Improving linearity

[Johnny Rod]

Can you flesh that out a bit? I'm sure Rod is also interested as HS is one of his specialities.

[chromatographer1]

Rod already is aware of the reasons, but when given a situation as above Rod will suggest the best improvement he can, given that situation.

Doing an 'elongated' injection splitless makes it VERY difficult to focus the more volatile components due to dilution and broadens the deposited sample plug, and increases the time contact of the sample components with reactive injector surfaces. I could go on but this is enough. These reasons don't affect toluene that much but they do for a lot of other analytes.

So why don't folks inject directly using a capillary system?

They don't know how to do it effectively, but it isn't that hard. Change injection system so the transfer line connects directly to the column.

But I am sure everyone is on board when I say it makes sense to inject all of the sample instead of 1/10 or maybe less of the sample.

My problem after I developed a generic HS analysis was getting the sample small enough I did not get peaks that were TOO BIG.

Such problems are nice ones to have, no?

best wishes,

Rod

[Mac2]

A nice problem indeed!!

How do I change the injection system so the column connects directly to the column?
Is there a procedure I can follow?

Typically we (the analysts in the lab) cannot "play" with the equipment as only the Agilent engineers are supposed to.

[chromatographer1]

It is simple: bypass the injector. Take the transfer line and pass it through the injector or the wall of the GC (there are usually holes built in for this, make sure it is insulated to avoid a cool spot) and connect it via a fused silica lined union (I bought mine from Restek) to the mega bore column or precolumn.

You will need to increase the sample flow and increase the head pressure to focus the sample plug. I used a 5 meter piece of 0.25mm ID FSOT tubing connected via another union after the analytical column, before the detector, to increase the pressure to 20-25 psi.

If you increase the flow to 20cc/min you will see chromatography similar to that I showed in my paper (Analytical Chemistry June 1995). Toluene was eluted in less than 6 minutes and I separated 18 solvents in that time span. Methanol was focused (with a very slight broadening) even using a 1ml sample loop as the sample peak width was 1/20 min wide (3 seconds) I was able to effectively use a 1.5mL loop when necessary, which was actually never.

best wishes,

Rod

[Mac2]

That's great Rod. I'll give it a go!!
Cheers

[aldehyde]

Correct, Agilent headspace should be run with a minimum of 10 mL/min flow through the headspace, in a split configuration (no purged packed inlets, no splitless injections.)

If you are using a configuration where you are adding carrier flow from the headspace (ie your transfer line is not plumbed into the inlet carrier supply) you MUST measure your effective split ratio using a flow meter.

As the post above me states, adding flow from the headspace will increase split ratio. Not adding flow from headspace will cause inlet flow to go backward through the headspace and you will see no peaks.

If I wanted to set up a headspace method with 1000:1 split ratio (200 mL split flow) I would decrease split ratio to 100:1 (20 mL split flow) and then make up the same split ratio by adjusting flow from headspace (increase headspace carrier flow until I measure a split flow of 200 mL/min with my flow meter. -- This gives an effective split ratio of 1000:1.)

If you "hard plumb" your headspace into the inlet carrier supply this is not a concern, as any pressure sent from the inlet's epc will pass through the headspace before making it to the inlet. There is no chance of back flow.

In this configuration it is still recommended to have a minimum of 10-15 mL/min total flow.

[TiagoBele]

Hello all!

Mac2, a time ago i was just having the same problem as you. But with a lot more of organic coumponds. So let's go:

An incubation of one hour at 85 degrees it's absolutely unnecessary. A time of incubation of 18 minutes at 70 degrees ( which by the way has been proved to be the best optmized temperature for headspace extraction ... se this application note : http://www.hpst.cz/sites/default/files/ ... -gc-ms.pdf ) is enough. A higher incubation time will make no diference for you ( of course there will be the time and money economy) , and at this temperature, the amount of water in vapor form will be very high, changing your results.

Toluene is very easy to work.
I work with a SIM mode at a 7890A/5975C of agilent. 10 mL of water in a 20 mL headspace vial.
( if you know some portuguese language i recomend this article : http://www.scielo.br/pdf/qn/v33n2/19.pdf (which proves that a 10 mL in a 20 mL headspace vial is the best proportion to work with. 5 mL in a 20 mL vial is not the optimal.)
Try to make all your dilutions at the same time. I hope you have an autosampler... =).

For more information, this topic may be helpfull. Rod also help me a lot in it.

viewtopic.php?f=2&t=21317

best regards!

=)