analysis, but it is debatable whether you will see any significant difference in performance by doing this.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Bruce Freeman on Wednesday, January 16, 2002 - 01:09 pm:

I'm not sure I see the logic of a very high purge flow. The typical inlet liner is less than 1 mL in volume, so a 10 mL purge can be expected to purge it of the preponderance of residual sample. At 30 mL/min, this would take only 20 seconds. Since Grob splitless injection requires a refocussing (solvent refocussing or thermal refocussing) at the head of the column anyway, the chromatography doesn't begin until somewhat after this.

The issue is not of keeping the inlet clean, but of purging out the trace of sample remaining after the bulk gets onto the column. Grob splitless injection ideally involves slow injection of the sample into, or slow evaporation of the sample in, the inlet, accompanied by a low flow rate of carrier through the inlet and onto the column. When something like 99% of the sample is on the column (and focussed there) you purge the inlet of the remaining sample, lest it tail out of the inlet and into the column throughout the chromatogram (which would happen if the purge flow were zero). Hence the purge flow is not a critical parameter. More important is the purge time. More important still is the refocussing mechanism (but that's a separate topic).

In short, you can reduce the purge flow (total flow minus septum purge flow minus column flow) to 30 mL/min or less and you'll likely see no problem from doing so.

By the way, the septum purge flow must be kept quite low. 3 mL/min is fine, but much higher than this is not acceptable. Since this flow occurs across the bottom of the septum and the top of the inlet liner, a high flow can actually aspirate the sample out the septum purge vent at the low flow conditions of splitless injection. The reason it normally does not do so is simply that the needle is long enough to get the sample well clear of the septum and septum purge flow.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By David McCalley on Thursday, January 17, 2002 - 05:55 am:

Jason-thanks, the pneumatics of GCs are different. I had forgotten the plumbing arrangement of HP/Agilent models. Certainly, the split flow can be turned off in some instruments rather than redirected.
Bruce-I assume you advocate slow injection since this tends to reduce the possibility of overflow of vapour out of the liner, as the sample can then be carried on to the column gradually? However, slow injection can give rise to a different problem-that is discrimination from the syringe needle, where the less volatile components are not expelled. Thus, discrimination is reduced by fast injection such as from fast autosamplers, where the syringe needle does not have time to heat up, allowing liquid sample to be injected. Of course, then you have to be careful that the vapour volume produced is smaller than that of your insert....
Again, some instruments allow a HIGH carrier flow through the system during the splitless period. This will transport the sample to the column more quickly, and for labile compounds, the residence time in the injector is reduced, reducing decomposition. I think splitless injection is technically very complex-the Grobs have written whole books discussing the above issues.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Bruce Freeman on Friday, January 18, 2002 - 07:04 am:

David - No I don't advocate a slow injection. Our HP 5890's don't give us that option. (Not sure about the 6890's.) We have excellent luck with a fast injection and a slow flow, providing the inlet temperature is not too high and there is a small piece of glass wool in the linlet liner.

This technique is straight out of the book by Konrad Grob, who did very clever and useful research on the dynamics of injection. (I think I mean, "Split and splitless injection in capillary gas chromatography : with some remarks on PTV injection", but I don't have my copy handy to check the title. Anyway, it appears there's a new edition by a slightly different title, as of last year.)

There are a number of important considerations for preventing overflow of the vapors under splitless conditions: Sample solvent, injection volume, inlet liner volume and restrictions, inlet temperature, initial column temperature, initial isothermal time, and septum purge flow. Any of these can adversely affect the splitless injection.

I agree generally with your statement of the general problem of slow injection, but should point out that these problems apply principally to mixtures of analytes of widely different boiling points. (Granted the problem is exacerbated by the low injection port temperatures used in splitless injection.) However, many of us work with well characterized samples, in which we are comparing an analyte, or analytes, in a sample to the same peak in a standard. Under these conditions, needle effects are less important. This is not to say they should be disregarded.

Posting is currently disabled in this topic. Contact your discussion moderator for more information.